Перейти к содержимому. | Перейти к навигации

Персональные инструменты
Вы здесь: Главная Издания Археологические вести Annotations of issues Археологические вести. Спб, 1998. Вып. 5. Аннотации.

Археологические вести. Спб, 1998. Вып. 5. Аннотации.




В.М. Массон. Археология и культурное наследие страны

V. M. Masson. Archaeology and the nation's cultural

Diversity and richness of the cultural legacy is the basic feature of a civilized society, an integrating factor in ethnic and national identity. Broadly speaking, cultural legacy includes such phenomena as mentality, which underlies moral norms and behavioural stereotypes, folklore systems, and a host of material elements. Artifacts are the people's materialized memory. An important part of culture's material part is archaeological heritage, which includes all sorts of archaeological objects: ruins of castles and fortresses, burial mounds, remains of ancient camp sites and cities, imposing monumental structures like dolmens or the British Stonehenge. These monuments, studied by professionals, concentarte a wealth of information about the ancient times and peoples. Sites which have not yet been studied constitute mankind's stock of invaluable information, and this is something that the ignorant modern technocracy fails to understand. Apart from the masterpieces of ancient art brought to light by excavations, the archaeological heritage includes mass artifacts which are largely impersonal and characterize the everyday life of simple people. Their origin and evolution are being studied by archaeology.

The notion of archaeological heritage originated in the 20th century along with the entire programme of respective activities known as archaeological heritage management. The information content of the archaeological heritage is enormous. Ecclesiastical archaeology, which was a prominent subdiscipline before the Revolution, is an example. This branch of archaeology focuses on material aspects of the Orthodox culture and is an important tool in reconstructing the historical past of the Orthodox nations. It helps reveal both the Christian specifics and more general historic processes. Thus, it has documented a considerable Byzantine influence on the stylistics of early Christian architecture in Scandinavia before Roman Catholic standards had been established there. The Saint-Petersburg Institute of the History of Material Culture (IIMK) has taken considerable efforts in reviving this important area of archaeological research (see Ecclesiastical Archaeology 1995).

The significance of cultural legacy for modern nations is tremendous, especially in the era of scientific revolution, when the universal distribution of electronic information systems is accompanied by the advent of new mass stereotypes resulting in a sort of global standardization and gradual decline of human individuality, both at the personal and at the national level. Century-old cultural traditions, behavioural and moral, are being lost. Ultimately, Durenmatt's Homo faber with his one-sided technical development is hardly more advanced than a trained ape. There is a grain of truth in the journalistic image of the future man as a Neandertal living in a comfortable cave supplied with modern audio and video equipment. Modern society must not shun progressive cultural traditions of the past, such as those developed in the Orient. However, while the Muslim tradition does contain a number of positive elements, it would evidently be futile to try and revive the ossified version of 7th century Islam.

Civilized modern nations are showing more and more readiness to preserve and use their cultural legacy, as evidenced by policies adopted with regard to archaeological monuments. The rise of prosperity in developed countries stimulated the growth of mass tourism and the emergence of tourist industry which has become an economic means of attracting public attention to the most representative material forms of past cultures. Respective legislation is becoming more and more efficient. In Europe, the first legislative act aimed at protecting archaeological monuments was the 1666 edict issued by the king of Sweden and declaring all antiquities to be the property of the crown. Since then, the protection of antiquities has become one of the major concerns of the legislators, as evidenced, for instance, by a thesis by E. Firt of Southampton University addressing Western European legislation on submerged sites. Several measures have been taken at the international level, primarily under UNESCO. In 1959, the International Committee on Monuments and Sites (ICOMOS) was organized, whose programme was outlined in the Venetian Chart. This document stringently regulates all restoration activities with a view to prevent the transformation of ancient ruins into quasi-restored eclectic modern constructions. In 1985, the International Committee on Archaeological Heritage was formed under ICOMOS (Price 1989). Heritage management was discussed at several international conferences, including the 1986 World Archaeological Congress in Southampton (Cleer 1989).

Basic methodological principles underlying practical activities are being developed. A concise analysis of the current situation was presented by Kristiansen in his paper for the Southampton congress (Kristiansen 1989). In his view, the heritage management can develop both ways: from political ideology to a scholarly stance and the other way round. Indeed, developing a sound management policy in the modern uneasy world is much like walking along the razor edge. Kristiansen notes that ethnic attribution of prehistoric cultures is in this way or other linked with ideological attitudes, either implicit or overtly political. As a result, heritage management activities become part of the political system (Kristiansen 1989: 23-24). We are well familiar with this situation since the transition from the U.S.S.R. to the C.I.S. was accompanied by numerous attempts to use the cultural legacy and the archaeological evidence in particular for stirring up inter-ethnic unrest often resulting in bloodshed. Simplistic determinist doctrines based on the notion of historical evolution as a succession of "formations" abounded in peremptory statements to the effect that such and such territory had "always" belonged to such and such people (whereas the actual pioneers were Neandertals or even their ancestors). Usually, such extreme views are popular either in developing nations or in countries with persistent totalitarian tendencies.

An educated approach to cultural legacy is a characteristic feature of a civilized society. This applies both to the legislation and to the moral atmosphere. A rather illustrative example is Denmark where a series of acts dealing with heritage management was passed, including the most important one, that of 1969. Under this act, 400 to 500 protective excavations, varying in scale but universally aimed at the preservation of nation's cultural legacy, are being carried out each year. Public moral standards with regard to these matters are amazingly high, largely due to a large-scale programme of public education in archaeology. Archaeological exhibitions function at 48 Danish museums, public lectures are being regularly given in a number of places, and a special archaeological magazine for secondary schools is being published. As a result, antiquities have become an object of national pride, and the necessity of their preservation is an idea firmly rooted in public mentality. Numerous societies of amateur archaeologists exist. Thefts of museum exhibits are exceptional. A large part of the exhibition "Twenty-Five Years of Archaeology in Denmark" (Vejle, 1993) consists of stands with artifacts brought by the amateurs. These include massive golden ornaments of the Roman period brought by primary school children. Doubtless, such facts are suggestive of a considerable progress achieved by the nation on its way toward a civilized 21st-century society.

With regard to the archaeological heritage management, Russia, like other members of C.I.S., has been overtaken by highly dangerous tendencies. Well-equipped gangs of looters empty the ancient burials and attempts are being made at setting up a black market in antiquities. A situation in Crimea, where these crimes are being committed quite openly, may provide an illustration. Unauthorized excavations have become even more frequent and are often no less dangerous. Legislative activities, which were started under the Soviet regime in an attempt to catch up with international standards have virtually come to a standstill.

Certain positive tendencies are also evident, though. Since the times of Perestroika, heritage management has been been attracting more and more attention (Masson 1989). Importantly, various nongovernmental structures have emerged which focus on the protection and study of endangered antiquities of high cultural value. Successfully functioning historical and cultural societies like "Povolzhye" in Samara and a similar association in Ekaterinburg draw on the existing legislative base and on certain executive practices of munucipal authorities. Whenever the governmental bodies make proper use of the existing system of tax privileges, private entrepreneurs sponsor protective and research activities. Shatalov Foundation for the Advancement of the Humanities and Education in Udmurtia is an example. A methodology of heritage management is being developed, as evidenced by the 1993 international conference in Tyumen jointly sponsored by IIMK, the World Archaeological Congress association, and the Tyumen Regional Museum (Problems in Cultural Origins 1993). Clearly, under the new decentralized infrastructure of Russian science and culture, state and public organizations must join efforts in solving these problems. Three groups of immediate tasks can be seen. First, a consistent legislative activity is required. Because the current parliamentary system proved rather inefficient at the state level, it would appear more promising to focus on the municipal level. Second, as noted by all the researchers, a universal registration of antiquities is on the agenda (Trotzig 1989). There were numerous attempts, both before and after the Revolution, to get this work started, yet so far the progress has been rather jerky, periods of hectic and largely spontaneous activities being followed by stagnation. Now as before, there is little coordination between scholarly and financial efforts taken by local and central structures. The third important group of tasks includes educatory efforts aimed at creating a public atmosphere favourable to heritage management. Now that millions of people are chasing profit under the motto "hard cash above all and right away", this is the most difficult problem. However, without even trying to solve it, Russia can hardly expect to ever acquire the status of a civilized nation.





В.П. Любин. Проблемы первоначального заселения человеком Кавказа и Евразии

V. P. Lyubin. On the initial peopling of Caucasus and Eurasia

The prehistory of the Caucasian isthmus is tightly linked with that of the adjacent regions of the Near East. In the Sarmatian, after Caucasus had become a peninsula of Western Asia, a rapid influx of savanna and steppe hypparionic fauna occurred, which included hominoids referred to as Udabnopithecines. Due to environmental changes in the early Pleistocene, the hypparions and the damans gradually became extinct and had been replaced by true elephants. Lower Pleistocene occurences with faunal remains are known in several places in Georgia, Armenia, and Northern Caucasus. One of them, Dmanisi, was a camp site of early Homo erectus.

Two facts concerning geographical position of Dmanisi and other later Lower and Middle Pleistocene Caucasian sites are worthy of note. Not a single site of that period is known in Colchis, Western Transcaucasia, and all these sites are situated within the middle part of the isthmus. The former fact may be due to the fact that from the beginning of Pliocene the Colchis became distinct as a stable refuge of relic florae, the moist tropical forests. Most of Central and Eastern Transcaucasia, on the contrary, was a half-open area of the savanna type (Chochiyeva, Mamatsashvili 1991) suitable for hypparionic and later faunae as well as for primates including humans. Indeed, from that the middle part of the isthmus is a submeridional zone of the Main Transcaucasian Transverse Elevation regarded by E. E. Milanovsky (1976) as the northern segment of the Afro-Arabian rift zone were the natural environments were favourable for hominid origins and evolution. The Caucasian segment includes the Nemrut-Ararat fault and cleft zone, the

Transcaucasian volcanic plateau, the Dziruli massif (Suram range), the Elbrus, Chegem and Kazbek-Tskhinvali zones of the Great Caucasus, the Pyatigorsk Laccoliths and the Stavropol' eminence (fig. 2).

The Transcaucasian plateau is of especial interest. Being a northern projection of Near Eastern plateaus and an extension of adjacent mountainous ecosystems, it was the first area to be inhabited by early Homo. It was here that the first humans came, together with, or after, the Near Eastern open-land fauna, and Dmanisi was one of their outposts (fig. 2).

The discovery of Dmanisi is the major sensation not just for Caucasus, but for Eurasia in general. Its numerous unique features include a very early date (1.8±0.1 m BP), large surface area (about 5,000 sq. m), excellent preservation of remains, subhorizontal position of strata, lime crust above the principal archaeological deposits (fig. 3), abundant faunal remains (Arch, meridionalis, Dicerorhinus etruscus, Equus, Struthio, etc.), archaic pebble industry (fig. 4), and a mandible of early Homo erectus (fig. 5) (Dzaparidze, Bosinski et al. 1989).

According to the Georgian paleontologist L. K. Gabunia, the Dmanisi mandible is similar to those of some Early Pleistocene hominids of Africa and Asia (Java). The African parallels include Olduvai OH 13 {Homo habilis), the earliest Homo erectus from Koobi Fora, Kenia (ER 730, ER 992, and possibly ER 15000); and, among the Javanese ones, "Pithecanthropus dubius" from Sangiran (Sangiran 9). "It is possible that the Dmanisi hominid belonged to a branch of early hominids which occupies the same position in Eurasia as that occupied by the forms intermediate between Homo habilis and Homo erectus in Africa or by the earliest Sangiran hominids in Java" (Gabunia, Vekua 1993).

Having originated in Africa 1.7-1.6 m years ago (Howell 1986), Homo erectus displayed exceptional dynamics by adapting to various environments and radically widening the human ecological niche. In Western Europe only late forms of H. erectus are known, dating from the Bruhnes period (Middle and Upper Pleistocene, or less than 600±100,000 years BP (Robroeks, Van Kolfschoten 1993, 1994). In China and Java, the chronological range of hominid remains is much broader, not only the Bruhnes but also the Matuyama period (Lower Pleistocene) being represented by Yuanmou, Gong-walin, Choukoutien, Longgupo and other sites in China, and Sangiran, Mojokerto and other localities in Java. The origins and absolute age of the Javan hominids, as well as their association with the artifacts, are widely debated. Recently, US and Indonesian specialists who used volcanic rock directly associated with remains of "Meganthropus" and the Mojokerto hominids have obtained absolute dates of 1.66±0.04 m and 1.81±0.07 m BP, which is close to the age of the earliest Homo (H. ergaster) from Koobi Fora. New datings make it possible that H. erectus originated outside Africa. His ancestor might have migrated from Africa before 1. 8 m years ago, that is, before the emergence there of the Acheulian with bifaces and cleavers. This may explain the sharp difference between the African and Asian cultures of H. erectus: lack of bifaces or cleavers in most Asian sites (Swisher et al. 1994).

Given all these data, Dmanisi may be regarded as the first evidence of the relationship between the African and Asian hominids. Its intermediate geographical position between Koobi Fora and Sangiran, and the close morphological similarity of the hominid remains found there probably attests to the routes of hominid migration from Africa into Asia. Humans might have migrated either along the sea coast (Singer and Wymer 1978) or by an inland route, from Dmanisi, rounding the Caspian sea, towards the Central Asiatic plains lying between the mountain systems, and further into China and Java (fig. 6).

Unlike the first migration, the distribution of the later forms of H. erectus in Asia, those associated with the typical Acheulian with bifaces and cleavers, was not a Transasiatic event. Sites with these tools are found in the Levant, South Arabia, Caucasus and, to a lesser extent, in territories situated to the east (Turkmenia, Kazakhstan, Kirghizia, and India).

The early colonization of Eurasia by humans associated with the earliest Acheulian industries is apparently evidenced by sites like Ubeydia, Jub-Jannin, and Latamna in the Levant. The earliest industry, that of Ubeydia, resembles Developed 0ldowan II of Eastern Africa (Bar-Yosef, Goren-Inbar 1993) and dates from the end of the Early Pleistocene, 1.4 m BP (Tchernov 1988). All the three assemblages represent the so-called Ubeydia-Latamna tradition with its characteristic pics, trihedrons, polyhedrons, spheroids, cleavers, bifaces with shovel-shaped distal ends, etc. (Stekelis 1966; Clark 1967, 1968; Hours 1975).

Pics, thrihedrons, polyhedrons, and spheroids are absent in the Caucasian Acheulian, either because people representing this tradition migrated here after the most archaic forms had been lost, or because they represented other facies of the Levantine Acheulian characterized mainly by bifaces of the Ras- Beirut type, or by bifaces and cleavers of the Jisr-Banat-Yakub type, or by the lack of any bifaces at all (see Hours 1985). The first territory to be populated by the late erectuses in this region was the southern fringes of the Caucasus (the Transcaucasian Plateau). It is only there that industries with numerous bifaces and cleavers are known (Satani-Dar, Dzhraber, Atis, Persati, Chikiani, Azykh, etc.; figs. 9, 10, 12, 14). Elsewhere in the Caucasus, the Acheulian bifaces are much less common. A route from Southern Georgia to Southern Ossetia appears to be the most likely one for the Acheulians who were moving in the direction of the Great Caucasian Range. Several sites are situated along this route, including those in the foothills, highlands and the Alpine zone of Southern Ossetia (Lashe-Balta, Kaleti, Tigva, Goristavi, and caves Kudaro I and III, and Tsona; figs. 8: 1; 9: 1, 2; 11: 1; 13; 15; 16: 2, 3). In the Northern Caucasus, Acheulian sites are known only in the Kuban basin (Treugol'naya Cave, Ignatenkov Kutok, Abadzekhskaya, and others; fig. 7: 1; 11: 2).

According to biochronology and absolute geochronology, the dispersal of the Acheulians in the Caucasus began during the second half of the Cromerian. The faunal remains from the Acheulian levels of Caucasian caves contain bones of animals belonging to the late stages of the Tiraspol and Singil associations. An important chrono-stratigraphical marker is provided by the discovery in level 7a of Treugol'naya of a microtine rodent close to Mimomys or intermediate between it and Arvicola. The ESR date of this level is 583±25,000 BP. The thermoluminescent date of the basal horizon of another cave, Kudaro III, is 560±112,000 BP.

Caucasus, and especially Transcaucasia, is the region where Acheulian proper, that is, Acheulian with bifaces, is distributed. In most industries, choppers and cleavers are represented as well (figs. 8; 9; 10: 2; 16). Bifaces of classical types (having regular linear outlines, biconvex in section, and processed with continuous bifacial retouch around their entire perimeter) are relatively rare (fig. 11: 2, 3; 12). The distinctive feature of some of them is their subrectangular outline (fig. 15). Some pieces (bifaces with "shoulders", i.e. with distal points, fashioned by notches) should be classified as "diverse" (fig. 14: 1-5, 8). In some industries, bifaces of nonclassical types predominate, including partial (fig. 14), and backed (fig. 13).

Because the Caucasian Acheulian has not been sufficiently published, it is difficult to compare it with its Levantine counterpart. However, Korobkov (1989) believes that some bifaces, cleavers, and choppers found in one of the sites on Mount Yashtukh, Abkhasia, resemble tools of the Ubeydia-Latamna tradition (fig. 7: 2; 8: 2). It is notable that like in the Caucasus the cleavers are also rare in the Levantine sites, except Jisr-Banat-Yakub, Israel (Gilead 1972). Another similarity is that rare elongated bifaces of African-Levantine types have been found in the Caucasus (fig. 11).

Between 600,000 and 500,000 BP, according to Robroeks and van Kolfschoten (1994), Homo erectus populated most of Western Europe. The same may be true of Causasus and possibly India. This may have been one of the largest migration waves in the population history of late Homo erectus.

As mentioned above, the Acheulian with bifaces is the main industrial tradition of the Caucasian Lower Palaeolithic. So far, however, it is still vague whether the Caucasian region and Levant belonged to the same province of biface industries (hypothesis of "iceberg top": the Caucasus is the northern extremity of this province) or the Caucasus is the detached "biface region" (hypothesis of "enclave" similar to that of Madras).

If neither Eastern Anatolia nor the areas between Levant and India yield any material linking all these areas, another hypothesis may be proposed: that of "transit zones" where the earliest nomads did not stay for a long time and left no traces of their presence.


М. В. Аникович, Б.А. Бредли, Е.Ю. Гиря. Технологический анализ стрелецких треугольных наконечников

M. V. Anikovich, B. A. Bradly, E. Yu. Girya. Early Upper Palaeolithic in the Russian Plain: Streletskayan flaked stone artefacts and technology

The artefact assemblages from early Upper Palaeolithic site in eastern European Russia contain flint tools of a more Middle Palaeolithic type. With these artefacts are bifacially thinned triangular forms that may represent the first use of this technology in the area, and perhaps anywhere in Europe. Early Upper Palaeolithic sites in Eastern

Europe show clear cultural variations that allow us to identify a series of archaeological cultures. The Kostenki-Streletskaya (Streletskayan) is of particular interest. It was originally distinguished by A. N. Rogachev (1957) on the materials from the Kostenki-Borshchevo region. It was not long before this material became well known beyond Russia (e.g. Klein 1969, Kozlowski J., Kozlowski S. 1975; McBurney 1976). In the 1970-s — 80-s one of Rogachev's students, M. V. Anikovich, continued to investigate this archaeological culture using new methods and techniques (Anikovich 1977, 1992; Rogachev & Anikovich 1984). The following is a summary of what the authors currently know about the Kostenki-Streletskaya archaeological (culture with a focus on bifacial technology of the flaked stone assemblages.

Typologically and chronologically, the Streletskayan derives directly from Middle Palaeolithic archaeological cultures in eastern European Russia and over its approximately 10 000 years of development, slowly takes on characteristics of Upper Palaeolithic assemblages. Of particular interest is the presence of a sophisticated biface thinning technology that persisted throughout the Streletskayan. Although it is probably derived from an earlier Middle Palaeolithic biface technology, the development of thinning methods clearly sets it apart. The use of carefully prepared platforms (including grinding and isolation) along with the possible pressure flaking and intentional heat treating of raw materials (observed only at Kostenki 1, Layer V) in the Early Upper Palaeolithic of Eastern Europe, indicates that this specialized technology was developed well before the much better known Solutrean technology of southwestern Europe. This may shed some light on the origin of Solutrean biface technologies.


В.В. Питулько, А.К. Каспаров. Древние охотники высокоширотной Арктики материальная культура и стратегия жизнеобеспечения

V. V. Pitul'ko, A. K. Kasparov. Ancient Arctic hunters: material culture and survival strategy

The article continues the discussion of the unique results obtained in the Siberian High Arctic during the excavations of the site located at Zhokhov island under 760 N. Data concerning the human occupation of the area in the Early Holocene were obtained over two field seasons (1989 and 1990). The series of radiocarbon dates from three laboratories show the age mid-value to be about 7.800 BP. A large amount of various artifacts were collected. Excavated materials show a developed technological level of the ancient aboriginal culture, which has distinct Mesolithic features. However, the combination of a microblade prismatic core industry and polished axes is unusual for an east Siberian culture of that period. Hunting equipment is represented by fine artifacts made from antler, bone and fossil mammoth ivory. One fishing spear, spear points, and arrow and dart points were excavated; most had flint-blade inserts (fig. 1, 2). Thanks to the preservative properties of permafrost, the wooden artifacts (fig. 1: 811) and home utensils (fig. 3, 4: 2, 3) were in exceptionally good condition. One special artifact must be mentioned: there was a large fragment of sledge-runner having rather progressive constructive elements. Both the postcranial and cranial skeleton remains identified as a dog, were also found and, probably, the dog traction was known as far back as 7.800 BP. Thus the adaptive capabilities of these ancient aborigines appear to be greater than usually thought. Inasmuch as the archaeological data of the site were published and some aspects of the culture were previously discussed in details, the focus of this article is on the faunal remains extracted from the cultural stratum during the excavations, which give important information for the reconstructions of the aboriginal survival strategy, seasonality, etc. The main food resource appears to be reindeer. However, the quantity of the polar bear bones is almost as high, while other species are in much lower numbers (table 1). The bones, both the polar bear and the reindeer, are heavily fragmented as a result of marrow splitting, cooking, etc. The Zhokhov site is the only ancient settlement where the bone remains of polar bear amount to 43.8% of the total, indicating a unique hunting specialization of the aborigines. Polar bears have rarely been mentioned as food source in ancient far northern sites, although they were sporadically hunted and eaten by the Arctic peoples. But usually the aborigines, both prehistoric and historic, subsisted mainly on reindeers or sea-mammals. Zhokhov island was no exception. The bone frequencies and specific fracturing of the bones indicate that it was a living site. We assume that the site was visited sporadically. The examination of bone remains allows to recognize some procedures of butchering. The data do not make it possible to determine the total duration of habitation of the site. Evidently, it is no more than precision of the radicarbon method of dating.


В.И. Тимофеев, Г.И. Зайцева, Г. Посснерт. Радиоуглеродная хронология цедмарской неолитической культуры в Юго-Восточной Прибалтике

V. I. Timofeev, G. l. Zaitzeva, G. Possnert. A radiocarbon chronology of Zedmar Neolithic culture, South-Eastern Baltic Area

A summary of radiocarbon dates for Zedmar neolithic culture, whose sites are situated in the inland part of southeastern Baltic area, Kaliningrad Region, Russia, and northeastern Poland, is presented. So far, four peat-bog sites have been excavated: Zedmar A, Zedmar D, and Utinoye Boloto I in the Kaliningrad Region (V. I. Timofeev), and Dudka I in Suwalki voivodship, Poland (W. Guminsky and J. Fedorchuk). Important new findings have been reported for Zedmar D, where the 1988 excavation area has revealed two stages of settlement, the later one being associated with the remains of wooden constructions (possibly platforms) discovered in the upper part of the layer whose basal part contained finds attributed to the Zedmar Culture. Remains of posts or piles were found which were likely details of a platform. The wood from the upper part of the constructions and the piles has been dated at 4,300-3,800 BP. The dates for the main. Neolithic, level are much earlier. Importantly, "accelerated" dates for carbonized food remains from the inner part of Zedmar type potsherds are in the range of 5,400-4,900 BP (see fig. 1 for dates related to two principal technological groups of ceramics, one tempered with crushed shells and organic matter, another with crushed stone). The two stages in the life of the site are separated by a break of at least 500 years (fig. 2). Also, a series of dates is available for the stratified site of Zedmar A, where three layers were unearthed, ranging from 5,500 BP to 4,900 BP. Similar dates have been obtained for Utinoye Ozero I and Dudka I sites. All known Zedmar sites, then, fall in a narrow interval of 5,600/5,500 BP to 4,800 BP. A comparison with the adjacent Neolithic centres of the Polish Plain, which were shown to have influenced the Zedmar sites, has revealed that the latter may be synchronous with the Brzesc-Kujawian group of the Late Lengyel Culture and the earliest stages of the Funnel Beaker Culture (fig. 3).


И. ван дер Плихт. Калибровка радиоуглероднои временной шкалы

J. van der Plicht. Calibration of the radiocarbon time scale

Introduction. The isotope 14C (radiocarbon) is continuously produced in the earth's atmosphere by cosmic radiation. Radiocarbon is radioactive and decays with a half life of 5730 years. This is shown in fig. 1, where the 14C concentration (relative to the original concentration) is plotted as a function of time. A stationary state of production, distribution between the main carbon reservoirs (atmosphere, ocean and biosphere) results in a constant 14C concentration in atmospheric CO2. The CO2, including the isotope 14C is taken up by plants through photosynthesis, and next also by other organisms. Living plants, animals and people therefore have in general the same 14C content as the atmosphere they live in. The carbon exchange with the environment ceases after death, whereupon the 14C concentration diminishes in time due to radioactive decay. Thus, the age of carbon-containing remains of organic matter (trees, peat, bone, charcoal, etc.) can be determined by measuring the residual amount of 14C left in the sample. This is the basic principle of the radiocarbon dating method.

The method as described above is based on the following assumptions, as will be obvious from fig. 1: 1. the half life should be accurately known; 2. the atmospheric 14C concentration at time t = 0 has a constant value throughout prehistoric time. One of the complications in the radiocarbon dating method is that these two basic assumptions are not completely true. This means that although the radiocarbon concentration can be measured accurately, the radiocarbon years will be related in a complicated way to historical ages. Nevertheless the radiocarbon dating method can be made valid by the process of calibration, as will be described below.

First, the half life is not very accurately known: 5730±40 years (Godwin 1962). Furthermore the half life used for age calculations is still the original Libby value of 5568 years (Libby 1955). The reason for using the wrong half life as convention is that all radiocarbon dates calculated this way have the same meaning. They are reported in BP (Before Present) where Present is defined as 1950 AD. The reason for this choice is that the 14C radioactivity standard corresponds to the value for that year.

Second, it is now known that the 14C concentration of atmospheric CO2 has not always been the same in the past. In tree rings, natural variations of the atmospheric 14C02 abundance were discovered on a time scale of one decade to a few centuries (de Vries 1958). Later it was discovered that these variations can be attributed to variations in solar activity (Stuiver, 1965), which in turn influence the production of 14C in the atmosphere. Also changes of the geomagnetic field strength influences the production of 14C in the atmosphere (Bucha 1970). This is understandable because both solar activity and geomagnetic field strength determine the amount of cosmic radiation impinging on the earth. In addition, the atmospheric 14C02 concentration also depends on exchange between the atmosphere and ocean. This exchange was certainly different from today during glacial times.

These violations of the basic assumptions for the radiocarbon calibration method cause radiocarbon years (in BP) to be different from calendar years (BC/AD). This means that there is no simple relationship such as AD = 1950-BP. Therefore the 14C method has to be calibrated: the relation between BP and BC/AD has to be established. This can be done by measuring 14C ages of samples which are also dated by other independent means.

Calibration of the Holocene. The ideal samples for calibration are tree rings because they can be dated absolutely by means of dendrochronology. Tree-ring chronologies are now available back into the Preboreal; a 14C calibration curve has been constructed from 9440 BC to the present.

A plot showing all the tree-ring calibration points and - for illustrative purposes - the so-called Libby line (AD = 1950-BP) is shown in fig. 2. The most prominent feature is the large deviation of the tree-ring data from the Libby line throughout most of the Holocene. This long-term trend can be explained by changes in the geomagnetic field (Bucha 1970). Furthermore, medium-term variations, the so-called wiggles (Suess 1970) can be recognized which are due to solar fluctuations. A clear example is shown in fig. 3 (de Jong et al. 1989).

All calibration data are published in 2 special volumes of the journal Radiocarbon (Stuiver, Kra 1986; Stuiver, Long, Kra 1993). There is a recommended part, 1950 AD - 2500 BC measured in 20 tree-ring year intervals (Stuiver, Pearson 1986; Pearson, Stuiver 1986).

Apparently, the concentration of 14C in the atmosphere was 10. 000 years ago such that radiocarbon years (BP) were about 1.000 years younger than calendar years (BC) - see fig. 2. Calibration means finding the correct calendar date corresponding to the measured BP-age through a curve such as shown in fig. 2. This calibration procedure takes into account the varying 14C concentration in the atmosphere (and terrestrial samples) throughout time. Also the dependency on a precise half-life value is then taken into account automatically.

It is the wiggly shape of the calibration data that makes calibrating complicated. For instance, wiggles can cause a 14C date to correspond to more than one calendar date. The radiocarbon age of, for instance, 4725 BP corresponds to calibrated ages of about 3625, 3500 or 3400 BC (see fig. 3). Calibration becomes non-straightforward if we also take into account the measurement error. Recently programs working on Personal Computers have been distributed for calibration purposes. We mention here only the Groningen program (van der Plicht 1993), written for IBM-compatible PC's.

An example of a calibration performed with the Groningen program is shown in fig. 4. The 14C result to be calibrated is 2300±40 BP. Two plots are shown in fig. 4. In the top part, the relevant part of the calibration curve (spline function) with the calibration data points from Stuiver and Pearson (1986), the recommended curve, are shown: along the y-axis, the Gaussian probability distribution of the 14C age is shown; along the x-axis, the calibrated probability function calculated. The latter probability distribution is also shown in the bottom part of fig. 4, which can be used for interpretation in a quantitative way. The lines in this graph indicate the 1δ (68.3%) and 2δ (95.4%) confidence intervals. The intercepts indicated by the vertical lines are to be taken as calibrated ranges. In the example of fig. 4 this means that the calibrated age ranges are 402-366 and 276-264 cal BC (1δ) 410-352 and 312-208 cal BC (2δ) For mathematical principles of the method, we refer to Dehling and van der Plicht (1993).

Calibration of the Late Pleistocene. The present tree-ring calibration data extend to 9440 BC, as measured for Preboreal European pine (Kramer, Becker 1993). It is not likely that this chronology can be pushed much farther back in time since the climate becomes too cold. There are trees found on Tasmania with late Pleistocene ages (13.000 BP), but there is no dendrochronology so the absolute ages are not known (Barbetti et al. 1992). So, in order to calibrate the radiocarbon time scale beyond the dendro-dated tree-ring limit, other dating methods have to be applied. The 3 most important calibration records are 14C measurements on samples also dated (presumably absolute) by varve counting, uranium isotopes, and thermoluminescence (TL). Whether these other dating methods are absolute remains a question of debate.

Annually layered varves (laminated sediments in glacier melt lakes) can be dated by varve counting, and by 14C dating of terrestrial macrofossils (seeds, leafs, insects) deposited in the varved sediment. Absolute varve dating is clearly only possible if the record is truly continuous, i.e. there is no hiatus, or the hiatus is exactly known in time; 14C dating is only possible by AMS because of sample size. At present there are records from Sweden (Wohlfarth et al. 1995), Poland (Lake Gosciaz, Goslar et al. 1995), Switzerland (Lake Soppensee, Hajdas et al. 1993), and Germany (Lake Holzmaar, Hajdas et al. 1995). The longest record existing thus far is from Japan (Lake Suigetsu). The data published thus far go back 12.500 BC (Kitagawa et al. 1995). The varved records, measured for their radiocarbon content are shown in fig. 5. The records overlap with the tree-ring data, and extend this curve into the Boiling. The major trend is that the calibration data tend to curve back to the Libby line, i. e. the deviation between radiocarbon years (BP) and calendar years disappears.

This is in contradiction with calibration data obtained from corals, dated by both 14C and U isotopes (see fig. 6). Here the radiocarbon years remain younger than the calendar years to 20.000 years ago, corresponding to a larger 14C concentration in the atmosphere than the present. The coral data sets as shown in fig. 6 are based on two assumptions: 1) the U-isotope dates can be considered absolute, and 2) the 14C dates represent atmospheric values (Bard et al. 1993; Lawrence Edwards et al. 1993).

The second assumption needs some explanation. The samples are corals and are thus of marine origin, and not atmospheric. Due to upwelling of deep and old water, the surface ocean has an apparent age of about 400 radiocarbon years (the so-called reservoir effect). All marine samples therefore have an apparent age of 400 years. This reservoir correction is applied to the coral data as shown in fig. 6. It is not known if the reservoir effect during glacial times was also 400 years. Nevertheless the coral data are considered to be the best calibration record to date. The general tendency of the coral calibration data set is in agreement with atmospheric 14C production calculations, based on measurements of the geomagnetic field in the past, in particular the radiocarbon time scale (Mazaud et al. 1992).

The varve data set is terrestrial (unless water plants have been used) but here the problem is the absolute scale. Annual lamination is not guaranteed, especially in late Glacial times; missing varve years in theory can explain the deviation between the varve and the coral data sets. Indeed recently the Swedish varves have been shifted by 500 years towards the coral data set (Wohlfarth et al. 1996).

Discussion. Variations of the 14C concentration in atmospheric CO2 complicate the process of dating in terms of historical ages. The 14C time scale has to be calibrated versus another independent dating method, preferably an absolute one. In general we can distinguish the following time periods:

1. Modern times (20th century AD). The atmospheric, 14C02 is here dominated by anthropogenic effects: burning of fossil fuels, and nuclear explosions. The atmospheric 14C02 is obtained from tree rings and (since 1950) from direct atmospheric measurements.

2. (Pre)historical times during the Holocene (the last 10.000 years). For the major part of the Holocene, the 14C scale is calibrated by means of absolutely dated tree rings (dendrochronology). The 14C variations are understood in terms of geomagnetic field variations during this time, as well as fluctuations in solar activity (fig. 2-3). Radiocarbon dates (in BP) can be calibrated by means of the tree-ring calibration curve to obtain historical ages (BC/AD) but due to wiggles the result can be ambiguous: a 14C date can correspond to several possibilities in calibrated ages (fig. 4).

3. Late Glacial maximum to Late Glacial (20. 000-10. 000 years ago). Here calibration by means of dendrochronology is impossible. Calibration data sets are obtained for marine samples (corals) by U-isotope dating and - from 14.500 years ago - for varved sediments (fig. 5, 6). Both methods disagree beyond ca. 10.000 BC; each method has its own assumptions and (dis)advantage. For calibration purposes the curve obtained from the corals is generally used, since these data can be explained by geomagnetic variations in the past. Strictly speaking there is no curve such as for the Holocene trees; the time resolution is large, and wiggles in atmospheric 14C02 are not detectable in marine samples.

Large fluctuations in atmospheric 14C have been observed near the Pleistocene/Holocene boundary. These are likely related to changes in CO2 exchange between ocean and atmosphere during the glacial/interglacial transition. For calibration of archaeological samples, the data sets therefore are of limited values.

4. Late Pleistocene (more than 20.000 years ago). For the time period 20.000-50.000 years ago, very limited calibration information is available. There are some measurements comparing 14C dates with the TL-dating method. The general trend is that the 14C ages remain younger than the TL ages (considered absolute), to an amount of about 3000 years 35000 years ago. Also this is consistent with palaeomagnetic information. Because of the large errors involved in the measurements, the value for archaeological calibration is very limited.

In summary, the varying 14C activity in the atmosphere during the 14C time scale (the last 50.000 years) are the cause of a complicated relation between radiocarbon years (BP) and historical ages (BC/AD); 14C dates are thus not absolute. Nevertheless the 14C method has proven to be of great use for archaeological dating, because the radiocarbon dates and historical ages are connected through a calibration curve. This curve is known in detail back to 9440 BC. For earlier times, only the general trend of the calibration curve is known. The 14C dates are 2000-3000 years younger than absolute dates in the late Pleistocene: 20.000 years ago, the radiocarbon ages are only about 18.000 BP.


Ю.Е. Березкин Н.Ф. Соловьева. Парадные помещения Илгынлы-депе (предварительная типология)

Yu. E. Berezkin, N. F. Solovyova. The main rooms at Ilginly-depe (a tentative typology)

The Chalcolithic site of Ilginly-depe (fig. 1), 14 ha in surface area, is situated in southern Turkmenistan, 240 km southeast of Ashgabat. In the fourth millennium B.C., the alluvial loamy valley around Ilginly-depe provided a bed for the Meana-Chai river. Apparently after the river had changed its course, the inhabitants of the village migrated to the nearby site of Altyndepe.

Radiocarbon dates, even calibrated ones, underestimate the age of Ilginly-depe by 300-500 years, although they agree with local stratigraphy. The reason for that is not clear. The Geoksyur-type pottery, found in two uppermost construction levels of Ilginly-depe (fig. 16: 1, 2, 8) was also discovered in the lower level of Shahr-i-Sokhta, Seistan, together with Proto-Elamite artifacts (Tosi 1971: fig. 3; Amiet, Tosi 1978; Piperno 1986), suggesting that Ilginly-depe was abandoned in 3 100—2 900 B. C. Given that the total depth of the deposits is about 14 m, the site must have been founded in the late fifth or early fourth millennium B.C.

The IIMK Kara Kum Archaeological Expedition, headed by V. M. Masson, started excavating the site in 1985. In 1985-89, about a dozen excavation areas were laid down, three of which became the principal ones.

Excavation 3 (N. F. Solovyova), 15 m by 14 m, is located in the southwestern part of the hill where the depth of the deposits probably attains 9—10 m. By 1994, five construction levels had been excavated, the first of which (the upper one) has not been preserved in some parts of the area (fig. 2: 1, 2, 3, 4; 4). Data on level VI, excavated in 1995, has not been included in the present article.

Excavation 5 (about 30 m by 30 m) is situated on the most elevated area in the northern half of the hill. The works here were mainly supervised by Yu. E. Berezkin, who was also responsible for works in excavation 4. The deposits were studied down to the depth of 1.5 m. In the field reports, plans and sections, the constructive levels are numbered according to local stratigraphy. Judging by the ceramics, levels I IB, IIA, IB, and IA in excavation 5 correspond to levels IV (evidently its beginning), late III, early III, and late II, respectively, in excavation 3. This correlation, however, is tentative, and the precise chronological correspondence between separate households in excavation 5 itself is not always clear. No materials related to level I or late level II in excavation 3 have been found in excavation 5, but during this period the most elevated part of the settlement was abandoned.

Excavation 4, about 70 m by 20 m, extends from the highest part of the hill down the slope from northwest to southeast. The difference between the heights is more than 3.5 m, but the deposits have been excavated to the depth of no more than 1.2— 1.35 m. No integral stratigraphic pattern has emerged. The upper levels I and II (roughly corresponding to the respective levels in excavation 3) have been preserved mainly in the northwestern part of the excavation. Structures of level III have been cleared out in the southwestern part.

The remaining excavation areas are small, and works on them were done by various researchers over one or two field seasons. Excavation 1 is situated on the southeastern projection of Ilginly-depe separated from the rest of the hill by a shallow gully. It is roughly contemporaneous with level III in excavation 3 and is overlaid by later deposits which lack architectural remains. Excavation 7 is located 60 m northwest of excavation 3. Its architecture is apparently contemporaneous with that of level III in excavation 3, while the unique collective burial was made in level II or possibly I.

The architecture of Ilginly-depe attests to the existence of a standardized layout. Its main feature are complexes which are square in plan and usually adjoin (on one or two sides) other structures and/or the street wall. The entrance is situated in the centre of the wall (any wall probably except the northwestern one), according to the direction of the main winds. To the left of the entrance, inside the structures, there is either a projection of the wall or a "box". "Boxes" are blind chambers filled with pieces of mud bricks, potsherds, and refuse. Possibly the walls of the boxes did not reach the ceiling and were gradually heightened as these "dustbins" were being filled. The floors of the main rooms are painted black. Wide entrances connecting the main rooms with the courtyards had thresholds. The step bearing was usually on the right (when viewed from outside). Other rooms of the same household, if present, also had doors connecting them with the courtyard. The courtyard had a passage connecting it with the street. Large households had two courtyards, one main, another economic.

Square structures with inner projections and floors painted black were first excavated at the sites of the Geoksyur Oasis where they were associated with Yalangach and Geoksyur-type pottery. None of the more western sites in the Kopet-Dag plain provide any parallels to the architecture of Ilginly-depe.

In the present article, the complexes are denoted by a combination of the number of the main room with that of the excavation (59.4 means a complex including room 59 in excavation 4). If the room has a double numbering, it receives a smaller number (e.g. 3 instead of 3/4). Several small structures which possibly also belong to this category but have not been completely excavated or are poorly preserved, are not mentioned. Data on the size and general layout of the complexes is presented in table 1. Surface areas are rounded off to 0. 5 sq. m.. linear dimensions are in metres. Surface areas of structures which possibly (or probably) belong to the same household as does the main complex but have separate doors connecting them with the courtyard, are not included in the total estimate. "Right" and "left" refer to the view from outside.

(1) Number of complex; (2) approximate dimensions; (3) construction level; (4) surface area of the main room; (5) direction of entrance; (6) number of benches (clay or wooden, as in 26. 3) in the main room; (7) number of large vessels (khums) standing on the floor or fallen down or number of pits for them; (8) type of structure to the left of the entrance (box or projection of the wall); (9) type of hearth, such as the following: "altar" (round hearthaltar shaped like a shallow pit in the floor about 1 m in diameter in the left part of the main room nearest to the door, opposite the box, if present, coated with clay and having a smaller pit in the centre); "hearth" (pit situated in the same place but without the clay coating); and "stove" near the wall; (10) layout of the left corner nearest to the door: "box" (box reaching the left side wall); "chamber" (the space behind the box is separated from the rest of the room by a wall with an aperture, small threshold, etc).; "cubicle" (the floor behind the box merges with floor of the room; cases where neither the box nor the cubicle are present are denoted by a minus sign); (11) presence and number of narrow chambers flanking the back wall of the main room or the wall to the right of the entrance, as in 9. 5.

A few complexes are characterized by the following features:

(12) Wide aperture ("window") between the main room and the right narrow chamber (26. 3, 9. 5, 26. 5); (13) agglomeration of ritual objects on the floor behind the "window" (26. 3, 26. 5); (14) the floor of the right narrow chamber adjoining the main room from behind is situated much lower than that of the left chamber (26. 3, 15. 4, 59. 4, 26. 5); (15) the black-painted podium that existed at the late stage of the complex on the place of the right narrow chamber; open on the side facing the main room and painted black (26. 5, 38. 3); (16) adobe structure, square in plan, in the corner of the main room to the right of the entrance (9. 5, 18. 5, 58. 5); (17) painted and plastic decoration of a certain type on the wall of the box (14. 3, 26. 3, 38. 3, 26. 5, 58. 5, and probably 10. 5); (18) small stone female figurine or, possibly instead of it, elongated stone or mud brick dug into the floor of the main room (all these are always placed at some angle to the threshold (9. 5, 10. 5, 12. 4, 26. 5, 59. 4; see also 55. 5, described together with 18. 5); (19) mud brick structure in the left side chamber, situated along its long axis (38. 3, 26. 5, possible 15. 4 and 59. 4); (20) bowl, puttied into the floor or into the wall, in the corner where the box adjoins the wall (10. 5, 58. 5); (21) brazier bowl instead of the mere pit in the centre of the hearth-altar (9. 5, 30. 5); (22) agglomeration of potsherds from fine vessels in the top layer of the fill of the box; it is the thickest near the back side and tapers off toward the middle of the box. This feature is present in three adjacent rooms cleared out in excavation 5: 26. 5, 58. 5 (where the clay with which the potsherds were mixed, was charred), and 10. 5.

Complex 38. 3, level V (fig. 9—11), is the earliest among those excavated by the end of the 1994 season. Its surface area is 47. 0 sq. m., and the surface area of the entire household which, apart from main room 38, included two subsidiary rooms (37 and 39), the main courtyard (44), and the utility courtyard (42) is more than 120 sq. m. Near the northeastern wall of the main room, there was a long and low (height, 0.3 m) platform. At the early stage, there was a hearth in the centre of the room. After the repair, the hearth pit was covered with a large stone, three smaller ones were placed above, and all the four were puttied with clay. The resulting elevation provided a base for a round Structure, possibly an altar, the side surface of which was painted black and red. Subsidiary chamber 39 was situated along the northwestern wall of the complex. On its floor, an unused clay mould for casting an axe-adze was found. After the repair, the floor of this semi-basement was raised by 0. 5 m, and the new compartment was connected with the main room by means of a "window". Finally, the entire inner space together with the wall and the "window" was filled with layers of plaster mixed with black paint, and a podium, open toward the main room, resulted. An architectural detail in the form of a protome of a bull, had been thrown down on layers of plaster in front of the podium and was found there. It was made of mud bricks, coated with paste, and shaped like a stylized fore part of a bull's body. Small clay horns were broken off when the sculpture fell down. Its surface was painted black several times. On the upper layer of paint, eight coiling snakes were carved. Near the northern corner of the box, several copper artifacts had been left: fragment of a knife, point, and socketed axe-adze (Solovyova et al. 1994: figs. 1-4, 19). The surface of the covered main courtyard no. 44 (late stage) was dressed with clay several times and painted black. The walls were plastered, and a seat made of mud bricks and unbaked clay was constructed near the southwestern wall. In the middle of the courtyard, a base of a rectangular adobe construction which apparently supported the roof, has been preserved.

Complex 26. 3, level IV (fig. 6—8), was constructed at the place of complex 38.3 and is somewhat smaller (about 42 sq. m. ). It was the centre of a household which also included the covered main courtyard 27, two subsidiary rooms 28 and 29, two rooms, 34 and 35, likely destined for habitation, and utility courtyard 36. Inside the main room, charred remains of two benches covered with red ocher were found. Eight painted large vessels (khums) stood in special pits. They were arranged in two rows parallel to the long bench along the northeastern and southwestern walls. A wide aperture connected the room with the side chamber. Part of the wall below the aperture was painted black all over. A design consisting of three rows of white circles, eight in each row, was carved against the upper layer of paint. An adobe disk, 0.6 m in diameter, possibly an altar, was found on the floor near the painted design. A painted vessel, the mouth of which had the same diameter as had the circles in the design, lay on the disk. Two painted cups lay on the floor beside the altar. Close to one of them stood two large stylized figurines made of unbaked clay. Between the altar and the bench, remains of two objects shaped like round bases of columns were found. One is made of the trunk of a tree, another of unbaked clay mixed with charcoal and potsherds. Both are painted red on the sides. Walls of covered couryard 27 had the same plastic decoration as had the walls of the main room. Two seats made of mud bricks and unbaked clay (there was only one seat before the repair) were built near the southwestern wall (fig. 7).

The abandonment of complex 26. 3 was preceded by a number of ritual actions. In contrast to the situation in complex 38. 3, all the details of the interior were preserved, all the objects were left on their places (or specially placed), and several artifacts were put on the floor. These included three copper objects (a mirror, a spearhead, and a fragment of a knife, see Solovyova et al. 1994: figs. 1-6, 8, 15), and the upper part of a terracotta figurine. The floor was strewn with grains of cereals and covered with firewood. After that, the structure was burnt down, after which the floor was first dressed with clay and then covered with pieces of plaster. Walls were destroyed, and a fragment of a twisted copper pin was placed on top of the northwestern wall (Solovyova et al. 1994: fig. 2-2). The entrance and the remaining part of the structure were bricked up (bricks were taken from the walls of the same structure).

Complex 14. 3, level III (fig. 5) postdates complex 26. 3 in excavation 3. Its surface area is 51 sq. m. Apart from the main room, the household included two semibasements 24 and 25 destined for utilitarian needs, main courtyard 13 and possibly certain structures situated behind the northeastern wall of the complex (constructively, however, these are not related to the complex).

In the floor of the main room, along the northeastern and southwestern walls, 9 pits for the vessels were made. One more was situated between the northwestern wall and the smaller side of the largest of the three benches in this room. All the three are made of mud bricks coated with a thick layer of paste. The edges of the seats and four legs of each bench in this room, as well as benches in other complexes of levels III and II, were made in the same technique, apparently in imitation of wooden benches. The space between the legs was painted black, while the legs and the seats were painted red. The southeastern side of the bench to the right of the entrance is shaped like a semicircle encompassing the painted khum dug into the floor. A rectangular adobe structure, plastered on the sides, was added to the central bench. It apparently supported the roof.

The box was a rectangular mass of debris and fragments of architectural details.

Complex 18. 3, level II (fig. 4) is the uppermost and the worst preserved level in excavation 3. The surface area of the main room is 47 sq. m. Near one of the two benches, situated closer to the centre, there lay two copper awls and a fragment of a copper knife (Solovyova et al. 1994: fig. 2-9, 17). To the right of the central bench, two painted khums were found. In the left half of the room, two stone vessels whose bottoms had been intentionally broken off were puttied into the floor. At the same place, a large hearth altar, 1.5 m in diameter, was constructed. In the southern corner, on the lower floor, low plastered walls separated a chamber with a miniature entrance and a miniature step bearing near it.

Complex 58. 5, level IVA (fig. 17) is one of the smallest (its surface area is 18 sq. m. ). On the southeast, it is delimited by the street wall, and on the southwest it borders on complex 10.5 from which it is separated by a wall. Apart from fragments of a khum, which had been placed in a pit near the western corner of the room, an adobe object whose base had been puttied into the floor was found near the southwestern wall. Before the complex was destroyed, all the objects in the room had been broken.

Complex 10. 5, level IVA, was situated between complexes 58. 5, 26. 5, and the street. Its surface area was 27 sq. m. (fig. 17). The area to the northwest of room 10 has not been excavated down to that level. The surface area of the house together with the courtyard is 50 sq. m. The inside of the structure was damaged by pits dug down from level II. The unique feature of this complex is a round tower, partly situated within the box and partly protruding into courtyard 23. It was made of mud bricks set on edge and its floor was also made of bricks.

Within the preserved part (up to the height of about 1 m) there was no aperture. A khum was found on the floor. The box was nearly completely destroyed on the side facing the room, but its bottom part, slanting inwards and painted black, suggests that the decoration was the same as in complexes 26. 5 and 58. 5 (fig. 18). A fragment of a large grinder had been dug into the floor in a vertical position near the entrance into the main room. Near the northwestern wall, there were three pits for the khums. A small adobe cubicle in the eastern corner contained a copper blade (the largest one to be found in Ilginly-depe) and two copper perforators (Solovyova et al. 1994: figs. 1: 1, 2: 10, 12).

Complex 10. 5, level IIIB (fig. 19). After the repair, the surface area of the complex attained 30 sq. m. (55 sq. m. if the courtyard is included). The complex was burnt down together with the adjoining complex 26. 5. Judging by the charcoal, the planks (or logs) were no less than 10 cm thick and belonged to 3-4 species of deciduous trees. In the courtyard, between the square tower and the wall of complex 26, thousands of charred grains of barley were found. Inside the room and in the courtyard, isolated parts of at least three human skeletons were discovered (burials 67-69), and two agglomerations of human bones including skulls (burials 35, 36). One of the burials was accompanied by a Yalangach-type bowl. No traces of fire were discovered on the bones.

Complex 26. 5, level IVA-IIIB (fig. 17) is contemporaneous with both stages of existence of complex 10. 5, but was not reconstructed. In some respects it resembles sanctuaries 38. 3 and 26. 3, but is smaller by one third (53. 5 sq. m.). Along the axis of the left of the two subsidiary chambers, there was a narrow structure 60 cm high. In its upper part, remains of sections were found, apparently of the same type as those in a well-preserved structure in chamber 37 of complex 38. 5. The second chamber was initially a narrow "basement" (fig. 17) and then a podium, painted black and connected with the main room by means of a "window" (fig. 19). Near the walls, there were 11 pits for the khums. In four of them, bottoms of khums were found. Of the five mortars for breaking up ocher, three lay on the floor, one inside a khum, and one was puttied into the floor. The box had a painted and plastic decoration (fig. 18). In front of the door, a female figurine was dug into the floor and wedged with a small stone. About half a liter of wheat grains were strewn nearby. Elsewhere on the floor, only remains of barley grains were found. As in previous samples, determined by Z. Yanushevich and E. Sergusheva, the wheat was Triticum estivocompactum, and the barley was Hordeum vulgare v. coeleste with an admixture of H. vulgare.

A triple aperture, 35-40 cm high, connected both chambers of complex 26. 5 with the main room. Neither a hearth nor clay benches were found in the room. In the fill, remains of a beam, apparently made of plane-tree, at least 2 m long and 25 cm by 15 cm in section, were found. The beam had been moved away from its original place. Two of its wider surfaces and both narrow ones were painted with red ocher. Before the complex was abandoned, a small copper blade had been placed near each of the three apertures connecting the room will the chambers (one of the two apertures leading to the right chamber had been bricked up). Near the "window" there were agglomerations of stone tools and broken bowls two of which belonged to the Yalangach type and one had a round bottom, was almost triangular in section and had an unusual painted design. The bowl and one of the stone vessels apparently had projections (possibly zoomorphous) on their rims; these projections were broken off. A large copper blade (Solovyova et al. 1994: fig. 1: 2) lay near the wall of room 54. The charcoal was found all over room 26 and in the left side chamber. On the charcoal layer, isolated skulls and postcranial bones of adults and children were found (burials 59, 60, 61, 70, and 71), and skeletons of two infants aged about 1 month (burial 58) and 4-5 months (burial 64). The position of the skeletons was unusual: the deceased lay supine, and their hands were behind their heads. Apart from a few phalanges of adults, none of the bones bore traces of fire.

Complex 12. 4, level III (fig. 12). This is a small structure, 13 sq. m. in surface area, with a single room. It adjoins a massive wall. A female figurine, dug into the floor near the entrance, resembles figurines or elongated stones found in other complexes of levels IV and III.

Complex 18. 5, level IIIA. After the destruction of complex 58. 5, its box and the adjacent part of the house wall were used as partitions inside the rectangular structure 55. 5, about 20 sq. m. in surface area, 42 sq. m. together with the courtyard (fig. 20). The floors of the rooms were painted black. Near the entrance, on the outer side, a female figurine made of limestone was dug into the ground in the vertical position.

Complex 18. 5, 32 sq. m. in surface area, emerged in the process of the reconstruction of complex 55. 5 and the destruction of complex 10. 5 which was replaced by a courtyard partly paved with potsherds. The utilitarian part of complex 18. 5 could also have extended in the northeastern direction up to the bend of the street. At the early stage (level IIIA, see fig. 20), there was no hearth in the main room (no. 18), and neither were there any clay benches. The brickwork to the left of the entrance apparently delimited the box. Its wall facing the room has not been preserved, but the floor of the room did not extend beyond that line. The box had apparently been destroyed by burials of infants aged from 2-4 to 6-12 months (burials 79-86). In the fill of the box, isolated bones of a two-year-old child and a facial part of a skull of an adult were found (burials 87, 88).

Complex 18. 5, late level IIIA or early IIB (fig. 20). At that stage, the walls of room 18 were not rebuilt. Two copper blades had been left in a wide shallow pit in the northeastern part of the room under the new floor. The box was destroyed, and the brickwork was turned into a projection. In the left half of the room, a hearth-altar was built, and to the right of the entrance, a square structure 115 cm by 100 cm, the base of which was made of bricks placed on edge and arranged (as in complex 58. 5) in three rows parallel to the wall with the entrance. Near the southeastern wall, possibly under the latest coating, there were five copper needles which had been either placed in a needle-case or tied together; in the eastern corner, four spindle-whorls lay together on the floor, while the fifth one was placed near the bench.

Complex 9. 5, level IIIB (fig. 19). This household, 105 sq. m. in surface area, has been completely excavated. The main structure together with the side chambers had a surface area of 55 sq. m. Complex 9. 5 is the only one which contained just one clay bench. In the left half of the room, there was a hearth-altar with a bowl dug into the hearth pit, and in the right half, a rectangular clay structure, 120 cm by 85 cm, on twelve legs (Masson et al. 1994: fig. 3: 2). The khum in the northern corner had been broken before the complex was abandoned. Its conical bottom was filled with potsherds from the upper parts, a layer of coating was added above, and a stone was put on top. Near the door connecting the structure with the courtyard, an elongated stone was dug into the floor. Before the complex was destroyed, several copper tools had been left in the side chambers: a copper blade (Solovyova et al. 1994: fig. 1: 3) and an unused whetstone were found under an upturned bowl in room 21, and a copper chisel, a small copper point, two small bone awls, and two dozens of various stone tools were found in a special cubicle in room 14.

Complex 1. 7, level III. The southwestern part of the structure, where the utility chambers may have been situated, has not been preserved. The surface area of the structure was either 51 sq. m. (if the chambers did exist) or about 62 sq. m. if there were none. Two red clay benches were found in the room. The entrance was apparently situated in the southeastern wall near the projection; however, the aperture has not been discovered. A small cubicle in the southern corner resembles that in complex 18. 3. At least ten intact spindle-whorls were found in the complex.

Complex 15. 4, level III (fig. 12). The surface area of the structure is 73. 5 sq. m. The surrounding area has not been excavated down to the respective level. Although the box has been well preserved, the plastic decoration has not been found. In the bottom part of the fill of the box, the top and the right part of the bottom part of a large clay figurine were found (Masson 1993: 25). At some stage, there were three red benches in the room. By the time when the structure was abandoned, the central bench had been destroyed by two thirds; none of its fragments were found nearby, but the outlines of the bench have been traced on the floor. In the eastern corner of the room, there were fragments of a khum, a copper rod, and an agglomeration of various stone artifacts (spindle-whorls, a broken weight, and tools for processing grain, skins, paint, and metal).

Complex 3. 1, level III (fig. 3). The structure, 81 sq. m. in surface area, is situated on the southeastern edge of the site. Most of the right wall has been destroyed. On the upper floor, near the place where the central of the two clay benches had adjoined the wall, a copper blade and a stone bead were found, and in the corner of one of the narrow chambers, there was a red polished bowl turned upside down.

Complex 25. 5, level IIIA (fig. 20). After the destruction of complex 26. 5, this area was apparently abandoned for the entire period of functioning of complex 9. 5. After the latter had been destroyed, remains of walls of complex 26. 5 sticking out of the debris were demolished, and structure 25. 5, 33 sq. m. in surface area, was built on that place. Neither the benches nor the hearth have been found inside.

Complex 30. 5, level IIIA (fig. 20). Its surface area is at least 41 sq. m., and the complex is contemporaneous with 25. 2. Due to its position in the surface layer, its preservation is incomplete. While the two main red benches were constructed at the same time as was the black floor (the only one in the complex), the bench opposite to the entrance, and apparently one that adjoined the right side wall and has not been preserved, were built later. A bowl was dug into the pit of the hearth-altar.

Complex 3. 4, level II (fig. 12). The surface area of the structure can be estimated at about 35-40 sq. m. The right half has not been preserved.

Complex 36. 4, late level II (fig. 12). Its surface area is about 45 sq. m. Because it was also situated in the surface layer, its southeastern part, including the hearth (if there was one) has not been preserved. The main pair of benches, covered with ocher, was constructed simultaneously with the single black floor. Apparently there was one more bench opposite to the entrance, but most of it has been destroyed.

Complex 34. 4, early level II (fig. 12). Because the floor in the southern corner of room 34 tapers off and appears on the surface, and the area in the northern corner below room 36 has not been cleared out, the general layout of the complex can not be reconstructed in detail. The complex is contemporaneous with utility structure 22. A khum lying on the side, a zoomorphous limestone mortar, and a copper blade were found inside (Masson et al. 1994: fig. 6: 2; 1: 18).

Complex 59. 4, level II (fig. 14). The surface area of this complex is 76.5 sq. m. Its back wall faced street 70. In front of it, there was main courtyard 45, 7 m by 14 m, overlaid by several layers of adobe and coating, not by pavement, suggesting that the courtyard was covered (fig. 15, sections a-d-a', c-c'). In the northeastern half of the courtyard, a square structure 10-15 cm high and 165-185 cm in width was cleared out. It was coated with clay four or five times; layers of coating are separated by thin layers of ash. Apparently fire (not a heavy one, though) was made here several times.

A complete set of clay benches, four in number, has been preserved in this sanctuary: the central one (fig. 16: 15), another one, adjoining the wall and connected with the first one (fig. 16: 16), one along the right (northeastern) wall (fig. 16: 17), and one connected with it and facing the entrance (fig. 16: 18). All are painted with ocher. In the main room, near the door connecting it with the courtyard, a vertically positioned mud brick, possibly an idol, was protruding out of the floor. Neither the pits for the vessels nor the vessels themselves nor even their fragments have been found in the building. On the place of the northeastern wall of the complex and immediately behind it, three dozens of small hearths, partly overlapping, were discovered. Stratigraphically, they are related to the period of the destruction and ritual burial of the structure and correspond to the layer of ashes and charcoal which overlaid the entire western half of the main courtyard (fig. 15, sections d-d', d-1, b-b'). In this layer, fragments of a khum with a Geoksyur-type painting were scattered (fig. 16: 1, 2). The structure to the southeast of the main courtyard was apparently a residence house. Four niches in the wall of room 41 (see Berezkin 1990 for a photo) have no parallels in the Aeneolithic of Turkmenia. South of the residence house, utility courtyard 52 was situated. In the wall separating it from main courtyard 45, there was a stove destined for manufacturing or domestic purposes.

Complex 59. 4, level IB (fig. 13). An earlier sanctuary was first filled with loose matter up to the level of the benches and then bricked up. As in courtyard 45, the mud brick was used. The new building was smaller (57. 5 sq. m. ) and had two successive floor levels, one situated 10 cm above another (fig. 15, section a-d). The benches (fig. 16: 13, 14) had no side rolls or separate legs. The benches on the upper floor were painted white (with a solution containing plaster or lime) rather than red. They were twice more narrow and lower than the usual ones, and trapezoid in section, widening toward the bottom. Apart from the two main benches, a third one was apparently present, but most of it has not been preserved. Under the floor of one of the utility chambers, a leg of a clay table was found (fig. 16: 3).

Conclusions. Some of the features differentiating the complexes are chronological. These apparently include the decor of the boxes and the types of the hearths. A specific plastic and painted decoration (a black horizontal band slanted inward and an upright projection rectangular in section and tapering toward the floor) are not found in complexes later than level III, but are present in all complexes of level IV and earlier ones regardless of their size (38. 3, 26. 3, 26. 5, 10. 5, and 58. 5). In level III, this decor has been registered in sanctuary 14. 3 but not in sanctuary 15. 4. In other contemporaneous complexes there are either no boxes or the plaster has fallen off from them. The hearth-altars have not been found in levels earlier than III but are present in all well-preserved complexes beginning from late III (it is impossible to say anything about early III since complex 10. 5 is heavily damaged). In completely preserved rooms of level IV (26. 3, 26. 5, and 58. 5), there are no hearths whatever, but in complex 38. 3 of level V a round pit with traces of fire on its edges was found in the middle of the room. In level III, a transition from wooden benches to clay ones occurred. However, only complex 26. 3, where charred remains of seats with remnants of paint were discovered, indicates that wooden benches existed in level IV and even earlier. Beginning from level III, clay benches are present in all large and middle-sized complexes and even in some small ones (18. 5). Another element found (only twice, though) no later than level IV is the podium, open toward the room, on the place of the subsidiary chamber. Its likely prototype is the "sufa" dating from the times of Namazga I in structure 1 at Dashlyji-depe (Khlopin 1960: 141, fig. 6, 9; 1963: 15). More remote parallels are provided by platforms of houses of Jeitun Neolithic culture and of sites of Zaghe and Sang-e-Chaxmaq in Northern Iran (Masuda 1974, 1976; Negahban 1979).

The principal feature of the evolution of clay benches was the progressive loss of details which did not accord with properties of the new material. In level III, the benches still follow the wooden prototypes in all their details. In level II, the plastic parts become schematic and lose their former elegant outlines, and in level I they disappear altogether. The last to disappear was the red ocher paint.

The number of utility vessels present in the complexes was evidently related to the destination of the latter. In this respect, complexes nos. 38. 3, 26. 3, 14. 3, and 26. 5 are quite distinct from the remaining ones since 9-12 pits for large vessels were found in each of them.

Decrepit "sanctuaries" were apparently always "buried" according to certain rules, and prestige objects were left behind on the floors or in cashes. Ruins of many complexes are "sealed" with mud bricks. Some had apparently been burnt down. In complex 26. 5 and in the "cult building" at Geoksyur I, human remains were found inside the rooms destroyed by fire (Sarianidi 1962: 49-51).


Н.Ф. Соловьева. Стенопись Илгынлы-депе

N. F. Solovyova. The wall-painting of Ilginly-depe

During the last decade, the IIMK Kara-Kum Expedition has been excavating the Ilginly-Depe site, South Turkmenistan, one of the largest in the entire Irano-Turkmenian area, dating from the fifth to the third millennia BC (Masson et al. 1994).

The excavations have brought to light unique architectural associations, tentatively described as sanctuaries. They are square blocks of structures usually including the main room with several subsidiary rooms and, sometimes, living and utility annexes. In front of their facades, covered courtyards delimited by the side walls were situated. The main rooms have a regular layout and a rich decor which includes floors painted black, wall panels, wooden and clay benches painted with ochre, altars and reliefs. Numerous unique artifacts were found inside (Berezkin 1992).

By late 1995, six construction horizons were revealed in excavation area № 3. In each horizon except the upper one, which was preserved in the northwestern part of the area only, the space in the centre is occupied by a sanctuary constructed directly at the place of an earlier ritual structure. In some respects, the sanctuaries are unique among the synchronous structures of the Near and Middle East. Their large size, certain elements of their decor (adobe seats in horizon IV, a sculptural pro-tome of a bull in horizon V) and artifacts found inside, such as a copper socketed axe-adze and a clay mould for a similar type of axe, distinguish these sanctuaries from those excavated elsewhere in Ilginly-Depe (Solovyova 1994). The most important discovery, however, is the wall painting, the fragments of which were found in the main rooms of the sanctuaries in excavation area № 3.

The inhabitants of the site evidently attached considerable importance to the colour decoration of the interiors. The floors were usually painted black (where no black paint was found, it had apparently been lost). Bright red benches contrasted with black floors. The bottom parts of thoroughly plastered walls were decorated with low black panels. Sometimes only the wall facing the entrance had a panel. Above the panels, walls were either whitewashed or left unpainted. Sometimes they were covered with black paint all over, and occasionally a design was painted on the black surface. Designs were made either in black against the light background or in light yellow, the colour of unbaked clay. The painters used a technique of sgraffito, which was unique at that time. The surface of the wall was covered with at least two layers of paint differing in colour. Then an outline drawing was made on the outer coating and some parts of it were incised to reveal a ground of a contrasting colour. Walls, including those with designs, were plastered. The plaster consisted of unbaked clay mixed with chopped straw. Over the plaster, a thin coating of black paint was applied and polished. The paint was a mixture of clay, water, and a large amount of ground charcoal. During each repair, the walls were either just repainted or first replastered and then painted anew.

In excavation area № 3, fragments of painted designs were discovered in main rooms of construction horizons IV (room № 26), V (room № 38), and VI (room № 47). In all the three instances, the patterns were found on the wall opposite to the entrance or on an architectural detail connected with it. The first to be found was a poorly preserved fragment of a design on the northwestern wall of room № 26 (horizon IV). It consisted of three rows of identical light-coloured circles painted against the black background (fig. 1). The space covered with the design was situated below the wall aperture between the two upright projections and had no horizontal projection which delimited the lower part of the wall elsewhere. This area was painted black as was the frame of the aperture. In the bottom part, the paint merged with that of the floor. The pattern, made against the upper layer of black paint, cut through all the earlier layers down to the unpainted plaster ground. The diameter of the circles coincides with that of the mouth of an unpainted jug which lay on the floor beside the wall. Perhaps the circles are imprints of its rim. The remaining part of the northwestern wall and all other walls in room № 26 were marked by a horizontal projection situated 0.2 m above the wall and were painted black from the floor up to the projection; above this level, they were whitewashed.

On the floor of room № 38 in horizon V, near the northwestern wall, a schematic protome of a bull made of unbaked clay was found. Perhaps the sculpture originally stood on the wall marking the left edge of an aperture similar to the one described above. Then, for some reason, it was thrown down. On the right side of the protome, against the upper paint layer, eight creeping snakes situated one above another are drawn (fig. 2). Their outlines, differing in size, were first incised with a sharp tool and then all layers of black paint were removed and the plaster ground was revealed. For that reason, the reptiles are light-coloured. The relief and the painting of the northwestern wall in room № 38 are similar to those of the respective wall in room № 26 of horizon IV.

In both cases, the patterns consisting of a succession of similar elements were painted just once, during the last repair of the walls, and had not been destroyed before the structures were abandoned.

On the northwestern wall (preserved up to the height of 0.7 m) of room № 47 in horizon VI, which was the main room of the earliest among the sanctuaries excavated by late 1995, seven layers of plaster were found, five of which (№№ 2-6 beginning from the earliest one) were painted in a manner traditional for Ilginly-Depe (fig. 3). Near the northern corner of the room, the lower part of a wide aperture similar to those in the northwestern walls of the main rooms in horizons IV and V, was revealed. The space below the aperture was separated from the remaining part of the wall by two upright projections, and the horizontal projection in the bottom part of the wall was lacking. The space between the two upright projections was left unpainted. After each of the seven repairs, the entire area below the aperture remained black. In this part, the black paint of the wall merged with the same paint covering the floor. Above, the black paint covered the lower part of the frame of the aperture and the floor of the utility room № 46 situated behind the wall, its floor being almost level with the aperture. At the height of 0.1-0.15 m above the floor, a horizontal projection has been traced on all the layers of plaster on the remaining part of the wall. Below it, a narrow unpainted strip of the wall separated the black-painted floor from the wall of the same colour.

The lower layer of the plaster, unpainted, provided the ground; the upper, 7th, layer was intentionally applied to hide the pattern. It was painted black from the floor up to the horizontal projection. The first and the last plaster layers are 5-8 mm thick, and the unpainted ones, 1-3 mm thick. The elements of the pattern, repeated many times, are the same in nearly all layers: a crescent with horns turned upwards, a stylized tree, a circle, and a snake. The first two elements are lacking in the third layer, and in the fourth and fifth layer, a mythological character is added (fig. 4).

The spatial layout of the composition, too, is similar: while the larger part of the painting (the left one) is filled with circles variously combined with crescents and a tree with a mythological character under it, the right part of the wall below the aperture is always painted with snakes situated one above another and creeping to the left. The upright projections separated by an unpainted space, the narrow unpainted stripe of the wall near the floor, and the lintel of the doorway leading to room no. 45 delimit the main composition. In contrast, the uninterrupted black paint in that part of the wall where snakes are represented, the lower part of the frame of the aperture, and the floors in rooms nos. 46 and 47, convey an impression that snakes are associated with the black colour. The number of snakes and their size vary from layer to layer, but their heads are always directed to the left, toward the remaining part of the composition.

The preserved fragment of the first layer in the left part of the painting, as the other layers, is mainly filled with rows of circles differing in size: the lower they are situated, the smaller they are. Below them, there are crescents lying horns up (in one place, their succession is interrupted by small circles). On the right, four crescents are painted pairwise one above another, and on the left, three crescents, arranged likewise. On the right, rows of circles and crescents are flanked with stylized trees growing one above another. Each has a straight trunk, roots, a top, and three pairs of symmetrical circular branches (fig. 4: 1).

In the second layer, all the crescents are shifted to the left and arranged in at least two rows, one above another. On the right, a tree is depicted. Between it and the crescent, four rows of circles are preserved, the lower one consisting of small circles only; in the second row on the right side there are two small circles instead of a large one (fig. 4: 11).

The painting of the third layer is similar to that of the middle part of the second one. However, there are no crescents or trees (fig. 4: 1 II).

Compositions of the fourth and fifth layers are virtually identical, except that circles are somewhat displaced (fig. 4: IV, V); on the left margin, an outline of a large crescent is preserved (fig. 4: IV, V). As usual, most of the space is filled with rows of small and large circles. On the right, there is a tree similar to those on earlier layers but higher. Beside it, a mythological character is depicted standing upright. It combines anthropomorphous and zoomorphous features (fig. 6). Its head has not been preserved, its fore (or upper) limbs are uplifted, and the figure in the fourth layer has a tail or a phallos. Judging by the direction of fibers of the plaster, both the tree and the mythological character were bas-reliefs whose shapes were cut away during the application of the next layer of plaster. To the left of the tree, below the last circle in the second row, an upright object, elongated-oval in shape, is represented. Apparently the lower part of a similar object is preserved in the centre of the composition on the first layer. Above the top of the tree, there is a poorly preserved element shaped like a comma with arch directed upward. This may be a root of another tree the trunk and branches of which have been destroyed.

The wall painting discovered at Ilginly-Depe is the fourth to be found on the agricultural sites of southern Turkmenia. In the 50s and 60s, fragments of polychrome wall painting in the geometrical style were discovered in the Aeneolithic horizons of the northern hill at Anau and Yassy-Depe (Khlopin 1963: 9, 13, tables IV, XI). At the Jeitun Culture site of Pessejik-Depe, fragments of painted representations of humans and animals resembling those of Chatal-Huyuk were found (Berdyev 1972: 38-39). However, in terms of subject and manner, the Ilginly-Depe wall painting has no parallels among these few finds. Moreover, while all the few fragments known from the agricultural sites, regardless of subject, are made in the traditional manner of monochrome or polychrome painting, the sgraffito art of Ilginly-Depe is unique.


В.А. Самар. Дремайловское поселение эпохи поздней бронзы на юге Украины

V. A. Samar. Dremaylovka: a Late Bronze age settlement in the Ukraine

Dremaylovka site is situated between the villages of Zmeyevka and Novoberislav, Borislav district, Kherson oblast', 4 km upstream from Berislav, on the left slope of a gully now submerged in Kakhovka reservoir. The site was discovered by N. P. Olenkovsky who dated it to the Late Bronze Age. The preserved part of the site is a 250-metre-long strip stretched along the east-to-west line along the bank of the reservoir. Formerly the settlement was situated on three adjacent promontories separated by ancient gullies (figs. 1, 2). The depth of the archaeological deposits varies, being small in the eastern part of the site and increasing in the western direction where stone foundations of houses were found in the profile of the bank.

The cultural sequence consists of a gray layer overlaid by turf with alluvial sand, shells, and a few potsherds dating from various periods from the Sabatinovka to the Scythian times. The cultural layer contains artifacts belonging to the Sabatinovka Culture and a small number of later ones. The third layer includes the foundations of walls.

Construction techniques. Remains of two residence structures were discovered (№ 1 in excavation areas I and II, № 2 in areas III and IV). Neither has been completely excavated.

Structure № 2 is in the centre of the site. It is partly submerged and transected by the steep bank of the reservoir. A wall, situated along the west-to-east axis and dividing the structure, has been unearthed. It is made of large (up to 1 m in size) limestone slabs placed on edge and slightly dug into the ground. The gaps between them are quite small and are filled in with gravel. Inside the house, potsherds dating to the later Sabatinovka and Belozerka periods were found. The ceramic assemblage is identical with the one from structure № 1.

Structure № 1 was apparently situated in the westernmost part of the site. About 800 sq m of the interior surface have been unearthed (fig. 2). The house consists of several rooms and is functionally divided into two parts, northern and southern. The northern part was destined for domestic activities, and the southern one for utilitarian purposes. The northern part (rooms №№ 1, 2, 5, and 12) has a regular layout, the foundations are made of very large slabs, carefully fitted together. Because the surface on which the structure is situated is gently sloping at 1.5 m in the southern and western direction, the foundations of walls in the southern part are constructed on a small artificial cushion made of subsoil clay and loam. During the construction of rooms, situated along the west-to-east axis, another technique was used. Rooms №№ 1 and 2 are on the same level; when the western wall of room № 5 was being constructed, a natural elevation of the construction layer was used: a small ledge was made on it and the foundation was laid on the ledge. Different levels of the foundations attest to people's attempts to cope with the sloping surface rather than to various stages of construction. Doorways were made in the northern part of the house.

Near the entrance to the house, a quern made of fine-grained limestone was found. Near the entrance to room № 2, a concentration of potsherds (about 100) was discovered and a fragment of a bronze artifact. On the floor of the corridor, a small goldsmith's anvil, a bronze awl, several granite hammerstones, a fragment of the working edge of a polished axe, and polishers made of potsherds were found.

In room № 2 in the northern part of the house, three utility pits were found (№№ 3, 5, 6). One of them (No. 6) was used as a cellar. In the same room, remains of a hearth were discovered in the form of a pavement made of smooth flat pieces of chert. Three doorways only were registered in house № 1 (fig. 2).

Foundations of walls are of two types, both of them well known in the steppe zone: (a) irregular, one to five layers high and one layer wide, and (b) three layers high and two layers wide. The first type was used in the foundations of the inner walls of the houses. The basal parts of such foundation were made of large rugged blocks of limestone and shell-rock with smaller stones on top. In many instances, the space between the stones was filled in with gravel. The grout was made of clay mixed with black earth, river sand and crushed bones of animals. In some cases, undestructed corners of the inner walls were discovered. They were buttressed with larger stone slabs, placed vertically. Parts of the foundations of the inner walls subjected to extra stress (entrances to the niches etc.) were buttressed in the same way. Safe parts were made of many layers of small stones, in keeping with the traditional technique. The second type was used in the foundations of outer walls.

In the floors of various rooms of house № 1, utilitarian pits №№ 3, 5, 6, and 13 were found (fig. 2). In room № 3, there were three small pits placed close to each other. Their function is not clear, although in one of them, narrowing towards the bottom, vessels could have been placed. Utility pits have been found in other Bronze Age sites in the Ukraine (Stepovoye, Voronovka II, and others). None of them, however, were discovered inside a synchronous structure at Zmeyevka (excavation area III).

Apart from utilitarian pits, three types of ritual ones were discovered. The first type is represented by hollows dug in the centres of three out of four rooms of the northern part (fig. 2, pits №№ 2, 7, 14). Each of them contained clay "rolls". These pits were apparently used in domestic fertility rites (Otroshchenko 1990: 6).

Ritual pits of the second type are altars №№ 1 and 4 (fig. 2) which were dug before the foundations of the house had been constructed. On the bottom of pit № 1, skeletons of two dogs lying on their sides were discovered. In the filling of this pit, somewhat above its bottom, a tortoise shell was found. On the bottom of altar № 4, there was a dog's skeleton which had been first dismembered and then reassembled. Alongside of it lay split long bones of a cow and a horse bearing traces of dog's teeth.

Construction sacrifices, including those of dogs which were being buried under the foundations of houses, had been practiced in the steppe zone since the Aeneolithic (Bibikov 1953: 197-198; Telegin 1973: 37, 44-45). As suggested by finds at Dremaylovka, the turtle, too, was used in sacrificial rites. These animals, like horses, cows, sheep, and even humans, were being sacrificed by the Indo-Arians (Klein 1987). The dog, believed to be the purest and the most sacred animal, could have replaced humans as objects of the sacrifice rites.

Ritual pits of the third type are situated under the southeastern corner of room № 9 and the southwestern corner of room № 5. Both contained intact vessels. The one found under room № 9 is a usual jar, and that under room № 5, a miniature biconical vessel with ochre-coloured surface (fig. 3, 11). Both look somewhat archaic and had probably gone out of use before having assumed a ritual function.

The ceramic assemblage of Dremaylovka is quite rich. In house № l, over 6,000 potsherds were found, and in house № 2, over 500. Assemblages from rooms №№ 1 and 11 of house № 1 stand out as being especially diverse. All the pottery is hand-made of clay tempered with crushed sherds, ground quartzite, limestone, and mica. The surface is smoothed. Most vessels have dense, well-baked walls. The firing was uneven, the colour of the vessels varying from black to terracotta, usually dark, grayish-brown or yellow-brown. In many instances the outer surface is covered with slip. Some sherds contain a heavy admixture of sand. Because its quality is quite low, this ware was apparently used for non-utilitarian purposes. The surface of some cooking vessels was smoothed with a denticulate stamp.

The most common vessels are pots, which belong to several types (fig. 3). The first one is represented by thick-walled large pots. Their necks and shoulders are buttressed with 2-3 rolls ornamented with incisions and round impressions (fig. 3, 14-17).

Pots of the second type have strongly everted collar-like rims and convex bodies. Necks of many pots are ornamented with incised horizontal lines with oblique hatching between them (fig. 3, 1-3). Typologically, this category derives from Belozerka ceramics.

All pots of the third type are fragmented. Their distinctive feature is the absence of neck (fig. 3, 6-8).

Pots of the fourth type have vertical necks below which there is a pattern consisting of pierced holes and long oblique incised lines (fig. 3, 4-5).

The fifth type is represented by well-profiled vessels having globular bodies and sharply everted rims and decorated with a roll which is rounded-subtriangular in section and is situated below the rim.

Jars are represented by three types. Those of the first type are open jars with rounded or truncated rims which are sometimes everted. This pottery is typical of the early Timber Grave period.

Jars of the second type are closed and have elongated slightly convex bodies without distinct shoulders. The rim is rounded or tapering, and below it there is a 1-3-cm-thick roll, plain or ornamented with oblique notches made with either a paddle or a stamp. These vessels are well-fired and often covered with slip.

The third type comprises jars with straight slightly sloping sides. Under the rim, a roll was applied which was rounded or sharpened in section and sometimes decorated.

Most vessels belong to the second or third type, and they are typical of the Sabatinovka Culture.

Also, there is a series of small jars some of which are ornamented with stamp impressions. Apparently they may be regarded as children's ware.

The kitchen ware is represented by pans (fig. 4, 18) decorated with oblique notches along the edge of the rim. The few large ewers are well-made.

The rims of the hemispherical bowls are either truncated (fig. 4, 15) or rounded (fig. 4, 16). Judging by the fragments, some bowls were conical (fig. 4, 17).

The table ware is represented by scoops, beakers, and straight-rimmed bowls. The collection includes some scoop handles (fig, 4, 9-10). In the eroded fill of house № 1, a large fragment of a horned scoop handle (fig. 11) was discovered, a unique find for a Southeastern European site.

The group of clay artifacts comprises "rolls" which were found in the pits (fig. 4, 14), spindle-whorls (biconical, fig. 4, 8, or, in one instance, flat, fig. 4, 6), a pellet (fig. 4, 7), and flanged braziers with trimmed edges (fig. 4, 13).

Among the bone artifacts, there are stamps for making impressions on skins (fig. 4, 1), a denticulate stamp for decorating vessels (fig. 4, 3), and a pendant made of a horse's phalanx (fig. 4, 4). There are also tools made of sheep's astragali with traces of wear on the sides (two are pierced), polishers made of horse's hoofs, and retouchers.

Inside house № 1, two bronze awls were found (fig. 3, 9-10), and what was probably a rim of a bronze vessel.

The collection of stone artifacts comprises querns, whetstones, percussors, retouchers, polishers, and an anvil. Also, a hammer with two working surfaces was found; it was probably attached to the handle by means of a leather strap.

Faunistic remains are quite numerous (11, 374 bones from at least 158 individuals). Most bones are those of domestic animals, the distribution being as follows:



Number of bones


Number of individuals



3 024

















































Chronology. The layout of the houses, the construction techniques, and the artifacts suggest that the settlement existed during the late Sabatinovka period (Samar 1990, 1992).

The closest parallels to house № l are many-chambered constructions at Voronovka II (Vanchtigov et al. 1991), as well as houses revealed by excavation III at Zmeyevka (Burakov 1961: 33) and excavation I at Anatolyevka (Chernyakov 1985). All these were built according to a regular layout using identical construction techniques. The feature shared by all of them is the presence of apse-like walls.

The principal dating criterion is ceramics. Such features as pots being much more numerous than jars, the use of red and black high-quality polishing, a large number of braziers with trimmed edges, and the custom of applying round or elongated pellets to the shoulders of the vessels, are indicative of the late Sabatinovka period. Again, some fragments of polished beakers are similar to Belozerka types, and there is a large number of scoop handles including those with knobs and plain flat ones like those from the Sabatinovka assemblage. The unique horned handle is typical of the Lower Danube area. A similar one, found in Malul Rosu, Dobruja, in association with pre-Babadag pottery, is believed to attest to the influence of Vatina Culture distributed in Oltenia, Banat, and northwestern Yugoslavia (Vanchugov et al. 1991: 50). The only parallel in the Northern Pontic comes from Voronovka II.

The houses at Dremaylovka. then, must be dated to the late Sabatinovka period. Based on recent findings suggesting that Sabatinovka is older than previously believed (Chernykh et al. ), the likely date of Dremaylovka is 13th or early 12th century BC.

House № 1 situated in the western part of the site belongs to the earliest period of Dremaylovka and the adjacent site of Zmeyevka (which, as the horizontal stratigraphy indicates, developed in parallel, see Samar 1992: 25). The finds from excavation areas I and II at Zmeyevka, situated in the eastern part of Dremaylovka - Zmeyevka complex, date from the early Belozerka period.


М.Е. Килуновская, Вл.А. Семенов. Оленные камни Тувы

M. Kilunovskaya, Vl. A. Semenov. Tuvinian stag stones, part 1. Recent finds, typology and cultural attribution

Stag stones are works of monumental Scythian art distributed in northwestern Mongolia, Trans-Baikal region, highland Altai, and Tuva. More of them have recently been found during the excavations of a large early Scythian mound on the Kholash River, Mongun-Taiga, in the Ulug-Khem valley, and in Uyuk Hollow. In the present article, they are described in a broader context.

Stag stones from the Kholash River (fig. 1: 1, 2) are the first to be found in Mongun-Taiga. They are carved from granite and depict reindeer in the Mongolo-Trans-Baikalian style. The dimensions of one of them are 170 cm by 65 cm by 20 cm. The representation in carved on one side of the slab, in the upper part of a broad plane which had been intentionally trimmed toward the top. A solar sign (circle) is represented, and three silhouettes of reindeer with stretched beak-shaped muzzles are directed toward it. The antler is shown in one animal only. Two others are hornless, and in all the three the fore leg is reduced in size (fig. 1: 1).

The second stela, rectangular in section and tapering off toward the top, has a protrusion marking its front side. On the lateral sides, circles are carved. The right one is 15 cm in diameter. Two reindeer figures are directed toward it, shown in profile with stretched beak-like muzzles. Both are hornless, have almond-shaped eyes and humps, which is a typical feature of such representations. On the left side of the stela, 180 cm by 33 cm by 35 cm, only a ring, 7 cm in diameter, is carved (fig. 1: 2).

Similar stelae representing reindeer of the Mongolo-Trans-Baikalian type without any additional attributes, have been found elsewhere in Tuva. The most interesting one was found by us in 1989 near Lake Beloye, Turan-Uyuk Hollow. It lay on a ploughed field near a chain of Scythian type mounds. The gray sandstone slab, unseparated into zones, bears representations of five reindeer, inscribed into one another in such a way that the hump of the one animal parallels the bend of another one's neck. The muzzles are disproportionately elongated. Ears are leaf-shaped, and eyes are large and almond-shaped. Large antlers with two branches directed forwards are spread above the backs. Legs are reduced (fig. 1: 4).

In 1989, a reindeer stela was found on the slope of Mount Kosh-Pei near the village of Arzhan (fig. 2). It is unique in that it contains many elements peculiar to the stelae which have been atributed to different types by various researchers. The stela is separated into three zones. In the upper one, separated by a "necklace" (a succession of oval pits), three oblique stripes are carved on the face side while the lateral sides bear representations of earrings with bell-shaped pendants and points inside the rings. These earrings have close parallels in an Aldy-Bel' type association from Saryg-Bulun burial ground (mound 1, burial 1), dating from 600-400 BC.

In the central part, figures of animals are situated. On the left side, there are seven flying reindeer with beak-shaped muzzles and reduced legs. Antlers are either absent or just one bifurcated branch is carved. Above them, there are two identical coiled panthers, one inscribed into another. In terms of style and details, they are similar to that shown on the unique bronze horse plaque from the Arzhan mound. The larger of the two is even of the same size (25 cm in diameter) as that on the plaque. The back side bears a carved representation of a coiled carnivore. Below it, a horse with bent legs is represented. On the right lateral side, a pentagonal shield with a circular boss in the centre is carved. On the left side, there is a representation of a socketed pick on a long handle with an oval butt. Similar picks were found in China where they date from the period no later than early Western Chou (9th century BC).

In the lower part of the stela, below the belt shown by two broad bands, weapons are represented. On the facial side there is a dagger with a straight guard and a ring-shaped top, attached by means of a short belt with two branches, possibly a hook. The blade has a rib, and a similar dagger is depicted on a reindeer stela from Arzhan dating from the early Scythian period. On the right side, there is knife in a sheath with a top shaped like "an arch on support" and a trapezoid thickening on the end. Beside the knife there is a rectangular hone, and beside it a bow in a gorytos (part of it has been lost). The bow has a curved end and is similar to a sigmoid Scythian bow. The gorytos is elongated and has a trapezoid protrusion in the lower part. Such gorytes predominate on stag stones in Mongolia and are not known from associations. The last carving, a very interesting one, is that of a boar. In terms of style, in conforms to the canons of Scythian art. The manner, however, is unique. It is a bas-relief, whereas all representations in the monumental art of central Asia are counterreliefs or reliefs en creux. The reindeer stela from Kosh-Pei, then, combines elements that may be dated from the 10th to the 5th century BC. Judging by the style in which animals (boar, horse, panthers, and stylized reindeer) are represented, it should be dated to the early Scythian time (700-500 BC).

Stelae of the Mongolo-Trans-Baikalian type coexist in Mongolia with stelae made in a conventionally realistic manner, which we describe as Turanian, since the most characteristic specimens of that type were found in the Turan-Uyuk Hollow. A fragment of a stela from Arzhan (fig. 4: 2) belongs to this type. No similar stelae are known in Tuva. Animal representations cover the entire lower belt of the stela. Such a belt-like arrangement of carvings in the lower part below the belt with weapons is rare on stag stones of this type. All the representations are small (ca. 10 cm), and this, too, is not typical of the stag stones. Animals are carved in two manners, silhouette and outline.

Several monumental stelae with realistic representations of reindeer are now in the collections of Minussinsk and Tuva regional museums. They are typical monuments with multifigure compositions made in the Scythian animal style and representations of early Scythian weapons (figs. 4: 1, 4; fig. 8).

Many stelae were reused in later periods, as evidenced by a fragment found in a Kyrgyz mound near Mount Charga (fig. 5: 2). It bears a short runic inscription. The stela is flat (85 cm by 65 cm by 15 cm) and is made of coarse-grained dark sandstone. The bottom is broken away. On the narrow facial side, there is a broad band and three oblique lines. On the left side, there are three silhouttes of reindeer, heavily damaged by lichen. Large branching antlers are seen, and massive bodies with triangular humps. The muzzles are realistic, the ears leaf-shaped, the eyes drop-shaped, and the mouths half-open. The left side is better preserved and accommodates three mountain goats standing one below another, and a large boar perpendicular to them, legs directed to the front side of the stela. All carvings are silhouettes, and lines are not sharp, unlike those on the stag stones.

Another unique stela (fig. 5: 1) was found near Mount Charga. It is made of sandstone (height, 1.8 m) and represents an anthropomorphous mask. The stela has been thoroughly carved and is cigarshaped, oval in section, and reveals certain features which are typical of stag stones in general. It is divided into three zones by two broad horizontal lines, a "necklace" and a "belt". Above the "necklace", there are three oblique lines which are long and encircle the stone, appearing on other sides in a collar-like fashion. Above them, the outline of an anthropomorphous mask is carved. The lower part of the face is stressed, and the chin is depressed. The eyes are roundish, have dots inside, and are connected with the V-shaped nose. The mouth is shown by a line, and its corners are bent downwards. Both the nose and the mouth are similar to those represented on the Altai stelae, specifically on the Chuy specimen. Eyebrows are shaped as connected arches. Ears, drawn like arches with small circles inside, are separated from the face.

The anthropomorphous mask carved on the Charga stela is paralleled by the earliest representations on the stag stones of Mongolia and Altai. It has certain stylistic peculiarities, including the double outline of the eyes, nose, and ears, which make it similar to the Okunev masks. Between the necklace and the belt in the central part of the stela, a pair of carnivores, with both canine and feline features, is represented. Beside them there are other animals: a goat, a reindeer, and remains of some other figures. Perhaps the procession of animals encircled the entire stela, as on the Arzhan specimen. All the animal figures are outlines carved in the percussion technique. The insides of their bodies are filled in with lines and scrolls on the croup typical of the Scythian style. Overall, the animals as well as the anthropomorphous mask should be dated to the early stage of the Scytho-Siberian art (800-600 BC). In our opinion, the formation of the canon was a gradual process. Early stag stones have representations which are similar to petroglyphs. These are present on stelae found near Mount Charga, near the Uyuk River (fig. 3), and in the Ulug-Khorum burial mound (fig. 7). On one of the stelae from Ulug-Khorum (the fifth stone in the kerb), a horse figure is elegantly carved. A similar figure is present on a reindeer stela found by Grach in the Sagly Valley, and on a stela from Saygyn with reindeer stylized in an ornamental fashion (fig. 1: 3).

Several stag stones were discovered during the excavation of kereksurs. One of them was found in 1987 in the Khorum mound (fig. 6) on the right bank of the Ulug-Khem. Seven stelae with representations were found by Grach during the excavations of Ulug-Khorum kereksur, the Sagly Valley. The Ulug-Khorum mound represents a certain stage reflecting the world model of the people who built the mounds. This tradition includes cist burials on the old ground surface, the construction of cromlechs around the central grave, drawings on the walls of the cists inside and outside the graves as well as on the stelae and on the slabs belonging to the constructions. All these are similar to the Saga-holm burial mound, Smaland, Sweden, where figures of animals, resembling the central Asian ones in terms of style, are represented on the cromlech. The structure of these monuments is paralleled by the layout of Bronze Age settlements such as Arkaim, Chelyabinsk Region (Zdanovich, 1992: 79-84), Demircihuyuk, Anatolia, and others.

In Tuva, stag stones belong to the principal elements of the Scythian type culture. Three main types can be distinguished: (a) stelae with realistic representations of reindeer (Turan, fig. 4: 1, 2, 4; fig. 5: fig. 8); (b) those with ornamentally stylized representations (Mongolo-Trans-Baikalian, fig. 1; fig. 2); and (c) intermediate ones (Sushin, fig. 4: 3). All these types coexist in Tuva and date from the early Scythian period.


С.С. Миняев. Бронзовая пластина-пряжка из Дырестуйского могильника

S. S. Minyaev. A bronze plaque buckle from the Dyrestuy burial ground

Studies into Hsiung-nu history and culture have a longstanding tradition at the Saint-Petersburg Institute of History of Material Culture. Over the recent years, these studies have been pursued by the Trans-Baikal Archaeological Project which, jointly with the Research Centre of the Buriatian Department for the Protection of Monuments, is focusing on a comprehensive field work at one of Hsiung-nu key sites, the Dyrestuy burial ground in the southwestern Trans-Baikal area (Jida district, Republic of Buriatia). As large parts of the site were being completely excavated, numerous ground burials were discovered which had no marks on the modern surface and for that reason had not been looted. The most important artefacts found in these graves are objects of art, first and foremost bronze ornamented plaques which were attached to the belts. Some information concerning these plaques has been published (Davydova and Minyaev 1988, 1993; Minyaev 1995). In the present article, one more composite belt, one from burial № 128, is described. It includes a bronze plaque which has no parallels among the known Hsiung-nu bronzes.

Burial № 128 belongs to group 1, of which twelve burials have been previously excavated. Grave № 128, like the adjacent grave № 127, was apparently related to the central mound, № 44 (fig. 1; see Minyaev 1993 for information on the general layout of the burial ground). Grave № 128 was not marked by any external features. The grave spot, rectangular with rounded corners, was detected at the depth of 30 cm during the excavation of a large area of the burial ground. The grave was filled with a sticky grey loam and detritus, some pieces of which were 15 cm in diameter. Inside the grave, remains of a coffin, 175 cm by 45 cm, were found, the walls being preserved at a height of about 20 cm. The lid had completely dissociated and was represented by a layer of blackish decay. Most of the walls, too, had decayed, and a thin layer of decay had remained of the bottom. The long walls protruded beyond the short ones, suggesting that they were joined by mortises and tenons.

The buried person was lying supine in an extended position, head directed towards NNE. The preservation of bones was poor, the skull had been crushed, and the ribs had decayed. Near the right zygomatic bone, a turquoise pendant was found, and to the left of the mandible and below it, there was a necklace consisting of small beads made of violet clay and glass. On the left forearm lay a corroded rectangular iron object, probably a tanged knife (fig. 2).

Under the left forearm, closer to the hand, a bronze plaque buckle, subrectangular in outline, was lying face down. It was fastened to a wooden base which had almost completely decayed. The buckle is 10.5 cm high, the width of its upper rounded part being 6.8 cm, and that of the lower trapezoid half, 5.7 cm (fig. 3). The central element of the scene represented on the plaque is a stylized tree with a feline carnivore hiding in its crown and apparently ready to jump (bunches of hair on its ears are suggestive of a lynx). The animal is shown en face in the upper part of the plaque, all the four paws clasping the branches and the trunk hidden in the foliage. In the lower part of the plaque, under the tree, a running mountain ram is represented. Its head with sharply twisted horns is shown en face in the middle of the trunk, which is rendered in profile; the legs are bent below and touch the lower frame of the picture. All the details are well elaborated.

It is for the first time that a plaque with this scene was found in an unlooted grave, thus precluding any doubts as to its Hsiung-nu attribution. Among the random finds from Siberia, there are two fragments of plaques with apparently similar representations. One was discovered near Izykh Mountain, Yenissey Valley (Minussinsk Museum, № 9094; Devlet 1980: table 28, 109), another belonged to Kossogol hoard, now at Krasnoyarsk Museum; Nashchokin 1967; Devlet 1980: table 28, 110). According to E. Banker, two similar plaques, both intact, apparently obtained by pirate excavations in Northern China, were exhibited at an antiquarian auction in Honkong. Their further fate is unknown.

The spectral analysis of the Dyrestuy plaque made by A. Kosolapov at Saint-Petersburg Hermitage showed that it had been manufactured of arsenious bronze with a minute admixture of other elements (the content of arsenic is about 11%, tin and lead making up less than 1%). As the previous analyses demonstrated, a similar composition was typical of Yenissey bronzes made after Hsiung-nu prototypes in the 2nd and 1st centuries BC (Minyaev 1983). The plaque fragment from the Kossogol hoard mentioned above (perhaps similar to the Dyrestuy specimen) was also made of arsenious bronze with about 5% antimony. The Izykh fragment, however, was cast of stannous-plumbic bronze, which was not characteristic of Minussinsk metallurgy of that period. This is just another indication that similar bronze artifacts were manufactured in various metallurgic centres and distributed over vast territories (Minyaev 1980).


Е.Я. Рогов, И.В. Тункина. Расписная и чернолаковая керамика из некрополя Панское I

E. Ya. Rogov, I. V. Tunkina. Painted and black—lacquered ceramics from the necropolis of Panskoye I

Painted and black-lacquered pottery excavated in 1969-86 from the necropolis of Panskoye I, northwestern Crimea, is described.

These vessels make up more than a half of the ceramic assemblage of the necropolis. Their total number is 68, some are intact and some fragmented; most were found in the graves. A few vessels were discovered among the artifacts that had remained after the burial feasts, and some were found either in or between the burial mounds.

Traditional terms common in the Russian and Western literature are used (Robinson 1950; Sparkles, Talcott 1970). Typological attributions were made with reference to Ivanov's classification (Apollonia 1963).

Most of the assemblage belongs to the mass production of Attic workshops. There are only few high-quality vessels: a kylix made by the "master of Jena" (№ 58), a krater of an unusual form (№ 48), an alabastron (№ 100), and an amphoriskos (№ 101). A few vessels were made not in Attic but in some other centres, which have not yet been localized.

The most numerous vessels are lekythoi (25 of them were found). Kylikes are on the second place (16 specimens). There are four one-handled bowls, three skyphoses and salt-cellars, two mugs, one oinochoe, one alabastron, one amphoriskos, one saucer, and one fish plate. Graffiti were found on five kylikes, six kantharoi, two one-handled bowls, and one salt-cellar.

Chronologically, the assemblage is rather uniform. The vast majority of the specimens dates from the first half of the 4th century B.C., and some from about 350-325 B.C. A few vessels (№№ 52, 87, 91, and 93) were made in the second half of the 5th century. They were apparently brought by the founders of the site. Only three vessels (№№ 82, 83, and 96) can be securely dated to late 4th - early 3rd century B.C. This is partly due to the fact that only few graves of that period have been excavated so far. Also, at the end of the 4th century significant changes in the burial rite occurred, burial goods became less abundant, and pottery was less often placed in the graves.


И.Н. Храпунов, В.П. Власов. Новое Кизил-Кобинское поселение в горном Крыму

I. N. Khrapunov, V. P. Vlasov. A New Kizil— Koba site in Highland Crimea

Kizil-Koba Culture was first described in 1926 by Bonch-Osmolovsky who associated it with the Taurians, the Crimean aborigines mentioned in Greek sources. Although sites representing this culture have been studied by several archaeologists including Schultz, Strzelecki, Kris, Leskov, Shchepinsky, and Kolotukhin, the available information is scarce, especially with regard to settlements. The present publication addresses materials from a recently discovered site studied by us.

The site is situated on the slope of Shpil Mountain, western Dolgorukov range, 0.5 km southeast of Druzhnoye, 18-20 km southeast of Simferopol, near the source of the Maly Salgir (fig. 1). This is not the only Kizil-Koba site known in the area. It was discovered in the course of excavations of a 3rd-4th century AD burial ground. Based on outward features alone it is impossible to determine its size. Within the excavation, which is more than 5000 sq m in surface area, 104 utility pits dug into the virgin soil were found. Beyond them, there are no habitation materials. No structures were revealed. All pits are circular in plan, and their section is barrel-shaped, bell-shaped, cup-shaped or cylindrical. Their diameter at the level of the subsoil is 1.0-1.7 m, and their depth ranges from 0.3 m to 1.3 m. One of them contained a burial of a child. Artifacts most commonly found in pits are fragments of hand-made vessels.

Vessels are of the following types: pots (figs. 2; 3: 1-5, 7, 8, 10, 11), ewers (fig. 4: 1-4, 6, 9), colanders (fig. 4: 8), bowls (fig. 4: 5, 7), cups (fig. 5: 1, 2, 4, 8, 10-14), beakers (fig. 5: 3, 5-7, 9), and scoops (fig. 3: 9, 12). One of the vessels resembles a Greek guttus (fig. 3: 6). Some vessels are decorated with rolls, and the rims occasionally have pits or knobs on them. Many cups, beakers, and ewers have a carved design (fig. 6).

The second group of pottery comprises Greek vessels, mostly made in Herakleia. Fragments of several dozens of amphorae were found in nine pits along with hand-made ceramics. Four vessels have englyphic stamps on their necks (fig. 7).

Figurines were poorly baked. Some were anthropomorphous (fig. 8: 9, 12), some zoomorphous (fig. 8: 6-8). The collection also includes globular (fig. 8: 10), biconic (fig. 8: 11), and elongated (fig. 8: 14) clay objects, as well as two conical objects with side openings (fig. 8: 13).

The tradition of making small clay artifacts likely originated in the forest-steppe zone of Ukraine and Moldavia in the Bronze Age and continued into the Early Iron Age. So far, clay figurines have not been found on other Kizil-Koba sites. However, poorly baked clay artifacts were excavated from Bronze Age settlements, as well as from Bosporan and late Scythian sites, and from a sanctuary near Yalta. All of them were made by people who practiced a developed agriculture and animal breeding. In cases where these artifacts belonged to assemblages, they were associated with sanctuaries, hearths, or stoves. All specialists agree that they were used for ritual purposes. This apparently also applies to clay artifacts made by Taurians who, as most modern writers believe, were represented by Kizil-Koba Culture (Khrapunov 1993). It may be speculated that they were used in the hearth and fertility cults.

Other finds from the pits include three bronze arrowheads (fig. 8: 1-3), several clay spindlewhorls, and flint tools.

Shpil is unique in having yielded more fragments of Greek amphorae than have all other Kizil-Koba sites taken together. Generally, the vessels date from 4th century BC. However, arrowheads, which were found in other utility pits, are earlier (late 7th—early 6th century BC, according to Melyukova's typological scheme). The most natural explanation, that the site existed from the 6th to the 4th centuries, is not supported by horizontal stratigraphy. We suggest that the date of this site be established primarily on the basis of fragments of amphorae, which are quite numerous and are distributed throughout the excavated area.

In terms of the construction of the utility pits and finds from their fill, mainly hand-made ceramics, Shpil is a typical Kizil-Koba site. Its uniqueness is due to an unusually large number of amphorae and the presence of anthropomorphous, zoomorphous, and other poorly baked clay artifacts.


Е.В. Власова, С.Л. Соловьев. Новые исследования хоры и некрополя Нимфея 1992-1995 гг.

E. V. Vlasova, S. L. Solovyov. New research of chora and necropolis of Nymphaeum: 1992-1995

The article is devoted to recent research of Nymphaeum - the Ionian colony, which was located on the western coast of the Kerch Strait, 17 kilometres south-west of Panticapaeum, the capital of the Bosporus state. The archaeological study of Nymphaeum has been going on for more than 100 years. However, according to a longheld tradition in classical archaeology, excavations of the Nymphaeum city centre have far surpassed in scale and volume any information gathered about its outlying areas.

Since 1992, the study of the outlying sites of the Nymphaeum city-centre has included the recently created Archaeological mission of the State Hermitage Museum. The basic goals of the mission was projected the archaeological map of the region, the boundaries and the spatial organisation of the outlying areas, typology of the population centres on the chora, studying the culture of the rural population. From the last year the research of Nymphaeum necropolis has become one of the topics of the project.

First steps toward successfully realising the project have already been taken. Large sections of the outlying areas along the lakes' coast and two ancient riverbeds, which extended toward each other, have been subjected to thorough examination during two years of archaeological study. In antiquity these riverbed channels possibly served as a single western boundary for the inner Nymphaeum chora.

In the ancient period two rather large population centres were located at the mouths of these river-bed channels. They probably controlled distinct regions of the western outlying areas of Nymphaeum. In the south-western sector of the chora such a role was apparently played by the settlement Ogon'ki 1, about which we unfortunately know very little. Then, alongside the settlement was discovered the burial mounds "Three Brothers", which is widely known among scholars owing to its rich and unique finds.

We know far less about the other site, which was located in the north-western part of the inner chora of Nymphaeum. The first and only mention of this site was by N. P. Kondakov in a report on excavations carried out near Nymphaeum in 1876. The large size of this settlement, and the numerous traces of ancient activity found there, led the researcher to the erroneous conclusion that the site represented the remains of ancient Tiritake. After years of oblivion, this major population point of ancient times (which was located on both sides of the old channel) was recently revealed and examined anew. The site also received the codename Churubash 1. Probably this site played a vital role in the north-western frontier of the Nymphaeum citystate.

The largest group of rural settlements of Nymphaeum was found to the south of the ancient city itself, and was situated virtually on the seacoast or very close to it.

The Geroevka-2 site was the one of these settlements excavated by The Hermitage Archaeological mission during three years (1992-94). More than 600 sq. m of the ancient settlement were opened up. Excavations showed that permanent settlement of the area occurred not only during the Classical period, but continued until the early Middle Ages. Well-represented materials of both periods were living and household buildings, several burial complexes and numerous material finds as well. It is worth noting that discovering of such diverse archaeological materials, which reflect virtually all forms of everyday life, is a very rare event in the study of Northern Pontic Sea rural settlements.

Of all the living and domestic, complexes discovered at the site, five storage pits and two dugouts dated from earliest period. In the early period of life in the settlement, a burial mound was situated on the section between the riverbed and the living zone.

Peculiarities of topography and purpose of uncovered objects allow to sugges the following reasons concerning character of the investigated part of the site. Firstly, that it was the separate estate, belonging, probably, to one large family and consisting of one dwelling and domestic structures -storehouses and storage pits for storing grain. The inhabited zone, apparenth, was separated by a fence from arable lands. In case of sudden danger the vacant pan of protected territory of a settlement became probable by a place for the home cattle. Moreover, nearby located allied graves were under constant looking after and protection from desecrating.

Analysing the particular features of the material culture of the settlement, it should be noted first that its principles were completely atypical of ancient Greek way of life. Second, the specific weight of hand-made pottery is almost a whole order higher than that ordinarily found in ancient Greek sites of the time. Third, on the whole the burial rites cannot be called characteristic of Hellenic burial practices of the Late Classical epoch. The presence of bearers of such traditions was conditional on the dispersion of natives within that area.

The settlement Geroevka-2 was abandoned at the beginning of the last quarter of the 4th century BC. The long, subsequent abandonment of the settlement was apparently interrupted only around the turn of the 4th-5th centuries AD. On the slopes of the river-bed, people who came here built a farmstead. The farmstead probably had a square form, the sides of which measured 20 m. The living rooms of the farmstead were cut into the northern slope of the streambed, and the domestic areas were cut into the southern slope. The wall construction allows us to assume that a second floor was built over the northern set of rooms. Alongside this farmstead there most likely was a pen for livestock. Many finds have been revealed during the excavating of this building and parts of the pen. Most of these finds date from the early Middle Ages, and are fragments of amphorae of the 4th-6th centuries AD.

Since 1995 the Hermitage archaeological mission after twenty-years interruption renewed excavations on Nymphaeum's necropolis. It occupied a large area around 16 sq. km. During two years of our excavations on the area about 600 sq. m forty-two graves were discovered. More than half of them were already plundered. Burials date to the second half of the 6th century BC - first centuries AD. In the same time burial mounds were constructed on the necropolis. Their earthen embankments were strengthened by stone kerbs.

Most of burials were inhumations in pits of a different design and dimension. Only two graves for its type were cremation. Most bodies were laid supine with a head to the North or the East. Only some of them were flexed and were oriented otherwise.

Average age of buried people was from 30 to 50 years old, and three skeletons belonged to the children 11-12 years old. Both sexes were represented virtually in an equal measure. In the attitude of chronology the graves were divided on five groups.

In the beginning of Medieval epoch the territory, before engaged by necropolis, has become resettled by the people. They obviously were the representatives of the natives in the Northern Black Sea region. On the necropolis the dug-out dwelling and series of storage pits concerned to that time. In its filling a plenty of finds was discovered, among which fragments of hand-made vessels numerically prevailed. Dating these materials does not fall outside the limits of the 4th - 6th centuries BC.


В.Л. Янин. Две древнерусские вислые печати начала XI в.

V. L. Yanin. Two old Russian pendent seals

The seals, which evidenced the introduction of official documents, made their first large-scale appearance in Russia in mid-11th century, when the sons of Yaroslav the Wise were ruling in Kiev. The event is well documented by sphragistic data. Altogether 14 seals of Izyaslav-Dmitry Yaroslavich are known so far (Янин 1970, т. 1: 166-167; Сотникова 1987: 158, рис. 1; Коваленко, Молчанов 1992: 75-76, 1993: 208-211), nine of Svyatoslav-Nikolay Yaroslavich (Янин 1970: t. 1: 167-168; №. 10- 13; Сотникова 1987: 167, рис.; Коваленко, Молчанов 1993: 209-211, рис. 1-3; Коваленко, Куза, Орлов 1980: 281), one of Vyacheslav-Mercury Yaroslavich (Янин 1970, t. 1: 168, №. 14), and 13 of Vsevolod-Andrey Yaroslavich (Янин 1970: т. 1: 168-170, №. 15-22; 1975: 64- 66; Szemioth, Wasilewski 1969: 84-85, tab. IV; Коваленко 1982: 245- 246, рис. 1; Булкин В., Булкин Вас, Смирнов, Ратнер, 1979: 430-431). Only two seals have been preserved from the earlier period: one, found in Kiev in 1912, was issued during the rule of Svyatoslav Igorevich, who died in 972, another, found in Novgorod in 1953, dates from the times of Izyaslav Vladimirovich (died in 1001) (Янин 1970, t. 2: 166, №. 1-2; Молчанов 1988: 50-52). Early seals were exceptional, and so apparently were the deeds which they authenticated. The find of two more earliest seals, then, is all the more important.

In late June 1994, horizons dating from the 1st quarter of the 11th century were being excavated by P. G. Gaydukov in Troitsk-X area in Novgorod. In the previous field season, the work was finished at the lowest level of layer 14 at the depth of 280 cm. In the same layer, at the depth of 275 cm, a hoard of Western European coins was found, the latest of which were struck in the late 1020s or somewhat earlier. Consequently, the hoard was buried no later than the 1030s. Tier 25, with which the burial of the hoard is related, has a firmly established date of 1025-1055 (Гайдуков, Янин 1995: 26-27). Tier 15, excavated in the beginning of 1994 field season, can so far be broadly dated to the 1st third of the 11th century (a more precise estimate will be possible after the dendrochronological analysis has been made).

On 28 June, a large lead seal was found in square 1131 at the depth of 293-300 cm. Its diameter is 32-36 mm, the fields are smaller (23-25 mm), and their edges are marked with beaded circles (fig. 1). One side bears a half-length representation of St. George with a spear at his right shoulder and a shield in his left hand. His head is surrounded with a beaded nimbus. The manner in which the young curly-headed and beardless saint is depicted is quite standard. On his sides, there are inscriptions, arranged in a columnwise fashion.

The other side of the seal is quite remarkable. It bears a half-length representation of a man dressed in a coat which is fastened with a massive round fibula near the right shoulder; he wears a high pointed helmet with a bead on top. The character is young (judging by the lack of beard) and has a pointed moustache. His name and title are indicated columnwise on his sides.

Both sides of the seal were beaten with some blunt tool, so not all letters are legible. The name Yaroslav, however, is seen quite clearly. The letter "O" is written separately on top of the column in a very elegant manner, making it similar to the traditional omicron in the word Ο ΑΓΙΟС, "Saint". The meaning of the inscription in the right column, too, is quite clear: "Prince of Russia", although some important parts of the inscription have been damaged.

The identification of the ruler represented on the seal, too, is quite easy. During the period to which, judging by the stratigraphic context, the find belonged, there was only one prince by the name of Yaroslav: the son of Vladimir Svyatoslavich, who christened Russia. Yaroslav's baptismal name was George, as evidenced by written records (Киевлянин 1850, т. III: 66; Соловьев 1988, кн. 1, тт. 1-2: 319-320), by the fact that he had founded St. George's monastery in Kiev (ПСРЛ 1908: стб. 139; 1926, т. 1, вып. 1: стб. 151), and by the coins that were struck in Novgorod during his rule and bear the inscription "Yaroslav's silver" and a representation of St. George (Сотникова, Спасский 1983: 196-198, №. 222-224) In his mature years he was referred to as Yaroslav the Wise.

The precise dating of the Novgorod seal is a more difficult matter. Yaroslav Vladimirovich lived a long life. He was the ruler of Novgorod until 1019, and from 1019 until his death in 1054 he "held the Russian land" in Kiev. To what period of his life, then, does the seal belong, and where was it made, in Novgorod or in Kiev?

The chronicles do not indicate Yaroslav's year of birth. What they do say is that he was Vladimir's fourth son (born after Vysheslav, Izyaslav, and Svyatopolk). It is also known that when Vladimir allotted principalities to his children, his eldest son Vysheslav received Novgorod, whereas Polotsk was alloted to Izyaslav, Turov to Svyatopolk, and Rostov to Yaroslav.

After Vysheslav's death, Novgorod was transferred to Yaroslav, while Boris assumed the rule in Rostov (НПЛ 1950: 159) after having recognized Svyatopolk's seniority (ПСРЛ 1926, т. 1, Вып. 1: стб. 132). Svyatopolk was born after 980 (ПСРЛ 1926, т. 1, вып. 1: стб. 78). Yaroslav died in 1054, allegedly at the age of 76 (НПЛ 1950: 182), suggesting that he was born about 978 or 979. Because the chronicle is rather inexact with regard to dates before 1000 AD, Yaroslav and Svyatopolk should be regarded as persons of roughly the same age who could both have aspired to the rule in Kiev after Vladimir's death. The chronicle does not give the year of Vysheslav's death or the beginning of Yaroslav's rule in Novgorod. However, with regard to Yaroslav's accession to the throne in Kiev in 1016, it is said that his rule in Novgorod had lasted for 28 years (НПЛ 1950: 142), implying that it began in 988 or 989. This, however, leaves no time for his rule in Rostov. We must evidently agree with S. M. Solovyev who hypothesized that 28 years were the duration of Yaroslav's rule in both northern cities, first in Rostov, and then in Novgorod (Solovyev 1988, кн. 1, тт. 1-2: 31 l). According to Tatishchev (1963: т. 2: 70), Vysheslav died in 1010. Nothing, however, supports this claim.

The first reliable evidence concerning Yaroslav's rule in Novgorod dates from 1014 when he refused to pay a traditional monetary tribute to Kiev. On 15 July next year, Vladimir died (shortly after having decided to set out for a punitive campaign against Novgorod). Svyatopolk, who had succeeded him on the throne in Kiev, murdered his brothers Boris, Gleb, and Svyatoslav, but was defeated by Yaroslav who led the Novgorodian troops. Although on 14 August 1017 Yaroslav had to flee to Novgorod having left Kiev to Svyatopolk and Boleslav, in 1019 he eventually recaptured Kiev with the help of the Novgorodians and ruled there until his death on 20 February 1054.

This does not mean that he never returned to Novgorod. A record under 1030 says that Yaroslav defeated Chud' (the Finnish-speaking tribes who lived on Novgorod's territory), founded the city of Yuryev (now Tartu), and promoted literacy in Novgorod (ПСРЛ 1851, t. 5: 136). In 1036, he again visited Novgorod and appointed his son Vladimir to rule there, but returned to Kiev with the Novgorodian troops after having learnt that Kiev had been besieged by the Pechenegs (ПСРЛ 1926, т. 1, вып. 1: стб. 150-151).

Yaroslav's seal could thus have been used to authenticate a deed during any period of his rule, no matter whether he reigned in Kiev or in Novgorod. For someone in Novgorod to receive a deed with his seal, it was by no means necessary for the prince himself to visit the city.

The date of the seal, then, may be specified only on the basis of stylistic characteristics of the seal itself. Its principal features are, first, that there are two representation, that of St. George and that of the prince himself, and second, that the saint on the seal is shown in the same way as on the Novgorod coins struck in the times of Yaroslav (fig. 2).

It has already been attempted to find the prototype of "Yaroslav's silver". Kunik (1860) was the first to observe that the representation of St. George on these coins could not have been borrowed from the Byzantine coins, on which the figure of the saint appears only in the 12th century. Kunik (1860: 114) suggested that the likely prototype was Yaroslav's seal which probably represented his patron saint.

Kunik's hypothesis remained speculative until the publication of the monograph on Byzantine sphragistics (Schlumberger 1884). Using this study, Tolstoy (1891, т. 2: 80-81) pointed to numerous parallels to the representation of St. George on Novgorodian coins among the 10th and 11th century Byzantine bulls. As to the absence of Yaroslav's seal, Tolstoy made the following prediction: "It can hardly be doubted that Yaroslav the Wise did have a seal; the established fact that the type of his coin had been copied from a seal, confirms this and allows us to reconstruct the appearance of that seal with some degree of certainty. A zealous inculcator of the Byzantine culture, as was Yaroslav, judging by the chronicles, should have possessed a seal complying with a Byzantine standard and taste. As the parallels suggest, this seal could have had a half-length representation of St. George on the face side and an inscription, or its Russian translation, on the reverse side (Толстой 1891, т. 2: 80-81).

At first sight, Tolstoy's prediction came true in a most brilliant way in 1962 when a seal of Yaroslav's son Izyaslav-Dimitry was found in Novgorod. It was used at the time when he was the ruler of the city (1052-54) (fig. 3). The seal definitely imitates "Yaroslav's silver" since it bears the representation of St. Dimitry, the divine patron of the owner, on the face side, and the prince's emblem encircled by a Greek formula meaning a wish of good luck on the reverse side (Янин 1970, т. 1: 166, No. 3).

In the light of Kunik's and Tolstoy's views, one might expect that the Novgorodian seal of Yaroslav would have St. George's representation on one side and the prince's emblem on the other.

However, this expectation is not upheld by the newly-found seal. While Izyaslav's bull directly follows the Novgorodian tradition of "Yaroslav's silver", Yaroslav's seal falls in a different context.

Tolstoy (1893) has drawn numerous parallels between the coins issued in the times of Vladimir Svyatoslavich and Svyatopolk the Cursed and those struck in Byzanty in the 10th and early 11th century. The latter type formed in mid-9th century during the rule of Michael III and existed uninterruptedly until the 1070s (Zacos, Veglery 1972, vol. 1: 56-91; Лихачев 1991: табл. LXXVI, 8-10, 13-15, табл. LXXVII, 1-4). However, as the Byzantine sphragistic data (which became available after Tolstoy's work was published) clearly indicate, the composition of the first Russian coins (Vladimir's "zlatniki" and "srebrenniki" of the I type) had derived not from the Byzantine coins, but from the 10th century Byzantine imperial bulls which bear the representation of the emperor with insignia on the obverse side and a half-length figure of Christ on the reverse (fig. 4). This is precisely how the first coins of Vladimir Svyatoslavich looked: the prince's portrait on the face side, and the half-length figure of Christ on the reverse (fig. 5).

Moreover, while during most of the 10th century, the emperor was represented with an orb in his right hand, in the times of Basil II and Constantine VIII (976-1025) the orb was replaced with a cross on a long staff. This is exactly the type of cross we see in the prince's right hand on Vladimir's zlatniki and srebrenniki of the I type (Сотникова, Спасский 1983: 115-139, № 1-51). Such coins never co-occur in hoards with the other three types of srebrenniki (those of Vladimir and Svyatopolk), whereas the remaining coins mentioned above circulated jointly. Hence it follows that Vladimir's coins of the I type are much earlier than the other srebrenniki and had gone out of use by the time when the latter were issued. If, given the presence of Christian symbols on all the Old Russian coins, those of the I type are dated to the first years following Russia's christening, then Vladimir's srebrenniki of the other types should be dated to the last years of his rule, as suggested by their joint circulation with the coins of Svyatopolk (Янин 1956: 168-169).

At that stage, the appearance of the Russian coins changed. Christ's portrait was replaced by the prince's emblem which had moved into the centre of the field, whereas on Vladimir's coins of type I it was modestly placed above the prince's shoulder. The second notable feature is the appearance of a nimbus around the prince's head. Although, as demonstrated by Stephani (1863), the nimbus could have symbolized not only holiness but supreme power as well, it is hard to escape the impression that what we observe here is a either a manifestation of a naive slyness or, more likely, an intentional device emphasizing the prince's exceptional role in the new religious system.

One more feature should be mentioned: the intraduction of the idea of a heavenly patronate, as evidenced by the representation of St. George on the Novgorodian coins of Yaroslav and that of St. Peter on two types of srebrenniki struck in the times of Svyatopolk (Сотникова, Спасский 1983: 191-196, № 206-218)(fig. 6), and by the supplementation of the traditional legend with the words "Saint Basil's" on one variation of Vladimir's srebrennik of the IV type, the latest among his issues (Сотникова, Спасский 1983: 179-180, № 175).

If Yaroslav's seal found in Novgorod is seen in this context, a number of important features emerges which are more indicative of the Kiev tradition than of the Novgorodian one. First, the prince himself is represented, in line with similar representations on the coins of Vladimir and Svyatopolk. Second, Yaroslav's name is written with a separate "O" alluding, as it were, to the nimbus on the portraits of Vladimir and Svyatopolk. Third, while the Novgorodian seal of Izyaslav-Dimitry Yaroslavich follows the "Novgorodian" tradition of "Yaroslav's silver", the seals of Svyatoslav-Nikolay Yaroslavich, which were issued in Chernigov and Kiev and bear both the prince's portrait (without the nimbus) and that of St. Nicholas (fig. 7) (Янин 1970:, т. 1: 167-168, № 10-13), are basically similar to the bull of Yaroslav Vladimirovich, thus speaking in favour of its relationship with the Kiev tradition.

Apparently the new seal was made at the time of Yaroslav's final ascension to the throne in Kiev in 1019, when he awarded ten grivnas to each Novgorodian who helped him to defeat Svyatopolk (НПЛ: 175-176). Perhaps the rewards were not restricted to this sum, and something else was granted. The awards must have been confirmed by documents which were authenticated with seals such as the one described here. Notably, it was found on the territory of an estate that in the 12th century was owned by a large boyar clan many of whose members were prominent persons and even elected posadniks (governors) (Янин, Зализняк 1993: 6-15). The fact that Yaroslav's seal was found there suggests that already in the early 11th century the estate belonged to the representatives of the city elite.

The find is remarkable in the cultural aspect as well. While Vladimir's and Svyatopolk's portraits (admittedly schematic ones, but still conveying some of their distinctive features) on their coins have long been familiar to the scholars, not a single representation of Yaroslav the Wise made during his lifetime was available. Such a portrait was doubtless present in St. Sophia's cathedral in Kiev where all his family was represented. However, this fresco has been partly destroyed, and the remaining part shows female family members only (Каргер 1954: 143-180). Pisarenko (1994: 13-17) has suggested, without being able to prove it, that the central character in the bear hunting scene painted in the staircase tower of St. Sophia is none other than Yaroslav. Finally, Gerasimov (1940: 72-76) has made a sculptural reconstruction using Yaroslav' skull, and his work has become widely known.

The portrait on the seal bears no resemblance to Gerasimov's reconstruction. This is not surprising since the anthropologist strove to create an image of a sage who died in his eighth decade. Signs of old age are accentuated on the reconstruction; the hair and beard leave too little space to compare the face with that on the seal, which shows a young man whose bellicose bristling moustache is certainly true to life. Apart from the final result, however, Gerasimov had depicted the successive stages of the reconstruction process, the most important being the one where the head had been completed but not yet supplied with hair. The published photograph of that stage (Герасимов 1940: 74, рис. 18г) conveys the impression of a tough rapacious face with a hooked nose, a personality which, at least in psychological terms, is rather similar to the man shown on the seal. It would be interesting to try and superimpose both representations.

The photograph of the second bull, which is also being published for the first time, was provided by P. P. Tolochko, whom I owe my sincere gratitude. The leaden seal, 27-32 mm in diameter (fig. 8), was found in the Kiev citadel during the excavations of 1934-37. Like the other finds from these excavations, it had been unavailable to the scholars ever since F. N. Molchanovsky, who headed the project, was arrested on political charges in 1937 and all his scientific materials were confiscated. Recently, one coin found by Molchanovsky (Vladimir's srebrennik of type I) was published (Сотникова, Спасский 1983: 129, № 19-2).

The seal bears a half-length representation of a man in a scale corselet. There is a crown decorated with a cross on his head, a nimbus around it, and an orb in his right hand. On his right side there is a vertical inscription which is illegible. The field is enclosed by a beaded circle. On the reverse side, there is a half-length representation of St. Peter making a blessing gesture with his right hand. On his sides there are inscriptions, arranged columnwise. The left one is legible.

The typical features of the bull are quite distinct. Clearly, the representation of man with insignia imitates that on the Byzantine imperial seals, the closest ones being those of Constantine VII (944-959) and Basil II (976-1015) (Zacos, Veglery 1972: Nos. 70, 71, 75). However, unlike his prototypes, the man on the Kiev seal has a nimbus, in accord with the Russian sphragistic tradition. He should, then, be regarded as the prince who possessed the moulds in which the bull was cast.

The prince's baptismal name, Peter, is alluded to by the representation of his patron saint on the reverse side. The same name, as was mentioned above, is present in apparently the same quality on two types of Svyatopolk's srebrenniki, implying that his baptismal name was Peter. Only one more 11th century prince is known to have had the same baptismal name: Yaropolk Izyaslavich, the grandson of

Yaroslav the Wise. However, he ruled in the 1070s and 1080s, long after this sphragistic type had been replaced by new ones (Янин 1963: 142-164).

The seal of Svyatopolk the Cursed from the Kiev citadel provides the closest parallel to Yaroslav's seal found in Novgorod. Both bear portraits of the owner of the bull and his holy patron. Both were made during a short time interval, and each can thus be used for crosschecking the conclusions based on its counterpart.


М.В. Шорин. Культовые камни Приильменья (по материалам Новгородской области)

M. V. Shorin. The sacred rocks of the Novgorod region

The article focuses on one of the least studied categories of archaeological monuments: the sacred rocks. Because the cult of rocks was part of many religious systems, these objects are a valuable source of information concerning ancient rites. Their study may shed light upon the old Slavic ritualism and its relationship with pre-Christian mythology as well as with the Christian ideology.

Sacred rocks occur all over the forest zone of European Russia and also in Belorussia and in the Baltic states. The present author has collected information on more than 120 sacred rocks of the Novgorod Region (fig. 1). Most of them are situated near the water sources (lakes, rivers, streams, and springs). Sometimes springs and trees in the vicinity of the rocks are also worshipped. One of the examples is a complex situated near the village of Gory. Apart from the rock, above which a chapel has been constructed, it includes a spring and a stone cross.

The rocks described in this article are large granite boulders of glacial origin. Their natural surface is smooth, but in most cases it is covered with carved signs. Some natural depressions are interpreted by the local people as tracks of animals or humans. Artificial anthropomorphous signs are carved representations of footprints or handprints (fig. 5). Artificial depressions, cylindrical (fig. 3), hemispherical (fig. 10), rectangular, or irregular in shape (fig. 2), are common. Some rocks bear carved representations of four-pointed or six-pointed crosses or signs shaped like a horseshoe.

Especially important is the rock known as Shcheglets (fig. 4). A. A. Formozov, who has studied unique carvings on it, finds their nearest parallels among the Bronze Age rock art of Sweden (Formozov 1965: 130-139).

According to the nature of representations, the sacred rocks can be classified into the following groups:

(1) Rocks bearing representations of footprints and handprints (60);

(2) Rocks with artificial depressions (22);

(3) Rocks with crosses (12);

(4) Rocks without carvings (23).

It is evidently not enough to register the size of the rocks. The adjacent space, too, must be studied. So far, in the Novgorod Region, these studies have been very limited in scope. In two instances, ash and charred stones were found near the sacred rocks, and in one of these cases, a pavement made of tightly adjusted slabs was discovered. Among the ash, fragments of vessels dating from the 16th century were found (Mil'kov 1982: 10-12, 18-19). Similar finds were made during the excavations near the sacred rocks in the Baltic area, specifically in Lithuania where the assemblages associated with the sacred rocks (pits with charcoal and ash, and stone pavements) date back to the 16th - 18th centuries (Urbanavicius 1972: 83; 1977: 79-89). It may be suggested that the rocks belonged to complex ritual associations, and that the making of fires was part of the ritual.

The fact that artifacts associated with the rocks are late does not in the least imply that the cult itself is recent. Far more likely, the rocks have been worshipped since much earlier times, but nothing has been left from the rites performed around them. In fact, the ground around some rocks with carvings is absolutely sterile.

This makes it necessary to look for a broader archaeological context in which the sacred rocks could be placed. So far, our attempts at finding such a context have failed. Because in some regions only late medieval sites are known, it is only possible to suggest that the cult existed in that period (Shorin 1986. 1986a, 1991).

It would be very important to link the cult of rocks with folklore and ethnographic evidence. Field studies have shown that the rocks were still being worshipped in the early 20th century. In some places, the cult still exists. "Footprints" on the rocks are being attributed to Christ, Virgin Mary, Paraskeue (Friday), or some local saint. The rock itself, as well as the water taken from the depressions in it, were believed to have miraculous healing properties. A tradition of offerings existed (money, food, wool, etc. were being left near the rocks). In rare instances, rites having little if anything to do with Christianity were performed nearby.

While some rocks are still regarded as sacred, there are many others which are no longer worshipped. No direct evidence suggesting that they were worshipped in the past is available. Near some rocks, festivals were held. On the nights before Ivan Kupala's (St. John the Baptist's day, a festival very popular with Russian peasants), the flattened top of a huge boulder near the village of Kamen' provided enough space for twelve pairs of dancers (fig. 7). These festivals apparently derived from pre-Christian rituals.

However, many rocks bearing depressions, "footprints" and "handprints" are believed to be "evil". According to traditional beliefs, they are haunted by evil spirits, and are referred to as Devil's rocks.

Such a dual attitude is explained by the complex history of the cult. Over many centuries, both the Orthodox church and the state attempted to eradicate remnants of pagan beliefs and objects associated with them. In places where these attempts had succeeded, the ritual practice was interrupted and the rocks were regarded as "evil". In other places, however, the ancient tradition continued and gradually merged with the Christian custom of worshipping holy places. Pre-Christian beliefs were transformed, but the Christian ones, too, were reinterpreted. Eventually, a specific brand of "folk Orthodoxy" emerged.

With regard to ancient beliefs, one should recollect the central Indo-European myth of the fight between the Thunderer and his adversary. Several Russian writers (V. V. Ivanov, V. N. Toporov, and B. A. Uspensky) have reconstructed this myth and traced its later transformation.

Several important features make it possible to link some of the sacred rocks with the cult of god Veles. They include the following: the rocks are related to the animal cult (many of them bear "animal tracks", and were believed to protect animals from evil spirits); also, they are associated with material wealth (cf. tales of treasures), healing, blindness as a punishment and the recovery of sight; they are often situated close to the water sources; and finally, some were called Volos (Veles), Elesina (Our Lady), Wet Rock, etc.

Especially noteworthy is a group of rocks named Devil's Rocks or Serpent's Rocks. These are known in Russia, Lithuania, Latvia, and Belorussia. The legends associated with them are common over the entire Balto-Slavic area and describe a conflict between Devil, or Serpent, on the one hand, and Perkunas, Perun, or Elijah, on the other.

Another group includes rocks known as "Blue". It has been demonstrated that this epithet does not refer to an actual colour. Rather, it is a symbolic attribute denoting relationship with the other world and death. This relationship is also inherent in Veles (Shorin 1988: 94-97).

Another important observation is that only left feet and hands are represented. The left side, too, was associated with the other world and with Veles who, being the god of the dead, was invisible for the living. His presence (and thereby his strength) was manifested only through his footprints. In the pre-Christian times, then, the sacred rocks were personified as Veles.

It may be suggested that rocks were worshipped in the Northwest from times immemorial. After the Slavs had colonized the region, the Veles cult was added to the ancient beliefs, and after the advent of Christianity it was transformed and became an integral part of folk Orthodoxy.


А.Р. Артемьев. Скрамасакс из псковского некрополя X — начала XI вв.

R. Artem'ev. 10th—early 11th centuries scramasax from the necropolis of Pskov

In summer 1985, excavations carried out by the Pskov archaeological project headed by V. V. Sedov, revealed a battle knife of the scramasax type (fig. 1: 1) in trench VI, in a horison overlying the subsoil, near burials Nos. 42 and 42a of the city necropolis. Apparently the knife belonged to the peripheral part of the burial mound which had been constructed above these burials and was later covered by the 12th - 20th century cultural layers. Because nothing except cremated remains was discovered in either of the two burials, the date of the mound is evidently the same as that of the entire cemetery: 10th - early 11th century.

The total length of the scramasax is 37.5 cm, the length of the blade, 28 cm, and its width, 2.4 cm (the point of the tang is bent). The specimen, then, does not fully comply with the criteria outlined by J. Petersen who believed that the length of the scramasax should be no less than 30 cm while the blade must be at least 3 cm wide. Nevertheless no doubts have been expressed as to the military attribution of the knives whose blade width is below 3 cm. According to P.-L. Lehtosalo-Hilander, the standards that determined the dimensions of the scramasaxes in Western Scandinavia were different from those in the Baltic area. The narrow-bladed knives are especially numerous in Finland which was probably the core area of their distribution. Also, they were found in Sweden (Gotland, Birka), Latvia, and Estonia.

Altogether thirteen 10th-century scramasaxes have so far been found in Russia. One more was discovered in the necropolis of Chersonesus in a grave which was undoubtedly that of a Russian warrior. Two specimens come from the burial mounds of the southeastern Ladoga area, two from Yaroslavl' area, the Volga basin (fig. 1: 4), three from Gnezdovo (fig. 1: 3), and five from Shestovitsy (fig. 1: 2).

Clearly, scramasaxes of the Finno-Scandinavian type were seldom if ever used by the Russian warriors. A list provided by Kirpichnikov (1966) contains information on 109 9th - early 11th century swords, 290 10th - 11th century spears and 211 axes, but only 13 10th-century scramasaxes. All of the latter, except those from Pskov and Chersonesus, come either from burial mounds or from regions with the highest frequency of Scandinavian burials in Russia (fig. 2). According to Stalsberg, the frequency of Scandinavian burials may range from 3.7% to 8.4% in Mikhailovsky burial ground, from 1.3% to 8.5% in Bol'shoy Timerevsky, and from 2.5% to 7.7% in Gnezdovo. Avdusin believes that more than 100 burials in Gnezdovo are Scandinavian Bliefeld has claimed that about 20 of 147 burial mounds in Shestovitsy published by him (i.e. 13%) are Scandinavian. Importantly, they comprise four out of five mounds in which scramasaxes were found. Burials with scramasaxes in Zalyushchik and Gorka Nikol'skaya, Southeastern Ladoga area, are doubtless Scandinavian. In the Pskov necropolis, where the number of burials excavated by 1991 had amounted to 65, Scandinavian graves are also present, and it is likely that the scramasax marks one of them.

Kirpichnikov is probably right when he claims that in 11th and 12th century Russia, most battle knives did not differ from the usual domestic ones. However, in our opinion, it would be premature to assert that knives especially destined for military purposes were few. Among the materials from the northern Russian towns, there is a series of narrow-bladed knives whose average length is 25 cm; details of their sheaths have also been found. Because it may hardly be assumed that they were used for domestic activities, they must have been battle knives. This group will be dealt with in a separate article.


С.Л. Санкина. О скандинавском присутствии на Русском Севере могильник Куреваниха-2 (к историографии норманской проблемы в антропологии)

S. L. Sankina. On the biological evidence of the Scandinavian presence in Northern Russia (with special reference to Kurevanikha—2 burial ground)

Although Scandinavian artifacts and even assemblages have been found in many burial grounds on the Old Russian territory, the skeletal series representing the Norse immigrants are very few. Most Scandinavian artifacts date from the 9th to the 11th century. Many writers believe that Russia's intense contacts with the Vikings did not extend beyond these chronological boundaries.

At present, medieval cranial series suggestive of the Western European (hence, Norse) presence are known from three burial grounds: Shestovitsy near Chernigov, Old Ladoga, and Kurevahikha-2 on the Mologa. Scandinavian artifacts, however, were found in one of these places only (Shestovitsy).

Cranially, most Germanic-speaking populations are very peculiar and can be easily distinguished from the Eastern Slavs. Most series representing the Germanic speakers are characterized by a combination of a long and low cranial vault, high face, high orbits, and a relatively narrow nose. This combination was found in all Scandinavian series.

The Baits as well as the populations inhabiting the western part of Old Russia and some Finnish speaking peoples (Estonians, Karelians, Komi, and Vepses) have high brain cases and low orbits. This makes them quite distinct from the Germanic populations. Other Finns, most populations of Old Russia, Southern Slavs, and most Western Slavs are intermediate and also different from the speakers of Germanic languages.

T. I. Alexeyeva (1973) has introduced four indexes aimed at differentiating the Germanic populations from their eastern neighbours. These include: the ratio between cranial height (Martin 17) and the unweighted average of cranial length (1) and breadth (8); the vertical facio-cerebral index (48: 17); the relationship between orbital height (52) and cranial height (17); and the ratio of nasal breadth (54) to facial breadth (45). These indexes set the Germanic and Eastern Slavic groups apart almost without overlap. Norse series from the territory of Old Russia fall within the Germanic range of variation (Table 1). However, they have a slightly higher relative nasal breadth, possibly due to incipient admixture.

In the 8th — 11th centuries, the most intense contacts of the Old Russian population with the Vikings took place in the area southeast of Lake Ladoga. Old Ladoga was the outpost of the Slavic colonization in the northern region. Its population consisted of various ethnic groups. Judging by the archaeological data, the Scandinavians who lived in Ladoga buried their dead in Plakun. Most of these burials, however, are cremations. Some 11th-12th century mounds on the northern cemetery of Ladoga have also been excavated, but the skeletal material has not been preserved. Contemporaneous skeletons excavated by V. I. Ravdonikas and G. P. Grozdilov from the ground burials at Zemlyanoe Gorodishche are the only early group from Old Ladoga available for osteometric studies.

Based on archaeological criteria, it is impossible to determine the ethnic attribution of this group since the burials were made according to the Christian rite and contained no artifacts. The skulls, however, exhibit the whole complex of "Germanic" traits. Also, these people, unlike Russians or Finns, were high-statured and had sharply profiled faces.

Because many of the deceased were children, the cemetery must have belonged to the permanent residents of Old Ladoga. Using stratigraphical data, two subgroups, an earlier one and a later one, were separated. Significant differences between them were found. While the earlier subgroup consisted of young tall males, the later one was more diverse in terms of demography. Cranially, being still "Germanic" in appearance, it deviates from the earlier subgroup in the direction of the Russian complex. Although both subgroups fall within the Scandinavian range of variation, it may be suggested that hybridization did occur.

Thanks to the works of the Northern Archaeological Expedition, one more series of skulls apparently representing Norse immigrants is now available. It was excavated by A. N. Bashen'kin in 1990 from the burial mounds near Kurevanikha in the Mologa basin (Kurevanikha-2). The burials, dating from late 12th — early 13th centuries, were made according to the Christian rite and, like those in Old Ladoga, contained almost no artifacts. In 1966, A. Nikitin excavated another group of mounds in the same area. The burials contained 10th — 11th century Scandinavian assemblages; the skeletons have not been preserved.

The series from Kurevanikha-2 consists of only five well-preserved male skulls (Table 2). Having low vaults and high faces, this group is quite distinct from the local populations, both Russian and Finnish (Table 3). The brain case, however, is rather short, in contrast to the Germanic series. But, given that the group is rather late, the brachycephalization phenomenon should be taken into account, although some admixture is also possible.

People from Kurevanikha, like those from Old Ladoga but unlike most Russians and Finns, had sharply profiled faces and high nasal bridges. The nasal protrusion angle, however, is not large, which is a local trait.

According to the results of the multivariate statistical analysis, the Kurevanikha group clearly falls into the "Scandinavian" cluster (fig. 1). The nearest parallels to this group, as well as to that from Old Ladoga, are series from Sweden and Iceland. The Shestovitsy group, too, joins the Scandinavian cluster and displays some similarity (not a close one, though) to series from Sweden and Britain.

The dates of the three morphologically Norse groups from Russia are different. Shestovitsy dates from the 10th century, the time when the Vikings actively interacted with the Slavs. Old Ladoga is later (11th — 12th centuries, the final stage of these contacts). Kurevanikha is the latest (12th — 13th centuries, the period when the direct influence of the Vikings was no longer felt). Each of the three shows some traces of hybridization with the local groups. This process, however, had not proceeded far enough to obliterate the characteristic biological features of the Norse settlers, and this applies even to the latest group. All the above suggests that the Norse colonies in Old Russia were rather stable, at least in terms of biological distinctness.




В.А. Алёкшин. Неандерталь Крапина, Монте Чирчео. Ритуалы в среднем палеолите

V. A. Alekshin. Neandertal, Krapina and Monte Circeo. Rituals in the Middle Palaeolithic

In August 1856, workers who were clearing from clay the Feldhofer rock shelter at Neandertal near Dusseldorf, Germany, struck on human bones, which they threw outside together with the clay. Later the bones got into the hands of the Protestant pastor K. Fuhlrott, who published the find (Fuhlrott 1859). The workers told him that the bones, situated in anatomical order, lay at the depth of 60 cm from the surface in a horizontal position, along the axis of the shelter, and the skull was oriented toward the entrance, which faced north. Schaafhausen (1888: 7-8) was apparently the first to suggest that the shelter contained a burial of a Neandertal male. This interpretation may be correct since, apart from the skull, many postcranial bones were preserved. The same opinion was expressed by F. May (1986: 17). Contrary to Smirnov's view (Смирнов 1991: 53), the absence of habitation material in the shelter does not rule out the presence of a burial. The Regourdou burial was also made in a shelter which was not a site of long duration (Алёкшин 1995: 199).

One should hardly take at face value the testimony of workers who claimed that the bones at Feldhofer were found in anatomical order. Because there is evidence suggestive of post mortem manipulation (specifically, cuts were found on the skull and on the right scapula, see Czarnetski 1977: 215-218), it is possible that the body was dismembered and only its parts were buried. There is reason to believe that the skeleton was actually situated along the axis of the shelter, perpendicular to the entrance.

The deposits in the Feldhofer shelter supposedly date back to early Wurm (Gieseler 1971: 193). The fact that body was arranged perpendicular to the entrance supports this date. In Western Europe, this position was typical only of burials dating from the middle of Wurm II (60,000-50,000 BP). Burials situated parallel to the entrance evidently appear later (Алёкшин 1995: 207).

About 800 fragments of Neandertal bones, including damaged skulls, were found in Krapina, north of Zagreb, Croatia. While the number of individuals estimated on the basis of bone fragments is 20-30 (Trinkaus 1978: 156; Smith 1976: 285), the teeth belonged to at least 80 persons (Wolpoff 1979: 104). Because most bones are fragmented, Gorjanovic-Kramberger (1901: 196) believed that the site represented remains of a cannibalistic feast. This view has been widely cited. In 1970-s several anthropologists, who had examined the Krapina collection, came to the conclusion that the long bones were intentionally cracked, and some of them bore traces of defleshing by stone tools. Based on this evidence, they concluded that cannibalism in Krapina may be regarded as an established fact (Smith 1976: 287; Ullrich 1982: 258). Apparently they did not take into account that most Krapina remains are not charred. Only 6.8% of cranial bones and 0.5% of postcranial bones bear traces of fire.

Another hypothesis was put forward by E. Trinkaus. Because Krapina bones have been neither damaged by carnivores nor smashed, Trinkaus believes they come either from disturbed graves or from burials that contained dismembered bodies. Traces of fire cannot by themselves attest to cannibalism (Trinkaus 1985). Trinkaus's view seemed to have been upheld by M. Russell (1987, 1987a). According to her, the traces of stone tools on Krapina bones are similar to those seen on the bones of certain American aborigines whose skeletons were de-fleshed in accordance with traditional burial rites. Also, Russell claimed that there are no indications thai long bones from Krapina were smashed to extract bone marrow. Later, however, she revised her views and concluded that the evidence suggesting that Krapina skeletons were dismembered was not as conclusive as she previously believed. Evidence in favor of defleshment does exist, but it could as well be explained by cannibalism (Russell et al. 1991). This revision indicates that anthropologists, too, sometimes lack reliable criteria for interpreting traces on Neandertal bones. Attempts at solving the mystery on the basis of archaeological context are doomed to failure since excavations done with the help of dynamite should not be regarded as thoroughly documented. The issue, then, remains unresolved although Trinkaus's hypothesis does seem attractive and may even be considered as the most likely one at present, given that reburial has been registered at several Mousterian sites, such as La Ferrasie, Shanidar, Skhul, and Qafzeh.

F. Bordes, who dated Krapina to the Wurm glacial, believed that the industry of certain layers resembled the Mousterian of La Quina (Bordes 1981. 82). If so, the probable date would be that stage of Wurm interstadial for which dismembered incomplete Neandertal burials are typical (Алёкшин 1995: 207).

On 24 February 1939, during construction works, a skull of a Neandertal male was found in Guattari rock shelter, Monte Circeo, Italy. A. Guattari, who found the skull, saw it lying on the surface; the mandible was lacking. He could not help taking it out to have a closer look at it. Then he put it back in its original position, as he believed.

The archaeologist A. Blanc, who started excavating the shelter on the next day, reached several conclusions: (1) the skull was that of a male aged 40-50 who was killed by a blow that crushed the right temporal bone and the right orbit; (2) the foramen magnum was artificially enlarged; (3) the skull was placed in the shelter and a ring of stones was made around it (fig. 1A). According to Blanc, the find attested to ritual cannibalism in the Mousterian (Blanc 1939; Sergi 1938-39, 1974: 55; Tarli et al. 1991: 449-450).

The idea of cannibalism was not supported by other specialists, who believe that the skull was intentionally exposed (Schott 1979: 3; Смирнов 1991: 29, note 11).

Regrettably, it is impossible to establish the location of the skull within the shelter. Guattari could have made a mistake while putting the skull on a heap of stones.

Recent archaeological studies have demonstrated that the skull initially rested on the left side (fig. IB). No traces of a blow which might have caused death are discernible. Neither are there any marks suggestive of defleshment. The foramen magnum appears to be intact (Giacobini 1990-91; Tarli et al. 1991; Toth and White 1990-91; White and Toth 1991).

Some writers believe that the skull was damaged because it lay on the surface (Tarli et al. 1991). Others suggest that it might have been gnawed by hyenas, although they admit that no apparent traces of their teeth are present (Toth and White 1990-91; White and Toth 1991). The upper level of deposits does contain remains suggestive of a hyena den (Piperno, Giacobini 1990-91; Stiner 1991). However, since no marks left by teeth have been found on the skull, Mussi concluded that it was brought to the shelter when no soft tissues had remained on it, so it did not attract the hyenas' attention (Mussi 1988: 97-98).

Is it possible, then, that the skull was detached from the dissociated corpse and exposed for ritual purposes? In that case, defleshment would have left no traces (Tarli et al. 1991; Алёкшин 1994: 138). This interpretation sounds quite plausible. People of the Middle Paleolithic apparently used skulls in some rites. In three instances (Shanidar, Kebara, and Regourdou) skulls were extracted from the graves after the funeral (Алёкшин 1993: 6; 1994: 138; 1995: 201). All these skulls were those of adult males. The Guattari rock shelter, situated in a remote place which is difficult to access, could have been visited only occasionally and it could have served as a sanctuary. Unfortunately, by having taken the skull out of its original context, Guattari depreciated the find, the date of which is estimated at 57-51,000 BP (Schwartz, Cezar 1988: 13; Schwarcz et al. 1991: 313-316).


С.А. Васильев. Стадиализм в палеолитоведении 60 лет спустя

S. A. Vasil'ev. Stadialism and the palaeolithic studies: a glance after 60 years

It is the hallmark of contemporary archaeology that a special attention is given to the study of growth of archaeological knowledge, to the changing approaches in the investigation of human past, shaped by different internal and external factors. This general tendency is of fundamental importance for the developmental history of Russian archaeology which has not yet become the subject of study. The stadial concept of the Palaeolithic culture was pioneered by scholars from the State Academy for the Material Culture History in the early 1930-s. Tire role of Ravdonikas, one of the leaders of Russian prehistory of those days, in the formation of this theory is difficult to overestimate. It is fair to say that the stadial concept was generically linked to the late evolutionism. There are important differences, however. Stadialism emphatically introduced the historical orientation in archaeology, trying to interpret the developmental stages in sociological terms. The stadial concept dominated in Russian prehistorical studies of the 1930-s— 1950-s, when local periodization schemes were put forward for the Palaeolithic of Eastern Europe (Efimenko, Boriskovsky), Caucasus (Zamiatnin), and Siberia (Sosnovsky, Okladnikov).

In the 1960-s — 1980-s stadial theory was strongly challenged and even discredited by the local culture approach pioneered by Rogachev and supported by Grigor'ev. This new concept rejected stadialism as "sociological sketchiness". It was postulated that only through the study of local peculiar features of the Palaeolithic in different regions we could achieve a real progress in prehistoric explorations. However, later the emphasis on the identification of new local cultures diminished due to the number of factors: practical impossibility to delineate spatial and temporal boundaries of cultures, complex and subtle character of variability of lithic industries, etc.

How can one now evaluate the stadialism heritage, what aspects of this theory could be regarded as relevant to contemporary debates? I argue that in spite of a lot of outdated concepts (including the impact of Marxist and Marrist schemes), stadial approach pioneered some ideas of much current interest.

First and foremost, the stadial concept included the notion of the inner temporal structure in the Palaeolithic culture development. Current Russian Palaeolithic archaeology witnesses the renewal of the keen interest to the periodization of the Upper Palaeolithic, which was left aside during the last decades (c.f., for example, recent works of Grigor'ev). It is worthy of note that stadialism did not reject the local diversity of the "concrete variants" of stages as well as a need to study in detail the local evolutionary successions. Ravdonikas, following the early views of Efimenko, made the point, for example, that the local differentiation of types in the Upper Palaeolithic posed a problem of coexistence of different culture variants.

Furthermore, stadialists advanced for the first time the problem of culture change without population shifts. In spite of repeated attempts to renew migratory concepts, it seems impossible to view the Palaeolithic culture history as a persistent chain of migratory waves. The adherents of stadialism saw the main cause of culture change in subsistence advances, the diversification of the social structure patterns of prehistoric societies.

A strong tendency toward a global-scale comparative studies of the Palaeolithic is among the most attractive aspects of stadialism. This focus on worldwide culture processes obviously stimulated the study of large culture areas of the Upper Palaeolithic developed in Russian prehistory (Zamiatnin, Okladnikov, Grigor'ev).

Last but not least - it was within the stadial theory framework that the problem of functional variability between and within sites was put forward by Boriskovsky.

Thus in spite of a considerable progress achieved in Palaeolithic studies during the last decades, the stadial concept enriched prehistory by new ideas and hypotheses. The analysis of the Palaeolithic culture development as stages and intra-stage differentiation is very promising. I argue in agreement with Gladilin and Sitlivyi that both the stadial and local-culture approaches to the Palaeolithic may be regarded as complementary, not mutually exclusive, aspects of the process of culture development, like two sides of a coin.


В.М. Массон. Эпоха древнейших великих степных обществ

V. M. Masson. The epoch of the earliest great steppe societies

In the Bronze Age, societies had formed in the Eurasian steppes which, in genetical and typological terms, were precursors of the nomadic Scythian society. There is good reason to describe this period as the age of the earliest steppe societies. At present, three large periods can be seen in their development. The first one covers most of the 4th millennium ВС, having terminated around 2,600 ВС. The second one lasted approximately from 2,600 ВС until 1,700 ВС, and the third, from 1,700 to 900 ВС. The Russian archaeologists usually describe the first period as the Eneolithic. In the Western European system, the three periods are referred to as Bronze I, Bronze II, and Bronze III, respectively.

It is for the first period that the most important new evidence had been gained over the recent decades. In the Eneolithic, conditions were emerging for the subsequent formation of animal-breeding societies in the steppe zone. One of the main features of that period was the Maikop cultural impact on the Pontic steppes. In southern Ukraine, following the discovery of the lower level at Mikhailovka, a number of similar sites had been described as a separate culture. In the steppes of the Dnieper Basin, the Post-Mariupol' Culture was separated, and in the Volga area, the Khvalynsk Culture. In the northern Kazakhstanian steppes situated in the remote periphery of the steppe belt, the settlement of the first horse-breeders, Batay, had been discovered.

All the societies represented by these sites had inherited from the Late Stone Age a very important feature: stability of the economy largely based on fishing. Settlements and nomadic camps cluster along the banks, only a few penetrating into the open steppe. Clay used for making pottery was tempered with ground shell. Animal-breeding was gradually becoming the most important part of economy. At Derievka, a Sredniy Stog Culture site, nearly half of the bone remains are those of horses, and ritual burials of horse skulls were found at the site. Cheek pieces made of antler attest to the sophistication of horse controlling devices. In burial mounds of the Post-Mariupol' Culture, bones of bulls, including skulls, were found, and in graves of Khvalynsk burial ground, there were bones of bulls, horses, and sheep. At Batay site, northern Kazakhstan, 99.9% of the animal bones are those of horses. Several double-holed bone plates from Batay were examined at the Use-Wear Laboratory of IIMK and found to be primitive cheek-pieces, judging by the nature of the wear. Most likely, people who lived in northern Kazakhstan at that time controlled herds of horses grazing at large. The Batay society, then, can be best described as that of early horse-breeders and fishers.

Conditions had thus formed for a large-scale and steady opening up of the turfy and cereal Eurasian steppes; indeed, the process had essentially started. The society possessed a number of domestic animal breeds. Breeding was dependent on local conditions. The horse was turning into the principal means of conveyance, crucial for the steppe societies. The formation of steppe lifestyles was underway. Mounds were being erected above the graves, making the cemeteries visible in the open land. This marked the emergence of the burial mound ritual which played a central role in the life of the ancient Old World societies.

Close relationships between the steppe people and the highly developed societies of the adjacent regions, Tripolye in the west, pre-Maikop and Maikop in the southeast, were quite important. These were the areas from where early metallurgical techniques were spreading across the steppe. Two burials of bronze-founders, with moulds for casting socketed axes of northern Caucasian appearance (fig. 1), are known for the Post-Mariupol' Culture. Imported vessels of Tripolye and Maikop types, which played a prestigious role in ancient societies, are frequently found in the steppe zone. No such links have been registered for the Botay Culture, which was possibly the reason why it had declined without having given rise to any subsequent cultures.

In the second period, the great steppe societies in the proper sense emerge. The earliest one was related to the Pit Grave Culture. Judging by all the available evidence, the core area of the standard Pit Grave Culture was the Volga Basin. Its spread to the west was due both to migrations and to the fact that new cultural and behavioural standards were easily adopted by various tribes. Regional peculiarities were largely defined by a variety of local traditions which had flowed into the cultural mainstream.

In that period, all across the area from the Volga to the Dniester, groups of burial mounds spread as a symbol of the steppe societies. Animal-breeding economy varied according to local ecological niches. Vast steppe areas were being rapidly opened up. Numerous mounds attest to the increased population size and a longer duration of settlement. Distinctive features of the burial rite emerge which survive for a considerable time. They include putting a cart beside the grave or, less often, inside it, constructing stone and possibly wooden stelae above the burials and probably also on the mounds, and restoring the facial features of the dead using plaster masks. Heavy four-wheeled carts were an innovation especially important for the steppes. Fortified sites evidence military conflicts. The society represented by the Pit Grave Culture was apparently poor but dynamic. These people were actively advancing in the western direction, having penetrated into Moldavia, Romania, Bulgaria and eventually gone as far as Hungary.

Basically the same features were typical of another great society of the steppes: one represented by the Catacomb Culture. Its core area was evidently different from that of the Pit Grave Culture. However, the mosaic of its local varieties (most researchers prefer to speak of several Catacomb-type cultures) recalls the earlier pattern in that it combines local traditions, including those of the Pit Grave Culture, with the new standards of the burial rite and prestigious ceramics. Local metallurgy and metal-working was actively developing, and professional artisans were being buried with their tool kits. Carts became more sophisticated, and several of their varieties are known (fig. 2). Burials attest to the formation of the elite.

The distinctive features of the third period (1,700-900 ВС) were maximal mobility of the steppe tribes, followed by stagnation and decline. The process of cultural interactions continued in the steppe zone, cultures were splitting and new territorial associations with their specific peculiarities were emerging. Especially noteworthy was the outburst of cultural dynamics in the Volga-Ural area where progress in metallurgy and metal-working was especially striking. Stannic bronzes were used, metal artifacts were cast with the help of stone models, and the "blind socket" method of founding had been introduced.

In sociological terms, archaeological sites of the steppe zone evidence the development of structures typical of the early complex societies (fig. 3), with their military aristocracy and a well-organized labour force engaged in building well-designed settlements under a centralized supervision. All over the steppe region, burials of the noblemen (chariotry) are being found. Inside the graves there are numerous cheek-pieces and remains of horses themselves: complete skeletons and their parts (Sintashta, Utevka, and other mound groups in the Volga area). Oval fortresses like Arkaim in the southern Urals attest both to the military conflicts and to a centralized organization.

Political development takes a military-aristocratic course. Judging by the absence of leaders' burials, the steppe societies were ruled by oligarchies. The earliest evidence of the early complex society is related to the Pit Grave Culture. Under these conditions, the third great community of the steppes emerges: the Timber Grave Culture. Its bearers migrated from the Volga Basin in the western direction, having reached the right-bank Dnieper area and assimilated the local populations. The bronze metallurgy, stimulated by cultural influences from the Volga-Ural region, was flourishing. The bipartite social structure is retained, although the impressive tombs of the chariotry disappear.

While stability, which promoted population growth, was also characteristic of the period following the Timber Grave period, cultural differentiation of the steppe societies increased during that time. Carpatho-Balcanic metal trade develops as a specific form of Late Bronze Age import. Nothing, however, is suggestive of a development toward urbanization. Rural populations continue to follow local traditions quite different from urban standards. It may be concluded that the opening up of all the ecological niches available in the respective economic system and the achievement of stability and homeostasis were followed by cultural and social stagnation. Being deprived of their inherent dynamics, cultures gradually decline, but not simultaneously. The process might have been stimulated by natural cataclysms, although these could not be the only cause of the changes. The Bronze Age of the steppes terminates, and the era of the early nomads begins, marked by the appearance of the Cimmerians in the western regions.


Б.Я. Ставиский. "Кушанская проблема" и археология Средней Азии (1968-1993 гг.) (некоторые итоги)

J. Stawisky. The Kushanian problem and the archaeology of Central Asia, 1968-93: some results

More than a quarter of a century ago, from September 27 to October 5, 1968, an international conference on the history, archaeology, and culture of Central Asia during the Kushanian Period was held in Dushanbe. The event itself and the subsequent publication of its proceedings (Central Asia in the Age of the Kushans, 1974, 1975) gave a strong impetus to further studies, archaeological ones in particular.

General issues in Kushanian history.

Significance of the Kushanian period for the history of Central Asia. New information gained over the recent three decades has modified many views concerning the impact of the Kushanian Age on the subsequent history of Central Asian culture and art. New post-Kushanian sites have been discovered, including Kuyevkurgan, a 5th century castle or fortified estate. Its excavations have revealed a submural clay sculpture which, in terms of technology and, to a large extent, style and imagery, continues the Kushanian traditions (Annayev 1984, 1988; Culture and Art of Ancient Uzbekistan 1991: 26-30).

Excavations at Penjikent have disclosed 5th and 6th century archaeological assemblages containing new works of monumental art performed in the Kushanian manner, as well as representations of Buddhist and Sivaite characters also linking the early medieval Sogdian art with Kushanian traditions (Belenitsky 1973: 52-54; Marshak, Raspopova 1988: 142, 143; Skoda 1992, tab. 70: 2-71). Also, the architecture and monumental decoration of some ritual structures of the post-Kushanian period apparently reflect Buddhist canons of the Kushanian age. This applies both to 4th - 6th century Mervia and 6th - 8th century Tokharistan, whose Buddhist architecture derives from the Buddhist tradition of Kushanian Bactria, and to the period of AD 400-900 in the Chu River valley, an area where Buddhism was apparently introduced from Sinkiang (Goryacheva, Peregudova 1994).

Northern boundaries of the Kushanian state and its sphere of influence in Central Asia.

This is one of the key issues in the Kushanian archaeology. In his talk at the 1968 conference, D. Sirkar noted that available written sources contain no direct evidence suggesting that areas north of the Oxus had ever belonged to the Kushanian empire (Central Asia in the Kushanian Period 1975: 412, 413). Allchin (1957) and Fischer (1958) also believed that Central Asia was not a Kushanian province. Seymal (1978) apparently subscribed to that view when he used an artificial term "Tran-soxiana" to denote territories north of the Oxus.

In 1961, I suggested that the Kushanian state included northern (right-bank) Bactria, or northern Tokharistan (the name has been used with reference to northern Afghanistan and southern Central Asia since at least AD 300), as well as Sogd, Khorezm, and Chach (Shash). In my view, the question of whether or not Ferghana belonged to the Kushanians, could not be resolved because the evidence was insufficient (Stavisky 1961). M. Mas-son, however, believed that the Kushanian state (formerly Tokharistan) reached only as far north and east as the lower Kashka-Darya valley. Areas to the north and to the east of it allegedly belonged to Sogd and Kangyuy (M. Masson 1968, 1975). My idea that most of the Central Asian Interfluve was under the Kushanian rule was thus countered by a no less hypothetical statement concerning the powerful Kangyuy whose rule was thought to extend to the entire central and northern Central Asia including Chach, Khorezm, Ferghana, and Sogd. M. Masson claimed that the Kushanian state did not reach further north than Bactria- Tokharistan, only the lower Kashka-Darya area and a narrow strip along the middle Amu-Darya valley being included.

Apart from these two hypotheses, it was claimed that the Kushanian state did not include Ferghana (Gorbunova 1986: 272, 273) or Khorezm (Weinberg 1977: 87-89, 177-183). Negmatov and Saltovskaya (1975: 263, 264) maintained that Ustrushana (part of Sogd bordering on Ferghana and Chach) was controlled by the Kushanians. Rtveladze (1986: 34-39), who surveyed a defence wall in the highlands near Darband and correctly identified it with the famous "Iron Gates" on the border between Bactria-Tokharistan and Sogd, believed that it had marked the northern border of the Kushanian state (see also Pugachenkova, Rtveladze 1990: 56-59). However, because "long walls" of that period had been registered around cities like Samarkand and Bokhara, and in the Mervian oasis, they did not necessarily mark the borders of great states (unlike the Great China Wall or the Roman walls). Doubtless, at least during the Kushanian period, northern Bactria together with southern (left-bank) Bactria was part of the Kushanian state not only culturally but also politically (Stavisky 1976, 1977).

The same may hardly apply to Khorezm, Sogd, Ferghana, or Chach. Their relationships with the Kushanians are more difficult to assess. However, the fact that Khorezm had its own coinage while the Sogdians used various types of coins should not be taken to disprove the idea that these territories were politically or at least economically and/or culturally dependent on the Kushanians. V. Masson's view regarding the Kushanian empire as a system surrounded by buffer dependencies and his idea that the Kushanians extended their control over the Central Asiatic interfluve (V. Masson 1981: 36, 37) appear to be the most correct ones. V. Masson believes that the Kushanians controlled the entire region between Amu-Darya and Syr-Darya except Khorezm, and that according to the inscription of Shapur I, their rule might have reached as far as Chach. He also thinks that the political ties of the Kushanians with Sogd and Ferghana must have been strong enough to enable a free transfer of Kushanian troops across the Zarafshan and Ferghana valleys (V. Masson 1981: 36, 37). But even if these views are correct, the political history of the territories north and east of Bactria-Tokharistan is still far from being well understood.

Northern Bactria-Tokharistan in the Kushanian period.

Growth of irrigated territories, increase in the number of sites, urbanization level. Rtveladze's studies have revealed an exceptionally intense settlement of Northern Bactria and especially its northwestern part during the Kushanian era, far above anything observed during the preceding periods (Pugachenkova, Rtveladze 1990: 65).

His comparison of the number of Achaemenian (600-300 ВС) and Kushanian (AD 0-300) sites in the six principal river valleys of northern Bactria is illustrative indeed: 7 against 76 in Surkhan-Darya. 2 against 45 in Kafirnighan, 5 against 38 in Sherabad, 4 against 27 in Vakhsh, 3 against 14 in Amu-Darya, and none against 9 in Parkhar. The total number of Achaemenian sites in northern Bactria-Tokharistan is 21 against 209 Kushanian ones, a ten-fold increase.

The surface area of irrigated land, too, had increased in the Kushanian period, and new agricultural oasises had emerged in some valleys. According to Mandelstam (1964: 23-25; 1966: 7, 149-153; see also Stavisky 1977: 64, 65), a channel at least 15 km long was built in the right-bank Bishkent valley, and several villages were founded on its banks, the largest one (2.5 ha in surface area) being Khan-Gaza. In the same period, several large irrigation constructions were built. Specifically, traces of a channel from the Kafirnighan and pits remaining from a line of wells running across the watershed were discovered. It was apparently planned to transfer water from the Kafirnighan to the Bishkent valley, but the attempt failed, and the underground channel connecting the wells had not been constructed.

An even larger agricultural oasis emerged in the left-bank part of the lower Vakhsh valley. According to Seymal (1971: 47; Stavisky 1977: 67-70), a channel with several branches was built there. The system had made it possible to irrigate 40, 000 ha of fertile land which was previously uncultivated. More than a dozen Kushanian sites have been discovered in the area.

According to very incomplete data of the mid-705, at least 13 urban settlements existed during the Kushanian period in the Balkhab valley, Central Bactria, no less than ten in the Surkhan-Darya valley, and at least three in the lower Vakhsh valley. The total number of urban centres in these three valleys, then, was no less than 25. They included huge fortified settlements, the surface area of each being several hundred hectares: Balkh, Termez-Tarmita, and Shakhrinaus in the upper Surkhan-Darya valley (Stavisky 1977: 84, 85).

Archaeological sequences and associations. Chronology.

One of the major achievements of Kushanian studies was that stratigraphical sequences were established for Bactria-Tokharistan (V. Masson 1985: 255-257, 397-401, tab. CIX-CXIII). Also, associations characterizing the Kushanian period as a whole and its separate stages were described. According to V. Masson's scheme, there are three such associations: Khalchayan, Dalverzin, and Zar-Tepe. His descriptions are based on all sorts of artifacts: pottery, metal, construction debris, etc. (V. Masson 1985: 255-257, 397-401, tab. CIX-CXIII).

The dating of these associations is a disputed issue. I, like most scholars, am convinced that the notorious "Kanishka date" should be set at late 1st or early 2nd century AD, but no later than AD 166 (Stavisky 1977: 27). The Khalchayan association contains bronze coins imitating those of the Greco-Bactrian king Heliocles, and coins of so-called Soter Megas, minted apparently in the end of Kudzula Kadfiz's (Kadfiz's I) rule (M. Masson 1950; Cribb 1993: 107, 108, 133), at any rate before Kanishka's accession, that is, at the initial stage of Kushanian history. The early date of Khalchayan is also supported by the fact that stratigraphically it immediately postdates the Greco-Bactrian association (V. Masson 1976: 9, 10).

There is little doubt that the second, Dalverzin, association, which in many respects follows the Khalchayan tradition, coincides with the peak of the Kushanian state. The latest association, Zar-Tepe, succeeds Dalverzin both stratigraphically and typologically. According to the commonly accepted view, it coincides with the final stage of the Kushanian period or even with the Kushano-Sasanian times. There is disagreement, however, with respect to the absolute chronology of Zar-Tepe. While some scholars date it to the late 3rd or 4th century, others believe 5th century to be a more likely estimate. The assessment of the upper border of Zar-Tepe depends on when Kushano-Sasanian and Sasano-Kushanian coins circulated in Bactria-Tokharistan. As the evidence from the Buddhist ritual centre on Kara-Tepe hill, Old Termez, suggests, these coins were succeeded by local ones with an anchor-shaped tamga around AD 400, and this date is upheld by data from Zar-Tepe (Zavyalov 1990: 177).

Ideology: art, cults and religion, literacy.

Kushanian art represents a mixture of several tendencies. One is the so-called dynastic tendency which emerged in Bactria in the early Kushanian period. Another one, which may be described as Greco-Buddhist, or Romano-Buddhist, originated in Gandhara. Also, a local, Bactrian (Tokharistanian) tradition is present. Apart from these, elements and motifs characteristic of Classical (Greco-Roman), steppe (Scytho-Sarmatian), Near Eastern, Iranian, and Indian art are clearly recognizable. Having absorbed and transformed all these, the Kushanian art of Bactria-Tokharistan became a highly peculiar blend differing from anything seen in other provinces of the Kushanian empire or the Buddhist world in general.

Important new evidence has been gained over the last three decades concerning religion and cults. It has become evident that it was only during the Kushanian period that Buddhism had spread over the region (fig. 1). It was neither the only nor even the official religion; rather, it coexisted with old Avestian-type Iranian cults, sometimes referred to as Mazdaist, as well as with the official dynastic cult of Kushanian rulers, with the ancestor cults, and with cults of the Great Mother Goddess and those of other female and male deities, with canonical Iranian Zoroastrism, with survivals of Greco-Roman beliefs, with Sivaism, probably with Manichaeism, and possibly with Christianity. This is evidenced by non-Buddhist ritual constructions related to the "dynastic" cult in Khalchayan (Pugachenkova 1966, 1971), by a wall painting representing Siva in Dilberjin (Kruglikova 1974: 44-48), by clay figurines of the "Bactrian goddess" (Pugachenkova, Rtveladze, Belyaeva et al. 1978: 75-90), as well by as other terracottas (Meshkeris 1991: 57-65), by certain inscriptions, including the Brahmi inscriptions from Kara-Tepe with the word "Mahesvara", one of Siva's names (Grek 1972: 117), and finally by written sources (Stavisky 1977: 173-209).

Thanks to the inscriptions discovered mostly during the last three decades, more information has been gained concerning the types of script used in Bactria-Tokharistan in the Kushanian times. At the early Kushanian stage, a Greek script inherited from the Greco-Bactrian state existed which, like the Greek language, was used by the Kushanian administration and probably by the Hellenistic (or Hellenized) officials in the Kushanian service. This is evidenced by early Kushanian coins and by recently discovered inscriptions both from the southern (left- bank) and the northern (right-bank) Bactria (Kampyr-Tepe and Yavan). A 1st - 2nd century AD inscription on a rock in Kara-Kamar, southwestern Uzbekistan, is quite unique. It was left by a certain Rhinos, evidently a Roman subject who had come from Parthia, like the Roman warrior named Gaius Rex of the XV Pannonian Legion who had marked his visit to the ritual caves of Kara-Kamar, possibly a Mitraist sanctuary, by making an inscription in Latin (Rtveladze 1990; Ustinova 1990: 145-147).

According to Livshits (1979: 95), another script occasionally used by the Kushanians was the Bactro-Aramean script. Aramean, inherited from the Achaemenian official tradition, was used to transcribe Bactrian (Eastern Iranian) speech. Livshits believes that one of the Greco-Bactrian documents from Ay-Khanum and two Kushanian inscriptions from Jiga-Tepe, the Balkh oasis, and Fayaz-Tepe, Termez, were written in Bactro-Aramean. The idea, however, is questioned by Fussman (1988: 46).

After the Kushanian empire had attained its heyday (marked by the famous Kanishka's accession to the throne), the Kushanian script, also called Bactrian, or Bactro-Greek, using Livshits's term, became the principal script of Bactria-Tokharistan, both left-bank and right-bank. Apart from coins, it was used in inscriptions carved on stone, probably official ones, reporting on constructive works, like the well-known inscriptions in Surkh-Kotal (Steblin-Kamensky 1981: 326-328, 333, 334) and Dilberjin on the left bank (Livshits, Kruglikova 1979: 98-112) or the Ayrtam inscription on the right bank (Turgunov, Livshits, Rtveladze 1981). The same script was used in Kampyr-Tepe, 30 km west of Termez, where remains of a text (or several texts) written in Indian ink on birch bark, papyrus, or palm leaves were found (Rtveladze 1984: 88, 91, 104-106) as well as inscriptions on potsherds written in Indian ink or scratched on clay before firing (Pugachenkova, Rtveladze 1990: 97-99).

The "Kushanian" script, along with Indian ones (see below) was used in proclamative and donative inscriptions and in texts containing owners' names. All these were found during the excavations of the Buddhist sites of Kara-Tepe, formerly Kushanian Tarmita (Termez), and Fayaz-Tepe (Stavisky 1977a: 217, 218). Examples of this script are also present on clay vessels for storing grain which were found in Dushanbe (Livshits 1969: 60, note 64), as well as on vessels from Tepai-Shakh, southwestern Tajikistan (Litvinsky, Sedov 1983: 149), Khalchayan (Pugachenkova 1966: 59, 60, fig. 35, left), Dalverzin-Tepe, Yalangush-Tepe in Bandykhan oasis, Zar-Tepe west of Termez, Mirzakul-Tepe in the fringes of modern Termez (Pugachenkova, Rtveladze 1990: 97, 98), Zang-Tepe near Termez, and Kafyr-Kala in the Vakhsh valley (the latter ones are possibly early medieval, see Livshits 1969: 73-75, 1976: 163, notes 4, 5). Italic Bactrian graffiti scratched on the walls of the already abandoned Buddhist caves in Kara-Tepe were apparently made in the late Kushanian or post-Kushanian period (see Stavisky 1977a: 218, where references to works by Livshits and Harmatta are given).

In Kara-Tepe, Fayaz-Tepe, and in the basement of one more Buddhist construction situated in Old Termez, on the territory now occupied by a dump, many inscriptions made in Indian scripts, Pracrit (Kharosthi) and Sanscrit (Brahmi) were found (Vorobyova-Desyatovskaya 1983: 22-49, 51 -63, 89). An especially important discovery was made in Dalverzin-Tepe: a hoard containing ten golden bars with punched inscriptions in Kharosthi stating their weight, owner, and place from where they were received (Vorobyova-Desyatovskaya 1976, 1983: 87, 88).

Another script, discovered by Fussman in Afghanistan, one which he described as an unknown script of Aramean origin, was also found by him on a Kushanian potsherd from Khalchayan (Fussman 1974, 1988: 49, 50). The following inscriptions are referred to by Livshits (1976: 165, 166) and Vertogradova (1982a: 160-167) as examples of an unknown script: (a) on a silver bowl from Issyk burial mound, southern Kazakhstan, most likely 3rd or 2nd century ВС; (b) on an ostracon from Tekkuz-Tepa, the Vakhsh valley, southern Tajikistan, possibly

Kushanian; (с) on three fragments of a Kushanian vessel from Kara-Tepe; (d) on a fragment of a definitely Kushanian vessel from Khatyn-Rabat. While the Kara-Tepe inscription is evidently monolingual, that from Khatyn-Rabat was bilingual since several Kharosthi letters are present on the inner surface of the fragment (Pugachenkova 1968: 32, 33) and several signs made in an allegedly unknown script on the outer side (first published by Vertogradova).

Rtveladze believes that two more examples of this script are known from southern Uzbekistan: (a) an inscription scratched before baking on the mouth of a khum found in Kosh-Tepe; and (b) an inscription carved on a fragment of a limestone artifact from Barat-Tepe (Rtveladze, Livshits 1985: 34). According to Vertogradova, two periods can be traced in the development of this script: the early one (3d and 2nd centuries ВС) and the late one (first centuries AD). She also believes that the script was used for recording texts in various languages or dialects, including the Saka or the Yuechi (according to Livshits and Rtveladze). The last one to be mentioned is the "Middle Persian" script, used in some parts of Bactria-Tokharistan during, or immediately after, the Sasanian conquest, that is, in the Kushano-Sasanian period. It is represented by the Kara-Tepe graffiti (Lukonin 1969: 40-45; Stavisky 1975: 8, fig. 10) and by the Dalverzin potsherds (Livshits, Nikitin 1990: 66-72).

Seven or eight scripts, then, are known to have been used in Bactria-Tokharistan in the Kushanian times: Greek, Latin, "Kushanian" (or "Bactrian"), Kharosthi, Brahmi, so-called "unknown script of Aramean origin", Middle Persian, and possibly "Bactro-Aramean". The Greek script and the Greek language were widely distributed in Bactria-Tokharistan before Kanishka's accession, in the early years of his rule, and maybe even later. The "Kushanian" script that had been used for recording local Eastern Iranian speech since Kanishka's times was doubtless autochtonous. Latin, Kharosthi, Brahmi, and Middle Persian were introduced from outside.

In Kanishka's times, the level of literacy among the Buddhists of Termez was high, since more than 150 potsherds with Indian inscriptions were found there, including bilingual ones, Pracrito-Bactrian and Sanscrito-Bactrian (likely attesting to the existence of translators' schools in Termez). After the Sasanian conquest, the proportion of literate persons must have remained the same. Some people who visited the abandoned Buddhist constructions at that time made graffiti on their walls (over twenty have been discovered). Most of them, written in Bactrian italics, were probably left by residents of Tarmita or other Tokharistanian localities, although some Indian and Middle Persian inscriptions are present as well. The former ones might have been made by Indian worshippers or tradesmen, the latter ones by Sasanian officials and possibly warriors.

Overall, studies done over the three recent decades have greatly enhanced our knowledge of Kushanian history.


Н.Г. Горбунова. Об одном типе бронзовых зеркал ("бактрийские"? "сарматские"?)

N. G. Gorbunova. Bronze mirrors of the "Bactrian" or "Sarmatian" type

Several writers have addressed the technology and origins of bronze mirrors often found in female burials. The typology of the Sarmatian mirrors was developed by Khazanov (1963), and that of the central Asian ones by Litvinsky (1978). Both researchers have described disk-shaped mirrors with a roll along the rim, conical knob in the centre, and a short rod-like handle (Khazanov's type VIII, and Litvinsky's type 1/3, see fig. 1: 9). Both believed that these mirrors were of Oriental origin (Khazanov, 1963: 65; Litvinsky, 1978: 85). Litvinsky has even proposed to call them Bactrian, and the idea was supported by Skripkin (1990, 152-153). According to another view, these mirrors were introduced to Bactria by the nomadic tribes (although they have also been found in Tup-khona burial ground). Mandelstam (1975: 141) has noted that they are less common in central Asia than in the steppe. Moshkova (1989a: 188), too, tends to the idea of the steppe (Sarmatian) origin, although she does not exclude the possibility of Oriental provenance. The issue was recently addressed by Zadneprovsky (1993: 88-93), who rejected the term "Bactrian type" and suggested that the mirrors (fig. 1: 1) had been brought to India during the Yuechi invasion of Bactria. He proposed to describe them as "Yuechi-Kushanian" (in India, they were found in a 100 ВС - 100 AD Sako-Parthian horizon at Taxila). Zadneprovsky, however, believes that they could have been manufactured in the Kushanian state.

Because more mirrors have been recently discovered, I would like to revisit the problem. I have listed all the finds (see table 1 for description, and fig. 2 for the map, where the finds are numbered as in table 1; places where the number of mirrors is unknown are marked by crosses).

Having pooled the data for Central Asia and northern Afghanistan, we note that the mirrors date from the late 2nd century ВС (sometimes 3d century ВС) to the 4th and 5th centuries AD. Three chronological groups may be tentatively defined.

The early group (late 2nd century ВС - 1st and 2nd centuries AD) includes three mirrors from Tilla-tepe, northern Afghanistan (fig. 1: 5, 8), two from the necropolis of Tulkhar, one from BM V burial ground, one from BM VI (fig. 1: 2), an unknown number from Ksirov (all these localities are in Bactria), two from Lyavandak burial ground (fig. 1: 3, 4), an unknown number from Kuyu-Mazar burial ground, Sogdia, one from the fortified settlement of Koy-Krylgan-kala, Khorezmia, one from Juzkuduk 2 burial ground, three from Kulkuduk burial ground, the central Kyzyl Kum (fig. 1: 6), one from Kuruksay burial ground, western Ferghana, and one from a burial at Burgulyuk, Tashkent district. The entire group, then, totals 18 specimens plus an unknown number from Ksirov and Kuyu-Mazar.

The intermediate group (1st century ВС - 3rd century AD) includes two mirrors from the ground burials of Tup-Khona, one from a burial vault at the necropolis of Dushanbe, one from BM IV burial ground (all these are from Bactria), one from BM V mound 15, southwestern Ferghana, two from Altyn-asar 4 burial ground, and two from the fortified settlements of Tompak-asar and Bedaik-Asar, the Lower Syr Darya basin, thus totaling nine specimens.

The late group (3rd - 5th centuries AD) includes one mirror from a burial vault at Mizdakhan necropolis, Khorezmia, one from Jun burial ground, Tashkent district (fig. 1: 7), and three from a burial pit at Ak-tepe II settlement, Bactria, a total of five.

Eight of the 18 mirrors of the early group (plus an unknown number from Ksirov) were found in Bactria, at least three in Sogdia, four in the central Kyzyl Kum, two in the Ferghana- Tashkent area, and one in Khorezmia. In the intermediate group, four mirrors are from Bactria, one from western Ferghana, and four from the Lower Syr Darya basin. It should be stressed that the three specimens from Bactria come definitely not from nomadic burials, and two from the Lower Syr Darya were found on a fortified settlement. In the late group, three mirrors are from Bactria, one is from Khoresmia, and one from the Tashkent district. Specimens from Bactria, then, clearly predominate in all the three groups (15+ of 33+).

It is especially noteworthy that mirrors of this type were found in the ancient town of Taxila, Pakistan. They come from the Saka-Parthian level dated at 100 ВС - 100 AD. Descriptions and illustrations given by Marshall (1951, V. II: 584-585 and pi. 182-208 and 211b) are not quite accurate. Apparently all the 28 mirrors belonged to this type. Regrettably, Marshall claims that all mirrors were made of copper, which contradicts the results of analysis (page 567), according to which, two mirrors of an unspecified type are made of bronze with a high content of tin (fig. 2). Also, Marshall describes ten bone and ivory handles. Apart from 28 mirrors from Taxila, Zadneprovsky describes five more specimens from India, three of which have been dated to the 2nd century ВС. At least 33 mirrors from India, then, are known. One mirror (undocumented) was found in Kunduz, Afghanistan (Litvinsky 1973: 83), and one in Hasani Mahalla, Iran (Datlaman III 1968: 17, pl. LXIII: 6).

The total number of mirrors of the early group found in south central Asia and India is about 50, and 10+ come from other parts of central Asia, closer to Amu Darya and Syr Darya.

Mirrors of the intermediate and late groups were also found in the same localities. However, the Lower Volga basin, the core area of the Sarmatian culture, has yielded only 20 specimens, none of them being earlier that 1st century AD (Skripkin, 1990: 152). Only three mirrors are known from more eastern territories: one in Gorokhov and two in Sargat burials of the forest Trans-Uralian area and Western Siberia (Mogil'nikov, 1992: 290, 304, tab. 119: 9, 126: 25, 34). These finds have been explained by links with Sarmatians, and the same applies to finds in the areas west of the Volga basin.

Despite the general similarity between all the mirrors of this type, certain differences are seen in their details, such as size and thickness of the roll, shape and size of the conical knob, appearence of the rim (smooth or scalloped), presence or absence of engraving. These differences have been traced on four Sarmatian mirrors from the collection of the Hermitage Museum: those from Kalinovsky burial ground (mound 55/8, coll. № 2806/61; fig. 1: 10), Bol'shoy Armavirsky mound (coll. № 2242/46), mound № 1 near Armavir (coll. No. 2240/1), and Troyany (coll. № 1035/3). The knobs and rolls on these mirrors are different.

The spectral analysis of mirrors from Tulkhar and Babashov burial ground (Bogdanova-Berezovskaya, 1966: 226-231; 1975: 193-208) and Ak-tepe II (Ravich, Shemakhanskaya, 1987: 144-146), and the Sarmatian mirrors (Mosjkova, Ryndina, 1975: 123, 125; Bartseva, 1993: 96) has shown that the content of tin in the bronze of which they are made ranges from low to high. The spectral emission and X-ray fluorescent analysis of metal samples from four 1st—3rd century AD mirrors found at Babashov burial ground and belonging to a different type, and from four Sarmatian mirrors mentioned above, was performed by M. Dneprovskaya and A. Sizov at the Hermitage Museum Laboratory of Technical Methods, and the results nearly matched those of Bogdanova-Berezovskaya (judging by the tin content, one of the mirrors was evidently made in Bactria). The Sarmatian mirrors contain a fairly high proportion of tin (see Addendum for the results). However, because the number of mirrors examined is small, no definite conclusions can be made.

All the mirrors, including those analyzed at the laboratory and those which have been registered according to illustrations and descriptions, are rather carelessly manufactured. Their facial side is sometimes rough. Ravich and Shemakhanskaya describe the specimens from Ak-tepe II as massive, rough, and unpolished. It is possible that the mirrors, especially those found in graves, have never been used in everyday life (Ravich and Shemakhanskaya are of the same opinion). Some graves contain broken mirrors, which evidently served ritual purposes (Litvinsky 1978: 106-111). Litvinsky has correctly noted that some terracotta figurines of women holding mirrors have been found in Central Asia; clearly, they have nothing to do with the nomads.

All these data suggest that mirrors of the type described by us are of Oriental (Central Asian, Bactrian) origin and were used not only by the nomads but by the agriculturalists as well. Their appearance in the Sarmatian culture was due to the influence from Central Asia, not vice versa. They became fashionable during the Middle Sarmatian period, but in the end of it they were replaced by pendant mirrors (Moshkova, 1989b: 200). In Central Asia, however, they survived for a much longer time whereas the pendant mirrors had never been widely used in this region.

After this article had been finished, a publication by Levina and Ravich appeared describing bronze mirrors from Jety-asar burial grounds (1995). Mirrors of this type, which they call "Bactrian", are few (only four, found in one burial ground and on two settlements).


Т.С. Нунан. Возникновение Киева как важного европейского торгового и ремесленного центра домонгольского периода

T. S. Noonan. The emergence of Kiev as a major European commercial and industrial center during the pre-Mongol era

Introduction. Research on Kiev's commerce during the pre-Mongol era has traditionally focused upon the city's foreign trade. However, archaeological excavations have shown that in the period of ca. 1100-ca. 1240 Kiev became a major exporter of glassware (bracelets, beads, rings, window glass, vessels), glazed pottery (vessels, decorated tiles, pisanki or ceramic eggs), expensive jewelry (inlaid enamel, decorations of niello, granulation, filigree work), religious goods (bronze enkolpions), rose slate spindle whorls, and amber items (beads, pendants, rings, crosses) to other parts of Rus' (Noonan 1991: 102-146). During the pre-Mongol period Kiev had emerged as a major European commercial and industrial center exporting the products of its workshops throughout the Rus' world and beyond. Kiev's development as a major medieval industrial center brings into question the idea that the city had begun to decline starting in the mid twelfth century. To resolve this issue, it is necessary to examine in detail Kiev's transformation into a major European industrial center. This study is a preliminary effort at such an analysis.

Considerable research on craft production in Kiev was done during the pre-revolutionary and Soviet periods. Much of this work up to World War II is discussed in B. A. Rybakov's book on The Handicrafts of Medieval Rus' (Рыбаков 1948). Subsequently, numerous studies on medieval Rus' crafts and craft production were published by a number of Soviet scholars. Many workshops were uncovered and various studies explored how different goods were made. One of the greatest contributions of Soviet scholarship on Kievan Rus' was to illuminate its urban life and especially craft production. Unfortunately, this research has been largely ignored by American scholarship whose interest in the Kievan era focuses upon other topics (Crowther 1969; Bibliography 1956 etc. ). Aside from a few translations of Soviet research, there are relatively few English-language works on Kievan towns and craft production (Tikhomirov 1959; Thompson 1967; Callmer 1981: 29-52). This lack of interest in pre-revolutionary and Soviet research on craft production has had some unfortunate results. For example, in his widely used text on early Rus', published in 1948, George Vemadsky claimed that there was no evidence that glass was produced in Kievan Russia (Vemadsky 1948: 116). But, already in 1907-1908, V. V. Khvoika had uncovered the remains of glassmaking and enameling workshops of the eleventh-thirteenth centuries in the detinets area of Kiev (Хвойко 1913: 71; Корзухина 1956: 324-329, 339). The failure to incorporate Khvoika's research into a standard English-language text on Kievan Rus' has helped to perpetuate an image of Rus' backwardness already in the pre-Mongol era. In sum, we can no longer ignore the fact that Kiev and other Rus' towns were important and advanced industrial centers already in the pre-Mongol era.

Pre-Mongol Workshops in Kiev. One way to examine the emergence of Kiev as a major medieval industrial center is to consider the remains of workshops excavated in the city. P. P. Tolochko provided a map showing thirty-one sites in the city where craftshops of the pre-Mongol era had been found up through the mid 19, 70-s (Толочко 1983: 138, 139, рис. 65). In examining the data published by the mid-1980-s, I found evidence for some sixty-one distinct craft workshops in pre-Mongol Kiev (see Table A). The list in Table A is incomplete. Some workshops already known by the mid 1980-s may have been omitted inadvertently since it is sometimes very difficult to determine what precisely constitutes a craft workshop. Some crafts do not leave any tell-tale evidence of craft production like distinctive kilns or forges. Thus, it is not always possible to distinguish amongst a craft workshop, a warehouse where goods were stored, and the residence of a merchants dealing in these goods. In addition, we sometimes find the by-products of craft production without distinctive signs of specialized tools, i. e., slag may be uncovered in a dwelling having no signs of a forge. There is no objective and mechanical way to identify all craft workshops easily. It is also impossible to date most craft workshops from pre-Mongol Kiev with great precision. Workshops are normally dated generally according to the stratum or strata in which they are found and not all archaeologists agree on these general datings. Narrow datings are rare (see Table A, Nos. 4, 14, 34). Thus, the dates for most workshops are not exact and should be used with care.

An examination of the sixty-one workshops raises several questions about the degree of specialization in craft production. Did blacksmiths, for example, make all types of iron goods or did they specialize in everyday tools, special instruments, weapons, harness parts, domestic goods, or items of clothing? Did one potter produce both expensive glazed ware and cheap kitchen crockery? How extensive was specialization and did it increase over time? Unfortunately, archaeology does not provide definitive answers to these questions. Some workshops suggest real specialization in goods made from slate (spindle whorls), bone, amber, bronze, etc. Other workshops suggest less specialization. Glassmaking and jewelry, for instance, were often very closely connected crafts. It is not always clear if a jeweler made his own glass or obtained it from a glassmaker. Some workshops suggest the existence of several diverse crafts at the same site. At the so-called workshop of the artist (Table A, No. 37), specialized woodworking tools and pieces of unworked and half-finished amber were uncovered. At one twelfth-century workshop (Table A, No. 51) archaeologists found evidence for the production of amber goods, glassmaking, and the production of bronze goods. The sixty-one workshops raise more questions than they answer regarding the degree of craft specialization in Kiev.

We must also remember that only a relatively small part of pre-Mongol Kiev has been excavated (Каргер 1958: вклейка; Новое в археологии 1981: 9). If less than five percent of the city's medieval area has been excavated more or less randomly and sixty or so workshops were found, then it is possible to project a potential total of ca. 1200 craft workshops from pre-Mongol Kiev. There are thus good reasons to believe that the sixty-one craft workshops in our study only represent the tip of the iceberg.

Since Kiev is a major modern city, archaeologists have not been free to excavate in many areas that they might consider as very promising. Table A only has evidence for three pottery workshops while it also contains thirteen blacksmiths' shops. The archaeological evidence points to massive production of wheel made pottery and yet our data fails to reflect this. Soviet scholars concluded, quite reasonably, that the potters' quarter of medieval Kiev has still not been uncovered (Новое в археологии 1981: 284). The sixty-one workshops in Table A are by no means a representative sample of the total craft production in Kiev.

The above discussion shows that there are serious problems in using craft workshops as an indicator of Kiev's industrial production in the pre-Mongol period. Despite these difficulties, the sixty-one workshops in Table A provide an insight, however flawed or distorted, into Kiev's development as a major medieval European industrial center.

The Branches of Kiev's Industry. Kiev's pre-Mongol workshops can be divided into several basic groups: blacksmiths' shops, potters' shops, jewelry shops, glassmaking shops, bronze and copper smiths' shops, bonecarving workshops, stonecarving workshops, amber workshops, plinf workshops, and indeterminable workshops. The few "unique" workshops can be connected with one of these major categories. For example, the workshop for baking lime (Table A, No. 58) was probably linked with the construction of such projects as St. Sophia's Cathedral (Новое в археологии 1981: 348). Therefore, this workshop was probably connected with the plinf workshop (Table A, Nos. 30-31) which functioned at about the same time and place. The only workshop standing by itself was the one for dressing skins (Table A, No. 59).

Two charts show the distribution of workshops by category. The second of these charts takes into account data showing that one workshop may have been used for production in more than one craft. Around twenty percent of the workshops in pre-Mongol Kiev were blacksmiths' shops and another twenty percent produced jewelry. Our unscientific sample thus suggests that blacksmithing and jewelry production were the two most numerous craft industries. In the next cluster, glassmaking had eleven percent, bonecarving and amber working each had ten percent, and construction workshops had eight percent. Finally, pottery, bronze and copper smithing, stonecarving, indeterminable, and others each had five percent or less.

The two most numerous crafts in our sample are not necessarily what one would expect. It is not surprising to find blacksmithing among the most widespread crafts. On the other hand, it is surprising that jewelry production was one of the most numerous crafts in Kiev. There must have existed a substantial rich class to patronize such workshops. Furthermore, a significant part of Kiev's population, beyond the very wealthy, must have possessed the means and desire to acquire jewelry. Finally, there were large markets elsewhere in Rus' for Kiev's jewelry. While blacksmiths produced everyday items, extensive jewelry production reflected a demand for everyday adornments, the city's affluence, and the existence of export markets. The production of fine jewelry using niello, granulation and filigree work shows the sophistication of Kiev's jewelry industry as well as the transfer from abroad of the technologies needed to make such fine jewelry. Kiev's craftsmen mastered these imported technologies and employed them in their growing production. Kiev's jewelry industry supplied a variety of goods to consumers of diverse classes living in both Kiev and throughout the Rus' lands.

Pottery, bronze and copper smithing, and stonecarving were among the least numerous crafts. It is not surprising to find the latter two crafts in this category. Bronze and copper could not compete with wood, iron, clay and, later, glass as materials. Stonecarving was limited by the fact that, as a rule, only churches and dwellings of the elite were made of stone. The only exception was the production of spindle whorls made from the rose slate found in Ovruch (Рыбаков 1948: 188-202). But, as noted above, it is surprising to find so few pottery workshops in Table A when excavations in Kiev have revealed huge quantities of shards from local wheel-made pottery. We can only assume that once the potters' section of Kiev has been discovered, pottery, like blacksmithing, will turn out to have been one of the most numerous crafts.

Glassmaking, bonecarving, amber working, and construction (plinfs, lime) were all crafts that were neither very numerous nor very few in our sample. The presence of a number of glassmaking and amber workshops may seem somewhat surprising. These workshops symbolize Kiev's emergence as a major industrial center. Along with good jewelry and fine pottery, the production of large quantities of glass and amber items in Kiev was intended to supply both the demands of Kiev's affluent populace and that of the elites in towns throughout Rus'. Large numbers of glass bracelets made in Kiev were exported to all parts of Rus' to meet the demands of urban women.

The Growth and Evolution of Craft Production. Table В (Russ. Б) shows how many of the fifty-one dated workshops in pre-Mongol Kiev were operating at any one time. Only a few workshops functioned during the first part of the tenth century and they represented fairly basic crafts. There is no evidence that Kiev's industries were either diversified or technologically advanced at this time. Two construction workshops may have begun to function by the mid-tenth century. For most of the tenth century, craft production in Kiev was modest in scope and basic in nature. A dramatic increase in the number of craft workshops took place in the late tenth and early eleventh century. The marked expansion of Kiev's workshops came, in part, from a growth in basic and/or already extant crafts. But, the industrial growth of this period was brought about by the appearance of new and/or more advanced crafts. Around five glassmaking workshops sprang up ca. 1000. Given the connections between glassmaking and jewelry, Kiev's emergence as a center for the production of fine jewelry, expensive ceramics, and glassware dates to ca. 1000. One or two workshops for the production of amber goods date to this time as well. Kiev's emergence as a major industrial center thus dates to about the same time as the initial conversion of the Rus' ruling elite to Orthodoxy.

In an earlier study, I suggested that the wealth which Kiev's increasingly sedentarized ruling class derived from tribute and the Byzantine trade combined with the closer contacts with Byzantium brought by conversion all created a group of affluent people with a propensity toward conspicuous consumption. Imports could be both expensive and difficult to transport. Thus, the easiest way to satisfy the new tastes was to produce these luxury goods in Kiev itself. It was easier to bring in Byzantine glass-makers who would introduce glassmaking into Kiev than to transport all the glass from Constantinople (Noonan 1988: 105-11 1; Noonan 1989). The dramatic increase in the number and type of Kiev's workshops ca. 1000 can easily obscure the fact that some earlier workshops in the areas of blacksmithing, jewelry, stonecarving, indeterminable, and construction now ceased operation. These developments highlight the dynamic character of Kiev's industries. As new workshops appeared or as projects like the construction of major churches were completed, older workshops died out while some new workshops proved short-lived.

The end of so many workshops during the early eleventh century led to a decline of thirty percent in the total number of functioning workshops ca. 1025 (from 30 to 21). The number of workshops stabilized at this point and remained at around twenty for the rest of the century. But, the turnover among workshops continued. Between ca. 1025 and ca. 1100, several workshops ceased to exist but no real trend is apparent. No new crafts appeared and there was no marked growth or decline in the number of workshops.

A second period of marked growth dates to around 1100. New blacksmith (3), jewelry (6), glassmaking (1 or 2), bronzemaking (1), bonecarving (1), construction (1), spindle whorl (1), and amber (2) workshops appeared at around this time. But, six workshops also ceased their existence at this same period. Table B, which shows an all-time peak of thirty-seven workshops ca. 1100, exaggerates the overall growth since it reflects both new and closing workshops. There was a real net gain of ten workshops or forty-eight percent. In any event, there was a substantial growth in the number of Kiev's workshops ca. 1100. This growth reflects what can be called a period of mature growth. In the well-established crafts such as blacksmithing, there was stability or only small growth. In glassmaking, one of Kiev's early high-tech industries, there was even a decline. At the same time, the period ca. 1100 saw a dramatic growth in the number of jewelry and amber workshops. Six new jewelry workshops appeared while the four existing workshops continued to function. In fact, all of these ten workshops were still operating at the time of the Mongol conquest in 1240. Based upon our sample, jewelry was the most notable growth industry of the later Kievan era and was characterized by a very strong consumer demand. The number of amber workshops doubled ca. 1100 reflecting what must have been an expanding market for amber crosses, pendants, beads, etc.

Three of the workshops to appear ca. 1100 represent new crafts in our sample. The workshop producing building stones (Table A, № 47) seems to reflect new building in Vladimir's City while the workshop making spindle whorls (Table A, № 48) indicates that Kiev's artisans had begun to mass-produce a highly popular everyday item used throughout Rus'. A new foundry producing bronze goods (Table A, № 36) suggests a modest growth in demand for bronze items. High-tech crafts based on technology transfer were not the only industries to have a large domestic market.

In brief, some crafts had reached their maturity ca. 1100 and did not share in the growth of workshops at this period. Jewelry and amber production experienced a sharp growth reflecting an increased demand for luxury goods while new workshops making everyday items also seemed to have prospered. The growth in workshops which occurred ca. 1100 was therefore selective; some crafts grew markedly, some grew modestly, and some remained stable or declined slightly.

Following the period of mature growth ca. 1100, the number of workshops in Kiev remained relatively constant for the rest of the twelfth century. One new blacksmithing shop opened but two closed while a new bonecarving shop appeared.

The third and final era of significant change in the number of workshops took place in the early thirteenth century. Unlike the prior two periods of great change (ca. 1000 and ca. 1100), the early 1200s witnessed a net decline of six craft workshops. Shops in blacksmithing (1), jewelry (2), glassmaking (2), amber (2) closed while only one new workshop (jewelry) began operation. These figures raise the question of whether Kiev's industrial growth had now peaked and, in fact, entered a period of decline prior to the Mongol conquest. The drop in the number of workshops occurred in both high-tech and low-tech crafts. The decline also encompassed some of Kiev's key export industries, e.g., jewelry, glassmaking, and amber goods. It is possible that by ca. 1200 Kiev was already exporting its products to every town and hamlet in Rus' where a demand existed, i.e., the limits of the market for Kiev's products had been reached. Furthermore, in glass-making and jewelry, local workshops producing the same goods made in Kiev had begun to appear throughout Rus', i.e., Kiev's craftsmen faced increased local competition. Kiev's industrial production may have moved from a period of mature growth to the initial stages of decline in the early 1200s.

On the other hand, the decline in the number of workshops ca. 1200 may not represent any real decline in Kiev's place as the leading industrial center of the Rus' lands. Of the fifty-one total dated workshops from the pre-Mongol era, twenty-three or forty-five percent were still functioning at the time of the Mongol conquest and these twenty-three workshops cover the entire spectrum of Kiev's craft production from high-tech to low-tech industries. On the eve of the Mongol conquest, Kiev was still a major European industrial center producing goods both for its own consumption and for the larger Rus' market.

The workshops which closed ca. 1200 may have been those that were unable to operate in the more competitive conditions of that period. The demise of some less efficient workshops in Kiev and the growth of competitive workshops elsewhere in Rus' does not necessarily mean that Kiev was on the decline as an industrial center. Kiev was probably experiencing the loss of its former monopoly position in the production of certain goods. In short, it remains for future research to determine if Kiev had entered a period of industrial decline ca. 1200. The Mongol conquest of Kiev in 1240 makes it impossible, at this time, to ascertain if a period of industrial decline had begun about a half-century earlier. But, it is striking that only one new workshop can be dated to ca. 1200.

Trends in Particular Industries. In addition to the analysis of the overall number of workshops, it is also useful to trace the evolution of specific industries in Kiev. However, only three crafts (blacksmithing, glassmaking, and jewelry production) have a sufficient number of workshops from which to construct a profile for the pre-Mongol era. Fortunately, these three crafts represent three different types of industries. Blacksmithing was a basic craft not requiring advanced technology. Glassmaking was a semi-luxury craft based on technology transfer which matured early. Jewelry was also a luxury craft based in part of foreign technology transfer but it matured late and continued at a high level down to the Mongol conquest.

Table С (Russ. B) shows the number of blacksmithing workshops extant in Kiev during the pre-Mongol era. Several blacksmithing workshops existed before Kiev's industrial take-off ca. 1000. The number of shops doubled during the first period of high growth ca. 1000 but then levelled off and remained relatively constant. There was only a modest increase during the period of mature growth ca. 1100 and a minimal decline ca. 1200. Having grown along with Kiev's overall economy following Vladimir's conversion, blacksmithing became very stable. Seven of the thirteen known blacksmith workshops were in the podol. Blacksmithing was not concentrated in the detinets or Iaroslav's City where the city's elite resided.

Table D (Russ. Г) shows the number of glass-making workshops in Kiev during the pre-Mongol period. Glassmaking was non-existent in Kiev prior to the period of initial growth. However, after glass-making developed rapidly ca. 1000, the number of workshops remained very stable over the next two centuries. A significant decline in the number of workshops began ca. 1200. Glassmaking workshops were found in all parts of the city. The demand for glassware encompassed all parts of Kiev's society and extended throughout Rus'.

Table E (Russ. Д) shows the number of jewelry workshops in Kiev during the pre-Mongol era. The earliest jewelry workshop dates to the tenth century. There was a definite growth in the number of workshops ca. 1000. But, the dramatic growth in jewelry production came ca. 1100. From this time to 1240, the number of workshops remained relatively constant. Compared with glassmaking, jewelry production matured later and maintained its market better in the first half of the thirteenth century. There was also greater turnover among glassmaking workshops than jewelry establishments. While jewelry workshops were located throughout Kiev, half were found in the detinets. Despite the large market for Kiev's jewelry elsewhere in Rus', jewelry workshops had a special affinity for the original heart of the city where many of its affluent elite resided.

Conclusions. An analysis of Kiev's sixty-one known workshops is crucial for the study of two fundamental issues: 1) what crafts existed in Kiev and what was the relative importance of these crafts, and 2) how did Kiev's industrial production evolve over time both for individual crafts and for crafts as a whole. This study tries to address these questions by suggesting some new and potentially fruitful ways of approaching the development of Kiev as a major European industrial center in the pre-Mongol era. While the methods employed here can provide important insights they also have clear limitations. Nevertheless, given the paucity of written sources, there is much to be learned from the methodology used here. At the same time, all conclusions suggested by our data should be considered tentative. Our analysis has suggested working hypotheses that need to be tested and revised as research continues and further excavations in Kiev provide new data. Finally, this study of Kiev needs to be expanded in two ways. First, we need to explore the production of Kiev's workshops, i. e., how many goods were made in the city's workshops. Second, the patterns from Kiev must be compared with those from other Rus' cities of the Kievan era. In particular, a comparable study of Novgorod is essential since dendrochronology makes the dating of its workshops more precise. We need to ask if there was an all-Rus' model of industrial growth or were there different models of which Kiev's was only one?