1During the 2009 and 2010 campaigns, work in Zone 3 focused on two significant rooms (3.8 and 3.1) of the building located on the summit of the hill and both already touched in 2007 and 2008 (Gaignerot-Driessen 2009). Excavation revealed a LM IIIB shrine (Room 3.8) and a large Hall (Room 3.1) to the North, presenting successive phases of occupation (figs. 5.1-5.5).

1. The shrine (Room 3.8)

2Room 3.8 (internal dimensions: 2.6 m E-W x 3.3 m N-S; 8.58 m ²) is located immediately to the East of Room 3.4 and to the South of Room 3.1 (fig. 5.1). Two small square spaces (3.9 and 3.10) of ca. 1 m ² each without entrance are situated to the North of Room 3.8 and separate it from the South wall (C 18) of Room 3.1 (fig. 5.8). They were found empty, apart for some pebbles, small fragments of shell and a crab pincer found in space 3.9. A triangular stone (ca. 0.6 m x 0.33 m) with a polished flat surface sits close to the centre of the North wall (C27) in Room 3.8. In the narrow, 0.20 m wide space between the stone and the wall were found three kalathoi (09-03-0514-OB004-5 and 09-03-0516-OB001) in situ with a lithic tool (09-08-0514-OB006) in one of them (09-03-0516-OB001), while two other very similar kalathoi (09-03-0514-OB002-3) were found in front of the triangular stone to the West. (fig. 5.7). Moreover, fragments of a sixth one (09-03-0511-OB001) were one meter further to the South. One of the kalathoi (09-03-0514-OB002) most likely fell from the mouth of a snake tube (09-03-0514-OB001) lying to its West (figs. 5.7-9). At the very end of the 2008 campaign, a first tube (08-03-0515-OB001), without handles but with small horns of consecration, had already been recovered slightly further to the South (Gaignerot-Driessen 2009: 125 and fig. 6.16). To the East of the triangular stone and against wall C 27 were found, from West to East, a lamp handle (09-03-0514-OB007), a large triton (09-03-0514-OB008) and an inverted high conical cup (diam. base: 0.6 m; h: 0.9 m) (09-03-0514-OB009) (fig. 5.7). In front of this collection of objects, to the South and following a semi-circular arrangement, were a discoid object made of schist (09-03-0514-OB010), a fragment of triton (09-03-0516-OB002), a jug, a stirrup jar and a fragmentary vase (09-03-0514-OB011-13) (fig. 5.8 and fig. 8.3b). In this area, traces of burned earth as well as burned pottery were encountered. A deer antler tine2 (09-03-0506-OB001) (fig. 5.9), a stone vase fragment (09-03-O506-OB002) and several joining pieces of a shallow everted bowl (09-03-0506-OB003) were also recovered from Room 3.8 (fig. 5.6).

3Apart from the triangular stone, Room 3.8 shows a series of other architectural features (fig. 5.8). A low, oblique bench (C29) runs against the East wall (C9) whereas an empty space in the North-east corner may have been used as a pot stand. The vases found broken but complete in the North part of the room may have fallen from this bench. To the South, against the West wall (C14) of Room 3.8 was a gourna with two circular depressions (C27) with, in front of it, four lithic tools and two bases of conical cups. Fragments of bone and shells were found scattered around the gourna and may constitute the rest of offerings prepared in this area. A bit further North, more lithic tools were collected and an accumulation of small stones against wall C14 may suggest another installation. A low wall (C28), on the same line as the gourna, divides the room in two parts. South of it, the foot of a goblet (09-03-0513-OB001), a lithic tool and a vase (09-03-515-OB003) were recovered (fig. 5.6).

4A small sounding under this deposit produced a stony level including a short LM IIIA kylix foot (09-03-0516-OB003) attesting an earlier occupation. A large piece of conglomerate forms an N-S line with the triangular stone which was carefully wedged with small stones at its base. A superficial cleaning of the North-west access to Room 3.8 stopped on an earlier fill.

5The finds made in this room (two snake tubes, six kalathoi, two triton shells, a deer antler tine, three conical cups) (fig. 5.9) and their arrangement as well as the architectural features observed (triangular stone, bench) (figs. 5.7-8) suggest that this space was used for religious activities and formed a shrine containing ritual equipment organized around the triangular stone which may has been used as an offering table or a platform (Gesell 1985:47-54 and Chart VI, 147). It is however difficult to call this stone an altar since no traces of burning were visible on it (Bretschneider in print 2011). The situation rather echoes the offering table found in the northeast corner of the bench sanctuary at Gournia (4 by 3 m): four snake tubes were upright around it and a fifth one was standing on top of it (Gesell 1985: 72). One crucial element that seems to be missing and would have made the identification as a LM IIIB shrine secure is a goddess figure with upraised arms. One could argue that before the site was destroyed and abandoned, the idol was carried out in time. There are many examples, however, that show that such figures usually are attested in larger numbers: at Vronda (Gesell et al. 1991, 1995), Karphi (Gesell 1985: 81), Halasmenos (Tsipopoulou 2009: 124-127), Kephala Vassilikis (Eliopoulos 1998: 307-309), to mention the closest sites where their presence is attested, a number of complete or fragmentary examples were recovered within LM IIIB to LM IIIC shrines as well as fragments scattered in their proximity or coming from pits or dumps. Oddly enough, after four campaigns of intensive excavations at Sissi, not even a small fragment of a goddess with upraised arms has been recovered. Given the situation in other contemporary sites located in the area, the absence of goddesses with upraised arms at Sissi perhaps needs a more convincing scenario than simply presuming they were taken away before the destruction. Indeed, ritual equipment commonly connected to goddesses (snake tubes and kalathoi) was actually stored or on display in the context of Room 3.8 which was clearly not domestic, while other snake tubes were recovered elsewhere, one in Room 5.13 (Zone 5; 09-05-0974-OB002; see Devolder in this volume), probably fallen from an upper floor, and another one in Room 4.11 (Zone 4; 09-04-0742-OB001; see Letesson in this volume), close to the entrance of Room 3.8. If it is accepted that Room 3.8 is a shrine without containing figures with upraised arms, then perhaps the identity of the figures themselves needs to be questioned. Sissi is indeed not the only site where the female figure with upraised arms seems to be missing from contexts where all other ritual equipment has been recovered. Examples such as Aghia Triada (building H), Karphi (Room 58 in the Priest’s House), Kommos (Room 4 in the House Shrine), Kephala Kondrou, Katsamba and Koumasa (Gesell 1985:41-42, 74-75, 81-82, 102) can be compared to the case at Sissi. Tsipopoulou (2009) recently suggested that the sets of figures and snake tubes found in the LM IIIC shrine at Halasmenos were votive offerings representing different clans or gene in a context of regional competition. Their absence at Sissi from a clear religious context then suggests that these figures with upraised arms were not the image of a divinity (Prent 2005: 181) but should rather be seen as representations of the worshippers themselves3.

6The singularity and importance of the Sissi shrine is also underlined by another fact: Room 3.8 is not an isolated, free-standing structure similar to other LM IIIA2-B and LM IIIC ‘town’ shrines. On the contrary, it is included into an important architectural complex located on the summit of the hill and adjacent to the main room (3.1) of the building, although not communicating with it. In a recent synthetic article, B. Hallager (2009) claims that domestic shrines did not exist on Crete during the period in question. Reconsidering the examples of rooms dedicated to a religious function in residential buildings as listed by Gesell (1985: 47), she opposes their identification as domestic shrines. This, she argues, is because of a lack of cult objects other than snake tubes and the latter are assumed to be moveable (Katsamba, Kommos, Kephala Khondrou); in other cases such as Knossos, she finds the context too dubious or interprets the finds as dumps or storage places for cult equipment (Palaikastro, Karphi) (B. Hallager 2009: 110-116). Her arguments do not seem to hold for Room 3.8 at Sissi and its interpretation as a shrine since the type and number of finds as well as the architectural features appears typical for a religious context. Moreover, their contextual association clearly indicates that the space was neither a dump nor a simple storage place for cult equipment. B. Hallager’s (2009: 121) claim that no room solely dedicated to the purpose of worship can be found in LM IIIA2-IIIC Crete seems no longer tenable because of the Sissi evidence. It may be added that even for Eliopoulos (2004: 82) the shrine at Gournia ‘is not an isolated, single-room public shrine’ and rather ‘belonged to a larger building context’. The exact function of the shrine at Sissi however depends on a better knowledge and understanding of the whole building. In this regard, the excavation of the large room 3.1 could perhaps shed some light.

2. The hall (Room 3.1)

7For its size and position, Room 3.1 is the main and central space of the building located on the summit of the Kefali hill. In 2007, when the East part of the room was excavated, a monumental threshold (1.80 m long) as well as a possible wine or olive press (C5) and a ceramic deposit were found (Gaignerot-Driessen 2009: 115-121). During the 2010 campaign, the rest of the room was excavated so that the size of the LM IIIB hall became clear (external dimensions: 9 m E-W by 8.62 m N-S; 77.58 m²; internal dimensions: 7.90 m E-W by 7.50 m N-S; 59.25 m²) and the presence of two axially placed column bases (fig. 5.10).

8The threshold in the East wall C1 and the two column bases are located on a central E-W axis at relatively regular intervals (threshold-East Column base: 2.44 m; East Column base-West Column base: 2 m; West Column base-West Wall D27: 2,44 m). The West base (C33) consists of an irregular, roughly circular block (max. diam.: 0.66m) of which the base is wedged by small stones (fig. 5.14). The East base (C29) is perfectly circular with a dressed and polished cylindrical surface, and the lower unworked part of the stone being embedded below floor level. The careful work on C29, very similar to e. g. a Neopalatial example from Kato Zakro (Shaw 2009: 83 and fig. 146, 277), in comparison with base C33 suggests it was reused. Moreover, fragments of a small stirrup jar (10-03-0544-OB001) were buried at its base, perhaps as a foundation deposit (fig. 5.11). A similar practice was already noticed elsewhere during the excavation of the building: a stirrup jar was found in 2007 at foundation level of the monumental threshold, and a deposit of shells, pebbles, bones, four miniature jugs, a tripod leg and a spout were found in space 3.7 (Gaignerot-Driessen 2009: 121-123), moreover, a vase covered with pithos fragments on which fragments of a triton shell were found and containing an inverted conical cup formed a foundation deposit to the West of Wall D11 (Letesson 2009: 133-134 and figs. 6.26 and 6.27, 133). Finally, in Room 3.1 a vase (10-03-0547-OB001) was left against the first course of Wall D22, at its eastern end (fig. 5.12). Around base C29, a few lithic tools (10-03-0524-OB001-6) (fig. 5.10) and some fragments of obsidian blades were recovered. Along Wall D22 and to the North of base C29, a tool made of pumice (10-03-0527-OB001) (fig. 5.13) was found.

9Evidence for destruction by fire was observed around the column bases, perhaps of the wooden columns, but no traces of ash which could suggest the presence of a hearth were recovered between the bases. A fragmentary clay surface (fig. 5.14) was, however, recognized between the bases slightly above floor level where there were also large fragments of two tubular vessels with one open end and one perforated domed end (10-03-0536-OB001 A-B) (fig. 5.16), perhaps fallen from above. These pots which have not yet been restored, may have been used as chimney pots to evacuate the smoke for a potential fire place between the two columns. The use of a portable hearth would explain the absence of ash from the clay surface. Large fragments of a pithos with guilloche decor (10-03-0531-OB007) (fig. 5.15) were also found in the area between the two column bases and Wall D22. Fragments of two other as yet unidentified vases (10-03-0531-OB005-6) were also recovered from the destruction layer in Room 3.1. It must be noted, however, that, in comparison with the Eastern part of the room excavated in 2007, the rest of Room 3.1 was almost empty (fig. 5.10). Room 4.9, which contained prestige vessels, was maybe the Hall’s pantry (see Letesson in this volume and Langohr, fig. 8.4 in this volume).

10A doorway between Room 4.8, where a fragment of the pithos from Room 3.1 was recovered (see Letesson in this volume), and Room 3.1 was found blocked in wall D22. On removal, a 1 m long stone threshold contemporary with the column bases was uncovered (fig. 5.17). Along Wall D22, a patch of fallen plaster was found. The other doorways to the South-East and to the North-East were also found blocked.

11In the South-west corner of room 3.1, a carefully constructed bin (C31) measuring 1.04 m by 0.58 m was found (fig. 5.10 and 5.18). It contained 10 lithic tools of different size and shape (10-03-0533-OB001-7, 11; 10-03-0540-OB001-2), as well as two fragments of millstones (10-03-0533-OB008-9) possibly reused as grinding stones (fig. 5.13). To the East next to the bin were sitting two slabs (C34): one is a fragmented ammouda (0.40 m by 0.40 m max.) with a circular depression, the other an irregular sideropetra block (0.29 m by 0.35 m) (figs. 5.10 and 5.20). We assume these slabs to be connected to the lithic equipment found in the bin and to be respectively a worktop and a seat. A semi-circular stone structure (C30) (diam.: 0.55 m; depth: 0.15 m) built against Wall C18, to the West of the South-east entrance into Room 3.1, may have been used as a pot stand (figs. 5.10 and 5.20).

12The South-west quarter of the room is covered with a well-preserved pebble floor that seems to belong to the same phase as the column bases (figs. 5.19-21). A similar floor was also found in Room 4.8 in 2009 (see Letesson in this volume). The excavation of the rest of the room revealed a thick layer of white-greenish local clay, in which the two column bases were set.

13North-east of the gourna (C6) found in 2007 next to the platform (‘press’) C5 against the northeast corner, another gourna (C35) was found in 2010 (fig. 5.17). Pieces of pumice, two fragments of obsidian blades (10-03-0545-OB003 and 7), two loom weights (10-03-0545-OB003) and the base of a miniature goblet in quartz (10-03-0545-OB004) were also recovered from this area (fig. 5.10). Against Wall C1, between the platform and Wall D22, the floor was paved with large stones, probably used in connection with the platform. It is worth noting that the floor level of the platform and the two gournes (21.47 m asl) in the eastern part of the room against Wall C1 is about 0.20 m lower than the level of the pebble floor, thresholds and column bases (21.66 m asl) in the western part. It is not unlikely that the platform already belonged to an earlier phase of the building but was preserved and reused during the LM IIIB period. This would have implied the digging out of the area of the platform up to the mouth of the two gournes and would explain the difference of level recorded here. South of column base C29 the foundations of a N-S running wall forming an angle with Wall C18 were found at lower level, appearing slightly lower and at the eastern limit of the pebble floor (fig. 5.10). The pottery recovered from the fill appears Neopalatial. It may be noted that the west part of Wall C18 is less regular and slightly differently oriented from this point onwards and preserves one more course than the eastern part. This may suggest that the eastern part of Wall C18 formed a later addition.

14The evidence of the floor levels and wall construction seems to suggest that, during LM IIIB, an existing earlier room with a pebble floor may have been enlarged to create a Hall, also incorporating an earlier installation (“the press”). At this stage, a bin, a pot stand, two stone thresholds and two column bases flanking a possible fireplace may have been added (fig. 5.21). These different architectural characteristics confirm the importance of Room 3.1 and it clearly appears that serious efforts were made during LM IIIB to incorporate this Hall at Sissi. The location of the room on the very top of the hill, the two stone thresholds, the two stone column bases, and the axial arrangement of the space all give a monumental aspect to Room 3.1. From the North, a line can be drawn that goes through the column base of Room 4.8, the threshold in Wall D22 and the column base C29 in Room 3.1. Entering the Hall from the East, the threshold in Wall C1 lined up with the two column bases which must also have had a visual impact on any visitor.

3. Conclusions

15An interesting parallel for Room 3.1 can be found at LM IIIC Kephala Vassilikis, where a large rectangular room with internal dimensions 7.50 m by 5 m (E6 or the ‘Hall of the Hearth’) presents a similar axial plan with two column bases flanking a clay hearth and a stone-lined pit built against the North-west corner as at Sissi (Eliopoulos 1998: 305). Room E3, or the ‘Hall of the Altar’, is adjacent to the North but does not directly communicate with Room E6. Benches and a platform run along all of its walls; in the middle of the room is a low, table-like stone construction with an unworked stone incorporated in its northern end and interpreted as a baetyl by the excavator (Eliopoulos 1998: 306-307). Only some fragments of a figure with upraised arms were found in E3. Snake tubes, a votive pinax, cups, kalathoi and figures with upraised arms were, however, recovered to the South of the ‘Hall of the Hearth’, in Rooms E4 and E5 (Eliopoulos 1998: 307-309). The so-called ‘Temple complex’ at Kephala Vassiliki hence shares several characteristics (architecture and finds) with the Sissi building, even if at Sissi the religious function seems thus far restricted to Room 3.8.

16Despite the proximity of the shrine, there is nothing to suggest a religious function for Room 3.1. The installation (“press”) in the North-east corner as well as the bin and working slabs in the South-west corner suggest that domestic industrial activities were performed in Room 3.1. The size, location, organization and finds of Room 3.1 clearly show, however, that it was more than an ordinary domestic space. The Sissi Hall seems indeed to reflect a message of power and seems to have been designed to, occasionally at least, allow the gathering of an important number of people.

17The nature of the building and the existence of the Hall in particular may have some impact on an attempt of interpreting Room 3.8. Above, I argued that Room 3.8 was an example of a domestic shrine. This designation is perhaps no longer appropriate, because it creates confusion between social and architectural characteristics. Taking the cult evidence on LM III Crete together we should perhaps no longer consider a distinction between on the one hand public and on the other hand domestic shrines but between independent structures and rooms incorporated into a residential building. Room 3.8 is an incorporated shrine, which does not imply it was a private one. It is, at the moment, too difficult to know which level of social structure is reflected by the building on the summit of the hill, its Hall and its Shrine: does it result from the initiative of a single individual, of a household or of a kinship group (see Mazarakis-Ainian 1997: 393)? Perhaps Room 3.8 at Sissi may for the moment being be called a ‘communal incorporated shrine’.