V. G. Moiseev, Museum of Anthropology and Ethnography (Kunstkamera) of the Russian Academy of Sciences

3 Universitetskaya Emb., Saint Petersburg, 199034, Russia

E-mail: vmoiseyev@mail.ru

UDC 572

Introduction

The problem of the origin of carriers of the Okhotsk culture of the first millennium AD since its isolation in the 1930s remains one of the most relevant in the population history of the Russian Far East and the Japanese Archipelago. The isolation of this culture from the civilizations of that time limits the role of written sources in elucidating its genesis, as a result of which archaeological and paleoanthropological data play a key role.

According to archaeological sources, the Okhotsk culture was formed in the 5th century AD on the basis of the Susu culture in the southern part of Sakhalin. The latter, according to A. A. Vasilevsky (2002), is related to the North Sakhalin culture, but has adopted some features of the Epidzemon culture, whose carriers repeatedly penetrated Sakhalin from Hokkaido. The researcher emphasizes the cultural heterogeneity of the Okhotsk people, believing that the Okhotsk culture is a sequence of related local archaeological cultures that coexisted in time and space. The cultural impact of the mainland cultures associated with the Tungus-Manchu tribes (Mohe culture) on the Okhotsk population is indisputable.

Immediately after the formation of the Okhotsk culture, namely from the fifth century AD, its expansion began on the territory of Hokkaido, where it quickly spread along the narrow coastline of the Sea of Okhotsk. Most of the Okhotsk monuments are located on the eastern and northern coasts of the island and on the Kuril Islands (Fig. 1); only a few are committed

Fig. 1. Location of the main monuments of the Okhotsk culture. Monuments whose craniological series were used in the analysis: 1 - Moyoro; 2 - Hamanaka; 3-Omisaki.

In the more southern regions of Hokkaido (Matsumura et al., 2006), this localization of Okhotsk sites is explained by the high level of specialization of the economy of its carriers, who practiced hunting large marine mammals. In addition to archaeological data, the results of isotope analysis (Yoneda, 2006) also point to the leading role of this area of specialization in the hunting base of the Okhotsk culture carriers.

It is assumed that the fates of local groups of Hokkaido Hunters were different. Among Japanese archaeologists, there is a widespread opinion that the eastern groups of Okhotsk are more acculturated than the northern ones by the local Proto-Jain population representing the Satsumon culture. Acculturation resulted in the Tobinatai culture of eastern Hokkaido. In the northern regions of the island, the Okhotsk people were displaced to Sakhalin by proto-harvesters (Ohyi, 1975).

Anthropologists have expressed several points of view regarding the origin and genetic connections of the bearers of the Okhotsk culture. The Eskimos (Suzuki, 1958) and Aleuts (Kodama, 1948; Ito, 1951) were among the possible" relatives " of the Hunters. At present, perhaps the most widespread opinion is that they are related to the modern Amur peoples, as well as to the Nivkhs (Yamaguchi, 1975, 1981; Ishida, 1988, 1994; Ishida et al., 1994; Kozintsev, 1990, 1992b). There was also a possibility of mestissation of the Okhotsk people with the Proto-Jain population of Hokkaido (Ko, 1965, 1971; Kozintsev, 1990, 1992b).

One of the main obstacles to studying the anthropological features of the native Okhotsk culture was the lack of craniological material. Only in recent years, after careful processing and registration of the anthropological collections of Hokkaido University (Sapporo), it was possible to study, perhaps, the most representative craniological series - from Moyoro (near Abashiri), as well as less numerous series - from Hamanaka (Rebun Island) and Omisaki (near Wakkanai). Previous studies of these materials were carried out using a rather short craniometric program (Kodama, 1948; Ito, 1951; Ito, 1965), which led to limited possibilities for using the results obtained. New efficient feature systems developed in recent years, as well as new methods of statistical processing of the material, make us turn again to these collections.

At present, it is becoming increasingly clear that different systems of anthropological traits are not equally informative when studying different stages of population history. I had to make sure of this when analyzing anthropological materials from Western and Southern Siberia, as well as from Eastern Europe [Moiseev, 2001, 2004]. It turned out that each feature system "tells" its own story. If the features within the systems are informative, then any of these stories are generally reliable, although not sufficient. In other words, there is no" perfect " system of traits that fully reflects the history of a particular human population. Differences in information received from different systems can be minimized by integrating data from different systems. The advantages of this approach have been demonstrated in a number of studies [Kozintsev, Gromov, Moiseyev, 1999; Kozintsev, 2004; Moiseev, 2001, 2004].

The purpose of this article is to identify the kinship relationships of local groups of native speakers of the Okhotsk culture of Hokkaido based on the integration of data on two systems of anthropological features that are most often used in the study of paleoanthropological material - craniometry and cranioscopy.

Program, material and methodology

Skulls of representatives of the Okhotsk culture, as well as Ainu and native speakers of the Jomon culture of Hokkaido, stored at Hokkaido University and Sapporo Medical College, were studied by me according to the craniometric program, as well as according to the cranioscopic program of A. G. Kozintsev (Kozintsev, 1988; Kozintsev, 1992a). 14 metric and five non-metric indicators were selected for the integration intergroup analysis (Tables 1, 2). From the cranioscopic model, the following parameters were used:

Table 1. Frequencies of cranioscopic signs in Okhotsk groups

Note. ZI is the occipital index, KSS is the sphenoid-maxillary suture, ZSS is the posteroscullar suture, PSU II is the suborbital pattern of type II, IPNSH is the index of the transverse palatine suture, NO is the supraorbital openings. The number of observations is shown in parentheses.

* Frequency calculated using the regression equation.

Table 2. Average values of craniometric features in the Okhotsk groups

Note. N is the number of observations, and X is the average value.

Table 3. Materials that served as the basis of the study

However, at this stage, the occipital index was excluded, since this trait, effectively distinguishing European and Asian populations, does not have a clearly defined gradient in Asia. In the case of craniometric features, data were used only for male skulls; cranioscopic data were summed up without taking into account gender; for KVSH, half-sums of values for male and female skulls were calculated.

Carriers of the Okhotsk culture are represented in three series. Two of them (Omisaki and Hamanaka) belong to the northern group of Okhotsk monuments, and one (Moyoro) belongs to the eastern group (see Figure 1). For comparative statistical analysis, we used 12 contemporary series from the Far East, as well as a combined series of representatives of the Jomon culture from the Japanese Archipelago (Table 3).

The statistical analysis consisted of two stages. At the first stage, craniometric and cranioscopic data (after conversion to radians) were processed separately using canonical analysis and principal component analysis, respectively. The vectors obtained as a result of these two analyses have fundamentally similar properties and can be used as new features based on the intergroup correlation of the initial indicators.

At the second stage, several first vectors were taken from the obtained canonical vectors (CV) and principal components (GC), which together reflect at least 70% of the total variability for each of the systems. In this case, this criterion is met by the first three KV and the first two GC. These vectors were again processed by the principal component method. Since both GC and HF are orthogonal, the new integral GC (IGC) results are based only on intersystem correlations.

Results and discussion

The first two vectors within each of the systems - IQ and IQ (46 and 50% of the total variance, respectively) - describe a fundamentally similar direction of inter-population differentiation. The correlation coefficient between them is quite high and amounts to 0.77 at p < 0.05 (Table 4). Both vectors separate the Dzemons and both groups of Ainu from the rest of the groups. It should be noted that in the case of I GC, there is a slight overlap in the values of Ainu and Okhotsk (Fig. 2). The distinct anthropological distinctiveness of the Dzemons and their similarity to the Ainu, revealed by this analysis, is a completely predictable result, which once again confirms the previously established fact of genetic continuity between these groups (see, for example, [Kozintsev, 1990]). The combination of measurement features characteristic of the Dzemons and Ainu includes large values of the smallest width of the forehead and the angle of protrusion of the nose, a relatively high bridge of the nose, as well as small values of the zygomaxillary angle, the height of the face and nose. Cranioscopically, these groups are characterized by a high frequency of HSS, IPNSH, as well as a low frequency of NO and PSU.

The Okhotsk groups are somewhat different in terms of their location within the boundaries of I GC and I KB. If the cranioscopic signs indicate a strong deviation of the Okhotsk people from all Mongoloid groups towards the Dzemonskaya group, then this shift is not so pronounced according to craniometric signs. In this case, the Okhotsk people are not opposed to the Mongols, but occupy an intermediate position between the Dzemons and the Amur groups, like the Eskimos, Aleuts and Chukchi. In both cases, the Amur peoples, being as far removed from the Dzemons as possible, are characterized by the most pronounced Mongoloid features. Despite these differences, the results of the analysis of both systems of features generally do not contradict each other.

To verify the latter conclusion, another canonical analysis was performed using individual craniometric data available to me for the four series I measured: the Okhotsk series from Moyoro, Hamanaka, and Omisaki, and the Ainu series from Hokkaido. The analysis was carried out on 28 criteria: 1, 8, 17, 5, 45, 43, 46, 40, 9, 77, Zm, SC, SS, 29, 30, 31, 48, 55, 54, 51, 52, 26, 27, 28, 32, 72, 74, 75(1). In the absence of a particular measurement, the average value for the group was taken for the individual.

The ranges of IQ values in the Ainu series and in all groups of Okhotsk people largely overlap (Figure 3).However, the Moyoro group has the greatest number of cases of overlap of IQ values with the Ainu (17% of individuals in this series). In the Hamanaka and Omisaki groups, such cases are 16 and 13%, respectively. Thus, the presence of Proto-Jain admixture in the Okhotsk people is confirmed. Its value according to the results of the analysis of individual data averaged about 15%.

The differences in the nature of the distribution of IQ values in different groups of hunters are noteworthy. So, while the individuals from Hamanaka who fall into the transgression zone with the Ainu are quite different from other members of their group, in Moyoro such a gap is not observed, the distribution area of individuals is more compact here. Perhaps this indicates the duration of the period of mestizo breeding of hunters with

See Table 4. Correlation coefficients between GC (cranioscopy) and CV (craniometry)

*P<0,05.

2. Arrangement of groups in the space of I GC and I KV.

3. Location of individuals in the local Okhotsk groups and the Ainu group with Hokkaido in the space of the first and second KV.

See Table 5. Correlation coefficients between GC (cranioscopy) and KB (craniometry) with the first three IGCS

4. Arrangement of groups in the space of I and II IGCS. Refer to Figure 2 for the technical description.

proto-mines in the Moyoro area. The Omisaki series is more similar in this respect to the Moyoro group.

Let's return to the results of the main analysis. The above-mentioned IQ and IQ are the only vectors with a high correlation between them (Table 4). Consequently, the remaining vectors represent relatively independent directions of inter-population differentiation, which once again confirms the thesis about the uniqueness of various systems of traits in terms of historical information. In order to decide which of these areas are most significant, we will refer to the results of integration analysis.

As might be expected, I IGC (37% of the total variability) is essentially a linear combination of I KB and I GC (Table 5). It contrasts the Dzemonian and Ainu series with the others (Figure 4). Among the Okhotsk groups, the Moyoro group is characterized by the most pronounced Dzemonian craniological specificity. Of all the series included in the analysis, it is the closest to the Ainu. This result corresponds to the popular idea in Japanese archaeology that the eastern groups of Hunters had the closest contacts with Proto-Jain representatives of the Satsumon culture (Ohyi, 1975; Amano, 1979). However, the Ainu specificity in the Moyoro series is still quite moderate, and in general, the Moyoro group is closer to the rest of the Okhotsk groups than to the Ainu. This indicates that the specific weight of the Ainu component in this group is significantly lower than that of the Okhotsk component itself.

Thus, the intermediate result of the integrative analysis confirms the opinion of a number of researchers about the presence of a small Proto-Jain admixture in the composition of Okhotsk [Ito, 1965, 1971; Kozintsev, 1990, 1992b]. It is important to note that the only meaning of the I IGK is to contrast the Dzemons and Ainu with other groups. Series that are equidistant from the above two do not necessarily have to be related to each other. To find out their genetic connections, it is necessary to consider other areas of differentiation.

The second IGC (27% of the total variability) is mainly determined by the II KV and II GC (Table 5). About a quarter of its variation falls on one group - the Aleuts. Since the problem of the origin of the Aleuts is not included

5. Arrangement of groups in the space of I and III IGCS. Refer to Figure 2 for the technical description.

among the objectives of this study, I will not discuss here the possible reasons for their craniological specificity. I will only note that in the light of the available genetic data [Rubicz et al., 2003], the result obtained does not look random. For the present study, it is more important that all the Okhotsk groups are located at the opposite end of the second IGC from the Aleuts. This result does not agree with the opinion that the Okhotsk people are related to the Aleutians (Kodama, 1948; Ito, 1951). If we still assume the possibility of such a relationship, we will have to assume that the differences between the Aleuts and hunters were the result of population processes that took place after the migration of hunters to Hokkaido, and this is unlikely. Most likely, the medieval ancestors of the Aleuts were not related to the ancestors of the Hunters.

The third IGC (20% of the total variance) is very significant for the Okhotsk problem. This IGC is almost completely determined by craniometric data, and the largest contribution is made by the III KV (Table 5). The third IGC contrasts the Arctic groups (Eskimos, Chukchi, and Aleuts) with the Amur groups, including the Nivkhs (Figure 5). The former are characterized by negative values, the latter by positive ones. The Dzemons and both Ainu series occupy a neutral position (let me remind you that their originality is manifested in the first vectors).

The Okhotsk people also correspond to intermediate values of the III IGC, but, unlike the Dzemons and Ainu, they do not show specificity for other significant vectors, which indicates their Mestizo origin. According to the third IGC, the Okhotsk people are heterogeneous. The Hamanak series is characterized by a fairly distinct shift towards the Amur groups, and the Omisaki series - towards the Arctic ones. The Moyoro series occupies an intermediate position. The "Arctic" trend found in the Omisaki group is confirmed by the fact that this series, being below the regression line, showed a similar direction of intersystem differences with the Arctic groups when comparing the first KV and the First GC (see Figure 2). The same trend is observed in the Dzemons and both groups of Ainu. If this is not accidental, then the most plausible hypothesis is that protomorphic traits are conserved in the aborigines of Japan, as well as in the Eskimos and Chukchi.

So, we can assume that local groups of Hokkaido Hunters differ in the ratio of the Amur and Arctic components, which may explain the anthropological variability within the Okhotsk population. This circumstance becomes particularly important in connection with the opinion of A. A. Vasilevsky about the illegality of using the concept of "Okhotsk culture", since the monuments attributed to it belong to several related, but completely independent cultures [1999]. It is premature to comment on this point of view from the standpoint of physical anthropology, since I have practically no Okhotsk material from Sakhalin at my disposal. So far, we can only note that the craniological differences between the three Okhotsk series with Hokkaido are quite large and exceed the differences between the Eskimos and Chukchi or between the majority of the Amur peoples (Table 6).

Nevertheless, the distances between the Okhotsk series are smaller than those between them and any other group involved in the analysis, so they are grouped together in clustering (Figure 6). Apparently, local groups of Hokkaido Okhotsk people had a common origin, and later the differences between them increased due to isolation or mismatch in the directions of ethnic ties. Although it is currently impossible to accurately date each Okhotsk burial, it seems unlikely that the observed intergroup discrepancies are caused by the different timing of the groups. The available dates for monuments largely overlap, and the use of cumulative series should offset the effect of different times, if any.

Commenting on the results of the cluster analysis, it should be noted that one should probably not attach special importance to the association of Okhotsk groups with Chukchi and Eskimos. This result is defined in-

Table 6. Euclidean distances between groups

Figure 6. Dendrogram obtained by clustering Euclidean distances using the unweighted pair-group method.

It is largely random and is related to the peculiarities of the method used, which does not allow us to reflect the intermediate position of mestizo groups. As it was shown above, the Okhotsk people occupy an intermediate position between the Amur and Arctic groups.

Conclusions

The main results of the integration analysis of cranioscopic and craniometric features are as follows: 1. Carriers of the Okhotsk culture (or Okhotsk cultures) were formed as a result of mestizo breeding, which involved Arctic populations similar to the Eskimos and Chukchi, and ancient groups of Tungus-Manchus who penetrated the territory of Southern Sakhalin from the lower Amur in the first centuries AD.

2. The Aleuts show a pronounced craniometric and cranioscopic specificity, the features of which are not observed in the Okhotsk groups. This makes it possible to exclude the Aleuts from the number of close relatives of the Okhotsk people.

3. The anthropological distinctiveness of the local Okhotsk groups of Hokkaido may reflect differences in the specific weights of the Amur and Arctic components in their composition.

4. The eastern groups of Hokkaido Hunters have a more pronounced Proto-Hawaiian component than the northern ones.

Acknowledgements

My sincere thanks go to Prof. T. Amano (Hokkaido University), Prof. H. Ishida (Okinawa University), and Dr. H. Matsumura (Sapporo Medical College) for the opportunity to work with craniological collections and scientific advice, and to Dr. H. Ono (Hokkaido University), without whose practical help my work in Japan would not have been possible. I am deeply grateful to A. G. Kozintsev for his critical comments and assistance in the work.

List of literature

Amano T. Okhotsku bunka no tenkai to chiikisa (Distribution and regional differences of the Okhotsk culture) // Bull, of the Institute for the Study of North Eurasian Cultures. - Sapporo: Hokkaido University Press, 1979. - Vol. 12. - P. 75 - 92. (in English).

Vasilevsky A. A. Okhotsk problem and its modern reading in Russia and Japan // Integratsiya arkheologicheskikh i etnograficheskikh issledovaniy [Integration of archaeological and ethnographic research]. un-t. - 1999. - p. 129-133.

Василевский А. А. Сусуя и эпидзёмон // The 5th Open Symposium of the Hokkaido University Museum: Okhotsk Culture Formation, Metamorphosis and Ending: Japan and Russia Cooperative Symposium. - Sapporo: Masaaki SUVA, 2002. - P. 89 - 95.

Debets G. F. Antropologicheskie issledovaniya v Kamchatskoi oblasti Anthropological research in the Kamchatka region//TIEH. - M.; L., 1951. - Vol. 17. - 264 p.

Ito S. Human skulls from the Moyoro site / / Koukogaku zassi. - 1951. - Issue 37. - pp. 214-217. (in English).

Kodama S. Moyoko Kaizuka (Moyoro Shell Heap Site). - Sapporo: Hokkaido Genshi Bunka Ken Kuyukai, 1948. - 112 p. (in English).

Kozintsev A. G. Ethnic cranioscopy: Racial variability of the skull sutures of a modern person. - L.: Nauka, 1988. - 168 p.

Kozintsev A. G. Kety, uraltsy, "americanoids": Integration of craniological data / / Paleoanthropology, ethnic anthropology, ethnogenesis: To the 75th anniversary of Ilya Iosifovich Gokhman / Ed.by A. G. Kozintsev. St. Petersburg: MAE RAS Publishing House, 2004, pp. 172-185.

Moiseev V. G. Severnaya Evraziya: yazykovaya differentsiatsiya i dannye fizicheskoi antropologii [Northern Eurasia: linguistic differentiation and data from Physical Anthropology]. - 2001. - N 4-P. 154-159.

Moiseev, V. G., Experience in integrating four systems of anthropological features (craniometry, cranioscopy, odontology, and dermatoglyphics), in Paleoanthropology, Ethnic Anthropology, and Ethnogenesis: To the 75th Anniversary of Ilya Iosifovich Gokhman, ed. by A. G. Kozintsev. St. Petersburg: MAE RAS Publishing House, 2004, pp. 186-200.

Yamaguchi B. Kraniologicheskie osobennosti nositelei okhotskoi kul'tury [Craniological features of carriers of the Okhotsk culture]. Dorumen, 1975, vol. 6, pp. 47-90 (in Russian).

Yamaguchi B. Chelovecheskie ostanki s Hokkaido [Human remains from Hokkaido]. Jinrugaku koza, 1981, vol. 5, pp. 137-156 (in Japanese).

Ishida H. Morphological Studies of Okhotsk Crania from Omisaki, Hokkaido // J. of the Anthropological Society of Nippon. - 1988. - N 96 (1). - P. 17 - 45.

Ishida H. Skeletal morphology of the Okhotsk People on Sakhalin Island //Anthropological Science. - 1994. - N 102(3). - P. 257 - 269.

Ishida H, Hanihara T., Kondo O., Ohshima N. A human skeleton of the early phase of the Okhotsk Culture unearthed at the Hamanaka-2 Site, Rebun Island, Hokkaido // Anthropological Science. - 1994. - N 102 (4). - P. 363 - 378.

Ito S. The stratigraphical changes of the skulls from Moyoro shell heap // Okajimas Folia Anatomica Japonica. - 1965. - Vol. 40. - P. 679 - 690.

Ito S. Regional differences of the Ainu skulls with reference to the non-Ainu characteristics of the skulls from Moyoro // Hokkaido Medical J. - 1971. - Vol. 46. - P. 247 - 268.

Kozintsev A. G. Ainu, Japanese, their ancestors and neighbours: Cranioscopic data // J. of the Anthropological Society of Nippon. - 1990. - N 98 (3). - P. 247 - 267.

Kozintsev A. G. Ethnic epigenetics: A new approach // Homo. - 1992а. - N43 (3). - P. 213 - 244.

Kozintsev A. G. Prehistoric and recent populations of Japan: Multivariate analysis of cranioscopic data // Arctic Anthropology. - 1992б. - N 29 (1). - P. 104 - 111.

Kozintsev A. G., Gromov A. V., Moiseyev V. G. Collateral relatives of American Indians among the Bronze Age populations of Siberia? //Amer. J. of Physical Anthropology. - 1999. - N 108(2). - P. 193 - 204.

Matsumura H., Hudson M. J., Koshida K., Minakawa Y. Embodying Okhotsk Ethnicity: Human Skeletal Remains from the Aonae Dune Site, Okushiri Island, Hokkaido //Asian Perspectives. - 2006. - N 45 (1). - P. 1 - 23.

Ohyi H. The Okhotsk Culture, a Maritime Culture of the Southern Okhotsk Sea Region // Prehistoric Maritime Adaptations of the Circumpolar Zone / Ed. by W. Fitzhugh. - The Hague; P.: Mouton Publishers, 1975. - P. 123 - 158.

Rubicz R., Schurr T. G., Babb P. L., Crawford M. H. Mtochondrial DNA Variation and the Origins of the Aleuts // Human Biology. - 2003. - N 75 (6). - P. 809 - 935.

Suzuki H. Physische Anthropologic in Japan // Homo. -1958. - N 9. - P. 37 - 47.

Yoneda M. Diet of the Okhotsk People Analyzed by isotopic Methods // Fife and Roots of the Okhotsk People. - Sapporo: Hokkaido University Press, 2006. - P. 51 - 57.

The article was submitted to the Editorial Board on 13.02.07.

Permanent link to this publication: