1Note portant sur l’auteur1

2This chapter constitutes a preliminary account of some of the Late Minoan pottery found on the Bouffo during the first two excavation campaigns2. Consequently, the funerary area (zone 1), used throughout a long period before the Neopalatial period, is not covered by the present report. This said, Late Minoan ceramics comprise most of the pottery hitherto found on the Kephali, with a majority of excavated contexts outside the cemetery, in zones 2 to 63. As yet, the stratigraphic units and the respective pottery have not been fully studied so all observations given here are preliminary. This explains why remarks on ceramic fabrics are kept to a minimum and are limited to some general observations made during the course of a rapid strewing of the ceramic material during the excavation campaigns. This account simply presents some preliminary comments on the shapes and the decoration of the Late Minoan pottery, with some suggestions for initial dating based on a few comparanda with some known Minoan assemblages. In anticipation of a full ceramic analysis, the recognition of typological and chronological pottery groups diagnostic for Sissi is still in its infancy. This first assessment of Late Minoan ceramics found on the Kephali is an attempt to provisionally place this already very promising settlement in its broader chronological and regional contexts. I first consider the Neopalatial pottery and then the Final Palatial and Post-palatial pottery.

1. The Neopalatial Pottery

1.1. Zone 2

3In the current stage of excavations, Zone 2 – the northwest middle terrace – is certainly the most promising area of the settlement for defining pottery consumption of the inhabitants of the Kephali during the Neopalatial period. A first glance at the assemblages from the annexes of the ‘Davaras Building’ suggests that the settlement suffered a disruption during the Late Minoan IB phase. Its nature still needs to be clarified but it preserved some remarkably well-fired pottery in space 2.1. The sherd material found in the ‘Davaras Building’ proper is not discussed here as its excavated contexts were relatively poor in ceramics, both where quantity and preservation are concerned4. In contrast, the compartmentalized structure added to the east of the latter provided good deposits, particularly the cellar situated south (2.1).

4The large quantity of pottery collected from this partly subterranean space 2.1 comprises several interesting deposits, in which subtle distinctions can be identified5. A small floor deposit found in the northwest corner of the room was sealed by considerable tumble, hence crushing the pottery below. A coarse vat was standing in situ on the bedrock, while an oval-mouthed amphora, some drinking and other closed vessels were scattered around in fragments, probably fallen during the destruction event (fig. 9.1).

5An interesting issue will be the definition of the chronological and taphonomical links between this seemingly primary deposit in the northwest corner of space 2.1 and the thick layers comprising an important and probably coherent assemblage of pottery collected in the rest of the same room before reaching the floor. Within the room was found a lot of well-fired, fine pottery consisting mainly of freshly broken but incomplete vessels (fig. 9.2). We assume that they represent redeposited destruction debris dumped here from a nearby area6.

6An interesting discovery among the ceramic material found in room 2.1 is an S-profile handled cup (fig. 9.2, 9.3a). Once comprehensively defined, this recurrent diagnostic shape will constitute without doubt a good chronological marker within the development of the site in general and, together with the other associated and more common Neopalatial drinking shapes, will be useful to illustrate consumption practices of the Sissi population during the Neopalatial period in particular. Often made in a well-fired, fine fabric it shows a special surface treatment: the cup was dipped upside down in the paint which habitually drips on the external and internal walls (fig. 9.3a). The oblique fixing of the strap handle is almost as characteristic as the repetitive ogival profile of the body. The S-profile shape and the dipped decoration technique are not unknown at other Minoan sites7. Significantly more exceptional is their common combination and the systematic presence of a handle as evidenced at Sissi. One also notes that in some cases, once the cup has been dipped, several horizontal painted lines are added from the base to the upper part of the body, in white or dark brown8. This drinking shape is tentatively and provisionally interpreted as locally inspired. We imagine its production at or in the region of Sissi. This assumption is based on the fact that dipped painted S-profile cups are particularly well represented within at least this horizon of the Neopalatial occupation of zone 2. Moreover, this type of cup seems at present not to have been commonly consumed at the nearby palatial site of Malia, nor in other Minoan sites. This hypothesis nevertheless needs confirmation in the future since the forthcoming publication of Quartier Nu at Malia will include a few complete handled S-profile cups treated in the dipped decoration technique coming from Neopalatial contexts (A. Farnoux and J. Driessen, pers. com.). It should not be forgotten that a proper Neopalatial ceramic sequence has still not been clearly defined at Malia, apart from O. Pelon’s work in Quartier Epsilon (Pelon 1970)9. In this connection, a particular category of cups identified by the excavators at Malia should be considered, the tasses galbées (esp. Pelon 1970: 80-81, 83-84, 150-151, 156; van Effenterre 1980: 530-532, fig. 762-769; A. Van de Moortel, pers. com). Under this label, a set of MM IIIA-LM IB handled cups are classified of which the profile develops along an axis of variation extending between a concave-convex cup on a raised narrow base to the common rounded cup with a flaring lower part. Some small-sized (H.: 6-8 cm), mostly monochrome examples are not entirely different from the specific profile of the Sissi cups (Deshayes & Dessenne 1959: 42, pl. VIII: 3, n° e and f [n° f is dipped painted and shows drips of red paint from the rim]; Pelon 1970: pl. XXVIII: 6 (a-b); van Effenterre 1980: fig. 762, 765). However, variations in shape for the tasse galbée do not seem to be paralleled by a proper chronological phasing and tangible developments (van Effenterre 1980: 532). Finally, at Mochlos, a few handled exemplars of the common LM IB ogival cup are mentioned by T. Brogan and K. Barnard, who also identified a few similar specimens in other contemporary Eastern Cretan contexts, as at Gournia and Prophetes Elias-Praissos (Barnard & Brogan 2003: 43, fig. 5 and pl. 7 [IB.197, IB. 198])10. In sum, the linkage of this seemingly local shape with the broader Neopalatial pattern is not straightforward. This prevents for the moment a clear assessment of its place within the larger Neopalatial chronological horizon.

7Other types of cup profiles exist in the same space 2.1 assemblage, linking the deposit in general with other contemporary Minoan traditions, such as conical cups and rounded-profile cups (fig. 9.3b). Where detailed chronological attributions and local developments of these different drinking shapes are concerned, precisions cannot yet be provided. However, where the conical cups are concerned, we noticed a good proportion of compact specimens with a convex profile or straight flaring walls (fig. 9.3c). Such morphological characteristics may be compared to type C in Van de Moortel’s conical cup typology for the Advanced and Final LM IA and LM IB phases at Kommos (Van de Moortel 1997: 57-59, 65-66, 71-72, fig. 8-10)11. According to her, the LM IA Final examples should have straight sides and straight rounded rims. In LM IB, their fabric is said to become very fine and soft, their bases better raised and their proportions somewhat more uniform (Van de Moortel 1997: 66). Since we are as yet unable to identify such recurrent features in the material of zone 2, it is too early to use the conical cups as chronological markers.

8This said, a Late Minoan IB date for the last use of the cellar 2.1 and its destruction may be suggested on the basis of the discovery of the false neck of a finely decorated three-handled stirrup jar found in the middle of the room, beneath a big stone, fallen on the bedrock floor (fig. 9.4). Finely decorated three-handled stirrup jars are mainly represented in LM IB deposits12 but, especially at Knossos, they are said to occur already in LM IA contexts (Mountjoy 2003: 63, fig. 4.5 [62], 68-69, fig. 4.8 [85-88]) and the form continues to be found in LM II levels (Warren 1983: fig. 5, fig. 18). However, and although we know nothing about the profile and the decoration of the Sissi example which could help to date it, a LM IB date is proposed for two reasons. First, the fabric refinement and small dimensions (the top disk of the false neck has a D.: 4 cm) of this example do not agree with the more commonly coarser and taller ovoid storage examples of LM IA (Mountjoy 2003: 68, 89)13. Secondly, not a single LM II stylistic feature has been preliminarily recognized in the space 2.1 deposit. Following this reasoning, the finely decorated amphora illustrated in figure 9.2 is here also tentatively assigned a LM IB date. A relatively good comparison where fabric refinement, shape and size are concerned has been found at Mochlos (Barnard & Brogan 2003: 71, fig. 31, pl. 18 [IB.381]).

9In space 2.2, situated to the north at ground floor level, a few complete vessels were found, smashed on the paved floor. Among these is a kind of pulled-rim bowl in a very fine but over-fired fabric (fig. 9.5). It is to date the only example on the site. It has not yet been restored. Pulled-rim bowls are well known in East Crete, particularly at the site of Palaikastro where they occur from LM II onwards and where this shape has definite local and regional LM IB antecedents, mainly in the ogival cup14. The Sissi example shows two small horizontal handles. The ceramic material found in the compartments abutting the east wall of the ‘Davaras Building’ seems to be entirely Neopalatial. The contemporaneity of the space 2.2 deposit with the secondary destruction deposit in space 2.1 is not excluded. However, the definition of this distinctive shape – the pulled-rim bowl – in the chronological local sequence still needs to be specified.

1.2. Zone 4

10The only other excavated zone on the Bouffo where Neopalatial pottery has thus far been identified is zone 4.

11It consists essentially of sherd material but a small deposit was found in some kind of drain at the intersection of walls D1 and D1815. This comprised an S-profile cup and a beaked jug. Both are much worn but they seem monochrome, possibly with dipped decoration, as far as the cup is concerned (fig. 9.6). The S-profile cup and beaked jug are not self-evidently diagnostic of the Neopalatial period. In addition, the S-profile cup slightly differs from the numerous morphologically comparable LM I examples found in zone 2 and described above, by its bigger dimensions, its coarser and less well fired fabric and, possibly, its surface treatment. Finally, the stratigraphy of this part of zone 4 is very complex and disturbed so the Neopalatial date is provisional. But since none of the (preliminarily identified) Final Palatial or numerous and secure Post-palatial levels excavated at Sissi have provided any evidence for whole or fragmented S-profile cups of the types attested in zones 2 and 4, this attribution is a good possibility.

12Moreover, close to the northwest angle of the summit, in the still only partially excavated room 4.9, the only comparable examples for the remarkably well-fired LM I pottery of zone 2 were found. This includes a couple of extremely fine sherds from drinking vessels decorated in the Tortoise-shell ripple style (fig. 9.7) which suggest that, already from the MM IIIB and/or LM IA period onwards, good quality ceramics were consumed at Sissi.

2. The Final Palatial and Post-palatial Pottery

13The excavations on the summit of the hill have yielded great quantities of LM III pottery, both as sherds and complete pots. The closeness of the surface layers is probably the reason why many deposits were affected. Still, several extraordinary well-preserved destruction deposits have also been brought to light, especially in zones 3 and 5. These come from the last floor levels, a few centimeters below the modern surface. The post-LM IB ceramic sequence promises to be very interesting, both where the local scale and the regional perspective are concerned. The main deposits belong to a LM IIIB destructive event, occurring possibly in an advanced stage of this phase but it is still too early to make definite statements about the precise moment. In any case, this destruction was followed by a very long abandonment of this centuries-old site. On a regional scale, it will be significant to relate the Sissi settlement to the LM IIIA2-B occupation of Quartier Nu at Malia, on the one hand (Driessen & Farnoux 1994), and with the LM (IIIB)-IIIC upland sites (Anavlochos, Selena and Papoura Mountains), on the other hand (Nowicki 2000: 147-173). In addition, although not yet well documented, the hill was also occupied during the LM IIIA phase and apparently from its beginning if not from LM II onwards.

2.1. Zone 316

14All the rooms excavated to date in zone 3 suffered from a violent destruction leading to their abandonment. Indeed, every single room was found full of archaeological material left in place, complete, but heavily broken. The most plausible hypothesis at present is that this reflects a single event. Some stylistic observations and comparisons can already be made and a preliminary chronological framework proposed.

15In the excavated part of room 3.1, large and coarse containers were associated with fine drinking and pouring vessels. The only pithos mended up and restored to date is interesting in many respects. Firstly, the fabric is unique as far as we can tell (fig. 9.8). The bright reddish-orange paste has obvious whitish and shiny small inclusions which give a gritty feel to the vessel’s surface. Secondly, the morphological and decorative characteristics of this almost one-meter high pithos allow some comparisons with LM III East Cretan examples17. The most relevant form to which this pithos can be related is form 8 in Christakis’ typology, mostly reported from LM IIIB-C contexts, and the LM III variant of forms 85-88 (respectively Christakis 2005: 7, fig. 3 and 16, 33, fig. 19). According to Christakis, these last examples were mainly produced in East-central and East Crete, especially in the Gulf of Mirabello (Christakis 2005: 16). A petrographical analysis by E. Nodarou highlighted that the production of amphorae and pithoid jars in the distinctive “Mirabello fabric” is carried on during the LM IIIA-B period18. Her limited analysis allows the recognition of exports towards the sites of Chrysokamino and Petras, but hitherto not to Mochlos (Nodarou 2007).

16A complete profile of a deep bowl with high straight walls and a slightly marked rim suggests a LM IIIB date (fig. 9.9)19. However, the monochrome surface treatment is more at home in the transitional LH IIIB/LH IIIC period and even especially at the beginning of LH IIIC on the Mainland (Vitale 2006: 180-181, fig. 3:4, 193, table 2; Mountjoy 1999: 37), while the situation has not been resolved for Crete20. At Malia, in Quartier Nu, the deep bowl is virtually absent from the LM IIIA2-IIIB deposits21. Since the occupation of the latter site is said to have ended around the middle or early in the second half of the LM IIIB period (Driessen & Farnoux 1994: 63), Sissi may have outlived its neighbour.

17This said, a terminus ante quem in LM IIIB for this deposit is suggested by some other vases. The presence of at least one fragmentary transport/storage piriform stirrup jar with octopus decoration fits better in a pre-LM IIIC context22. Besides, a finely decorated stirrup jar with a relatively squat profile found near the threshold of room 3.1 is characteristic for LM IIIB (fig. 9.10). It is made in a buff fabric and the decoration on its shoulder consists of Minoan flowers and concentric arches. The colour of the paste is not especially whitish which suggests it is not Chaniote, contrary to another complete example also found in zone 3 (see below)23.

18A final interesting but fragmentary vessel found in this room – not yet restored – is an open container resembling a bathtub (07-03-0419-OB005). It has a very pale buff fabric and is decorated in dark paint with octopus and floral motifs. The same very pale buff clay was used for other large containers (jars, pithoi) stored in the same space. A coherent program of petrographic analysis is planned in the future. Macroscopic observations on fabrics will help to define specific issues and questions, such as the potential presence of a potter workshop at Sissi or nearby24. Indeed, very pale buff clay pockets have been identified in the direct environment of the settlement, less than 500 m away25.

19It is likely that this pottery deposit was almost directly on a Neopalatial level, as four complete conical cups found just below the floor of the destruction layer suggest (fig. 9.11). Indeed, the profile, more convex than the generally outward walls of the LM III conical cups, suggests a Neopalatial dating (see above). More evidence is needed, however, to reconstruct the LM I phase of the building.

20Room 3.3 also provided a rich destruction deposit with vases smashed in situ. Besides several large storage vessels, and notably a huge tripod cooking pot in a particularly coarse fabric with important traces of burning (08-03-0451-OB004), there were several unusual and finely decorated vases. The most impressive piece is evidently an almost complete bell krater (fig. 9.12a). Most scholars who have seen this vase agree with a late LM IIIB date26. The clumsy decoration is made of simple abstract motives organized in panels (fig. 9.13). Good comparanda are relatively difficult to find27. It is interesting to note that its base was completely removed, most probably before firing, since the removal after the firing would certainly have damaged and broken the rest of the vase. Other examples are known: E. Hatzaki has remarked this practice at Knossos and also suggests its removal before firing28. On the Mainland, at Mycenae, this practice has been observed for deep bowls (Koehl 2006: ill. 5). The reason for this removal is unclear for the moment. Close by was a large shallow bowl with an atypical shape for which no close comparison has at present been found (fig. 9.12b).

21The same floor deposit included an unusual vessel, with a concave profile and a pierced bottom, a high-walled lid topped with a curved handle29 and at least one fragmented shallow bowl with two horizontal handles (fig. 9.14). All are in a buff fabric. Each of these vases has a parallel in neighbouring Quartier Nu at Malia30. Again, a full study of the LM IIIB deposits will allow us to define if and how these two settlements relate.

22The space between rooms 3.3 and 3.4 also preserved some interesting and complete vases, notably one complete high-stemmed kylix (08-03-0491-OB001) and a finely decorated stirrup jar (fig. 9.15). The latter is made in a whitish fabric. Its profile is very similar to the stirrup jar mentioned above, found near the threshold of room 3.1. The decoration applied on the shoulder is different, however, and quite extraordinary. It consists of a linear and dotted pattern, perhaps with some marine inspiration, not unlike that seen on a stirrup jar found in the Dikastiria cemetery at Khania (fig. 9.16). The import of a “Khania workshop” product at Sissi is a distinct possibility.

2.2. Zone 5

23The excavated deposits in zone 5 are less well preserved than those in the building on the summit. Most of the material comes from the tumble that is underlying the surface layers and it is not always easy to distinguish the fills from the destruction deposits. The removal of some of this tumble has nonetheless provided us with a good sample of diagnostic sherds of the pottery consumed on the Kephali during the LM III period including cups, miniature stirrup jars, kylikes and champagne cups (fig. 9.17).

24Near the northeast border of zone 5, the surface layer was very thin and immediately overlay an archaeological deposit consisting of a dense packing of sherds. Preliminary analysis suggests that this layer is the result of leveling or even a rubbish area. It is of interest that these top levels, especially those of spaces 5.8 to 5.10, comprised the only but verygood examples of LM IIIA1 ceramics hitherto found on the site(fig. 9.18).

25The best Post-palatial deposit in Building E comes from space 5.1331. Below a thick tumble, the first pieces of a well-preserved primary destruction deposit were collected. Besides several fragments of a larnax, the second example after that from zone 3, a broken but almost complete poppy capsule rhyton was excavated (fig. 9.19), an unusual shape (Koehl 2006: 126). Two examples are known from the LM IIIA1-B Mochlos Limenaria cemetery (Nicgorski 1999: 537-541, pl. CX), where they are dated to LM IIIA2 (Smith 2002: 453), and two other chance founds were made at Maroulas, in central-west Crete (Mavriyannaki 1973: 85-88, n° 1-2). A last unpublished and fragmentary example comes from an LM IIIA2 context in Quartier Nu at Malia32.

26Taking into account the presence of more LM IIIA1 and IIIA2 pottery (including the poppy capsule rhyton, champagne cups, shallow bowls, short-stem kylikes, plain pyxides, ladles) as well as LM IIIB vases in zone 5, it cannot be excluded at this point that this area was continuously occupied during LM III. This needs to be tested in the future, however.

2.3. Zone 6

27Even more than zone 5, zone 6 was very exposed to erosion which explains the bad conservation of the ceramic material. Analysis of its assemblages is still needed to define whether the material found was re-deposited material or whether primary deposits were destroyed by erosion. For instance, several joining sherds of a finely decorated stirrup jar were found directly to the south of wall F1 and make up an almost complete profile, but the breaks of the sherds are seriously eroded (fig. 9.20).

28A final observation for closing this preliminary analysis of Late Minoan ceramics at Sissi concerns the presence of coarse transport/storage octopus stirrup jars in the LM III buildings. Several isolated sherds were found in a superficial disturbed context in zone 6 (08-06-2002-OB001), while good primary destruction deposits in rooms 3.1 and 4.9 sealed some well-preserved even if incomplete examples (fig. 9.21). These containers belong to the last phase of occupation in this area. As noted above, the production and consumption of octopus stirrup jars are usually seen against the background of a LM IIIA2-IIIB Post-palatial economy (Haskell 2005). All Sissi examples examined until now are made in the same pale buff fabric and are decorated with a red to dark brown paint. At Malia, in Quartier Nu, some similar transport stirrup jars are inscribed with Linear B while a petrographic analysis conducted on some examples (both inscribed and not inscribed) suggested a South-central Crete provenance, from the region between the Eastern Mesara and the village of Myrtos33. With this in mind and taking into account the apparent importance of Sissi during the LM IIIB period, the continuation of excavations and study could put this coastal settlement as an important node in the regional exchange network34.