New Developments in Italian Landscape Archaeology: Theory and methodology of field survey Land evaluation and landscape perception Pottery production and distribution. Proceedings of a three-day conference held at the University of Groningen, April 13-15, 2000 9781841714691, 9781407324869

Table of contents :
Front Cover
Title Page
Copyright
Table of contents
Preface
Plates I-IV
1 Introduction: Advances in Regional Archaeological Research in the Mediterranean
2 General Papers
3 Comparative Settlement Archaeology
4 Understanding Digital Archaeological Landscapes
5 Change in Pottery Technology and Production in the Light of Urbanization and Colonization
6 Potential Land Evaluation in Archaeology
7 Landscape Perception in Archaeology: the Urban and Colonial Experience

Citation preview

BAR S1091 2002

New Developments in Italian Landscape Archaeology

ATTEMA ET AL (Eds): NEW DEVELOPMENTS IN ITALIAN LANSCAPE ARCHAEOLOGY

Edited by

Peter Attema Gert-Jan Burgers Ester van Joolen Martijn van Leusen Benoît Mater

BAR International Series 1091 2002

B A R VanLeusen 1091 cover.indd 1

08/09/2009 14:38:32

ISBN 9781841714691 paperback ISBN 9781407324869 e-format DOI https://doi.org/10.30861/9781841714691 A catalogue record for this book is available from the British Library

BAR

PUBLISHING

Table of contents Preface

I

Plates I-IV

II

1

Introduction: Advances in Regional Archaeological Research in the Mediterranean - Douwe Yntema

1

2

General papers

7

2.1

The Aims of the RPC Project - Gert-Jan Burgers

7

2.2

A Short History of Dutch Research in the Mediterranean - Marianne Kleibrink

13

2.3

Two Challenges for Landscape Archaeology - Peter Attema

18

2.4

Settlement Pattern Analysis and Demographic Modeling - John Bintliff

28

2.5

Some Current Approaches to Protohistoric Centralization and Urbanization in Italy - Alessandro Vanzetti

36

2.6

Greek Colonization and Romanization From a Native Perspective - Francesco D’Andria

52

2.7

Pottery Production and Metallurgy - Andrea Zifferero

60

3

Comparative Settlement Archaeology

69

Convenor: Gert-Jan Burgers

4

3.1

Introduction

69

3.2

List of Participants

69

3.3

Working Papers

69

3.4

Discussion

111

Understanding Digital Archaeological Landscapes

125

Convenor: Martijn van Leusen

5

4.1

Introduction

125

4.2

Participants

125

4.3

Position Statements

125

4.4

Discussions

133

Change in Pottery Technology and Production in the Light of Urbanization and Colonization

153

Convenor: Benoît Mater 5.1

Introduction

153

5.2

List of Participants

153

5.3

Some Reflections on the Meanings of Pottery Within Landscape and Settlement Archaeology -

5.4

6

Benoît Mater & Maria Beatrice Annis

155

Post-scripts

169

Potential Land Evaluation in Archaeology

185

Convenor: Ester van Joolen 6.1

Introduction

185

6.2

Speakers and Participants

185

6.3

Aims of the Workshop and Issues For Discussion

185

6.4

Land Evaluation in Archaeology, a Way to Enrich the Reconstruction of Past Agrarian Landscape;

6.5

a case study of Iran - Abbas Farshad

189

Palaeo-soils and Prehistoric Agriculture in Central and Southern Italy - Gaetano Forni

196

7

6.6

Some Considerations on the Use of Archaeological Land Evaluation - Philip Verhagen

200

6.7

Summary of the Discussions

205

6.8

Conclusions

207

Landscape Perception in Archaeology: the Urban and Colonial Experience

211

Convenor: Peter Attema 7.1

Introduction

213

7.2

The Sacred Landscape of the Sibaritide: veneration of ancestors, nymphs and deities - Marianne Kleibrink

213

7.3

Aspects of Urbanization and Ethnic Identity in the Middle Tiber Valley - Gabriele Cifani

221

7.4

Lucanian Sanctuaries and Cultural Interaction - Helle Horsnæs

229

7.5

Perception des hommes et perception de la terre en Campanie septentrionale à l’époque

7.6

de la conquête (V° av. J. - C. / III° av. J. - C.) - Jean Paul Vallat

235

The Geography of the Ritual Landscape in Complex Societies - Andrea Zifferero

246

Preface This volume, based on the proceedings of the three-day international conference ‘Regional Pathways to Complexity’ held in April 2000 at the University of Groningen (the Netherlands), brings together international expert contributions on a broad range of common themes in Mediterranean landscape archaeology. It includes the proceedings of five workshops in which the invited participants have explored in depth the problems and potential of a number of topics which we consider central to the pursuit of landscape archaeology: the comparison of settlement histories across projects and regions, the methods and methodology involved in analysing regional settlement data, the relationship between pottery technology and production and societal processes such as urbanisation and colonisation, the potential of land use models based on land evaluation techniques, and the archaeological study of past landscape perception from an urban and colonial perspective. To students of Italian archaeology, and Mediterranean archaeology in general, the papers and workshop discussions will serve as an excellent introduction to these subjects and their complexities, conveying the state-of-the-art in Mediterranean landscape archaeology. We hope that the present volume will further our goal of developing suitable methods for studying the development of indigenous societies in the context of early urbanization and Greek and Roman colonization. The need for such methods has become increasingly evident since, in the mid-1980s, archaeologists conducting regional research in Italy and other parts of the Mediterranean have begun to realise that indigenous societies played active rather than passive roles opposite the dominant cultures of the Greeks, Etruscans, and Romans. These societies should therefore be studied in their own right as well as in the context of the pressures and opportunities offered by the Greek and Roman colonization. Landscape archaeology can, because of its ability to monitor long-term regional trends of settlements and landuse, fulfill a central role in providing a framework for the study of the patchwork of largely rural societies that characterized Italy in the first millennium BC. As editors we have ourselves been involved for many years in landscape archaeological research projects in Italy, and have found the inavailability of a single-volume overview of the field a great disadvantage. We hope that this volume will prove to be both useful and enjoyable to current and future students of Mediterranean archaeology. It now remains our pleasant duty to thank all those who have helped make the conference, and these proceedings, a success. We thank the participants for the enthusiasm and speed with which they have replied to our pre-conference request for papers and their contributions to the workshop discussions; students from the two organising departments for their invaluable help during the preparation and running of the conference; the Royal Netherlands Academy of Arts and Sciences (KNAW), the Netherlands Organisation for Scientific Research (NWO), and the Groningen Institute of Archaeology (GIA) for underwriting the conference costs; and NWO together with the Archaeological Institute of the Free University of Amsterdam (AIVU) for subventing the preparation and publication of this volume. The Editors

Plate I: Ancient topography of Italy

Plate II: Ancient topography of southern Etruria and Lazio

Plate III: Ancient topography of the Salento peninsula

Plate IV: Topography of the northern Sibaritide

8

The aims of the RPC-project

1 Introduction: Advances in Regional Archaeological Research in the Mediterranean

the past are not just represented archaeologically by larger habitation centers and impressive sanctuaries, but also by villages, farmsteads, huts, small ritual places, fields etc. These less obtrusive remains – though perhaps less spectacular to the wider public - could, of course, not be neglected by those who wished to study the interaction between various aspects of ancient societies and the ’machinery’ behind them1. Therefore, the study of ‘less important’ sites became an absolute necessity, especially because agriculture and animal husbandry appeared to be by far the most important economic activities taking place in pre-industrial societies. As a result of this, a good insight into the rural sector of societies became vital to the archaeologist’s construction (still ‘reconstruction’ at that time) of the socio-political and economic patterns of the past. A substantial change in theoretical approach to the past, therefore, resulted in a new complex of archaeological research strategies, and necessitated the development of a new set of research tools.

Douwe Yntema

Over the past thirty years, archaeologists have conducted regional research programs in many areas of Europe, and not just there: comparable research has also been undertaken, and is still being carried out, in other parts of the world - especially in northern and central America. This fairly sudden and great popularity of regional research is closely linked, it seems, with the rise of the New Archaeology in the late 1960s and 1970s. It should be remembered that one of the major aims of this particular theoretical approach to societies of the past was ‘the explanation of change’. Archaeologists discovered that the method of excavation – usually limited to one or two sites only, because of its highintensity and time consuming character - was not ideally suited to produce the answers to the wide-ranging questions asked in this particular approach of the past. Since the excavated area of any given site represents in most cases only a relatively small percentage of its total surface area, it is unlikely to contribute substantially to our understanding of the settlement as a whole or to produce an adequate data set for the broad research themes formulated by the New Archaeologists - let alone that such excavations suggested explanations for change. In order to find answers to this type of question the research area had to be enlarged very considerably. The rise of regional research programs is therefore probably characteristic of the change-over from an archaeology primarily studying artefacts and artefact assemblages, towards one basically interested in the mechanisms that generated ancient artefacts, patterns in the excavations, and changes in the landscape. At about the same time it was also realized that the larger settlement or sanctuary that was usually chosen as the object of excavation, was decidedly not the only feature from distant pasts that could be studied by archaeologists. Although these larger sites promised the most spectacular results in terms of artefactual evidence, the unraveling of (for instance) the changing social and economic structures of an ancient society was judged to be even more desirable - artefacts had become less important. Many societies of

1

2

At the same time as these major changes in archaeology, occurred the birth of what was to become a close associate of the regional research program: the field survey. Instead of the detailed information about a single small area produced by excavation, field surveys generated more general types of information concerning larger areas. Their Mediterranean version owed much to the old-school prospections in the antiquarian tradition, to regional inventories, to Italian topographic studies of more or less comparable nature, and to the rescue operations in south Etruria initiated by Brian Ward Perkins of the British School at Rome2. Systematic forms of field walking and recovery of artefacts were soon introduced and bias factors of the new fieldwork methods were amply discussed at conferences and symposia. The precise methods used during these field activities, however, differed vastly from team to team3. Each group improved its methods as a result of these discussions, but also learned by trial and error: soils, angle of inclination of the terrain, erosion/colluviation, plowing intensities and types of vegetation differed from region to region. Every research team, therefore, had to adapt its strategies to the specific circumstances within its research area, and solve its own particular set of problems. The ’invention’ of the field survey was in some ways a god-sent: this new archaeological method of fieldwork seemed considerably less costly than the slow and timeconsuming excavations. This was particularly welcome news in the later 1970s and early 1980s, when many Euro-

The so-called subsystems (e.g., social, economic, symbolic/projective, etc.) found in several New Archaeology publications convey the impression of being the wheels of a machine (the system or society). The machine accelerates only when changes occur in two or more subsystems; see, for instance, Renfrew 1972: 485 ff. (The multiplier effect in action). Trends in regional archaeological projects in the Mediterranean have recently been discussed by, for instance, Graeme Barker (1996). For an early and basic paper on regional inventories

3

1

(which directly or indirectly had considerable influence in Scandinavia and the Netherlands in the 1960s and 1970s), see Tode 1926; for the topographical studies, see, for instance, the Forma Italiae series; for a synthetic report on Ward Perkins’ initiatives in South Etruria, see Potter 1979. See, for instance, the volume ‘Archaeological Survey in the Mediterranean Area’ (Keller & Rupp 1983), and especially John Cherry’s concluding paper in this same volume: ‘Frogs round the Pond’ (Cherry 1983).

RPC 2000 conference proceedings

pean university departments experienced severe budgets cuts in a climate of post-1973-oil-crisis austerity. Since this happened mainly in the northern and north-western part of Europe it is therefore perhaps not a complete coincidence that the tool of the archaeological field survey became especially popular in those regions. In any case, the idea spread to many other parts of the English-writing world and soon appeared to be at home in Great Britain, northern America and parts of Scandinavia and the Low Countries4. In these early stages, archaeological field survey may indeed have been considered to be a cheap alternative to cost-intensive excavations. Soon, however, it became clear that whilst it could never substitute for excavation as a basic tool for collecting information about the past, it was at least a very good complementary technique (if not particularly cheap). Initially the main focus of the various field survey projects was often on the rural aspects of ancient societies in the Mediterranean. This was quite a logical development, closely connected - as we have seen above - with the rise of regional research programs. Rural aspects of the past, and the ancient countryside in general, had been almost completely neglected by archaeologists who had focused on the larger sites instead. Town and countryside, moreover, appeared to be closely linked in the ancient world: e.g., the Greek town and the surrounding countryside (ch?ra) together made up the polis. These so-called ‘rural’ field surveys resulted in the discovery and mapping of thousands of surface concentrations of ancient artefacts (often called ‘foci of human activity’), the interpretation of which was hotly debated. There was, however, considerable consensus among survey archaeologists that the ancient landscapes of the Mediterranean had been littered with agricultural units such as groups of shacks, one-family farmsteads, hamlets and large centrally managed rural estates such as Roman villas. Soon it was discovered that small cemeteries and sanctuaries were also part of these rural landscapes. The next step in the development of field surveys was the ’return to the city’ (e.g. Bintliff & Snodgrass 1988). Urban excavations rarely uncover more than a few percent of the total settlement area. These excavated areas, moreover, are rarely representative of the total settlement area. In fact, they were often selected because they promised ‘good results’, i.e. public buildings and other more or less spectacular finds. It was quite clear that in order to gain insight into the dynamics and history of a particular settlement and into the character of that settlement at a particular stage of its development, excavation was not the most suitable method for recovering the most crucial evidence. Important information on urban centers that had not been

built over in recent times could be obtained by applying the by then well-know method of field survey to the towns of antiquity, i.e. by systematically picking up artefacts that lay in the fields covering the remains of the abandoned town. Such procedures could shed light on the growth or decline of larger settlements and might also help in discovering functional diversity between various parts of the settlement (public spaces, residential quarters, sacred areas, storage areas etc.). The enormous quantity of artefacts on the surface of these urban sites offered a new challenge to field archaeologists. In order to deal with it, fine-meshed systems of intra-site subdivisions were developed5. Nowadays, field survey is far removed from the early ‘site hunting’ methods: most field surveys are non-site surveys. The surface of a given landscape is perceived as a continuum in which areas of substantial artefact densities (the so-called ‘foci of human activity’) alternate with areas showing much lower artefact densities: the present landscape in its entirety has become the object of research of the archaeologist. The basic aim of the field survey is to construct images of the landscapes of the past in which both wild nature and manmade nature with their larger and smaller settlements, cemeteries, sanctuaries, tilled fields, grazing areas, lagoons and forests have their place. This also means that attention no longer focuses exclusively on settlements (second stage, ‘urban’ surveys) and areas suitable for agriculture (which were especially popular in the first stage of field survey). So-called ‘marginal’ areas such as mountains and marshes became an object of research since these ‘wild’ areas had also been used by the societies of the past. Woods could be foraging areas for herds of swine, and marshes could offer pasture for sheep and cattle. Both could be hunting areas, but might also have been the home of the good or evil powers that played an important role in the everyday life of ancient communities. So field survey became an important tool in landscape archaeology; it generated questions not only about why people lived at a specific spot (site catchment and site location analysis; already practiced in the 1970s), but also about why people avoided particular spots in the landscape. In the case of ‘avoidance’, land evaluation techniques were sometimes helpful.

The Mediterranean countries of Europe were slow to adopt the new method of field survey. In German-speaking Europe field surveys are still fairly unpopular: the activities of Franziska Lang (Freie Universität Berlin) at Stratos and those of Hans Lohmann in the Miletos area are the exceptions to the rule. For instance, Bintliff & Snodgrass 1988 (Thespiae and Haliartos); Alcock 1991 (Phlius); Perkins & Walker 1990 (Doganella, Albeg-

na valley); Attema 1993 (sites in the Pontine region, southeast of Rome); Yntema 1993b (Valesio); Burgers 1998 (sites in the Brindisi district). Cf. M.R. Jarman, C. Vita Finzi & E.S. Higgs 1972; a fairly influential book at this time was Claudio Vita Finzi’s Archaeological Sites in their Setting (Vita Finzi 1978).

4

5

At present, field survey is only one of several approaches used in regional studies. It is often accompanied and supplemented by other activities, such as archaeological excavations, inventories of old finds (archives, private collections, amateur activities) and geophysical or anthropological research. A series of such activities together, and the integrated databases they produce, can result in dynamic

2

Introduction

images of societies of the past for a particular region. The research questions asked often concern the Braudelian middle range (the conjonctures)6, resulting in publications that sometimes cover a particular region or district from the Neolithic to the present7. Landscapes play an important role in these studies8: initially, the landscape was mostly considered to be the ‘scenery’ against which human actions took place (Braudel’s longue durée). But more recent studies attempt to explain how ancient societies interact with the landscapes they populate: how people adapt themselves to landscapes, and what strategies they devise in order to adapt landscapes to their own aims. Both ancient societies and ancient landscapes appear to be actors in a play about the past. The cognitive archaeology of the later 1980s, and the Cultural Heritage Studies of the 1990s, moreover, have made us aware of the fact that social and cultural significance is applied to landscapes by the societies of both the past and the present. Ancient landscapes, therefore, can also be studied as symbolic constructs (cf. Attema 1996). Regional projects enable research into such topics as ‘regional pathways to complexity’ and ‘the decline and decomposition of complex regional societies’. Since the composition of these images currently requires a great deal of interpretation of evidence coming from various sources, and is strongly influenced by the theoretical model used by the research group concerned, the pictures that are eventually presented for a given region are undoubtedly biased. Though they are all based on large quantities of evidence, they are basically the mental constructs of the researchers. In recent years a large number of regional research projects has been carried out in the Mediterranean. As has been said above, they supply a wide variety of images of regional change. However, we should not be content with a series of regional images. The next logical step to be taken is the comparison of various regional developments and the integration of these regional pictures into supra-regional images. It might, for instance, be quite interesting to compare the changes that took place in the area around lateRoman Rome to those occurring in the area surrounding late-Roman Constantinopolis. Ideally, of course, research into both late-Roman towns themselves should be carried out within the same project. It would be equally interesting if we could make a comparison between the developments in various districts of Italy for a particular period, and compare this probably highly variegated ‘Italian’ picture to similar images of a group of Micro-Asiatic districts. But already it is clear that we are in serious trouble if we wish to compare developments in two or more regions and

draw up supra-regional pictures of change. Such a supraregional comparison is necessarily based on images constructed by different teams with different approaches. Susan Alcock’s Graecia capta represents an excellent effort in creating pictures of the Roman landscapes of Greece (Alcock 1993), but it is patently clear from the texts that the general image is constructed on the basis of regional images that display a substantial variety in research methods, in types of research, in research intensity, in theoretical and methodological approaches and in ways of presenting the research results. All these elements severely hamper the presentation of an overall picture of Roman Greece. Research carried out in different regions and by different teams, therefore, has to be made compatible in order to integrate the results of regional research programs and produce supra-regional images of the past that give an insight into the long-term development of larger areas. If, for instance, we wish to create images of Italy in Roman-Imperial times without relying exclusively on ancient written sources or major monuments, then we should agree on at least some of the basic procedures in regional research projects. This observation holds true especially for field surveys. One of the major problems here is that field survey reports, more often than not, are very incomplete: they mostly contain unenlightening descriptions of the methods used. Field surveys are, for instance, almost invariably described as ‘systematic’, but it is quite unclear what this is supposed to mean; and nobody, of course, carries out an unsystematic field survey. A second and very common omission in field survey reports is the presentation of only a selection of the finds9. Publications regarding regional projects often supply a narrative about the societies of the past. But whilst it is almost standard procedure in excavation reports to publish artefacts on which the dating and other types of interpretation are founded, such crucial pieces of evidence are sadly missing in most field survey reports. Those who wish to check datings or evaluate the general developments as sketched in the regional image created by the author(s) are badly served: for lack of artefactual data the results of a very high percentage of field surveys cannot be calibrated or checked. Perhaps that uncanny feature, shared by many field survey reports, has been caused by the formerly almost universal belief that surveys can be repeated when necessary. It may, in fact, be true that field surveys are nondestructive (especially when we leave the finds on the surface after having recorded them), but the idea of repeatability now seems rather naïve in view of the intensive forms of soil improvement practiced in the Mediterranean, often with European Community funds. In the Mediterranean,

Fernand Braudel’s ideas played and continue to play a major role in this type of research (especially Braudel 1972); for the influence of the Annales School, see Bintliff (ed) 1991. See, for instance, C. Renfrew & M. Wagstaff 1982. The word ‘landscape’ is used here in the strict, traditional sense,

not in the metaphoric way used in Alcock’s Graecia Capta (Alcock 1993). Cf. Alcock’s mottos in her conclusive remarks to one of the Aix POPULUS colloquia: ‘Have you hugged your pot person today?’ and ‘Be explicit’ (Alcock 2000: 265-266).

6

7 8

9

3

RPC 2000 conference proceedings

the traces of societies of the past are vanishing with breathtaking speed, on the surface as well as in the subsoil10. For these very reasons we should think very seriously about establishing standard procedures and asking for a standardized set of information for each field survey, both from ourselves and from our colleagues. Only then will we be able to compare and integrate the results of various regional research projects on a basic level, and form the basis on which to draw up supra-regional images and discover supra-regional parallelisms or inter-regional divergences. Such a standardization in reporting results and presenting data should, of course, not take the form of a compulsive set of strict regulations. If only we succeed in drawing up a few basic rules for field procedures and reporting, much would be gained. We should recall that are there no strict laws dictating how an excavation should be carried out either, but interpretations based on very different theoretical approaches and data recovered with vastly different excavation methods are still being compared all the time.

tionalization and colonization, exploitation and marginalization’, started in 1997. It was sponsored by NWO (Netherlands Organization of Scientific Research) and the participating Universities (Groningen University and Amsterdam Free University), and ended in 2002. Its aims, participants and first results will be discussed in Burgers’ paper below (chapter 2.1). An international RPC conference was organized in April 2000 in order to exchange views with colleagues who were or had recently been involved in very similar research activities in other parts of Italy. Our own attempts to compare the increasingly complex societies in three Italian districts had revealed that such a comparison was hampered by a substantial set of methodical and methodological problems - even though the approaches and methods used by both the Groningen and Amsterdam teams were fairly similar. Since other regional projects carried out by nonDutch teams have tended to used rather more dissimilar methods and strategies12, one of the major aims of the conference has been to spot and inventory the methodical problems that come to light when we try to put a series of regional research programs into a larger supra-regional framework. Since we wished to avoid that the complications arising from this comparative approach reached almost global dimensions, we decided that the conference should focus geographically on Italy (and not on a larger segment of the Mediterranean), and chronologically on the 1st millennium BC. Might we indeed discern parallelisms, convergences and divergences in various regions of Italy on the basis of a larger series of fairly recent regional research projects? At the very least we expected that there were useful lessons to be learned in creating a higher degree of compatibility between future regional research programs. The volume that now lies before you presents the proceedings of the Regional Pathways to Complexity conference; it contains both the general and introductory papers presented in the opening session, and detailed accounts of the often very lively workshop sessions in which individual themes were pursued. We hope that its contents may help all of us to take the next step in the closely linked fields of landscape archaeology and regional archaeology: the construction of long-term, supra-regional pictures of the past.

In 1995 the Netherlands Organization for Scientific Research (NWO) decided to fund archaeological research groups that had amassed and processed substantial regional databases, intending to provide an extra stimulus for synthetic studies concerning these relatively wellresearched regions. The general framework devised for this program required that applicants should formulate projects aiming to give insight into the interplay between human settlement and landscape (both wild nature and manmade landscapes), preferably in larger regions and over a substantial period11. The project team was to consist of one or two senior researchers and three or four junior researchers, all financed by NWO. In the fall of 1995 three of these ‘multiple’ projects were approved. One studied long-term developments in the southern part of the Netherlands from the Bronze Age to the Middle Ages (Meuse-Demer-Scheldt area) and was carried out at Leiden University and Amsterdam University; a second studied the early Middle Ages around the North Sea, focusing especially on the northern part of the Netherlands (the ‘Frisia’ project, carried out at Groningen University and Amsterdam University); whilst the third project targeted regional research in Italy. The latter project, entitled ‘Regional Pathways to Complexity’ (RPC) with the subtitle ‘Towards an archaeology of early urbanization in Italian landscapes: three coastal landscapes of central and southern Italy between interna-

10

A check-up on 20 small concentrations of artifacts (all smaller than 0.2 ha) discovered during the Oria field survey (1981-1983) revealed that 17 of these could still be traced in 1985, that 9 survived into the 1990s (check-up in 1992) and that only two of them seemed to have survived in the year 2000. All of the large scatters in the same area, presumably referring to Roman villas, were of course still highly recognizable. For the report on this ‘rural’ field survey, see D.G. Yntema 1993a.

11

12

4

The general framework of the research program ‘Settlement and Landscape in Archaeology’ drawn up by the author, was accepted by the Foundation for Historical Research (SHW) of the Netherlands Organization for Scientific Research (NWO) in the spring of 1995. It should be observed that a third Dutch group working in Italy (Leiden University) used very different field survey methods; for the most recent relevant publication, see Van de Velde 2000.

Introduction

References

Francovich, R. & H. Patterson (eds) 2000, Extracting Meaning from Ploughsoil Assemblages. The Archaeology of Mediterranean Landscapes 5. Oxford: Oxbow Books. Jarman, M.R., C. Vita Finzi & E.S. Higgs 1972, Site Catchment Analysis in Archaeology, in: P.J. Ucko, G.W. Dimbley & R. Tringham (eds), Man, Settlement and Urbanism, London: 6166. Keller, D.R. & D.W. Rupp (eds) 1983, Archaeological Survey in the Mediterranean Area, BAR International Series 155. Oxford. Perkins, P. & L. Walker 1990, Survey of an Etruscan city at Doganella in the Albegna valley, Papers British School at Rome 58:1-143 Potter, T.W. 1979, The Changing Landscape of South Etruria. London: Dent. Renfrew, C. 1972, The Emergence of Civilization. London: Methuen. Renfrew, C. & M. Wagstaff (eds) 1982, An Island Polity: the Archaeology of Exploitation in Melos, Cambridge: Cambridge University Press. Tode, A. 1926, Organisation und praktische Durchführung einer allgemeinen archäologischen Landesaufnahme, Vorgeschichtliches Jahrbuch 3:10-21. Van der Velde, P. 2000, An extensive alternative to intensive survey: point sampling in the Riu Mannu Survey Project, Sardinia, Journal of Mediterranean Archaeology 14(1):24-52. Vita Finzi, C. 1978, Archaeological Sites in their Setting. London: Thames & Hudson. Yntema, D.G. 1993a, In Search of an Ancient Countryside. The Free University Field Survey at Oria, Province of Brindisi, South Italy (1981-1983), Amsterdam: Thesis Publishers. Yntema, D.G. 1993b, The Settlement of Valesio, Southern Italy. Final Report on the Field Survey in the Town Area, Bulletin Antieke Beschaving 68:49-70.

Alcock, S.E. 1991, Urban survey and the polis of Phlius, Hesperia 60:421-463. Alcock, S.E. 1993, Graecia Capta. The Landscapes of Roman Greece, Cambridge, Cambridge University Press. Alcock, S.E. 2000, Extracting meaning from ploughsoil assemblages: concluding remarks, in: Francovich & Patterson (eds) 2000:265-266. Attema, P.A.J. 1993, An archaeological survey in the Pontine Region. Groningen. Attema, P.A.J. 1996, Inside and outside the landscape. Perceptions of the Pontine Region in Central Italy, Archaeological Dialogues 3(2):176-195. Attema, P.A.J., G.-J. Burgers, M. Kleibrink & D.G. Yntema 1998, Case Studies in indigenous developments in early Italian centralization and urbanization: a Dutch perspective, European Journal of Archaeology 1(3):327-381. Barker, G. 1996, Regional archaeological projects. Trends and traditions in Mediterranean Europe, Archaeological Dialogues 3(2):160-175. Bintliff, J. & A. Snodgrass 1988, Mediterranean Survey and the City, Antiquity 62 (1988, no. 234, march):57-71. Bintliff, J. (ed) 1991, The Annales School and Archaeology, Leicester/London, Leicester University Press. Braudel, F. 1972, The Mediterranean and the Mediterranean World in the Age of Philip II (revised edition, trans. S. Reynolds). London: Collins. Burgers, G.-J.L.M. 1998, Constructing Messapian Landscapes. Settlement Dynamics, Social Organization and Culture Contact, Dutch Monographs on Ancient History and Archaeology, vol XVIII. Amsterdam: Gieben Publishers. Cherry, J. 1983, Frogs round the Pond: Perspectives on Current Archaeological Survey Projects in the Mediterranean Region, in Keller & Rupp 1983:375-416.

5

RPC 2000 conference proceedings

6

The aims of the RPC-project

2

General Papers

2.1

The Aims of the RPC Project

regions are denied a significant role in the social process, and reduced to mere passive receivers. The RPC project views analytically the very Greek and Roman impact itself, and moreover studies it as one of numerous factors determining the regional processes of centralization and urbanization. Others include the geographical setting and the natural environment, the technological and organizational level of societies and the social strategies played out in the regional arenas (Attema et al. 1998-1/2). In this way, we wish to emphasize that the processes we study are complex: highly differentiated in space, time and intensity. Investigation is directed to the myriad interrelated changes occurring in the cultural domains of society; such as the economy, ritual, funerary practice and socio-political relationships. The basis for inference in these social domains are the results of our study of transformations in physical landscape, as well as in patterns of land use, settlement structure and artefact production, and organization. These transformations typically occur on the time scale of the longue durée, when they are environmental, or on the time scale of the conjoncture, the socio-economic cycles that societies are subject to. In the project I have outlined these dynamics are observed from a regional and a macroregional standpoint, so as to detect the interaction between internally-driven regional processes and the wider interregional frameworks they are part of.

Gert-Jan Burgers 2.1.1 Introduction The full title of the RPC project is Regional Pathways to Complexity, Landscape and Settlement Dynamics in early Italy. The primary aim of this project is to study processes of centralization, early urbanization and colonization in three Italian landscapes during roughly the 1st millennium BC. The regional perspective closely fits in with the tradition, well established in Dutch archaeology, of regional projects, both in the Netherlands and in the Mediterranean. In 1996, this tradition culminated in a research program proposed by the Netherlands Organization for Scientific Research (NWO), entitled ‘Settlement and Landscape in Archaeology’. Three projects were subsidized, two concerned with northwestern Europe and one with the Mediterranean, this last is the RPC project. The project, a joint undertaking by the Groningen Archaeological Institute (GIA) and the Archaeological Institute of the Vrije Universiteit of Amsterdam (AIVU), started in the late summer of 1997, and has completed its planned initial period of four years. It brings together researchers specializing in settlement and landscape archaeology, environmental reconstruction, artefact analysis and spatial analysis. There are six staff; four PhD students appointed to carry out primary research on interrelated topics: Esther van Joolen and Martijn van Leusen at the GIA, and Froukje Veenman and Benoit Mater at the AIVU. Prof. Peter Attema of the GIA, and the author on behalf of the AIVU, are responsible for the day-to-day management of the project and for the final integration of the various studies. Other senior scholars involved in the project are Prof. Marianne Kleibrink and Dr. Bert Nijboer of the GIA, and Prof. Douwe Yntema of the AIVU. Researchers in complementary disciplines are collaborating with us. An international advisory board has been appointed.

Figure 1 shows the three regions constituting the core research areas of the RPC project: the Pontine Region in Latium, the Salento Isthmus in Apulia and the Sibaritide in Calabria. There is a rich tradition of Dutch archaeological

2.1.2 Comparative regional archaeology We are aiming at a multidisciplinary and comparative assessment of regional trajectories towards complexity in protohistoric Italy, and at the same time investigating responses to and interaction with Greek and Roman colonialism. A major incentive is the wish to revise the Greek- and Roman oriented approaches to ancient urbanization in Italy. Of these it is the urban success of Rome and of the Italiote Greek poleis that attracts most attention. By contrast, the emergence of urbanism in the greater part of Italy is generally seen in the diffusionist perspective; it is often believed to have been enhanced by the pervasive strength of Greek and Roman colonization and urban culture. In these traditional perspectives of Hellenization and Romanization, the indigenous peoples of most Italic

Figure 1: The research areas of the RPC project.

7

RPC 2000 conference proceedings

Figure 2: Digital elevation model of the Pontine region. 1-15: areas surveyed by the Groningen Institute of Archaeology in the period 1987-1999. 1: Lanuvium survey (1995), 2: Segni survey (1997), 3: Cisterna survey (1990), 4: Cora transect survey (1987), 5: Ninfa survey (1998-9), 6: Norba survey (1995), 7: Caracupa/Valvisciolo site survey (1988), 8: Olmobello survey (1992), 9, 10: Norba transect survey (1987), 11: Contrada Casali site survey (1988), 12: Sezze survey (1994), 13: Selva Forcella site survey (1995), 14: Sezze transect survey (1987), 15: Fogliano survey (1998-9)

2.1.3 The Pontine Region Figure 2 shows the digital elevation model of the Pontine Region, which has been surveyed over the past decade in the context of the so-called Pontine Region Project, directed by Prof. Peter Attema. The project includes transect surveys, urban surveys and rural surveys of catchment areas of Roman colonies. It aims to define long-term developments in settlement patterns and land use of the Pontine Region from the late Bronze Age until late Antiquity. Thus, it sets out to create a context for the study of urbanization and Roman colonization in this region. The study area lies between 60 and 80 km south of Rome. Three main physiographic units can be discerned in this landscape. The first is that of the Pontine plain, an

fieldwork in all three regions. The datasets generated by projects still ongoing constitute the starting point of the RPC research. The aims of these projects, funded by the participating universities, by the Netherlands Organization of Scientific Research (NWO) and by the Royal Netherlands Academy of Sciences (KNAW), closely coincide with those of the RPC team. Moreover, they were developed in collaboration with Italian superintendencies and universities, notably the superintendencies of Latium, Calabria and Apulia and the Scuola di Specializzazione in Archeologia Classica e Medievale of the university of Lecce. We sincerely thank our Italian colleagues for their constant collaboration and hospitality. The projects will be briefly summarized in the following. 8

General Papers

Figure 3: Digital elevation model of the Salento Isthmus, with indication of the AIVU survey areas.

region surveys make up a total area of some 45 km2, from sample areas on the lower slopes of the volcanic Alban hills (e.g. at Lanuvium), the alluvial parts of the Pontine plain (e.g. at Valvisciolo and Sezze) and the slopes and uplands of the Lepini mountains. In the context of the RPC project these were supplemented by intensive surveys in the beach ridge/lagoonal area along the coast in 1998 (Attema, van Joolen & van Leusen 2001). These latter surveys have been combined with geomorphologic, vegetation and sedimentologic research as well as with land evaluation techniques. The data collected so far in the various landscape units give us a fair idea of the major transformations in settlement and land use in the Alban hills and the Pontine region.

almost flat coastal area with ancient beach ridges. In the northwest this plain is delimited by the volcanic area of the Alban hills, in the northeast by the limestone mountain range of the Monti Lepini. The settlement data for these areas consist, broadly speaking, of four types of records. A first dataset is constituted by inventories of incidental finds reported to the archaeological superintendence. A second by controlled sets of topographical data mapped by Italian topographers and published in the so-called Forma Italiae series. A third and major dataset for the region is the reported results of excavations, notably those at Satricum carried out by the GIA and the University of Amsterdam, directed respectively by Prof. Marianne Kleibrink and by Dr. Marijke Gnade (see notably Maaskant-Kleibrink 1987, 1992; Gnade 1992; 2000). Long-term excavation and documentary research into the protohistoric site of Satricum have brought invaluable insight into the development of an early Latin settlement, within the framework of regional socioeconomic and religious dynamics. Our fourth and final source is controlled data from the systematic surveys carried out by the Pontine Region Project, and the Agro Pontino Survey of the University of Amsterdam; and the data from Italian survey work (see especially Voorrips et al. 1991; Attema 1993). The Pontine

1

2.1.4 The Salento Isthmus Figure 3 shows the digital elevation model of the second of the three regions central to the RPC project, the Salento isthmus in Apulia. This is a ca. 80 km wide stretch of land between Taranto and Brindisi connecting the Salento peninsula to Italy. For this region the RPC project draws largely on the databases elaborated in the context of the ongoing Brindisino program of the Archaeological Institute of the Vrije Universiteit of Amsterdam (AIVU)1. Prof. Johannes Boersma and Prof. Douwe Yntema began research here in 1981.

Antieke Beschaving and in Taras. Notiziario delle attività di tutela.

Preliminary reports of the fieldwork carried out in the context of the Brindisino-project are published annually in the Bulletin

9

RPC 2000 conference proceedings

The Brindisino forms the eastern half of the Salento Isthmus. The larger part is a slightly undulating plain. Starting from the Adriatic, it rises very gradually to approximately 60m above sea level. This landscape consists mainly of light arable soils. By contrast, in the immediate coastal zones one finds an alternating landscape of dunes, low cliffs and lagoons. To the south, the plain merges into the more hilly, calcareous landscape of the Serre Leccesi. To the west and north, the Brindisi district encompasses part of the hard limestone spurs of the Murge uplands, a plateau which gradually rises up to the Apennine mountain chain. The general theme of the Brindisino-project is the Romanization of the region. This process is studied from a long-term diachronic perspective with ample research devoted to the social dynamics of the indigenous group in the pre-Roman period from as early as the late Bronze Age. A major aim of the project is to study the relationship between social dynamics and the organization of settlement and landscape. To that aim, in the 1980’s and 1990’s various teams carried out a series of intensive field surveys, complemented by environmental research.

the streams entering the plain from the mountains. The Sibaritide has been the subject of research by the GIA since 1990. In this project attention is focused on a comparison of the indigenous colonization in the Bronze and Early Iron Ages with Greek colonization movements in the Sybaris plain. In line with the RPC theme, the general aim of the project is to retrieve the non-dominant archaeological history of the native societies from the dominant archaeology of the Greeks. The database we work with in this region is again made up from a variety of sources. The main object of study has been a system of Bronze Age hilltop sites. One of these, at the top of the Timpone della Motta, near Francavilla Marittima, is a sanctuary, with settlements of huts and houses occupying three lower terraces. This site is currently under excavation by Prof. Marianne Kleibrink (Maaskant Kleibrink 1993; Kleibrink 1998). Aside from the records of these excavations, the datasets comprise a vast inventory of sites and scatters compiled by Italian topographers. We also make use of a set of data on the protohistoric periods collected in fairly recent times by Prof. Renato Peroni’s team (Peroni et al. 1994). To this set we may add a number of regional surveys carried out by the GIA in the course of excavations at the Timpone della Motta. Finally, to add to these, in 2000 the RPC team carried out problem-oriented intensive surveys on the terraced landscape in the transitional zone between the plain and the foothills, and extensive surveys of the inland mountainous zone (Van Leusen & Attema, in press).

The field walking carried out so far covers a total surface of some 100km2, incorporating environmental zones throughout the ca. 600 km2 Brindisi region. The areas investigated are shown in figure 3. The field walking ranges from total coverage surveys of catchment areas of major ancient towns, like Oria and Valesio (Burgers 1991; Yntema 1993), to urban surveys like those at Muro Tenente and Muro Maurizio (Burgers 1998) and transect surveys crosscutting various physiographic units, as at Ostuni in the Murge area (Attema, Burgers & van Leusen forthcoming). The latter surveys were started in 1999 in the context of the RPC project, and were continued in 2000. The survey results have been supplemented with the archaeological data extracted from archival and museum research, and from publications of sporadic find reports and incidental forms of field walking, such as those inventoried by Profs. Lorenzo and Stefania Quilici-Gigli in the Brindisino (Quilici & Quilici-Gigli 1975). Both our surveys and the sporadic find reports were frequently tested by augering and by excavations of well-preserved stratigraphies. Further, in the context of the regional project large-scale excavations were carried out at a number of large fortified sites, where surveys had been conducted, for example Valesio and Muro Tenente (Boersma & Yntema 1987; Boersma 1995; van Alberda et al. 1999). These excavations provide the necessary refinements of chronological, functional and spatial insight.

2.1.6 The research objectives The above-discussed research projects provide the major data sets with which to meet the RPC aims, as well as those of individual, closely-related PhD studies. As stated, the project is devised as an integration of four of such studies. I will now briefly introduce them. They are further discussed in the various workshop sections. GIS methodology and applications form the core of the PhD project of Martijn van Leusen. All the regional site records just mentioned are being inventoried in a single site database for the purpose of analysis in a Geographical Information System. Using information on settlement distribution and land use potential provided by other project members, his research aims to use and develop locational modeling techniques derived from catchment, visibility, and network analysis and to model the dynamic processes of colonization and urbanization occurring in the study areas. Furthermore, Martijn van Leusen has been coorganizing the problem-oriented fieldwork with which we try to make up for gaps in the regional archaeological records. These notably concern the archaeology of marginal or extensive exploitation areas such as wetland and highlands in antiquity. These areas are the lagoonal Fogliano area in the Pontine region, the Murge plateau near Ostuni in the Brindisi region, and the highlands and terraces in the vicinity of Francavilla Marittima in the Sybaris region. Martijn is elaborating the data of these surveys, integrating them in his GIS-research2.

2.1.5 The Sibaritide Figure 4 shows the digital elevation model of the third study area of the RPC project, a region known as the Sibaritide in Calabria, South Italy. The model shows a quite different landscape from that of the two other regions. It largely consists of an alluvial plain. To the north, this plain is delimited by a crescent-shaped mountain range. The plain gives limited access to the mountainous hinterland. Inland routes are largely restricted to the wide riverbeds of 10

General Papers

knowledge, locational factors and the socio-political and economic contexts are analyzed within the wider framework of the project. The primary goal of the research is to distinguish between ‘internal’ progress and external stimuli in pottery technology and production - which of these speeded up or slowed down technological improvement in pottery production, or possibly even led to a decline4. The PhD project by Froukje Veenman is aimed at the reconstruction of actual (as opposed to potential) land-use. Different types of pastoral land use are reconstructed on the basis of data from off-site find patterns, ethnographic parallels, and archaeo-zoological analysis. In this study, the influence of pastoralism on the structuring of the landscape is emphasized. The extent and significance of pastoralism is reconstructed on the basis of an analysis of published bone data from twenty south- and central Italian sites. Historical and ethnographic research must shed light on the pastoral strategies used in the three study areas. Exploitation patterns emerging from ethnographic and historic sources are used to interpret the bone data and to build GIS models for land use from the Bronze Age to the late Empire. Finally, changes in land use are interpreted in the light of contemporary processes of urbanization and colonization5.

Figure 4: Digital elevation model of the Sibaritide, with indication of the GIA survey areas. 1: Monte Sellaro survey (1998), 2: Foothill survey (1996), 3: Francavilla Marittima survey (1995), 4: Sibaritide survey (2000). Grid size: 10km.

The four studies discussed above, investigating aspects of centralization and urbanization processes, will be integrated with the analyses of the settlement data available for the three regions and the fieldwork carried out by the RPC team. From the outset these four studies have adopted a comparative approach. As emphasized, the comparative approach is central to the project, especially when we consider the observed regional differentiation. In this way, we put forward our contribution to the debate on internally-driven versus externally-triggered regional developments. The discussion within the RPC team focuses on such questions as: • what major changes in settlement behavior and land use occurred, and what changes in craft technology and organization can we detect? • in what way did landscape organization change, and what were the ecological consequences? • What do these changes tell us about the character and intensity of regional centralization and urbanization processes and about the influence of colonization on the indigenous societies? • How did they affect the social construction of the landscape and the ways in which the landscape was perceived? • How did internally driven regional processes interact with the supra-regional frameworks they were embedded in?

The PhD project of Esther van Joolen is aimed at the reconstruction of past potential land use. With that objective, she carries out a land evaluation study adapted from that of the FAO, integrating it with test inspections of a series of new pollen cores for the study areas. The FAO system, recently adapted for archaeological use, is being extended by Esther van Joolen to handle land-usage and land qualities specific to the study areas. The project employs the so-called Automated Land Evaluation System (ALES) as well as GIS software in order to produce landsuitability maps. It also uses information on past actual land-use and agricultural technology, provided by herself and the other RPC-members3. The theme of the PhD project by Benoît Mater is the scale, organization and technological development of pottery production in the context of early urbanization and colonial movement. She uses ceramic analysis in order to reconstruct local pottery systems through time. The regional cases are subjected to socio-economic analysis applying ethnoarchaeological models to pottery technology and production at various stages of complexity. Access to technological

2

3

P.M. van Leusen 2002, Pattern to Process, methodological investigations into the formation and interpretation of large-scale patterns in archaeological landscapes, PhD thesis, University of Groningen. Van Joolen’s PhD thesis, on Potential land evaluation in archaeology (a reconstruction of ancient landscapes for land utilization types in Italy during the 1st millennium BC), will be submitted to the University of Groningen in the spring of 2003.

4

5

11

Mater’s PhD thesis, on Pre- and protohistoric pottery production in the course of urbanization and colonization in three Italian regions, will be submitted at the Free University of Amsterdam in the spring of 2003. F. Veenman 2002, Reconstructing the Pasture, PhD thesis, Free University of Amsterdam.

RPC 2000 conference proceedings

Burgers, G.-J. 1991, The Valesio Field Survey, Bulletin Antieke Beschaving 66: 124-131. Burgers, G.-J. 1998, Constructing Messapian Landscapes. Settlement Dynamics, Social Organization and Culture Contact in the Margins of Graeco-Roman Italy. Amsterdam: Gieben. Gnade, M. 1992, The Southwest Necropolis of Satricum, Amsterdam. Gnade, M. 2000, Satricum in the Post-Archaic Period. A case study of the interpretation of archaeological remains as indicators of ethno-cultural identity, University of Amsterdam (PhD thesis). Haagsma, B.J. 1996, Survey in de Sibaritide, Calabrië. Een preliminair verslag van drie campagnes. In Tijdschrift voor Mediterrane Archeologie 17: 47-52. Kleibrink, M. 1998, Enotrian and Greek Settlers in the Sibaritide, in: Attema, Burgers, Kleibrink & Yntema 1998-2: 330-342. Leusen, P.M. van & P.A.J. Attema in press, Regional Archaeological Patterns in the Sibaritide: preliminary results of the RPC field survey campaign 2000, Paleohistoria 42/43 (2000/2001). Maaskant-Kleibrink, M. 1987, Settlement Excavations at Borgo Le Ferriere . Vol.1: The Campaigns 1979, 1980, 1981. Groningen: Forsten Maaskant-Kleibrink, M. 1992, Settlement Excavations at Borgo Le Ferriere . Vol.2: The Campaigns 1983, 1985, 1987. Groningen: Forsten. Maaskant-Kleibrink, M. 1993, Religious activities on the Timpone della Motta, Francavilla Marittima - and the identification of Lagaria, Bulletin Antieke Beschaving 68: 1-47. Peroni, R., F. Trucco, C. Belardelli et al. 1994, Enotri e Micenei nella Sibaritide, Taranto. Quilici, L. & S. Quilici-Gigli 1975, Repertorio dei beni culturali archeologici della provincia di Brindisi, Fasano. Voorrips, A., S.H. Loving & H. Kamermans (eds), The Agro Pontino Survey Project. Methods and Preliminary Results (Studies in Prae en Protohistorie 6), Amsterdam. Yntema, D.G. 1993, In Search of an Ancient Countryside. The Amsterdam Free University Field Survey at Oria, Province of Brindisi (Italy). Amsterdam: Thesis Publishers. Yntema, D.G. 1993-2, The Settlement of Valesio, Southern Italy. Final Report on the Field Survey, Bulletin Antieke Beschaving 68: 49-70.

The research is organized around two main poles, each comprising two complementary themes: on the one hand settlement and land use studies based on archaeological and ethnographical research, and on the other landscape and technology evaluation founded on palaeo-environmental research and technological studies. Applying the same criteria, methods and research organization to each of the three regions, and using a joint GIS, it is hoped that we may begin to understand the factors generating the processes under study and so come closer to a landscape archaeology that will contribute to the definition of a theory of centralization and urbanization in early Italy.

References Alberda K. van, G.-J. Burgers, H. Burgers, D. Karel & D. Yntema 1999, Muro Tenente. Centro messapico nel territorio di Mesagne, a cura di A. Nitti, Manduria. Attema, P.A.J. 1993, An Archaeological Survey in the Pontine Region, a Contribution to the Early Settlement History of South Lazio, Groningen (PhD thesis). Attema, P.A.J., G.-J. Burgers, M. Kleibrink & D. Yntema 1998-1, Centralisation, early urbanisation and colonisation in a regional context, Dutch excavations and landscape archaeology in central and southern Italy, Saguntum 31: 125132. Attema, P.A.J., G.-J. Burgers, M. Kleibrink & D. Yntema 1998-2, Case studies in indigenous developments in early Italian centralization and urbanization, a Dutch perspective, Journal of European Archaeology 1(3): 326-381. Attema, P.A.J., G.-J. Burgers & M. van Leusen (forthcoming), Walking the Murge: interim report of the Ostuni Field Survey (Apulia, Southern Italy), Studi di Antichità 2002. Attema, P., E. van Joolen & M. van Leusen 2001, A Marginal Landscape: Field work on the beach ridge complex near Fogliano (South Lazio), Palaeohistoria 41/42 (1999/2000). Boersma, J.S. & D.G. Yntema 1987, Valesio. History of an Apulian Settlement from the Iron Age to the Late-Roman Period, Fasano di Puglia. Boersma, J.S. 1995, Mutatio Valentia. The late Roman Baths at Valesio, Salento, Amsterdam.

12

General Papers

2.2

A Short History of Dutch Research in the Mediterranean

ed from the fifties in Italy, first as an assistant to Buchner on Ischia, and subsequently to Paola Zancani Montuoro, excavating the Heraion near Paestum and later the Athenaion at Francavilla Marittima (Stoop 1971, 1972, 1977, 1983-1990). Many of these archaeologists are now deceased and it is fitting, at least to our mind, that tumuli be erected for them.

Marianne Kleibrink Asked to open this conference, which I sincerely hope will become one of the landmarks of Dutch Mediterranean archaeology, a strong desire arose to first look back, to try and place this conference in perspective. I hope you will allow me that flash back; after that I shall take the opportunity to formulate a few points on the new tasks for fieldworkers of today in the Mediterranean.

2.2.3 The second generation The second generation of field-archaeologists, now retired, semi-retired or about to retire, either followed up those pioneers’ initiatives or started new ones: Bakhuizen’s work led for instance to work by Hans Boersma in Ostia and Apulia (Boersma 1975, 1985, 1990, 1995) as well as by Reinder Reinders in Halos (Reinders 1990). In the meantime Dutch involvement with Roman wall-painting had led to work on Pompeii by Wim Peeters and Fred Bastet (Bastet 1958; Bastet & de Vos 1979; Peters 1963, 1971, 1993; de Waele & Moormann 1984). Two of the experts working with Vermaseren, Herman Geertman and Maria Beatrice Annis, married and continued their studies of, respectively, Roman monuments (Geertman 1975) and Sardinia; Stoop’s initiatives resulted in fieldwork by Coen Stibbe (on the work of Coen Stibbe see Stibbe 1980 and Gnade 1991) and myself at Satricum (Maaskant-Kleibrink 1987, 1992), followed by Groningen campaigns at Francavilla Marittima (Maaskant-Kleibrink 1972, 1974-1976, 1993, 1996, 19961997, 2000a, 2000b; Attema, Delvigne, Drost & Kleibrink 1997/1998). The story of the start of the Dutch Institute’s and Groningen University’s campaigns at Satricum (Borgo le Ferriere, 60 km south of Rome) is too good an example to keep back of how chance has played its part. The work at Satricum began because in the middle of the 1970’s the walls and foundations of the famous holiday resort of Queen Juliana, the ‘Elefante felice’ near Porto Ercole on the promontory of Monte Argentario, showed dangerous cracks and an Italian architect was asked to estimate the damage. This man was convinced that near Porto Ercole an Etruscan town had existed. He persuaded people around the Queen to call in archaeologists from the Dutch Institute in Rome to evaluate his idea. As Maria Stoop was the only Dutch archaeologist with sufficient experience in Italian archaeology, she was asked to accompany Coen Stibbe, the just appointed archaeologist at the Dutch Institute, to investigate the matter. The short excavation undertaken by Stoop, in which Marjan Galestin and myself from Groningen University also participated, resulted in an interesting medieval house with a corner tower and higher up lots of Roman amphorae. In other words, there was no Etruscan town in that particular spot. The excavation at Monte Argentario, however, had attracted Dutch and Italian ambassadors as well as directors of Universities, who all decided that in future it would be a good thing for the Dutch Institute in Rome and the Dutch Universities to carry out excavations in Italy. Fortunately, in the meantime archaeologists around Massimo Pallottino had decided to invite the foreign institutes in Rome to participate in rescue excavations in Latium, because the uncontrolled growth of

The 20th century and Dutch fieldwork in the classical world The Dutch Departments of Classical Archaeology until long after the Second World War hardly ever undertook fieldwork in the Classical lands. The Dutch School in Rome was very much a stronghold of historians, especially church- and art-historians. A School in Athens was lacking altogether. Back in those days archaeologists were either interpreting archaeological objects by means of classical texts, the work on the Attic Choes by professor Van Hoorn comes to mind (van Hoorn 1951) or working with art historical principles like for instance the volumes on Pompeian wall painting by Prof. Hendrik G. Beyen demonstrate (Beyen 1938, 1958). After Carl Wilhelm Volgraff, who worked early in the 20th century with the French School in Athens (Volgraff 1928, 1931), only the much admired Amsterdam professor Caroline Henriette Emilie Haspels – widely known because of her work on the Attic lekythoi (Haspels 1936) - was in far away Phrygia. However, her fieldwork made little impact in Holland, probably because the publication appeared so late (Haspels 1971). 2.2.1

2.2.2 The pioneer generation of field archaeologists Fortunately from the 1960’s onwards a number of Dutch archaeologists pursued their own fascinations in the Mediterranean: of this first generation, which we may duly call the “pioneers”, we mention: professor Jan Willem Salomonson, director and promoter of fieldwork in Tunisia (e.g. at Uzita, see Van der Werff 1982) and specialist on African mosaics (Salomonson 1965; 1979); Maarten Jozef Vermaseren who excavated underneath the S. Prisca in Rome (Vermaseren & van Essen 1959, 1965) and built a famous network of historians, mostly, but also of archaeologists, who published the iconographic and textual evidence on oriental religions in the Roman Empire (see the series M.J. Vermaseren, Etudes préliminaires aux religions orientales dans l’Empire romain, Brill, Leiden and London, for the work of Vermaseren see De Boer & Edridge 1978); Cees Bakhuizen who was interested in Chalkis and was quick to understand the importance of minerals to developments in the Late Bronze Age and Early Iron Age. He also was the first to introduce survey methods to Dutch archaeologists in the Mediterranean (Bakhuizen 1970, 1976, 1985). And last but not least Piet Stoop who excavat13

RPC 2000 conference proceedings

Rome threatened many protohistorical sites. The English traditionally worked in Etruria and had no opportunity to participate, so it was the Scandinavians who started at Ficana and the Dutch at Satricum (Satricum 1982, 1985).

young archaeologists in field work. The initiative of Ko Feye to maintain the post of secretary has now fortunately led to the Dutch School in Athens, being opened in a respectable building – at last.

Those were the optimistic 1970’s, and the 2nd generation of archaeologists has paid for that optimism. In the Netherlands the Classical archaeologists were either part of the Classical Languages Departments which provided education in classical culture, but were unwilling to pay for fieldwork, or belonged to Art Historical Departments which lacked sympathy for the study of ‘ugly’ material. Consequently, fieldwork archaeologists did not have the backing of draughtsmen, conservation archaeologists, physical anthropologists or laboratories. Moreover, there was no equipment for measuring or for photography. The process of first obtaining and then educating the necessary fieldstaff, as well as the transmission of fieldwork know-how of the Mediterranean to students became an exhausting affair without much result. This is because in the early 1990’s, the few extra staff members we had got hold of were sacked during the severe cutbacks at the Universities. Now, fortunately, most Departments of Pre-and Protohistory (or in ugly words ‘national’ archaeology) in Dutch Universities have finally been forced to lift the ban on ‘foreign’ archaeologies, thus also on Classical and Mediterranean fieldwork, and it may be hoped that by combining the efforts of several archaeologies fieldwork in the Mediterranean by the Dutch Universities will lose its amateurish character. For much of the research undertaken in the ’70 and ‘80’s praise must go to the Netherlands Research Foundation, which did support field campaigns and researchers in Italy and Greece not, as in the early days, with Landrovers and cameras, but with the fairly regular funding of excavation projects. Hans Boersma, very much part of the second generation of field archaeologists, started after his university education, as I did, with a job in the Royal Coin Cabinet at The Hague, busying himself with Roman coins (Boersma 1963; van Gelder & Boersma 1967) while I studied engraved gem stones (Maaskant-Kleibrink 1978). That somber Hague environment provoked dreams in him of Roman villae and in me of protohistorical peoples in Italy, dreams we pursued as University professors in later years. This small personal tale may serve to illustrate that in those times one was supposed to have a specialist field of research in classical archaeology, fieldwork ambitions were a holiday extra. As said, the fieldwork of this second generation in the Mediterranean was only halfheartedly backed by the Universities. The support of the Dutch Institute in Rome was in most cases altogether lacking – a sad outcome of the 5year terms for the appointed archaeologists as the program of each faded away when the job in Rome was over. The researchers in Greece only could fall back on the help of a Secretary – who worked a few hours for the Dutch archaeologists. This must have been frustrating for the director of the Dutch Survey School in Athens, professor Te Riele, who at Lavda in the Peloponnesos educated quite a few

2.2.4 The third generation Now I come to the most important part of this talk: the third generation of Dutch archaeologists working in the Mediterranean, the generation in full swing and fortunately involving not an occasional handful of archaeologists as in the past, but all the Universities with Classical departments, each with fixed projects (sadly the excellent work by Salomonson and Bakhuizen did not induce the University of Utrecht to keep the archaeology department after it decided it no longer needed Classical Languages and Cultures in its programs). First to be mentioned certainly is professor Douwe Yntema, who belongs to the aforementioned generation as well, because he started his fieldwork very early and assisted Bakhuizen and later, Boersma. Another 2nd generation characteristic is that his ‘true’ field of research, ancient pottery, lies in mainstream archaeology (Yntema 1985). This famous researcher of the South-Italian countryside and its cultural and economical processes, must be regarded further as an important promoter of Dutch fieldwork throughout the Mediterranean region (Yntema 1993). The second we name is professor Peter Attema, who began his fieldwork as a student in the Groningen Satricum campaigns and started the Groningen surveys in the Pontine region (Attema 1993). The University of Amsterdam continued working at Satricum under the directorship of Marijke Gnade (Gnade 2000), while Professor Crouwel, another archaeologist with interests in mainstream archaeology, obtained permission to carry out fieldwork in the Peloponnese (Crouwel 1979). The University of Nijmegen, committed to Pompeii since the work of Wim Peters, Jos de Waele and Eric Moormann, carries on with its Pompeiian evaluations (Mols 1994 (1999)). The experience of the Leiden archaeologists profited Dr. Peter van Dommelen who works in Sardinia (van Dommelen 1998). Leiden University’s fieldwork, moreover, has recently been much strengthened by the appointment of professor John Bintliff, whose work in Greece is widely known and reported. It is a well-known saying among anthropologists that the third generation of a country’s immigrants must be regarded as autochthonous. This means that the recent, third generation of Dutch field archaeologists is fully at home in the Mediterranean. Our Dutch excavators and field archaeologists are indeed University-educated in Mediterranean field archaeology and, although they still suffer from the same malfunding and academic misapprehensions of what archaeological fieldwork is about, they are certainly better established in international archaeological enterprise in the Mediterranean. Also, the equipment in the Institutes now is more or less adequate. For all that, adverse new developments in the Universities and in the Mediterranean add up to inferior working conditions. In the Dutch Universities there are very few posts available for Mediterranean 14

General Papers

was taken up in the 19th century and split the Mediterranean into culturally advanced and well-liked areas, e.g. Athens and Corinth, and barbarian areas; which resulted in archaeological research of the dominant areas (by the American and British Schools, but if the Dutch had had a chance they would have been there too) and a steady loss of the archaeology of other important areas. We must remember that there still is, in Southern Italy and Sicily, an entire secret archaeology living off the export of ancient objects. In view of these things, we must not fool ourselves into thinking that what previous generations of Classical archaeologists made of Mediterranean archaeology was good, even if these archaeologists offered much of their time, money and even health to the cause. They were too caught up in the rivalry between European and American governments to discover key sites of classical culture.

archaeology, although the teaching traditionally involves three curricula – ancient art-history, Classical culture and field archaeology – and is therefore heavier than in comparable disciplines. It has never been compensated by more staff. Neither has the fact that obtaining one’s data from fieldwork is a more lengthy process than obtaining them from a library. Consequently, many of our best Mediterranean archeologists live large parts of their lives as nomads – fighting to survive from one project or national research grant to another. For the students some things have changed for the better, fieldwork in the Mediterranean nowadays is a regular part of the study program, but others have changed for the worst: the 4-year curricula of Dutch universities make it impossible to schedule sufficient information technology as well as exams in the ancient languages and Italian and/or New Greek, and stretches of time in the field or/and in the Institutes in Rome and Athens. The fieldwork programs are interesting, so most students tend to choose them, consequently Mediterranean archaeology in Holland is again unbalanced – lacking now in historical education.

The strategy of study proposed by Dutch archaeologists of the 3rd and 4th generation offers the hope of new perceptions: they propose to put aside the old evolutionary models, which in Italy have always reigned: by this I mean the postulated development of humanity from a pre-Neolithic hunter-gatherer existence, to that of Neolithic farming communities, interrupted by eneolithic warrior-pastoralists, followed by Bronze Age pastoralists-cum-settlers and an Iron Age existence of town-and-country peoples.

It has been hard to install Mediterranean archaeological fieldwork in Dutch University departments; the work is often misunderstood as is much other archaeology. The misunderstanding for the larger part stems from the popular misunderstanding of history. For most people history is a fixed thing, a thing each nation worth its salt possesses and a thing you can read about in books or put away in a museum. This leads to the idea that the archaeologist only has to discover the past, or in other words, that she/he only has to uncover a ready thing under her/his feet. It still is not common knowledge that this 19th century, fixed idea of history has led straight to the racial wars and murders of today. Archaeology should teach people the many changeable aspects of past cultures. Archaeology, like other disciplines, has to be constructed and deconstructed all the time. It is also not common knowledge among archaeologists, the Mediterranean ones included, that their ideas of ‘pure’ cultures, for instance those they have been reconstructing for Greek and Roman males, have done much damage. Among other things they have promoted misogynistic cultural attitudes. In my opinion unbiased, regional research, as proposed by the RPC group, is a good means of showing how broad a variety of arrangements for organizing their life human societies have contrived. Any new concept of history, to which archaeology can contribute, must beware of dominant story telling, superior cultures, star people or regions, and humbly start again.

Some RPC researchers will look into land-use and pastoralism without preconceived ideas about evolution, and I hope their workshops will inspire successful discussion of what it is possible to undertake in this field of research and what the main questions ought to be. Other researchers will be looking into pottery technologies, not so as to construct pure typo-chronological series to promote a dominant strain of culture, but to study different technologies and uses of pottery. Fortunately, in the past everybody used pots, not just Greeks and Romans, so we may be sure that from this entryway to the complex reality of ancient societies much can be learned. It will not be easy to develop methodologies which can be used to good result. Experience has shown me what enormous amounts of time are consumed by pottery study, time the universities nowadays have no patience with. So everything depends on the development of good methodology. The GIS data and systems used in this project should make it possible to show what occupational patterns are demonstrated in the several societies under scrutiny. I am aware of the fact that this will be a long process, since every Classical archaeologist already carries a map of the ancient world in his or her head: there is much sea on it, there are two centers: Athens in the East and Rome in the West, and if you look better you see other coastal towns - either Greek or Roman. What you do not see is the native people, the poorly marked areas of the hinterland. What you also do not see are the enormous gaps, for instance the 7th century BC gap in the south of Italy and the 5th century gap in the Roman region. We have been asking, what is in these gaps? Was everybody moving and fighting as the ancient literature

2.2.5 New tasks A new start must bring to the archaeologist ways and means of studying the peoples of the past without constructing dominant genders or dominant ethnicities – and this will not be easy, because the archaeological record, and especially that of the Mediterranean, is full of bias. The Mediterranean is the region where the idea of pure and dominant Graeco-Roman culture was created; this idea 15

RPC 2000 conference proceedings

Dommelen, P.A.R. van 1998, On colonial grounds: a comparative study of colonialism and rural settlement in the first millennium BC west central Sardinia, Leiden. Geertman, H.A.A.P. 1975, More Veterum, il ‘Liber Pontificalis’ e gli edifici ecclesiastici di Roma nella tarda antichtà e nell’alto medioevo, Groningen. Gelder, H.E. van & J.S. Boersma 1967, Munten in Muntvondsten, Bussum. Gnade, M. (ed) 1999, Stips votiva, papers presented to C.M. Stibbe, Amsterdam. Gnade, M. 2000, Satricum in the Post-Archaic period: a case study of the interpretation of archaeological remains as indicators of ethno-cultural identity, Amsterdam. Haspels, C.H.E. 1936, Attic Black-figured Lekythoi, Paris. Haspels, C.H.E. 1971, The Highlands of Phrygia, 2 volumes, Princeton. Hoorn, G. van 1951, Choes and Anthesteria, Leiden. Maaskant-Kleibrink, M. 1978, Engraved Gems in the Royal Coin Cabinet, The Hague, The Greek, Etruscan and Roman Collections, Den Haag. Maaskant-Kleibrink, M. 1987, Settlement Excavations at Borgo le Ferriere vol. I: the campaigns 1979, 1980, 1981, Groningen. Maaskant-Kleibrink, M. 1992, Settlement Excavations at Borgo le Ferriere vol. II: the campaigns 1983, 1985 and 1987, Groningen. Maaskant-Kleibrink, M. 1972, Abitato sulle pendici della Motta, Atti e Memorie della Società Magna Grecia n.s. 11-12: 75-80. Maaskant-Kleibrink, M. 1974-1976, Abitato sull’altopiano meridionale della Motta, Atti e Memorie della Società Magna Grecia n.s. 15-17: 169-74. Maaskant-Kleibrink, M. 1993, Religious activities on the Timpone della Motta, Francavilla Marittima - and the identification of Lagaria, Bulletin Antieke Beschaving 68: 1-47. Maaskant Kleibrink, M. 1996, Le scoperte recenti sul Timpone della Motta, I Greci in Occidente. I santuari della Magna Grecia, Napoli: 198ff. Maaskant-Kleibrink, M. 1996-1997, Dark Age or Ferro I? A tentative answer for the Sibaritide and Metapontine plains, in: Debating Dark Ages, CAECVLVS 3, Papers on Mediterranean Archaeology: 63-91. Maaskant-Kleibrink, M. 2000a, Enotri, Greci e I primi culti nell’Athenaion a Francavilla Marittima, Magna Graecia 35: 18-30. Maaskant-Kleibrink, M. 2000b, Early Cults in the Athenaion at Francavilla Marittima as Evidence for a Pre-Colonial Circulation of nostoi Stories, in: V. Gassner, M. Kerschner, U. Muss & G. Wlach (eds), Die Ägäis und das westliche Mittelmeer, Beziehungen und Wechselwirkungen 8. bis 5.Jh.v.Chr. (Akten des Symposions, Wien 1999), Verlag der Österreichischen Akademie der Wissenschaften, Wien: 165185. Mols, S.T.A.M. 1999, Wooden Furniture in Herculaneum: form, technique and function, Circumvesuviana, Nijmegen. Peters, W.J.T. 1963, Landscape in Romano-Campanian mural painting, Assen. Peters, W.J.T. 1971, Van Prima Porta naar Pompeii: de tuin in de Romeinse schilderkunst van de laatste jaren van de republiek en de eerste eeuwen van de keizertijd, Nijmegen.

suggests, and did European city life emerge from that fighting – or, was everybody ill and was this the first period of the malaria-bug – as has been suggested by historians? Summing up again - now that the research is liberated from fixed ancient histories, and ancient colonization has been liberated from recent European colonialism, the questions are almost too many to answer in the span of one generation. So, as you will have gathered from my words, the title “Regional pathways to complexity” to my mind must be changed into the title “New pathways to regional complexity” – and I very much hope that the next generations of Mediterranean field archaeologists will be able to devise paths of research with which we may obtain a clearer idea of the initiatives and peoples which led to the complex organization of town-life, institutional religion, controlled production and others, which we still live by.

References Attema, P.A.J. 1993, An archaeological survey in the Pontine region, 2 vols., Groningen. Attema, P.A.J., J.J. Delvigne, E.E. Drost & M. Kleibrink 19971998, Habitation on plateau I of the Timpone della Motta, Francavilla Marittima, Italy, Palaeohistoria 39/40: 375-412 Bakhuizen, S.C. 1970, Salganeus and the fortifications on its mountains, Utrecht. Bakhuizen, S.C. 1976, Chalcis-in-Euboea: iron and Chalcidians abroad: Chalcidian studies 3, Leiden. Bakhuizen, S.C. 1985, Studies in the topography of Chalcis on Euboea, Leiden. Bastet, F.L. 1958, De datum van het grote hypogaeum bij de Porta Maggiore te Rome, Leiden. Bastet, F.L. & M. de Vos 1979, Proposta per una classificazione del terzo stile pompeiano, ‘s Gravenhage. Beyen, H.G. 1938, Die Pompejanische Wanddekoration vom zweiten bis zum vierten Stil, Den Haag. Beyen, H.G. 1958, Das stilistische und chronologische Verhaeltnis der letzten drei pompejanische Stile, Den Haag. Boer, M.B. de & T.A. Edridge 1978, Hommages à Maarten J. Vermaseren: recueil d’études offert par les auteurs de la série Etudes préliminaries aux religions orientales dans l’Empire romain à Maarten J. Vermaseren à l’occasion de son soixantième anniversaire le 7 avril 1978, Leiden. Boersma, J.S. 1963, De Romeinse muntvondsten in de provincie Noord-Brabant, Den Haag. Boersma, J.S. 1975, Domus aedificatur [de bouwprocedure van het romeinse stadshuis], Amsterdam. Boersma, J.S. 1985, Amoenissima Civitas: Block V.ii at Ostia: descriptions and analysis of its visible remains, Assen. Boersma, J.S. 1990, Oria and Valesio: Dutch archaeological investigations in the Brindisi region of southern Italy. Mededelingen van de afdeling Letterkunde, Koninklijke Akademie van Wetenschappen nieuwe reeks 53(3), Amsterdam. Boersma, J.S. 1995, Mutatio Valentia: the late Roman baths at Valesio, Salento Amsterdam. Crouwel, J.H. 1979, Wheeled vehicles and ridden animals in the ancient Near East, Leiden.

16

General Papers

Peters, W.J.T. 1993, La casa di Marcus Lucretius Fronto a Pompei e le sue pitture, Amsterdam. Reinders, H.R. 1990, New Halos: a Hellenistic Town in Thessalia, Greece, Utrecht; Versterkte nederzettingen, C.J.C. Reuvenslezing, Stichting Nederlandse Archeologie. Salomonson, J.W. 1965, La mosaique aux chevaux de l’antiquarium de Carthage, Archeologische Studien van het Nederlands Historisch Instituut te Rome, Den Haag. Salomonson, J.W. 1979, Voluptatem spectandi non perdat sed mutet: observations sur l’iconographie du martyre en Afrique romain, Verhandelingen Koninklijke Akademie van Wetenschappen, Amsterdam. Satricum 1982 = Satricum, una città Latina, exhibition catalogue Comune di Latina cinquantenario 1932-1982, Firenze. Satricum 1985 = Attema, P.A.J. et al. (eds), Nieuw Licht op een Oude Stad: Italiaanse en Nederlandse opgravingen in Satricum, exhibition catalogue, Rome. Stibbe, C.M. (ed) 1980, Lapis Satricanus: archaeological, epigraphical, linguistic and historical aspects of the new inscription from Satricum, Archeologische Studiën van het NIR, The Hague. Stoop, M.W. 1971, Francavilla Marittima. Acropoli sulla Motta, Atti e Memorie della Società Magna Grecia 11-12: 108-16. Stoop, M.W. 1972, Santuario di Atena sul Timpone della Motta: bronzi, terrecotte e ceramiche, Atti e Memorie della Società Magna Grecia n.s. 11-12: 37-66.

Stoop, M.W. 1977, Francavilla Marittima, Acropoli sulla Motta, Atti e Memorie della Società Magna Grecia n.s. 15-17: 107-67. Stoop, M.W. 1983-1990, Note sugli scavi nel santuario di Atena sul Timpone della Motta (Francavilla Marittima - Calabria), Bulletin Antieke Beschaving 58-65. Van der Werff, J.H. 1982, Uzita: vondstmateriaal uit een antieke nederzetting in Midden-Tunesië, Utrecht. Vermaseren, M.J., van Essen, C.C. 1959, Gids van Santa Prisca op de Aventijn, Leiden. Vermaseren, M.J, van Essen, C.C. 1965, The excavations in the Mithraeum of the church of Santa Prisca in Rome, Leiden. Volgraff, C.W. 1928, Opgravingen te Argos, Mededelingen der Koninklijke Akademie van Wetenschappen 66: 87-107. Volgraff, C.W. 1931, Nieuwe opgravingen te Argos, Mededelingen der Koninklijke Akademie van Wetenschappen 72: 71-124. Waele, J.A. de & E.M. Moormann 1984, Om de tuin geleid: een feestbundel aangeboden aan prof. Dr. W.J.Th. Peters ter gelegenheid van zijn vijfenzestigste verjaardag, Nijmegen. Yntema, D.G. 1985, The matt-painted pottery of Southern Italy. A general survey of the matt-painted pottery styles of Southern Italy during the Final Bronze Age and the Iron Age, Utrecht. Yntema, D.G. 1993, In search of an ancient countryside: the Amsterdam Free University field survey at Oria province of Brindisi, South Italy (1981-1983), Amsterdam.

17

RPC 2000 conference proceedings

2.3

Two Challenges for Landscape Archaeology

Generally speaking, two different strands may be discerned in the regional approach to the archaeological record in Italy. One is landscape archaeology, the other regional settlement pattern analysis. In the Italian, and the Mediterranean context, landscape archaeology is strongly bound up with topographical survey and systematic field walking. Often, if not always, both topographical survey and systematic field walking data will be integrated with environmental information, such as geological, pedological and land-use maps. Less integrated use is made of historical maps, although, as I will argue in the final part of this paper, these may contribute to a deeper understanding of the longue durée of a given region. Landscape archaeology is an approach to archaeological investigation that looks at the landscape in much detail, furnishing data ranging from complex urban sites to pottery scatters of only a few meters in diameter or even single sherds. The landscape archaeologist will first analyze the finds in their present-day environmental context, while seeking to correct for the various biases involved. The final aim is an interpretation of the evidence in the context of past environmental and social conditions. In the present context regional settlement pattern analysis means those studies based on site inventories mapped on a macro-regional scale rather than those limited to parts of the landscape. These studies deal with the principal sites in a broader, regional perspective, rather than with all traces of human activity in a small area of the landscape. Analyses based on the former type of data will typically focus on the political and economic dimensions of the landscape, aiming at establishing site hierarchies and territoriality, catchments and carrying capacity. The environmental data involved will often be of a rather generic type, for example the 1:50,000 geological map sheets of Italy. It is an important tendency of these studies slowly to become integrated with the more detailed data from landscape archaeology and excavation of rural structures (as in Cifani, this volume).

munity with new, often very detailed data on settlement and landscape organization in protohistoric and early historic Italy. This wealth of data cries out for synthesis on various levels. Barker and Lloyd’s volume on Roman landscapes in the mediterranean region of 1991 was an important step in this direction, as it brought together a range of survey projects, but a true comparison was not then achieved (Barker & Lloyd 1991). The further step to actual comparison between, and synthesis of, survey regions has recently been taken by projects such as the Tiber Valley Project (Patterson & Millett 1998), the Lecce project on the indigenous settlements of Puglia and Basilicata (D’Andria 1999), the Regional Pathways to Complexity Project (Attema et al. 1999), and Vallat’s work on Campania (this volume, section 7.5), all of which are represented in this volume. These projects have in common that they aim to collate data sets from different survey projects, and inventories of topographical data from systematic investigations, to be analyzed in one geographical database. The methodological problems involved in such work are being described in a dissertation by Martijn van Leusen (Van Leusen 2002) and have been commented upon by myself in a recent case study on the comparison of the Sibaritide and the Pontine Region with allusions to various other regions in Central and South Italy (Attema 2001). It is clear that comparative regional archaeology is not an easy task, as the data stem from widely divergent sources going back to the first decades of the previous century, and because the field strategies by which they have been obtained, varied even within one project. The wealth of data from modern surveys in combination with the data from the rich topographical tradition in Italy, has slowly begun to undermine traditional views on urbanization and colonization in Italy (D’Andria, this volume). These core processes are definitely no longer seen as being limited to the Greek and Roman cities and their territories (as they used to be), and the indigenous world of the Italic peoples is now slowly revealing itself as a serious partner in the making of Magna Graecia and Roman Italy. The vast amount of data generated by landscape archaeology hides an enormous potential in this respect, and it is certainly true that the methods and results of landscape archaeology are insufficiently incorporated into mainstream Italian archaeology for reasons that I will explain below. Here lie challenges for the adherents of Italian landscape archaeology.

Landscape archaeology and regional settlement pattern analysis have acquired an important position in presentday Italian archaeological research, especially in those studies concentrating on centralization processes during the later middle and late bronze age, and forms of early urbanization and Greek and Roman colonization during the first millennium BC. Systematic topographical work in the tradition of the Forma Italiae, started by the pioneer Giuseppe Lugli in the 1920’s, and numerous systematic surveys all over Italy, conducted in the spirit of Ward Perkins’ systematic surveys in South Etruria just after the second World War, have provided the archaeological com-

2.3.2 Two challenges As it is, I can see two challenges for landscape archaeology or, more generally, for the regional approach to the archaeological record that, in my view, should be part of its academic agenda. The first challenge concerns the need to develop a comparative regional archaeology aimed at formulating an urbanization theory based on the results of landscape archaeological projects. In combination with other sources, this should replace the Greek and Roman colonial urbanization paradigm. This paradigm change has already been set in motion with the shift from the study of urbanization as a supra-regional dynamic driven by colo-

Peter Attema 2.3.1

Introduction

18

General Papers

nization, to urbanization as a strictly regional and largely internally driven phenomenon (e.g. Burgers’ study of the Brindisi region, 1998). This emphasis on internal dynamics and confinement to the region as an analytical unit is doubtless a very positive development, but it has created a theoretical problem. Contrary to the colonial approach of Magna Graecia archaeology and Roman classical archaeology that provided a supra-regional theory of settlement dynamics in Italy, regionalism fails to explain such supraregional processes, unless, of course, it can develop an appropriate theory and methodology based on the comparative regional approach. To solve this problem, it will need to collate data from survey projects and regional settlement inventories on a grand scale for its main building blocks.

single sherd findspots, while soil maps were compiled to carry out a detailed land evaluation, and a pollen core was studied in order to reconstruct past vegetation conditions (Attema, van Joolen & van Leusen 2001). Moreover, an ethnographic study was conducted of sub-recent settlement systems known from literature and from historical maps at an earlier stage and incorporated in the present study (Attema 1993). To obtain an understanding of the effects of modern land transformations and the way these have affected the archaeological record, the relief and hydrography of the pre-war landscape were studied, resulting in a hypothesis as to the specific locational factors of Bronze Age/Iron Age settlement, to be tested in campaigns to come (Feiken & van Leusen 2001). I highlight this type of detailed landscape archaeological studies of marginal micro-regions, because I feel that, in future, they may add important information to an understanding of how early urbanization transformed outlying areas as well, and teach us how these areas responded to processes in the core areas, where early urbanization and Romanization took place (in this case the Latial Iron age and Archaic cultures in the Alban hills between the 10th and 5th centuries BC, and the subsequent Romanization brought about by Rome and its Latin allies). Similar very intensive surveys were carried out by the RPC team in the Murge of Puglia near present-day Ostuni, a carstic landscape that developed in the margins of the Brindisino plain where urbanization took off at the end of the Iron Age (Burgers 1998). Surprisingly, the parts of the Murge landscape surveyed by us were intensively used during the Middle and Late Bronze age transition (Attema, Burgers & van Leusen forthcoming). The most recent intensive surveys by the RPC team were carried out in 2000 in the Sibaritide in the catchment of the site of Francavilla Marittima, where attempts were made to locate a possible dispersed protohistorical pattern related to the major protohistorical sites. As was noted in the earlier surveys various visibility factors play an important role in the under-representation of the protohistorical landscape in the archaeological record (cf. Bintliff’s studies on the “hidden landscape” of Boeotia in Bintliff et al. 1999 and the discussion in Journal of Mediterranean Archaeology 13(1): 100-123). In this sense these highly intensive surveys may act as an important corrective to the broad patterns present in the landscapes of antiquity in which the Classical and Roman periods dominate. I have referred in the introduction to analysis of these broad patterns as regional settlement pattern analysis.

The second challenge for landscape archaeology concerns the potential of regional settlement and environmental data for mapping the social dimensions of the landscape through the reconstruction of collective perceptions. The nature of the settlement pattern and the ways in which the landscape was organized and exploited, should not only be the object of political and economic analysis, but may also provide insight into the different social groups inhabiting the various geographical parts of a landscape. A typical Mediterranean landscape is made up of a coastal zone where urban and rural areas concentrate, and more sparsely settled marshlands and mountainous hinterlands. The archaeology of the latter areas is sadly underdeveloped, and it is landscape archaeology in particular that is slowly making up for that, implying that other social groups in the landscape once living on the periphery of the urbanized world are now being mapped too. The potential of landscape archaeology to map this periphery of the ancient Italian social landscape will add significantly to a theory of centralization, urbanization and colonization that is not solely one of facts and figures, but takes into account the changing mentalities of the various social groups that dwelt in the landscapes of antiquity. In the following I will comment on both challenges, realizing that both are strongly interlinked or, as Graeme Barker and David Mattingly phrased it in the editorial to the POPULUS series of conference proceedings on the archaeology of Mediterranean landscapes (Barker & Mattingly 1999/2000:vii), “The greatest challenge of inter-disciplinary landscape archaeology in the Mediterranean in the coming years will be how to bridge the divide between the ecological approaches of the natural sciences to past landscapes, on the one hand, and the concerns of social archaeologists on the other with the interface between human actions and landscape”.

An example of regional settlement pattern analysis from the RPC project is the inventory of sites known in the Sibaritide between the Bronze age and the Roman period. Figure 2 shows a crude overview of the protohistorical settlement pattern in the Sibaritide collated by the RPC project, with the protohistorical settlement system concentrating on the foothills, while figure 3 shows the Archaic and later patterns filling out the classical landscape (based primarily on data sets from the Carta Archeologica 1969 and Peroni & Trucco 1994). The long term change depicted

2.3.3

Landscape archaeology and regional settlement pattern analysis The work done by the RPC team at Fogliano in the Pontine Region is a good example of the spatial resolution and multidisciplinary approach of present-day landscape archaeology (figure 1). An intensive survey of the ancient beach ridges near present-day Borgo Grappa mapped even 19

RPC 2000 conference proceedings

Figure 1: Summary map of the distribution of protohistorical sherds in the Fogliano survey.

here in a very crude form may be interpreted as one of a shift in the political and economic center of gravity from the zone of the foothills, where the protohistorical sites controlled both the plain and the inland transhumance routes, to the plain, where urbanization started in the 6th century BC and where an urban way of life radiated into the foothills under the influence of Greek colonization (Attema, in press). The fact that rural infill is not visible in the plain of the Sibaritide itself, is due to alluviation that has rendered the archaeological record in the plain invisible at the surface. This is a serious bias in the analysis not only of the Sibaritide, but other alluvial plains in Italy as well. Comparison with coastal plains, which have suffered less from sedimentation, may offer valuable analogies here (e.g. parts of the Metapontine coastal flat). Nor should we forget that the archaeology of the mountains is strongly biased as well, since many sites are hidden by vegetation in areas that will

never be under the plough. A pilot study based on the interpretation of air photography, suggests that, out of a sample area of 30 km2 including both plain and mountains, 62 % would be surveyable (Rooke 2001). Deducting from this some 15% “open macchia” and ca. 18% grass-covered fields with low visibility, an average ca. 29% of the sample area (but varying from ca. 10% in the mountainous zone to ca. 40% in the plain and foothills) would be surveyable in practical terms. Assessment of these factors is imperative if we want to evaluate the actual visibility of past settlement patterns in the landscapes we study. Clearly we need both types of study to understand the implications of urbanization, that is, both the detailed studies of micro-regions like the Fogliano example in the Pontine Region mentioned earlier (which uses the possibilities of landscape archaeology to a great extent), and the study of the more generic patterns of regional settlement analysis, as shown in the example of the Sibaritide, in 20

General Papers

Figure 2: The protohistorical settlement pattern in the Sibaritide. White boxes: main sites, grey diamonds: minor sites. The area of the Quilici survey is outlined in white. Grid size: 10 km.

2.3.4

The region as the analytical unit: advantages and limitations Landscape archaeological studies and regional settlement studies in combination thus offer the possibility of describing the early settlement history of the landscapes of Italy from the Braudelian perspective of the longue durée, and of analyzing the changes that occurring in the middle longterm. These changes may be repetitive, and may lead us to discern specific cyclical patterns in the relation between certain landscape units and human settlement and land-use behavior. These patterns, in turn, may be related to specif-

which all existing topographical information is collated to describe broad developments (preferably checked by intensive survey in sample areas). When interlinked, these studies can be used to model the long-term settlement dynamics of Mediterranean regions, highlighting both the dramatic overall changes; centralization, urbanization and colonization in the core areas, and the more subtle changes in the margins, i.e. in the marshlands and the mountains so typical of Mediterranean landscape (see also Horden & Purcell 2000).

21

RPC 2000 conference proceedings

Figure 3: The classical settlement pattern of the Sibaritide, with main protohistorical sites added for reference. Black diamonds: Hellenistic-Roman sites, grey diamonds: Roman villae. The area of the Quilici survey is outlined in white. Grid size: 10 km.

demonstrate that early Roman colonization of the ager Signinus did indeed build on a developed archaic settlement system and infrastructure (Attema, in prep.). Although there was a clear shift in centralized power from the plain to the hilltop settlement of Signium in the Roman Republican period, the background to this change was one of long-term rural infill of the fertile soils of the Sacco valley. Rather than with sudden change in the rural patterns of the ager Signinus, we deal here with the modification and transformation of these patterns. Detailed landscape studies in combination with regional settlement pattern analy-

ic historical processes. And this seems to be precisely the point at which the colonial paradigm of urbanization was deficient. When the Greeks came, there was already ongoing long-term development in the Sibaritide, and if we analyze the territories of newly founded Roman colonies, there often already was a developed infrastructure on which Roman colonists built further. To elaborate the latter, we may look at the case of the Roman colony of Signium in Lazio. Here both regional pattern analysis done by the Italian archaeologists N. Cassieri and A. Lutazzi (1985) and our own observations resulting from field walking, 22

General Papers

approaches have played a role in the paradigm change that is currently taking place in academic practice. This change can be described as one leading away from a fairly uniform colonialist archaeology to a wide range of regional and local archaeologies. The superiority of Greek and Roman culture is no longer taken for granted in these studies and local cultures are now seen as worth studying in their own right. They are considered to have actively modified if not resisted Greek and Roman influence, instead of having been dominated by it. This paradigm change is not only being applauded by many academic researchers but also reinforces the local amateur archaeologists’ opinion that every landscape has its own particular story to tell. Figure 4 shows Nino LaRocca, mayor of the small village of Alessandria Carreta in Calabria, interviewing a local farmer. He is one of the many local amateur archaeologists all over Italy who share the intimate knowledge of “their” landscape with academic landscape archaeologists. Larocca’s help is indispensable to the effort of the RPC team to map the more inaccessible areas of the Sibaritide. This is done in order that the broad patterns of settlement and landscape organization in the Sibaritide be informed with data on the mountainous zones. In the case of the Sibaritide we must suppose a very strong connection between the protohistoric settlement pattern and transhumance routes. Clearly, the RPC team, as so many other landscape projects, has opted for the region as the analytical unit, highlighting both indigenous settlement and land-use patterns, and colonial influence in the region from the late 8th/7th century BC onwards. The tendency to focus on the region as an analytical unit in a diachronic perspective has proven instrumental in changing the colonial paradigm. But one can also raise a point against this “regionalization” promoted by the many landscape archaeological projects that we have seen in the last decades. Most project teams are introspective and have little interest in relating their results to those of sister projects and little time for it. While the importance of the individual settlement histories is beyond doubt, we all know that regions do not function in isolation and that they are interlinked in many ways — politically, ethnically, ritually, by land and by sea. These unifying aspects are strongly present in the colonial paradigm through the study of high culture expressions such as Greek and Roman architecture, sophisticated urban planning and specific groups of high-status colonial objects. These studies brought together the coastal regions of Italy within one broad network of colonial development, from which resulted a clear theory of urbanization with colonization at its core. From the “regionalist” perspective of landscape archaeology and regional settlement pattern analysis, no substitute theory for the colonialist one can be expected unless we start comparing and interlinking regional developments. This constitutes the first of the two challenges that I outlined at the beginning of this paper, namely the need for a diachronic comparative regional archaeology in Italian landscape studies.

Figure 4: Nino Larocca showing some local Bronze Age finds.

sis thus reveal certain continuity in the landscape against which political change could be portrayed. This political change was symbolized and made effective in the heavily fortified Roman colony of Signium. It is important to say that landscape archaeology and regional settlement pattern analysis will treat Greek and Roman colonies on the same analytical level as the socalled indigenous settlements usually characterized by a much longer settlement history, or in other words analysis from a geographical point of view. And rather than explaining changes in terms of events (such as the founding of a Greek or Roman city) and political and cultural domination, as in the colonial paradigm, regional analysis, by the very nature of its data, will emphasize the gradual transformation and modification of settlement structures through time. In this approach Greek and Roman colonization have become ingredients in the long-term process of the early urbanization of Italy. These ingredients were added at different points in time, and under different circumstances at different places, but as I have argued earlier in this paper, they were not always the prime causes of urbanization. Because of the long-term perspective of landscape archaeology and regional settlement pattern analysis with their emphasis on transformation and modification, both 23

RPC 2000 conference proceedings

We may conclude that, as the regionalist paradigm in itself, being introspective, will be unable to produce a new theory on urbanization, a comparative regional archaeology is the only way to fill the theoretical void that the demise of the colonial paradigm has brought about. I will elaborate on this argument somewhat further below. 2.3.5 Urbanization instead of colonization? In traditional colonial theory, developments in the sphere of urbanization in central and southern Italy were brought together under the common heading of “colonization”, a top-down approach that explained changes as brought about by colonial initiative. Now that the heading “colonization” will not do anymore, we might consider the heading “urbanization” as a ready candidate to take over its role as a theme for the study of settlement developments in protohistoric and early historic Italy. The logical point of departure here are the geographical sciences, an option that was, of course, already brought to the fore in the 1970’s in for instance the volume Man, Settlement and Urbanism (1972) where the methodology of present-day landscape archeology and regional settlement analysis is closely reflected in the geographical approach then advocated by Wheatly. Discussing possible approaches to urbanism (or urbanization), Wheatly mentioned in his contribution under the heading “ecological theories of urban development” how the functionally-interrelated basic components of the urban complex are environment, population, technology, and social organization, with the last conceived as “an adaptation to the unavoidable circumstance that individuals are interdependent and that the collectivity of individuals must cope with concrete environmental conditions. In this view the study of urbanism (or urbanization) involves, according to him, specification of the precise technological, demographic and environmental conditions under which various urban forms (and in our case urbanization) of organizations may be expected to appear and - once established - to develop at given rates (Wheatly 1972). This is of course a very processual option in our present post-processual era, but the option may turn out to be a necessary first step in answering fundamental questions as; how region-specific are the developments we are mapping? Can we identify a conditioning role for landscape and technology in regional landscapes? At what points in time were certain constraints overcome, making a substantial demographic growth possible, generating new forms of settlement organization and land-use, and making possible the agricultural colonization of the alluvial coastal plains? It is clear that besides a quantitative approach towards our survey and settlement data, comparative land evaluation and research into technological developments are also needed in landscape archaeological projects. Why is this so important? Landscape infill in large parts of 4th century BC central and southern Italy suggests a rather uniform process, while in the foregoing periods differences between regions appear to be much larger. Whilst, for example, rural infill of the coastal plain of Metapontum was already strong in the 6th

Figure 5a: Interpretation of land use after the 1851 map by the Military Geographical Institute of Vienna (from Attema 1993, Figure 12).

Figure 5b: Interpretation of the infrastructure of the 1851 map, showing an intricate network of paths and hut compounds (from Attema 1993, Figure 13).

24

General Papers

century BC (Carter 1994), in the area around Poseidonia it only developed in the 4th century BC (Horsnaes 1999). Could it be that up to the 4th century BC, regions tended to develop quite independently, after which they were absorbed in a broader-scale development in which regional networks became part of larger supra-regional urban-based networks? Would this coincide with the period in which buffer zones between early-urbanized regions were being filled in with urban and rural settlement? Would this development relate to improvements in the infrastructure drawing together such regions, and if so in what way? To formulate and test hypotheses of this kind one cannot do without a methodologically sound comparative regional archaeology of the settlement data collected in the various landscapes of central and southern Italy. The workshop on comparative settlement archaeology organized by Gert-Jan Burgers and the workshop on methodology and Geographical Information Systems organized by Martijn van Leusen set out to investigate many of the problems implicated in the comparison of topographical and survey data, while the workshop on land evaluation by Ester van Joolen sets out to stress the importance of an intimate comparative and diachronic knowledge of landscape development and agricultural technology. Benoît Mater’s workshop on pottery production draws attention to the increasingly complex socio-economic networks involved in pottery production and distribution during this period as an expression of integration into supra-regional developments. Comparison of the results of regional projects has the potential to generate the data necessary for building a new theory on the early urbanization of Italy to eventually replace the colonialist paradigm. A characteristic of this theory will be that it will accommodate regional developments as well as overarching processes, indigenous development as well as colonial input.

The biased interest in ancient interventions in the landscape has a long tradition dating back to the Renaissance, especially where it concerns ancient land reclamation projects. This constitutes probably the most drastic form of landscape transformation, as it often implies profound changes in the hydrography, infrastructure, settlement and land-use patterns. Implementation in the past of rational types of land-use has often resulted in clear signs in the present landscape. Air photo interpretation in particular has revealed the increasing rationalization of land use that was part of the early urbanization process. I may refer here to both the Greek and Roman land division patterns that have been identified in the Mediterranean. We may take the developments that took place in the ager of Terracina (Pontine Region, Lazio) in Roman Republican times as an example. Already at an early stage in the Roman Republican period the Valle di Terracina was centuriated and peopled with colonists, while the adjoining slopes of the Monti Lepini were filled in with Roman villae specializing in olive culture. Now we can imagine how, at the same time that the physical contrast between the urbanized landscape units of developing towns such as Terracina and the peripheral marginal world of the Pontine marsh increased, the social contrast increased as well between the town and country- dwellers of Terracina on the one hand, and those that continued to live off a marsh economy on the other hand. These social differences in the landscape need to be mapped. This takes me back to our surveys on the ancient beach ridges of Fogliano near Terracina. The results of these surveys have shown that the ancient beach ridges with the adjoining marsh did indeed remain a marginal area during the Roman period, while urban development at Terracina was strong and its related countryside rapidly urbanizing. Observations of such contrasts between adjacent landscape units can, on the basis of the differences in settlement pattern and related land-use and material culture, be related to different social groups dwelling in the landscape. To study such processes in more detail, we may also turn to textual information and ethnographical analogy. How did ancient writers perceive the various landscapes we study, and can we reconstruct mentalities from their texts that fit our spatial data? To reinforce the latter aspect we may also look at historical cartographic data from the pre-industrial landscape, and search for analogies of types of land-use in the marginal areas, and related ways of dwelling in and exploitation of such landscape units. Figures 5a and b show the intricate network of paths and fishing settlements in an area that was called the Macchia di Terracina in historical maps (figure 5a is the original mapping from 1851, 5b a digitized version of that map). Interpretations of historical maps furnish analogies for the settlement system and type of land use, in this case for the scattered protohistorical findspots in the Fogliano area shown in figure 1. The small numbers of sherds mapped in the Fogliano area may reflect what little archaeological remains small-scale low-status settlement will leave behind. In fact, what we see happening in the Fogliano area is a cyclical pattern - both the protohistorical phase and the sub-recent

2.3.6

Mapping and visualizing social dimensions and perceptions of the landscape My second challenge for landscape archaeology is more of a post-processual nature, as it concerns more specifically the duty of Mediterranean landscape archaeology and regional settlement pattern analysis to map the social dimensions of the landscape. As I stated in the introduction to this paper, it may touch upon cognitive aspects of the landscape through the reconstruction of its collective perception. I must emphasize that I am not referring here to the reconstruction of individual experiences of humans with regard to the phenomenology of landscape, but rather to the changing collective perception by groups of people of their environment as a result of the physical changes brought about in the landscape through ever more compelling forms of landscape organization. It is especially the highly organized parts of the classical landscape of town and country that traditionally have been the object of archaeological study. And, as I have already stated, we should not forget the ‘marginal’ worlds of the marsh and the mountains that in the course of the ongoing urbanization of the plains acquired an economically and socially peripheral position. 25

RPC 2000 conference proceedings

phase of marginal settlement were followed by colonization; Roman colonization in the classical period, the fascist reclamation in the modern era (Attema 1993, 1996). In sum, both data sets, one documenting the ongoing rationalization of the landscape by means of dominant and often colonial features like centuriation, improvements in the hydrography and town planning, the other documenting the marginal landscape, are needed to recreate the social landscape of antiquity and to measure the influence of urbanization on the various social groups present in the landscape. The merit of landscape archaeology in this respect, is its ability to map the spatial correlates of this process.

cal sense. The next step then is to relate these separate visualizations into a narrative sequence, so as to reconstruct the changing collective perceptions of the landscape. The latter step can hardly be taken without the aid of textual evidence and historical cartography. In accordance with Ian Morris’ recent plea for an archaeology as cultural history (Morris 2000), one can indeed imagine a landscape archaeology that is informed by the long-term cultural geography of the landscape.

References Attema, P.A.J. 1993, An archaeological survey in the Pontine Region, a contribution to the early settlement history of South Lazio, 900-100 BC, Groningen (Ph.D. thesis). Attema, P.A.J. 1996, Inside and outside the landscape, perceptions of the Pontine Region in Central Italy, Archaeological Dialogues 3(2): 176-195. Attema, P.A.J. et al. 1999, Regional Pathways to Complexity: landscape and settlement dynamics in early Italy, in: C. Fabech & J. Ringtved (eds), Settlement and Landscape, Proceedings of a conference in Århus, Denmark, May 4-7, 1998: 475-477. Attema, P.A.J. 1999b, Cartography and Landscape Perception: A case Study from Central Italy, in: M. Gillings, D. Mattingly & J. van Dalen (eds), Geographical Information Systems and Landscape Archaeology (The Archaeology of Mediterranean Landscapes 3), Oxford, Oxbow Books: 23-34. Attema, P.A.J. 2001, Early Urban and Colonised Regions of Central and South Italy: A case study in comparative landscape archaeology, in: Darvill, T. & M. Gojda (eds), One Land, Many Landscapes (BAR International Series 987). Attema, P.A.J., in prep., Early Roman Colonialism in South Lazio, a survey of three landscapes. Attema, P.A.J., E. van Joolen & P.M. van Leusen 2001, A Marginal Landscape: Field work on the beach ridge complex near Fogliano (South Lazio). In: PaleoHistoria 40/41 (2001). Attema, P.A.J., Burgers, G.-J. & M. van Leusen, M. forthcoming: Walking the Murge: interim report of the Ostuni Field Survey (Apulia, Southern Italy), Studi di Antichità 11 (2002). Barker, G. & D. Mattingly (eds) 1999/2000, The Archaeology of Mediterranean Landscapes vols. I-IV. Oxford: Oxbow books. Bintliff, J., Ph. Howard & A. Snodgrass 1999, The Hidden Landscape of Prehistoric Greece, Journal of Mediterranean Archaeology 12(2): 139-168. Burgers, G.-J.L.M. 1998, Constructing Messapian Landscapes, settlement dynamics, social organization and culture contact in the margins of graeco-roman Italy, Amsterdam: Gieben. Carta Archeologica 1969, Carta archeologica della piana di Sibari (redatta da L. Quilici, S. Quilici, C. Pala & G.M. De Rossi), Atti e Memorie della società Magna Grecia, Roma. Carter J.C. 1994, Sanctuaries in the chora of Metaponto, in: Alcock, S.E. & R. Osborne (eds), Placing the Gods, Clarendon Press, Oxford: 161-198. Cassieri, N. & A. Lutazzi 1985, Note di topografia sul territorio tra Segni e Paliano, Quaderni di Archeologia Etrusco-Italica 11 (Archeologia Laziale VII): 202-209. D’Andria, F. 1999, Ricerche recenti sugli insediamenti indigeni di

I end this paper by going briefly into the cartographic representation of collective perceptions of the landscape or “mental maps”. How can we visualize the social dimensions of the landscape? It is worthwhile to taking a look at historical thematic maps for inspiration. At the time cartographers became interested in the recreation of the ancient landscape in the contemporary geographical mould, it became possible to map ethnic groups in the landscape and to locate their settlements in the landscape on the basis of the ancient texts (Attema 2000). These early cartographers were the first to furnish Livy’s texts and other ancient sources with spatial correlates. Admittedly, the first historical thematic maps depicting settlement and environment in close visual relationship did so on a very thin scientific basis and indeed form a great contrast with our present-day visualizations of the ancient landscape with dots for sites in sometimes highly technical maps. We should, however, be able to produce other types of maps as well. On the basis of our detailed settlement data, vegetation reconstructions, erosion and sedimentation history, and the close study of environmental data from the ancient sources, it should be quite possible to capture the “grand view” of the historical thematic map once again, but now on a quantitative basis and with precise spatial correlates. How much of the regional landscape was really subject to urbanization in the 4th c. BC and how much territory was exactly occupied by built-up urban areas and how much by agricultural land? Which parts of the landscape were (or became) isolated from the core areas? Can we visualize in this way the social contrast between the marginal world of marsh and mountains and the civilized world of town and country, and model these social changes through time? How did the ritual landscape relate to the landscape of town and country through time? Geographical Information Systems have the power to create visualizations of the landscape that bring out the social processes we study in precise geographical frameworks. By modeling the data to such an extent that we really fill in our regional landscapes with settlement, infrastructure and land use patterns in the plains, marshes and mountains of the mediterranean landscape, we may not only conjure up the holistic view of the ancient landscape that was present in the historical thematic maps, but also arrive at a deeper understanding of what early urbanization brought about in the protohistorical landscape in a physi26

General Papers

Puglia e Basilicata, in: La Forma della Città e del Territorio, Esperienze metodologiche e risultati a confronto, Atti dell’Incontro di studio - S. Maria Capua Vetere 27-28 novembre 1998, Roma: 103-118. Feiken, H. & M. van Leusen 2001, Interpreting Field Survey Results in the Light of Historic Relief Change: the Fogliano beach ridges (south Lazio, Italy), in: Stančič, Z. & T. Veljanovski (eds), Computing Archaeology for Understanding the Past. Proceedings of the CAA2000 conference (BAR International Series 931). Oxford: BAR Publishing: 205-211.

patterns in archaeological landscapes, PhD thesis, University of Groningen.

Morris, I. 2000, Archaeology as Cultural History, Blackwell Publishers, Oxford. Patterson, H. & M. Millett 1998, The Tiber Valley Project, Papers of the British School at Rome, 1998:1-20 Peroni, R. & F. Trucco 1994, Enotri e Micenei nella Sibaritide, Taranto. Rooke, M. 2001, GIS & Survey. Unpublished work, Groningen Institute of Archaeology. Wheatly, P. 1972, The Concept of Urbanism, in: Ucko, P. et al. (ed), Man, Settlement, and Urbanism. Proceedings of a meeting of the Research Seminar in Archaeology and Related Subjects held at the Institute of Archaeology, London University. London: Duckworth.

Horden, P. & N. Purcell 2000, The Corrupting Sea: a study of Mediterranean history, Oxford. Horsnæs, H.W. 1999, The Cultural Development in NorthWestern Lucania, c. 600-273 BC, University of Copenhagen (Ph.D. thesis). Leusen, P.M. van 2002, Pattern to Process, methodological investigations into the formation and interpretation of large-scale

27

RPC 2000 conference proceedings

2.4

Settlement Pattern Analysis and Demographic Modeling

of the British School at Rome in South Etruria during the 1950s-60s (Potter 1979) and the Minnesota Expedition to Greek Messenia in the 1960s-70s (McDonald & Rapp 1972). As is now well-known, within a short period there arose a further development in topographical landscape research, a ‘New Wave’ (Cherry 1983; Bintliff 1994b) - in which the aim was to accomplish field-by-field cover of small sectors or whole blocs of a regional landscape, using fieldwalkers placed within such close distances that all but the smallest activity-traces on the surface would be observable by one or more of the team; this is often referred to as ‘intensive’ as opposed to the previous ‘extensive’ method of reconnaissance. In retrospect, I see that the methodology we devised enabling us to intensify observation was ready to be applied before we realized what it entailed in data processing and interpretation. An unparalleled explosion of new information required far more time to sort and interpret, and prepare for publication than expected. A new complication arose from the delay in project completion, in that the impetus towards improvement in the young subdiscipline of field survey meant that projects working with older data found themselves with an array of additional questions and analytical procedures to incorporate into their plans. Nonetheless the ‘New Wave’ projects gave us a Mediterranean densely-filled with ‘sites’ or dots on the map of a remarkable richness in space and time, and summoned up a very lively discourse among historians, anthropologists and excavation-based archaeologists. By the 1990s, when many projects of the ‘Golden Age’ of field-by-field survey were fully or partly published, questions could rightly be raised as to what was the precise status of the numerical and spatial information on these distribution maps which emerged from regional surveys all around the Mediterranean. Doubts were raised by practitioners themselves, and by other scholars, notably in Ancient History. Two problems were highlighted: first, it was much easier to find a rich spread of surface sites than to evaluate what each one represented in social, economic, chronological etc. terms. Second, that even the vast increase in sites per square kilometer represented only a random sample of the original settlement and activity systems in a given landscape - what was not being found and how significant was it? Some success had already been achieved through refinements in local contextual procedures, for example the linking of visible sites and distributions to the geomorphologic maps and histories of the region under study (most notably by the Southwest Argolid project in Greece – van Andel et al. 1986), or the recalculating of figures on past populations to allow for coarse chronologies (on the Melos project, also in the Aegean – Cherry 1979), other refinements were the adoption of filtering mechanisms to allow for variations in vegetation cover, and totalcounting of all surface artefacts during fieldwalking to include ‘offsite’ as well as ‘site’ activity traces (a procedural link pioneered on our own Boeotia project in Greece – Bintliff & Snodgrass 1985), and we had some success in correcting the bias introduced through differential ability to recognize and date surface artefacts (as for example with the balancing of imported fineware chronologies for sites

John Bintliff The wave of intensive and semi-intensive field survey projects sweeping across the entire circum-Mediterranean region since the 1970s, has infilled many regional landscapes with dots representing settlement and activity networks of every period from Paleolithic to Post-Medieval. From the beginning these projects have been carried out in the unshaken belief that sites can be identified with comparative ease, that their size can be determined for each chronological phase and that therefore the pattern of demographic change in the moyenne and longue durée can be reconstructed. Careful study of the publications and ongoing experience of projects being prepared for publication make it difficult to accept these assumptions of an earlier wave. Albert Ammermann (pers. comm.) characterizes the most recent phase of methodological rethinking in field survey as one in which regional project directors must cease to see surface data as simpler and more direct evidence on regional settlement dynamics than excavation data - and recognize that surface data are just as difficult to understand as excavated artefact assemblages. Relevant aspects of this new ‘problematique’ focus on the taphonomy of surface assemblages, on the serious need for refining ceramic and lithic typologies, on the inadequacy of regional sampling strategies, and on the limited number of publications evaluating the relationship of surface finds to deposition and its governing conditions (full-time occupation, temporary-seasonal occupation, storage or specialized activity loci, burial and cult foci), not forgetting the often considerable effects of offsite deposition on surface find study even within ‘sites’. A final aspect to be noted is the relatively small body of applied theory concerned with pattern and process in long-term settlement dynamics. The longest tradition in the Mediterranean - that of German scholarship with its Landeskunde approaches of the preWar and early post-War era - has suffered unaccountable neglect, whilst other approaches are only just beginning to be investigated and experimented with. In this paper I shall elaborate on the issues just raised and discuss new ways to work on the closely-related tasks of settlement analysis and demographic reconstruction using Mediterranean survey data. 2.4.1 The historical development of intensive survey Topographical landscape research has almost as long a history as Archaeology itself: one thinks for example of the field trips and observations of Antiquarians such as Stukeley in the 17th century (Greene 1995:21-23), and of the more detailed recording of monuments along exhaustively explored country routes in late 19th century Greece to be found in the work of Lolling and other scholarly topographers (Lolling, reissued 1989). Archaeological field survey in the Mediterranean lands only came of age with the planned regional coverages pioneered by teams such as that 28

General Papers

zone, some 600 km2, has been surveyed, from which some 69 farms and other rural sites can be identified for the Roman Imperial period. At around 1 site per 8 square kilometers, it is very tempting to relegate this region to extreme underdevelopment in socioeconomic terms compared to contemporary settlement and land use systems in Lycia. The extraordinarily small size of the associated urban foci in this part of Tunisia - a few hectares, might seem to bear this out. Nothing could be further from the truth in fact. Recent research has shown that low-density, widely dispersed rural sites in this region reflect a specialized form of economic production geared to long-distance exchange - with unusually-extensive olive groves - each tree far from its neighbor to enhance its productivity - matched to very large estates. If differences in regional economic organization (an aspect of the Roman Empire increasingly apparent from archaeological research - cf. Fulford 1987) help to give a quite new reading of site density variation as revealed by survey, more difficult and even perhaps intractable problems can result from the limitations of chronological determination for a period map of sites. John Cherry back in 1979 (op. cit.) made a very plausible case for deconstructing a map of Early Bronze Age rural sites - seemingly a full landscape of family farms - on the Aegean island of Melos, by suggesting that only a handful of farms were likely to be in use at any one time (based on the immense time span of the ceramics dating the sites and their probable individual life span of 1-2 generations).

with local domestic and coarseware assemblages – cf. Cambi & Fentress 1989 for the Albegna survey). 2.4.2 The need for a new Quellenkritik Despite the undeniable success of the New Wave surveys of the 1970s-90s in plotting on landscapes plentiful sites of almost all periods, the corrections and methodological changes brought into that tradition in its more recent phase have only served to reveal far more deep-seated problems in the recovery and interpretation of surface survey data (cf. in general, Bintliff, Kuna & Venclova 2000). Tackling these problems is all the more important, since scholars want to utilize the growing number of New Wave survey publications for comparisons between regional patterns on an interregional scale (a fascinating project notably explored for the first time in the Roman Landscapes conference - Barker & Lloyd 1990), with later examples including Sue Alcock’s monograph on Roman Greece (Alcock 1993) and her study of Eastern Mediterranean surveys (Alcock 1994), and my own paper on long-term growth trajectories in the Aegean (Bintliff 1997a). I shall discuss in turn a number of topics arising from the data brought into current survey theory discourse as a result of the New Wave surveys, complex and difficult though it is. 2.4.3 Density and quantity are not enough... Two case studies in Mediterranean survey bring out rather well the need to challenge and replace the assumption that the density of activity foci or ‘sites’ can be used as a reasonable indication of economic complexity or cultural florescence. This is a significant revisionary statement, since one of the key points of New Wave survey was its success in multiplying regional site densities many times over in contrast to earlier extensive and topographic research. The first of the case studies is the recent remarkable results from the survey of a polis / city state territory in Lycia (S.W. Turkey) - that of Kyaneai, by a German team led by Frank Kolb (Kolb 1993, 1995, 1996, 1998). This is some 138 km2 and can well be described as very marginal karst landscape. However, the great colonization and infill of the region which occurred between Hellenistic and Roman times, followed by retraction to very low levels of population in all subsequent eras (including recent times) has ensured that the Greco-Roman town and country survive in enviably fine upstanding monumental condition (urban and rural house structures, rural tomb monuments, etc.). A site density of some 5 per km2 is typical of recovery results from the richest farming landscapes elsewhere in the Mediterranean. Although there is historical evidence to suggest some degree of export from the region, the land potential and communication problems would argue that the primary activity underlying this ancient intensification of occupancy was a growing internal regional market. In dramatic contrast are the results obtained by another recent intensive survey, this time in Tunisia, but also emphasizing Greco-Roman antiquity - the Segermes survey (Dietz, Ladjimi Sebai & Ben Hassen 1995). A far larger

Another form of quantitative revisionism concerns an even more fundamental reading of survey data - the equation of numbers of dated sherds or lithics collected from a site with intensity of activity per period represented. Here my own recent analysis of the Boeotia Project survey in Central Greece has caused me to rethink our understanding of that landscape in ways totally divergent from our views at earlier stages of this regional project (cf. Bintliff & Howard 1999; Bintliff, Howard & Snodgrass 1999; Bintliff 2000). The final publication of this long-lived survey program (it began in 1978) will commence with a volume devoted to a small sector - some 5 kilometers square - which forms a part of the rural hinterland of the ancient city of Thespiae. This Leondari South-East / Thespiae South (LSE/THS) district produced 18 rural sites, one being Medieval, seventeen being Greco-Roman, and none being recognizably prehistoric. The publication of this slight group would have been easy enough till recently, since grid collection on all the sites and reasonably-large sherd collections would have allowed one to produce a map of occupation sites per period across the landscape, where significant numbers of finds was read as occupation, low numbers as temporary use or ‘offsite’ (a vague way of leaving that data out of further analysis). Indeed this has been the norm for other Greek New Wave surveys up to the most recent to be published. And yet distinctive and problematic aspects of the LSE/THS data argue against such a straightforward path and indeed several years have been spent dealing with those methodological and interpretative difficulties before we now feel confident enough to publish the group. 29

RPC 2000 conference proceedings

The first obstacle to easy reading of the district is the nature of the ‘offsite’. Continuous counting of sherd densities in line-walking allowed us to create a total density map of the whole sector, corrected for visibility variations. The entire district is an unbroken carpet of sherds, with an average of well over 2000 pieces per hectare. A dated sample collected throughout the 5-plus square kilometers surveyed revealed a further remarkable statistic - something like 75-80% of all this material (of the order of some 1.5 million surface sherds by extrapolation) belongs to just one chronological phase - the broad Classical Greek period. We have rehearsed our arguments to explain such offsite densities elsewhere (Bintliff & Snodgrass 1988b; Snodgrass 1994), but the data now available leave no other possibility than to ascribe this to radial manuring out of the large city of Thespiae, a phenomenon closely matched in Tony Wilkinson’s studies of agricultural manuring in the historic and prehistoric Near East (Wilkinson 1989). The scale of impact here is also very comparable - with major urban sites in the Near East creating manuring haloes in the form of dense carpets up to several kilometers out from their perimeters. GIS study using an access surface out from the city of Thespiae, merging distance with slope angle, and allowing for major differences in soil fertility, also finds very good agreement with variations in the density of the LSE/THS carpet. The specific problem raised by what now still seems a manageable mass of data relates rather to the implications for site study in the district. Since the manure carpet is ubiquitous, and considerable, sherds collected onsite cannot automatically be assumed to reflect site-activity, but could have resulted from manuring at times when the fields incorporating the site were merely in agricultural use. Indeed when we formalized this challenge into a ‘residual analysis’ - predicting the typical density of offsite finds in the area of the site were the site not to have been in use we found that many of our sites had finds per period that were not elevated above offsite expectation. Naturally this result generally arose with sherds onsite of Classical Greek date - and we have concluded that even where the commonest finds onsite were of that period, in several cases the absence of densities above local offsite must mean that the ‘sites’ were agricultural fields in that era, only achieving occupation status in subsequent Roman, Late Roman or Medieval times, when often smaller numbers of sherds still represented vastly-increased densities compared to the modest to poor representation of those periods in the local and general offsite. A not-unexpected corollary of the site-offsite problem is posed by what is a relatively common site type in the survey of the Greek Mediterranean - the small rural family-clan cemetery. We have been able to show that in the LSE/THS sector at least, the very-localized and low density scatters of fine and special purpose wares revealing the presence of such tomb groups usually are at or below the density of contemporary urban manuring carpets. It is the qualitative features that seem to have allowed identification in the first place, and since such scatters can be so small that only one walker may pass on or near these sites, it must be

highly likely that many if not most elude detection in standard 15 m or greater interval fieldwalking. One very positive result of the manipulations required to resolve occupation or otherwise for Classical Greek sites has been a much greater resolution of the significance of other period finds on the sites of this district. We have been able to suggest differences in the density and shape of scatters for Roman and Late Roman times which could indicate variable site function - from full intensive occupation through small scale and possibly seasonal or temporary use, down to minimally-heightened activity foci within an agricultural or pastoral land use pattern. Here I would underline the necessity of continuous landscape offsite recording to clarify the depositional context of onsite finds, but also add that raw numbers of finds per period is not enough alone to indicate what was happening in the site locality. I have already illustrated this in the special case of cemeteries, but we are now able to point to an additional complication: even when a small collection of finds onsite for a particular period may be little different to expected offsite finds for that phase, leading to an initial evaluation that at that time the site cannot be shown to be more than a zone of agricultural manuring or very light temporary activity, the spatial properties of low-number onsite finds can reverse such an interpretation. The reason for this is as follows: although over the whole site area absolute numbers of finds for a particular period may be merely at local offsite level for that phase, when their dispersal over the site grid is shown - those small numbers can turn out to be highly focussed in just one limited part of the grid - where their density is significantly-higher than in the offsite. The clear implication of such complex empirical observations is that the days of surveying without detailed offsite counting and collecting are over, nor can onsite collections rely any more on ungridded grab samples. Quite a different but no less significant revision of procedures seems now clearly called-for when we come to consider the way in which prehistoric surface sites make their existence manifest to fieldwalkers. Even New Wave surveys can be shown to achieve very large increases in site densities for most historic periods but far less significant multipliers for the much longer periods of prehistoric occupation. Indeed the calculable population and land use picture one might extract from the data for the prehistoric phases of most surveys seems far too limited to be plausible, although such surveys normally can point to welldefined sites with abundant sherds and lithics from the later prehistoric farming eras of Neolithic, Bronze or Iron Ages. Now it can be shown from exhaustive study of small surveyed landscapes (in Boeotia for example, – Bintliff, Howard & Snodgrass 1999, or the Agro Pontino – Attema, van Joolen & van Leusen 2001), that these clearly-identifiable later prehistoric sites are probably the exception, with other kinds of barely- to hitherto un-recognized sites forming the norm for Mediterranean landscapes. The arguments can be summarized as follows: much of the ceramic material of these eras is coarse or low-fired, and in the many millennia since deposition such pottery has a very 30

General Papers

low survival rate in surface deposits prone to natural weathering and regular cultivation, in contrast to the harder wares of historic times. The larger and longer-occupied a later prehistoric site may be, the greater chance that its presence reaches the normal level of site recognition typical for historic sites. But if a prehistoric site is relatively short-lived and/ or small, then the sherd material is normally as slight as a mere few pieces or even a single piece of pottery visible to a fieldwalker in a single fieldwalking transect. Fortunately, a confusion with prehistoric manuring can be ruled out, as a result of insights obtained by Czech colleagues on similar settlement systems (Kuna et al. 1993; Kuna 2000), where it has been shown convincingly that current surface finds have to be the result of modern plowing into settlement or burial strata, since the survival of coarse scatters – had they been only placed from the beginning in the ploughsoil - would be well-nigh impossible. Now this problem will only be critical if whole regions and periods of prehistory had a settlement pattern in which farms or hamlets were common – at these times, according to the arguments, their settlement traces will be so limited as to be ignored by modern field surveys as ‘offsite’. In fact it has become clear that much of Europe was typified by such a mode of settlement in Neolithic to earlier Bronze Age times, and locally also into the earlier Iron Age, with villages or multi-period sites being in the minority and tending to become typical only as a later horizon in most regions. This model has been tested on the Boeotia LSE/THS sector discussed above, with the result that although initially no prehistoric occupation sites were identified in the area, but instead a group of nucleated village-hamlet sites at regular and wide distances around it, we now would recognize a very dense cover of small, one-to-two generation farms of later Neolithic to Bronze Age date across the whole district - on the basis of very small scatters of ceramics found throughout the area.

Roman sites seem to have far higher proportions of storage and processing vessels. Such investigations have been pioneered in Aegean survey by Todd Whitelaw for the data emanating from the Kea survey (Whitelaw 2000). 2.4.4 Refinements to demographic reconstructions In the excitement generated by the vastly increased site database produced by the New Wave surveys, working assumptions were made in order to give an impression of population change as reflected in regional survey data. Most project teams have been aware of the weaknesses of seeing such ‘guesstimates’ as factual, but it is now right to move on to more refined palaeodemographic interpretations. We have already referred to John Cherry’s pioneer deconstruction of Bronze Age settlement maps, but it has still to be admitted that little progress has been made in dealing with the problem that in most parts of the Mediterranean, surface finds of later prehistoric ceramics and lithics are generally only assignable to such long periods of time that the resultant phase maps cannot claim to offer settlement or activity pictures likely to belong to the same points in time. Put bluntly, maps of 20 activity foci could be a short-lived subphase, with all sites in contemporary use, within a potential time span of 1000 years for a pot or lithic assemblage, or alternatively, might reflect a society of one or two families shifting base around a wide territory over a far longer period of time. Within the historic periods conditions are more favorable for progressive refinement of dating. Some long-term projects such as Metapontum in southern Italy (Carter 1990) have been able to link excavation of urban and rural sites to associated rural survey, and at the same time to subdivide the typological series of ceramics - even for domestic wares - to much shorter phases than the norm for Mediterranean survey projects, where surface finds are often only assignable in bulk to periods of 300-400 years. Thus at Metapontum claims can be made for rural site abandonments and reoccupations during phases as short as a century - a situation most surveys would be unable to document. In fact on most surveys, activity evidenced at a surface site within a phase of maximum several hundred years’ duration is commonly read as equal to continuous use throughout the period. In the absence of such advanced typochronologies, more traditional links between ancient sources for population levels and the evidence of survey continue to be made (cf. my own attempts on behalf of the relationship between archival data for ancient Boeotian populations and the intensive and extensive evidence from town and country in Classical Greek times – Bintliff 1997c). The recent growth of interest in post-Roman landscape history has opened up greater opportunities for close-matching of survey data and historical records of greater detail and accuracy than Greco-Roman records. Thus work on Venetian (cf. the Argolid project – Jameson, Runnels & van Andel 1994) and Ottoman tax registers and cadasters (Bintliff 1995, 1996; Kiel 1997), has opened up exciting opportunities to compare the surface traces of abandoned settlements of late Medieval or early Post-Medieval date with archives offer-

Another area of rethinking also concerns density measurements for ceramics or lithics in surface collections. Martin Millett has suggested that variations in pottery supply or less commonly consumption, between periods, will have the effect of creating significant fluctuations in apparent activity across regional landscapes (Millett 1991). In most Mediterranean landscapes however it seems up to now that the bulk of total assemblages were made within the region, so that consumption would be a more likely factor to investigate (the opposite being true though where surveys have relied on exotic imports to date site occupation - as with African Red Slip in Late Roman Italy - cf. Cambi & Fentress 1989). The Boeotia Project LSE/ THS district for example seems to provide evidence for unusually-enhanced pottery consumption on sites of Classical Greek as opposed to later periods, and this extends to the range of recognized forms in use. One practice which can be identified as contributing to this is the social importance of formal dining and associated tableware in contemporary society, whereas for example Roman, and even more, Late 31

RPC 2000 conference proceedings

worked examples for us to develop and refine today (the pioneer analysis of the settlement chambers of Eastern Crete by Lehmann 1939, or of those of the Early Medieval Netherlands by Heidinga 1987, spring to mind). More recent, micro-studies of the long-term sequent occupance of single modern parishes can also serve as models – for example the exhaustive survey and excavations of Lunel in Languedoc by Claude Raynaud (Raynaud 1990) or our own total survey of the Valley of the Muses in Boeotia (Bintliff 1996, 2000). Equally notable is the development out of Landeskunde models of the Czech group of landscape archaeologists deploying what they term ‘community area theory’ (Kuna 1991; Neustupny 1991) – which has strong links to Siedlungskammer geography.

ing very detailed information on population size and structure as well as economic production. 2.4.5 Settlement patterning One aspect of landscape study increasingly neglected by archaeology since the beginning of the 1980s (not coincidentally in parallel with the displacement of law-seeking new archaeology by anti-positivistic post-processual archaeology) is that of regularities in the spatial patterning in settlements. Since the New Wave surveys matured over the same period, unsurprisingly they have suffered from theoretical poverty on this topic. The approaches available from the earlier tradition of Spatial Analysis are however of vital significance to the interpretation of long-term as well as synchronous settlement systems within each regional landscape. They focus on regularities in the spacing, number and size of rural settlements of equivalent status, and on the similar properties to be associated with settlement hierarchies. To study such phenomena survey methods have to be adequate to recover large and meaningful segments of such past systems, and this means first and foremost that strip or small bloc surveys rarely work. One needs contiguous blocs of the size of at least several modern communes / parishes, preferably also incorporating their relevant central-place(s). It has first to be mentioned why such ‘old-fashioned’ locational approaches deserve reintroduction into archaeology (a topic I have discussed elsewhere – Bintliff 1997b). The reason is very simple – regularities remain a normal feature of recovered settlement distributions, and the geographical literature of the 1950s-1970s dealing with such phenomena were empirically-based and robust, and have not gone away with the rise of more fashionable cognitive and phenomenological approaches. Nonetheless, the critique of New Geography lay in the valid argument that it relied too much on automatic mechanisms for settlement geography and made little or no allowance for historical uniqueness and the role of human action. As a result, the rebirth of such generalizing approaches needs a less deterministic underpinning, which can now be found in the non-linear dynamics of Chaos-Complexity Theory (Lewin 1993; Reed & Harvey 1992; van der Leeuw & McGlade 1997; Bintliff 1997b). Very summarily, Chaos-Complexity argues that there is no predictable shape to an aspect of past society, such as rural settlement systems, however there are strong but not overwhelming tendencies to the recurrent and cross-cultural emergence of similar forms of life. The actual realization of a settlement system or hierarchy will always therefore represent a specific dialectical compromise between such general shaping factors and the particular opportunities and constraints offered by a given landscape and its given society within a given preceding and unique historical trajectory. I have been much influenced by earlier applications of landscape geography in which some of these cross-cultural elements have been explored. The German Landeskunde tradition for example, and in particular the concept of favored zones of settlement (often separated into natural cells by less favored areas) or Siedlungskammer offers

What all these geographical approaches share in common is their grounding in empirical evidence for regularities in settlement systems across time and space. The fact that genuine ethnic or social memory continuities are frequently inapplicable, plus the cross-cultural parallels, suggest strongly that the tendencies to recurrent form are created by basic tendencies in human spatial and social behavior which are not tied to specific cultural systems. Amongst recognized relevant factors (cf. in general, Bintliff 1999a) are: 1. Catchment constraints for land use – least effort principles generally restrict the radius of land use from home bases, and there may be agreement across many societies regarding the scale of preferred territory with related forms of exploitation; 2. A limited range of territory shapes dependent on whether favored resources are generally distributed or layered directionally across the landscape (so that circular or strip territories might be common, respectively); 3. Siedlungskammer may be defined through natural barriers or resource pockets, creating constraints on the positioning of all rural sites in a region, either through direct limits within such chambers or through the effects of chambers on adjacent settlements in less constrained environments (cf. the long-term nucleated-settlement niches of the Greek province of Boeotia – Bintliff 1994a);. 4. Cycles of population growth and decline can often be linked to increasing density of settlements, and this replication and infill can produce nested series of regular spacings and territory sizes through relatively simple and even cognized processes of internal colonization; 5. Tendencies to recurrent forms of settlement system may be associated with cross-cultural regularities in social organization. Thus it can be argued that in many farming societies with little or no social ranking, the decreasing effectiveness of face-to-face community relations when settlements exceed 100-200 people can cause recurrent fission and the creation of similarly small scale satellite hamlets across the landscape. In contrast, it can be argued that although overcoming such problems, so that settlements can rise to many 32

General Papers

Medieval times (Brown, 1995). Interestingly, whereas Roman market centers compare well to Early Medieval networks, implying a relatively-poor flow of surpluses across and out of the region, the High Medieval pattern forms a far denser and better integrated network which is held to represent the effective emergence of a long-distance commercial economy up to the national level (Britnell 1995).

hundred or even several thousand inhabitants, requires the emergence of other forms of social organization (ranking or horizontal social segmentation), the resulting large communities have the potential (realized in widely-varying times and places) to develop into small polity-city state networks; 6. In tandem often but sometimes independently, recurrent regularities in the placing of district ‘centralplaces’, especially market towns, have long been identified and analyzed by geographers. Archaeology has neglected these properties in the last generation, but there is much to be learned by testing for the regional operation of recognized recurrent characteristics of simple central-place systems (often a day-return radius appears limiting on participation by rural inhabitants, creating radii of servicing of 2-3 hours travel and intercenter distances of 10-30 or so kilometers); 7. A promising area of analysis, frequently closely-tied to the emergence of central-place and/ or market town networks, focuses on the parasitic rise of local large settlements with the power to grow on the surplus products of a surrounding cluster of lesser rural settlements (cf. Wilkinson 1994). Differences can be observed between hierarchies of this sort where military power and tribute typify the system of dependency, and those where factors of marketing and economic or service specialization tie such hierarchies closer to modern forms of settlement hierarchy.

References Alcock, S.E. 1993, Graecia Capta. The Landscapes of Roman Greece. Cambridge: Cambridge University Press. Alcock, S.E. 1994, Breaking up the Hellenistic world: survey and society, in: I. Morris (ed), Classical Greece. Ancient Histories and Modern Archaeologies, Cambridge: Cambridge University Press: 171-190. Attema, P., E. van Joolen & P.M. van Leusen 2001, A Marginal Landscape: Field work on the beach ridge complex near Fogliano (South Lazio), Palaeohistoria 41/42 (1999/2000): 149-162. Barker, G. & J. Lloyd (eds) 1991, Roman Landscapes. Archaeological Survey in the Mediterranean Region. London: British School at Rome. Bintliff, J.L. 1994a, Territorial Behaviour and the Natural History of the Greek Polis, in: E. Olshausen & H. Sonnabend (eds), Stuttgarter Kolloquium Zur Historischen Geographie Des Altertums 4, Amsterdam: Hakkert Verlag: 207-49, Plates 1973. Bintliff, J.L. 1994b, The History of the Greek Countryside: As the Wave Breaks, Prospects for Future Research, in: P.N. Doukellis & L.G. Mendoni (eds), Structures Rurales Et Sociétés Antiques, Paris: Les Belles Lettres: 7-15. Bintliff, J.L. 1995, The Two Transitions: Current Research on the Origins of the Traditional Village in Central Greece, in: J.L. Bintliff & H. Hamerow (eds), Europe between Late Antiquity and the Middle Ages. Recent Archaeological and Historical Research in Western and Southern Europe, Oxford: BAR Publishing. BAR International Series 617: 111-30.

At present it seems reasonable to suggest that there is a general tendency in the long-term for settlement hierarchies of a relatively parasitic nature to give way to networks where large, upper-level settlements provide useful economic and social services complementary to what is available at lower levels of the settlement hierarchy. Much attention therefore is being focused on the timing and spread of commercial and economizing forces into the countryside, with reference to geographical classics such as the 19th and early 20th century studies of Von Thünen and Christaller. Recent work has indeed suggested that the hinterlands of ‘megalopoleis’ such as ancient Athens or Rome were already being transformed into urban market-focussed specialized production zones with new kinds of ‘suburban’ settlement (Bintliff 1994a; Morley 1996), whilst even small regional market towns may well be acting in defined central-place niches in antiquity (for the Roman town of Iol Caesarea cf. Bintliff 1997b after Leveau). The sophisticated research program using survey and excavation to investigate the Gallo-Roman settlement systems in Southern France, initially carried out under the auspices of the Archeomedes Project (Durand-Dastès et al. 1998), uses elaborate statistical and computerized database management and spatial analyses to delineate such recurrent networks of hierarchised sites and place them into interpretations informed by geographical theory. In England, it has even proved possible to study the comparative degree of commercialization of the rural market centers through a comparative analysis of the number and spacing of such centers across the same landscape in Roman, earlier Medieval and then High

Bintliff, J.L. 1996, The Archaeological Survey of the Valley of the Muses and Its Significance for Boeotian History, in: A. Hurst & A. Schachter (eds), La Montagne Des Muses, Geneva: Librairie Droz: 193-224. Bintliff, J.L. 1997a, Regional Survey, Demography, and the Rise of Complex Societies in the Ancient Aegean: Core-Periphery, Neo-Malthusian, and Other Interpretive Models, Journal of Field Archaeology 24: 1-38. Bintliff, J.L. 1997b, Catastrophe, Chaos and Complexity: The Death, Decay and Rebirth of Towns from Antiquity to Today, Journal of European Archaeology 5: 67-90. Bintliff, J.L. 1997c, Further Considerations on the Population of Ancient Boeotia, in: Bintliff, J.L. (ed), Recent Developments in the History and Archaeology of Central Greece, Oxford: BAR Publishing: 231-52. Bintliff, J.L. 1999a, Ch.13: Settlement and Territory, in: Barker, G. & A. Grant (eds), Companion Encyclopedia of Archaeology, London: Routledge: 505-45. Bintliff, J.L. 1999b, Structure and Contingency. Evolutionary Processes in Life and Human Society. London: Cassell.

33

RPC 2000 conference proceedings

Greek Countryside. The Southern Argolid from Prehistory to the Present Day. Stanford, California: Stanford University Press. Kiel, M. 1997, The rise and decline of Turkish Boeotia, 15th-19th century, in: Bintliff, J.L. (ed), Recent Developments in the History and Archaeology of Central Greece: 315-358. Oxford: BAR International Series 666. Kolb, F. (ed) 1993, 1995, 1996, 1998, Lykische Studies 1-4. Asia Minor Studien 9, 18, 24 & 29. Bonn: Rudolf Habelt Verlag. Kuna, M. 1991, The Structuring of Prehistoric Landscape, Antiquity 65: 332-47. Kuna, M., M. Zvelebil, P.J. Foster & D. Dreslerova 1993, Field Survey and Landscape Archaeology Research Design: Methodology of a Regional Field Survey in Bohemia, Pamatky Archeologicke 84(2): 110-30. Kuna, M. 2000, Surface Artefact Studies in the Czech Republic, in: J. Bintliff, M. Kuna & N. Venclova (eds), The Future of Archaeological Field Survey in Europe, Sheffield: Sheffield Academic Press: 29-44. Lehmann, H. 1939, Die Siedlungsräume Ostkretas, Geographische Zeitschrift 45: 212-28. Lewin, R. 1993, Complexity. Life at the Edge of Chaos. London: J.M. Dent. Lolling, H.G. 1989, Reisenotizen Aus Griechenland 1876 Und 1877. Berlin: Reimer Verlag. McDonald, W.A. & G.R. Rapp (eds) 1972, The Minnesota Messenia Expedition. Reconstructing a Bronze Age Regional Environment. Minneapolis: University of Minnesota. Millett, M. 1991, Pottery: population or supply patterns?, in: G. Barker & J. Lloyd (eds), Roman Landscapes. Archaeological Survey in the Mediterranean Region, London: British School at Rome, 18-26. Morley, N. 1996, Metropolis and Hinterland. The City of Rome and the Italian Economy, 200 BC - AD 200. Cambridge: Cambridge University Press. Neustupny, E. 1991, Community areas of prehistoric farmers in Bohemia, Antiquity 65: 326-331. Potter, T.W. 1979, The Changing Landscape of South Etruria. London: Elek. Raynaud, C. 1990, Le Village Gallo-Romain et Médiéval de Lunel-Viel (Hérault). Paris: Annales Littéraires de l’Université de Besançon. Reed, M. & D.L. Harvey 1992, The New Science and the Old: Complexity and Realism in the Social Sciences, Journal for the Theory of Social Research 22: 353-80. Snodgrass, A. 1994, Response: The Archaeological Aspect, in: I. Morris (ed), Classical Greece. Ancient Histories and Modern Archaeologies: 197-200. Cambridge: Cambridge University Press. van Andel, T.H., C.N. Runnels & K.O. Pope 1986, Five Thousand Years of Land Use and Abuse in the Southern Argolid, Greece, Hesperia 55(1): 103-28. van der Leeuw, S.E. & J. McGlade (eds) 1997, Time, Process and Structured Transformation in Archaeology. London: Routledge. Whitelaw, T. 2000, Reconstructing a Classical Landscape with Figures: Some Interpretive Explorations in North West Keos, in: R. Francovich & H. Patterson (eds), Extracting Meaning from Ploughsoil Assemblages: 227-243. Oxford: Oxbow Books.

Bintliff, J.L. 2000, Deconstructing ‘The Sense of Place’? Settlement systems, field survey, and the historic record: a case-study from Central Greece, Proceedings of the Prehistoric Society 66: 123-149. Bintliff, J.L. & P. Howard 1999, Studying needles in haystacks. Surface survey and the rural landscape of Central Greece in Roman times, Pharos. Journal of the Netherlands Institute at Athens VII: 51-91. Bintliff, J.L., P. Howard & A.M. Snodgrass 1999, The hidden landscape of Prehistoric Greece, Journal of Mediterranean Archaeology 12(2): 139-168. Bintliff, J., M. Kuna & N. Venclova (eds) 2000, The Future of Archaeological Field Survey in Europe. Sheffield: Sheffield Academic Press. Bintliff, J.L. & A.M. Snodgrass 1985, The Boeotia Survey, a Preliminary Report: The First Four Years, Journal of Field Archaeology 12: 123-61. Bintliff, J.L. & A.M. Snodgrass 1988a, The End of the Roman Countryside: A View from the East, in: R.F.J.Jones, J.H.F. Bloemers, S.L. Dyson & M. Biddle (eds), First Millennium Papers: Western Europe in the First Millennium Ad, Oxford: British Archaeological Reports: 175-217. Bintliff, J.L. & A.M. Snodgrass 1988b, Off-Site Pottery Distributions: A Regional and Interregional Perspective, Current Anthropology 29: 506-13. Britnell, R.H. 1995, Commercialisation and Economic Development in England, 1000-1300, in: R.H. Britnell & B.M.S. Campbell (eds), A Commercialising Economy. England 1086 to c.1300, Manchester and New York: Manchester University Press: 7-26. Brown, A.E. (ed) 1995, Roman Small Towns in Eastern England and Beyond. Oxford: Oxbow Books. Cambi, F. & E. Fentress 1989, Villas to castles: First millennium AD demography in the Albegna Valley, in: Randsborg, K. (ed), The Birth of Europe, Roma: Analecta Romana Instituti Danici, Suppl. XVI: 74-86. Carter, J.C. (ed) 1990, The Pantanello Necropolis 1982-1989. An Interim Report. Austin: The University of Texas at Austin. Cherry, J.F. 1979, Four Problems in Cycladic Prehistory, in: Davis, J. & J.F. Cherry (eds) Papers in Cycladic Prehistory, Los Angeles: University of California: 22-47. Cherry, J.F. 1983, Frogs Round the Pond: Perspectives on Current Archaeological Survey Projects in the Mediterranean Region, in: D.R. Keller & D.W. Rupp (eds), Archaeological Survey in the Mediterranean Area, Oxford: British Archaeological Reports: 375-416. Dietz, S., L. Ladjimi Sebai & H. Ben Hassen (eds) 1995, Africa Proconsularis. Regional Studies in the Segermes Valley of Northern Tunisia, I-II. Århus. Durand-Dastès, F. et al. 1998, Des Oppida aux Métropoles. Archéologues et Géographes en Vallée du Rhône. Paris: Anthropos. Fulford, M. 1987, Economic Interdependence among Urban Communities of the Roman Mediterranean, World Archaeology 19: 58-75. Greene, K. 1995, Archaeology: An Introduction. London: Routledge. Heidinga, H.A. 1997, Medieval Settlement and Economy North of the Lower Rhine. Assen/Maastricht: Van Gorcum. Jameson, M.H., C.N. Runnels & T.H. van Andel (eds) 1994, A

34

General Papers

Wilkinson, T.J. 1989, Extensive Sherd Scatters and Land-Use Intensity: Some Recent Results, Journal of Field Archaeology 16: 31-46. Wilkinson, T.J. 1994, The Structure and Dynamics of DryFarming States in Upper Mesopotamia, Current Anthropology 35: 483-520.

35

RPC 2000 conference proceedings

2.5

Some Current Approaches to Protohistoric Centralization and Urbanization in Italy Alessandro Vanzetti

2.5.1 Introduction The geographical variety of the Italian landscape and the differing access of communities to resources (especially ores) have played their part in forming distinct regional socio-cultural systems which passed through the process of (proto-) urbanization at different times (figure 1). As will become clear below, the concept of the “proto-urban centre” as adopted in the Italian debate (Peroni & di Gennaro 1986) roughly corresponds to Renfrew’s definition of the Early State (Renfrew 1975). The key-problem is whether developments in the later Bronze Age and early Iron Age in Italy indicating early centralization and (proto-) urbanization were modeled after the Greek experiment or were largely contemporaneous and independent. For instance, in Etruria the transition towards urban structures was approximately contemporary with that in Greece, and recent research suggests it was part of an overall Mediterranean process. Etruria is therefore at the center of today’s debate on protohistoric centralization and urbanization in Italy. The study of the formative period of Etruscan urban centers is, in current Italian research practice, situated at the crossroads between Etruscology (a branch of Classical Archaeology) and the much younger discipline of Protohistory. Thus we have two scenarios, strictly bound to the respective approaches. Study perspectives, such as “primitivism” and “modernism”, or the degree to which written texts (of much younger date) are referred to, have played significant roles in the formation of these alternatives. Summarized (E = Etruscology, P = Protohistory) these are: • a gradual, long-term development before regional socio-cultural systems underwent the external stimulus of Greek colonization (E), as opposed to early and fairly rapid internal Italic transformation (P); • strong asymmetry between Etruria and the other Italic regions (E), versus weak asymmetry (P); • an emphasis on ethnic-cultural values (E), as opposed to social values (P), as the cause of early proto-urbanization outside Etruria, in cases where such complexity is associated with “Villanovan” characteristics (since in Etruria the Villanovan society is the forerunner of the Etruscan one).

Figure 1: Main Italian areas discussed in the paper.

research is interested in regional developments (Latium, Apulia, Sibaritide) more than in general modeling (Attema et al. 1998, Voorrips et al. 1991, Yntema 1991). In this paper a number of sample areas will be discussed in the light of the ideas and interpretations of two main schools; • the Roman school of protohistory (headed by R. Peroni, with ideas partly shared by A.M. Bietti Sestieri). The main sample area discussed here will be Etruria, while some observations are made on Latium and the Sibaritide as well. • the Padua school with its emphasis on geology and the environment and a clear preference for the theories and methods of the New Archaeology, while maintaining relevant inputs from Italian theory building (as for instance Etruscology). Their main study area is the Veneto. 2.5.2 Etruria Four intersecting and competing approaches to the origins of Etruscan cities may be distinguished: • Etruscology, focusing on urban centers and their necropoleis, pays special attention to the orientalizing period and Greek influence (e.g. Cristofani 1986). • the Roman school of topography, which has provided important data on regional settlement (e.g. Guaitoli 1981). • Ward Perkins’ English school, cradle of modern Landscape Archaeology, which has stimulated studies result-

Landscape Archaeology, the English contribution, is important to the debate. At first its views were substantially in agreement with those of Etruscology (Ward-Perkins et al. 1961, 1968, Potter 1979), but now the results of Landscape Archaeology tend to be translated into the terms of historical process using the Braudelian temporal paradigm (Barker 1995a, 1995b, Bintliff 1997). Dutch as well as English archaeology focuses steadily on man’s impact on the landscape using the analytical tool of Land Evaluation. The very title of this conference makes clear that Dutch 36

General Papers

Figure 2: Settlement evolution in Southern Etruria from the final Bronze age (left) to the early Iron age (right). From di Gennaro 2000: figures 5 and 7.

2.5.2.1

Early nucleated centers or gradual synoikism (problems 2 and 3) The Roman school of protohistory (P for short) emphasizes the sudden and revolutionary character of the transition from the final Bronze Age settlement system to the large centers of the early Iron Age. This transition would have been characterized by the formation of large protourban centers with territories of 1000 to 2000 km2, concentrating a population that in the preceding period was dispersed in about 20-30 settlements controlling small territories of between 20-50 km2 (figure 2; synthesis in Di Gennaro 2000). A Roman School-type social interpretation would claim that the transition was caused by the introduction of land ownership in parts of the territory at least. This would place the first signs of urbanization well before the period of Greek colonization, and even before the period of precolonization, and conflicts with the Etruscologic view that Greek colonization was the main force for change in the Iron Age. The latter see urbanization as a gradual process in which small sites continued to exist while large centers were being formed, and the results of Ward Perkins’ surveys appeared to support their views. Originally the plateaus on which the large early Iron Age centers were formed would have been settled by independent small

ing in detailed topographical knowledge of the centers and their territories and promoted environmental studies (e.g. Ward-Perkins et al. 1968). • the Roman school of protohistory, which promotes territorial studies and directs attention to the settlements on the heights and dominant plateaus (e.g. di Gennaro 1982, Peroni & di Gennaro 1986). While no one doubts that a fundamental restructuring of the settlement system took place during the final Bronze Age and early Iron Age, when the dense system of hilltop villages was replaced by settlement on large plateaus – sites of the future Etruscan cities -, there remain three problems characterizing the study of this transition. These focus on: 1 uniformity in the types of settlement of the final Bronze Age 2 speed of transformation of a diffuse settlement system into a concentrated (more or less nucleated) one 3 the way in which large plateaus were settled in the course of the Iron Age.

37

RPC 2000 conference proceedings

Figure 3: The plateaus of the proto-urban centers of Tarquinia, Veio and Vulci. The surface clusters of settlement finds of the early Iron age and the corresponding cemeteries are outlined. From Pacciarelli 1991b, Bietti Sestieri 1997: figure 3.

communities in hamlets with their own necropoleis (figure 3). A process of “synoikism” would have unified the hamlets into larger centers. Synoikism seems, however, a wrong word here, as it implies a sudden concentration at one place of various settlements that before were located at considerable distance from each other, rather than the imperceptible fusion of neighboring sites (Peroni 2000). Site formation processes are a pressing problem in the Etruscan case (see Arnoldus-Huyzenveld et al. 1990): is it permissible to interpret the discontinuous traces of archaeological material at the surface as distinct sites (hamlets) of the early Iron Age? And is the evidence of distinct necropoleis good support (since nucleated settlements too may have distinct necropoleis along the roads of communication: Cardosa 1995, Bietti Sestieri 1997, di Gennaro, Guidi 2000)?

Alessandro Guidi (P) demonstrated even in 1985 that at Veio the surface scatters were random and did not fit a nearest-neighbor model (Guidi 1989). Thus the pattern would probably indicate a full occupation of the plateau of Veio. Di Gennaro and Schiappelli (P) have reached the same conclusion after re-studying data from Veio collected by the British School at Rome (pers. comm.; di Gennaro, Amoroso & Schiappelli in press). Mandolesi and Pacciarelli (P) surveyed respectively Tarquinia and Vulci (Mandolesi 1999, Pacciarelli 1991a). In both cases the early Iron Age material was found dispersed all over the plateau, indicating the feasibility of the hypothesis. Visibility studies are of obvious importance here, along with study of the processes of soil movement and erosion (today as in the past) in order to explain why finds 38

General Papers

Figure 4: Hypothesis of gradual selection of protohistoric settlements in the territory of Tarquinia. From: Mandolesi 1999: figure 82.

3 a third assumption is that of continuity, which holds that the absence or poor representation of chronological periods bracketed between preceding and following phases, is due to sample bias and/or to poor knowledge of the material record of certain periods (e.g. recent Bronze Age in Latium).

are generally more abundant on the rims of the plateaus. If surveys of the large plateaus, the locations of future Etruscan sites, yield evidence supporting the existence of unified centers dating from the early Iron Age, then should the dispersed presence of final Bronze Age materials on these same plateaus be seen in the same light? And if the plateaus already knew settlements in the final Bronze Age, what was their relation to contemporary sites dispersed in the countryside? Was the ‘proto-urban’ transition thus, after all, gradual? However this may be, a further retro-dating would argue even more strongly against the traditional ideas of Etruscology, placing as the early dating would do, the origins of the ‘proto-city’ further from the colonial phase.

To validate all three assumptions a serious extension of settlement studies is needed going beyond simply surveying and, notably, will have to involve excavation. Meanwhile, the quality of interpretations of settlement evolution must be principally judged in terms of their comprehensive historical coherence. 2.5.2.3 The case of Tarquinia Recently A. Mandolesi (1999) carried out a settlement study of the plateau and territory of Tarquinia for the early Bronze Age – early Iron Age period, using published, archival, and new survey data. From his work the importance of the teleological argument in the thought of the Roman school of protohistory becomes clear. Some topics raised by the author are: • the selection of the dominant elevations in the landscape for settling, a choice which is especially evident from the later phases of the middle Bronze Age • the interpretation of cases where lower zones are chosen for settlement from the recent Bronze Age onwards, as either driven by demographic expansion from a neighboring plateau site, or chosen with the aim of installing an industrial activity area • the recurring hypothesis of some sort of federal relationship between different settlements.

2.5.2.2

Uniformity of the settlement model in the Final Bronze Age (problem 1) Territorial research in Italy has been dominated by extensive surveying, often in areas of poor visibility; as we have seen, the formation processes of the surface record are complex; furthermore, excavation is often lacking, or limited in its interpretative potential. Those problems undermine attempts at thorough historical reconstruction. The Roman school of protohistory has tried to by-pass these problems by operating on the following basic assumptions: 1 It is generally accepted that the geomorphologic unit is a good estimate for the extent of the ancient settlement (even if only a few fragments have been found and no excavations have been carried out). We may refer to this as the assumption of unity of settlement 2 there is a tendency to believe that finds on the slopes (and sometimes on the valley floors) have been displaced from the associated plateaus. We may refer to this as the assumption of settlement

In this scheme, Mandolesi proposes Tarquinia as the outcome of a millennium-long process of change in the settlement system within a substantially (and teleologically) stable geographical configuration (figure 4). Developments in the final Bronze Age settlement system are seen as decisive for the unification of Tarquinia. The proposal is interesting, but the study is hampered by its local perspective and a lack of sufficient data to allow the possibility of proof. For instance, F. di Gennaro (2000), drawing on the local hydrography, has proposed that the territorial configura-

The above have been very little discussed in the past, suggesting as they do a more varied patterning of the settlement systems (difficult to prove or assume as a rule anyway, when relying on existing data). These two assumptions appear to be the biggest source of doubt as regards the interpretations of the Roman school; they have been applied, for instance, to both the Etruscan region and the Sibaritide. 39

RPC 2000 conference proceedings

Mediterranean process: as the outcome of the various experiences, especially stimulated by the Near East, that were lived through after the demise of the Mycenaean world. It was a process in which both Greece and Italy took part (Bietti Sestieri 1997:399). Development towards urban complexity in Etruria was among the earliest in Italy, stimulated by the general Mediterranean tendency and by sudden internal growth. These developments caused a chainreaction. 2.5.3 Etruria and Latium Developments in Latium were different. One does not see there a clear process of selection and concentration in particular sites. The main sites tended to be occupied for long periods, and to grow in stages (Peroni 1989, Bietti Sestieri 1992, di Gennaro & Guidi 2000); anyway, one can cite in the early Iron age the quick rise of Rome to proto-urban status with regional power (Carandini 1997, Guidi 2000). A. Guidi (1985) contrasted the two situations using rank/size diagrams to evaluate the dimensional relation between sites and the integration of territorial systems, stating that “the incorporation of Rome into the Etruscan state was a decisive element in the creation of a highly integrated system in the whole Latial area”. This would have happened between the 8th and 7th centuries BC. The present author, however, suggests that it is possible to read in the rank/size diagrams tendencies towards a hierarchical structure and a substantial integration of the settlement system as early as the Latial IIB/III phase, i.e. before that incorporation, which therefore cannot have been the primary cause. It is rather Rome’s spatial developments in the Latial phase IIa/start IIb (the transfer of the necropoleis from the Forum to the Esquiline as Rome expanded from the nucleus Campidoglio/Palatine to include the Forum and the Esquiline) which indicates the start both of the protourban phase and of the Latial integration which in the end led to competition with Etruria (Peroni 1988, Bietti Sestieri 1992, Bettelli 1997, in a peculiar perspective Carandini 1997). The time-lag for the start of proto-urban process in both regions is therefore short: in traditional chronology, the start of the 9th century in Etruria and second half of the 9th century in Latium. As far as socio-political developments in Latium are concerned, A.M. Bietti Sestieri (1992, 2000) has concluded from her study of the necropoleis of Osteria dell’Osa and Castiglione, that the community of Gabii whose dead are buried in these necropoleis was still tribally organized in the Latial IIA and IIB phases. Gabii, in her opinion, was not yet a unitary proto-urban settlement. Both R. Peroni (1989) and A. Carandini (1997), however, are of the opinion that Gabii was already organized as a chiefdom and possibly was already a unitary proto-urban settlement, though perhaps not densely settled. For Bietti Sestieri the development of Rome was favored above Gabii and the rest of Latium by its position near Etruria, but the latter caught up through the influence of Rome. Peroni and Carandini do not see such distinct differences in social structure, nor do they think that Rome’s development was heavily dependent on Etruria.

Figure 5: Final Bronze age settlements in Southern Etruria, set against the areas of the early Iron age proto-urban territories (the future proto-urban centers are marked by oblique lines); the hatched lines indicate four possible “federal” groups defined using the local hydrography. From di Gennaro 2000: figure 6.

tion of the Tarquinia area in protohistory may have been different from that of historical times (figure 5). Let’s come to other aspects of Mandolesi’s study. As discussed before, data from Tarquinia contribute relevantly to criticism of the traditional Etruscological standpoint regarding the unification of small villages on the large Etruscan settlement plateaus. The presence of some final Bronze Age sherds from the main plateau suggests an earlier, more dynamic and accelerated process of urban coalescence than does the classic interpretation of the Roman school of protohistory (still to be clarified in detail). The creation of the proto-urban center involves the genesis of other settlement nuclei in the Tarquinia area (Tarquinia Civita, Castellina, Monterozzi) and on the coast (the Saline area), in a complex pattern of landscape use (figure 6). This early dating of the proto-urbanization phase and the integrated pattern implied in Etruria contrasts, for example, with M. Rendeli’s (1991) point of view. He associated Etruscan urbanization with Greek influence and casts serious doubts on early Iron Age “complexity” (see critics in Carandini 1997:457-464). It is the present author’s contention that in backdating proto-urbanization in Etruria one must view it as part of a 40

General Papers

Figure 6: The territory of the proto-urban center of Tarquinia in the early Iron age. From Mandolesi 1999: figure 84.

While the importance of the connection with Etruria is still debated (was there a form of dependence through a chain reaction?), itt must be concluded, in any case, that even in Latium centralization and (proto-) urbanization preceded the pre-colonial Hellenic influence.

ilarities (material culture, funerary rites) to the Etruria of the Villanovan period. In particular; Pontecagnano, Sala Consilina, Capua in Campania; Fermo in the Marche; Verucchio and Bologna in Emilia-Romagna. The question is whether their cultural traits correspond to the Etruscan ethnos (e.g. Torelli 2000). This remains to be proved rather than simply assumed. And if there is talk of a transfer of cultural information and/or persons, one should try to evaluate which level of affinity with Etruscan culture signi-

2.5.4 Etruria and Italy There has been a decade-long discussion on the significance to be attributed to those areas showing cultural sim41

RPC 2000 conference proceedings

Figure 7: Spatial arrangement of known settlements in the plain of Sybaris in the early and middle Bronze Age, in the early Iron Age, and in the Archaic age. The supposed territory of Torre Mordillo in the early Iron age is filled in gray. From Peroni 1994, Vanzetti 2000.

42

General Papers

fies political dominance (and even colonization), and which signifies other forms of close interrelationship. It is the author’s contention that a transfer of cultural information ultimately cannot be distinguished from a transfer of persons, and that in the final analysis there will not be a great difference between strong acculturation and colonization.

Bronze Age of southern Italy (Bietti Sestieri 1988). Ironically, Bietti Sestieri’s “ex Etruria lux” intrusion in the Veneto has led to widespread discontent and contrast among the scholars of the Veneto, but the debate is now open, and must still reach the phases of conflict-rejection implied by Bietti Sestieri’s model itself.

The current tendency in Etruscology (Bartoloni 2000) and partly also in Protohistory (de Marinis 1994) is to revive the direct solutions proposed by the historical-cultural school (relying on the material facts), and hypotheses of continuity based on historiographic data (retroactive validity of the sources). Peroni’s ideas regarding the nature of the relationship between Villanovan Etruria and outlying areas are that Pontecagnano was strategically planned to control the lower Tyrrhenean area, while Fermo and Sala Consilina were not (Peroni 1996:430-6, 454-6). All other centers with Villanovan characteristics outside Etruria may have acquired these characteristics through their early development towards proto-urban status. According to the model of proto-urban development, this did not take place entirely outside of Etruscan influence. Peroni has put forward the hypothesis that proto-urbanization was linked to the coming into being of partial private property, and this would have been the driving factor in the process (Negroni Catacchio & Peroni 1979, Pacciarelli 1991b).

2.5.5

A situation without full-grown (indigenous) protourban outcomes: Calabria Three zones in Calabria have been the subject of detailed territorial studies; the Sibaritide (Peroni and colleagues: Peroni 1994, Peroni & di Gennaro 1986, Vanzetti 2000); the territory of Kroton (D. Marino, 1998); the promontory of Tropea (M. Pacciarelli, R. Varricchio and a local archaeological group: e.g. Pacciarelli 1989-90). The present author here discusses the Sibaritide. In Calabria developments towards proto-urbanization, although less advanced than in Etruria, were cut short by the Greek conquest and the foundation of their colonies. In the case of Kroton, the Greek colony was sited on the remains of the indigenous settlement. Our poor knowledge of the indigenous remains makes it impossible to judge the level of development of the community living there, though it may have been more advanced than that in the Sibaritide. In the Sibaritide, reconstruction of the settlement patterns (figure 7) is based on extensive surveys, which were primarily targeted at the Pleistocene hilltops and terraces favored by the protohistoric people. The data have been integrated with the results of excavations on the sites of Broglio di Trebisacce, Torre Mordillo near Spezzano Albanese, Timpone della Motta at Francavilla Marittima, and Cittavetere di Saracena. In the middle Bronze Age the distribution of settlements over the Pleistocene terraces is very even (one every 3-5 km) and characterized by the conditions necessary for dry farming and tree cultures (olives, later perhaps walnuts), a stable hydrogeology, and the use of naturally defended places. Towards the final Bronze Age a process of concentration and selection took place and only the larger and best-defended sites survive. The intervening phase of contact with the Mycenaeans (recent Bronze Age) does not correspond with a particular phase of contraction or expansion, as it did not in the later pre-colonial period. The processes of selection and expansion seem to be internally driven. In the course of the Iron Age there was renewed expansion resulting from developments in the final Bronze Age; this occurred above all in the center of the plain, and is interpreted as the outcome of a drive for capillary occupation of the territories belonging to the more important centers. This occurred in the form of a competitive process of expansion and developing local supremacy, which continued in parallel with the pre-colonial phase and was eventually interrupted by the arrival of the Greeks.

Bietti Sestieri (1997, 2000) reads the early cultural relationship between Villanovan Etruria and the outlying areas as one directed and controlled by Etruria in order to sustain, by means of personal investment, the productive and distributive metallurgical network. In a slightly later phase she discerns a progression towards dominance for Etruria at a national level, beginning in the final Bronze Age with the rise of centers such as Frattesina (socio-economic control model), and completed in the early Iron Age with the foundation of the Villanovan centers in Etruria itself and in the outlying areas (colonial model). Colonna (1994) has stated cautiously that Frattesina most likely marks the route along which the Etruscans penetrated Italy, assuming they came from the Aegean. Of course Bietti Sestieri’s position here is that of autochthonous origins - the precise opposite to Colonna’s. In her view, the Villanovan impact on the Veneto was so deep and prolonged because of the crisis of the palafitte/terramare cultural complex and the consequent retardation of the process of political structuring of the communities of the Veneto. On the other hand, the appearance of the Paleo-Venetian facies in an advanced phase of the early Iron Age marks the disappearance of “cultural” elements of the Villanovan facies in the plain north of the river Po. According to Bietti Sestieri, this might even have taken the form of an organized reaction against the presence in their territory of population nuclei of external provenance, or systematically linked to the Villanovan or Tyrrhenean communities in the Po area. The present author recognizes here an interpretative mechanism that he describes as intrusion-contrast-conflictrejection, a mechanism also applied by Bietti Sestieri to the so-called ‘Mycenaean connection’ in the middle and late

R. Peroni (Peroni 1994, Peroni & di Gennaro 1986) has analyzed the relations between the settlements on the basis of: • the defining of a hierarchy among larger and smaller centers using as criteria the size of the plateau/hilltop bearing remains; its dominance over adjacent territo43

RPC 2000 conference proceedings

Figure 8: Hypothetical model of the distribution and movement of sites in the hill system of the Vicentino during the settlement cycle of the middle and recent Bronze age; notice the cyclical displacement of location hypothesized for settlements in the hilly area. From Balista & Leonardi 1985: figure 9.4.

ries; its distance from other larger and smaller sites; and its duration of occupation. • the hypothesis that the settlements wielded effective control over the surrounding countryside, the limits whereof are assumed to correspond with the more important physical geographical boundaries (rivers, mountain crests). These territories can be approximately indicated using geographical models such as Thiessen polygons. • the proposition of “historical” mechanisms in order to explain movement from one site to another (cf. Mandolesi 1999, but for single situations).

and is spread over various areas. As early as the final Bronze Age the acropolis is well defended by an earthwork (vallum et fossa). On the acropolis are important dwellings and storage rooms, and indications of metalworking (iron working occurring as early as the end of the Bronze age). The difficulty of defining a site’s territory and its place in a hierarchy can only be solved by more excavations on both the larger and the smaller sites within the hypothetical territory. One must also have a keen eye for the typological and archaeometric studies of the material, that may indicate the existence of relationships. A specific study of these aspects was carried out by Levi et al. in 1999. It was found that some large storage jars (dolii) of the recent Bronze Age (and a few of the final Bronze Age) circulated intensively in the Sibaritide, probably because of their content, while at Broglio during the Iron Age up to 50% of the fine ware was imported from the south and central Sibaritide. This indicates specialization and the existence of a regional or sub-regional market, and a trend towards regional unification.

Of course, the same problems play a role here as were already noted for Etruria. How can we ascertain the real extent of the ancient settlements and their phases of contraction and expansion, and in what way have erosion, deposition and modern land-use affected the archaeological record? Experience at Broglio di Trebisacce, from surface examination and excavation, leads the present author to state that it is practically impossible to estimate the original population of such a site quantitatively; it can only be modeled. The site of Broglio di Trebisacce is 11 ha in extent

A rank-size analysis by the present author (Vanzetti 2000b) on the basis of Peroni’s territorial analysis (1994) demon44

General Papers

strates that, while the dimensional relationships between the settlements show a constantly convex structure and therefore indicate a non-hierarchical internal relationship (accepting Peroni’s territorial division and using the total plateau size associated with each site), the rank-size structure tends to log-normality in the transition to the Iron Age. If this fact is not considered very convincing, it can, nevertheless, serve as an indication of the general tendencies of the system; the Greek intervention, with the founding of Sybaris and the destruction of almost all the villages surrounding the plain, generated a strongly centripetal force in the rank-size distribution (known as the “primate” type). This type of system collapse, that appears to have occurred as the territory came under colonial domination, is morphologically analogous to that demonstrated for south Etruria in the period of proto-urbanization at the beginning of the Iron Age. That which was not completed as an indigenous development, became so by the superimposition of the Greek element.

2000 for a recent standpoint) - their research clearly has had an impact at the national level (expressed in the respect it is shown, rather than in a general familiarity with, or agreement to, its contents). The middle and lower plain of the Veneto up to its hilly and mountainous area is the territory studied, thus comprising a fair number of physical-geographical units that are more or less seen as one system in protohistory. The research material consists mainly of archival data (bibliography / Soprintendenza records), and of local excavations; but in 1986 Armando De Guio started a program of systematic survey (Balista et al. 1998) that has resulted in various projects throughout the research area. As there is no pretension that all the sites in the area from any one period are known, statistical evaluation plays an important part. Moreover, it is assumed that sites may have had a temporary character. More-over, the hypothesis of cyclical occupation of specific locations is put forward on the basis of data from stratigraphic research and from that gathered (by excavation or survey) at single sites, viewed in relation to the distribution of remains from various phases (figure 8). Montebello Vicentino is one good example, where strata of occupation and colluviation due to slope erosion alternate with incipient development of pedogenetic sections. This is related to the topographical dislocation of successive occupational phases; the recent Bronze Age and the final Bronze Age on the sloping terraces, the early Iron Age close to the plain; the advanced Iron Age, after a period of abandonment, returning to the slopes.

The relation between Greek and native in the Sibaritide has recently been reviewed by M. Maaskant-Kleibrink (forthcoming; Bulletin Antieke Beschaving) on the basis of a reconsideration of the data from Francavilla Marittima. She attempts to show there was no conflict between the incomers and the long-standing communities, while the current opinion (sustained by both Peroni and the present author) is that there was conflict, not so much in the pre-colonial period, but in the colonial aftermath. It seems indeed plausible that the Greek role in the indigenous dynamic of the pre-colonial period was cooperative and will have profoundly influenced the local communities, but just because of that, it must have had to do with an ideological brainwashing prior to colonial domination, which will have stifled local energy. If some communities, such as that of Francavilla, were able to continue developing in another form, they were the exception rather than the rule, at the root of which must have stood the sanctuary which endured from pre-colonial to colonial times.

This approach can be summarized as follows: sites are treated as either present or absent, rather than as entities whose extent can be determined, and which had a reciprocal relation to a territory. The School of Padua (excepting De Guio: De Guio et al. 1986) gives more weight to locational choices and overall density of occupation than to socio-political factors. Analyses of the Thiessen polygon type, or rank/size diagrams, receive little attention, nor do the relationships between individual sites in controlling a territory. Socio-political factors merit attention mainly on an aggregate level, on the basis of the concentration or abandonment of sites in particular zones. • There is no assumption of continuity, but rather a hypothesis of cyclical occupation. Absence of certain phases, in this view, presents no problem. • There is no assumption of settlement. Naturally defended sites are not immediately accorded a dominant socio-political role, and it is held that there is a great variety of locational choices subject to change over time. There is no reason to bring into organic relation sites on hilltops and sites in open positions (again excepting De Guio: De Guio et al. 1986). • The assumption of unity is applied, but not problematized.

The extension of the Sibaritide research remains a rare case in Italy, and many problems are still to be solved: • difficulties in sampling and in estimating whether the data are representative of the actual ancient situation; • analyses are based on targeted (non-random) surveys and on reconstructions of the territories and their extent, although these have only been marginally tested; • the insufficient quantity of data for each individual site (effective chronological span; spatial and functional interpretation). 2.5.6

A situation with a slower evolutionary rhythm: the Veneto The study of the evolution of the settlement systems of protohistorical Veneto is the principal research subject of the so-called School of Padua. Although this school’s publications have often been by single authors and are sometimes difficult to access - being published within differing circulation networks (Bagolan & Leonardi 2000, De Guio

The first analysis (Balista et al. 1982) tended to define settlement models above all through: • a sound definition of the geological/environmental context and of the potential resources, supported by data from excavations; 45

RPC 2000 conference proceedings

It becomes clear that the approach of the school of Padua is more environmental and less historical than that of the Roman school, more empirical and less deductive, with a good balance between data and critical observation. In the early Paduan publication (Balista et al. 1982) the complex analyses, however, hardly rose above the descriptive level. More significant elaborations of a historical model were only developed later, stimulated by the national debate on the concept of proto-urbanization, as for instance in Capuis et al. (1984), and more clearly in a later article by G. Leonardi (1992). The latter, on the basis of abundant archival data in two sample areas (the SE Veneto plain and the hills and mountains of the Vicentino, see figure 9), revealed the transitional period between the recent Bronze Age and the final Bronze Age to be one of a profound crisis, as in the whole of northern Italy. In the following period, i.e. the oldest phase of the Iron Age (traditionally 9th century BC), the number of settlements in the SE plain remained low, testifying to stable development, while a remarkable increase took place in the hills of the Vicentino. The two areas then saw a shared fall in settlement numbers in the recent phase of the early Iron Age (traditionally 8th century BC), which was followed (7th-6th century BC) by an increase, first in the SE plain, then in the Vicentino, along the same lines. To understand this development, one must consider the extent of the sites: from the final Bronze Age onwards sites of tens of hectares appear in the SE part of the plain, apparently replacing one another until the proto-urban center of Este materializes in the 8th century BC (cf. figure 9). In chronological order, these are Frattesina di Fratta Polesine, whose florescence was in the final Bronze Age; Montagnana, reaching its apogee in the beginnings of the Iron Age, and Este, which seems to have grown out of the partial displacement of a number of nearby hamlets, giving birth to a large unitary configuration in the recent phase of the early Iron Age (traditionally 8th century BC). The fall in settlement numbers of the traditional 8th century BC noted by Leonardi can therefore be related to this process of protourbanization. Not only Este, but also other sites in the Veneto plain (Padua and perhaps Oppeano and Gazzo, outside of Leonardi’s sampling areas) grew in extent in the same period. The renewed later expansion seems to have been driven by the settlements in the plain as the outcome of a process of stabilization of the political system. Whether the stable situation in the plain paired with the settlement expansion in the Vicentino at the outset of the early Iron Age, was even then dependent on the existence of integrated relations between the hilly area and the larger centers like Montagnana, which preceded Este and Padua, or whether it indicates a basically independent development of two political/economic systems, remains a question. The present author concludes that, while the wide-ranging questions originally posed have now been reduced in scope, the growing interest in a historical perspective makes the results of the work of the school of Padua interact with the analyses of the Roman School.

Figure 9: Upper, comparison between the quantity of settlements in the SE plain of Veneto and in the Vicentino, from the recent Bronze age to romanization. Lower, model of territorial evolution in the areas of Montagnana, Este and Padova (plain) from the recent Bronze age to the middle Iron age, and developments during romanization. Graphs use traditional dates expressed in centuries. From Leonardi 1992: figs. 27 and 28.

• a computer-based (analytical, sensu Clarke) evaluation of the association between various categories of localization and of potential resources used by the settlements; • an evaluation of settlement density in terms of the total number of sites and changes in birth/death rates during various phases. 46

General Papers

2.5.7

Conclusions

settlements that came into being in the recent phase of the early Iron Age (the same phase as that of the ascendancy of Rome), in combination with growing settlement densities like those described for Etruria and the Veneto.

2.5.7.1 Evolutionary dynamics of settlement Generally speaking, well-defined models exist for each phase in protohistoric Italy. In the course of the middle and recent Bronze Age, there was an increase in the settlement density of the territory, followed in the final Bronze Age by a crisis. This corresponded to a reduction in the number of settlements and a further selection of favored geomorphological location characteristics. The sites now active, often those most important in the preceding network, became the reference points in ever-growing territories. On a more detailed level, the alternation of periods of crisis and of expansion allows us to define the principal settlement cycles, with each area taking different paths. In the context of this paper the dynamics related to the protourban developments of the early Iron Age are important above all. It is now time to return briefly to the concept of protourbanization used by Italian scholars of protohistory. The fundamental observation to make, for whoever wants a clear-cut definition like the one proposed by G. Childe (1950) for the city, is that the basic criterion is simply one of dimensions - a proto-city is so because of its dimensions, and therefore its supposed population should be sufficiently high as to postulate organizational forms of a complexity that make it an embryonic city (Peroni 1989, Leonardi 1992, de Marinis 1994). Of importance here is the teleological argument, which rests on the empirical evidence that the large agglomerates defined as ‘proto-urban’ often, as in the Etruscan case, became true cities in the historical period. Observations on necropolis’ arrangement and on production organization have been sometimes quoted as confirmations. Apart from cyclical expansion and contraction, true synoikism may have been an important factor in the formation of proto-urban centers. Here we use synoikism in the Greek sense, of the transferring of various settlements from different parts of the territory to a single larger site, instead of the gradual unification of intermingling neighboring settlements into one settlement. In Etruria, and possibly also in the Veneto, the growth of proto-urban sites corresponds to the concentration of a large part of the population in the major centers and to the abolition of pre-existing territorial ties. Following this phase of depopulation, controlled anyway by the main center, a hierarchical process of repopulation took place at various times, led by the center itself within its territory, but possibly also affecting local expansion in relatively independent marginal areas. This cycle was, however, largely blocked in Latium, without phases of general collapse of the system. As in the Veneto, this may be due to its slower rhythm of development, but in Latium it was sustained through a more marked continuity, probably caused by accentuated competition between many centers with “proto-urban potential” in a restricted territory, which did not allow rapid unification in one center. One may read a period of competition for territorial control between various Latial proto-urban centers into the number of minor

The phases of clear and incisive external contacts - Mycenaean in the middle and recent Bronze Age, extending into the initial phase of the final Bronze Age, and Greek precolonial in the early Iron Age (the supposed contacts with Cyprus in between these two phases being less concrete or, in any case, less incisive in the structure of the communities) - seem to coincide with phases of settlement expansion rather than reduction. At the same time, the end of the contact period did not necessarily generate a serious crisis (cf. the Sibaritide in the final Bronze Age). It looks as if we may hypothesize that external contacts were, above all, established at times which were in themselves already expansive. Because of this, the actual contact is more of a consequence of the process towards complexity, than one of its causes. Regional development trajectories towards complexity differed in time, with doubtless an early role for Etruria; a role embedded in the general process of proto-urbanization at the Mediterranean scale. The Etruscan developments, though different in nature, did not happen after the Greek developments. To explain this process, the theory of Peer Polity Interaction is invoked by some, but the present author does not agree with this approach. An important point in the Italian debate is the possible central role Etruria has played at the national level. If we accept that the Etruscan development started a chain-reaction that led to waves of proto-urbanization through a process that one could describe as “diffusion”, then we still have to decide whether this happened by direct political intervention in the marginal areas from the start of the Iron Age, or even in the course of the final Bronze Age (Bietti Sestieri 1997). Such a reading would imply that we are dealing with a local Italian core-periphery model. Evidently it is not easy to escape the concepts of cultural inequality that have been hypothesized for the Greek-Italian relationship; only the actor has changed. Rather, there are at least two distant actors on a high level of social development in the course of the early Iron Age - Greece and Etruria (but perhaps Cyprus/Phoenicia should be seen as a third actor, independently of Greece). These two may have accelerated processes towards complexity that were already in motion, as in the Veneto and Latium, under the influence of the very same Mediterranean medium-term tendencies. 2.5.7.2 Databases and research necessities Important data can be retrieved from literature and archives (the Veneto is only one example), especially for specific parts of North and Central Italy, where there is a long ecclesiastical, aristocratic and bourgeois tradition of interest in the material culture of antiquity, now reflected in the activities of local amateur archaeological groups. Apart from the processing of existing data sets with new methodologies and using the growing body of data from 47

RPC 2000 conference proceedings

2.5.7.3

excavations, it is through test excavations and surveys, especially of the intensive type and when based on geomorphologic and pedological analyses and corrected for the relation between surface and sub-surface, that progress can be made.

Core research points and the Italian perspective on the problem

Problems and current solutions Geomorphologic conditions are very heterogeneous in Italy, causing variable visibility of the palaeo-surfaces (e.g. erosion and alluviation along the Apennine range, implying abraded hilltops and covered valleys). This causes obvious problems in defining sites in terms of density and extent; in estimating settlement density and population size, we need to realize that we only have minimum quantities to start from. The Roman school has tried to evade this problem by their three assumptions (of settlement, unity, and continuity). The School of Padua, on the other hand, for various reasons has not adopted all three assumptions. In any case, relying on actual ‘raw’ site data cannot be straightforward.

The need for systematic surveys to integrate research biases Today, the data sets best used for wide areas are those from targeted surveys (Italian research in Southern Etruria and the Sibaritide are cases in point). These need, however, to be evaluated by research on the conservation of palaeo-surfaces and depositional processes. A re-evaluation of data from the South Etruria survey (cf. Patterson, in press), the systematic surveys of the territories of Rome (Bietti Sestieri 1984), of the Gubbio basin (Malone & Stoddart 1994), the Pontine plain (Voorrips et al. 1991) and the Biferno valley (Barker 1995a) is useful when discussing today’s prevailing hypotheses, provided that it can offer solutions to the problems of chronological comparison and site definition in relation to the three assumptions of the Roman school (assumption of settlement, assumption of unity, and assumption of continuity).

The “historical” perspective of Italian archaeology Firstly there is the national tendency to put the “individual” historical subject at the center of attention, even when a middle- or long-term perspective is pursued. Secondly there is the discipline of Etruscology in Italy, which has incorporated the formative phases of the Etruscan civilization (Proto-Villanovan and Villanovan), as well as the other civilizations of pre-Roman Italy (in an implicit core-periphery model) into a teleological explanation. Consequently these civilizations are studied from a historiographic perspective. Furthermore, Etruscology tends to uphold views on urbanization derived from Greek poleogenesis.

Integration of environmental data Italian research often lacks the support of adequate palaeoenvironmental studies and also sufficient indications for the reconstruction of palaeo-economies (scarcity of significant samples from excavations, lack of tradition and interdisciplinary studies, funds). Even projects geared to this integration have only partly achieved their objective (Veneto, Balista et al. 1982, 1988; territory of Rome, Bietti Sestieri (ed) 1984). In Dutch research, however, there has been a solid interest in landscape dynamics (Attema et al. 1998; Voorrips et al. 1991), and a matching of data should prove significant. Geomorphologic, sedimentary and pedogenetic aspects In the Biferno valley surveys (Barker 1995a) there is a remarkably effective treatment of the human impact on the hydrogeological equilibrium (although admittedly it takes a very broad view). These surveys consider fundamental problems generally ignored by the Italian schools (apart from such northern examples as Nisbet & Macphail 1983, Coltorti & Dal Ri 1985, Balista & Leonardi 1985). We need to be able to distinguish between phenomena active in the medium and long terms and other, short-term phenomena, and to define the measure of synchronicity.

The above points have generated many of the ideas relevant to the discussion presented in this paper: – the relevance of the concept of site and the importance accorded to its historical “individuality” (related to the three assumptions of the Roman school); – the resulting teleological approach, including the variation applied to Tarquinia; – the possibility of doing research such as that of Carandini on the origins of Rome, which posits a strong interrelationship between archaeological data and historical or mythographical sources; – the search for proposed interpretations aiming at an overall historical coherence on at least the national level, or, the consciousness of the necessity to confront one’s data with neighboring consolidated situations in order to arrive at satisfactory interpretations.

Distribution of artefacts In order to distinguish separate areas of influence and to point out possible territories dependent on one center, one must study the distribution of related artefact typologies (as a reflection of information transfer) or the distribution of the artefacts themselves (through archaeometric analyses coupled to observations through autopsy), evaluating their circulation. One such study of the Sibaritide is unfortunately still an isolated case (Levi et al. 1999, with a general synthesis for whole Italy). Italian research interest in this area is possibly lagging behind our environmental concern.

The strong historical and historiographic perspective of Italian archaeologists may be contrasted to the widespread tendency among Italian ancient historians not to accept archaeological intrusions - unless these are guided and controlled by history through a substantial dominance by the sources, or at least by historical/artistic data. One of the problems for the discipline is therefore, paradoxically, to demonstrate the intrinsic validity of its data in the historical arena. That is why, if the present author has assumed here that having a historical perspective was a positive aspect of Italian archaeology because it is part of its desire 48

General Papers

al di là dell’Appennino (Atti della Giornata di Studio, Pavia 17.06.1995), Biblioteca di Athenaeum 38: 15-46. Balista, C. & G. Leonardi 1985, Hill slope evolution: pre- and protohistoric occupation in the Veneto, in S. Stoddart & C. Malone (eds), Papers in Italian archaeology IV, i - the human landscape, BAR International Series 243, Oxford: 135-152. Balista, C., A. De Guio, G. Leonardi & A. Ruta Serafini 1982, La frequentazione protostorica nel territorio vicentino: metodologia analitica ed elementi preliminari di lettura interpretativa, Dialoghi di Archeologia 4(2): 113-136. Balista, C., M. Bagolan, F. Cafiero, A. De Guio, S.T. Levi, A. Vanzetti, R. Whitehouse & J. Wilkins 1998, Bronze-Age “fossil landscapes” in the Po Plain, Northern Italy, in B. Hänsel (ed), Mensch und Umwelt in der Bronzezeit Europas (Proceedings of the Meeting, Berlin 17-19.3.1997), Kiel: 493499. Barker, G. 1995a, A Mediterranean Valley - Landscape archaeology and Annales history in the Biferno valley, Leicester UP, London & New York. Barker, G. 1995b, Landscape Archaeology in Italy - Goals for the 1990s, in N. Christie (ed), Settlement and economy in Italy 1500 BC - 1500 AD, Papers of the fifth Conference of Italian archaeology, Oxford: 1-11. Bartoloni, G. 2000, Le origini e la diffusione della cultura villanoviana, in M. Torelli (ed), Gli Etruschi, catalogo della mostra, Venezia: 53-71. Bettelli, M. 1997, Roma - La città prima della città: i tempi di una nascita, Studia archaeologica 86, Roma. Bietti Sestieri, A.M. (ed) 1984, Preistoria e Protostoria nel territorio di Roma, Lavori e Studi di Archeologia 3, Roma. Bietti Sestieri, A.M. 1988, The “Mycenean connection” and its impact on the central Mediterranean societies, Dialoghi di Archeologia 6(1): 23-51. Bietti Sestieri, A.M. 1992, The Iron Age community of Osteria dell’Osa: a study of socio-political development in Central Thyrrenian Italy, Cambridge. Bietti Sestieri, A.M. 1997, Italy in Europe in the Early Iron Age, Proceedings of the Prehistoric Society 63: 371-402. Bietti Sestieri, A.M. 2000, The role of archaeological and historical data in the reconstruction of Italian protohistory, in Ancient Italy in its Mediterranean setting - Studies in honour of E. Macnamara, Accordia Specialist Studies on the Mediterranean, London: 13-31. Bintliff, J. 1997, Regional survey, demography, and the rise of complex societies in the ancient Aegean: core-periphery, neoMalthusian, and other interpretive models, Journal of Field Archaeology 24(1): 1-38. Capuis, L., A. De Guio & G. Leonardi 1984, Il popolamento in epoca protostorica, in Misurare la terra: centuriazione e coloni nel mondo romano - Il caso veneto, Modena: 38-52. Carandini, A. 1997, La nascita di Roma, Torino. Cardosa, M. 1995, Rapporto tra abitati e necopoli tra Bronzo finale ed età del ferro, in N. Negroni Catacchio (ed), Preistoria e Protostoria in Etruria, Atti del II incontro di studi (Farnese, 21-23.05.1993), Milano: 189-197. Childe, V.G. 1950, The urban revolution, The Town Planning Review 21: 3-17. Colonna, G. 1994, intervento in discussione al XXXIII Convegno di Studi sulla Magna Grecia (Taranto 8-13.10.1993): 170-172.

for knowledge, rigor, and embracing coherence, he still maintains that if Italian archaeology wants to be productive it needs to be manifest, without pre-conceived positions, and integrated as discussed in this paper. Moreover, a simpler “statistical” perspective can often sufficiently elucidate the situation, and allow a cautious and preliminary integration with the historical perspective. On the other hand, allowing the historical analysis of archaeological data to prevail may result in radical investigative limitations, as a purely historiographic perspective assigns the greatest role in proof and guidance to the written sources. Against the principle of authority These urgent research needs have now been outlined. In conclusion I want to emphasize that the basic requirement for the validity of investigations is that they can be verified, a requirement not always fulfilled in Italy, and one which raises three fundamental issues: • the need for explicit declaration of theoretical position and basic concepts; • the need for systematic publication of data (cartography, stratigraphy, diagnostic artefacts etc.) • the need for explicit application of analyses, which can be replicated and verified.

Acknowledgments During the preparation of the original paper help was provided by D. Gatti, A. Schiappelli, F. Veronese and L. Zaghetto, and preliminary versions of this article were read by M.A. Bianchi Fossati, M.A. Castagna, A. De Guio, F. Ferranti. I am indebted to R. Peroni for brief, stimulating discussions regarding the definitive version (setting apart our usual exchange); suggestions for the illustrations were provided by F. di Gennaro and A. Schiappelli; and P.A.J. Attema has kindly, but constantly, pressed me to finish the paper, and has condensed and translated it into English. Without all of them the article wouldn’t ever have been written, but the author’s opinions, especially those which can perhaps give rise to criticism, are not necessarily shared by these colleagues. Part of the discussion of the Veneto case study derives from a seminar organized by the author with the participants of the excavations at Broglio di Trebisacce, with the title “The remains of the day: the Padua school”.

References Arnoldus Huyzendveld, A., F. Bistolfi, A. Guidi & A. Zifferero 1990, Cures Sabini: risultati della settima campagna di scavo, Archeologia laziale X(2), Quaderni di Archeologia etruscoitalica 19, Roma: 293-301. Attema, P.A.J. , G.-J. Burgers, M. Kleibrink & D.G.Yntema 1998, Case studies in indigenous development in early Italian centralization and urbanization: a Dutch perspective, European Journal of Archaeology 1(3): 326-381. Bagolan, M. & G. Leonardi 2000, Il Bronzo finale nel Veneto, in M. Harari & M. Pearce (eds), Il protovillanoviano al di qua e

49

RPC 2000 conference proceedings

Maaskant-Kleibrink, M. 2000, Enotri, Greci e i primi culti nell’”Athenaion” a Francavilla Marittima, Magna Graecia XXXV(1-2): 18-30. Malone, C. & S. Stoddart (eds) 1994, Territory, Time and State – The archaeological development of the Gubbio basin, Cambridge. Mandolesi, A. 1999, La “prima Tarquinia” - L’insediamento protostorico sulla Civita e nel territorio circostante, Grandi contesti e problemi della protostoria italiana 2, Firenze. Marino, D. 1998, Aspetti dell’insediamento nella Calabria centroorientale tra età del bronzo recente e prima età del ferro, in N. Negroni Catacchio (ed), Preistoria e Protostoria in Etruria, Atti del III incontro di studi (Manciano-Farnese, 1214.05.1995), Firenze: 287-300. Negroni Catacchio, N. & R. Peroni 1979, Le ultime pagine di F. Rittatore sul “Protovillanoviano”, in Il Bronzo finale in Italia, Atti XXI Riunione Scientifica IIPP (Firenze 21-23.10.1977), Firenze: 27-44. Nisbet, R. & R.I. Macphail 1983, Organizzazione del territorio e terrazzamenti preistorici nell’Italia settentrionale, Quaderni della Soprintendenza archeologica del Piemonte 2: 43-57 e tavv. XV-XVII. Pacciarelli, M.1989-90, Ricerche nel promontorio del Poro e considerazioni sugli insediamenti del primo Ferro in Calabria meridionale, Rivista Storica Calabrese X-XI(1-4): 9-39. Pacciarelli, M. 1991a, Ricerche topografiche a Vulci: dati e problemi relativi all’origine delle città medio-tirreniche, Studi Etruschi 56: 11-48. Pacciarelli, M. 1991b, Territorio, insediamento, comunità in Etruria meridionale agli esordi del processo di urbanizzazione, Scienze dell’Antichità 5: 163-208. Patterson, H. (ed) in press, Approaches to regional archaeology in the middle Tiber valley, Acts of the Workshop (British School at Rome, Roma 1998). Peroni, R. 1988, Comunità e insediamento in Italia fra età del bronzo e prima età del ferro, in A. Momigliano & A. Schiavone (eds), Storia di Roma, I: Roma in Italia, Torino: 7-37. Peroni, R. 1989, Protostoria dell’Italia continentale, Popoli e Civiltà dell’Italia antica 9, Roma. Peroni, R. 1994a, Le comunità enotrie della Sibaritide ed i loro rapporti con i navigatori egei, in R. Peroni & F. Trucco (eds), Enotri e Micenei nella Sibaritide, vol.II, Taranto: 831-879. Peroni, R. 1996, L’Italia alle soglie della storia, Roma-Bari. Peroni, R. 2000, Formazione e sviluppo dei centri protourbani medio-tirrenici, in A. Carandini & R. Cappelli (eds), Roma. Romolo, Remo e la fondazione della città, catalogo dell mostra, Roma. Peroni, R. & F. di Gennaro 1986, Aspetti regionali dello sviluppo dell’insediamento protostorico nell’Italia centro-meridionale alla luce dei dati archeologici e ambientali, Dialoghi di Archeologia III s., 4/2: 193-200. Potter, T.W. 1979, The changing landscape of South Etruria, London. Rendeli, M.1991, Sulla nascita delle comunità urbane in Etruria meridionale, Annali di Archeologia e Storia antica XIII: 9-45. Renfrew, C. 1975, Trade as action at a distance: questions of integration and communication, in J.A. Sabloff & C.C. Lamberg-Karlovsky (eds), Ancient civilization and trade, Albuquerque: 3-59.

Coltorti, M. & L. Dal Ri 1985, The human impact on the landscape: some examples from the Adige valley, in S. Stoddart & C. Malone (eds), Papers in Italian archaeology IV, i - the human landscape, BAR International Series 243, Oxford: 105-134. Cristofani, M. 1986, Economia e società, in G. Pugliese Carratelli (ed), Rasenna. Storia e civiltà degli Etruschi, Milano: 77-156. De Guio, A. 2000, Ex occidente lux: linee di un percorso critico di rivisitazione del bronzo finale nel Veneto, in M. Harari & M. Pearce (ed), Il protovillanoviano al di qua e al di là dell’Appennino (Atti della Giornata di Studio, Pavia 17.06.1995), Biblioteca di Athenaeum 38: 259-357. De Guio, A., S.P. Evans & A. Ruta Serafini 1986, Marginalità territoriale ed evoluzione del “paesaggio del potere”: un caso di studio nel Veneto, Quaderni di Archeologia del Veneto 2: 160-172. De Marinis, R. 1994, La prima età del Ferro nell’Italia settentrionale, Bullettino di Paletnologia Italiana 85: 405-429. Di Gennaro, F. 1982, Organizzazione del territorio nell’Etruria meridionale protostorica: applicazione di un modello grafico, Dialoghi di Archeologia 2: 102-112. Di Gennaro, F. 2000, “Paesaggi di potere”: l’Etruria meridionale in età protostorica, in G. Camassa, A. De Guio & F. Veronese (eds), Paesaggi di potere: problemi e prospettive (Atti del seminario, Udine 16-17.05.1996), Quaderni di Eutopia 2, Roma: 95-119. Di Gennaro, F. & A. Guidi 2000, Il Bronzo finale dell’Italia centrale. Considerazioni e prospettive di indagine, in M. Harari & M. Pearce (eds), Il protovillanoviano al di qua e al di là dell’Appennino (Atti della Giornata di Studio, Pavia 17.06.1995), Biblioteca di Athenaeum 38: 99-131. Di Gennaro, F., A. Amoroso & A. Schiappelli in press, Un confronto tra gli organismi protostatali delle due sponde del Tevere. Le prime fasi di Veio e di Crustumerium, in H. Patterson (ed) Approaches to regional archaeology in the middle Tiber valley, Atti del Workshop (British School at Rome, Roma 1998). Guaitoli, M. 1981, Notizie preliminari su recenti ricognizioni svolte in seminari dell’Istituto, Quaderni dell’Istituto di Topografia Antica dell’Università di Roma IX: 79-87. Guidi, A. 1985, An application of the Rank-Size rule to protohistoric settlements in the middle Thyrrenian area, in S. Stoddart & C. Malone (eds), Papers in Italian archaeology IV, iii – patterns in Protohistory, BAR International Series 245, Oxford: 217-242. Guidi, A. 1989, Alcune osservazioni sull’origine delle città etrusche, in Atti del II Congresso Internazionale Etrusco (Firenze, 26.05.-2.06.1985) I, Roma: 285-292. Guidi, A. 2000, Il Lazio e la Sabina tra la tarda età del bronzo e l’età del ferro, in G. Camassa, A. De Guio & F. Veronese (eds), Paesaggi di potere: problemi e prospettive (Atti del seminario, Udine 16-17.05.1996), Quaderni di Eutopia 2, Roma: 85-94. Leonardi, G. 1992, Assunzione e analisi dei dati territoriali in funzione della valutazione della diacronia e delle modalità del popolamento, in M. Bernardi (ed) Archeologia del paesaggio (IV ciclo di lezioni sulla ricerca applicata in Archeologia, Certosa di Pontignano, 14-26.01.1991), Firenze: 25-66. Levi, S.T. et al. 1999, Produzione e circolazione della ceramica nella Sibaritide protostorica I. impasto e dolii, Prima di Sibari 1, Firenze.

50

General Papers

Torelli, M. 2000, Presentazione, in M. Torelli (ed), Gli Etruschi, catalogo della mostra, Venezia. Vanzetti, A. 2000, Costruzione e problemi dei “paesaggi di potere” nella Sibaritide (Calabria) dall’età del bronzo alla prima età del ferro, in G. Camassa, A. De Guio & F. Veronese (eds), Paesaggi di potere: problemi e prospettive (Atti del seminario, Udine 1617.05.1996), Quaderni di Eutopia 2, Roma: 153-187. Voorrips, A., S.H. Loving & H. Kamermans (eds) 1991, The Agro Pontino Survey Project, Studies in Prae- en Protohistorie 6, Amsterdam.

Ward-Perkins, J. et al. 1961, Veii. The historical topography of the ancient city, Papers of the British School at Rome 29: 1-121. Ward Perkins, J. et al. 1968, The Ager Veientanus, North and East of Rome, Papers of the British School at Rome 36. Yntema, D.G. 1991, The birth of a town: the settlement of Oria in Southern Italy from the Iron age into the Hellenistic period, Mededelingen van het Nederlands Instituut te Rome 50: 102-114.

51

RPC 2000 conference proceedings

2.6

régimes distincts, je fais le choix d’étudier les interrelations, les chevauchements, les entrelacs” (Amselle 1999:25). In 1989 Ruth Whitehouse and John Wilkins published an article on “Greeks and natives in southeast Italy: approaches to the archaeological evidence”, in a volume on Center and Periphery (1989). They criticized pro-Greek prejudice and the concept of Hellenization as “something that other societies will acquire through simple exposurelike measles (but nicer!)” (Whitehouse & Wilkins 1989:103). Recent contributions in the Accordia Research Papers have taken the same line, proposing an interesting systematic revision and the application of the “peer polity interaction” model to Greek-indigenous relations in Southern Italy, but often the validity and completeness of the archaeological record used has not been verified (Herring 1991). Current research and systematic analysis appear to have changed considerably, in comparison with Whitehouse and Wilkins’ research of ten years ago. In particular, the recent development in studies on settlement systems in the indigenous world of Southern Italy is leading to a completely new way of reconstructing the dynamics of transformation in these societies1. More sophisticated excavation techniques together with a clear definition of synchronic and diachronic work, have helped us to recognize cultural processes of transformation; “…material culture is not a repetitive static system …synchronic reconstruction (the analysis of individual archaeological contexts) is the necessary base for historical (diachronic) narrative” (Bietti Sestieri 1997:373). By combining single data in an accurate description of spatial relations, we can trace the endogenous dynamics of internal evolution within different cultures. Yet we must at the same time beware of applying excessively theoretical schemes to individually evolving dynamics: “phases of stability, devolution and collapse mark the trajectories of groups characterized by different degrees of socio-political complexity” (Bietti Sestieri 1997:371). In Southern Italy, native settlements reflect very advanced experiences of urbanization marked by strong destructive and involutionary processes (D’Andria 1991). But these were followed by long periods during which such sites lose archaeological visibility, in 5th century BC Messapia for example, as can be seen in Cavallino and perhaps Oria too. Computer management and the use of GIS systems make a systematic treatment of the settlement dynamics possible. The University of Lecce is working on a census of native settlements in Southern Italy, building up a data bank, which up to now includes 255 complex settlements (ranging from hut villages of only a few hectares to fortified centers covering more than 150 hectares, most of them in the Puglia region) (D’Andria 1997, 1999). If we look at a map of all the native settlements of Southern Italy, the numbers make it evident that there is yet enormous potential for research, with respect to the Greek colonies too (figure 1). The histogram in figure 2 also indicates that there has been an impressive amount of research in, for example, the south of Apulia and the north of Basilicata. In the latter are found Lavello and Serra di Vaglio, both of

Greek Colonization and Romanization From a Native Perspective Francesco D’Andria

Greek colonization from the 8th century BC and, many centuries after, the Roman conquest, represent two crucial phenomena of transformation of the Italian peninsula, a region characterized by a complexity of peoples, living in territories which are neither geographically nor environmentally homogeneous. Our knowledge of the indigenous cultures, the Etruscans excepted, has been inhibited by a particular tradition of studies based on the concepts of Hellenization and Romanization. Compared to the magnificent monuments of the Greek colonies, the archaeology of the native populations, particularly in Southern Italy and Sicily, has been considered the expression of subordinate cultures: “ellos son como niños”, to quote a study by S. Gruzinski and A. Rouveret (1976). It reflects a eurocentric scheme of interpretation, in which a classic Greek model is mixed with a latent colonialism, “… on attribue à une culture grecque (idéalisée pour la cause), le monopole de la Raison, de l’Ordre et de la Logique tandis que les mentalités indigènes sont rejetées dans l’irrationnel et le fantastique…” (Rouveret & Gruzinski 1976:4-5). The need for a deeper understanding of the indigenous Italic world was voiced at the 1971 Taranto Conference on Le genti non-greche della Magna Grecia (Atti Taranto 1971) and was taken up again in 1981 in the Cortona Conference on Forme di contatto e processi di trasformazione nelle società antiche (Atti Cortona 1981). Here, criticism of the concept of acculturation cleared the ground for a reexamination of the term Hellenization. It was considered to be too much centered on the Greek model. Thus, among leading scholars a new perspective gradually emerged on the essential role of interaction among coexisting societies as a factor in cultural change and evolution (cf. Flannery 1972, 1983). By the time of the 1997 Taranto Conference on Confini e frontiera nella grecità d’Occidente (Atti Taranto 1997) systematic revision had occurred, but a predominantly Hellenic-centered vision remained. During this conference the problem of defining Greek and indigenous societies within fixed parameters was confronted. Discussion focused on world-system and center-periphery models and on the concept of ethnicity, showing how ethnic identity is socially constructed rather than intrinsically determined (Lomas 1993; Bintliff 1997). These discussions brought out new points of view, from which ethnic groups could no longer be viewed as self-contained but, following the logic of cross-breeding, as hybrid systems open to interaction with others. This is the perspective of C. Meillassoux’ ensemble symplectique (Meillassoux 1986), and of the J.L. Amselle’s Logiques métisses, where “Plutôt que de concevoir les ethnies comme des univers clos situés les uns à côté des autres, les systèmes politiques comme des entités nettement séparées, les conceptions religieuses comme des mondes bien délimités, les types d’économie comme des 52

General Papers

which have been subjected to extensive research. In Messapia a peak in the number of settlements only partially investigated is due to an increase in number and scope of surveys of the territory (figures 3-4); in the Brindisi region, a group of Dutch researchers, directed by D. Yntema and G.J. Burgers, has played a major role in this field2. Analyzing our sites file, we have gathered data on the size of the settlements and with this have drawn up maps, indicating their distribution. Walled centers, exceeding 100 hectares, turn out to be concentrated in Puglia, while those in Basilicata, Calabria and Campania, south of the river Sele, are generally smaller than 50 hectares. This phenomenon can be understood in the light of geographical and geomorphologic characteristics. An example is Montescaglioso, a central place, surrounded by smaller centers and located on the border of the Metapontum chora (figure 6). From the three-dimensional diagram, we can trace the course of the medieval walls and the wider circuit of fortifications pertaining to the Enotrian settlement, some wellpreserved parts of which have recently come to light (D’Andria 1999:114 ff., figs. 18-19). As for the Sinni valley, which was an important axis of communication between the Ionian and Tyrrhenian Seas, we have tried to sift through the accumulation of topographical information, found in the Forma Italiae, and to connect what we find with other characteristics of the territory in the GIS system, such as altitude, lithology, and soil fertility. The Forma Italiae is a remarkable achievement in Italian archaeology, but the abundant information it provides makes it very difficult to get a clear idea of the synchrony of the settlements3. With the census of the settlements and applications of GIS system we can begin a study of the pre-Roman Sinni valley settlement hierarchy, in which the site of Chiaromonte assumes a central role (figure 5; Barra Bagnasco 1999). Unfortunately nothing is known about the spatial organization of the center, because only the necropolis and the sanctuary have been systematically investigated. The position of the settlements around the territory of Chiaromonte reveals the coherence of a system in which all strategic and economic potential was used. The wealth of Greek vases discovered in the tombs of the Sinni valley has previously been interpreted as ‘Greek penetration’ into the inland area (mesogaia); now it is possible to link this evidence to the dynamics of settlement evolution and to spatial and social interrelations. We can now draw up a picture of a complex native society, of the organization of the land and of the hierarchy of the settlements. This also holds good as far as Lucanian and Apulian grave goods are concerned. Precious though these grave goods may be, previously any interpretation of them in social terms was hampered by the absence of context. Systematic investigations in the Salento region offer new elements for the reconstruction of a settlement system developing in the early part of the Iron Age. This system was characterized by villages, which became more numerous in the second half of the 8th century BC due to the dynamics I have attributed to demographic growth. The role of Otranto as a trading post, controlling the maritime routes, which linked the Aegean to Central Europe by way

Figure 1: Distribution map of Greek (dark dots) and native (bright dots) settlements in Southern Italy (4th - 3rd centuries BC).

of the Adriatic, was of fundamental importance in this (D’Andria 1995). The large number of Corinthian imports into this area shows what Greek interests were. The imports can be compared to those found in places like the Greek colony of Pithecussa, in the Tyrrhenian (figure 6). In Salento, the 6th century BC settlement system is characterized by three dominant centers; Oria, Cavallino and Ugento. This situation is very different from that of the 4th and 3rd centuries BC, when there was an increase in dominant settlements larger than 100 hectares, defining a cantonal type of organization of the territory. On figure 4, the larger stars show the ratio between the estimated size of the centers and the effective level of exploration in the area; in Nardò and Ugento, where the areas surrounded with fortifications are larger than 100 hectares, no systematic investigations have been conducted up to now. In the Salento region settlement structures can be defined as becoming increasingly close to endogenous development dynamics, reflecting a society very different from that of the Greek colonies. The dominance of singlefamily groups and the appearance of chiefs, reveal a socie-

Figure 2: Histogram showing the degree of research on native settlements in Southern Italy. Ampio = thorough, Superficio = superficial, Parziale = partial.

53

RPC 2000 conference proceedings

ty that was strongly hierarchical. Recent research shows the different ways in which these elites used the residential areas. Virtual computer reconstructions allow us to visualize better the spatial organization of different settlement areas. Elite residential areas are generally found at the center of the settlement, as in the case of Vaste (figure 9; cf. D’Andria 1990:57, carta 1), and as is now also attested at Muro Tenente (van Alberda et al. 1999:53 ff.). Similarly, in Basilicata, inside the Enotrian center of Difesa San Biagio, near Montescaglioso, a palatial structure has been excavated in the upper part of the hill (figure 10; D’Andria & Roubis 1999). Recent researches and excavations have highlighted the use of the orthogonal building organization even in some native settlements, in the period between the 4th and 3rd centuries BC. I am referring in particular to the Lucanian center of Laos, the Enotrian center of Pomarico (figure 7), the fortified building excavated near Acquarica, in Salento (figure 12; between Lecce and Rocavecchia; D’Andria 1999:115 ff.), and the quarter discovered by the Dutch team at Muro Tenente, in the province of Brindisi (van Alberda et al. 1999:69-72). Once again, we should beware of considering these a widespread phenomenon of Hellenization in the indigenous world. We deal with a period in which the techniques of regular division of land had been acquired by the native world as well. It is better to regard these examples as part of the process of redefinition of the territory, in a period (4th and 3rd centuries BC) in which demographic growth made the agricultural use of new land a necessity. It is in this context that we should consider also the settlements of Pomarico, encapsulated within a system dominated by the central place of Montescaglioso, and Acquarica near Lecce, in Messapia, where a building with an orthogonal plan has been excavated which can be interpreted as a fortified residential house for elite groups who controlled the agricultural activities in the surrounding territory. Studies of the importation of Greek objects, in particular pottery, in native societies, have traditionally interpreted these phenomena as proof of Greek commercial expansion and as a result of the “profound Hellenization” of these societies. An example of this is the interpretation of the famous site of L’Incoronata, near Metaponto, on the right bank of the river Basento. Here, on top of native settlement buildings, excavations have brought to light square Greek oikoi and deposits of precious objects dating to the later period of life, in the mid 7th century BC; large locally produced, decorated vases like the deinoi, with figures of horses, or the relief vases like the perirrhanterion, with mythic fighting (Orlandini 1986). Referring once again to the inevitable Greek venture in the mesogaia, an Eastern Greek presence has been spoken of in the context of Incoronata. In this vein, the settlement was considered an outpost for exchanges with the indigenous world, with characteristics of a ‘craftsman - trading center’. However,

Figure 3: Southern Apulia. Distribution map of Messapian settlements of the 4th and 3rd centuries BC, with estimated extensions in hectares.

Figure 4: Southern Apulia. Distribution map of Messapian settlements of the 4th and 3rd centuries BC, with estimated extensions in hectares, in relation to the degree of research. Ampio = thorough, Sup = superficial, Parz = partial.

1

2

Modern settlement archaeology in Southern Italy has an important starting point in the Roccagloriosa excavations (Gualtieri 1987).

54

Very important in this respect is the publication of the surveys carried out by Dutch teams in the Brindisi area (Yntema 1993; Burgers 1998).

General Papers

excavations in the indigenous hinterland have not revealed any Greek pottery produced at Incoronata, as they have in the other Greek colonies of Crotone and Taranto. Clearly this ceramic production is to be seen as belonging to the Greek elite world in the coastal settlement of Siris, and should be related to a particular phase of tension in political institutions, caused by the presence of the Achaeans in the area and the foundation of a new colony, Metapontum, following radically different patterns of territorial occupation (Atti Policoro 1986). Findings of precious Attic, Corinthian and Eastern Greek vases in the native necropoleis show these were constantly present from the 6th century BC; in this case too, interpretation tends towards Greek ‘commercial flow’. The impressive, decorated Attic vases in the Apulian settlements along the Adriatic coast have been interpreted from such a perspective. But this does not take into account the choice of objects, made to meet the demand of native elites, who incorporated them into funerary rites that varied from place to place, even in settlements very close to each other. In some cases the production of Attic workshops can even be shown to have been conditioned by native demand. For example, Athenian artisans modeled red figured vases with myths in a typical Greek style, but with shapes drawn from the native ceramic repertoire (D’Andria 1988). The depiction of images and myths on these objects is a confirmation of the possessor’s status and the sign of his autonomous choice within the symbolic systems drawn up by dominant barbarian groups, as has been demonstrated by Catherine Morgan for Etruria (Arafat & Morgan 1994) and Gert-Jan Burgers for Messapia (Burgers 1998:183 ff.). The native use of wine, together with objects connected to it, has also often been described as the acquisition by natives of the Greek symposium. By studying the contexts in which these objects were found, for example in archaic Messapian residential areas like at S. Vito dei Normanni or Cavallino, Grazia Semeraro has been able to identify large numbers of Greek transport-amphora for wine. Adopting ethnological explanatory models, like that of the “workfeast”, which has already been used by Dietler for Celtic society inland of Marseilles, a similar interpretation in commensal politics has been proposed also for the Messapians of the 6th century BC (Semeraro 1997) The native perspective also played an important role in the course of the Roman conquest of the south, when native elites established alliances with the ruling Roman classes. The complexity of these alliances provoked a complete overturning of the land systems, particularly at the end of the 3rd century BC, after the Hannibalic war. The radical transformation of the landscapes and the psychological impact this had on the population living on this land is of great importance. The huge Messapian fortifications were mainly taken away, leaving a landscape of ruins, crossed by paths belonging to the new farming system.

Figure 5: Valle del Sinni. Native settlements, analysis of distances and soil fertility.

Figure 6: Southern Italy and Sicily. Distribution map of Late Corinthian Geometric wares.

3

In the Sinni valley a systematic survey has been carried out by L. Quilici and S. Quilici-Gigli (1997).

55

RPC 2000 conference proceedings

Figure 7: Pomarico Vecchio. Plan of the settlement, with remains of the fortification walls and regularly structured quarter (4th/3rd centuries BC).

In this sector too, the latest contributions tend to go beyond a fixed concept of Romanization as the influx of a dominant culture. The systems of the indigenes are considered to have been transformed through a process of exchange between Rome and their communities (cf. Millet 1990, David 1994, Lomas 1993). The concept of ethnicity investigates the ways in which every town attempted to build up its own identity within the colony or municipium structure, with results differing from town to town. Nicola Terrenato dealt with this subject in his work on ‘The Romanisation of Italy: global acculturation or cultural bricolage’, where he pointed out the influence that nationalist historiographers in Italy and Germany have had on the concept of the Romanization of Italy (Terrenato 1997). Recent studies of individual contexts, on the contrary, show “the strong diversity of outcomes of Romanization and also a greater role for ethnic identities than previously assumed”. Terrenato’s proposal was to analyze the phenomenon through the identification of contrasting subsets of society, for example of elite-commoners, urban-rural, public-private etc. Within local communities these tend to be affected by Romanization in different ways. The great Greek cities, which had allied themselves with Hannibal, forfeited their influence to native centers, and a city like Taranto was replaced by Brindisi on the foundation of the Latin colony there in 244 BC. In the south of Apulia the new order took hold with the circulation of bronzes from the Brindisi mint that gradually replaced the Tarantine coins.

Yntema has shown that the process of incorporating the Brindisi region into the Roman state, during the 3rd century BC, corresponds to the integration of the area in a Mediterranean system of interregional contacts, and was a continuation of what had happened previously; “Changes in site hierarchy and mentality changes in the field of craft, trade, commerce and agriculture as well as social changes owe much to the dialogue between the native culture of the Brindisi area and the emerging Mediterranean koinè culture, with its numerous local variants (among them also a central-italic Roman variant) of the Middle Hellenistic period (Yntema 1995)”. A new discovery in the Salento area, at Vaste (Lecce) last year, provides us with fresh elements for the understanding of the new role of Rome. A sanctuary was uncovered in the center of the Messapian settlement, directly beneath the modern square (figure 8a). The pottery discovered here indicates that cult activities took place at the spot during the 3rd century. These were followed by its abandonment during the 2nd century BC. It has been recognized as belonging to a chthonic cult, with the chasma ghes, the bothroi and the hypogeum megaron room, containing an altar with holes for the libation, very similar to the Greek ones, for a goddess whose Messapian name was OXXO (figure 8b). There are many elements here recalling the world of Latium: the sacrifice of dogs similar to the rituals during Robigalia feasts and the use of pans for ritual cooking which are different in form from the local traditional ones; 56

General Papers

Figure 8: Vaste (Lecce), piazza Dante. a: Chthonic cult place, with demarcation walls, bothroi and megaron (second half of 3rd century BC). Among the finds b: Cup with painted inscription mentioning ‘oxxo’ (3rd century BC); c: Limestone head of a female divinity, with traces of paint (second half of 3rd century BC)

the former are a type from Latium and archaeometric tests will show if they were imports from Central Italy. Of the limestone cult statue only the head has been found, and once again it is very different from Tarantine sculpture (figure 8c). In contrast, we can see a connection with the limestone capitals belonging to a Roman republican temple in the Latin colony of Brindisi. During the 2nd century BC, Metapontum too was reduced to being a small coastal settlement (the so-called castrum), due to a radical change in settlement hierarchy. The municipium, which would correspond to the center of Montescaglioso, took over the central role. Various findings have shown that it was a center with public buildings,

which were standing here after the social war: there is a mosaic inscription mentioning duumviri, telamons which might have been set into the wall of a theatre or other building. In conclusion, I would like to refer to a particular situation, that of Sicily, where pro-Greek prejudice has long played a negative role not only in studies of the indigenous cultures, but also in Romanization studies. The Sican, Sicule and Elymnian cultural contexts are generally considered passive under the impact of the Greek colonial venture, and we are only at the start of promising research into the ethnicity processes within these cultures. Classical archaeologists perceive the Roman conquest as the end of the Greek colonies 57

RPC 2000 conference proceedings

and of the objects of their acculturation (the natives). However, the flexibility of the Sicilian system in the Roman world tends to be overlooked. For example, the great monumental complexes like Solunto, Segesta, Monte Iato, and Morgantina, in particular in the west of Sicily, have been dated to periods during which the Greek presence was evident, as in the time of Timoleon or Agathocles, between the 4th and 3rd centuries BC. However, recent researches show that these cities were strongly developed in the 2nd and 1st centuries BC, within the new system promoted by the constitution of the roman Provincia of Sicily (Wilson 1990:67 ff. for roman republican phases). In later studies, a dating of the 2nd half of the 2nd century BC has been proposed on the basis of chronological elements from stratigraphic excavations, for the theatre of Segesta, and also for similar buildings in Solunto, Monte Iato and Tindari4. The game of identities speeds up here: to strengthen direct ties with Rome, Segesta linked itself successfully to the Trojan tradition as this was promoted by the native town elite, belonging to the ethnos of the Elymnians. And it is on account of this, the insertion of Sicily into the Roman system that we can date the ’Hellenization’ of Segesta, an indigenous center, which after endogenous dynamics of development adopted as model the Hellenistic city, in its characteristic form - its monuments, language and magistratures.

Bintliff, J. 1997, Regional Survey, Demography, and the Rise of Complex Societies in the Ancient Aegean: Core-Periphery, Neo Malthusian, and other Interpretative Models, Journal of Field Archaeology 24: 1-36. Burgers, G.-J.L.M. 1998, Constructing Messapian Landscapes, Settlement Dynamics, Social Organisation and Culture Contact in the Margins of Graeco-Roman Italy, Amsterdam: Gieben. D’Andria, F. 1988, Messapi e Peuceti, in: Italia omnium terrarum alumna, Milano 1988: 686ff. D’Andria, F. (ed) 1990, Archeologia dei Messapi, Bari. D’Andria, F. 1991, Insediamenti e territorio: l’età storica, in: I Messapi. Atti del XXX Convegno di Studi sulla Magna Grecia, Taranto 1990, Napoli: 393-478. D’Andria, F. 1995, Corinto e l’Occidente: la costa adriatica, in Atti del Convegno di Studi sulla Magna Grecia, Taranto 1994, Taranto: 457-508. D’Andria, F. (ed) 1997, Metodologie di catalogazione dei Beni Archeologici, Beni Archeologici-Conoscenza Technologie, 1.1., Lecce-Bari. D’Andria, F. 1997, Ricerche archeologiche sul teatro di Segesta, in: II Giornate Internazionali di Studi sull’Area Elima, Gibellina 1994, Pisa: 429-450. D’Andria, F. 1999, Ricerche recenti sugli insediamenti indigeni di Puglia e Basilicata, in: La forma della città e del territorio. Atti Conv. S. Maria Capua Vetere 1998, Roma: 103-118. D’Andria, F. & D. Roubis, 1999, L’insediamento indigeno di Difesa San Biagio a Montescaglioso, II campagna di scavo 1996, Siris 1: 123-155. David, J.M. 1994, La romanisation de l’Italie, Paris. Flannery, K.V. 1972, The cultural evolution of Civilizations, Annual Review of Ecology 3: 399ff. Flannery, K.V. 1983, Archaeology and ethnology in the context of divergent evolution, in: K.V. Flannery & J. Marcus (eds), The Cloud People, New York: 361-362. Gruzinski, S. & A. Rouveret 1976, Histoire et acculturation dans le Mexique colonial et l’Italie méridionale avant la Romanisation, in: Mélanges de l'Ecole Française de RomeAntiquités 9-61. Gualtieri, M. 1987, Fortifications and Settlement organisation: an example from pre-roman Italy, World Archaeology 19: 30-46. Herring, E. 1991, Socio-political change in the south Italian Iron Age and Classical periods: an application of the peer polity interaction model, in: Accordia Research Papers 2: 31-55. Lomas, K. 1993, The city in southeast Italy. Ancient topography and the evolution of urban settlement 600-300 B.C., in: Accordia Research Papers 4: 63-77. Lomas, K. 1993, Rome and the Western Greeks, 350 B.C.-A.D. 200, London. Meillassoux, C. 1986, Anthropologie de l’esclavage, Paris. Millet, M. 1990, The Romanisation of Britain, Cambridge. Orlandini, P. (ed) 1986, I Greci sul Basento. Cat. Mostra, Como. Quilici, L. & S. Quilici-Gigli 1997, Considerazioni metodologiche sulla cartografia dei Beni archeologici: introduzione alle ricerche topografiche nella Valle del Sinni, in Atlante di topografia antica 6: 231-260.

References Alberda, K. van, G.-J. Burgers, H. Burgers, D. Karel & D. Yntema 1999, Muro Tenente, centro messapico nel territorio di Mesagne, Manduria. Amselle, J.L. 1999, Anthropologie de la frontière et de l’identité ethnique et culturelle: un itinéraire intellectuel, in: Atti Taranto 1997: 25ff. Arafat, K. & C. Morgan 1994, Athens, Etruria and the Heunenburg: mutual misconceptions in the study of Greekbarbarian relations, in: I. Morris (ed), Classical Greece: ancient histories and modern archaeologies, Cambridge: 108-134. Atti Cortona 1981 = Forme di contatto e processi di trasformazione nelle società antiche. Atti Conv. Cortona, maggio 1981, PisaRoma 1983. Atti Policoro 1986 = Siris-Polieion. Fonti letterarie e nuova documentazione archeologica, Atti Conv. Policoro 1984, Galatina 1986. Atti Taranto 1971 = Le genti non-greche della Magna Grecia. Atti del XI Convegno di Studi sulla Magna Grecia, Taranto 1971, Napoli 1972. Atti Taranto 1997 = Confini e frontiera nella grecità d’Occidente. Atti del XXXVII Convegno di Studi sulla Magna Grecia, Taranto 1997, Napoli 1999. Barra Bagnasco, M. 1999, Sistemi insediativi nella Basilicata dal Sinni al Bradano, tra il IV e il III sec. a.C., in: Magna Grecia e Sicilia, Atti Conv. Messina 1996, Messina: 39-58. Bietti Sestieri, A.M. 1997, Italy in Europe in the Early Iron Age, Proceedings of the Prehistoric Society 63: 373ff.

4

A preliminary presentation of stratigraphic researches in the Segesta theatre is to be found in D’Andria 1997.

58

General Papers

Semeraro, G. 1997, Eu neusí. Ceramica greca e società nel Salento arcaico, Lecce-Bari. Terrenato, N. 1997, The Romanisation of Italy: global acculturation or cultural bricolage, in: Forcey, C., J. Hawthorne & R. Witcher (eds), Proceedings of the Theoretical Roman Archaeology Conference, Oxford. Whitehouse, R. & J. Wilkins 1989, Greeks and natives in southeast Italy: approaches to the archaeological evidence, in: T. Champion (ed), Centre and Periphery. Comparative studies in archaeology, London: 102-126.

Wilson, R.J.A. 1990, Sicily under the Roman Empire, Warminster. Yntema, D.G. 1993, In Search of an Ancient Countryside, The Amsterdam Free University Field Survey at Oria, Province of Brindisi, South Italy (1981-1983), Amsterdam: Thesis Publishers. Yntema, D.G. 1995, Romanization in the Brindisino, southern Italy: a preliminary report, Bulletin Antieke Beschaving 70: 153-177.

59

RPC 2000 conference proceedings

2.7

Pottery Production and Metallurgy

may be viewed as a contribution to that approach. In it I will focus on the production and circulation within urban territorial organizations of local coarse ware products. A central question deals with the differences, if any, between such pottery used in central places (cities, markets and sanctuaries) and pottery used in rural contexts (e.g. open sites). I will treat specifically the use of domestic pottery in ritual contexts, both from the perspective of the archaeological and the literary sources.

Andrea Zifferero 2.7.1 Introduction The first part of this paper contains some observations on the role of domestic pottery in defining both the territorial and ethnic dimension of complex societies. Focusing particularly on Tyrrhenian Italy, it gives an analysis of the role of domestic pottery within specific food systems developed by local communities. The food system based on the mixed use of meat and cereal products led to the adoption of different vessel shapes, which may usefully be interpreted as possible markers of ethnicity. A very wide chronological span usually helps in defining cultural/ethnic changes in regions or historical territories: the perspective proposed here deals mainly with the protohistoric and preRoman periods (Long & Scarpi 1989). The field of domestic pottery is huge, and is being studied at different levels: apart from the technological and archaeometric approach, which has recently begun to focus on fabrics and uses of coarse wares, the possible connections between nourishment and vessels is now also being investigated, especially by Roman and Medieval archaeologists. In the field of pre-Roman archaeology the latter approach receives unfortunately little attention. A regional perspective on the development of metallurgical technologies is the goal of the second part of the report. The new data on pre-roman mining activities brought to light by recent research in the area of the Colline Metallifere (Tuscany), are reviewed here. These deal with different methods of extracting copper and lead from sulfide ores and iron from oxide ores. The idea is that the minero-genetic and geological setting of ore in different regions stimulated the transfer of metallurgical technologies from more advanced mining and metallurgical areas to less advanced regions. The archaeological evidence will be considered a marker of such changes.

2.7.2.2 Lines of research The analysis of domestic pottery in protohistoric and preRoman Italy is detailed, but follows different thematic lines of research (among the most up-to-date contributions, see Santoro Bianchi & Fabbri 1997, Nijboer 1998). This is due to the different traditions of Italian and foreign schools of thought, whose interests range from production, recently enriched by the archaeometric methods of investigation, to chrono-typological approaches, mainly devoted to the analysis of the formal evolution of shapes. The interest in the study of domestic coarse wares appears especially limited when compared to the study of the more attractive funerary pottery, so apt for the study of social status and funerary customs in past communities. The following points may help in the assessment of current problems: • Recent Italian studies on European protohistory focus on problems of classification of shapes, in connection with their denomination. They look at the organization and development of the Bronze and early Iron Age communities, and consider the impact of processes of colonization in south Italy and the formation of early urban forms in central Tyrrhenian Italy and subsequent urbanization in northern Italy (Cocchi Genick 1999). This trend was already apparent in attempts to lay down common criteria for the nomenclature of pre-Roman pottery, necessitated by a program for the cataloguing of pottery (Parise Badoni 2000) formulated by the Italian Ministry of Cultural Heritage.

2.7.2 Domestic pottery, food systems and ethnicity Each ancient society in protohistoric and pre-Roman Italy developed its own specific food system mirrored by the domestic pottery. The question I will consider here is how we may carry out comparative analyses that look for analogies and differences between systems in time and space, in such a way that they reflect developments in territorial organization. And related to this, whether it is possible to use such data to point out ethnic differences in pre-Roman cultures. The socio-political and economic context here is early state organization in the Tyrrhenian and the emergence of a dependent complex rural landscape.

• The chrono-typological analysis of protohistoric pottery dominates other interpretations of the pots. I do not want to criticize this type of analysis, but would like to draw attention to other possible approaches. An excellent example is the pottery analysis reported in the final publication of the excavation of the Bronze Age site of Sorgenti della Nova, in the Fiora valley (Viterbo, Latium) (Negroni Catacchio 1995). Here the study of aspects of the food system suggested by the shapes offers a new perspective on the analysis of protohistoric pottery.

2.7.2.1 Domestic pottery in urban territorial networks Scholars of pre-Roman archaeology study the high standard pottery especially to shed light on trade routes and cultural contacts and changes. Much work still remains to be done on the coarse wares and the following case study

Coarse ware classes from the pre-Roman period have only relatively recently become the object of more sophisticated chrono-typological studies, and still are far removed from an organic typological assessment of regional production and shapes (among the most far-reaching studies, e.g. 60

General Papers

vessel type occurs frequently. More of a problem is of course identifying the composition of the bread, considering the different level (quantitative and qualitative) of agricultural technology in Etruria and Latium vetus (Rendeli 1993). The palaeobotanical sources indicate the cultivation of more advanced species of cereals in Etruria than in the agriculture of Latium vetus, where spelt persisted. The baking vessel continued in use apparently up to the 5th century BC in southern Etruria, and with a slight typological development, in the 4th - 3rd centuries BC in Latium vetus, when it was probably substituted by the common and public baking ovens. The case of the “bacini di S. Omobono” in fact offers a good application of the proposed model: in this case the connection between social rank and rite is confirmed both by the literary sources and the archaeological data. The literary sources propose a link between the making and the offering of liba to the Mater Matuta, by the Roman matres familiae, on the occasion of the feast of the goddess. The feast would coincide with the Matralia, about the middle of June (Dumèzil 1977). This is a datum confirmed by the presence of this kind of basin in the archaeological deposits of the Fortuna and Mater Matuta sanctuary of S. Omobono in Rome (Colonna 1963-64, Roma 1989). Also, a broken specimen of this vessel shape was contained in the HH 11-12 grave in the cemetery of Quattro Fontanili at Veii, dating to the 8th century BC, the contents of which rank among the richest and most notable of the entire cemetery (Cavallotti Batcharova 1965, Guidi 1993). The case described here clearly demonstrates the link between the shape of a vessel, and a primary activity (the baking of bread) and the way this may be associated with ritual and domestic usage in protohistoric and pre-Roman communities. A similar link may be reconstructed through the analysis of the cylindrical-ovoid jar. This shape has a long tradition in the central Tyrrhenian area during the protohistoric and pre-Roman period. Many references in the sources

Carafa 1995). As stated, the emphasis remains on classes of decorated fine ware. 2.7.2.3

A case study from the central Tyrrhenian area on cooking stands, baking covers and jars A recent discussion about the status of the study of domestic pottery has led to an attempt to reconstruct aspects of the food system of the central Tyrrhenian communities. The aim is to create an interpretation model of the different functions of pottery, respectively as a marker of food habits, a marker of ethnicity and therefore of food acculturation, a marker of ritual activities, and finally as a marker of social rank (Zifferero, forthcoming a) (figure 1). The model may be used in many cases: e.g. as a marker of the prevailing food system (and therefore as an expression of ethnicity and changes occurring in a community’s food system), as in the recently proposed case of the Etruscan and Latin cooking stands (Zifferero 2000). Such a change, restricted to the Etruscan area, occurred in the canonic model established in the middle-Tyrrhenian area from the late Bronze Age to the early Iron Age. In Etruria in the 8th century BC, a Greek shape (attested at Pythekoussai), completely substituted the old model. The more conservative Latin and Sabine area, roughly coinciding with the territories lying on the left bank of the Tiber, stuck to the old shape (Scheffer 1981) (figure 2). The advantage of working with very detailed distribution maps, which allow for the delineation of cultural frontiers is evident here: slight typological differences may act as territorial indicators of political control. A second case bearing on social ranking and the ritual interpretation of pot-forms concerns a vessel shape interpreted as a portable oven to cook bread in. This shape, consisting of a medium sized basin with upright walls and plain bottom, and decorated with four tongue handles, has long been identified as a type developed in the central Tyrrhenian area in the early Iron Age. It became typical of the material culture of Latium Vetus and southern Etruria with only slight differences in the shape. A proposal that the so-called “bacini di S. Omobono” (as the first editor named them) were the forerunners of the classical testa and klibani, has recently been confirmed (Cubberley et al. 1988, Zifferero forthcoming, a). The function of this tool follows its shape: the bread was kneaded in the form of a libum, a sort of thin and circular cake: the libum was made in Latium Vetus with spelt flour (Triticum dicoccum), not containing yeast, a typical product of the food system of the regal period and early Republican Rome (André 1961, Fayer 1982). The cooking process began with the exposure of the testum to the fire: after heating it, the libum was put on the fireplace with the basin on it: the cooking was done by refraction of heat. A very similar way of cooking bread has been observed in ethno-archaeological studies in the Ligurian area (Mannoni 1965). Interestingly, slight contemporaneous differences in the shape of the tongue handles can be observed in Etrurian pots during the period between the second half of the 7th and the 5th centuries BC when this

Figure 1: An interpretative model for pre-Roman domestic pottery.

61

RPC 2000 conference proceedings

Marta L.di Bracciano

Roma

Figure 2: Distribution map of cooking stands in the middle Tyrrhenian area, from the Late Iron Age to the Hellenistic period (second half of the 8th-4th centuries BC), according to the typology of Scheffer 1981. Source: Zifferero 2000. Symbols: 1 = modern centre; 2 = ancient centre; cooking stand 1 = type IIA; cooking stand 2 = type IIB; cooking stand 3 = type IIC; cooking stand 4 = type ID. Topography: 1 = Roselle; 2 = Doganella; 3 = Vulci; 4 = Volsinii; 5 = Tarquinia; 6 = Acquarossa; 7 = Allumiere, La Farnesiana, 8 = settlements of the district of Allumiere and Tolfa; 9 = S. Marinella, Castellina del Marangone; 10 = Tolfa, Pian Conserva; 11 = Canale Monterano, Piana di Stigliano; 12 = Canale Monterano, Frassineta-Franco; 13 = S.Giovenale; 12 = Pyrgi and its district; 15 = Cerveteri; 16 = settlements of the district of Caere; 17 = Nepi, Fosso del Pavone; 18 = Falerii veteres; 19 = Veio; 20 = Veio, Casale Pian Roseto; 21 = Roma, Malagrotta; 22 = Roma, Acquafredda; 23 = Ficana; 24 = Castel di Decima; 25 = Roma, Acqua Acetosa-Laurentina; 26 = Roma; 27 = Antemnae; 28 = Fidene; 29 = Crustumerium; 30 = Cures Sabini; 31 = Lavinium; 32 = Ardea; 33 = Ariccia and settlements of the Alban district; 34 = Gabii; 35 = settlements of the district of Tivoli; 36 = Satricum; 37 = Velletri; 38 = Palestrina; 39 = Anagni.

demonstrate that it was a useful vessel in the food system of regal and early Republican Rome. Pultem (a thin soup of cereals), a fundamental part of the archaic Roman cuisine (Andrè 1961, Fayer 1982) was cooked in it. The internal organs of animals from sacrifices devoted to agrarian deities (Zifferero forthcoming, a) as well as the usual pultes were boiled in the jar. The example is useful as it shows how our data on ancient pottery and its uses (i.e. for ceremonial cooking) can be related to archaeozoological as well

as to palaeobotanical finds (figure 4). These data are useful for defining aspects and problems of cultural change. There are many ways indeed, to answer the question why a ceramic type changed. The influence exerted by technological progress, or by acculturation from colonial contact, or more general acculturation, may all lead to changes influencing the development of shapes and types. But it remains to be stressed that, in the majority of cases, the analysis of change does not consider (or only in a very limited way), 62

General Papers

Figure 3: Distribution map of baking covers with four tongue handles, from the Orientalizing period to the Late Archaic phase (second half of the 7th - first half of the 5th century BC). Source: Zifferero 2000. Symbols: 1 = modern centre; 2 = ancient centre; 3 = cover with curved tongue handles; 4 = cover with plain tongue handles; 5 = cover with triangular, plain tongue handles. The empty symbols mark uncertain cases. Topography: 1 = Vulci; 2 = Bolsena, Barano; 3 = Volsinii; 4 = Tarquinia; 5 = Pyrgi; 6 = Canale Monterano, Piana di Stigliano; 7 = San Giovenale; 8 = Cerveteri; 9 = Falerii veteres; 10 = Magliano Sabina; 11 = Veio; 12 = Veio, Casale Pian Roseto; 13 = Roma, Acquafredda; 14 = Ficana; 15 = Castel di Decima; 16 = Roma, Acqua Acetosa-Laurentina; 17 = Roma; 18 = Roma, Tor Vergata; 19 = Roma, La Rustica; 20 = Antemnae; 21 = Fidene; 22 = Ficulea; 23 = Crustumerium; 24 = Cures Sabini; 25 = Lavinium; 26 = Ardea; 27 = Ariccia and settlements of the Alban district; 28 = Gabii; 29 = settlements of the district of Tivoli ; 30 = Palombara Sabina, Cretone; 31 = Satricum; 32 = Velletri; 33 = Palestrina; 34 = Artena, Muracci di Crepaddosso; 35 = Segni; 36 = Colleferro, Colli S. Pietro; 37 = Anagni; 38 = Sora.

developments occurring in the food system. A study of Rome in the regal period is considered briefly in the following. In the case of Rome a statistical analysis of domestic pottery data by Paolo Carafa (1995) registers a change at the end of the 6th century BC in shapes and mode of production. In this period corresponding with the concluding years of the Etruscan monarchy at Rome, there was a substantial decrease in the number of shapes, including that of the cylindrical-ovoid jar, combined with the contemporary

introduction of a new fabric, the “impasto augitico” class, characterized in open shapes, mostly represented by basins (Matteucci 1986, Rossi Diana & Clementini 1988). A possible explanation of this phenomenon is the introduction of a new and more advanced species of wheat for the making of bread, namely Triticum aestivum or Triticum compactum that substituted the local spelt. The introduction of new mortaria shapes in “impasto augitico” derived from Etruria, and is perhaps to be connected with the practice of frumentationes (i.e. distributions of grain to 63

RPC 2000 conference proceedings

Figure 4: Hypothetical scheme considering the role of domestic pottery in connection with pork meat in the Roman food system, between the 8th and 7th century BC. Source: Zifferero forthcoming, a.

Figure 6: Interpretative scheme, showing the approach to the exploitation of sulfide ore in the pre-industrial age in the Monti di Campiglia, between the Valle del Temperino and the Gran Cava. After a drawing in Bianchi et al. 1996, n. 38 (A. Casini). Symbols: 1 = pits, shafts and galleries probably mined in the Etruscan and Roman periods, from the surface to the core of the ore, at a relevant depth; 2 = galleries and open cast exploitation areas, excavated in the Renaissance period. The upper section (phase A) reproduces a typical pre-Industrial cultivation approach, different from later excavations (phase B).

appear to be those dedicated to the comparison of the protohistoric production of Mediterranean metallurgy, with a general assessment of types, between the Bronze and early Iron Ages (Giardino 1995). Some contributions have dealt with the relationship between regional metallurgical production and its region, as in the case of the relation between the Fiora valley and northeastern Italy (see bibliography collected in Zanini 1997). From recent assessments of the state of the art of archaeometallurgical studies, we may note the following issues: • The detection of a trend in the use of copper sulfides (mainly chalcopyrite), to produce the so-called “black copper”, in the early Bronze Age (Casagrande et al. 1992). Archaeological evidence demonstrates that the exploitation of ores in combination with metallurgical activities in the Colline Metallifere had already taken place in the Chalcolithic (Fedeli 1995). • New evidence attesting to the metallurgical production of lead at the time of trading contacts with Mycenaeans in the late Bronze Age, has been found in southern Latium and in the Fiora valley (Angle et al. 1992, Poggiani Keller 1999). • The presence of iron ore in smelting activities documented at many final Bronze Age sites in central and southern Italy testifies to an earlier date for iron production (Delpino 1993).

Figure 5: Hypothetical scheme considering the role of domestic pottery in connection with cereals in the Roman food system, between the 6th and 5th century BC. Source: Zifferero forthcoming, a.

the urban population). This practice may have been introduced in Latium vetus together with the new species of wheat, probably in connection with the practice of baking with yeast (figure 5). 2.7.3 Metallurgy The introduction of regional perspectives on the development of metallurgical technologies is the goal of the second part of this paper. The new data on pre-Roman mining activities that has emerged from recent research in the area of the Colline Metallifere (Tuscany) is reviewed here, as it sheds light on different attempts to extract copper and lead from sulfide ores and iron from oxide ores. The idea is that the different minero-genetic and geological complexes, in which ores are found, stimulated the transfer of metallurgical technologies from more advanced mining and metallurgical areas to less advanced regions. The archaeological evidence will be considered as an indicator of such changes. Recent research on metallurgy has focused both on techniques of archaeometallurgy and on the exploitation of ores. Among various contributions, the most relevant

I want to emphasize here the importance of regional developments in metallurgical activities, i.e. in the relation between extractive techniques and metallurgical processes. In this respect, the new evidence on the Ligurian copper 64

General Papers

Figure 7: Interpretative drawing of the section (A) and plan (B) of the Temperino mine (Campiglia Marittima). The 19th - 20th century cut of the main gallery has broken through into the Etruscan and Roman system of pits and shafts. These works, aimed at the exploitation of the ore, have been refilled by the ancient miners with sterile material that subsequently became compacted. The modern gallery has made the ancient works visible from the bottom of the ore deposits: the arrows mark the position of the Roman/Etruscan pits. After a drawing in Bianchi et al. 1996, n. 40 (A. Casini - M. Mellini). Symbols: 1 = massive limestone; 2 = red limestone; 3 = sulfide ore (skarn); 4 = porphyry; 5 = ancient mining activity. In the plan B: 1 = gallery cut in the 1950’s; 2 = small gallery; 3 = modern work that broke through into the ancient system of pits and shafts; 4 = part of the modern mine dug through the system of “rooms and pillars”; 5 = ancient works filled with sterile ore; 6 = skarn; 7 = part of the gallery mined into the limestone.

Figure 8: Section and plan of the Buche al Ferro (R1) (Castagneto Carducci) (see figure 2.7.10 no. 1). Source: Cascone 1993.

mining site of the Libiola, dating back to the Chalcolithic, and from the Sardinian sulfide mines, may be analyzed in parallel with the activity of sulfides smelting (with special attention to copper and lead), carried out in Nuragic Sardinia (Campana et al. 1996, Lo Schiavo 1996, Giardino 1998). The influence exerted by Aegean metallurgy on the metal production of Sardinia is well known (firstly, by the presence of Cypriot copper in the form of portions of oxhide ingots, and secondly by the exchange of technological expertise on metallurgical practices) (Lo Schiavo et al. 1990). Particularly relevant is the study of exchange in technical knowledge in the northern Tyrrhenian area. Here archaeological and archaeometallurgical work-in-progress on the exploitation of the Monti di Campiglia ores has lead to new hypotheses (Tanelli 1993, Tanelli et al. 1994, Mascaro & Cuteri 1995). This project aims at a diachronic study of metallurgy in the main phases of human occupation of the area from the protohistorical period up to our own day (Casini & Cascone 1997) (figures 6-7). From our perspective the early phases of exploitation are the most interesting. The presence of copper carbonates and of copper, lead and silver sulfides apparently stimulated the min-

ing and metallurgical activity as early as the Chalcolithic period, for metallurgical activity in copper has been detected in the northern area of that mining basin (Fedeli 1995). More intensive use of the mining resources of the basin has been documented for the Etruscan period (6th - 3rd centuries BC), the period when the city of Populonia controlled the Campigliese district where the iron ore from Elba was processed (Bartoloni 1991, Casini 1993) (figures 8-9). It is interesting to note the connection in the late Iron Age and Etruscan period between mines and sites as a new type of fortified settlement was established in the 6th century BC to improve control of the area (Zifferero forthcoming, b) (figure 10). In the Roman and Medieval periods this district knew remarkable mining activity, but due to the different socioeconomical context in which it and metallurgical activities took place, a significant change came about in the settlement pattern (Casini & Francovich 1993). A recent new perspective in the study of this period is the integrated analysis of metallurgical activities through the investigation of Medieval sulfide slag recovered from the site of Rocca San Silvestro (Benvenuti et al. 1992, see also Cucini Tizzoni & Tizzoni 1992). This research carried out in parallel with the analysis of copper and lead/silver ore, has lead to the 65

RPC 2000 conference proceedings

Figure 9: Plan and section of the Buche al Ferro 2 (R2) (Castagneto Carducci) (see figure 2.7.10 no. 1). Source: Cascone 1993.

hypothesis of the so-called “single cycle”, developed by the metallurgists to separate mixed sulfide ores, (the “skarn”), in order to obtain black copper and lead. The silver was obtained as a by-product of lead through a process of cupellation (Guideri 1998) (figure 11). This process requires a technological approach different from others, as it is strictly dependent on the nature of the ores. This raises new questions in the archaeometallurgical field and demands a precise assessment of the problems connected with the successful (or unsuccessful) metallurgical approach to ores. A more precise documentation of such processes may therefore explain the reason and nature of some contacts, apparently paradoxical, such as the presence of Cypriot copper in Nuragic Sardinia, a land extremely rich in copper (Lo Schiavo et al. 1990, Carancini 1996). Recently I have formulated the hypothesis that the Villanova communities of the Populonia and Vetulonia area (the latter controlling an area very rich in ore sulfide deposits in the Colline Metallifere) profited from the transfer of the more advanced Sardinian metallurgy of Nuragic Sardinia to the Etruscan communities to exploit their very mixed sulfide ores (Zifferero forthcoming, b). It is an observation building on suggestions made by C. Giardino (1998), and which has found confirmation in the late appearance of lead objects (8th century BC), presumably

Figure 10: Map of the pre-Roman settlement system in the area of the mining basin of the Monti di Campiglia. Source: Zifferero forthcoming, b. Symbols: 1 = modern centre; 2 = open-cast modern quarry; 3 = pre-Roman site; 4 = pre-Roman site with traces of slag, or quarry; 5 = necropolis; 6-7 = ancient mines; 8 = Hellenistic site. Topography: 1 = Buche al Ferro (Castagneto Carducci); 2 = Buca della Guardia (Castagneto Carducci); 3 = Podere I Cancellini (Castagneto Carducci); 4 = Cava Solvay (San Vincenzo); 5 = Cava Solvay (= CM 123.1) (San Vincenzo); 6 = Buca degli Spagnoli (San Vincenzo); 7 = I Manienti, sommità (San Vincenzo); 8 = Cava di Monte Calvi (Campiglia Marittima); 9 = Poggio Le Strette (San Vincenzo); 10 = Podere San Dazio (San Vincenzo); 11 = Monte Rombolo, sommità (Campiglia Marittima); 12 = Campo alle Buche (Campiglia Marittima); 13 = Fornelli di Monte Rombolo (Campiglia Marittima); 14 = Cava di Valle Lanzi (Campiglia Marittima); 15 = Buca della Caprareccia (Campiglia Marittima); 16 = Buca di Poggio all’Aione (Campiglia Marittima); 17 = Buca dell’Aquila (Campiglia Marittima); 18 = Discarica della Gran Cava (Campiglia Marittima); 19 = metallurgical site of Val Fucinaia and Capattoli (Campiglia Marittima); 20 = Monte Spinosa, slopes (Campiglia Marittima); 21 = Cava Maffei (Campiglia Marittima); 22 = Monte Spinosa, summit (Campiglia Marittima); 23 = Buca del Moro e Buca del Morello (Campiglia Marittima); 24 = Campiglia Marittima, via Burattelli; 25 = Campiglia Vecchia (Campiglia Marittima); 26 = Monte Valerio (Campiglia Marittima); 27 = Cava di Monte Valerio (Campiglia Marittima).

66

General Papers

Bartoloni, G. 1991, Populonium etruscorum quodam hoc tantum in litore. Aspetti e carattere di una comunità costiera nella prima età del ferro, in: Studi in onore di Massimo Pallottino, Archeologia Classica XLIII: 1-37. Benvenuti, M., Francovich, R., Guideri, S., Mascaro, I. & Tanelli, G. 1992, Le scorie metallurgiche medievali di Rocca San Silvestro, in: L’Appennino settentrionale (76e Riunione della Società Geografica Italiana), Firenze: 295-297. Bianchi, G., Boldrini, E., Casini, A., Cicali, C., Guideri, S. & Zifferero, A. 1996, San Silvestro. Parco Archeominerario: Museo del Parco e Museo della Miniera. Pannelli, Firenze. Campana, N., Maggi, R., Stos Gale, Z. & Houghton, J. 1996, Miniere e metallurgia in Liguria fra IV millennio e IV secolo a. C. Un excursus, in: Piola Caselli, F. & Piana Agostinetti, P. (eds), La miniera, l’uomo e l’ambiente. Fonti e metodi a confronto per la storia delle attività minerarie e metallurgiche in Italia, Firenze: 15-52. Carafa, P. 1995, Officine ceramiche di età regia. Produzione di ceramica in impasto a Roma dalla fine dell’VIII alla fine del VI secolo a. C., Roma. Carancini, G.L. 1996, Metallurgia e società nell’Italia protostorica, in: Piola Caselli, F. & Piana Agostinetti, P. (eds), La miniera, l’uomo e l’ambiente. Fonti e metodi a confronto per la storia delle attività minerarie e metallurgiche in Italia, Firenze; 287304. Casagrande, A., Garagnani, G.L., Poli, G. & Spinedi, P. 1992, Considerazioni sulla metallurgia del rame nell’antichità: studio delle inclusioni in pani, panelle e lingotti, in: Antonacci Sanpaolo, E. (ed), Archeometallurgia. Ricerche e prospettive, Bologna: 149-157. Cascone, G. 1993, La zona speleologica del Massiccio di Monte Calvi. Primo contributo alla sua conoscenza, in: La scienza della terra nell’area della Provincia di Livorno a sud del fiume Cecina (Quaderni del Museo di Storia Naturale di Livorno, 13, suppl. 2): 183-212. Cascone, G. & Casini, A. 1997, Metodologia per lo studio delle attività minerarie antiche nei Monti di Campiglia Marittima, in: Zanini, A.1997, Dal bronzo al ferro. Il II millennio a.C. nella Toscana centro-occidentale, Pisa: 21-23. Casini, A. 1993, Archeologia di un territorio minerario: i Monti di Campiglia, in: La scienza della terra nell’area della Provincia di Livorno a sud del fiume Cecina (Quaderni del Museo di Storia Naturale di Livorno 13, suppl. 2): 303-314. Casini, A. & Francovich, R. 1993, Problemi di archeologia mineraria nella Toscana medievale: il caso di Rocca San Silvestro, in: Les Techniques Miniéres de l’Antiquité au XVIII Siècle, (Actes 113° Congr. Nat. Soc. Savantes), Editions du CTHS, Paris: 249-265. Cavallotti Batcharova, A. 1965, Veio (Isola Farnese). Continuazione degli scavi nella necropoli villanoviana in località “Quattro Fontanili”, in: Notizie Scavi: 49-225. Cucini Tizzoni, C. & Tizzoni, M. 1992, Le antiche scorie del Golfo di Follonica (Toscana). Una proposta di tipologia, Notiziario Museo Civico Milano, supplemento IX. Cocchi Genick, D. (ed) 1999, Criteri di nomenclatura e di terminologia inerente alla definizione delle forme vascolari del Neolitico/Eneolitico e del Bronzo/Ferro, Firenze. Colonna, G. 1963-64, Area sacra di S. Omobono. La ceramica di impasto posteriore agli inizi dell’età del ferro, in: Bullettino

Figura 11: Diagram showing the hypothetical “single cycle” method for obtaining metal from the sulfide ore. The highly mixed polymetallic nature of the sulfides in the Monti di Campiglia made a manual separation of the copper ore from the lead/silver ore very difficult. This cycle allowed an operational sequence aiming to gain various metals from the same ore. Elaborated from a drawing in Bianchi et al. 1996, n. 29 (S. Guideri).

related to the local ores, in the area surrounding the Colline Metallifere. Further advances in the technology of iron may have been based on the more advanced level of knowledge of the Chalcidians, whose skill in the making of swords has been widely testified to by the literary sources. The introduction of the processes of carburation and tempering to strengthen the metal may have been caused by contacts with the Euboian colonists who settled in the lower Tyrrhenian area in the cities of southern Etruria (Mele 1981, La Salvia 1997).

References Andrè, J. 1961, L’alimentation et la cuisine a Rome, Paris. Angle, M., Caneva, C., Conti, A.M., Dottarelli, R., Gianni, A., Giardino, C. & Persiani, C. 1992, Casale Nuovo (LT) e la tarda età del bronzo nel Lazio meridionale, in: La Sardegna nel Mediterraneo tra il bronzo medio e il bronzo recente, Cagliari: 265-303.

67

RPC 2000 conference proceedings

della Commissione Archeologica Comunale di Roma 79: 3-32. Cubberley, A.L., Lloyd, J.A. & Roberts, P.C. 1988, Testa and klibani: the baking covers of classical Italy, in: Papers of the British School at Rome 56: 98-119. Delpino, F. 1993, Apporti egei nell’avvio di attività siderurgiche nell’Italia antica?, in: Bullettino di Paletnologia Italiana 84: 481-490. Dumézil, G. 1977, La religione romana arcaica. Milano. Fayer, C. 1982, Aspetti di vita quotidiana nella Roma arcaica. Dalle origini all’età monarchica, Roma. Fedeli, F. 1995, Scavo di un insediamento eneolitico nel distretto minerario del Campigliese (LI), in: Negroni Catacchio, N. (ed), Preistoria e Protostoria in Etruria 2. Tipologia delle necropoli e rituali di deposizione, Milano: 73-81. Giardino, C. 1995, Il Mediterraneo occidentale fra XIV e VIII secolo a. C. Cerchie minerarie e metallurgiche (BAR International Series 612), Oxford. Giardino, C. 1998, I metalli nel mondo antico. Introduzione all’archeometallurgia, Roma-Bari. Guideri, S. 1998, The Production of Metals for Coinage in Medieval Tuscany. The Technological Context, in: Milliken, S., Vidale, M. (eds), Papers from the EAA Third Annual Meeting at Ravenna 1997, vol. IV: Craft Specialization: Operational Sequences and Beyond (BAR International Series 720): 153-157. Guidi, A. 1993, La necropoli veiente dei Quattro Fontanili nel quadro della fase recente della prima età del ferro italiana, Firenze. La Salvia, V. 1997, La fabbricazione delle spade delle grandi invasioni. Per la storia del “processo diretto” nella lavorazione del ferro, in: Quaderni Medievali 44: 28-54. Longo, O. & Scarpi, P. (eds) 1989, Homo edens. Regimi, miti e pratiche dell’alimentazione nella civiltà del Mediterraneo. Milano. Lo Schiavo, F., Maddin, R., Merkel, J., Muhly, J.D. & Stech, T. 1990, Analisi metallurgiche e statistiche sui lingotti di rame della Sardegna (Quaderni della Soprintendenza Archeologica di Sassari e Nuoro 17), Ozieri. Lo Schiavo, F. 1996, Miniere e metallurgia in Sardegna: la ricerca archeologica dal presente al passato, in: Piola Caselli, F. & Piana Agostinetti, P. (eds), La miniera, l’uomo e l’ambiente. Fonti e metodi a confronto per la storia delle attività minerarie e metallurgiche in Italia, Firenze: 187-206. Mannoni, T. 1965, Il testo e la sua diffusione nella Liguria di Levante, in: Bollettino Ligustico 17 (1-2): 59-64. Mascaro, I. & Cuteri, F. 1995, Colline Metallifere. Inventario del patrimonio minerario e mineralogico in Toscana: aspetti naturalistici e storico-geografici (Regione Toscana, Dipartimento Ambiente), Firenze. Matteucci, P. 1986, L’uso dei mortai di terracotta nell’alimentazione antica, Studi Classici e Orientali 36: 239-277. Mele, A. 1981, I Ciclopi, Calcodonte e la metallurgia calcidese, in: Nouvelle Contribution à l’étude de la société et de la

colonisation eubéennes (Cahiers du Centre Jean Bernard 6), Napoli: 9-33. Negroni Catacchio, N. 1995, I materiali. Ipotesi sulla destinazione d’uso, in: Negroni Catacchio, N. (ed), Sorgenti della Nova. L’abitato del Bronzo Finale, Firenze: 397-403. Nijboer, A.J. 1998, From Household Productions to Workshops. Archaeological Evidence for Economic Transformations, Premonetary Exchange and Urbanisation in Central Italy from 800 to 400 BC, Groningen. Parise Badoni, F. (ed) 2000, Ceramiche d’impasto dell’età orientalizzante in Italia. Dizionario terminologico, Roma. Poggiani Keller, F. 1999, Scarceta di Manciano (GR). Un centro abitativo e artigianale dell’età del bronzo sulle rive del Fiora, Manciano. Rendeli, M. 1993, Città aperte. Ambiente e paesaggio rurale organizzato nell’Etruria meridionale costiera durante l’età orientalizzante e arcaica, Roma. Roma 1989, Il viver quotidiano in Roma arcaica, (catalogo della mostra) Roma. Rossi, D. & Clementini, M. 1988, Nuove considerazioni sul tipo del bacino di impasto augitico, in: Atti Accademia dei Lincei. Rendiconti Morali XLIII: 39-72. Santoro Bianchi, S. & Fabbri, B. (eds) 1997, Il contributo delle analisi archeometriche allo studio delle ceramiche grezze e comuni: il rapporto forma/funzione/impasto. Bologna. Scheffer, C. 1981, Acquarossa II.1. Cooking and cooking stands in Italy 1400-400 BC, Stockholm. Tanelli, G. 1993, I minerali e le miniere del Campigliese, in: La scienza della terra nell’area della Provincia di Livorno a sud del fiume Cecina (Quaderni del Museo di Storia Naturale di Livorno 13, suppl. 2): 165-182. Tanelli, G., Morelli, F., & Benvenuti, M. 1994, I minerali del Campigliese: beni ambientali, culturali e industriali, in: Bollettino Società Geologica Italiana 112: 715-728. Zanini, A. (ed) 1997, Dal bronzo al ferro. Il II millennio a.C. nella Toscana centro-occidentale, Pisa. Zifferero, A. 2000, La ceramica preromana come indicatore di processi socio-economici: il caso dell’Italia medio-tirrenica, in: Francovich, R. & Patterson, H. (eds), Extracting Meaning from Ploughsoil Assemblages, Oxford: 147-159. Zifferero, A. forthcoming (a), Ceramica preromana e sistemi alimentari: elementi per una ricerca, in: Further approaches to regional archaeology in the Middle Tiber Valley, British School at Rome, 1998. Zifferero, A. forthcoming (b), Estrazione, lavorazione e commercio dei minerali tra Sardegna ed Etruria: per una definizione dei meccanismi e delle direttrici di scambio nel Mar Tirreno, in: Etruria e Sardegna centro-settentrionale tra l’età del bronzo finale e l’arcaismo, atti del XXI Convegno di Studi Etruschi ed Italici, Alghero, 1998.

68

3

Comparative Settlement Archaeology

spective it must be concluded that, although the ‘breaking up’ of the Italian landscape has contributed to the breakdown of traditional generalizations, few attempts have been made to formulate new historical questions and supraregional syntheses. And yet, so many regional data have become available for it. As a matter of fact, the regionalization inherent in recent fieldwork programs has led rather to an overemphasis on internally-driven regional stimuli for change, from environmental constraints to elite social strategies, at the cost of factors related to the integration of local societies into interregional networks of exchange, power or mentalities. Especially odd does this seem in the case of Italian urbanization studies, since with urbanization and growing social complexity in general, the rate of integration tends to increase and is therefore likely more and more to determine regional dynamics. This issue was confronted in the second part of the workshop. Discussion focused on the kind of questions that can or should be asked at the supra-regional level and on the applicability of models to explain the observed similarities and divergences in regional trajectories. Below, the transcribed version of the discussions is presented in paragraph 3.4.

Convenor: Gert-Jan Burgers 3.1

Introduction

The general aim of the workshop on comparative settlement archaeology was to compare and synthesize the results of recent fieldwork, and in particular of survey projects across a range of Italian regions. The comparison focuses on the major theme of the RPC project, i.e. 1st millennium BC urbanization in Central and Southern Italy. In the plenary session of this conference it was emphasized that recent trends in Italian archaeology are moving away from the traditional perception of this process as a rather homogeneous one, triggered in particular by dominant Greek and Roman colonial and urban forces. In contrast, from the perspective of regional settlement and landscape archaeology, urbanization in ancient Italy is increasingly being approached as a highly differentiated phenomenon. It is now acknowledged that there existed a range of variations in regional trajectories towards urbanism. No more do we draw on such historical events as the founding of the Greek colonies for our principal explanatory device. Instead of viewing events as major causes, we approach them as elements in long and medium-term cycles of moderate change in the organization of regional landscapes, settlement systems and societies. This recent line of research has set in motion numerous regional fieldwork projects, incorporating excavation, field survey and environmental data for the study of such changes. The first aim of the present workshop is to evaluate some of these projects, to discuss the databases available and to assess what questions can be asked at the regional level. Accordingly, participants were invited on the basis of their involvement in regional fieldwork projects in Italy, or their interpretative contributions to the theme of regional settlement archaeology. As preparation, they were asked to present a working paper including: (1) a short chronological overview of 1st millennium BC settlement trends in the respective fieldwork area(s) or study region(s), (2) an evaluation of the nature and intensity of the research methods employed in the various regions, (3) comments on how representative were the data sets thus acquired, and (4) a discussion of the explanatory frameworks used to interpret the data sets. The working papers received are presented in section 3.3. Most of them are structured as mentioned above. However, some have been rewritten to a different structure since the conference. Jean-Pierre Vallat has worked his contribution into a full paper, which has been included in chapter 7 of this volume. The second major aim of this workshop goes beyond the regional level of research. It deals with the comparative regional archaeology approach that we wish to develop in the context of the RPC project. From this comparative per-

3.2

List of Participants

Marianne Kleibrink (Groningen Institute of Archaeology, University of Groningen, Poststraat 6, 9712ER Groningen, The Netherlands; [email protected]) is Professor in Classical Archaeology and Numismatics at the University of Groningen. Since the late 1970’s she has directed excavations at the protohistorical site of Satricum in Lazio. In 1989 she took up the excavations of the Timpone della Motta at Francavilla Marittima in Calabria, where she excavates the indigenous, and later Greek, sanctuary and part of the Bronze Age-to-Archaic settlement. Steve Thompson (Institute of Classical Archaeology, Waggener Hall 123, University of Texas at Austin, Austin, TX 78702; [email protected]) is Research Scientist at the Institute of Classical Archaeology of the University of Texas at Austin. He received his PhD in Anthropology in 1999 from the University of Virginia. As part of his dissertation research, an analysis of Hellenization as a process of transformation of regional socio-political and economic systems, he directed the Morgantina Archaeological Survey in central Sicily. In his current position at the University of Texas he is preparing for publication the results of four seasons (1981-1984) of survey in the chora of colonial Greek Metaponto in southern Italy. He is also currently directing an additional survey in the Metapontino as well as excavations in the chora of Greek Chersonesos on the Black Sea. Gert-Jan Burgers (Archaeological Institute, Vrije Universiteit, De Boelelaan 1105, 1081 HV Amsterdam, The Netherlands; [email protected]) is lecturer and research fellow at the Amsterdam Vrije Universiteit. Since 1992 he has been involved in the Brindisino Project with Johannes 69

RPC 2000 conference proceedings

Boersma and Douwe Yntema. As part of his dissertation research on settlement, landscape and social transformations in ancient Salento, he has directed surveys on the Salento isthmus. He is presently directing excavations at the Hellenistic fortified sites of Muro Tenente and Li Castelli di San Pancrazio, Salentino, in the same region. He is also coordinator with Peter Attema of the Regional Pathways to Complexity Project.

Helen Patterson (British School at Rome, Piazzale W. Churchill 5, 00197, Rome, Italy; [email protected]) took her degree in Archaeology at the University of Reading. Her doctorate was undertaken at the University of Sheffield on The later Roman and early medieval pottery from San Vincenzo al Volturno, Molise: production and distribution in central and southern Italy, AD 400-1100. As coordinator for the Siena element of the EU funded POPULUS project she extended her interests to the analysis of methodological approaches to Mediterranean field survey assemblages. In 1996 she was appointed Assistant Director (Archaeology) at the British School at Rome. As such, she directs the School’s archaeological research initiative, the Tiber Valley Project.

Frank Vermeulen (Vakgroep Archeologie, Universiteit van Gent, Blandijnberg 2, 9000 Gent, Belgium; [email protected]) is Professor in Classical Archaeology at Ghent University. His main earlier research projects are the Romanisation of sandy Flanders (regional research based on systematic field walking and excavations in Northern Belgium; 1981-1991), a geo-archaeological survey and excavations in the territory of ancient Pessinus (Central Turkey, 1990-1996) and a geo-archaeological study, based on aerial photography, fieldwork and GIStechnology, of protohistoric and Roman roads and land divisions in the north Gaulish Civitas Menapiorum (WBelgium/NW-France, 1997-1999). He is currently directing a new survey-project (aerial survey, systematic field walking) in the Marche region, in collaboration with the University of Macerata, titled: “The Potenza valley survey. From Acculturation to Social Complexity in Antiquity: A Regional Geo-Archaeological and Historical Approach”

Grazia Semeraro (Scuola di Specializzazione in Archeologia Classica e Medievale, Università di Lecce, Via D. Birago, 64, 73100 Lecce, Italy; [email protected]) took her degree in Archaeology at the University of Lecce (Italy) in 1983. Following her doctorate on the distribution of imported Greek pottery in Southern Apulia, she is now a researcher at the University of Lecce, teaching the archaeology of Magna Graecia and applied informatics both there and at the School of Specialization in Classical and Medieval Archaeology, University of Lecce. She is presently co-ordinating the activities of the Laboratory of Archaeological Informatics at the University of Lecce. Since 1984 she has also worked in Turkey as a member of the Italian Archaeological Mission at Hierapolis (Phrygia). She is currently responsible for excavations in the archaic settlement at S.Vito dei Normanni (Brindisi, Italy) and in a research project at Tas Silg (Malta). Her primary research interests concern problems of interaction and exchange between Greeks and natives in Southern Italy, and computer-assisted applications in archaeological research.

Rob Witcher (British School at Rome, Piazzale W. Churchill 5, 00197, Rome, Italy; [email protected]) is a Leverhulme Research Fellow, working on the Tiber Valley Project at The British School at Rome. He completed his Doctoral thesis, entitled Modelling Roman Imperialism: Landscape and Settlement Change in Italy in 1999 at Leicester University, UK, with Prof. D. Mattingly and Prof. C. Ruggles. Research interests include Landscape Archaeology, especially the theories and methodologies of surface survey techniques; the history and archaeology of pre-Roman/Roman Italy, particularly the nature of imperialism/colonialism; the theories of spatial representation, the history of cartography, maps and itineraries; notions of spatio-temporal perception and experience.

Francesco D’Andria (Scuola di Specializzazione in Archeologia Classica e Medievale, Università di Lecce, Via D. Birago, 64, 73100 Lecce, Italy; [email protected]) is professor in Classical Archaeology and director of the Scuola di Specializzazione in Archeologia Classica e Medievale of the Università di Lecce. Under his guidance, the Scuola has developed into a center of expertise on the archaeology of Magna Graecia, and especially of the Salento region, not only through a series of well defined excavation projects (notably at Cavallino, Otranto and Vaste), but also by the use of advanced data recording and processing techniques. Prof. D’Andria also has a Chair in Roman Archaeology and Art History at the Scuola di Specializzazione in Archeologia di Matera, and is director of the Italian Archaeological Mission at Hierapolis (Phrygia), in Turkey.

Helle Winge Horsnæs (Ndr. Fasanvej 201 st.tv, DK-2000 Frederiksberg, Denmark; [email protected]) is curator of antique coins of the Royal Collection of Coins and Medals, at the National Museum in Copenhagen. She studied classical archaeology at the University of Copenhagen and at the Istituto Universitario Orientale, Naples. In 1989 she received a mag.art., with a thesis on the Campano-Etruscan necropolis at Arenosola (SA) (parts published as articles). In 1999 she successfully defended her PhD dissertation on The Cultural Development in Northwestern Lucania c. 600-273 BC, parts of which will be published in the ARID Supplementum series in 2001. In 1994 she arranged (with S. Houby-Nielsen and H. Damgaard Andersen) a seminar on Urbanization in the Mediterranean in the 9th to 6th centuries BC in Copenhagen - later published as the Acta Hyperborea 7, 1997.

Alastair Small (Department of Classics, Edinburgh University, David Hume Tower, George Square, Edinburgh EH8 9JX, Great Britain; [email protected]) is an honorary fellow of the Department of Classics, Edinburgh University, Scotland and Professor Emeritus of the Department of History and Classics, University of Alber70

Comparative Settlement Archaeology

Department of History and Classics of the University of Alberta, Canada, he has directed the excavations of the Canadian Archaeological Mission at Roccagloriosa (province of Salerno, Italy) from 1989 until 1992. From 1990 until 1999 he has also co-directed the Upper Bradano Valley Survey Project, as well as excavations at the site of Oppido Lucano. Since 1992 he is co-director of the excavations of a Roman villa at Ossaia (Cortona, province of Arezzo, Italy).

ta, Edmonton (Canada). He has been involved in problems of settlement archaeology in South Italy since 1965, especially at the sites of Botromagno, Monte Irsi, and San Giovanni di Ruoti. He directed several field surveys intended primarily to describe the environments of these sites, but over time a secondary objective - to reconstruct changing patterns of settlement and land use over time - has become particularly important. He is currently editing a monograph on several field surveys carried out by S.P. Vinson, H. Aldridge, C. Chapman and others in the late 1960s and early 1970s in the vicinity of Gravina di Puglia.

Jean-Pierre Vallat (Université de Paris 13 UFR des Lettres et Sciences Humaines, Avenue Jean-Baptiste Clément, 93440 Villetaneuse, Paris, France; [email protected]) is Professor in Roman History at the Université Paris-Nord.

John Bintliff (Faculteit Archeologie, Rijksuniversiteit Leiden (RUL), Reuvensplaats 4, 2300 RA Leiden; [email protected]) is Professor in Classical Archaeology at the RUL. He studied Archaeology and Anthropology at Cambridge University (BA, Ma), setting a long-term interest in Environmental and Landscape Archaeology and in Archaeological Theory. His PhD thesis was also submitted at Cambridge - Natural Environment and Human Settlement in Prehistoric Greece (published by BAR, Oxford in 1977). In 1978 he began a highly-fruitful collaboration with Professor Anthony Snodgrass of Cambridge University: a regional fieldwork project in Greece - the Boeotia Project, which conducted annual field survey programs almost without interruption from 1979 to 1991. His interests in the history, theory and philosophy of archaeology as a discipline have continued to be reflected in a series of papers and edited volumes up to the present-time. Landscape history has also continued as a major interest. Gabriele Cifani (Via F. Lucchini 16, 00136 Roma, Italy; [email protected]) is presently writing his PhD thesis at the university of La Sapienza at Rome on the urbanization of central Italy. Alessandro Vanzetti (Museo Pigorini, Roma, Italy; [email protected]) is presently appointed at the Museo Pigorini and detached to the La Sapienza University at Rome where he carries out research. He is closely involved in the excavations at Broglio di Trebisacce in Calabria directed by Prof. R. Peroni. Elisabeth van ‘t Lindenhout (Groningen Institute of Archaeology, University of Groningen, Poststraat 6, 9712ER Groningen, The Netherlands; [email protected]) is temporarily appointed as lecturer in archaeology at the University of Groningen. Presently she is finishing her PhD thesis on the early urbanization and architecture of Lazio between the 8th and 4th centuries BC Maurizio Gualtieri (Università degli Studi di Perugia, Istituto di Studi comparati sulle società antiche, Via Armonica, 3, 06123 Perugia, Italy; [email protected]) received his PhD in Classical Archaeology at the University of Pennsylvania, Philadelphia. He is currently associated professor at the Dipartimento di Studi Storico-Artistici of the Università degli Studi of Perugia. As a former professor at the 71

RPC 2000 conference proceedings

3.3

Working Papers

The primary data sets of interest to the project are those provided by intensive field surveys, supported by environmental research, problem-oriented excavations and inventories of published sources (excavations and incidental prospections). The surveys carried out in the 1980’s (notably the Oria and Valesio surveys) covered a total surface of some 81 km2, incorporating various environmental zones throughout the ca. 600 km2 Brindisi plain. Allowing for minor shifts in methodology and in the composition of teams and their direction, no insurmountable differences have arisen to prevent a reliable comparison of the survey results. As a matter of fact, data compatibility was one of the aims of the projects. As is the case with most Mediterranean survey projects, the various surveys in the Brindisino first focused in particular on rural settlement patterns. Although some of the larger sites in the region have recently been the scene of small-scale excavations, the majority of them still constitute a terra incognita, except for some sporadic reports on incidental finds. This holds good for larger defended sites as well as for sites without fortifications. Since the larger settlements are a crucial factor in any understanding of settlement hierarchy and related phenomena, a decision was made to tackle this hiatus in the data sets and to incorporate four of such site areas and their peripheries into the regional survey project (early 1990’s). In the late 1990’s it was decided to confront another bias in the regional data set, i.e. the uneven intensity of research activities between the Brindisi plain and the relatively marginally exploited hills of the so-called Murgeplateau, north of the Brindisi plain, which hitherto has seen only sporadic, incidental investigations. In the Murge area surveys were launched in 1999 and 2000, in the context of the RPC project. The survey results are complemented by archaeological data from archival and museum research, and from publications of sporadic find reports and incidental field walking. Whereas field observations of archaeological sites have been recorded from the 16th century onwards, the surge of interest in ancient topography during the last decades is responsible for the location of many new archaeological find spots. They have been inventoried in a number of publications. Of particular interest is the compilation of the Repertorio dei beni culturali archeologici della provincia di Brindisi, a comprehensive catalogue of all finds hitherto reported in the Brindisino (Quilici & Quilici-Gigli 1975). New discoveries are frequently published in such series as the Notiziario delle attività di tutela, the annual supplement to the journal of Taras. They have all been inventoried in the RPC site database, to be GIS-manipulated. These particular data pose serious problems to site interpretation from the perspective of a regional landscape approach. Most of them have been unsystematically collected and recorded, and the data collections are seriously biased by the activities of (amateur) archaeologists prospecting a given area, and by their personal research interests. For instance, they avoid certain areas because of a preconceived absence of Classical/Roman sites, whereas other areas attract them

3.3.1 The Salento Isthmus Project (Apulia) Gert-Jan Burgers General information on the Salento Isthmus project The Salento Isthmus is one of the three regions central to the RPC project. It is the ca. 50 km wide stretch of land connecting the Salento peninsula (the heel of the Italian boot) to Italy. For this region the RPC project draws on the databases elaborated in the context of the ongoing Brindisino research program which the Archaeological Institute of the Free University of Amsterdam (AIVU) has been carrying out since 1981 (the Brindisino forms the eastern half of the Salento Isthmus). Both the aims and methods of this program coincide with those of the RPC project. Its general theme is the Romanization of the indigenous societies of the Brindisi region. This process is studied from a long-term diachronic perspective, with ample research being devoted to the social development of the indigenous peoples in the pre-Roman period from as early as the Late Bronze Age. A major aim of the program is to study the relationship between such social development and the organization of settlement and landscape. To that aim, in the 1980s and 1990s our team carried out a series of intensive, systematic field surveys in different landscape units (notably the Oria and Valesio surveys and a set of urban surveys), and a number of large-scale excavations on both urban and rural sites (notably at Valesio and Muro Tenente). This research has also done much to clarify the still poorly- understood formation of the many fortified towns in the region, a process that is commonly referred to as urbanization. It has contributed to our investigating this process as one not only of settlement nucleation, or of the emergence of an urban way of life, but rather as a process involving a much wider range of social developments, including the physical and symbolic rearrangement of the regional landscape in its totality. The larger part of the Brindisi region as defined in this project is made up of a slightly undulating plain. From the Adriatic it rises very gradually inland to an average 60 m above sea level. The landscape consists of mainly light arable and clayey soils in Pleistocene and Pliocene sands. By contrast, in the immediate coastal zones one finds an alternating landscape of dunes, lagoons and low cliffs. To the south, the plain merges into the more accentuated, calcareous landscape of the Serre Leccesi. To the west and north the hard limestone spurs of the Murge surround the Brindisi district. These hills gradually rise up to the Apennine mountain chain, which divides Italy into a Tyrrhenian and an Adriatic half. Regional settlement trends (see table 1) Evaluation of research biases 72

Comparative Settlement Archaeology

Period

Settlement trend

Site no. Oria survey

Site no. Valesio survey

Final Bronze Age / Early Iron Age (12th9th BC)

* Limited number of rather dispersed sites, located mainly on coastal promontories; few external contacts. Inland Salento almost uninhabited.

1 (Oria)

None

Early Iron Age (8th7th BC)

* Settlement expansion in most of the peninsula; existing sites expand, contemporaneous with an internal colonization of previously marginally exploited inland areas; emergence of settlement hierarchy (based mainly on size); all sites are dispersed hut settlements; * Greek ‘trading’ posts (Taras, Brindisi, Otranto) and settling colonies (Taras).

One large dispersed site (Oria) – c. 90 ha, two small outlying sites

One large dispersed site (Valesio) – c. 20 ha

Archaic/Classical period (6th – mid 4th BC)

* Tarantino: from late 7th century: emergence of a fully hierarchical settlement system: formation of urban Taras and related rural territory with isolated farmsteads, hamlets and fortified strongholds * Brindisino/Salento: from mid/late 6th century: early urbanization; consolidation of regional settlement patterns; hierarchies now become more pronounced, manifested materially in processes of fortification/nucleation and the emergence of more rational lay-out at a limited number of major settlements; adjacent to these: emergence of tribal sanctuaries

One large, nucleated settlement (Oria), 40-50 ha, fortifications around central hill, extra-mural sanctuary, one rural site

One large, more nucleated site (Valesio), c. 20 ha, with nearby sanctuary

Early Hellenistic period (late 4th-3rd BC)

* Tarantino: stabilization of settlement configuration/hierarchy, further intensification of agriculture/expansion of rural settlement * Brindisino/Salento: Expansion, fortification, nucleation and more rational organization of most Archaic/Classical sites (hierarchies expressed in differences in extent of walled areas; between 5 and > 100 ha). Contemporaneously, rural territories are filled in with small isolated farms and hamlets, even in marginal areas; agricultural intensification and demographic growth

More nucleated Oria surrounded by fortifications (c. 120 ha), extra-mural sanctuary; dispersed rural pattern: 46 rural sites, mostly small farms, some hamlets

More nucleated Valesio surrounded by fortifications (c. 80 ha), extra-mural sanctuary; dispersed rural pattern; 40-50 sites, mostly small farms and sheds

Late Hellenistic / Republican period (2nd-1st BC)

* Tarantino: general decline of both urban and rural settlement * Brindisino: Latin colony of Brundisium (founded c. 245 BC) becomes regional central place; Emergence of zonation of declining economic profitability; Many outlying fortified sites are virtually abandoned, save those along the axis formed by the Via Appia leading to Brundisium. Number of rural sites halved in all survey areas. In some areas, however, the extant sites clearly expand; Agricultural specialisation focused on external market (via Brindisi). Amphora production centres emerge at/in vicinity of Brindisi.

Nucleated Oria (extent?); dispersed rural pattern of some 20 larger farms and some hamlets

Largely abandoned hamlet Valesio. Dispersed rural pattern of some 20 small farms

Table 1: 1st Millenium BC settlement trends for the Brindisino region, Apulia.

73

RPC 2000 conference proceedings

for the reverse reason. Moreover, prospectors are generally biased towards a single element of interest to them, such as fortifications, architectural fragments, graves or roman sigillata wares. To cope with these distortions, in the context of the intensive, systematic surveys carried out by the Brindisino project, representative test cases were formulated to verify the incidentally recovered data and to compare them to the survey results. Moreover, in this way discrepancies in site location, related to incidental recording, could be measured and corrected. The well-known biases inherent in survey procedures (visibility, accessibility, agricultural practices, survey crew, collecting and analyzing procedures etc.) will not be discussed during the workshop. As far as possible, they have been accounted for in the survey strategy and in the calibration of data. Moreover, survey results have been frequently tested by drillings and the excavation of extant stratigraphies. Furthermore, in the context of the regional project large-scale excavations have been carried out at a number of sites, where surveys had been conducted (Valesio, Muro Tenente, San Pancrazio). They provide the necessary refinement of insight into chronological, functional and spatial aspects.

munities. In particular, the focus is on the role of changing power relations, for our project views competition for internal power positions as a force motivating change in local society. Thus, the development and manifestation of competitive political-economic and ideological strategies is investigated, notably with regard to the organization of settlement and landscape. To give an example, the appearance of an urban lay-out or the building of fortifications in pre-Roman Salento, previously explained in the context of Hellenization as indicating the passive character of culturally inferior indigenous societies, can be read as socially meaningful, and can be explained in terms of internal socio-political mechanisms, resulting in the active participation of indigenous societies in the intensification of Mediterranean exchange networks, and the willed introduction of foreign material culture, concepts and techniques into their communities. Supra-regional explanatory frameworks The focus on endogenous change discussed above does not mean that the processes are studied in regional isolation. As a matter of fact, the example given emphasizes that the preRoman urbanization process in indigenous Salento was closely related to the strengthening of the region’s participation in interregional exchange systems and its integration into wider socio-political, cultural and economic networks. The same holds good for the early-Roman, late-Republican phase, when the region was progressively integrated into an expanding state and market system. It is imperative, therefore, to study the specific trajectory of the Salento region in its interaction with a wider, macro-regional framework. As for the pre-Roman period, such a framework is constituted by all Magna Graecia, including the south-Italian indigenous regions, and by mainland Greece. It is in these contexts that the applicability of theories on colonialism and models such as peer polity and core-periphery can be evaluated, adjusting for differences in time, space and intensity. As for the Roman period, the integration of the region into the Roman orbit begs an evaluation of theories on imperialism, and equally of acculturation and integration models.

Explanation on the regional level A major aim of the project is to study 1st millennium BC processes of centralization and urbanization in our research area. A significant contribution to this research is made by field survey. Whereas for a long time classical archaeologists studying ancient urbanism in Italy focused mainly on the physical characteristics of single sites, the regional scope of the surveys has made it possible to include the wider landscape study and thus to explain regional settlement patterns/hierarchies in the context of the relationship between town, countryside and marginal areas. This leads to questions on the town’s role as economic center, and the spread of market exchange, on the town’s consumer role versus the supply role of the countryside, and on the political and religious relationship between the town and the wider landscape. Secondly, the ability of the surveys to monitor longterm trends in the occupational history of the region, has given us the opportunity of studying such questions from a diachronic perspective, to study Braudelian conjonctures, and to give ample attention to the centralization processes preceding the spread of urbanism. Thirdly, the surveys’ spatial focus opens up the study of endogenous, regional processes as devices for explaining centralization and urbanization. Until recently, the latter had been explained from a macro-regional perspective, as a sign of the diffusive strength of Hellenization and Romanisation, with colonization seen as a prime-mover. In our project, as already stated, the multi-faceted relationship between emergent towns and the local landscape within the study region is a primary research object. Similarly, site (and off-site) data are interpreted in terms of regional economic, agricultural and demographic cycles in relation to specific environmental factors. Moreover, attention is devoted to internal social dynamics in the local com-

Selected publications Alberda, K. van, G.-J. Burgers, H. Burgers, D. Karel & D.G. Yntema 1999, Muro Tenente. Centro messapico nel territorio di Mesagne, Manduria. Attema, P.A.J., G.-J. Burgers, M. Kleibrink & D.G. Yntema 1998, Centralisation, early urbanisation and colonisation in a regional context, dutch excavations and landscape archaeology in central and southern Italy, Saguntum 31: 125-132. Attema, P.A.J., G.-J. Burgers, M. Kleibrink & D.G. Yntema 1998, Case studies in indigenous developments in early Italian centralization and urbanization, a Dutch perspective, Journal of European Archaeology 1 (3): 326-381. Boersma, J.S. 1990, Oria and Valesio. Dutch Archaeological Investigations in the Brindisi Region of Southern Italy, Mededelingen KNAW, afd. Letterkunde (new series) 53 (3).

74

Comparative Settlement Archaeology

Boersma, J.S. 1995, Mutatio Valentia. The Late Roman Baths at Valesio, Salento, Amsterdam: Thesis Publishers. Boersma, J.S. & D.G. Yntema 1987, Valesio. History of an Apulian Settlement from the Iron Age to the Late-Roman Period, Fasano di Puglia. Burgers, G.-J. 1994, The Salento Isthmus Project. Second Interim Report, Bulletin Antieke Beschaving 69: 145-154. Burgers, G.-J. 1998, Constructing Messapian Landscapes: Settlement Dynamics, Social Organization and Culture Contact in the Margins of Graeco-Roman Italy, Amsterdam: Gieben.

Yntema, D.G. 1991, The Birth of a Town: The Settlement of Oria in Southern Italy from the Iron Age into the Hellenistic Period, Mededelingen Rome 50: 102-114. Yntema, D.G. 1993, In Search of an Ancient Countryside. The Free University Field Survey at Oria, Province of Brindisi, South Italy (1981-1983), Amsterdam: Thesis Publishers Yntema, D.G. 1995, Romanization in the Brindisino, Southern Italy: A Preliminary Report, Bulletin Antieke Beschaving 70: 53-77.

75

RPC 2000 conference proceedings

3.3.2 The Metapontino and Morgantina Archaeological Survey Projects (Basilicata and Sicily)

material affinities with the so-called “Ausonian” assemblages first documented by L. Bernabò Brea on Lipari and believed to reflect the late prehistoric migration(s) of Italic populations into Sicily from the Italian peninsula. In eastern Sicily, Morgantina is one of a series of Ausonian sites that, with the exception of Milazzo in the northeast, are located around the margins of the Catania Plain (Paternò, Lentini, Mulino della Badia/Grammichele). No material recovered during survey could be unambiguously assigned to the Late Bronze Age. This may, however, be due to difficulties in recognizing the largely undecorated wares characteristic of the period. On the other hand, survey evidence combined with the accumulated archaeological record of eastern Sicily suggests that the region, along with much of the Catania Plain, was largely abandoned during the Middle and Recent Bronze Ages (c. 1500 – 1000 BC), when population appears to have been heavily concentrated within the southeastern corner of the island. Ausonian settlement beginning in the Final Bronze Age, then, appears to have been moving into a largely unoccupied area and population levels may have been quite low early in this process.

Steve Thompson 3.3.2.1 Settlement trends in east-central Sicily My work here was centered on the site of Morgantina, a large hilltop settlement occupied continuously for the entire course of the 1st millennium BC and the focus of extensive American excavations (Princeton U., U. Illinois, U. Virginia, Wesleyan U.) since 1955. I intensively surveyed 40 km2 systematically distributed throughout a 150 km2 study universe centered on Morgantina. Field Methods Survey units consisted of 0.5 x 2 km transects systematically spaced at staggered 2 km intervals across a 150 km2 region. The long axis of survey transects intentionally crosscut the prevailing east-west grain of local topography to ensure that all topographic zones were covered by survey. Transects were subdivided into “tracts”, typically defined by modern field boundaries and characterized by homogeneous surface visibility, land use, and topography. Standardized forms were completed for each tract surveyed, including walker interval, visibility, land use, topographic setting, and a subjective evaluation of off-site surface artefact density. Routine survey typically was at 15 meter intervals. All potential chronologically diagnostic artefacts, all feature shards (rims, bases, handles), all prehistoric pottery, and all lithics were collected during routine tract survey and bagged as a group according to tract number. Sites were defined on the basis of comparatively higher surface artefact density or by the presence of anomalous artefact types and/or features. Site mapping often incorporated a method of artefact density mapping in which counts of surface artefacts (typically total counts but sometimes broken down into separate classes) were recorded at 5 – 10 m intervals along walked lines spaced 5 – 10 m apart. On-site artefact collection relied on close interval walking (walker interval recorded on site forms) and focused on all potentially diagnostic artefacts, feature shards, and a subjectively determined sample of representative or unusual fabrics. Site maps showing the location and boundaries of sites and distance to nearby landmarks were made for each site. All sites and tracts surveyed were plotted (using compass, pacing, and measuring tapes) on 1:10,000 field maps photo-enlarged from 1:25,000 IGM maps. Tract and site boundaries were subsequently digitized (using Arc/Info) with line and polygon attributes coded for a range of attributes. Artefact databases were joined to the GIS using provenience ID. The GIS then determined site and tract sizes.

Final Bronze/Iron Age (900 – 600 BC) Excavations at Morgantina have not been able to determine the size of the settlement during this period, though it appears to have covered an area of ca. 4 – 5 hectares and to have consisted of sub-rectangular dwellings sunk into the slopes of the hill and constructed of wattle, daub, and thatch. No organized site plan is known, and the structure of the settlement appears to have consisted of scattered dwellings each oriented in accordance with local topography. Slight indications of clusters of dwellings across the site may reflect excavation biases, topographical constraints, or social organization. A total of 67 fragments of Final Bronze/Iron Age pottery were discovered during survey. Most of this consists of matt-painted wares. No imported Greek pottery of the late eighth-seventh centuries was collected during survey. Final Bronze/Iron Age material was collected from 16 discrete find spots. All but 4 of these find spots were defined as sites due to the density of surface material encountered, though most of these sites also contained remains of other periods. In the other 4 cases, Final Bronze/Iron Age pottery was collected during routine tract walking and these have been designated as off-site finds. Eleven (68%) of these find spots, including all 4 off-site locations, yielded only one or two diagnostic shards. Finds of 4 and 5 Final Bronze/Iron Age shards were made at two locations, while the other three find spots all contained between 10 and 20 diagnostic shards. Three of the 16 find spots are located within 2 km of the contemporaneous Morgantina settlement and these may be near access routes to the site. The remaining 13 find spots all occur 5 or more km from Morgantina and are clustered into 3 distinct groups. These 3 clusters of Final Bronze/Iron Age material are each associated with high elevations and overlook river valleys not visible from Morgantina. It is difficult to establish site areas accurately for

Late Bronze Age (ca. 1250 – 800 BC) Excavated evidence of settlement at Morgantina begins ca. 1000 BC. The settlement at Morgantina shows strong 76

Comparative Settlement Archaeology

these find spots, but most appear to be relatively small (< 0.5 ha.). The density of finds at three of these sites may indicate permanent occupation, while the remainder seem indicative of more ephemeral activity. I interpret these remains as evidence of the emergence of, minimally, a second tier of settlement in the local settlement hierarchy centered on Morgantina. The locations of these sites typically 5 or more km from Morgantina suggests an expansion of production into more distant areas, thus an expansion in the overall catchment zone exploited from Morgantina. Given the high elevations of these sites and that they typically overlook land not directly visible from Morgantina, it may be that these sites also served surveillance functions.

the heightened availability of iron farming implements needed to turn these comparatively heavy soils. I argue that the intensification of regional production during the Archaic-Early Classical period was driven by a desire to raise levels of surplus production and that this desire can only be understood in terms of local politico-economic systems. Late Classical - early Hellenistic period (400 – 200 BC) This period represents a high point both at Morgantina and in the region at large. The central settlement at Morgantina expanded to cover an (walled) area of more than 60 ha. and the site’s agora was the focus of many monumental constructions. During survey, Late Classical-Early Hellenistic material was collected from 56 locales, including 47 sites and 9 tracts. Site sizes are broadly similar to those recorded for the preceding period. On-site diagnostic artefact densities show a weakly bimodal curve, with a greater percentage of higher density sites than during the preceding period. Find spots are widely scattered throughout the entire study universe. With one exception, again a site with a ritual function, higher density sites with larger surface areas are found only at distances of 5 or more km from Morgantina. One of these sites yielded the only cut stone (olive/grape?) press bed found during survey, suggesting that these large outlying sites may have played a role in the specialized processing of agricultural products. Sites with areas less than 0.9 ha. and with medium to high artefact densities (>20/ha) seem best interpreted as farmsteads. More than 80% of all sites in this class are located within a band 2-5 km distant from Morgantina. Land within 2 km from Morgantina, it would seem, typically was not occupied by permanent residences but was probably farmed by groups residing in the city. The greater number and density of sites interpreted as single-family farms over that of the preceding period may be indicative of intensified rural production. It may also be the case (based on some literary sources) that during the 4th-3rd centuries Morgantina and its territory was formally colonized by Greek settlers and, therefore, the regional settlement pattern recovered by survey reflects an inherently Greek (as opposed to native Sicilian) one.

Archaic - early Classical period (600 – 400 BC) Artefacts of this period were recovered from 52 locations, consisting of 34 sites and 18 tracts. Roughly 2/3 of material dated to this period belongs to the local Siculo-Geometric matt-painted ceramic tradition. Off-site locations typically yielded only one or two artefacts. Archaic-Early Classical sites typically range (over a very smooth curve) in size from 0.15 ha to 1.5 ha. Following corrections for variations in surface visibility and sampling fraction, ArchaicEarly Classical site find spots also can be compared in terms of diagnostic artefact density. The resultant curve is clearly bimodal, centered upon a low-density group ( 5 times per 10 years) Regular floods (every year)

It is obvious that even with a well-documented geological record, classification of areas in especially classes F1 or F2 will be impossible to make in a prehistoric context; apart from that, the degree of importance of the periodicity of floods may have been different in the past. Much more 201

RPC 2000 conference proceedings

a

Natural irrigation by inundation (regadío) of areas close to the river bed of the Rio Aguas, available to all (pre-) historic societies; b Irrigation of the rest of the floodplain by means of canals, probably introduced in the Roman Imperial period; and c Irrigation of the lower hill slopes by means of a combination of water conduits (known as acequias) and terracing, introduced in the Andalusian period.

have been applied in order to obtain the higher production levels needed. During the Andalusian period, a complex terrace and hydraulic system was introduced, in order to be able to settle mountain slopes previously unavailable for settlement. The system seems to have worked satisfactorily during the later Arab period as well, allowing for sufficient food production for the settlements as well as surplus production of olives and (silk production-related) mulberry trees, and at the same time reducing environmental degradation. It is interesting to observe that after the Spanish reconquista the hydraulic infrastructure was abandoned because of ‘insufficient productivity’ according to contemporary sources. This points to the fact again that the use of technological developments is highly dependent on social and economic circumstances.

As a first step, a basic land evaluation of the area was carried out for each of the three irrigation types. The distribution of land use in 1978 (when agriculture in the area was not yet fully modernized) was used to determine where each of the three types of irrigation was to be found in the recent past. The results of this analysis were extrapolated to the whole area, as not all of it was used for agriculture in recent times. Not surprisingly, the most probable areas for natural irrigation were found close to the riverbed of the Rio Aguas, whereas areas suitable for irrigation of type b) and c) were found at increasingly large distances from the river. The suitability maps were then combined with data on settlement distribution and estimates of demographic development through (pre-) history in order to model hypothetical agricultural production zones for each period under a hypothesis of self-sufficiency. The models produced were used to judge whether potential food production problems might have occurred, and if irrigation could have been applied to counteract these problems. From archaeological evidence it is clear that during the Neolithic and Chalcolithic no artificial irrigation was applied; the modeled pattern of cultivation zones conformed to this evidence, as settlements are found close to water resources and needing relatively modest amounts of land for crop cultivation. However, a sharp population increase during the Late Bronze Age (the Argaric period) meant that much larger areas were needed for cultivation. The demographic rise was coupled to a period of drier and hotter climatic conditions, and these combined factors evidently led to agricultural production problems. Argaric society seems to have tried to counteract the problem by introducing a monoculture of barley, which is a highly drought resistant crop, and could therefore be grown in areas not well suited for other crops. However, this strategy seems to have led to nutritional problems, large-scale deforestation in the river valley and possibly the eventual collapse of Argaric society when it could no longer cope with the environmental problems. It is not clear whether artificial irrigation was technologically speaking an option to the Argaric people, but it is certainly a strategy that for some reason was not pursued. A gradual but steady population decline ensued. It was not until the Roman Imperial period that the area was again used for extensive agriculture. The modeling showed that the most suitable soils should almost all have been taken into cultivation in order to feed the population. However, it is well known that the Roman Empire furthered surplus food production. This implies that artificial irrigation must

6.6.4 The human perception of suitability Farmers in the Atlantic zone of Costa Rica, when asked if they could distinguish between good and bad soils, answered that one could easily recognize the good, black soils, suitable for almost any crop and the bad, red soils that were only suitable for grazing. In essence, they applied a very simple form of land evaluation; formalized land evaluation methods are basically just tools to express the modern perception of land suitability in an unambiguous way. In the past, farmers will have judged the quality of land by using their own experience of what was good or bad for a certain type of land use. These judgments may have been different from our judgements on land quality, for different factors may have been considered. These include; whether surplus production was aimed at or not, what the level of labor input and technological investment needed, and whether population centers needed to be close to the zones of food production. In addition, their understanding of crop requirements may have been different from ours. The Roman perception of land suitability A glimpse of the way people thought about land suitability in the past can be obtained from Roman agronomical texts. Favory et al. (1995) have analyzed the texts of Columella and Pliny the Elder in order to see if the Romans’ view of agricultural suitability was very different from modern views. One thing immediately evident is the emphasis placed on light, easily workable soils. Columella distinguishes four types of soil: • Rich and light soils, easy to work, giving the highest return with the least effort • Rich and compact soils, that require hard labor but compensate this with good returns • Humid soils, that do not need much labor, and can still produce a crop • Dry, compact and poor soils, that are difficult to work and do not yield a good return This very basic description formed the basis of the Roman land taxation system, as has been demonstrated for the Tricastin area in the Middle Rhône Valley (Berger et al. 1997, Van der Leeuw 1998). 202

Potential Land Evaluation in Archaeology

Pliny the Elder, in his Natural History, goes far beyond Columella’s soil classification. In his encyclopedic text, he describes what the requirements for many crops are, and the characteristics of many soil types, and the best crops to plant in them. The text does have errors and omissions in it, but it proved possible to compare the soil descriptions provided by Pliny with the categories distinguished on modern soil maps. And so it is possible to perform a land evaluation more or less as Pliny would have done it, had he soil maps available. One of the examples of the way Pliny described soils is given here, translated from Favory et al. (1995):

Asparagus Requirements according to modern agronomers: Needs a sandy soil, cool but well drained, neither humid nor calcareous. Resistant to winter frost, but sensitive to spring frost. Can be cultivated on almost any location, except on high ground. Needs deep plowing and manuring. Requirements according to Pliny: Sow it in a humid or deep soil that has been dug up. Saturate it with manure. Weed often. The best-suited soils for the cultivation of asparagus are the soils of the gardens of Ravenna.

Humidus Ager; Locum humidiore; Humidum solum

The Roman and modern views agree on the need for a deep soil that is manured, but Pliny insists on a humid soil, whereas modern agronomers do not. Apart from that, the number of criteria used by modern agronomers is much larger and includes requirements of climate, chemical composition and texture of the soil (note however that Pliny’s definition of humid soils probably already implies a sandy texture). A land-evaluation for asparagus based on modern criteria would probably restrict the area suited to asparagus cultivation, and might classify as suitable soil types different from those evaluated on Pliny’s criteria.

Translated as Most probable soil type Effect on crops Crops

humid soil recent alluvial soils, sandy and humid but not hydromorphic almond trees become sterile or even die emmer, asparagus, olives, elm trees; almond trees and alfalfa to be avoided Advice dig up the soil for cultivation of asparagus; market garden culture Hydrology close to water To be recognized by vicinity of water; vegetation type Topography plains Vegetation grassland Comparable soils rich soils (not suited for almonds), deep soils (asparagus), dry soils (elm trees), grassland soils (not suited for alfalfa), very or slightly humid soils, clayey soils (olives) Opposed soils dry soils on hillsides (different cultivation for elm trees), irrigated dry soils (well suited for alfalfa), warm and hard clay soils (well suited for almond trees), dry clayey soils (different planting season for olives)

6.6.5 Conclusions Land evaluation can play a role in the study of (pre-) historic societies as a method of establishing the amount of land available for subsistence production. However, from the examples given in this paper it will be clear that a modern land evaluation appraisal would not be suitable for use in archaeology. Even though the basics of crop requirements and soil characteristics are the same now as in the past, environmental and human factors may radically change the outcome of an archaeological land evaluation. The result of such a land evaluation should be to reconstruct the perception of (pre-) historic societies with regard to land suitability, and not that of we moderns. The interplay of environmental factors, technological development and social and economic structures, determines which areas prehistoric societies will have regarded as suitable for crop cultivation (or for other uses). It may be impossible to recover exactly what was in the minds of (pre-) historic peoples when they assessed the suitability of soils, but it is to be hoped that the examples given here give an indication at least of how to pursue land evaluation in an archaeological context.

From the comparison of Pliny’s soil descriptions with modern agronomic texts it can be concluded that the Romans were not overly concerned with the concept of soil fertility. Nowadays, fertile soils are associated with the profitable cultivation of fruit trees, but the Romans thought fertile soils were those suited for the cultivation of the traditional crops of wheat, olives and vines, which formed the basis of the Roman subsistence economy. On the other hand, the Romans placed a much stronger emphasis than modern agronomers on the workability of soils, not surprising in an era when manual labor was still very important in order to grow any kind of crop successfully. Crop requirements as recognized by Pliny may also be a little bit different from our modern requirements, as is illustrated by this example:

References Berger, J.-F., F. Favory, T. Odiot & M.-P. Zannier 1997, ‘Pédologie et agrologie antique dans le Tricastin central (DrômeVaucluse), d’après les textes agronomiques et épigraphiques latins et les données géoarchéologiques’ in: Bravard, J.P., G. Chouquer & J. Burnouf (eds), La dynamique des paysages protohistoriques, antiques, médiévaux et modernes. XVIIe Rencontres Internationales d’Archéologie et d’Histoire d’Antibes. Editions APCDA, Sophia-Antipolis: 127-154.

203

RPC 2000 conference proceedings

du Rhône dans l’Antiquité (part II). University of Cambridge, Cambridge: 73-114. Finke, P., J. Hardink, J. Sevink, R. Sewuster & S. Stoddart 1994, ‘The dissection of a Bronze and Early Iron Age landscape’ in Malone, C. & S. Stoddart (eds) Territory, Time and State. The archaeological development of the Gubbio Basin. Cambridge University Press, Cambridge. Kamermans, H. 1993, Archeologie en landevaluatie in de Agro Pontino (Lazio, Italië). Universiteit van Amsterdam. Amsterdam. PhD thesis. Kamermans, H. 2000, ‘Land evaluation as predictive modelling: a deductive approach’ in Lock, G. (ed), Beyond the Map. Archaeology and Spatial Technologies. NATO Science Series, Series A: Life Sciences, vol. 321. IOS Press / Ohmsha, Amsterdam: 124-146. Leeuw, S.E. van der (ed), 1998, The Archaeomedes Project: Understanding the natural and anthropogenic causes of soil degradation and desertification in the Mediterranean Basin. Research Results. Office for Official Publications of the European Communities, Luxemburg. Verhagen, P., S. Gili, R. Micó & R. Risch 1999, ‘Modelling prehistoric land use distribution in the Rio Aguas valley (province of Almería, S.E. Spain)’, in Dingwall, L., S. Exon, V. Gaffney, S. Laflin & M. van Leusen (eds), Archaeology in the age of the Internet – CAA97. Computer Applications and Quantitative Methods in Archaeology 25th Anniversary Conference, University of Birmingham. British Archaeological Reports, International Series 750. BAR Publishing, Oxford. Download.

Boerma, J.A.K. 1989, ‘Land evaluation in prehistoric perspective: some observations’ in: Haex, O.C.M., H.H. Curvers & P.M.M.G. Akkermans (eds), To the Euphrates and beyond. Rotterdam: 17-28. Castro, P.V., R.W. Chapman, S. Gili, V. Lull, R. Micò, S. Montón, C. Rihuete, R. Risch & M.E. Sanahuja-Yll 1998, Aguas Project: Palaeoclimatic reconstruction and the dynamics of human settlement and land-use in the area of the middle Aguas (Almería), in the south-east of the Iberian Peninsula. Research results. Office for Official Publications of the European Communities, Luxemburg. Chadwick, A.J. 1978, ‘A computer simulation of Mycenaean settlement’, in: Hodder, I. (ed), Simulation studies in archaeology. Cambridge University Press, Cambridge: 47-57. Dalla Bona, L. 1994, Ontario Ministry of Natural Resources Archaeological Predictive Modelling Project. Center for Archaeological Resource Prediction, Lakehead University, Thunder Bay (Ontario). Doorn, P.K. 1993, ‘Geographical Location and Interaction Models and the Reconstruction of Historical Settlement and Communication: The Example of Aetolia, Central Greece.’ Historical Social Research 18: 22-35. FAO 1976, A Framework for Land Evaluation. ILRI Publication 22/FAO Soils Bulletin 32, Rome. Favory, F., J.-J. Girardot & M.-P. Zannier 1995, ‘La perception des sols et des plantes chez les agronomes romains’ in Van der Leeuw, S.E. (ed), The Archaeomedes Project: Understanding the natural and anthropogenic causes of soil degradation and desertification in the Mediterranean Basin. Volume 3: Dégradation et impact humain dans la moyenne et basse vallée

204

Potential Land Evaluation in Archaeology

6.7

Summary of the Discussions

6.7.1

Reconstructed landscapes, land qualities and land characteristics

ments are considered, importance is attributed especially to land characteristics, such as texture and soil depth. Soil type is nowhere emphasized (Van Joolen). We must keep in mind that it also depends on the aims of the classification system, Delvigne remarks. When using the limiting factor system, land characteristics such as presence and depth of water table, bedrock at the surface, and texture, are basic items to be looked upon.

Abandoned fields; erosion and/or degradation Verhagen comments that from the Archaeomedes research in southeast Spain it was evident that terraces have been used for a very long time; the terraces were in use as water collecting structures, with minimal erosion hazard. It was a very efficient system for small-scale agriculture during the Andalusian period. The abandonment of agriculture on terraces after the reconquista caused serious erosion in the study area. This set off severe bad-land formation, inhibiting the regeneration of woodland. However, vegetation cover can regenerate very quickly under more favorable circumstances, thereby reducing the fields’ vulnerability to erosion. Attema adds that, if terraces are constructed for olive groves (as already has been proven by pollen analysis in the Pontine region) and afterwards left abandoned, erosion can take place. But, as Verhagen explains, olive groves in the past usually had undergrowth, decreasing the erosion rate. According to Farshad, when a field is left abandoned, vegetation usually comes back and reduces erosion. But a field can degrade when no one takes care for it after abandonment. Erosion also depends on soil type, and position in the landscape. Nevertheless abandoned fields do not experience severe erosion (by water), because of regenerated vegetation. Veenman introduces a very interesting aspect from her recent research on the degradation of vegetation in the Mediterranean. She concludes that present-day agricultural fields are never fully abandoned. Animals (goats, cows) always have their effect on the vegetation cover. Farshad remarks that it is difficult to formulate a general statement about erosion rate. A lot of detailed information is needed.

Reconstructed landscapes and their potential land use maps In his lecture, Verhagen presents a method for modeling the extent of agricultural territories to create potential land use maps in an area in southeast Spain. The main aim of the research was to integrate archaeological data, especially on the people’s diet, and the way they cultivated their crops, into a model of the possible configuration of the agricultural fields in the area, which has been continuously occupied from the Neolithic period until now. According to the geological situation, four categories of farming system can be adopted: • Agriculture on the very fertile floodplains • Dry farming, which can be divided into – a system with irrigation by means of simple canals diverted from the streams, and – a system with hydrological structures and terraces, that enables people to cultivate at higher elevations • Dry farming, with very low productivity To find out where these farming systems might have been applied, a lot of information had to be digitized, such as the geology, the elevation, and the distance to the streams in the area. The information thus acquired was compared to the modern land use map. The result was a sort of potential land use map. With the map of land potential, and the maps of the discovered settlements, we tried to model the amount of agricultural land needed for each settlement. Next, a relative attractivity index for the different types of agriculture was calculated for each pixel on the map. Human diet was mainly determined on the basis of macro remains, and with complex calculations on e.g.: how much of these crops were needed to feed the people. The yields were based on production figures from Spanish agronomists. Probably, they used modern, pre-industrial yield figures, both for dry farming and irrigated farming. In the research, a few assumptions had to be made; among them, that each settlement had tried to be self-sufficient, and that each settlement had free access to the surrounding lands. For each historical period, Verhagen shows the potential land use maps.

Land characteristics; importance of texture versus soils in land evaluation According to Farshad, soil development is dynamic, but mostly, it is not to be measured over a hundred years or so. This is why soils are often considered a stable element. But it may also be that in some cases soils develop in 500 years, or even less (for example man-made top soils, plaggen soils, and saline or alkaline soils in arid regions, as a result of overirrigation). So, again, it is difficult to make a general statement. Although a lot of detailed information is needed to come to soundly based conclusions, often a few indicators can be employed and cross-checked to achieve this result. Fertility (associated with soil type) can change a great deal in a few thousand years. This is also true for soil structure. Considering that many factors are involved in soil formation (the Jenny equation), one cannot use soil or any soil-related property as a fixed item (parameter). So when decisions are made about using soils or texture and/or structure in archaeological land evaluation, they are specific to the location and cannot be taken as a general rule. Delvigne remarks, that a farmer is usually more interested in texture than in soil type. And when crop require-

6.7.2 Land utilization types (LUTs) Technology in Italy: from Bronze Age till Roman Age (This paragraph comprises the two issues mentioned in section 6.3.3.2) During the workshop, professor Forni first 205

RPC 2000 conference proceedings

Discussion What information do we have in the context of historic land evaluation? Farshad: it can be put in the framework of workability (in the author’s research, probably the most important land quality). Workability comprises slope class, surface stoniness, rockiness, texture (clay type, loose or firm), and consistence (loose, friable, hard or firm).

gave a detailed lecture about the evolution of agriculture and ploughs in Italy from the Neolithic till the Roman Times. This historic evidence is made up of various discoveries, such as real tools, inscriptions in caves or images on situlae, votive models, real iron ploughshares, and reliefs on funeral ash urns. Early indications of agriculture, together with preparation of the fields are fossil furrows date back to the Eneolithic in, for instance, the Aosta valley. Forni illustrated these discoveries in full detail, assisted by several slides. Most of the important items of the lecture are presented in Forni’s written contribution. Professor Forni summarizes that “one of the great inventions in the agricultural history of Italy is, of course, the plough. During the Neolithic and the Eneolithic (periods starting about 5000 BC), the wooden plough was introduced for light soils. To cultivate the clayey Pleistocene soils, this plough was used as a complementary tool alongside, for example, the zappa (hoe), and slash-and-burn techniques. In the Iron Age (epoch that started around 1000 BC), the plough with the iron ploughshare was introduced. Greek influence seems to have been very decisive, for example, on mathematical thinking, and land division. The Iron Age can be seen as a revolutionary period in agriculture. Until the industrial revolution, the light plough was commonly used in Italy and the rest of the Mediterranean. The heavy plough was very much located in the northern part of Italy.”

Despite general opinion in the literature, there is no direct relation between heavy soils and plough type. In fact, it is a question of labor and time. For example, in the period before the Iron Age, the fields had to be treated by slash and burn methods first to loosen the soil. Next other tools were used to break the soil, which was heavy and clayey. Only after that the ground could be ploughed with a wooden plowshare. However, with the introduction of the iron plowshare it wasn’t necessary anymore to treat the ground before plowing. So, working with heavy ploughs or iron plowshares, doubled or tripled the amount of land that could be cultivated, compared with that of the Bronze Age. Farshad: we must realize that in ancient times population pressure was not as high as today. This is a question of different scales. To feed the people, not all the land was needed. So infertile fields or those marginally suitable for agriculture were left abandoned. Only when population pressure increased, were these fields taken into use.

Horses in agriculture in ancient Italy Around 700 or 800 BC, agriculture faced a transition from oxen to horses. The advantage of horses is that they were faster and could be steered more easily (Forni). According to German and French researchers, horses were only introduced into Mediterranean agriculture in the Middle Ages, but according to Forni there is evidence that they were used already during the time indicated. Attema adds that there is also evidence from aristocratic graves of this time, that horses pulled carts.

6.7.3

Classification possibilities

Perception of ancient people on agrarian suitability of fields Verhagen clearly illustrates this issue in his lecture and the contribution above on the results of the Archaeomedes Project. One of the aims of the project was to compare present-day views of the suitability of agricultural areas on modern soil maps, with the perception of Roman people on soil suitability. Verhagen: perception of the environment is influenced by the way we think we can make use of it. The Romans were as aware as we are of the agricultural potential of different soil types. But, for the French researchers, it turned out to be very difficult to determine which criteria the Romans used. Fortunately, two Roman agronomers provide us with important agricultural information: Columella presents a very simple scheme of land suitability, and Pliny the elder gives a lot of information on the soil requirements of crops. One factor of significance which can be deduced from their writings, is that the amount of effort (manpower) needed to develop a soil, was regarded as much more important than soil fertility (this in contrast to modern insights on suitability); another factor is that the olive and the vine had to be the most profitable so they had to be grown on the best ground.

Forni distinguishes two main types of light plough A type of plough of which the power was concentrated at one point. The implement was especially constructed for terrain which had not been cultivated before. This kind of plough could be lifted up easily, when the farmer hit a stone in the field. But it asked more power from the person behind it and was hard to steer. The power of the second type of plough is more horizontally apportioned; it was used in terrain which had previously been cultivated. It was easier to work, and needed less power. On the basis of inscriptions it seems that the first was used in the early periods but there is no statistical evidence for it. It is also possible that both ploughs existed contemporaneously for a very long time. However, in Roman Times, the second type of plough prevailed. After about AD 200, the heavy plough was introduced in the north of Italy only. The advantage of the heavy plough with its cart was that it could plough the earth much faster than before.

Discussion Verhagen: in contrast to modern agricultural opinion, ancient farmers probably did not consider fertility the most important factor in the determination of soil suitability. In this context, workability is mentioned much more in 206

Potential Land Evaluation in Archaeology

Agrarian technology in ancient Italy In prehistorical land evaluation, different kinds of farmingsystem such as small, self-subsistence farms, large estates producing for sale, dry and irrigated farming, and slash and burn systems in mountainous areas (Spurr 1986) as well as pastoralism itself must be examined. As becomes clear from Forni’s contribution, agricultural technology developed significantly over a very long time in Italy. But, within the scope of archaeological land evaluation, it is an important conclusion that in all periods people were able to cultivate all kinds of soils. But to cultivate a clayey soil in, for example, the Bronze Age probably required considerably more manpower and time than in later epochs, when farmers could work with iron plowshares. So, when applying the ‘least effort theory’, suitable soil availability did increase from the Bronze Age up to Roman Times, because preparing a field for cultivation required less effort, and the use of an iron plowshare saved time. Technology also comprises knowledge of soil improvement techniques in terms of fertility, irrigation, drainage and reclamation. This knowledge increased through time, and of course continuously improved soil suitability.

the literature next to land characteristics such as water content, depth of soil, light and heavy soil, and structure. Sometimes color is mentioned, merely as a diagnostic indicator of a specific kind of soil. Forni adds that there is evidence that the Romans had the knowledge to analyze soils. They even performed experiments with the physical and chemical properties of soils. Attema remarks that the information given above is especially applicable to (large) villa cultures. Other types of farming also existed, such as small-subsistence farming. Settlement patterns have to be analyzed first, before decisions can be made about the socio-economic situation in a certain region, with corresponding needs.

6.8

Conclusions

As Verhagen rightly notices in his contribution, land evaluation in archaeology is quite a new procedure, and only a few of the problems encountered have received proper attention. Land evaluation can be carried out in different ways with different inputs and related outputs. Van Joolen uses the method to create land-use maps, which indicate the potential suitability of an area to different kinds of agricultural system in different historical periods. However, Farshad applied the potential land evaluation method rather differently, as he explains in his contribution. In the research, he uses a parallel approach by comparing more directly parts of reconstructed landscapes with actual, real archaeological findings. In the contribution below, only those issues are mentioned which have been topics at the workshop, followed by the simplified model for Italian historical potential land evaluation.

6.8.1

Workability and perception of suitability In the first place, we must imagine ourselves in the position of ancient farmers. As becomes clear from the contribution of Verhagen, workability must be one of the most important limiting factors (or the highest in factor rating) in a prehistorical land suitability classification. So, workability is a limiting factor, but it changes through time, depending on the improvement of agricultural tools (for example from wooden to iron plowshares), the management requirements of the crops planted (cereals have other requirements than have olives), and available time and manpower. Important characteristics affecting workability are: slope percentage, stoniness and rockiness at the surface, and soil texture (Dent & Young 1981, Kamermans 1993, Farshad 1997).

Input requirements for a historical potential land evaluation

Landscape dynamics As became clear from the discussion on the dynamics of landscapes, no strict rules can be provided. The history of the surface erosion of abandoned fields depends on such factors as climatic change, soil type, geographical position, the type of crops cultivated, intercropping (for example olive trees with an undergrowth of cereals), the presence of animals in the fields, the rate of vegetation regeneration, and field management (manuring, terracing). In every case, all factors have to be examined before drawing conclusions on erosion, deposition and/or degradation. The argument on the importance of soil texture as against soil type in land evaluation generates divergent opinions. One view states that soils are stable land characteristics, and present-day soil types can be used (in most cases) as a reference for ancient soil types in potential land evaluation for archaeology. On the other hand, fertility can change rapidly and in that case so would soil type. Also, ancient farmers probably were more interested in soil texture than are we moderns.

Despite the problems and uncertainties in using land evaluation for archaeological purposes, it is the author’s opinion that the method can be useful in creating hypothetical land-use maps. But the conditions stated below must be fulfilled: • Landscape reconstruction must receive proper attention. Of course, depending on its aims, the scale and intensity of the research must be adjusted, • Investigation into man’s perception of agrarian suitability in the past, • Thorough examination of every (reconstructed) land utilization type, including management and soil requirements of ancient crops (LURs), the field preparation capabilities of farming equipment, knowledge of ancient hydraulic systems and the application of natural fertilizers, • Reconstruction of past climates, especially identification of periods with increasing or decreasing precipita207

RPC 2000 conference proceedings

6.8.2

A simplified model for Italian land evaluation in archaeology In the author’s thesis (in prep.), a more detailed model will be published. The scheme below (tables 1 – 5) shows a summary of the agreed important items, which must be included in a potential land evaluation for archaeology. It is specifically meant as an addition to the FAO framework for land evaluation (1976).

tion and relatively hot and cold (colder) periods. These factors influence, for instance, vegetation growth capabilities, and erosion hazards, • All potential land-use maps must be compared with independent archaeological information.

Important issues

Related constitutive components

Landscape reconstruction

Assessment of areas susceptible to erosion or deposition Assessment of areas and soils, which may have been degraded Dating of soil layers: radiocarbon dating, dating with archaeological discoveries Assessment of present-day human influence on the landscape Reconstruction of ancient waterways Pollen analysis; dry or wet periods, cold or warm periods Workability (must get high priority, next to the other qualities mentioned in Dent & Young 1981) Relief, slope class, soil depth, soil texture, (clay type) stoniness and rockiness at the surface and in the topsoil, consistency of the soil, soil classification

Climate reconstruction Land qualities Land characteristics

Table 1: Investigating the research area

Important issues

Related constitutive components

Types of LUTs; landuse requirements Crop requirements

Research in literature: Columella and Pliny the elder

Technology

Modern research on the cultivation of genetically unchanged ancient crops Research in literature: Columella and Pliny the elder Development of agricultural tools Development of techniques: irrigation, drainage, terracing Development of knowledge of fertility: manuring According to Forni: seven periods in the history of Italian agricultural technology: I Period of pre-agricultural tools II Period of pre-ploughs (6000-4000 BC) III Period of the introduction of the plough; animal traction (4000-800 BC) IV Period of the use of agricultural tools made of iron (800-0 BC) V Period of the use of ploughs with wheels and mouldboards (0-1600 AD) VI Introduction of American plants VII Use of the tractor

Table 2: Determination of relevant land utilization types in the past

In this step, the requirements of the LUTs are compared with the capabilities of the land. In the classification procedure, we must remind ourselves that ancient peoples thought differently about land suitability. This difference in perception has to be of great importance in classifying the land. Table 3: Matching and preparation of maps

208

Potential Land Evaluation in Archaeology

Important issues

Related constitutive components

Demographic studies Socio-economic studies Reconstructed infrastructure Distance to ancient waterways

Human diet (evidence from excavations) Settlement dynamics (surveys)

Table 4: Economic and social analysis

Important issues

Related constitutive components

Pollen analysis

Reconstruction of vegetation (regeneration of forests, influence of eating animals) and farming activities (cultivation of wheats, olives) Reconstruction of vegetation and pastoralism Farming activities

Macro remains Excavations

Table 5: Testing the hypothesis

209

RPC 2000 conference proceedings

210

7

Landscape Perception in Archaeology: the Urban and Colonial Experience

pretations that take as their starting point “a minute analysis of all the material related to a given site”. And this is probably where regional geographical analysis and material culture studies can profitably meet. As Marianne Kleibrink shows, the ritual configuration of the landscape and associated artifacts is deeply rooted in the long-term history of the Italian landscape. In her contribution on the sacred landscape of the Sibaritide (section 7.2) she goes as far back as the Neolithic and early to middle bronze ages, hypothesising the existence of landscape units that were reserved for “ancestors” buried in caves overlooking these units, and that therefore remained unsettled. A highly significant change would have taken place in this configuration in the late bronze age and iron ages, when the ritual landscape became more closely tied to the settled landscape - as for instance reflected in the establishment of cemeteries near the settlements. To test the cognitive spatial models outlined in these papers on ritual landscapes, systematic landscape archaeology can be of great help. In the case of the Sibaritide we may ask if the tabù landscape units really existed (and can also be attested elsewhere), or whether the absence of protohistoric settlement debris reflects a bias in the archaeological record. Landscape archaeology has a huge contribution to make in this sense, through fieldwork aimed at uncovering the less visible remains of the protohistorical landscape. It has been observed that, whereas substantial protohistorical sites will not go unnoticed, the thin spread of protohistorical sherds characteristic of a settlement pattern consisting of small sites is extremely hard to detect (see Bintliff this volume section 2.4, with references; see lately for Italy, Attema et al. 2001 and Attema & Van Leusen forthcoming). Likewise, landscape archeology will need to test the spatial models that have been postulated for (proto-) urban societies, as landscapes characterized by an opposition between the cultivated and the uncultivated; a model that can only be tested by means of the combination of intensive survey and ecological research. One complicating factor, however, is that relatively few landscape units outside those that were central to the Classical Landscapes of Antiquity have been subjected to intensive survey. This means that landscape archaeologists need to shift the emphasis from fieldwork in the elevated coastal plains, the hilly countryside and the fertile valleys to less accessible areas such as the higher mountainous zones, river deltas and former marshlands. These are typically landscape units that have remained archaeologically unexplored in many regions, but which may well reveal aspects of society in the geographical and temporal margins of the Classical Landscapes of Antiquity. In a lecture at the occasion of the Fifth Conference of Italian Archaeology, Graeme Barker had already identified the development of “strategies to investigate such landscape units in order to acquire a more complete and certainly more complex picture of settlement and land use dynamics in antiquity” as one of the goals for the landscape archaeology of Italy in the 1990s (Barker 1995). It seems that this observation remains valid for the first decade of the present millennium as well, our current advantage being that new fieldwork

Convenor: Peter Attema 7.1

Introduction

The point of departure of the workshop on landscape perception in archaeology, as stated in the initial conference announcement, is that landscape archaeology should not be limited to the physical reconstruction of the human landscape - the factual description of its settlement patterns through time and the cataloguing of the associated artifacts and ecofacts. It was proposed that, fundamental though this is for our understanding of high level political and economic processes such as urbanization and colonization, landscape archaeology should also attempt to provide scenarios of how groups of people may have perceived (changes in) their environment - groups living both in the “core areas” and in the “periphery” of urbanization, people both arriving as newcomers from outside and already there for generations (as is particularly acute in the analysis of colonial situations). It was observed that landscape archaeology is biased towards obtaining data on those groups that are directly included in these high level processes, while those living on the social or geographical fringes of these processes receive relatively little attention. Our picture of the (formation of) the Classical Landscapes of Antiquity will remain incomplete unless the temporal and geographical fringes, and models of possible interaction between various ethnic and social groups, colonists and indigenous peoples and other possible categories, become subjects of field investigations. For the reconstruction of the human perception of the landscape, geography is particularly important, since we may expect that the forces of urbanization and colonization created formal and/or fluid boundaries in the landscape between the urban, the rural and the marginal spheres; they thereby profoundly altered the pre- and protohistorical conception of the landscape (which likewise needs to be reconstructed). Andrea Zifferero’s article on the geography of the ritual landscape in complex societies (section 7.6) presents us with a number of cases in which formal ritual boundaries are hypothesized around (proto)urban settlements in Central and South Italy. Key concept here is that of the sacred marker, indicating liminal zones between city and country, between cultivated and uncultivated areas, and between political territories including those of the Greek colonies. In her paper on Lucanian sanctuaries and cultural interaction (section 7.4) Helle Horsnaes warns us, however, not to jump to conclusions when it comes to the “ethnic” interpretation of cult places as either indigenous or Greek, and this may severely affect the proposed models in for instance the case of colonial territories. She pleads for inter211

RPC 2000 conference proceedings

between the 6th and 5th century BC can be interpreted as the result of social and political struggles within the various emerging polities, “where the dichotomy between aristocracy and middle classes is reflected in the organisation of the territory”. Cifani’s research demonstrates ways to read social structure and class into regional settlement patterns. The papers in combination thus show the interpretative potential of landscape archaeology when it comes to landscape perception as well as interaction of social groups in the landscape, but they also indicate the need for further fieldwork in the less-well investigated parts of the Italian landscapes.

strategies and methods of analysis have been developed in the mean time to attain his goal. The importance of acquiring new data for landscape reconstruction in all its aspects is also underscored by Jean Paul Vallat in his paper on “Perception des hommes et perception de la terre en Campanie septentrionale à l’époque de la conquête” (section 7.5). Vallat reminds us that the Roman conquest of northern Campania “gave rise to a huge conflict between the Roman and the indigenous peoples, but, at the same time, also between rural settlers and Nature: streams, rivers, forests, and marshes are difficult realities to fight, and Campania Felix is not a constant of agricultural life under Roman rule”. What we need to investigate are precisely the deltas, forests, marshes and mountains and the way these were or were not incorporated in society. An adequate understanding of the human ecology of these areas is crucial, as is stressed by Peregrine Horden and Nicholas Purcell at various places in their recent study of Mediterranean history (Horden & Purcell 2000). The need for detailed knowledge of the new political, economic and ritual configuration that resulted from the (proto-) urbanization and Greek and Roman colonization of the Italian landscape, and of how this might have articulated with concepts of ethnicity and social identity, was a second theme discussed in the workshop. The implications of urbanization as a force in the creation and modification of ethnic and social identities is borne out well in Gabriele Cifani’s paper on the relation between urbanization and ethnic identity in the Middle Tiber valley (section 7.3); Cifani concludes that the changes in the landscapes of the Middle Tiber Valley

References Attema, P., E. van Joolen & M. van Leusen, 2001, A Marginal Landscape, Field work on the beach ridge complex near Fogliano (South Lazio), Palaeohistoria 40/41 (1998/1999): 149-162. Attema, P. & M. van Leusen, forthcoming, Regional Archaeological Patterns in the Sibaritide: Preliminary results of the RPC field survey campaign 2000, Palaeohistoria 42/43 (2000/2001). Barker, G., 1995, Landscape Archaeology in Italy – Goals for the 1990s, in Christie, N. (ed), Settlement and Economy in Italy, 1500 BC to AD 1500, Papers of the Fifth Conference of Italian Archaeology, Oxbow Books, Oxford. Horden, P. & N. Purcell, 2000, The Corrupting Sea, a study of mediterranean history, Blackwell, Oxford.

212

Landscape Perception in Archaeology

7.2

The Sacred Landscape of the Sibaritide: veneration of ancestors, nymphs and deities

caves with separate entrances: Sant’Angelo nr. III has been systematically excavated (Tinè 1964), the neighbouring chambers not; but it is known that in Sant’Angelo nr. II human bones were present. This was also the case with the Grotta dell’Antenato, a collapsed chamber, discovered by Nino Larocca, originally belonging to the central Sant’Angelo system. Although Santo Tinè claims that the caves were domestic, Ruth Whitehouse convincingly argued for a ritual use (Whitehouse 1992:39-40). The Sant’Angelo caves are too small and the pottery too fine for household wares: the Neolithic pottery is finely decorated and from the Bronze Age there are a very large tray and a beautifully decorated bowl as well as fragments of well-burnished drinking cups. In the Grotta Pavolella (or Grotta degli Scheletri), two strata of Neolithic human bones were found. One contained unburned human bones, the other burnt remains of at least 20 individuals, mostly children and young adults, cremated in situ. Children did not produce offspring and for that reason were not worth cultic attention (Pearson 1999:459). Their cremated remains therefore point to offerings. The Pavolella cave continued to be in use in the Copper Age during which dismembered, unburnt, human bones were displayed on its floor and in niches (Whitehouse 1992:69). The special display in the caves of disarticulated human remains (Whitehouse 1992:68ff.), cremated youngsters or children, green axes (Leighton & Dixon 1992; Whitehouse 1992:76ff.) and fine pottery (for the neolithic Malone 1985; Whitehouse 1992:73ff.; for the Bronze Age Cocchi Genick et al. 1995, Cocchi Genick 1998), are reasons to connect them to ritual and not to formal burial. In ritual burial, human skeleton parts are often de-individualised by dismembering, and separately used to articulate the cave chambers, skulls playing a special role. As axes, swords or other stone and metal implements accompany the skeleton parts - often miniaturised or broken to prevent use in daily life - the dedicators as well as the deceased must have been high-ranking men. The pattern here described is not unique to Southern Italy but is also met with in Central Italy, especially in Campania, Lazio and Toscane. In the Grotta della Carbonara for instance a skull filled with ashes was exhibited upside down (Cocchi Genick 1998:366). Other caves contained human skulls in the midst of skulls of badgers and/or dogs (Cocchi Genick 1998:362). In the Belvedere-caves carbonised seeds were found and elsewhere berries and seeds are known to have been dedicated near the skulls (Cocchi Genick 1998:366). Ancestor veneration asks for this kind of ritual with human bones (Pearson 1999). From such ancestor veneration in caves, relatively far from the coeval communities, one may conclude that the identification of groups of people with specific territories was already present in the Neolithic and intensified during the Bronze Age: “Ancestor cults are localised and are characteristic of farming communities which rely on seasonal mobilization of communal labour, requiring unity among the living” (Pearson 1999:158). The dedication of fruits, cereals and seeds as well as the presence of animal bones, always in complicated mixtures of

Marianne Kleibrink This study concerns the sacred history of the Sibaritide, the modern name for the wide coastal plain around Sibari (the ancient Greek colony of Sybaris) and its hilly hinterland, situated on Italy’s east coast, below the heel. In this region long before the veneration of sky gods was introduced by overseas merchants, the native families dedicated objects at sacred places in the landscape that were thought to be inhabited by supernatural powers. 7.2.1

Cave cults and landscape perception during the Neolithic and Bronze Age periods In the Sibaritide most Neolithic settlement material has been found at open-air sites, e.g. at S.Gada near Morano Calabro, S.Maria del Castello at Castrovillari and on the plateau of Torre Mordillo. The most important Neolithic village with clear phases of the 6th and 3rd millennium BC discovered so far is at modern Favella della Corte, located in a wetland near the confluence of the Crati and Coscile, not far from Sibari. The site measures circa 4 ha. And here a hut from the earliest phase with a wicker floor was recently discovered (Tinè, Tinè & Traverso 1994). Neolithic Southern Italian settlement excavations have rendered few human bones. Burial in or near settlements was evidently practised with only few deceased and these were interred without any gifts. The caves Deposition took place more often in caves, relatively far from the settlements, where together with Neolithic material, many human bones have been found. The finds in the Grotte di Sant’Angelo (Tinè 1964, 1987; Whitehouse 1992: 39-40; Larocca 1991:61ff.) and the Grotta Pavolella (Carancini & Guerzoni 1987:783-793; Whitehouse 1992: no. 15), caves in the S. Marco mountain range near Cassano all’Ionio, as well as in the Grotta del Pozzo and Grotta Sirena at Frascineto (Paladino & Trajano 1989:78), resemble in many ways finds from caves elsewhere in Southern and Central Italy, for instance the Grotta Latronico (Potenza: Whitehouse 1992: no. 13) or the Grotta di Pertosa (Whitehouse 1992:62, 71) or the Romito cave at Papasidero (Cremonesi 1987:1-37; Graziosi 1962:12-20; Whitehouse 1992:162). The latter caves were used during the Neolithic, Eneolithic and Bronze Age periods, although there is no evidence of continuous use from one period to the other. For the caves in the Sibaritide neither continuous use nor gaps in the chronology could be established, because archaeological research has been unsystematic. The Grotte di Sant’Angelo and adjacent caves are karstic features in the so-called ‘Muraglione’, a natural wall of limestone rising circa 590 m above modern Cassano allo Ionio. The system consists of at least five interconnected 213

RPC 2000 conference proceedings

wild and cultivated/domesticated species, is a further indication that the cave cults were connected with early farming communities. From the widespread use of caves for ancestral ritual, usually far away from contemporary settlements, it may be deduced that the focus of each territory was the mountain where ancestor rites were concentrated. Barker discovered that Bronze Age settlement in the Tyrrhenian area, opposite the Ionian Sibaritide, also clustered around caves with cults (Barker 1985:68). In Central Italy likewise the Belvedere caves acted as cult centres both to the more central and more marginal settlements. Barker’s model of geographical units centering on cult caves seems to fit the central Sibaritide very well, with the “Muraglione” at its center. A novelty of the Late Bronze Age is the use of twin caves. At a number of sites a new cave was taken into use next to the one containing the material from older periods. We may suppose that the ‘neolithic’ cave cults continued over a long period: firstly because elsewhere in Italy this is very much the case and secondly because - apart from a few ‘a grotticella’ burials (burials in artificial caves) (La Torre 1995:175ff.) - no other burial system before the Final Bronze Age/Early Iron Age is known for the Sibaritide.

were rather mobile and at least seasonally settled on the higher mountain slopes behind the semicircle of permanent villages. Between the MBA sites of Santa Maria del Castello, Francavilla Marittima, Cerchiara and Villapiana recent surveys of the Groningen Institute together with Nino Larocca, the sites of ‘Ostello’ at Civita, ‘Timpa delle Fave’ at Castrovillari, ‘Terra Masseta’ and ‘Carlomagno’ on the slopes of the Sellaro are considered smaller dwelling places (internal reports GIA 2000-2001). Ideologically, all inhabitants will have perceived their space as much larger than their settlement territory, at least mentally stretching it to the caves in the mountains with the ancestral dead. The Muraglione at Cassano allo Ionio and settlement in the Sibaritide In the above, attention has been drawn to the central role of ancestor rites in the Sant’Angelo caves for the early farming communities. Further proof of the dominant position of the S.Marco caves in the central Sibaritide comes from a reconstruction of the old landscape and a more stringent analysis of MBA settlement. The plain of Sibari then demonstrated more contrast than today. Its coastal marshes and lagoons must, together with the multiple fluvial fans, have formed an interesting marshy landscape intersected by low conglomerate rocks with shrubs. This must have been especially marked around the Raganello and the Coscile, which over time formed several streambeds. Molluscs and parts of tortoises in the excavations of Francavilla Maritttima and Broglio are also indications of this type of landscape (A. Tagliacozzo in Peroni et al. 1994). The Muraglione, a massive limestone “wall” above Cassano allo Ionio, towering to circa 600 meters, must have been an even more impressive contrast to the coastal marshes than today. Below it, terraces of various heights were created by the activities of the sea and rivers. These are arranged in three quarters of a circle and naturally stepped in height (Vezzani 1968). Elsewhere the pattern of terraces is less orderly because interrupted by many old and recent alluvial fans. Nonetheless the wide and wellarranged terraces below the Muraglione seem not to have attracted much settlement before the very last period of the Bronze Age and /or the beginning of the Iron Age (Quilici et al. 1969, the sites from circa n. 50 and following all seem to date to the historical periods; Peroni et al. 1994, nrs. 31, 32 date them at the earliest from the Final Bronze Age; the Groningen University field walks in this area did not turn up protohistorical material, except in the caves: internal reports GIA, Attema & Van Leusen forthcoming). Elsewhere in the Sibaritide, especially in the neighbourhood of rivers and springs, both occurring in this area, such terraces did attract early settlement. Thus the absence of settlement around the S.Marco contrasts strongly with the pattern found elsewhere. Also, the terraces in this area did not carry the characteristic brown soils resulting from long human usage (Claudio Ballista in Peroni & Vanzetti 1998, on the basis of the work by Heilmann 1972 in the Crati valley). In colour the reddish terraces below the Muraglione in the 1960’s still resembled the Macchiabate, an extensive

Middle Bronze Age settlement in the Sibaritide Interestingly for the Sibaritide, a special kind of territoriality in the MBA period is indeed evident from the settlement pattern that we know well, thanks to the research by Renato Peroni and his group (recently Peroni et al. 1994:840ff). The MBA settlement pattern is remarkable: from Cape Spúlico to Cape Trionto no less than 13 newly inhabited sites follow perfectly the crescent shape of the foothills around the coastal plain. All sites are on low hills, which dominate the old transport routes along the rivers as well as the old north/south route, the ‘pedemontana’, along the foothills. A few inland sites also are on low hills, overlooking the wide valleys created by the main rivers of the Sibaritide, the Crati and Coscile. The settlements are 3/5 km apart, a reasonable walking distance for farmers from their hut compounds to their fields and pastures. Thus the MBA pattern confirms to the site-catchment theory of Eric Higgs and Claudio Vita-Finzi (Renfrew & Bahn 1991:224ff.). Because many settlements are situated on old fluvio-marine terraces, consisting of conglomerate and/or sand, thus covered by light soils, Peroni plausibly argues that MBA society in the Sibaritide preferred terraces on which dry farming of cereals and vegetables could be practised (Peroni et al. 1994:837). Thus from the MBA onwards people will have lived in hut-compounds with gardens. These hut villages all had their own separate territories, bordered by a river on each side. The streams will have supplied fresh water as well as the connection to the sea and the mountains. The animal bones from the few excavated sites indicate that the villagers mainly practised mixed farming. These communities presumably were dominated by a body of village elders, among other things performing ancestral rites. Smaller sites in the higher mountainous areas, recently discovered, indicate that some inhabitants 214

Landscape Perception in Archaeology

burial area of the Early Iron Age, which through time also remained largely unturned. The only possible explanation, in keeping with the ritual burials in the caves of the Muraglione, seems to me that this area was sacred, and that it was set apart for the ancestors. A further indication for a different, almost ‘tabu’, use of this area stems from the triangular, evenly spaced, arrangement of the earliest hut-villages in the central Sibaritide: Torre Mordillo, Castrovillari and Francavilla Marittima. The cave-assemblages as yet known from the Sibaritide are only just sufficient to detect ancestor cults. Also, the surveys in the S.Marco area and the foothills around the Sibarite plain are only just sufficient to detect an occupational discrepancy. In both areas further systematic study and research is necessary.

stronger and gradually the old pottery and metal shapes were substituted by new ones. Gradually the Sibaritide became filled in with a totally different type of society, judging from the new burial customs. There is the problem that the few burial indications we have are not a reliable indication of the number of inhabitants. Mario Torelli was first in stating that in Italy, before the 8th century BC, differential burial was practised and that the enormous rise in the number of tombs at Quattro Fontanili in Etruria was the result of allowing far more individuals to be formally buried than before (Torelli 1981:55; Hoekstra 1996-’97:4763). It took a long time in Italian archaeology before the implications of this option sank in. From finds of FBA fibulae in the Sibaritide, the interesting fact emerges that the sites, which grew to be important during the Iron Age, were all occupied during the Final Bronze Age. Unfortunately the known fibulae from this period come from clandestine excavations and old private collections, but generally their provenance is secure (Lo Schiavo & Peroni 1979). Comparisons with fibulae from Timmari and Sicily indicate that the burials they may have accompanied, most likely cremations, belong to only two or three generations. Supposedly, these FBA burials were of a few elite families who were not able to keep their position later on.

From ancestral to princely powers During the Late Bronze Age in a number of settlements important changes took place, while on a number of dominant hills in the landscape new settlements appeared. The best evidence up to now for the socio-economic and political changes in the Sibaritide comes from the excavations at Broglio di Trebisacce (Peroni et al. 1994; Peroni & Vanzetti 1998). The excavation yielded ceremonial drinking sets, either made in an Italo-Mycenaean style or made in ItaloGrey Ware or again in the local impasto tradition, as well as storage jars of large dimensions, which among other things will have contained olive oil (drinking sets: Castagna 2000:68-69; jars: Levi 1999). These items indicate that in this village a ‘princely’ form of living was copied from the Aegean. The finds at Broglio of a number of separate cellars with large storage jars from the last century of the Bronze Age, the period after the sack of Mycenae, may be interpreted as evidence for the spread among aristocratic families of these higher social forms of living. The new social forms were based on the controlled production of much-valued substances, presumably olive oil and wine. The rise of an aristocratic ruling class must have caused a general neglect of the old ancestral caves and brought about not only social but also religious change. From the extensive import and production of ceremonial pottery and its relative concentration in the native huts of standing, it is clear that the chiefs now preferred ceremonies in their daily surroundings instead of in far away mountain caves. The new pattern was characterized by the redistribution of power and cult. A change in ritual took place. Rather secret acts with limited participation gave way to rites in full daylight, having effect because the special sets of ceremonial dinner services could be observed by many.

Hoards In this period another complex, depositional system, in which axes, but also swords and other costly metal implements were important, was in use at conspicuous sites in mountains, such as lakes and springs. This happened not far away from the settlements as with the MBA ancestral caves, but nearer to them. The Sellaro (1240m high) with its twin peaks is the most conspicuous mountain overlooking the central plain of Sybaris. This calcareous mountain contains many caves. In the 1950’s on its lower slopes a large quarry was exploited and a hoard of asce ad occhio (shaft hole axes) was found by a workman (Procopio 1953:217 ss.). This find was a repetition of similar finds in at least two other different locations nearby. These hoards unfortunately were dispersed and none, or perhaps only a few objects, have surfaced (Museo Civico di Cosenza). To the commonly accepted economic interpretation that these hoards are the caches of itinerant bronze smiths (Bietti Sestieri 1973:383-424, 1981:223-264), for many have been added ritualistic interpretations (Bradley 2000; Hoekstra forthcoming). The Sellaro axe depositions were made at places from which they are difficult to retrieve, therefore they are most likely to have been used for ritual depositional acts. In other words, the formal disposal of costly metal objects, which started in the ancestral caves continued, but was now directed to sites away from the caves, usually springs, lakes, clefts, or other conspicuous spots in the landscape. In the LBA, at such spots, almost exclusively axes and swords were buried, which since the Neolithic had been connected with powerful males. This is an indication that they were the ones who gained from these rituals. The chiefs – identifiable in the settlements by the cellars with storage jars, will have practised such ceremonies to

7.2.2 The caves of the nymphs: burial, hoards and settlement in the Sibaritide during the Final Bronze Age Cremation burials The Final Bronze Age and first period of the Iron Age are not yet clearly defined phases for the Ionian coastal areas of Southern Italy. At the end of the Bronze Age influences from the Proto-Villanovan and Villanovan cultures became 215

RPC 2000 conference proceedings

demonstrate their affluence, besides, it served to keep the metal objects rare and valued. Thus it is no surprise that the lower slopes of the Sellaro contained several similar hoards, and we may suspect that the sulphorous ‘Grotta delle Ninfe’ originally attracted the hoards. Such spots, uninhabitable because of the malodorous fumes, were elsewhere also predominantly used for disposal ceremonies directed at natural phenomena (Bradley 2000). In later Latin myths such spots were considered the dwelling places of nymphs. What the names were of the powers responsible for the miraculous waters is unknown. The sulphorous springs are at the boundary between Villapiana and Francavilla Marittima, and this may be significant too.

corredi in Italian) dating from the Ferro 1b phase, to which the burials discovered at S.Lorenzo di Spezzano also belong, next the corredi excavated by Pasqui in 1888 at Torre Mordillo (phase 2a) and subsequently the ones from the Macchiabate necropolis at Francavilla Marittima dating from the Ferro 2a/b periods. Fairly abruptly from the 9th century BC onwards the formal burial of many of the deceased members of the native families was practised in all villages. As formal burial included the giving of gifts to the dead the fairly sudden appearance of the sets of grave gifts in the Early Iron Age may be considered an appropriate material expression of the transformation of the deceased from an important member of the community into a family ancestor. Thus in a symbolic way the layout of the cemetery mirrors the layout of the settlement. The Macchiabate necropolis comprising at least 250 individual tombs was excavated in the 1960’s by Paola Zancani Montuoro and fully published later on (Zancani Montuoro 1970-‘71, 1974-76, 1977-79, 1980-81, 1983-84). The tombs are constructed of river stones and contain indigenous burials, mostly inhumations laid out on their back, first with contracted, later with stretched legs. The Macchiabate necropolis comprises burial groups, some of which were excavated in the 1960’s by Paola Zancani-Montuoro (i.e. Strada, Cerchio Reale, Temparella): many are still to be excavated. The existence of various burial groups, and the distribution of the burials within these, is highly suggestive of the continued importance of kinship in the general layout of the cemetery. Apparently family groups reserved parts of the cemetery for the burial of their dead. Careful study of the successive Temparella burials allowed an insight into the transformation of the indigenous society from a competitive group of aristocratic families to a more amalgated village population, with the extended family as unit in a politics that endured for over three centuries (800 - 500 BC), and which stood up under foreign pressure (Kleibrink 2000a, 2000b and forthcoming). Competition among the families is especially evident from the various burials during the 8th century BC. A number of dead are buried with many grave gifts, others practically without any. As an example we may mention Temparella tomb 63, the tomb of an aristocratic ‘princess’ dating from the 8th century BC. The woman was buried with a scepter in her right hand and many complicated bronze pendants, as well as her personal jewellery. Similar competitive burial systems are known in other indigenous societies of Central and Southern Italy and are linked to the rise of an aristocracy. The Macchiabate burials initially demonstrate a similar rise of aristocracy. This process was, however, truncated by influences from abroad. These induced the indigenous families of the 8th century BC to perform special rites on top of the Timpone della Motta. Here in a large absidial native hut, a special hearth/altar received gifts, and large standing looms with monumental loomweights were operated by aristocratic women. The natives thereafter dedicated the more costly objects to this altar and no longer buried them with their ancestral dead.

Settlement in the Final Bronze Age The Final Bronze and Early Iron Ages in the Sibaritide show a conspicuous rise in the number of archaeological finds that can be related to demographic expansion. A steady growth in population will have been a major incentive for a new organisation of subsistence and settlement. In this period the settlements in the Sibaritide, became much wider-spaced over the landscape. The FBA pattern shows a system of six major settlements each dominating a segment of the crescent range of foothills around the plain. Now, unlike that of previous centuries, the choice of locations was no longer exclusively determined by a favourable position in relation to the means of subsistence. For the first time, socio-political considerations influenced the choice: some formerly attractive locations were abandoned and only those locations were selected for settling that were at an acceptable social distance from the next community. The following sites became dominant: Amendolara, controlling a territory between the streams of the Ferro and Straface; Broglio di Trebisacce controlling a territory between the Pagliara and Saraceno; Francavilla Marittima dominating the land between the Caldanelle and the Raganello; Castrovillari (Madonna del Castello) controlling the inner valley of the Coscile; Torre Mordillo controlling the area between the Eiano and the Crati; and Casale Rizzo/Serra Castello (recently, Peroni et al. 1994). These settlements are on terraced hills measuring over 7 ha and situated in positions very dominant over the surrounding territory. Torre Mordillo and Broglio di Trebisacce were fortified. This strongly politicized landscape emerging during the 12th century BC, suggests mounting social pressures within the communities, which lasted until the 9th century BC, when gradually a different structure emerged (see below).

7.2.3

The first Gods in the Sibaritide

Burial customs The Early Iron Age history of the native population is almost exclusively known from necropoleis. From south to north, at Castiglione di Paludi, Torre Mordillo and Francavilla Marittima, large Iron Age necropoleis, where inhumation was practised, have been found. The one at Paludi containing burials with pots and metal gifts (referred to as 216

Landscape Perception in Archaeology

The indigenous sanctuary on the top of the Timpone della Motta At Francavilla Marittima the native, female jewellery known from the 8th century BC tombs, but also iron swords, were found in the ancient sanctuary on top of the Timpone Motta, as already indicated near an altar (Kleibrink 1993:1-47; 1996:198 ff.; 1996-’97; 1998:20-30; 2000a,b). In small crevices in the conglomerate rock near a rounded and slightly raised burned part of the rock, quantities of fibulae, digital spirals, finger and arm rings, amber beads, and faience scarabs were found. The spatial distribution and stratigraphy were clear: these objects must have been dedicated at the nearby hearth-altar. This eschara will have received far more such dedications, which ended up in a sacred deposit nearby. This deposit is the famous and much plundered deposit I; the finds stolen from it and dispersed over many collections will be published by specialists from the Getty Museum and the Bern and Groningen Institute, and are now being prepared for return to Italy. Dr John Papadopoulos will publish the bronzes from the Geometric period stolen from Deposit I, which I consider belong to the altar on the Timpone della Motta. Another important find on the top of the Timpone della Motta at Francavilla Marittima also related to the altar, was a thick stratum of ashes. The fragments of pottery, from matt-painted trays (scodelle) and closed pots, as well as from large impasto containers, studied by Maria Sangineto, indicated that the altar was used during the final 9th century BC (the period of the Geometrico Antico) and the 8th century BC (Geometrico Medio) (Kleibrink & Sangineto 1999:1-61). Many fragments of animal bones were found with the pottery. It is especially noteworthy that no metal jewellery was among the finds from the ash layer; also all pottery fragments found in the ashes were unburnt. As stated above, the metal objects were found around the altar itself. A large apsidal hut may have contained the hearth/altar as well as a large standing loom, which was in function on the Timpone Motta (Building V phase 1). It is also possible that the hearth/altar was an open-air feature, located on the outside of the apsidal building. Many weights, beautifully decorated with labyrinth motifs, were excavated in 1996 and 2000 (Kleibrink 1996:198 ss.; 1998:20-30; 2000a,b). The weaving platform had left post-holes and small trenches cut in the rock, together allowing a squarish weaving stool to be reconstructed. There are two rows of loom weights, one of which is closed at the sides by heavier, undecorated weights, which the world over are used at the seams of the wrap-weighed loom. The impasto ceramics found together with the weights dates the weaving hut to the late Early and Middle Geometric period. A timber temple (building V phase 2) was constructed over the apsidal building at the end of the 8th century BC, together with two other parallel long buildings. Thus at Francavilla Mma a very monumental sanctuary was constructed in the last decades of the 8th century BC. It consisted of a long rectangular megaron (nr. III) flanked by two still larger ones, nr. II excavated by Maria Stoop in the 1960’s (Stoop 1983:17-38), and the newly excavated nr. V.

The technique of construction of these timber temples with their double rows of posts cut into the bedrock clearly is native. The roof construction must have been indigenous too, since no early roof tiles have been found. This sanctuary must have been built by the aristocratic families buried at Macchiabate, as it appeared that the same objects were present in the tombs as well as in the altar zone. Moreover the natives constructed their huts in the same way, with large posts put in holes of circa 50 cm in depth and a diameter of circa 50 cm. An indigenous long house with postholes on the Motta’s top, as well as another large oval hut underneath the Casa dei Pithoi on plateau III and a recently discovered hut on plateau I of the Timpone della Motta prove the point. It may be concluded that at Francavilla Greeks and Oenotrians together will have constructed an impressive temple-sanctuary in the last decades of the 8th century BC, on a place that was already special in an earlier period, as the altar and the weaving hut indicate. Many objects point to the presence of aristocratic indigenous women. This means that the religious impulse is not necessarily Greek or missionary. Monumental temples constructed with double rows of wooden posts fixed in rock cut holes found at Francavilla Marittima are as yet unknown anywhere in the Mediterranean. In the second half of the 7th century BC temple V was reconstructed (building V phase 3): its posts were taken out and closed with yellow clayey soil that covers the entire floor of the building. This yellow layer is thickest where the post-holes are and in the thicker areas small pots, pyxides, aryballoi and many hydriae were buried in sets. This burial of the first timber temple and the reconstruction of a new building is a clear moment of change, indeed of colonial Greek enterprise. 7.2.4 Conclusion The results from recent excavations and field surveys in the Sibaritide as well, demonstrate that the Oenotrian civilisation was much stronger and more homogeneous than current archaeology recognises (Maaskant-Kleibrink 1996‘7:63-91). The many regularly spaced Oenotrian hilltop settlements and adjacent necropoleis recently discovered in the Sibaritide, but also in Basilicata, demonstrate that in this part of Italy a civilisation with a strong cultural unity existed during the Iron Age. This civilisation flourishing in a large part of Italy’s Mezzogiorno is best called Oenotria. Unfortunately leading explanations, for instance those formulated by Piero Guzzo and François de Polignac deny the impact of the Oenotrians on the Southern Italian landscape: in Guzzo’s view it were the Greek colonies like Metapontion, Siris/Herakleia and Sybaris, all founded in coastal plains, which attracted the Oenotrians to live in the villages on the hilltops around the plains (Guzzo 1982:146151; 1983:237-251). In De Polignac’s view the Greeks founded sanctuaries along the border of their colonial territories to offer indigenous societies an opportunity to display wealth and influence, thereby civilising them (De Polignac 1984). These theories allow for an Oenotrian development only during the 7th century BC and under 217

RPC 2000 conference proceedings

Larocca, F. 1991, Le grotte di Calabria, Martina Franca. La Torre, G.F. 1995, Guide Archeologiche, Preistoria e Protostoria in Italia, La fascia tirrenica da Napoli a Reggio Calabria, Forlì. Leighton, R. & J.E. Dixon 1992, Jade and greenstone in the Prehistory of Sicily and Southern Italy, Oxford Journal of Archaeology 11(2): 179-200. Levi, S.T. 1999, Produzione e circolazione della ceramica nella Sibaritide protostorica. 1. Impasto e dolii. Firenze. Lo Schiavo, F. & R. Peroni 1979, Il Bronzo Finale in Italia, Atti della XXI Riunione Scientifica, Il Bronzo Finale in Italia, Firenze 21-23 ottobre 1977: 552-569. Kleibrink, M. 1993, Religious activities on the Timpone della Motta, Francavilla Marittima - and the identification of Lagaria, Bulletin Antieke Beschaving 68: 1-47. Kleibrink, M. 1996, Le scoperte recenti sul Timpone della Motta, I Greci in Occidente. I santuari della Magna Grecia, Electa, Napoli: 198 ff. Maaskant-Kleibrink, M. 1996-’7, Dark Age or Ferro I? A tentative answer for the Sibaritide and Metapontine plains. In: Debating Dark Ages, CAECVLVS 3 (Papers on Mediterranean Archaeology): 63-91. Maaskant-Kleibrink, M. 2000a, Enotri, Greci e i primi culti nell’Athenaion a Francavilla Marittima, Magna Graecia XXXV: 18-30. Maaskant-Kleibrink, M. 2000b, Early Cults in the Athenaion at Francavilla marittima as Evidence for a pre-Colonial Circulation nostoi Stories, Die Aegaeis und das westliche Mittelmeer, Beziehungen und Wechselwirkungen 8. bis 5. Jh. v. Chr. (ed. Friedrich Krinzinger), Vienna. Kleibrink, M. forthcoming, In Search for Sybaris: an Evaluation of Historical and Archaeological Evidence, BABesch 2002. Kleibrink, M. & Sangineto, M. 1999, L’insediamento enotrio su Timpone della Motta I, la ceramica geometrica dall’edificio V, Francavilla Marittima, Bulletin Antieke Beschaving 72: 1-61. Malone, C. 1985, Pots, Prestige and Ritual in Neolithic Southern Italy. In: Malone, C. & S. Stoddart (eds) Papers in Italian Archaeology IV, part 11. British Archaeological Reports, International Series 244: 118-51. Paladino, A. & Trajano, G. 1989, Calabria Citeriore, Archeologia in Provincia di Cosenza, Trebisacce. Pearson, M.P. 1999, The Archaeology of Death and Burial, Frome. Peroni, R., F. Trucco, C. Belardelli et al. 1994, Enotri e Micenei nella Sibaritide, Taranto. Peroni, R. & A. Vanzetti 1998, Broglio di Trebisacce 1990-1994, Elementi e problemi nuovi dalle recenti campagne di scavo, Soveria Mannelli. Procopio, G. 1953, Cerchiara di Calabria, ripostiglio di acette bronzee dell’Età del ferro, Bolletino della Preistoria Italiana n.s. 8: 153-154. Quilici, L. et al. 1969, Carta archeologica della Piana di Sibari, Atti e Memorie della Societa Magna Grecia 9-10, Roma. Renfrew, C. & P. Bahn 1991, Archaeology, Theories, Methods and Practice, Thames and Hudson. Tinè, V. 1964, La Grotta di S.Angelo III a Cassano Ionio. Atti e Memorie della Società Magna Grecia n.s. 5: 11-55. Tinè, S. 1987, Il Neolithico, Storia della Calabria antica I, Reggio Calabria/Roma: 39-63. Tinè, S., V. Tinè & A. Traverso 1994, Il villaggio di Favella della Corte e la neolitizzazione dellla Sibaritide, Atti Taranto 1994. Torelli, M. 1981, Storia degli Etruschi, Roma/Bari.

Greek domination. In my opinion the vitality and coherence of the Oenotrian culture can be demonstrated if not only settlement patterns are reviewed as has repeatedly been done, but if more attention is paid to the sacred landscape, which emerges when sufficient attention is paid to the artefact assemblages buried in the landscape over time.

References Barker, G. 1985, Prehistoric farming in Europe, Cambridge, New York, etc. Atti Taranto 1994, Sibari e la Sibaritide, Taranto (Atti del Trentaduesimo Convegno di studi sulla Magna Grecia, Taranto-Sibari, 7-12 ottobre 1992). Bietti Sestieri, A.M. 1973, The metal industry of continental Italy, 13th to 11th century BC, and its connection with the Aegean, Proceedings of the Prehistoric Society 39: 383-424. Bietti Sestieri, A.M. 1981, Produzione e scambio nell’Italia protostorica. Alcune ipotesi sul ruolo dell’industria metallurgica nell’Etruria mineraria alla fine dell’Età del Bronzo. In: L’Etruria Mineraria (Atti del XII Convegno di studi Etruschi ed Italici, Firenze etc., 16-120 giugnio 1979): 223-264. Bradley, R. 2000, An Archaeology of Natural Places, London & New York. Carancini, G.I. & R.P. Guerzoni 1987, Gli scavi nella grotta Pavolella presso Cassano allo Ionio (Cs), Atti XXVI Riunione Scientifica Istituto Italiano di Preistoria e Protostoria, Firenze: 783-793. Castagna, M.A. 2000, La Casa dei bevitori, Archeo 186 (XVI, 8): 6-71. Cocchi Genick, D., I. Damiani, I. Macchiarola, R. Peroni & R.R. Poggiani Keller 1995, Aspetti culturali della media eta` del bronzo nell’Italia centro-meridionale, Florence. Cocchi Genick, D. 1998, L’antica età del bronzo nell’Italia centrale, Florence. Cremonesi, G. 1987, Il Paleolitico, Storia della Calabria antica, I, Roma/Reggio Calabria: 1-37. De Polignac, F. 1984, La naissance de la cité grecque, Paris. Graziosi, P. 1962, La scoperta di incisioni rupestri di tipo paleolitico nella grotta del Romito presso Papasidero in Calabria, Klearchos 13-14: 12-20. Guzzo, P. 1982, Modificazioni dell’ambiente e della cultura tra VIII e VII secolo sulla costa ionica d’Italia, Dialoghi d’Archeologia 2: 146-151. Guzzo, P. 1983, La Sibaritide e Sibari nel VII sec. a.C., Atti del Convegno Internazionale, Grecia, Italia e Sicilia nell’VIII e VII sec. a.C., Atene (5-20 ott 1979), Annuario della Scuola Archeologica di Atene e Missioni italiane in oriente, vol. 60 n.s. 44 (1982): 237-251. Heilmann, P.G.F. 1972, On the Formation of Red Soils in the Lower Crati-basin, S. Italy (thesis Utrecht), Meppel. Hoekstra, T.R. 1996-’7, Life and Death in South Etruria. The social rhetoric of cemeteries and hoards. In: Debating Dark Ages, CAECVLVS 3, Papers on Mediterranean Archaeology, Groningen: 47-63. Hoekstra, T.R. forthcoming, Buried Wealth and its Social Significance. Metal Hoards and Tombs in Italy, XII-VIIIth Centuries BC, Groningen thesis.

218

Landscape Perception in Archaeology

Stoop, M.W. 1983, Note sugli scavi nel santuario di Atena sul Timpone della Motta (Francavilla Marittima Calabria) 4, Bulletin Antieke Beschaving 58: 17-38. Vezzani, L. 1986, I terreni plio-pleistocenici del basso Crati, Cosenza. Whitehouse, R.D. 1992, Underground Religion, Cult and Culture in Prehistoric Italy, Accordia Specialist Series on Italy, Herring E., R.D. Whitehouse & J.B. Wilkins (eds), London. Zancani Montuoro, P. 1970-’71, Necropoli di Macchiabate, Coppa di Bronzo Sbalzata, Atti e Memorie della Società Magna Grecia n.s. 11-12 [1970-71]: 9-33. Zancani Montuoro, P. 1974-76, Necropoli. Tre Notabili Oinotri dell’VIII sec. A.C. Atti e Memorie della Società Magna Grecia n.s. 15-17 [1974-76]: 10-106.

Zancani Montuoro, P. 1977-79, Francavilla Marittima. Necropoli di Macchiabate. Saggi e Scoperte in Zone varie. Atti e Memorie della Società Magna Grecia n.s. 18-20 [1977-79]: 7-91. Zancani Montuoro, P. 1980-81, Francavilla Marittima a) Necropoli e ceramico a Macchiabate. Zona T. (Temparella), Atti e Memorie della Società Magna Grecia n.s. 21-23 [19801981]: 7-130. Zancani Montuoro, P. 1983-84, Francavilla Marittima. Necropoli di Macchiabate. Zona T. (Temparella, continuazione), Atti e Memorie della Società Magna Grecia n.s. 24-25 [1983-1984]: 7-110.

219

RPC 2000 conference proceedings

7.3

Aspects of Urbanization and Ethnic Identity in the Middle Tiber Valley

in the Sabine area especially we have no evidence of settlements for the 9th century BC (Filippi, Pacciarelli 1991) and in the Faliscan area only a few fortified settlements survive this demographic transition; we can also note that they are located near natural boundaries and seem to delineate the first northern frontier of the early state of Veii (figure 2) (Pacciarelli 1991; Ceci & Cifarelli 1992; 1995; di Gennaro 1995). The Volsinian area shows less radical transformation: the central place of Volsinii (now Orvieto) occupied an 80 hectare fortified area on a rocky plateau (Tamburini 1992:14-21) and seems to have had a weaker influence over its territory in comparison with the other early Tyrrhenian states such as Veii, Caere, Tarquinia and Vulci; in fact the territory of the Volsinii is characterized by more villages in the 9th century and the transition from the Bronze Age to the Iron Age is less evident too (Colonna 1978; Carandini 1997:345, n. 109; Tamburini 1998:57). However, two new settlements suggest the existence of a “state” frontier: the fortified village of Monte Piombone (Wendt et al. 1994:119-138), and the open village of Gran Carro, on Lake Bolsena (Tamburini 1995), which appears to be connected to the food supply system of the Volsinii and which also marks the end of an important road system from Lake Bolsena to the Tiber valley used from the Bronze Age onwards (Casi & Tamburini 1999:270-271, figure 12). On the left side of the Tiber the Umbrian and Sabine areas seem to adopt the proto-urban system only at the end of the 8th and the beginning of the 7th century BC. Terni in Umbria, with an area of 50 hectares (Roncalli 1988:398399; Rienzi 1997:76) and Cures in Sabina, with an area of 30 hectares (Guidi et al. 1996; Zifferero et al. 1996), are the main settlements, but for both regions the organization of the territory suggests not so much the large state communities of the Etruscan area (figure 3) but more the idea of a medium-sized city-state system, whose boundaries generally lay within a radius of a one-day walk from the central town (see for a definition: Nichols & Charlton 1997:8-9). As a result of this transformation, during the beginning of the Iron Age, the Tiber became a political and cultural boundary between different ethnic groups2. The Orientalizing period (810-580 BC) is characterised by the development of the secondary centers in the Etruscan area (Iaia & Mandolesi 1993), where aristocratic people came to power, as shown also by the burial customs (Zifferero 1991); at the midpoint of the 6th century the process of state formation in central Italy was complete, and two types of organization can be clearly seen: the first

Gabriele Cifani If we glance at a physical map of Italy we can observe that the Tiber Valley is the only natural pass through the Apennine Mountains connecting the central and northern parts of the peninsula. The importance of this region for trade opportunity is stressed also by the presence of different ethnic groups (the Umbrians, Etruscans, Faliscans, Sabines and Latins; figure 1) who divided the area into territories from the beginning of the Iron Age, using the river Tiber as a conventional frontier (Colonna 1986; Bartoloni 1986). The Middle Tiber Valley represents the area of contact between not less than four of these peoples: the Etruscans of Volsinii, the Faliscans, the Umbrians and the Sabines, and thus provides an opportunity to observe several regional pathways to complexity by comparing the ethnic groups1. In this paper I will focus on two aspects: 1) The different patterns of urbanization in the Etruscan area of Tiber, Umbria and Sabina between the 9th and 8th centuries BC 2) The particular process of rural colonization in the Etruscan state of Volsinii during the second half of the 6th and the beginning of the 5th century BC In central Italy during the last phases of the Bronze Age (11th - 10th centuries BC) the territory was organized on the basis of several small fortified villages, between 2 and 5 hectares in size, each having a territory, reconstructed by the application of Thiessen’s polygons, of a few square km (di Gennaro 1982, 1986, 1988; Guidi 1998). The landscape is considerably fragmented by the small villages, and the Tiber seems already to play an important role in attracting settlers. The model of society hypothesized for this period is the chiefdom (Bietti Sestieri 1998:17-18). This homogeneous situation was drastically modified at the beginning of the 9th century BC, when we observe the earliest differentiation of future ethnic groups within the territories. In fact the formation of the 175 hectare protourban center of Veii, in the lower Tiber valley (Guaitoli 1981; Pacciarelli 1991:183-185), caused a general depopulation of the surrounding region of the Faliscans and Sabines:

1

This paper offers a synthesis of data from a PhD work about the Middle Tiber Valley (Cifani 1999). Essential recent works about the landscape archaeology of this region: Territory of Volsinii: Bloch 1972; Colonna 1973, 1999; Tamburini 1998; Casi & Tamburini 1999; Stopponi 1999; Wendt et al. 1994; Territory of Veii (including Ager Faliscus and Ager Capenas): Potter 1976, 1979; Morselli 1980; Colonna 1988; Camilli & Vitali Rosati 1995;

2

220

Mazzi 1995; De Santis 1997; Patterson & Millett 1998; Tartara 1999; Southern Umbrian Area: Roncalli 1988; Rienzi 1997; Monacchi 1999; Sabine Tiberine Area: Muzzioli 1980; Filippi & Pacciarelli 1991; Santoro 1996, 1997; Mari 1992, 1996. Colonna 1986; in addition, regarding the relationship between ethnicity and territories of pre-roman communities, see Bradley 1997 and the stimulating paper by M. Cristofani (1999).

Landscape Perception in Archaeology

Figure 1: The ethnic communities of the Middle Tiber Valley (from Colonna 1986).

being the large territorial state, with a hierarchical system of settlement and one central place, typical of Rome and Southern Etruscan peoples, and the second the city-state, with a small territory (figure 4). A rank-size diagram of the settlements, divided into ethnic areas of the Middle Tiber Valley in the archaic period, shows more precisely this phenomenon: • The Volsinian area reveals a three-level structure of the territory, with one central place of 80 hectares, a middle group of settlements the size of which ranges from 18 to 35 hectares, and finally a large group of small settlements, 1 to 5 hectares in size (figure 5). • The Veientan area with the Ager Faliscus shows a similar organization (figure 6). • A completely different diagram however is seen for the Umbrians and Sabines: in both cases the graph shows few elements of diversity and the trend is quite homo-

geneous (figures 7 and 8). For the Sabine area we can see diversity among the central places and secondary centers as, for example, between Cures and Campo del Pozzo. We may also hypothesize a settlement hierarchy for the Umbrian area. In both cases there is evidence of a cluster of “peer” city-states, located at an average distance. In the case of the Sabina this is 15 km as the crow flies, in the case of Umbria 25 km. The territory is subdivided into small territories rather than large ones as in the Volsinian and Veientan state territories. For the Etruscan lands however, we may note a contrast between on the one hand urban and central places, where a middle class of craftsmen and traders rises up progressively, and on the other hand the peripheries where aristocratic groups keep and improve their power based on traditional agriculture, sheep-rearing and tolls. This economic and social contrast probably became a political engagement 221

RPC 2000 conference proceedings

Figure 2: Settlements of the Middle Tiber Valley in the first phase of the Early Iron Age. 1: Orvieto [Volsinii]; 3: Castellonchio; 4: Poggio di Sermugnano; 5: Bagnoregio; 10: Civita del Fosso di Arlena; 11: Gran Carro; 12: Montefiascone; 13: Fondaccio; 18: Monte Piombone; 24: Castellaccio; 28: La Ferriera; 34: Capena; 38: Todi; 40: Terni; 41: Amelia; 47: Poggio Sommavilla (?); 48: Campo del Pozzo (?); 52: Monte S. Angelo; 53: Veio.

within states; an account of such social struggles in the 6th century BC is offered by the literary tradition of archaic Rome (Cornell 1995:173-197) but, at least as far as the Volsinian area is concerned, landscape archaeology can offer evidence too (Colonna 1985; Chierici 1999). From the beginning of the 6th century, the town of Volsinii reveals a thorough social transformation as is indicated by the isonomic organization of the necropolis of “Crocifisso del Tufo” (Colonna 1985:110) and by the evi-

dence of a rational plan of the town with a grid of underground drains. One of these was recently investigated by stratigraphic excavations, which confirmed its use in the 6th century BC (Cenciaioli 1991). In parallel with the urban reform of Volsinii, we can observe, from the second half of the 6th century BC onwards, the progressive decline of many secondary centers in the territory. The best-known examples are Acquarossa and Monte Piombone, where Swedish investigations revealed the end of 222

Landscape Perception in Archaeology

Figure 3: Settlements of the Middle Tiber Valley in the second half of the 8th century BC 1: Orvieto [Volsinii]; 3: Castellonchio; 4: Poggio di Sermugnano; 5: Bagnoregio; 6: Civita di Grotte di Castro; 10: Civita del Fosso di Arlena; 12: Montefiascone; 13: Fondaccio (?); 17: Acquarossa; 18: Monte Piombone; 21: Orte; 23: Vignanello; 24: Castellaccio; 25: Corchiano; 28: La Ferriera; 29: Nepi; 30: Falerii; 31: Monte Lombrica; 32: Narce; 33: Nazzano Romano; 34: Capena; 38: Todi; 40: Terni; 41: Amelia; 43: Otricoli; 45: Magliano Sabina; 47: Poggio Sommavilla; 48: Campo del Pozzo; 49: Cures; 50: Montelibretti; 51: Eretum; 53: Veio.

the centers in the first half of the 6th century BC (Viterbo 1986:32-33; Wendt et al. 1994:119-138). In the region around Lake Bolsena, two important settlements were abandoned at the end of the 6th century BC: Civita di Grotte di Castro and Civita del Fosso di Arlena (Tamburini 1998:69-82 and 78-88). There was, however, a third phenomenon occurring in parallel with the above: the foundation of new small forti-

fied settlements, which started during the 6th century. There are two examples of these: Monterado, east of Lake Bolsena, and Castellaro, on the Tiber near the SE corner of the Volsinian state. Monterado has been known since the end of the 11th century BC; this 2-hectare settlement is located on a volcanic hill and fortified by a ring of walls with a system of underground drains (Barnabei et al. 223

RPC 2000 conference proceedings

Figure 4: Settlements of The Middle Tiber Valley in the second half of the 6th century BC. A = Frontier line between the Etruscan territories of Volsinians and Veientans. 1: Orvieto [Volsinii]; 3: Castellonchio; 4: Poggio di Sermugnano; 5: Bagnoregio; 6: Civita di Grotte di Castro; 7: Barano; 8: Bolsena; 9: Monterado; 10: Civita del Fosso di Arlena; 12: Montefiascone; 14: Celleno; 16: Castellaro; 20: Pian Miano / Monte Casoli; 21: Orte; 22: Gallese; 23: Vignanello; 25: Corchiano; 27: Sutri (?); 29: Nepi; 30: Falerii; 32: Narce; 33: Nazzano Romano; 34: Capena; 36: Vaccareccia; 38: Todi; 40: Terni; 41: Amelia; 43: Otricoli; 45: Magliano Sabina; 46: Foglia (?); 47: Poggio Sommavilla; 48: Campo del Pozzo; 49: Cures; 50: Montelibretti; 51: Eretum; 53: Veii.

shards dating from the 6th to 3rd centuries BC. The pottery of Castellaro was produced at Volsinii, as is proved by the shape and the quality of grey bucchero, and some of its inhabitants probably came from the urban population of Volsinii, as shown by two fragments of grey bucchero inscribed with letters similar to other “graffiti” of urban origin (Cifani 1996-1999). Monterado and Castellaro are

1894:44; Gamurrini et al. 1972:17; Quilici & Quilici Gigli 1974). Castellaro has only been recently discovered by survey and is presented for the first time on this occasion (figure 4, n. 16). It is a small fortified settlement of 2 hectares on the top of a small volcanic hill, like Monterado, which appears to have been artificially regularized. A survey was carried out in 1999 after deep ploughing turned up several 224

Landscape Perception in Archaeology

Figure 5: Territory of Volsinii. Histogram of settlement sizes for the late archaic period (540-480 BC). 1: Orvieto [Volsinii] (85 ha); 20: Pian Miano / Monte Casoli (25 ha); 39: Copio (25 ha); 6: Civita di Grotte di Castro (20 ha); 5: Bagnoregio (18 ha); 10: Civita del Fosso di Arlena (6,3 ha); 7: Barano (5 ha); 4: Poggio di Sermugnano (3 ha); 16: Castellaro (3 ha); 9: Monterado (2,5 ha).

very similar to each other and represent two examples of new foundations or “colonies” supported within its territory by Volsinii, which aimed to control strategic keypoints and deprive the old aristocratic centers of authority and economic importance. In addition, these fortified settlements were political landmarks for people living in isolated farmhouses. The rural demography of this area has been fairly well established by surveys, and offers evidence of many open settlements in the territory; however, only two archaic farmhouses have been explored by the Soprintendenza: Pianello near Castiglione in Teverina (D’Atri 1986), and Girella near Monterado (Cagiano de Azevedo 1974). These are both complex buildings with stone walls and tiled roofs; their shape and size suggest that they were probably used by no more than two families. The farmhouses are evidence of the desire of Volsinii to colonize its territory, offering land to town-dwellers. As sponsor and developer of colonies, Volsinii’s interest in such technical infrastructure as rural drainage systems, well-known in archaic Latium and Etruria (Judson & Kahane 1963) can be understood. These would probably have been tested in the urban setting before being applied in agriculture. Obviously these changes did not occur in the same way for the whole Middle Tiber Valley, because in general the rise of the middle classes in the 6th century BC developed differently. For example, the Ager Faliscus, as part of the Veientan territory, shows continuity of form among the

secondary centers and no evidence of crisis inside the aristocratic groups, who continued to be buried each with several imported attic vases, as shown by the necropoleis of Narce, Falerii, Corchiano and Vignanello (Torelli 1990:218). This phenomenon is, however, more evident in Sabina and Umbria, where the local aristocracies seemed to preserve their role, as shown by the rich tombs with chariots of the 6th and 5th centuries BC at Todi (Emiliozzi 1997:319, nn. 92-93; Torelli 1982:57) and Spoleto (Emiliozzi 1997:179-190, 319, n. 87), and where the towns have revealed little evidence of public architecture (Roncalli 1988:397-405). To conclude we can say that the long-term changes in the landscape of the Middle Tiber Valley, between the 6th and 5th centuries BC, can be interpreted as the result of social and politic struggles within the states. Here the dichotomy between aristocracy and middle classes is reflected in the organisation of the territory. In addition, the result of the analysis of the evolution of settlements in relation to ethnic groups stresses the importance of the influence of local traditions and ideologies on the organization of territory, as against economic and environmental factors. This is also why it is important, from the point of view of research criteria, to state that post-processual archaeology (Johnson 1999:98-115; Barker 1999:24) should influence the methodology of landscape research by means of a holistic approach which considers all the archaeological evidence and not just that of the settlements; only in this way can we try to explain some of the many pathways to complexity. 225

RPC 2000 conference proceedings

Figure 6: Territory of Veii. Histogram of settlement sizes for the late archaic period (540-480 BC). 53: Veii (175 ha); 34: Capena (60 ha); 30: Falerii (45 ha); 32: Narce (35 ha); 25: Corchiano (30 ha); 29: Nepi (12 ha); 23: Vignanello (10 ha); 21: Orte (8 ha); 33: Nazzano Romano (8 ha); 22: Gallese (7 ha); 36: Vacchereccia (1 ha).

Figure 7: Umbrian Area. Histogram of settlement sizes for the late archaic period (540-480 BC). 40: Terni (50 ha); 38: Todi (35 ha); 41: Amelia (32 ha).

Figure 8: Sabine Area. Histogram of settlement sizes for the late archaic period (540-480 BC). 49: Cures (30 ha); 47: Poggio Sommavilla (25 ha); 51: Eretum (18 ha); 48: Campo del Pozzo (10 ha); 50: Montelibretti (10 ha); 45: Magliano Sabina (8 ha).

Acknowledgements The study of urbanization in central Italy is part of my PhD thesis carried out at the “La Sapienza” University of Rome between 1996 and 1999. For helpful debates I thank Prof. G. Colonna, A. Carandini, A. Guidi and Drs. F. di

Gennaro, S. Stoddart, N. Terrenato, A. Zifferero. I would also like to thank the Soprintendenza Archeologica dell’Etruria Meridonale and particularly the officer Dr. Valeria D’Atri for granting permission to study the unpublished site of Castellaro. 226

Landscape Perception in Archaeology

References

Cornell, T.J. 1995, The beginnings of Rome. Italy and Rome from the Bronze Age to the Punic Wars (c. 1000-264 BC), LondonNew York. Cristofani, M. 1999, Litterazione e processi di autoidentificazione etnica fra le genti dell’Italia arcaica, in La Colonisation Greque en Mèditerranée occidentale (actes de la rencontre scientifique en hommage à Georges Vallet, Rome-Naples 1995), Paris: 345-360. D’Atri, V. 1986 [1988], Castiglione in Teverina (Viterbo), in Studi Etruschi 54: 352-355. De Santis, A. 1997, Alcune considerazioni sul territorio veiente in età orientalizzante ed arcaica, in G. Bartoloni (ed), Le necropoli arcaiche di Veio. Giornata di studio in memoria di Massimo Pallottino, Roma: 101-143. Di Gennaro, F. 1982, Organizzazione del territorio nell’Etruria meridionale protostorica: applicazione di un modello grafico, in Dialoghi di Archeologia n.s. 4(2): 102-112. Di Gennaro, F. 1986, Forme di insediamento tra Tevere e Fiora dal bronzo finale al principio dell’età del ferro, Firenze. Di Gennaro, F. 1988, Il popolamento dell’Etruria meridionale e le caratteristiche degli insediamenti tra l’età del Bronzo e l’età del Ferro, in Etruria meridionale. Conoscenza, conservazione e fruizione (Atti Convegno a Viterbo 1985), Roma: 59-82. Di Gennaro, F. 1995, La Ferriera di Sutri, in N. Negroni Catacchio (ed), Preistoria e Protostoria in Etruria, Atti del Secondo Incontro di Studi, Milano: 277-280. Emiliozzi, A. (ed), 1997, Carri da guerra e principi etruschi (catalogo della mostra, Viterbo 1997), Roma. Filippi, G. & M. Pacciarelli 1991, Materiali protostorici dalla Sabina Tiberina. L’età del bronzo e la prima età del ferro tra il Farfa ed il Nera (Quaderni del Museo Civico Archeologico di Magliano Sabina 1), Magliano Sabina. Gamurrini, G.F., A. Cozza, A. Pasqui & R. Mengarelli 1972, Carta Archeologica d’Italia (1881-1897), Firenze. Guaitoli, M. 1981, Notizie preliminari su recenti ricognizioni svolte in seminari dell’Istituto, in Quaderni dell’Istituto di Topografia 9: 79-87. Guidi, A. 1998, The emergence of the State in central and northern Italy, in Acta Archaeologica 69: 139-161. Guidi, A., F. Bistolfi, A. Zifferero, O. Colazingari, M. T. Fulgenzi, A. Arnoldus-Huyzenveld & M. Ruffo 1996, Cures Sabini: lo scavo, le strutture, la cultura materiale, le attività economiche, in Identità e Civiltà dei Sabini (atti del XVIII Convegno di Studi Etruschi e Italici, Rieti - Magliano Sabina 1993), Firenze: 143-204. Iaia, C. & A. Mandolesi 1993, Topografia dell’insediamento dell’VIII secolo a. C. in Etruria meridionale, in Journal of Ancient Topography 3: 17-48. Johnson, M. 1999, Archaeological Theory. An Introduction, Oxford. Judson, S. & A. Kahane 1963, Underground drainageways in southern Etruria and northern Latium, in Papers of the British School at Rome 31: 74-99. Mari, Z. 1992 [1993], Note topografiche su alcuni centri protostorico-arcaici fra Lazio e Sabina, in Studi Etruschi 58: 17-52. Mari, Z. 1996, Insediamenti arcaici nella Sabina meridionale, in Identità e Civiltà dei Sabini (atti del XVIII Convegno di Studi Etruschi e Italici, Rieti - Magliano Sabina 1993), Firenze: 297-323. Mazzi, M.C. (ed) 1995, Capena ed il suo territorio, Roma.

Barker, G. 1999, Hunting and farming in prehistoric Italy: changing perspectives on landscape and society, in Papers of the British School at Rome 67: 1-36. Barnabei, F., A. Cozza, G.F. Gamurrini & A. Pasqui 1894, Antichità del territorio falisco esposte nel Museo nazionale Romano a Valle Giulia, Monumenti Antichi dei Lincei 4. Bartoloni, G. 1986, I Latini e il Tevere, in Quaderni del Centro di Studi per l’Archeologia Etrusco Italica 12: 98-110. Bietti Sestieri, A.M. 1998, L’Italia in Europa nella prima età del ferro: una proposta di ricostruzione storica, in Archeologia Classica 50: 1-67. Bloch, R. 1972, Recherches archéologiques en territoire volsinien de la protohistoire à la civilisation Étrusque, Paris. Bradley, G. 1997, Iguvines, Umbrians and Romans: ethnic identity in central Italy, in T. Cornell, K. Lomas (eds), Gender and Ethnicity in ancient Italy, London: 53-67. Cagiano de Azevedo, M. 1974, Bagnoregio (Viterbo). Scavo in località Girella, in Notizie degli Scavi: 21-37. Camilli, A. & B.Vitali Rosati 1995, Nuove ricerche nell’agro capenate, in N. Christie (ed), Settlement and Economy in Italy 1500 BC to A.D. 1500, (Papers of the Fifth Conference of Italian Archaeology, Oxford, 1992), Oxford: 403-412. Carandini, A. 1997, La nascita di Roma. Dei, lari, eroi e uomini all’alba di una civiltà, Torino. Casi, C. & P. Tamburini 1999, Rapporti tra geomorfologia e insediamenti nel distretto lacustre volsiniese tra l’età del bronzo e il periodo etrusco, in Annali della Fondazione per il Museo “Claudio Faina”: 259-279. Ceci, F. & F.M. Cifarelli 1992, Aspects de l’occupation du sol dans le sud de l’Etrurie au IX siècle avant Jésus-Christ, in Actes Colloque international de Lons-le-Saunier (1990), Nancy-Paris: 445-458. Ceci, F. & F.M. Cifarelli 1995, La fase antica della prima età del Ferro in Etruria meridionale: aggiornamenti, in N. Negroni Catacchio (ed), Preistoria e Protostoria in Etruria, Atti del Secondo Incontro di Studi, Milano: 281-283. Cenciaioli, L. 1991, Cunicoli di drenaggio ad Orvieto, in M. Bergamini (ed), Gli Etruschi maestri di idraulica, Perugia: 169176. Chierici, A. 1999, Corredi con armi, guerra e società a Orvieto, in Annali della Fondazione per il Museo “Claudio Faina” 6: 183221. Cifani, G. 1996-1999, Dinamiche insediative e territoriali nella Media Valle Tiberina dal IX al III secolo a.C. (Tesi di Dottorato in Archeologia, Etruscologia), Università di Roma “La Sapienza”. Colonna, G. 1973, Ricerche sull’Etruria interna volsiniese, in Studi Etruschi 41: 45-72. Colonna, G. 1978, La posizione di Bagnoregio nell’antico territorio volsiniese, in Doctor Seraphicus 25: 43-52. Colonna, G. 1985, Società e cultura a Volsinii, in Annali della Fondazione per il Museo “Claudio Faina” 2: 37-53. Colonna, G. 1986, Il Tevere e gli Etruschi, in Quaderni del Centro di Studi per l’Archeologia Etrusco Italica 12, Roma: 90-97. Colonna, G. 1987, I Latini e gli altri popoli del Lazio, in Italia Omnium Terrarum Alumna, Milano: 409-528. Colonna, G. 1999, Volsinii e la Val di Lago, in Annali della Fondazione per il Museo “Claudio Faina” 6: 9-29.

227

RPC 2000 conference proceedings

Monacchi, D. 1999, Storia e assetto in età antica del territorio in cui ricade la villa di Poggio Gramignano, in D.N. Soren (ed), A roman villa and a late roman infant cemetery. Excavation at Poggio Gramignano (Lugnano in Teverina), Roma: 23-42. Morselli, C. 1980, Sutrium, Firenze. Muzzioli, M.P. 1980, Cures Sabini, Firenze. Nichols, D.L. & T.H. Charlton (eds) 1997, The Archaeology of City-States, Washington. Pacciarelli, M. 1991 [1994], Territorio, insediamento, comunità in Etruria meridionale agli esordi del processo di urbanizzazione, in Scienze dell’Antichità 5: 163-208. Patterson, H. & M. Millett 1998, The Tiber Valley Project. Papers of The British School at Rome 66: 1-20. Potter, T.W. 1976, A Faliscan Town in South Etruria, London. Potter, T.W. 1979, The Changing Landscape of Southern Etruria, London. Quilici, L. & S. Quilici Gigli 1974, Bagnoregio (Viterbo) – Saggi di scavo in località Monterado, in Notizie degli Scavi di Antichità: 38-55. Rienzi, P. 1997, Terni dalla Prima Età del Ferro alla conquista romana (VIII-II sec.a.C.), in V. Pirro (ed), Interamna Nahartium. Materiali per il museo archeologico di Terni, Terni: 59-135. Roncalli, F. 1988, Gli Umbri, in Italia omnium terrarum alumna, Milano: 373-407. Santoro, P. 1996, Il museo di Magliano Sabina nella prospettiva delle ricerche sulle culture della Sabina, in Identità e Civiltà dei Sabini (atti del XVIII Convegno di Studi Etruschi e Italici, Rieti - Magliano Sabina 1993), Firenze: 275-285. Santoro, P. (ed), 1997, Magliano. Origini e sviluppo dell’insediamento, Pisa-Roma. Stopponi, S. 1999, Contributo alla conoscenza del territorio orvietano, in Annali della Fondazione per il Museo “Claudio Faina” 6: 41-76.

Tamburini, P. 1988-1989, Orvieto e il territorio volsiniese nella prima età del ferro, in Bollettino dell’Istituto Storico Artistico orvietano 44-45: 7-79. Tamburini, P. (ed) 1995, Un abitato villanoviano perilacustre. Il sul lago di Bolsena (1959-1985), Roma. Tamburini, P. 1998, Un museo e il suo territorio. Il Museo Territoriale del Lago di Bolsena 1. Dalle origini al periodo etrusco, Bolsena. Tartara, P. 1999, Torrinpietra (IGM 149 I NO), Firenze. Torelli, M. 1982, La società della frontiera, in M. Bergamini, G. Comez (ed), Verso un museo della città. Mostra degli interventi sul patrimonio archeologico, storico, artistico di Todi, Todi: 54-58. Torelli, M. 1990, Storia degli Etruschi, Roma-Bari (3° ed.). VITERBO 1986, Architettura etrusca nel Viterbese. Ricerche svedesi a S. Giovenale e Acquarossa 1956-1968 (Catalogo della Mostra a Viterbo) Roma. Wendt, L., M.B. Lundgreen, P. Roos, E. Rystedt, M. Strandberg Olofsson & C.B. Persson 1994, Acquarossa VII. Trial trenches, tombs and surface finds, Stockholm. Zifferero, A. 1991, Forme di possesso e tumuli orientalizzanti nell’Italia centrale tirrenica, in E. Herring, R. Whitehouse, J. Wilkins (eds), Papers of the fourth Conference of Italian Archaeology, London: 107-134. Zifferero, A., O. Colazingari, M.T. Fulgenzi, F. Bistolfi & A. Guidi 1996, Cultura materiale e sistemi insediamentali nella sabina tiberina, in A.M. Bietti Sestieri, V. Kruta (eds), The Iron Age in Europe (Papers of the XIII Congress of the International Union of Pre and Protohistoric Sciences, Forlì 1996, vol. 12), Forlì: 91-105.

228

Landscape Perception in Archaeology

7.4

Lucanian Sanctuaries and Cultural Interaction

sanctuary of Satriano the material was found in the fill of the natural open pool of the spring, and it can be argued that it was deposited as votives. In the other cases the material can only be interpreted as evidence for Archaic cults through the composition of the finds or by the fact that the areas became cultic sites in the 4th century. In all cases the votive groups are found in an area where indigenous tombs and settlement have also been found, but it should be noted that they do not seem to be situated inside a habitation area. The famous Mefitis sanctuary at Ansanto in the Hirpinian Mountains is sometimes referred to as an Archaic or even a 6th century Italic sanctuary (Bottini & Rainini 1976). This dating is based on the votives. 8 objects - all figurines - are dated in the 6th or the 5th century in the catalogue of the publication: nos. 1-3 (bronze figurines; nos. 8, 18 and 19 (wooden figurines) and nos. 24-25 (terracotta figurines). The date relies on Colonna’s classification of bronze votives (Colonna 1970), especially the comparison between Ansanto nos. 1-2 with Colonna nos. 344 and 348, but it is important to note that Colonna did an attempt to classify the bronzes with the intention of forming production groups. The sequence was, according to Colonna, “attendibile da un punto di vista cronologico soltanto nelle sue linee generale”. Furthermore Colonna’s “periodo arcaico” was dated 525-375 BC. Thus the evidence for an Archaic cult at Ansanto is unequivocal, but the site is still of importance for its long history (at least from the 4th century BC into the Roman imperial period) where votives seem to have been deposited continuously in the sulphurous volcanic spring.

Helle Horsnæs 7.4.1 Introduction Along the coastlines of Southern Italy a large number of Archaic Greek sanctuaries are placed like pearls on a string. The sanctuaries were often situated at the border of the chora, and they were from early on adorned with splendid buildings. They were magnificent landmarks. These Western Greek sanctuaries formed the empirical basis for F. de Polignac’s theory of the importance of the sanctuaries in the formation of the polis (De Polignac 1984). In the period contemporary with the formation of the Greek apoikiai the situation in indigenous Lucania was in striking contast to this. In the vast area extremely few finds antedating the fourth century BC can be associated with non-burial cultic activity, and these finds are mainly restricted to sherds found in areas with later cultic activity. Yet, the tombs found in inland Lucania are evidence that the inhabitants of this area performed rituals connected with the burials. This supports my opinion that humans have always had some sort of religious belief, and that they have performed rituals and ceremonies accordingly. A diachronic study of the evidence of cult and rituals in Lucania, based primarily on contemporary archaeological and epigraphical evidence focusing on the development of visible features of cultic activity, will not only chart the development of the physical evidence of Lucanian cult, but also reveal much information on the indigenous peoples’ response to first Greek, and later Roman presence in the area.

7.4.3 The ‘Lucanian’ period During the 4th century the picture gradually changed, and at least 20 sites have yielded material that can be interpreted as indigenous cultic assemblages. The topographical position divides these sites into two groups. One is a small group consisting of buildings with cultic connotations within - or even in the centre of - an inhabited area, the other and much larger group consists of extramural sites near a natural spring. In both groups there may be artificial installations connected with the sanctuary intended to conduct or collect water (canals, cisterns), and where a central cultic building is found it is quasi-square rather than oblong / rectangular. There are often secondary buildings related to the area, and in some instances one of them is a two-room hall complex. Altars, hearths or the remains of a possible cult statue are found at a number of sites. It is possible to divide the 4th-3rd century sanctuaries into three types. The first consists of large regional sanctuaries centred on a holy spring. Examples are Macchia di Rossano and Valle d’Ansanto. Both may have been in use before the 4th century BC, and they certainly remained in existence well into the Roman period (the cult of Rossano was even transferred to the Roman foundation Potentia). The second type of sanctuary is centred on an extremely small square building, situated in the open space of a courtyard that may be surrounded by halls and/or porticoes and connected with water installations. Examples of

7.4.2 The ‘Archaic’ period We can assume that the lack of archaeological material reflecting cultic activity in the eighth to sixth centuries must be due to the fact that the inhabitants of Lucania worshipped their gods in places that are archaeologically invisible and with rites that have left few traces. The nature of the finds from the late sixth century onwards, now slowly emerging, seems to support this view. Indeed, the finds related to cultic sites that may go back to the late 6th century are restricted to a few votive groups - Satriano Loc. Torre (E. Greco 1991) and North-west sanctuary (Holloway 1970), Rivello Loc. Colla (G. Greco 1990), Timmari (Lo Porto 1991), Garaguso (Hano, Hanoune & Morel 1971) and Palinuro (Naumann & Neutsch 1960) - of pottery mostly, and in some cases terracotta figurines having a distinct Greek character. Palinuro and Rivello are situated on the Tyrrhenian coast, while Satriano, Garaguso and Timmari are inland sites. Satriano controls the mountain route from the Tanagro Valley to the upper Basento, while Garaguso and Timmari are situated close to the Salandrella and Bradano Rivers leading to the Ionic Sea. These sites have all yielded material traditionally dated in the late 6th and 5th centuries (Ionic-type cups, cups of Bloesch type C and/or local geometric ware). In the case of the north-west 229

RPC 2000 conference proceedings

this type are Roccagloriosa Complex A (with water canals, and votives and altar in situ) and Pomarico Vecchio Big Building (with altar table and statue base found in a cistern situated in front of what seems to be the remains of a small building). In both these cases the cult site is clearly intramural and seems to be in a central position to the inhabited area. The preliminary reports on the “Casa della Rampa” in Laos hint that this complex was very similar to the Big Building in Pomarico, but in the case of the “Casa della Rampa” at Laos there is no mention of a small square building (E. Greco 1996 and poster presented at the EAA Annual Meeting in Ravenna 1997). Armento shows the remains of two successive square buildings, water installations, and the remains of a bronze statue of Heracles, all found in an area that eventually became enclosed by halls and porticoes. It seems to be situated close to an inhabited area, but the topographical relations between the sanctuary and the habitations are not clear. Furthermore the 3rd century rural complex at Banzi is closely comparable to this type. The small square building is situated in a fenced area, next to the farm-house. The third type is still hard to define, and may dissolve to nothing when more material becomes available. At Satriano, San Chirico and Rivello somewhat larger shrines have been found. The information on Rivello and especially San Chirico is of a preliminary character, but there are similarities between the three complexes found at Satriano (loc. Torre), Rivello, and most recently San Chirico Nuovo that cannot be wholly overlooked. Most important are the combination of a square building, halls/porticoes, and the presence of terracotta figurines. These three sites may therefore constitute another type of sanctuary, where the shrines were not combined with an altar. Still, Satriano the best known of these complexes - has also some similarities with Armento, in particular the bi-partite hall with finds that seems related to the male sphere (votive weapons) and a number of thymiateria. It should be noted that Satriano and Rivello are two of the sites yielding material from the late 6th century onwards. The overlaps between the characteristics of the three types of sanctuaries are extremely important. They may on one hand warn that this classification - based on a relatively low number of often badly known sites - must be seen as preliminary; but they may also provide us with the clue that these sites all belong to the same material culture: that the similarities may be due to their common forerunners, while the differences may be explained as reflections of different “functions”. As mentioned above Lucanian cult buildings appear at a relatively late stage, from the late 4th century onwards. This, however, does not imply that cult buildings did not exist before that, as the number of pre-4th century Lucanian buildings is in general very low. But the topographical position of the large sanctuaries of the Hellenistic/Republican periods, combined with the general focus on artificial water installations in connection with cultic areas, and the find circumstances of the earliest assemblages that may be interpreted as “cultic”, suggest that - apart from the burial rituals taking place at the grave, visible through the choice and posi-

tion of the grave goods and in the way of burying - until the late 5th century religious gatherings often took place outside the inhabited area, in a grove and/or close to a spring, and offerings may have consisted of both pottery and objects made of perishable material. Among the perishable materials there may have been objects of leather or wood (for example wooden statuettes like the ones from Ansanto), as well as contents in the pottery (beverages, corn). 7.4.4 The identification of the cults The “function” of the sanctuaries - or rather: “who worshipped which deity, and how did they do it?” - is one of the most complex questions. In the late 20th century the discussion of cults in indigenous areas of Southern Italy was based mainly on Mario Torelli’s article from 1977: Greci ed indigeni in Magna Grecia: Ideologia religiosa e rapporti di classe. The premise for Torelli’s article is important: “L’apparizione dei Greci nelle aree di colonizzazione ... si manifesta ... nel segno della superiorità militare, tecnologica, politica, o ... di un livello di sviluppo di forze produttive più complesso rispetto a quello del mondo indigeno ... . L’elemento indigeno ...è ... ideologicamente subalterno rispetto a quello greco e l’interferenza ideologica ... è stata certamente ridottissima” (Torelli 1977, esp. p. 49). Torelli firmly believed in a Greek superiority in relation to the indigenous peoples in all aspects of culture, and his study was therefore concentrated on the Greek influence on indigenous cults. He saw the earliest evidence for indigenous sanctuaries in the finds of Greek style architectural terracottas at inland sites, and the votive “deposits” with predominantly Greek material were seen as evidence for the cult also being of Greek type. This view seems to imply that the indigenous peoples of Southern Italy had no religious beliefs of their own, a view I feel, that is difficult to maintain. Instead we must first try to find out as much as possible about local cultic activity, and after this, discuss to what degree the cultic expressions of the Lucanians (and other peoples of Italy for that matter) are a result of Greek influence. In a third instance it may be worth studying the influence of local cults in the Greek sanctuaries of the apoikiai. When we ask what the votives can tell us about the cult, we expose a general methodological problem. The identification of the cults in Lucania is often based on the votives: female statuettes are regarded as referring to a goddess, miniature weapons as referring to a god. The fact that the votive terracottas depict Greek types, has led many scholars to a rather mechanical identification of the Lucanian divinities with the Greek gods. A figure depicting a woman with a shield becomes an “Athena”, an enthroned woman wearing a polos becomes the Paestan Hera, a nude becomes Aphrodite. Literary and archaeological sources are not in accordance in their interpretation of many of the terracottas. The literary sources on the Ansanto sanctuary clearly indicate that the Italic goddess Mefitis was worshipped in the sanctuary. She had obvious chthonic aspects, at least in this particular sanctuary otherwise known for its vulcanic springs with a strong sulphuric odour, and as being the 230

Landscape Perception in Archaeology

entrance to Hades in Vergil’s Aeneid. What did Mefitis look like? This is clearly seen in the LIMC entry on Mefitis, where all the examples are taken from the Ansanto figurines: the primitive xoana, the terracotta “Athena”, “Paestan Hera”, and “Aphrodite”. If all these figurines are to be interpreted as representations of Mefitis, she obviously did not have a fixed iconography. The very important epigraphical evidence from Macchia di Rossano, dating from ca. 350 BC to the end of the Republican period, indicates that the sanctuary was dedicated to Mefitis, but it is also clear that other gods and goddesses were honoured at Rossano. This may explain the great variety of figurines from Ansanto. As at Rossano, Mefitis may be regarded as not the only, but the most important deity. Thus the sanctuary may have been known by the name of the most important god, and the mention of only one god cannot be decisive for excluding polylatry. Similarly, the Abella cippus mentions only Heracles, but other deities may easily have been venerated at the same place, as was the case with the sanctuary of Cerres at Agnone. There is another possible interpretation of the variety of figurine types in the sanctuaries. In a recent summary of the inscriptions from Macchia di Rossano, Lejeune has proposed that the Lucanian pantheon be seen in the light of the “conception trifontionelle” defined by the structuralist Dumezil (Lejeune 1990). This model implies that the main goddess of the sanctuary, Mefitis, is seen as a “third function” goddess, guarding the sphere of fertility, but also including chthonic aspects. If Mefitis is protecting the fertility of both man, beast and crops, the topographical situation of the Lucanian sanctuaries close to springs as well as to trade or transhumance routes becomes evident. Likewise it is clear why a great variety of votive types could be used in one sanctuary: Mefitis cannot simply be identified with Demeter, Hera, Athena or Aphrodite. She contained aspects of all of them. Whether the variety of figurine types in a sanctuary should be interpreted as depicting a deity with many aspects or as evidence for the worship of a plurality of (female) deities, this explanation does not account for other types of votives commonly found in the Lucanian sanctuaries. One group that easily springs to the eye is the group of (miniature) weapons. Many miniature weapons were found in Armento interpreted as a (among others?) Heracles sanctuary because of fragments of a bronze (cult?) statue of Heracles found there. Although originally part of the Greek pantheon, Heracles seems quickly to have gained great popularity throughout the Italian Peninsula (Radke 1965:140-142), in particular in the Samnite area, where bronze figurines depicting Heracles are very common (Van Wonterghem 1992). In Campania the cult of Heracles is documented at Abella and Nola by the Cippus Abellanus (Vetter 1953, no. 1) and in Lucania both by the finds of miniature Heracles clubs used as votives and by the occasional use of a Heracles figure as belt clasps on the “Samnite” belts. What, then, was the role of Heracles in Lucania, or for that matter in Campania and Samnium? Van Wonterghem

has noted two major aspects of Heracles. On one hand he is connected with transhumance, protecting the herds and the springs. On the other he may have taken the place of the warrior god Mamers (related to the Roman Mars) in the southern part of Abruzzo/Molise, where Mamers figurines give way to Heracles figurines (Van Wonterghem 1992:321). In the first case Heracles can be seen as the consort of Mefitis, that is: with a fertility function according to the scheme of Dumezil. But it is also possible to argue that Mamers in some area has taken over the iconography of Heracles so that what we see in Lucania is a fusion of the two of them. Thus there would be no discrepancy in the finding of votive weapons in a Heracles sanctuary. 7.4.5 Dualism in Lucanian cult There are examples of two “sets” of votive types being found in such a close connection that they clearly belong to the same sanctuary. The oldest example is that of the archaic material from the two groups at Garaguso (Altieri and Autera): the first group contained a large number of votive terracottas, while the second group was mainly metal objects (Hano, Hanoune & Morel 1971). It has been suggested by Torelli (1977) that the two groups at Garaguso represent two different classes of worshipper. Another bipartition can be seen in the 4th century, where two groups containing different material have been found at Banzi/Fontana dei Monaci. Masseria (1991) has argued that the miniature weapons, tools and jewels found in the metal deposit are deposited for their symbolic value: the weapons and tools representing objects qualifying an adult man, and the jewels pertaining to the adult woman. Thus, the votives in the altar were deposited as part of a transition rite from adolescent to adult. The other group was found closer to the spring, and Masseria believes that this area was used for ritual purification, for example in connection with marriage - in itself a rite de passage. A division can be seen at Satriano, but it is interpreted differently. The large hall was divided into a square room with traces of a hearth, and a larger room in which were found two terracotta figurines, a thymiaterion, and some iron weapons. The shrine southwest of the hall contained figurines and pottery. E. Greco (1991) suggests that the two contexts represented male and female. Some 20 m further to the southwest a small, unroofed enclosure was found. The enclosure seems to have been empty, but around it lay fragments of thymiateria and an unguentarium: and we must believe that some sort of ritual took place there. Also in the open area between the shrine and the enclosure votive terracottas (85 fragments) and pottery were found, and on the stairs leading from the open area to the hall lay a roughly worked pillar. Greco interpreted the pillar as the base for a bronze phial found close by, comparing it with representations of a pillar used as the base for an egg in Dionysian scenes in South Italian red figured vase painting. Although the three contexts here mentioned have a number of features in common, three very different explanatory models have been employed. In Garaguso and Satriano the two types have been explained as differences in either the class or gender of the worshippers, while the 231

RPC 2000 conference proceedings

finds from Banzi have been explained as evidence for two different rituals. In this connection we must be aware that different theoretical and methodological approaches to the same material may lead to different conclusions. A broader empirical basis is desirable in order to conduct a comparative study of votive deposits.

sure. The sanctuary is normally described as a Greek sanctuary dedicated to Demeter (Cipriani 1989). In the light of the ambiguity of the finds from Loc. Fonte and the important recent find of a 5th century Lucanian cemetery at Loc. Gaudo only 600 m north of Poseidonia, we may ask whether the Albanella sanctuary was intended as a non-Greek cult site from the beginning. If so, the quasi-square form of the enclosure may become an important sign. The interpretation of the meaning of the finds from these sanctuaries has of course enormous influences on both the question of the ethnic composition of the inhabitants of the Greek colonies and on the definition of the chora.

7.4.6 Cult and ethnicity Were the cults in Lucania indigenous or Greek? Were the rituals undertaken by Greeks or by indigenous inhabitants? Could it be an indigenous cult dressed up in a Greek outfit? And if indigenous, is it then possible to interpret the cult in the light of Greek practices/customs, just because the votives are Greek? Is it possible to make an ethnic division of cults based on the votive types present in a given sanctuary? Is it possible to make an analysis of the votives from this point of view? A study of the minor sanctuaries in the area surrounding the Greek apoikia Poseidonia elucidates the problems involved in these questions. The votives found at Loc. Fonte, ca. 10 km east-northeast of Poseidonia, are described as of the same type as the ones found in Poseidonia itself and in the Heraion at Foce del Sele, and Loc. Fonte is normally regarded as part of Poseidonia’s chora (Greco, Stazio & Vallet 1987). Consequently Loc. Fonte could be interpreted as a Poseidoniate border sanctuary. But close to the sanctuary two archaic tombs with indigenous grave-goods have been found. The combination of Greek votives and indigenous tombs is thus closely comparable with the situation at sites such as Garaguso and Timmari. These sites are normally considered to be in indigenous territory, and the “Greek-ness” of the votives there is underlined in contrast to the “localness” of the tombs. We must consider whether the use of Greek type votives is a result of a conscious choice of a specific type indicating the ethnicity of the worshippers, or whether it reflects the facility with which one could get mass-produced votive terracottas “made in Poseidonia” or “made in Metaponto”? And again: can - or must - we use the same interpretative models for different sites? The possibility that indigenous cults used expressions very similar to the Greek cults, for instance Greek votive figurines, makes it extremely difficult to argue whether a specific sanctuary should be interpreted as a Greek border sanctuary or an indigenous cult site. In the case of Loc. Fonte, I think it is very unsafe to apply an ethnic (Poseidoniate or indigenous) to the sanctuary, and not far from Loc. Fonte, the sanctuary at Albanella presents similar problems. The oldest finds from Albanella are dated to the 5th century (Cipriani 1989). In this early phase the sanctuary consisted of six hearths (focolari) and two (votive) pits containing burnt offerings, and depositions of objects (pottery, a hammer head, an iron fragment (interpreted as part of a lance), and a terracotta mask) inside the almost square enclosure (the so-called strato 9). However, the majority of the objects from Albanella, in particular the many terracotta figurines, are dated to the 4th century, i.e. in the period after the Lucanian take-over of Poseidonia. The 4th century material was found in a layer covering the whole enclo-

7.4.7 Italic influence on Greek cult It remains a question whether Lucanian cult ever inspired (formerly) Greek areas. Here the case of Poseidonia is extremely interesting. Poseidonia was founded as a Greek settlement around 600 BC, but it is generally agreed that it became Lucanian in the late 5th century. The sanctuaries of the Greek city continued to be in use, and the Lucanian take over seems to have happened quite peacefully. Only in the Foce del Sele sanctuary is there evidence of dramatic change, but by the end of the 5th century the area had been cleaned and new buildings erected. Best known is the Square building. The very un-Greek form of this building and its possible function have been variously interpreted, and there is a standing discussion whether the building is regarded as sacred or profane (G. Greco 1997, with all relevant references). Giovanna Greco has recently argued that the Square Building was used by the young girls who had to weave the new peplos for the goddess. Her arguments are founded on the fact that one of the largest groups of material found in the Square building were the ca. 90 loom weights, and a comparison with evidence for peploforia in Greek sanctuaries. G. Greco’s interpretation draws on Greek rather than indigenous parallels, but loom weights were also found in considerable numbers in Timmari and Ansanto, and especially in the latter sanctuary Greek influence seems negligible. The emerging pattern of 4th century Lucanian sanctuaries with (quasi-) square buildings (Satriano; Rivello/Colla; and the very small shrines at Armento, Roccagloriosa, Pomarico/Big Building, Lavello/Gravetta) underlines the importance of this structure for the Lucanian cult, and I wonder whether the Square Building in the Foce del Sele sanctuary may be seen as an example of a Lucanian building type dressed up in Greek masonry. If so, also the interpretation of the finds will have to be reconsidered. 7.4.8 The Romanization of Italic cults The Roman political take-over of Southern Italy in the second quarter of the 3rd century seems to have had little impact on the material culture of the area. In some places the cult ceased during the 3rd century. At Pomarico Vecchio the cultic area of the court of the Big Building was deliberately dismantled, but the small shrine at Banzi was not 232

Landscape Perception in Archaeology

erected until the 3rd century and at Armento the shrine was renewed in this period. In some places, notably the large “regional” sanctuaries Macchia di Rossano and Valle d’Ansanto but also at Armento, building activity even intensified after the Roman conquest, and the terraces and halls at these sites may be regarded as weak reflections of the more magnificent Hellenistic terrace complexes. That the cultic activity taking place at these sanctuaries was still of a clear Italic, rather than Greek or Roman, nature is implicit in the use of Oscan well into the 1st century BC for a large number of the inscriptions from the Macchia di Rossano santuary - indeed the majority of the Oscan inscriptions post-date the Roman conquest. It has been suggested that the monumentalization of the Italic sanctuaries in this period should be seen as a political statement. The monumentalization of the sanctuaries underlined the importance of ancient local cults, and became a means for the leaders of the Italic peoples to protest against Rome’s growing influence. The same may apply to the banquet hall recently excavated in Tricarico, which may signal the emergence of the first known type of public building in Lucania. The hall was in use from the first half of the 3rd century, and was destroyed in the end of the third century (De Cazanove 1996). This banquet hall may be related to the almost contemporary buildings described as hestiatoria in Buccino (Johannowsky 1985; Johannowsky 1990) and Pompei (below the Casa delle Forme di Creta; see De Caro 1992:76). These structures have previously been related to a non-Greek banquet tradition parallel to the one suggested for Tricarico (De Caro 1992:76 and nn. 46-47; De Cazanove 1996:920-921). Again, however, we must remember that this is a period of growing monumentalization throughout the Mediterranean, and we should keep in mind that although Rome was the nominal master of Southern Italy from ca. 270 BC, the extent of the Roman presence was negligible well into the 2nd century BC. The turning point was probably the Hannibalic War, when Rome suddenly found trouble very close to its doorstep. The foundation of a number of coloniae along the coastline of Southern Italy in the early years of the 2nd century can be seen as Rome’s attempt to meet a future threat and to guard the coastline, as well as a manifestation of Roman presence directed at the indigenous peoples and the Greek cities, both of which had seceded from Rome in 218-211 BC. There is no reason to believe that Rome sought to destroy pre-Roman cults, on the contrary it is much more likely that Rome took in local cults and encouraged continuity. The Mefitis cult of Macchia di Rossano seems to have been administered since the Social War from Roman Potentia (founded 2nd century), and at some later point the cult was probably transferred to Potentia itself (Lejeune 1990:39; De Lachenal 1992). In some cases the existence of a pre-Roman cult may even have had some importance in the positioning of a new colony. Among the earliest Roman colonies in the hinterland, Grumentum was founded ex novo in the 3rd century. It was later connected with Potentia by the Via Herculia, and thus became a crossroads for inland trade in Lucania.

In connection with the question of the religious landscape, it may be of some importance to note, firstly that a preRoman sanctuary must have existed at San Marco, south of the Roman town, where a votive deposit has been found containing some one hundred votives (mainly female terracotta busts and figurines, but also among other finds four coins) from the second half of the 4th and the early 3rd centuries BC, secondly that Grumentum has yielded one of the relatively few Latin dedications to Mefitis (Lejeune 1990:47). 7.4.9 Conclusion In the preceding lines I have tried to discuss some of the points raised by Peter Attema in his presentation of the workshop, in the form of a discussion of some aspects of Lucanian cult and sanctuaries: the outline of the development of Lucanian sanctuaries, the physical expression of the sanctuary, the variety of interpretations of votives. I have kept the focus on aspects elucidating the theoretical and methodological problems involved in the discussion of indigenous expressions of cult. To me it is very important that ideally any interpretation should take as its starting point a minute analysis of all the material related to a given site, and as a second stage undertake the comparison with related sites. Only when this empirical basis exists is it possible to apply a theoretical approach to it. This underlines the necessity of good full publications of the corpora of material, of which the volumes of the Corpus delle stipe votive in Italia have set an example. In this case the analysis of the empirical material has shown that a number of features (notably the figurines) found in sanctuaries in Lucania are of Greek type. We must question this material from various angles. We cannot mechanically take over interpretations based on our knowledge of the Greek world, but must be very aware that the meaning of an object may change when it is transported from its original context. A well-known approach to the Classical world is the back projection to pre-historic periods, of situations better known from historical sources. I believe that in the case of Southern Italy it is possible to use the material from the Hellenistic/Republican period to illuminate the Archaic period, because there is an unbroken cultural continuity from the eighth century onwards (as argued in Horsnæs 1999), and because the material from the pre- and protohistoric phases seems to tell the same story as in the later periods. A third approach to the discussion is the application (or adaptation) to the material of “global theory”. This was done successfully by Lejeune (1990) in his interpretation of the inscriptions of the Macchia di Rossano sanctuary, and it seems possible to extend his views to the more general interpretation of the Lucanian pantheon.

233

RPC 2000 conference proceedings

References

Horsnæs, H.W. 1999, The Cultural Development in NorthWestern Lucania c. 600-273 BC, Copenhagen. Johannowsky, W. 1985, Nuove scoperte a Volcei e nel suo territorio, Rassegna Storica Salernitana n.s.: 237-243. Johannowsky, W. 1990, Volcei, in Tagliente, M. (ed): Italici in Magna Grecia. Lingua, insediamenti e strutture, Leukania, Monografie di Archeologia della Basilicata 3, Venosa: 35-40. Lachenal, L. de (ed) 1992, Da Leukania a Lucania. La Lucania centro-orientale fra Pirro e i Giulio-Claudii, Catalogo della mostra Venosa, Venosa 1992. Lejeune, M. 1990, Méfitis d’après les dédicaces lucaniennes de Rossano di Vaglio, Bibliothèque des Cahiers de l’Institut de Linguistique de Louvain-la-Neuve 51, Louvain-la-Neuve. Lo Porto, F.G. 1991, Timmari. L’abitato, le necropoli, la stipe votiva, Rome. Masseria, C. 1991, Banzi - l’area sacra in loc. “Fontana dei Monaci”, in M. Salvatore (ed), Museo Archeologico Nazionale di Venosa, Matera: 84-85. Naumann, R. & B. Neutsch 1960, Palinuro. Ergebnisse der Ausgrabungen II. Nekropole, Terrassenzone und Einzelfunde, RM-EH 4, Heidelberg. Polignac, F. de 1984, La naissance de la cité grecque: cultes, espace et société VIIIe-VIIe siècles avant J.-C., Paris. Radke, G. 1965, Die Götter Altitaliens (Fontes et commentationes 3), Münster. Torelli, M. 1977, Greci ed indigeni in Magna Grecia: Ideologia religiosa e rapporti di classe, Studi Storici (Istituto Gramsci) 18: 45-61. Vetter, E. 1953, Handbuch der italische Dialekte I, Heidelberg. Wonterghem, F. van 1992, Il culto di Ercole fra i popoli oscosabellici, in: Héraclès. D’une rive à l’autre de la Méditerranée, Bruxelles: 319-351.

Bottini, A. & I. Rainini 1976, Valle d’Ansanto. Rocca San Felice (Avellino). Il deposito votivo del santuario di Mefite, Notizie degli Scavi: 359-524. De Caro, S. 1992, Lo sviluppo urbanistico di Pompei, Atti e Memoria Grecia 3a ser. 1: 67-89. De Cazanove, O. 1996, Un édifice de repas communautaires en Lucanie interne, Mélanges de l’Ecole Française de Rome Antiquités 108(2): 901-94. Cipriani, M. 1989, S. Nicola di Albanella. Scavo di un santuario campestre nel territorio di Poseidonia Paestum, Roma. Colonna, G. 1970, Bronzi votivi umbro-sabelliche a figure umana. I, Periodo arcaico, Firenze. Greco, E. 1991, In Lucania: ruolo dei sessi e istituzioni politico religiose (a proposito del santuario di Torre di Satriano), Dialoghi di Archeologia, 3a ser., 1-2: 75-83. Greco, E. 1996, Laos, colonia di Sibari, in D’Andria, F. & K. Mannino (eds), Ricerche sulla casa in Magna Grecia e in Sicilia, Università di Lecce, Scuola di Specializzazione in Archeologia Classica e Medioevale, Archeologia e Storia 5, Galatina: 127-132. Greco, E., A. Stazio & G. Vallet 1987, Paestum, Città e territorio nelle colonie greche d’occidente I, Taranto. Greco, G. 1990, L’area sacra di Colla, in AAVV: A sud di Velia: 69-75. Greco, G. 1997, Des étoffes pour Héra, in J. de La Genière (ed): Héra. Images, espaces, cultes. Actes du Colloque International du Centre de Recherches Archéologiques de l’Universié de Lille III et de l’Association P.R.A.C. (Lille, 29-30 novembre 1993). Collection du Centre Jean Bérard 15, Naples: 185-199. Hano, M., R. Hanoune & J.-P Morel 1971, Garaguso (Matera): Relazione preliminare sugli scavi del 1970, Notizie degli Scavi 1971: 424-438. Holloway, R.R. 1970, Satrianum, The archaeological investigation conducted by Brown University.

234

Landscape Perception in Archaeology

7.5

Perception des hommes et perception de la terre en Campanie septentrionale à l’époque de la conquête (V° avant J. - C. / III° avant J. - C.)

a lot of sites from the first period under examination; the third is the diversity in life-span of these sites. Some were occupied for centuries, others only lived in for a few tens of years. This means we must describe the perception of the landscape and of men, using a sociological and anthropological approach.

Jean Paul Vallat

7.5.1 Introduction A lire les Silves de Stace et l’Histoire naturelle de Pline, on peut se demander si, à la même époque, dans le courant du premier siècle de notre ère, la réalité de la terre campanienne est bien la même et si les pentes du Mont Massique demeurent largement boisées ou si elles sont couvertes de vignes1. Là où l’on doit s’acharner pour vaincre une nature hostile, des forêts épaisses, afin de faire passer une voie côtière entre montagne et mer, dans les parages de la colonie de Sinuessa, là aussi s’étendent des sommets à la plaine, en passant par de riants coteaux, les vins les plus fameux, Falerne, Faustinianus, mais aussi les vins plus médiocres, Massique notamment, de la Campania Felix. Etrange contraste qui pourtant n’invalide ni l’un ni l’autre des deux témoignages. La Campanie est une région où la main de l’homme a tant fait, tout fait, peut-être, pour maîtriser l’espace, vaincre forêts et marais, rétablir la circulation des eaux dans les embouchures ensablées, drainer et assécher, colmater, éviter la remontée des nappes phréatiques et trouver l’eau, surveiller les pentes de collines et lutter contre l’érosion, échapper aux tremblements de terre et craindre les volcans. Si Tite Live parle d’infesta regio lors de l’arrivée des premiers colons, au IV° siècle, et si l’imaginaire collectif garde encore présent aujourd’hui le formidable impact des combats entre Romains et Samnites qui conduisirent dans les années 340/290 avant J.-C. à la devotio de trois générations de conquérants romains, les Decii, c’est que la conquête ne fut ni facile ni rapide. Les hommes et les terres ne se laissèrent vaincre facilement ni les uns, ni les autres. Le paysage campanien, au nord du Volturne n’a cessé de changer, dans le temps long, à l’échelle des siècles, voire de plusieurs siècles, mais aussi à l’échelle d’une vie d’hommes, de quelques décennies. Ceci est vrai pour l’Antiquité comme pour la période très contemporaine. Là où, il y a quarante ans, au sommet des collines de Roccamonfina ou du Massique, des bergers naissaient dans des cabanes ou dans de petits villages, au milieu des champs de fèves, de haricots, de blé, il y avait dix ans plutôt, à la fin de la seconde guerre mondiale d’impénétrables bosquets et de nombreux bois, domaine de chasse apprécié de l’aristocratie et de la grande bourgeoisie napolitaine. Il y a aujourd’hui de maigres bosquets et un maquis, quelques oliviers et des vignes repartant à l’assaut des crêtes calcaires escarpées au-dessus de la mer. De même, dans l’Antiquité, la génération de Decius Mus n’a pas vu le même paysage que celle d’Hannibal et ce dernier aurait été surpris des cultures en place au temps de Varron. Les textes médiévaux et moder-

Abstract Campania has been studied from many angles; wars, preRoman and Roman cities, forms of colonization, institutions, social differentiation. There are many historical texts and there exists a huge documentation dealing with Campania dating from Antiquity. From the year 1978, substantial field surveys have covered North Campania, from the coast of Sinuessa to the hills of Cales, besides a few excavations or “sondages”. The present paper is supported by a study of 414 sites, of which 272 were identified as villae, farms, villages, sanctuaries, cemeteries or figlinae, using a 4D software package. The data can be consulted, using multiple criteria, by any researcher. Questions may be asked as to how many sites, farms, villas, villages, tombs, sanctuaries appeared in a certain period. Or queries as to the sites that show a certain type of ceramics or opus, or the farms that, in any particular period, were situated at a certain altitude and had certain characteristics of water supply. In the present section on the perception of the landscape, this particular paper deals with the perception of the landscape during the Roman conquest. From my point of view, this not only entails the perception of land by men, either indigenous or Roman, but also the perception of men by men. What did Samnites, Sidicini, Aurunci, Ausones and Roman people think of each other? The conquest gave rise to a huge conflict between the Roman and indigenous peoples, but, at the same time, also between rural settlers and Nature: streams, rivers, forests, marshes are difficult realities to fight, and Campania Felix is not a constant of agricultural life under Roman rule. Public and private land represent two of the best realities of appropriation, and determine in large part the relations between the Roman state, cities, temples, and private owners during the 4th to the 3rd centuries BC. It is this period of the Roman conquest that we will examine in this paper. What are the main changes in agrarian life that took place? This analysis tries to introduce the possibility of diverse models of development of social and economic rural life in Campania. Settlement forms and patterns were not only complex in the Roman Republican period and under the Empire. Also in the early 5th and 4th centuries BC, we already see a system of villages, farms, sanctuaries, and the beginning of a hierarchy among these. The evidence for this increases and gets more complex during the 3rd century BC and we may point out three basic facts: one is the incredible growth in occupation of the available land by a range of diverse farms, villae and villages, the second is the disappearance of

1

Stace, Silvae, 4,3,50-51; Pline, N.H., 14, 63.

235

RPC 2000 conference proceedings

Figure 1: Types de prospection en campanie entre 1979 et 1999.

nes relatant les conflits de bergers pour des parcours dans la plaine de Cellole, entre Suessa et Sinuessa, entre Garigliano et Volturne, traduisent la profonde évolution du paysage qui s’est développée de la fin de l’époque romaine jusqu’aux XIV°-XVI ° siècles. En Campanie comme souvent ailleurs, c’est l’homme et sa maîtrise des terroirs et finages qui ont fait la terre. Ce n’est pas un “don des Dieux”. Après cette rapide mise au point nous voudrions examiner d’abord la façon dont évolue l’occupation du sol, sur le plan de l’installation de l’habitat, du “remplissage” des finages, au moment de la conquête romaine.

Il s’agit de s’entendre sur la nécessité d’un langage commun de description et d’enregistrement des données pour pouvoir comparer les travaux des uns et des autres. Les critères qui permettent aux uns ou aux autres de parler de ferme, de village, de villae sont souvent plus subjectifs qu’objectifs, même si les fiches d’enregistrement des données en prospection se sont faites de plus en plus complexes. De toute façon, l’explicitation des critères qui conduisent à telle ou telle définition doit se faire pratiquement au cas par cas, car ni la taille des sites, ni leur matériel, ni même leur organisation interne (pars urbana, pars rustica…) ne suffisent à justifier de telle ou telle appellation. Une villa d’Etrurie n’est pas semblable à une villa d’Apulie et ressemble fort peu à une villa syrienne ou gauloise. Il faudrait bien poser, sinon résoudre, les problèmes de datation des sites, qui, notamment en prospection restent importants. Ils tiennent à la nature des échantillons de céramique récoltés, au problème des céramiques communes, mais aussi aux critères de datation de l’opus des murs, à la vitesse de diffusion des techniques. L’incertum ou le réticulé en milieu rural en Campanie est-il employé simultanément avec celui des villes ou bien y-a-t-il un décalage dans

7.5.2

Rythme et type d’implantation humaine au V° siècle avant J.-C. Avant de présenter le mode d’implantation des sites en Campanie, dans la phase de la conquête romaine, je voudrais insister sur quelques points que John Bintliff et moimême avons largement abordé lors de ce colloque et auparavant (Bintliff 1991, Leveau et al. 1995) et qui, depuis les travaux de D. Potter, d’A. Carandini ou de G. Barker sont d’ordre méthodologique (Potter 1979, Carandini et al. 1985, Barker 1995): 236

Landscape Perception in Archaeology

Figure 2: Rural settlement in Campania: Massiccia region.

le temps? Quel décalage y-a-t-il entre l’utilisation de cette technique en milieu urbain campanien et à Rome (De Caro & Greco 1981, Tremblements de terre 1986). A quelles causes sont dus les décalages spatiaux ou temporels éventuels? Les techniques ont pu avoir du mal à se diffuser faute d’une mobilité des architectes et des artisans. Elles ont pu être rejetées par tel milieu urbain ou rural pour des raisons de goût ou de moyens financiers. Les problèmes résident aussi dans le rapport entre sondages et prospections, dans la technique des prospections et dans le matériel divers “remonté” en surface selon que l’on mène les recherches en plaine, à flanc de coteaux, ou au sommet d’une colline. Ils tiennent encore aux dates “mythiques”, comme les guerres Samnites ou puniques, les crises gracquiennes, les guerres civiles qui ont souvent influencé fortement les interprétations de matériel archéologique (Hayes 1972, Morel 1981, Tchernia 1986, 1996). Il convient aussi d’étudier les rapports entre les diverses formes d’occupation rurale (fermes, villae, villages, nécropoles, sanctuaires) qui, dans le cadre des prospections, des fouilles, des sources archéologiques, littéraires et épigraphiques sont souvent, ou ont été longtemps victimes de visions mécanistes qui mettaient en avant l’exclusion de tel

type d’occupation par rapport à tel autre, selon les époques. Pour faire bref, nous dirions que dans maintes recherches, la petite ferme “colonaire” était propre à l’époque de la conquête et disparaissait avec la crise des guerres puniques, tandis que la villa triomphait entre second siècle avant notre ère et second siècle après, pour laisser ensuite place soit aux villae palatiales, soit aux villages de l’Empire tardif et du Moyen Âge. Aujourd’hui cette approche semble s’estomper progressivement et la recherche sur les villae archaïques ou sur le village romain républicain ou sur la petite ferme impériale changent peu à peu notre vision des campagnes romaines. En prospection, les datations de sites répondent à des critères quelque peu différents des critères utilisés en archéologie stratigraphique. Le contexte d’une couche, son identification comme couche de destruction, de remblai, d’occupation, ne s’appliquent pas. C’est pourquoi, les dates dévolues aux sites ne peuvent que tenir compte de la date de la céramique la plus ancienne et de la date de la céramique la plus récente. Mais ces données schématiques sont corrigées par le poids des échantillons, la répartition des quantités de matériel propre à chaque époque considérée. La cartogra237

RPC 2000 conference proceedings

phie historique des sites peut en être singulièrement affectée2. Cependant, cet inconvénient est compensé par l’aspect statistique des données : les erreurs éventuelles introduites dans la datation d’un site ne peuvent se reproduire sur tous les sites voisins. C’est pourquoi, l’aspect quantitatif est la première chose que nous allons examiner. Le tableau ci joint montre la répartition des sites entre deux dates (500-250 avant J.-C.) qui n’ont valeur que de repère abstrait et aléatoire et n’ont de caractère ni archéologique ni historique. Ces dates ne correspondent pas à un certain type de céramique (comme par exemple pourrait l’être, dans les années 515/500, le matériel de la phase VI des fouilles de Capoue qui permet de voir le passage du bucchero à la céramique à vernis noir campanienne, ou comme, dans les années 250/180 avant J.-C., la forme 2732 d1 des classifications de Morel pourrait permettre de cerner le passage du III° au II° siècle avant J.-C.) (Johannowsky 1983, n. 48, p. 67, mais avec des attestations du bucchero jusque vers 450.) Elles ne sont pas non plus liées à l’intervention de telle ou telle puissance régionale en Campanie (Grecs, Etrusques, Samnites, même si on pourrait utiliser la date de 497 pour fixer les rapports entre Romains et Grecs de Cumes, ou de 524 pour dater la présence étrusque à Capoue, et le conflit entre Grecs de Cumes et Etrusques de Campanie, ou de 473 pour évoquer les rapports entre Grecs et Samnites, ou encore de 340/296 pour tenir compte des phases initiales de l’occupation romaine) (Frederiksen 1984, chapitre 3, 4, 5, 6; Humbert 1978). Les dates aléatoires et le long terme ont l’avantage d’éviter la sur interprétation de données archéologiques à la lumière d’événements historiques. Tout d’abord on peut remarquer le remplissage régulier de l’espace et le fort dynamisme du IV° siècle. Si l’on passe de 35 à 62 sites en 250 ans, le IV° siècle à lui seul voit 39 créations de sites soit “un pic” considérable par rapport au siècle qui le précède (2 fois moins de créations). Le demisiècle qui suit, entre 300 et 250 est presque aussi dynamique puisqu’une quinzaine de bâtiments nouveaux apparaissent en 50 ans. C’est un site nouveau qui naît tous les 6 ans aux V° siècles (15 en 100 ans), mais ce en est plus de 1 tous les 3 ans au IV° siècle (39 en 100 ans), comme dans la première moitié du III° siècle (15 en 50 ans). La perception de l’espace par les propriétaires ou exploitants qui s’installent en est profondément modifiée. Le IV° siècle donne l’impression d’une effervescence dont on peut se demander si elle n’est due qu’à des facteurs internes (les Campaniens, les indigènes acquérant de nouvelles maisons rurales) ou si elle puise dans la migration, la colonisation romaine qui historiquement se renforce entre 340 et 318 de deux colonies, à Suessa et Cales, et d’un ager assignatus dans le Falerne. Mais les chiffres cachent bien des aspects de la réalité. Passer de 35 à 47 puis de 47 à 59 et enfin de 59 à 62 sites ne s’est pas fait de façon linéaire car s’il y a eu des créations, il y a

2

eu aussi des disparitions de sites. Au moment où l’on passe de 35 à 47 sites, s’il y a 15 sites nouveaux, soit 1 tous les 7 ans, cela signifie aussi qu’il y a eu 3 disparitions. Lorsque l’on passe de 47 à 59 sites, si 39 sites sont nouveaux, cela signifie que 27 ont disparu, soit 1 tous les quatre ans. Il en va de même entre 300 et 250, car les quinze créations sont accompagnées de 12 disparitions soit 1 tous les quatre ans également. Cette mobilité dans l’occupation de l’espace est importante à saisir. Elle donne une vision géographique, spatiale, économique, mais aussi sociologique du rapport à la terre. Le remplissage du territoire est moins rapide que ne le laisserait présager le nombre de sites nouveaux. Mais, la construction d’habitats ruraux, l’installation de nouvelles familles de propriétaires ou d’exploitants, ou l’acquisition et la gestion des terres qui entourent villae et fermes, par de vieilles familles désireuses d’investir provoquent une tendance à la pression foncière, à la faim ou non de terres que le seul examen des vestiges archéologiques traduit imparfaitement. Géographiquement, on ressent, du haut des collines campaniennes quelle autre maison rurale les acquéreurs perçoivent à leur horizon plus ou moins lointain, en cette fin du V° siècle. Cette mobilité du paysage montre avec quel voisin on peut s’allier, entrer en conflit, échanger des biens, des produits, éventuellement des dots. Elle ne rend pas compte pour autant de la diversité des formes de l’implantation dans le territoire, phénomène que nous allons tenter de détailler maintenant. La base de données de l’ensemble des sites campaniens (414 enregistrements, dont 272 pour des sites identifiés comme villae, fermes, villages, sanctuaires, nécropoles, fours de potiers ou ateliers d’amphores) permet d’appréhender ainsi l’installation des 35 bâtiments ruraux et sites des années 500. Habitat rural L’habitat est composé d’une villa, d’une ferme et d’un village, respectivement Cortigiani (M54), Monte Cicoli (M27) et Masseria Irace (S12). Dans le cas de Cortigiani, il est difficile de dire, en l’absence de fouilles, quelle forme prend la villa à cette époque et si le matériel qui permet une datation haute de l’occupation du site est ou non la preuve que le site est, dès cette époque, une villa, ou s’il ne l’est devenu que plus tard, au cours de l’époque républicaine de son occupation. Cette problématique de la villa archaïque a longtemps été écartée, mais les fouilles récentes de la périphérie de Rome (Carandini et al. 1997) tendraient à faire remonter assez haut la chronologie de tels sites. Les prospections et fouilles des marais Pontins et la reprise des prospections et fouilles dans la Vallée du Tibre apporteront d’utiles données à cette question. Dans la vallée du Biferno, Capracotta et Matrice semblent mériter la dénomination de villae “Samnites” plutôt dans les années 350 avant J.-C. qu’avant (Barker 1995). La villa de Cortiginani, construite en opus polygonal et située à 161 mètres d’altitude, ne

Les références aux sites campaniens que l’on trouve dans cet article reprennent celles de notre base de données et de nos cartes. La cartographie et la nomenclature de la Campanie septentri-

onale sont divisées en 5 grands secteurs correspondant aux cartes topographiques de Carinola (C), Mondragone (M), Suessa (S), San Andrea (SA), Pignataro Maggiore (PiMag).

238

Landscape Perception in Archaeology

Figure 3: Rural settlement in Campania: Cales region

paraît pas occupée après les années 100 de notre ère. La ferme de Monte Cicoli, à 220 m. d’altitude, manifeste une occupation qui semble continuer jusqu’à la fin du MoyenÂge, tandis que le village d’Irace, à 60 m d’altitude, est abandonné dès la fin du III° siècle avant J.-C. Ces trois évolutions différentes ont, en partie une explication dans leur rapport au territoire: Cortigiani fait partie d’une crête calcaire sur le flanc occidental et méridional du Massique qui a connu tout au long de la période protohistorique et historique une densité d’occupation sans cesse croissante, même si créations de sites et abandons ne se sont pas effectuées de façon linéaire. La rupture de pente, la possibilité d’exploiter aussi bien les forêts et bosquets du Massique et du Petrino que les pentes à la terre riche et bien drainée, près de sources et au flanc de collines qui bien qu’escarpées sont facilement accessibles depuis la plaine de Carinola, ont sans doute largement contribué à l’implantation de Cortigiani. D’ailleurs, tout près de cette villa, à 200 m. et à 130 m., à l’époque protohistorique puis au début de l’époque romaine d’autres signes d’occupation sont attestés. Il y a donc là, à la fois, un site et une situation privilégiés. On peut en dire presque autant de Cicoli, point de contrôle et de passage entre Latium et Campanie, surplombant l’endroit où vient s’implanter, en 296, la colonie de Sinuessa. La

ferme a joui de l’approvisionnement en eau par les sources de collines, par des résurgences et par la construction d’un aqueduc romain situé à 100 m. d’altitude qui court tout au long de cette courbe de niveau. Bien adossée aux collines de l’Incaldana et du Petrinum, où Cicéron aimait à jouir de son “deversorium”, la ferme est en étroit rapport avec les autres sites de la périphérie de la colonie romaine. Massacré par la spéculation immobilière et l’implantation illégale de l’hôtel Bagni Solfurei, ce site mériterait une étude fine pour pouvoir déterminer dans quel contexte il a été occupé pendant des centaines d’années, et quelles relations il entretient avec les autres fermes et villae d’époque plus tardive, ainsi qu’avec les établissements thermaux réputés pour leurs eaux sulfureuses dans l’Antiquité (Bagni Solfurei, M37; Pizzuto, M38). Quant au village de Masseria Irace, sa naissance, son développement, sa disparition posent tout le problème de l’habitat indigène, de la place de Suessa au IV° siècle, du rapport entre village pré romain et ville romaine au début de la conquête. Irace semble avoir précédé l’implantation de la colonie et avoir disparu lorsque le prestige et le développement de celle-ci à la fin du III° siècle lui ont porté ombrage. Mais on pourrait aussi y voir l’aspect conjoncturel des guerres puniques et du passage d’Hannibal “sur les crêtes riantes du Massique”, voire la concurrence 239

RPC 2000 conference proceedings

de la Via Appia qui, le long de la côte, et non plus par l’intérieur de la Campanie, assure au début du III° siècle, de part la naissance de la fondation de la colonie de Sinuessa, le développement de la façade occidentale de la plaine de Cellole, jusqu’à Sinuessa. Ceci est d’autant plus tentant que tout le piémont septentrional et occidental du Massique semble connaître assoupissement voire déclin, peut-être lié à une crise oléicole, dans cette période du milieu de la république, alors que le flanc méridional débute un essor vertigineux auquel le vin n’est pas étranger.

au cours des siècles suivants). La plupart disparaissent vers la fin du III° siècle avant J.-C. (Crocelle, San Guido, Cofanari, Veterolla, Arivito). Ces nécropoles n’ont souvent livré qu’un matériel pauvre, d’une ou deux sépultures. Leur répartition géographique est d’abord caractérisée par l’occupation de sites de faible altitude, en général au-dessous des 150 m. et souvent en plaine entre 5 et 30 m. d’altitude. C’est un aspect différent de la position des trois habitat étudiés qui, eux, sont tous au-dessus des 60 m. voire des 200 m. d’altitude. La situation des nécropoles est ensuite liée à la proximité de l’axe routier du Nord du Massique, celui qu’emprunte l’Appia primitive, par Suessa et Cascano, et qui reprend vraisemblablement un parcours protohistorique sur lequel se trouve également le village d’Irace (Ponte Ronaco, Rio delle Cammarelle, Domenico, Casanova). Mais d’autres tombes se trouvent dans la plaine de Carinola, entre Mont Massique et fleuve Savone (Francesco, San Guido, Cofanari) ce qui peut paraître plus étonnant, puisque au V° siècle aucun habitat ne paraît exister dans ce secteur (Cortigiani est nettement plus au sud et plus en altitude) et que les voies de desserte des champs ne sont pas connues. On peut se demander, comme c’est d’ailleurs le cas pour les habitats éventuels de colons de l’ager falernus au IV° siècle, si les traces de l’habitat ne sont pas situées sur de basses terrasses du fleuve, masquées aujourd’hui par des terrasses plus récentes, ou enfouies sous des épandages de colluvions arrachées au Massique. Les bonifications d’époque moderne dans la zone de Carinola ont tellement transformé le paysage, reconstitué des pentes de façon à ce que les affluents du Savone puissent rejoindre le fleuve, que la recherche sur les nécropoles et habitats archaïques, dans ce secteur, ne peuvent se satisfaire ni des prospections superficielles ni même aériennes qui ont été menées. Il en va de même des sites funéraires de Crocelle, Lenze, Veterolla. Lenze est d’ailleurs le site où l’on a identifié une dizaine de sépultures, et où l’occupation paraît limitée au VI° siècle. Même si, en général, en Campanie, les nécropoles et les tombes éparpillées sont bien mieux étudiées pour la période archaïque que les habitats, dans les dix cas que nous présentons, aucune fouille ne permet de tirer plus de conclusions sur la nature culturelle et cultuelle des sépultures, encore moins de déterminer si elles appartiennent à des indigènes, à des Samnites, voire à des Grecs ou à des Etrusques. Enfin l’un des sites est un sanctuaire, Panetelle (M159) (Johannowsky 1970-1971, 1974)4. Ce temple d’embouchure, comme celui de dea marica sur le Garigliano, détermine peut-être à la fois une frontière occidentale et septentrionale du territoire campanien, marquages face à la

Autres sites Vingt et un sites sont de nature indéterminée, c’est-à-dire que l’on y trouve de la céramique éparpillée mais pas de structures permettant une identification précise du site. La proportion de 21/35 sites au total soit 60% n’a rien de surprenant si l’on considère que sur 414 sites toute période confondue, 272 soit 68% sont de nature indéterminée. C’est une proportion que l’on retrouve dans la plupart des prospections de grande envergure, en Etrurie, dans le Biferno, par exemple, en Grèce ou en Syrie également, et seule une combinaison de sondages et de fouilles peut abaisser ce chiffre et permettre de mieux connaître la nature des sites. Les sites des années 500 ne sont donc pas moins facilement identifiables que ceux des périodes suivantes. En revanche on manque de typologie comparative et de corpus pour pouvoir détecter des formes d’occupation du sol et d’habitat qui ne répondent pas aux critères connus pour l’époque romaine (pierre, maçonnerie, tuile, signinum et mosaïque). Les “cabanes”, l’habitat de pisé, avec des poteaux, au sol en terre battue ne rentrent que progressivement, et au fur et à mesure que l’on se pose la question de les rechercher, dans les critères d’identification pour l’habitat d’époque pré romaine (Gasperetti & Crimaco 1993, 1995-1996)3. Une dizaine de sites sont des nécropoles ou tombes (Ponte Ronaco S16, Rio delle Cammarelle S24, Domenico C25; Casanova C27, Francesco C31, San Guido C44, Cofanari C45, Arivito M170, Crocelle M14, Lenze M94, Veterolla M 152) ce qui là encore reste une proportion tout à fait comparable à l’ensemble de l’échantillon des 414 sites. En général la période d’occupation de ces nécropoles paraît relativement plus court que celui des habitats que nous venons d’examiner car nombre d’entre eux, occupés dès le VIII° siècle ne le sont plus après le V° siècle (Lenze, sans doute disparu à la fin du VI° siècle, Francesco, Domenico, Casanova, où seul ce dernier site peut sembler se prolonger avec un léger déplacement de quelques dizaines de mètres

3

D’une part, une cabanne du VI° siècle abandonnée dès 550, sur le site de Sinuessa montrerait qu’à la différence de la zone de Suessa, la côte près de Sinuessa, occupée par des populations aurunques avait déjà été abandonné lorsque la colonie romaine s’installe. Mais les chercheurs font remarquer que «l’exiguïté de la fouille» rend prémature toute conclusion. D’autre part, une épave de bateaux et des aménagements à –9 mètres sous la plage actuelle dans les villae maritimes romaines de ce secteur tendent

4

240

à prouver l’ampleur du changement de paysage intervenu entre l’époque préromane et romaine. La situation du temple de Panetelle par rapport à la ville de Sinuessa et l’étude du castrum de la colonie mettent en évidence une ligne de côte entre San Limato, Panetelle, et les Vasche plus éloignée qu’aujourd’hui du site antique, où il est partiellement englouti. Une nécropole s’étendait d’ailleurs à l’ouest de la ville jusqu’au I° s. av. J.-C.

Landscape Perception in Archaeology

mer et marquage face au Latium. Son occupation dure jusqu’au milieu du premier siècle de notre ère comme nous le verrons plus loin. La situation du V° siècle avant J.-C. correspond à un territoire pré romain encore relativement peu maîtrisé, où l’habitat est rare. De l’un à l’autre, les habitants ne se perçoivent guère. Aucun site relevant de la production de céramique, de tuiles, d’amphores n’est reconnu. Espaces en friches, forêts et bosquets, étendues dévolues aux marais, zones inondables du Garigliano et du Volturne, du Savone et des ruisseaux adjacents (Fontanelle, Rio Roda, etc.), ligne de cote plus à l’Ouest qu’aujourd’hui, plage plus basse, modifieraient largement la perception du paysage qu’ont eu les Romains. Cette situation ne serait pas contradictoire avec les données recueillies dans certaines fouilles de Pompéi où la forêt, le hêtre notamment, paraît descendre jusqu’au niveau de la mer, indiquant un climat plus humide et une densité végétale arborée encore forte aux VI° et V° siècle avant J.-C (Bonghi Jovino 1985)

urbains consommateurs. On ne sait, bien entendu, et on ne saura sans doute jamais, sauf inscription sur tuiles, sur amphores, ou sur tout autre support aisément identifiable comme appartenant à la villa, quelles sont les familles qui se sont installées entre V° et III° avant J.-C. Ces familles sont elles indigènes, romaines, romano-indigènes? Sont –elles restées propriétaires des villae aussi longtemps que le laissent entendre les traces d’occupation? Bien entendu il est tentant de penser que des propriétaires tôt installés ont gardé longtemps leur patrimoine rural, traversant les crises conjoncturelles, qui, nous allons le voir affectent d’autres sites. En revanche, ce que l’on peut constater c’est l’importance de la situation de ces villae: deux d’entre elles seulement sont placées dans la plaine, à 20 et 45 m. d’altitude (Francolise et San Rocco). Les sept autres ont privilégié les cotes altitudinales élevées, entre 150 et 270 mètres. Quatre correspondent au territoire de l’ager calenus (Calvi Risorta avec deux sites, Giano Vetusto et Francolise avec un site respectivement), un au finage de Sinuessa (San Rocco). L’appartenance des autres est plus difficile à établir. L’exemple de San Rocco est intéressant car si le matériel le plus ancien retrouvé sur ce site date du début du V° siècle, cela signifie que l’on a, là, sur le flanc occidental du Massique un site qui, occupé ensuite jusque vers 250 après J.C., se présente peut-être déjà sous la forme d’une villa, ou au moins d’une ferme, avant la fondation de la colonie de Sinuessa. Elle renforcerait, vers le sud, la présence déjà attestée de la villa de Monte Cicoli. San Rocco et sa voisine pourraient appartenir donc soit à une forme indigène d’occupation du sol, de la part des Aurunques, soit au cadastre de Suessa, fondée au milieu du IV° siècle et sur le territoire de laquelle vint s’implanter, ensuite, vers 296, Sinuessa. La position aux marges du finage, à la fois au sud et sur les premiers contreforts du Petrino, aux pieds du Massique, pourrait accréditer l’idée que la zone est encore “infesta regio” comme l’écrit Tite Live, dangereuse, parce que marécageuse ou/et de possession conflictuelle entre indigènes et colons romains. De plus, l’Appia passant encore pour quelques décennies par Suessa et Cascano, au nord du Massique, la villa serait assez isolée, séparée des colons du Falerne par les collines escarpées. Quant aux villae de Cortigiani, Finocchiaro, Cesariello, Gran Celsa leur appartenance aux trois territoires de Sinuessa, Suessa ou de l’ager falernus peut se discuter et elles peuvent dépendre aussi bien de l’un que des autres. En effet, si la ligne de partage des eaux et la disposition des crêtes montagnardes ferait plutôt pencher pour Forum populi et l’ager Falernus, les textes des gromatici veteres5 montrent de tels conflits pour le partage de ces zones entre les colonies de Sinuessa et Suessa qu’il n’est pas certain qu’au début de la république et de l’occupation romaine ce ne soit pas à Suessa, entre 340 et 296, puis à Sinuessa, après 296, que cette fraction de collines ait été attribuée, alors que Forum Populi et les assi-

7.5.3

Rythme et type d’implantation humaine au III° siècle avant J.-C. Si l’on examine la situation deux siècles plus tard, vers 250, alors que trois colonies romaines ont été fondées à Cales, Suessa et Sinuessa, que l’ager falernus est assigné, et que la cité de Teanum est probablement alliée de Rome, on constate que le territoire septentrional de la Campanie est désormais occupé par 62 sites, soit deux fois plus que précédemment. Bien entendu, cette densification s’est accompagnée, là encore, d’abandons. Mais la qualité de l’occupation rurale a également beaucoup évolué. Habitat rural Neuf villae font désormais partie du paysage. Si celle de Cortigiani (M 54) est encore occupée et le reste trois siècles, les nouveaux sites de San Rocco (M 115), Cesariello (C 22), Finocchiaro (C 29), Gran Celsa (S 37), Calvi Risorta (2 villae, PiMag 12 et 16), Giano vetusto (PiMag 13), Francolise (PiMag 24), ont tous eu une espérance de vie de plusieurs siècles, et n’ont disparu qu’entre la fin du II° siècle et le milieu du VI° siècle de notre ère. C’est une donnée tout à fait originale de nos recherches que nous n’avions pas perçu en 1979 et que P. Arthur, en reprenant nos travaux, n’avait pas mis en avant non plus (Vallat 1987, Arthur 1991). Or, il semble que les sites de villae occupés les plutôt soient parmi ceux qui ont la durée de vie la plus longue. Nous avons émis à ce sujet diverses hypothèses qui ne s’excluent d’ailleurs pas les unes les autres. Tout d’abord, les installations nouvelles prennent place dans un paysage “peu rempli” où le choix des terres doit rester relativement aisé, la pression foncière n’étant pas encore à son maximum. Terroirs divers, bonnes positions par rapport aux routes, proximité relative des colonies et des centres

5

Frontin, La., 2, 48 “Ut in Campania in Suessano culti habent in Monte Massico plagas silvarum determinatas”. Ce texte révèle très vraisemblablement une situation du Bas Empire, le cas

épineux de Suessa étant aussi évoqué par Agennius Urbicus, La. 79, 13 tandis que le texte général de Frontin, La. 15, 1-8 peut très bien s’appliquer à cette cité :”silvae in montibus”.

241

RPC 2000 conference proceedings

gnations du Falerne devaient se limiter à la plaine “usque ad Volturnum”. Ainsi, les villae apparaissent comme étroitement liées au statut des territoires indigènes ou coloniaux mais aussi aux qualités agricoles variées des terroirs où elles s’implantent, et enfin à la recherche d’abri face à des inondations, de meilleures terres de coteaux face à des plaines insalubres et marécageuses, voire en quête de position défensive face à des incursions possibles d’ennemis. Leur construction est soit en opus polygonal (Cortigiani, Gran Celsa) soit en incertum. Une seule dispose d’un pressoir (Finocchiaro) dont la date d’installation ne peut être précisée en l’état actuel des recherches. Cinq sites sont des fermes: Cicoli, déjà en place durant la période précédente, Giano Vetusto (PiMag6 et 7), Terragrande (M 69), Creta (C 18) et sont situées entre 140 et 225 m. d’altitude, sauf Terragrande (50m). Elles appartiennent aux territoires de Cales, de Sinuessa, du Falerne, de Sessa. Elles participent à un remplissage progressif du paysage, à une avancée de la maîtrise agricole des terres. Comme les villae, les fermes construites à cette époque restent occupées jusque sous l’Empire. Cicoli reste occupée jusqu’à la fin du Moyen–Âge. Trois autres sites ne disparaissent qu’entre le I° siècle (Creta, Giano Vetusto PiMag7), et le II° siècle après J.-C. (Giano Vetusto PiMag5). Seule Terragrande disparaît antérieurement, sans doute durant le second triumvirat. Dans les cas de Terragrande (villa placée à 50 m. d’altitude), on peut expliquer la durée de vie soit en invoquant la difficulté de la recherche liée à l’alluvionnement et aux colluvions du secteur de plaine entre Carinola et le fleuve Savone, soit en tenant compte de la fragilité de ce paysage très lié à la main de l’homme et où l’eau et l’érosion aérienne peuvent rendre précaire la présence de l’occupation rurale. Les fermes de l’ager calenus (Giano Vetusto) sont à l’abri de tels phénomènes, placées haut en altitude (225 et 250 m.) dans un territoire où la densification se poursuit tout au long de la République, sous la forme de villae (Calvi Risorta PiMag19) ou de fermes (Rochetta e Croce, PiMag9; Calvi Risorta, PiMag5). Cependant, dans l’ensemble, la durée de vie des fermes est plus courte que celle des villae, même si on doit constater que sondages et fouilles sont trop lacunaires pour saisir combien de fermes se sont éventuellement transformées en villae, phénomène que l’on perçoit aussi bien en Etrurie que dans la vallée du Biferno. En revanche, la fragilité des fermes campaniennes par rapport aux villae ne paraît pas contestable. Trois villages sont désormais présents: deux sont situés autour de Sessa et un aux limites entre ager sinuessanus et ager falernus. Près de Sessa, il s’agit d’Irace qui est toujours là mais disparaît à la fin du III° siècle, alors que Cascano (C5) nouvellement créé, sans doute vers 300 avant J.-C., à 210 mètres d’altitude, manifeste des traces d’occupation jusqu’au début du VI° siècle de notre ère. Enfin, au sud, le Cimitero di Mondragone (M124), à 30 m. d’altitude, est occupé sans doute au même moment et lui aussi reste occu-

pé mais jusqu’au milieu du VI° siècle. C’est le site du pagus sarclanus de la République et de l’Empire qui émerge, là. Géographiquement et sociologiquement, on peut constater quelques phénomènes intéressants. Dans la proximité de la colonie de Sessa, les villages d’Irace et de Cascano ainsi que la colonie de Suessa, s’articulent désormais avec un front pionnier vers le sud et l’est où des fermes peu nombreuses (Creta) et des villae plus conséquentes (Cortigiani, Finocchiaro, Cesariello, Gran Celsa) remplissent l’espace interstitiel. Désormais, les propriétaires et occupants de ces sites se voient d’un habitat rural à l’autre, ce qui n’était pas le cas deux siècles plutôt. Mais, on peut supposer que le terroir est encore largement mis en valeur à partir des habitats groupés villageois et urbains. En revanche, autour de Sinuessa, colonie qui, rappelons-le vient juste d’être créée au moment où nous avons choisi d’arrêter cette étude (dans les années 250) et dans l’ager falernus (en l’absence de certitudes archéologiques quant à l’occupation de Forum Populi avant les années 200 avant J.-C.), le village de Cimitero di Mondragone, les villae de Cortigiani et San Rocco, viennent marquer le paysage avec dans leur champ de vision les fermes de Terragrande et Cicoli. Le système apparaît plus équilibré que près de Sessa: villages, fermes et villae entrent sans doute plus rapidement en concurrence pour l’accès aux bonnes terres, et le poids des colons de l’ager falernus et sinuessanus a dû être considérable dans l’articulation entre les différentes formes de mise en valeur des terroirs. Mais, la situation dans les années 500-300 était telle que le paysage apparaissait alors presque vide, et que les années 300 font figure de véritable front pionnier à développement extrêmement rapide. Enfin, autour de Cales, fermes et villae occupent plutôt les hauteurs orientales du finage de la colonie. Seule la villa de Francolise tient une position de plaine (Cotton et Métreaux 1985, Cotton 1979)6 entre ager falernus et ager calenus de la plus haute importance sur le plan de la qualité des terres agricoles, mais aussi des voies de communication conduisant de Sinuessa et Suessa vers Cales et Capua. Il ne semble pas, en l’état actuel de nos connaissances que des villages anciens ou d’époque romaine soient attestés dans ce secteur nord oriental de la Campanie. Est-ce dû à des raisons historiques tenant aux populations Aurunques, Ausones, Sidicines? On pourrait alors se demander pourquoi elles ont eu un comportement différent en territoire de Suessa Aurunca. Est-ce dû à la disparition très rapide du seul village un tant soit peu attesté, occupé entre l’Âge du Bronze et l’Âge du Fer (Calvi Risorta, PiMag 4 altitude 130 mètres) (Compatangelo 1985, site 4, page 22)7? Est-ce dû au poids rapide de la colonie de Cales et au rôle urbain indigène que joue aussi, à peine plus au nord, la ville de Teanum Sidicinum (Teano)? Les familles qui détiennent ou cultivent les terres afférentes à l’ensemble de ces habitats ruraux, qu’elles soient anciennes ou récentes, indigènes, romaines ou romano-

6

7

Le site étudié ici fait partie de ceux qui précèdent l’installation des villae de San Rocco et Posto.

242

La nature du matériel, l’extension de l’épandage sur 5300 m2, rendent difficile l’interprétation du site.

Landscape Perception in Archaeology

indigènes ont, au milieu du troisième siècle, une position tout à fait privilégiée dans la communauté rurale. La pression foncière n’est pas encore très forte, même si, sur certains coteaux, les habitats s’échelonnent désormais tous les 500 mètres. L’étagement des maisons rurales n’est pas encore tel que l’on ne pourra pas jouir à la fois de terroirs variés. L’eau est mieux maîtrisée grâce à la colonisation des plaines par les trois grandes colonies et par les colons de l’ager falernus. Quel est le rapport entre occupation de terres par les villages et les villes et propriétés foncières des villae et fermes? Y-a-t-il concurrence ou complémentarité? Quel rôle jouent les colonies dans la confiscation des terres des vaincus et dans le lotissement des colons? Les colons sontils en mesure d’exploiter leurs lots à partir des villes coloniales ou des villages? Quelles sont les hiérarchies qui se sont mises en place entre villae et fermes? Une distinction s’est-elle peu à peu établie entre ceux qui ont gardé leurs lots, ceux qui les ont vendus, ceux qui, grâce à un patrimoine familial antérieur, ont acheté d’autres terres, expliquant l’éclosion de fermes et de villae, en dehors des hameaux et des villes de colonisation?

même titre que l’Heraion du Sele ou le temple de Dea Marica sur le Garigliano (Mingazzini 1938), l’effacement de la limite entre Latium et Campanie rend vraisemblablement caduque la position de ce sanctuaire. Mais il peut y avoir aussi dans cet abandon une raison cultuelle, culturelle liée à la désaffection pour le plus ancien des temples indigènes ou campaniens de ce secteur géographique. En revanche, celui de Soppegna (C 48)8, né dans les années 400 et disparu vers 150 avant J.-C., appartient à cette floraison de sanctuaires qui de Capoue à Cales et Teano irradie tout l’intérieur de la Campanie, et sur laquelle la documentation se fait de plus en plus complète (Blazquez 1961, 1963, 196869, Losada Nunez 1993, Ciaghi 1983, bibliographie faite à partir de ce livre, Scatozza Höricht 1987, Della Torre & Ciaghi 1980). Dans le cas de Soppegna, l’abandon doit être compris en fonction d’une explication globale de l’évolution cultuelle et culturelle de l’ère orientale de la Campanie, de la concurrence entre Romains, Sidicins et Campaniens, de la place respective des cultes urbains et ruraux. D’ailleurs, dans le territoire de Cales, les sanctuaires de Fontana Regina (PiMag 29), à 110 mètres d’altitude, abandonné vers 150 avant J.-C. également, et de Rochetta et Croce (PiMag 2) à 220 m. d’altitude, abandonné avant la guerre sociale, ne résistent pas plus que celui de Soppegna. Seul Montemaggiore (PiMag 30), à 530 mètres d’altitude, est fréquenté longtemps, sans doute jusque vers 200 après J.-C. Est-ce en raison de son isolement, de sa quasi-inaccessibilité, d’un culte particulier? Il est difficile actuellement de répondre.

Autres sites Alors qu’au V° siècle on avait compté vingt et un sites indéterminés, ils sont trente au III° siècle (30/62) soit 50% contre 60% précédemment (21/35) et contre 68% dans le total de l’échantillon (272/414). Les sites du III° siècle sont donc relativement bien connus et les structures de plus en plus identifiables, notamment avec le développement de l’opus polygonal et de l’incertum, des sols en signinum, des cisternes étanches, des tuiles, voire avec l’apparition d’éléments de décor de plus en plus nombreux (mosaïques, colonnes). Aussi, les sites où n’apparaît que de la céramique éparpillée sont –ils de moins en moins nombreux. Si l’on avait identifié antérieurement une dizaine de nécropoles, ce chiffre reste pratiquement constant, en très légère augmentation, sans doute en relation avec l’augmentation de la population que nous constatons par la densité des sites ruraux et l’éclosion de trois sites urbains et de quelques villages. Elles sont d’ailleurs réparties très également entre plaines et collines, près des villes, villages, fermes et villae, et traduisent donc bien l’image de “remplissage” des territoires que nous avons perçues en examinant l’habitat rural et les sanctuaires campaniens. Enfin, l’unique sanctuaire du V° siècle est désormais renforcé de cinq autres. Panetelle (M159), dont l’occupation se poursuit jusqu’au milieu du premier siècle après notre ère, disparaît peut-être au moment où l’extension de Sinuessa vers le sud, la floraison de la zone artisanale le long de la côte et l’organisation semi-urbaine, semi-villageoise du territoire autour de Forum Populi, de vicus sarclanus, de vicus caedicius, d’Urbana ne rendent plus utile le maintien du sanctuaire d’embouchure du Savone? Si Panetelle est un site d’embouchure et de frontière, peut-être, au

8

7.5.4 Conclusion En ce milieu du III° siècle, 15% des sites sont des villae (9/62 sites) alors que, deux siècles auparavant, elles ne représentaient que 3% du total (1/35). Le phénomène de la villa ne connaît pas encore l’ampleur qu’il prendra dans ce secteur de la Campanie entre 100 avant J.-C. et 100 après J.-C. (90/272 sites soit 30%) mais il est amorcé. Il y a cinq fois plus de fermes que précédemment. Elles représentent 8% du total des sites (5/62). C’est beaucoup moins qu’à la fin de la république (63/272 sites soit 25%) mais là encore, comme dans le cas des villae, le mouvement a débuté. En revanche, leur présence est plus fragile que celle des villae. La plupart des sites de fermes nées au début de la République ne passent pas le cap du premier siècle de notre ère. Seule Cicoli, la première créée, comme on l’a vu plus haut, demeure dans la longue durée de plus d’un millénaire. C’est une nouvelle confirmation, semble-t-il, de la prééminence des sites les plus anciens au sein des finages. A partir de 250 avant J.-C. et sans doute encore plus peu de temps avant la guerre sociale, l’appréhension du territoire se fait totalement différente. Dans un espace désormais densément occupé, où émerge un site tous les 500 mètres et parfois beaucoup moins, où chaque crête est occupée, où les sites s’étagent de la plaine au sommet des collines entre la mer et les contre-

La fouille du sanctuaire de fondo Ruozzo à Teano dont le toponyme est Soppegna a été présentée succinctement par Morel 1992.

243

RPC 2000 conference proceedings

forts de l’Apennin, la concurrence pour la terre se fait âpre, violente, et les sites nés à la fin de la République ont souvent une durée de vie courte. Il semble qu’il en soit ainsi parce qu’installés sur les dernières terres libres, entraînés dans le jeu des spéculations viticoles, oléicoles ou maraîchères destinées aux villes, et à un marché de luxe, les exploitations agricoles ont beaucoup moins bien résisté que les habitats installés à la fin de l’époque indigène et au début de la colonisation romaine. Nous avons analysé ailleurs, et avec d’autres arguments tirés des sources littéraires et épigraphiques, cette hypothèse de l’organisation agraire et du fonctionnement des rapports à la terre. Nous voulions ici seulement en rappeler et approfondir les éléments à propos de la période de transition entre protohistoire et période historique. Dans le cadre de notre colloque, le nouveau paysage et les nouvelles implantations rurales que nous venons de percevoir peuvent se raccrocher à une perspective plus large qui englobe l’ensemble des processus de transformation qui touchent le sud du Latium et le nord de la Campanie entre V° et III° siècles avant notre ère. En 312 l’Appia affronte deux grandes zones de marais (ager pomptinus et ager falernus) et doit tenir compte de la colonisation de Terracine (vers 329-318 avant J.-C., tribu Oufentina), de Suessa (tribu Aemilia), de Minturno, de Sinuessa (tribu Teretina) entre 313 et 296 avant J.-C., et de l’ager falernus (340-318 avant J.C., tribu Falerna). Or, les recherches dans la zone pontine d’Aricia/Cisterna (Quilici 1992) montrent que la colonisation est antérieure à la voie, infirmant les recherches de De la Blanchère qui pensait que l’Appia s’installait en territoire désert et abandonné, en déclin depuis l’installation des Volsques. Partout le tracé de l’Appia est postérieur à l’œuvre de bonification (Ponte di Mele). Cette zone a connu des sanctuaires, des sites sans rapport avec l’Appia, à l’époque de la République naissante. Dans la zone de Sinuessa la situation n’est pas loin d’être la même. L’Appia est tardive, liée à la fondation de la colonie au début du III° siècle. La centuriation de l’ager falernus là encore semble antérieure à l’Appia: la voie a pour assise un banc de tuf, puis elle longe le flanc sud du Massique, avant de remonter vers le Nord au lieu de filer droit vers Capoue. Conformément à ce que dit Tite Live elle permet sans doute de bonifier l’infesta regio et a pu servir de levée et de digue. A l’époque de Domitien les forêts sont encore nombreuses et la construction de la voie qui désormais sous le nom de domitienne va longer la côte nécessite une grande quantité de bois, ce qui rend indispensable l’attaque du massif forestier du Mons Massicus et d’autres collines. Sans doute jusqu’à Liternum et Cumes, des travaux pour contenir les eaux du Savone et du Volturne ont-ils été nécessaires, comprenant digues et levées de terre comparables aux travaux engagés dans le secteur de Terracine quelques siècles plutôt. Or, les fouilles récentes sur le littoral de Sinuessa (villa Palmieri, Tranzo etc.) ont montré que les niveaux du III° siècle sont à – 9 mètres du niveau de sol actuel et à – 3 m du niveau de rivage. La ville de Sinuessa connaît entre III° et II° siècle avant notre ère une expansion péri urbaine, un essor de villae maritimes (avec jardin et irrigation) antérieur à l’essor artisanal des fours d’amphores Dressel 1 B, soit dans les années qui précèdent 135 avant J.-C. Travaux de bonification, déforesta-

tion, digues, occupation du territoire par un habitat rural diversifié et hiérarchisé, mise en valeur très diverse des terroirs, attestent de l’intérêt que les Romains ont porté à ces zones du Sud du Latium et du Nord de la Campanie, qui étaient loin d’être des déserts, lors de leur confiscation par Rome!

References Arthur, P. 1991, Romans in northern Campania, British School at Rome 1, Rome. Barker, G. 1995, A Mediterranean Valley, Leicester. Bintliff, J. 1991, The Annales School and Archaeology, Leicester. Blazquez, J.-M. 1961, Terracotas del santuario de Cales (Calvi), Campania, Zephyrus 12: 25-42. Blazquez, J.-M. 1963, Terracotas del santuario de Cales (Calvi), Campania, AEA 30-6: 20-39. Blazquez, J.-M. 1968-69, Terracotas del santuario de Cales (Calvi), Campania, Zephyrus 19-20: 107-113. Bonghi Jovino, M. 1985, Nuove ricerche a Pompei, regio VI, insula V, Milan. Carandini, A. et al. 1985, Settefinestre, Modena. Carandini, A. et al. 1997, La villa dell’auditorium dall’età arcaica all’età imperiale, Mitteilungen des Deutsches Archäologisches Instituts 104: 118-148. Ciaghi, S. 1983, Le terrecotte figurate da Cales del Museo nazionale di Napoli, Studia archeologica 64, Rome. Compatangelo, R. 1985, L’ager calenus, Naples. Cotton, M.A. & G.P.R. Métreaux 1985, The San Rocco villa at Francolise, Rome. Cotton, M.A. 1979, The Late Republican villa at Posto, Francolise, London. De Caro, S. & A. Greco 1981, Campania, guide archeologiche, Laterza, Rome. Della Torre, O. & S. Ciaghi 1980, Terrecotte figurate ed architettoniche del Muse Nazionale di Napoli. Terreccote figurate da Capua, 1, Naples, 1980. Frederiksen, M. 1984, Campania, British School at Rome, Oxford. Gasperetti, G. & L. Crimaco 1993, Sessa Aurunca (Caserta). Località Pertecale, Bolletino di Archeologia 22: 23-28. Gasperetti, G. & L. Crimaco 1995-1996, Bolletino di Archeologia subacquea II-III: 1-2. Hayes, J.W. 1972, Late roman pottery, London. Humbert, M. 1978, Municipium et civitas sine suffragio, Ecole Française de Rome 36, Rome. Johannowsky, W. 1974, Fasti Archeologici XXIV-XXV, 4961. Johannowsky, W. 1970-1971, Dialoghi di Archeologia 4-5: 467490. Johannowsky, W. 1983, Materiali di età arcaica dalla Campania, Monumenti antichi della Magna Grecia IV, Naples. Leveau, P., P. Sillères & J.-P. Vallat 1995, Les campagnes de la méditerranée romaine, Paris. Losada Nunez, A. 1993, Cabezas votivas feminas del santuario de Cales, Campania. Estudio y analisis tipologico, Boletin del Museo Arquologico Nacional (Madrid): 37-46. Mingazzini, P. 1938, Il santuario della dea Marica alle foci del garigliano, MAAL 37: 684-690.

244

Landscape Perception in Archaeology

Morel, J.-P. 1981, La céramique campanienne. Les formes, Befar 244, Paris. Morel, J.-P. 1992, Ex- voto par transformation, ex- voto par destination (à propos du dépôt votif de Fondo Ruozzo à Teano), Mélanges P. Lévêque, 6, Paris: 247-260. Potter, T.W. 1979, The Changing Landscpae of South Etruria, Londres. Scatozza Höricht, L.A. 1987, Le terrecotte figurate di Cuma del Museo Archeologico Nazionale di Napoli, Studia archeologica 49, Rome. Tchernia, A. 1986, Le vin de l’Italie romaine, Befar 261, Paris. Tchernia, A. 1996, Les céramiques communes de Campanie et de Narbonnaise (Ier s. av.J.-C.-IIe s. ap.J.-C.), Naples. Tremblements de terre, éruptions volcaniques et vie des hommes dans la campanie antique, Naples, 1986. Quilici, S. 1992, Opere di bonifica in relazione a tracciati viari, Tecnica stradale, Rome: 73-83. Vallat, J.-P. 1987, Le paysage agraire du piémont du Massique in Chouquer G. et al. (eds), Structures agraires en Italie centroméridionale, Ecole Française de Rome 10: 315-377.

Périodes/Sites

-700/-350 -350/-200

villae de la zone du Massique fermes de la zone du Massique hameau, village, petite ville du Massique Total de l’habitat en % des 100 habitats de cette zone

1

10 18 6 34 34%

1 2 2%

villae de la zone de Cales fermes de la zone de Cales village, hameau, petite ville de la zone de Cales Total de l’habitat en % des 18 habitats de cette zone

5 1 2 0 0 1 7 6% 40%

Annexe 1: Tableau du nombre d’habitats selon leur nature entre VIII° et III° siècle

Année

-500

-400

Nombre de sites “vivants” 35 47 Dont nombre de sites nouveaux 4 15 Dynamisme faible moyen

-300

-250

59 62 39 15 fort moyen

Annexe 2: Tableau des sites entre VI° et III° siècles avant J.C.

245

RPC 2000 conference proceedings

7.6

The Geography of the Ritual Landscape in Complex Societies

Magna Graecia It is interesting to consider the link between the topographical features of the Greek sanctuaries and the geomorphology of the countryside. An extra-urban sanctuary in its location is very visible from the plain around the town: it would have seemed a powerful token of the authority of the city over its surroundings, symbolically imposing civic order as opposed to the natural order of woodland or uncultivated areas (De Polignac 1991:27-52). At the same time, De Polignac emphasises the concept of liminality in his analysis of the colonies of Magna Graecia and Sicily. The sanctuary is therefore both the visible marker of an established balance and integration of the city with its rural landscape, and the sign of the extent of civilisation expressed in the agricultural landscape as opposed to the wilderness of mountain ranges and woodland areas, or even the counterposition of the earth and the sea. In such perspective, a strong symbolic value has been attributed to the Heraion at the mouth of the river Sele, representing the northern edge of the chora of Poseidonia, facing the Etruscan communities on the right bank of the river (De Polignac 1991:95-127, Greco 1992:63-81, 156162). The Greek world probably offers a better view of such phenomena than cultural regions outside Greek colonial influence (roughly speaking the Italic and Etruscan regions, with a specific emphasis on central Italy). More recent studies have been devoted to the so-called “areas of respect” of the southern Italian and Sicilian colonies, e.g. the non-assigned spaces within the walled urban perimeter. These have been connected with the need of the population living in the chora to settle inside the city in cases of sudden danger. These analyses attempt to draw a quantitative picture of the colonial chorai settlement system, emphasising the particular juridical status of the polis. They clearly wish to counterbalance the apparent topographical disorder with a theory of the division of the space, brought about by the act of dividing the cultivable land among its inhabitants. Results from surveys have shown that sanctuaries were also focal points in the exchange networks between small farms in local colonial settlement systems (Muggia 1997).

Andrea Zifferero 7.6.1

Introduction: the liminal role of sacred areas and sanctuaries A remarkable number of studies on the “political” structure of the rural landscapes dominated by urban centres in pre-Roman Italy have focused on the importance of sacred areas (or, on a different scale, of sanctuaries) in territorial organisation. Both the location and function of sacred areas may be viewed as markers of the origins and development of territories, also shedding light on the “topographical history” of the urban centres (Molinos & Zifferero 1998). The most significant data, from the colonial districts of southern Italy and Sicily, reflect a more advanced phase of study. The rich information is due to the more integrated knowledge of the rural landscape, whose settlement system has been correctly associated with the foundation of the Greek colonies. The “liminal” function of sacred areas, defined as the influence exerted by the sacred area in a determined peripheral position in a political territory, has been stressed above all by P.G. Guzzo (1987) in a paper in which he successfully proposed a model describing the typology of sanctuaries in the Magna Graecia area. According to Guzzo, the general historical and topographical characteristics of the colonisation process are an expression of its economic nature (i.e. the colonies are planned in their productive, mainly agricultural, context). The sanctuaries then acquire the role of signalling a limit (sometimes forming a real physical frontier). We may discern the following categories: a The sacred area as a marker of the limit between the city and the country; b The sacred area as a marker of the limit between the cultivated and uncultivated areas of the country (unploughed fields, woods); c The sacred area as a marker of the limit between the territory of a colony (used for agricultural purposes or only partly utilised) and the territories of neighbouring colonies;

The Greek evidence thus gives us a good perspective on the complex role of sacred areas and sanctuaries. For example, the strategic position on a frontier facing a potential enemy, gave rise not only to agrarian cults, but also to cults with a strong accent on defence or warlike activities.

A general definition of the liminal role of sacred areas and sanctuaries may be summarised as follows: the sacred area and sanctuary may establish a limit between two different spheres of influence, either internally within one territory (e.g. city and country, urban and suburban area), or externally (e.g. between Greek and Italic communities). It may also promote contact between different social and political groups with a mutual interest in economic and productive activities (thus promoting the economy of the polis in a wider sense). In the latter case the sanctuary may belong to the category of frontier sanctuaries.

Central Italy Recent studies dealing with pre-Roman Italy have focused on the middle Tyrrhenian area, analysing the organisation of the landscape in close connection with the increasing importance of urban centres in southern Etruria (figure 1; Rendeli 1993). The emergence of a rural network has been related by M. Rendeli to the influence of the urban centres which promoted and substantially controlled the development of the countryside. This process was the starting point for a “state theory”, which was applied in this region during the Iron Age and was in full practice in the 6th cen246

Landscape Perception in Archaeology

tury BC, with the defining of frontiers, to show the extent of territories dependent on the cities (Zifferero 1995, Guidi 2000:194-223). Investigation of the countryside has resulted in the identification of a network of small farms, which has extended the typology of local settlement systems, though leaving some problems still unsolved (e.g. the distribution and role of sacred areas and sanctuaries inside the tissue of territories). The hiatus in our present knowledge hinders every attempt to reconstruct the configuration of the landscape; I do not mean the modern and partial picture emerging from surveys and modelled by modern attitudes and biases of interpretation, but the configuration of the former pre-Roman landscape, as planned by the ancient communities and modelled by their inhabitants (see, as an example of a model of complexity, the chronological steps in the evolution of the countryside of Caere, pictured in Enei 1992, 1993; figures 1: 4; 2). This gap has been partially filled by a new methodological approach, aimed at highlighting the concept of “liminality” in the organisation of space. However, this approach operates on a different scale and focuses on the various components of the individual settlement system and the structure of the necropoleis (Riva & Stoddart 1996). Dealing with the ratio of the distribution of sacred areas, particular attention has been paid to the presence of even small-scale sacred areas, just outside the perimeter of the plateau, along the gentle slopes leading to the valleys surrounding the urban area, or close to the edges of the plateau facing it. These sanctuaries have been identified as suburban sanctuaries (in many cases in southern Etruria) and probably signified a perimetral limit of that part of the ager closely linked to the city. According to C. Riva and S. Stoddart (1996:95), “Farmland in this area could be cultivated by occupants of the city itself without more than logistical locations for storage of tools and produce. The area was instead densely occupied by ritual monuments, which were frequently visible from the city itself. Intervisibility was a key element in the organisation of this stretch of the landscape, assisted and constrained by the local dissected topography.” A second line of sanctuaries, usually placed beyond the suburban areas occupied by the necropoleis, marked the boundaries between the city and the more peripheral sectors of the ager. General information about the ratio of the latter division (i.e. the shape and the limits of the external ager as delimited by the position of sacred areas within the rural landscape) is scarce for the Etruscan area, as compared to the relatively abundant and still increasing evidence relating to the density and extent of the late 7th-6th century open sites. Some difficulties in interpretation derive from the disparate and scattered level of knowledge of the countryside. Though a general reading of the landscape offers some good perspectives, the problems connected with the recognition of sacred areas in field survey are still overwhelming and are impeding our capacity for an integrated interpretation of the sanctuaries within a rural settlement framework. This apparent lack of communication among practitioners of the discipline has led to hypercritical exchanges

in the debate over the objective identification of sacred areas (e.g. the presence of architectural decorations such as antefixes in a surface scatter of sherds would not be enough to identify it with a sacred area). This has limited the analysis of the relation between sacral expressions and certain natural phenomena (e.g. hot and cold springs, sulphur water spouts, caves and so on). A further step in the analysis of the sacred areas of central Italy is the current research on the presence of sanctuaries along the peripheral boundaries (frontiers) of a territory, for it is known that Etruscans and Italics put the physical limits of the ager (even if extremely far from the core of the territory, represented by the urban centres) under the protection of the gods. Recent studies on political frontiers in antiquity have highlighted the role played by political/social sciences and economic geography in the development of “frontier history” (Ampolo 1998). In recent years the archaeology of complexity began to conceive of frontiers in antiquity as articulated: not fixed lines, but progressively increasing or reducing the extent of the territory bounded, always depending on a state authority, being formally a component of the organisation of its ager (Guidi 1998). In this context, sacred areas or sanctuaries contribute to the shaping of the pattern of ancient city boundaries and state frontiers: we have to consider that the different historical, topographical and juridical conditions of cities and towns exerted a strong influence in each single case. Etruria The Etruscan area has produced evidence of various kinds, depending on the available knowledge of the connection between the cities and their own ager (Colonna 1985a:149159). In the case of the Caeretan ager in southern Etruria, the model of A. Ruiz and M. Molinos (1989) was applied with the aim of detecting the nature and articulation of the northwestern frontier that separates the territories controlled by Caere and Tarquinii (Figures 1: 3-4; 3). This study investigated the protohistoric and pre-Roman settlement system of a large district, extending from the Tyrrhenian coast to the watershed of the Tolfa mountain range (figure 4; Zifferero 1995, forthcoming). The existence of a “mixed frontier” was investigated departing from geomorphology (the mountain range) and the local settlement pattern (mainly oppida on tufa plateaux), as expressed by a line-up of small sanctuaries (perpendicular to the coast) on the “critical points” represented by the routes leading into the mountain range. The archaeological record seems to point to the end of the 7th century BC as the terminus post quem for the birth of this frontier. The evidence for the existence of sacred areas is complemented by votive offerings dating from that period, although the occupation of some mountain settlements, possibly with the function of controlling the zone, seems to have begun earlier (D’Ercole et al. 1998). The organisation of the frontier practically overlaps the florescence of the rural settlement system. The two phenomena seem to be highly integrated in a densely populated landscape. 247

RPC 2000 conference proceedings

Figure 1: The most relevant centres of the middle Tyrrhenian Italy, mentioned in the text. Symbols: 1 = ancient centre; 2 = modern centre. Topography: 1 = Vulci; 2 = Volsinii; 3 = Tarquinii; 4 = Caere; 5 = Veii; 6 = Falerii Veteres; 7 = Ficana; 8 = Castel di Decima; 9 = Acqua Acetosa-Laurentina; 10 = Antemnae; 11 = Fidenae; 12 = Crustumerium; 13 = Cures Sabini; 14 = Lavinium; 15 =Ardea; 16 = Aricia; 17 = Gabii; 18 = Tibur; 19 = Satricum; 20 = Velitrae; 21 = Praeneste; 22 = Anagnia.

The more extensive data from the 4th century BC set this frontier in a different historical perspective. During the war between Tarquinii and Rome (whose territories practically flank each other) Caere’s role was one of substantial neutrality (Naso 1999). This case study remains at the moment quite isolated, although another possible case may be the identification of the boundaries of the ager controlled by the city of Veii (figure 1: 5). Recent and accurate analyses produced a review and classification of the sites facing the lower Tiber valley and the part of the former Etruscan territory along the right bank of the river, conquered by the Romans (De Santis 1997). These boundaries seem to be guarded by a network

of pagi situated on tuff plateaux of limited extent, and controlled by local aristocratic groups involved in the exploitation and control of arable land. The archaeological evidence consists mostly of surface finds and necropoleis, dating from the 8th to the 6th century BC. In the Archaic period this settlement system seems to have developed into a rural landscape based on open sites, although in this picture the possible role exerted by the sacred areas is still to be depicted. More detailed information comes from the northern edge of the territory controlled by Vulci: although a critical analysis is still lacking, a frontier based on a strategic distribution of oppida and sacred areas in the Albegna valley 248

Landscape Perception in Archaeology

+

π

Ω

0

•

Figure 2: The Etruscan landscape of the southwestern sector of the ager caeretanus (7th-6th centuries BC). Symbols: = small rural sites; = large rural sites; = small necropoleis; = cult place; = incedental finds; = large necropoleis; = Caere, urban area. Source: Enei 1992.

would have extended from the coastal sanctuary of Talamone up to the internal sites of Ghiaccioforte and Saturnia (figure 1: 1). This process is likely to have evolved in a period ranging from the early 7th to the 3rd century BC Roman conquest (Colonna 1977). A recent contribution concerning the site of Doganella has emphasised the activity of craft workshops (storage vessels and metal objects production) identified on the plateau. This activity gives the centre a strong economic and exchange function, in the framework of the commercial policy of Vulci, directed at the markets of central and northern Etruria and, above all, the maritime routes to the upper Tyrrhenian sea and the Ligurian sea (Perkins & Walker 1990). On the basis of his analysis of the development of the agri controlled by the cities of southern Etruria, M. Rendeli points to the middle of the 6th century BC as the period in which this system had come to full maturity, including the “crystallisation” of the cult places in the rural landscape. This trend seems to have been quite

general in the settlement system of the middle Tyrrhenian area (Rendeli 1993:356-367). Due to the different nature and process of formation of some cities of central Etruria, they achieved a full urban shape only very late, in comparison with those formed in the poleogenesis of the Tyrrhenian area. The case of Chiusi offers abundant evidence, particularly relevant in terms of sanctuaries, which has not yet been analysed within the framework of a political system (Rastrelli 1998, 2000). As for central Etruria, clear evidence enables us to articulate in the outline of a strategic territory the area around the Chiana valley between the centres of Chiusi, Cortona and Perugia. A very attractive hypothesis was advanced by G. Colonna, dealing with the existence of a templum, i.e. an area with a sacred orientation, at the Trasimeno Lake, based on the position of votive offerings found along its banks, revealing the presence of sacred areas (Colonna 1976-77). This deliberate organisation of the geographic space in central Etruria becomes visible when plotting a suburban line marked by 249

RPC 2000 conference proceedings

Figure 3: Model for the identification of frontiers, proposed by A. Ruiz Rodriguez and M. Molinos (1989).

the so-called tular stelae, and it sometimes corresponds with the find-spots of votive offerings (Colonna 1988).

that such sacred areas represent a projection of the town (i.e. of the social and economic network created by the town) on its ager; c) Most sanctuaries seem to have been active from the archaic period onwards. In the territory of the Aurunci, their origins date back to the early Iron Age. The author seems to exclude this model from the influence exerted by Greek colonies in southern Italy, because the data suggest an earlier date than theirs.

Evidence from the Italic cultural areas The evidence from the Italic cultural areas appears stronger in a new perspective on the organisation of territories in northern Campania, thanks to the work of P. Carafa (1998). This scholar has brought to light a complex network of sanctuaries, marking successive boundaries in the territories of the Aurunci (in the upper and middle valley of the Volturno River) related to the Etruscan town of Capua. It is interesting to note how the disposition of these sanctuaries shows features common to other groups: a) The evidence indicates that the typology of the “frontier sanctuaries” is also represented in northern Campania. This is not surprising as their presence was to be expected in the organisation of the territory of the Etruscan centres. It does, however, constitute a new result in the archaeological study of the Italic culture areas, and acquires a strong significance given the level of current knowledge of the articulation of such centres, in comparison with the more detailed topographical investigations in the neighbouring Greek colonies (Cuma). b) The topography of these sanctuaries seems to respect dispositional rules common to the cases we have already proposed. There is a first line of sacred areas around the town, placed at the distance of about 2.5 km from the centre. A second line of sanctuaries seems to be placed at a distance of about 5 km from the centre. The author notes analogies with Rome and some Etruscan cities. Since the existence of the sanctuaries is supported by the archaeological data, Carafa could point to a synchronicity between the origins of the town and the foundation of the sanctuary, confirming

All the considered cases show a substantial continuity in cult up to the period of full Romanization: a factor clearly showing the longue durée trend of these sanctuaries, in spite of the changes occurring in the social and economic structure of the area up to the Social War. A similar tendency is visible in the middle Adriatic section of the Italian peninsula, in southern Abruzzo, among the river valleys of Pescara and Trigno, where research in progress has identified a specific function characteristic of the sacred areas: as markers of inter-ethnic boundaries. It is very interesting to observe the relation of these boundaries to the settlement system: the substantial lack of an urban settlement system here induced the creation of inter-ethnic boundaries in a land where the population was distributed per pagos. It is evident that the sanctuary functioned as a social, economic and religious meeting point for the inhabitants of the region (D’Ercole et al. 1997). 7.6.2 Centre and periphery In the following some possible lines of research on liminality are given. When studying the connection between centre and periphery, one should consider the presence of the agri gentilicii (i.e. portions of cultivable land in and around the urban area), always viewed as a distinguishing feature 250

Landscape Perception in Archaeology

Figure 4: Hypothetical organization of the frontier in the sector between Tarquinii and Caere, in the 4th - first half of the 3rd centuries BC. In the centre is the Tolfa mountain range, with the accesses marked by the letter V. Symbols: 1 = settlement on tuff plateau; 2 = isolated chamber tomb or necropolis; 3 = cult area; 4 = main routes. Topography: 1 = Centocelle (Tarquinia); 2 = Ripa Maiale (Allumiere); 3 = Casale dell’Aretta (Allumiere); 4 = Bufalareccia (Allumiere); 5 = Grotta Porcina (Vetralla); 6 = Blera; 7 = S. Giuliano (Barbarano Romano); 8 = Luni sul Mignone (Blera); 9 = Selvasecca (Blera); 10 = S. Giovenale (Blera); 11 = Grasceta dei Cavallari (Tolfa); 12 = Cava della Scaglia (Civitavecchia); 13 = Poggio Granarolo (Civitavecchia); 14 = Pisciarelli (Civitavecchia); 15 = Castellina del Marangone (S. Marinella); 16 = foce del Marangone (S. Marinella); 17 = Punta della Vipera (S. Marinella); 18 = Pian Cisterna (Tolfa); 19 = Pian Conserva (Tolfa); 20 = Pian dei Santi (Tolfa); 21 = Rota (Tolfa); 22 = Monterano (Canale Monterano); 23 = Bagni di Stigliano (Canale Monterano); 24 = Poggio Spiantacase, q. 305 (Allumiere ); 25 = Pyrgi (S. Marinella); 26 = Sasso di Furbara (Cerveteri); 27 = Procoio di Ceri (Cerveteri); 28 = Ceri (Cerveteri). Source: Zifferero 1995.

251

RPC 2000 conference proceedings

of Etruscan proto-urban centres and particularly of Rome during the protohistoric and regal period. The role played by Roman or allied aristocracies on peripheral portions of the ager romanus (see, for instance, the group of the Titini latrones, controlling the territory of the oppidum of Acqua Acetosa-Laurentina), according to G. Colonna (1991), proposes anew the suggested model of a “mixed frontier”, as in the case of Caere, based on oppida and sacred areas at the critical points (here related to the routes leading to Rome) (figure 1: 9; 5). Strabo (5, 3, 2) reports that the Ambarvalia ceremony began at the sacred area called Festoi, coinciding with the 6th mile of the via Laurentina: a place archaeologically still unknown but thought to be connected with nearby Acqua Acetosa-Laurentina. In the case of Rome, the role played by the non-Roman local aristocracies in consolidating the territory of the centre is quite evident in the regal period. The settling of gentes from various conquered areas of Latium vetus in Rome was important for their integration within the network of the Roman aristocracy. The coming of the Sabines to the city in the time of Titus Tatius, king of Cures, or the new settlement of the most distinguished gentes from the destroyed Alba Longa, promoted by the king Tullus Hostilius, in the area of mons Caelius, are currently sustained as historical realities by recent critics (Colonna 1996a, Carandini 1997:34-345). The different approaches of scholars to distinctions among property and the control of land has given rise to a discussion of the connection between the physical growth of Rome and the creation and development of political organisation among the urban population (curiae). We are, however, still far from having sufficient knowledge of the position of aristocratic landed estates in the suburbs of Rome and elsewhere outside the core of the Latin centres of Latium vetus (Capogrossi Colognesi 1988, Carandini 1997:298-304). A very stimulating passage in Livy (II, 5; 1) deals with the former landed estates of the Tarquinii, between the urbs and the river Tiber. The senate, after the defeat of the last Tarquin, devoted all the ager Tarquiniorum to Mars: the area then became the Campus Martius. Its use, during the regal period, was both agrarian (the community of Rome had permission from the senate to cut the crop growing on these properties) and “political” as, farming apart, the area was set aside for games and chariot races (La Rocca 1984, Thuillier 1985). In comparison with the Roman situation, the scarcity of written sources on Etruria has obliged scholars to make full use of archaeological data. What appears from the analysis of the relation between aristocracy and the control of territories is the probable existence of agri gentilicii, most of all in the Orientalizing period. These were marked by the tumuli of the gentes emerging in the urban centres (Zifferero 1991). The pressing question, therefore, is the following: is it possible to trace, through the interpretation of archaeological data, the passage from aristocratic control of land in the middle Tyrrhenian area to full urban control based on an

enlarged population, by looking specifically at sacred areas? The formal lack of contact between these two topics, land control and sacred areas, is only apparent. On the one hand, the distribution of some aristocratic markers, such as tumuli, shares with cult places some “liminal” aspects, i.e. peripheral or boundary location with respect to settlements. On the other, a connection between funerary sacra gentilicia and the civil sacra of enlarged bodies of people living in and outside cities, towns and oppida may be established (see below). Some aspects of this complex network of links have been investigated by I. E. Edlund (1987) and C. Riva and S. Stoddart (1996), as previously quoted. They concern a line-up of suburban cults, forming a physical division between the city and the countryside. But certainly there are other possible limit-indicating landmarks within the configuration of a territory. To investigate these, a “holistic” approach seems appropriate, one that involves aspects of geomorphology, rural settlement, the presence of boundaries or frontiers, and phenomena of cultural diffusion or expansion. The holistic approach is exemplified in the following case studies: the Alban Hills, Rome and Caere. The Alban Hills In 1987, the geographer C. Cerreti furnished a new and stimulating perspective on the relations between sanctuaries and territories in Latium vetus. In his study this scholar analysed the location of the most celebrated cults, with a special focus on the area of the Alban Hills (figure 6). He stresses the strong division between agrarian cults on the one hand and woodland cults on the other. Although this opposition had already been observed by scholars of religious history, Cerreti’s approach is inspired by economic geography. The target of the research is the link between cults and the topography of the region. Primary attention is devoted to woodland areas, loci religiosi, for their natural quality, because they are extra-territorial, and no-man’s land, as opposed to the cultivated agri of the centres of ancient Latium. Accordingly, the opposition between the sacred features of the luci, silvae and saltus (usually put as a whole under the protection of deities such as Diana, Faunus/Silvanus, or Fauna/Bona Dea), and the agrarian cults is particularly emphasised, depending on their position inside the ager. Cerreti observes, moreover, that the agrarian cults tend to be in a liminal position of the ager: the lustratio of the boundaries of the ager romanus antiquus itself suggests indeed the existence of a defined perimeter, a sort of sacred limit signalling the extension and the power of the agrarian culture preceding the woods. Applied to the geography of Latium vetus, these insights lead the present author to consider the existence of a hierarchy of sacra that stands in relation to the geomorphology of the region. It appears that different strands of research seem to point to the same conclusion: namely that the hierarchy was based both on the distribution and the differences in altitude of the cults. The hierarchy of the Alban Hills was certainly based on the presence of Iup252

Landscape Perception in Archaeology

Figure 5: The extent of the ager romanus antiquus according to A. Alföldi, with integrations by G. Colonna. Topography: 1 = Dea Dia; 2 = Acqua Acetosa-Laurentina; 3 = Fossae Cluiliae; 4 = Fortuna Muliebris; 5 = La Rustica; 6 = Antemnae; a = Prima Porta; b = Colle S.Agata; c = Acquafredda; d = Pantan di Grano; e = Monte Roncione. Source: Colonna 1991.

piter Latiaris (probably derived from an original panLatin local deity, whose identity has been recognised in the figure of Indiges), and on the cults of Diana, deity of the wood, established in various luci of the same mountain range (figure 6: 38; Ampolo 1989, Cecamore 1996). The extra-urban character of the cult of Diana at Nemi is evident from the literary sources, especially in such details as the nomination of a rex nemorensis - generally a slave or fugitive who could claim this position after having defeated the rex currently in charge (figure 6: 39). It was a ritual practice that, in the archaic period, was completely out of the territorial control of the local pre-eminent centres, i.e. Aricia, Lanuvium and Tusculum (figure 6: 33, 44, 41; Coarelli 1987) . Another piece of indirect but meaningful evidence is the dedication of the shrine of Diana in Rome, during the reign of Servius Tullius, on the Aventine hill, a plateau which was outside the boundary of the city, and marked by the pomerium (figure 6: 11; Ghini 1995, Zevi 1995).

The hypothesis, then, could be that there existed a strong religious organisation in the Alban Hills, reputedly the core of the Latin culture. We may deal then with a hierarchy that respected the original religious structure of the Latin territory, one that was based on an Iuppiter cultus practiced on the top of the Monte Cavo (Figure 6: 38). To Iuppiter Latiaris, the most celebrated divinity of Latium vetus, were dedicated regular feasts of all the Latin communities: the antiquity of such a deity and the role his cult played in the future ritual of Rome has been often remembered by the literary sources. It is possible then to think of a hierarchy based on three different levels: the first level would correspond to the upper zone of the Alban Hills (at an average height of 800 metres above sea-level, usually at the summits of the range) devoted to Iuppiter; the second, devoted mainly to Diana/Artemis and Fortuna, corresponding to the central zone of the mountain range and the vast area comprising the lakes; the third level would correspond to the area of the foothills and thus with the territory of the 253

RPC 2000 conference proceedings

Figure 6: Simplified map of the main Latin centres and cult places between Rome and the Alban mountain range. Symbols: 1 = ancient centre; 2 = cult area; 3 = cult area or sanctuary devoted to Diana; 4 = presumable extent of the area of the alban mountain range possibly devoted to the cults of Diana/Fortuna. Topography: 1 = Ostia; 2 = Ficana; 3 = sanctuary of the Dea Dia on the via Campana; 4 = sanctuary of Fors Fortuna at the 1st mile of the via Campana; 5 = Tarentum; 6 = city of Rome within the Servian wall; 7 = sanctuary of Hercules at the 1st mile of the via Tiburtina; 8 = sanctuary of the Spes vetus at the 1st mile of the via Labicana; 9 = sanctuary of Minerva (?) at the 1st mile of the via Latina; 10 = sanctuary of Mars at the 1st mile of the via Appia; 11 = temple of Diana on the Aventinus; 12 = sanctuary of the Fortuna muliebris at the 4th mile of the via Latina; 13 = Castel di Decima; 14 = Acqua Acetosa-Laurentina; 15 = Fossae Cluiliae; 16 = Bovillae; 17 = sanctuary of Pater Indiges at the mouth of the Numicus; 18 = Lavinium; 19 = cult place at Tor Tignosa; 20 = cult place at Santa Palomba; 21 = cult place at Valle Caia; 22 = cult place at Pescarella; 23 = Ardea; 24 = Aphrodisium at Banditella; 25 = cult place at La Fossa; 26 = cult place at Tor San Lorenzo; 27 = cult place at Colle Cane; 28 = cult place at Ottanta Rubbie; 29 = cult place at Caffarella Nova; 30 = cult place at Campoleone Scalo; 31 = Lucus Ferentinae; 32 = Castel Savello; 33 = Aricia; 34 = cult place at Casaletto; 35 = cult place at Albano; 36 = presumed site of Alba Longa; 37 = presumed temple on the south-western bank of the Albano Lake; 38 = sanctuary of Iuppiter Latiaris; 39 = temple of Diana in nemore aricino; 40 = presumed sanctuary of Diana at Corne; 41 = Tusculum; 42 = Gabii; 43 = Praeneste; 44 = Lanuvium; 45 = temple of Iuno Sospita; 46 = cult place at the Parata spring; 47 = cult place at Soleluna; 48 = cult place at Colle Ottone; 49 = cult place at Colle della Salvia; 50 = Velitrae; 51 = sanctuary of Fortuna in Monte Algido (?); 52 = cult place at Cigliolo; 53 = Cori; 54 = Artena.

Latin urban centres, and would take in cults with a markedly agrarian character. This hierarchical organisation of the Alban Hills has also been emphasised by C. Ampolo, who provides a detailed map of the cults there devoted to the goddess Diana. Ampolo recognised two major sites,

corresponding probably to woodland areas, in the socalled nemus aricinum and in the sanctuary of Diana at Corne, near Tusculum (figure 6: 39-40; (Ampolo 1989). The location of these cults indicates a regular distribution of the sanctuaries over the landscape, but apparently 254

Landscape Perception in Archaeology

Figure 7: The position of the fanum Valetudinis et Bonae Deae in the southeastern slopes of Monte Cimino. Symbols: 1 = ancient centre; 2 = ancient centre destroyed by the Romans; 3 = cult area. Topography: 1 = fanum Valetudinis et Bonae Deae; 2 = original position of the ara Larum; 3 = presumed position of a lucus, according to the toponomastic relics; 4 = Sutrium; 5 = Nepet; 6 = Falerii novi; 7 = Falerii veteres. Source: Gasperini 1988a (elaborated on a drawing by M. Chighine).

excludes a large zone located on the eastern sector of the range (shadowed on figure 6). This is the area where some authors locate the cults of both Fortuna and Diana in Algido, recorded by Livy (XXI, 62, 7-8) and Horace (Carm. Saec. 69-72) (Melis & Quilici Gigli 1983:19-24). The cult of Diana could therefore mark, at least in the Archaic period, the limit between the woods and the fields, including the zones of the mountain range (at a height roughly between 400 and 800 metres above the sea-level), which were reputedly “undivided and common”, and related to the territories of the centres along the foothills of the Alban Hills (although the proximity of the centres is

apparent from the toponyms that identified the sacred areas: lucus Dianae in nemore aricino). In this respect we might consider a passage in Cicero (pro Mil. 31, 85), who records this mountain group as being devoted to Iuppiter Latiaris, and therefore not liable to everyday human settlement (Grandazzi 1986:64-66, Zevi 1995). The sacred character of the tops of particularly relevant mountain ranges has also been assumed in other cases, in both Italic and Etruscan regions. In both areas there are some very high mountains, whose top and sloping sides were devoted mainly to Iuppiter. It is sufficient to quote 255

RPC 2000 conference proceedings

In most of the cases mentioned, we may thus expect the existence of a “sacred map”, adapted to and at the same time depending on the religious topography of each site, reproducing on earth by landscape features, as it were, a divine templum. C. Cecamore (1996) has recently offered a relevant interpretation of the religious spatial configuration of the Alban Hills: that it consists of an auguraculum corresponding to the top of Monte Cavo (figure 6: 38). If true, this would be an excellent and isolated case of an extra-urban auguraculum, functioning within the religious and geographical spatial organisation of the early Latin communities. The central position of Monte Cavo in Latium vetus has always been emphasised: the fact that the region was sacred to Iuppiter may have to do with the original mythographic and historical position of Alba Longa. Two main hypotheses have resulted from archaeological research on this legendary centre. The first locates Alba Longa in the hills surrounding the southeastern bank of Lake Albano with a crown of protohistoric settlements around it (Colle dei Cappuccini and Tofetti, close to the modern town of Albano) (figure 6: 36; Chiarucci 1996, 2000) the second sees Alba Longa in the sparse settlement system of the upper part of the Alban Hills region (Grandazzi 1986). The literary sources relate quite clearly a former dependence of the cult of Mons Albanus from Alba; the same term “Alba” should then be attributed to the district reaching to the core of the Alban Hills, and so to the presence of Iuppiter Latiaris. Moreover, the sources consider Alba to be the mother city of all the Latin populi: if Albula is in fact, as suggested by C. De Simone (1975), one of the original “Latin” names of the Tiber, there is even a formal coincidence between the earlier control of Alba and the surrounding Latin region.

Figure 8: The sanctuaries placed at the 1st mile on the routes outside the Servian walls. Topography: 1 = Fors Fortuna; 2 = Mars; 3 = Minerva (?); 4 = Spes; 5 = Hercules; 6 = Anna Perenna; 7 = Tarentum. A = Antemnae. Source: Colonna 1991 (from a drawing by L. Quilici).

the cases of Monte Cimino as the place where the cult of Iuppiter Ciminius was practiced and that of Monte Amiata (between Tuscany and Latium) a place dedicated to Iuppiter (Gasperini 1988a, Cambi 1996, forthcoming). On hierarchy, it is worth remembering the cult of Soranus Apollo, to whom apparently the whole of Monte Soratte in the middle Tiber valley was dedicated, an area facing the Sabine communities settled on the opposite bank of the river. The recent (re-) discovery at Nepi of an altar dedicated to Diana Compotens sheds further light on the distribution of cults practised on the tuff highlands of the Faliscan region (Brocato & Fontana 1989, Di Stefano Manzella 1992, Camilli 1998). The best-documented case, however, is the sanctuary at the Arcella on the southeastern slopes of Monte Cimino, in northern Latium (Figure 7). It is feasible to project on this mountain the existence of an organised religious topography: the top of the elevation was dedicated to Iuppiter Ciminius, as suggested by the epigraphic sources (CIL, XI 2688), while the middle zone still features the original positions of altars (one carved on a local rock) indicating the sacral influence of Valetudo and Bona Dea which extended to the slopes of the same mountain range (Figure 7: 1), further there was a woodland cult place devoted to the Lares (figure 7: 2; Ross Taylor 1923:163, Gasperini 1988a).

Rome Rome is probably the most interesting and best-investigated case. This is due to the richness of the literary and epigraphic sources that, together with archaeological data, contribute to a reconstruction of profound historical nature (figures 6: 6; 8; amongst the vast bibliography, see above all Scheid 1987, Colonna 1991). The relationship between the cults and the topography of Rome has been tackled in a number of recent analyses, some of which have taken a broad perspective. The situation in the case of Rome is extremely dynamic from a historical point of view. The particular liminal position of some tribes in the Roman territory has been stressed by C. Ampolo (1987); he observed the relation between names of tribes expressing “national” characteristics (e.g. the tribe Romilia) and their position within the site boundaries of Rome, facing Etruscan territory. A more articulated perspective has been offered by A. Carandini (1997), in a particularly complex picture illustrating step by step the growth of the city of Rome. Here we will try to analyze some developments in the enlargement of the ager romanus antiquus, emphasizing the role of frontiers, and examining trends in the distribution and topographical positioning of the sacred areas. The history of the ager romanus antiquus (i.e. the territory 256

Landscape Perception in Archaeology

romanus (figure 9; Carandini 1997: 445-456 and fig. XIII, Carafa 2000). Such a vision of the “rounded” shape of Rome has its roots in the classical reconstruction carried out by A. Alföldi, and re-elaborated by G. Colonna (1991), of the connection between the ager and the distribution of tribes (figure 5). It is particularly interesting to note the model of expansion proposed by A. Carandini, which is actually not based on a casual, but on a planned incorporation of portions of land of the conquered populi within the framework of Rome and its close suburbs. In creating the rural tribes, Rome would have respected the presence of these groups and would have taken account of the topographical background of the tribes and even the trajectories of geographical origin of some groups. It would have done so to emphasise Rome’s former relationships with the conquered Latin centres in the spatial organisation of Rome. The defining of two concentric lines (the internal line drawn by the sanctuaries placed at the 1st mile from the Servian walls and the external line traced by the sanctuaries at a distance between the 4th and the 6th mile from Rome), reflects a mature organisation limited by a frontier on the edge and marking at the same time the division between the suburbs and the ager (figures 5-6, 8-9). The two division lines express, however, two different objectives for signalling the space. The ceremonies of lustratio at the external edges of the ager, for example, express the agricultural vocation of the Roman territory through the ritual of the Ambarvalia already from the early regal phase onwards. The possibility of a territorial organisation even before the Romulan conquest has been cautiously explored by Colonna (1991), and was decisively carried forward by A. Carandini (1997:481-487).

Figure 9: A model proposing the shape and the structure of Rome: the boundaries separate the areas of the city and mark different steps of growth. Source: Carandini 1997.

essentially acquired during the reign of Romulus, both on the left and right side of the Tiber, and taken from the Latin and Etruscan ethne) is a case well known to the critics (figure 5; Carafa 2000). Further study could involve the following topics: a) The ager romanus antiquus is a very good example of a frontier organised with limits marked by sacred areas at the critical points of the territory (i.e. on the routes), and protected by oppida in the proximity of such critical points; b) The sources are very helpful in such a reconstruction, offering good quality information on the cults and ceremonies that took place in an area restricted to between the 4th and the 6th mile from Rome; c) The strong agricultural imprint on some of these cults has recently been emphasised along with the idea that the frontier serves to distinguish a densely populated from a less populated or woodland area. d) Thanks to A. Carandini’s and P. Carafa’s work we have improved our knowledge of the shape of Rome and its territory. They propose a model expanding from the centre, and which distinguishes pre-urban, proto-urban and fully urban phases. It comprehends the crown of pagi and the external character of the area of the ager

According to the literary sources, the historical framework envisions a progressive “erosion” of the territory of Alba caused by the Roman conquest. After having consolidated in the Romulean age (second half of the 8th century) the part of its ager facing the Alban Hills, the city, according to tradition, defeated and destroyed Alba during the reign of Tullus Hostilius. This conquest would have been “planned” through an individual contest between six young warriors, belonging to distinguished gentes from Alba and Rome. A topographical fixed point has to be put firmly at the fossae Cluiliae, a limit (deriving in the sources alternatively from cluere, ‘to purify’, or from Cluilius, the name of the Alban king conducting the war against Rome) of the Roman territory already reached in the Romulean age, on the 5th mile of the route leading to Alba (figure 6: 15; sources in Grandazzi 1986:88). From an archaeological point of view, G. Colonna (1996a) has put a particular emphasis on this area. He observed the overlapping of the fossae Cluiliae with the site of provenance of some bronze furniture related to a late 6th century chariot, said to come from the estate of Roma Vecchia along the via Appia, discovered in the 18th century possibly in connection with more archaic metal tools (Buranelli 1997). The original 257

RPC 2000 conference proceedings

Figure 10: Distribution of the architectural terracottas (circle) and of the ex voto (triangle) in the urban and suburban areas of Caere. Filled symbols mark precise positions. 1 = Vigna Parrocchiale; 2 = Vigna Marini Vitalini; 3 = Vigna Ramella; 4 = Orto Consalvi; 5 = Casaccia; 6 = Madonnella; 7 = Campo del Purgatorio; 8 = Vignale; 9 = Felcetello; 10 = Vigna Rosi upon Valle Zuccara; 11 = Vigneto Pescini; 12 = Vigna Turiani; 13 = Terreno Renzi; 14-15 = S. Spirito; 16 = Vigna Grande; 17 = Vigna della Casa; 18 = S. Antonio; 19 = Vigna Falconieri; 20 = Vignaccia-Granarone; 21 = temple of Manganello; 22 = Via degli Inferi; 23 = Via della Bufolareccia; 24 = Valle Zuccara; 25-27 = Valle della Mola. Source: Nardi 1989.

258

Landscape Perception in Archaeology

Figure 11: Simplified map of the main centres and cult areas in the territory controlled by the city of Caere: in the centre is the Tolfa mountain range. Symbols: 1 = settlement on tuff plateau; 2 = necropolis; 3 = cult area. Topography: 1 = Porto Clementino (Tarquinia); 2 = Tarquinii; 3 = Cava della Scaglia (Civitavecchia); 4 = Pisciarelli (Civitavecchia); 5 = Poggio Granarolo (Civitavecchia); 6 = Centocelle (Tarquinia); 7 = Casale dell’Aretta (Allumiere); 8 = Bufalareccia (Allumiere); 9 = Ripa Maiale (Allumiere); 10 = Monte Rovello (Allumiere); 11 = Luni sul Mignone (Blera); 12 = Selvasecca (?) (Blera); 13 = Grotta Porcina (Vetralla); 14 = Castellina del Marangone (S. Marinella); 15 = foce del Marangone (S. Marinella); 16 = Punta della Vipera (S. Marinella); 17 = Poggio Castelsecco (S. Marinella); 18 = Poggio Spiantacase, q. 305 (Allumiere); 19 = La Fontanaccia (Allumiere); 20 = Pian dei Santi (Tolfa); 21 = Pian Conserva (Tolfa); 22 = Pian Cisterna (Tolfa); 23 = Grasceta dei Cavallari (Tolfa); 24 = S. Giovenale (Blera); 25 = Civitella Cesi (Blera); 26 = Blera; 27 = S. Giuliano (Barbarano Romano); 28 = Pyrgi (S. Marinella); 29 = aquae Caeretanae; 30 = Sasso di Furbara (Cerveteri); 31 = Pian Curiano (Tolfa); 32 = Bagni di Stigliano (Canale Monterano); 33 = Rota (Tolfa); 34 = Poggio S. Pietro, q. 164 (Tolfa); 35 = Monterano (Canale Monterano); 36 = Monte Sassano = Monte Calvario (Canale Monterano); 37 = Monte Tosto (Cerveteri); 38 = Griciano (Cerveteri); 39 = Caere; 40 = Monte Abatoncino (Cerveteri); 41 = Procoio di Ceri (Cerveteri); 42 = Ponte del Lupo (Roma); 43 = Castel Campanile (Roma); 44 = Ceri (Cerveteri); 45 = Bagni di Vicarello (Bracciano); 46 = Trevignano Romano; 47 = Sutrium.

context of these objects was probably funerary: the author observes the correspondences between this material in its presence here, the proximity of some Roman mausolei that are shaped like tumuli, their position close to the fossae Cluiliae, and the Livian narrative (I.25). Colonna relates it to the treatment of the dead Horatii and Curiatii, who were buried in separate tombs placed on the site of the contest.

The same source adds an interesting detail to the tale of the duel: during the phases preceding the fight between the six warriors, the albanus rex, Cluilius, died, leaving his name to the site where (we suppose) he may also have been buried (I.23). Once more, the combined evidence points toward the contemporary presence of limits and tombs. The sacred character of the boundary running along the 259

RPC 2000 conference proceedings

0

5 cm

Figure 12: Drawing of the coarseware crater dedicated to Diana, from the villa at La Fontanaccia (Allumiere), with details of inscriptions on the neck and handles; probably datable to the 1st century AD. Source: Vitali Rosati 1994.

sanctuary, probably related to the nearby community of Blera (Zifferero 1995).

fossae Cluiliae can indirectly be presumed through the rite of the clarigatio (i.e. the formal declaration of war), carried out by the pater patratus, leading the collegium of the Fetiales (Serv. VA IX.52, Livy I.24, I.32.6-14): this ceremony had to take place at the border of the territory controlled by the enemy, whose conquest was augured by throwing a lance into it. Considering in this case both literary sources and the archaeological data, we should not doubt the presence of the Romulean boundaries of the ager romanus at the fossae Cluiliae. It is perhaps possible to add some more evidence to the hypothesis of the interpretation of significant graves as markers of boundaries, as already discussed in the Etruscan complex of the Grotta Porcina (Vetralla), where the presence of an aristocratic mound will have stimulated the formation of a boundary as marked by a late 6th century

After having definitively defeated the Albans, the Romans plundered and destroyed their centre (with the exception of the temples and sanctuaries), integrating the population on the mons Caelius and taking control of a part of the territory of this community (Grandazzi 1986). According to G. Colonna, the extension of Roman territory in the first half of the 7th century BC was limited to the foothills of the mountain range, and coincided with the sacred area of the lucus Ferentinae (1995:40). The importance of this lucus has been emphasised in the literary sources, and it is assumed to have been the mouth of the effluent of the Albano Lake. Recent discoveries allow a more precise identification with the area of the so-called lacus Turni, 260

Landscape Perception in Archaeology

closely flanking Monte Savello (figure 6: 31-32; Ampolo 1981, Colonna 1985b, Chiarucci 1996). Formal control exerted by the nearby centre of Aricia seems highly probable. This place became the federal meeting point of the communities belonging to the Latin League. It was frequented especially during the events following the collapse of the Roman monarchy at the end of the 6th century BC. The historical background to these events has been recently validated with both epigraphic and archaeological evidence by C. Ampolo (1981, 1990) (figure 6: 33).

been connected with fertility cults (Granino Cecere 1992, Lega 1995): it is, moreover, interesting to observe that some of them might indicate the boundaries of Lavinium and Ardea in the direction of the Alban mountain range (as suggested in Melis & Quilici Gigli 1982:34). These rural sacred areas seem to be organised in a quite regular line, parallel to the coast, as if to mark an interruption in the continuity of the plain (figure 6: 19-22; 29-30). The location of these sanctuaries, between Tor Tignosa and Campoleone, has certainly been determined by the influence of the facing Latin centres Lavinium, Ardea, Aricia and Lanuvium (figure 6: 18, 23, 33, 44). Special attention must be paid to the cults in the region located near hot or cold sulphur springs (Camponeschi & Nolasco 1982:259-262, 368-370). The case of the sacred area placed at Tor Tignosa in a zone characterised by remarkable sulphur ores, testifies to a cult dedicated to Lar Aenia: this district, around the modern settlement of Solferata, has been identified as the mythical Albunea, a sacred place rich in sulphur springs that sheltered an oracle of Faunus. The liminal nature of this district emerges from the figure of Faunus, whose person is linked to the woods and the wilderness, as has been recently emphasised by A. Carandini (1997:82-84). It is located close to the cult dedicated to Aeneas, founder of the nearby centre of Lavinium (figure 6: 19; Guarducci 1971). We can imagine the suggestive and liminal sulphurous landscape to have been a sacred district with a strong peripheral character. The deified figure of the Trojan hero, then, would give expression to the supremacy of the urban civilisation and of agriculture over the wild nature of woods. Such features could give to this zone the nature of an ecological boundary, one that fits a possible limit between the territory controlled by Lavinium and its neighbours (it is, for example, located midway between Lavinium and the supposed site of Alba Longa).

These data give rise to the following observations: a) The role of the lucus Ferentinae is certainly important because it is in a “Latin” territory, most probably at the northwestern boundaries of the territory of Aricia, if we give credit to the hypothesis of the “liminal” character of the woodland areas. b) The manuscript tradition also defines the place as porta Ferentina (Plut., Rom. 24.2). This version has been rejected and corrected by C. Ampolo (1981:220, 1989:164, Ampolo & Manfredini 1988:332) into the current denomination of lucus Ferentinae. While admitting the remote possibility of a variant to the manuscript tradition and while excluding a reference to an eventual Roman urban gate, what could the porta Ferentina be? One possible explanation would be that the porta Ferentina was the access to a templum, whose gate had to be sub monte albano: a sort of religious district consisting of the Alban mountain range and legitimised by the presence of Iuppiter Latiaris on the central hilltop (in the sense already proposed by Cecamore 1996). There are some arguments supporting this hypothesis: firstly, the possible sacred connection between this lucus and the Albano Lake (where a presumed sanctuary has been recovered on its south-eastern banks: (Chiarucci 1996; figure 6: 37); secondly, there is mention of the caput Ferentinae in connection with the obscure practice of praetor ad portam, mentioned by the sources (Fest. 276 L.). It probably originated from the “liminal” nature of that place (Ampolo 1990:127). It might then make sense to believe the porta Ferentina to be one of the accesses to the sacred region of the Latin communities.

Caere The case of Caere is one of the most stimulating in southern Etruria: recent research has focused on the urban area, with a parallel analysis of its ager. The research was carried out by various scholars with differing approaches and intent, as such influencing the quality of information (Nardi 1988, 1989; Enei 1992, 1993). It has resulted in an enormous bulk of data, especially concerning the organisation of the ager. In a very detailed contribution, G. Nardi (1989:53) published a distribution map of the suggested sacred areas which have come to light in the urban area on the basis of survey data and in a few cases, based on data from previous excavations at the site (figure 10). Compared with the urban sacred areas, the picture of the ager is not equally detailed, with the exception, of course, of the sanctuary of Pyrgi, one of the harbours of Caere and dedicated to Uni/Astarte, Ilithyia/Leukothea, Suri and Cavatha (figure 11: 28; Colonna 1996b). Other cases are equally interesting: the sacred area of Monte Tosto, in close contact with the Orientalizing monumental mound from the end of the 8th century BC, has been interpreted as a sign of the “tyrannical” character of the power wielded by the city: this fact lends a particular significance to a site with a strong aristocratic character

In addition to this, there is still the third “level” of cults, situated in the foothills but closely connected with the Latin plain. The existence and distribution of most of the cults can probably be explained by the need to attribute limits to the plain and fertile zones, controlled by the centres lying south-east of Rome (mainly Lavinium and Ardea), these being at the same time in contact with the coast (figure 6: 18, 23). In this vast area, roughly contained between the modern towns of Pomezia, Ardea, Velletri and Albano, there are a relevant number of sacred areas, mostly dating to the Hellenistic period, but sometimes dating back to the late archaic phase. These flourished as a consequence of their direct contact with the fields (Melis & Quilici Gigli 1982, 1983, Chiarucci & Gizzi 1989, Bedini 1991). For the most part, these sanctuaries have indeed 261

RPC 2000 conference proceedings

ticular points in the territory and linked to the organisation of the ager: archaeologically, they are visible from the beginning of the 6th century BC onwards; in many cases their apparent rarity is due to the poor condition of surface finds. At Caere, the sacred areas mark the limits of former enlargements of the city and its ager: most parts of them show a continuity up to the Hellenistic age, signalling the different phases of development of the rural settlement system. This relationship between the centre and immediate periphery emerges particularly from both the archaeological record and the location of these presumed sanctuaries, apparently distinguishing a first level of the ager (i.e. the so-called suburban area, extending to the second line of sanctuaries, as such indicating probably the limit of the peri-urban cemeteries which were under the direct use and control of the city. This might be the case with the sanctuary of Monte Abatoncino; figure 11: 40). The rural (oppidani) sanctuaries (such as those of Griciano (?) and Procoio di Ceri, respectively at the western and eastern sides of the city; figure 11: 38, 41) give access to a wider zone, less directly controlled by the city and most probably organised by the secondary centres. It is a densely populated area, presenting different types of settlement. Apart from the sites on tuff plateaux (e.g. Castel Campanile; figure 11: 43), it is interesting to observe how, in a relatively limited area, large rural sites, large necropoleis, small rural sites and small necropoleis co-exist from the second half of the 7th to the 6th century BC (Enei 1993). The sacred areas linked to harbours and lagoons share the function of protecting the coasts, allowing contact between the local people and foreign traders or sailors. They are rich in foreign cultural features and could represent a more complex category than we deduce from the known evidence.

(figure 11: 37; Torelli 1981). A synthetic study of the ager caeretanus has not appeared yet. M. Rendeli (1993:283-367) has mainly collected the published evidence, and does not include the recent data brought to light by the surveys of F. Enei (1992, 1993:34-43), which contain new material about sanctuaries (figure 2). A general comment on a plan including most of the known cult places of the area (in a longue durée perspective) may provide some points for further discussion (figure 11). A concise and rough classification of the most significant categories of sacred areas may be organised as follows: a) Urban sanctuaries: identified on the site of the centre, these have been recently mapped through a comparison between old and new data (figure 10; Nardi 1989) ; b) Suburban sanctuaries: placed very close to the urban plateau, these mark the so called “area of respect” of the centre (Colonna 1985a, Nardi 1989, Merlino & Mirenda 1990); c) Sanctuaries marking the “area of respect of the ager”: these are usually placed in connection with or in proximity to the peri-urban cemeteries and may signal a limit within the main centre (inhabited area and surrounding necropoleis) and the ager (controlled by secondary centres) (Colonna 1985a, Riva & Stoddart 1996, Enei 1993); d) Sanctuaries in the ager, or “oppidani sanctuaries”: these are widely distributed in the territory politically controlled by Caere, and are controlled by the secondary centres (oppida), to which these cult places seem to be related (Colonna 1985a, Nardi 1988, Enei 1993, Gentili 1999); e) Sanctuaries linked mainly to natural phenomena (above all cold or hot springs, associated with sulphur eruptions): their nature and distribution have been emphasised by L. Gasperini (1988b); they are present in the ager of Caere, with cults in some cases dating back to the Archaic period (figure 11: 29, 32); (Gasperini 1976, Cosentino & Sabbatini Tumolesi 1989). f) Sanctuaries belonging to harbours and probably placed along the banks of lagoons: besides the well-known case of Pyrgi, the hypothesis of the presence of lagoons near the other harbours of Caere is underscored by the discovery of a possible Archaic sacred area at Ponte del Lupo along the via Aurelia, in the area facing the coast among Palo and Marina di San Nicola corresponding to the site of Alsium (figure 11: 42; Enei 1993). g) Frontier sanctuaries: these protect the outer boundaries of the ager, usually separating two polities: the northwestern area of the territory of Caere has revealed a dense presence of sacred areas, close to the ecological boundary formed by the Tolfa mountain range (Zifferero 1995). An organisation of boundaries can be presumed in that part of the territory of Caere adjacent to the Veian-controlled area, traditionally identified as a buffer zone corresponding to the Arrone valley (figure 11: 5, 7-9, 12-13, 15-16; Tartara 1999). Discussion Sacred areas may be organised as a network, placed at par-

The above model should be applicable to the territory traditionally recognised as under the influence of Caere: it shows how it is possible to give an interpretation of the ritual landcape, if one takes into account a precise map of the division of the ager based on the sacred areas. Such a reading may lead to a more integrated interpretation of the ancient landscape: it is interesting to observe that Rome has been analysed using a model departing from the round shape of the centre and of its surrounding landscape, respectively formed by montes and colles, pagi and ager. In the case of the ager caeretanus the ratio of the distribution of sanctuaries seems to follow successive limits, starting from the coastal line: a) The first set of cult places taken together re-stated the border of the land and of the sea, according to the hypothesis of Riva and Stoddart (1996), and the set also marked the territory, for those who approached that part of it, placed under the influence of the city. b) The second set delimited the coastal plain, densely scattered with rural sites, and extended up to the line signalled by the emergence of the tuff plateaux. The evidence for this division is quite scarce, but should con262

Landscape Perception in Archaeology

Bedini, A. (ed) 1991, Santa Palomba. Archeologia e storia, Roma. Brocato, P. & S. Fontana 1989, La riscoperta di un’iscrizione sacra presso il Casale L’Umiltà (Nepi - VT), in Ricognizioni Archeologiche 5: 29-37. Buranelli, F. 1997, Il carro di Roma dalla Tenuta di Roma Vecchia sull’Appia Antica, in Emiliozzi, A. (ed), Carri da guerra e principi etruschi, Roma: 191-193. Cambi, F. (ed) 1996, Carta archeologica della provincia di Siena. II. Il Monte Amiata (Abbadia San Salvatore), Siena. Cambi, F. forthcoming, Culti e storie locali, in La gestione del territorio. Memoria, partecipazione, sviluppo della ricerca, proceedings of the conference, Saluzzo 2000. Camilli, A. 1998, Interpretando l’assenza: note sul ruolo politico ed economico dell’ager publicus nell’espansione romana, in Pearce, M. & M. Tosi (eds), Papers from the EAA Third Annual Meeting at Ravenna 1997. I: Pre- and Protohistory, BAR International Series 717: 250-255. Camponeschi, B. & B. Nolasco 1982, Le risorse naturali della Regione Lazio 7. Roma e i Colli Albani, Roma. Capogrossi Colognesi, L. 1988, La città e la sua terra, in Storia di Roma I, Torino: 263-289. Carafa, P. 1998, Le frontiere degli dei. Osservazioni sui santuari di confine nella Campania antica”, in Pearce, M. & M. Tosi (eds), Papers from the EAA Third Annual Meeting at Ravenna 1997. I: Pre- and Protohistory, BAR International Series 717: 211-222. Carafa, P. 2000, Le guerre e le conquiste di Romolo”, in Carandini, A. & R. Cappelli (eds), Roma. Romolo, Remo e la fondazione della città, Milano: 340-342. Carandini, A. 1997, La nascita di Roma. Dèi, Lari, eroi e uomini all’alba di una civiltà, Torino. Cecamore, C. 1996, Nuovi spunti sul santuario di Iuppiter Latiaris attraverso la documentazione d’archivio”, in Pasqualini, A. (ed), Alba Longa. Mito, storia, archeologia, Roma: 49-66. Cerreti, C. 1987, Assetto territoriale e religione nel Lazio protostorico. Note preliminari per una ricerca, in Rivista Geografica Italiana 94: 1-29. Chiarucci, P. 1996, Viabilità arcaica e luoghi di culto nell’area albana, in Pasqualini, A. (ed), Alba Longa. Mito, storia, archeologia, Roma: 317-333. Chiarucci, P. 2000, Alba Longa, in Carandini, A. & R. Cappelli (eds), Roma. Romolo, Remo e la fondazione della città, Milano: 219-221. Chiarucci, P. & T. Gizzi 1989, Nuovo culto di campagna ai margini dell’Albanum, in Documenta Albana 11: 21-30. Coarelli, F. 1987, I santuari del Lazio in età repubblicana, Roma. Colonna, G. 1976-1977, La dea etrusca Cel e i santuari del Trasimeno, in Rivista Storica dell’Antichità 6-7: 45-62. Colonna, G. 1977, La presenza di Vulci nelle valli del Fiora e dell’Albegna prima del IV secolo a.C., in La civiltà arcaica di Vulci e la sua espansione (Atti X Convegno di Studi Etruschi e Italici), Firenze: 189-213. Colonna, G. (ed) 1985a, Santuari d’Etruria, Milano. Colonna, G. 1985b, Il Lucus Ferentinae ritrovato?, in Archeologia Laziale 7: 40-43. Colonna, G. 1988, Il lessico istituzionale etrusco e la formazione della città (specialmente in Emilia Romagna), in La formazione della città preromana in Emilia Romagna (atti del convegno), Bologna: 15-36.

sist mainly of the sanctuaries placed at the beginning of the hill system (both on the western side, consisting of the much more pronounced hills forming the Ceriti mountain range, and on the eastern side, characterised by elongated tuff plateaux incised by the Sanguinara, Cupino and Tre Denari streams (figure 11: 18, 30). These sanctuaries probably mark the line of the hilly woodland areas, coinciding with an average height of about 200 metres above sea level; it is probably the zone of the lucus devoted to Silvanus, remembered by Virgil (Aen. VIII 597-602): this zone is characterised by a less organised rural pattern, dominated by a system of small oppida, in which open sites are rare. c) The third and last set probably concerns the zone of the mountain cults (at an average height of about 500 metres above sea level). The information on this is scarce and limited to some cults of the Roman age, which probably had their roots in the archaic period. But, even here we have information suggesting a hierarchy of sacred areas in the form of a cult of Diana, located at a lower height. In the proximity of the M. La Tolfaccia/villa della Fontanaccia (Allumiere), a coarse-ware crater was found with an exceptionally important Latin inscription, referring to the cult of this goddess, whose sanctuary ought to be located not far from the villa, placed on a terrace at the base of the Monte La Tolfaccia (figures 11: 19; 12; Vitali Rosati 1994). The presumed northeastern boundary of the ager caeretanus may also testify to a hierarchy of deities: I refer here to the altars of Monte Calvario (previously known as Monte Sassano), carved in the tuff rock (figure 11: 36; Gasperini 1984). The impression is that this mountain area, lying between the Mignone valley and the Sabatino Lake, and representing in the pre-Roman age a boundary between the former caeretanus, veiens, and Faliscan territories, would have been peripheral from an economic point of view, but certainly strategically important for access to the Tiber valley.

References Ampolo, C. 1981, Ricerche sulla lega latina, in La Parola del Passato 36: 219-233. Ampolo, C. 1987, Roma arcaica fra Latini ed Etruschi: aspetti politici e sociali, in Cristofani, M. (ed), Etruria e Lazio arcaico (atti dell’incontro di studio), Roma: 75-87. Ampolo, C. 1989, Boschi sacri e culti federali: l’esempio del Lazio, in Les Bois sacrés (Actes du Colloque International du Centre Jean Bérard), Naples: 159-167. Ampolo, C. 1990, Roma arcaica ed i Latini nel V secolo, in Crise et transformation des sociétés archaïques de l’Italie antique au Ve siècle av. J.C., Rome: 117-133. Ampolo, C. 1998, Frontiere politiche e culturali, in Pearce, M. & M. Tosi (eds), Papers from the EAA Third Annual Meeting at Ravenna 1997. I: Pre- and Protohistory, BAR International Series 717: 179-183. Ampolo, C. & M. Manfredini (eds) 1988, Plutarco. Le vite di Teseo e di Romolo, Milano.

263

RPC 2000 conference proceedings

Colonna, G. 1991, Acqua Acetosa Laurentina, l’ager Romanus antiquus e i santuari del I miglio, in Scienze dell’Antichità. Storia, Archeologia, Antropologia 5: 209-232. Colonna, G. 1995, Gli scavi del 1852 ad Ardea e l’identificazione dell’Aphrodision, in Archeologia Classica 47: 1-67. Colonna, G. 1996a, Roma arcaica, i suoi sepolcreti e le vie per i Colli Albani, in Pasqualini, A. (ed), Alba Longa. Mito, storia, archeologia, Roma: 335-354. Colonna, G. 1996b, Pyrgi, in Studi Etruschi LXI (1995): 440-446. Cosentino, R. & P. Sabbatini Tumolesi 1989, L’edificio termale delle Aquae Caeretanae, in Miscellanea Ceretana I, (Quaderni di Archeologia Etrusco-Italica 17), Roma: 95-112. De Polignac, F. 1991, La nascita della città greca, (trad.ital.), Milano. D’Ercole, V., V. Orfanelli & P. Riccitelli 1997, L’Abruzzo meridionale in età sannitica”, in Campanelli, A. & A. Faustoferri (eds), I luoghi degli dei. Sacro e natura nell’Abruzzo italico, Pescara: 21-24. D’Ercole, V., D. Testa & A. Zifferero 1998, Nuovi dati dalla stipe votiva di Ripa Maiale (Allumiere - Roma), in Preistoria e protostoria in Etruria (atti del terzo incontro di studio), Firenze: 529-532. De Simone, C. 1975, Il nome del Tevere. Contributo per la storia delle più antiche relazioni tra genti latino-italiche ed etrusche, in Studi Etruschi 43: 119-157. De Santis, A. 1997, Alcune considerazioni sul territorio veiente in età orientalizzante e arcaica, in Bartoloni, G. (ed), Le necropoli arcaiche di Veio, Roma: 101- 141. Di Stefano Manzella, I. 1992, Nuova dedica a Soranus Apollo e altre iscrizioni dal Soratte, in Mélanges de l’École Française de Rome - Antiquités 104(1): 159-167. Edlund, I.E.M. 1987, The gods and the place: location and function of sanctuaries in the countryside of Etruria and Magna Graecia (700-400 B.C.), Stockholm. Enei, F. 1992, Ricognizioni archeologiche nell’Ager Caeretanus: rapporto preliminare, in Herring, E., R. Whitehouse & J. Wilkins (eds), Papers of the Fourth Conference of Italian Archaeology 3. New Developments 1, London: 71-90. Enei, F. 1993, Cerveteri. Ricognizioni archeologiche nel territorio di una città etrusca, Ladispoli. Gasperini, L. 1976, Scoperte archeologiche a Stigliano (Canale Monterano), (Quaderni della “Forum Clodii” 3), Bracciano. Gasperini, L. 1984, Il monumento rupestre di Numerio Pullio nel Foroclodiense, in Archeologia Classica 36: 361-374. Gasperini, L. 1988a, Il santuario romano delle acque all’Arcella di Canepina (VT), Roma. Gasperini, L. 1988b, Gli Etruschi e le sorgenti termali, in Etruria meridionale: conoscenza, conservazione, fruizione, Roma: 27-35. Gentili, M.D. 1999, I santuari territoriali in età etrusca, in Pani Ermini, L. & S. Del Lungo (eds), Leopoli-Cencelle I. Le preesistenze, Roma: 77-87. Ghini, G. 1995, Il santuario di Diana a Nemi (RM): nuove ricerche, in Christie, N. (ed), Settlement and Economy in Italy 1500 BC to AD 1500, Oxford: 143-154. Grandazzi, A. 1986, La localisation d’Albe, in Mélanges de l’École Française de Rome - Antiquités 98(1): 47-90. Granino Cecere, M.G. 1992, Epigrafia dei santuari rurali del Latium vetus, in Mélanges de l’École Française de Rome Antiquités 104(1): 125-143.

Greco, E. 1992, Archeologia della Magna Grecia, Roma-Bari. Guarducci, M. 1971, Enea e Vesta, in RM 78: 73-118. Guidi, A. 1998, The emergence of the state in central and northern Italy, in Acta Archaeologica 69: 139-161. Guidi, A. 2000, Preistoria della complessità sociale, Roma-Bari. Guzzo, P.G. 1987, Schema per la categoria interpretativa del “santuario di frontiera”, in Scienze dell’Antichità. Storia, Archeologia, Antropologia 1: 373-379. La Rocca, E. 1984, La riva a mezzaluna. Culti, agoni e monumenti funerari presso il Tevere nel Campo Marzio occidentale, (Studi e Materiali del Museo della Civiltà Romana, 11), Roma. Lega, C. 1995, Topografia dei culti delle divinità protettrici dell’agricoltura e del lavoro dei campi nel suburbio di Roma, in Quilici, L. & S. Quilici Gigli (eds), Agricoltura e commerci nell’Italia antica (Atlante tematico di topografia antica - 1° supplemento), Roma: 115-125. Melis, F. & S. Quilici Gigli 1982, Luoghi di culto nel territorio di Ardea, in Archeologia Classica 34: 1-37. Melis, F. & S. Quilici Gigli 1983, Votivi e luoghi di culto nella campagna di Velletri, in Archeologia Classica 35: 1-44. Merlino, M. & T. Mirenda 1990, Caere, in Maffei, A. & F. Nastasi (eds), Caere e il suo territorio. Da Agylla a Centumcellae, Roma: 4-56. Molinos, M. & A. Zifferero (eds) 1998, Political and Cultural Frontiers, in Pearce, M. & M. Tosi (eds), Papers from the EAA Third Annual Meeting at Ravenna 1997. I: Pre- and Protohistory, BAR International Series 717: 177-258. Muggia, A. 1997, L’area di rispetto nelle colonie magno-greche e siceliote. Studio di antropologia della forma urbana, Palermo. Nardi, G. 1988, I caratteri naturali e la viabilità antica, in Cristofani, M., G. Nardi & M.A. Rizzo, Caere I. Il parco archeologico, Roma: 11-27. Nardi, G. 1989, Appunti sui santuari urbani, in Miscellanea Ceretana I, (Quaderni di Archeologia Etrusco-Italica, 17) Roma: 51-68. Naso, A. 1999, Un oppidum etrusco sul sito di Cencelle, in Pani Ermini, L. & Del Lungo, S. (eds), Leopoli-Cencelle I. Le preesistenze, Roma: 70-76. Perkins, Ph. & L. Walker 1990, Survey of an etruscan city at Doganella, in the Albegna valley, in Papers of the British School at Rome 58: 1-134. Rastrelli, A. 1998, Luoghi sacri ai confini del territorio di Chiusi, in Pearce, M. & M. Tosi (eds), Papers from the EAA Third Annual Meeting at Ravenna 1997. I: Pre- and Protohistory, BAR International Series 717: 202-207. Rastrelli, A. 2000 (ed), Chiusi etrusca, Chiusi. Rendeli, M. 1993, Città aperte. Ambiente e paesaggio rurale organizzato nell’Etruria meridionale costiera durante l’età orientalizzante e arcaica, Roma. Riva, C. & S. Stoddart 1996, Ritual Landscapes in Archaic Etruria, in Wilkins, J.B. (ed), Approaches to the Study of Ritual. Italy and the Ancient Mediterranean, London: 91-109. Ross Taylor, L. 1923, Local cults in Etruria, Rome. Ruiz Rodriguez, A. & M. Molinos 1989, Fronteras: un caso del siglo VI a.n.e., in Fronteras (Arqueologia Espacial 13): 121-135. Scheid, J. 1987, Les sanctuaires de confins dans la Rome antique. Réalité et permanence d’une représentation idéale de l’espace romain, in L’Urbs. Espace urbain et histoire (Ier siècle av. J.C. - III siècle ap. I.C.), Rome: 583-595.

264

Landscape Perception in Archaeology

Tartara, P. 1999, Torrimpietra, Firenze. Thuillier, J.P. 1985, Les jeux athlétiques dans la civilisation étrusque, (Bibliothèque des Écoles Françaises d’Athènes et de Rome 256), Rome. Torelli, M. 1981, Delitto religioso. Qualche indizio sulla situazione in Etruria, in Le délit religieux dans la cité antique, Rome: 1-7. Vitali Rosati, B. 1994, Iscrizione votiva dalla villa in località La Fontanaccia ad Allumiere, in Archeologia Uomo Territorio 13: 87-91. Zevi, F. 1997, I santuari “federali” del Lazio: qualche appunto, in Eutopia IV (2): 123-142.

Zifferero, A. 1991, Forme di possesso della terra e tumuli orientalizzanti nell’Italia centrale tirrenica, in Herring, E., R. Whitehouse & J. Wilkins (eds), Papers of the Fourth Conference of Italian Archaeology 1. The Archaeology of Power 1, London: 107-134. Zifferero, A. 1995, Economia, divinità e frontiera: sul ruolo di alcuni santuari di confine in Etruria meridionale, in Ostraka 2: 333-350. Zifferero, A. forthcoming, La geografia del sacro nelle società complesse: ipotesi per una ricerca sull’Italia medio-tirrenica preromana”, in Molinos, M. & A. Zifferero (eds), Primi Popoli d’Europa, Abaco Edizioni, Forlì.

265