Archaeological landscapes of the Bekaa: past results and future prospects

Archaeological Landscapes of the Bekaa Past Results and Future Prospects Paul Newson* Abstract This exploratory paper makes an assessment of our knowledge of the archaeological landscapes of the Bekaa Valley. As an exemplar the paper utilises as a framework the seminal fieldwork study of Leon Marfoe in assessing what we understand about the central Bekaa in the Graeco-Roman period. Marfoe’s model of settlement development for this region and period is comprehensive and complex, however as the paper reveals, modern techniques of landscape archaeology can provide more detailed information and a more nuanced understanding of the settlement history. The paper indicates that the Bekaa is a rich archaeological palimpsest for future research. Keywords Landscape archaeology, Bekaa Valley, Graeco-Roman, settlement development, Lebanon Introduction The Bekaa Valley is a fascinating region of modern Lebanon, at once spectacularly impressive in its visual impact: it gives an illusion of a certain separateness, a sense intensified when approaching from the coast, as one descends down the bordering mountain slopes of Mount Lebanon into the plain below (Figure 1). The valley floor is a flattish plain, 100 kilometres or so long and only eight to ten kilometres wide, sandwiched between two high and imposing mountain chains – the Lebanon and the Anti-Lebanon (Horden and Purcell 2000: 54; Rey-Coquais 1967: 21). It is this essential physical geography of the Bekaa Valley region which engenders these notions of remoteness, difference and otherness. The resultant environment provokes reactions in the visitor and has helped shape the lives of the valley inhabitants: the physical characteristics help create * Associate Professor, Department of History and Archaeology, American University of Beirut, Lebanon mental maps and a sense of place unlike those of the rest of Lebanon, which has, ultimately, an outlook to the Mediterranean Sea. This has long been noted; for example, the geographer Richard Thoumin wrote ‘one experiences the deep impression of being in a region separate from the rest of the world’ (‘on ressent l’impression très vive de se trouver en une région séparée du reste du monde.’) (1936: 273). And Ussama Makdisi commented on how Western travellers viewed the wide expanse of the Bekaa as ‘a portal into a different world’ (1998: 145). Whilst seemingly separate from the adjacent areas, the plains’ structure is open-ended, and it lies almost parallel to the coastline, which has allowed it to serve as a communications corridor between the north and the south. Also, several mountain passes over both defining ranges have permitted reasonable east-west links (Rey-Coquais 1967: 26-27). 2 Paul Newson Figure 1: Bekaa landscape: view west from the Upper Ferzol valley (the foothills of Mount Lebanon) to the plain and the Anti-Lebanon beyond (photo Hassan el-Hajj) The special position of the Bekaa as a zone of transformation has attracted scholars who have sought to establish theories that explain the region’s special nature and why it has developed in particular ways (exemplifying diverse studies which highlight the exceptional nature of the Bekaa: Abou Diwan and Doumit 2016; Chatty 1975, 2013; Sommer 2001). Most prominent of these is the work of Leon Marfoe, an archaeologist who undertook an innovative archaeological site survey in the early 1970s across the valley floor and whose PhD developed influential theories on the long-term ecology and settlement history of the region (1978, 1979). Taking Marfoe’s seminal work as a starting point, this paper will explore issues of the landscape archaeology of this region. Using the Graeco-Roman period as an example, I will discuss some of the key findings of the long-term settlement history of the region and its potential in the future. As it is a large subject, I will concentrate on the region of the central Bekaa, which can be said to extend from the Chtaura-Anjar region at the southern end up to the headwaters of the River Orontes (Nahr el-Assi) in the north (Figure 2). This has long been suspected of being the core region of a programme of Roman settlement development when it was part of the territory of Beirut (Newson 2015; Sommer 2001). Berytus LVI, 2016 Figure 2: The Bekaa region, areas and sites mentioned in the text How the Bekaa has been seen In terms of how the Bekaa has been investigated archaeologically, the initial attraction was undoubtedly the monumental temples at Baalbek. These had been noted by Arab travellers and geographers throughout the early and medieval Archaeological Landscapes of the Bekaa – Past Results and Future Prospects Islamic and into the Ottoman periods, along with information on the geography, ecology and the socio-cultural aspects of the valley (Jidejian 1975: 72-78). By the late 17th century CE, European travellers were drawn to the region specifically to record the remains of the temples in the town (Jidejian 1975: 1-10). The first serious study of these temple ruins were undertaken in 1751 CE by James Dawkins and Robert Wood (Wood 1757). Their huge volume of work on Baalbek and also Palmyra proved to be highly influential both on architectural fashions in Europe and perceptions of the orient (Makdisi 1998; Rey-Coquais 1967: 17). Other visitors followed and began to describe the customs and culture of the Bekaa more generally and also individual aspects of ancient material culture that they encountered, especially inscriptions, but also various ruins, particularly of Graeco-Roman temples and structures such as rock-cut tombs. Visitors included Volney, who emphasized in Orientalist detail the context of the temples at Baalbek, and Burckhardt, who included descriptions of the Bekaa in an account of a journey to the Lebanon from Damascus (Burckhardt 1822: 1-50; Makdisi 1998: 141-44; Volney 1787: 232-48). The central Bekaa Valley has been the subject of a variety of archaeological work over the past sixty years. This work has been very valuable in providing details of various aspects of the settlement history of the Bekaa, although a high percentage of this work has naturally focussed on understanding the important religious site of Baalbek. In this regard, one of the first fundamental studies beyond Baalbek was the documentation of many of the monumental Graeco-Roman temple sanctuaries (Krencker and Zschietzschmann 1938). Beyond this, there has been some attempt through archaeological excavations to determine the development of the many prominent tells situated on the valley floor. On this point, the continuing study of the tell settlement at Kamid el-Loz in the southern Bekaa Valley has been significant (e.g., Hachmann 1989; Heinz et al. 2010). A further series of tell excavations has been those undertaken at the central Bekaa site of Tell el-Ghassil by the American University of Beirut Museum (Baramki 1961, 1964, 1966; Doumet-Serhal 1996). Other work has concentrated on exploring the evidence found in the valley for human activity during earlier prehistoric periods prior to the development of tells, particularly that of the Neolithic (notably Copeland 1969; Copeland and Wescombe 1965, 1966; Kuschke 1978; Kuschke et al. 1976; Schroeder 1970, 1991). In terms of the periods of settlement after the tell sites were abandoned, greatest attention has been paid to two extraordinary sites of the Classical and early Islamic periods. The earlier of these two sites and one which has merited continual research has been the settlement and sanctuary site at Baalbek. Recent work has continued to document and excavate the main temple sanctuary complex of Baalbek, and to begin an exploration of the environs (van Ess 2008). An additional historic site of international importance which has also been the subject of research is that of the Umayyad settlement of Anjar (located in the eastern central Bekaa). Whilst the interior of the Umayyad settlement has been the subject of intense excavation and detailed assessments of the architecture and the site purpose, the territory and context of this settlement has been largely ignored (e.g., Chéhab M 1963; Chéhab H 1993; Finster 2007; Hillenbrand 1999). Current orthodoxy The current orthodoxy on long-term settlement history is founded upon the pioneering survey work of Leon Marfoe in the 1970s (Marfoe 1978), the first half of which, dealing with the early periods up to the Iron Age, was re-edited and republished in the 1990s (Marfoe 1995, 1998). Subsequent to this work using the theories of Clifford Geertz, he developed a complex model of socio-political organization which he termed the ‘Integrative Transformation’ (Marfoe 1979, 1982). This model built upon Marfoe’s broad brushstroke approach to the long-term pattern of settlement history, which resulted from the many site visits he made along the length of the Bekaa (Marfoe 1978). The model essentially explained changing settlement patterns in the Bekaa as a consequence of political structural development Berytus LVI, 2016 3 4 Paul Newson based on the agglomeration and dispersal of centralised political units. Important to Marfoe’s theory was the influence played by the complex pattern of localized environments, shaped by different combinations of natural elements. Influencing factors include: the mountain topography which impacts on, for example, the degree of exposure to prevailing winds and sunlight, and variations in precipitation due to rain-shadow effects; the underlying limestone geology ensures an uneven distribution of surface water leading to a reliance on springs in most areas, but specific geology in some regions has led to the creation of lakes and marshy zones (Horden and Purcell 2000: 56). Consequently, in terms of land use and settlement types, the multiple microenvironments thus encouraged a range of local adaptations. So to give some examples, in some locations, especially within the areas with strong perennial karstic springs, irrigation farming regimes were established, while in progressively drier regions of the plain, cereal dry-farming to steppe pastoralism developed, and the mountain slope regions experienced the growth of terraced farming and vertical transhumance (Marfoe 1978: 787-88). On a regional level, the Bekaa throughout many periods was envisaged by Marfoe as a distinct no man’s land between different political entities, be it Egyptians and Hittites in the Bronze Age; Seleucid and Ptolemaic kingdoms in the Hellenistic period; or Fatimids and Hamdanids in the Early Middle Ages (Horden and Purcell 2000: 87-88; Marfoe 1978: 18). So for a large part of its history, the Bekaa has acted as a narrow communications corridor between dominant political entities situated at the northern and southern ends, but was itself never fully dominated by one or the other as a result of the mountainous environs and the patchy nature of populations across the valley. Taking into account these two intersecting factors – local ecosystems and regional sociopolitical processes – Marfoe envisaged a threelevel development in land use and settlement within the Bekaa. The first lasted from the end of the Palaeolithic until the fourth millennium Berytus LVI, 2016 BCE and was denoted in the plain by a gradual specialization into early forms of agriculture and the establishment of some permanent settlements in areas with the maximum access to all potential food sources, be it hunted, collected or farmed (Marfoe 1978: 789-90). The second level (from the fourth millennium BCE to the Hellenistic period) was centred round the advent of urban forms of settlement, which reflected increasing populations as a result of specialization in agricultural techniques and efficiency and in human occupations (Marfoe 1978: 790-93). During this period, the population totals in the valley fluctuated in periods of growth and decline, a process which continued throughout the third level of settlement development, which Marfoe characterized as beginning in the early Hellenistic period and continuing until the end of the Ottoman period (1978: 797-99). This third level was exemplified by the exploitation and settlement of previously little utilised zones in the valley region, particularly those zones at higher altitudes with the emergence of terrace agriculture and vertical transhumant economies. A key period within this third phase was from the beginning of the fourth century BCE to the eighth century CE – i.e., from the Hellenistic to the early Islamic. This was an important period of both political and technical developments that introduced the basic frameworks which in many ways governed the settlement and land use parameters for the rest of the period and continued until the changes of the 20th century. Many of these innovations can be seen to stem from the point when substantial parts of the valley were incorporated and developed as a Roman colony at the very end of the first century BCE. Therefore, because of its potential part in the formation of the modern landscape, as constituted by Marfoe, the Graeco-Roman period will form the focus for the rest of the paper. In summarizing Marfoe’s survey of settlement for the period, we can identify a number of points: 1. He saw difficulty in pinpointing Hellenistic settlement with the result that he regarded it as ‘ambiguous’ (Marfoe 1978: 620). Though the evidence was unclear, Marfoe postulated Archaeological Landscapes of the Bekaa – Past Results and Future Prospects increases in population and urbanization, with an exodus from the plain and the central Bekaa region (1978: 625). 2. For the Roman period, the evidence from his survey pointed to settlement being at its greatest. From this, he suggested that Roman colonial settlement was particularly strong in the western foothills or piedmont of the Bekaa, the region in which Hosn Niha is situated (Marfoe 1978: 647). 3. For the Byzantine period, Marfoe saw a decline in settlement with a significant focus shift from small sites to big sites, which he interpreted as a result of depopulation due to over-taxation, population decline, and peasant impoverishment amongst other reasons (1978: 675-76). 4. For the subsequent early Islamic period, Marfoe saw a near absence of sites in light of political uncertainty, which was not fully overcome until the medieval period and the resultant settlement expansion of the Mamluk era (1978: 696-97, 720). As a result of his survey findings, Marfoe’s long-term settlement trajectory for the GraecoRoman period marked it as a distinct one, a pattern which goes against the grain of the results of many other Near Eastern surveys. This other survey work has tended to establish the height of settlement in the Graeco-Roman era to be during the Byzantine period: with a density in settlement (and population) not reached again until very recently. This expansion and increased density of settlement has been put forward using as clear case studies the marginal regions of the Hauran and limestone massif in modern day Syria (Decker 2009; Tate 1997). Figure 3 shows the comparisons of Marfoe’s survey with some other surveys of marginal regions of the Near East in terms of number of sites for each period. Consequently, the received picture of the Bekaa is one at odds with most other regions of the Near East over a comparable timescale (cf Decker 2007 for other surveys on the frontiers of Mesopotamia). Is this a true reflection of the settlement sequence? Figure 3: Comparison of Mafoe survey site numbers by period with the results of some other notable archaeological surveys of marginal zones of the Near East. The information is compiled from the results of the following surveys: Birecik-Carchemish – Algaze et al. 1994; Limes Arabicus - Parker 1992; Southern Ghors - MacDonald 1992; Upper Khabur – Lyonnet 1996; Wadi el-Hasa - MacDonald 1988 Berytus LVI, 2016 5 6 Paul Newson Marfoe put forward a number of hypotheses to explain this particular settlement trajectory from the Hellenistic to the early Islamic. The Bekaa region’s incorporation into the Roman Empire linked it to regional markets and saw the investment of capital. New, large-scale investment led to the development of previously neglected agricultural regions, particularly for orchard cultivation, terrace agriculture and increasing transhumance as sheep and goat made use of higher altitude natural pastures (1978: 797). By the Late Roman period the success of this expansion had resulted in a large, dense population, which led to more and more marginal land being cultivated. Simultaneously, empirewide social changes brought about higher taxes and rents, and the concentration of landholdings (Marfoe 1978: 813-14). The cumulative effect of these processes increased the numbers of impoverished rural dwellers and led eventually to the abandonment of settlements and cultivated land. Marfoe proposed that the remaining rural population, in order to maximise profits and also avoid the higher taxes, specialized in animal husbandry (1978: 815). At the same time, the development of camel pastoralism in the semi-arid steppe regions to the east of the Bekaa resulted in increased pressures on the availability of pasture within the Bekaa as sheep and goat herders could no longer utilize winter pastures in the steppe plains to the north and east of the Bekaa (Marfoe 1978: 815). The cycle of increased impoverishment of sedentary agriculturalists, subsequent farmland abandonment, and takeover by pastoralists continued until the overall population had decreased and the dominant settlement and land use patterns within the Bekaa had been completely transformed in the Byzantine period (Marfoe 1978: 816). Marfoe suggested this general pattern of expansion and contraction had been repeated in other periods: often as a result of similar processes of political and social change which, aided by the inherent ecological constraints of the Bekaa, in turn created cycles of land use changes and the accompanying settlement expansion and contraction. In formulating this rational and credible hypothesis, Marfoe collected evidence from Berytus LVI, 2016 known and visible sites through what might be called today a ground-truthing exercise. Sites were known through previous surveys and excavations, and also a number were found marked on the 1:20,000 topographic maps and available aerial photographs (Marfoe 1978: 865). In making a site visit, surface pottery was collected using a basic systematic collection method, and site size estimated. Where possible, the diagnostic pottery was then broadly dated through reference to similar stratified excavated examples (Marfoe 1978: 829-30). This methodology was very successful in the sense that a Bekaa-wide range of sites of different types and sizes was sampled giving us the most complete picture of the archaeological resource of the Bekaa as a whole (Figure 4). Conspicuous in the archaeological catalogue are the large prominent tell sites of the central plain region. This is a reflection of a number of issues beginning with the undoubted importance of such sites as the main settlement foci of the Bekaa for much of human history in the last 10,000 years. However, it is clear that given the resources and the time available, Marfoe’s survey had to be large in scale and broad in scope. This meant that he relied on published work for many of the sites prior to the Bronze Age, and areas above the 1500 m contour level were not included (1979: 866). In completing the fieldwork over such a large region, there was by necessity a heavy reliance on sites which were clearly visible and published on the 1:20,000 scale maps. Since Marfoe’s survey was completed in the 1970s, fieldwork methodologies have developed exponentially, creating the current discipline of landscape archaeology. Growing numbers of fieldwork studies of areas in the Near East have helped highlight the limitations of Marfoe’s pioneering study. There is now an acceptance of the importance of scale and, in order to obtain a holistic picture of settlement development, archaeologists should not only take full account of intra-site developments but also inter-site relationships as well. For example, as has been recognized through more recent studies, the draw of and emphasis on tell sites has created ‘Landscapes of Tells’ in many areas of the Near East ignoring Archaeological Landscapes of the Bekaa – Past Results and Future Prospects Figure 4: A map of Marfoe’s sites for the Central Bekaa region. The 1500 m contour level (red line) is the demarcation line for Marfoe’s survey, but note the general absence of sites above the 1250 m contour. The following symbols represent particular site types: circles - modern villages; squares – tell sites; triangles – other archaeological sites other, less conspicuous occupation sites inbetween (Wilkinson 2003: 123). The poor knowledge of the relationship of settlement change to material culture change has also been shown to be responsible for gaps in settlement, as demonstrated by the transformation of settlement types in the Early Bronze IV and Persian periods in Jordan (Banning 1996). Recent work Shortly after Marfoe’s fieldwork survey had been initiated, Lebanon descended into fifteen long years of civil war, and it was not really possible to undertake any archaeological work until the dawning of the current century. The majority of this new phase of work has been completed at the large important sites such as Baalbek and Kamid el-Loz. There has been some survey and site recording work attempted in the continuing difficult political environment which still persists in the Bekaa Valley (Figure 5). A crucial element of this work has been completed by the Department of Antiquities of Lebanon as part of CRM (Cultural Resource Management) and development recording. Conspicuous among the published survey work has been the preliminary results of Bonatz and others conducted at the beginning of this century to the east of Kamid el-Loz in the foothills of the Anti-Lebanon (Bonatz et al. 2002). A second survey was carried out in the immediate environs of Baalbek by FischerGenz and Ehrig, and though this was interrupted by the 2006 war, the results as with the earlier survey by Bonatz reveal a very different picture of settlement and land use for the region (FischerGenz and Ehrig 2005). Berytus LVI, 2016 7 8 Paul Newson Figure 5: Map of the Bekaa valley with location of recent survey work areas: A = Baalbek Hinterland Survey (Fischer-Genz and Ehrig 2005); B = Kamid el-Loz and Anti-Lebanon Survey (Bonatz et al. 2002); C = Central Bekaa Region – Niha and Ferzol Valleys (Newson and Young 2014, 2015). Black dots are sites recorded in Marfoe’s survey (1978) Both surveys concentrated their work on small sample areas and so were able to use more intensive methods than Marfoe for recording archaeological information. Bonatz’s team walked the majority of their survey area (c. 150 km2), with particular attention on areas away from the mountain tops and very steep slopes (2002: 287). The Fischer-Genz and Ehrig survey covered a similar area (c. 120 km2) using a combination of ‘extensive survey’ – locating sites with information from maps and local informants and in some test areas walking a series of short intensive transects (2005: 138). Perhaps the key outcome from each of these short-lived small area surveys relates to the high density of Berytus LVI, 2016 archaeological remains that can be documented within different topographical environments on the margins of the Bekaa Valley, specifically in these instances from among the piedmont regions of the western slopes of the Anti-Lebanon. Both surveys documented a range of sites with different functions, from dwelling to agricultural production, and dating to all the main periods from the prehistoric to the Islamic. As one might expect from a more intensive approach to fieldwork, the density of sites in both areas is far greater than one would suppose from Marfoe’s work. Within the Baalbek region, of the 13 substantial settlement sites recorded and represented on the Bekaa hinterland survey map, Archaeological Landscapes of the Bekaa – Past Results and Future Prospects only five were previously noted by Marfoe in his survey (Fischer-Genz and Ehrig 2005: 137; Fig. 21). The Bonatz survey has published more extensive information on the types of site they recorded in their survey. They classify a variety of different settlement sites, from temporary nomadic sites (e.g., Sites 35, 52 and 53) to longterm farmhouse and/or hamlet settlement sites (e.g., Sites 19, 30 and 45) (Bonatz et al. 2002: 292-97). In addition, there were numerous other site types, such as Roman religious sites (Sites 36 and 38) and watermills of the Islamic periods (Sites 26-28, 34), amongst others (Bonatz et al. 2002: 292-95). Each survey also recorded what they describe as tell sites: Tell Kabb el-Kroûm (Site 14) (Bonatz et al. 2002: 288; Fig. 7) and Site 33 (Fischer-Genz and Ehrig 2005: 137). These small tell sites are important for they reveal Bronze Age (and later) occupations away from the plain at strategic points controlling access along narrow valleys of the Anti-Lebanon. Whilst only preliminary summaries of the surveys have been published, both these surveys thus effectively demonstrate the intensive occupation of the land in regions which had previously been classified as somewhat marginal. Secondly, they also effectively provide evidence which seems in certain ways to contradict the survey results and subsequently some of the conclusions reached by Marfoe. For example, Fischer-Genz and Ehrig point out that their survey results indicating prehistoric settlement beyond the plain overturn Marfoe’s notion of forest cover inhibiting settlement exploitation of the slopes of the Anti-Lebanon (Fischer-Genz and Ehrig 2005: 138). Thirdly, on a wider scale, both surveys have recorded sites from all periods, some of which have previously been under-represented in Marfoe’s seminal work. This is particularly true for off-tell sites of the Bronze and Iron Ages, and for both surveys, especially the Hellenistic. In the Bonatz survey, Hellenistic sites proved to be 17 in number, second only to Islamic sites (with 24) (Bonatz et al. 2002: 300). Clearly, a more intensive investigation of a smaller area using modern methods and approaches to landscape survey can provide new insights to the settlement development and land use within the Bekaa. These surveys also help stress the point that Marfoe’s model of the Bekaa would benefit from a reassessment. New work The prospect of investigating a vital region in the Bekaa, the central western piedmont, has provided an opportunity to apply more systematic methodologies in recording archaeological remains and surface material. An important intention of the Central Bekaa Project (CBP), an ongoing fieldwork project by this author, was to document the morphology of settlements in the Graeco-Roman period and explore the relationships between settlements and their environments. This project has also provided a basis with which to explore in depth some of the issues raised by Marfoe, who listed only a limited number of sites for this region (on the whole just village sites – the list of information for his sites in this area can be seen in Table 1; Figure 6). The initial impetus to the project was an invitation from the Department of Antiquities to document the surface archaeology of the remains of the village associated with and dominated by the stillstanding remains of a Roman period monumental temple sanctuary at Hosn Niha (Newson and Young 2014, 2015). Since then, work has expanded down the Niha Valley to the environs of the village of Niha and into the parallel Ferzol Valley (Wadi el-Habis) 2 km or so to the southeast. Berytus LVI, 2016 9 10 Paul Newson Figure 6: Map of sites from Marfoe’s survey within the western piedmont region, that is, above the 1000 m contour line (see Table 1). Key sites include: 199 = Niha; 200 = Hosn Niha; 201 = Ferzol; 205 = Le Laure (Nabaa el-Habis) Hosn Niha The site at Hosn Niha was chosen for a number of reasons. The monumental scale of the large temple at Hosn Niha and its well-preserved and picturesque bearing have ensured that the site has attracted attention from visitors, both tourists and archaeologists (Figure 7). The majority of previous work there has been focussed on documenting the main religious sanctuary, which includes a later Byzantine basilica church (Aliquot 2009: 303-06; Steinsapir 2005: 7880). Long known to be associated with this large religious sanctuary are the remains of a settlement. Unfortunately, a high percentage of the settlement has been badly damaged in a long series of illegal excavations going back at least to the 19th century CE, if not earlier. It was known that some of this damage was carried out during the civil war period, so as well as recording any archaeological features, a significant aim of the project was to make an assessment of the extent of the damage (Newson and Young 2015). Berytus LVI, 2016 Prior to our project, Jean Yasmine had completed a series of site studies and an excellent examination of the structures related to the religious complex (2005, 2006, 2009, 2013). As part of this work, Yasmine also sketched a brief outline of the visible remains of the associated village and collected some pottery sherds exposed in sections or brought to the surface in the illegal excavations (2006: 1318, 47-48). Most of the pottery sherds were found to be locally sourced coarsewares, and at the time these pottery types were little known from excavated contexts elsewhere. Because of this fact, Yasmine’s pottery expert Abdallah Allamedine could only make a preliminary analysis. In this analysis, the majority of the diagnostic pottery was dated to either the first century or the sixth century CE (Yasmine 2006: 57-58). From such analyses as were made, it was suggested that the village was founded in the first century CE as part of the Roman colony of Beirut. That is, it was founded by veteran legionary soldiers and their families who, as far as we know, were Archaeological Landscapes of the Bekaa – Past Results and Future Prospects 11 Figure 7: The large Graeco-Roman temple at Hosn Niha given parcels of agricultural land in the central Bekaa Valley. Although the pottery did not reflect this, as little was dated to the second century CE, it was assumed the settlement’s prosperity reached its peak during this century, as it was during this period that the large temple was constructed. The building of a church and at least one large farmstead in the Byzantine period suggests the settlement continued to function into the later period. From the pottery, it was assumed that the settlement was completely abandoned with the coming of Islam in the mid-seventh century CE (Yasmine 2006: 48). Therefore, to obtain a more comprehensive assessment of the village remains, it was necessary to employ systematic methods of enquiry. Through the use of these approaches the work began to achieve a number of specific aims including: 1. A record of the range and spatial location of different structures and the extent of their preservation; 2. The acquisition of a more nuanced picture of the relationship between the village and the religious complex of temples and the succeeding church; 3. The reconstruction of the spatial and chronological development of the village; 4. And finally the placement of the settlement within its regional context. Although the work at Hosn Niha is ongoing, the first season in late 2011 was spent assessing the extent of the visible structural remains and their spatial relationships. The village remains were found to extend across some 600 metres southwest–north-east following the contour line of the slope and some 100 metres down to the foot of the slope (Figure 8). Within this six hectare area a range of features was recorded by differential GPS including various tombs, structures and exposed wall fragments of buried structures and terraces (Newson and Young 2014). As there appeared to be few complete structural remains on the surface, a series of 50-metre transects was undertaken to record surface density distributions of pottery and to collect a sample of diagnostic sherds from across the village site. This sample diagnostic pottery is currently being processed (by Dr Hanna Hamel), and further work is required on its assessment, particularly in light of potential future work at the site. However, some preliminary remarks can be made. Berytus LVI, 2016 12 Paul Newson A high percentage of the pottery recovered has been storage vessels, which follows Yasmine’s findings and reflects the intense agricultural nature of the settlement. Nevertheless, there are some finewares of different periods and there is some differential spatial patterning of pottery types and periods beginning to emerge across the site. As a consequence, it appears that the history of settlement at the site is perhaps more complex than either Marfoe or Yasmine predicted. Certainly it is amply clear settlement that extended into the Byzantine, perhaps the Early Islamic, which Marfoe did not envisage. Having said this, in line with the views of Yasmine, there is as yet no definite sign of settlement prior to the first century CE with an absence of evidence so far for early Hellenistic or earlier pottery, although some Eastern Sigillata A has been collected that could date to the first century BCE. This would seem to confirm that large-scale sedentary occupation at this relatively high altitude location could be directly related to the institution of the Roman colony. As perceived by Marfoe, this entailed the subsequent increased exploitation of formerly Figure 8: The key site remains at Hosn Niha Berytus LVI, 2016 marginal areas by Roman smallholders for the production of olive and vine (1978: 661-63). Undoubtedly, the main focus of settlement is that of the Roman and Byzantine period. Given such a long period of time, over 600 years at least, it is probably not surprising that the evidence so far collected would seem to suggest there were changes in the spatial morphology of the village. The surface pottery indicates that for the Roman period the settlement was a more dense agglomeration at the foot of the slope in the south-western area of the site. By the time of the Byzantine period, this compactness had been replaced by a more diffuse form of occupation spread over a greater area (Newson and Young 2015: 459). The type of dwelling also seems to have evolved between the two periods. In the Roman period, the village form would appear to be a concentration on many small dwellings, densely clustered with some less dense occupation on the margins. In the Byzantine period, there were more dispersed, large farmhouse types, as demonstrated by the substantial well-built Archaeological Landscapes of the Bekaa – Past Results and Future Prospects 13 remains of HN016 (Newson and Young 2014: 262; Yasmine 2006: 14-15). The numbers of these large farm houses may have been relatively few. As stated earlier, an additional problem is the current knowledge of the local pottery types. Particularly problematic is the extent to which the Byzantine pottery types continued into the early Islamic period. Whilst for the Bekaa this is at present unclear, what is very clear is that following a hiatus there was a reoccupation of part of the settlement in the early medieval period. The surface collection of diagnostics has identified a particular concentration of characteristic and glazed pottery sherds dating to the 13th and 14th centuries CE in the vicinity of the large stone construction in the south-western part of the site (HN003) (Newson and Young 2014: 265-66). In reviewing the evidence gained, particular questions spring to mind; notably, does this represent a continuation of occupation from the early Islamic period, or is this a reoccupation of the most favourable part of the site? Was the large stone ‘Tower construction’ (HN003) perhaps an early medieval fortified farm that guarded a pass over the mountain, the focus of a renewed or rejuvenated settlement? The interesting results from Hosn Niha in terms of its settlement development history and size have invited comparisons with other settlements to see if these results were similar or different. There was as yet no directly comparable rural site of a similar size and function in the central Bekaa Valley with which to make valid comparisons in terms of settlement history. Kamid el-Loz provides substantial evidence for occupation in the Hellenistic and the Roman period, but there is a shift away from this ancient tell site during the Byzantine period (Heinz 2016: 47). Hence in terms of the trajectory of settlement development and occupation, there are major differences apparent between these sites. On the other side of Mount Sannine, in the region of the Metn, one of the few rural sites in terms of comparable size and geographic situation to Hosn Niha has been investigated. This is the village site of Ej-Jaouze, the archaeological results of which are in the process of being published by the director Lina Nacouzi and others (Nacouzi et al. 2004) (Figure 2). The site of Ej-Jaouze, which lies at an altitude of 1320 m, seems to exhibit a similar occupation sequence for the Graeco-Roman period onwards. From the structural evidence at this site, it is clear there was a contrast in settlement structure between the small dense structures of the Roman period and the Byzantine period, which is characterised by the construction of a single large rectangular farmstead: a farmstead that appears very similar in design to the exposed rectangular structure ‘le grande residence,’ as named by Yasmine at Hosn Niha (HN016 - Figure 8). A further indication is that the surface pottery assemblage from Ej-Jaouze, which consists of pottery from the Roman and Byzantine periods followed by a lacuna in sherds until pottery of the 12th and 14th centuries, reveals a reoccupation of the site, just as at Hosn Niha. Niha Perhaps the best location for comparisons to Hosn Niha of village size, development history and occupation is the adjoining site of the village of Niha, two kilometres to the south-east, at an altitude of 1100 m (Figure 9). This too has a monumental Graeco-Roman temple sanctuary whose remains date to the first and second centuries CE and whose architecture is similar in style to that of Hosn Niha (Freyburger 1999; Krencker and Zschietzschmann 1938, 105-37; Steinsapir 2005: 74-78). Often found associated with the temple sanctuary are numerous statues, altars and other worked stone (Fani 2009). Niha village is also very important for the GraecoRoman period in terms of the number of Greek and Latin inscriptions that have been recorded in its environs. A high proportion of these inscriptions are memorials to burials, while others are altar and statue dedications (Figure 10). In terms of the Bekaa region, these form the best collection of inscriptions for this period outside Baalbek, and have been used frequently to provide insights into the society and governance of the Roman colony (Ghadban 1987; Rey-Coquais 1987). Berytus LVI, 2016 14 Paul Newson Figure 9: The Niha valley looking south towards the village of Niha, to the right of the village is the reconstructed large Graeco-Roman temple As with Hosn Niha, the temple sanctuary at Niha, and particularly the large temple, has been the subject of continued research on different aspects and of a protracted series of excavations and restorations in the 1940s and 1960s (summaries: Aliquot 2009: 300-03; Donceel 1966: 255-56; Yasmine 2009: 141-44). Unfortunately, the detailed evidence from the excavations, such as the pottery finds, have not been published. Consequently, there are gaps in the story of the sanctuary development, and very little is known about the village associated with it. A key hindrance to studying Niha village development is the occupation of the site today by the modern village. However, within and around the village are enclosed areas of open land or what might be designated fields, some of which are cultivated as vineyards or orchards or olive groves, and others of which are used as gardens or remain waste land. These areas have offered opportunities for the collection of surface material with which to compare the results from Hosn Niha. villagers, the addition or removal of topsoil from and to other regions around the village. Though some of these claims may have been aimed at dissuading archaeologists from investigating their land, others could have some truth in them. There has certainly been a great deal of remodelling of the Niha village landscape, for example, with new houses sometimes built on large, high platforms and the construction of roads on extended causeways; all of which has necessitated an extensive amount of terrace cutting into the steep-sided slopes of the narrow valley that Niha village occupies. The open areas were sampled in a systematic way so that densities of surface pottery could be calculated and could act as a check on soil changes. So transects of 10 metres apart were walked along the longest axis of each field, followed by a general collection of diagnostics. As with Hosn Niha, any archaeological feature, such as rock-cut tombs and ancient terracing, as well as information on the position of worked stones were recorded. There are some potential problems in collecting surface material from such fields, including the levelling of areas and, as reported by some Preliminary results have proved to be very promising. The collection of surface pottery has revealed interesting patterns in terms of their Berytus LVI, 2016 Archaeological Landscapes of the Bekaa – Past Results and Future Prospects 15 density across the village (Figure 11). The densest areas of pottery are located to the north and south of the temple sanctuary (Fields NH 2, 5, 6 and 12). The remaining dense areas of pottery generally line the course of the stream bed of the Nahr (river) Niha which bisects the village (Fields NH 12-15, 21, 33). Otherwise apart from the large fields NH 24 and 25 which abut the edge of the village in a small side valley, surface sherd density is very low. This is true as one moves further away from the present-day village core and also, interestingly, for the steeply terraced fields of the valley slopes which tightly Figure 10: The reconstructed large temple at Niha encompass the village. The lack of sherds on the terrace fields and on fields away from the village would suggest that higher seem to allude to some form of earlier activity/ density areas are a sign of occupation – i.e., where occupation of the site. the villagers lived. This would strongly suggest that the current village core is built more or less Further work on the ancient village. The need to go beyond the known site and to As for the periods in which this village was try and expand knowledge of the relationship inhabited, the post-Graeco-Roman period pottery between core village settlement sites and the, at has yet to be studied, but cursory analyses reveal present, relatively unknown inter-site regions of some patterning in sherd distribution (Figure 12). the Bekaa has led to a recent expansion of the The fields at the north end of the village (e.g., Central Bekaa Project (Figure 13). As a contrast NH 1, 2, 5 and 6) are dense in Graeco-Roman and comparison to the Niha Valley, the nearby wares, particularly finewares of Eastern Sigillata Ferzol Valley provides a good opportunity to A forms of the very end of the first century explore such inter-site relationships in depth. As BCE and early first century CE. This contrasts with Niha Valley, the Ferzol Valley is dominated sharply with fields immediately south of the by the site of a Graeco-Roman village which temple and along the course of the Nahr Niha, also has its own monumental temple sanctuary where, although there are some early Graecocomplex (Aliquot 2009: 306; Taylor 1971: 33). Roman sherds, the majority are from the Islamic The Ferzol Valley (also called the Wadi el-Habis) periods and particularly apparent are the glazed is wider and more densely farmed than the Niha wares of the 13th-14th centuries CE. Of more Valley (Figure 14). A programme of intense intriguing interest are a number of earlier sherds, systematic fieldwalking was begun in 2016 at occasional sherds of the third-fourth millennium different locations within the valley to sample Chalcolithic, Bronze and Iron Ages which were different valley environments and to look for located in fields on the eastern and southern edges evidence of small sites and other occupational of the built-up core of the village (NH14, 15, activity outside the large long-term settlement 21 and 33). Although only very few in number, sites as epitomised by the village sites at Niha the consistent location of these earlier sherds, and Ferzol. Therefore, for the first season, four and their being from successive periods would areas were fieldwalked: Berytus LVI, 2016 16 Paul Newson Figure 11: Density map of surface pottery in and around the village of Niha. There is a much higher density of sherds within the core village area, especially along the course of the Nahr Niha, and in the environs of the Graeco-Roman temple sanctuary Berytus LVI, 2016 Archaeological Landscapes of the Bekaa – Past Results and Future Prospects 17 Figure 12: Map of Niha showing pattern of sherds relating to major periods. This is based on a preliminary assessment of the surface pottery. The distribution highlights a number of issues which require further work. Although the numbers are very few, there is evidence for some activity in the valley in the period before the Hellenistic. Secondly, there is a significant concentration of early Roman pottery in the regions to the north of the present village. Thirdly, pottery numbers drop significantly as one moves away from the village centre (see Figure 11), in this central region, medieval Islamic pottery dominates, although there are also significant numbers of Roman ceramic sherds Berytus LVI, 2016 18 Paul Newson 1. The upper high valley area beyond the escarpment which divides it from the rest of the valley. monastery, includes the site of numerous rock-cut tombs and a karstic spring (Newson 2015; Nordiguian 2005: 46-49). 2. The region of Nabaa el-Habis or, as commonly called in academic publications, ‘La Laure’ (French - hermetic monastery), immediately at the foot of the escarpment which, besides the supposed hermetic 3. A sample of fields midway between ‘La Laure’ and the village of Ferzol. 4. A sample of open areas closest to the location of the Graeco-Roman temple sanctuary. Figure 13: Map of fieldwalked areas of the Ferzol valley. Initial results suggest areas of dense pottery within Areas 2 and 4. Area 3 has significant coverage of pottery, while Area 1 has very little. Further systematic fieldwalking is planned Berytus LVI, 2016 Archaeological Landscapes of the Bekaa – Past Results and Future Prospects 19 Figure 14: The Ferzol valley (Wadi el Habis) looking southwest towards the Bekaa valley taken from ‘La Laure’. This valley is wider than the neighbouring Niha valley, and is unsurprisingly much more densely cultivated with many fine vineyards The results are at a preliminary stage, but again the collection of surface material in a systematic manner is already providing valuable information. The density of sherds varies between the different sample areas, with very few sherds from area 1 compared to area 3, indicating that the upper region was an area of little activity. Within area 3, very dense areas of surface sherds have indicated the presence of isolated, small occupation sites, including at least one for the Hellenistic period, which is likely to be related to a farmstead of some sort. There is pottery from multiple periods from both ‘La Laure’ and the village, especially of the Islamic and Graeco-Roman periods, suggesting as with the Niha Valley, long-term occupation over many periods. Furthermore, initial impressions of the very high percentage of early medieval Islamic pottery at the ‘Laure’ site would seem to suggest that at this time a thriving community was prospering in this location (Figure 15). Conclusion This paper has sought to highlight and discuss just a few of the issues concerned with our understanding of the Bekaa. It has shown that relatively little work has so far been accomplished beyond intra-site studies of pivotal sites within the Bekaa plain. Although the magisterial work of Marfoe remains the basis for any assessment of long-term settlement, the application of modern landscape methodologies is beginning to clarify our views. Many of Marfoe’s hypotheses are being confirmed whilst others will have to be re-thought. Systematic surface investigation in the central Bekaa Valley is beginning to show interesting results and is changing perceived notions of the rural settlement history of the Bekaa. It is hoped this work will continue in the near future and allow us to obtain a more nuanced view of the occupation sequence within this region, Berytus LVI, 2016 20 Paul Newson Figure 15: The rock-cut limestone scarp of ‘La Laure’ in the Ferzol valley. This is a rich zone of rockcut tombs; some were converted to troglodyte habitations in later periods and traditionally formed a monastery particularly for evidence of the cycle of population continuity or rupture between different settlement periods between the Late Byzantine and the early Islamic periods. It could be that what the archaeology at sites such as Hosn Niha is revealing is a characteristic settlement history for the upland regions of the Lebanon, in which relatively high altitude areas are only settled as a result of particular socio-cultural conditions prevalent in the lower lying regions. This tends to confirm Marfoe’s thesis on some levels, but it also highlights some of the limitations of his study in concentrating on the Bekaa plain and regions below the 1500 m contour. Berytus LVI, 2016 Future work will necessarily need to investigate the relationship between the plain and the mountain, and to use the whole raft of tools available to current landscape archaeologists (e.g., remote sensing technologies, geophysical work, diatom studies and residue analyses) to account for new concerns from levels of scale to environment exploitation (Altaweel 2016: 1706). By completing such studies, we will be able to better assess the extent to which this important region conforms or otherwise to other regions in the Near East, and not just in settlement trajectories but in many other ways as well. Archaeological Landscapes of the Bekaa – Past Results and Future Prospects 21 Marfoe Site Number Marfoe Site Name (Other Name) Site Type Source of Information Marfoe Palaeolithic Environment Zone 196 El Karak (Karak Nuh) Modern village: Tomb Archaeological remains UCAT 197 Qsarnaba Modern village: Temple Archaeological remains 198 Temnine el Faouqa Temple 199 Niha 200 Late Bronze Hellenistic Islamic (Medieval) Ottoman X X X PPW X X X Archaeological remains PPW X Modern village: Temple Archaeological remains PPW X X X Hosn Niha Temple Archaeological remains PPW X 201 El Fourzol (Ferzol) Modern village: Temple Archaeological remains PPW X X 202 Ain es Sefli Roman site Archaeological remains PPW 203 Bir el Joubb Unknown Archaeological remains PPW 204 Balbassane Unknown Archaeological remains PPW 205 Nebaa el Habis Tombs Archaeological remains PPW *230 Jebel Baniya Palaeolithic tool scatter Archaeological remains MS-MW 246 Mseidej Roman site Archaeological remains RPS 428 Zahle Modern town Text reference PPW 429 Maallaqa Modern village Text reference AAF 430 Nabi Ayla Modern village Text reference PPW X 443 Bednayel Modern village Text reference PPW X 444 Beit Chama Modern village Text reference PPW X 445 Kfar Dabach Modern village Text reference RPS X X Roman X Byzantine X X X X X X X Table 1: Shows a summary of Marfoe’s site catalog for the Western Piedmont of the Bekaa (See Figure 6). For each site the known dated material information is shown, and the Marfoe environment zone of each site location (Marfoe 1978). The Environment Zone abbreviations stand for: AAF – Active Alluvial Fan; MS-MW – Mountain Slope/Mountain Wadi; PPW – Perennial Piedmont Wadis; RPS – Rocky Piedmont Slopes; UCAT – Upper Colluvial-Alluvial Terraces. An * for the site number 230 (Jebel Baniya) means the site is outside Marfoe’s Survey Area Acknowledgements The fieldwork mentioned in this paper has been funded by generous grants from the University Research Board of AUB. I would also like to thank the Directorate General of Antiquities, Lebanon, and especially the Director General Sarkis El Khoury and Raffi Gerjian for their continued support, as well as the people of Ferzol and Niha. And not least, I am grateful to the late Professor Tony Wilkinson for interesting discussions of his work and time with Leon Marfoe in the Bekaa and its importance as an exceptional phenomena of the Near East. Berytus LVI, 2016 22 Paul Newson Bibliography Abou Diwan G and Doumit J 2016 Ancient wetlands of the Biqāʿ: A buffer zone between the hinterlands of Sidon and Berytus in the Roman period. BAAL (2012) 16: 215-52. Algaze G, Breuninger R and Knudstad J 1994 The Tigris-Euphrates Archaeological Reconnaissance Project: Final report of the Birecik and Carchemish Dam survey areas. Anatolica 20: 1-96. Aliquot J 2009 La vie religieuse au Liban sous l’empire romain. BAH (Bibliothèque Archéologique et Historique) 189. Beirut: IFPO. Altaweel M 2016 Review of T J Wilkinson, E Peltenburg and E B Wilkinson (eds) Carchemish in context: The Land of Carchemish Project 2006-2010. Antiquity 90: 1705-07. Banning E 1996 Highlands and lowlands: Problems and survey frameworks for rural archaeology in the Near East. BASOR 301: 25-45. Baramki D 1961 Preliminary report on the excavations at Tell el Ghassil. BMB 16: 87-102. Baramki D 1964 Second preliminary report on the excavations at Tell el Ghassil. BMB 17: 47-103. Baramki D 1966 Third preliminary report on the excavations at Tell el Ghassil. BMB 19: 29-49. Bonatz D, Ali, N and Jauss C 2002 Preliminary remarks on an archaeological survey in the AntiLebanon. BAAL 6: 283-307. Burckhardt J L 1822 Travels in Syria and the Holy Land. London: John Murray. Chatty D 1975 From Camel to Truck: A study of the pastoral economy of Al Fadl and the Al Hassena in the Beqaa Valley of Lebanon. PhD dissertation, University of Califonia, Los Angeles. Chatty D 2013 From Camel to Truck: The bedouin in the modern world. 2nd ed. Cambridge: The White Horse Press. Chéhab H K 1993 On the identification of ʿAnjar (ʿAyn al-Jarr) as an Umayyad foundation. Muqarnas 10: 42-48. Chéhab M 1963 The Umayyad Palace at ʿAnjar. Ars Orientalia 5: 17-25. Copeland L 1969 Neolithic village sites in the South Beqaa, Lebanon. MUSJ 45: 85-114. Berytus LVI, 2016 Copeland L and Wescombe P 1965 Inventory of Stone-Age sites in Lebanon, Part I. MUSJ 41: 30-175. Copeland L and Wescombe P 1966 Inventory of Stone-Age sites in Lebanon, Part II. MUSJ 42: 1-174. Decker M 2007 Frontier settlement and economy in the Byzantine East. DOPap 61: 217-67. Decker M 2009 Tilling the Hateful Earth. Agricultural Production and Trade in the Late Antique East. Oxford: Oxford University Press. Donceel R 1966 Recherches et travaux archéologiques récents au Liban (1962-1965). L’Antiquité classique 35: 222-61. Doumet-Serhal C 1996 Les fouilles de Tell El-Ghassil de 1972 à 1974: Étude du matérial. BAH (Bibliothèque Archéologique et Historique) 146. Beirut: IFPO. Fani Z 2009 Le tympan ouest et les couronnes du temple a de Niha. Topoi 16: 153-67. Finster B 2007 Researches in ʿAnjar II. Preliminary report on the ornaments of ʿAnjar. BAAL 11: 143-65. Fischer-Genz B and Ehrig H 2005 First results of the archaeological survey project in the territory of ancient Heliopolis-Baalbek. BAAL 9: 135-38. Freyburger K S 1999 Les temples de Niha. Témoins de cultes locaux d’influence romaine au Liban. Topoi 9: 569-77. Ghadban C 1987 Observations sur le statut des terres et l’organisation des villages dans le Beqaʿ hellénistique et romaine. In E Frézouls (ed) Sociétés urbaines, sociétés rurales dans l’Asie Mineure et la Syrie hellénistiques et romaines: actes du colloque organisé à Strasbourg (novembre 1985) par l’Institut et le Groupe de Recherche d’histoire romaine et le Centre de recherche sur le Proche-Orient et la Grèce. Strasbourg: AECR, 217-238. Hachmann R 1989 Preliminary remarks on an archaeological survey in the Anti-Lebanon, 1963-1981. German excavations in Lebanon. Part One. Berytus 37: 5-187. Heinz M 2016 Kamid el-Loz: Imperialism, urbanism and communication – Facets of history and a concise note on our current research concept. Berytus 55: 43-78. Heinz M with Kulemann-Ossen S, van Lengerich L, Leschke C, Nieling J, von Rüden C, Wagner E 2010 Kamid el-Loz. Intermediary between Cultures: More Than 10 Years of Archaeological Research in Archaeological Landscapes of the Bekaa – Past Results and Future Prospects 23 Kamid el-Loz (1997-2007). BAAL Hors-Série VII. Beirut: Ministry of Culture and Directorate general of Antiquities of Lebanon. Marfoe L 1979 The integrative transformation: Patterns of sociopolitical organisation in southern Syria. BASOR 234: 1-42. Hillenbrand R 1999 ʿAnjar and early Islamic urbanism. In G P Brogiolo and B Ward-Perkins (eds) The Idea and Ideal of the Town between Late Antiquity and the Early Middle Ages. Leiden: E. J. Brill, 59-98. Marfoe L 1982 Empire and ethnicity in Syrian society: From archaeology to historical sociology ‘revisited’. In Archéologie au Levant: recueil à Roger Saidah. Collection de la Maison de l’Orient Mediterranean no. 12. Lyon: Maison de l’Orient, 463-79. Horden P and Purcell N 2000 The Corrupting Sea: A Study of Mediterranean History. Oxford: Blackwell. Jidejian N 1975 Baalbek: Heliopolis “City of the Sun”. Beirut: Dar el-Machreq. Krencker D and Zschietzschmann W 1938 Römische Tempel in Syrien, nach Aufnahmen und Untersuchungen von Mitgliedern der deutschen Baalbekexpedition 1901-1904. Berlin: W. de Gruyter. Kuschke A 1978 Preliminary remarks on an archaeological survey in the Northern Biqa. BMB 30: 43-46. Kuschke A, Mittmann S and Müller U 1976 Archäologischer Survey in der nördlichen Biqa., Herbst 1972. Report on a Prehistoric Survey in the Northern Biqa. TAVO Beihefte, Reihe B11. Wiesbaden: Dr Ludwig Reichert. Lyonnet B 1996 Settlement pattern in the Upper Khabur (N.E. Syria), from the Achaemenids to the ʿAbbasid period: Methods and preliminary results from a survey. In K Bartl and S R Hauser (eds) Continuity and Change in Northern Mesopotamia from the Hellenistic to the Early Islamic Period. Berliner Beiträge zum vorderen Orient, Band 17. Berlin: Dietrich Reimer, 349-61. MacDonald B 1988 The Wadi el-Hasa Archaeological Survey of the Kerak Plateau 1979-1983, West Central Jordan. Waterloo, Ontario: Wilfred Laurier University. MacDonald B 1992 The Southern Ghors and Northeast ʿArabah Archaeological Survey. Sheffield Archaeological Monographs 5. Sheffield: J.R. Collis Publications. Makdisi U 1998 The “Rediscovery” of Baalbek: A metaphor for empire in the nineteenth century. In H Sader, T Scheffler and A Neuwirth (eds) Baalbek: Image and Monument 1898-1998. Beiruter Texte und Studien, Band 69. Beirut: Franz Steiner, 137-56. Marfoe L 1978 Between Qadesh and Kumidi: A History of Frontier Settlement and Land Use in the Biqa, Lebanon. PhD dissertation, University of Chicago. Marfoe L 1995 Kāmid el-Lōz 13. The Prehistoric and Early Historic Context of the Site: Catalog and Commentary. SBZA Band 41. Bonn: Rudolf Habelt. Marfoe L 1998 Kāmid el-Lōz 14. Settlement History of the Biqāʻ up to the Iron Age. SBZA Band 53. Bonn: Rudolf Habelt. Nacouzi L with Alpi F and Pieri D 2004 Ej-Jaouzé (Metn, Liban): Mission de 2003. BAAL 8: 211-62. Newson P 2015 Greco-Roman burial practices in the Bekaa Valley, Lebanon, and its adjacent uplands. Journal of Eastern Mediterranean Archaeology and Heritage Studies 3: 349-71. Newson P and Young R 2014 Archaeological assessment of the site at Hosn Niha: Preliminary report 2011-2012. BAAL (2011) 15: 257-80. Newson P and Young R 2015 The archaeology of conflict damaged sites: Hosn Niha in the Biqaʿ Valley, Lebanon. Antiquity 89: 449-63. Nordiguian L 2005 Temples de l’époque romaine du Liban. Beirut: Presses de l’Université Saint-Joseph. Parker S T 1992 The Limes and settlement patterns in central Jordan in the Roman and Byzantine periods. In S Tell, G Bisheh, F Zayddine, K ʿAmr and M Zaghloul (eds) SHAJ IV. Amman, Jordan: Department of Antiquities, 321-24. Rey-Coquais J-P 1967 Inscriptions Grecques et Latines de la Syrie. Tome VI. Baalbek et Biqāʿ. BAH 78. Paris: Paul Geuthner. Rey-Coquais J-P 1987 Des montagnes au désert: Baetocécé, le pagus Augustus de Niha, la Ghouta à l’Est de Damas. In E Frézouls (ed) Sociétés urbaines, sociétés rurales dans l’Asie Mineure et la Syrie hellénistiques et romaines: actes du colloque organisé à Strasbourg (novembre 1985) par l’Institut et le Groupe de Recherche d’histoire romaine et le Centre de recherche sur le Proche-Orient et la Grèce. Strasbourg: AECR, 191-216. Berytus LVI, 2016 24 Paul Newson Schroeder H B 1970 A prehistoric survey in the Northern Bekaa Valley. BMB 23: 193-204. Schroeder H B 1991 Natufian in the Central Bekaa Valley, Lebanon. In O Bar-Yosef and F R. Valla (eds) The Natufian Culture in the Levant. Archaeological Series 1. Ann Arbor, MI: International Monographs in Prehistory, 43-80. Volney C F 1787 Travels Through Egypt and Syria in the Years 1783, 1784, and 1785. Volume 2. London: G. G. J. and J. Robinson. Wilkinson T J 2003 Archaeological Landscapes of the Near East. Tucson, AZ: University of Arizona Press. Sommer M 2001 The Bekaa Valley in Antiquity. A regional-historic survey. BAAL 5: 71-91. Wood R 1757 The Ruins of Baalbec, otherwise Heliopolis in Coelosyria. London. Accessed 20 November 2016: http://digi.ub.uni-heidelberg.de/diglit/wood1757 Steinsapir A I 2005 Rural Sanctuaries in Roman Syria: The creation of a sacred landscape. BAR IS 1431. Oxford: British Archaeological Reports. Yasmine J 2005 Remaniements de temples d’époque romaine; les cas de Niha et de Faqra. L’apport de l’étude métrologique. BAAL 9: 301-16. Tate G 1997 The Syrian countryside during the Roman era. In S E Alcock (ed) The Early Roman Empire in the East. Oxford: Oxbow: 55-72. Yasmine J 2006 Complexes cultuels ruraux d’époque romaine dans la Beqaa (Liban): le cas de HosnNiha problèmes d’arcitecture et d’archéologie. PhD dissertation, University of Paris I Sorbonne. Taylor G 1971 The Roman Temples of Lebanon: Les Temples Romains au Liban. 2nd ed. Beirut: Dar el-Machreq. Thoumin R 1936 Geographie humaine de la Syrie centrale. Paris: Librairie Ernest Roux. Van Ess M (ed) 2008 Baalbek/Helipolis: Results of Archaeological and Architectural Research 20022005. BAAL, Hors Série IV. Beirut: Ministère de la Culture, Direction Génerale des Antiquités. Berytus LVI, 2016 Yasmine J 2009 Transformations monumentales de sanctuaires et de temples ruraux antiques: les cas de Hardine et de Niha. Topoi 16: 121-52. Yasmine J 2013 The Niha sites (Lebanon) cultural landscape: A 3D model of sanctuaries and their context. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Volume XL-5/W2: 699-704. Accessed 20 November 2016: http://www.int-arch-photogramm-remote-sensspatial-inf-sci.net/XL-5-W2/699/2013/