Besenyos, Ogur and Oguz
Alans and Ases
|Russian Version needs a translation||
|The first Türks|
W.B. Henning(1908 - 1967)
THE FIRST INDO-EUROPEANS IN HISTORY
Society and History. Essays in Honour Karl August Wittfogel
// Ed. by G.L.Ulmen, Hague-Paris-New York, 1978, pp. 215-230
This article is a part of citations from the works of W.B. Henning, an outstanding philologist of Iranian languages, with bearings on the history of the Türkic peoples. For us, the article is valued because it provides a historical outline of the Guties, one of the oldest horse nomadic tribes that came into the pages of history. W.B. Henning is attempting to create an Indo-European paradigm for the horse nomads, and in doing that delves into material that would otherwise remain unexplored.
In philological aspect, the W.B. Henning's idea is contrafactual, since adding Gutian to the Indo-European family would add another oddball: the Indo-European family is flexive, and Gutian is agglutinative. The listing of agglutinative languages and references can be found at A.Toth Are all agglutinative languages related to one another?:
The table shows perfectly clear that the 5 agglutinative cases highlighted in bold conflict with the body of 441 flexive Indo-European languages, and the only way to fit them into the family is by pretending or postulating that morphological typology is irrelevant. This pretension is debunked by philologists who stipulate that “typology is hierarchically superior to genetic linguistics” (R.Anttila, Historical and Comparative Linguistics, 2nd ed. Amsterdam and Philadelphia 1989, p. 318), that languages can change, but only in the frame of their typological constraints, and that typological structures are inheritable, a position held by most non–IE–centered linguists. We see on the map that Lulu/Luluba/Lullubi/Uluba was noted next to the Gutian, and they both were ethnologically identical intruding tribes.
In his Indo-European exploits, W.B. Henning glosses over the tribe called Tukriş (Turuk, Tuγri), who were the same nomads as Guties, and were their immediate neighbors. W.B. Henning then connects the misnomer Tokharians (Yuezhi/Yüe-chih) with the Guties, and double-loops it to Tuγri. In the contents of the Türkic history, the association with the name Türk is far more profound than the connection with the oases dwellers in the Taklamakan's now desertified steppe.The Guties and Tukriş in the Mesopotamia appear in the company of the Türkic tribes Kangar, Subartu (Suvar/Chuvash/Savir etc.) and Kumans. To an inbaised eye this connection appear salient, the same company appears on the pages of history again in again, under slightly varying appelatives, for the next two millennia. Even more facinating, the name Tukriş extends not only eastward from Mesopotamia, it also figures west of it. In the Egyptian references to the tribe that in ancient Greek was called Lydian, the Egyptians used the tribal endonym and called them Tursha in the records dated to 1234-1225 BC (V.V Shevoroshkin, Lydian Language, Moscow, 1967). We can't be more greatful to W.B. Henning and Indo-European studies in general for opening a fascinating porthole into continuity and distribution of the peculiar name Türk that reached our days.
The posting skips “Tokharian” details by using a smaller font. Empty bracket  indicate placeholders for omitted characters. Page numbers are shown at the beginning of the page. Page breaks in continuous text are indicated by //. Some diacritics are dropped, to avoid font conflicts. The subheadings in bold blue, bold highlighting, and (Pinyin transcriptions) are added for the posting. Diacritical c is shown as ch, diacritical s is shown as sh, and χ retained as x to denote voiced h.
One of the Indo-European
tribes that sought a home in Asia is distinguished from all others by its apparent lack
of energy and vitality: that is the tribe whose descendants spoke the languages known as
“Tokharian”, By nature isolationist, they chose to make
permanent settlements in a remote region in the heart of Central Asia — a chain of oases
along the northern edge of the Great Desert of Chinese Turkestan, at the foot of the
T'ien-shan mountains - for which there was little or no competition. When the Chinese
first encountered this tribe in the 1st century BC, they did not even recognize it as
constituting a distinct ethnic unit and failed to give it a name, such as they had for
their neighbors to the North and East. Even today, after the enormous growth of our
knowledge of Central Asian history in this century, we still have no name for this tribe
and even its linguistic designation “Tokharian” is in dispute. In order to avoid
confusion, I will not refer to “the Tokharians” but to “the speakers of Tokharian”
(i.e. fake Tokharian, the Kuchean and Karashar).
Such considerations, it seems to me, make it advisable to regard the speakers of Tokharian
(i.e. fake Tokharian, the Kuchean and Karashar)
as a mere defeated remnant, an enfeebled splinter of a once large and powerful
nation that like its brethren went through the usual phases of war, adventure, and
conquest, but ultimately proved unsuccessful. If this supposition holds, there is a good
chance that the activities of this lost nation will have impinged on one or another of
the great civilizations of antiquity, and will therefore be recorded in our historical
material; it is the object of this paper to trace the early history of that long extinct
nation (which provisionally we shall call the “Proto-Tokharians”).
First: their number. Tokharian is linguistically isolated, it has no close relatives and is therefore the sole survivor of a language-group on a par with e.g. Germanic or Indo-Iranian. This circumstance renders it probable that the speakers of this language-group were not less numerous than the speakers of some of those other groups. Or positively expressed, the Proto-Tokharians constituted a large and populous nation, divided into numerous tribes speaking divergent dialects.
Second: the place of their appearance. This is easily determined. All other Indo-Europeans of Asia, sooner or later, come within the view of the civilizations emanating from Babylonia. There is no good reason for excepting the Proto-Tokharians (i.e. proto-fake Tokharians, the Kucheans and Karashars) from the common fate; they will have made their debut somewhere in the Near and Middle East.
Third: the time of their appearance. The assumption to be made is inevitably contentious. Comparativists have asserted that, in spite of its late appearance, Tokharian (i.e. fake Tokharian, the Kuchean and Karashar) is a relatively archaic form of Indo-European. 3 This claim implies that the speakers of this group separated from their Indo-European brethren at a comparatively early date. They should accordingly have set out on their migrations rather early, and should have appeared within the Babylonian sphere of influence also rather early. Earlier, at any rate, than the Indo-Iranians, who spoke a highly developed (therefore probably later) form of Indo-European. Moreover, as some of the Indo-Iranians after their division into Iranians and Indo-Aryans 4 appeared in Mesopotamia about 1500 BC, we should expect the Proto-Tokharians (i.e. Gutians) about 2000 BC or even earlier.
If, armed with these assumptions as our working hypothesis, we look through the pages of history, we find one nation - one nation only - that perfectly fulfills all three conditions, which, therefore, entitles us to recognize it as the “Proto-Tokharians”. Its name was Guti; the initial is also spelled with q (a voiceless back velar or pharyngal), but the spelling with g is the original one. The closing -i is part of the name, for the Akkadian case-endings are added to it, nom. Gutium etc. Guti (or Gutium, as some scholars prefer) was valid for the nation, considered as an entity, but also for the territory it occupied (This philological logic is as well unsound, since accumulation of case-endings in transmission from one language to another is a common-place occurrence: Guz /Guzlar > Gut-i > Gut-i-um, a la foc > foc-us > foc-us-es, reht > rect-um > rect-um-s). It persisted as a geographical term long after the nation itself had disappeared. Such multiple acceptance of a national name was common in antiquity; for example, Parsa meant the Persian nation, a man belonging to it, the country inhabited by it, and the capital of that country — four different meanings expressed by one and the same word, without a change in its grammatical status.
The Guti appeared towards the end of the first Semitic dynasty ruling over Babylonia,
the dynasty of Agade (Akkad), in the reign of Naram-Sin, one of the great conquerors in
Babylonian history. He struggled to secure the northern limits of the Babylonian plain,
the mountainous country on the border of present-day Persia. One suspects that the
reason, or at any rate a reason, for his efforts in that direction was pressure caused by
the influx of new tribes. After a long rule (of 37 years), in the course of which he
carried the power of Babylon far beyond the limits attained by his predecessors, he
suffered a great defeat at the hands of the Guti, who thereupon conquered the whole of
Babylonia and kept it in subjection for a hundred years.
The mere fact that the great Naram-Sin, at the height of his powers, could be overthrown by this previously unknown nation, is a measure of its vigor and freshness. At least, it has been widely believed that the end of Naram-Sin was due to the Guti. 5 There is a direct statement to that effect in a latish text, the “Weidrier Chronicle”, which tells of the wrath of God Marduk against Naram-Sin:
For the second time he [Marduk] set against him the hordes or Gutium, and gave his kingship away to the hordes of Gutium. The Gutium, sullen fellows who do not know the fear of God, who do not know how to apply the rules and precepts of the cult properly, etc. 6
There are, however, some difficulties which I need not go into; many of these are caused by the habit of Babylonian chronologers of treating contemporaneous kings or dynasties as successive to each other.
A difficulty that deserves mentioning is the isolated appearance of Gutium in a passage relating to far earlier times - a story about a legendary prehistoric king, Lugal-arini-munclii by name. In it he is credited with having conquered many countries, among them Gutium. The list of these countries clearly bespeaks its late origin. Thus the text has been generally and rightly regarded as an item of mythology rather than history. 7
The invasion of the Guti, to whom the Babylonians referred in terms associated with the way in which civilized nations talked of obnoxious nomadic invaders all the world over, caused a setback to the development of Babylonian civilization. Not surprisingly, there are not many records from the time of Guti rule, so that it becomes difficult to form a clear idea of the actual conditions.
The length of the period of Guti domination (which had “caused . . . long grass to grow on the highways of the land”) 8 has been calculated as 91 years by Jacobsen, in his edition of the Sumerian kinglist, but as 124 or 125 years in some of his sources; at any rate, it lasted for a considerable time. At the end of the period, the last king of the Guti, Tirigan, was overthrown and killed by Utu-hegal, the ruler of Uruk, who characterized Gutium as:
a viper of the hills, an enemy of the gods, who had carried the kingship of Sumer off to the mountains, who had filled Sumer with evil. 9
These words, from an inscription by Utu-hegal, show already clearly that the Guti never
truly settled in Babylonia, but merely exploited and gradually ruined that country.
As yet I have carefully avoided stating an absolute date for the events sketched here. The reason for this reticence is the general “uncertainty of all dates in the early history of Babylonia. While there is a good system of relative chronology, in absolute terms its points are shifted every few years, as a rule in a downward direction, it would be consonant with present-day opinion if we assumed that the time of Guti rule centered about the year 2100 BC. 10 (I.e. 2150-2050 BC)
The Guti descended upon Babylonia from the hills of Western Persia. Their name remained attached to a specific district, the neighborhood of the Lower Zab (Little Zab, 35.25°N 43.5°E), which may thus be presumed to have formed their headquarters during their rule of the lowlands. That they had freshly arrived there within the life-time of Naram-Sin seems very probable. Actually, the opinion that they were then recent intruders into the scene of the Middle East has enjoyed some support from the late German Assyriologist A. Ungnad, a fact that is particularly valuable here because that eminent scholar was reluctant to admit any such intrusions. His work on Subartu (presumptive ancestors of Chuvashes, from Subar/Suvar, Slavic Severyans), published in Berlin in 1936, is a most useful and highly competent guide in the type of problem that concerns us here. Regrettably it is marred by racist prejudices, such as one associates with the time and place of its publication.
Ungnad 11 thought it possible that the place of ultimate origin of the Guti was somewhere in “the East” (by which he meant Russian Turkestan) (By 1978 the “Russian Turkestan” was long gone, replaced by the Stalinist concoction of 5 Central Asian “Socialist Republics”); there is no actual evidence in favor of such a hypothesis. When one considers the location of the headquarters of that nation, one may be more readily inclined to believe that it had arrived by way of the Caucasus. Such an assumption finds some support in a parallel situation of later history.
If we regard the Guti as “Proto-Tokharians”, their nearest relatives among the Indo-Europeans would be the Hittite nations of Asia Minor l2 who (so the archaeologists insist) had reached their historic settlements through the Caucasus. The distribution of the Guti at the Lower Zab and the Hittites (et al.) in Asia Minor recalls the outcome of a Volkerwanderung at the end of the 8th century. At that time two savage nations of northern barbarians, the Kimmerians and the Scythians, invaded the Near East. Both emerged from Southern Russia, and both passed through the Caucasus. The Kimmerians (to believe Herodotus) crossed that mountain range by the central pass, the “Gate of the Alans”, and ultimately flooded all Asia Minor. The Scythians followed the eastern passage, the Pass of Derbend, on the shores of the Caspian Sea, brought ruin on Media, and ended up not far from the place which had once been the headquarters of the Guti. As such tribal movements are apt to take the same direction and follow the same routes, it is probable that the Guti, too, had come from Southern Russia through the Pass of Derbend.
After their ultimate defeat in Babylonia, the Guti withdrew from the scene. A small remnant, perhaps a part of the nation that had adopted more settled habits, was left behind in the Lower Zab region. It is mentioned from time to time in later centuries as one of those noisome hill tribes on the northern frontier, that need to be punished by a military expedition. However, that remnant no longer constituted a serious danger: evidently, the bulk of the powerful Guti had departed, gone beyond the horizon of the Babylonian observers (It is nearly impossible for the sedentary people to expel the nomads; some nomadic leftover always remains in place after a general withdrawal).
Barbarous nomadic invaders have frequently adopted a different course. They often settled among their new subjects, conceived a liking for the gifts of civilization, dispersed, intermarried, gradually merged in the population, and with their distinctiveness lost their energy and warlike habits. The Guti, however, never settled down in Babylonia, they remained barbarians, and preserved their identity as well as their mobility. Hence, when conditions turned adverse, they were still capable of breaking up camp and setting out on the road to fresh worlds to conquer and despoil. 13 (That is a patented biased nonsense, including the flow direction of the civilization. All nomadic conquerors, at all times, set up their settlements separate from the indigenous sedentary population, without exception. A nomadic tribe, rich with cattle, would never settle down for hard labor of farming and land bondage. Individually it is a different matter, but the young W.B. Henning here speculated in generalities)
There is no need for us to follow the fate of the remnant Guti in the Persian foothills. A single fact suffices to illuminate the change in their position: Gutians, in succeeding centuries, were apt to figure as merchandise in the slave market. Incidentally, the trade in Guti slaves may supply a pointer to their having been physically distinct from the general run of people in Babylonia. However, it depends on the interpretation of a single word, 14 and as its meaning has been disputed, it may be wiser not to rely on it. It is in any case not seriously in doubt that the Guti differed ethnically from the people of the lowlands. A direct statement by the great Babylonian king Hammurapi (ca. 1890-1750 BC), made in describing his conquest of the border nations, indicates that their languages differed from those familiar to him: “whose hills are distant and whose language is queer”. 15 (Few would doubt the difference in the appearance of the Semitic and Kurgan people. That had nothing to do with the slave trade: as somewhat later history points out, the nomadic tribes preferred to hunt for the slaves among the vulnerable stationary farming populations, the example of nomads selling nomads as slaves are far in between; even in the worst case of Kimak Kaganate, the victims were settled pastoral communities, not the fluid nomadic tribes)
Hammurapi, in the inscription just quoted, names four such border-nations. The first
has been restored as “Elam”, but this is uncertain. The others are Gutium, Subartu, and
Tukriş. Leaving aside Subartu (the subject of Ungnad's book cited above), we find the
Guti here associated with the Tukriş
(Tukrish), a nation comparatively rarely mentioned, whose
settlements (it seems) adjoined the Guti territory in the east or south-east. As with the
Guti, there is an old reference to their having been in Western Persia in remote
antiquity, in a story from Boghazkoy about a prehistoric Elamite king (Autaluma), in
which a king of Tukriş (Tukrish)”, Kiklipatalli, is mentioned; it has been generally accepted as a
further instance of mythical history. If we leave aside that isolated reference, the
Tukriş (Tukrish) would appear to have been contemporaneous with their Guti neighbors. The sibilant
at the end of the name is presumably an indigenous case-ending, so that it is actually
Tukri. For reasons that will become clear later, I assume that Tukri and Guti were two
closely allied brother nations that came together to Western Persia and who left it
together shortly before the end of the third
millennium. (Why would young W.B. Henning “assume” a visitor
nation when there is an indication of prehistoric ingenuity is as questionable as missing
the screaming parallels Guti - Tukri < Guzes - Türks,
especially when the horse husbandry nomadic tradition of Guti, Tukri, Guzes, and Türks
make that parallel obvious)
Where did they go from Persia? The first part of this paper has already indicated how I propose to answer that question: they migrated to the heart of Asia, and there took over huge territories, including most of Chinese Turkestan, a portion of China proper, and a part of Kan-su (Gansu) to the west of the Huang-ho (Huanghe). Some of their tribes settled in permanent habitations, others clung to the nomadic life. Their dispersal over a very large area loosened the bonds between them, so that, two thousand years or so later, the various sections were no longer aware that they shared a common ancestry. Their languages, after so great a lapse of time, very likely had become mutually unintelligible (This proposal is aimed at producing a historyline consistent with the conclusion of the author's hypothesis, which was proven to be wrong. The name Kucha is a local dialectal form of Kushan. The Guties-Guzes in the form Uzes/Udes are later known in the Caucasus, in the Western Central Asia in the Late Antique time, in the Eastern Central Asia in the Middle Ages, in the Middle East during Middle Ages and Late Middle Ages, their name simply means generic “tribes”; the Türks are known as a Saka tribe in the Early Antique time, as a Hunnic tribe in the Antique time, then a component of their own Türkic Kaganates in the Late Antique and Early Middle Ages; the Suvars are later known in the Caucasus in the Antique time, as their own state in the N. Caucasus steppes in the Late Antique time, as members of the Western Türkic Kaganate, as members of the Bulgar confederation, as Middle Age state that rivaled the incipient Rus, and as members of the Bulgar state in the pre-Mongol time; all these ethnical names are associated with what the modern terminology calls “Türkic” ethnos).
This is the heart of the theory that I wish to propound. Possibly the archaeologists may welcome a theory that involves a considerable movement of people from Persia to the limits of China as early as the close of the third millennium BC. Similarities in various artifacts and in the ornamentation of pottery of prehistoric Persia and Ancient China have been pointed out long ago. They formed the principal subject of Ernst Herzfeld's last important work, Iran in the Ancient East. After studying the comparative material illustrated in that volume, even one who cannot claim to be an archaeologist is driven to assert that some of the similarities are so remarkable, in their details and complexities, that independence is impossible. Actually, the Tokharians themselves have been mentioned as the agents of communication here, but on the basis of a different, more vague, and less satisfactory historical construction. 17 (Archeology does prove the similarities between bronze technology brought to China by Zhou-circle horse nomadic tribes during 1600 BC Yin Shang period, to Central Europe during 3000 BC Kurgan expansion period, and found in the Guti area in the Middle East ca. 2000 BC. Again, the cited speculations of Ernst Herzfeld on Tarim basin “Tokharian” farmers are drawn from the thin air, with a farcical notion that the tribes of free cowboys would somehow get inspired to turn into a tribe of meekly oasis subsistent farmers living under a heel of their nomadic brethrens)
Before turning to the settled sector of the invaders of Central Asia, we shall deal with their nomadic brethren, who continued to pursue the way of life of the Persian Guti and therefore were the appropriate vessels of the national spirit. The nomads went furthest to the East and set up a powerful empire, which centered upon Western Kan-su (Gansu). There they came under the observation of the Chinese, who referred to them by their old national name, as Guti. The two characters (月支/月氏) which were fixed as standard orthographs for that name shortly before 200 BC are pronounced as Yüe-chih in modern Mandarin (Among various phonetic reconstructions, “Guti” does not show up, but there is an Old Chinese “Tokwar” for “Tokhar”, confirmed by corroborating historical evidence).
All students of Ancient Chinese history are familiar with the Yüe-chih, and know what a
desperate search there has been for the Western equivalent of that name; that such an
equivalent existed has been taken for granted, and rightly so. The search has produced so
many equations that it is scarcely reasonable to propose a new one, yet all of them have
been found wanting in one particular or another. Some differences between the Chinese
form and the proposed Western equivalent were glossed over. In this respect, at least,
our new equation is preferable to all others: Guti and Yüe-chih are perfect equivalents.
In the latest edition of his dictionary, 18 Karlgren gives ngiwot-tsie as the Middle Chinese pronunciation of Yüe-chih 月氏 (月支) and *ngiwat-tie(g) as the corresponding Old Chinese form. Middle Chinese means AD 600, while Old Chinese is defined by Karlgren as “early Chou (Zhou)”, which may mean 1000 BC or a little later. What we need to know is the pronunciation current when the orthography of the foreign name was established, i.e. about 200 BC, and as that point in time bisects the years between the dates set for Old and Middle Chinese, the appropriate pronunciation may be assumed to have been intermediary between the two forms supplied.
In the first syllable (which is the crucial one), there is virtually no difference, so that we may adopt either form. In the second syllable, the closing g is believed to have disappeared long before 200 BC, and the difference in its initial, from palatal t' to palatal affricate, is very slight. If the name represents foreign Guti, one would assume that its t had been assibilated or palatalized in front of i, a requirement that incidentally is in consonance with the surviving Tokharian languages in both of which t has turned into ĉ (ch) before i.
Further remarks are necessary on the first syllable. In Middle Chinese representations of foreign words, an initial nasal may render a foreign voiced stop as well as a nasal and a closing t may represent a foreign -r as well as a dental; groups of internal sonants have to be simplified. Hence the regular Central Asian equivalent of the first syllable is gwar or gur in medieval times. 19 In antiquity, however, final -t was never used for -r, so that the form appropriate for about 200 BC would be gwat or gut for the first syllable, gwat-t'i or gut-t'i for the whole name. This agrees well enough with Guti, especially if we allow for the tendency to palatalization mentioned above. Such a reconstruction has indeed been made by none other than Bernhard Karlgren, the eminent Swedish Sinologist, who has contributed more than any other scholar to the elucidation of ancient Chinese phonology. In a statement supplied to Sten Konow at his request, Karlgren postulated foreign gwat/gwot or gat/got or gut etc. + ti (or possibly ĉi [chi]?) etc. 20 (Thank you, Mssrs. B. Karlgren and W.B. Henning, [and also Mssr. C. Beckwith, 2009, p. 5, footnote #16, and Appendix B pp. 380-383], it is very helpful to equate Yuezhi/Yüe-chih and “Tokwar/Tokhar” and at the same time “Guti/Guti”, one can only wonder what kind of philological tricks are beyond possible, like in the popular paradox “Can the Almighty create an obstacle that he/she can't overcome?” For philologists the answer is definite “No” )
It may be possible to decide which of these variants is the correct one. Considerable efforts have been made to trace references to the Yüe-chih in such Chinese sources as are earlier in origin than the end of the 3rd century. The various names claimed as attributable have been fully discussed by the late G. Haloun in his admirable article, Zur Üe-tsi-Frage. 21 An assured form of the name, found from the 4th century onwards, is Yüe-chih 22 禺知 which yields Old Chinese *ngiu-tie(g); it surely indicates that the common spelling should be interpreted as Gu(t)-t'i rather than Gwat-.
There is no need to discuss all the identifications that have hitherto been advanced for Yüe-chih, but a few remarks on recent proposals may not be superfluous. First, the identification made by Haloun. I find myself at variance with the conclusions presented by him in his article on the Yüe-chih problem. Influenced by the work of Tibeto-Burman-Chinese comparative philology, he posited a form: ? ŋiwat'ia or ? zŋiwat'ia, with reservations about the initial to be assumed. That form, he suggested, might represent foreign *zguja(ka) which should be the name of the Scythians.
This opinion is faced by difficulties in both directions. On the Chinese side, the initial sibilant (which would be essential to the proposal) is uncertainly grounded; it rests on a comparison with Tibetan zla(-ba) 'moon'. If it once existed, one may well doubt that it still existed as late as 200 BC; indeed, Haloun himself placed two asterisks before the form with z-. On the Western side, the name of the Scythians is found only in Greek (Σκυθαι); Akkadian (ašguzai [ashguzai]) and Biblical Hebrew (אשכנז). Its initial was šk, not zg; that at least is certain. The nature of the third consonant cannot be determined satisfactorily; there is no way to bridge the gap between the Greek aspirate t and the Akkadian-Hebrew voiced sibilant; *zguja is thus unacceptable.
The other explanation to which I should like to draw attention is that advanced by E. G. Pulleyblank in the course of his recent study on the reconstruction of Old Chinese. 23 The western equivalent favored by Pulleyblank is 'Ιατιοι, a tribal name mentioned but once in Greek literature, in Ptolemy's Geography (vi 12 # 4), This comparison was made first by Marquait (in 1901), who boldly identified 'Ιατιοι with ''Ασιοι (of Strabo) and Asiani (of Trogus Pompejus) and Πασιανοι (of Strabo) which he emended into Γασιανοι. 24 A striking feature in the equation as formulated by Pulleyblank is the suggestion that of the first syllable of Yüe-chih both the initial velar and the central -u/w- can be disregarded, so that yüe may represent foreign iat. This opinion contrasts with that expressed by Karlgren, who regarded precisely these two elements as the essence of that syllable. In support of his case Pulleyblank adduces two or three other foreign names in Chinese garb, but none of the equations seems entirely convincing. 25 A possible pointer to the value to be attached to an initial velar nasal may be seen in the careful manner in which foreigners represented that Chinese sound; there is an ancient example of it in the Sogdian rendering of the name of the city of Yeh 鄴, once pronounced as ŋiop-: here the Sogdians wrote 'nkp-, using three of their letters to reproduce the nasal. 26
To return now to the history of the
Yüe-chih. According to Haloun, references to them
may be traced as far back as the 7th century; the more remote in time, the more vague and
the less sharply defined they become, until they cease altogether. Uncertainty does not
allow us to infer that the Yüe-chih had not arrived yet, but it may indicate that the
western approaches were beyond the horizon of the Chinese in great antiquity, While we cannot
tell when the Yüe-chih reached Kan-su (Gansu), we can tell when they left it. Here for once history truly repeats itself.
After having enjoyed the fruits of their empire for a long period, the Yüe-chih were overthrown by the rising power of the Hsiung-nu (Huns) in the first half of the second .century BC. Their last king was killed and the main body of the nation, still nomadic and mobile, turned its back upon civilization, broke up camp and departed seeking fresh pastures. However, they left behind several small groups that had settled in the mountains fringing the southern edge of Kan-su (Gansu). The Chinese named those that departed the “Great Yüe-chih'”, those that stayed the “Little Yüe-chih”.
When the Great Yüe-chih withdrew, they set out in the general direction of Persia. One might be inclined to say that they retraced their steps. Possibly a dim memory had been preserved among them of a pleasant land far to the West whence their ancestors had come. However, conditions had greatly changed and the renewed migration never reached as far as Persia. Some 30 or 40 years later it came to a halt in Bactria, the northern part of present-day Afghanistan, Here the Great Yüe-chih established a new empire (which was to be their last), dominated the surrounding countries and generally made a nuisance of themselves to their neighbors.
We come now to a remarkable circumstance, while the Chinese continue to call that nation Yüe-chih or more precisely Great Yüe-chih, their new neighbors use an altogether different name: Toxar, i.e. 'Tokharian'. Indians, Persians, Sogdians, Greeks — everybody called the Yüe-chih by this new name, and Bactria itself came to be known as Toxaristan 'land of the Tokharians'. It is as if that nation had changed its name on the way, had left the fringes of China as Yüe-chih and arrived in Bactria as Tokharians. However, further investigation has shown that the seemingly new name was used for the Yüe-chih already in their old home in Kan-su (Gansu). The remnant “Little Yüe-chih” there was also known as Toxar (Tuxar). The Tibetans spoke of the Thod-kar living in the north-eastern corner of their country; and through the silk-trade various related forms (the town Θογαρα, the tribe Θαγουροι, the mountain Θαγουροιορος) reached the pages of Ptolemy's Geography. Because of such evidence it has long been recognized that the Yüe-chih and the Tokharians, in both the East and the West, must have been virtually identical.
In order to account for the origin of the double name, I would propose having recourse to the ancient Tukri. I have postulated above that the Guti and the Tukri, two brother nations, set out from Persia together in the distant past. Let us assume that these two groups fused in the course of the centuries into a new whole, which could claim either this or that name. As some two thousand years have elapsed, we should not be unduly surprised at the slight alteration in form of the ancient name Tukri.
Actually, all the new forms associated with the Yüe-chih (Θογαρ-, Θαγουρ-, Toxap-, Tuxar-, Təxwar-, Tukhar-) have preserved the consonantal skeleton (dental + velar + r) and the old u-sonant appears in every specimen of the name.
Powerful support for the opinion expressed in the preceding paragraph is supplied by the nomenclature attached to the “speakers of the Tokharian languages”, the groups settled at the foot of the T'ien-shan mountains, which we considered first and to which we now return in our concluding argument.
It will be recalled that (in the Tarim basin) there were two dialects, “Tokharian A” in the East, “Tokharian B” in the West (in the kingdom of Kuči [Kuchi]). 27 (The W.B. Henning own objections against the use of the fake appellative “Tokharian” to the Arsi and Karashar languages see W.B. Henning The name of “Tokharian” language) It was because of the eastern dialect that the name “Tokharian” has been introduced for these languages by modern scholars. In a bilingual text, preserved in Tokharian A and in Uygur Turkish, the name of the other language was stated to be Tuγri in the Turkish version. That form Tuγri (twγry) was wrongly vocalized and regarded as representing the Tokharian name familiar from the story of the Yüe-chih. Strictly speaking, the appellation “Tokharian” for these languages is erroneous.
There can hardly be any doubt now that Tuγri is the same name as ancient Tukri-š (Tukrish) The form is wonderfully well preserved, even including the stem-vowel -i, the only difference being a slight alteration in the quality of the second consonant.
More than a quarter of a century has passed since I published an article 28 in which I made an effort to establish Tuγri as the correct pronunciation and argued that at the time when it was in use (about AD 700) it differed substantially from the Tokharian name (Tuxar), indeed sufficiently so to exclude confusion. While this argument has been widely accepted, I have sometimes been credited also with proving that the two names, Tuγri and Tuxar, had no connection with one another. This was never my intention. On the contrary, I regard it as certain that both derive from one and the same prototype, which we can now identify as ancient Tukri.
Once the name of the Tukri is found to be preserved for the speakers of the Tokharian (i.e. Tuγri = fake Tokharian, the Kuchean and Karashar) languages, we are bound to ask whether the name of their twin, the Guti, has also survived in the same area. In order to arrive at an answer, we should consider the changes that Guti would have undergone if it had been a Tokharian (i.e. Tuγri = fake Tokharian, the Kuchean and Karashar) word. It is well established that g before u becomes k in either Tokharian dialect (i.e. Tuγri = fake Tokharian, the Kuchean and Karashar), and t before i 29 becomes č (ch): therefore Guti should have become Kuči (Kuchi). This we know is the name of the western part of the Tokharian language (i.e. Tuγri = fake Tokharian, the Kuchean and Karashar) area, the home of Tokharian B (Kuči/Kuchi).
It would be tempting to infer that Tokharian B, the language of Kuči, continued the
language of the Guti, while Tokharian A, named Tuγri, continued the language of the
Tukri. This would be a neat solution, but one I fear would be too simple.
In fact, Tuγri occurs also as a political and
geographical term, and so used serves to cover the whole area of Tokharian speech
(Tarim basin). There are references to the “Four Tuγri Land”, which term apparently designates the four kingdoms of
For long it was not known what form the name Kuči assumed in either of the Tokharian languages (i.e. Tuγri = fake Tokharian, the Kuchean and Karashar). Recently, however, Professor Werner Winter discovered the name in an unpublished Kuchean document, in the guise of kusinne; it accompanies forms of orotstse walo 'great king'. 32 In the adjective, “of Kuči”, the first syllable has been reduced through the progression of the stress accent and the second consonant has turned into a palatal sibilant through contact with the initial. The resulting form closely resembles the above-mentioned name of the North-Eastern kingdoms. It also resembles the long-known medieval Turkish name of Kuči, 33 i.e. Küsän. 34
There are thus strong arguments in favor of the assumption that the names Kuči and Tuγri once embraced the whole of the speakers of the Tokharian languages (i.e. Tuγri = fake Tokharian, the Kuchean and Karashar). 35 This is in perfect consonance with the observations made previously about the Yüe-chih or Toxar and equally reflects the fusion of the twin nations, the Guti and the Tukri. If they were indeed the Proto-Tokharians, as I have endeavored to argue in this paper, they could claim to be the first Indo-Europeans (i.e Türkic people) in history; for their appearance in Persia (i.e. Mesopotamia, there was no Persia in the 21st c. BC whatsoever) precedes the arrival of the Hittite nations in Asia Minor.
Linguistic evidence in support of identifying the Guti (and Tukriş) with the
Proto-Tokharians (Proto-Karashars/Kucheans) can hardly be expected; for virtually nothing is known of their language. Yet the personal names, especially those of the Guti kings, may supply some hints.
Jacobsen, Sum. King-List, 207 n. 40, wrote “that (e)š is a grammatical ending is suggested by the frequency with which it occurs at the end of Gutian proper names; cf. Inimabak-eš, Inkišu-(e)s, Igešau-(e)š, and Iarlaga-(e)š”. Or, one might suggest, -š by itself was such an ending, presumably of the Nom. Sg. masc. (= I.-E. -s), there being no need to change the actual In-ki-šuš (var. In-gi4-sus; Jacobsen 118-19, line 30), I-ge4-eš-a-uš (line 35), Ia-ar-la-ga-aš (line 32 var,) from -uš and -aš into -u-(e)š, -a-(e)š; cf. also Tukri-š explained above. The names thus reveal the existence of different types of vocalic stems, ending in a/e, i, and u.
Ungnad, Subartu, 83 n. 2, wrote about one A-ra-ha-ab, an enemy of Ammiditana, the great-grandson of Hammurapi, in these terms: “Sein Name ist jedenfalls nicht indogermanisch.....Vielmehr 1st ein Zusammenhang mit gutaischen Namen unabweisbar; vgl. Lasirab ..., Nikillagab, Jarlagab (3. bzw. 8. Konig des Reiches von Gutium), Jakulaba ..., Jadihab ...”. Instead of Lasirab (La-si? -ra? -ah) Gelb proposed and Jacobsen (118-19, line 41) accepted Laerab; and instead of Nikillagab Jacobsen (line 31) read Sarlagab (Sar-lagab-la-gab, var. za-ar-la-ga-ba). This Sarlagab Jacobsen, p. 207, compared wiih Sarlag (Sar-la-ag), the name of a Guti king not otherwise included in the King List; assuming that one and the same ruler was meant to be recognized “in the name Sarlag an abbreviated form of this Sarlagab”. We might prefer to invert that statement and regard Sarlag as the stem, Sarlagab as a form lengthened by a declensional ending. That -ab was such an ending, is strongly suggested by the existence of Iarlagab (Ia-ar-la-gab, line 36, cf. 38) by the side of Iarlagaš. If the latter constituted the nominative in the Guti language, it is probable that the frequent forms in -ab will have been genitives; for in the kind of sentence from which foreigners could have abstracted the forms of names (e. g. “king X. the son of king Y. orders . . .”) nominatives and genitives will have predominated. Incidentally, A-ra-ha-ab, far from being “jedenfalls nicht indogermanisch”, is now seen to belong to A-ra-hu, Old Pers. Araxa, in the inscr. of Bisitun, an Armenian acc. to Darius.
Beside Iarlagas and Iarlagab there is a cognate name which occurs as
Ia-ar-la-ga-an = Iarlagan outside the King List, see Jacobsen p. 150 n. 34, who proposed to identify it
with [ia?]-ar-la-ga-an-da, King List line 47 (p. 120—21); according to Jacobsen we have
here two different ways of spelling *Iarlagand. One ventures to suggest that -an (or
possibly -and) constituted a further declensional ending; cf. also Tirigan, the name of
the last Guti king. In view of the poverty of the Sumerian phonemic system and the wealth
in consonants of the more ancient I.-E. languages, it is conceivable that S- and
I- in Sarlag(ab) and Iarlag(ab) are merely different approximations to a consonant alien to
Sumerian (a palatal sibilant? ).
If indeed the royal names contain various case-forms, a comparison with the actual Tokharian (i.e. Tuγri = fake Tokharian, the Kuchean and Karashar) case-system is suggestive:
It should be explained that in Tokharian (i.e. Tuγri = fake Tokharian, the Kuchean and Karashar)
The Tokharian (i.e. Tuγri = fake Tokharian, the Kuchean and Karashar) forms quoted here in comparison are not selected at random from a vast repertory, but constitute the basic forms (the whole of the “primary” declension) of the appropriate class of nouns, those designating “vernunftbegabte Wesen”. The endings -(a)m of the oblique case (on which the “secondary” declension is built) and -ap/-yap of the genitive singular are confined to nouns of that type. 36 The uniqueness of the ending of the genitive within Indo-European enhances the probability that the agreement owes nothing to coincidence.
1 The essentials of this paper were presented to the Collegium Onentologicum of Berkeley
as the First Lessing-Olschki Memorial Lecture in November, 1962.
11 Ungnad, loc. cit. , 151 n. .5.
Besenyos, Ogur and Oguz
Alans and Ases