Navigation – Plan du site

AccueilNuméros52VariaRevisiting the Middle Paleolithic...

Varia

Revisiting the Middle Paleolithic site of Dashsalakhly Cave (Azerbaijan, South Caucasus): Cave and Lithic Assemblage

Réexamen du site du Paléolithique moyen de la grotte de Dashsalakhly (Azerbaïdjan, Sud-Caucase) : la grotte et l’industrie lithique
Yoshihiro Nishiaki, Azad Zeynalov, Ulviya Safarova et Yaqub Mammadov

Résumés

La grotte de Dashsalakhly est un site paléolithique moyen rare mis au jour en Azerbaïdjan. Bien que le site ait été fouillé pour la première fois en 1958, ses découvertes n’ont pas été évaluées dans une perspective moderne dans le contexte plus large du Paléolithique moyen du Caucase du Sud. Cette étude présente un résumé des travaux de terrain menés en 2023, après une interruption de 65 ans, ainsi qu’une nouvelle analyse des artefacts lithiques mis au jour en 1958. Les résultats confirment que cette grotte abritait autrefois une industrie du Paléolithique moyen caractérisée par la production courante de supports allongés utilisant la méthode Levallois. Ses caractéristiques techno-typologiques suggèrent que l’industrie lithique de cette grotte est plus proche du Moustérien des régions du Zagros et du Levant que de celui du Grand Caucase.

Haut de page

Texte intégral

We would like to thank Dr. Farhad Guliyev (Director of the Institute of Archaeology and Anthropology, Baku), Dr. Seadat Aliyeva (Avey State Historical-Cultural Reserve) and Mr. Musa Mursaqulov (Keshikchidagh State Historical-Cultural Reserve, Gazakh, Azerbaijan) for their permission and practical assistance in our research. Dr. Hitoto Nakara of Aoyma Gakuin University in Japan assisted with the preparation of the drawings. Financial support for this study was provided by the Japan Society for the Promotion of Science (grant numbers 23H00690 and 24H00001) and Mitsubishi Foundation (grant number 202320018).

Introduction

1Research on the Middle Paleolithic of Southwest Asia is a focal subject in Paleolithic archaeology and paleoanthropology. This interest stems from its direct relevance to our understanding of the replacement and/or interaction between Homo sapiens, who originated in Africa approximately 300 ka, and Neanderthals, who evolved on the European continent around the same time (Bae et al. 2017; Nishiaki and Akazawa eds. 2018; Bergström et al. 2021). Research in the South Caucasus, situated at the northern edge of Southwest Asia, holds similar relevance. Its strategic geographic position at the crossroads between Europe and Asia has attracted researchers investigating the dispersal of Neanderthals from Europe to the East and their contact with modern humans (Ghasidian et al. 2023; Doronicheva et al. 2024).

2This study aims to provide new data on the Middle Paleolithic lithic industry in the South Caucasus. Despite increasing availability of new data from previously unknown sites, the evaluation of the extant datasets is also required to explain the cultural processes, which are assumed to be complex, that occurred during the interaction and/or replacement between these two hominin groups in the Middle Paleolithic.

3To enrich databases on Middle Paleolithic occurrences in the South Caucasus, this paper reports our research on Dashsalakhly Cave, Azerbaijan, and its lithic assemblage (fig. 1). Dashsalakhly Cave is located approximately 20 km west of Gazakh and less than 1 km from Azerbaijan’s international border with Armenia. Its excavation, carried out in 1958 by M. Huseynov, reported a Middle Paleolithic deposit approximately 30–35 cm thick above the bedrock, overlain by Bronze Age and later deposits (Huseynov 2010: 57–58 and 177–178). Despite its rarity as a Middle Paleolithic cave site in Azerbaijan, the site and its lithic assemblage have not been re-evaluated from a modern perspective because of political instability in the area since the early 1990s.

Fig. 1 – Map showing the Paleolithic sites related to the text.

Fig. 1 – Map showing the Paleolithic sites related to the text.

Map Y. Nishiaki.

The 2023 field investigations at Dashsalakhly, West Azerbaijan

4The Caucasus are characterized by two massive mountain ranges, which run in parallel from east to west between the Black Sea and the Caspian Sea, and the foothills between them. The Greater Caucasus in the north, over 5,000 m above sea level, has served as a natural barrier between Europe and Asia. On the other hand, the gentler Lesser Caucasus Mountain range in the south allows easier cultural interaction on both sides. The South Caucasus, the subject of this study, refers to the Lesser Caucasus Mountains and surrounding foothills, plateaus, and plains. It is often regarded as the northern edge of Southwest Asia.

5Over the last two decades, reliable stratigraphic evidence and radiometric dates that were previously unavailable have been produced. In particular, optically stimulated luminescence (OSL), thermoluminescence (TL), and electronic spin resonance (ESR) absolute dating methods have provided fresh insights into known Middle Paleolithic sites excavated decades ago (e.g. Adler and Tushabramishvil 2004; Pinhasi et al. 2012; Mercier et al. 2010). Some remarkable results from Azerbaijan include those from the Azykh (Fernández-Jalvo et al. eds. 2016) and Gazma Caves (Zeynalov et al. 2023). We also re-excavated the Middle Paleolithic Damjili Cave in western Azerbaijan from 2016 to 2022 to investigate its potential using a modern method (Nishiaki et al. 2025; fig. 1).

6Building on these efforts, we attempted to shed new light on Dashsalakhly Cave and its lithic evidence during the August 2023 season. This cave is located at the western foot of Avey Mountain, just 1.5 km west of Damjili Cave (fig. 2). Avey Mountain, located 889 m above sea level, is a small mountain on the northern edge of the Lesser Caucasus Mountains. Dashsalakhly Cave is situated on the left bank of a small valley that runs westward near its source (fig. 2). Although the cave was documented as facing open land in 1958 (fig. 3: 1), in 2023 it was deeply buried in a forest with vegetation that had developed when the cave was inaccessible owing to political instability (fig. 3: 2). Its geographic coordinates are 41º08’39.89”N and 45º12’55.63”E, and the altitude of the cave bedrock is 685 m (Oregon 750TJ Garmin Ltd).

Fig. 2 – Location of the Dashsalakhly and Damjili Caves in Avey Mountain.

Fig. 2 – Location of the Dashsalakhly and Damjili Caves in Avey Mountain.

Map Y. Nishiaki.

Fig. 3 – General view of the Dashsalakhly Cave.

Fig. 3 – General view of the Dashsalakhly Cave.

1. distant view in 1958; 2. view of the cave from the east (2023); 3. the cave entrance; 4. view from inside the cave.

Photo 1: Huseynov 2010; photos 2–4: Y. Nishiaki.

7Huseynov (2010: 57–60) described the cave opening as approximately 5 m wide, 2 m high, and 17 m long. This description was confirmed in our fieldwork, except the length of the cave which was found to be approximately 12 m. Considering that the terrace area was distributed at approximately 5 m in an open area, Huseynov included the terrace in his estimate of the cave length. Additionally, we identified a step on the bedrock on the southern side of the cave mouth (figs. 3: 3 and 4: 4). During the 2023 season, several sounding pits were opened to determine whether any primary deposits remained; however, they yielded only secondary sediments. Nevertheless, we conducted excavations in an area of 3 m × 5 m (fig. 4: 1) on the terrace, where relatively thick sediments (up to 50 cm) were identified. However, the excavations failed to reveal primary cultural deposits. It is likely that almost all cultural layers of Dashsalakhly Cave were removed during the 1958 excavation, leaving no in situ Middle Paleolithic deposits.

Fig. 4 – Plan and sections of the Dashsalakhly Cave.

Fig. 4 – Plan and sections of the Dashsalakhly Cave.

1. plan: the shaded area indicates squares excavated in 2023; 2. sections: the arrows show the position of the upper charred layer (see fig. 5: 2).

Y. Nishiaki.

8Furthermore, our 2023 season’s research confirmed the accuracy of Huseynov’s (2010) stratigraphic description. First, black tar-like soot remains were observed on both sides of the cave wall, approximately 30 cm above the bedrock (fig. 5: 1). These remains are distributed approximately 2.5 m along the length of the cave from the mouth inwards. This corresponds with the known features of large fireplaces in the Middle Paleolithic, as described by Huseynov (2010: 60). Second, a clear-cut line on the wall showed different weathering patterns. This line was located approximately 1.7 m above the bedrock (fig. 5: 2), likely signifying the tops of the original deposits that existed before the 1958 excavation (Huseynov 2010: 60).

Fig. 5 – Traces of the charred layers visible on the Dashsalakhly cave wall of.

Fig. 5 – Traces of the charred layers visible on the Dashsalakhly cave wall of.

1. the charred layer close to the bedrock; 2. the upper charred layer.

Y. Nishiaki.

The Middle Paleolithic lithic assemblage of Dashsalakhly cave

9In addition to the aforementioned fieldwork, our 2023 research included a reanalysis of lithic materials unearthed during the 1958 excavation. According to Huseynov (2010: 237), 326 Middle Paleolithic lithic artifacts were recovered. In this study, we analyzed 253 items—approximately 80% of the original item count—stored at the Institute of Archeology and Anthropology in Baku (table 1). The remaining specimens were exhibited at museums, including the Baku National History Museum, the Archaeological Museum in Baku, and the Gazakh History Museum. Although some of these artifacts were briefly accessible to the authors, this paper refers to the Institute of Archeology and Anthropology collection, which was examined in more detail.

Table 1 – General inventory of the Dashsalakhly lithic assemblage.

Raw
material

Andesite

Gabro/
daicite

Flint

Chalcedony

Obsidian

Total

Total (%)


Cores

7 (2.8)

Flake cores, single-platform

0

0

0

0

1

1

Blade cores, single-platform

0

1

1

0

0

2

Levallois flake cores, unidirectional

0

1

1

0

0

2

Levallois flake cores, centripetal

0

2

0

0

0

2


Core-management pieces

14 (5.5)

Levallois debordant

0

1

0

0

0

1

Core-edge flakes

1

5

2

0

3

11

Core front flakes

1

0

1

0

0

2


Debitage

Non-Levallois

93 (36.8)

Cortical flakes

3

3

0

0

0

6

Partially cortical flakes

6

15

1

0

0

22

Flakes

5

26

4

0

7

42

Partially cortical blades

1

2

0

0

0

3

Naturally backed blades

1

5

0

0

0

6

Blades

4

6

1

0

3

14

Levallois

34 (13.4)

Levallois flakes

4

7

0

0

3

14

Levallois points

0

3

0

0

0

3

Levallois elongated points

0

1

0

0

0

1

Levallois blades

5

11

0

0

0

16

Retouched tools

14

26

11

1

13

65

65 (25.7)

Chunks

4

0

1

1

0

6

6 (2.4)

Chips and fragments

2

25

5

0

2

34

34 (13.4)

Total

51 (20.2)

140 (55.3)

28 (11.1)

2 (0.8)

32 (12.6)

253

253 (100.0)

10Lithic artifacts tend to be poorly preserved. They often exhibit post-depositional surface modifications such as heavy weathering and double patina. Therefore, the available lithic collections appear to include materials from secondary contexts. This interpretation is reinforced by the fact that the Middle Paleolithic artifacts were discovered in a 30–35 cm-thick layer on the bedrock, directly overlain by Holocene deposits without intermediate ones between them. This stratigraphic finding indicates that the deposits containing lithic artifacts were damaged by natural agents for a long period of time. Huseynov (2010: 57) stated that there is only “one cultural layer.” This implies the integrity of the recovered lithic artifacts’ techno-morphological features.

Raw material

11Huseynov (2010: 177) reported that the lithic assemblage of Dashsalakhly Cave is composed of a variety of raw materials including jasper, hornfels, gabbro-diabase, siliceous limestone, andesite, and obsidian. Our reanalysis utilized a more general classification of rock types (table 1). Although their sources were not specified, we suspect that most raw material groups were locally available. According to our survey of the Avey Mountains (Kadowaki 2021; Nishiaki and Shimogama 2025), knappable materials, such as andesite and dacite have been identified approximately 2 km south of Dashsalakhly Cave.

12A detailed field survey could not be conducted in the current conditions; hence the raw material use was not investigated in detail. Nevertheless, obsidian is undoubtedly a non-local raw material. We discovered that obsidian was utilized more commonly than was reported: 5% in the original report (Huseynov 2010: 177) but 13% in the present study (table 1). The 32 obsidian pieces included only one core (fig. 6: 5), suggesting that most of the obsidian artifacts brought to the site were either finished or prepared off-site. Ten obsidian pieces were subjected to provenance analyses (Zeynalov et al. 2025). The results identified at least six different sources distributed across the modern territories of Georgia, Turkey, and Armenia, with the most distant source being Sarkamis in Northeast Turkey, which is approximately 260 km from Dashsalakhly Cave. However, this does not mean that all obsidian pieces were brought into the cave from primary sources through direct acquisition trips or exchanges by Middle Paleolithic communities. We should consider the possibility that obsidian was acquired as a secondary raw material in local riverbeds at the Agstafa and Kura Rivers, within a 20-km distance of the cave (Huseynov 2010: 177).

Fig. 6 – Cores and core management pieces from Dashsalakhly.

Fig. 6 – Cores and core management pieces from Dashsalakhly.

1. Levallois flake core, unidirectional recurrent type; 2. Levallois blade core, unidirectional recurrent type; 3–4. Levallois flake cores, centripetal recurrent type; 5. non-Levallois obsidian flake core, prismatic type; 6. non-Levallois blade core, prismatic type; 7. non-Levallois blade core, narrow-faced type; 8. overshot flake with a truncated-faceted butt; 9. Levallois débordant blade; 10. core-tablet flake.

Y. Nishiaki.

13As for the non-obsidian raw material types, the Dashsalakhly lithic assemblage includes “chunks” (table 1), which we define as preliminary splits of larger nodules for further reduction or test knapping. Their occurrence indicates that at least some raw material blocks were brought to the settlement outside the cave after the initial pretreatment. The occurrence of cores, core management pieces, cortical flakes/blades, and small debitage fragments (table 1) do indicate on-site knapping. Nevertheless, the excavated lithic assemblage primarily consists of finished tool blanks and retouched tools. This indicates that it was primarily derived from brought-in materials that were processed or manufactured off-site.

Blank production technology

14Four of the seven cores were classified as Levallois, showing recurrent unidirectional (fig. 6: 1–2) or centripetal (fig. 6: 3–4) blank removal. The remaining three cores included one non-Levallois single-platform core for flakes (fig. 6: 5) and two cores for blade production (fig. 6: 6–7). Two of these non-Levallois cores exhibited semi-prismatic forms (fig. 6: 5–6), whereas the other had a burinated form with blank removal traces from a narrow face of the core (fig. 6: 7). The core management pieces in the Dashsalakhly assemblage correspond to the observed combination of Levallois and non-Levallois core reduction technologies. These include Levallois débordant pieces (Boëda 1994) detached to maintain the lateral convexity of the cores (figs. 6: 9 and 7: 11), naturally backed blades and flakes (fig. 7: 12), and flakes refreshing the working surface of the single-platform cores through large removal (fig. 6: 8). There is also a core-platform tablet with a single-platform semi-prismatic core (fig. 6: 10).

15Overall, the available assemblage indicates the common use of Levallois technology for blank production (fig. 7). The Bordesian–Levallois index (IL) amounts to 32.6 (63/193). The Levallois blanks often exhibited unidirectional or unidirectional convergent dorsal scars (table 2). Blanks with bidirectionally flaked dorsal scars were rare, which was consistent with the absence of bidirectionally reduced cores. Despite the occurrence of Levallois cores showing centripetal flaking scars, the blank assemblage rarely exhibited a centripetal pattern (fig. 7; table 2). This discrepancy may reflect the exhausted nature of the Levallois cores, merely representing flaking patterns in the final stage of core reduction.

Table 2 – Dorsal-scar types for the Levallois blanks in the Dashsalakhly lithic assemblage.

Blank
types

Uni-directional

Convergent

Crossed

Bi-directional

Centripetal

Total (%)

Levallois flakes

9

11

5

0

3

28 (45.9)

Levallois points

1

5

1

0

0

7 (11.5)

Levallois elongated points

1

3

1

0

0

5 (8.2)

Levallois blades

9

9

1

1

1

21 (34.4)

Total (%)

20 (32.8)

28 (45.9)

8 (13.1)

1 (1.6)

4 (6.6)

61 (100.0)

Fig. 7 – Levallois and related pieces from Dashsalakhly.

Fig. 7 – Levallois and related pieces from Dashsalakhly.

1–3. Levallois flakes; 4. Levallois point; 5–6. Levallois elongated points; 7–10. Levallois blades; 11. Levallois débordant flake; 12. naturally backed blade.

Y. Nishiaki.

16It is also worth mentioning that the Levallois components of Dashsalakhly Cave, whether retouched or not, principally consist of flake and blade forms (fig. 7: 1–3 and 7–10; tables 1–3), whereas point forms are rare (fig. 7: 4–6).

Table 3 – Size of the complete Levallois blanks from Dashsalakhly.

mm

Max

Min

Mean

S.D.

Levallois flakes (n=24)

Length

85

20

51.3

14.98

Width

53

13

35.2

8.20

Thickness

12.8

4.2

8.1

2.40

Levallois points (n=7)

Length

72

31

51.9

18.11

Width

48

22

33.4

9.29

Thickness

14.5

6.5

10.04

2.73

Levallois elongated points (n=5)

Length

82

61

69.6

8.56

Width

35

28

30.0

2.92

Thickness

11.6

6.9

8.8

1.83

Levallois blades (n=11)

Length

88

54

69.4

12.52

Width

39

21

27.5

5.93

Thickness

12.2

4.8

8.8

4.28

Levallois total (n=47)

Length

88

20

57.5

16.43

Width

53

13

32.6

7.98

Thickness

14.5

6.9

8.4

2.42

17Another important technological feature was the common production of blade blanks with a blade index (ILam) of 31.6 (61/193). The proportion of blades among the Levallois products was even higher at 42.6% (26/61; fig. 7: 5–10). The blade blanks were rather robust pieces struck with a hard hammer.

18Blanks, whether Levallois or not, are commonly produced from cores with faceted striking platforms. The Bordesian faceting index (IF) and its strict index (IFs) are 56.9 (91/160) and 31.3 (50/160), respectively (table 4). Faceting was particularly popular in Levallois blank production (table 4). However, as indicated by the low IFs, faceted platforms with deliberate faceting directly prior to a blank production blow are not necessarily common. Furthermore, blanks with chapeau-de-gendarme-type butts, which are often associated with Levallois point production (Bordes 1961), were hardly identified. A straight platform appeared to be the norm especially for non-Levallois blades in the study’s assemblage. This may be related to the rarity of Levallois points in Dashsalakhly Cave.

Table 4 – Platform types for the Dashsalakhly lithic assemblage.

Blank types

Cortical

Plain

Dihedral

Faceted

Faceted, strict

Total

Levallois flakes

0

5

5

4

11

25

Levallois points

0

0

2

1

4

7

Levallois
elongated points

0

0

1

2

2

5

Levallois blades

0

0

8

0

7

15

Non-Levallois flakes

1

12

20

3

6

42

Non-Levallois blades

0

51

5

3

7

66

Total (%)

1 (0.6)

68 (42.5)

41 (25.6)

13 (8.1)

37 (23.1)

160 (100.0)

Retouched Tools

19Table 5 shows a breakdown of the tool types among the retouched artifacts. The most common were sidescrapers (31/66, 47.0%), whose retouch was often made on a single edge only (fig. 8: 2–4), followed by convergent scrapers (fig. 8: 5–8) and double scrapers (fig. 8: 11). Transversal and déjeté scrapers, whose retouch edges were slanted toward the flaking axis of the blank, were uncommon in the Dashsalakhly tool assemblage. Quina-type retouch was also virtually absent. These findings suggest that the retouch intensity was low, thereby implying that the cave’s occupational intensity was also low (Dibble 1987). The large number of lightly retouched Levallois and non-Levallois blanks in the tool assemblage also indicated a low retouch intensity (table 4).

Fig. 8 – Retouched tools from Dashsalakhly.

Fig. 8 – Retouched tools from Dashsalakhly.

1. retouched elongated Levallois point; 2. obsidian single-convex sidescraper with basal thinning; 3. single convex sidescraper; 4. single straight sidescrapers; 5, 7–8. convergent convex sidescraper; 6. convergent alternate-convex sidescraper; 9. backed knife; 10. Mousterian point with truncated faceted butt; 11. double convex sidescraper with truncated faceted butt; 12. obsidian flake with basal thinning; 13. pick.

Y. Nishiaki.

20The Mousterian (fig. 8: 10) and retouched Levallois points (fig. 8: 1) were less conspicuous but important tool groups. They often exhibited fractures at the tip, in the form of burinating (fig. 9: 1), bending (fig. 9: 2), or flute-like fractures (fig. 9: 3, 4). Given the similarities between these fractures and impact fractures corresponding to the use as projectiles (Groman-Yaroslavski et al. 2016), these scars deserve detailed future study using a microscope.

Fig. 9 – Pointed tools with impact fractures in the Dashsalakhly lithic assemblage.

Fig. 9 – Pointed tools with impact fractures in the Dashsalakhly lithic assemblage.

1. burinated fracture; 2. ending fracture; 3–4. flute-like fractures.

Y. Nishiaki.

21There were other miscellaneous tool types, such as retouched Levallois and non-Levallois pieces, notches, truncations, and a pick (fig. 8: 13). A few Upper Paleolithic tool types were also discovered, but at a negligible frequency (table 5; fig. 8: 9).

22Notably, a small number of retouched tools exhibited the truncated-faceted technique (fig. 8: 10–11; Schroeder 1968; Nishiaki 1985). Furthermore, we identified at least two specimens showing basal thinning retouch on the ventral surface of the proximal end (fig. 8: 2, 12). This retouch was presumably for hafting the proximal ends of these tools.

Discussion

23The 2023 fieldwork study revealed that virtually all Middle Paleolithic sediments of the Dashsalakhly Cave were removed by the original excavations in 1958. Therefore, the recovered lithic assemblage should be interpreted using only the 1958 material stored at the Institute of Archaeology and Anthropology, Baku.

24Although the excavator reports that all those were situated in a thin layer on the bedrock, many lithic artifacts of Dashsalakhly Cave exhibited secondary damage. Therefore, they may not represent a stratigraphic integrity. However, it is difficult to assess this interpretation because of the complete depletion of the original archaeological deposits. In the present study, the available lithic artifacts are dealt with as a single assemblage.

25The assemblage includes a large proportion of finished tools and standardized blanks, but a relatively small number of by-products from their manufacturing processes. This corroborates the excavator’s view that many of the artifacts were brought into the cave after processing at locations outside the cave (Huseynov 2010: 177). Retouched tools and Levallois blanks comprised nearly 40% of the total in the Dashsalakhly lithic assemblage (table 1). Moreover, the retouched pieces generally exhibited light retouch, indicating a low retouch intensity. This is evidenced by the retouched Levallois pieces and single/double sidescrapers (table 4). It is also worth mentioning that the recovered assemblage contained cores, chunks from the initial raw material splitting, and debris such as chips and fragments (table 1). However, these occurrences were less common than in the Middle Paleolithic residential camps (see Nishiaki et al. 2012). The Dashsalakhly communities might have moved around with finished tools and tool blanks, as well as cores and their raw materials for knapping at subsequent stations. When a usewear analysis of the recovered lithic materials is performed in future, the purpose of visits to the cave, possibly the procurement of specific resources through hunting, can be better understood (see above; fig. 9).

26From a broader perspective, the primary question concerns the cultural position of the Dashsalakhly lithic industry in the Middle Paleolithic Caucasus. Although our understanding of the cultural framework of the Caucasian Middle Paleolithic has become more refined in recent decades, the available picture is still preliminary. This is due to the lack of detailed technological descriptions of known lithic assemblages and their radiometric dates. Nevertheless, most researchers agree that at least two cultural provinces existed in the Middle Paleolithic Caucasus (Doronicheva et al. 2023). The first was the Eastern Micoquian cultural complex, in the western Greater Caucasus and the Russian Plain to the north. Typical sites include the Mezmaiskaya Cave, a Neanderthal fossil-bearing site dated from 70–40 ka. The second was the Zagros Mousterian, in the eastern part of the Greater Caucasus and further south, which is also considered to be a product of the Neanderthals.

27The Dashsalakhly industry is clearly different from the Eastern Micoquian. It contains none of the hallmark tools of the Eastern Micoquian, such as small handaxes, bifacial asymmetrical scrapers, and knives. Instead, sidescrapers with unifacial retouch made on flakes and blades are common. Technologically, the Dashsalakhly industry commonly used Levallois technology, which was unpopular in the Eastern Micoquian but a distinguishing characteristic of the Zagros Mousterian. In other words, the Dashsalakhly industry can be assigned to a group of the Zagros Mousterian.

28The Zagros Mousterian was originally defined by Skinner (1965) as a Middle Paleolithic industry distributed in the eastern wing of the Fertile Crescent of Southwest Asia. He first distinguished it from the Levantine Mousterian through the common practice of manufacturing sidescrapers on non-Levallois flakes. However, this simple scheme has been challenged since the 1980s. A reanalysis of the lithic assemblage of Bistun Cave, Iran, which is the key material used by Skinner to define the Zagros Mousterian, revealed a common use of Levallois technology and blade blank production. Moreover, the truncated-faceted technique for blank modification, which was also popular in the Levantine Mousterian, was identified as an important component of the technological repertoire of the Zagros Mousterian (Dibble 1984). In recent years, research has increasingly revealed a great deal of regional variability in the Middle Paleolithic in the Zagros. For example, lithic assemblages characterized by elongated bifacial foliate production have been identified in Southern Zagros (Biglari et al. 2009), whereas assemblages with many unretouched Levallois products but few sidescrapers have been identified in the east (Nasab and Hashemi 2016). In other words, the Zagros Mousterian cannot be considered monolithic, as is the case for the Levantine Mousterian (Nishiaki and Akazawa eds. 2018). It follows that the Zagros Mousterian in the South Caucasus, to which the Dashsalakhly industry belongs, may have considerable variability. This possibility should also be considered in the time-space context of the Dashsalakhly industry in the Middle Paleolithic Caucasus.

29The following aspects of the Dashsalakhly industry deserve careful examination. First, the common production of blade blanks is one of the industry’s characteristic traits. Blade-rich Middle Paleolithic assemblages have been reported in the South Caucasus. The earliest known assemblage is the Djruchula-Kudaro industry (Meignen and Tushabramishvili 2010). Thick blades were produced using a combination of Levallois and Laminar methods (Boëda 1994). TL dates from Djruchula Cave, Georgia, suggest a period of 250–130 ka (Mercier et al. 2010). These technological features and radiometric dates have led some authors to relate the Djruchula-Kudaro industry to the early Middle Paleolithic of the Levant (Meignen and Tushabramishvili 2010). Blade-rich early Middle Paleolithic assemblages have been recognized at many sites across the Levant, such as Hayonim, Abu Sif, and Hummal, and more recently at Misliya and Dederiyeh (fig. 1). Levallois and Laminar reduction systems coexisted in these assemblages in various proportions. These systems were oriented toward the production of elongated blanks, which are often retouched into elongated points (Meignen and Tushabramishvili 2010; Nishiaki et al. 2022; Meignen 2024: 260). Furthermore, the Middle Paleolithic lithic assemblages of the northern Zagros Mountains, such as Shanidar (Reynolds et al. 2022), Hazar Merd (Garrod 1930), Warwasi (Dibble and Holdaway 1993) and Bistun (Dibble 1984) are also known as blade-rich Middle Paleolithic industries (fig. 1). However, insufficient dating of these sites has prevented the definition of a relationship between the Middle Paleolithic blade industries of the Levant and the Zagros. Resuming excavations at Shanidar Cave, which was dated at 85 ka and earlier (Reynolds et al. 2022), is expected to shed new light on this issue.

30In the meantime, we shall examine the techno-typological variability of the Zagros Mousterian in the South Caucasus based on our findings from Dashsalakhly. The Djruchula-Kudaro and Dashsalakhly industries were found to be more dissimilar than similar. In the Djruchula-Kudaro industry, the production of elongated blanks largely used the Laminar system (Meignen and Tushabramishvili 2010), whereas Dashsalakhly assemblages predominantly resulted from Levallois methods. Consequently, the blades have different morphological characteristics. Typologically, unlike the Dashsalakhly assemblage, the bifacial retouching of the Djruchula-Kudaro industry is thought to represent a link to the Eastern Micoquian cultural complex.

31Thus, the Dashsalakhly industry should be compared with industries from the eastern and southern parts of the South Caucasus. In these regions, common blade production using Levallois technology, as observed in Dashsalakhly Cave, was not necessarily the norm. For example, Mousterian materials from Damjili Cave, situated between the finger calls with Dashsalakhly, indicate the existence of a flake-based industry (Huseynov 2010; Nishiaki 2025), in the same region. The original excavator regarded the Damjili industry to be earlier than the Dashsalakhly industry (Huseynov 2010: 177). A Mousterian industry comparable to the Damjili industry has been found in Layer 3 of the Azykh Cave, which was also placed in an earlier phase of the Middle Paleolithic on a techno-typological basis (Huseynov 2010). Recent re-excavations have confirmed the flake-based nature of the Azykh Cave Layer 3 (Asryan 2015). Furthermore, ESR dates of 180 and 100 ka have been assigned to this industry (Fernández-Jalvo et al. eds. 2016).

32The common use of Levallois for blade production at Dashsalakhly Cave probably represents the large variability of the Zagros Mousterian in the South Caucasus. Our limited database suggests that the most comparable assemblage seems to be from the Taglar Cave in South Azerbaijan. The Bordesian general indices of the Taglar assemblages, representing a long sequence, are 35–50 in IL, 25–40 in ILam and 60–70 in IF (Jafarov 1983), whereas the Dashsalakhly indices are 32.6, 31.6 and 56.9, respectively. The statistical results are highly comparable. Although the Taglar assemblages have not been dated, the original excavator considers them to be later than the Azykh Cave Layer 3 and Damjili Cave on a techno-typological and paleoenvironmental basis (Huseynov 2010). The Dashsalakhly and Taglar Caves are potentially comparable with the Hovk 1 Cave in Armenia, whose excavations yielded Levallois blade-rich assemblages. Importantly, TL dates ranging from 104 to 54 ka are also available (Pinhasi et al. 2011).

33The next issue concerns the typological variability. The small sample size of the Dashsalakhly community may not represent the entire range of tool types. Nevertheless, the available data indicated the dominance of sidescrapers in the tool assemblage. These were accompanied by only a small number of points, either retouched or unretouched (table 5). By contrast, the Zagros Mousterian of the South Caucasus included assemblages with many points at Yerevan Cave, Lusakert, Angeghakot I, Armenia and Gazma, Azerbaijan. For example, the proportions of points among the retouched tools from Angeghakot I and Gazma were 43.9% and 31.2%, respectively, with points outnumbering sidescrapers by 8.5% in Angeghakot I and 19.3% Gazma (Liage et al. 2006; Zeynalov et al. 2023). Another intriguing trait of this assemblage group is their relatively small tool size. This has led some authors to propose the term “Caucasian micro-Mousterian” (Liage et al. 2006). Additionally, we note that this industrial group often used a truncated-faceted technique. Pointed tools modified with this technique are specifically called “Yerevan points” (Liagre et al. 2006). In our opinion, it is not appropriate to define such a tool type, because the truncated-faceted technique has been widely applied to all forms of flake blanks, including those used as scrapers, cores, and more (Nishiaki 1985). The current literature dubs this technique as “Nahr Ibrahim core” when it is used for blank production (Solecki and Solecki 1970) or refers to “Taglar scrapers” for scrapers using this technique (Jafarov 1983). This technique is likely to have been used for a variety of purposes in the Middle Paleolithic and earlier (McPherron 2007), we did not examine this issue further to understand the characteristics of small-point industries.

Table 5 – Tool types for the Dashsalakhly lithic assemblage.

Types

LF

LP

LEP

LB

NLF

NLB

CEF

Total

Backed knives

1

0

0

1

0

0

0

2

Basal-thinned piece

1

0

0

0

0

0

0

1

Borers

1

0

0

0

1

0

0

2

Mousterian points

1

3

1

0

0

1

0

6

Mousterian point (TF)

1

0

0

0

0

0

0

1

Sidescrapers, single straight

1

0

0

1

3

1

1

7

Sidescrapers, single convex

1

0

0

1

5

3

1

11

Sidescrapers, single concave

1

0

0

0

1

0

0

2

Sidescrapers, convergent

1

0

0

0

0

0

0

1

Sidescrapers, convergent (TF)

1

0

0

0

0

0

0

1

Sidescrapers, double convex

1

0

0

0

0

0

0

1

Sidescrapers, double convex
(alternate)

1

0

0

0

0

0

0

1

Sidescrapers, double straight

0

0

0

0

1

0

0

1

Sidescrapers, double straight-convex

2

0

0

0

1

1

0

4

Sidescrapers, transversal

2

0

0

0

0

0

0

2

Truncation

0

0

0

0

1

0

0

1

Notches

0

0

0

0

2

0

0

2

Pick

0

0

0

0

1

0

0

1

Retouched Levallois blades

0

0

0

2

0

0

0

2

Retouched Levallois elongated points

0

0

3

0

0

0

0

3

Retouched Levallois point

0

1

0

0

0

0

0

1

Retouched blades

0

0

0

0

0

5

0

5

Retouched flakes

0

0

0

0

7

0

0

7

Total

16

4

4

5

23

11

2

65

LF: Levallois flake; LP: Levallois point; LEP: Levallois elongated point; LB: Levallois blade; NLF: non-Levallois flake; NLB: non-Levallois blade; CEF: core-edge flakes; TF: truncated-faceted.

34Available chronological data contribute to a late date being assigned to this industry. This small-point industry has been assigned to the late Middle Paleolithic, likely MIS3, through chronostratigraphic observations at Yerevan Cave and the geomorphological study at the Angeghakot I site (Liagre et al. 2006). Radiometric dating at Gazma Cave in Nachichevan District of Azerbaijan has been an important addition, with dates of approximately 50 ka obtained using the OSL method (Zeynalov et al. 2023). This corresponds to the aforementioned relative dating. Taken together, the Dashsalakhly industry, which is not affiliated with this small-point industry, is likely derived from an earlier period.

Conclusions

35This study examined the Dashsalakhly Cave, a rare Middle Paleolithic cave site excavated in Azerbaijan, considering its current state and lithic assemblage from the 1958 excavation. Our fieldwork in 2023 demonstrated that the cave was almost completely excavated in 1958, making it difficult to identify in situ deposits today. Nevertheless, this study accessed the 1958 lithic collection from Dashsalakhly, characterizing it as a Levallois-based elongated blank industry with a high frequency of sidescrapers. These techno-typological features matched the definition of the Zagros Mousterian. In other words, this study demonstrated the distribution of the Zagros Mousterian as far west as the western end of the modern territory of Azerbaijan.

36The Zagros Mousterian in the South Caucasus was first noted at the Taglar Cave, South Azerbaijan (Jafarov 1983). However, as in other Mousterian industries, it may contain a large amount of variability Our current knowledge indicates interaction with the Eastern Micoquian industry to the north (Doronicheva et al. 2023), the flake-based industry as represented by Azykh Cave Layer 3, the blade-rich industry with common use of Levallois technology as found at Dashsalakhly and Taglar Caves, and the small-point tool industry represented by the Yerevan and Gazma Caves. The Dashsalakhly industry is regarded as one of these variants, and there may have been more culturally distinct industries in the long Middle Paleolithic cultural sequence of the South Caucasus.

37Our current understanding of Middle Paleolithic cultural occurrences in the South Caucasus is insufficient. Thus, it is necessary to establish a cultural framework for Middle Paleolithic occurrences. To achieve this goal, it is important to accumulate solid data for each site and its lithic assemblages. It is anticipated that the data from Dashsalakhly Cave presented in this study will help provide a more comprehensive picture of Middle Paleolithic variability and its anthropological implications.

Haut de page

Bibliographie

Adler D. and Tushabramishvil N. 2004 – Middle Palaeolithic patterns of settlement and subsistence in the southern Caucasus. In: Conard J. (ed.), Settlement Dynamics of the Middle Paleolithic and Middle Stone Age: 91–132. Tubingen: Kerns Verlag.

Asryan L. 2015 – Azokh Cave Lithic Assemblage and their Contextualization in the Middle and Upper Pleistocene of Southwest Asia, PhD dissertation. Tarragona: University of Rovira i Virgili.

Bae C. J., Douka K., Petraglia M. D. 2017 – On the origin of modern humans: Asian perspectives. Science 358 (6368), [online] https://doi.org/10.1126/science.aai9067.

Bergström A., Stringer C., Hajdinjak H., Scerri E., Skoglund P. 2021 – Origins of modern human ancestry. Nature 590 (7845): 229–237.

Biglari F., Javeri M., Mashkour M., Yazdi M., Shidrang S., Tengberg M., Darvish J. 2009 – Test excavations at the Middle Paleolithic sites of Qaleh Bozi, Southwest of Central Iran: A preliminary report. In: Otte M., Biglari F., Jauber J. (eds.), Iran Paleolithic: 29–38. Oxford: Archaeopress.

Boëda E. 1994 – Le concept Levallois. Variabilité des méthodes. Paris: CNRS Éditions.

Bordes F. 1961 – Typologie du Paléolithique ancien et moyen. Paris: CNRS Éditions.

Dibble H. L. 1984 – The Mousterian industry from Bisitun cave (Iran). Paléorient 10 (2): 23–34.

Dibble H. L. 1987 – The interpretation of Middle Paleolithic scraper morphology. American Antiquity 52 (1): 109–117.

Dibble H. L and Holdaway S. J. 1993 – The Middle Paleolithic industries of Warwasi. In: Olszewski D. I. and Dibble H. L. (eds.), The Paleolithic Prehistory of the Zagros and Taurus: 75–99. Philadelphia: The University Museum, The University of Pennsylvania.

Doronicheva E. V., Golovanova L. V., Doronichev V. B., Kurbanov R. N. 2023 – Archaeological evidence for two culture diverse Neanderthal populations in the North Caucasus and contacts between them. PLoS ONE 18 (4): e0284093, [online] https://doi.org/10.1371/journal.pone.0284093.

Doronicheva E. V., Golovanova L. V., Doronichev V. B., Tregub T., Tselmovitch A., Korzinova A., Spasovskiy Y. N., Nedomolkin A. G., Volkov M. 2024 – Hominin occupation of the North-Central Caucasus during the Middle Paleolithic: New results from Saradj-Chuko Grotto and the state of research. PaleoAnthropology 2024 (1), [online] https://doi.org/10.48738/2024.iss1.

Fernández-Jalvo Y., King T., Yepiskoposyan L., Andrews P. (eds.) 2016 – Azokh Cave and the Transcaucasian Corridor. Cham: Springer International Publishing.

Garrod D. 1930 – The Paleotithic of southern Kurdistan: Excavations in the caves of Zarzi and Hazar Merd. American School of Prehistoric Research 1930: 9–43.

Ghasidian E., Kafash A., Kehl M., Yousefi M., Heydari-Guran S. 2023 – Modelling Neanderthals’ dispersal routes from Caucasus towards east. PLoS ONE 18 (2), e0281978, [online] https://doi.org/10.1371/journal.pone.0281978.

Groman-Yaroslavski I., Zaidner Y., Weinstein-Evron M. 2016 – Mousterian Abu Sif points: Foraging tools of the Early Middle Paleolithic site of Misliya Cave, Mount Carmel, Israel. Journal of Archaeological Science 7: 312–323.

Huseynov M. 2010 – The Lower Paleolithic of Azerbaijan [in Azerbaijani with english summary]. Baku: National Academy of Sciences of Azerbaijan.

Jafarov A. K. 1983 – Mousterian Culture of Azerbaijan based on Materials from Taglar Cave. Baku: Elm.

Kadowaki S. 2021 – Neolithic chipped stone artifacts from Hacı Elamxanlı Tepe. In: Nishiaki Y., Guliyev F., Kadowaki S. (eds.), Hacı Elamxanlı Tepe: The Archaeological Investigations of an Early Neolithic Settlement in West Azerbaijan: 107–132. Berlin: Ex oriente.

Liagre J., Gasparyan B., Ollivier V., Nahapetyan S. 2006 – Angeghakot 1 (Armenia) and the identification of the Mousterian cultural facies of “Yerevan points” type in the Southern Caucasus. Paléorient 32 (1): 5–18.

McPherron S. P. 2007 – Tools versus Cores: Alternative Approaches to Stone Tool Analysis. Neumünster: Cambridge Scholars Publishing.

Meignen L. 2024 – Technological, cultural, and behavioural changes in the Levant from the Late Lower Palaeolithic to the Mid-Middle Palaeolithic: Contribution from the Hayonim sequence. In: Meignen L. and Bar-Yosef O. (eds.), Hayonim Cave: From the Early to the Middle Palaeolithic in the Levant (Israel): 255–280. Leiden: Sidestone Press.

Meignen L. and Tushabramishvil N. 2010 – Djruchula Cave, on the southern slopes of the Great Caucasus: An extension of the Near Eastern Middle Paleolithic blady phenomenon to the North. Paléorient 32 (2): 81–104.

Mercier N, Valladas H., Meignen L., Joron J.-L., Tushabramishvili N., Adler D. S. 2010 – Dating the Early Middle Palaeolithic laminar industry from Djruchula Cave, Republic of Georgia. Paléorient 36 (2): 163–173.

Nasab H. V. and Hashemi M. 2016 – Playas and Middle Paleolithic settlement of the Iranian Central Desert: The discovery of the Chah-e Jam Middle Paleolithic site. Quaternary International 408: 140–152.

Nishiaki Y. 1985 – Truncated-faceted flakes from Levantine Mousterian assemblages. Bulletin of Department of Archaeology, the University of Tokyo 4: 215–226.

Nishiaki Y. 2025 – Middle Palaeolithic lithic artifacts from the Damjili Cave. In: Nishiaki Y., Zeynalov A., Mammadov Y. (eds.), Damjili Cave: Investigating the Late Pleistocene to Holocene Human History in the Southern Caucasus: 77–84. Oxford: Oxbow Books.

Nishiaki Y. and Akazawa T. (eds.) 2018 – The Middle and Upper Paleolithic Archeology of the Levant and Beyond. Singapore: Springer Nature.

Nishiaki Y. and Shimogama K. 2025 – Site reconnaissance survey in Avey Mountain and its vicinity. In: Nishiaki Y., Zeynalov A., Mammadov Y. (eds.), Damjili Cave: Investigating the Late Pleistocene to Holocene Human History in the Southern Caucasus: 9–21. Oxford: Oxbow Books.

Nishiaki Y., Kanjou Y., Akazawa T. 2022 – The Early Middle Palaeolithic lithic industry of Dederiyeh Cave, Northwest Syria. L’Anthropologie 126 (3): 103028, [online] https://doi.org/10.1016/j.anthro.2022.103028.

Nishiaki Y., Kanjou Y., Muhesen S., Akazawa T. 2012 – Temporal variability of Late Levantine Mousterian assemblages from Dederiyeh Cave, Syria. Journal of Eurasian Prehistory 9 (1-2): 3–27.

Nishiaki Y., Zeynalov A., Mammadov Y. (eds.) 2025 – Damjili Cave: Investigating the Late Pleistocene to Holocene Human History in the Southern Caucasus. Oxford: Oxbow Books.

Pinhasi R., Higham T. F., Golovanova L. V. and Doronichev V. B. 2011 – Revised age of late Neanderthal occupation and the end of the Middle Paleolithic in the northern Caucasus. Proceedings of the National Academy of Sciences 108 (21): 8611–8616.

Pinhasi R, Nioradze M., Tushabramishvili M., Lordkipanidze D., Pleurdeau D., Moncel M.-H. 2012 – New chronology for the Middle Paleolithic of the Southern Caucasus suggests early demise of Neanderthals in this region. Journal of Human Evolution 63: 770–780.

Reynolds T., Hunt C., Hill E., Tilby E., Pomeroy E., Burke A., Barker G. 2022 – Le Moustérien du Zagros. Une vision synthétique à partir de la grotte de Shanidar. The Zagros Mousterian: The view from Shanidar Cave. L’Anthropologie 126 (3): 103045, [online] https://doi.org/10.1016/j.anthro.2022.103045.

Schroeder H. B. 1968 – The Lithic Industries from Jerf Ajla and their Bearing om the Problem of a Middle to Upper Paleolithic Transition, PhD dissertation. New York: Columbia University.

Skinner J. 1965 – The Flake Industries of Southwest Asia: A Typological Study. PhD dissertation. New York: Columbia University.

Solecki R. L. and Solecki R. S. 1970 – A new secondary flaking technique at the Nahr Ibrahim Cave, Lebanon. Bulletin du musée de Beyrouth 23: 129–134.

Zeynalov A. A., Anoikin A. A., Kulakov S. A., Otcherednoy A. K., Kurbanov R. N. 2023 – Gazma Cave. A final Middle Paleolithic site in Azerbaijan: Paleogeography, chronology, Archaeology, Ethnology & Anthropology of Eurasia 51 (3): 40–49.

Zeynalov A. A., Kuzminoikin Y. V., Glascock M. D., Fataliyev R. A. 2025 – Procurement and exchange of obsidian in the Middle Palaeolithic of western Azerbaijan (South Caucasus). Journal of Archaeological Science: Reports 66: 105313, [online] https://doi.org/10.1016/j.jasrep.2025.105313.

Haut de page

Table des illustrations

Titre Fig. 1 – Map showing the Paleolithic sites related to the text.
Crédits Map Y. Nishiaki.
URL http://journals.openedition.org/paleorient/docannexe/image/6087/img-1.jpg
Fichier image/jpeg, 317k
Titre Fig. 2 – Location of the Dashsalakhly and Damjili Caves in Avey Mountain.
Crédits Map Y. Nishiaki.
URL http://journals.openedition.org/paleorient/docannexe/image/6087/img-2.jpg
Fichier image/jpeg, 256k
Titre Fig. 3 – General view of the Dashsalakhly Cave.
Légende 1. distant view in 1958; 2. view of the cave from the east (2023); 3. the cave entrance; 4. view from inside the cave.
URL http://journals.openedition.org/paleorient/docannexe/image/6087/img-3.jpg
Fichier image/jpeg, 11M
Titre Fig. 4 – Plan and sections of the Dashsalakhly Cave.
Légende 1. plan: the shaded area indicates squares excavated in 2023; 2. sections: the arrows show the position of the upper charred layer (see fig. 5: 2).
Crédits Y. Nishiaki.
URL http://journals.openedition.org/paleorient/docannexe/image/6087/img-4.jpg
Fichier image/jpeg, 1,2M
Titre Fig. 5 – Traces of the charred layers visible on the Dashsalakhly cave wall of.
Légende 1. the charred layer close to the bedrock; 2. the upper charred layer.
Crédits Y. Nishiaki.
URL http://journals.openedition.org/paleorient/docannexe/image/6087/img-5.jpg
Fichier image/jpeg, 7,5M
Titre Fig. 6 – Cores and core management pieces from Dashsalakhly.
Légende 1. Levallois flake core, unidirectional recurrent type; 2. Levallois blade core, unidirectional recurrent type; 3–4. Levallois flake cores, centripetal recurrent type; 5. non-Levallois obsidian flake core, prismatic type; 6. non-Levallois blade core, prismatic type; 7. non-Levallois blade core, narrow-faced type; 8. overshot flake with a truncated-faceted butt; 9. Levallois débordant blade; 10. core-tablet flake.
URL http://journals.openedition.org/paleorient/docannexe/image/6087/img-6.jpg
Fichier image/jpeg, 766k
Titre Fig. 7 – Levallois and related pieces from Dashsalakhly.
Légende 1–3. Levallois flakes; 4. Levallois point; 5–6. Levallois elongated points; 7–10. Levallois blades; 11. Levallois débordant flake; 12. naturally backed blade.
URL http://journals.openedition.org/paleorient/docannexe/image/6087/img-7.jpg
Fichier image/jpeg, 733k
Titre Fig. 8 – Retouched tools from Dashsalakhly.
Légende 1. retouched elongated Levallois point; 2. obsidian single-convex sidescraper with basal thinning; 3. single convex sidescraper; 4. single straight sidescrapers; 5, 7–8. convergent convex sidescraper; 6. convergent alternate-convex sidescraper; 9. backed knife; 10. Mousterian point with truncated faceted butt; 11. double convex sidescraper with truncated faceted butt; 12. obsidian flake with basal thinning; 13. pick.
URL http://journals.openedition.org/paleorient/docannexe/image/6087/img-8.jpg
Fichier image/jpeg, 2,0M
Titre Fig. 9 – Pointed tools with impact fractures in the Dashsalakhly lithic assemblage.
Légende 1. burinated fracture; 2. ending fracture; 3–4. flute-like fractures.
URL http://journals.openedition.org/paleorient/docannexe/image/6087/img-9.jpg
Fichier image/jpeg, 1,3M
Haut de page

Pour citer cet article

Référence électronique

Yoshihiro Nishiaki, Azad Zeynalov, Ulviya Safarova et Yaqub Mammadov, « Revisiting the Middle Paleolithic site of Dashsalakhly Cave (Azerbaijan, South Caucasus): Cave and Lithic Assemblage »Paléorient [En ligne], 52 | 2026, mis en ligne le 03 juin 2026, consulté le 05 juin 2026. URL : http://journals.openedition.org/paleorient/6087 ; DOI : https://doi.org/10.4000/16bvs

Haut de page

Auteurs

Yoshihiro Nishiaki

The University Museum, The University of Tokyo, Tokyo – Japan

nishiaki@um.u-tokyo.ac.jp

Articles du même auteur

Azad Zeynalov

Institute of Archaeology and Anthropology & Museum of Natural History, Azerbaijan National Academy of Sciences, Baku – Azerbaijan

Azykh1960@gmail.com

Ulviya Safarova

Institute of Archaeology and Anthropology, Azerbaijan National Academy of Sciences, Baku – Azerbaijan | Department of History and Archaeology, Khazar University, Baku – Azerbaijan

samkhet@mail.ru

Yaqub Mammadov

Institute of Archaeology and Anthropology, Azerbaijan National Academy of Sciences, Baku – Azerbaijan | Department of History and Archaeology, Khazar University, Baku – Azerbaijan

yaqub_amea@mail.ru

Haut de page

Droits d’auteur

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 susceptibles d’être soumis à des autorisations d’usage spécifiques.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search