1.1. Conformation and function

2With a diameter of c. 10 metres, Structure 10 (St 10) was by far the largest excavated structure found on the site. It was vertically cut into the light-orange Pleistocene carbonated colluvium (SU 1002, east half) and the havara (SU 1007, west half; fig. 5-3, A; Pantazis 1967, Schrimer 1998). The upper limit of the structure outlined an almost perfect circle, with only two irregularities: a 10-20 cm slight rise to the east and a much more consequential one, made up of brown-grey sediment, to the northeast (fig. 5-3, B), which turned out to be an entrance after further excavations. The fill was hollowed out by abundant post-PPNA pits (chap. 4). Pit depth of St 10 decreased from 1 m in the north to several centimetres at its southern limits, in relation to its visible upper limit (fig. 5-3, C). As the pit and its fills were obliquely truncated by the post-PPNA erosion of the slope, these measurements under evaluate the original depth. In particular, the south limit now visible at ground level was initially cut into the havara, just like the north one, although the northern part of the structure was undoubtedly deeper than the southern part.

Fig. 5-3 – General views of Communal building St 10: A, towards the north, first stripping of the southeast and southwest quadrants in 2011; B, towards the south, last clearing of the whole structure in 2012 (in the foreground, the north trench showing the stones of the Sotira house wall); C, towards the west, before the excavation of the north third.

3Due to the fact that excavations were forced to stop at the end of the 2012 season, it has not been possible to comprehensively study the northern area of the original building. In the other areas (fig. 5-4), the floor was flat and horizontal around the central post hole and in the northeast quadrant, down to the entrance. But in the two southern quadrants and in the southern part of the northwest quadrant, it gradually rose in a centripetal way to c. 70 cm, creating a sort of low-profile symmetrical amphitheatre, on both sides of the axes defined by the entrance. In addition, this elevation appeared to be steeper and narrower to the west than to the south. This organisation showed no congruence with the delimitation of the two different Pleistocene sediments in which the building was dug: havara to the west and carbonated colluvium to the east.

Fig. 5-4 – Topographic record, section and 3D photogrammetric reconstruction of St 10 at the end of the 2011 season, when the north part was not yet excavated; the forced end to the excavation at the end of 2012 made it impossible to record such images for the whole building (3D processing: “osmbundler”, 3D viewing: MeshLab).

4The excavated surface was surrounded by a 30-cm-wide and 30-cm-deep perimeter trench (St 47). This foundation trench was actually only visible in the southwest quadrant (fig. 5-4) and in the north-south and east-west profiles (see below). In the three other quadrants, its presence was only assumed, as it was masked by a thick earthen wall (St 172, 179, 184, 188) pressing against the substratum. For this reason, we interpreted this perimeter trench as the foundation trench of the building’s perimeter earthen wall. On the east and west sides, this foundation trench abruptly widened to 50 cm in squares ZX/40 and G/38 (fig. 5-5; see also fig. 5-20), possibly to reinforce the wall. In the southernmost part, which was the shallowest, the foundation trench was interrupted along almost 2 m. This may result from the combination of slope erosion and bioturbations by plant roots, rather than from the presence of a southern interruption of the wall, i.e., a second entrance. No other evidence – such as a higher density of remarkable deposits (large flint arrowheads, Spondylus cylinder beads, perforated marine shells) or a discontinuity in the perimeter elevation – confirmed the presence of a second entrance.

Fig. 5-5 – Portion of the foundation trench (St 47) situated in the extreme southwest of St 10 (squares ZX-ZY/40-41), still partially filled with the building earth of the perimeter wall: A, sudden widening of the trench, probably corresponding to a thickening of the wall; B, very large flint arrowhead, a remarkable deposit hidden in the base of the wall (St 93); C, oblong pit grouping pairs of post holes, not yet excavated (St 92).

5The eight oblong pits arranged in a semi-circular chain 0.5 to 1 m inside the peripheral trench/wall (fig. 5-5) provided evidence of the use of pairs of posts to strengthen the weight-bearing capacity of the perimeter wall (see below). Together with the perimeter wall and the central post hole(s), these pits clearly indicate the presence of an admittedly light but continuous 31 m² roof.

6St 10 was interpreted as a communal building similar to the ones found in numerous continental PPN villages (Vigne et al. 2012, 2017b) on account of the following characteristics. The building was circular and large, buried, and required huge and necessarily collective efforts to excavate c. 70 m³ and build a roof of more than 31 m². The intentional organisation of the floor suggests a place designed for social interaction. In addition, a high density of remarkable deposits and ritual fire places were unearthed in the building and will be described in detail later in this chapter.

1.2. General organisation of the sediment fills

1.2.2. East-west and north-south profiles

9Under the 0.10-0.15-m-thick arable soil, both the east-west and north-south profiles showed a complex succession of SU (fig. 5-6). Even though they were often interrupted by general erosion or by later pottery Neolithic pits, most of the SU appeared in the form of layers extending to a large part or even across the whole of the St 10 pit. This indicated, contrary to what we proposed in Vigne et al. (2012) on the sole basis of the 2011 excavations, that the successive activities highlighted by these SU developed in the same 10-m-large circle from beginning to end, without any reduction in the surface of the building. Consequently, we assume that this occurred in a sufficiently short lapse of time for the transmission of memories of previous uses, and that these different phases were probably functionally similar.

Fig. 5-6 – East-west (A) and north-south (B) profiles of the Communal building fills (St 10), along lines 36/37 (view toward the north) and B/C (view toward the west) of the grid pattern.

10On the other hand, both profiles also indicated an alternation of two sediment categories, named following Karkanas and Efstratiou (2009): (i) either formal or unformal constructed floors (in French, respectively: chape and (sol d’épandage) composed of thick and well-levelled layers of very clayey, compact and generally homogeneous silts (Mylona et al. 2017, see chap. 9), such as, emblematically, SU 10.6 or 10.2; (ii) embankments (remblais) or stony floors (radiers) composed of accumulations of building earth very rich in stones, flint knapping products or tools, fragments of macro-tools and poorly fragmented animal bones. The latter were overlain by the former and can therefore by interpreted as stony floors serving as substructures (soubassements) for the constructed earthen floors (sols construits). This suggests a succession of several buildings lying on the rubble of the previous ones.

11Altogether, these general observations indicate a succession of renovations of the same building within the same initial 10 m pit, and over a relatively short duration of time.

12In more detail, seven phases can be distinguished in both profiles.

1. PPNA phase 1 (“Building 1”, green in fig. 5-6) was mostly visible in the north-south profile, because the east-west profile was surveyed in 2011 before the exhaustive excavation of the few centimetres covering the Pleistocene substratum (namely SU 10.10). Phase 1 comprised the perimeter wall foundation (SU 10.7, St 47), visible at the two extremities of both profiles, a 5-10‑cm-thick layer of homogeneous pure building earth (formal floor) coating the Pleistocene St 10 base (SU 10.9, 10.10, 10.14) hollowed by the central post hole (St 66, reworked during phase 2) and a complex arrangement in row 35-36, 1.5 m to the north of the St 10 limit (fig. 5-6, B). It was composed of a very large stone (St 200), covered by a thick embankment rising 35 cm above the floor. As the northern limit was removed by the vast Sotira pit (St 190); the north profile of this structure is unknown, and we cannot determine whether it was a bench or a pillar. In this first phase, only the perimeter wall and this bench/pillar structure display the above-described classical superimposition “substructure/constructed floor”. Here, the substratum (SU 1002 and 1007) plays the role of a substructure under the formal earthen floor.
2. PPNA phase 2 (“Building 2”, formerly named “Intermediate building” in Vigne et al. 2012, 2017b and Mylona et al. 2017; orange in fig. 5-6). Conversely, in this phase, the stony substructure was composed of many successive deposits (SU 10.11-13, 10.15-16 and maybe 10.3), all made up of stony building earth (Mylona et al. 2017; chap. 9). This constituted a 20-30-cm-thick layer under another 20-30-cm-thick layer of building earth floor (SU 10.6, 10.28). At both ends of the east-west profile (fig. 5-6, A), the layers obliquely covered strips of building earth (SU 10.8), probably related to the perimeter wall (see below). This succession thickened at the limit of rows 35 and 36 (St 166, fig. 5-6, B), just above the “Building 1” bench/pillar, suggesting the superimposition of a second bench/pillar. At the limit of rows ZZ and A, on the east-west profile, the two general layers (SU 10.13/10.16 and SU 10.6) were interrupted by an 80-cm-wide structure (St 171) lying on a stony base (SU 10.17), suggesting the presence of another pillar. In rows 38 and 39, the phase 2 sediment interrupted the continuity of phase 1 sediments at the level of the central post hole (St 66), indicating the reuse of the latter (fig. 5-6, B). This was confirmed by the presence of four large stones, probably used for wedging, in the profile.
3. PPNA phase 3 (“Building 3”, formerly named “Building 2” in Vigne et al. 2012, 2017b and Mylona et al. 2017; yellow in fig. 5-6). This phase was again composed of two deposits, i.e., a 10-15-cm-thick stony floor (SU 10.5) topped by a 5-20-cm-thick more or less “pure building earth” floor (SU 10.2, 10.4). The west and south limits were absent due to post-PPN (west) and historical (south) slope erosions. The earth here was darker than in phase 2. In the east-west profile (fig. 5-6, A), this succession appeared to be horizontal and regular, leaning to the east against the SU 10.8 earthen formation of phase 2, which was itself connected to the perimeter wall (row G). The only irregularity was a very dense mass of clayey earth at the boundary of rows ZZ and A (St 134), located just above the probable west pillar of phase 2, and maybe its replication in phase 3. The situation was much more complex in the north-south profile (fig. 5-6, B), where this succession was disturbed, from the north to the south: by an admittedly discreet widening (St 153), similar to and superimposed on the benches/pillars described at the limit of rows 35 and 36 for phases 1 and 2. Slightly north of the central post hole position, a sub-quadrangular pillar (St 54) lay on a thin stony floor (SU 10.5), at least 40 cm high and 55 cm wide from north-south. In rows 38 and 39, south of this pillar, the two components of the floor (SU 10.2B and A) thicken to 40-45 cm, forming a depression rather than a horizontal layer, with extensive bioturbation.
4. PPNA phase 4 (not yet identified in Vigne et al. 2012, 2017b; blue in fig. 5-6) was composed of a 5-cm-thick layer of building earth rich in flint flakes (stony floor SU 10.19) topped by a 5-15-cm-thick pure earth layer with limited Sotira contaminations (embankment and floor SU 10.20). These layers were limited to the three rows B, C and D in the east-west profile and to a very small formation in the north-south profile. Micromorphological observations (chap. 9) suggested that they are the only remains of a fourth building, abandoned then washed away by erosion. Contrary to the preceding phase, there is no evidence that phase 4 extended over the entire area of St 10.
5. Sotira pits and colluviums (grey in fig. 5-6). In the north-south profile (St 145, 190) 30- to 80-cm-deep pits perforated the deposits of the four preceding phases, sometimes across a large surface (St 190: 2.25 m north-south). In a sometimes--complex succession of filling phases (St 190), in conjunction with more or less substantial quantities of reworked PPNA flint and -macro-tools, they all provided small quantities of Sotira pottery sherds and caprine or fallow deer bones (chap. 4).
6. Post-Neolithic carbonated colluvium deposits mixing PPNA and Sotira materials (SU 1006, SU 10.1; deep part of SU 1000) in stratigraphic discordance above the sedimentary formations. Several historical cultivation pits were dug into SU 10.1. They deposited over a strong erosion phase of the slope and before the last centuries BC.
7. The uppermost sediment deposits grouped together a series of pits (east-west profile) and colluviums related to historical cultivation activities. Some rare fragments of wheel-thrown potsherds and terracotta objects suggest the Archaic Period (chap. 4).

2.1. The entrance of “Building 1” (phase 1) and the external circulation level

14In the two western square metres of the east trench, we found a portion of earthen wall (St 175) in continuity with the perimeter wall (St 72) remains, forming a 120° angle towards the exterior (fig. 5-7). Almost 3 m further north, the perimeter earthen wall (St 179/185) symmetrically turned outwards at c. 120°. These two outward turns of the perimeter wall suggested a 3-m-wide entrance clearly dating to the first phase of the construction and use of the Communal building. Inside the building, the micromorphological analyses of the formal floor showed marked -structuration due to intense trampling, suggesting the presence of an earthen ramp (Mylona et al. 2017; chap. 9). However, as we could not complete the excavation, we could not verify the existence of another earthen arrangement further outwards, to reduce this surprisingly wide entrance.

Fig. 5-7 – Nadiral view of the north-eastern half of St 10 and the entrance zone (A) and its interpretative schema (B; same colour codes as for fig. 5-6). The scale is 1 m long and oriented north-south.

15The east-west profile of the phase 1 entrance system was also studied in the profile of the East trench (appendix4-3), in continuity with the east termination of the east-west St 10 profile of fig. 5-6 A. It did not reveal any additional earthen features potentially related to this first entrance system (fig. 5-8), with the possible exception of the 40-50-cm-thick and apparently rectilinear earthen wall (St 144), possibly part of another unexcavated building, to the east. This wall was located 1 m outside the Communal building, directly on the Pleistocene colluvium (SU 1002).

Fig. 5-8 – Detail of the east-west profile of the St 10 entrance area: relationships between the St 10 fills and the sediment accumulation between its northeast limit and the St 144 earthen wall.

16Whatever the function of this wall and its relationship with the entrance system, it is interesting to note that the products of its deterioration lying against its west face (SU 2023 and 2024, fig. 5-8) covered a thin basal horizontal clayey silt layer (SU 2005), also lying directly on the substratum. Micromorphological analyses of this SU revealed trampling marks resulting from frequently used surfaces in alternating moist and dry weather conditions, without any indicator of external conditions (i.e., runoff features). This points to a semi-protected/roofed space (Mylona et al. 2017; chap. 9). This level was overlain by several other similar levels (SU 2004 and SU 2003: with only some slight traces of trampling in the lower part of SU 2004), and appeared to be the earliest known level of human circulation around the Communal building, in the entrance area. It was 40-50 cm higher than the “Building 1” floor, indicating the presence of either two to three steps or a surface sloping downwards from the exterior to the floor of the building. SU 10.46 (fig. 5.7) may possibly constitute the remains of this structure, also connected to building 2 where the only two dog mandibles from the site were found (chap. 27).

2.2. The entrance of “Buildings 2 and 3” (phaseS 2-3)

17Between phases 1 and 2, the perimeter wall was reorganised on both edges of the entrance. To the southeast, after a demolition phase which only left 30-40 cm of the initial wall (St 72) height, a new wall (St 117) was built on this level and on a small external step resulting from a slight widening of the initial digging of St 10 (fig. 5-9). This formed the slight bulging of the circle mentioned at the beginning of this chapter (1.1.). At the northwest edge of the primitive entrance, the initial earthen wall was also partially levelled out, before the phase 2 wall (here St 185) was superimposed on the outward turning termination of the former (here St 179; fig. 5-10).

Fig. 5-9 – Sub-nadiral view of the Communal building perimeter wall, south of the entrance. After partial demolition of the initial wall (St 72), another slightly more peripheral wall (St 117) was superimposed on the former and prolonged towards the north, contributing to the reduction of the entrance width. The scale is 1 m long and its left extremity is oriented to the north.

Fig. 5-10 – Base of the north wall of the Communal building entrance where the partly levelled out original wall (St 179) was overlain by the north wall of the entrance of “Buildings 2-3’s” (St 185), containing abundant limestone blocks and flint flakes. The stony base of a probable earth pillar/post wedging (St 178) leant against these walls (right).

18Against this wall, at the very place where it turned outwards, we found a more or less circular 25-cm-wide accumulation of large stones (over 10 cm long) packed in the building earth (fig. 5-10; see also fig. 5-7), which probably indicates a pillar based on the stony floor or a post wedging (or both: post-framed pillar). The upper part had been eroded. It was not possible to determine whether this pillar existed during phase 1 or if it was only built in phase 2.

Fig. 5-11 – View from the west of the entrance to “Buildings 2-3” with the four pillars delimiting the corridor. North is on the left.

Fig. 5-12 – Map of the different components of the entrance of “Buildings 2-3”.

Fig. 5-13 – Oblique view of the bases of the two external pillars of the Communal building entrance, taken towards the southeast

19In addition to this reorganisation of the two wall termina--tions, a series of features were identified in “Building 2” (formerly “intermediate” building in Vigne et al. 2011a, 2017b). These were either still in use or renovated/completed for “Building 3”. They were (fig. 5-7, 5-11 and 5-12; see details in appendix5-3):

• The southern half of the phase 1 entrance was blocked by a 1.5-m-long and 0.8-wide northwest/southeast earthen structure. We only found the basal stony floor (SU 2018/2010) of this structure and some remarkable deposits of exceptionally large flint flakes. Its north extremity was prolonged over 1.5 m long to the northeast until St 180 (fig. 5-12) by a 50-60-cm-wide wall, again with only the basal stony floor intact. This thick SU 2018/2010 lay directly on the above-mentioned circulation layers relating to the phase 1 entrance (SU 2003-2005, fig. 5-8).
• Parallel to this southeast wall and in conjunction with it, earthen wall St 185 (fig. 5-7, 5-12) delimitated a 1.5-m-long hallway to the northwest.
• In the four corners of this hallway, accumulations of 15-25 large stones packed in the building earth (St 178, fig. 5-10; St 180 and 181, fig. 5-13; SU 2022) suggest the presence of pillars. None of these could be excavated. However, the presence of a 15-cm post hole close to St 181 (St 182, fig. 5-13) implied that these pillars were reinforced with wooden posts.
• The hallway floor formed a c. 1 m² square prolonged to the northeast by a short corridor narrowed to 50-60 cm by the two external pillars. To the southwest, it also seemed to have been narrowed between the two innermost pillars (St 178 and SU 2022), in such a way that only one person could have passed through the hallway at a time. The floor was covered with 1-3 cm of building earth, representing either a circulation level or plaster, in continuity with a probable access ramp. The presence of several Sotira pits in this area made it difficult to confirm that this space was separated from the exterior by a small step (fig. 5-13). The robustness of the four pillars and the presence of the two lateral walls suggest that it was covered with a roof.

2.3. Conclusion

20Even though the apparently wide-open entrance structure of phase 1 was more or less obscured by the phase 2 entrance, the latter, which was also used for phase 3, appeared to result from a considerable reorganisation aiming to reduce access. The latter entrance system was composed of a narrow 1.5-m-long corridor protected by lateral walls and a roof.

3.1. Features related to the architecture of the building or internal earthen arrangements

3.1.1. Peripheral pairs of post holes

23Four oblong pits (typically 65 x 30 cm), located 50-70 cm inside and parallel to the perimeter wall, were regularly spaced in the southwest (St 127, 92, 102) and southeast quadrants (St 105). Several other features (St 106-107, St 112-113) suggest that there were also three similar regularly spaced pits in the latter quadrant, and two others in the north (St 195 and 197). It is thus possible that a total of 10-12 pits surrounded the building, and not just the four clearly visible oblong pits in the southwest quadrant. However, we cannot rule out the possibility that this system was restricted to the southern half of the building.

24Unfortunately, only three of these pits could be excavated. Each was composed of two holes wedged with stones or fragments of macro-tools (fig. 5-15), in which two pairs of posts with a diameter of 10-15 cm were fixed. These were probably tied to each other with perishable matter. In all three cases, remarkable deposits (picrolite, spondyle) were associated with these pits.

Fig. 5-15 – Nadiral views of one of the typical peripheral paired post holes (St 92) from the Communal building, taken in three successive phases of the excavation (A-C); D, Reconstruction of the profile.

25As already suggested above, this system probably reinforced the peripheral roof structure, at least in the southern half.

3.1.2. Central post holes

26Three large post holes were found in the centre of the St 10 circle. They were at least 20 cm deep, but the excavation stopped before reaching the base. The observations made in building 800 (chap. 6) suggest that they could have continued down to 70 cm. The most central one (St 109, fig. 5-16) was 60 x 40 cm. This oval shape probably resulted from the removal of the post, which was visible in the north-south profile. This unusually large post was wedged by numerous stones, the largest one being 25 x 25 x 25 cm. A second post hole (St 119: 25 x 30 cm), partially coalescing with the first, was found 15 cm further to the north. It may be interpreted either as a reinforcement of the first (paired posts?) or as the replacement of the first one after removal, with another post playing the role of the central post hole for “Building 2 or 3”. A third post hole of almost the same size (St 110: 40 x 30 cm) was located 50-60 cm to the south of the first one. In contrast to St 119, it is too far away to be considered as a post paired with the latter. It is probably the central post hole of “Building 2 or 3” (or even 4).

Fig. 5-16 – A, Oblique view of the central post holes (St 109 in the centre, St 110 on the right and St 119 on the left) of the Communal building, at the beginning of the excavation of the southern half of the fills. B, Detail of the N-S profile at the level of the central pole holes, showing the large wedging stones removed by the (successive?) removal of the post(s).

27It seems clear that the central post was removed and replaced or at least repositioned during each renovation phase.

3.1.3. Other stake, pole or post holes and putative pillars

28About 15 other circular soil spots in the Pleistocene floor probably indicate stake, pole or post holes. They could not be excavated. As for the central posts, they are most likely related to “Buildings 2-4” but we cannot exclude that some of them may have been dug by Sotira populations. However, a cluster of ten spots grouped in squares ZY-A/39-41 (fig. 5-14, B) suggests a specific arrangement connected with “Building 1”. More precisely, four or five of them (St 84, St 94-96 and maybe St 97), sometimes large, created an almost straight line following the east-west diameter, suggesting a 3-m-long partition wall. It was impossible to assert if this played a role in supporting the roof.

29The mapping of the largest stones found on the Pleistocene floor (fig. 5-17) brought a complementary view, albeit biased by erosion (absence of large stones in the south?) or by the subsequent removal of stones. At least three stone clusters suggested the presence of either a large post hole, that could not be detected as the excavation was unfinished, or of an earthen pillar on a stone base (possibly framed by a wooden post). Two of the clusters were already visible in the diametric profiles (fig. 5-6): north of the central post in the NS profile and between rows ZZ and A in the EW profile. The first may be connected with the central post. The third one contained not only by far the largest stone found in St 10 (St200, fig. 5-18), but also a series of stones creating a NS elongated pillar base, or even a partition wall base. In addition, as mentioned above, we could not determine if the entrance’s north pillar St 178 already existed during phase 1.

Fig. 5-17 – Schematic plan of the largest limestone blocks found on the Pleistocene floor of the Communal building. The dotted circle indicates the location of the putative pillars.

Fig. 5-18 – View of the north and northeast arrangements of the Communal building visible at the end of the excavation.

3.1.4. Earthen benches

30We could not rule out the possibility that the slight elevation of the peripheral floor in the southwest and southeast quadrants may have been accentuated by a thickening in the building earth as a result of the making of peripheral benches. However, this was not clearly visible in either the north-south profile (fig. 5-6 B) or during the excavation.

31Conversely, a 30-50-cm-high stone and earth accumulation named SU 10.39 in the NS profile (fig. 5-19) and 10.35/10.51 in the northeast quadrant (fig. 5-18) created a definite elevation. We presume that this was a bench because it was elongated over 3-4 m east-west, with a still relatively abrupt south edge in spite of erosion (fig. 5-19) and it was randomly covered with a 2-3 cm thick coating of fine clayey earth. Of course, on account of the large Sotira pit, we could not determine whether this feature extended to the north wall or not. West of this -excavation, however, it clearly extended too far north to be the base of the north pillar. The base of the latter was therefore probably included in this north bench.

Fig. 5-19 – Oblique view of the probable north bench of “Building 1”, cut at the level of the large stone St 200

3.2. Fire features

3.2.1. Hearth plates

32Four hearth plates were located in the south half (fig. 5-14: St 74, 78, 116, 121). Only two of them could be studied in detail.

33St 74 was a 40 x 35 cm oblong baked clay plate, lying directly on the local Pleistocene substratum, which was rubified in some places around the plate (fig. 5-20, A). The hearth plate was covered by the plaster of the earliest floor (SU 10.10).

34St 78 (fig. 5-20, B) was located 50 cm inside the western thickening of the peripheral foundation trench. It was composed of two rounded baked clay plates similar to St 74 and a rounded rubified surface. As it was not excavated, it was impossible to determine whether the latter was a small hearth pit or just the deep layer of an eroded third baked clay plate. The southernmost of these three features slightly covered the edge of the nearby pair of post holes (St 127).

Fig. 5-20 – Hearth plates. A, St 74 after light spraying with water; B, The three components (a, b, c) of St 78, near the paired post holes (St 127) and the western thickening of the peripheral foundation trench (St 47) of the Communal building.

35These features were too well delimited to result from a generalised building fire. They should therefore be interpreted as temporary hearths set on a building floor previously covered with a thin layer of smoothed clay. This could suggest ritual fires made either during the first stages of building (foundation rituals), or during the use of the building, since St 78a partly covered the nearby pair of post holes. It is difficult to describe these features further without an exhaustive excavation of the building floor.

3.2.2. Fire pits

36Three small and shallow fire pits were identified. Two of them were located in the northwest quadrant (fig. 5-14: St 64 and 79), very close to each other, and one in the southeast (St 113).

37Only St 64 could be excavated (fig. 5-21). It consisted of a central 5-cm-thick zone of grey earth, sub-rectangular in shape (35 cm N-S, 20 cm E-W). The western part was made up of a mix of silt and pinkish-grey concretions and the eastern part of bluish limestone blocks, coloured as a result of heating, packed in grey silt. Some small charcoals were found in the eastern and western parts and cereal seed imprints were preserved in the baked clay (chap. 23). This enigmatic fire pit seemed to be associated with a small rounded shallow pit coated with clay and filled with a grey silt rich in flint (St 203). It was excavated into the Pleistocene substratum through the basal coating of the building (SU 10.9).

Fig. 5-21 – Sub-nadiral view (A) and plan and profiles (B) of the hearth pit St 64 and its surroundings. SU 64.1: compact rubified earth; SU 64.2: semi-compact grey silt; SU 64.3: grey-pink silt with carbonated concretions and rare charcoal; SU 203.1: rather loose thin grey silt; SU 203. 2: compact clayey building earth.

38If we assume that these two features were functionally associated, they may represent (together with the nearby small pit St 208 and the other fire pit St 79?) a small temporary unit for craft or food consumption during the first phase of utilisation of the Communal building.

3.3. Other remarkable pits

3.3.1. “Mosaic” pits

39Two 35-cm-large bowl-shaped pits were respectively located 1 m north (St 81) and west (St 82) of the central pole (fig. 5-14). Only the first one was excavated (fig. 5-22). It was c. 10 cm deep, filled with grey-brown silt, earthen clods, animal bones and flint. A minimum number of 90 small 3-4 cm calibrated flat calcareous stones covered the pit base and walls in a joined-up manner. No evidence of joining or topping with plaster or clay was found, ruling out the use of these pits for storing liquid. There is no evidence of fire. These enigmatic pits, also found in the PPNB of Shillourokambos (Guilaine et al. 2011, 2021) and in several other places at Klimonas, may have been used for ritual or craft purposes.

Fig. 5-22 – “Mosaic” bowl-shaped pit St 81: A, East-West profile of the fills viewed towards the north; B, Sub-nadiral view of the bottom (North is to the top); C, Top and bottom map and profile. The reference level of the profile is at 100.32 m

3.3.2. Caches and other pits with remarkable deposits

40Except for the numerous previously mentioned rounded post holes, ten larger, not rounded pit openings were detected when stripping the building floor (fig. 5-14). They were located beneath the north bench (St 196), near the central post (St 120 and 204) or near the small fireplace St 64 (St 129, 203 and 208). In addition, five sub-quadrangular or circular-shaped pits were grouped in a cluster (St 98, 99, 100, 101, 115). It is unlikely that they were Sotira or historical pits, as most of them appeared to be overlain by the basal layer of the building fill (SU 10.10).

41Only two could be excavated: St 203, described above as possibly associated with craft or food activities, and St 120. The latter was a 30 x 40 cm sub-quadrangular, very shallow pit (10 cm), roughly surrounded by c. 20 small calcareous stones and filled with a loose grey silt with small charcoals (fig. 5-23, B). It contained a well-crafted and complete handstone, broken in situ (i.e., post-depositionally; fig. 5-23, A), and a fragment of another handstone, also in green rock. At the base of the pit, the grey silt was impregnated by pedological carbonate concretions (fig. 5-23, C).

Fig. 5-23 – The only remarkable deposit pit that could be excavated in the initial floor of the Communal building (St 120), at three stages of excavation.

42In addition to these medium-sized pits, and the remarkable objects found in the peripheral foundation trench (see above: 1.1.), several remarkable objects were found in some very small and shallow pits dug into the rocky floor. The limits of these pits were not always clear. One of the pits (St 75), 10 cm wide and 9 cm deep, contained a pendant made from a perforated fragment of a left scallop valve. In two other depressions, located in a zone of high-density structures, we unearthed a group of four-five large flint flakes or tools (St 86 and 87, fig. 5-20, B). A remarkable flint arrowhead (St 48) was also found on the floor, but was part of the base of SU 10.8, which probably derived from the demolition of the first building wall. This remarkable deposit was probably placed in the earthen wall and landed on the floor when the wall collapsed.

3.4. Organisation of the features in the circular space of St 10

43As already proposed by Vigne et al. (2012), the many features observed on and in the St 10 basal floor suggest a well-planned architectural organisation (fig. 5-24).

Fig. 5-24 – An attempt to analyse the architectural organisation of the first phase of the Communal building.

44The entrance appears to define a structuring axis. It serves as an axis of symmetry for the peripheral elevations and coincides with the axis outlined by the three central post holes and the cluster of probable remarkable deposit pits in the southwest quadrant. One of the paired peripheral posts was also aligned with this entrance axis. As the entrance axis formed a 45° angle in relation to the cardinal points, it was also connected in some way to the two N-S and E-W axes. The E-W axis also seemed to play a second major role: the south limit of the entrance, east and west hearth plates, the line of large post holes suggesting a partition in the west half, and the western “mosaic” pit were located on this E-W axis. The N-S axis appeared to be less structuring, as only the north pillar and the north “mosaic” pit are aligned with it.

45Astronomical references cannot be excluded, but the purpose of these axes appears to be connected to the location of the building and to technical constraints. The northeast location of the entrance, which was common to all the building entrances on the site (chap. 6 and 7), was opposed to the dominant winds. The north-south axis created a sense of slope and strongly constrained the architecture by creating dissymmetry between the north and the south roof-bearing systems, aligned with the direction of rainwater runoff down the slope.

46A probable second structuring system intersected that of the three main axes, in the form of two circles, in addition to the perimeter wall. The internal circle was delimited by the admittedly blurred internal edge of the north earthen benches, and also by the indistinct slope break where the floor had begun to rise up to the peripheral wall in the south and west directions. Most if not all the pits putatively interpreted as ritual- or craft/food- related were located in this circle. It can therefore be considered as the delimitation of the building’s central zone. The paired post holes in the southwest and southeast quadrants delineated a second circle, with an intermediate radius between the former one and the perimeter wall, i.e., c. 4.5 m. At least in the south half of the building, this peristyle circle played a significant architectonic role.

47The presence of the three pairs of post holes in the southwest quadrant suggests that the building may also have been structured by two secondary diametric axes per quadrant. Does this mean that the whole building was divided into twelve 30° sectors? No clear evidence to support this hypothesis was found in the three other quadrants.

3.5. Discussions and interpretations of the first phase of the Communal building occupation

48The high degree of deterioration of the earthen features and the unfinished excavation of the building deprived us of several key observations. In spite of this, we collected enough information to attempt to reconstruct the main aspects of the first occupation phase of the Communal building. It is appropriate to define them at this stage of the chapter as the N-S and E-S profiles of the St 10 fill demonstrated that this general organisation structures the following phases, described below.

3.5.1. Circulation level around the building

49The circulation level observed against the eastern edge of the building, close to the southern limit of the entrance, unquestionably corresponds to the first stage of occupation (fig. 5-8 and 5-25, A) and provides a unique opportunity to discuss the buried depth of the building. This level was located c. 0.6 m above that of the central floor of the building. We also know that, to the north, the floor was at least 1.2 m below the modern erosion level, suggesting that the building was originally dug down to 1.2-1.5 m. These two observations indicate a 0.6-0.8 m difference in elevation between the circulation level at the northern limit of the building and the one close to the entrance, i.e., 4-5 m further to the south. This means that the slope was very steep at the northern limit of the entrance (fig. 5-25, B), probably because it was artificially dug to open the entrance.

50On the other hand, as the southern extremity of the building floor was 0.15-0.12 m below the modern surface and 0.7 m higher than in the centre, we can estimate that the PPNA circulation level around the southern limit of the building was 0.3-0.4 m above the modern level of the terrace (fig. 5-25). We can deduce, therefore, that the slope between the entrance to the southern limit of the building was very gentle or even flat and that the southern part was dug out to a depth of c. 0.5 m, i.e., twice as shallow as the north one. The estimated excavated volume for St 10, previously assessed at around 70 m³ in Vigne et al. (2012), can now be confirmed to be between 67 to 78.5 m³.

51This confirms that the building was half-buried but also that it was clearly visible in the village, as it was raised 70-100 cm above ground level.

Fig. 5-25 – Reconstruction of the first phase of occupation of the Communal building, based on an East-West profile (A) and a 3D reconstruction (B).

4.1. The stratigraphic units

62The fill of “Building 2” was divided into two clearly visible groups in the N-S and E-W profiles (fig. 5-6).

4.1.1. Lower part of the fills: substructures

63The lower part was composed of heterogeneous clayey silts rich in centimetric to decimetric limestones, with highly fragmented flint flakes and animal bones. Micromorphological analyses showed several more or less localised trampling levels (Mylona et al. 2017; chap. 9), probably resulting from people renovating the building. These SU can be interpreted as -demolition products of earthen buildings re-used as embankments for levelling the St 10 pit, i.e., substructures. These stratigraphic units were roughly concentrically arranged (fig. 5-26). In the centre, the very stony SU 10.3 (that could alternatively be considered as a refection level of “Building 1”; chap. 9) and the more clayey and indurated SU 10.13 formed a vast 20-30-cm-thick filling, with a horizontal top (see fig. 5-6, A). To the south, around this locally very bioturbated accumulation, a crescent of SU 10.11, with abundant broken flint, probably represented a lateral variation of the same embankment. However, it decreased in thickness to the south. Externally, it was replaced by SU 10.8, composed of a lighter building earth with fewer stones (fig. 5-26). To the east, SU 10.8 was composed of some isolated blocks of building earth (St 65, 69, 71) lying on SU 10.3. These became increasingly aggregated southwards, and formed a continuous layer overlying SU 10.3. Conversely, to the west, SU 10.8 lay directly on the coating of the “Building 1” floor. This composite unit probably resulted from the natural or intentional collapse of the “Building 1” wall on one hand, and from the remains of the putative south benches of the same building on the other.

Fig. 5-26 – Oblique view of the stripping of the lower sedimentary components of “Building 2”, composed of concentric stratigraphic units creating a thick level embankment. The left extremity of the scale points to the north.

64We can conclude that this thick substructure was at least partly made with demolition products from “Building 1”, either used as such (SU 10.8) or remixed with water, stones and sediments rich in already broken (i.e., in secondary deposits) flint and animal bones (chap. 24).

4.1.2. Upper part: constructed earthen floor(s)

65The upper sedimentary units relating to “Building 2” were mostly composed of a 15-20-cm-thick layer of homogeneous building earth (SU 10.6), with much fewer stones than in the underlying embankment. The numerous flint, animal bone, -macro-tool and stone vessel fragments, as well as some large charcoals, were badly broken, suggesting that this material did not come from demolition but was specifically used for making this floor with first-hand raw materials. When the upper limit was not too badly eroded or bioturbated, it was remarkably horizontal, flat and even smooth (fig. 5-27). In the profiles, this formation either appears to be homogeneous from top to bottom, or is sometimes divided into two layers separated by small stones. In this latter case, the upper layer is not very thick, irregular in thickness and less homogeneous (“SU 10.6 sup” or 10.28). It may correspond either to products from the demolition of “Building 2”, lying on the floor of the building itself or a repair level of the building’s floor, as suggested by micromorphological observations (chap. 9). The presence of photoliths suggests that this floor was covered with plants or maybe mats (Mylona et al. 2017, chap. 9 and 22). Few micro-residues of human activity were observed in the floor, which could point to activities with low residue production.

Fig. 5-27 – Sub-nadiral view of the first stripping of the “Building 2” floor in the north sector of St 10.

66This floor was c. 30 cm above that of “Building 1”, i.e., “Building 2” was still buried about 1 m to the north and 0.2 m to the south. It spread out over a surface of c. 40 m². To the northeast, it came to a standstill against, from the north to the south, the north branch of the peripheral wall, the northeast pillar (St 178, fig. 5-10), the entrance area (see 2.2) and the south branch of the peripheral wall. It is important to recall that, at the contact of this floor, both the north and south branches of the “Building 1” wall had been renovated by a massive addition of building earth (St 182 overlying St 179 and St 117 overlying St 47, respectively, fig. 5-10 and 5-14; see also chap. 9).

4.2. Benches and general organisation of “Building 2”

67To the north, floor SU 10.6 runs into the abrupt 10-cm-high elevation of the same good-quality earthen material (St 166, fig. 5-28). The east prolongation of this feature was mostly destroyed or almost interrupted by the large Sotira excavations (St 191, fig. 5-27), but the preserved strips in St 166 and St 172 were clearly in continuity with each other. Thus, it appeared that the bench curved to the north then aligned with the north delimitation wall of the building entrance. The west prolongation of St 166 could only be followed for a short distance, between the north pillar and the large Sotira silo (St 150). This low but abrupt elevation could therefore have been almost 4 m long from west to east, and can be interpreted as the remains of a low bench. The south edge of this bench would be more or less located in line with and above that of “Building 1”.

Fig. 5-28 – Northern limit of the “Building 2” floor, leaning against a short portion of the elevation of the north bench, between the north pillar and the Sotira silo.

68To the northwest, the abrupt vertical contact between the floor and SU 10.15 and 10.16, located beyond the northwest pillar (fig. 5-6, A), suggested that this north bench extended out to the peripheral wall in this area. The sediments of these SU were, however, richer in stones and more heterogeneous than those of St 166 and SU 10.6, suggesting that they might represent stony floors and that they could have been covered by a fine, now eroded, building earth. It should be noted that these SU lay from place to place on blocks of demolition products derived from “Building 1” (SU 10.8, fig. 5-6, A). Thus, they were not reused features but were built specifically for “Building 2”.

69The delimitation of the “Building 2” floor to the west and southwest was unclear as a result of many bioturbations (already visible in the underlying embankment, fig. 5-26). However, based on the north-south profile (fig. 5-6, B), and on observations made during the excavation, it appeared that it spread south until row 42, where it was also confined by an earthen feature (SU 10.12). It was impossible to delimit the southern side more precisely on account of the historical pit chain St 7 (fig. 5-29, A). However, it is noteworthy that in the middle of row 39 in the north-south profile, the southern side rose sharply by about 20 cm, which could correspond to the edge of a southern bench.

70This suggests a rather similar general organisation to that of “Building 1” (fig. 5-29, B), albeit with a much narrower entrance and a central space almost completely surrounded by probably slightly lower benches (c. 20 high). It is possible that the south bench could have been more than 3 m long, reducing the central space in comparison to that of “Building 1”.

Fig. 5-29 – Plan of the extension of SU 10.6 and the attempt to reconstruct the general organisation of “Building 2”.

4.3. Stakes, posts and post holes; pillars

71In addition to the central post discussed above (3.1.2., fig. 5-16) and to the four posts of the entrance (2.2., fig. 5-10 and 5-12), only seven stake holes (sometimes questionable) were identified for “Building 2” (fig. 5-30). It should, however, be kept in mind that their attribution to this building rather than to “Building 3” was sometimes uncertain (see appendix5-2). They were clustered against the east wall (St 51, 61, 52, 63), isolated to the south (St 27) or to the north (St 192), or aligned east-west (St 58, 59, 60). None of them could be easily interpreted, except the latter, which may replicate the alignment of “Building 1” interpreted as a frame for a possible partition wall (see supplementary information in appendix5-5).

Fig. 5-30 – Location of the stake and post holes, of the putative pillars and the different pits in “Building 2”.

72The three stone accumulations packed in building earth are of particular interest here and can be interpreted as pillar bases. We already described the north-easternmost accumulation, which possibly served as the entrance’s fourth pillar (St 178, fig. 5-10). To the northwest, the continuity of the bench (between St 166 and SU 15-16) was broken by a 50 x 43 cm accumulation of 4-12 cm stones mixed with flint and large microtool fragments (fig. 5-31). At a depth of 28 cm, it was embedded in the underlying embankment (SU 10.13).

Fig. 5-31 – Base of the north-western pillar (St 171) of “Building 2”, truncated in the north by the Sotira silo. On the right, the surface of the floor, delimited to the north by the rest of the bench (St 166).

73Between the northwest and the northeast pillars, a large block of massive limestone with flat upper and lower faces (St 209) could have been used for setting a post. It was associated with an accumulation of stones breaking the continuity between the east and the west parts of SU 10.6 (fig. 5-28 and 5-31). This suggested the existence of a third pillar, in line with and above the north pillar of “Building 1”. Large stones and important accumulations of building earth north of the central post holes (fig. 5-6, B) indicated that the central pillar, already suspected in “Building 1”, had been restored and maybe strengthened for “Building 2”.

Fig. 5-32 – Pits in “Building 2“: A, B, St 26; C, St 28.

4.4. Pits and remarkable deposit

74We only recorded three pits in the “Building 2” floor, at least partly because a large southern part of its surface was eroded. Two of them were rounded and shallow (10 cm) without any distinctive elements in them (fig. 5-30 and 5-32). The third one (St 57) may correspond to the remains of a small hearth pit, but had largely been destroyed by the St 15 Sotira silo.

75A fourth very small pit deserves a special mention (St 186, fig. 5-33). Located close to the north bench edge, behind the north pillar (fig. 5-30) this 20 x 10 cm pit was 4 cm deep, filled by a dark grey-brown silt and contained a square fragment of left human parietal (chap. 32). The pit was excavated in SU 10.38, i.e., the upper part of the north bench of “Building 1”, but was just a few centimetres below the surface of the north bench of “Building 2”. It was therefore unclear as to what phase it should be related to. Nonetheless, this deposit was truly remarkable as only 13 human bones have been found at the whole site, mostly skull bones.

Fig. 5-33 – Remarkable deposit in the north bench – a human parietal fragment (A) in a small pit (B).

5.1. Floor and embankment

76“Building 3” was eroded in the south sector but much better preserved in the north. Like “Building 2”, it was characterised by a substructure layer (stony floor SU 10.5) topped by an earthen floor (SU 10.2, fig. 5-6, A).

77The 10-12-cm-thick and dense stony floor was directly laid on the ruins or last renovation of “Building 2” (SU 10.6, SU 10.6 sup, 10.28 and 10.31; fig. 5-34, A-B). It was made up of small limestone blocks of 3-8 cm, with quite sharp edges, packed in building earth and accompanied by numerous centimetric to decimetric flint fragments (fig. 5-34, D). In some places, there were concentrations of several bones and burnt ochre pieces. The surface was not carefully levelled: it outlined a depression in square B/36 and at the level of the central post, as already noted in the north-south profile (fig. 5-6, B). The latter -irregularity probably results from the removal of the central post of “Building 2”.

Fig. 5-34 – Sub-nadiral view (A) and plan of the components (B) of the stony floor of “Building 3”. C, Oblique view of the superposition (in the raw B berm) between stony floor SU 10.5 and earthen floor SU 10.2, overhung by the bench St 153 (left, i.e., north). D, Detail of the composition of the stony floor with limestone and flint blocks.

78In the central part (B/38-39, fig. 5-34, B), including the central post, limestone and flint pieces were less densely distributed. The external limits of the stone layer were not circular but roughly pentagonal, and randomly organised following the entrance symmetry axis mentioned in “Building 1”.

79The earthen floor (SU 10.2) was composed of a fine, homogeneous and dense brownish-grey clayey silt with very few stones and archaeological items. The floor surface was horizontal and smoothed (fig. 5-34, C). It displayed significant thickness variations of less than 10 cm in E/36 to 18 cm in B/36, depending on the irregularities of the top of the underlying embankment. It was only well preserved on a small surface of c. 8 m², in the north zone (fig. 5-35) and in the ZZ-G/37 trench (“yellow layer”). South of this trench, it was almost eroded but still detectable at the surface of the embankment over 6 m², to the central posts (fig. 5-34, B).

Fig. 5-35 – Sub-nadiral view of the second stripping of the St 10 north sector, showing the earthen floor of “Building 3”, bordered in the north by the north bench, perforated by many Sotira pits.

80Micromorphological analyses evidenced at least three phases of floor repair by refilling, leaving no doubt about the presence of plant litters or matting. They also suggest, as for “Building 2”, a spatialisation of activities in “Building 3”, probably including food preparation and consumption. Running water marks were also observed (Mylona et al. 2017; chap. 9).

81The average depth from the floor surface was at a +0.95 m elevation, whereas that of the havara at the north limit of St 10 was at +1.13 m. “Building 3” was thus still buried but only by 20-30 cm on the north side. The southern side was not buried.

5.2. Bench(es)

82As for “Building 2”, the floor was delimited to the north by an abrupt elevation of c. 5-15 cm of fine smoothed building earth (fig. 5-35 and 5-36, St 153). This formation of compact earth also lay on substructure SU 10.5 and was located almost in line above the bench of “Building 2” (St 166). Like the latter, it only subsisted on a 30-cm-wide band because its north extension was destroyed by the large Sotira pits in squares A-E/35-36. This bench edge spread more or less straight in an E-W direction over a length of 3 m. Towards the east, its extension was blurred by a small fault line and by another Sotira hollow (St 147).

83Further to the east, as well as in the northwest quadrant, stony floor SU 10.5 disappeared (fig. 5-6, A) and SU 10.2 gave way to a slightly different SU (SU 10.4, fig. 5-34), a light-grey clayey silt with limestone blocks several centimetres long and small quantities of archaeological material. Following the same reasoning as for SU 10.15-16 of “Building 2”, just beneath it, this SU 10.4 may be considered as the deep part of the eastern and western extensions of the north bench, the upper part of which may have been clipped by erosion. If we opt for this admittedly tenuous interpretation, we have to consider that, for “Building 3” like for “Building 2”, a 1-m-wide bench ran at least 5-6 m all along the north wall, starting from the entrance. The apparent linearity of the section visible on fig. 5-35 and 5-36 suggests that the whole bench system might have created a polygon rather than a circle. This observation is in accordance with the above-hypothesised pentagonal delimitation of the stony embankment (fig 5-34) and, in turn, strengthens this supposition.

Fig. 5-36 – Oblique view, taken from the east to the west, of the preserved narrow band of the north bench of “Building 3” (St 153). The two large rounded notches visible on the right of the bench result from Sotira pits (St 190-191). In the upper left angle, Sotira pit St 145 cuts into the floor of “Building 3”.

5.3. Other internal features

84Supplementary photos and drawings of the internal features relating to “Building 3” are available in appendix 5-6.

85The most important internal structure was a central pillar, clearly visible in the N-S profile (fig. 5-6 B, St 54) and located 50-80 cm north of the putative central post hole (fig. 5-37). This was a massive sub-quadrangular block of brown homogeneous and compact building earth laid on the central area of the substructure, with sparse stones. It was preserved to a height of 40-50 cm and measured 110-120 cm long from east-west. The north-south dimension cannot be more accurately assessed over more than 40 cm, because test trench ZZ-G/37 was excavated before this earthen pillar was recognised as such and destroyed its northern termination. Still, some of the vertical limits of the pillar appeared to have been smoothed. At the pillar’s south foot, on a preserved strip of smoothed floor (SU 10.2), lay an earthen layer (SU 10.2 base), probably deriving from the deterioration of the pillar (melting: fig. 5-38, A). Along the south limit of the pillar, a post was planted (St 55, fig. 5-38 A,B). A small circular depression 20 cm further to the north may be the base of another post hole. This pillar may have been wood framed.

Fig. 5-37 – Location of the internal features of “Building 3” and an attempt to reconstruct its general organisation.

86The earthen dome of St 134, located above the N-W pillar of “Building 2” was very eroded (fig. 5-35) but clearly visible in the E-W profile (fig. 5-6, A). Conversely, no clear evidence of a north pillar superimposed on that of “Building 2” was found, except if the big stone (St 209, fig. 5-27 and 5-31) is considered to be part of “Building 3” rather than part of “Building 2”.

87No other stakes or post holes were found, probably owing to the high degree of erosion of “Building 3”, and only one small pit (St 29, fig. 5-38) could be located, at the limit of the putative south bench.

8.1. General organisation

94These successive versions of the Klimonas Communal buildings have much in common with PPNA communal buildings on the Continent: perfect circular shape, larger than the surrounding buildings, (semi-)buried, radial symmetry, peripheral wall and benches, roof, caches of remarkable items including some human skull bones (e.g. Aurenche and Kozłowski 1999, Schmidt 2000, Stordeur et al. 2000, Yartah 2004, 2005, Ibañez 2008, Coqueugniot 2011, 2014, Goring-Morris and Beler-Cohen 2011, Abbès 2014, Stordeur 2015, Hadad 2017). The two to four large pillars assumed to have been built in all three phases may even recall for example, the pillars of the E or D type buildings of Nemrik 9 (Kozłowski and Kempisty 1989) and may prefigure the architectural model of the largest buildings of Tenta or Khirokitia (Le Brun 1994, Todd 2001). The south peristyle also recalls some continental late PPNA or early PPNB architectures (Tell Abr’, Yartah 2004, 2005; Jerf El Ahmar, Stordeur 2015). A lateral descending entrance has been described at several sites in Central Syria (Abbès 2014) and Dja’de, where the ramp was as wide as at Klimonas (Coqueugniot 2014). Foodstuff storage and the practice of special artisanal or food activities inside the communal building have been reported at Jerf El’Ahmar (Stordeur 2015). Even the reconstruction/renovation of several successive buildings, sometimes after partial destruction, are common PPN practices (Hadad 2019).

95The evolution of the entrance size, reduced between phases 1 and 2, is also of particular interest. Of course, this may result from the necessity to control the risk of runoff water entering and flooding the lower parts of the building. If this was the main motivation, it resulted in facilitated control of the people authorised to enter the building. This confirms that, as on the continent (Stordeur 2015), this building was among other things, an influential means of dividing society into at least two unequal parts from a social-political point of view.

96The Klimonas Communal building grouped together diverse continental traits, mostly from Southeast Anatolia and across the Euphrates Valley. Without doubt, this building was one of the multiple versions of PPNA communal buildings known throughout the Levant and Southeast Anatolia, indicating that a large part of the socio-cultural PPNA package was transferred to Cyprus at the latest at the beginning of the 9^th millennium cal BC (Vigne et al. 2012, 2017b, 2019).

8.2. Large surface and shallow buried depth

97The 10 m diameter of the Klimonas building appears to be rather large in comparison to north Levantine and Southeast Anatolian standards. Diameters fluctuate between 5-6 m in the Palmyra area (Abbes 2014, Stordeur 2015), 7-8 m in the middle Euphrates Valley (Stordeur et al. 2000, Ibañez 2008, Coqueugniot 2014, Stordeur 2014, 2015), with a maximum of 10-12 m at Tell Abr’ (Yartah 2004, 2005), 5-8 m in the Jezireh and the high Euphrates-Tigris Plateaux (Kortik Tepe, Özkaya and Coşkun 2009; Nemrik, Kozłowski 1989, Qermez Dere, Watkins, 1990; M’lefaat, Kozłowski et al. 1998; Halam Çemi, Rosenberg et al. 1998). Significantly larger buildings were only found in Göbeckli (buildings D and maybe C; Schmidt 2000).

98A 10 m diameter is more common in the south Levant (Nahal Oren, Noy et al. 1973; Gilgal I, Noy 1989; Netiv Hagdud, Bar Yosef et al. 1991; see also Aurenche and Kozlowski 1999, Goring-Morris and Belfer-Cohen 2011, 2014), especially in Jordan where the largest specimen was found at Wadi Faynan (Jordan, 18 m, Mithen et al. 2011).

99The large size of the Klimonas Communal building may thus indicate a south Levantine influence.

100Taking into account the many similarities with the north Levant and Anatolia, it seems, however, that the large size of the Klimonas Communal building is probably characteristic of the building itself, rather than a south Levantine influence. As suggested by R. Hadad (2017, 2019), it should be considered as a counterpart of the shallow burying depth, which, this time, clearly differs from the continent, where buildings were generally buried to a depth of 1.8-2 m (Stordeur 2015). Our estimation of an excavated volume of 67-78.5 m³ should be compared with the 90-100 m³ extracted from the largest buildings at Jerf El Ahmar (50 m² x 2 m deep): according to Hadad (2017). The difference between the two can be estimated at 50 additional days/person and probably 30% more as it is more difficult to extract excavation spoil from 2 m down than from 1 m deep.

101However, the shallow burying of the Klimonas building finds some equivalents in central Anatolia (e.g., Ashıklı, Şahin et al. 2019).

8.3. Architectural fragility

102In comparison to most of the continental standards, the main differences are the near absence of stones, except for wedging the largest posts or pillars. The material used everywhere is basic cob, without recourse to any more sophisticated form of building earth, such as adobe (chap. 9). This specific use of cob was also mentioned at Asprokremnos (Manning et al. 2010, McCartney 2017). The use of cob could have been counterbalanced by building more massive walls or radial pillars, such as the ones of the Dja’de building (Coqueugniot 2014), and by a higher-quality cob (chap. 9). However, the 30-cm-thick perimeter earthen wall was thinner than most of its continental counterparts, which were 50-100 cm thick or sometimes more, and the huge radial pillars of Dja’de, Jerf El Ahmar or Mureybet were replaced by apparently weaker pillars. This suggests a lighter roof, but also, together with the relatively low burial depth and the rather large surface, a more fragile architectural model at Klimonas. R. Hadad (2017, 2019) remarked that these traits were also distinctive of the Jordan building of Wadi Faynan, and suggested a kind of “resistance” of peripheral areas to core area standards.