CHAPTER III.

THE BARBARIAN BACKGROUND

I

THE LANDS BETWEEN THE ROMAN EMPIRE AND CHINA

 

ROME was not overthrown in a day. It was the work of centuries that produced the great tidal wave which in the dawn of the Middle Ages was to sweep away those solid barriers (as they seemed) which the Roman Empire had erected wherever its frontiers stretched. For centuries before this decisive event, the great plains of Eastern Europe and the steppes or high plateaux of central Asia, which one day were to impart the final impetus, obscure as they are and often silent, were in fact the scene of vast upheavals of peoples which presaged ruin for distant lands. It is truer to say that one knows that it happened than how it happened, for certainly the scanty archaeological or historical evidence at our disposal for the period before the middle of the fourth century (the subject of this chapter) for the most part is neither very clear nor very conclusive. The task is to bring order out of this chaos as best one can, and to elucidate the handful of facts which represents the sum of our present knowledge.

There is one fact, however, which should be made clear from the first. The many territories comprised in the area between the Roman and Chinese Empires, though they lack unity either geographical or political, were nevertheless engaged throughout their whole extent in a perpetual travail which reacted on almost every one of their peoples, diverse as they were in origin and language, since almost all experienced indirectly the effects of events in which one or other of them was directly involved. Each impulse starting from one end of this immense ‘entre-deux’ passes from group to group and produces adjustments which affect the whole mass.

One group alone seems to stand as an exception to this rule, and watches the passage of the centuries unmoved, namely the central group of peoples of Indo-European language already long settled as farmers on the narrow strips of alluvial land which they inhabit to this day, extending round the inner circumference of the Tarim basin and as far as the northern slopes of the Tien-shan in Dzungaria. These interesting peoples, industrious and not unskilful, have watched the conquerors pass, and a succession of conquerors has, in effect, passed along the edges of the neighbouring deserts. But they themselves have cared for none of these things, indifferent to everything except the cultivation of their soil and always ready to accept any overlord whose demands did not exceed the act of submission and the payment of a tribute: even had they had the will to resist they lacked the power, scattered as they were in their little cities on an attenuated line more than a thousand miles long.

This group is today one of the best known of any, thanks to the notable discoveries of Aurel Stein, Grünwedel, A. von Le Coq, Pelliot and Hackin, who have excavated considerable remains of their ancient civilizations (though none, unfortunately, belonging to the period now under review), and have drawn from their hiding­places very many precious manuscripts which illustrate the intellectual side of their culture. Here indeed, in the very heart of Asia, survived the culture of the ancient Sacae from beyond the Oxus, of whom the most westerly branches spoke the East-Iranian language, while their kinsmen in the North-East, at Kucha, Karashahr and Turfan, spoke a different language formerly known as Tocharish but now more correctly called the Kuchean or Turfanese language, which seems to be nearer to the pure Indo-European group. But this interesting group of peoples does not itself play any active part in our period. Extraordinarily impressionable as it is, it engages our interest here only because it helps us to reconstruct some of the links in the long chain which runs over great deserts and high mountain-passes, and joins together the different populations scattered over those vast lands between the two Empires of Rome and China. For the Indo-Europeans of this central zone, which corresponds roughly to Chinese Turkestan, form a stable mass which hitherto has defied alike the nomads of the steppes and the armies of China; but they are also a pole of attraction, because the great routes across Asia go through their lands, and thus, while they lose nothing of their own individuality, they are a possible connecting-link between the Western and Eastern civilizations.

The chief of these Asiatic routes is the famous ‘Silk-route,’ known to us mainly through what is said of it by the Greek geo­grapher Ptolemy in the second century of our era, a route which for some time linked up the Chinese Empire with the world of Parthia and Rome. Starting in Syria it climbed to the Iranian highlands byway of Edessa, Nisibis, Ecbatana, Rhagae, Bactra and the mountains of the Comedae (the modern Kumedh, in the Pamirs), to reach, between Roshan and Ferghana, the silk-market at the spot that bore the name of the ‘Stone Tower’ where the caravans from the Levant exchanged their wares with the caravans from China. Ptolemy, relying on Marinus of Tyre, even relates how a Graeco-Syrian merchant (whom some take to have been a subject of the Parthian Empire), Maes Titianus by name, had dispatched agents along this route as far as the city of the ‘Seres.’ This city may belong to Kan-su, whether Si-an-fu, or even the capital of the Later Han, Lo-yang or Ho-nan-fu. About its identity geographers are not yet agreed, as is true also of the Roman names of the places which marked the course of the ‘Silk­route,’ Issedon Scythica and Issedon Serica, which maybe Kashgar and Kucha, or Kucha and Lou-Ian, north of the Lop-Nor. It was along this route that Buddhist missionaries from North-West India and Afghanistan, which then formed the Indo-Scythian Empire, brought to the Tarim basin the elements of what is called the Graeco-Buddhist civilization. From the first century to the fifth, Indian monks, in fact, unceasingly made their way by the passes of the Pamirs from Kashgar to Tun-huang, whether by Yarkand and Khotan on the south or by Kucha, Karashahr and Turfan on the north, as they pressed on to preach the gospel of Buddha, at first in all the Indo-European oases of the Tarim and later in China itself. These brought with them, as they had brought to the Indo-Scythian Empire of the Punjab and Afghanistan, that Graeco-Roman art in which they then found their means of expression.

It may, moreover, be observed how other elements of Graeco­romanization became added at the same time to the Graeco- Roman images of Buddha that were imported into this Tarim region. These elements were brought by trade all along the route from Antioch to Si-an-fu. Sir Aurel Stein, in his exploration of the oases mentioned above, has found—though for a period earlier than the fourth century—striking evidence of this double in­fluence. At Rawak, near Khotan, he has discovered bas-reliefs of the first century of our era carved in stucco with figures of bodhi­sattvas, notable for their truly hellenic nobility and harmonious proportions. At Rawak, too, and also at Yotkan (formerly Khotan) and in the valley of the Niya between Khotan and the Lop-Nor, he has found Roman sealings of the same period representing Pallas Athena armed with the thunderbolt and wearing the aegis and also Zeus, an Eros, a Heracles, and four-horsed chariots, finally Indo­Scythian coins from Afghanistan. At Miran, south of the Lop-Nor, classical influence is yet more clearly visible and displays the particular effect of Roman Asia Minor. Among the fragments of frescoes brought from this region by Sir Aurel Stein may be noted a Buddha followed by his monks which is in a purely Roman tradi­tion, beardless ‘angels’ or genii, some winged some wingless, in red mantles that recall the art of Pompeii, and figures also beard­less and wearing on their heads the Phrygian cap which gives them the appearance of Mithras. These frescoes, which belong to the third century of our era, afford striking analogies with the painting of Roman Syria and the Fayum of Imperial times. One of them bears an inscription in Indian characters which gives the name of the painter Tita, which may well be an indianized form of Titus.

II.

THE CIVILIZATION OF THE STEPPES

Noteworthy as are the facts that have been described, it would doubtless be rash to regard them as the dominating factor in deciding what were the influences that counted for most in the history of the relations between Asia and Europe in the period under review.

Indeed, the lands between China and the Roman frontiers are, above all, the meeting-point of two great streams of culture: on the one hand the Sarmatian, certainly the stronger of the two, rising near the Roman frontiers or, more exactly, from the regions occupied by the nomad Goths, and on the other hand the Turko- Mongol stream, rising near the frontiers of the Chinese Empire or, more exactly, from around the modern Ordos in lands occupied at that time by the eastern Hsiung-nu. The meeting of these two streams, reinforced by tributaries which in the same way rose either in the West or in the East, produced a kind of hybrid civili­zation common, as it seems, to all the nomads between the Roman Empire and China; and this, in fact, is an early symptom, and a clear one, of that close intercourse between widely differing peoples which has been suggested above, and which may be em­phasized here.

To get a clear view of it, one should turn to this same central region of Asia, not indeed to the settled Indo-European popula­tion of the Tarim, which remained unaffected by the various nomad civilizations, but to the Turko-Mongol peoples who frequented the high pastures round or between the upper Irtysh and the Tarbagatai Mountains on the one hand, and the upper Orkhon and the northern bend of the Huang-ho on the other. It is here that one can get the best impression of that art which for want of a better name is called or miscalled ‘the art of the steppes.’ In the period under review, it had long since acquired its essential character; but it continued to develop in matters of detail, for the sufficient reason that the nomads who were its exponents were always on the move. It appears most plainly in bronze plates for harness or armour, standard-staffs with the most interesting stylized animal decoration depicting, among other things, in arresting foreshortening, deer (cervidae) or wild beasts, strangely entwined in mortal combat; and finally, jewels and ornaments originally encrusted with glass beads, though usually only the sockets still survive. Naturally the products of this art are nearly always very difficult to date, but there seems very little doubt that in the third and fourth centuries it was still very far from being worked out, and throughout this period we encounter it constantly, with local variants or perhaps with one or other of its component elements predominating, over the length and breadth of the regions between China and the Roman Empire. The discoveries of the last fifty years, some of them very recent, allow us now to define, though still very imperfectly, some of its typical manifestations, and they fall into four main groups.

The Sarmatian group, which comes first, is close to the lands inhabited by the Goths or by peoples of Graeco-Roman civilization, and hence naturally shows clear traces of having been directly influenced by Hellenistic art. Iranian art, too, made its influence felt upon it, but in spite of the combination of influences this group furnishes (in the south-west) the farthest outpost of ‘the art of the steppes.’ This point has been developed in another chapter devoted to the Sarmatian peoples, both those of them who became amalgamated with the Goths and those who, farther East, kept their independence under the name of Alans; but it will perhaps be useful to emphasize the great interest of some of the treasures found at Novocherkask and now thought (with high probability) to belong to the third century of our era. One of the most interesting is a diadem of gold, with decoration in pearls, garnets and amethysts: in the centre is a large Hellenistic or Roman cameo, but on the upper rim are cervidae and trees showing the taste and manner of all the ornaments which are most characteristic of the art of the steppes. The thighs of the animals are hollowed into pear-shaped sockets intended for precious or semiprecious stones in a style that appears again on a silver belt from Maikop and on a number of objects worked in precious metal from Siberia and especially from the region of Lake Baikal. Other Sarmatian objects of the second and third centuries which have been found in proximity to these, such as scabbard-ornaments in the form of sledges or ringed sword-pommels, have been matched by similar discoveries in Eastern Asia, which indeed are no doubt their prototypes.

The second group of discoveries belongs to the countries on both sides of the Ural Mountains, in the neighbourhood of Perm and notably at Kachka in the heart of the Finno-Ugrian country, or, again, farther south in the province of Orenburg, or finally on the other side of the pass of Ekaterinburg near Shadrinsk. The finds in these parts, some of them at least dating from the third and early fourth centuries, are closely related to the products of Sarmatian art; they include buckles, necklaces, gold and silver rings, fibulae, swords, filigree earrings and enamelled glass beads. Comparable with these finds are those of Pianibor near Sarapul about two hundred and twenty miles below Perm on the River Kama, where the conical earrings (made of a metal thread wound into spirals), the pendants in die shape of bird or horse, or the bronze brooches shaped like epaulettes, characteristic as they are, have points in common with the finds of the Ural district.

Much farther east, on the upper Yenisei and more particularly near Minusinsk, various excavations, unhappily without method, have brought to light incidentally a respectable number of objects that can be assigned probably to the second, third and fourth cen­turies of our era. Especially noteworthy are knives of bronze or iron, excellent bronze bowls, and pieces of armour or harness decorated in the animal style so characteristic of the civilization of the steppes.

Finally, on the borders of China round the bend of the Huang- ho and in Ordos, a fourth group, closely related to the third, shows more perfect and more highly developed examples of this same animal style of art, and among them excellent plaques with polycephalic animals and with human heads, to be dated probably to the third century, and also bronze or iron knives and cylindrical bowls not unlike those of Minusinsk.

To this highly schematic picture should be added some mention at least of various sporadic groups such as that revealed by the finds at Kosibejevo in the province of Tambov to the east of the upper Don, or the group from the province of Kaluga on the River Oka south of Moscow. Both yield a great number of objects be­longing almost certainly either to the third century or perhaps (in Kaluga at any rate) to the fourth, and introduce us to an art which stands between the Sarmato-Gothic and that of the countries occu­pied by the Germans south of the Baltic. Fibulae from Kosibejevo, or those in the shape of triangles or horseshoes from Kaluga, neck­laces with perforated terminal discs (at Kosibejevo), or enamelled jewels, bracelets and crescents—all these things, and others, make up a curious hybrid art that certainly implies a regular traffic across the barbarian hinterland between the Baltic provinces and the countries of the Black Sea.

It is this that gives to certain products of the art which the Russian excavations are slowly bringing to light their air of be­longing together in a way that often seems disconcerting. But through all this tangle of interacting influences one thread can be followed by the most casual eye: no one can deny the extra­ordinary continuity of an art of which on the one hand the manifestations appear sufficiently varied to prevent confusion (the Ural and Minusinsk groups, for example, are perfectly distinct), while, on the other hand, its extremes, so to speak, meet in the oddest way, as one can see in the Sarmatian group and the group from Ordos, where the same armour-plates and lance-handles repeat themselves in the same animal style.

The first general impression that one gets from this common art is that it is a stereotyped art. In our period its subjects seem already fixed and time-honoured, the result of the meeting and blending of many ancestral influences. It has been affected, or even shaped (in degrees and proportions varying with place or period), by the ancient civilizations of Greece and Iran and China; yet it is perhaps in our period most of all that it shows itself as at once homogeneous and mature. It has its realism, but above all it relies on simplification, as can best be seen from a glance at those little heads of foxes or asses reduced to their essential features, or at the pole-tops in the shape of cervidae that are so common in our museums. Moreover, its principal aim is decoration. Even in those dramatic contests of animals in which some of the artists excel, even when they portray the terrifying spectacle of these ferocious creatures tearing their prey with sharp fangs or twisting their bodies with all the power of their muscles, still they cannot resist the temptation to frame the scene, or even to obscure it, with a regular network of curved lines. The antlers of the cervidae or the horses’manes issue in spirals or ringlets; horns and tails merge into foliage or into heads of birds or gryphons; the nostrils curl round in spirals. The animal loses itself in its own decoration, dense as of tropical undergrowth. Decorative indeed is this art of the steppes, in its very essence, decoration is its one purpose, and one must admit that it succeeds. Even the stylization in which it takes such delight is only another means to this end.

It may well be asked whether there is no other conclusion to be drawn from a study of the works which it has produced, whether the historian is not justified in trying to look beyond the passive material of the archaeological finds to a glimpse of that life itself which he is denied by the absence of contemporary texts, though the want is satisfied in the periods immediately before and after. If it is true that a particular type of art reveals a particular type of culture, one can see reflected here the features of those motley hordes of nomads, sprung from different stocks but leading the same kind of life, the peoples who in the third and fourth centuries roamed the wide steppes between Rome and China. They are horsemen, tireless horsemen, like the nomads who for centuries are to pour into China and Europe: always and everywhere pieces of harness are among the objects that meet our eye. Nomads and drovers with no fixed abode (and consequently no cities), they drive before them their herds of horses, cattle and sheep, from the steppe to the hills and from hills to steppe in time with the seasons. They are bandits, like all their kind, and they go armed, as ready to attack others as to defend themselves, with their bows and keen arrows—arrowheads are plentiful among the finds—with their swords and their long lances, decorated with these stylized animals which were used perhaps as insignia or as totems. They live by hunting the wild beasts that abound in the desert, they pitch the tents of felt used by Turko-Mongol peoples from time immemorial, they are followed by their women wearing gay dresses and ornaments sparkling with glass beads (those beads which we know today only by their empty sockets) and no doubt riding with their children and belongings in the same chariots in which they are to appear later. From the picture of the Hsiung-nu of Mongolia which survives from the second century b.c. in the Chinese chronicle of the Han dynasty to the picture of the Huns of the Danube frontier in Ammianus Marcellinus at the end of the fourth century of our era, the same character lives and survives, as does the art which is its product.

III.

 THE WESTWARD EBB OF THE BARBARIANS OF THE STEPPES

Among and between these nomads, then, there were perpetual eddies. Asia was in flux, and the nations of the steppes were ever ready to take the tide in their affairs which might lead them on to better lands and to booty. The stronger hordes, those with a more energetic ruler to bring them to greater victories, succeeded in subduing the weaker one by one, and enrolling them in their turn to join in assailing fresh tribes or winning fresh booty unless there arose, from China, a strong power to impose order on all alike.

To keep out all these nomads the Chinese emperors had reared a whole system of defences, a veritable limes, which was initiated, in 121 bc, by the great Han monarch Wu-ti. This limes, which in more than one feature recalls some sectors of the Roman limes, e.g. in the area of the Agri Decumates, had been formed, since 108 bc, of a continuous line of small forts and military colonies from the present city of Su-chou, in Kan-su, as far as Tun-huang, on the border of modern Chinese Turkestan, in the direction of the Tarim. Along this line the Chinese were thenceforward incessantly engaged in holding in check the barbarians, whose chief element was the Turko-Mongol nomad tribes whom they called without distinction by the name Hsiung-nu, probably the same tribes who were called Huns by the people of the West. It was at the western end of this line, on the other hand, that they strove to secure the control of the ‘Silk-route’ and the suzerainty over the Tarim basin, which the Hsiung-nu had never ceased to challenge. They, therefore, sent expeditions, of which the most famous had been under the second—the ‘later’—Han dynasty, when two great Chinese generals, Tou Ku and Pan Ch’ao, had been enabled to crush the Hsiung-nu in a.d. 73 near Lake Barkol and then in 75 near Yar, some distance from Turfan, after a counter-offensive of the barbarians who had succeeded in conquering Kashgaria. The last quarter of the first century had been filled with unceasing struggles of the two generals, above all of Pan Ch’ao, to keep the ‘Silk-route’ free and to control it. Indeed, when Pan Ch’ao was made governor-general at Kucha in the closing years of the century, he seems to have turned his eyes to the countries of the West, beyond the Tarim, if it is true that he sent his lieutenant Kan Ying to make a reconnaissance to the borders of the Parthian Empire and charged him in a.d. 97 to collect information about the distant Ta-ts’in, the Roman Empire.

But despite the efforts of Pan Ch’ao and, later, those of his son Pan Yong to maintain the Chinese positions in these regions, the troops of the Celestial Empire were compelled, in the course of the second century, to make a deep withdrawal from the Tarim and then from the limes, in order to return to the defensive, and even this defensive was maintained with difficulty when, in the last quarter of the century, the Han dynasty, weakened by risings of which the most serious was that of the ‘Yellow-hats’ that broke out in Chihli and Ho-nan in 184, sank into anarchy. Throughout all the closing years of the century and the beginning of the third century rebel military chiefs were making themselves masters in the northern, southern and western provinces. It was the beginning of the unhappy period known as ‘the Three Kingdoms’ (Wei, Wu and Shu), and for nearly a century China was a prey to civil war and consequently unable to police the steppes. The nomads, also, had a free field in Central Asia. This was the great period of the Tungus-Mongol (or Mongol-Tungus) hordes called the Sien-pi, from the middle of the second century masters of Mongolia proper, where they had overwhelmed most of the Hsiung-nu. From there they had succeeded in extending their power by degrees from the peninsula of Liao-tung (north-east of the gulf of Chihli) as far as the Gobi desert, driving back such of the Hsiung-nu as refused to submit to them, some towards the Altai Mountains, others to the borders of China (of which some of these Hsiung-nu had for quite a time become the more or less loyal allies) hard by the Great Wall, in Ordos and to the north of the province of Shan-si.

These very troubled times, in which one gets the impression that the whole of Asia was in ferment, lasted into the second half of the third century. Then the unity of China was restored after a fashion by the Ssu-ma family, who usurped the imperial throne under the name of the Chin dynasty. Once again, as in the Han period, the prestige of China began to make itself felt in Central Asia, where, in their anxiety to escape conquest by the Sien-pi, the Indo-European princes of the Tarim basin did. homage to the Celestial Empire (ad 285)—their saviour, as they hoped. There was also a tendency for diplomatic relations to be formed at that time between the Chinese Empire and Rome by way of the Tarim region. The annalists, at least, indicate that in a.d. 284 the Chin emperor received presents sent by the Ta-ts’in (the Roman Empire). But this was only a pause. The Celestial Empire, so far from saving others, could only save itself at the cost of admitting within the Great Wall as foederati those of the Hsiung-nu who had retired southwards before the Sien-pi. They settled more especially to the north of Shan-si, and, as in the Roman Empire, these foreign protectors were apt to be dangerous if the central govern­ment should weaken again, inasmuch as the Chinese, like the Romans, in the course of the third century were obliged to strengthen their army with a number of Hsiung-nu chiefs or generals, who received Chinese titles.

At the beginning of the fourth century the decline of the Chin dynasty hastened the catastrophe, One of the Hsiung-nu generals who had been admitted into the Empire, Liu-yuan, established his authority over all the Hsiung-nu foederati and installed himself in ad 303—4 at T’ai-yuan, the capital of Shan-si, where he proclaimed himself Emperor in 308. In 311 his son Liu-ts’ung, a kind of ferocious genius, invaded the province of Ho-nan, sacked the Imperial capital Lo-yang, captured the feeble Emperor Chin Huai-ti himself, and had him executed two years later. This was the signal for barbarians from near and far to fall upon China as their prey and to fight over her provinces and her plunder: among the most formidable were the Hsiung-nu, Sien-pi and T’o-pa (another group of Turk or Mongol tribes). The details of this struggle are both bloody and obscure, but they are of much less importance than the stark developments which they produced: the Chinese Empire was systematically dismembered, and its northern provinces Chihli, Shan-si, Shen-si, Shan-tung and Ho­nan were torn from it. This vast and brutal operation served to occupy all the forces of the steppes, a fact which probably explains the temporary lull in the movements of peoples on the eastern frontiers of the Roman Empire in Europe.

During the fourth century, however, the situation gradually changes, and becomes more complex. On the one hand, the wars of neighbouring ethnic groups, and within the groups them­selves, started fresh movements which had their repercussions in the steppes; but also the tracts of Mongolia, partially emptied of Sien-pi when they conquered Chihli and Shan-tung (where they split up in exhaustion), were now occupied by new hordes, pro­bably of Mongols, and doubtless from the extremities of the modern .Manchuria. They were the Juan-juan, whose swift ex­pansion in Mongolia became in less than fifty years a serious threat to all the Hsiung-nu who still remained: these remnants had breathed more freely since the descent of the Sien-pi upon China, but they now began to feel the pressure of the new arrivals, whose chiefs asserted their superiority by rejecting the Hsiung-nu title of shan-yu in favour of the purely Mongol title of khan, and proceeded to make themselves masters of all Mongolia and extend their power from the gulf of Corea to the Altai Mountains. Moreover, they drove before them to the south-west, so as to press upon Sogdiana and Bactria, other peoples who now enter the stage of history, notably a group nearly related to the Juan-juan, the Hephthalites (the Yeta of the Chinese), who, under the name of Huns which they shared with the Hsiung-nu, were destined to cross swords with Persia and India. Similarly some of the bar­barians established in Shen-si in north-western China, seeing no easy conquests in the eastern provinces, set out again in the direc­tion of Central Asia, where one of their chiefs, Fu Chien, later (in ad 382) ruled over lands extending to Karashahr and Kucha. In this way the passage became barred on every side.

It was at this time that the groups of the Hsiung-nu, who at the beginning of the third century had been pushed back by the Sien-pi towards the Altai Mountains, advanced towards the ex­treme west of Asia. The details of the fearful struggle of nomads that ensued are obscure; but its consequences are clear. Driven from the Altai range the Hsiung-nu ended by crossing the steppes which extend to the north of Lake Balkash and the Aral Sea. There two choices lay before them, the route to the south-west towards the Jaxartes valley and the rich lands of Sogdiana and Bactria, or that to the west straight on towards the Volga. But actually at this time, their freedom of choice was restricted. The Jaxartes valley had been occupied for more than three centuries by the Yüeh-chih (the Indo-Scythians mentioned by Greek historians), who had been driven from Mongolia or near it about 170 bc, and after a rough passage had finally settled here, where they acquired a veneer of Greek and Hindu culture and founded a powerful State which at one period, in the first century of our era, had for a time extended beyond the upper Jaxartes valley itself over the valley of the Oxus and the Hindu Kush, the lands watered by the Indus, Pamir, Kashmir, the Punjab, and the plain of the Ganges above Benares. Certainly this was already ancient history, and partially effaced by their reverses from the be­ginning of the third century onwards in obscure wars against the Sassanids of Persia and the Gupta princes of India; but hard pressed as they were by their enemies in the south, the Yüeh-chih could still hold the regions of the Jaxartes securely enough, and this being so the western Hsiung-nu had no alternative but to keep straight on. About a.d. 355 they advanced westwards with the intention of forcing their way at all costs across the great plain of Russia.

This was the date at which the ‘great invasions’ of the West begin. The Hsiung-nu, who had hitherto made history exclusively in the East, now suddenly proceeded, under their new name of Huns, to contribute their chapter to the history of Europe.

To study the nations of the steppes, though it may set us more problems than it resolves, does answer some questions. The barbarian invasions of the early Middle Ages are in a sense no more than a sequel of the struggles which for so long had had the steppes of Asia for their main battlefield, or at any rate the only battlefield of which historical texts (the Chinese annalists) have anything to tell us. Our literary sources in general tell us of the developments in Mongolia and on the Yellow River and of those on the banks of the Danube; but between the two tales there is a gap. Here archaeology comes to our aid and supplies the missing link in our chain of evidence: it is certainly not pure coincidence that the early medieval art in Europe known as ‘barbarian art’ is really only the continuation, almost without change, of what we have hitherto called ‘the art of the steppes.’

It is impossible without going far beyond the chronological limits of this chapter to give this fact its full and proper emphasis, though the researches of the experts increasingly illustrate it. But one need only remember that it is from the soil of Hungary and Wallachia, from tombs in the lands occupied by our European Huns after they had broken down the barrier of the Goths, that perhaps the most numerous and certainly the most characteristic specimens have been recovered of an art that, with different manifestations and traditions, reappears in the tombs of the Goths, Franks, Burgundians, Vandals, Lombards, and of all the Germanic peoples of the West. Bronze cauldrons of the Huns identical with those of Minusinsk and Ordos have been found by the River Kapos in the heart of Hungary; the same knives also; and the jewels in glass­ware, the fibulae, the clasps, the perforated baldric-plates, the pins with animal heads which characterized the art of early medieval Europe are nothing if not the ancient art of the steppes. The steppes, it seems, have overflowed over Europe. The historian is thus Justified in crossing the arbitrary boundaries of two so-called continents which are really the complement of each other, in quest of the principle of continuity without which history itself becomes a riddle without an answer.