[665] _Geogr. Journ._ 1895, II. p. 460.

[666] _Journ. Anthrop. Soc._ 1897, p. 47 sq.

[667] _Ibid._ p. 58.

[668] Ripley and Dana, _Amer. Cyc._ IX. 538.

[669] _Shogun_ from _Sho_ = general, and _gun_ = army, hence Commander-in-chief; _Mikado_ from _mi_ = sublime, and _kado_ = gate, with which cf. the "Sublime Porte" (J. J. Rein, _j.a.pan nach Reisen u.

Studien,_ 1881, I. p. 245). But Mikado has become somewhat antiquated, being now generally replaced by the t.i.tle _Kotei,_ "Emperor."

[670] Keane's _Asia_, I. p. 487.

CHAPTER IX

THE NORTHERN MONGOLS (_continued_)

The Finno-Turki Peoples--a.s.similation to the Caucasic Type--Turki Cradle--Ural-Altaian Invasions--The Scythians--Parthians and Turkomans--Ma.s.sagetae and Yue-chi--Indo-Scythians and Graeco-Baktrians--Dahae, Jat, and Rajput Origins--The "White Huns"--The Uigurs--Orkhon Inscriptions--The a.s.sena Turki Dynasty--Toghuz-Uigur Empire--Kashgarian and Sungarian Populations--The Oghuz Turks and their Migrations--Seljuks and Osmanli--The Yakuts--The Kirghiz--Kazak and Kossack--The Kara-Kirghiz--The Finnish Peoples--Former and Present Domain--Late Westward Spread of the Finns--The Bronze and Iron Ages in the Finnish Lands--The Baltic Finns--Relations to Goths, Letts, and Slavs--Finno-Russ Origins--Tavastian and Karelian Finns--The Kwaens--The Lapps--Samoyeds and Permian Finns--Lapp Origins and Migrations--Temperament--Religion--The Volga Finns--The Votyak Pagans--Human Sacrifices--The Bulgars--Origins and Migrations--An Ethnical Transformation--Great and Little Bulgaria--Avars and Magyars--Magyar Origins and early Records--Present Position of the Magyars--Ethnical and Linguistic Relations in Eastern Europe.

In a very broad way all the western branches of the North Mongol division may be comprised under the collective designation of Finno-Turki Mongols. Jointly they const.i.tute a well-marked section of the family, being distinguished from the eastern section by several features which they have in common, and the most important of which is unquestionably a much larger infusion of Caucasic blood than is seen in any of the Mongolo-Tungusic groups. So p.r.o.nounced is this feature amongst many Finnish as well as Turkish peoples, that some anthropologists have felt inclined to deny any direct connection between the eastern and western divisions of Mongolian man and to regard the Baltic Finns, for instance, rather as "Allophylian Whites" than as original members of the yellow race. Prichard, to whom we owe this now nearly obsolete term "Allophylian," held this view[671], and even Sayce is "more than doubtful whether we can cla.s.s the Mongols physiologically with the Turkish-Tatars [the Turki peoples], or the Ugro-Finns[672]."

It may, indeed, be allowed that at present the great majority of the Finno-Turki populations occupy a position amongst the varieties of mankind which is extremely perplexing for the strict systematist. When the whole division is brought under survey, every shade of transition is observed between the Siberian Samoyeds of the Finnic branch and the steppe Kirghiz of the Turki branch on the one hand, both of whom show Mongol characters in an exaggerated form, and on the other the Osmanli Turks and Hungarian Magyars, most of whom may be regarded as typical Caucasians. Moreover, the difficulty is increased by the fact, already pointed out, that these mixed Mongolo-Caucasic characters occur not only amongst the late historic groups, but also amongst the earliest known groups--"Chudes," Usuns, Uigurs and others--who may be called Proto-Finnish and Proto-Turki peoples. But precisely herein lies the solution of the problem. Most of the region now held by Turki and Finnish nations was originally occupied by long-headed Caucasic men of the late Stone Ages (see above). Then followed the Proto-Mongol intruders from the Tibetan table-land, who partly submerged, partly intermingled with their neolithic neighbours, many thus acquiring those mixed characters by which they have been distinguished from the earliest historic times. Later, further interminglings took place according as the Finno-Turki hordes, leaving their original seats in the Altai and surrounding regions, advanced westwards and came more and more into contact with the European populations of Caucasic type.

We may therefore conclude that the majority of the Finno-Turki were almost from the first a somewhat mixed race, and that during historic times the original Mongol element has gradually yielded to the Caucasic in the direction from east to west. Such is the picture now presented by these heterogeneous populations, who in their primeval eastern seats are still mostly typical Mongols, but have been more and more a.s.similated to the European type in their new Anatolian, Baltic, Danubian, and Balkan homes.

Observant travellers have often been impressed by this progressive conformity of the Mongolo-Turki to Europeans. During his westward journey through Central Asia Younghusband, on pa.s.sing from Mongolia to Eastern Turkestan, found that the people, though tall and fine-looking, had at first more of the Mongol cast of feature than he had expected.

"Their faces, however, though somewhat round, were slightly more elongated than the Mongol, and there was considerably more intelligence about them. But there was more roundness, less intelligence, less sharpness in the outlines than is seen in the inhabitants of Kashgar and Yarkand." Then he adds: "As I proceeded westwards I noticed a gradual, scarcely perceptible, change from the round of a Mongolian type to a sharper and yet more sharp type of feature.... As we get farther away from Mongolia, we notice that the faces become gradually longer and narrower; and farther west still, among some of the inhabitants of Afghan Turkestan, we see that the Tartar or Mongol type of feature is almost entirely lost[673]." To complete the picture it need only be added that still further west, in Asia Minor, the Balkan Peninsula, Hungary, and Finland, the Mongol features are often entirely lost. "The Turks of the west have so much Aryan and Semitic blood in them, that the last vestiges of their original physical characters have been lost, and their language alone indicates their previous descent[674]."

Before they were broken up and dispersed over half the northern hemisphere by Mongol pressure from the east, the primitive Turki tribes dwelt, according to Howorth, mainly between the Ulugh-dagh mountains and the Orkhon river in Mongolia, that is, along the southern slopes and spurs of the Altai-Sayan system from the head waters of the Irtysh to the valleys draining north to Lake Baikal. But the Turki cradle is shifted farther east by Richthofen, who thinks that their true home lay between the Amur, the Lena, and the Selenga, where at one time they had their camping-grounds in close proximity to their Mongol and Tungus kinsmen. There is nothing to show that the Yakuts, who are admittedly of Turki stock, ever migrated to their present northern homes in the Lena basin, which has more probably always been their native land[675].

But when they come within the horizon of history the Turki are already a numerous nation, with a north-western and south-eastern division[676], which may well have jointly occupied the whole region from the Irtysh to the Lena, and both views may thus be reconciled. In any case the Turki domain lay west of the Mongol, and the Altai uplands, taken in the widest sense, may still be regarded as the most probable zone of specialisation for the Turki physical type. The typical characteristics are a yellowish white complexion, a high brachycephalic head, often almost cuboid, due to parieto-occipital flattening (especially noticeable among the Yakuts), an elongated oval face, with straight, somewhat prominent nose, and non-Mongolian eyes. The stature is moderate, with an average of 1.675 m. (5 ft. 6 in.), and a tendency to stoutness.

Intermediate between the typical Turki and the Mongols Hamy places the Uzbegs, Kirghiz, Bashkirs, and Nogais; and between the Turks and Finns those extremely mixed groups of East Russia commonly but wrongly called "Tartars," as well as other transitions between Turk, Slav, Greek, Arab, Osmanli of Constantinople, Kurugli of Algeria and others, whose study shows the extreme difficulty of accurately determining the limits of the Yellow and the White races[677].

a.n.a.logous difficulties recur in the study of the Northern (Siberian) groups--Samoyeds, Ostyaks, Voguls and other Ugrians--who present great individual variations, leading almost without a break from the Mongol to the Lapp, from the Lapp to the Finn, from Finn to Slav and Teuton. Thus may be shown a series of observations continuous between the most typical Mongol, and those aberrant Mongolo-Caucasic groups which answer to Prichard's "Allophylian races." Thus also is confirmed by a study of details the above broad generalisation in which I have endeavoured to determine the relation of the Finno-Turki peoples to the primary Mongol and Caucasic divisions.

Peisker's description of the Scythian invasions of Irania[678] may be taken as typical of the whole area, and explains the complexity of the ethnological problems. The steppes and deserts of Central Asia are an impa.s.sable barrier for the South Asiatics, the Aryans, but not for the North Asiatic, the Altaian; for him they are an open country, providing him with the indispensable winter pastures. On the other hand, for the South Asiatic Aryan these deserts are an object of terror, and besides he is not impelled towards them as he has winter pastures near at hand.

It is this difference in the distance of summer and winter pastures that makes the North Asiatic Altaian an ever-wandering herdsman, and the grazing part of the Indo-European race cattle-rearers settled in limited districts. Thus, while the native Iranian must halt before the trackless region of steppes and deserts and cannot follow the well-mounted robber-nomad thither, Iran itself is the object of greatest longing to the nomadic Altaian. Here he can plunder and enslave to his heart's delight, and if he succeeds in maintaining himself for a considerable time among the Aryans, he learns the language of the subjugated people and, by mingling with them, loses his Mongol characteristics more and more. If the Iranian is now fortunate enough to shake off the yoke, the dispossessed iranised Altaian intruder inflicts himself upon other lands. So it was with the Scythians. Leaving their families behind in the South Russian steppes, the Scythians invaded Media _c._ B.C. 630, and advanced into Mesopotamia as far as Egypt.

In Media they took Median wives and learned the Median language. After being driven out by Cyaxares, on their return, some 28 years later, they met with a new generation, the offspring of the wives and daughters whom they had left behind, and slaves of an alien race. A hundred and fifty years later Hippocrates remarked their yellowish red complexion, corpulence, smooth skins, and their consequent eunuch-like appearance--all typically Mongol characteristics. Hippocrates was the most celebrated physician and natural philosopher of the ancient world.

His evidence is unshakeable and cannot be invalidated by the Aryan speech of the Scythians. Their Mongol type was innate in them, whereas their Iranian speech was acquired and is no refutation of Hippocrates'

testimony. On the later Greek vases from South Russian excavations they already appear strongly demongolised and the Altaian is only suggested by their hair, which is as stiff as a horse's mane--hence Aristotle's epithet [Greek: euthytriches]--the characteristic that survives longest among all Ural-Altaian hybrid peoples.

E. H. Parker unfortunately lent the weight of his authority to the statement that the word "Turko" [Turki] "goes no farther back than the fifth century of our era," and that "so far as recorded history is concerned the name of Turk dates from this time[679]." But Turki tribes bearing this national name had penetrated into East Europe hundreds of years before that time, and were already seated on the Tanais (Don) about the new era. They are mentioned by name both by Pomponius Mela[680] and by Pliny[681], and to the same connection belonged, beyond all doubt, the warlike _Parthians_, who 300 years earlier were already seated on the confines of Iran and Turan, routed the legions of Cra.s.sus and Antony, and for five centuries (250 B.C.-229 A.D.) usurped the throne of the "King of Kings," holding sway from the Euphrates to the Ganges, and from the Caspian to the Indian Ocean. Direct descendants of the Parthians are the fierce Turkoman nomads, who for ages terrorised over all the settled populations encircling the Aralo-Caspian depression. Their power has at last been broken by the Russians, but they are still politically dominant in Persia[682]. They have thus been for many ages in the closest contact with Caucasic Iranians, with the result that the present Turkoman type is shown by J. L. Yavorsky's observations to be extremely variable[683].

Both the Parthians and the _Ma.s.sagetae_ have been identified with the _Yue-chi_, who figured so largely in the annals of the Han dynasties, and are above mentioned as having been driven west to Sungaria by the Hiung-nu after the erection of the Great Wall. It has been said that, could we follow the peregrinations of the Yue-chi bands from their early seats at the foot of the Kinghan mountains to their disappearance amid the snows of the Western Himalayas, we should hold the key to the solution of the obscure problems a.s.sociated with the migrations of the Mongolo-Turki hordes since the torrent of invasion was diverted westwards by Shih Hw.a.n.g Ti's mighty barrier. One point, however, seems clear enough, that the Yue-chi were a different people both from the Parthians who had already occupied Hyrcania (Khorasan) at least in the third century B.C., if not earlier, and from the Ma.s.sagetae. For the latter were seated on the Yaxartes (Sir-darya) in the time of Cyrus (sixth century B.C.), whereas the Yue-chi still dwelt east of Lake Lob (Tarim basin) in the third century. After their defeat by the Hiung-nu and the Usuns (201 and 165 B.C.), they withdrew to Sogdiana (Transoxiana), reduced the _Ta-Hia_ of Baktria, and in 126 B.C.

overthrew the Graeco-Baktrian kingdom, which had been founded after the death of Alexander towards the close of the fourth century. But in the Kabul valley, south of the Hindu-Kush, the Greeks still held their ground for over 100 years, until Kadphises I., king of the Kushans--a branch of the Yue-chi--after uniting the whole nation in a single Indo-Scythian state, extended his conquests to Kabul and succeeded Hermaeus, last of the Greek dynasty (40-20 B.C.?). Kadphises' son Kadaphes (10 A.D.) added to his empire a great part of North India, where his successors of the Yue-chi dynasty reigned from the middle of the first to the end of the fourth century A.D. Here they are supposed by some authorities to be still represented by the _Jats_ and _Rajputs_, and even Prichard allows that the supposition "does not appear altogether preposterous," although "the physical characters of the Jats are very different from those attributed to the Yuetschi [Yue-chi] and the kindred tribes [Suns, Kushans, etc.] by the writers cited by Klaproth and Abel Remusat, who say that they are of sanguine complexion with blue eyes[684]."

We now know that these characters present little difficulty when the composite origin of the Turki people is borne in mind. On the other hand it is interesting to note that the above-mentioned Ta-Hia have by some been identified with the warlike Scythian Dahae[685], and these with the Dehiya or Dhe one of the great divisions of the Indian Jats. But if Rawlinson[686] is right, the term _Dahae_ was not racial but social, meaning _rustici_,--the peasantry as opposed to the nomads; hence the Dahae are heard of everywhere throughout Irania, just as _Dehwar_[687]

is still the common designation of the Tajik (Persian) peasantry in Afghanistan and Baluchistan. This is also the view taken by de Ujfalvy, who identifies the Ta-Hia, not with the Scythian Dahae, or with any other particular tribe, but with the peaceful rural population of Baktriana[688], whose reduction by the Yue-chi, possibly Strabo's Tokhari, was followed by the overthrow of the Graeco-Baktrians. The solution of the puzzling Yue-chi-Jat problem would therefore seem to be that the Dehiya and other Jats, always an agricultural people, are descended from the old Iranian peasantry of Baktriana, some of whom followed the fortunes of their Greek rulers into Kabul valley, while others accompanied the conquering Yue-chi founders of the Indo-Scythian empire into northern India.

Then followed the overthrow of the Yue-chi themselves by the _Ye-tha_ (_Ye-tha-i-li-to_) of the Chinese records, that is, the _Ephthalites_, or so-called "White Huns," of the Greek and Arab writers, who about 425 A.D. overran Transoxiana, and soon afterwards penetrated through the mountain pa.s.ses into the Kabul and Indus valleys. Although confused by some contemporary writers (Zosimus, Am. Marcellinus) with Attila's Huns, M. Drouin has made it clear that the Ye-tha were not Huns (Mongols) at all, but, like the Yue-chi, a Turki people, who were driven westwards about the same time as the Hiung-nu by the Yuan-yuans (see above). Of Hun they had little but the name, and the more accurate Procopius was aware that they differed entirely from "the Huns known to us, not being nomads, but settled for a long time in a fertile region." He speaks also of their white colour and regular features, and their sedentary life[689] as in the Chinese accounts, where they are described as warlike conquerors of twenty kingdoms, as far as that of the A-si (Arsacides, Parthians), and in their customs resembling the Tu-Kiu (Turks), being in fact "of the same race." On the ruins of the Indo-Scythian (Yue-chi) empire, the White Huns ruled in India and the surrounding lands from 425 to the middle of the sixth century. A little later came the Arabs, who in 706 captured Samarkand, and under the Aba.s.sides were supreme in Central Asia till scattered to the winds by the Oghuz Turki hordes.

From all this it has been suggested that--while the Baktrian peasants entered India as settlers, and are now represented by the agricultural Jats--the Yue-chi and Ye-tha, both of fair Turki stock, came as conquerors, and are now represented by the Rajputs, "Sons of Kings," the warrior and land-owning race of northern India. It is significant that these Thakur, "feudal lords," mostly trace their genealogies from about the beginning of the seventh century, as if they had become Hinduized soon after the fall of the foreign Ye-tha dynasty, while on the other hand "the country legends abound with instances of the conflict between the Rajput and the Brahman in prehistoric times[690]." This supports the conjecture that the Rajputs entered India, not as "Aryans" of the Kshatriya or military caste, as is commonly a.s.sumed, but as aliens (Turki), the avowed foes of the true Aryans, that is, the Brahman or theocratic (priestly) caste. Thus also is explained the intimate a.s.sociation of the Rajputs and the Jats from the first--the Rajputs being the Turki leaders of the invasions; and the Jats their peaceful Baktrian subjects following in their wake.

The theory that the haughty Rajputs are of unsullied "Aryan blood" is scarcely any longer held even by the Rajputs themselves; they are undoubtedly of mixed origin. But the definite physical type which H. H.

Risley[691] describes as characteristic of Rajputs and Jats in the Kashmir Valley, Punjab and Rajputana, shows them to be wavy-haired dark-skinned dolichocephals, linked rather with the "Caucasic" than the "Mongolian" division.

Nearly related to the White Huns were the _Uigurs_, the _Kao-che_ of the Chinese annals, who may claim to be the first Turki nation that founded a relatively civilised State in Central Asia. Before the general commotion caused by the westward pressure of the Hiung-nu, they appear to have dwelt in eastern Turkestan (Kashgaria) between the Usuns and the Sacae, and here they had already made considerable progress under Buddhist influences about the fourth or fifth century of the new era.

Later, the Buddhist missionaries from Tibet were replaced by Christian (Nestorian) evangelists from western Asia, who in the seventh century reduced the Uigur language to written form, adapting for the purpose the Syriac alphabet, which was afterwards borrowed by the Mongols and the Manchus.

This Syriac script--which, as shown by the authentic inscription of Si-ngan-fu, was introduced into China in 635 A.D.--is not to be confused with that of the Orkhon inscriptions[692] dating from 732 A.D., and bearing a certain resemblance to some of the Runic characters, as also to the Korean, at least in form, but never in sound. Yet although differing from the Uiguric, Prof. Thomsen, who has successfully deciphered the Orkhon text, thinks that this script may also be derived, at least indirectly through some of the Iranian varieties, from the same Aramean (Syriac) form of the Semitic alphabet that gave birth to the Uiguric[693].

It is more important to note that all the non-Chinese inscriptions are in the Turki language, while the Chinese text refers by name to the father, the grandfather, and the great-grandfather of the reigning Khan Bilga, which takes us back nearly to the time when Sinjibu (Dizabul), Great Khan of the Altai Turks, was visited by the Byzantine envoy, Zimarchus, in 569 A.D. In the still extant report of this emba.s.sy[694]

the Turks ([Greek: Tourkoi]) are mentioned by name, and are described as nomads who dwelt in tents mounted on wagons, burnt the dead, and raised to their memory monuments, statues, and cairns with as many stones as the foes slain by the deceased in battle. It is also stated that they had a peculiar writing system, which must have been that of these Orkhon inscriptions, the Uiguric having apparently been introduced somewhat later.

Originally the Uigurs comprised nineteen clans, which at a remote period already formed two great sections:--the On-Uigur ("Ten Uigurs") in the south, and the Toghuz-Uigur ("Nine Uigurs") in the north. The former had penetrated westwards to the Aral Sea[695] as early as the second century A.D., and many of them undoubtedly took part in Attila's invasion of Europe.

Later, all these Western Uigurs, mentioned amongst the hordes that hara.s.sed the Eastern Empire in the fifth and sixth centuries, in a.s.sociation especially with the Turki Avars, disappear from history, being merged in the Ugrian and other Finnish peoples of the Volga basin.

The Toghuz section also, after throwing off the yoke of the Mongol or Tungus Geugen (Jeu-Jen) in the fifth century, were for a time submerged in the vast empire of the Altai Turks, founded in 552 by Tumen of the House of a.s.sena (A-shi-na), who was the first to a.s.sume the t.i.tle of Kha-Khan, "Great Khan," and whose dynasty ruled over the united Turki and Mongol peoples from the Pacific to the Caspian, and from the Frozen Ocean to the confines of China and Tibet. Both the above-mentioned Sinjibu, who received the Byzantine envoy, and the Bilga Khan of the Orkhon _stele_, belonged to this dynasty, which was replaced in 774 by Pei-lo (Huei-hu), chief of the Toghuz-Uigurs. This is how we are to understand the statement that all the Turki peoples who during the somewhat unstable rule of the a.s.sena dynasty from 552 to 774 had undergone many vicissitudes, and about 580 were even broken into two great sections (Eastern Turks of the Karakoram region and Western Turks of the Tarim basin), were again united in one vast political system under the Toghuz-Uigurs. These are henceforth known in history simply as Uigurs, the On branch having, as stated, long disappeared in the West.

The centre of their power seems to have oscillated between Karakoram and Turfan in Eastern Turkestan, the extensive ruins of which have been explored by D. A. Klements, Sven Hedin and M. A. Stein. Their vast dominions were gradually dismembered, first by the _Hakas_, or _Ki-li-Kisse_, precursors of the present Kirghiz, who overran the eastern (Orkhon) districts about 840, and then by the Muhammadans of Mawar-en-Nahar (Transoxiana), who overthrew the "Lion Kings," as the Uigur Khans of Turfan were called, and set up several petty Mussulman states in Eastern Turkestan. Later they fell under the yoke of the Kara-Khitais, and were amongst the first to join the devastating hordes of Jenghiz-Khan; their name, which henceforth vanishes from history[696], has been popularly recognised under the form of "Ogres,"

in fable and nursery tales, but the derivation lacks historical foundation.

At present the heterogeneous populations of the Tarim basin (Kashgaria, Eastern Turkestan), where the various elements have been intermingled, offer a striking contrast to those of the Ili valley (Sungaria), where one invading horde has succeeded and been superimposed on another. Hence the complexity of the Kashgarian type, in which the original "horse-like face" everywhere crops out, absorbing the later Mongolo-Turki arrivals.

But in Sungaria the Kalmuk, Chinese, Dungan, Taranchi, and Kirghiz groups are all still sharply distinguished and perceptible at a glance.

"Amongst the Kashgarians--a term as vague ethnically as 'Aryan'--Richthofen has determined the successive presence of the Su, Yue-chi, and Usun hordes, as described in the early Chinese chronicles[697]."

The recent explorations of M. A. Stein have thrown some light on the ethnology of this region, and a preliminary survey of results was prepared and published by T. A. Joyce. He concludes that the original inhabitants were of Alpine type, with, in the west, traces of the Indo-Afghan, and that the Mongolian has had very little influence upon the population[698].

In close proximity to the Toghuz-Uigurs dwelt the _Oghuz_ (_Ghuz, Uz_), for whom eponymous heroes have been provided in the legendary records of the Eastern Turks, although all these terms would appear to be merely shortened forms of Toghuz[699]. But whether true Uigurs, or a distinct branch of the Turki people, the Ghuz, as they are commonly called by the Arab writers, began their westward migrations about the year 780.

After occupying Transoxiana, where they are now represented by the Uzbegs[700] of Bokhara and surrounding lands, they gradually spread as conquerors over all the northern parts of Irania, Asia Minor, Syria, the Russian and Caucasian steppes, Ukrainia, Dacia, and the Balkan Peninsula. In most of these lands they formed fresh ethnical combinations both with the Caucasic aborigines, and with many kindred Turki as well as Mongol peoples, some of whom were settled in these regions since neolithic times, while others had either accompanied Attila's expeditions, or followed in his wake (Pechenegs, Komans, Alans, Kipchaks, Kara-Kalpaks), or else arrived later in company with Jenghiz-Khan and his successors (Kazan and Nogai "Tatars"[701]).

In Russia, Rumania (Dacia), and most of the Balkan Peninsula these Mongolo-Turki blends have been again submerged by the dominant Slav and Rumanian peoples (Great and Little Russians, Servo-Croatians, Montenegrini, Moldavians, and Walachians). But in south-western Asia they still const.i.tute perhaps the majority of the population between the Indus and Constantinople, in many places forming numerous compact communities, in which the Mongolo-Turki physical and mental characters are conspicuous. Such, besides the already mentioned Turkomans of Parthian lineage, are all the nomad and many of the settled inhabitants of Khiva, Ferghana, Karategin, Bokhara, generally comprised under the name of Uzbegs and "Sartes." Such also are the Turki peoples of Afghan Turkestan, and of the neighbouring uplands (Hazaras and Aimaks who claim Mongol descent, though now of Persian speech); the Aderbaijani and many other more scattered groups in Persia; the Nogai and k.u.muk tribes of Caucasia, and especially most of the nomad and settled agricultural populations of Asia Minor. The Anatolian peasantry form, in fact, the most numerous and compact division of the Turki family still surviving in any part of their vast domain between the Bosporus and the Lena.

Out of this prolific Oghuz stock arose many renowned chiefs, founders of vast but somewhat unstable empires, such as those of the Gasnevides, who ruled from Persia to the Indus; the Seljuks, who first wrested the Asiatic provinces from Byzantium; the Osmanli, so named from Othman, the Arabised form of Athman, who prepared the way for Orkhan (1326-60), true builder of the Ottoman power, which has alone survived the shipwreck of all the historical Turki states. The vicissitudes of these monarchies, looked on perhaps with too kindly an eye by Gibbon, belong to the domain of history, and it will suffice here to state that from the ethnical standpoint the chief interest centres in that of the Seljukides, covering the period from about the middle of the eleventh to the middle of the thirteenth century. It was under Togrul-beg of this dynasty (1038-63) that "the whole body of the Turkish nation embraced with fervour and sincerity the religion of Mahomet[702]." A little later began the permanent Turki occupation of Asia Minor, where, after the conquest of Armenia (1065-68) and the overthrow of the Byzantine emperor Roma.n.u.s Diogenes (1071), numerous military settlements, followed by nomad Turkoman encampments, were established by the great Seljuk rulers, Alp Arslan and Malek Shah (1063-92), at all the strategical points.

These first arrivals were joined later by others fleeing before the Mongol hosts led by Jenghiz-Khan's successors down to the time of Timur-beg. But the Christians (Greeks and earlier aborigines) were not exterminated, and we read that, while great numbers apostatised, "many thousand children were marked by the knife of circ.u.mcision; and many thousand captives were devoted to the service or the pleasures of their masters" (_ib._). In other words, the already mixed Turki intruders were yet more modified by further interminglings with the earlier inhabitants of Asia Minor. Those who, following the fortunes of the Othman dynasty, crossed the Bosporus and settled in Rumelia and some other parts of the Balkan Peninsula, now prefer to call themselves _Osmanli_, even repudiating the national name "Turk" still retained with pride by the ruder peasant cla.s.ses of Asia Minor. The latter are often spoken of as "Seljuk Turks," as if there were some racial difference between them and the European Osmanli, and for the distinction there is some foundation.

As pointed out by Arminius Vambery[703], the Osmanli have been influenced and modified by their closer a.s.sociation with the Christian populations of the Balkan lands, while in Anatolia the Seljuks have been able better to preserve the national type and temperament. The true Turki spirit ("das Turkentum") survives especially in the provinces of Lykaonia and Kappadokia, where the few surviving natives were not only Islamised but ethnically fused, whereas in Europe most of them (Bosnians, Albanians) were only Islamised, and here the Turki element has always been slight.

At present the original Turki type and temperament are perhaps best preserved amongst the remote _Yakuts_ of the Lena, and the _Kirghiz_ groups (_Kirghiz Kazaks_ and _Kara Kirghiz_) of the West Siberian steppe and the Pamir uplands. The Turki connection of the Yakuts, about which some unnecessary doubts had been raised, has been set at rest by V. A.

Sierochevsky[704], who, however, describes them as now a very mixed people, owing to alliances with the Tunguses and Russians. They are of short stature, averaging scarcely 5 ft. 4 in., and this observer thought their dark but not brilliant black eyes, deeply sunk in narrow orbits, gave them more of a Red Indian than of a Mongol cast. They are almost the only progressive aboriginal people in Siberia, although numbering not more than 200,000 souls, concentrated chiefly along the river banks on the plateau between the Lena and the Aldan.

In the Yakuts we have an extreme instance of the capacity of man to adapt himself to the _milieu_. They not merely exist, but thrive and display a considerable degree of energy and enterprise in the coldest region on the globe. Within the isothermal of -72 Fahr., Verkhoyansk, in the heart of their territory, is alone included, for the period from November to February, and in this temperature, at which the quicksilver freezes, the Yakut children may be seen gambolling naked in the snow. In midwinter R. Kennan met some of these "men of iron," as Wrangel calls them, airily arrayed in nothing but a shirt and a sheepskin, lounging about as if in the enjoyment of the balmy zephyrs of some genial sub-tropical zone.

Although nearly all are Orthodox Christians, or at least baptized as such, they are mere Shamanists at heart, still conjuring the powers of nature, but offering no worship to a supreme deity, of whom they have a vague notion, though he is too far off to hear, or too good to need their supplications. The world of good and evil spirits, however, has been enriched by accessions from the Russian calendar and pandemonium.

Thanks to their commercial spirit, the Yakut language, a very pure Turki idiom, is even more widespread than the race, having become a general medium of intercourse for Tungus, Russian, Mongol and other traders throughout East Siberia, from Irkutsk to the Sea of Okhotsk, and from the Chinese frontier to the Arctic Ocean[705].