In modern archaeology, Middle Palaeolithic cultures in Europe and Western Asia are primarily associated with Neanderthal, and - during the emergence and spread of industries of the Late/Upper Paleolithic - with the dispersion of anatomically modern humans (AMH) across the Eurasian continent. The process of cultural change was presumably initiated by the dispersion of archaic Homo Sapiens from Africa (Out of Africa ... 2010; Groucutt et al. 2015). Starting from about 50 ka ago, in various regions of Eurasia, changes are recorded in key elements of material culture associated with the emergence of new techniques for the splitting of stones, specific types of tools, the widespread use of bone material and the creation of non-utilitarian items (jewellery, art etc.). Establishing the causes, nature, routes of dispersion and especially the timing of this phenomenon is one of the central problems of world archaeology. In this context, the identification, at the end of the 20th century, of the stone industries of the initial Upper Palaeolithic (IUP) is one of the notable achievements of Palaeolithic archaeology. These complexes were initially characterized through materials from Boker-Takhtit and the Ksar Akil grotto in Levant (Douka et al 2013), and have since been identified in many regions of Eurasia (Vishnyatsky 2008; The Early ... 1988).
There are several known centers of IUP industry: Central and Eastern Europe, the Middle East, and some regions of Central Asia. Whereas it is indisputable that similar lithic technologies can be found in all of these areas, it is not self-evident that they represent a unified cultural phenomenon. An alternative possibility is convergence, which commonly responses to adapting Levallois technology in the production of blade blanks, or some combination of multiple local origins with subsequent dispersal. In almost all cases, chronology is the key for establishing the origin(s), direction and speed of technological spread, and for understanding the nature and timing of interaction with the “aboriginal” cultures in different territories.
In Central Asia, complexes associated with the early stages of the Upper Palaeolithic have been discovered relatively recently, and they seem to be less numerous than in the western part of the continent. Most of these sites are concentrated in northern and northeastern border – Altai, Transbaikalia and Northern Mongolia (Kara-Bom, Ust-Karakol-1, Kamenka, Podzvonkaya, Tolbor-4 etc.) (Derevianko et al. 1998; Derevianko et al. 2003; Derevianko et al. 2007; Rybin 2015). All these assemblages are oriented towards mass production of blades, have common primary splitting technologies and same forms of specific tools, usually include non-utilitarian items. This variant of the IUP was called Siberian-Mongolian or Asian (Rybin, 2015). The question of its origin is currently open. Two main hypotheses are currently being discussed. This could be either migration of carriers of these industries from Western Asia, about 50 ka ago (Rybin, 2015) or formation of local cultural features around the same time in Gorny Altai as a result of autochthonous development based on the final Middle Paleolithic industries recorded in Denisova Cave and Kara-Bom (Derevianko et al 2003). In the context of solving this problem, each new stratified site, outside the known area of distribution of the IUP in the region brings additional information in support of a particular hypothesis, especially if there is a reliable chronology. Until recently, vast areas of Central Asia were excluded from the IUP-problems due to the absence of any sites, but in recent years the situation has changed dramatically.
The archaeological sites of Kazakhstan occupy a broad territory connecting several large historical and cultural regions (South Central Asia, Siberia, Northern China and Eastern Europe). This region is distinguished by specific geographic conditions: a pronounced continental arid climate, and dominant denudation processes. This has resulted in the small number of Stone Age sites identified in the region, mostly represented by surface localities. Upper Palaeolithic materials occurring in situ are present at a number of sites in South Kazakhstan - Maibulak, im. Ch. Valikhanova, Rakhat et al. [Taymagambetov, Ozherelyev, 2009; Ozherelyev et al., 2019; Fitzsimmons et al., 2017]. There are several sites with stratified Late Palaeolithic industries in Central and Northern Kazakhstan (Batpak-7, Ekibastuz-15, Ekibastuz-18) (Taimagambetov and Ozherelyev 2009). In the eastern part of Kazakhstan, and despite its proximity to the Russian Altai (known to be rich in Palaeolithic sites) stratified complexes of the Upper Paleolithic were practically unknown until recently. However, in 2016 a joint Russia-Kazakhstan expedition discovered the Ushbulak multilayer site (Fig. 1A), and this site has now become key to studying the Upper Palaeolithic stages of the region (Shunkov et al. 2017).
Geomorphology. The Ushbulak site is located 1500 m asl at the foot of the southwestern slope of the Saur Ridge, in the northeastern end of the Shiliktinskaya valley (Fig. 1A) at the junction of the intermountain Shiliktinskaya depression with the Ridge. Tectonic uplift is recent and gives rise to a young tectonically-controlled relief (Geology of the USSR ... 1967).
The base of the southwestern macroslope of the Saur Ridge is characterized by significant slopes (from 15 to 25º), eroded by deeply incised gorges with steep sides and usually perennial watercourses; the Ushbulak site is found in one of these gorges. The transition from the southwestern macroslope of the Saur ridge to the Shiliktinskaya depression is morphologically clearly expressed by a sharp change in slope. A gently sloping piedmont surface begins at the foot of the ridge, formed by a mantle of deluvial-proluvial deposits and merged colluvial fans emerging onto the plain.
Modern relief-forming processes are largely determined by the landscape-climatic conditions of the region: continental climate, cold winters and hot summers (Gidrogeologiya SSSR… 1971). Southern windward slopes of the Saur ridge receive more than 700 mm of precipitation per year, and snow reserves are up to 300 mm; these cause a violent spring flood and, as a result, intense erosional activity within mountainous areas, accompanied by an active removal of material within the depression.
Archaeological material has been found both within and on both sides of the upstream channel within the ravine formed by flow from the Vostochny spring (Fig. 1C). The ravine cuts 6 m into the surface of the foothill plain; it has a V-shaped profile with a narrow bottom up to 3–5 m wide. Water emerges from loose sediments at the very foot of the slope of the Saur ridge, with a flow of ~6 l/s. The spring is located in a northwestern trending fault zone which controls massive intrusions of granodiorites and diorites; the water source appears to be associated with fractures in the intrusive complex.
Stratigraphy. The sedimentological sequence of the Ushbulak site reflects the successive replacement of alluvial-proluvial activity by processes of predominantly deluvial-slope movement and accumulation. The composite section, based on the description of two excavations and a series of exploratory pits on both sides of the stream, contains eight main lithological layers (Fig. 2 and Table 1).
L8 (6.8 (visible) - 5.9 m): Poorly sorted rubble-gravel sediments with the inclusion of randomly oriented single blocks in a heavy loamy pore-type aggregate. The petrographic composition is predominantly of local rocks but contains material exotic to the modern catchment geology. Archaeologically sterile. Presumed to represent ancient proluvial-mudflow accumulation.
L7 (5.9 – 5.6 m): Rubble-gravel with sandy-loamy pore-type filler of ochre-brown colour. Petrographically crushed stone, 90% from rocks of the nearest sources. Three generations of sediment were identified, occurring with a dip of about 5–7º, resulting in insets within each other and in the underlying layer 8. Genetically, layer 7 is sediment from a small stream with a variable hydrological regime, washing over the underlying coarse-detrital sediments.
L6 (5.6 – 4.6 m): Represented by two generations of sediment. The lower layer is gray heavy loams with lenses of coarse-grained sands at the bottom of the layer. The upper deposit is gray sandy loam 40-50 cm thick with thin lenticular inclusions of light, organic reach loam, mainly of proluvial origin. Genetically, the interlayers of heavy loams and organically rich sandy loams of layer 6 were deposited by a low-flow stream without a morphologically pronounced channel giving rise to peats in some areas. There is a tendency for the inclination of the sediments to increase from 2–3º at the bottom of layer 6, to 5–6º towards the top. From the lithology and occurrence of archaeological material, eight sub-horizons have been identified.
L5 (4.6 – 3.5 m): A unit of interbedded sands brown-gray and rusty-ochre, silty sandy loam and loam, with abundant weathered granite gravel and inclusions of fine granite rubble. There are signs of erosional deposition of rubble layers at the base of layer 5 into the top of the underlying sediments (layer 6), reflecting an increase in proluvial activity.
L4 (3.4 – 2.7 m): Fine-grained sands and ochre sandy loams, underlain by an interlayer of poorly sorted gravel material with a sandy loam filler of variable thickness. They are represented by two horizons of a genetically single proluvial complex in zone of active accumulation.
L3 (2.7 – 1.6): Light sandy loams, pale-and gray-brown, with interlayers enriched in gravel-sandy material. Three horizons are distinguished, reflecting the transition from predominantly proluvial to deluvial processes.
L2 (1.6 – 0.4 m): Light gray silty sandy loam, with abundant gravel and saprolitized rubble. Three horizons are distinguished, genetically related to the activity of proluvial and colluvial processes of varying degrees of intensity.
L1 (0.4 – 0.0 m): Modern soil.
Archaeology. At present, there are two detailed excavations and 12 pits with a total area of 40 m2 at the site (Fig. 1B). Archaeological material has been recorded in seven main lithological layers (layers 7-1). Based on the technical and typological characteristics of the artefacts, their stratigraphic position, as well as the accompanying faunal remains, four cultural and chronological complexes were identified at the site: the initial stages of the Upper Paleolithic (layers 7.2–5.2), the advanced Upper Paleolithic (layers 5.1–4), the Final Upper Paleolithic (layers 3.3–2.1), and the Palaeometal Epoch (layer 1) (Shunkov et al. 2019).
The primary reduction in the lithic industry of the lower layers of the site (~16 thousand artefacts) (Fig. 3) is characterized by the complete predominance of two-platform cores of counter-blade splitting; mainly elongated spalls, some large, up to 30 cm long; and extensive use of picketage in the preparation of the splitting zone. The tool kit includes end-scrapers, intensively retouched blades, truncated-faceted and spine-shaped items, as well as specific tool forms: blades with an interception, items with a ventral thinning of distal edge. A number of features allows confident attribution of this complex to the initial stage of the Upper Palaeolithic (Anoikin et al. 2019).
The upper layers are significantly poorer in archaeological material (~1300 artifacts), although they also show artefacts of characteristic lithic production (see Fig. 3). For example, in the primary reduction in layers 5.1–4, end-face and single-platform double-front forms of cores for micro-blades and small blades are recorded. Layer 3 is characterized by variations in small-lamellar and micro-lamellar double-flat cores with counter-cleavage and unidirectional forms, and in the most recent materials (layer 2), the appearance of prismatic cleavage is recorded. The tool kit is rather monotonous: single end-scrapers of various modifications, retouched knives and micro-blades (Anoikin et al. 2019).
The Ushbulak site appears to be unique in the region, with industries of different stages of the Upper Palaeolithic represented in a conformable occurrence. This makes it possible to reconstruct the chronology of the presence and evolution/replacement of these cultures throughout the second half of the Late Pleistocene. Especially because of these unique characteristics compared to other archaeological sites of Central Asia, one of the most important tasks is to develop a reliable chronology for the accumulation of the site, and particularly the cultural horizons (Kurbanov et al. 2021). We have undertaken detailed luminescence dating using optically stimulated luminescence (OSL) from quartz and infrared stimulated luminescence (IRSL) from K-rich feldspars, supplemented by AMS 14C dating of charcoal and bone material from layer 6.