Now You See Them, Now You Don’t: Biased Macro-scale Site Distribution Maps and Their Influence on Interpretations

Organisers Iza Romanowska (University of Southampton) and Jonas Danckers (Universities of Bologna and Leuven, FWO)

It is common knowledge that the visibility of archaeological sites is largely influenced by their structural properties (e.g. large vs. small sites, low vs. high density sites) and post-depositional processes (e.g. organic vs. inorganic material, surface sites vs. deeply stratified sites). Although archaeologists are aware of this representation problem, it can be argued that during the history of our discipline, certain periods and regions have received repeatedly more attention than others, precisely because of their exceptional archaeological visibility (e.g. surface sites, wetland sites, large sites, architecturally elaborated sites, etc.). Consequently, site distribution maps on a regional or supra-regional scale do not always simply reflect representative ‘patterns of the past’, but can be seriously biased by the ‘research intensity of the present’ or/and regional differences in preservation and the ease of detection of sites (taphonomical factors).

Again, although archaeologists are aware of this ‘modern’ bias in representation, the difference between well and lesser known regions and periods did and still does, often implicitly, influence our interpretations (e.g. a ‘crisis’ for a less investigated period, a ‘golden era’ for a well investigated period; climatic explanations for over- or underrepresentation, economic importance or a ‘demographic boom’ for well documented areas and phases, migrationist explanations for the ‘appearance’ of a well documented phase, etc.). It can thus be argued that a growing awareness of the problem of archaeological representation on the macro-scale has not always led to an equivalent change on the interpretational level.

This session wishes to address this conundrum and compare how archaeologists cope with it. For example, the implementation of the Valetta treaty resulted recently in a serious expansion of the scale and intensification of archaeological research in many European countries. The related use of a more random and often methodologically more thought-out sampling strategy smoothed out some ‘historical’ representation biases on the macro-scale. However, does it suffice to simply change our methodologies in order to alter the traditional interpretations based on earlier representation biases? Or are there other ways of coping with this problem? Can new distribution maps change traditional grand narratives based on older ones? Or are academic power structures stronger than a flood of new ‘dots on a map’?

We invite speakers to focus on this problem on a more theoretical level or with specific case-studies from all over the world.

No more tears. Notes on the spatial structure of development-led archaeology

Speaker: Marc Vander Linden (University of Leicester)

Over the past two decades, development-led archaeology has generated a quantity of data and a spatial coverage far larger than all research-driven work carried out over a century. Yet, the value of this information is often questioned for several reasons, such as quality management issues and biases in the spatial distribution of the corresponding work. Some of these limits are well-known (e.g. linear projects), but others remain to be explored (e.g. spatial correlates of changes in legislative frameworks).

Using an extensive database covering several countries, this paper will explore the spatial structure of development-led archaeology in contrasted environments. Although distribution biases can be shown, their nature can be explained and accounted for by an intimate knowledge of the various legal systems. In this sense, traditional cautionary tales can and must be overwritten in order to stress the positive impact of development-led archaeology upon our knowledge of the past.

Where is the Lower Palaeolithic hiding? Spatial distribution of Lower Palaeolithic sites in Central and Eastern Europe or the lack of thereof

Speaker: Iza Romanowska (University of Southampton)

The pattern of spatial distribution of Lower Palaeolithic (LP) sites east of the Rhine is peculiar. The sites are rare, they do not come in clusters, and they do not seem to be associated with ancient river terraces. This is a robust pattern that has been recognized but not addressed as a distinct research topic so far. It may represent either a real past phenomenon such as climate variability, different dispersal routes ‘out of Africa’ or simply reflects modern research bias. A new alternative is suggested here based on recent developments in geological mapping. It will be argued that an uninterrupted mantel of glacial derived silt (loess) sealing interglacial soil levels may be covering traces of Lower Palaeolithic human activity at significant depths throughout most of Central and Eastern Europe.

Dance of the immaterial bodies: turf, stone and lime buildings of the Western Isles

Speaker: Mark Thacker (University of Edinburgh)

The evidential bias, which privileges monumental material survival, results in Western Isles Medieval and later rural settlements and buildings which are not well represented by plan-forms and settlement maps. In particular, the widespread pre-Improvement use of turf has left little upstanding remains and interpretations extrapolated from later surviving stone buildings of similar plan-form. Ultimately, the visible speaks for the invisible. The differential survival and use of lime, stone or turf in the Western Isles, however, has wider implications. Materials sourced from the wider landscape to animate mobile buildings and settlements in the Western Isles, evade and question modern, static, cartographically inscribed meanings and the distinctions between us and them, outside and inside, visible and invisible.

This paper will present a comparative discussion of this period in Western Isles settlement, in terms of the negotiation of hybridised identities within a networked landscape of relationality. The tension between implicated movement, or dance, and visual metaphors of stasis will be explored. It is argued that a social analysis of the materiality, temporality, mobility and performance of these buildings, subverts previously ahistorical interpretations based upon maps, plan-forms, structure and economy, but also brings to these buildings a relationship of greater significance.

Let’s go to the Plain, there is nearly nobody! The origins of the terramare (Northern Italy) & problems of archaeological representation

Speaker: Jonas Danckers (Catholic University Leuven,  University of Bologna & Research Foundation Flanders)

During the Middle and Late Bronze Age (BM – BR, ca. 1650-1150 BC), the Central Po Plain (Northern Italy), was characterised by the presence of large, regularly organised, ditched settlements. The ‘rise’ of these so-calledterramare has traditionally been explained by a colonisation of the Plain by people immigrating from elsewhere (Cardarelli 2009). It can be argued however that this part of the terramare grand narrative has been influenced by differences in site numbers caused by variable archaeological visibility. Already in the 19th century, the overrepresentation of the pluristratigraphical terramare in the Plain and Bronze Age wetland sites around the North-Italian lakes allowed Pigorini to develop a nationalistic theory according to which Italy was gradually colonised from the North. Less easily observable pre-terramaricoli sites were often lacking still and migration constituted an ideal ‘deus ex machina’ explanation for cultural change. Since the resumption of the studies in the 1980s, it is rather the ‘unnatural boom in settlement density’ during the BM2 that has been stressed, seen as the result of ‘an influx of people from outside’.

The author intends to argue that asking the question ‘Why are these BM2 sites better represented?’ is crucial for revising this theory. BM2 sites are archaeologically better visible because they were (from then on) occupied for longer time spans and often feature embankments. These phenomena can possibly be seen as the result of the introduction of irrigation techniques (resulting archaeologically in a change from low to high density sites). Addressing the representativity issue can help us, in other words, to reflect on what exactly triggered social change.

Prehistoric patterns in a present day landscape. Understanding the distribution pattern of the Bronze Age barrow in north-western Belgium

Speaker: Jeroen De Reu (Ghent University)

The Bronze Age barrows occur in large numbers in the landscape of north-western Belgium. Although their distribution pattern is widespread, it is possible to roughly delimitate micro-areas characterized by a high density of barrows. However, other areas are characterized by a low density, typified by (a) rather isolated (group of) monuments or by a complete absence of monuments. The main questions are how these patterns can be explained and which were the parameters influencing the selection of a suitable barrow location in the landscape. However, beforehand it is important to gain insights in the influence of the prospection method, the present-day landscape and the land-use patterns on the observed distribution patterns. The aim is to detect whether a certain observed pattern corresponds with the historical reality or whether it is biased by current land-use patterns, the limitations of the survey technique or the state of research. In some areas, these patterns are a result of a bias in the database, while in other areas this corresponds with a historic reality.

Roman Wine & Oil Presses Everywhere & Almost Nowhere: Historical Fact or Archaeological Fiction?

Speaker: Dimitri Van Limbergen (Pisa University)
Email: &

Are different regional patterns of press density simply concurrent with different regional economic realities in Roman times, or are there also other mechanisms at play? For instance, press densities ranging from 0.5 to 1.7 presses/km² have been recorded in the North Syrian limestone hills for the Late Roman and Early Byzantine periods. Then again, at present, the average ratio presses/km² for Late Republican and Early Imperial central Adriatic Italy (Picenum et Ager Gallicus) only amounts to a mere 0.003. As a result, the latter region has long been considered economically marginal, while the former area has repeatedly been labelled as a paragon of Late Antique rural economic expansion. Surely, this Syrian-Italian quantitative discrepancy is in part related to unequal levels of regional economic growth in wine and oil production. However, it is less clear how different land regimes or different forms of rural habitation in the past have influenced the archaeological detectability of these presses. Furthermore, which proportion of all past pressing machinery we ultimately recover also depends on a ‘hotch-potch’ of specific environmental conditions, varying local construction materials in antiquity and regionally confined post-antique processes. The author highlights the impact of said phenomena on the possible over- or underrepresentation of presses in these two regions of the Mediterranean.

Pre-Colombian tumuli, pit-houses & caves: the case for a distributional archaeology of the southern Brazilian highlands.

Speaker: Phil Riris (University of Southampton)

In the last five decades, archaeological research on the pre-Colombian populations of southern Brazil and northern Argentina has taken on a variety of influences from French, American and Austrian scholars. Traditionally research agendas included taxonomic methods, detecting diffusionism and building normative culture-historical frameworks (Lopez Mazz 1999; Politis 2003). More recently, the effects of post-processual British landscape archaeology and the widespread adoption of geographical information systems are becoming evident in published work (e.g. Dias 2007; Saldanha 2008). While approaches to the material record have changed, the archaeological site, as a discrete space-time entity, has maintained a pre-eminent position. Consequently, it can be argued that regional frameworks for southern Brazil have been conditioned by the importance accorded to “dots on a map” highlighted in the session abstract – individual archaeological sites and their locations forming distributions amenable to analysis. Following the lead of critiques developed elsewhere (Dunnell and Dancy 1983; Ebert 1992) this paper will examine: a) how archaeologists’ perception of what constitutes an analytically significant “site” influences our models of pre-contact society and b) a potential direction for future research to cover some of the middle ground lost between the site level and regional frameworks, drawing on an example from north-eastern Argentina.

To dot or not to dot? Mapping & interpreting Saharan archaeology with satellite remote sensing

Speaker: Martin Sterry (University of Leicester)

In the last five years satellite imagery of huge swathes of North Africa and the Saharan desert has become freely or cheaply available. Consequently the number of known archaeological sites has risen at an exponential level as projects have been able to increase both their scale and intensity of investigation. Yet interpreting this flood of data by means of conventional distribution maps is problematic. In many areas the archaeological data are contiguous and complex. Settlements take a vast myriad of forms that problematise typologies whilst a range of shelters, huts and other buildings sit amongst fields, roads and irrigation systems that link these sites together. Equally, hundreds of thousands of burial cairns can be identified singly, in small, large and dispersed groups and along landscape features that stretch for more than 100km, defying simple categorisation into cemeteries. Within this palimpsest dating equally becomes contentious. Underlying these problems is the central question of: is the ‘dot on the map’, the ‘site’ still a useful concept? And if not, how do we approach the archaeology without them? This paper will seek to discuss how these issues have been confronted in recent work in the southern Fazzan, Libya.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s