TOLBOR PALEOLITHIC HOARD (NORTHERN MONGOLIA)*

The article presents the results of studying a unique complex of stone artifacts (57 flakes) found in the valley of the Ikh-Tulbariin-Gol river (Northern Mongolia). The context of the discovery and technical and typological characteristics allow us to diagnose it as a treasure trove of stone tools left by Upper Paleolithic hunter-gatherers. Such a complex was first found on the territory of Mongolia. Selective traceological analysis of artifacts from the treasure indicates that they could have been used for working with wood. Additional information about the possible marking role of the treasure in the system of Stone Age monuments in the valley of the Ikh-Tulbariin-Gol river is provided by maps compiled on the basis of GPS data and terrain computer modeling programs.

Keywords: Northern Mongolia, Paleolithic, treasure trove, technology, tracology, hunter-gatherers.

Introduction

The valley of the Ikh-Tulbariin-Gol River (right tributary of the Selenga River, Northern Mongolia) (Figure 1) is a very productive research site for a joint Russian-Mongolian-American expedition to study Stone Age monuments. The first explorations conducted here in 2002 revealed more than a dozen promising sites with artifacts from the Late Pleistocene and Early Holocene periods. In subsequent years, two multi-layered stratified monuments were studied permanently: Tolbor-4 (2004-2007) and Tolbor-15 (2008-2011). The main results of these studies were published [Derevyanko et al., 2007; Gladyshev, Olsen, Tabarev, 2010; Gladyshev et al., 2010; Gladyshev, Gunchinsuren, Rybin et al., 2011; Gladyshev, Tabarev, 2009; Gladyshev, Tabarev, Olsen, 2010].

* This work was supported by the Russian Foundation for Basic Research, project No. 12-06-00037a "Technological portraits of Upper Paleolithic industries in Mongolia". The authors express their sincere gratitude to their colleagues Prof. K. Akoshima (Tohoku University, Japan) and Prof. M. Kornfeld (University of Wyoming, USA) for valuable advice and comments on various subjects of the publication, as well as to the artist A. V. Abdulmanova for the graphics for the article.

1. Map of Mongolia showing the position of the valley of the Ikh-Tulbariin-Gol river.

In 2010, in order to organize the numbering of previously known monuments and search for new ones, a detailed survey of the left side of the Ikh-Tulbariin-Gol River valley in its middle and lower reaches was carried out. This section of the valley is represented by a series of terraced deluvial plumes cut by deep canyons, ravines and dried-up streambeds. Lifting material was recorded at almost all levels-from the high floodplain to the rear seams and watershed. Intensive testing of sites with the highest concentration of artifacts resulted in the identification of three sites (Tolbor-16 - 18) with thick (up to 3 m) cultural deposits.

Of particular interest is the original closed complex of 57 large flakes, which, judging by the location and nature of the material, is a treasure trove of Paleolithic products, the first known on the territory of Mongolia. Preliminary data on the complex were published in several articles (Gladyshev, Olsen, and Tabarev, 2010; Gladyshev, Popov, and Tabarev, 2010; Gladyshev, Gunchinsuren, and Popov et al., 2011; Gladyshev et al., 2011). A detailed study of the material and additional local studies conducted in 2011-2012 make it possible to provide specialists with a more detailed description of the Tolbor Paleolithic treasure.

General technical and typological characteristics of the treasure

The complex, called the Tolbor Paleolithic treasure, was discovered during a survey of the slopes of a deep (up to 15-17 m) ravine with traces of watercourses in the spring season. Judging by the development of the riverbed and the degree of rolling of pebbles, in the past there could have been a fairly powerful stream (the left tributary of the Ikh-Tulbariin-Gol river). Unimpressive and sparse lifting material was collected on the slopes. In the two pits laid on the most level areas, virtually nothing was found; the thickness of loose sediments, which are usually associated with the artifacts of the final Pleistocene, was 15-25 cm in them.

When viewing the shallow plume from the southern side of the ravine (49°14 ' 07 "N, 102°56' 04 " E, height above the river's edge approx. 40 m) (Figs. 2, 7), on a small area clearly visible con-

2. Tolbor Paleolithic clade.

1 - location of the treasure, south-eastern exposure; 2-accumulation of flakes before clearing.

centering of large flakes. Several of them were washed to the surface by the previous day's downpour, but the bulk of them were in the layer (Figs. 2, 2). After preliminary clearing of the site, a small excavation (50x50 cm) was marked out and the complex was dismantled. All the artefacts were located in a gray-brown loess layer in the range of 1-15 cm below the daytime surface on an area of approximately 30x30 cm (Figs. 3, 1). The density of the cluster suggests that initially they were neatly stacked in some kind of container (bag) and partially shifted as the organic material decomposed. Unfortunately, the stratigraphy does not allow us to state unequivocally whether a special depression was created for the treasure, or whether it was placed in a natural fold on the slope and covered with turf. It is possible that the integrity of the complex was minimally damaged, and two or three artifacts were lost due to the already mentioned rainstorm or were thrown down the slope by grazing goats, but this does not detract from the significance of the find and the value of information.

The complex consists of 57 flakes-billets of dense dark gray siliceous shale (Figs. 3, 2), typical of this river valley and materials related to the Paleolithic (Tolbor-4 and - 15).

In addition, all the flakes are more or less covered with a carbonate crust, which is an indirect sign of the Pleistocene age of the artifacts. Thus, according to its raw material characteristics, the complex is monolithic and local.

According to formal criteria, only 12% of blanks have a length more than 2 times greater than the width (see the table). However, they can not be attributed to plates, because there is no characteristic cut of the dorsal side, no parallelism of the edges, or traces of the prepared site. All blanks in the complex under consideration are flakes. It should be noted that" plate hoards " are known on the territory of Eurasia. The best example is a complex of 130 plates found in 1970 in Grand Presigny (Western France) (Pelegrin, 2006).

According to the classification adopted for working with the Paleolithic industry of Tolbor monuments (see, for example: Gladyshev, Gunchinsuren, Rybin et al., 2011), large flakes are considered to have at least one of the parameters (length, width) of 5 cm or more. Such flakes in the complex are 95 % (53 specimens), the rest (4 specimens, 5 %) are average (one of the parameters is 2.5 cm or more). Moreover, in 28 % (16 specimens) of flakes, one of the parameters exceeds 10 cm, which, according to some experts, allows them to be classified as macro (Figure 4). In 48 specimens (84%), the sum of two measurements is 10 cm or more (with a thickness of no more than 1.5 cm)., this is also an indicative feature of the complex. This percentage of medium, large, and macro-splinters is completely out of character for monuments such as workshops or workshop sites. This suggests that this is the result of a purposeful selection of blanks obtained by splitting large cores. This assumption is confirmed by another fact: only two chips (3.5%) have preserved areas with a pebble crust, the vast majority of blanks-

3. Tolbor Paleolithic clade. 1 - view after clearing; 2 - contents of the treasure.

Main dimensional characteristics of artifacts from the Tolbor Paleolithic hoard

Figure 4. The largest artifact (macro split) in the treasure trove.

these are not primary, but secondary flakes removed after preliminary preparation of the chipping front. The complex also does not have any clearly applied flakes, i.e. not all chips are represented, but only those that correspond to certain parameters. 57 % (34 specimens) of the flakes completely or partially preserved the impact pad. In all cases, without exception, it is not faceted, the impact (most likely, a rather heavy stone bump) was applied on the natural surface at an angle of 55-65°. Neither as part of the complex, nor in the immediate vicinity

there are no traces of debiting (small flakes, splinters, flakes, etc.) that inevitably accompany workshops and work sites in short-term parking lots. There are also no whole or broken pebbles that can be interpreted as bumpers, no blanks of nuclei, no depleted nuclei, or pieces of raw materials. Morphologically distinct tools or fragments of them are also missing. The blanks were deliberately selected and brought to this place. All this indicates the specific nature of the complex and the possibility of interpreting it as a treasure trove.

It should also be noted that the flakes under consideration do not contain traces of ochre or other dyes typical, for example, of the Clovis tip hoards of the Paleoindean period in North America (Tabarev, 1999, 2009). Ochre in these contexts, according to researchers, indicates the ritual nature of the complex or the prestigious nature of technologies (Hayden, 1998).

Results of functional analysis of artifacts from the treasure trove

First of all, it should be noted that until now, extensive collections from the Paleolithic monuments of Mongolia have not been the object of traceological study. Selective examination of artifacts from the Tolbor Paleolithic hoard was, in fact, the first experience of functional analysis of materials from this region.

A small Olympus BH microscope was used in the x100, x200, and x400 magnification modes. Six flakes were selected for analysis (N 7, 12, 22, 29, 44, 54), on which there were traces of regional episodic retouching (Fig. 5). The analysis of artifacts gave very interesting results. Three of the flakes showed slight but legible traces of recycling. For their classification, we used a model developed and successfully applied by Japanese specialists from Tohoku University (Sendai) [Serizawa, Kajiwara, Akoshima, 1982]. Wear on artifacts N 7 (along the distal end from the dorsal and ventral sides), 29 (along the notch from the ventral side) and 54 (along the edge from the ventral side) (Fig. 6) corresponds to type B with a parallel arrangement of traces and bright polishing, which indicates the contact of the stone product with wood. Minor wear is also recorded along the edge of the notch at the distal end of the N22 chip, but it was not possible to identify the contact material. Presumably it can also be a tree. No traces of disposal were found on artifacts N 12 and 44.

Thus, individual flakes from the complex could be used occasionally. Whether it was before they became part of the" treasure", or after, it is difficult to say. In any case, the facts of the use of artifacts from the hoards of stone products in the archaeological literature are given. For example, such data were obtained during excavations (1970-1990s) and analysis of materials from the Nogava monument (pref. Miyagi, Honshu Island, Japan), dated in the range of 11.0-10.3 Ka BP (Tabarev, 2011; Kelly, 1972; The Nogawa Site..., 1996). Recesses were recorded within the proposed housing structure, which contained compact sets of stone products (complexes). There are approximately 450 artifacts in Complex 1 -

Fig. 5. Flakes selected for tracological analysis (numbers according to the table).

6. Flakes with identified traces of use and micrographs of these areas.

four large flat bifacial products (more than 20 cm along the largest axis) were placed on a pile of flakes. According to Japanese experts, they could be stored in a container made of bark or a bag made of leather (Kanomata, 2006, 2010). Bifacial products are nuclei for obtaining flakes. The flakes were used as end scrapers, and analysis of the traces of disposal on them indicates woodwork. This complex is interpreted as a "hunting set". In complex 2, there were more than 50 large and medium-sized flakes. They were used as scrapers, and traceological analysis indicates that animal skins were processed (Kanomata, 2010).

Another similar example is a treasure trove of 171 flakes at the Ustinovka-3 site in Primorye. The nature of the accumulation of artifacts suggests their initial presence in a container made of soft material placed between stones (Kononenko, 1996, p. 122). Five flakes showed traces of intense harmony (Kononenko, 2001; Hunter-Gatherers..., 2003, p. 104).

The difference between the Tolbor hoard and these complexes is one - it was found outside the boundaries of the parking lot or settlement and, in our opinion, illustrates an element of a slightly different strategy. Partial confirmation of this was obtained by analyzing the location of the treasure in the Ikh-Tulbariin-Gol river basin.

Location of the treasure in the system of Paleolithic monuments in the valley of the Ikh-Tulbariin-Gol river (preliminary conclusions)

One of the most important tasks of the 2011 field season was the compilation of detailed computer reports.

maps of the location of Stone Age monuments in the Ikh-Tulbariin-Gol river basin. It was performed using a GPS navigator and the Trimble Juno SB GPS and ArcGIS software package. This work, in addition to the planned results, also led to one very interesting observation concerning the Tolbor Paleolithic hoard.

Analysis of maps with relief features of the right tributaries of the Selenga River showed that the shortest (approx. 7 km) and most comfortable way in the western direction from the valley of the Ikh-Tulbariin-Gol River to the neighboring valley of the Altaatyn-Gol River is located in the middle course of the Ikh-Tulbariin-Gol River just opposite the plume with the location of the Tolbor treasure. Walking along this route was accompanied by the collection of expressive lifting material and detailed mapping of the area.

It turned out that, despite the fairly dense forest, from the highest point of the watershed, even the naked eye can clearly see the section of the left bank of the Ikh-Tulbariin-Gol River and the location of the Tolbor Paleolithic treasure. A computer model of this section compiled from GPS measurements (Fig. 7) clearly demonstrates:

- the treasure is located almost exactly in the eastern direction relative to the highest point of the watershed;

- the location of the treasure offers a view of the entire middle course of the Ikh-Tulbariin Gol river, the path to the neighboring valley and the highest point of the watershed;

- the watershed crossing connects the valleys of two right tributaries of the Selenga River into one system. A preliminary survey of the neighboring valley has already confirmed the presence of a number of Stone Age sites and the prospects for stationary studies (Bolorbat et al., 2012).

The Tolbor Paleolithic hoard was not accompanied by organic remains, so its dating today is possible only on the basis of the above technical and typological indicators. The absence of any direct or indirect signs of plate and microplate technologies makes it impossible to link the complex with either the early Upper Paleolithic or Late Paleolithic industries of Northern Mongolia. Its pronounced flake character corresponds to a fairly wide chronological period - 25-15 Ka BP (Gladyshev et al., 2011).

The functional purpose of the treasure, in our opinion, corresponds to the peculiarities of hunting strategies of the inhabitants of the middle course of the Selenga River in the Late Paleolithic. Most likely, this is not a ritual complex, but a hunter's set. A number of researchers suggest that the creation of hoards consisting of blanks for hunting tools (flakes, plates, bifaces) is a reaction to climatic features and reflects increasing mobility [Kanomata, 2006, 2010; Kornfeld and Tabarev, 2009]. In this case, the appearance of the Tolbor treasure is associated with a relatively cold time in the region and less comfortable conditions for the search and exploitation of stone raw materials.

At the same time, the features of the location, the connection with the transit area between two river valleys allow us to talk about the symbolic, marking function of the treasure. Almost all hoards of stone products from the Paleoindean period in North America were associated in one way or another with some notable relief elements - a rock, a grotto, a river bend, etc. [Tabarev, 1999, 2009; Frison and Bradley, 1999; Kilby, 2008].

In any case, the discovery of a treasure trove of Paleolithic artifacts in Mongolia is extremely important in itself. It expands both our understanding of hunter-gatherer behavioral patterns in the region and the geography of Paleolithic hoards known in Eurasia.

Fig. 7. Computer model of the Ikh-Tulbariin-Gol River valley section, a - treasure trove; b - surfaces visible from the location of the treasure; c - the highest point of the watershed; d - Stone Age monuments; e - road; d-Ikh-Tulbariin-Gol River; f, w - horizontal lines after 500 and 100 m, respectively.

List of literature

Bolorbat Ts., Gladyshev S. A., Gunchinsuren B., Odsuren D., Tabarev A.V., Hatsenovich A.M. Arkheologicheskie razvedki v bassein reki Altaanyn-gol (Severnaya Mongolia) [Archaeological exploration in the Altaanyn-gol River basin (Northern Mongolia)]. Problemy arkheologii, etnografii, antropologii Sibiri i sopredel'nykh territorii: mat-ly of the final session of the Institute of Archeology and Ethnography SB RAS 2012. Novosibirsk: Publishing House of IAET SB RAS, 2012, vol. XVIII. - P. 25-29.

Gladyshev S. A., Gunchinsuren B., Popov A. N., Tabarev A.V., Bolorbat Ts., Odsuren D. New Stone Age monuments in the valley of the Ikh-Tulbariin-Gol river, Northern Mongolia (based on 2010 materials) / / Archeologijn sudlal. - 2011. - Vol. XXX. - pp. 51-64.

Gladyshev S. A., Gunchinsuren B., Rybin E. P., Tabarev A.V., Bolorbat Ts., Odsuren D., Olsen D. Results of the study of the multilayer paleolithic site Tulber-15 in 2010 / / Archeologijn sudlal. - 2011. - Vol. XXX. - pp. 21-50.

Gladyshev S. A., Olsen D., Tabarev A.V. Novye dannye po verkhnepaleoliticheskim kompleksam Severnoi Mongolii i problema rasprostraneniya plastinchatykh i mikroplastinchatykh industrii v dalnevostochnom regione [New data on Upper Paleolithic complexes of Northern Mongolia and the problem of spreading lamellar and microplastic rocks in the Far Eastern region].Sakhalinsk: Sakha State University, 2010, pp. 171-179.

Gladyshev S. A., Olsen D., Tabarev A.V., Kuzmin Ya. V. Chronology and periodization of Upper Paleolithic monuments of Mongolia / / Archeology, Ethnography and Anthropology of Eurasia. - 2010. - N 3. - p. 35-42.

Gladyshev S. A., Popov A. N., Tabarev A.V. Novye verkhnepaleoliticheskie pamyatniki Severnoi Mongol'ii (po razvedkam 2010 g.) [New Upper Paleolithic monuments of Northern Mongolia (based on exploration in 2010)]. Problemy arkheologii, etnografii, antropologii Sibiri i sopredel'nykh territorii: mat-ly of the final session of the Institute of Archeology and Ethnography of the Siberian Branch of the Russian Academy of Sciences, Novosibirsk: Izd., 2010. - Vol. XVI. - pp. 41-46.

Derevyanko, A. P., Zenin, A. N., Rybin, E. P., Gladyshev, S. A., Tsybankov, A. A., Olsen, D., Tseveendorzh, D., and Gunchinsuren, B., Technology of stone splitting at the early Upper Paleolithic stage of Northern Mongolia (Tolbor-4 site), in Archeology, Ethnography, and Anthropology of Eurasia. - 2007. - N 1. - p. 16-38.

Kononenko P. A. Ustinovka-3 site and the problem of transition from Paleolithic to Neolithic in Primorye / / Late Paleolithic - Early Neolithic of East Asia and North America: Materials of the International Conference-Vladivostok: Dalpress, 1996. - pp. 117-136.

Kononenko P. A. Ecology and dynamics of archaeological cultures in the valley of the Mirror River at the end of the Pleistocene-beginning of the Holocene (Ustinovsky complex, Russian Far East) / / Archeology, Ethnography and Anthropology of Eurasia. - 2001. - N 1. - p. 40-59.

Hunter-gatherers of the Sea of Japan basin at the turn of the Pleistocene-Holocene. Novosibirsk: IAET SB RAS Publ., 2003, 176 p. (in Russian)

Tabarev A.V. Paleoindean treasures-caches of North America / / Humanities in Siberia. - 1999. - N3. - p. 84-87.

Tabarev A.V. The case of hidden tips (hoards-caches of stone products on the territory of North America) / / Stratum plus. - 2005 - 2009. - N 1: Middle Paleolithic: In Search for Dynamics / Middle Paleolithic: In Search for Dynamics. - pp. 300-333.

Tabarev A.V. Hoards of stone products in the complexes of the original Dzemon, the Japanese Archipelago. - Vladivostok: Publishing House of the Far East feder. un-ta,2011, pp. 18-47.

Frison G., Bradley B. The Fenn Cache, Clovis Weapon and Tools. - Santa Fe: One Horse Land & Cattle Co., 1999. - 111 p.

Gladyshev S., Popov A., Tabarev A., Olsen J.W., Gunchinsuren B. First Known Paleolithic Cache in Mongolia // Current Research in the Pleistocene. - 2011. - Vol. 28. - P. 58 - 60.

Gladyshev S. A., Tabarev A. V. New Data on the Early Upper Paleolithic of Northern Mongolia // Current Research in the Pleistocene. - 2009. - Vol. 26. - P. 17 - 18.

Gladyshev S.A., Tabarev A.V., Olsen J.W. Origin and Evolution of the Late Paleolithic Microindustry in Northern Mongolia // Current Research in the Pleistocene. - 2010. - Vol. 27. - P. 38-40.

Hayden B. Practical and Prestige Technologies: The Evolution of Material Systems // J. of Archaeological Method and Theory. - 1998. - Vol. 5, N1. - P. 1 - 55.

Kanomata Y. Technique and Chronology of the Upper Palaeolithic Industries in the Tohoku District // Palaeolithic Archaeology. - 2006. - N 68. - P. 1 - 15 (in Japanese with English Summary).

Kanomata Y. Functional Analysis of Stone Tools Put into the Cache-pit at Nogawa Site from the Final Stage of the Pleistocene // J. of the Japanese Archaeological Association. - 2010. - N 30. - P. 47 - 64 (in Japanese with English Summary).

Kelly Ch.T. The Earliest Cultures in Japan // Monumenta Nipponica. - 1972. - Vol. 27, N 2. - P. 143 - 147.

Kilby D.J. An Investigation of Clovis Caches: Content, Function, and Technological Organization: PhD Dissertation / The University of New Mexico. - Albuquerque, 2008. - 266 p.

Kornfeld M., Tabarev A. The French Connection? Or is It? // Current Research in the Pleistocene. - 2009. - Vol. 26. - P. 90 - 92.

Pelegrin J. Long Blade Technology in the Old World: An Experimental Approach and Some Archaeological Results // Skilled Production and Social Reproduction: Aspects of Traditional Stone-Tool Technologies. - Uppsala: Uppsala University, 2006. - P. 37 - 68.

Serizawa C., Kajiwara H., Akoshima K. Experimental Study of Microwear Traces and Its Potentiality // Archaeology and Natural Sciences. - 1982. - Vol. 14. - P. 67 - 87 (in Japanese with English Summary).

The Nogawa Site at Miyagino Ward, Sendai City // The Board of Education of Sendai City. - Sendai: Sendai City Cultural Properties Research Reports, 1996. - N205. - P. 1 - 145 (in Japanese with English Summary).

The article was submitted to the Editorial Board on 21.12.12, and the final version was published on 24.12.12.

Permanent link to this publication: