Ancestral Veneration and the Settlement History of Tell Sukas

Palestine Exploration Quarterly, 141, 2 (2009), 138–151 ANCESTRAL VENERATION AND THE SETTLEMENT – KA –S HISTORY OF TELL SU Michael Brown Scholarship of the Bronze-to-Iron Age transition in the northern Levantine littoral has traditionally stressed a major disjuncture in settlement and societal development c. 1200 BCE, concurrent with the destruction and abandonment of the Ugaritic capital at Ras Shamra. Based upon a reinterpretation of the archaeological remains at Tell Sūkās, this study argues for a degree of regional continuity in cultural lineage from the midsecond to early first millennium BCE, based upon the proposed identification of a previously unknown example of ancestral veneration. introduction Tell Sūkās is a multi-period (c. 2000 bce–c. 400 ce) site of approximately 0.8 ha in size, located on the coastal fringe of the Gabla Plain in southern Syria (Fig. 1). The site consists of a central tell that overlooks two natural harbours directly to the north and south of the main settlement. On the far shore of the southern harbour an associated outdoor sanctuary complex has also been excavated at Mīna Sūkās (Fig. 2). With the exception of two soundings made in 1934 by E. O. Forrer, the site was excavated between 1958 and 1963 (not including 1962) by the Carlsberg Expedition to Phoenicia under the directorship of P. J. Riis.1 On the basis of epigraphic evidence, Tell Sūkās has been identified with ancient Souksi, which places it within the boundaries of the kingdom of Siyannu-Ušhnatu at the close of the Late Bronze Age. In common with other eastern Mediterranean coastal sites including Enkomi and nearby Ras Shamra, the stratigraphic sequence of Tell Sūkās shows abrupt charcoal and/or ash deposits, interpreted by the excavators as evidence of urban conflagration occurring c. 1170 bce (Riis 1970, 24).2 Unlike its larger contemporary to the north, however, there is no conclusive evidence for even a hiatus in settlement at Tell Sūkās (Lund 1986, 40–42), and the Early Iron Age remains instead display extensive redevelopment and expansion. The site is therefore essential to an understanding of events on the Syrian littoral for the period in question, owing to its continued occupation post-‘destruction’ c. 1170 bce. In the centre of the tell, a Middle Bronze Age collective grave (Tomb IV) was excavated containing 41 individuals, the vast majority of whom were represented by disarticulated skeletal components which had been subject to secondary treatment. The internment site was associated with contemporary and immediately subsequent phases of architecture. While the fragmentary state and poor preservation of these buildings makes a characterisation of their function difficult, they are most probably to be regarded as private dwellings (Riis 1970, 126; Lund 1986, 185). This paper presents a re-interpretation of the architectural remains overlying the grave site, based upon a distributional analysis of associated artefacts. In examining the respective periodic patterning of deposition across the Bronze-to-Iron Age divide, it is intended to facilitate a more balanced assessment of the relative degrees of continuity and change in the Address correspondence to Michael Brown, School of History, Classics and Archaeology, University of Edinburgh, Old High School, Infirmary St., Edinburgh EH1 1LT, Scotland, email: M.G.Brown-2@sms.ed.ac. uk © Palestine Exploration Fund 2009 doi: 10.1179/174313008x437837 a n c e s t ral veneratio n at tell sūk ās 139 Fig. 1. The northern Levantine region with principle sites mentioned in the text. application of material culture in coastal Syria following the disruptions of the late second millennium bce. methodology The focus of this analysis is the spatial patterning of deposition at Tell Sūkās during the periods designated by Riis et al. as J (Late Bronze Age: c. 1600–1170 bce) and H2 (Early Iron Age: c. 1170–850 bce). Excavation of the Late Bronze Age (J) strata was undertaken only on the basis of, ‘where it was held to be indispensable’, in order to comprehend settlement development (Riis 1970, 20). This situation contrasts with the strategy adopted in relation to the Iron Age (H) remains, which were a primary focus of investigation for the expedition in order to establish ‘a safer chronology of the Iron Age culture in Phoenicia’, and with a view to ‘elucidate the relations between the Near East and Greece during the same period’ (Riis 1970, 10). While in the interests of clarity it would have been desirable to analyse material derived solely from the J and H2 periods, in order to compare the respective spatial and compositional patterning of artefacts directly before and after the disruptions of the late 12th century bce, the disturbed stratigraphy of the site precluded the excavators from reliably 140 pales tine ex ploration quarte rl y, 141, 2 , 2 009 Fig. 2. Enhanced satellite view of Tell Sūkās facing east from the Mediterranean coast. (Courtesy of Google Earth™ mapping service.) assigning all items to a particular Iron Age sub-division. As such, there is a potential inclusion of material dating from up to 175 years after the primary H2 focus of inquiry. Although the number of objects involved is likely to be very low because of the removal of those items specified as belonging to the later H1 period (c. 850–675 bce), there is nonetheless the theoretical risk of sample ‘contamination’. However, based upon the presumption that the majority of artefacts included in the J and modified H categories belong to the final phase of occupation in both periods, a comparative analysis is valid as each sample population largely consists of in situ abrupt abandonment material (Riis 1970, 27, 29). The two sample populations used in the following analysis thus consist of all those artefacts representing individual items or parts thereof, which have been catalogued in the Sūkās excavation reports as pertaining to the Late Bronze Age and Early Iron Age. Any errors or inconsistencies in stratigraphic attribution made by the expedition are accordingly carried over into this analysis.3 Each artefact included in the periodic samples is further identified by its find spot and its typological categorisation. The resolution of the former measurement allows each object to be located within a 5 m2 area of the excavation grid, based upon their 10 m2 grid square and compass quadrant designations (for example the easternmost area excavated is to be found in the southwest corner of square G down and 19 across, thus receiving the catalogue entry G19SW).4 The typological categorisation of all artefacts used in this study, in common with the strata designations and find localisations, is that established in the expedition’s Sūkās reports. The Near Eastern pottery and associated artefacts (of overwhelmingly Levantine origin) were discussed by Buhl (1983) in Sūkās VII, and the Aegean material was published by Ploug (1973) in Sūkās II. While these volumes, when taken together with the strata designation amendments noted in Sūkās VIII, can be considered to be as full a representation as can presently be attained from the publication series, the Cypriot material has not yet received dedicated treatment, and has only been given more limited attention in the site excavation narratives contained within Sūkās I (Riis 1970, 20–40) and Sūkās VIII (Lund 1986, 11–51). As the original intention of these volumes was to present the stratigraphy and architectural a n c e s t ral veneratio n at tell sūk ās 141 remains, and to date these on the basis of the finds, priority was given to items of possible chronological significance. It should therefore be noted that the artefact categories used in this study do not constitute a fully comprehensive, but rather a representative sample population, in their present published form. The combined sample population for the period J and modified H artefact groups consists of 142 temporally, spatially and typologically provenanced objects as defined by the criteria outlined above. The largest group is the period H assemblage, which accounts for 80% (114 objects) of the total artefact population, with the remaining 20% (28 objects) forming the J period assemblage. With the limitations inherent in the two sample populations in mind, together with the generally low artefact quantities involved, the following is therefore restricted to a qualitative assessment of the distribution and degree of continuity in find localisations, together with a compositional ratio comparison for and between the J and modified H period artefact groups. The spatial distribution of find localisations is shown in Fig. 3 (Late Bronze Age period J) and Fig. 4 (Early Iron Age period H). The extent of the exposed area in which material from the respective strata was retrieved is indicated by the shaded areas in Fig. 3 (J — 312.5 sq m) and Fig. 4 (H — 525 sq m). All objects included in the periodic sample groups are listed by artefact class in Table 1. tell sūkās A comparison of the sample distribution maps (Fig. 3–4) shows that the Early Iron Age (H) assemblage is more widely spread than that of the Late Bronze Age (J). While this is primarily the result of the varying degrees to which the respective strata have received investigative attention, rather than demonstrating actual trends in settlement development, of potentially more interest is the discernable concentration of finds from both periods in grid quadrant G11SW (Fig. 3–4). This is likely to be significant because of the earlier presence in this area of a large Middle Bronze Age (c. 2000–1600 bce) collective grave (Tomb IV), spread across three burial levels with a vertical aperture of c. 2.5m in diameter (Thrane 1978). While the relatively poor preservation of the Late Bronze Age (J) strata due to subsequent construction activity (Lund 1986, 13) means that the architectural relationship between this feature and the remains directly overlying it is not entirely clear, there is evidence to suggest a degree of continuity in form between the two periods. To quote Lund (1986, 15), ‘The “northern-wall” (which traversed the area of G11SW on an orientation of east–north-east/west–south-west) had obviously been reutilised and similar reuse of earlier structures probably occurred elsewhere, as manifested by the preservation of the orientation of the earlier walls in the new complex’. Furthermore, in the subsequent architectural horizon pertaining to the Early Iron Age (H2), the ‘extensive reuse of earlier walls’ (Lund 1986, 24) in the incumbent Complex V again shows a significant incorporation of components from the preceding Late Bronze Age building, thereby extending the possible duration of some kind of continuity of place related to the grave site across the Bronze-to-Iron Age divide (see Fig. 5 for diachronic representation of G11SW). The incomplete excavation of the Early Iron Age building and the underlying Bronze Age structures (of which the respective remains in G11SW are a part), unfortunately means that neither can be fully understood in terms of its architectural function or role within the wider contemporary settlement plan. It is interesting to note, however, that while the Early Iron Age (H2) redevelopment of the site incorporated a renewal of structural features dating from the preceding Late Bronze Age (Riis 1970, 34; Lund 1986, 24, 28, 42), the subsequent Late Iron Age (H1) horizon demonstrates markedly less continuity with the preceding phase 142 pales tine ex ploration quarte rl y, 141, 2 , 2 009 Fig. 3. Spatial distribution of Late Bronze Age (J) sample. (Map based upon that established by E. Fugmann, Chief Architect — Carlsberg Expedition to Phoenicia.) in terms of either architecture (Lund 1986, 32) or depositional patterning. This break with tradition at the close of the H2 period may correspond with another possible ‘destruction’ at Tell Sūkās during the 9th century bce, attributed by Riis (1970, 40) to the repeated Assyrian invasions under Shalmanesar III in the years 858–844 bce.5 The respective compositional make-up of the two sample populations also shows significant aspects of both continuity and change between the Late Bronze and Early Iron Age periods. In the Late Bronze Age (J) sample the ‘indigenous’ Near Eastern artefacts account for the vast majority of finds at 76% of the total assemblage, with the Cypriot objects a n c e s t ral veneratio n at tell sūk ās 143 Fig. 4. Spatial distribution of Early Iron Age (H) sample. (Map based upon that established by E. Fugmann, Chief Architect — Carlsberg Expedition to Phoenicia.) forming the second largest category at 14%. In the Early Iron Age (H) sample the proportion of Near Eastern objects drops to 69%, while the number of Cypriot items increases to 27%. While it should be noted that the priority given to recording finds of chronological significance could have potentially led to an over-representation of Cypriot material, thereby precluding a reliable determination of compositional significance in quantitative terms, this increase of 13% in the H sample when compared with the J assemblage suggests an escalation in the procurement and utilisation of Cypriot type pottery at Tell Sūkās in the Early Iron Age. 144 pales tine ex ploration quarte rl y, 141, 2 , 2 009 Table 1. Tell Sūkās periodic samples by artefact class. ARTEFACT CLASS Near Eastern pottery and other objects Near Eastern pottery Lamps Wall brackets Metal objects including weapons and tools Beads, pendants, and amulets Cylinder seals, stamp seals and sealing Scarabs, scarab impression and scaraboids Objects shaped as human beings and animals Bone and ivory objects Stone objects Miscellanea Aegean Pottery and terracottas Minoan and Mycenaean vases and figurines Cypriot pottery White painted pendant line style Base ring ware White slip II ware White painted I or II ware White painted III ware(?) Cypriot unclassified LBA (J) QUANTITY EARLY IRON AGE (H) QUANTITY 15 0 0 0 0 1 1 5 0 0 0 35 2 3 13 15 1 1 2 2 4 1 3 4 1 1 1 0 0 1 0 9 15 3 3 1 Fig. 5. From Bronze-to-Early Iron Age in G11SW. a) Architectural context of the Middle Bronze Age collective grave (Riis 1970, 12); b) Late Bronze Age architectural horizon (Riis 1970, 13); c) Early Iron Age architectural horizon (Riis 1970, 27). All figures 1:100. (Drawings by E. Simony, G. Simony and E. Fugmann.) In the case of the Syrian littoral, the occurrence of Aegean and/or Cypriot artefacts has consistently been viewed as significant in that it supposedly alludes to the activities of Aegean and/or Cypriot peoples. The ‘suspect’ nature in this context of Aegean pottery in particular has meant that there is a lingering tendency to view its presence as being indicative of visiting traders, or the establishment of embedded pottery manufacturing enclaves a n c e s t ral veneratio n at tell sūk ās 145 (e.g., Iacovou 1988, 84).6 However, the Aegean artefacts form the smallest group in both periods, accounting for 10% in the Late Bronze Age (J) sample, and a mere 4% in the Early Iron Age (H) assemblage, suggesting that the Aegean market was only of peripheral importance to the settlement’s inhabitants at this time.7 This apparent decline in imports during the Early Iron Age when compared to the Late Bronze Age may reflect a degree of import substitution rather than a cessation in the use of these prestige commodities in mortuary and associated contexts (van Wijngaarden 2005, 114–115, 123 esp. Tell Dan), through a process of ‘hybridisation’ in which previously exotic Aegean and Cypriot forms were assimilated into the local ceramic repertoire. Worthy of mention in this regard is that of the small amount of Minoan and Mycenaean artefacts to have been found from either period at Tell Sūkās, all are classified by the excavators as being either vases or figurines, raising the possibility that these items were regarded as some form of ‘exotica’ rather than having a more utilitarian value. Regrettably, the Aegean and Cypriot material found in an Early Iron Age context is described in insufficient detail to determine the extent to which these objects are ‘heirlooms’ recycled from the immediately underlying Late Bronze age deposits. Changes are also evident in the material composition of the Early Iron Age (H) sample when compared to the Late Bronze Age (J) assemblage (Table 1). In general terms, the Iron Age material is more diverse, incorporating metal objects (entirely absent from the Late Bronze Age J sample), and a greater percentage of non-utilitarian decorative goods. These include items of personal adornment (beads, pendants and amulets) which are again missing from the Late Bronze Age (J) assemblage, while in the Early Iron Age (H) they account for 13% of the total. If as proposed the spatial patterning identified above in G11SW does represent a disproportionate association of material finds directly over the site of the Middle Bronze Age collective grave, this would seem to suggest a potential continuity in related function, and by extension depositional behaviour, throughout the Late Bronze Age and into the Early Iron Age. It is proposed that the pattern of artefact deposition may indicate that acts of ancestral veneration were taking place in this area, which in turn would seem to denote continuity in cultural tradition, and by extension population lineage (Lund 1986, 187), following the disruptions of c. 1170 bce. In the absence of direct corroboratory evidence, it should be stressed that this interpretation of the spatial feature identified in G11SW remains highly speculative. The practice of ancestral veneration is, however, arguably suggested by both the proximity of internments, and the evident survival of comparatively large quantities of pottery in this area, which could indicate its containment in some form of environment of preservation such as a sanctified enclosure. The conservative retention of architectural components from the Middle Bronze Age through to the Early Iron Age may also indicate the conscious incorporation of this ‘sacred’ space into subsequent domestic structures. From the grave itself there is evidence for continuing and frequent contact with the deceased post-internment, up until the tomb is sealed at the close of the Middle Bronze Age, ‘The confused character of the grave, and the often very bewildering situations shown by the many bones found in haphazard positions seem to reflect a very intensive ancient secondary displacement of bones and antiquities alike’ (Thrane 1978, 3).8 Fragmentary animal and human bones were also found in disturbed Late Bronze Age and Early Iron Age contexts in G11SW and immediately to the south (Lund 1986, 15, 28). If the proposed interpretation is correct, then these artefacts may represent a latent act of visitation in the form of a subsequent association of deceased members of the community, with accompanying offerings of food and drink to the ancestors, above the original tomb site. 146 pales tine ex ploration quarte rl y, 141, 2 , 2 009 In the eastern sector of the tell, Riis (1970, 38–40) differentiated the Bronze-to-Early Iron Age remains of Complex IV from the surrounding habitation quarter, suggesting that they may have served at least in part a sacral purpose. In support of this hypothesis he drew architectural parallels between this structure with its associated courtyard and pit, and the Bronze Age ‘patrician’ houses of Tell Beit Mirsim and Megiddo (Oren 1992). Adjacent to the south harbour at Mīna Sūkās a late second–early first millennium bce open-air sanctuary complex was also excavated that exhibited numerous features in common with contemporary cultic sites in Cyprus at Athienou and Soloi (Riis et al. 1996, 27–28). It was surmised that this sanctuary was most likely dedicated to the same deity (probably Melqart) who was the subject of worship in the later Graeco–Phoenician temple occupying this site during the 6th–1st centuries bce (Riis et al. 1996, 7; Riis 1979). Of the Late Bronze Age and Early Iron Age items found, several of the smaller vessels were identified as having contained liquid and/or food deposits, leading the excavators to conclude, ‘it seems indisputable that the pottery deposits served a cultic purpose’ (Riis et al. 1996, 14). Both these examples of ritual behaviour would indicate that the evidence put forward in favour of ancestral veneration in G11SW should be evaluated within the context of a wider tradition of cultic activity at the site. Potential candidates for vessels used in libation rituals at Tell Sūkās include the numerous Late Bronze Age and Early Iron Age examples of Cypriote White Slip II ware bowls found in the immediate vicinity of G11SW (e.g., Lund 1986, 17, 28) and at the Mīna Sūkās sanctuary (Fig. 6). It should be noted, however, that the specific typology of vessels employed in this context would have been subject to broader trends of ceramic consumption, reflecting changes over time in the value assigned to different wares, meaning that no specific style of vessel is diagnostic of venerative behaviour. Rather, considerations of function as opposed to decoration are germane to the identification of such activities, yet have seldom been explored, and are essential for a more holistic assessment of the degree of continuity in patterns of consumption across the Bronze-to-Iron Age divide. ancestral veneration From the wider region there are numerous other broadly contemporaneous examples of ritualistic practices, which lend apparent support to the proposed identification of venerative activity. At Ras Shamra, which owing to its relative cultural homogeneity with Tell Sūkās potentially offers us the most reliable source of parallels with respect to ideational beliefs concerning the deceased, many of the ordinary houses, in addition to the royal palaces, incorporated corbel-vaulted tombs beneath their floors that were designed to accommodate multiple internments (Salles 1995). We understand something of the cultic practices associated with ‘royal’ internments from Ugaritic text RS 34.126 which details the rites performed Fig. 6. Cypriote White Slip II ware bowl from Mīna Sūkās. Rim diameter c. 18cm (Riis et al. 1996, 58). (Drawing by K. Jacobsen and A. Nilsson.) a n c e s t ral veneratio n at tell sūk ās 147 to commemorate the accession of Ammurapi III (the last king of Ugarit) c. 1200 bce, including the invocation of the cult of the dead in honour of his late father Niqmaddu III. In this liturgy the ancestors of the king are called upon to participate, offerings are presented, and blessings requested as a form of lineage legitimation for the new monarch (Levine and de Tarragon 1984). Although this ‘royal’ ritual would undoubtedly have differed in many respects from that associated with the burial of less eminent members of society, the physical similarity of all the intramural tombs excavated at Ras Shamra would seem to suggest a degree of social accord with regard to mortuary practices. While there is as yet no scholarly consensus as to whether this veneration of ancestors at Ugarit amounted to their worship as ‘household gods’, there is general agreement that the deceased continued to play an important and active role in society (Lewis 1989). In terms of specific evidence for the invocation of ancestor cults above tomb sites, such as that proposed for Tell Sūkās, Schaeffer (1939, 46–56) interpreted the presence of pits with associated ceramic pipes and guttering in and around the tombs at Minet el-Beida and Ras Shamra as representing the functional paraphernalia of libation rituals. While this interpretation has been critiqued by Pitard (1994) upon the basis that these features are perhaps more plausibly to be identified as utilitarian drainage systems, the conspicuous access arrangements for these tombs leaves open the possibility that some form of ancestral cult was practised in this manner (Schloen 2001, 346). In the mid third–early second millennium bce votive pilgrimages to the tombs of dead kings by their newly married successors are documented at Ebla (Archi 1988), and the kispum ceremony, the sharing of meals between the living and the dead (Tsukimoto 1985), is known from Mari (Durand and Guichard 1997, 63–70). Contemporary archaeological evidence for elite ancestor veneration as a means of legitimising social hierarchy has been documented at Umm el-Marra (Schwartz 2007). On the Levantine coast of Cyprus ritual feasting with the deceased has been proposed for Late Bronze Age Tomb 66 at Enkomi (Crewe forthcoming). The kispu ritual as funerary repast is also depicted on numerous Syro-Hittite funerary monuments of the early first millennium bce (Bonatz 2000). Excavation of the monumental Early Bronze Age tomb 302 at Tell Jerablus Tahtani near Carchemish has provided definitive evidence for tomb visitations, and concurrent ritualistic deposition, directly over the site of earlier internments. Such evidence is found in the form of discrete concentrations of objects that are markedly distinct from the cultural material associated with the underlying mortuary assemblage, and represents a form of ancestral commemoration discernible by precisely the disproportionate agglomeration of artefacts that characterises the Sūkās record. Although the exact nature of this venerative behaviour is unclear, the sheer architectural prominence of the tomb within its landscape context (outside the fortification wall of the tell facing the Euphrates) would seem to speak strongly of a desire, ‘to establish the status of the donors by reference to the power of the dead and illustrious ancestors’ (Peltenburg 1999, 433). At Qatna, the re-excavation of the Royal Palace and discovery of its ‘royal tombs’ has provided the most explicit archaeological example of ancestral veneration (c. 1340 bce) discovered to date (al-Maqdissi et al. 2003). Amongst the many spectacular finds recorded, including a golden hand that may have been part of a libation ‘arm’, the south and west chambers incorporated benches arranged along their walls. While some of these were bare, others shelved numerous pottery bowls and storage vessels. The excavators identified these as having contained food, thereby demonstrating the literal and symbolic sharing of kispu with ancestors. To quote Pfälzner (2006, 22), ‘Many generations were thus brought together in tombs, dining together, with the living sitting on their benches, and the dead lying on their biers and their sarcophagi’. It is proposed that the disproportionately large quantity of Late Bronze Age and Early Iron Age pottery fragments found above the collective grave site 148 pales tine ex ploration quarte rl y, 141, 2 , 2 009 at Tell Sūkās could represent the remains of vessels that once held similar offerings for the dead. The stratigraphical evidence from G11SW would further suggest a continuity of this practice across the Bronze-to-Iron Age divide. discussion When considering the reasons for the divergent fate of Tell Sūkās when compared to that of other coastal emporia in the region, particularly Ras Shamra which was by contrast abandoned and not immediately resettled following the disruptions of c. 1180 bce,9 a possible explanation may lie in the different political positions that the two settlements held at the close of the Late Bronze Age. From epigraphic sources Tell Sūkās is generally identified with ancient Souksi, which according to Akkadian text PRU 4 (lines 71–76) was part of the kingdom of Siyannu-Ušhnatu located to the south of Ugarit (Riis et al. 2004, 78–79; Astour 1979). The text itself dates from the second half of the 14th century bce, and is in the form of a treaty between king Niqmepa of Ugarit and king Abdicanati of Siyannu-Ušhnatu. Prior to this document the latter had been a vassal of Niqmepa of Ugarit. Abdicanati was evidently not content with this situation, and following an appeal to the Hittite king, instead became a direct dependency of the Hittite vice-king in Carchemish, with the resulting treaty detailing the territorial possessions of each polity (Singer 1999, 662–666). Concerning the respective experiences of the two kingdoms in terms of the events of c. 1200 bce, Liverani (1995, 51) has observed, ‘on a l’impression que le sort de la moitié nord du royaume (c’est-à-dire l’Ougarit au sens strict) et le sort de la moitié sud (c’est-à-dire le Siyannu-Ushnatu) ont été différents’. He goes on to suggest a possible political explanation for this situation stating, ‘Est-ce que Siyannu a survécu à l’invasion grâce à sa dépendance politique de Karkémish?’. Other possible reasons for the renewal and continued survival of Tell Sūkās on into the Iron Age following its possible sacking c. 1170 bce include preferential trading links (Bell 2005), and its comparatively minor status when compared with Ugarit and many other settlements in the region which may have diverted the more sustained attention of would be aggressors. Elsewhere in the Gabla plain the very limited extent of excavation precludes a detailed assessment of the relative degree of continuity in patterns of settlement and societal development at this time. At Tell Siyannu, the probable capital of the kingdom of Siyannu-Ušhnatu, preliminary investigations have revealed limited archaeological evidence of the Late Bronze Age town (Bounni and al-Maqdissi 1998). Evidence of Late Bronze Age and Early Iron Age occupation has also been recorded in soundings at Tell Darūk (ancient Ušnatu?) and Arab al-Mulk (Oldenburg and Rohweder 1981). Discontinuous Late Bronze Age settlement has similarly been documented for Tell Iris (al-Maqdissi and Souleiman 2004). On-going excavations at Tell Tweini (ancient Ugaritic Gibala) have revealed a continuous sequence of Late Bronze-to-Early Iron Age occupation (Bretscheider et al. 2004; 2008). In more general terms the continued use of toponyms from the Late Bronze Age to the present day also attests to a degree of continuity in the lineage of the region’s population. conclusions A review of the excavation data from Tell Sūkās suggests the possibility of a continuous tradition of ancestral veneration, of potentially up to 750 years in duration, from the close of the Middle Bronze Age through to the Early Iron Age.10 The justification for this conclusion comes primarily from the proposed temporal continuity in function above the collective grave site in G11SW. The identification of this feature offers further support to the chief excavator’s conclusion that, ‘The general character of the Iron Age finds betrays a local a n c e s t ral veneratio n at tell sūk ās 149 Syro-Phoenician culture’ (Riis 1970, 126). As such, it would seem likely that while the disruptions of c. 1170 bce were significant in that they may have provided the impetus behind the widespread renewal and innovation in architectural features at the beginning of the Early Iron Age, it is highly unlikely that this juncture marked the permanent displacement of one population group by another, be they Sea-Peoples (Yon 1992), Arameans (Lawson Younger 2007), or otherwise. The evidence for continuity in mortuary behaviour at Tell Sūkās discussed above contrasts sharply with the pronounced divergence known in other aspects of religious behaviour that emerge at the end of the second millennium bce. In particular, changes at this juncture in the pantheon(s) of the Phoenician coastal centres of the Levant show a significant fragmentation of ideational practice, with the increasing prominence of tutelary city-gods (including Melqart) at the expense of older pan-Canaanite deities (Xella 1986). That the proposed tradition of veneration at Tell Sūkās was able to transcend this wider context of social and ideological dislocation, exemplified by the abandonment of nearby Ras Shamra, arguably speaks of the continuing power of association between the ancestors of Tell Sūkās and their Early Iron Age descendants. acknowledgements I would like to thank the numerous reviewers who suggested many corrections and improvements to this study. I also wish to express my gratitude to John Lund for his kind words of encouragement and invaluable comments on an earlier draft. The responsibility for any errors, and the opinions expressed, remain solely my own. Permission to reproduce figures from the excavation was granted by Peder Mortensen (Chairman, Carlsberg Expeditions to Syria). notes 1 The excavations at Tell Sūkās have been published 6 With the notable exception of Minet el-Beida thus far in ten final volumes; Riis 1970; Ploug 1973; Riis (ancient Mahadu), epigraphic evidence suggests only and Thrane 1974; Thrane 1978; Alexandersen 1978; a very limited ‘foreign’ mercantile population associated Riis 1979; Buhl 1983; Lund 1986; Oldenburg 1991; Riis with the primary urban centres. With particular et al. 1996. reference to Ras Shamra see Yannai (1983, 209–10). 2 The interpretation of charcoal and/or ash lens 7 For the 6th century bce, Riis (1982, 259) has deposits as evidence for violent events of epoch-making assessed that Eastern Greek sherds may constitute significance has played a prominent role in the study of less than 10% of the total ceramic finds from Tell the Syrian littoral and adjacent regions during the late second millennium bce. These ‘destruction’ levels have Sūkās. 8 Thrane (1978, 49) suggests that the tomb may have variously been interpreted as evidence for conflagration resulting from natural disasters (e.g., Schaeffer 1948, been sealed at the end of the Middle Bronze Age 560), or as having occurred as the result of human (c. 1600 bce) for practical reasons in order to combat the odour of decomposition coming from the grave. agency (e.g., Liverani 1987, 69). 3 An example of a potential sampling bias is a slight While a change in cultic practice cannot be ruled out, difference between Sūkās I (Riis 1970) and VIII (Lund it may also have simply become impractical to continue 1986) in so far as Riis at times referred to finds made performing mortuary visitations in such a confined in the upper strata (i.e. out of context) in his period space. See ibid. for discussion of secondary treatment. chapters. Lund generally refrained from doing so, but 9 A temporary ‘squatter’ re-occupation at Ras listed more finds without direct chronological relevance Shamra subsequent to its abandonment c. 1180 bce has (John Lund, personal communication 2004). been proposed by Yon (1992, 118–119), although this 4 A minority of the finds deemed to be of particular interpretation has been convincingly critiqued by significance were given point-specific coordinates that Schloen (2001, 340). Much better evidence for ‘intruallow for their exact positioning within the excavation sive’ re-settlement exists at nearby Ras Ibn Hani, where grid. These items are however insufficient in number to immediately following the destruction of the site locally allow for the identification of generalised trends based upon their spatial distribution when viewed in isolation, made LCIIIC ware appears. For discussion see Lagarce and have accordingly been included in the wider and Lagarce (1988) and Caubet (1992). 10 A comparable longevity of cultic practice in periodic samples using the reduced common definition excess of five hundred years has been recorded at the of 5 m2. 5 A contemporary discontinuity in settlement has monumental necromantic installation in late thirdbeen noted at Tell Tweini (Bretscheider et al. 2008, early second millennium bce Urkesh (Buccellati and Kelly-Buccellati 2004, 20-29). 39). 150 pales tine ex ploration quarte rl y, 141, 2 , 2 009 bibliography Alexandersen, V., 1978, Sūkās V: A Study of Teeth and Jaws from a Middle Bronze Age Collective Grave on Tall Sūkās. Publications of the Carlsberg Expedition to Phoenicia 6 (Copenhagen). Archi, A., 1988, ‘Cult of the ancestors and the tutelary god at Ebla’, in Y. Arbeitmen (ed.), Fucus: A Semitic/Afrasian Gathering in Remembrance of Albert Ehrman. Current Issues in Linguistic Theory 58 (Amsterdam), 103–112. Astour, M.C., 1979, ‘The kingdom of Siyannu-Ushantu’, Ugarit Forschungen 11, 11–28. Bell, C., 2005, ‘Intra-regional variation in long distance trading relationships on the northern Levantine coast — the key to site survival?’, in C. Briault, J. Green, A. Kaldelis and A. Stellatou (eds.), SOMA 2003. BAR International Series 1391 (Oxford), 9–14. Bonatz, D., 2000, ‘Syro-Hittite funerary monuments: a phenomenon of tradition or innovation?’, in G. Bunnens (ed.), Essays on Syria in the Iron Age. Ancient Near Eastern Supplement 7 (Louvain), 189–210. Bounni, A. and M. al-Maqdissi., 1998, ‘Compte rendu de la cinquième campagne de fouilles à Tell Sianu su la côte Syrienne’, Altorientalische Forschungen 25(2), 254–64. Bretscheider, J., M. al-Maqdissi, K. Vansteenhuyse, J. Driessen and K. Van Lerberghe., 2004, ‘Tell Tweini, ancient Gabala, in the Bronze Age’, Ägypten und Levante 14, 215–30. Bretschneider J., K. Van Lerberghe, K. Vansteenhuyse and M. al-Maqdissi, 2008. ‘The Late Bronze and Iron Age in the Jebleh region: A view from Tell Tweini’, in H. Kühne, R. M. Czichon and F. J. Kreppner (eds.), Proceedings of the 4th International Congress on Archaeology of the Ancient Near East (Wiesbaden), 33–46. Buccellati, G. and M. Kelly-Buccellati., 2004, ‘Der monumentale Palasthof von Tall Mozan/Urkeš und die stratigraphische Geschichte des ābi’, Mitteilungen der Deutschen Orient-Gesellschaft 136, 13–39. Buhl, M-L., 1983, Sūkās VII: The Near Eastern Pottery and Objects of Other Materials from the Upper Strata. Publications of the Carlsberg Expedition to Phoenicia 9 (Copenhagen). Caubet, A., 1992, ‘Reoccupation of the Syrian coast after the destruction of the “Crisis Years”’, in W. A. Ward and M. S. Joukowsky (eds.), The Crisis Years: the 12th Century B.C. — From Beyond to the Danube to the Tigris (Dubuque), 123–131. Crewe, L., (forthcoming), ‘Feasting with the dead. Tomb 66 at Enkomi’, in T. Kiely (ed.), 2009, Cyprus in the British Museum: Papers in Honour of Veronica Tatton-Brown (London). Durand, J-M., and M. Guichard., 1997, ‘Les rituels de Mari’, in D. Charpin and J-M. Durand (eds.), Florilegium Marianum III: Recueil d’etudes à la memoire de Marie-Therese Barrelet. Memoires de NABU 4 (Paris), 19–78. Iacovou, M., 1988, The Pictorial Pottery of Eleventh Century B.C. Cyprus. Studies in Mediterranean Archaeology and Literature 78 (Jonsered). Lagarce, J. and E. Lagarce., 1988, ‘The intrusion of the Sea People and their acculturation: A parallel between Palestinian and Ras Ibn Hani data’, in S. Shaath (ed.), Proceedings of the First International Symposium on Palestinian Antiquities held at Palestine Archaeological Centre, Aleppo, 1981. Studies in the History and Archaeology of Palestine 3 (Aleppo), 137–169. Lawson Younger Jnr, K., 2007, ‘The Late Bronze Age / Iron Age transition and the origins of the Arameans’, in K. Lawson Younger Jnr. (ed.), Ugarit at Seventy Five: Proceedings of the International Symposium held at Trinity International University, Illinois, 2005. (Winona Lake), 131–174. Levine, B.A., and J-M. de Tarragon., 1984, ‘Dead kings and Rephaim: the patrons of the Ugaritic Dynasty’, Journal of the American Oriental Society 104, 649–59. Lewis, T. J., 1989, Cults of the Dead in Ancient Israel and Ugarit. Harvard Semitic Monographs 39 (Atlanta). Liverani, M., 1987, ‘The collapse of the Near Eastern regional system at the end of the Bronze Age: the case of Syria’, in M. J. Rowlands, M. Larsen and K. Kristiansen (eds.), Centre and Periphery in the Ancient World (Cambridge), 66–73. Liverani, M., 1995, ‘Le royaume d’Ougarit’, in M. Yon, M, Sznycer and P. Bordreuil (eds.), Le pays d’Ougarit autour de 1200 av. J.-C.: Acts of the International Colloquium held Paris, 1993. Ras Shamra-Ougarit 11 (Paris), 47–54. Lund, J., 1986, Sūkās VIII: The Habitation Quarters. Publications of the Carlsberg Expedition to Phoenicia 10 (Copenhagen). al-Maqdissi, M., H. Dohmann-Pfälzner, P. Pfälzner and A. Suleiman., 2003, ‘Das königliche Hypogäum von Qatna’, Mitteilungen der Deutschen Orient-Gesellschaft 135, 189–218. al-Maqdissi, M. and A. Souleiman., 2004, ‘Tell Iris’, in G. Galliano and Y. Calvert (eds.), Le royaume d’Ougarit. Aux origines de l’alphabet (Lyon), 64. Oldenburg, E. and J. Rohweder., 1981, The Excavations at Tall Darūk and Arab al-Mulk. Publications of the Carlsberg Expedition to Phoenicia 8 (Copenhagen). Oldenburg, E., 1991. Sūkās IX: The Chalcolithic and Early Bronze Age Periods. Publications of the Carlsberg Expedition to Phoenicia 11 (Copenhagen). Oren, E., 1992, ‘Palaces and patrician houses in the Middle and Late Bronze Ages’, in A. Kempinski, R. Reich, H. Katzenstein and J. Aviram (eds.) The Architecture of Ancient Israel (Jerusalem), 105–120. Peltenburg, E.J., 1999, ‘The living and the ancestors: Early Bronze Age mortuary practices at Jerablus Tahtani’, in G. del Olmo Lete and J. L. M. Fenollós (eds.), Archaeology of the Upper Syrian Euphrates: The Tishrin Dam Area. Aula Orientalis-Supplementa 15 (Barcelona), 427–442. Pfälzner, P., 2006, ‘Syria’s royal tombs uncovered’, Current World Archaeology 15, 12–22. Pitard, W.T., 1994, ‘The “libation installations” of the tombs at Ugarit’, Biblical Archaeologist 57, 20–37. Ploug, G., 1973, Sūkās II: The Aegean, Corinthian and Eastern Greek Pottery and Terracottas. Publications of the Carlsberg Expedition to Phoenicia 2 (Copenhagen). a n c e s t ral veneratio n at tell sūk ās 151 Riis, P. J., 1970, Sūkās I: The North-East Sanctuary and the First Settling of Greeks in Syria and Palestine. Publications of the Carlsberg Expedition to Phoenicia 1 (Copenhagen). Riis, P. J. and H. Thrane., 1974, Sūkās III: The Neolithic Periods. Publications of the Carlsberg Expedition to Phoenicia 3 (Copenhagen). Riis, P. J., 1979, Sūkās VI: The Graeco-Phoenician Cemetery and Sanctuary at the Southern Harbour. Publications of the Carlsberg Expedition to Phoenicia 7 (Copenhagen). Riis, P. J., 1982, ‘Die Griechen in Phönizien’, in H.G. Niemeyer (ed.), Phönizier im Westen: Proceedings of the international symposium ‘Die phönizische Expansion im westlichen Mittelmeerraum’ held Cologne, 1979. Madrider Beiträge 8 (Mainz am Rhein), 256–260. Riis, P. J., J. Jensen, M-L. Buhl and B. Otzen., 1996, Sūkās X: The Bronze and Early Iron Age Remains at the Southern Harbour. Publications of the Carlsberg Expedition to Phoenicia 12 (Copenhagen). Riis, P. J., I. Thuesen, J. Lund and T. Riis., 2004, Topographical Studies in the Gabla Plain. Publications of the Carlsberg Expedition to Phoenicia 13 (Copenhagen). Salles, J–F., 1995, ‘Rituel mortuaire et rituel social à Ras Shamra/Ougarit’, in S. Campbell and A. Green (eds.), The Archaeology of Death in the Ancient Near East. Oxbow Monograph 51 (Oxford), 171–184. Schaeffer, C. F-A., 1939, The Cuneiform Texts of Ras Shamra-Ugarit. The Schweich Lectures (London). Schaeffer, C. F-A., 1948, Stratigraphie Comparée et Chronologie de l’Asie Occidentale (London). Schloen, J. D., 2001, The House of the Father as Fact and Symbol: Patrimonialism in Ugarit and the Ancient Near East. Studies in the Archaeology and History of the Levant 2 (Winona Lake). Schwartz, G. M., 2007, ‘Status, ideology and memory in Third Millennium Syria: “Royal” tombs at Umm el-Marra’, in N. Laneri (ed.), Performing Death: Social Analyses of Funerary Traditions in the Ancient Near East and Mediterranean. (Chicago), 39–68. Singer, I., 1999, ‘A political history of Ugarit’, in W. Watson and N. Wyatt (eds.), Handbook of Ugaritic Studies (Leiden), 603–733. Thrane, H., 1978, Sūkās IV: A Middle Bronze Age Collective Grave on Tall Sūkās. Publications of the Carlsberg Expedition to Phoenicia 5 (Copenhagen). Tsukimoto, A., 1985, Untersuchungen zur Totenpflege (kispum) im alten Mesopotamien. Alter Orient und Altes Testament 216 (Kevelaer). van Wijngaarden, G. J., 2002, Use and Appreciation of Mycenaean Pottery in the Levant, Cyprus and Italy c. 1600–1200 B.C. (Amsterdam). Yannai, A., 1983, ‘Studies on trade between the Levant and the Aegean in the 14th to 12th centuries B.C.’, unpublished Ph.D. dissertation (Linacre College Oxford). Yon, M., 1992, ‘The end of the kingdom of Ugarit’, in W. A. Ward and M. S. Joukowsky (eds.), The Crisis Years: the 12th Century B.C. — From Beyond to the Danube to the Tigris (Dubuque), 111–122. Xella, P., 1986, ‘Le polythéisme phénicien’, in C. Bonnet, C. E. Lipinski and P. Marchetti (eds.), Religio Phoenicia. Studia Phoenicia 4 (Namur), 29–39.