REGIONAL ARCHAEOLOGICAL PATTERNS IN THE SIBARITIDE PRELIMINARY RESULTS OF THE RPC FIELD SURVEY CAMPAIGN 20001

MARTIJN VAN LEUSEN & PETER ATTEMA Grol/ingel/ II/slilllle af Archaeology, Grol/il/gel/, Nelherlal/ds

ABSTRACT: In the late summer of 2000 members of the Groningen Institute of Archaeology carried out a three-week archaeological field survey in the Sibaritide (, South ) within the framework of the Regional Pathways to Complexity project, a joint undet-taking by the archaeological instihltes of Groningen University and the Free University of Amsterdam. The survey area is located opposite the protohistorical site ofFrancavilla Marittima that is currently being excavated by the Groningen Institute of Archaeology and h'ansects a series of marine and fluvial terraces along the river Raganello. The survey covered 125 hectares intensively and 315 hectares extensively. The aims of the survey were twofold; to compare the results obtained by modern survey methods such as adopted by the RPC project with those from earlier region-wide topographical surveys in order to identifypossibie biases in the older regional data sets; and to add to the em'lier piecemeal site catchment surveys carried out by the Groningen Institute of Archaeology as part of the excavations in the period 1991-2000. FolIowing a description of the survey and finds processing methods employed, the results of the survey are presented and discussed in two main chronological phases (protohistory and Hellenis­ tic/Roman). Finally, implications fo r future research in the catc1unent of Francavilla Marittima are laid out.

KEY WORDS: Archaeological survey, South Italy, Francavilla Marittima, protohistory, Bronze Age, lron Age, Hellenistic period, Roman period, settlement patterns, bias modeIling.

l. INTRODUCTION application ofcareful 's ource criticism' ofthe available archaeological data. Fieldwork, especiaIly survey, has This article presents a preliminary report on the Oc­ been used in an attempt to fill gaps and assess biases tobel ', 2000 fi eld walking survey campaign of the in these data. Regional Pathways to Complexity (RPC) proj ect in the Sibaritide (nOl-thern Calabria, Italy), fo cusing on 2. THE SIBARITIDE the shldy of spatial and chronological pattems in the archaeological record and their interpretation in terms of regional settlement history. The Sibaritide is an alluvial plain on the Gulf of From 1991 onwards, the Groningen Instihlte of Taranto (part of the Ionian Sea), bounded on the Archaeology has conducted annual excavations at the landward side by the Pollino massif(to the north) and hilltop sanctuary and settlement areas of Timpone the Sila mountain range (to the south). It is named Motta at Francavilla Marittima, with occasional limited after the ancient Greek colony of Sybaris, which lies field surveys in the immediate neighbourhood of that in the approximate centre of its coastline. Since, on site (Attema et al., 1998, 2000; Maaskant-Kleibrink, the one hand, its history of settlement and land use 2000 (with references to em'l ier publications)). In 1997, has been largely determined by the possibilities af­ the Sibaritide became one of three regions being fo rded by the physical and natural environment and, studied in the context of the RPC proj ect run jointly on the other, its archaeological record has been con­ by the archaeology departments of the University of ditioned (through the medium of recent land use and Groningen and the Free University ofAmsterdam (both land cover) by that same environment, a brief review in the Netherlands).2 The central aim ofthe RPC proj ect of the geology, c1imate, and land use of the study area is to study regional processes of centralisation, will be given here. urbanisation, and colonisation which have occurred from the Iron Age onwards, and especiaIly to evaluate 2.1. Geology the relative importance of internal and external contributing fa ctors to these processes. Among several The Sibaritide is a roughly triangular coastal lowland, approaches toward that aim is the detailed shldy ofthe fo rmed mainly in the Quatemary by the accumula­ relations between the regional landscape and its tion of erosion products from the Pollino mountain settlement history at various spatial and temporal range to the nOl-th and the Sila mountains to the south. scales. It requires the modeIling of landscape A series of stepped fo ssil coastlines, up to an alti­ characteristics within a GIS environment, and the tude of several hundred meters above the present sea 398 M. VAN LEUSEN & P. ATTEMA

15km (

lonian Sea croti

A .Syboris

c I II)

Fig. I. Major landscape units of the Sibaritide. The coastal and alluvial plains (D) are surrounded by terraced (A) and non-terraced (B) marine and fluvial sands and conglomerates which merge into the lower slopes of the Pollino and Sila ranges (C). level, can be recognised in these loose sediments and ever, have in historie times been minimised by regu­ conglomerates. Each is represented by a relatively latory works such as river bank reinforcements. steeply sloping 'cliff and a weakly sloping (5%) fan­ Given the velY dy namic geology of the Sibaritide, like 'terrace' . These stepped coastlines possibly were in which for example the Greek colony at Sybaris lies the result of the interplay of tectonic upliftof the area below six to eight meters of alluvial deposits, the and changing sea levels due to climate changes in archaeological record of the plain can only be inter­ the Quaternary. Fluvial erosion and deposition must preted with reference to palaeogeographical recon­ have accompanied these changes in climate through structions. However, the late Holocene deposits are changes in vegetation cover, river regime and sea le­ genera Ily too truck for a manual auguring progralllille vel. As vegetation cover increased at the onset of the to succeed, although the mechanical augerings taken Holocene, the rate at which sediment was transported in the course of the 1960s search for Sybaris dem­ must have slowed down. However, the sedimentation onstrate that a dedicated coring program is not im­ rate in the coastal plain could still have been con­ possibie (Rainey & Lerici, 1967).3 Turning to the siderable in human terms, and the alluvial deposition mountainous inland parts of the Sibaritide, geomor­ of several meters of sediments could easily have phology and land use are such that surveys of large changed local hydrogeography. Erosion in the hin­ contiguous areas are impossible. These r,>,IO zones are terland and sedimentation in the coastal lowlands are therefore generally unsuitable for survey, and the ongoing proces ses, though now mostly as a conse­ oppOltunistic study of available sections is the only quence of violent winter storms. Their effects, how- low-budget option open to researchers. The location Regional archae% gical paltems in the SibC/l'itide 399

of our survey transect across the transitional zone, and small almond groves, but a considerable amount consisting mainly of the aforementioned marine and of land especiaIly in and around the dry gullies is still fluvial telTaces of the Raganello river, was therefore unused and covered in the local thorny Illaechia. The largely determined by the geomorphological stmc­ prevalence of the latter increases with elevation, ture of the region. However, as will be made clear whilst the access ibi lity decreases. A marked contrast below, there were other archaeological reasons to exists with the land utilisation types mapped in the concentrate research in this zone as well. early 1950s (CNR 1956a, b); at that time, less inten­ The legend of the most recent detailed geologi­ sive forms of agriculture such as olive groves and caI map of the area (CGC 1969) allows us to estab­ grassland occurred over larger parts of the survey lish that there are four main formations in the study zone, at the expense of the arable (macchia appears transect. From lowest to highest these are: not to have been mapped). l) Alluvial deposits held in place by vegetation Ol' artificial works; morphologically, these consist of a single alluvial fan emanating from the Raganello 3. RESEARCH HISTORY valley; 2) Yellow-red sands and pebbly sands associated FolIowing the discovery of Sybaris and its Hellenis­ with conglomerates consisting of well-cemented tic and·Roman successor towns through the combined rounded pebbles; this fonnation inc1udes clayey efforts of tbe University of Pennsylvania Museum intercalations and, locally, banks of yellow-green clay s and the Lerici foundation (Lerici, 1960: pp. 303-337; and silts; morphologically the formation consists of Rainey & Lerici, 1967; Rainey, 1969: pp. 26 1-273), three terrace leveIs. The formation is highly permeable an approximately 21 by 24 km (500 kIn2) area4 in the but can resist erosion well locally, depending on the centre of the region was surveyed by a team under degree of cementation of the deposits; tbe direction of the young Lorenzo Quilici, who is 3) Grey-blue clays, often intercalated with sand now a senior professor at the University of Bologna. lenses, and conglomerate sands. This formation occurs Quilici aimed both to pro vide a context for the ex­ only in one area neal' Lauropoli, is easily eroded and is cavations starting at Sybaris in 1969 and to record prone to slumping especiaIly where sandy layers surface archaeology in advance of land development abound; the permeability of the deposits is generally schemes funded by the Italian Cassa per il Mezzo­ low; and giorno (De Rossi et al., 1969: p. 147). In what fol­ 4) Well-cemented coarsely stratif ied conglomerates lows we sball refer to this survey, for brevity's sake, consisting of larg e calcareous pebbles and rounded as the 'Quilici survey', and to its results as the 'Qui­ sandstones, on occasion of pebbly Ol' large-grained lici data set'. FolIowing their desktop study of the sand s, locally with silty Ol' clayey lenses. With a records held by the Soprintendenza archeologica generally high permeability, the local sensitivity to delia Calabria, the study area was divided into three erosion of this formation depends on its degree of zones, each 'topograpbically' surveyed by one Ol' two cementation. members of the team. In addition to revisiting and evaluating the known sites, many new sites and mo­ 2.2. Climate and land use/land cover numents were discovered as well. The results were record ed on a total of 858 forms and 23 sheets of the A fairly detailed discussion of the current climate and 1: l 0,000 map series published by the 'Cassa per il land use of the Sibaritide can be found in D' Angelo Mezzogiorno', but were considerably condensed for and Orazie Vallino (1994: pp. 785-792), who note publication. The whereabouts of the original survey that the climate of the terraces and foothills is more archive are currently unkIlOwn. The nearly 800 sites pleasant than that of the coastal plain, with more of archaeological interest collected from documented frequent rains because of the surrounding mountain, evidence Ol' recorded in the fi eld by this team still and less oppressive heat in the summer and autumn provide the bulk of the archaeological record for the because of the wind and altitude. The highland is region - later additions being of a piecemeal nature even more wet and cool, and is suitable for trans­ - and are therefore the primary subject of the study humance in summer. The larger part of the survey repOlted here (see fig. 2). transect nowadays is given over to arable land and a Additional work on the protohistoric settlement mixture of old and new olive groves, the latter giv­ bistory of the Sibaritide was published by the exca­ ing rise to extensive soil disturbance (deep plough­ vator of Broglio di , Renato Peroni, based ing, leveIling) and cultivation of previously marginal on a series of surveys conducted since 1979 by the zones (especially steep slopes). The arable is appar­ University of Rome in conjunction with the Super­ ently used for grain crops and, in at least one re­ intendency for Calabria (Bergonzi et al., 1982; Pe­ corded case, maize for which a previous vineyard had roni, 1994; Peroni & Trucco, 1994).5 By contrast, recently been uprooted. Other lesser types of land use very littIe j)t'imary research has been done on either in the survey zone include grassland, a rubbish dump, the classical and Hellenistic Greek period, Ol' the 400 M. VAN LEUSEN & P. ATTEMA

* Protohistoric • Archaic .... Hellenistic D Hellenistic-Roman • Roman Imperial :-: Medieval

Fig. 2. Distribution of settlements by period from the Quilici survey (after De Rossi et al., 1969: fig. 3). Dithered areas: alluvial deposits. The rectangle indicates the area in which the RPC survey took place. subsequent late Republican and Imperial Roman pe­ largely unpublished preliminary and exploratory sur­ riod in the Sibaritide outside of Sybaris itself; here veys in the immediate surroundings of the Timpone we must rely on the general literature regarding Mag­ Motta, extending on occasion to neighbouring Monte na Graecia and the Roman Empire. Since 1991, the Sellaw(Feiken & Weterings, 1998) and nlliher along Groningen Institute of Archaeology has been in­ the foot slopes towards Broglio di Trebisacce (Haags­ volved in excavations at the protohistoric/Archaic ma, 1996). hilltop sanctuary of and its In 1994, Haagsma and Delvigne conducted un­ settlement. These excavations have given rise to systematic fieldchecks on settlement sites identified Regional archaeological patterns in the Sibaritide 401

by the Quilici survey between the Timpone Motta 1ectivity inherent in the methodology emp10yed fo r and Broglio (Haagsma & Delvigne, unpublished the survey. notes). These revealed that the lowest-lying of the Accordingly, the main aim of the SIBA'OO sur­ marine terraces were aiready settled in the Middle vey campaign has been to assess the influence these Bronze Age while the upper patiS of the coastal plain factors had on the quality of the extant archaeologi­ were not; a Late Bronze Age site was fo und on the cal record, that is, primarily of the Quilici data set. Timpa del Castello,. a rocky outerop above the present : Since we were unsuccessful in obtaining the neces­ village of Ftancavilla Marittima.6 Their survey yield­ sary infOlmation, this assessment had to be obtained ed no identifiable Iron Age Ol' Roman material, but through a combination of targeted desktop and fie1d did locate some of the 6th century BC Archaic ill1- studies. pasto missing from the earlier Quilici survey results. In order to assess the influence of land use/land Based on this work Haagsma (1996) suggested that cover around the time the Quilici survey took place, Roman intensive land use may have avoided the a land utilisation map depicting the situation of the fo othills and may have been concenh'ated in the coas­ early 1950s at a scale of 1:200,000 was digitised and tal plain. In 1995, and again in 1998, several small georeferenced (CNR 1956a, b).7 This map shows that field surveys were conducted by students of Kleibrink the Sibaritide plain was dominated at that time by in settlement and cemetery zones around the fo ot of arable (seminativo asciutto) but that a fa ir amount the Timpone Motta. The 1995 survey by Kleibrink of grazing land (pascolo ed incolto prodottivo) was and Waterbolk (reported in Haagsma, 1996) fo und also present; olive groves and arable dominated the a lot of material then interpreted as having been fo othills (to which fruit- and vineyards may be added 'washed off the Timpone', but only one small Ar­ especiaIly in the area around the town ofCassano allo chaic site; Haagsma and Attema also located a 6th Ionio); while the higher slopes of the Pollino were century wall segment and kilns associated with a mostly down to wood- and grazing land. settlement on the lowest plateau of the Timpone delia The frequency of Quilici sites with regard to land Motta. A subsequent reassessment of the stored mate­ use was tested fo r deviations from randomness, us­ ria I from this survey by Attema and De Haas resulted ing the x2-test after removing all alluvial areas be­ in the definition of several new sites, including the low 25 m as!. Table l A lists the preliminary results Archaic/Classical extension to the lron Age cemetety of this test, which indicate that there is a strong of Macchiabate belonging to the settlement on the correlation between land use and site density.8 As we Timpone della Motta (De Haas, 2001: pp. 18-19). can see, the frequency of sites in woodland (cats 13 Despite covering only very small areas and fe w and 14 combined) is low with only nine observed sites, these surveys provide significant additional sites where 22 would have been expected given that information in the fo rm of evidence fo r periods 'mis­ sing' from the Quilici data set, such as the Archaic 3.3 percent ofthe study area is covered in woodland. (6th century BC), and fo r continuity and change in Similarly, 52 observed sites in mixed vineyard-olive the proces ses of centralisation and acculturation be­ grove (cat. 9) against an expected number of 24 rep­ tween the indigenous Oenotrians and the Greek colo­ resents a significant deviation from randomness. Of nists. the land use types that occur most frequently, dry arable with trees (cat. 2) is slightly fa voured with 56 sites against an expected 39, while productive grass­ 4. PATTERNS AND BIASES and uncultivated lands (cat. 19) is slightly avoided with 75 sites against an expected 103. However, in The interpretation of regional archaeological records order to satisfy all of the requirements fo r this test, in tenns of a history of settlement and landscape must some land use types must be grouped together so that be informed by an assessment of the biases that might the expected munber of sites per category is less than concea1 Ol' produce patterning in those records. A five in les s than 20% of categories (see fig. 3). cursory look at the geological map of the Sibaritide The results of a second x2-test (table IB) confirm shows that alluviation of the coastal p1ain must have that land use and site location correlate very signifi­ been a fa ctor of major significance, resulting in an cantly (far above the 0. 1 % level of significance). almost total lack of find spots below a quota of 25 However, they also show that none ofthe three most m asl and in the alluvial plains of the and Cos­ frequent land use types (dry arable, olive groves, and cile rivers extending into the hinterland (dithered rough grazing lands) has had a dramatically positive areas in fig. 2). Other fa ctors which previous stud­ 61' negative influence on the number of sites discov­ ies have suggested to be a priori causes of signifi­ ered. Interpretation of these results is not straight­ ca nt biases in the archaeological record are: the land fo rward, not just because land use at the time of the use and land cover (abbreviated in what fo llows to Quilici survey may have been significantly different 'LULC') at the time of the survey, the accessibility from the land use that was mapped in 1956, but also given the contemporary infrastructure, and the se- because the clustered nature of archaeological site 402 M. VAN LEUSEN & P. ATTEMA

Table I. Reslllts ofa x' test of 666 QlIilici sites against 1956 land lItilisation types. 751 sites in region of 350 km'.

Site characteristics % cover Expected siles Actllal sites X'

(A) Ro\\' res/llls

(O) No data 85 (I) Seminativo (ascilltto) 46.4 309. 1 325 0.8 17 (2) Seminativo arborato (ascilltto) 5.9 39. 1 56 7.348 (3) Seminativo irriguo 0.6 4.2 4 0.010 (4) Seminativo arborato irrigllo 0.5 3.3 O 3.318 (7) Vigneto 2. 1 13.8 12 0.23 1 (8) Uliveto 20.5 136.8 122 1.597 (9) Vigneto-Uliveto 3.5 23.6 52 34.225 (IO) Agrumeto 0.3 2.0 2 0.000 (Il ) Frutteto (frlltta polposa) 0. 1 0.5 O 0.494 (13) Bosco cedllo 1.1 7.2 I 5.387 (14) Bosco d'alto fusto 2.2 14.7 8 3.042 (19) Pascolo ed incolto prodllttivo 15.5 102.9 75 7.568 (20) Sterile 0.2 l.2 O 1.240 (21 ) Insediamenti ed altre forme 0.8 5.2 9 2.854 (99) Acqlle 0.4 2.4 O 2.449

Totals 100.0 666.0 666 70.579

(B) Res/llls (ifterreclassificalioll

(O) No data 85 (I & 2) Seminativo ascilltto 52.6 350. 1 381 2.727 (3 & 4) Seminativo irriguo l.1 7.6 4 1.683 (7) Vigneto 2.1 13.9 12 0.250 (8) Uliveto 20.7 137.5 122 1.756 (9) Vigneto-Uliveto 3.6 23.7 52 33.72 1 (IO & Il) Agrllmeto / Frlltteto 0.4 2.5 2 0.094 (13 & 14) Bosco 3.3 22.1 9 7.727 (19) Pascolo ed incolto prodllttivo 15.5 103.5 75 7.838 (2 1 ) Insediamenti ed altre forme 0.8 5.2 9 2.795

Totals 100.0 666.0 666 58.59 1

record s makes them notoriously subject to spatial of JI/terall/niu/II (De Rossi et al., 1969: pp. 9- 10).10 autocorrelation effects.9 Figure 3 shows the land use around the c1uster to be As an example of the fo rmer, it seems Cl priori the otherwise rare category of mixed vine- and ol­ unlikely that the Quilici team could have observed ive yards (table l B, cat. 9). In combination with the even as many as 75 sites under the very adverse fa ct that neither vineyards nor olive groves in mo­ visibility conditions of grass- and uncultivated land noculture (cats 7"nd 8) give rise to anything like the (see table lB, cat. 19). And indeed it may not be observed high frequency of archaeological site loca­ coincidental that the Quilici sites fa lling within the tions in the mixed unit, the high frequency must be SIBA '00 survey transect (see Section 7 below) ap­ caused by the spatial clustering of the sites rather than peal' to c1uster within one such zone of grass- and by patticularly favourable land use at the time of the uncultivated land. Perhaps these areas had aIready survey - whatever that may turnout to have been. been brought under the plough by the time the Quilici The distinct linear c1ustering of sites, which Qui­ team surveyed the area in the late 1960s. Access to Iici himself interpreted as an indication of the exist­ the original survey records will play a very impor­ ence of 'villages' along proto- and early historic tant rale in assessing the significance of the land use routes, could also be caused by biases in the Quilici changes that may have taken place in the 1960s. survey itself. With a survey universe (excluding the Quilici himself adduced an example of the latter coastal plain) of som e 350 km2 it is clear that the effect, interpreting one dense c1uster of sites located survey team could not possibly have systematically near the town of Eianina as part of a large roadside covered everything. It would instead have had to settlement within the territory of the Roman colony fo cus on those areas that were most Iikely (from Regional archaeological patterns in the Sibm'itide 403

•

• • I . • • I •• lonian Sea

t • ··•.. � . .. .-_· • • • • � r· • • • • • ••• • • • • .. �. • ... . • • • • • • . ."/• • • • • • • • .. • • • • . • • JI • • .. � ,

• • •

• • • •

9 13,14 c::::J20,21 5km mrrm10,11 c::::::::J l 9

Fig. 3. Distribution of Quilici sites across 1956 land utilization types. Legend: see table IB. Dashed line indicates approximate location of the 25 m elevation contour. previous finds) to contain sites and which could be talled roads digitised off the IGM 25V map series. reached within a reasonable amount of time by road The proximity of sites to infrash'ucture seems fa irly or track. Figure 4 again shows the Quilici data set, straightforward in some areas, especiaIly if we re­ this time against the background of a relief map, member that minor tracks were not digitised. and in relation to towns and meta lied and semi-me- 404 M. V AN LEUSEN & P. ATTEMA

5km

Fig. 4. Association of the Quilici data set (dots) with modem accessibility circa 1956 (paved mads only); 25 m elevation contour indi­ cated by dashed line.

These preliminary studies therefore indicate that 5. THE SURVEY: APPROACHES the potential of nnther bias 1110delling for the Quilici data set must be explored when the original research The approach taken by us was to select a represen­ archive has been located. The second method for tative transect through the transitional zone between assessing biases in the Quilici data set is through field the coastal plain and its mountainous hinterland, studies designed to detect archaeological materials consisting of marine and fluvialterraces (the firstand present in areas, or from periods, that Quilici and his second geological formations discussed above), and team might have ignored or been unable to detect. to see how the recorded sizes, locations, interpre- These studies are the subject of the next section. Regional archaeological patterns in the Sibaritide 405

Fig. 5. The SIBA2000 fieldwork transect (exeluding Monte San Nicola). Extensive (black) and intensive (white) survey areas. Plus signs: Quilici sites; circles: RPC sites and scalters. tations and dating of the sites mapped by Quilici and high-intensity survey of all available fieids within a his team compare with the current archaeological sur­ 5 by l km transect running parallel to the Raganello fa ce record. The transect was approximately 6 km river, and covering all fo ur main geomorphologic long and 1.5 km wide and located between the val­ units (fluvio-marine terraces). In this survey, agricul­ ley of the Raganello river and the modem town of tural fi eids were subdivided by pacing into units Lauropoli (fig. 5). The second geological fo rmation, approximately 50 by 50 m (0.25 ha) in size, and these which is morphologically divided into fo ur distinct were walked at 10 111 intervals, with all manmade levels Ol' 'terraces', is today used for the most inten­ materials collected except those that were of obvi­ sive fOrIns of agricultural exploitation, consisting of ously recent date. Further samples were collected if mostly arable fieids but with large sections given circul11stances (such as al1 increase in finds density) over to olive culture; due to the extremely restricted warranted. In order to assess the effect of differen­ surface visibility in the other three units, our survey tial vis ibi lit Y on the recovery rates, five fa ctors af­ took place almost entirely within this unit. Archaeo­ fe cting the recovery of archaeological materials were logical sites mapped by Quilici and his team are also recorded separately for each collection unit, on a exclusive to this unit. scale of l to 5: stoniness, shadiness, vegetation cover, The SIBA'OO survey was conducted using two tillage/dust, and amount of recent materiai on the teams working independently and studying the same surface; an estimate of the total visibility, again on ' general area at two different spa tia l scales. The first a scale or l to 5, was also made. team used metllods developed in recent years for a 406 M. VAN LEUSEN & P. ATTEMA

The second team conducted a much more ex­ based on the macroscopic properties of the ceramics tensive and site-oriented survey in an area directly - fabric, inclusions, colour, thickness and morpho­ neighbouring the intensive survey transect, and over­ logical characteristics. Since most of the surface lapping it in several parts in order to provide mate­ material is very fragmented and abraded, a primm'y rial fo r comparing the two. The main aim of this division into three clear-cut ware categories was survey was to check whether the broad settlement used: patterns mapped by Quilici and his team were up­ 1) Jmpasto, which is the handmade pottery charac­ held, by locating and describing any 'sites' occur­ teristic of the protohistoric periods. It is irregularly ring in the area; it therefore used agricultural fieids fired and often has a burnished surface; as its collection units, and a walker interval of 15 m. 2) Com'se wares, which are turnedor finished on a Once a site had been identified by this team, a more wheel and characteristic of the Archaic to Roman detailed survey was made at 5 m intervals in orde r periods. They are generally orange firing wares made to map finds density contours and make diagnostic of a clay with fine to coarse inclusions; collections. 3) Depurated wares, which are wheel-turned and Only two forms were used during the survey: a either pale- or orange firing and have either a hard Ol' a Unit Form to record properties of the collection unit soft powdery surface. Depending on the clay base and including bias factors and number, type, and sum­ firing characteristics, these wares can be classified into mary contents of the samples (terrned 'Bags') col­ various periods from the late Iron Age to sub-recent lected; and a Bag Form to record the contents of times. each of these samples. Unique Bag ID's are forrned by sub-numbers from unique Unit ID's; a water Classifications within this primal)' division are based proof ticket with the Bag ID was kept with the finds on characteristics relating to function (e.g. categOl)' throughout the processing. This system, developed 1 'tile' is a functional category; category 2a 'com'se out of previous administrative systems used at the thick ware' is functionally interpreted by the addi­ GIA and AIVU, is described in more detail elsewhere tion 'p ithos'). Sherds belonging to these broad func­ (Van Leusen, 2002: Chapter 8). Mapping of spatial tional classes are fa irly easily identified even in a collection units was done using a combination of in­ much-abraded state. The association of certain types dependent single-receiver GPS measurements and of tiles, pithoi and datable pottery sherds such as 1:10 ,000 scaled field maps; again, a more detailed black gloss and ferm sigillata may then indicate the discussion is available (Ryan & Van Leusen, 2000). existence of a fa rmstead at a particular time and place.

6. FINDS PROCESSING Categories JO, 13: bnpasto wares and 'grulI1i' JlI1pasto wares are hand-fOl'med, with coarse tempel' Rather than making any decisions about the signifi­ and varying types of surface finish. In principle they cance of the finds in the fieid, the intensive survey can be classifiedby fo rm, thickness, finish and deco­ team simply collected the fi nds fo r processing and rations into various wares, but the SIBA'OO material interpretation at the survey base. For the purpose of allows only a simple division to be made between tracking individual fi nds bags, a simple system of the well-burnished early to middle Iron Age ware check-boxes was used on the field fo rms for each found near Monte San Nicola and the undiagnostic collection unit. Processing at the survey base began and often abraded sherds of late Iron Age ware found with washing and drying, after which all finds were elsewhere. Grumi (burnt daub) appears as orange-red classified by the second author according to the sys­ low-fired crumbs in ploughed fi eids and by itself tem laid out in table 2. Both survey teams together cannot be dated; in association with impasto or dated collected some 210 kg of material (some 6000 finds) archaic wares it may be assigned to one of these from 610 collection units, ofwhich 70 kg were clas­ periods. sified during processing as largely recent Ol' sub­ recent tile/brick (subsequently discarded). Table 2 Categories Ja, 2a-c, 5a: Com'se wares provides the breakdown of these totals by material These are coarse wheel-made wares of red firing fab­ category. The column headed '%arch' gives the rela­ rics containing visible temper, production of which tive weights per category when considering only pre­ began in the Archaic period and continued through modern ceramics (totalling 129.2 kg). the Hellenistic period. Wares can be c1assified by fonn and thickness into tiles, storage, kitchen, and e 6. 1. System of classification table war s. A further fo rm-based subdivision of coarse wares is envisaged, but must await closer The main aim of the classification of ceramic wares study of the material (cf. Attema et al., 2000: Ap­ used in the SIBA'OO survey was to assign dates and pendix l) and comparison with collections from re­ functions to the material found. The classification is gional excavated contexts.11 In total, coarse material Regional archaeological pattel'l1s in the Sibaritide 407

Table 2. Material categories used in summary classifications of tbe RPC field survey campaign 2000 survcy finds, \Vithprelimin ary total counts, \Veights, and percentage by \Veight of pre-modem ceramics.

Cat. Description Count We ight (g) % arch

Ti/e 230 36300 la Coarse tile 144 22955 17.8 Ib Depurated antique tile 84 13220 10.2 d '(sub-) recent tile 2 125 2 Cow'se wares 1208 45321 2a Coarse tbick \Vare (pithos) 175 27970 21.6 2b Coarse medium thick \Vare 942 16820 13.0 2c Coat'se thin \Vare 91 531 0.4 3 Depuraled wares 2/34 23335 3a Depurated orange \Vare 457 4783 3.7 3b Depurated paIe \Vare 1677 18552 14.4 4 Fille ",ares 70 7/8 4a Black gloss/blanded \Vare 62 638 0.5 4b Tena sigillata 6 75 0.1 4c Thin hard painted fine \Vare 2 5 0.0 5 flldelerlllillale ",ares 2102 91712 5a Indeterminate coarse \Vare 300 8532 6.6 5b lndeterminate depurated \Vare 396 13567 10.5 5c Indeterminate (sub-) recent \Vare 1406 696 13 6 eOlllbed wares la 140 0./

Olher wares alld lIIaleria/s 7 Glazes \Vare 11 225 0.2 8 Hard red \Vare, medium thick 32 485 0.4 9 Lithics IO 75 IO Impasto \Vares 42 730 0.6 13 Grumi 72 2625 14 Glass 5 55 15 Wa ster 8 2760 16 Slag 19 62 15

Totals 5936 210696 100.1

(cats la, 2a-c) makes up almost 53% by weight, and form and, being undiagnostic, are then assigned to the Hellenistic to Roman material (cats I b, 3, 4, 5a, category 5b. b) some 46%. Wheel-turned coarse wares lacking 'ar­ chaic' characteristics of colour, tempel' and surface Categories 4 and 6: Fine and Combed wares finish may date fr om the Hellenistic or Roman peri­ The fi ne wares encountered during the survey con­ ods, but callnot be dated independently. sist for the most part of a paie Ol' orange firingdepu­ rated clay with a glossy surface decoration of blacle Categories l b, 3, 5b: Depurated wares slip, either (archaic/classical) banded ware or (Hel­ Depurated tiles and wheel-turned pottery wares are lenistic) black gloss ware. A majority of the sherds made out of levigated clays, to which temper may had, however, lost most or all of its surface finish; be added to achieve specific properties. These wares if these could not be identified as a fine ware by form range in colour fr om orange to paie buff; different (e.g. 'kylix') they were classifiedas category 5b 'in­ wares and forms may be distinguished on the basis determinate depurated ware'. Other fi ne wares oc­ of fO lln and thickness. Roof tiles, which usually have curring in very small numbers are terra sigillata and added temper, tend to be preserved in relatively large an unidentified thin painted ware. Only 10 sherds fr agments and can be distinguished by overall shape with a 'combed' decoratioll were found during the and rim profile.12 The remaining depurated material su'rvey. The material, which is hard and has either mostly consists of orange or paie firing amphora an orange or a paie colour, suggest a late (possibly fr agments, and a small amount of 'fine' wares which post-antique) date but no parallels have been found can easily be recognised by their surface treatment, yet. It is ex:pected that fu rther study of categories 4 fa bric, form, thickness, and decoration. All of these and 6 will result in a finer typo-chronological divi­ also occur in severely fr agmented and/or abraded sion of the material. 408 M. VAN LEUSEN & P. ATTEMA

CategOl)l 9: Lithics most fieids were full of bits of ploughed-up conglom­ No lithics were previously known from the study erate rock of a colour and shape often mimicking area, and indeed the SIBA'OO survey collected only potsherds. It is noteworthy that this material tended a very few, most ofwhich were judged to be probably to be found only when surveyors were sitting down of natural origin in view of their inegular negatives Ol' were looking especiaIly for it; we ean therefore and retouches.13 Only three are clearly intentional not be certain that the map shown in figure 6a is an flakes, one of which is a large white patinated flake accurate reflection of the distribution of this mate­ whose distal end was intentionally retouched into a rial. scraper. Although no firm date ean be given, the For both the coarse (fig. 6b) and the Hellenistic­ Levallois-like technique employed suggests an attri­ Roman material (fig. 6c), the corrected finds distri­ bution to the Middle Palaeolithic (however, the piece butions show a quite intensive use of all terrace edges is atypical). The other two flakes cannot be dated. (including mi nor 'internal' ones), except perhaps those that are very small or inaccessible. The two distribution patterns themselves are very similar, 7. RESULTS any differences being quantitative rather than quali­ tative: no concentrations of coarse wares occur away 7.1. The intensive survey from concentrations of Hellenistic-Roman material. The intensive survey teams covered some 125 hect­ It therefore remains to be determined whether the ares (552 collection units) spread over the four ma­ COal'se wares should be interpreted as evidence for rine-fluvial 'terraces' within the survey transect in a an early (6th-4th centUl'y BC) settlement expansion total of 121 personJdays (inclusive of adminish'ation, fo llowed by a Hellenistic phase of continuity, Ol' as GPS measurements, etc), for an average of l hect­ just another category of Hellenistic wares. In the ares per person/day. This is somewhat slower than latter case, littie direct evidence for Archaic-Classi­ the speeds achieved during similar surveys elsewhere cal rural settlement remains, and the first substantial (ef. Attema et al., 2002, in press) and was mainly due rural expansion in the foothills could have occurred to the extra time needed to cope with mapping prob­ as late as the late 4th-3rd century BC. Among the lems. Table 3 and figure 5 give an overview of the fine wares, one small group of 'banded ware' kylix fi eIds covered by the intensive survey. fragments has already been tentatively assigned to the In order to produce distribution maps by period, Archaic period on the basis of comparable material the raw counts per material category per collection present in the museum at , but this needs unit were corrected fo r unit area, percent coverage, further study. It may be noted further that the Hellenistic to and visibility, using the formula D = N * (100 1 C) Roman material in the survey transect consistently * (100 1 V) 1 A where D = corrected finds density, N points to dates in the range 325 BC-AD 100, and = raw finds count, C = estimated percent coverage, only rarely into the later Imperial period. This con­ V = estimated percent visibility, and A = unit area in hectares. It is recognised that this procedure, while tradicts received wisdom regarding the historie settle­ producing normalised finds densities expressed in ment dynamics in Calabria, referred to in Section 9. numbers per hectare, tends to exaggerate small den­ sity variations where low numbers of finds are in­ 7.2. The extensive survey volved; appropriate care must therefore be taken when interpreting the distribution maps reproduced The extensive survey team covered some 50 agricul­ here in figure 6. tural fieIds (3 15 hectares) in an estimated 38 personl The dish'ibution of protohistoric material shown days, for an average of 8.3 hectares per person/day. in figure 6a is characterised by small amounts of Table 4 lists and describes the sites located by this material in all surveyed fields, generally associated team and the sites mapped by Quilici in the same spatiaIly with the major terrace edges. The material area. tends to consist of undiagnostic impasto pottery, The extensive survey recorded a small concen­ except on the higher terraces neal' the Monte San tration of ill1pasto materiaI on site 3042 (extending Nicola where denser concentrations of more identi­ into field 3043), and another one of coarse impasto­ fiable burnished Bronze Age and Iron Age wares like material in field 3056. The large majority of the occur. Here the presence of a large protohistoric site material and sites were dated to the Hellenistic pe­ (Peroni & Trucco, 1994: pp. 819-820, site 31) was riod. Sites 3002/3003 appeared to consist of recently confirmed. ploughed-up high-quality material, possibly belong­ ' Coarse (unburnished) protohistoric impasto fab­ ing with Quilici site 126 (located some 100 m up­ rics are very difficult to find in arable fieids under slope but of which almost no remains could be the conditions prevalent in the Sibaritide fo othills. found); confirmed 4th-2nd c. BC sites include 3008, Their recovery was made extremely difficultbecause 3016, 3026, 3041, and 3055; generic Hellenistic ma- Regional archaeological pattems in the Sibaritide 409

Table 3. Preliminaty results of the RPC intensive fjeld survey campaign 2000, by fj eld.

Units Toponymic Notes

1001- 1031 Azienda La Silva We st An old olive grove with stands of oak, cut along the slope by a wide concrete-lined V­ shaped ditch and a U-shaped aqueduct (both probably dating to the 1930s). Quilici site 62 ('scatter ofHellenistic-Roman sherds, including BG') was fou nd again but appears 'smeared out' across a large part ofthe fieid; hvo out ofseveral finds concentrations are sufficiently distinct for individuation (scatter 20 in units 1020-23 and 1030-3 1, scatter 21 in units 100 1-04 and 1009- 1 I). Quilici site 63 ('scatter of Hellenistic-Roman material, including grey ware'), supposedly present at the westernedge ofthis fieid, could not be fo und despite specific searching. 1032- 1058 Azienda La Silva North Ploughed fieid. Part ofQuilici site 62 was found nem'est the Azienda buildings (scatter 20, unit 1045); the rest of field contains one other major (scatter 8, units 1048-49) and two minor finds concentrations (scatter II, unit 1052; scatter 12, unit 1050). Concentration of recent material in unit 1037. 1059- 1087 Ploughed fieid; one small site was located (scatter I, unit 1060). 1088- 1117 Aloisi Olive grove Yo ung olive grove, just harrowed; used concrete posts to set out grid; units 1108-09, 1088-89, 1094-97, 1 106--07 and I I 13-14 were surveyed by team 2 before harrowing; no sites present. 1118-1142 Sferra-cavallo Irregular fieid; found one small sherd concentration with impasto (scatler IO, unit 1133). 1143- 1208 Large ploughed field just east of Lauropoli; steep slopes into valley on hvo sides; contains fo ur sites (scatter 9, unit 1 168; scatter 7, units 1 186-87; scatter 5, unit 1202; scatter 6, unit 1207). 200 1- 2036 Aloisi North 200 1 -2022 irregular ploughed field with evidence of deep ploughing; mapped two concentrations of thin wares (scatter 15, unit 2001; scatter 16, units 2005-06), the latter contimting across a recent path into units 2025-7, and further disturbed by works relating to the driveway and garden ofthe house belonging to Dr Aloisi. 2023-2036 were surveyed in a young olive grove before harrowing. 2037- 2055 Irregular ploughed fieid; units 2038-39 contains a finds concentration (scatter 19) which might be a kiln but also con tai ns recent material; 2049 contains a sub-recent tile kiln. 2055- 2064 Irregular ploughed fieid; no remarks. 2065- 2089 Rubbish dump Elongated ploughed field with unploughed sections; contains a very diffu se scatter of coarse and thin wares in the topmost units (scatter 17, units 2079-8 1 and 2084-86). 2890- 2105 Ploughed field along Raganello valley; no remarks. 2106- 2125 Irregular ploughed field along Raganello valley with adjoining units along lower terrace slopes; fo und Quilici site 135 ('scatter of Hellenistic-Roman material, inc\uding some large sherds'; is scatter 18, units 2106-08 and 21 12-15), some impasto sherds, and off­ site material provenient from the next higher terrace. 2126- 2148 Group of irregular ploughed fieids along lower edge of terrace; three independent grids were used (for units 2126-35, 2136-37, and 2138-48); finds were concentrated along the north-eastern edge fa cing the Raganello valley (units 2126--27, 2126, 2128, 2136), but no distinct concentrations could be defined. 2149- 2161 Irregular ploughed fieid; no remarks. 2162- 2177 Ploughed fieId, sloping steeply toward the Raganello valley, plus units adjoining path along top of the slope; fo und only a sub-recent finds concentration approximately where Quilici site 134 ('diffu se scatter of Hellenistic-Roman material') should have been (unit 2176). 2178- 2184 Some small ploughed fieids on slope facing the Raganello valley, with unploughed section in between; no remarks. 2185- 2234 Rebecco (Masseria Frascino) lrregttlar ploughed fieids with steep terrace and gully slopes on their eastern side; no remarks. 2236- 2268 Four ploughed fieids on steep slope fa cing the Raganello valley; located a possibly Bronze Age / lron Age 'grumi' site on fa irly level bottom section ofslope (scatter 13, units 2263- 64 and 2268).

2270- 2275 Part ofa ploughed fieid, surveyed rather too quickly because offailing light (= low quality). No finds concentrations. 2276- 2289 Two irregular ploughed fieids on terrace along Raganello valley; no remarks. 2290- 2330 Three irregular ploughed fieids on terrace along Raganello valley; steep slopes on southern edges; located finds concentrations in hvo fieIds (scatter 4, unit 2306; scatter 14, units 23 12-13 and 23 16--17) . . 2331- 2344 Monte S Nicola Tlu'ee small irregular ploughed fieids just below the utUlamed hilltop on which stands the radio tower; containing one previously unknown Hellenistic site (scatter 2, unit 2344). Found part ofPeroni Bronze Age / lron Age site 31 on the saddle (scatter 3, units 2332 and 2334) and on the south-east slope of Monte S Nicola (grab sample). 410 M. VAN LEUSEN & P. ATTEMA

60 ,

6b

Fig. 6. Correeted finds density distributions resulting from the intensive survey. Grid spaeing: l km. (a) Protohistorie period (cats 10, 13); (b) 'Archaic' mate ria l (cats la, 2a-e); (c, ovedeat) Hellenistic to Roman material (cats Ib, 3, 4, 5a, b). Regional archaeological paffems in fhe Sibarifide 411

o 50 100 200

6c

Fig. 6. COlltillued.

terial occurs at sites 3025, 3029, 3039, 3047 and south-west of Cassano allo Ionio, and a few widely 3048; early Roman (2nd-l st c. BC) at sites 3032, scattered larger farms (fallorie padronali) which may 3039; and Roman Imperial wares (lst-3rd c. AD) are be recognised by architectural remains. The territory found at sites 3016 and 3018. as a whole was therefore littered with settlements (villas and villages) in the Roman period, c1usters of which 'define a fi nely branc11ed network of routes " 8. SETTLEMENT AND INFRASTRUCTURE until at least the second century AD. However, the material itself was described by Quilici as 'quite Whilst the results of the SIBA'OO survey should first unpretentious, even pOOl', indicating local production and fo remost be compared to those of the Quilici and use' - recalling Strabo's (VI, 253) judgement survey which they were intended to test (De Rossi of Sybaris in his time as 'barbaric'. et al., 1969: pp. 147-155), the general lack of rural At the regional scale, Quilici contrasted the pat­ settlement data pertaining to the coastal and alluvial tern of autarchic 'villages' to the north of the Crati­ plains of the Sibaritide means that we must also view Coscile with that of single fa rms to the south and them in the less specific geographical and historicai west, and wondered if this might not also express a context ofRoman Calabria, as provided by Accardo's diffe rence of social, even ethnic, units rather than recent study (2000). It bears repetition that most of being purely a consequence of the differing geomor­ the surface material fo und north of the Crati-Coscile phologic structures of these areas. He suggested (De was ascribed by Quilici either to the late Hellenistic Rossi et al., 1969: p. 149) that the difference may (tate Republican) period or to the Roman Imperial be related to the presence of 'colonial' landscapes period. Even sites and scatters not expressly dated of single villas emanating from the Roman colonies by him were to be ascribed to these three centuries of Interamnium (Eianina and the upper Coscile) and (150 BC-AD 150). Most of these appeared to repre­ Copia Thurii (the hills south of the Coscile), while sent small fa rms continuing into the Imperial period, the area of larger villages in the fo othills north of and tending to c1uster into 'villages ' located on the the Coscile represents a survival from the pre-Roman ' lowest parts of the foothills; there were also a Hellenistic society. If true, the Hellenistic settlement fe w small villages lying along the water-rich valleys pattern recovered by the SIBA'OO survey SilOUld 412 M. VAN LEUSEN & P. ATTEMA

Table 4. Site list for the RPC field survey campaign 2000 extensive survey. BG = black glazed ware; TS = term sigillata.

RPC ID Quilici ID Max N/m' Area (m) Notes

3002 126? 50 These two concentrations possibly constihIte a new site, but are more probably related to 3003 30 Quilici 126 (a 'Hellenistic-Roman scatter containing tile frs '). They contain both 'archaic' coarse ware and BG ring bases, pit/I Os sherds, and tiles, including one hal f of a terracotta mould fo r votive fe et. Site 'halo' extends into fieids 3004/3010, including Hellenistic imbrex, but mostly consisting of tiles with many quartz inclusions. Nearby in this same field is a large dump of stones, preslllllably ploughed up locally. 3008 127 IO IOxl5 Site contains 'archaic' com'se tile and pottery, Hellenistic and Republican roof tile, and 4nd-3rd c. BC pithos. Separate from this in field 3007, but concentrated along tile modem road bed, is some off-site material: rim and handle ofa late 4th c. BC sÅJ'phos, and a 100m weight. Logged by Quilici as a 'diffuse scatter of Hellenistic-Roman material'. 3016 125 35 50x30 Scatter of com'se 'archaic' ware and Hellenistic material located on terrace edge overloDking the Franceschiello valley, partly obscured by modem path and rubbish dump: BG pottery, pithos, fo ot of 3rd c. BC ka I/tharas, early 2nd c. plate. Site 'halo', extending over 100 m into field 3015, contains some concentrations of cobbles: Hellenistic-Roman material, material of late 4th c. BC BG cup and Corinthian B amphora, 2nd-3rd c. AD 'African' TS, and a com'se ware 100m weight. Logged by Quilici as a 'large dense Hellenistic-Roman scatter including BG'. 3018 128 35 50x75 Large scatter containing tile, storage pots, coarse and fine wares. Located Oll an elevated portion of the termin cut by a modem field boundary. It is ve ry homogeneous in fine wares and contains a/o several I rd c. AD Roman TS frs (CFTS fo rm 21.3), some with plastic decoration, and a column fragment. Diffu se material occurs off- site in fieids 30 17 and 30 19; dump ofstones, tile and dolium frs in easterntip of3019. Logged by Quilici as a 'Hellenistic-Roman scatter and villa with floor in apl/s spica/lllll (information from farmers)' . 3025 New 5 15xl5 Scatter of Hellenistic material, do/il/III, BG, 100m weight, all material of similar fa bric; also a number of river cobbles present. 3026 131? 75xl25 Large area of diffilse finds in northem corneroff ield; mostly amphora, but also containing 4th-3rd c. BC Corinthian material, and one 100m weight. Note: not intensively walked (perhaps off-site). Identification with Quilici 131 (a 'diffuse Hellenistic-Roman scatter containing an amphora handle'), which could not be fo und in the neighbouring field 3027, rests on the assumption that the latter may have been inaccurately mapped; the core of site 3026 is probably located in the neighbouring field to the north (not surveyed). 3029 New 2 2x2 Small scatter of Hellenistic tile, com'se and depurated wares. 3032 New 25 30x20 Vitrified kiln material, wasters, a 2nd c BC amphora spike and neck of a jar in pasta grigia, and 2nd-1st c. BC Roman rooftiles. 3039 New 5-10 IOx20 Scatter located on edge of terrace overlooking the Franceschiello valley. Material of 'archaic' coarse ware, Hellenistic cooking pots, Roman roof tiles, possibly TS. 3040 New 10xl0 Site contains a great many river cobbles; coarse 'archaic' tiles and thick-walled pottery present but not diagnostic, hence no clear dating evidence. May be related to site 304 1 nearby. 304 1 New 40x40 Site is strategicaIly placed, with a good view of the terrace. Contains Hellenistic tile, BG ceramics, Corinthian A/B, dolil/III, and recent fa brics. Note: visibility low because of bushes. 3042 New 20x l0 Very diffu se scatter of shapeless illlpasto sherds on terrace edge overlooking the Franceschiello valley; this material occurs lower down the valley slope in field 3043 as well. 3047 new 5-10 IOx20 Scatter of possibly Hellenistic material, contains no clear diagnostics. Depurated orange firing roof tiles with some inclusions; various medium and thin coarse wares; a few depurated orange firing sherds with some inclusions. 3048 130 90x30 Sizeable scat!er of Hellenistic material fo und on both sides of the modem road bed: coarse 'archaic' roof tiles and medium thick wares, I x BG, one coarse 100m weight. Note: very low visibility because of olive trees. Logged by Quilici as a 'diffu se Hellenistic­ Roman scatter'.

3055 2-5 40x20 Scatter of Hellenistic material collected rif ter intensive survey ( = scatter 6): coarse tiles, medium thick and thin wares. Site contains relatively many luxury items: late 4th c. BC Corinthian NB amphora frs, BG including fo ot of cup, and other depurated paie and orange firing sherds of table wares. 3056 10xlO Low-density scatter of coarse material, probably including Late Iran Age illlpasto, collected

rift er intensive survey ( = probably scatter 5). Com'se tiles, medium and thin wares, including one rim with comb motives on the lip. Regional al'chaeological patterns in the Sibal'itide 413 consist of sporadic villages along major thorough­ extensive survey in the future, and obtaining access to fa res. The location of roads connecting the coast to the original research. Although routes may well have inland regions would have been dictated by the major existed where he posits them, sites also occur well away val1eys and passes in the uplallds and mountains (De from these along all 'millor' terrace edges. With respect Rossi et al., 1969: p. 151); a system of fOltiets and to dating, the materials collected in the SIBA'OO survey watchtowers safeguarded Hellenistic Thurioi and may also suggest consistently em'lier dates (Archaic to early have done the same,for Archaic Sybaris, but so far Roman) than those suggested by Qullici (late no evidence fo r this has been fo und. Hellenistic to early Imperial); and these results may Although actual findspredating the late Hellenis­ tentatively be extrapolated across all ofthe Sibaritide. tic period were rare at the time of his survey, Quilici also attempted to reconstruct protohistoric patterns of settlement and, especiaIly, infrash'ucture (viabilita) 9.1. Protohistoric to Archaic in the Sibaritide (De Rossi et al., 1969: pp. 97, 152- 155). Several of his routes either cross or come near The adverse cOllditions for retrieval of protohistoric to the SIBA'OO survey area. Using an 18th century ceramics on the intensively worked terraces must be parallel, he postulates the existence in the early Iron taken into account when interpreting the archaeologi­ Age of a 'coastal' route fo lIowing the fo othills, con­ cal record of relatively well preserved upland proto­ necting the large protohistoric centres, and avoiding historic sites. There is currently no evidence that the (marshy) plain.14 A second route is postulated protohistoric (B A-EIA) settlement occurred any­ parallel to this along the lower lip of the lowest 'ter­ where in the survey transect but at the very highest race'; a third route fo llows the 'radial' morphology elevations (the Monte San Nicola at 500 m asl stands of the terrain and the line of a known Roman road some 150 m above the highest terrace). This can be from to and fu rther on to interpreted as confilmation of existing thought about Civita and, presumably, Thurioi/Copia in the plain; protohistoric settlement patterns, but it leaves unex­ a fourth route connecting Torre Mordillo, Cassano, plained why the most important hilltop settlements and Civita forms part of a postulated protohistoric should be located on hills overlooking the Sibaritide long-distance 'mountain route'. Quilici derived this plain. Could the needs ofpastoralism - winter graz­ protohistoric infrastructure from the locations of ing in the pia in, which at this time would have been known sites and the terrain morphology, interpret­ a heavily wooded and seasonally flooded marginal ing rivers as ban'iers, ridges as routes, and valley area with soils too stony or c1ayey fo r palaeotechnic flOOl'S as difficultterrain because of water and veg­ agricultural use - have been so vital as to be the etation. The peaks and gorges are orientation points, determining element in settlement type and location sulphurous springs meeting places. The fo rmation of choice? Altematively, the postulated 'ring' of proto­ routes under 'free' conditions, he maintains, would historic centralised settlements around the plain may, have been similar to that of the Middle Ages which in pmt, prove to be chimerical, the result of biases is preserved today in ancient mule paths. Some of in academic research and interpretation. One indica­ the latter (between Francavilla Marittima, San Lo­ tion that this may have been the case is the apparent renzo Bellizi, and ) were ex­ concenh'ation of protohistoric research on the hilltops plored during the SIBA'OO campaign and were fo und of the fa scia co/linara (see note 5). Vanzetti recently to have protohistoric sites along them, but it is not again stressed the tendency of the Roman school of yet proven that such sites c1uster on these routes. protohistoric studies to assume the presence of a Figure 2 shows the two main protohistoric routes settlement where the evidence permits rather than postulated by Quilici. suggests this (Vanzetti, 2000: pp. 7, 23; 2002). For example, Broglio di Trebisacce is seen as one single large settlement rather than as the two nuclei sepa­ 9. CONCLUSIONS rated by 250 m fo r which archaeological evidence exists. Equally, settlements are assumed to be as large The results of the SIBA'OO survey can successfullybe as the geomorphologic units (plateaux) they occupy, used to re-interpret the geographical and chJonological and 'missing' periods are assumed to be present even patterns of settlement suggested in the late 1960s by in the absence of evidence. Taken together, these Lorenzo Quilici. Both the intensive and the extensive assumptions may mask more complex realities in the surveys have recovered about twice the munber of sites protohistoric evidence. We fe el the possibie continu­ mapped by Quilici in the same area, despite the at.ion of the prehistoric settlement pattern into the deterioration of the soil archive in the intervening hinterland must be studied before further regional decades. The resultant site distribution suggests, but interpretations can be made. do es not prove, that Quilici's 'villages' are artefacts of For the Later Iron Age and Archaic periods, the research and visibility biases. A definite answer to this evidence from the survey is equivocal. Despite the question will require targeting a wider area for historicai evidence for the establishment of Sybaris 414 M. VAN LEUSEN & P. ATTEMA around 720 BC and the archaeological evidence for southern Italy were aiready under pressure from the continued use of the sanctuary and necropolis at equally expansive indigenous Bruttii by the early 3rd neighbouring Timpone Motta into the late 6th cen­ centUlY BC, and the Romans became involved when tury BC (Attema et al., 2000), no securely datable they were called in against a coalition fmmed by the materials from thi s period have been fo und in our Bruttii, the colony of Taras, and Pyn:hus of Epirus. survey. Given our experience with the very low vis­ We are told that, when this coalition was finally ibi litY of com'se impasto wares among the naturally defeated in 272 BC, much land in Calabria was al­ occurring stones in the survey area, we must conclude ready confiscated from the Bruttii; resistance flared that neither our own surveyors nor the Quilici team up again during the second Punic war and by the end were ab le to identify such material with any degree ofthe 3rd century the remaining Bruttii are enslaved. ofreliability; we cannot therefore infer anything from Both these wars and the advent of malarial disease its absence. Further fieldwork will be needed to pro­ caused a severe decline of the Greek city-states dur­ duce a more reliable distribution map of this mate­ ing that same period. ria!. For the Archaic, much will depend on a closer Following a period of military occupation and dating of the coarse wares, which make up more than administration, Roman colonies were established on half of the finds by weight, through association with ager publicus (and other towns revived Ol' refounded) datable fine wares Ol' through typological compari­ in Calabria in the COUl'se of the 2nd centuty BC - son with excavated material within the region. Copia Thurii being one of the first. From this time until the 2nd century AD, medium to large-sized agricultural villas spread in Calabria mainly in the 9.2. Hellenistic to Roman Imperial territories of these c% niae. The Via Popilia, extend­ No such visibility problem occurred with the classi­ ing the Via Appia from Capua to Rhegium and prob­ cai Hellenistic and Roman ceramic types. It appears ab ly built in 132-131 BC, ran through the Sibaritide from the results of our field work that the large-scale from Muranum in the n011h to Consentia in the south, spatial patterns mapped by Quilici in the 1960s are, probably fo lIowing the valleys of the middle Coscile at least fo r the transitional zone between the pia in and Crati rivers16 and with secondary roads to Copia. and the hinterland, partially correct. Large and small These roads and rivers must have been extremely sites of the Hellenistic-Roman period do occur in important in enabling the marketing of agricultural elongated clusters along 'maj or' terrace edges. How­ produce, and archaeological indicators of colonisa­ ever, such sites also appeal' to cluster locally along tion and Romanisation are therefore likely to cluster the edges of small valleys cut into the terraces; this on them.17 In the centuries AD, the last traces of the is at least the case along the eas tem edge of the instihltions of Magna Graecia were supplanted by Va /lone Organafa Idi Franceschie//o which runs just those of the Roman state, and areas of extensive to the east of Lauropoli. The possibility that Quilici 's (Iafijundia) and intensive (villa) cultivation devel­ site clusters are caused by, rather than merely co 1'­ oped. related to, modem accessibility and land use must However, besides the land designated fo r the co­ remain a hypothesis at this stage; it may eventually lonies, large tracts of land were also sold to rich be confirmed Ol' rejected when we obtain access to Romans who developed slave estates from about 100 his survey archive. Be. High land prices and the capital outlay needed Among the 'new' small classical sites identified fo r market-oriented production of wine and olive oil by the intensive survey, several are located within induced many small fa rmers to sell their land. A the clusters initially identified by Quilici while oth­ slave-based capitalist villa economy developed in ers are scattered all over. It was observed both by Calabria, as elsewhere, into the early Empire. From the QlIilici team and in the SIBA'OO survey15, that the beginning of the 4th century AD Calabria suf­ many of these Hellenistic-Roman ceramic scatters fe red from the general crisis of the Roman Empire, are relatively small and pOOl', with a 'standard' as­ and some 40% ofvillas were abandoned Ol' subsumed semblage consisting of some coarse and depurated into /afijillldiae fo unded on a combination of cereal storage and kitchen ware, a fe w bits of fineware in­ fa rming and pastoralism. By the end of that cenhn'y cluding Black Glaze, and one small pyramidal 100m these had developed into independent and fortified weight. To confirm this impression, a comparative power bases fo r the pofenfiores, fo reshadowing the study of the assemblages taken from several such fe udal esta tes of the middle ages. However, literm'y sites both in the Sibaritide and in the Brindisino is references indicate that the region in general contin­ being planned. ued to be productive into the 7th centuty AD. It now remains to put these results into the general Only one clearly Roman villa, possibly of the context of Roman Calabria (after Accardo, 2000). platform type, was (re-)Iocated (from Quilici site Despite the settlement expansion apparently ongoing 123) during our survey, confirming the results of the from the la te 4th century BC, the Greek colonies of 1995 survey on the terraces between Francavilla Regiollal a/'chaeological patterns in the Siba/'itide 415 and Broglio di Trebisacce and the pattern of widely half of229 I NO and 230 IV NO. This area is bounded by the spaced villas with no off-site distribution as observed fo llo\Ving co-ordinates - North: 44 1 0000; South: 4386000; in the survey by Lorenzo Quilici. While it is possibie East: 2650000; We st: 2628000. that a denser pattern ofvillas was located in the pia in, 5. Although this is not clearly stated in the published accounts, its depositional history currently precludes us from it appears thaI these surveys \Vere largely limited to thefascia cal/il/ara (foothill zone), and \Ve re intended to locate finding proofof this; a study of Medieval and sub­ protohistoric settlements similar to those at Amendolara, recent histori.cal documents may be helpful in back­ Francavilla Marittima, and Torre Mordillo. tracing Roman settlement patterns. The same goes fo r 6. The terminology used in this paper is the 'early' one advocated the Byzantine and medieval periods, fo r which no by Peroni, in which protohistory includes the MBA to lA physical evidence was recovered at all during the periods, and history begins \Vith the early Greek colonisations survey although several chapels and a monastery are of the 7th and 6th centuries BC. A more conservative known to have existed in the region. 18 Fuhll'e research chronology \Vouldhave protohistory begin with the late Iron should be targeted at those local geomorphologic Age (8th century BC), and the historic period \Vith the late units in the pia in for which there is evidence or at Classical and Hellenistic historians ofthe 5th and 4th centuries least a possibility that antique and later remains might BC. be observable. 7. Georeferencing is the procedure whereby digital map data are tied to an area of the Earth's surface by specitying the map projection and co-ordinate system. Here, the Gauss-Boaga 10. ACKNOWLEDGEMENTS projection and the Ttalian national co-ordinate grid according to European Dalum 1940 are being used. 8. The total X2 of 71 at 14 degrees of freedom is significant at Aspects of the research repOlied here have previously the .00 I level. been the subject of shldent projects at the GIA. We 9. Spatial autocorrelation refers to the fa ct that observations made would especially like to thank Kathelijne Kruidhof in close proximity to each other tend to have similar fo r her initial analysis of the Quilici data set, Erasmus geographical attributes; thus land use observations made at exchange student Lia Pala fo r helping prepare the lwo archaeological sites separated by only 100 m have a much Quilici maps fo r digitisation, Wielce de Neef fo r her higher chance of hlrning out to be identical than if the sites excellent MA thesis on the Iron Age to Archaic land \Ve re separated by a distance of 1000 m. use oftwo neighbouring Greelc colonies, and Michiel IO. Quilici identified the modem town of Castrovillari, on the Roolce, Jasper Huis in 't Veld, and Tymon de Haas upper reaclles of the Coscile river, with lnteramnium; but the fo r their preliminary analyses of the survey data distances, of 13 and 8 miles respectively, provided by the Antonine Itineraries and the Ta bula Peutingeriana for the collected since 1995. Further thanlcs must go to the stretches of the Via Popilia between Muranum (modem mayor and council of Francavilla Marittima fo r fur­ Murano), Interamnium, and Ta urasia (modernTa rsia) make it nishing us with excellent accommodations at the unlikely that this identificalion is correct. They instead indicate Timpone Motta site museum. Finally, we would like a location in the middle Coscile valley. to thank the Soprintendente, Silvana Luppino, fo r the I I. Subsequent closer inspection and cataloguing ofthe diagnostic interest she has shown in om work and fo r her guid­ material by scatler during the campaign ofSeptember/October ance in obtaining the necessary permissions. 200 I has confirmed the existence of sites \Vith an origin in lhe (Iater) archaic period, supporting an early dale fo r the associated com'se \Vares. Il. NOTES 12. These rim profiles \ViII be the subject of a closer study in the near fu ture. I. The campaign \Vas conducted as parI of a collaboralion pro­ 13. The nearest Neolilhic sile is at Favella delia Corte, so uth of ject \Vith the archaeological superintendency of Calabria. We lhe Crati river. The lithics collectecl by lhe SIBA2000 survey \Vould like to express our gratitude to the Soprintendente, \Vere stucliecl in April 200 I by Marcel Nieklls of ARC (Cen­ Dr.ssa Silvana Luppino, fo r arranging the necessary pennis­ trum voor Archeologische Research & Consultancy). sions. Some aspects ofthe SIBA2000 field campaign, such as 14. The use ofhistoric parallels in the reconstruction ofprehistoric the highland reconnaissance survey and the experiments \Vith routes isjustifieclon the assumption these routes are the 'most context-aware portable digital field assistants, are nol discussed natural ' and by the supposecl conservatism of their use; here (see Ryan & Van Leusen, 2002). compare Smal I et al., 1998: p. 338 for a similar argument 2. The RPC project (1997-2002) is part ofthe research program concerning transhumance routes in Apulia. 'Landscape and settlement' ofthe Netherlands Foundation for 15. Quilici remarks that "black gloss is encountered only rarely Scientific Research NWO, doss. nr. 360-61 .000. and general ly the assemblages are pOOl' with no evidence fo r 3. A recent manual auguring program in the plain, to depths of substantial constructions; there are no cisterns and graves are up to 7.5 m, has obtained absolute (radiocarbon) dates of up very unassuming as well" [my translation]. to 2100 BP fo r peaty layers, providingpast qllelll dates for the 16. Accardo (2000: pp. 30-39). Thought to have been navigable, more recent sedimentation phases (pers. comm. Jan Delvigne). the former fo r a distance of 18 km until the confluence \Vith 4. The study area con'esponds to the fo llo\Ving (partial) sheets the Esaro; the latter for 40 km, at least halfway to Consentia. of the IGM 1:25,000 map series, 1958 edition: 221 li NE & 17. Accordil1g to Accardo (2000: p. 56), villas in the territorium SE, 222 III NO & SO, the eastern half of221 Ir NO & SO, the of Thurii lie mostly in these valleys, are medium-sized and north-eastem quadrant of229 1 NE, and just over the northern specialise in the production of wine and oil. Almost all date 416 M. VAN LEUSEN & P. ATTEMA

from the 2nd centtll'y BC to the 2nd centtll'y AD; some conti­ Mediterralle Areheologie 17, pp. 47-52. nue into the 4th and 5th centuries AD. HAAS, T. DE, 2001. Over schervell, samples ell sites. Beoordelillg 18. One Byzantine chapel was located on top ofthe Timpone delia ell alla�)'se vall vOlldstpatrollell vall diverse sllrveys ill de Motta at Francavilla Marittima. Sibaritide. Unpublished report, University of Groningen. LERlCI, C.M., 1960. J II110vi metodi di prospeziolle areheologica alia seoperta delle civilta sepolte. Lerici, Milano. 12. REFERENCES LEUSEN, P. M. VAN, 2002. Pattems to process: Me thodologieal illvestigatiolls il/to the fo rmatiolI alld illte/pretatioll of spatiet! ACCARDO, S., 2000. Vi l/ae !'OmalIae lIel/ 'Ager Bmttil/s. 1/ pattems ill arehaeological lalldseapes. Ph.D. Thesis, Univer­

paesaggio rl/rale Calabrese dl/rallte il domillio ROllwllo (= sity of Groningen. Sttldia Archeologica, 107). Bretschneider, Roma. MAASKANT-KLEIBRINK, 2000. Early cults in the Athenaion at ATTEMA, P.,G.-J. BURGERS, M. KLEIBRINK & D. YNTEMA, Francavilla Marirtil11a as evidence fo r a pre-colonial circula­ 1998. Case sttldies in indigenous developments in early ltalian tion of 1I0stoi stories. In: V Gassner, M. Kerschner, V Muss & centralisation and urbanisation: a Dutch perspective. Ellropeall G. Wlach (eds), Die igiiis I/lld das \Vestliehe Mittelmee/; JOllmal of Arehaeology I, pp. 326-381. Beziellllllgen IIl1d We chsel\Virkllngell 8. bis 5. Jh t. V. C/II: AktelI ATTEMA, P. , G.-J. BURGERS & M.VAN LEUSEN, in press. The des Sy mposiolls Wi ell, 24. bis 27. Miirz 1999. Verlag der Osttmi field survey campaign 1999. Stlldi di Alltiehitå II. Osterreichischen Akadel11ie der Wissenschaften, Wien, pp. 165- ATTEMA, P.A.J., J. DELVIGNE, E. DROST & M. KLEIBRINK, 184. 2000. Habitation on plateau I of the hill Til11pone delia Morta NEEF, W. DE, 1998. Pa tronelI WII/ verallderillg. Lalldschapsgebmik (Francavilla Marirtima, ltaly): a preliminary report based on ell menselijke aetiviteit ill de MetapolItiIIo (Basilicata) ell de surveys, test pits and test trenches. Palaeohistoria 39/40 (1997/ V/akte vall Paestllm (Campallie) ill de ijzertijd ell de Griekse 1998), pp. 375-41 1. k% llisatieperiode (1000-400 V. ClII:). Unpublished M.A. the­ BERGONZI, G., V BUFFA, A. CARDARELLl, C. GJARDlNO, sis, University of Groningen. R. PERONI & L. VAGNETTI, 1982. Rieerc/I sIlIla protostoria PERONI, R., 1994. Le comunita enotrie delia Sibaritide ed i loro

delia sibaritide (= Cahiers du Centre Jean Berard Vll-VllI). rapporti con i navigatori egei. In:R. Peroni & E Trucco (eds), Institut Fratwais, Naples. Ellotri e Micellei IIelIa Sibaritide. Tarento, pp. 83 1-79. eGC 1969. Carta geologica delia Calabria. Foglio 22 l-II NE delia PERONI, R. & E TRUCCO (eds) 1994. Ellotri e Micenei IIelIa carta d'Italia 1:25.000 dell'IGM. Cassa per opere straordinarie Sibaritide. Tarento. di pubblico interesse nell'Italia l11eridionale. PUGLlESE CARRATELLl, G. (ed.), 1996. The \Vestem Greeks. CNR 1956a. Carta delllltilizzazione del suoli, 1:200.000, Foglio Bompiani. 19. Consiglio Nazionale delle Ricerche, direzione generale del RAlNEY, EG., 1969. The location of Archaic Greek Sybaris. AJA catasto and Touring Club ltaliano. 73, pp. 261-73. CNR 1956b. Carta delhitilizzaziolle del sIloIi, 1:200. 000, Foglio RAINEY, EG. & C.M. LERICI, 1967. The searehfor Sy baris: 1960- 20. Consiglio Nazionale delle Ricerche, direzione generale del 1965. Lerici, Roma. catasto and Touring Club Italiano. RYAN, N. & M. VAN LEUSEN, in press. Educating the digital D'ANGELO, S. & ECH. oRAzæ VALLlNO, 1994. La Sibaritide. fieldwork assistant. In: G. Burenhult & 1. Arvidsson (eds), Ar­

Lineamenti geografico-al11bientali ed insedial11ento ul11ano. In: chae% gical 1l/formaties: Plls/dllg the ellvelope (= BAR In­ R. Peroni & E Trucco (eds), Ellotri e Mieellei IIelIa Sibaritide. tern. series 106). pp. 40 1-412. Tarento, pp. 785-829. V ANZETTl, A., 2000. Costruzione e problem i dei "paesaggi di DE ROSS1, G.M., L. PALA, L. QUJLICI & S. QUILlCl-GlGLl, potere" nella Sibaritide (Calabria) dall 'eta del bronzo alia prima 1969. Carta archeologica delia piana di Sibari. Atti e Memorie eta del fe r ro. In: G. Camassa et al. (eds), Paesaggi di potere:

delia Societa Iv/aglla Greeia, N.S. 9-10, pp. 91-155. problemi e prospettive (= Quaderni di Eutopia 2). pp. 239- FEIKEN, R. & 1. WETERINGS, 1998. MOllte Selh/ro '98: eell sllr­ 283. vey vall de MOllte Sellaro. Unpublished re port, University of VANZETTI, A., 2002. Some current approaches to protohistoric Groningen. centralization and urbanization in Italy. In: PAJ. Attema et al.

HAAGSMA, BJ., 1996. Survey in de Sibaritide, CaIabrie. Een (eds), Ne\V developmellts ill Itet!ian lalldseape arehaeology ( = preliminair verslag van drie campagnes. Tij dsehri!t voor BAR Intern. series 1091).