Altai: Paleolithic

Introduction

The Altai range is located at the cross-roads between Central and Northeast Asia, between 49–52° N latitude and 32–88° E longitude. The Altai plain borders it in the north and the Ob River Basin opens to the west on the Siberian plain. On the northwest border stand the Salair and the Alatau ranges and in the southwest, the western Sayan range. In the south, the Mongolian Altai runs east to the Gobi Desert and joins the Kazakh steppe to the west. In the southwest, the Altai is separated from the Central Tian Shan by the plains of the eastern Balkash, by the Tarbaghatay range, and by the surrounding Zaisan. The plains reach the Xinjang and the Taklamakan desert via the Turpan-Ami depression and southern and Inner-Mongolia via the western edge of the Gobi desert.

The Altai Mountains are the result of a complex and contrasted history. Several periods of complete flattening (e.g., Early Mesozoic, Late Paleozoic) of the landscape have been recorded and the current setting was initiated during the Late Jurassic (Chlachula 2001). The Siberian Mountains are seen as the continuity of uplifting processes starting from the Baikal region, progressing from east to west and reaching the Altai-Sayan during the late Pliocene.

The Altai alpine ridges, such as the Chuya, Saylugem, Katun, and Kuray, include peaks with a maximum elevation of 4,506 m asl. The high plateaus of more than 2,000 m in elevation (e.g., Ukok, Chulyshman, Ulugan) represent a third of the Altai territory and intermediate mountain landscapes, ranging between 800 and 2,000 m asl, account for about half of it. These are mainly located in the northern and western regions and their relief is shaped by active hydrographic networks and by selective erosion. Intermediate mountain landscapes were formed by the erosion of the peneplains of large plateaus (Shahgedanova 2003). Two main lakes, the Markakol and the Teletskoe, are located respectively at the southern and northern edges of the Altai range. The current climate is continental with contrasting seasonal climatic inversions. The winters are generally cold with mean temperatures ranging from −16° C in the foothills to −36° C in the high mountains (Shahgedanova 2003). The minimum temperature is recorded in the Chuya depression, reaching −60° C. Precipitation is stronger in the northwestern part of the Altai, as the southeast is more arid (Chlachula 2001).

The earliest sedimentary deposits in the Anuy river valley were first attributed to the last interglacial, c. 120–110 ka, when karsts were formed due to the exposure of carboniferous formations (Baryshnikov & Maloletko 1997). Such a chronological attribution is consistent with the tectonic activity recorded at the end of the Middle Pleistocene that would be responsible for a deepening of the valley system by about 100 m the Anuy River and about 200 m in the Katun River. This view, however, has been challenged on the basis of radio-thermoluminescence (RTL) measurements providing ages of c. 225 ka for the pebble and gravel layers at the base of the Anuy alluvial plain (Derevianko et al. 2003).

During the first half of the twentieth century, the interest for Paleolithic research in the Altai was stimulated by the collection of surface artifacts at the confluence of the Katun and the Biya rivers and with excavations in the Byisk area. In 1954, S.I. Rudenko reported evidence of Paleolithic occupation at Ust-Kanskaya Cave, along the Charysh River. Rudenko quickly noted some Mousterian typological features on the lithic assemblage, and eventually attributed the human occupation to the Last Glacial based on the fauna recovered. A.P. Okladnikov recognized affinities with the Mousterian from Western Europe and Central Asia and until the discovery of Ulalinka site in 1961, Ust-Kanskaya was considered the most ancient Paleolithic site in the Altai. Okladnikov discovered and excavated important sites in the region, such as Strashnaya Cave, Denisova Cave, and Kara-Bom and remains a major figure in Altai Paleolithic archaeology. His work in Uzbekistan, the Baikal area, Mongolia, and the Far-East led him to consider the existence of a vast Sibero-Mongolian Levallois techno-complex (Fig. 1).

Altai: Paleolithic, Fig. 1

A.P. Okladnikov (left) and A.P. Derevianko (right) (Photo credits: IHMC RAS St. Petersburg, IAE SBRAS Novosibirsk)

Among his students, A.P. Derevianko appears as the most influential contributor to Paleolithic research in the Altai region. He participated in the discovery of the Anuy I open-air site in 1983 and Okladnikov Cave in 1984. At about the same time, excavations started at Maloyalomanskaya Cave, in the Katun basin. Derevianko discovered the site of Ust-Karakol 1 at the confluence of the Karakol and Anuy Rivers. The latter was first excavated in 1986 (sector 1) and excavations at the nearby site of Anuy II started in 1989. Since the beginning of the 1990s, the sites of Karama, Anuy III, and more recently, the Mousterian site of Chagyrskaya were discovered and his team. Following Okladnikov’s legacy, Derevianko undertook numerous international expeditions (e.g., Montenegro, Uzbekistan, Mongolia, Far-East) and developed many collaborations. This approach generated a rich data set upon which synthetic models of population dynamics could be built (Derevianko 2011) (Fig. 2).

Altai: Paleolithic, Fig. 2

Location of the main Paleolithic cave (stars) and open-air (circles) sites in the Altai (Adapted from Nasa Visible Earth). 1 Strashnaya Cave – 2 Chagyrskaya Cave – 3 Okladnikov Cave – 4 Iskra Cave – 5 Karama – 6 Anuy I-III – 7 Denisova Cave – 8 Kamminaya Cave – 9 Ust-Karakol 1 – 10 Ust-Kanskaya – 11 Kara-Bom – 12 Tiumechin 1–4 – 13 Maloyalomanskaya Cave – 14 Kara-Tenesh – 15 Biyka Caves – 16 Ulalinka

Paleolithic sites are mostly located in the northwestern and in the central part of Gorny-Altai, between the Alpine relief and the northern plain. They lie in intermediate mountain zones, generally between 300 and 1,200 m asl. In the northwest, Strashnaya Cave and Chagirskaya Cave are located in the Charysh Basin. Okladnikov Cave, Isrkra Cave, Karama, Anuy I-III, Denisova Cave, Ust-Karakol, and Kamminaya Cave are located in the Anuy basin and Ust-Kanskaya is located further south along the Charysh, in the Central Altai. Ulalinka was found near the city of Gorno-Altaisk, and the Biyka Cave complex and the site of Kara-Tenesh lie upstream along the Katun River. The Tiumechin complex and the Kara-Bom site are located near the Ursul river, and Maloyalomanskaya Cave can be found along the Mala Yaloman, a small tributary of the Katun. Only a few sites such as Barbughazy, Torgun, and Yustid are found in the southeast, and they are mainly attributed to the final stages of the Paleolithic.

Definition

Based on the Eurasian system of division, three main periods can be recognized during the Paleolithic of the Altai: the Lower, Middle, and Upper Paleolithic. The Lower Paleolithic corresponds to the first human occupation of the Altai that would start c. 800 ka. The Middle Paleolithic would start sometime at the end of the Middle Pleistocene and last until c. 50 ka. The Upper Paleolithic covers a time range from c. 50 ka to the end of the Pleistocene.

Key Issues/Current Debates

The Lower Paleolithic

According to Derevianko (2011), the Altai was first colonized by small populations of Homo erectus/ergaster starting from c. 800 ka, that subsequently disappeared from the region c. 500 ka. The site that would best represent this first wave of human occupation is Karama, along the Anuy River (Derevianko & Shunkov 2009). The diversity of the exotic flora is said to fit with a Middle Pleistocene attribution and the RTL dates of 643 ± 130 ka and 542 ± 110 ka have been obtained on the lower portion of the sequence (layers 8–14). Although the identifiable artifacts do not include handaxes, Derevianko and Shunkov tend to emphasize elements that would fit with an Acheulean attribution. The technology is described as cores on pebble blanks and flakes with subparallel dorsal pattern. The tool-kit includes various kinds of scrapers (including naturally backed examples), Clactonian notches, and choppers. More surprising is the reported presence of core-like endscrapers with abrupt retouch. Derevianko acknowledged that the first Lower Paleolithic occupation of the Altai is elusive and likely represents a short-term event. Discovered in 1961, the site of Ulalinka was originally presented by Okladnikov as evidence of Lower Paleolithic human occupation. The chronological attribution beyond the Bruhnes-Matuyama reversal and the authenticity of the lithic artifacts became, however, quickly controversial. The period following the initial peopling of the Altai corresponds to a gap in the archaeological record followed by what is interpreted as a replacement of population (Derevianko 2011). Based on assemblages from the lowermost layers in the Denisova Cave main chamber, Derevianko suggested that a new population settled in the Altai during the Middle Pleistocene, around 300 ka (Fig. 3).

Altai: Paleolithic, Fig. 3

Denisova Cave (Picture by N. Zwyns)

These newcomers bring the first evidence of Levallois and blade technology. Although first mentioned as Mousterian, the archaeological material is also seen as derived from a Late Acheuleo-Yabrudian (Derevianko & Postnov 2004).

The attribution of these assemblages to the Lower Paleolithic is mostly based on RTL dates. At one standard deviation, ages between 330 and 130 ka were obtained on the layer 22. Two inversions of magnetic polarity have been recorded and, following the dating of the layer, interpreted as Biwa I (220–176 ka) and Biwa II (330–266 ka). This contradicts estimations based on tectonic, geomorphology, and small mammals that suggest a formation of the karstic system too closer to the last Interglacial. Furthermore, some authors have warned that the RTL method follows different assumptions than the standard TL and that unbleached particles in cave sediments may be problematic for luminescence dating. In front of the cave, the lower part of the sequence is attributed to the last interglacial and the single magnetic inversion recorded is currently assigned to the Blake episode.

The Middle Paleolithic

The Altai Middle Paleolithic is described as belonging to two main variants. Based on Levallois indexes and on frequencies of Mousterian elements, Shunkov (2005) recognizes a Mousterian variant opposed to a Levallois-Mousterian variant. This variability was interpreted as reflecting different settlement patterns of a single MP tradition. Mousterian assemblages, which are only represented in cave sites, would represent long-term occupations. On the contrary, the Levallois-Mousterian seasonal occupations are associated with open-air contexts. A behavioral ecology approach was further developed by P. Wrinn (2010). Based on the occurrence/absence of formal tools or chips that would testify to tool rejuvenation or raw material management, he classifies MP occupations into three main categories: ephemeral/task specific, ephemeral generalized, and intermittent generalized occupation. His analysis of the fauna and of the lithic frequencies suggests a low intensity of human occupation in the region. The Altai is then seen as a refugium for hominins during cold phases.

The Levallois-Mousterian variant is mainly represented by the assemblages from the middle part of the section at Denisova Cave (layers 20–12), the lower and middle part of the sequence at Ust-Karakol I sector 1 (layers 19–12), and the lowermost cultural layers at Kara-Bom. In addition, the variant possibly occurs at Ust-Kanskaya and Strashnaya Cave. Kara-Bom and Ust-Karakol assemblages show a technology characterized by the production of Levallois points from tabular cores. The reduction is mostly unidirectional but also includes bidirectional instances, with occasional removals of debordant blanks to reshape the convexities of the core. The reduction system includes the production of elongated spalls. Although it should be supported by additional chronometric data, the Levallois-Mousterian would appear during the OIS5e, c. 120 ka and eventually disappear from the region at the beginning of OIS3.

The Mousterian variant is defined based on the material from Okladnikov and Chagirsakaya Cave. Recently, Derevianko and Markin suggested that these two sites illustrate the existence of a distinct MP facies, relatively late, and intrusive, which they call the Sibiryachikha variant (Derevianko & Markin 2011). It is characterized by a lack of Levallois elements and by a good representation of Mousterian retouched tools, such as various types of sidescrapers and notches. These assemblages seemingly appear at the beginning of OIS3, around 55–50 ka, and last at least until the middle part of OIS3, c. 37 ka. The Sibiryachikha assemblages are said to represent Neandertal populations moving across Central Asia (e.g., Teshik-Tash) and subsequently penetrating in the Altai under the demographic pressure of the spreading Modern Human populations.

In sum, two variants can be distinguished among the MP assemblages from the Altai. The Levallois-Mousterian is better represented in the archaeological record and seems to predate the appearance of a more elusive, and apparently intrusive Mousterian. How much seasonal, functional, or cultural factors are responsible for such variability is not yet clear.

The Upper Paleolithic

The local Levallois-Mousterian is said to have gradually evolved into two main UP variants, the Ust-Karakol and the Kara-Bom variants. Derevianko (2011) sees both UP trends as the result of an incipient evolution from a local Middle Paleolithic background. In his model, the latter process is used to support a scenario of multiregional emergence of modern human anatomical and behavioral features. Other authors have underlined the role of site-formation processes to explain the apparent transitional character of some assemblages or the regional discontinuity of human occupation (Wrinn 2010; Zwyns 2012). Following these views, the Altai data set may support to models in which population movements more consistent with the Out-of-Africa hypothesis.

The Kara-Bom variant (or Initial Upper Paleolithic) is defined on the basis of levels OH5 and OH6 from the eponymous site, but is also described at Ust-Karakol 1 (sector 1, OH5.5 and OH5.4), Kara-Tenesh, Maloyalomanskaya Cave, and in the Byike complex. The defining technological features are clearly expressed in the production of laminar blanks (Fig. 4).

Altai: Paleolithic, Fig. 4

Initial Upper Paleolithic (the Kara-Bom variant) from Ust-Karakol 1 sector 1. 1, 2, blade with inverse proximal retouch; 3, bidirectional blade core; 4, burin-core. (Drawings by N. Zwyns)

Asymmetrical blade cores were reduced from two opposed platforms. The main flaking surface is usually located on the broad face of the core and one of the narrow faces is used to reshape lateral convexities. Large and robust blade blanks are produced and retouched whereas thick technical side blades were turned into burin-cores to detach small laminar blanks (Zwyns et al. 2012). This technology is seemingly associated with an early appearance of ornaments, starting from c. 42 ka. The Kara-Bom trend has been considered as an example of local and gradual transformation of the technology observed in the underlying Middle Paleolithic layers (Derevianko 2011).

The Ust-Karakol variant (or Early Upper Paleolithic) first occurs during the middle phase of OIS3. It is marked by the production of small laminar blanks from narrow-fronted and carinated cores reduced by direct percussion. It is associated with the generalization of bone tools and fully developed forms of ornaments. The defining assemblages are Ust-Karakol 1 (sector 2) layers 11–8, but other assemblages from Anuy I-III, Strashnaya upper levels, Tyumechin-4, and Ushlep 6 are considered analogous. It is said to have evolved from a Levallois background assigned to OIS5e at Ust-Karakol 1, sector 2, (layers 19–18). Originally described as non-Levallois, the central chamber of Denisova Cave was later said to illustrate the same gradual development, between 100 and 30 ka, toward genuine UP (layers 11 and 9). Although the Ust-Karakol trend is sometimes listed as a possible source for the spread of microblade technology in Northeast Asia, the early UP assemblages display no clear evidence supporting an early use of pressure flaking.

A recent study based of the laminar technology of Kara-Bom and Ust-Karakol variant supports the existence of two main technical traditions at the beginning of the Upper Paleolithic (Zwyns 2012). By comparison with Europe and the Levant, the Kara-Bom variant is assigned to the Initial Upper Paleolithic (IUP) to describe the presence of derived UP features without necessarily implying a local transition. The IUP definition reinforces analogies proposed by numerous authors (e.g., Derevianko 2011) with sites from the Cis-Baikal (e.g., Makarovo-4), Trans-Baikal (e.g., Tolbaga, Barun-Alan, Khotyk, Kamenka A and C, Varvarina Gora, Podzvonkaya), or Mongolia (Tolbor 4). The IUP is followed by an Early Upper Paleolithic (EUP) phase with a shift toward developed UP features (Ust-Karakol variant). The latter correspond to a series of behavioral changes in terms of economy (emancipation of the bladelet production) knapping techniques (shift to soft hammer), but also probably in terms of weaponry (composite spears). Although they are poorly dated, the two phases appear separated by a significant time span and by climatic events such as Heinrich 4 (Fig. 5)

Altai: Paleolithic, Fig. 5

Main models of Upper Paleolithic emergence in the Altai: summary (see text for references)

.

According to the chronometric data, the Upper layers of Anui 2 document human presence during a time span that covers the end of OIS3 until the Late Glacial Maximum. It is not yet clear if human groups inhabited the lower valley of the Altai during the LGM or if the record illustrates a regional discontinuity in human occupation. It seems, however, that the occupational hiatus is more clearly marked than in the piedmonts. Starting from OIS 2, the blockage of river systems driven by the expansion of mountain glaciers led to the multiplication of lakes. Eventually, periodic drainage events may have caused catastrophic floods that may represent a source of bias in the archaeological record for the period between 26 and 13 ka.

Human Remains

At Okladnikov Cave, dental remains occur in stratum 2 with a left lower molar, but also in stratum 3 with a left lower premolar, a left lower molar, and a right lower molar. Postcranial remains are found in stratum 2 with an adult humerus, and in stratum 3 with a subadult humerus and a hand phalanx. The good preservation of the collagen extracted from the phalanx, from the subadult humerus, the femur, and from the adult humerus has led to the reconstruction of the mtDNA sequence which has been identified as Neandertal (Krause et al. 2007). The archaeological assemblage is Mousterian but the dating of the human occupation is far from clear. Direct dates would indicate that the Neandertal presence in the area lasted until at least 37 ka 14C BP.

At Strashnaya Cave, eight deciduous teeth presumably belonging to a single individual come from an unclear stratigraphic context that can be attributed to the UP. The specimens could not be assigned to a clear taxon. At Maloyalomanskaya Cave, the discovery of a single human tooth has been reported, but no detailed description has been published.

At Denisova Cave, two teeth have been found in the 1984 collection from the central chamber. Denisova 1 is an upper central incisor found in layer 12. Although it has been previously published as a human tooth, Viola et al. (2011) assign it to a worn incisor from a large bovid. Denisova 2 is a deciduous molar (right first lower) found in layer 22.1 that would date to at least OIS 5e. In 2000, layer 11.1 of the south gallery yielded a tooth belonging to a young adult and identified as a third or second upper molar (Viola et al. 2011). Denisova 3 is the proximal epiphysis of a juvenile manual phalanx uncovered in layer 11.2 of square D2 of the east gallery. The phalanx belongs to a distinct individual with an age evaluated at around 6–7 years old. Both mtDNA and nuclear DNA were extracted from these remains, resulting in the identification of a hitherto unknown archaic hominin (Krause et al. 2010; Reich et al. 2010). The nuclear DNA indicates that these hominins, referred to as “Denisovans,” belong to a lineage sharing a common origin with Neandertals that post-dates the split with MH ancestors. Initial morphological descriptions of the tooth noted a set of archaic features not seen in Neandertals or in early modern humans, further suggesting a distinct evolutionary history.

At the recently discovered Chagysrkaya Cave, Viola and colleagues (2011) described human fossils associated with layers 6b and 6c. Chargyrskaya 1 is a worn upper deciduous canine and Chagyrskaya 2 is an atlas fragment. Both fossils are associated with layer 6b. Chagyrskaya 3 is an upper premolar and Chagyrskaya 4 is a lower incisor. Both are worn and small and would fall outside the range of Neandertals. Nevertheless, the mtDNA is under reconstruction; although the results are not fully published, they seem to indicate an attribution to Neandertals (Viola et al. 2011). The rich lithic and fauna assemblage is assigned to the Mousterian and the excavation is still ongoing.

Future Directions

The recent identification of the Denisovans in the Altai opens up new perspectives for understanding the peopling of Asia. Genetic data suggest that, apart from some present-day populations from Melanesia and Australia, Denisovans did not contribute significantly to the genome of Eurasian populations (Reich et al. 2011). The current geographical distribution of the Denisovan genetic input lead to the hypothesis that the Altai might represent the northern edge of their territorial expansion. Where and when they did interbreed with MHs is still unclear and the current data do not rule out other sterile encounters.

The human fossils from Denisova Cave have not been directly dated. New radiocarbon dates were produced on cut-marked bones and bone tools that belong to the same layers as the Denisovan fossils. The results indicate post-depositional admixtures from subsequent and underlying layers (Reich et al. 2010). The chronological and cultural attributions of the Denisova hominins remain, therefore, uncertain. Although it shows the first occurrence of modern behaviors, the IUP assemblages predate the earliest known MH fossil from Siberia. According to the data at hand, the latter appears contemporaneous of the EUP. The Altai Neandertals are, up to now, the only taxon clearly associated with an archaeologically defined unit, namely, the MP of Mousterian variant. The makers of the Levallois-Mousterian variant are still unknown and the identity the IUP populations is yet to be found.

Finally, the apparent complexity of the Altai Paleolithic human occupation likely reflects a lack of reliable and comparable chronological data. Recent steps taken toward the identification of human remains in clear archaeological, taphonomic, environmental, and chronological contexts have already proven to be highly valuable. Further efforts to acquire high-resolution datasets are, nevertheless, essential in order to test current models. There is a few doubt that it will provide new keys for the understanding of the peopling dynamics in a region that stands as a gate between Central and Northeast Asia.

Cross-References

Fossil Records of Early Modern Humans

Mongolia: Paleolithic

Mousterian Industry Tradition

Neanderthals and Their Contemporaries