LEFKANDI Xeropolis - 2007

Lefkandi. I. S. Lemos (BSA/Oxford) reports on the 2007 excavation season in which both regions I and II were extensively investigated (Figs. 1, 2).
Region 1 area M. The EIA ‘megaron’ and its LH IIIC predecessor were further investigated (Fig. 3). The flooring/packing and other lower soils were removed down to the latest LH IIIC floors. The EIA building had only one floor surface − mostly mud-bricks or earth mixes: some additional postholes of this phase were discovered. The construction sequence of the building was visible at the S end of the W wall 65: the foundation of the main wall was one course deeper than that of the porch (which presumably weighed less). Slightly later, a further angled wall was added at the SW corner, built over a pebble spread that had accumulated there since LH IIIC. It is possible that a yard was added to the W side of the EIA ‘megaron’ (defined by walls 120 and 134), perhaps at the same time. More investigation is needed to clarify its date and function.

The LBA building is also long and rectangular in plan (at least 12m N−S x 5.5m E−W). At the S end, it was divided into 2 small compartments, of which the E one served as its entrance. A central room probably formed the main part of the building: there may have been another room at the N end, but hill erosion and the erection of a LGeo apsidal house have done much damage. The building was probably entered via a threshold placed just E of the central median line. There are slight indications of some sort of porch or walls in antis at the E end: postholes might indicate something similar at the W end too, but later walls cover much of the relevant area. The SE room is defined at the W by wall 153 and at the N by the E/W cross wall 149; the room had a further access through a doorway at the NE. Otherwise the SE room was completely disturbed and destroyed by the LGeo pit 13. Another small room immediately W of the SE room, the SW room (ca. 2.5m²), has flat-laid mud-bricks in the E half which might originally have stood up to 0.2m above the floor. If so, they could have formed a low platform (2.5m l. x 0.8m w.). This could have been used as support for furniture. The main room (7.5 x 5.5m) was remarkably empty; one median posthole and evidence of mud-brick furniture to the NW have been discovered so far. At the N end of the room, and in particular the E part of this area, a stone construction is visible running E−W. It has a roughly flat and uneven surface, sits on a few centimetres of soil above the latest floors of the room and is covered by a thick yellow clay spread that runs on beyond it and under the floors of the later EIA building. It is hard to decide whether the structure belongs to the LBA or EIA building, but considering the manner in which the latest LH IIIC floor to the N runs off the edge of the stone structure (because of the presence of even earlier walls below), it was likely built sometime during the end of the use of the LH IIIC structure. Its full extent to the W is masked by the block of soil left to support the ‘fire-box’ (see below). Finally, at the N limits of the building, the existence of another room is indicated by the E−W cross wall 88. Nothing more can be known here because of severe hill erosion at this end.

As indicated, the LH IIIC building contained only minimal internal furniture: some small bins of mud-brick against wall 78/79 and a few small postholes. The latest floor was essentially bare, but a considerable number of earlier floors were found in the NW of the main room (the only area exposed this year). Seven ‘pairs’ of floors have been excavated; these are made of red/orange mud-brick mix or yellow clay with an overlying layer of grey soil containing spots of carbon. Careful continuation of work here will be vital, as will a test into the underlying earlier LH IIIC phases.

In several places, LH IIIC walls were reused by the EIA builders (e.g., walls 77 and 88), although elsewhere earlier constructions were completely ignored. In some places the mud-brick floors of the EIA ‘megaron’ rest only a few centimetres above the floors of its predecessor, yet elsewhere, some 0.2−0.3m of an intervening one-phase fill may exist. These observations suggest that the major LH IIIC walls were reduced to their stone socles before the construction of the EIA building began, and that the interior space too was emptied by the later builders.

Within this period of transition, a ‘box’ made of mud-bricks (0.7m E−W x 0.4m N−S) was constructed at the N end and more or less on the medial axis of both the earlier and later buildings. Within the ‘box’ and spilling out from it was a very fine grey to black soil (of the sort often called ‘ashy’ and associated with decayed or burnt vegetable/organic matter). About half way down was part of a coarse pot (other parts of which were just outside the ‘box’) and some 6 murex shells that had been subjected to considerable heat. The whole find resembles a type of kiln. Further investigation might clarify the purpose of the ‘box’ and its date. A quantity of the ‘grey ash’ fill was intermixed with soils found over the LH IIIC floors, and part of the EIA floor only just covered the ashy spreads. Both these observations imply that the ‘box’ belongs more to the later building (the EIA ‘megaron’). The suggestion that it might be linked to possible ceremonies related to the construction of the ‘megaron’ requires further research.

Both the LH IIIC building and its EIA successor were oriented on a compass grid and are thus distinctively different from earlier LH IIIC structures (one of them visible only 0.005−0.10m below the floors). Moreover, the LH IIIC building, again like the later one, has a long, rectangular plan different from the known LH IIIC houses on Xeropolis. It is essential to establish that it was indeed built after Lefkandi phase 2b, which would set it in stark contrast to earlier but also contemporary complexes (such as that found in the nearby area P and in earlier excavations). Such important differences in orientation and plan have major implications for understanding the function of these buildings within the spatial organization of the settlement and for reconstructing the social order on Xeropolis during the LBA−EIA transition.

Area P. The multi-phased complex is now confirmed as being used from LH IIIC−LGeo and Ar. The LH IIIC structure revealed in 2006 belongs to Lefkandi phase 3. This discovery will probably clarify the report of the so-called S house excavated by Popham and Sackett (D. Evely ed., Lefkandi IV: The Bronze Age − the Late Helladic IIIC Settlement at Xeropolis [London, BSA, 2006], 41−42), of which our complex might even be part. It is also now clear that the building extended to the W, E and S, and includes part of the N side of area P excavated in 2004 and 2006. The complex is thus of considerable size. Providing that the preliminary dating of the structure is correct, it is interesting that its construction and size are not inferior to those of houses dated to phase 2 as defined by Popham and Sackett. Following a collapse of this structure, indicated by an infill of mud-bricks, the area was again occupied: new walls were constructed and old ones reused (for example wall 94 was partly reused, while wall 130/112 was most probably utilized as an internal division). Wall 141 to the S is also new. Pottery found in these levels could belong to the LBA−EIA transitional phases. During the PGeo period (roughly 10th Ct) the excavated part of the structure is ca. 5m w. and extends to both the W and E, but also outside the area investigated so far. N of the structure, an exterior space (perhaps a yard) laid with pebbles has produced a considerable amount of butchered animal bone: the area was probably used for butchery (Fig. 4).

Architectural details of the EIA phase of the complex were clarified this year. Access from the E was through a short corridor, over a pebbled surface and a stone threshold slab at its NE corner. The corridor leads into a small room (5m x 4m). Admission to a room beyond this to the W was through wall 133 where an entrance at the SW was marked by another stone threshold. This W unit has a further pair of hollows or shallow pits: it is not yet clear whether they were intended for storage or were used for timber roof supports. Either might suggest an apsidal end in this part of the building but such a reconstruction remains tentative. Preliminary study of the pottery suggests that the EIA structure had at least 2 phases of occupation. The earliest is marked by burnt debris in the E room, above which good quality pebble-laid floors were found. Both phases date within the PGeo and Sub-PGeo periods.

Over this area, partly running above the S wall 141, but orientated slightly differently (more W−N/E−S) is what appears to be part of the S external wall and the beginning of the apse of another structure. Little is preserved to go with this, but its construction differs from other walls and there is some evidence that it belongs to a LGeo II/Ar structure.

In region II areas S and R, excavation focused on the area around the ‘wall’ revealed in 2006 (Fig. 5). A test trench (2m w. x 15m l.) running N−S of the E baulk in area R was opened to investigate the zones N and S of the ‘wall’ (consisting of walls 118/152 and 121). In area S (to the E) the ‘wall’ was traced for a further 20m; it runs in a straight line (allowing for local slippage to the N) and while it is eroded in places, elsewhere it stands high. At its E end, a stone complex projecting to the N could be part of a bastion, although more work is required to understand the function of the structure and its relation to the ‘wall’. Similar projections have been noted in connection with the ‘wall-line’ to the S. The sequences of soils S of the ‘wall’ (walls 118/152) and at the E end of area S, copy those seen in area R.

In area R, excavation proceeded in the ‘hollow’ to the N of the ‘wall’ discovered in 2006 (wall 118), reaching LH IIIC soils. The exact interface between EIA and LBA levels was not identifiable everywhere: digging generally stopped at LH IIIC levels. More LBA and EIA figurines were recovered − mostly bulls, but with an almost complete PGeo horse figurine. A large amount of good quality pottery was found in LH IIIC levels; the most popular shape is the kylix.

Excavation to the S of the ‘wall’ in area R (wall 118) revealed a series of hard-packed surfaces of good quality datable to LH IIIC. Contrary to previous reconstructions, potential multiple phasing has been identified with the foundation trench itself (over 1m d.) reaching down to MH levels. To the S and within the LH IIIC hard surfacing lay wall 161, of good quality but different construction to wall 118. It runs parallel and some 3m S of it. Wall 161 too has various stages of repair/rebuild, as well as a perhaps even earlier version just to its N. Only a small part of it was uncovered this year. To test whether this 2nd ‘wall’ was also to be found in area S, we calculated where it might be in relation to the extension of the ‘wall’ there. Wall 177 was found at the extreme E end of area S in exactly the predicted position. Thus it appears that there were 2 ‘walls’ marking the unusual profile at the N side of Xeropolis in this central region. This supports the earlier suggestion that a major, long-term civil engineering project was undertaken in a part of the settlement which might have been the main approach from the mainland in the N. Further work is needed to clarify whether the 2 ‘walls’ were contemporaneous; clearly, however, the discovery of these 2 ‘walls’ implies a far more complex and impressive defensive system than hitherto suspected.

In area T an area W of the ‘wall’ was opened to investigate the ‘structure’ found at the NW corner of area R in 2006. Excavation exposed its E and S sides (the other sides are yet to be unearthed), notably increasing its size. Its construction technique is unique at Lefkandi and elsewhere. A stone socle was first laid, with care taken primarily in the construction of its interior face, where large rounded cobbles were set in a neat row and smaller ones packed in behind; then the ‘exterior’ face was lined with large rounded cobbles set vertically, with smaller stones placed behind to create an even surface. The result was to build an almost vertical exterior face (up to 0.4m h.). The vertical sherds and stones reported in 2006 appear now to have been set in a 2nd phase of use. Further excavation is required to understand whether this structure was part of a building or some other feature. Its use lies within LH IIIC and/or the beginning of the EIA.

Further W lies another ‘structure’, ca. 4m² (Fig. 6). The walls show at least 2 phases and were built with both small and flat stones and larger ones. The entrance, at the E side, was marked by flat slabs. The fill − of which only the later phase has been excavated − is generally free of mud-brick. If this is generally so, it is possible that the structure was a low wall only and was unroofed. A remarkable feature of this ‘structure’ is the number of circular ‘platforms’ within it, roughly on its central N/S axis. The largest is ca. 2m di., the smallest less than half that size. Another, even smaller ‘platform’, was found against and partly built into the S wall of the ‘structure’. This is a circular kerb of large rounded stones enclosing flat slabs. Over this construction, a layer of smaller pebbles and stones covered most of the circular outline of the ‘platforms’. The platforms’ can be compared to a similar construction in the E room of the Toumba building (M.R. Popham, P.G. Calligas, L.H. Sackett, Lefkandi II.2: The Protogeometric Building at Toumba: the Excavation, Architecture, and Finds [London, BSA, 1993], pl. 7).

The soils within the ‘structure’ and around the ‘platforms’ contain little bone and no domestic debris or ash. Among plentiful sherds outside the ‘structure’ and in the nearby area, the most popular shape is the krater (one example datable to Sub-Myc or EPGeo has a di. of ca. 0.7m). Inside the ‘structure’, a fair number of spindle-whorls and conical buttons were found. An exceptional item is a small anthropoid steatite pendant with incised decoration, similar to those found in 2006 and in the 1960s. Most of the pottery from the ‘structure’ and the surrounding area appears to belong to the EPGeo (ca. end 11th−E10th Ct). Between the area of the earlier ‘structure’ of unusual construction NW of the ‘wall’ and the ‘structure’ with the platforms further W, was a series of ‘enclosures’ dating to several phases (LH IIIC−PGeo). Excavation here remains to be completed