The author of the present paper originally intended to give, true with a delay, a critical review of Professor J. A. Lopatin's study on the Goldi ethnography, but during the writing some explanations of the author's points of view appeared to be absolutely necessary. Thus, the scope of the author's work naturally extended and he treated the problem of Goldi origin on a large scale which led him to a further step to connect the Goldi movement with that of other Tungus groups. Owing to that, some unpublished facts concerning the Tungus and Manchus taken from the author's material have also been set forth for illustrating some deductions resulting from his previous investigations. Moreover, the author had but limited library facilities and thus used the works that happened to be at hand. So for instance the publications of Professors Torji and Shiratori dealing with the problems of Manchuria could not be got. Generally speaking such a partial publication of deductions and facts owing to an occasion has many objections, especially from the point of view of authors who are naturally interested in being as much as possible convincing and prepared for criticism. In spite of this, the author publishes his deductions in the form of the forthcoming paper, hoping that before he publishes the material and deductions concerning: the problems at present treated, in an extended form, he will also obtain an encouraging and always welcomed criticism. In this consideration the author would like to find excuse for possible defects and prematureness of this publication.
Shanghai, December, 1926.
The Goldi [1] are one of the most interesting groups in the Far East owing to their closeness to the Manchus and, from an ethnographical viewpoint, their intermediary character with reference to the Northern Tungus. An ethnographical study published by J. A. Lopatin [2], together with previous investigations [3], provides some new data which permit to venture a comparison of the Goldi with the Manchus and other Tungus groups. His voluminous book containing a detailed map of the Goldi geographical distribution and several figures and photographs is the result of his several visits among the Goldi of the Amur and Ussuri rivers during 1912 to 1919, This attempt at giving a monographical survey of the Goldi ethnography is the first of its kind and as such it is worthy of specialists' attention.
The Goldi, known in some English writers' works under the Chinese name of Yu-pi-ta-tse, i.e., Fish Skin Barbarians, or, as wrongly used, Fish Skin Tartars [4], belong to the southern branch of the Tungus linguistic family [5]. Perhaps their language may be regarded as a branch of the Manchu spoken [6]. The present affinity of the Goldi with the Manchus is also beyond any doubt, as I [7] have shown, for the Manchus consider them as a kindred group which may be included into iče (iči) manau [8], i.e., the Modern Manchus. However, the Goldi, as will be shown, are not of the same origin as the Manchus living in Heilungkiang and Kirin. The Goldi language shows several elements characteristic of the Northern Tungus dialects and their ethnography in many respects shows the traces of Northern Tungus complex. Moreover, some Paleasiatic [9] influence has also to be counted as an element forming the present Goldi ethnographical features. A recent direct Chinese influence has lately also been exercised upon those of them who live in the vicinity of Chinese colonizers, i.e., especially in the mountainous region of the Ussuriland where the Chinese spread since the growing migration wave had passed the Ussuri and Sujfun rivers [10], also in the basin of the Sungari River. The Goldi have recently swallowed some Northern Tungus groups on the banks of the Amur River, which may also be connected with their origin and in a great degree with a recent process of Goldi spreading. In dependence on the above influences the Goldi form three main groups, i.e., (1) the Amur River Goldi, including the Goldi properly speaking, Olcha, Mangun, Kili (Kile), and Samagir; (2) the Ussuri River Goldi, nearly extinct, and (3) a very numerous group of the Sungari River Goldi. The first group is a resultant from Paleasiatic influence (Giliaks, etc.), Northern Tungus influence, and a recent Russian influence; the second group is under a strong Chinese influence introduced among them by Chinese hunters and trappers; and the third group is influenced by a direct contact with the Manchus more than any other, but it has recently fallen under a strong Chinese influence The last group, however, is not investigated at all and J. A. Lopatin's informations are rather incidental, being borrowed from various sources. Meanwhile this group may be regarded as having the opportunity of being the most typical of all groups for it is located in a territory very near, or even in itself, of the Manchu cradle, i.e., the middle and upper courses of the Sungari River and its tributaries.
From Chinese sources it is established that in the past various events took place in the present Goldi territory. However, the relationship between various groups known previously to the last Manchu dynasty are not clear at all. It may also be noted that the process of ethnical variations — the flourishing and decline of groups — in this part of Asia has also been recorded: various Paleasiatic groups, Koreans, Mongols, Tungus, also later the Chinese and Russians were struggling and pushing one another during the whole period since the neolithic period to our days. Archaeological excavations and incidental finds, as well as the records of various travellers, have already given the evidence showing that there were some periods of a very flourishing state of this territory and some others when depopulation and decline are evident. So, for instance, in the,middle course of the Amur River huge constructions, dated of the Kin dynasty, walled cities, cemeteries, etc., are very abundant, but the same region provides several remains of different stages of neolithic period. It is evident that with reference to the Goldi any historic succession cannot now be established and for instance the fact of an agriculturist population known to early Russians (XVIIth century) as dučery and various remains found in the present Goldi area do not provide reliable and positive data for establishing the kinship between that population and Goldi. So we are forcibly confined to hypotheses and theories, one of which is expounded in the present paper. Let us now proceed to the comparison of the Goldi with the Manchus, principally of Heilungkiang, the Ajgun District, who came into their present territory as late as the end of the XVIIth century from the Ninguta region, and on the other hand with the Northern Tungus of Manchuria and Mongolia, also Reindeer Tungus principally of Manchuria and Transbaikalia. In the forthcoming comparison I shall not deal with all ethnographical elements observed, but with those which are characteristic of those groups.
[1] The name «goldy» was borrowed by Russian travellers and traders from a Goldi group designating by this name another Goldi group. It was introduced by Russians into the scientific terminology where it has appropriated a recognized place as have many other conventional terms. Generally speaking they call themselves xödjö, xoзen, хodjen (according to Professor P. P. Schmidt, The Language of the Olchas, in publications of the Latvian University, VIII, 1923, Riga, p. 230 — «low», i.e., inhabitants of the low course of the Amur River), also many other names in dependence on the regional distinctions and regions occupied by different ethnical groups, e.g., nani,nanaj (local), manguni (the Amur River), akani (the Ussuri River Goldi), goldi (according to Professor Schmidt's suggestion, perhaps «middle», i.e., the middle course of the Amur River). This etymology of the name goldi, I believe, is right. In fact, golde+ suff. (Bir. dial.)—to keep the middle, to go in the middle of…; golo (Man. Sp.)—the middle, in general, the being between… (not only that of the river) and golo (Bir,)—the region, area (in Man. as well, also province, etc.), particularly, the area of a spirit's influence; at least, golde || golda (Mong. Cf. Professor A. Rudnev, Material on the Dialects of Eastern Mongolia, St. Petersburg, 1911)—to be in the middle. However, gilāmi (Bir.) means: the Goldi living on the banks of the Amur River, down the Sungari River, and those of the Ussuri River, as groups distinct from the Goldi of the Sungari River. I think, however, that the root gold belongs rather to the Mongol language. Moreover, their names are also varied by designations owed to their neighbours, e.g., čoldok (by Giliaks), olča (by Northern Tungus, Negidals), ačany (by early Russians), yu-pi-ta-tse (by the Chinese), kordekke (by the Japanese), janta. (by Giliaks), kil'i (a group of them by the Chinese), samagir, etc. Almost all these names were used by various writers at different periods, but only Goldi (gol'dy of Russians) has become a name for a general designation of all groups, i.e., it has become a scientific classificatory term. Thus an evident misunderstanding has here occurred and the principal cause of this misunderstanding was probably the lacking of a name by which Goldi designate themselves, as it is observed among the Manchus and some other Tungus groups living in the vicinity of the Manchus and Mongols. Unfortunately J. A. Lopatin did not clear up this problem and evidently has no opinion of his own. (Back)
[2] J. A. Lopatin, the Goldi of the Amur, Ussuri and Sungari Rivers. Essay of an Ethnographical Investigation (in Russian). Published by the Vladivostok Branch of the Amur Section of the Russian Geographical Society, Vol. XVII, Vladivostok, 1922. pp. 370, VI. (Back)
[3] Vide Supplementary Note 1. (Back)
[4] This term dates from the end of the XVIIth century when it was introduced by French fathers who, however, reserved it to a limited group of the Goldi (and partly perhaps Giliaks?). However, M. A. Frazer (Fish Skin Tartars, in the Journal of N.C.B.R.A.S., Vol. XXVI., No. 1, 1891-1892, Shanghai) in a detailed description of the Orochi of Port Imperial, based upon V. P. Margaritov's work On Orochi of Port Imperial, St. Petersburg, 1888, owing to a misunderstanding classifies them as Fish Skin Tartars which implies the confounding of them with the Goldi. As will be shown, such a classification of these groups cannot be justified. (Back)
[5] The affinity of the Goldi to Manchu language was already known to French fathers of the XVIIth century, who asserted that the yu-pi-ta-tse (ie., Goldi) language is a mixture of Manchu and that of Кe-tcheng-ta-tse (J. B. du Halde, C.J., Description geogra-phique, etc., de I'Empire de la Chine et de la Tartarie Chinoise, etc., Vol. IV. Paris, 1730, p. 12). The latter term is applied to various groups of the Northern Tungus, also probably to the Udehe (cf. infra, p. 175). Later on the study of this language inclined the linguists to consider it rather as a Manchu dialect (cf. Professor J. Z a x a r o v, Material for the Study of Goldi Language, in Bulletin of the Imp. Eus. Geo. Soc, 1876, fasc. 1, St. Petersburg; Professor W. L. Kotwicz, in Zivaja Starina, Vol. 70-71, 1909, St. Petersburg; Professor P. P. Schmidt, op. cit.). (Back)
[6] In so far as a preliminary analysis of the material dealing with the Manchu spoken (gathered by me in the Aigun District of Heilungkiang) permits to suppose. (Back)
[7] Social Organization of the Manchus, A Study of the Manchu clan Organization, Extra Vol. III of the N.C.B.R.A.S., Shanghai, 1924. (Back)
[8] з and dj are affricates very common in Manchu and Goldi. In Professor Schmidt's works they are transcribed dz and dz, which transcription I shall preserve when this author is quoted. (Back)
[9] I am using this term «Palseasiatic,» as a conventional term for designation of groups which, being taken as ethnographical complexes, do not belong to the Tungus, Mongol, Turk, Chinese, Koreans and Japanese ethnical groups. I agree with W. Jochelson (Archeological Investigations in the Aleutian Islands, Washington, 1925, p. 4) who rejects this term, but for the time being I like better to preserve it, for Ratzel's term «die Rand Volker» (marginal peoples), which W.Jochelson proposes, does not cover some groups of Siberia surely being remains of an early population (the Yenissy River Ostiaks, etc.) (Back).
[10] This process also the ethnography of the Chinese colonizers in general was investigated by a traveller who gave a good description of his observation (cf. V. K. Arseniev, The Chinese in the Ussuri Region. Published by the Amur Section of the Imp. Russ. Geo. Soc., in Russian, Habarovsk, 1914). This investigation may also provide the material for a chapter in a study on «the Chinese abroad » and in that on »description of ethnical groups transplanted into an alien environment (Back)», etc.
[11] Du Halde (op. cit. pp. 13-14) says that a Manchu group ilan xala (my transcription) of the Sungari River are mixed up with Yu-pi-ta-tse and use the original Yu-pi-ta-tse clothing. So they also use oxen, let us add like some early population of Manchuria (see below p. 163 footnote 85). According to du Halde, they lived in the Xurxa River basin and periodically migrated northward. It is not, however, clear whether he described the Goldi incorporated as ici maпaи, or he spoke about »goldified» Manchus, the latter supposition being little probable. Perhaps a part of the Goldi was not at all included into iči manau and remained as a semi-independent group, though incorporated into the military organization. (Back)