Version classiqueVersion mobile

The Human Face of Radiocarbon

 | 
Zoï Tsirtsoni

Rhodopes

Chapter 11. Late Chalcolithic Tatul

Krassimir Leshtakov, Nadezhda Todorova et Vanya Petrova
Traduction de Tatiana Stefanova

Texte intégral

Prolegomena2

  • 2 The Introduction (Prolegomena), the paragraph about the Late Chalcolithic layer – stratigraphy and (...)

1Tatul is one of the most popular prehistoric sites in Bulgaria. This is a result not only of its “megalithic” nature and the beautiful landscape, but also of the recent legend – very effective in terms of promotion – that this is the place where Orpheus was buried (fig. 1). Far from aiming to discuss those speculations, we are dealing here with an important problem of prehistory related to the 4th millennium BC. Tatul offers important data, which, once meticulously analyzed and interpreted, can help correctly formulate a number of questions, even if they do not yet provide their answers.

  • 3 Todorova 1986, p. 225‑227; Georgieva 1987.
  • 4 Özdoğan 1991, p. 219‑220, fig. 1; Draganov 1998, p. 217‑218, fig. 4; Parzinger 1993, p. 199; Parzin (...)
  • 5 For the religious hypothesis, see Raduncheva 2003, p. 163‑178; for the others, see Tsirtsoni, chapt (...)
  • 6 N. Ovcharov was the Director of the archaeological excavations; D. Kodzhamanova and Z. Dimitrov wer (...)

2After it became obvious that there is a hiatus in the sequence of all tells excavated so far in the Thracian plain, corresponding to a break in the occupation over a longer or shorter period in the 4th millennium BC, archaeologists concentrated their efforts on the quest for a settlement or a cemetery earlier than the Ezero A horizon, eventually dated back to the first half of the 4th millennium BC. The so-called “Transitional” period was documented with some degree of certainty in North Bulgaria3, but the region situated to the south of the Balkan range did not provide the same evidence. It was certain that no firm data were available from Upper Thrace. It was suggested that some indications from Eastern Thrace existed, but they were based on parallels between a few potsherds and the pottery yielded by the submerged Late Chalcolithic settlement in Sozopol4. Numerous hypotheses were proposed to answer the question: what caused this lack of information? They ranged from aggressive steppe invasions and the migration of large groups of people driven by cosmological and religious motivations, to climatic shifts and catastrophes5. Quite naturally the more rational colleagues directed their expectations to the neighbouring mountains, especially the Rhodopes. It was assumed that this relatively low mountain, with high diversity among its landscapes, would provide more suitable conditions for a continuation in settlement, economy and demography. Tatul was one of the sites which was expected to provide information for filling the gap in our knowledge on this subject. This is why we accepted, a few years ago, the invitation to take part in the archaeological excavations at Tatul6. Regretfully, it was not possible to excavate the Chalcolithic layer completely, due to circumstances that are beyond our control, and for that reason the results cannot be regarded as conclusive. Despite this fact, the archaeological excavations carried out in 2004‑2005 and in 2007 provided important information, which deserves to be presented to the academic community.

Fig. 1 – Tatul, general view from the north.

  • 7 We are not going to comment on the widespread thesis, according to which each rock formation, each (...)

3Considering the context of the Late Chalcolithic occupation of the Eastern Rhodope Mountains, a certain progress is evident in recent years in the archaeological map of the region (fig. 2: 1). Although a reliable stratigraphic sequence, or a locus classicus, is still not available, a considerable number of settlements (and according to some archaeologists, a great number of sanctuaries as well7) were added on the archaeological map. The eastern part of the Rhodopes is characterized by a low-mountain relief, mild climate, and abundant resources of timber, water, copper and gold. The temperatures are typical for the transitional Mediterranean area, and the soils along the larger river valleys are suitable for cultivation. These conditions favour a well-developed settlement network, different from those in Upper Thrace or on the southern slopes of the mountain and the Aegean coast. It is obvious that this network has its own specific features, which are yet to be studied; however, some of these features can be already summarized below.

  • 8 Leshtakov 1997, p. 130‑131; Todorova 2004, p. 342‑344.
  • 9 Raduncheva 2003, p. 194.
  • 10 Sites of Orlitsa and Varhari: see chapters 12 and 13 in this volume.
  • 11 Georgieva et al. 1999‑2000, p. 25; Borislavov 2001a.
  • 12 Borislavov 2001b, p. 79.
  • 13 Detev 1969.
  • 14 Detev 1966, p. 29.
  • 15 Gerdzhikova 1994, p. 76 (with references).
  • 16 Kamarev 2005, p. 373.
  • 17 Nehrizov 1994, p. 81.
  • 18 Ovcharov et al. 2004, and personal observations.

4First, there are no large tells in this region – not only along the mountain river valleys but also along the Maritsa River, flowing to the east of the Rhodopes, after its meander at the town of Simeonovgrad8. The only exception is Tell Sedlare on the Varbitsa River; it is 8‑9 m high and was inhabited during the Neolithic and the Late Chalcolithic9. Two Chalcolithic settlements were excavated within the frameworks of recent rescue excavation projects near Momchilgad and the village of Kirkovo10, and a third one on the Middle Arda River11. Field surveys or trial excavations provided information about several other sites in the same area12. Evidence from the northern foothills of the Rhodope Mountains was provided by the research carried out by P. Detev13; special attention has to be paid to the settlement situated on the bank of the Mechka River, near the town of Parvomaj14. The small-scale excavations carried out at the Pchelarovo hill settlement on the northeastern part of the mountain also yielded Late Chalcolithic artifacts15. Late Chalcolithic pottery and small finds were found during the rescue excavations at near the Alevi tomb (türbe) at the village of Zvinitsa16, whereas the remains of a Late Chalcolithic house were discovered during archaeological excavations at the village of Chomakovo17. Finally, the archaeological excavations at Perperikon, a site situated on a high hill above the Perpereshka River, also yielded pottery dating back to the end of Late Chalcolithic18.

Fig. 2. 1 – Map of the Eastern Rhodopes with the Late Chalcolithic and Final Chalcolithic sites mentioned in the text.

Fig. 2. 2 – Tatul, an orthophoto image.

  • 19 Delchev 2006, p. 131‑132.

5It is also worth mentioning the caves in the Eastern Rhodopes, some of which were visited in the Late Chalcolithic. Short galleries and low ceilings are typical features of these caves. In general they are not so attractive as the caves situated in the Western Rhodopes, such as Yagodina Cave, Haramiiska Dupka, or Gorni and Dolni Razh near the village of Trigrad. Archaeological excavations have so far been carried out in only one of the East Rhodopean caves, Samara Cave near the village of Ribino. It is situated in the valley of the Krumovitsa River. Although the pottery yielded by the excavations is not abundant, it provides enough grounds for the relative dating of the site19.

  • 20 The information is based on the results from an archaeological field survey carried out in 1983 by (...)
  • 21 Unpublished information from the excavations at Perperikon. The excavations were directed by N. Ovc (...)
  • 22 Shukerova 1996, p. 46.

6There is also evidence collected by field surveys. A Late Chalcolithic settlement was located in a small river valley near the thermal springs at the town of Dzhebel20. A second large settlement was found in the valley of the Perpereshka River, situated ca 5‑6 km upstream from Perperikon21. Chalcolithic pottery sherds were collected near the villages of Dolna Chobanka and Chomakovo22. Also, when our team visited the Harman Kaya site near the hamlet of Gasak in the village of Bivolyane, Momchilgrad region, we collected potsherds which, although limited in number, were typical for the end of the Late Chalcolithic period, and thus as a whole contemporary to the pottery from Tatul.

7As already said, the available information is still not sufficient for a precise description of the settlement pattern; nevertheless, some typical features can be defined. From a topographical point of view, the sites located so far can be classified as follows: caves, sites on river terraces (similar to the low tells), sites on relatively high rocky hills in the river valleys, and sites on high peaks, sites on naturally protected hill tops with steep slopes and cliffs. Tatul falls into the last group.

  • 23 See supra, n. 7.

8The presence of Late Chalcolithic pottery at Tatul has been known for a long time. Potsherds have been collected for many years on the steep slopes of the Kaya Basha (Kaya Başı) hill or yielded by treasure-hunter trenches. During the first excavations at the site carried out by I. Balkanski, the Chalcolithic layer was apparently not reached and the related pottery was not even mentioned in the field diary. There are no photos, drawings or any descriptions; there is no information about the context in which the Chalcolithic potsherds were found. Notwithstanding the lack of evidence or any logical argument, the site was claimed to be a Late Chalcolithic sanctuary23. The followers of this suggestion were very much impressed by the sculpted rocks on the Kaya Basha hill. Their imagination turned them into wandering tortoises, posing large felines, wriggling snakes and other reptiles, which are well known characters of the theriomorphic myths.

  • 24 Renfrew 1985, p. 2.
  • 25 Lin (pl. linove), i.e. large wooden or stone trough typical for the traditional Bulgarian culture, (...)
  • 26 According to the anthropological information collected from locals, many of the so-called Thracian (...)

9It is obvious that without archaeological excavations it is impossible to establish the period in which the rock carvings were made. However, often the desired is accepted as reality and a syndrome, which I will take the liberty to name “the Renfrew syndrome”24, is developed as a result. In many cases, beds cut into the rocks for stone walls have been interpreted as “stairways to heaven”, stone-carved winepresses (linove25) as sarcophagi, negatives of millstones as solar discs, and holes from stone-cutting awls as ritual stones with special holes for divination…26. It has to be explicitly stated that our archaeological excavations did not reveal any stone-carved features covered by a Chalcolithic layer that would have supported a date at the very end of the 5th or the first half of the 4th millennium BC.

The Late Chalcolithic layer. Stratigraphy and features

  • 27 The name of the hamlet is also mentioned in the publications as Veznitsa or Vezenitsa. There is als (...)
  • 28 Geographic coordinates 41° 32’ 29.96” N and 25° 32’ 43.72” E. The highest cliffs stand ca 400 m asl (...)

10The Kaya Basha hill, situated on a high promontory to the south of the Vezhnitsa hamlet27, dominates the valley of the Nanovitsa River (a right tributary of the Arda River) and is surrounded by a number of other well-defined high hills28. The place is naturally protected, and it is not surprising that it was chosen as the location of a fortified residence built in the medieval period (fig. 1; 2: 2).

  • 29 Balkanksi 1978, p. 10.

11The first archaeological excavations at Tatul in 197529 were preceded by reports noting that there were plenty of trenches on the plateau at the Kaya Basha hill. Some of them were made for removing stone blocks from the ancient walls, but others attest the activity of treasure hunters or amateur archaeologists (representatives of the local intelligentsia prior to the World War II). One of these trenches was made in a loose soil – the fill of a pit dating back to the Roman period or Late Antiquity. In 2005 it was found that the pit cut through the Late Chalcolithic layer, reaching the bedrock. Two years later we were able to document this layer and part of the related features.

  • 30 Ovcharov et al. 2008a, p. 35, fig. 1.
  • 31 The positions of the archaeological features and finds are designated by their location measured fr (...)
  • 32 The upper boundary of the Chalcolithic layer was documented at 391.70/391.65 m asl in squares F3-4 (...)
  • 33 Leshtakov & Tsirtsoni, forthcoming.

12The total area on which the Chalcolithic layer was recorded did not exceed 50 m2. Approximately one half of the area was thoroughly excavated down to the bedrock30. A thin layer of yellowish clay with small pieces of eroded rock, which did not yield any artefacts, was registered on top of the rock. This layer was reached at the bottom of pit 33 in squares E3-F3, at 391.50 m above sea level31. Its maximum thickness was ca 0.10 m. The clayey layer was overlaid by a layer of brown heterogeneous soil, yielding ceramic vessels and finds in situ. The sediment was relatively light in colour and mixed with a high number of small lumps of daub32. Burnt animal bones, as well as small white stones or single larger stones, were found in some places. On top of this layer was found the highly burnt debris of a big building, excavated over an area of ca 50 m2. The large blocks of fired daub from the building walls were covered by a thin layer of relatively loose brown soil, containing a high concentration of small pieces of fired daub, to which it owes its colour. The layer yielded a large number of potsherds and animal bones; a grinding stone was also discovered (fig. 3). It was sealed by a layer of very compact dark brown soil mixed with pieces of eroded rock. A large double hearth dated to the Bronze Age was discovered on top of the latter, at 392 m asl. Therefore, this is the hiatus that separates the Chalcolithic deposits from the upper cultural layer, dating back to the 3rd-2nd millennium BC33.

13An almost identical situation was provided by section 2 in square F4. The layer between 392 m and 391.85 m asl consisted of a brown, homogenous, fine and relatively loose soil. The pottery yielded dates back to the Bronze Age. A layer of compact dark brown clayey soil mixed with fragmented small stones and gravel was documented from 391.85 m asl downward; its thickness varied from 0.15 to 0.25 m. The pottery, which dates as a whole from the Late Chalcolithic period, was highly fragmented and some of the fragments had eroded surfaces. The high concentration of chipped-stone artefacts is worth mentioning. The colour of the soil gradually became lighter with increasing depth, whereas its structure became more homogenous and the concentration of small lumps of fired daub increased. At its base the compact fire-destruction layer of the building mentioned above was unearthed again (fig. 4).

Fig. 3 – Excavation in square F4. 1: concentration of pottery, stones and animal bones under the burnt debris; 2: ceramic vessel fragmented in situ with a cattle bone inside.

Fig. 4 – Field drawings from square F4. 1: Final Chalcolithic burnt building debris, the upper level; 2: concentration of pottery sherds, stones and animal bones under the burnt debris; 3: section of the Final Chalcolithic layer in sq. F4.

14The remains found at the site belong to walls built from clay and wood mainly, although a considerable number of small and middle-sized stones were also used. They were fired at a high temperature. Some of the fired pieces were large and had nicely smoothed surfaces. They were heavy and the clay was sandy, tempered with a great number of tiny stones, but almost no organics. Imprints of wooden posts up to 0.10 m in diameter were visible on some of the pieces, as well as imprints of wooden beams with rectangular sections. Large rounded and middle-sized crushed stones, fragments of grinding stones, as well as smaller stones and gravel, were found within the compact massive debris and the layer below, whereas several larger rock pieces (up to 0.50‑0.60 m long) were unearthed below the debris and at the bottom of the Chalcolithic layer. It clearly indicates that at least one of the walls was built with the compacted clay technique (pisé), and that, besides wooden beams and posts, stones of various sizes were also used in the construction, with larger ones at the foundation. The position of the clay fragments (on their face) allows the assumption that the wall had collapsed from northeast to southwest.

15Some of the daub pieces are lighter, due to the high amount of organic inclusions in the clay. They are more friable but also have nicely smoothed surfaces. Clay “slabs” 4‑5 cm thick and covered with fine plaster were also discovered. Single middle-sized (0.10-0.20 cm) rounded and crushed stones were found under the pieces of fired daub, and sometimes between them. The lightest fragments probably come from the outer frame of various structures, with nicely finished surfaces. They have rectangular sections with rounded edges and are 0.10‑0.15 cm thick; they are made from organic-tempered clay. Fragments bearing imprints of wattle or thin branches were not discovered.

16Here we should mention a particular feature of the stratigraphical sequence, which gives grounds for different hypotheses. The layer immediately on top of the fired debris and between the pieces of fired daub was heterogeneous; at some spots the soil was loose and brown in colour, with thin layers of more compact clayey soil and many small lumps of fired daub. This layer captures the attention, because it yielded a large number of small finds, as well as several large vessels found in situ. Fragments of a deep bowl with a neck and a vertical handle were discovered to the east of the compact debris in square F4, at 391.50 m asl. A concentration of potsherds was discovered in situ at the same level (391.55/391.50 m asl) in section 2. The above elements allow us to suggest that the building either had an attic fit to live in, or a second floor. An alternative suggestion is a stair-like construction whose floor would be adapted to the slope. If we assumed that it was a two-storey building, it would be logical to attribute pieces of lighter walls to the second floor, but there are no explicit stratigraphic proofs supporting this. In addition, the two-storey hypothesis does not seem very probable because the building was comparatively small and no clear traces of floor construction were preserved, but on the other hand, the slope was not steep enough to construct a stair-like building. In either case, the finds discovered in situ on top of the ruins and the general stratigraphic situation, provide grounds for multiple interpretations.

Pottery

17The majority of the ceramic vessels and sherds presented in this article come from a reliable context. They were discovered within the burnt building and in stratigraphically connected, tight concentrations. The greatest part of pottery was fragmented, but belongs to recognizable types. However, some of the pottery was secondarily fired and does not provide complete technological information.

Technology

18Sherds belonging to the fine ware group were very well made and their walls have a uniform thickness. The pots were probably formed by using moulds, as well as the so-called primitive potter’s wheel for the final stages. The ceramic vessels were well fired in general, a fact indicating high skills and a long tradition of pottery-making.

19The dominant ware is brown-coloured, in different shades and nuances, varying from light beige to dark brown. Considerably fewer are the gray-black and black wares. Usually the surface colour is even. Most of the vessels were made from medium-fine to medium-coarse clay, mainly tempered with minerals: fine quartz sand and tiny quartz inclusions, but sometimes chamotte and fine organic materials were also added.

20The surface of the majority of the vessels is coated by a finer or thicker slip or self-slip, which is nicely finished, most often matt and evenly burnished. Sherds with a shiny burnished surface are less numerous. The coarser wares, traditionally designated as kitchen-wares, often have unevenly smoothed or unevenly burnished surfaces. The rustication of the exterior by fine scratching (Besenstrich) is particularly typical. The storage vessels are often decorated with the so-called organized barbotine.

  • 34 Todorova, chapter 10 in this volume.

21The reddish-brown slipped ware, with matt burnished or fine rusticated surfaces and brittle cores, is typical for the pottery assemblage. A similar technological group has been identified in Dolno Dryanovo, where it comprises a large amount of the pottery assemblage34. A smaller, although distinctly identifiable, technological group comprises the fine light yellowish-ochre ware (mainly bowls) with matt, “soft”, porous surfaces. The pots are light, and made from clay tempered with white (limestone?) inclusions, black sand and chamotte.

Decorative techniques

  • 35 It seems probable that the technique of graphite-painting, as well as the raw material (graphite) w (...)

22Almost one third of the ceramic vessels and sherds are decorated. Various decorative techniques are represented, often in different combinations. The fine ware is more abundantly decorated, with the percentage of the graphite-painted sherds being the highest one. The graphite-paint is matte, gray-whitish and lacks the typical “metallic” lustre. Since in most cases the pattern is very poorly preserved and sometimes wiped-off to such an extent that it can only be detected as a negative image, it is very difficult to reconstruct the decorative themes35.

23Crusted decoration is common; the colour range comprises various shades of red, from bright carmine red to orange-red and various shades of bordeaux-red. Yellow, ochre, or cream (ivory) coloured motifs are also common. The pigment was applied on the pot surface after firing; most often it was laid in a very thin layer on a slightly rusticated surface and formed fields or bands with different shapes. But there are also sherds on which the pigment was laid in a thicker layer, creating a slight relief.

Morphological characteristics and parallels

  • 36 Todorova 1986, p. 110‑112.

24From a formal and typological point of view, the pottery assemblage is homogenous and does not show great diversity. Shapes include bowls, deep bowls, jars, cups and lids. The assemblage is dominated by hemispherical, spherical, or sphero-conical, usually simple open shapes with smooth profiles. The Late Chalcolithic vessels with a sharp carination typical for Northern Thrace36 are not represented at Tatul.

  • 37 Todorova, chapter 10 in this volume, fig. 8: 1, and the referenced parallels.

25The conical bowls with thickened rims and ca 30 cm diameters are the most numerous (fig. 5: 1‑4; 6: 3, 5). Deeper examples with an indentation on the exterior are typical for the site. Some of them are lavishly decorated on both sides with graphite-painted and crusted patterns. In the centre of the ornamental patterns are circles or spirals, consisting of parallel graphite-painted lines and wider bands coloured with cream/ivory or red pigment. The thickened rim and the exterior of the upper part of the body are covered with ochre-painted triangles and metopes of oblique parallel graphite-painted lines. On the exterior, the composition is completed by four symmetrical vertical fields, framed by plastic bands, or bands of thick ochre pigment (fig. 5: 1, 2). Such spiral-shaped and concentric motifs are also known from Dolno Dryanovo, Sozopol, Krivodol, etc., but the compositions there are simpler and restricted to the interior37.

26The bowls with in-turned and sometimes thickened rims (fig. 6: 8, 10‑13), as well as the bowls with a short cylindrical or slightly S-shaped upper part (fig. 6: 2, 6, 7‑9), are also very typical. Their decoration is modest and consists of simple graphite-painted patterns, plastic knobs and horizontal plastic bands.

  • 38 Dolno Dryanovo, Yagodina Cave, Klise Bair, Bezhanovo-Banunya, Devetaki Cave, Hotnitsa-Vodopada, Maa (...)
  • 39 Ilcheva 2009, p. 217, fig. 56: 1; p. 218, fig. 57: 1, 4; p. 228, fig. 67: 1, 7; p. 240, fig. 79: 7.

27The deep bowls provide the greatest diversity of types and variants. The typical deep bowls have globular-conical bodies with rounded carination and slightly everted thickened rims. The most numerous are the relatively small deep bowls with a rim diameter varying between 18 and 22 cm. The vessel walls are thin, slipped and matt burnished, sometimes with a rusticated exterior on the lower part. The majority of deep bowls are graphite-painted. The decorative patterns are simple, mainly linear: the most common consist of groups of 7‑10 oblique narrow graphite-painted lines (fig. 5: 5, 6; 7). In fact, these bowls are the most typical feature of the Final Chalcolithic pottery repertoire in the Balkans. Deep bowls with the same shape and decoration are found in a number of sites in the Rhodope Mountains, the Aegean region and present-day North Bulgaria, and all of them date back to the same period38. Such bowls are not present in the pottery repertoire of the “classic” phases of the Late Chalcolithic in Thrace. In contrast to specimens from the other sites, the graphite-painted patterns on the Tatul ceramic vessels are sometimes complemented by triangles with ochre pigment (fig. 5: 5, 6). The decoration on one of the bowls (fig. 7: 5), with oblique graphite-painted lines overlying wide horizontal channels, is typical for the pottery yielded by sites to the north of the Stara Planina Mountain39.

Fig. 5 – Final Chalcolithic pottery.

Fig. 6 – Final Chalcolithic pottery.

  • 40 I would like to thank K. Leshtakov, who was responsible for the prehistoric pottery and finds disco (...)
  • 41 No analysis is carried yet, thus it is not possible to determine the origin of pigment – organic or (...)

28Special attention has to be paid to the deep bowls with flat oval lugs on their rims. The shape is represented only by fragments. The lugs were probably two or four (fig. 7: 7‑9, 11, 12). It is a specific element, which has parallels so far only at Perperikon40. Faded traces of red ochre are still visible on one of these deep bowls (fig. 7: 9), and another one was painted with a thick layer of black pigment on both sides before firing. The motifs consist of parallel lines, which were not painted very precisely and were probably applied with a thick brush (fig. 7: 12). The black painting is exceptional for the Tatul pottery, and, for the time being, it is not known from other Chalcolithic sites in Thrace and the Rhodopes41.

  • 42 The pot was found fragmented in situ within the boundaries of the building in square F4, at 391.15  (...)
  • 43 Chapter 10 in this volume, p. 179, fig. 10; chapter 14 in this volume, p. 259, fig. 9.

29The next group comprises hemispherical vessels with short necks. The deep S-shaped bowls with a hemispherical or squat spherical body and a short, slightly flared neck are the most numerous ones (fig. 5: 7; 8). Some of the vessels have vertical strap handles, usually rising above the rim (fig. 8: 3; 9: 4), while others have a pair of vertical loop handles attached to the maximum diameter. The smaller thin-walled vessels are decorated with standard patterns: groups of parallel graphite-painted lines, or broad bands and triangles coloured with ochre (fig. 8: 2‑4). One of the best-preserved specimens deserves special attention, because of the abundant decoration of graphite-painted motifs, red pigment, a couple of plastic knobs and four groups of oval finger impressions (fig. 5: 7). The neck of the bowl is decorated with alternating triangles cross-hatched with graphite-painted and ochre lines; the body is decorated with S-shaped motifs and festoons, the latter consisting of three narrow graphite-painted lines and flanked with broader ochre painted bands42. However, most of the specimens of this type were not decorated, and the larger ones were less carefully made; their surface is rough, unevenly burnished with apparent stripes, or rusticated (Besenstrich). The S-shape bowls have exact parallels in the Dolno Dryanovo and Yagodina pottery43 and can be defined as typical for the Final Chalcolithic pottery assemblages of the Balkans in general (Sozopol, Rebarkovo, Hotnitsa-Vodopada, Klise Bair, etc.).

Fig. 7 – Final Chalcolithic pottery.

Fig. 8 – Final Chalcolithic pottery.

Fig. 9 – Final Chalcolithic pottery.

30The closed shapes represent a smaller part of the pottery assemblage. A number of sherds allow the reconstruction of jars with conical necks (fig. 10: 2) and probably large pots with vertical handles attached on their bellies. Several decorated sherds can be attributed to closed vessels, but no precise shape can be reconstructed (fig. 9: 8, 9, 11‑13).

  • 44 Todorova & Matsanova 2000, p. 338 (with references).
  • 45 Georgieva 1988, p. 123, fig. 13: 8.

31Storage jars are represented by two vessels of almost identical shape: a tall rounded body with a conical neck and a pair of vertical handles on the belly. The shape is known from Late Chalcolithic pottery assemblages, e.g. at Telish III, Yunatsite44, and the Galatin house45. The vessels from Tatul have more elongated proportions, as well as larger handles, in contrast to the parallels mentioned above. The surface is covered with horizontal barbotine lines. One of the pots (fig. 10: 5) was found fragmented in situ among and on top of large pieces of fired debris in square E5, whereas the second pot was retrieved from the partly excavated building in square F4 (fig. 10: 6).

  • 46 Unpublished, exhibited in the Stara Zagora Regional Historical Museum.
  • 47 Andrieşescu 1924, p. 86, fig. 14.
  • 48 Georgieva 2003, p. 231, fig. 7: 2.
  • 49 Chapter 10 in this volume, p. 181, fig. 13: 4, 5.

32The conical lids with two opposite oval openings (fig. 11: 2, 3) are known from the pottery assemblages of the last stages of the Late Chalcolithic in Thrace and have exact parallels in Starozagorski Mineralni Bani46, Sultana47, Tell Kozareva48, Dolno Dryanovo49, and others.

Fig. 10 – Final Chalcolithic pottery.

Decorative themes and stylistic features

33The decoration on the Tatul pottery requires special attention, for it presents a number of special features and well-defined stylistic particularities, which might be considered (at the present stage of research) both as cultural and as chronological indicators.

34The decorative patterns are dominated by simple linear motifs: groups of vertical, oblique or arch-shaped lines, cross-hatched triangles, etc.; the more complex motifs are spirals, ovals or festoons, consisting of groups of three or more parallel lines. In contrast to this general simplification of the motifs and the decorative patterns, some of the vessels are covered with very rich, even lavish decoration (fig. 5: 1, 2, 7; 9: 1).

  • 50 Very little is known about the Pchelarovo pottery and finds (Shukerova 1984, p. 77; Kamarev 2005). (...)
  • 51 Chapter 10 in this volume, p. 177, fig. 8.
  • 52 Draganov 1998, p. 207, fig. 2: 1, 2.
  • 53 Chohadzhiev 1991, p. 33, fig. 9: 11, 14.

35One of the most characteristic decorative features are groups of wider or narrower graphite-painted lines, combined with bands painted with red or whitish pigment (fig. 5: 2; 9: 8, 9, 11‑14). Similar patterns are known from Pchelarovo50, Dolno Dryanovo51, Sozopol52 and Vaksevo53.

  • 54 Chapter 10 in this volume, p. 181, fig. 12: 6.
  • 55 Draganov 1998, p. 207, fig. 2: 3, 4.
  • 56 Garašanin & Simoska 1976, T. VII, IX; Simoska et al. 1976, p. 75, T. III: 1.
  • 57 There are festoons in the lower part of the decorative compositions on a few ceramic sherds found a (...)

36A particularly popular feature, specific to this pottery assemblage, is the decoration on the lower part of the ceramic vessels with festoons in groups of four. The festoons consist of 3 to 7 narrow graphite-painted lines, flanked by bands painted in different colours. In all cases, they are complemented by relatively small plastic knobs (fig. 5: 7; 9: 1‑3, 5‑7). In stylistic terms, this pattern corresponds to the festoons found on ceramic vessels from Dolno Dryanovo54, Sozopol55, Šuplevac and Crnobuki56. However, no exact parallels for the organization or the execution of the decoration exist57.

37Another “standard theme” used in the decoration of the upper part of vessels consists of alternating triangles painted with a mineral pigment and triangles composed of narrow parallel graphite-painted lines (fig. 5: 1, 2, 4‑7; 9: 1, 10, 14).

  • 58 Todorova 1986, p. 112, 131.
  • 59 Georgieva 2003, p. 229, fig. 5: 1‑4.
  • 60 Andrieşescu 1924, p. 80‑81, fig. 7‑10.
  • 61 Dimitrov 2007, p. 149, fig. 4c; p. 151, fig. 5c, e; and personal observations.
  • 62 Valentinova, chapter 5 in this volume, p. 109.
  • 63 Ilcheva 2009, p. 225, pl. 64: 2, 3, 15.
  • 64 Chapter 10 in this volume, p. 179, fig. 11: 3.

38We should also point out some typical features related to the decoration of the coarser ware. The most common are plastic and barbotine decoration, followed by incision. The plastic bands with fingerprints, typical for the classic phases of the Late Chalcolithic, as well as the so-called “nail barbotine”58, are not present at all. The oval, single or double knobs, usually in pairs or groups of four, symmetrically arranged on the vessel’s maximum diameter, are the most common decorative element (fig. 9). The narrow plastic ribs, most often in groups of three (fig. 8: 10; 10: 4) or as frames of ornamental fields (fig. 5: 2), are also very typical. Such plain plastic bands are very common in the pottery assemblage of Tell Kozareva59 and Sultana60, but the closest parallels of the Tatul sherds are found in the pottery from the underwater excavations in Sozopol Bay61. The same plain plastic bands are found in Final Chalcolithic sites in present-day Northern Bulgaria (Bezhanovo-Banunya62, Shemshevo-Klise Bair63), but also in Dolno Dryanovo64.

  • 65 N. Todorova 2003; Gergov 2007a.
  • 66 Todorova & Avramova, chapter 14 in this volume, p. 259 and 260, fig. 10 (with references).

39Speaking more generally, the characteristic ornamental themes at Tatul differ considerably from those typical of the classic phases of the Late Chalcolithic in Thrace65. At the same time, the majority of the elements and stylistic features of the decoration are common for all Final Chalcolithic pottery assemblages in the Balkans and the North Aegean. For instance, the alternation of graphite-painted narrow parallel lines with wider bands and vertical cross-hatched metopes is also typical of the pottery yielded by Yagodina I, Devetaki Cave, Klise Bair, Perperikon, and others66. Simpler compositions are also the same, for example the alternating oblique parallel lines and hatched triangles on the upper part of open vessels. What is striking however at Tatul is the “standardization” of the ornamental themes, which suggests a well-established decorative style.

  • 67 On the terminology, see also in this volume chapter 10, p. 175, n. 20; chapter 14, p. 251, n. 9.

40In conclusion, it can be stated that the formal, typological and decorative characteristics of the Tatul Chalcolithic pottery differ considerably from those of the Late Chalcolithic pottery assemblages in Upper Thrace. The parallels are few, and they are related to the less indicative and less typical shapes or decorative patterns. Closer parallels are always connected to the latest phases of the “classic” Chalcolithic cultures known from Starozagorski Mineralni Bani, Tell Sultana, Tell Kozareva, etc. At the same time the closest parallels for the Tatul pottery come from sites dated to the final phases of the Balkan Chalcolithic67. Although similarities in shapes can be detected on a broader territory, the closest parallels come from sites located in the Eastern Rhodope Mountains, Perperikon and Pchelarovo. In contrast, analogies with the pottery assemblages from Western Rhodopes (Yagodina Cave, Dolno Dryanovo) are not that close.

Small Finds

41The small finds retrieved from the Chalcolithic layer at Tatul comprise several artefacts, grouped according to their material: stone, chipped stone, and clay.

Stone artefacts

42The only find that can be attributed to this period, based on a reliable stratigraphic context, is a quartz polisher with clearly visible use-wear traces.

Chipped-stone artefacts

  • 68 For further details about the chipped-stone collection, see Zlateva-Uzunova, forthcoming.
  • 69 Ilcheva 2009, p. 42‑45, pl. 1: 1‑6; pl. 7‑8.
  • 70 Valentinova & Gushterakliev 2008, p. 116, fig. 3.
  • 71 Gergov et al. 1985.

43The chipped-stone collection comprises 364 artefacts68. The raw material supply was, in general, based on local sources. Macroscopic analysis provided information about 23 varieties of chalcedony, opal-chalcedony, jasper, opal, zeolite, quartz, agate and flint. The most common tool types are the retouched flakes and the end-scrapers. There are also combined tools and backed tools with retouched blades and side-scrapers. The arrowheads are chronologically the most diagnostic artefacts; three complete specimens, a blank and three fragments were studied (fig. 12). Two of the arrow-points have a concave base, and the third one has a slightly convex base. This group has exact parallels in the collections from a number of Final Chalcolithic sites in Northern Bulgaria: Hotnitsa-Vodopada, Shemshevo-Klise Bair, Veliko Tarnovo-Kachitsa D69, Bezhanovo-Banunya70, and Telish71.

Clay artefacts

44The collection comprises spindle-whorls and a fragment of a loom-weight.

  • 72 Three of the artefacts came from a reliable stratigraphic context, a layer of burnt building debris (...)

45Four spindle-whorls can be dated back to the Late Chalcolithic; all of them were found in square F472. Two types can be distinguished: the disc-shaped and the short conical type, with a concave-conical sub-type. The first one is represented by a fragment of a spindle-whorl found in the building debris (fig. 11: 5). The second is represented by one specimen with a flat base, and two specimens with concave bases (fig. 11: 4, 6, 7). The “short conical” definition needs further specification in order to differentiate these artefacts from those typical for later periods: the Chalcolithic spindle-whorls have curved sides (most often concave but also sometimes convex), and the edge is rounded; the concave base is an additional specific feature. Specimens of both types are small, undecorated and weigh between 10 and 15 g. They are made from medium-fine clay, unevenly smoothed, without any further surface treatment. The first type has a very simple, non-diagnostic shape, and is difficult to date outside the context in which it was found. The second type is more interesting because, based on the available information, it can be considered as a diagnostic shape with limited chronological distribution.

Fig. 11 – Final Chalcolithic pottery (1‑3), spindle whorls (4‑7) and a loom-weight (8).

Fig. 12 – Flint arrowheads.

  • 73 I would like to thank N. Ovcharov (NIAM-BAS) and D. Kodzhamanova (Kardzhali Regional Historical Mus (...)
  • 74 Raichev 1991; Todorova & Avramova, chapter 14 in this volume, p. 265, fig. 13: 1‑9.
  • 75 Pernicheva 2000, fig. 12.18: 1‑6.
  • 76 Georgieva 1993a, p. 10, fig. 6: 9‑10.
  • 77 Vandova 2007, fig. 5: 1‑4.
  • 78 Garašanin & Simoska 1976, p. 17, pl. VIII: 26‑27; Simoska et al. 1976, p. 53, pl. IV: 6, 9.
  • 79 Trantalidou et al. 2005a, p. 50, fig. 2.
  • 80 Christmann 1996, p. 305‑306, pl. 162.
  • 81 Wace & Thompson 1912, p. 53.
  • 82 Adrymi-Sismani 2007, p. 77, pl. XII: s, and personal communication.
  • 83 Lambert 1981, fig. 393‑396, 258‑271.
  • 84 Immerwahr 1971, p. 18, pl. 15, no. 234‑236.
  • 85 Marinescu & Andreescu 2003‑2004, fig. 29: 4‑8.
  • 86 Haşotti & Popovici 1992, p. 40, pl. 9.

46Both types have exact parallels in their shape and size among the spindle-whorls from sites dated back to approximately the same period: Perperikon73 and Yagodina Cave74 in the Rhodopes, Kolarovo in Southwestern Bulgaria75, Negovantsi76 and Dervish Banya in Kyustendil77, Šuplevac and Crnobuki I in Pelagonia78, and Angitis Cave in Northern Greece79. There are also good parallels from other regions, such as Thessaly and Attica. Collections of similar spindle-whorls have been published from the Rachmani layer of Pefkakia Magoula80, from the site of Rachmani itself, including a group of spindle-whorls found in House Q81, and from the site of Mikrothives82. Disc-shaped and short conical spindle-whorls were found at sites in Attica as well, such as the Kitsos Cave near Lavrion83, the Ancient Agora of Athens84, and others. The lack of parallels from sites in Northern Bulgaria probably indicates a gap in the research, as parallels are known to the north of the Danube among the finds retrieved from the first archaeological campaigns at Piscul Cornisorului near Sălcuţa85 and among the spindle-whorls discovered in the Cernavoda I layer at Tell Hîrşova86.

  • 87 Raichev 1991, p. 51; Todorova & Avramova, chapter 14 in this volume, p. 265, fig. 13: 10, 11.
  • 88 Mikov & Dzhambazov 1960, fig. 70: c.
  • 89 Chapter 10 in this volume, p. 183, fig. 15: 3.
  • 90 Ilcheva 2009, p. 137, fig. 21b.
  • 91 Unpublished finds. Personal observations from the exposition in the museum of Volos.
  • 92 Papadopoulos et al. 2001, p. 58, 64, pl. 3.
  • 93 Elster 2003, p. 240, fig. 6: 19a, pl. 6: 9b, c; 6: 10a; 6: 11.
  • 94 Bernabò-Brea 1964, p. 658, pl. LXXXII: a.

47Besides spindle-whorls, the Tatul Chalcolithic layer yielded a unique fragment of a loom-weight: this is a small cylindrical artefact with a vertical opening and rounded edges (fig. 11: 8). It was made from coarse clay with a high amount of organic temper; it is badly fired, most probably in the fire which destroyed the house. Loom-weights similar in shape were found in sites dating back to the Late Chalcolithic in present-day Bulgaria and Greece. The closest parallels from Bulgaria are two specimens from Yagodina Cave87, and three more individual finds from Devetaki Cave88, Dolno Dryanovo89 and Hotnitsa-Vodopada90. A larger number of finds are known from Mikrothives in Thessaly91 and Agios Ioannis on Thasos92. The same shape also existed in the Early Bronze Age. It is worth mentioning that in the territory of present-day Northern Greece, the cylindrical loom-weights were used until the early phase of EBA – Sitagroi IV93 and Poliochni Blue94.

Relative and absolute chronology

48The characteristic features of the Tatul pottery assemblage and the parallels listed above provide grounds to believe that the site was occupied in the final phases of the Late Chalcolithic. This statement is convincingly supported by the radiocarbon dates. Three bone and antler samples discovered within the building in square F4 were analyzed. The first 14C date (Lyon-5510/SacA-13076) was obtained from a large mammal’s bone found inside the richly decorated deep bowl described above (supra, p. 197, and fig. 3: 2; 5: 7), together with a flint arrowhead (fig. 12: 1). The other two samples were taken from the same assemblage: one (Lyon-5621/SacA-13381) comes from a level with fragmented ceramic vessels and bones to the south of the above-mentioned bowl, whereas the other (Lyon-5509/SacA-13075) is from a tortoise shell found immediately above the bedrock.

49The samples were processed in the Radiocarbon Laboratory of UMR 5138 – Archaeology and Archaeometry at Lyon, France, within the framework of the “Balkans 4000” project. They were measured at the facilities of the Committee for Atomic Energy at Saclay (LMC14) with the AMS method, and provided three dates (table 1).

Table 1 – Radiocarbon dates from Tatul.

50As seen from this table, the dates form a relatively compact cluster, and two of them have almost identical values (5295 ± 35 BP and 5260 ± 35 BP). The third date has a higher value (5415 ± 35 BP), but the sample context does not provide reasons to believe that it reflects a real chronological distinction.

  • 95 Boyadzhiev 1998, p. 353‑354.
  • 96 It is important to note that dates with values around 5250/5200‑5150/5100 BP are also typical for t (...)
  • 97 Todorova, chapter 10 in this volume, p. 184‑186, table 2 and related comments.
  • 98 Distinguishing separate phases within the horizon 4200/4100‑3800/3750 cal BC is hampered by a serie (...)
  • 99 Hansen et al. 2008, p. 77‑79.
  • 100 Govedarica 2004, p. 72.
  • 101 Govedarica 2004, p. 82; Rassamakin 2011, p. 298. The dates mentioned are presented at 2σ as follows (...)
  • 102 Georgieva 2005, p. 154‑156.

51The dates from Tatul correspond well to the later subgroup within the frame of the second group of dates defined by Y. Boyadzhiev95, which covers a short time span at the very end of the Krivodol-Sălcuţa assemblage96. Dates with similar values have not been published so far from settlement mounds in Thrace and the KGK VI assemblage in general. This fact, together with the characteristics of the pottery assemblage discussed above, provide grounds to believe that the occupation at Tatul immediately followed the “classic” phases of the Chalcolithic in the Thracian Plain, thus falling into the same “final” stage, together with Dolno Dryanovo, Sozopol, Rebarkovo, Sălcuţa III and probably the beginning of Cernavoda I97. On the other hand, some specific features of the pottery assemblage and the date with the higher value from Tatul, can be considered as arguments in favour of a slightly earlier date for the site, at the very beginning of this provisional “chronological horizon”98. The published radiocarbon dates from Pietrele (Măgura Gorgana), which fall within the interval 4400‑4250 cal BC99, are even earlier and correspond to the latest stage of the “classic” Late Chalcolithic cultures. The parallels in the pottery allows the suggestion that Sultana Malu Roşu, Tell Kozareva and Starozagorski Mineralni Bani, already mentioned supra (p. 199), also fit the same time span. An interesting addition to this picture is provided by the date from burial 12 of the Decea Mureşului cemetery, which corresponds to the same period100. Other “ochre burials” from the territory of present-day Romania and Moldavia provide even earlier dates101. As far as the absolute chronology is concerned, these dates fully support P. Georgieva’s statement about a gradual penetration of ideas and/or people from the steppes in the East Balkan cultures, and an increase in contact in the late 5th millennium BC102.

  • 103 Tsirtsoni, chapter 15 in this volume, p. 280, table 2.

52The dates from Dikili Tash require special attention. In general, the long series of dates corresponds well to the dates from Tatul and Dolno Dryanovo. Two of the dates (from the latest phase of the Chalcolithic occupation) are completely identical to the ones from Tatul103.

Discussion

53Summarizing all the facts presented above, it can be stated that the excavated Late Chalcolithic remains at Kaya Basha hill belong to a burnt house, situated on a relatively flat area among the rocks on the southwestern slope. The debris remained intact since no buildings were constructed on this spot in the later periods. The Late Chalcolithic layer was sealed by sterile soil and overlapped by the Bronze Age features. This guarantees the stratigraphic position of the structures and makes the contextual information reliable. The bedrock rose gradually to the east and the northeast, reaching the cliffs, which were about 10 m higher, creating a barrier against the cold winds. Access to the occupied rock-platform was most probably made from the north, where the slope slanted. The present width of the platform is 3.50‑4 m and the length is ca 9‑10 m, restricting the maximum size of the building. It is impossible to make more precise conclusions, due to the erosion and the later disturbances. The direction in which the walls have fallen followed the inclination of the slope: the north wall fell to the southwest. It is worth mentioning that, in spite of the solid construction, no trampled or plastered floor was found. The large number of the small finds and potsherds found around and among the debris give clear indications that the fire broke out suddenly and the inhabitants did not have time to collect their items.

54It cannot be firmly stated that this was the only building on the hill. The area was densely built in later periods, and the earlier remains could have been removed during later construction activities. Although Late Chalcolithic ceramic sherds were found in all excavated areas, building debris were discovered in the southwest part of Kaya Basha only. It also has to be noted that there were no traces of any carving of the rock prior or contemporary to the building.

  • 104 Some archaeologists believe that the rich decoration with mineral (“crusted”) paints supports the c (...)
  • 105 Ovcharov et al. 2008a, p. 37‑43.

55Finally, it has to be emphasized that there is no reliable evidence about the existence of any ritual objects or structures. The available field data do not differ from those provided by regular settlements in Upper Thrace. The pottery and small finds form an inventory typical of an ordinary Chalcolithic house. Some of the ceramic vessels are lavishly decorated indeed, and it is tempting to interpret them as “special” – part of the paraphernalia of a sanctuary, or ritual vessels. However, this type of decoration is so common that it can be regarded as diagnostic for a specific chronological horizon, and the parallels come from utilitarian assemblages104. The same applies to the small finds. Those made from clay are related mainly to everyday life, namely spinning and weaving. There is no indication of an intentional deposition of objects, comparable to the examples from the Bronze Age layer at the same site105. The context in which the artefacts were found is perfectly ordinary, and does not provide reason to interpret the Late Chalcolithic site at Tatul in general, and the excavated house in particular, as a sanctuary. There are several suggestions about the function of the house, but it is difficult to provide strong arguments supporting any of them, as the house was not completely excavated. Besides, the adjacent area was not excavated, and the house cannot be placed in its cultural environment. Attention has to be paid to several strange facts. Despite the fairly solid construction of the house, no plastered or trampled floor was found; no heating structures or places for cooking food were excavated within the investigated area either. It can be claimed that we found only one of the rooms, and that the other(s), situated further to the west, had been destroyed by erosion. If that was the case, the missing heating structures could have been situated there. In either case, one negative fact is undisputable: no seasonality or cyclic variation related to the house was documented – an argument that supports its utilitarian function.

56The results from the pottery comparative analysis mark the place of Tatul at the beginning of the Final Chalcolithic in the Balkans. The same is supported by the absolute chronology. At the same time, the Tatul pottery assemblage displays a specific shape repertoire and decorative style, which characterize the culture of this stage of the Chalcolithic in the eastern part of the Rhodope Mountains. The particular features seen in the pottery seem to be the result of regional developments and local traditions, which maintained their specificities against the distinct trend toward a cultural unification. Probably this is part of the dynamic processes that developed in the Balkans during the late 5th and the early 4th millennium BC.

Notes

2 The Introduction (Prolegomena), the paragraph about the Late Chalcolithic layer – stratigraphy and features, and the Discussion, were written by K. Leshtakov. Pottery, and relative and absolute chronology were written by N. Todorova; small finds was written by V. Petrova. The flint artifacts were studied by R. Zlateva, to whom the authors would like to express their thanks. The text has been translated from Bulgarian by T. Stefanova..

3 Todorova 1986, p. 225‑227; Georgieva 1987.

4 Özdoğan 1991, p. 219‑220, fig. 1; Draganov 1998, p. 217‑218, fig. 4; Parzinger 1993, p. 199; Parzinger & Özdoğan 1995, p. 17.

5 For the religious hypothesis, see Raduncheva 2003, p. 163‑178; for the others, see Tsirtsoni, chapter 1 in this volume, p. 26‑27 (with references).

6 N. Ovcharov was the Director of the archaeological excavations; D. Kodzhamanova and Z. Dimitrov were Deputy Directors. The prehistoric layers were excavated under the direction of the authors of the current article. We are grateful for the possibility to publish the results related to the Late Chalcolithic occupation at Tatul. For preliminary reports, see Ovcharov et al. 2007, 2008a, 2008b.

7 We are not going to comment on the widespread thesis, according to which each rock formation, each rock-cut feature and each relatively high peak are sanctuaries (see Raduncheva 2003). Neither are we going to discuss the hypothesis stating that the megalithic culture emerged in the Chalcolithic, and even earlier. Such a discussion seems irrelevant, since no reliable evidence has been provided until now to support this theory.

8 Leshtakov 1997, p. 130‑131; Todorova 2004, p. 342‑344.

9 Raduncheva 2003, p. 194.

10 Sites of Orlitsa and Varhari: see chapters 12 and 13 in this volume.

11 Georgieva et al. 1999‑2000, p. 25; Borislavov 2001a.

12 Borislavov 2001b, p. 79.

13 Detev 1969.

14 Detev 1966, p. 29.

15 Gerdzhikova 1994, p. 76 (with references).

16 Kamarev 2005, p. 373.

17 Nehrizov 1994, p. 81.

18 Ovcharov et al. 2004, and personal observations.

19 Delchev 2006, p. 131‑132.

20 The information is based on the results from an archaeological field survey carried out in 1983 by a team directed by T. Spiridonov and R. Georgieva, and supported by the Institute of Thracology. The author was a member of the team. This is unpublished information and I would like to give thanks for the opportunity to use it.

21 Unpublished information from the excavations at Perperikon. The excavations were directed by N. Ovcharov; K. Leshtakov was a member of the team.

22 Shukerova 1996, p. 46.

23 See supra, n. 7.

24 Renfrew 1985, p. 2.

25 Lin (pl. linove), i.e. large wooden or stone trough typical for the traditional Bulgarian culture, which was used for pressing grapes in the wine making process.

26 According to the anthropological information collected from locals, many of the so-called Thracian cult-structures in the mountains were made recently, in the last centuries. The old generation still keeps memories of their purpose, “authors” and mode of usage.

27 The name of the hamlet is also mentioned in the publications as Veznitsa or Vezenitsa. There is also a completely fabricated toponym “the hamlet of Kayabashitsa”.

28 Geographic coordinates 41° 32’ 29.96” N and 25° 32’ 43.72” E. The highest cliffs stand ca 400 m asl. The easiest access to the site is along the north-northeast slope. The eastern and southern stairways documented during the excavations date back to a much later period and cannot be related to the prehistoric occupation. The rest of the slopes are very steep and there are almost vertical cliffs to the south-southeast. The ridge measures about 80 x 50 m (northeast-southwest).

29 Balkanksi 1978, p. 10.

30 Ovcharov et al. 2008a, p. 35, fig. 1.

31 The positions of the archaeological features and finds are designated by their location measured from sea level (in m asl = meters above sea level) and not from the surface, due to the complex topography of the site.

32 The upper boundary of the Chalcolithic layer was documented at 391.70/391.65 m asl in squares F3-4 and at 391.80 m asl in square E5. The trenches made to the north, northwest and east, that reached the bedrock did not provide evidence for a Chalcolithic layer in the rest of the hill area. The southern hill slopes were very steep and no cultural layer could be found there. Although Chalcolithic pottery sherds were discovered, they were few and did not come from a reliable archaeological context.

33 Leshtakov & Tsirtsoni, forthcoming.

34 Todorova, chapter 10 in this volume.

35 It seems probable that the technique of graphite-painting, as well as the raw material (graphite) which was used, were different from the ones used by the potters in other regions, e.g. in Northern Thrace and the Karanovo V‑VI sites, where the graphite-paint is thicker and more resistant and in most cases still preserves the typical metallic shine. However, any conclusion on this issue has to be based on a series of special analyses and studies; see also Todorova & Avramova, chapter 14 in this volume, p. 259, n. 24.

36 Todorova 1986, p. 110‑112.

37 Todorova, chapter 10 in this volume, fig. 8: 1, and the referenced parallels.

38 Dolno Dryanovo, Yagodina Cave, Klise Bair, Bezhanovo-Banunya, Devetaki Cave, Hotnitsa-Vodopada, Maaras Cave-Angitis, Thasos-Limenaria, the Cyclop’s Cave near Maroneia: the parallels are referred to and commented on in detail in the chapter about Dolno Dryanovo (chapter 10 in this volume).

39 Ilcheva 2009, p. 217, fig. 56: 1; p. 218, fig. 57: 1, 4; p. 228, fig. 67: 1, 7; p. 240, fig. 79: 7.

40 I would like to thank K. Leshtakov, who was responsible for the prehistoric pottery and finds discovered at the site, for the opportunity to study the Chalcolithic pottery from Perperikon and to use the information. A similar fragment was later identified at Dikili Tash; Tsirtsoni, chapter 15 in this volume, p. 296, fig. 27: d.

41 No analysis is carried yet, thus it is not possible to determine the origin of pigment – organic or mineral.

42 The pot was found fragmented in situ within the boundaries of the building in square F4, at 391.15 m in a layer of fine brown loamy soil with charcoal flecks and ashy streaks (supra, fig. 3: 2).

43 Chapter 10 in this volume, p. 179, fig. 10; chapter 14 in this volume, p. 259, fig. 9.

44 Todorova & Matsanova 2000, p. 338 (with references).

45 Georgieva 1988, p. 123, fig. 13: 8.

46 Unpublished, exhibited in the Stara Zagora Regional Historical Museum.

47 Andrieşescu 1924, p. 86, fig. 14.

48 Georgieva 2003, p. 231, fig. 7: 2.

49 Chapter 10 in this volume, p. 181, fig. 13: 4, 5.

50 Very little is known about the Pchelarovo pottery and finds (Shukerova 1984, p. 77; Kamarev 2005). I am indebted to D. Kodzhamanova (Kardzhali Regional Historical Museum) for the opportunity to study the collection.

51 Chapter 10 in this volume, p. 177, fig. 8.

52 Draganov 1998, p. 207, fig. 2: 1, 2.

53 Chohadzhiev 1991, p. 33, fig. 9: 11, 14.

54 Chapter 10 in this volume, p. 181, fig. 12: 6.

55 Draganov 1998, p. 207, fig. 2: 3, 4.

56 Garašanin & Simoska 1976, T. VII, IX; Simoska et al. 1976, p. 75, T. III: 1.

57 There are festoons in the lower part of the decorative compositions on a few ceramic sherds found at Perperikon.

58 Todorova 1986, p. 112, 131.

59 Georgieva 2003, p. 229, fig. 5: 1‑4.

60 Andrieşescu 1924, p. 80‑81, fig. 7‑10.

61 Dimitrov 2007, p. 149, fig. 4c; p. 151, fig. 5c, e; and personal observations.

62 Valentinova, chapter 5 in this volume, p. 109.

63 Ilcheva 2009, p. 225, pl. 64: 2, 3, 15.

64 Chapter 10 in this volume, p. 179, fig. 11: 3.

65 N. Todorova 2003; Gergov 2007a.

66 Todorova & Avramova, chapter 14 in this volume, p. 259 and 260, fig. 10 (with references).

67 On the terminology, see also in this volume chapter 10, p. 175, n. 20; chapter 14, p. 251, n. 9.

68 For further details about the chipped-stone collection, see Zlateva-Uzunova, forthcoming.

69 Ilcheva 2009, p. 42‑45, pl. 1: 1‑6; pl. 7‑8.

70 Valentinova & Gushterakliev 2008, p. 116, fig. 3.

71 Gergov et al. 1985.

72 Three of the artefacts came from a reliable stratigraphic context, a layer of burnt building debris. The fourth one was found very close to them in a secondary context (within the boundaries of a later pit); nevertheless, it was assigned to the same group based on formal and typological characteristics.

73 I would like to thank N. Ovcharov (NIAM-BAS) and D. Kodzhamanova (Kardzhali Regional Historical Museum), Directors of the Perperikon Excavation Project, for the opportunity to study the unpublished small finds from the site, kept in the Kardzhali Museum reserve.

74 Raichev 1991; Todorova & Avramova, chapter 14 in this volume, p. 265, fig. 13: 1‑9.

75 Pernicheva 2000, fig. 12.18: 1‑6.

76 Georgieva 1993a, p. 10, fig. 6: 9‑10.

77 Vandova 2007, fig. 5: 1‑4.

78 Garašanin & Simoska 1976, p. 17, pl. VIII: 26‑27; Simoska et al. 1976, p. 53, pl. IV: 6, 9.

79 Trantalidou et al. 2005a, p. 50, fig. 2.

80 Christmann 1996, p. 305‑306, pl. 162.

81 Wace & Thompson 1912, p. 53.

82 Adrymi-Sismani 2007, p. 77, pl. XII: s, and personal communication.

83 Lambert 1981, fig. 393‑396, 258‑271.

84 Immerwahr 1971, p. 18, pl. 15, no. 234‑236.

85 Marinescu & Andreescu 2003‑2004, fig. 29: 4‑8.

86 Haşotti & Popovici 1992, p. 40, pl. 9.

87 Raichev 1991, p. 51; Todorova & Avramova, chapter 14 in this volume, p. 265, fig. 13: 10, 11.

88 Mikov & Dzhambazov 1960, fig. 70: c.

89 Chapter 10 in this volume, p. 183, fig. 15: 3.

90 Ilcheva 2009, p. 137, fig. 21b.

91 Unpublished finds. Personal observations from the exposition in the museum of Volos.

92 Papadopoulos et al. 2001, p. 58, 64, pl. 3.

93 Elster 2003, p. 240, fig. 6: 19a, pl. 6: 9b, c; 6: 10a; 6: 11.

94 Bernabò-Brea 1964, p. 658, pl. LXXXII: a.

95 Boyadzhiev 1998, p. 353‑354.

96 It is important to note that dates with values around 5250/5200‑5150/5100 BP are also typical for the next group of dates, the third in Y. Boyadzhiev’s scheme, which refer to a later and different stage of the cultural development. The problem was commented on by Y. Boyadzhiev: ibid. Examples are provided by the dates from Haramiiska Dupka (Boyadzhiev 1995, p. 185; Todorova & Avramova, chapter 14 in this volume, p. 266, n. 75), Ostrovul Corbului, dated back to phase IIIb of Sălcuţa, nec. 2 (Radu 2003, p. 168, table 1: 5260 ± 60 BP), and from Peştera Ungurească-Bodrogkeresztúr, with values at 5260 ± 40 BP (GrN-29101) and 5350 ± 40 BP (GrN-29014): see Biagi & Voytek 2006, p. 189.

97 Todorova, chapter 10 in this volume, p. 184‑186, table 2 and related comments.

98 Distinguishing separate phases within the horizon 4200/4100‑3800/3750 cal BC is hampered by a series of objective difficulties. The excavated sites yielding artefacts dated to this period are few and distributed over a vast territory. The amount of stratigraphic information is scarce or completely absent; the published sites are even fewer, thus making conclusions on their relative chronology questionable. The attempts to distinguish phases based on absolute dates also face problems (see Boyadzhiev 1998, p. 352‑353).

99 Hansen et al. 2008, p. 77‑79.

100 Govedarica 2004, p. 72.

101 Govedarica 2004, p. 82; Rassamakin 2011, p. 298. The dates mentioned are presented at 2σ as follows: Decea Mureşului (KIA-368, 5380 ± 40 BP = 4335‑4215; 4202‑4136; 4122‑4085 cal BC); Giurgiuleşti (Ki-7037, 5560 ± 80 BP = 4560‑4240 cal BC), and Kainar (KIA-369, 5580 ± 50 BP = 4511‑4339 BC).

102 Georgieva 2005, p. 154‑156.

103 Tsirtsoni, chapter 15 in this volume, p. 280, table 2.

104 Some archaeologists believe that the rich decoration with mineral (“crusted”) paints supports the connection of pottery to ritual activities (Avramova 1992, p. 246; Gergov 2007a, p. 138). But this interpretation is partial, and lacks contextual analysis. The main argument rests on the non-durable character of those mineral substances, which would make the ceramic vessels unusable for household activities. However, other archaeologists note that the time-distance does not allow us to make final conclusions about the stability of the paints, and suggest that in the past their adherence on the vessels’ walls could have been far more stable (Vitelli 1999, p. 69). After all, from a theoretical point of view, the distinction between pottery used in everyday life and pottery used for rituals does not necessarily mean that the latter was involved in cult activities.

105 Ovcharov et al. 2008a, p. 37‑43.

Table des illustrations

Légende Fig. 1 – Tatul, general view from the north.
URL http://books.openedition.org/momeditions/docannexe/image/517/img-1.jpg
Fichier image/jpeg, 112k
Légende Fig. 2. 1 – Map of the Eastern Rhodopes with the Late Chalcolithic and Final Chalcolithic sites mentioned in the text.
URL http://books.openedition.org/momeditions/docannexe/image/517/img-2.jpg
Fichier image/jpeg, 92k
Légende Fig. 2. 2 – Tatul, an orthophoto image.
URL http://books.openedition.org/momeditions/docannexe/image/517/img-3.jpg
Fichier image/jpeg, 92k
Légende Fig. 3 – Excavation in square F4. 1: concentration of pottery, stones and animal bones under the burnt debris; 2: ceramic vessel fragmented in situ with a cattle bone inside.
URL http://books.openedition.org/momeditions/docannexe/image/517/img-4.jpg
Fichier image/jpeg, 284k
Légende Fig. 4 – Field drawings from square F4. 1: Final Chalcolithic burnt building debris, the upper level; 2: concentration of pottery sherds, stones and animal bones under the burnt debris; 3: section of the Final Chalcolithic layer in sq. F4.
URL http://books.openedition.org/momeditions/docannexe/image/517/img-5.jpg
Fichier image/jpeg, 332k
Légende Fig. 5 – Final Chalcolithic pottery.
URL http://books.openedition.org/momeditions/docannexe/image/517/img-6.jpg
Fichier image/jpeg, 192k
Légende Fig. 6 – Final Chalcolithic pottery.
URL http://books.openedition.org/momeditions/docannexe/image/517/img-7.jpg
Fichier image/jpeg, 160k
Légende Fig. 7 – Final Chalcolithic pottery.
URL http://books.openedition.org/momeditions/docannexe/image/517/img-8.jpg
Fichier image/jpeg, 192k
Légende Fig. 8 – Final Chalcolithic pottery.
URL http://books.openedition.org/momeditions/docannexe/image/517/img-9.jpg
Fichier image/jpeg, 304k
Légende Fig. 9 – Final Chalcolithic pottery.
URL http://books.openedition.org/momeditions/docannexe/image/517/img-10.jpg
Fichier image/jpeg, 128k
Légende Fig. 10 – Final Chalcolithic pottery.
URL http://books.openedition.org/momeditions/docannexe/image/517/img-11.jpg
Fichier image/jpeg, 216k
Légende Fig. 11 – Final Chalcolithic pottery (1‑3), spindle whorls (4‑7) and a loom-weight (8).
URL http://books.openedition.org/momeditions/docannexe/image/517/img-12.jpg
Fichier image/jpeg, 136k
Légende Fig. 12 – Flint arrowheads.
URL http://books.openedition.org/momeditions/docannexe/image/517/img-13.jpg
Fichier image/jpeg, 212k
Légende Table 1 – Radiocarbon dates from Tatul.
URL http://books.openedition.org/momeditions/docannexe/image/517/img-14.jpg
Fichier image/jpeg, 115k

CC-BY-NC-ND-4.0

Le texte seul est utilisable sous licence CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Cette publication numérique est issue d’un traitement automatique par reconnaissance optique de caractères.
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search