CHINA - ARCHAEOLOGY.
The study of prehistory in China is particularly difficult owing very largely to the geographical structure of the country. Three areas may be enumerated, the great prairie region, which in cludes much of Mongolia and part of western Manchuria, northern China, including the great plain and the lowlands of Manchuria, and thirdly, southern China. This latter region, which is in the main extremely mountainous has at present been little studied archaeologically, and its prehistory is almost unknown.
Thus there are many gaps in our knowledge, but none the less great credit is due to those explorers who have contributed the knowledge which we already possess. In the northern region the country is sparsely inhabited mostly by wandering nomad tribes. Distances are great and, except on the borders of China proper, few discoveries can be made without the aid of elaborate ex peditions such as that of the American Museum of Natural History, which discovered the Dinosaur's eggs in Gobi, and has already found abundant traces of Neolithic man in surface de posits. The southern part of this region and nearly all northern China is covered with fine wind-deposited loess. The discoveries of Peres Licent and Tielhard make it probable that the remains of early man are hidden beneath this yellow covering, which is so thick in some places that exploration is impossible except where the loess has been cut through by the action of water in river valleys, and the older strata have thus been exposed. Such canyons in the loess have proved in a few cases to be most fruitful. In the plain itself, however, these deep river valleys are non-existent for the streams run at a level which is actually higher than the plain. All along the fringe of the plain there are found the natural conditions necessary for the formation of deep val leys. The mountains of northern China also possess caves, so fruitful a source of finds for the prehistorian, which though so far only productive of a somewhat recent culture, offer promise of important discoveries.
This crude character and large size of the implements found are of considerable importance in making comparisons with other sites. The implements are difficult to place typologically. The majority suggest an Aurignacian industry, others are more remi niscent of the Magdalenian. The comparison between the finds made at the type site and on other hearths in the neighbourhood, at different levels, suggests that the area was inhabited by palaeo lithic man for the greater part of the time during which the loess was in the process of formation.
The second series of sites in the Ordos country is different from Shuitungkou. Finds have been made along the course of the Sara-osso-gol. This river, which is a tributary of the Hwang Ho, flows from the loess-covered mountains of the province of Shensi and traverses the south-eastern angle of the Ordos coun try. It has cut a ravine which is sometimes over 2ooft. deep. The sides of this ravine show clearly the succession of strata which have filled up the basin which formerly existed in this region at the beginning of quaternary times. Man lived on an old floor and since Palaeolithic times there has been the following succession of strata. First, dunes accumulated over the old floor, then a lake, then new dunes, then there was a new lake, until finally the level of the modern steppe was reached. Neo lithic finds have been made in the black soil at the top. In the levels associated with the presence of man the fauna is extremely rich, but implements are small and infrequent. The fauna in many ways resembles that from Shuitungkou. The woolly rhinoc eros is very abundant, remains of elephant were found, probably Elephas nomadicus, not the mammoth (Elephas primigenius). The wild ass and Bos primigenius also occur, while gazelles abound. Numerous antlers of deer were found, apparently used as some form of weapon or tool. A very large camel also ap peared. The big and small animals found in this deposit are in great contrast. The small mammals are without exception the same as those found to-day in the same regions. Among the big animals many types are represented which have either become totally extinct or are no longer found in the same area. These large animals show remarkable affinities with the fauna of Europe in Middle Quaternary times, and appear to have had a wide extension over northern and central Asia at the time of the formation of the loess. The implements are extremely small and present unusual forms. The largest implement is under Sin. long. This implement presents a close parallel to some of the scrapers from Shuitungkou. Similarly some of the smaller implements from that place suggest the Sara-osso-gol specimens. It seems probable that we are not dealing with different cultures. The Sara-osso-gol neighbourhood is singularly deficient in any form of pebbles and it is probably due to this geological condition and not a difference in culture that the explorers of the site were only able to collect a mere handful of implements, whereas on the other site their finds ran into many pounds weight. The Sara osso-gol culture appears to have had a wide extension in space. At present no implements have been found in the strata above the Palaeolithic. There is therefore a complete absence of any form of transitional period between the Palaeolithic age and the surface Neoliths.
The Palaeolithic industry has also been studied by the same authors in the loess of northern China. Owing to the rapid erosion to which this soil is subjected this has been possible, and two principal sites have been examined, Kingyan in Kansu and Yu fengtou in Shensi. All these sites probably belong to the same period, that is the base of the loess.
The distribution of the finds so far made is restricted to the southern part of the Ordos country, the north of eastern Kansu, and the north of Shensi. Nothing Palaeolithic has been discov ered either in eastern Gobi or beyond the north-western border of the Ordos country, possibly due to the fact that the correct strata have been eroded, but it is curious that animal bones have been found in deposited strata. In the more southerly regions no thorough search has as yet been made in those places where the base of the loess is exposed.
The chronology is as yet uncertain. a comparison of the fauna and of the types of the implements suggests that probably this early culture was approximately contemporary with the Mouster ian and the earlier part of the Aurignacian period. The im portance of the finds depends on several factors. First they show that the Palaeolithic industries spread right across the Eurasiatic continent, indeed some of the implements show analogies with implements found by Buxton in the north Arabian desert. Secondly, the finds at the base of the loess suggest that man was inhabiting China at a period before the deposition of that great yellow covering which is to-day the most characteristic feature of that country. The finds also help to throw light on the date of the loess.
The remarkable character of this culture is its apparently abrupt ending. The old floors are clearly defined. There are other strata superimposed upon them, but no traces of human handiwork between the Palaeolithic and the Neolithic have so far been discovered in situ. Up to the present the Mesolithic, i.e., the age between the Palaeolithic and the Neolithic is entirely absent with the exception of certain finds in Gobi. Nelson, a mem ber of the staff of the Third Asiatic Expedition of the American Museum of Natural History claims to have discovered certain implements on the surface of the desert which are Azilian in form, but as they are unrelated to any definite stratigraphy It is not possible at present to express any definite views on them.
Implements from other parts of China have also been claimed as being either Palaeolithic or Mesolithic but a closer examination has failed to substantiate the claims put forward, although at first sight some of them do bear a close resemblance to Palaeo lithic implements.
Far the richest finds have been made in the north. The Gobi region and much of north-eastern China and Manchuria have proved to be extremely rich in Neolithic remains. Surface finds are extremely abundant over most of the whole region. In some cases, especially in regard to arrow heads of stone there may well be some doubt as to their age, as the Chinese were well aware of the use of these implements, which may have continued to serve barbarous tribes until well within the historic period. In certain parts of Mongolia, however, implements have been found in situ, not as surface finds, and can therefore be definitely dated.
The Neolithic implements nearly always occur in a bed of black sands, from 3–I2ft. deep superimposed on the Quaternary sands, which in Mongolia replace the Chinese loess. The Neo lithic here appears to succeed the Palaeolithic without any transi tional period and is very widely spread. One of the most remark able sites so far explored is at Linsi (approximately ii6° E. N.). Here most of the implements are extremely rude, not unlike those of Palaeolithic date, but some fine and very delicately worked implements were also discovered. A few grindstones were found but no green polished celts. The pottery is very rude and was not made on a wheel. No painted pottery has been found on these Neolithic sites. Among the finds at Linsi certain flattened pear-shaped implements were found, and a similar instrument was obtained by Andersson at Kalgan. These stone implements are very important. Not dissimilar implements occur in other parts of Asia, but the closest parallel is North America. There can be little doubt that they were rude ploughshares and it is pos sible that they form another link between north-eastern Asia and North America. In any case there is every reason to believe that this Neolithic culture was extremely widely spread over a wide region in the Far East.
Some of the finds suggest that climatic changes have taken place since the Neolithic Age. If the implements are, as sug gested above, really ploughshares some at least come from regions which to-day are devoted to pastoral nomadism, and the presence of agricultural implements suggests that in Neolithic times the climate was somewhat different. Further, quite a large number of Neolithic implements have been found in regions which to-day are extremely poor steppe. They may be the remains of a merely temporary occupation, but the evidence is at least suggestive of a considerable change in climate.
The site at Yang Shao Tsun in Honan occupied a very big area, being over 600yd. long and about 5ooyd. broad. The de posits which are nearly 'oft. thick are slightly covered by arable soil. The stratum on which the remains rest has been in places excavated into large pits, often as much as 6f t. deep and of the same width. The deposits consist mainly of a mixture of soil and ashes, and apart from a few graves contains mostly the debris of an inhabited site. The fauna of the site has not yet been published in detail, but apart from the porcupine it does not appear to include any interesting wild animal. As in modern China the domestic pig appears to have been abundant.
The finds in Fengtien province in Manchuria were made in a small cave close to the railway between Tientsin and Mukden, about half way between those two towns. The cave was only about 2of t. long and 'oft. broad. It contained, however, over 6f t. of deposits, which consisted of a filling of clay alternating with three bands of charcoal which were extremely rich in ancient remains. In a layer, about half up, of broken and calcined human bones Black recognized the remains of about 45 individuals. Apart from differences in the implements due to the available raw material there is no reason to doubt that the Honan and the Fengtien sites belong to the same culture.
In Kansu, Andersson has explored certain parts especially in the west of the province round the city of Lanchow. He has found numerous sites, both inhabited and burying places. On the basis of the pottery he has grouped the sites into different sub periods of which the three lowest correspond to the sites from other parts of China, whereas the upper ones appear to belong to the Copper or early Bronze Age. The pottery from Kansu is both abundant and of very fine workmanship, whereas the im plements are both poor in workmanship and relatively infrequent. The best pottery has been found in the graves. Most of the bodies were found extended, the head being turned towards the north, but in two graves the skeletons had been placed on the left side in a contracted position. A large number of the bones were covered with red ochre.
At present no exact information appears to have been published on the associated fauna, but the domestic pig as in the other sites, to have been extremely abundant.
Except for the sites in Kansu no metal has as yet been found on any of these sites. Either the people were unacquainted with the use of metal, or alternately, and less likely, metal objects have so far escaped the eyes of the excavators. The artifacts may be divided into three classes, stone objects, bone objects and pottery.
Rings of stone and of mussel shell were also found. Their mean ing is somewhat obscure. They may have had a magico-religious purpose. Parallels to such stone rings can be found in various parts of the world, but at present they have not been sufficiently studied to throw any light on this intricate problem. Bone needles and awls have been found by Andersson in some abundance. One he believes to be almost identical with a specimen from the Danish kitchen-middens. The resemblance was probably due rather to the limitations imposed by the material than to any actual cul tural connection.
The most characteristic pottery, however, from the Chinese sites consists of a very fine ware decorated with painted designs. In some cases the surface of the pot has been treated with a fine washing of clay, technically known as a slip, bef ore the paint has been applied ; in other cases the paint has been put direct on to the finished surface of the pot. The colour of the paint is usually either red or black and red. An examination of the fabric of the pots has shown that they are the product of a relatively advanced industry. They are made of a loess clay, the most readily obtainable material in northern China, which is not avail able south of the Yangtze Kiang; so that future finds south of this river, even if of the same culture, are likely to differ in technical details. The clay has been very well washed and freed from plant remains and other impurities, and probably worked into a paste. To this mixture some quartzite or other flinty material was added to give strength to the pot and to prevent it collapsing during the process of manufacture. Some of the wares are of different colours. These differences are due to the amount of iron oxide and to various degrees of burning. It is probable that the addition of iron was made in order to affect the colouring, and possibly also the potters were able to control the firing, but on the whole these differences, striking as they are, must be considered largely as the result of semi-accidental factors. The classification by colours is therefore not an essential feature. The nature of the ornament is quite otherwise.
Most pottery was made on the wheel and the firing was carried out with considerable skill. The forms can seldom be recon structed owing to the fragmentary condition in which they were found, but at least three classes appear to be represented. The painted pottery includes, first, bowls, which are either hemispheri cal or rather flat and basin-like. Secondly, there is a group of tall pear-shaped pots. These have a comparatively narrow base and gradually increase in size for about two-thirds of their height, after which they again decrease in diameter, but are provided usually with a projecting rim. The unpainted ware includes a series of tripod jars of considerable interest. A vessel which prob ably had a narrow neck like so many water pots used in the East to-day was also found, but the exact shape of this vessel is uncer tain as it was much broken. Pots with pointed bottoms have also been found.
Tripod vessels are not uncommon in south-eastern Europe and occur as early as the first city of Troy. The pear-shaped vessels show an undoubted analogy with similar pots, sometimes painted, sometimes plain, which have been found in the Danube-Dnieper region, although there are certain differences in the form of the rim. On the other hand the small bowls show affinities with others found at Anau and Susa, in the earliest strata on those two sites (Anau I. and Susa I.) . The pots with pointed bottoms have a wide distribution, and parallels could be cited from Troy, Egypt, and even India, and therefore provide little evidence for the solu tion of the real connection of the culture. A further parallel with Troy is, however, found in a pot with a perforated bottom, a type of pot also found in Persia and Egypt. At present, parallels with other regions of the Far East are lacking, but some of the objects from Manchuria, described by Torii as pottery handles, may turn out to be the legs of tripod pots.
The decoration on the pottery throws further light on the prob lem. At a date which probably coincides with the 3rd millennium B.C. (though there are divergent opinions on this question) the use of painted pottery was widely spread over south-eastern Europe and western Asia. The technique is similar, but certain differences in design have led some writers to question the associa tion of all the Chinese finds with those of the West. There can be no doubt on general grounds that all the pottery belongs to the same family and that it has a very wide distribution in Eurasia, at the period covering the transition from the Stone to the Copper and Bronze ages. Although the distance is very great there are no marked geographical boundaries except deserts in the region which stretches from the Aral-Caspian basin to the Pacific Ocean. There have been trade routes across this region from time immemorial and there is reason to believe that in former times the deserts did not constitute the barriers which they do at present owing to different climatic conditions.
In the extreme West, painted pottery has been found on various sites in south-eastern Europe, namely at Tripolje (in south-west Russia), at Petreny (in Bessarabia), at Schipenitz (in Bukovina), and on various other sites in Transylvania, Galicia and Moldavia.
The painted pottery penetrates as far south as Thessaly and had a wide distribution over the whole of the area mentioned. Both in form and in the types of decoration the Chinese ware is very suggestive of this Danube-Dnieper culture. The pottery of Asia Minor on the whole is less reminiscent of the Chinese ware, but elsewhere in western Asia close parallels can be found.
In the southern area parallels can be found with the Chinese pottery in Mesopotamia, Persia and Baluchistan. The Persian sites include Susa, the capital of ancient Elam, and the mounds of Pusht-i-Kuh, about zoom. west of Susa, the most important site being Tepe Mussain. The earliest pottery from Susa may be divided into two classes, of which the first and most ancient shows certain resemblances with the eastern Chinese pottery, but it should be noted that copper implements are found in all strata at Susa. A closer parallel is found, however, at Tepe Mussain, which in date probably coincides with the transition from Susa I. to Susa II. In the pottery from this site all the motives are found which occur in the eastern Chinese pottery, except the meander, which is limited to the finds from Kansu.
The finds of painted pottery from Mesopotamia, notably at Kish and Jemdet-Nazr by Langdon, in the north and by Campbell, Thomson and Woolley in the south provide the necessary link between Anau and the Persian sites.
The pottery from Baluchistan, published by Noetling, resem bles very closely the Honan pottery, and indeed is closer to it than any other of the western material except possibly some of the Mesopotamian wares.
First, the mass of evidence is against the suggestion that cop per-working took its origin in China and it seems more probable that it began in the Near East, although the exact location is un known at present. Secondly, the evidence of graves makes it un likely that the spread of painted pottery was due to any large migration (see CHINA, Ethnology), as those skeletons which have been examined belong to a type which is not very different from the modern Chinese, whereas the users of painted pottery in the Near East belong to an entirely different racial type. In what ever way the use of painted pottery spread, whether by trade or by the introduction of a superior culture by conquest or other method, the people who introduced it were able only to spread the culture but not to alter the physical type of the Far East. Thirdly, a definite distinction must be made between the prehistoric pot tery from Kansu and from the rest of China, as hitherto dis covered ; they can hardly be in the same line of evolution. It seems unlikely that there should be two independent sources of invention of such characteristic pottery and it seems reasonable to suggest that both represent successive waves of migrations of culture from elsewhere. The more eastern pottery has characters of its own in addition to those features which it shares with the Near East. It may, therefore, represent a local development, due possibly to isolation, of a technique originally acquired from out side. Such an hypothesis might serve to explain the absence of metal objects in the sites examined.
Recent evidence, produced by Teilhard, may serve as a latest possible date, i.e., the painted pottery may be dated by what came after it, not by what came before. In Honan itself inscriptions written on shell and ivory have been found which appear to belong to the Shang or Yin dynasty. These finds may probably be dated about the middle of the i8th century B.C. (c. 176o B.c.). If this date is correct then in the earlier half of the 2nd millennium B.C. the inhabitants of mid-China not only possessed a system of writ ing in hieroglyphic characters, but also used ivories and richly decorated bronzes. Neither of these features appears in the painted pottery culture. Therefore the latest possible date for this period is the end of the 3rd millennium B.C. Even this date, however, must be received with a good deal of caution ; with our present evidence, however, it seems the most reasonable. Thus the latest finds from Kansu would belong to the end of the 3rd millen nium B.C. and the other painted pottery would be earlier; the culture probably existing in China during most of the 3rd millen nium B.C. (300o-2000 B.C.), though in view of the marked differ ences between the Chinese painted pottery and that from else where it may have had its beginning at an even earlier date. No absolute chronology can be formulated until further evidence has been provided, especially in China and in Sinkiang (Chinese Turkistan).
The evidence so far discussed provides abundant evidence of relations between prehistoric China and the West but throws no light on relations between that country and the Far East. Recent work by the Geological Survey of Indo-China has revealed the existence of a well developed Neolithic culture in that area, which extends to the south, probably over most of south-eastern Asia, but whose northerly extension, as far as it is known at present, appears to lie on the borderland of China. This Indo-Chinese pre historic culture appears to be entirely separate from any Chinese culture so far known and to constitute an entirely separate archae ological province. That there should have been no form of cul ture contact is unlikely, but at present it is quite impossible to suggest any definite analogies between this culture and the cul tures which have been discussed above. Possibly the geographical conditions and the dense vegetation of tropical south-eastern Asia provided an insuperable barrier, possibly the apparent lack of con nection is due entirely to our lack of suitable explored sites.
To the north, however, we have more exact information. Japan has always been closely connected geographically with China and there is every reason to believe that the painted pottery culture of China was not without its influence on Japan. Most of the Japanese Neolithic pottery depends for its ornament either on in cised patterns or on relief. It is also characterized at least in one stage by the presence of figurines of quite a different type from anything so far found in China. The development of this early Japanese ware can be traced throughout the Stone Age in Japan and appears to represent a culture which is entirely independent of China, although it shows certain resemblances to some of the early ware from Manchuria and Korea. Painted pottery, how ever, does occur, although rarely. It seems not unlikely that such specimens represent either imported wares or possibly a local imi tation of Chinese ware. It is probable therefore that there was a certain amount of relation between Japan and China in Chalco lithic times. On the other hand fairly extensive excavations have been made in Japan so that it is probable that if the painted ware culture ever existed in Japan it would have been discovered. But there are indications that there was a close connection between Neolithic Japan, at a certain stage and part of the mainland oppo site. North-eastern China may, therefore, possibly represent a different cultural province from the rest of northern China, but here again we must wait for further excavations before any defi nite light can be thrown on the question. There are indications that at least part of this region remained in the Neolithic stage of culture when China itself, or at least the plain of China had developed a high civilization.
Relations Between Chalcolithic Culture and Chinese Civilization.—Andersson, to whom the discovery of this chap ter in Chinese prehistory is due, suggests that many features of Chinese civilization are directly descended from the Chalcolithic culture. First, he has suggested that the Li character is directly descended from a pictographic representation of a tripod pot. Arne, however, seems inclined to think that although the develop ment is striking the tripod pot may have been introduced into China once more with true Chinese civilization, from the West. Secondly, the knife at present used by knife-sharpeners in Peking is alleged to be directly descended from a knife found in the Neo lithic sites. A type of knife, akin to this form is used to-day to reap Kaoliang (a type of millet). Thirdly, the adze used to-day in Peking appears to be descended from the stone celt by way of the bronze celt. The question of the origin of wheat in China is too controversial and not sufficiently supported by evidence to be included here, but it is at least suggestive that its earliest recorded use is in conjunction with painted pottery in the Near East, so that wheat-growing may be part of the same culture complex of which painted pottery is a conspicuous element.
We thus have evidence of a Palaeolithic phase, represented by implements and hearths of the middle and upper Palaeolithic periods. This is followed by a widely spread Neolithic. Presum ably at a later date, from analogies elsewhere, there was in China a widely spread culture associated especially with painted pot tery, a period which can probably be divided into two parts. But little in this culture can be considered to be definitely Chi nese. Certain elements of it seem to have survived in the Chinese culture of the historic period, and as far as "we know at present the possessors of this culture did not differ essentially in their physique from the inhabitants of modern north China. Side by side with this culture there seems to have persisted in the north eastern part of the Chinese dominions a type of Neolithic culture, which probably persisted among barbarous tribes well into the historic period.