張騫 Zhang Qian. The Secret Envoy of Han Emperor Wu in Search of the Arsi (Yuezhi) and the Fall of the Græco-Bactrian Kingdom. (Annotated Compilation of Eastern and Western Sources)

Chris M. Dorn’eich 

張騫 

Zhang Qian 

THE SECRET ENVOY 

OF HAN EMPEROR WU IN SEARCH OF THE ARSI (YUEZHI) 

AND THE FALL OF THE GRÆCO-BACTRIAN KINGDOM 

(ANNOTATED COMPILATION OF EASTERN AND WESTERN SOURCES) 

Berlin 2008

Dorn'eich · Zhang Qian

CHRIS M. DORN’EICH 2004 

“ The Bowang marquis, Zhang Qian” 

NEW STATUE IN FRONT OF THE ANCIENT GRAVE MOUND OUTSIDE CHENGGU (HANZHONG) 

— 1 —


張騫 


THE SECRET ENVOY 

OF HAN EMPEROR WU IN SEARCH OF THE ARSI (YUEZHI) 



Berlin 2008

I M P R E S S U M 

Zhang Qian. 

The Secret Envoy of Han Emperor Wu in Search of the Arsi (Yuezhi) 

And the Fall of the Græco-Bactrian Kingdom. 

Copyright © 2008 Christof Michael Dorn'eich, Berlin 

II

C O N T E N T S 

Summary IV 

1 — In what year did Zhang Qian reach the Oxus River ? 1 

2 — Are we entitled to equate ›Daxia‹ with Tachar/Tocharistan ? 29 

3 — How are we to understand the four names in Strabo’s list ? 73 

Bibliography 97 

Illustrations 107 

III

S U M M A R Y 

The following study grew out of comments I started to jot down after Professor Falk had 

given me a new article by FRANTZ GRENET: ›Nouvelles données sur la localisation des cinq 

“ yabghus” des Yuezhi‹, Journal asiatique (Paris) 294/2–2006, published 2007. When the author 

was so kind as to send me an off-print a little later, I read it once again with even greater interest. 

My reaction was that for my own better understanding I wanted to clarify: 

— the chronology of Zhang Qian’s mission; 

— the meaning and extent of Chinese 大夏 (Da–xia); 

— the correct reading of Chinese 月氏 (“Ar–si” in place of the mistaken “Yue–zhi”). 

My comments kept growing over the next six or seven months and in time I found out 

that the last topic was a very complex one and called for a separate paper. In the end, it was 

superceded by a topic which evolved from the others: the number of nomadic nations that 

ended Greek rule in Bactria. 

(1) CHRONOLOGY OF ZHANG QIAN. 

As far as Zhang Qian’s famous mission was concerned, I found it strange that for the 

year of his arrival at the Arsi court, then on the north side of the Oxus River, I found so many 

different figures. This was odd because the one and only source on this is the oldest Chinese 

history book, the ›Shiji‹, mainly chapter 123, where Zhang Qian’s ›Report‹, at least in part, is 

reproduced. He tells us therein that the Arsi had conquered the Daxia who were, however, 

without a king. Many later authors understood this to mean that the Arsi had not really taken 

over the lands south of the Oxus. 

If Zhang Qian had arrived years after the advent of the Arsi, this view would be admissible. 

But the following study shows that Zhang Qian arrived on the scene within a few 

months of the Arsi takeover. From this it follows that the Daxia had become subjects of the 

Arsi who were now fully involved in establishing a new order. They pointedly showed Zhang 

Qian the flourishing markets in 藍市 Lanshi, the old capital of Daxia. It was only in the very 

beginning that the Arsi, coming from Sogdiana, had preferred to establish the (rather provisional) 

court of their king (“ordos”) on the near (north) side of the Oxus. 

(2) MEANING OF 大夏 DAXIA. 

The great problem of the otherwise excellent Chinese sources is their fearful distortion of 

foreign names. This is so to the present day. What unsuspecting reader would guess that 美 

國 , the State of “Mei,” is the transcription of (A)me(rica) ? Since high antiquity, the Chinese 

transcribed foreign names in cumbersome ways and then abbreviated these drastically — 

and not always to the first two or three syllables of such a name. 

The present study shows that the old (wholly impossible) equation Daxia = Bactria has 

blocked the correct interpretation of the country, people and language of Daxia. Even the 

great CHAVANNES, undisputed authority in matters Chinese, fell into this trap, printing Ta-hia 

= Bactria. Later he did a fine translation of a chapter of the Tangshu and stated: “ Notice sur 

le T’ou-ho-lo (Tokharestan). Le T’ou-ho-lo ... c’est l’ancien territoire (du royaume) de Ta-hia.” 

Hence, Daxia was the ancient Tachar, the later Tocharistan. Tachar is not an equivalent 

of Bactria, it is only its eastern portion: this makes a decisive difference. The second chapter 

of the following study is based on the three crucial identifications: 

— 大夏 Daxia = Tachar (not Bactria); 

— 藍市 Lanshi = Darapsa (not Bactra); 

— 濮達 Puta = Bactra (not Pu•kalåvatð). 

All other assumptions are consequences of these three equations. They help us to understand 

that the Daxia of Zhang Qian are the Tochari of Trogus — which is not at all surprising 

as Trogus, too, states that the Asiani (Arsi) became the kings of the Tochari (Daxia). 

IV

These identifications also help us to realize that the Arsi, ruling Daxia from Lanshi in 

the times of the Former Han (206 BCE – 25 CE), are still ruling from there at the beginning of 

theLater Han (26 CE), i.e., over a hundred and fifty years later. This means that the Arsi had 

been confined to Tachar for a long time — held in check by their immediate western neighbors, 

the awesome Parthians, suzerains of a vassal Saka state in Bactra. The Parthians even 

tried to drive the Arsi out of Tachar again. Trogus, in the Epitome of Justin, tells us that Artabanus 

attacked the Tochari, in about 123 BCE. The Parthian king is killed in action and the 

situation remains undecided. About two generations later, in the first century BCE, the Arsi 

break through the Hindukush rampart and establish themselves in the Kabul Valley. A full 

century later still it is the founder of a new dynasty who unites all Arsi forces under his command. 

With this, he is finally in a position to attack the mighty Parthians and drive them out 

of three key positions: Kabul, Bactra and Taxila — in that order. 

(3) STRABO’S LIST OF FOUR. 

Zhang Qian clearly describes the Daxia as the indigenous population of Eastern Bactria. 

With this, it can be shown that the Daxia, or Tacharians, have never been conquering nomads. 

With Zhang Qian we know that the Tachari had dwelled in the land of their name 

since at least many centuries — and under a wide range of foreign invaders: Achaemenid 

Persians, Alexander the Great, Bactrian Greeks, Central Asian Sakas and then the Far 

Eastern Arsi, who all left their mark on the Tacharian language. 

It has often been repeated that Strabo lists four, but Trogus only two conquering nomad 

nations in connection with the fall of Greek Bactria — and that the Chinese sources know 

just one such nation, the Arsi. This study establishes the fact that the Chinese historians, too, 

speak of t w o conquering peoples. In the ›Hanshu‹ we are told something of prime importance 

– overlooked in all translations: the 塞王 Saiwang or “ Royal Sakas” had briefly ruled 

in Daxia/Tachar before they were evicted from this part of Bactria by the Arsi. 

With Trogus corroborated by the ›Shiji‹ and the ›Hanshu‹, Strabo’s vexed list becomes the 

main target for the concluding investigation. One name on that list has always been questioned. 

It will be shown that, in fact, t w o names do not belong on that list: the Pasiani and 

the Tachari. Strabo left an unpublished manuscript when he died. It contained hundreds of 

marginal notes. With this we may assume that Strabo had added the two names in the margins 

of his manuscript – as a gloss on his Asioi. The unknown editor of his Geography took it 

for granted that Strabo wanted to add these two names to his list — which so far included 

only the Asioi (Arsi) and the Sakaraukai (Saiwang). Strabo in Amaseia had used the very 

same source as before him Trogus in Rome: the ›Parthian History‹ of Apollodoros of Artemita. 

In the past, the fall of the Greek kingdom in Bactria has always been reconstructed in a 

way which remained in contradiction to this or that part of the historical evidence. Based on 

a step by step evaluation of both the Western and Eastern sources, quoted verbatim, this study 

outlines the complex sequence of historical happenings which lead to the destruction of 

Greek power north of the Hindukush. 

Thus, new insight is gained in a number of different topics. The more important are: 

— The genuine Tacharians have for centuries been firmly settled in Tachar/Tocharistan; 

— the conquering Arsi were confined to Tachar, or Eastern Bactria, for over 150 years, held in 

check by the more powerful Parthians; 

— in this long time the Arsi become known as the new (or pseudo-) Tacharians; 

— the balance of power, in favor of the Parthians so far, is only reversed by the mid-first century 

CE when a self-proclaimed Arsi king manages to evict the mighty Parthians from the 

Kohistan, Western Bactria, the Panjab, and finally the whole of the Indus Valley. With this, 

the foundations were laid for a new superpower in Central and South Asia: that of the Arsi 

under the Kushan dynasty. Ptolemy, in the later 2nd century CE, splashes the name Tachori 

(and variants) over all the places where the Arsi had been in the past three hundred years, 

culminating in his calling the last “ ordos” of the Arsi in the Far East — close to Han China, 

the Zhaowu 昭武 of the Chinese sources (modern Zhangye 張掖 ) — Qog£ra (Thogara). 

V

張騫 


THE SECRET MISSION 

OF HAN EMPEROR WU IN SEARCH OF THE RUZHI (YUEZHI) 



1. IN WHAT YEAR DID ZHANG QIAN REACH THE OXUS RIVER ? 

Ever since the publication, in 1738, of GOTTLIEB SIEGFRIED BAYER’s Historia Regni 

Graecorum Bactriani, St. Petersburg, the Hellenistic kingdom in distant Bactria has 

intrigued students and scholars of Asian history. The exact time and circumstances of 

the foundation of this ancient kingdom, in about the middle of the third century BCE, 

have always been hotly debated. But the collapse of this highly developed culture — a 

vibrant blend of Greek, Persian, Indian and local influences — about a century later 

has proved even more difficult to elucidate, beyond the fact that it was due, as is well 

known and universally accepted, to an onslaught of an uncertain number of nomadic 

peoples, bursting forth from the wide steppes in the north-east of Central Asia. 

Regarding the violent end of the Græco-Bactrian kingdom, north of the Hindukush 

Mountains in present North Afghanistan, we have very short classical Western sources: 

the extant Prologi of Pompeius Trogus and a few statements in the Geography of Strabo. 

Of the 44 books of Trogus’ World History, published in the time of Augustus, books 

41 and 42 primarily contained the history of the Parthians. But they also included remarks 

on the history of the latter’s eastern neighbors, the Græco-Bactrians. Trogus’ 

bulky work has been lost, however, only his just mentioned Prologi and a pitiful Epitome, 

done by a later hand and containing hardly more than one tenth of the original 

work, have come down to us. Also lost is the original source book for both Trogus and 

Strabo, namely the Parthian History by Apollodoros of Artemita. And lost, too, is Strabo’s 

other (and earlier) main work, his History, which may have contained a chapter 

on the history of the then eastern extremity of the Græco-Roman world. 

As far as our written sources are concerned, it is a fortunate fact that we have on 

the downfall of the Greeks in Bactria — for the first time in world history —, not only 

Western, but also Eastern sources to draw from. These are the Chinese Standard Histories 

正史 , mainly the first two, the Shiji 史記 and the Hanshu 漢書 . These two 

Chinese history books reproduce the precious report of our sole eyewitness on the 

scene, the Chinese emissary of Han Emperor Wu 漢武帝 : Zhang Qian 張騫 (d. 114 

BCE) by name — a man of outstanding abilities. To his sharp senses we owe a good 

number of first-hand observations which, albeit in an abridged form only, have come 

down to us. 

The present paper endeavors to extract from the ancient Chinese sources what 

Zhang Qian has to tell us about Bactria — and compare it with the knowledge from 

our classical Western sources. In this connection it is curious to note that the actual 

year in which Zhang Qian arrived at the shores of the Oxus River (modern Amu Darya) 

and in Bactria is still very much disputed among modern scholars. 

— 1 —

The Chinese sources, however, are unequivocal about this year. Various texts in the 

Shiji unmistakably state that Zhang Qian — the first Chinese envoy who traveled so far 

west — was sent out in a secret mission by Emperor Wu, and eventually returned to 

the Chinese capital Chang ’an 長安 in the spring of 126 BCE. 

It is narrated in the Shiji that this mission lasted 13 years. And Zhang Qian spent: 

— “more than 10 years” 十餘歲 in captivity with the Xiongnu; 

— “more than 1 year” 留歲餘 with the Ruzhi (Yuezhi) 月氏 in Daxia (i.e. in the eastern 

half of Bactria); 

— “more than 1 year” 歲餘 a second time as captive of the Xiongnu. 

To this we have to add short periods of time for four journeys: 

— starting from Longxi 隴西 , the border town, until being arrested by the Xiongnu; 

— escaping from the Xiongnu (near Shule, i.e. Kashgar) until reaching the 月氏 ; 

— returning from Daxia until being arrested by the Xiongnu again; 

— escaping from the ordos of the Xiongnu chanyu 單于 until reaching Chang‘an. 

All four journeys together must have lasted less than 1 year. Of the first we can surmise 

that it lasted only days or weeks. The second one across the Pamirs and Sogdiana may 

have lasted some 3–4 months. The third one 

is not so easy to estimate, but cannot have 

been shorter than 3 months. The fourth should have been a matter of weeks as the distance 

was short and Zhang Qian this time escaped in the company of Yudan 於單 , the 

deposed Xiongnu crown prince, and the two men were able to help each other effectively: 

Yudan in the Xiongnu Empire and Zhang Qian in Han China. 

With this information, it is clear that the historic mission started in the spring 

of 139 BCE — and not in 138, as even some Chinese and Japanese scholars believed 

or still believe. It is clear, therefore, that our Chinese Ulysses arrived at the ordos or 

court of the 月氏 in the company of just his Xiongnu servant Gan Fu 甘父 “after more 

than 10 years” as a captive of the Xiongnu and some three to four months traveling, i.e. 

in the summer of 129 — and not 128 as most modern texts erroneously tell us. This 

is a particularly important correction. 

Zhang Qian left Daxia in the fall of 128 and spent all of 127 with the Xiongnu again, 

and then escaped a second time in late winter or early spring of 126 BCE. 

In exile, the Xiongnu crown prince Yudan was made a “marquis” by Han Emperor 

Wu on May 2, 126 BCE, but soon afterwards he died. 

The third year ›yuan–shuo‹, fourth 

month, (day) ›bingzi‹, was (the start 

of) the first year of Yudan as marquis. 

In the fifth month he died. 

Shiji 20. 1031 

元朔三年四月丙子侯于單元 

年 

五月卒 

It is surely of particular importance to know the exact time of arrival of Zhang Qian 

at his destination — the court of the Ruzhi (Yuezhi) 月氏 , whom he found newly established 

on the north bank of the Oxus River, the modern Amu Darya. 

As for the Ruzhi 月氏 , a short remark on their name is due here. The reading 

“Ruzhi” for Chinese 月氏 , which I give in this article, is rather new and still widely unknown 

in the West. But as early as 1991: 92, the authoritative “Lexicon to the Shiji,” or 

Shiji Cidian 史記辭典 , decreed: 

月氏 , pronounced Ròuzhð —【月氏 (ròu zhð 肉支 )】... 

n for this is that 月氏 is in fact an ancient 肉氏 , to be read accordingly. 

The reaso 

The magnificent catalogue to the exposition Ursprünge der Seidenstrasse (“The Origins 

of the Silk Road”), which I saw here in Berlin in December 2007 and which is 

based on originally Chinese texts, states on page 286: 

... die Yuezhi (nach anderer Lesart: Rouzhi) ... 

— 2 —

The modern reading of the Chinese character 肉 (meat) is “ròu,” but in ancient 

times the reading was “rù” — which I prefer as it is closer to our extant Western 

names of the 月氏 , namely Rishi(ka), Asioi / Asiani, Arsi and ÅrÝi. For the same reason, 

I like to see in › 氏 ‹, read ›zhð‹ with Shiji commentator Zhang Shoujie 張守節 (8th c.), 

the closest-possible Chinese approximation to the sound of ›si‹, thus giving us Ru–si — 

so much closer to the above Western names than “Yue–si” can ever be. 

There is still more sound evidence for this identification. Whereas there are indications 

that the sound of › 月 ‹ was › i u ɐ t ‹ in Middle Chinese, this sound, with the 

help of Uighur-inherited pronunciations of Chinese characters, has been reconstructed 

recently as an older, or Tang-time, › u r / a r ‹ which was written ›wr‹ and ›’r‹ in Old Uighur 

script (Prof. SHÕGAITO 庄垣內 in a lecture in Berlin, March 2007). This, in all probability, 

suggests a perfectly fitting revised reading of 月氏 as Ar–si. 

As shown above, the historic date of Zhang Qian’s arrival at the Royal court of the 

Ruzhi 肉氏 (later spelled 月氏 , and for a long time incorrectly transcribed “Yuè–zhð” 

in Pinyin) is: Summer of 129 BCE. In other words: Zhang Qian arrived in Daxia 

within a few months of the final fall of the Greek kingdom of Bactria — which, as 

can be deduced from numismatic and other evidence, still existed in the year 130 BCE. 

To know this, is indeed of importance to clearly understand the Shiji’s description of 

Daxia, located directly to the east of the final Greek possessions around the capital 

Bactra. 

It is interesting to note that the year of Zhang Qian’s arrival at the Oxus River was 

correctly calculated by DE GUIGNES in 1759 and by BERNARD in 1973. In the more than 

two hundred years between the two eminent French Orientalists we find an astonishing 

range of incorrect calculations. 1759: 24, DE GUIGNES writes: 

J’ai dit plus haut que Tcham-kiao rentra dans la Chine l’an 126 avant J.C. Il avoit employé 

treize ans à faire ce long voyage; il étoit donc parti vers l’an 139 avant J.C. Mais comme 

il étoit resté pendant dix ans prisonnier chez les Huns, il n’a pû arriver chez les Yue-chi 

que vers l’an 129 ... 

I stop quoting this early study here because the author goes on to say that Zhang 

Qian 張騫 (Tcham–kiao) stayed with the Ruzhi 月氏 until the year 127 — & peut-être 

une partie de 126. This, of course, is clearly impossible. 1973: 111, BERNARD writes: 

Il est incontestable qu’en 129 av. J.-C. — la date du voyage de Chang K’ien est fixée de 

façon sûre par les annales chinoises — la Bactriane avait perdu son indépendance politique 

au profit des Yué-chi, mais elle gardait encore l’identité d’un état vassal ... 

Here we see why it is so important to know the exact time of Zhang Qian’s arrival at 

the Ruzhi 月氏 court: it is the closest terminus ante quem for the final destruction of 

Greek Bactria we know of. Yet, almost nowhere in our modern Western literature — as 

far as I can ascertain — are we told that Zhang Qian arrived at the Oxus and the court 

of the 月氏 so early. This is all the more surprising as we have, today, as two thousand 

years ago, just one primary source to guide us: the Shiji, or magnum opus of 

Sima Tan (d. 110) and his son, Sima Qian (145–c.86). In 1825: 115-116, RÉMUSAT writes: 

L’empereur … choisit pour son ambassadeur Tchhang-kian, qui partit, accompagné de 

quelques autres officiers, pour aller trouver les Youeï-chi dans le lieu où ils s’étoient retirés 

… Tchhang-kian avoit à traverser, pour venir dans la Transoxane, des contrées qui 

étoient au pouvoir des Hioung-nou. Ceux-ci eurent connoissance de l’objet de son voyage, 

et réussirent à lui couper le chemin. Lui et ses compagnons furent arrêtés et retenus dix 

ans prisonniers … 

Ils parvinrent à s’échapper, et vinrent d’abord dans le Ta-wan … En voyant Tchhangkian, 

ils eurent beaucoup de joie … ils s’empressèrent de lui donner toute sorte de facilités 

pour aller dans la Sogdiane. Ce fut là qu’il apprit que les Youeï-chi … s’étoient rendus maîtres 

de Ta-hia. L’ambassadeur les suivit jusque dans ce dernier pays, au midi de l’Oxus; 

mais il ne put obtenir d’eux de quitter une contrée fertile, riche, abondante en toute sorte 

— 3 —

de productions, pour revenir dans les déserts de la Tartarie faire la guerre aux Hioung-nou. 

Tchang-kian, fort mécontent du mauvais succès de sa négociation, et ayant encore perdu 

une année chez les Youeï-chi … il prit sa route à travers les montagnes du Tibet: mais cela 

ne lui servit de rien; les Hioung-nou, dont les courses s’étendoient jusque là, le prirent encore 

une fois, et le retinrent assez long-temps. Il parvint enfin à s’échapper, à la faveur des 

troubles qui suivirent la mort du Tchhen-iu régnant, et revint en Chine après treize ans 

d’absence, accompagné d’un seul de ses collègues, le reste de l’ambassade. 

Leaving aside a few minor flaws in this rendering of the story as told in Shiji 123, 

RÉMUSAT does not give his readers any idea about the absolute chronology of Zhang 

Qian’s historic mission. To do so, he would have had to state the name of the Xiongnu 

chanyu 單于 (emperor) who had died when Zhang Qian finally escaped in the company 

of that chanyu’s son and crown prince. To find out, RÉMUSAT would have had to 

read Shiji 110. We do not know whether he did. From this early translation, it is impossible 

to know in what year Zhang Qian arrived at the Oxus River. 

One year later, 1826: 57, KLAPROTH told the story this way: 

La nation de Yue tchi habitait alors entre l’extrémité occidentale de la province de Chen 

si, les Montagnes célestes et le Kuen lun, c’est-à-dire dans le pays que nous appelons à 

présent le Tangout, où elle avait formé un royaume puissant. En 165, les Hioung nous l’attaquèrent, 

la chassèrent à l’occident, ou elle se fixa en Transoxiane. 

L’empereur Wou ti rechercha l’alliance des Yue tchi, parcequ’il espérait qu’ils se réuniraient 

avec lui contre les Hioung nou. Le Tchhen yu ayant pénétré ce dessein chercha tous 

les moyens pour le faire échouer. 

Tchang khian s’était offert à l’empereur pour entreprendre le voyage en Transoxiane, et 

il avait demandé à être accompagné d’environ cent hommes; mais, en passant par le pays 

des Hioung nou, il fut arrêté avec sa suite et retenu prisonnier pendant dix ans; au bout de 

ce temps il trouva l’occasion de s’évader, et marcha du côté de l’ouest. Il trouva les Yue tchi 

dans leur nouveau pays. L’envoyé chinois y séjourna pendant plus d’un an, au bout duquel, 

repassant chez les Hioung nou, il fut fait de nouveau prisonnier; mais il s’échappa, et revint 

en Chine après treize ans d’absence. 

In the margin of his text, next to the line Tchang khian s’était…, KLAPROTH gives 

“126 av. J.-C.” This way it is left to the imagination of the reader whether this absolute 

year applies to the departure from Chang’an, the arrival at the Oxus, or the return to 

China of Zhang Qian. A few pages later KLAPROTH adds: 

Le voyage que le général chinois Tchang khian entreprit, en 126 avant notre ère, dans 

les pays occidentaux, avait pour but de susciter des ennemis aux Hioung nou. 

From this sentence, readers were led to believe that the year stated was that of the 

departure of Zhang Qian. We may note here that we have to combine the texts of the 

two translators to get close to what is actually said in Shiji 123. And it is interesting to 

see that from now on Zhang Qian will always be a general in this story — as if he had 

been undertaking a military mission. In reality, this secret mission was purely 

political. Zhang Qian was only made a general a few years after his return to China as 

a reward for his merits as an ambassador. 

In 1836: 37-38, RÉMUSAT writes in a foot note to his splendid translation, published 

posthumously, of the Foguoji 佛國記 , or “Memoirs of the Buddhist Kingdoms,” by the 

Chinese Buddhist monk and pilgrim to India in the years 399–414, Fa Xian 法顯 , 

edited by KLAPROTH, who also died before the final publication: 

Tchang khian, que DEGUIGNES, par erreur, a nommé Tchang kiao, est un général chinois 

qui, sous le règne de Wou ti de la dynastie des Han, l’an 122 avant J. C., fit la première expédition 

mémorable dans l’Asie centrale. On l’avait envoyé en ambassade chez les Yue ti, 

mais il avait été retenu par les Hioung nou, et gardé dix ans chez ces peuples. Il s’y était 

même marié et avait eu des enfants. Durant ce séjour, il avait acquis une connaissance 

étendue des contrées situées à l’occident de la Chine. Il finit par s’échapper et s’enfuit à 

— 4 —

plusieurs dizaines de journées du côté de l’ouest, jusque dans le Ta wan (Farghana). De là 

il passa dans le Khang kiu (la Sogdiane), le pays des Yue ti et celui des Dahæ [Daxia]. 

Pour éviter à son retour les obstacles qui l’avaient arrêté, il voulut passer au milieu des 

montagnes, par le pays des Khiang (le Tibet), mais il ne put éviter d’être encore pris par 

les Hioung nou … Il parvint à s’échapper de nouveau et revint en Chine après treize ans, 

n’ayant plus que deux compagnons, sur cent qui avaient formé sa suite à son départ. Les 

contrées qu’il avait visitées en personne étaient le Ta wan, le pays des grands Yue ti, celui 

des Ta hia (Dahæ) et le Khang kiu ou la Sogdiane. 

Comparing these texts with the Chinese original one realizes that the translators 

mixed their own comments into their renditions. What, then, did the Chinese text of 

Shiji 123 really say? 

To be sure, as early as 1828 one of the students of RÉMUSAT published a full and pioneering 

translation of this important chapter of the Shiji. He had done it under the 

close guidance of his teacher RÉMUSA T. His name was given as Brosset jeune (“Brosset 

jun ior”) — he was Monsieur Marie-Félicité Brosset who soon abandoned his sinological 

studies in favor of other Oriental languages. One reason may have been that his 

struggles with the Chinese language went largely unnoticed by the scholarly community 

o f Western Europe. Unfortunately, young Brosset’s French translation was reproduced 

without the original text. I include it here for the sake of easy comparison. 

( BROSSET 1828: 418–421) 

Les traces des Ta ouan (Fergana) sont con- 

depuis Tchang–kien, 

nues 

capitaine des Han, en l’année ›kien–youen‹ 

( 140 ans avant J.-C.). 

A cette époque, le fils du Ciel interrogeant 

des Hiong–nou qui s’étaient soumis, apprit 

que les Hiong–nou avaient battu les Youe– 

chi, et fait une coupe du crâne de leur roi; 

qu’enfin les Youe–chi s’étaient dispersés, la 

rage dans le cœur contre les Hiong–nou, 

sans vouloir faire la paix avec eux. 

A ce récit, l’empereur des Han, qui souhaitait 

détruire les barbares des environs, et 

pour réaliser ses projets de communica- 

par des caravanes qui traverseraient 

tions 

le pays des Hiong–nou, fit chercher des 

gens capables de cette commission. 

Kien, capitaine de la caravane des Youe– 

chi, et Tchang–y–chi kou–hou nou–kan–fou 

sortirent ensemble par Long–si, se portant 

vers les Hiong–nou; 

ceux-ci les arrêtèrent et les livrèrent au 

Tchen–yu (c’était alors Lao–chang). 

Le Tchen–yu les retint … 

Il les garda dix ans et leur donna des femmes. 

Mais Tchang kien, qui avait ses instructions 

des Han et ne les perdait pas de vue; 

se trouvant tous les jours plus libre au mi- 

des Hiong–nou, s’échappa avec ses 

lieu 

compagnons, se dirigeant vers les Youe–chi 

( ils émigrèrent vers la grande Bucharie, en 

l’an 139 avant Jésus-Christ); 

Shiji 123. 3157–3159 

大宛之跡見自張騫 

張騫漢中人 

建元中為郎 

是時天子問匈奴降者皆 

言匈奴破月氏王以其頭 

為飲器 

月氏遁逃而常怨仇匈奴 

無與共擊之 

漢方欲事滅胡聞此言因 

欲通使道必更匈奴中乃 

募能使者 

騫以郎應募使月氏與堂 

邑氏 ( 故 ) 胡奴甘父俱 

出隴西 

經匈奴匈奴得之傳詣單 

于單于留之 … 

留騫十餘歲與妻有子然 

騫持漢節不失 

居匈奴中益寬騫因與其 

— 5 —

et après quelques dixaines de jours de mar- 

il arriva à Ta ouan. 

che, 

Les gens du pays avaient entendu parler 

de la fertilité et des richesses des Han; 

mais, malgré tous leurs désirs, ils n’avaient 

pu nouer de communications. Ils virent 

Kien avec plaisir … 

Sur sa parole, le roi de Ta ouan lui donna 

des guides et des chevaux de poste, qui le 

menèrent à Kang–kiu (Samarkande). De là 

il fut remis à Ta–youe–chi. 

Le roi des Youe–chi avait été tué par les 

Hiong–nou, et son fils était sur le trône. 

Vainqueurs des Ta–hia (habitans du Candahar) 

les Youe–chi s’étaient fixés dans 

leur pays, gras et fertile, peu infesté de voleurs, 

et dont la population était paisible. 

En outre, depuis leur éloignement des Han, 

ils ne voulaient absolument plus obéir aux 

barbares. 

Kien pénétra, à travers les Youe–chi, à Ta– 

hia, et ne put obtenir des Youe–chi une lettre 

de soumission. 

Après un an de délai, revenant au mont 

Ping–nan, il voulut traverser le pays de 

Kiang; mais il fut repris par les Hiong–nou. 

Au bout d’un an, le Tchen–yu mourut. Le 

Ko–li–vang de la gauche battit l’héritier de 

la couronne, et se mit en sa place; l’intérieur 

du pays était en combustion. 

Kien, conjointement avec Hou–tsi et 

Tchang–y–fou, s’échappa et revint chez les 

Han (en l’année 127 avant J.-C.) ... 

屬亡鄉月氏西走數十日 

至大宛 

大宛聞漢之饒財欲通不 

得見騫喜 … 

大宛以為然遣騫為發導 

繹抵康居康居傳致大月 

氏 

大月氏王已為胡所殺立 

其太子為王 

既臣大夏而居地肥饒少 

寇志安樂又自以遠漢殊 

無報胡之心 

騫從月氏至大夏竟不能 

得月氏要領 

留歲餘還並南山欲從羌 

中歸復為匈奴所得 

留歲餘單于死左谷蠡王 

攻其太子自立國內亂 

騫與胡妻及堂邑父俱亡 

歸漢 … 

This then is what the two reputed Sinologists and one of their most ambitious stu- 

dents translate for those many Western Orientalists, Historians, Geographers etc. 

who — in this and over the next few generations — are unable to read Chinese them- 

selves. What do the latter make out of the translations and narratives ? 

RITTER, 2 1837: 545; 547, writes: 

Einfluß des chinesischen Reiches auf West-Asien unter der Dynastie der Han (163 vor bis 

196 nach Chr. Geburt). Tschangkians Entdeckung von Ferghana, Sogdiana, Bactrien und 

der Handelsstraße nach Indien, um das J. 122 

vor Chr.G. … Hier ist der Ort, unter diesem 

Kaiser seines chinesischen Generals, Tschangkian, dessen wir schon früher einmal gedachten 

(Asien I, S. 201, 195), genauer zu erwähnen, als des Entdeckers Sogdianas, des Cas- 

sondern als politischer Missionar, um das 

pischen Meeres und Indiens, nicht als Eroberer, 

Jahr 122 vor Chr. Geb. … 

It is not altogether clear, but one may guess that RITTER took the year 122 BCE as 

the time of discovery, i.e. the year of Zhang Qian’s arrival in the Far West. 

LASSEN in 1838: 250 writes: 

In diesen Szu hat man längst die Saker er kannt und es stimmt damit, dass die Saker 

sich schon vor dem Falle des Baktrischen Reiches eines Theils Sogdianas bemächtigt hatten 

… Die Yuetchi stossen die Szu weiter und nehmen die von ihnen besetzten Gebiete ein; 

— 6 —

die Szu nach Süden gedrängt finden Gelegenheit, sich des Landes Kipin zu bemächtigen, 

die nachrückenden Yuetschi nehmen das Land der Tahia. Ein Chinesischer General 

Tchamkiao war auf diesem Zuge bei den Yuetschi und das wohlbegründete Ereignis fällt 

in die Zeit unmittelbar vor 126 vor Chr. Geburt. 

In 1829 and in St. Petersburg, the first gr 

eat Russian Sinologist BICHURIN published 

a translation of Hanshu 96, which does not contain Zhang Qian’ s biography nor his 

mission to the 月氏 — these went into Hanshu 

61 —, but it does mention Zhang Qian’s 

name a few times and includes an updated 

description of the people of the 月氏 and 

its early history. This translation into a We 

stern language went almost unnoticed by 

Western scholars. An exception is SCHOTT w ho published a book review of it, here in 

Berlin. In 1841: 164-165; 169, SCHOTT states: 

… gab der Pater Jakinph (Hyacinth) Bitschurinskij, früher eine Zeitlang Archimandrit 

an dem Griechischen Kloster in Peking … bereits vor zwölf Jahren vorliegendes Werk hererdienstlich, 

besonders … da der Verfasser 

aus … aber seine Arbeit ist gleichwohl sehr v 

hier aus einer Quelle geschöpft hat, die bis jetzt keinem Europäischen Sinologen zugäng- 

Si-yü-tschuan (Kunde von den Si-yü, 

lich gewesen … Diese Beschreibung, im Originale 

westliche Grenz-Regionen) betitelt, bildet einen integrierenden Theil der Annalen jenes 

Kaiserhauses, welches die Pariser Bibliothek schwerlich besitzen dürfte; denn Abel-Remuie 

vorfinden … In seiner 18 Seiten starken 

sat hat seine Beiträge zur alten Geschichte Mittelasiens nur aus den Resumé’s entlehnt, 

die sich in Ma-tuan-lin’s kritischer Encyklopäd Vorrede macht Pater Hyacinth folgende Bemerkung[en] : 

… Aber zwei Jahrhunderte vor u. Z. stiftete ein nördliches Barbarenvolk, von den Chine- 

in Central-Asien, die das Reich 

sen Hiong-nu genannt, eine ungeheure Steppen-Monarchie 

der “Himmelssöhne” in langwierigen Kämpfen 

demüthigte, und der Chinesische Hof mussdenken, 

um diesen gefährlichen Feind un- 

te endlich auf ausserordentliche Maassregeln 

schädlich zu machen. Gefangene Hiong-nu sagten aus, auf der Landstrecke von der 

Grossen Mauer bis Chamul (Ha–mi) habe vor nicht gar langer Zeit ein mächtiges Volk — 

die Yue-tschi oder Yue-ti (Geten) — gewohnt, das aber, von den Hiong-nu verdrängt, ins ferne 

Abendland ausgewandert sei. 

Da schickte Kaiser Wu-ti (140 bis 85 vor Chr.), in der Hoffnung dieses Volk gegen die 

Hiong-nu aufzureizen, seinen General Tschang-kian als Bevollmächtigten an sie ab. Die 

Hiong-nu lauerten diesem Magnaten auf, und hielten ihn zehn Jahre lang in gefänglichem 

Gewahrsam, bis er endlich Gelegenheit fand zu entfliehen, und nun durch Fergana und 

Sogdiana zu den Yue-ti gelangte. Allein der Fürst dieser Nation, welcher die Ta-hia (Dacier) 

unterworfen und in ihrem Lande sich niedergelassen hatte, dachte in seinen schönen Besitzungen 

nicht mehr daran, sich an den Hiong-nu zu rächen. Tschang-kian verweilte hier 

einige Jahre, kehrte dann unverrichteter Sache zurück und fiel ein zweites Mal den Hiongnu 

in die Hände, aber Unruhen im Hiong-nu-Reiche verschafften ihm Gelegenheit, ein 

zweites Mal zu entrinnen; und so erreichte er (126 v.Ch.) endlich wieder seine Heimat … 

For the year of Zhang Qian’s return, BICHURIN’S calculation, 126 BCE, was the best 

so far. I have Bichurin’s Russian translation here before me, but regrettably not his 

preface. Hence I am unable to say, whether or not he had also calculated a definite year 

f or the arrival of the “general” at the court of the “Geten” (the Massagètes of RÉMUSAT 

1829: 220). Anyway, beyond KLAPROTH and SCHOTT, few scholars in the West read BI- 

CHURIN’S translation of Hanshu 96. Among those who did not read it is the famous 

French geographer, VIVIEN DE SAINT-MARTIN. In 1850: 261–262, 267 (foot notes); 265, 292– 

293 (main text), he writes: 

Cet officier se nommait Tchang-Khian. Parti de la cour impériale en l’année 126, il fut arrêté 

en chemin par les Hioung-nou, qui pénétrèrent l’objet de sa mission, et qui le retinrent 

parmi eux. Tchang-Khian, parvenu enfin à s’évader après dix années de captivité, ne put 

conséquemment arriver chez les Yué-tchi qu’en l’année 116, et en effet il les trouva bien établis 

dans la Transoxane, qu’ils possédaient depuis dix ans … 

— 7 —

Mais ce qu’il nous est surtout important de connaître plus en détail, c’est la nation même 

des Yué-tchi … Le Pline chinois, Ma-touan-lin, a réuni au XIII e siècle ces anciennes notions, 

encore augmentées de notions plus récentes, et en a formé un article spécial parmi 

ceux qu’il consacre aux nations de l’intérieur de l’Asie. Nous insérons ici la traduction de ce 

morceau, qu’a bien voulu nous fournir M. Stanislas Julien; elle complète et rectifie en 

beaucoup de passages essentiels celle qu’Abel Rémusat en a donnée … 

Abel Rémusat et Klaproth identifient constamment le Ta-hia des relations chinoises 

avec la Bactriane, c’est-à-dire avec la partie orientale du Khoraçân actuel. Ce rapprochement 

ne nous paraît pas exact. Nous ne voyons nulle raison de nous éloigner ici de la synonymie 

naturelle que nous fournit la situation des Dahæ dans l’ancienne géographie classique, 

sur la côte S.-E. de la mer Caspienne, au midi de l’ancienne embouchure de 

l’Oxus … 

Ce que nous voyons quant à présent avec certitude, c’est … qu’après avoir séjourné pendant 

trente ans environ dans les pâturages de la Dzoûngarie, les Yué-tchi furent contraints 

par un nouveau refoulement de pousser plus loin leur émigration; qu’ils descendirent alors, 

vers les années 130 à 126 avant notre ère, dans les steppes du nord du Jaxartès, et que 

bientôt après, franchissant ce grand fleuve, ils vinrent s’emparer, en l’année 126, des riches 

provinces qui avaient appartenu peu avant aux rois grecs de la Bactriane, entre le Jaxartès 

et l’Oxus; qu’ils y établirent dès lors leur domination exclusive … 

It was a disaster of sorts that the Western translations of the “Chinese sources” 

should start with the late “Encyclopedia” 文獻通考 of MA D UANLIN 馬端臨 instead of 

w ith the Chinese Standard Histories 正史 , which MA DUANLIN had reworked in a very 

superficial, confused, or at least confusing, manner. It took a long time to repair the 

damage. Later authors strongly warned against using MA DUANLIN indiscriminately. 

Once again I must quote LASSEN who, in 2 1874: 370-371, writes: 

Die Zeit dieses Ereignisses lässt sich mit ziemlicher Genauigkeit nach den Berichten 

über die Sendung des Chinesischen Generals Tchangkian zu den Jueïtchi feststellen. Der 

Kaiser Wuti aus der Familie der Han, welcher von 140—80 vor Chr. G. regierte, in der Absicht, 

die Hiungnu zu nöthigen, ihre Waffen gegen Westen zu richten und dadurch sein 

Reich von ihren fortwährenden räuberischen Einfällen zu befreien, beschloss, ein Bündnis 

mit ihren Feinden, den Jueïtchi, zu schliessen und sie zu einem Kriege gegen sie zu bewegen; 

er beauftragte den oben genannten General mit der Unterhandlung. Als dieser die 

Jueïtchi erreichte, fand er sie schon im Besitze von Tahia und nicht geneigt, sich an den 

Hiungnu zu rächen … Da sie ausserdem zu entfernt von den Chinesen wohnten, konnten 

sie sich nicht entschliessen, dem Tchangkian den Oberbefehl über ein Heer zu geben und 

in die raue und wüste Gegend ihrer früheren Wohnsitze zurückzukehren. Der Gesandte 

des Chinesischen Kaisers kehrte daher unverrichteter Sache in sein Vaterland zurück. 

Das Jahr seiner Rückkehr wird nicht übereinstimmend angegeben. Nach einer Angabe 

kehrte er im Jahre 126 vor Chr. G. zurück, nach einer andern 122. Der älteste Chinesische 

Geschichtsschreiber, bei welchem eine Bestimmung hierüber sich findet, Ssémathsien, 

lässt die Abreise zwischen den Jahren 140 und 134 vor Chr. G. stattfinden (in seinem Sséki, 

§ 123). Es bleibt daher zweifelhaft, ob die zwei Jahre, welche er bei den Jueïtchi zubrachte, 

von 130 oder 124 an zu zählen sind ... Da die Angabe, dass Tchangkian im Jahre 122 zurückkehrte, 

sich in einem aus Chinesischen Quellen geschöpften Werke findet, möchte sie 

als die richte betrachtet werden. 

LASSEN shows great respect for RÉMUSAT’s translation of the first of the famous 

Chinese Buddhist pilgrims who came to the holy land of India and wrote detailed repor 

ts: primary sources of the highest importance. In 1874 LASSEN copies RÉMUSAT’s 

mistake of 1836. But he also remarks that the year 122 is in clear contradiction to an- 

of RÉMUSAT’s notes, namely that Zhang Qian, after his return, was made a mili- 

other 

t ary commander in 123 BCE. LASSEN had not been told that Shiji 123 states in simple, 

u nmistakable terms Zhang Qian’s year of return (at least as far as the authors were 

concerned — how should they know that later readers would no longer be familiar 

— 8 —

with the Chinese calendar ?). LASSEN was caught between doubt and praise. One soluion 

to his dilemma could be that the figure 122, in fact, was a printer’s mistake for 

t 

127 — the year of return that “BROSSET jeune” published in 1828, worked out with his 

t eacher RÉMUSAT. With all its shortcomings BROSSET’s early translation remained the 

main entry point to the Chinese sources for the next author. 

In 1877: 448–452, VON RICHTHOFEN writes: 

Entdeckung der Länder am Oxus und Yaxartes durch Tshang-kiën (~128 v.Chr.). Als 

Hsia[o]-wu-ti (140 bis 86), der glücklichste der Han-Kaiser, zur Regierung kam, begannen 

die Hiungnu, die sich seit 160 ruhig verhalten hatten, abermals Einfälle in das Reich. Ein 

einsichtsvoller und kräftiger Regent, beschloss er, ihre Macht zu brechen und die Carawanenwege 

durch das von ihnen beherrschte Land für sich zu öffnen. Die Hiungnu hatten 

sich durch räuberische Einfälle eine Schreckensherrschaft über die Völker des Tarym- 

Beckens gesichert. Alle diese hatten ein Interesse an ihrer Niederwerfung; aber kein Volk 

konnte, wie man glaubte, in gleichem Maass Rache gegen sie brüten, wie die Yue-tshî; 

denn aus dem Schädel ihres im Jahre 157 erschlagenen Königs war ein Trinkgefäss gemacht 

worden. Sie mussten als Bundesgenossen gewonnen werden. 

Ein General Namens Tschang-kiën wurde beauftragt, sie in ihren neuen Wohnsitzen 

aufzusuchen. Seine Reise ist von hohem Interesse, denn sie ist die erste chinesische Expedition 

nach fernen Gegenden im Westen, von der wir Kunde haben. Wahrscheinlich war es 

in der That die erste; denn der Bericht hat die Färbung einer abenteuerlichen Entdeckungsreise 

nach ganz unbekannten Ländern (ich folge der Erzählung im 123sten Buch des 

Sse-ki von Sz’ma-tsiën nach der dankenswerthen Uebersetzung von Brosset ... 1828, p. 418– 

450, da dieser Bericht nur 40 Jahre nach der Aussendung von Tschang-kiën geschrieben 

wurde und in hohem Grade das Gepräge der ungeschminkten Wahrhaftigkeit trägt; eben- 

so benutze ich die von Brosset berechneten Jahreszahlen, nach welchen die Gesandtschaft 

im Jahre 127 zurückkehrte, also 139 auszog, während sie gewöhnlich, nach Ma-twan-lin, in 

die Jahre 136 bis 123 verlegt wird). 

Um das Jahr 139 verliess Tshang-kiën seine Heimath mit einem Uiguren Namens 

Tshung-i, welcher wahrscheinlich mit manchen Wegen in Central-Asien bekannt war, und 

einer Begleitung von 100 Mann. Nach zehnjähriger Gefangenschaft bei den Hiungnu entkamen 

sie und setzten ihre Reise nach dem Reich Ta-wan am Yaxartes fort, wo sie die Yuetshî 

vermutheten. Sie hörten, dass diese weiter, nach dem Oxus, in das Land der Ta-hiâ, 

gezogen seien ... Dort, berichtet er, fand er die Yue-tschî nördlich vom Fluss Wei (Oxus) 

wohnend ... Sie empfingen ihn gut, erklärten aber, dass ihr Land fruchtbar sei, und sie 

darin glücklich, friedlich und der Plünderung wenig ergeben lebten; sie konnten sich nicht 

entschliessen, in ihre früheren rauhen und öden Wohnsitze zurückzukehren, um die alten 

Feinde zu bekriegen. Das Nomadenleben hatten sie noch nicht abgelegt. 

Auf dem Rückweg kam Tschang-kiën nach dem Gebirge Ping-shan und wollte von da 

durch das Land der Kiang gehen, wurde jedoch von den Hiungnu gefangen genommen 

und entkam nach einem Jahr. Erst im Jahre 127 kehrte er mit Einem aus seinen 100 Begleitern 

an den kaiserlichen Hof zurück. Sein Hauptzweck war verfehlt. Er hatte die gewünschten 

Bundestruppen nicht mitgebracht. Aber er hatte Wichtigeres erreicht. Denn er 

konnte seinem Kaiser über die Existenz grosser Völker im fernen Westen berichten ... 

Nach Feststellung der Lage von Ta-wan lassen sich die Positionen der anderen Völker 

und Reiche annähernd bestimmen. Die Khang-kiu und Yen-tsai breiteten sich am Yaxartes 

abwärts aus. Die ersteren nomadisierten wahrscheinlich in den Gegenden von Taschkent, 

Tschemkent und Turkestan, während die Yen-tsai den Unterlauf des Stromgebietes bis 

zum Aralsee einnahmen. Die Khang-kiu hatten im Nordosten die Usun zu Nachbarn. Mit 

der Residenz am Issyk-kul, breiteten sich diese wahrscheinlich am Nordfuss des Alexandergebirges 

und des Karatau über Talas hinaus aus. Südwestlich von den drei grossen 

Reichen am mittleren und unteren Yaxartes folgten einige kleine Reiche, deren Namen uns 

nicht aufbewahrt sind. 

In dem Thal von Samarkand begann das ehemalige Gebiet der Ta-hiâ, von dessen 

nördlichem Theil nun die Yue-tshî Besitz genommen hatten. Die letzteren scheinen sich 

— 9 —

ebenso nach Westen, gegen das jetzige Bokhara, als nach Südwesten bis zum Oxus ausgebreitet 

zu haben, während das unkriegerische, verweichlichte Volk der Ta-hiâ die reichen 

Handelsplätze im Süden des Oxus nebst grossen Strecken auf dem rechten Ufer desselben 

inne hatte ... Die Yue-tshî breiteten sich aus und mögen die Ta-hiâ nach Westen gedrängt 

haben, da die Dahae oder Daoi der griechischen Schriftsteller am Kaspischen Meer wohnten 

... 

With so many contradictory explanations of one and the same source text, it was 

about time for another — closer — look at Shiji 123 by those who read Chinese. 

SPECHT does, and, in 1883: 348, explains: 

Les Yué-tchi, ou Indo-Scythes, qui habitaient primitivement entre le pays des Thun- 

Hoang et le mont Ki-lian (les monts Célestes), furent vaincus, en 201 et en 165 avant notre 

ère, par les Hioung-nou. Ils s’enfuirent au-delà des Ta-Ouan, battirent les Ta-hia de la Bactriane 

dans l’ouest, et les subjuguèrent. Leur roi fixa sa résidence au nord de l’Oxus; c’est 

dans cette contrée que Tchang-kian, ambassadeur chinois, les trouva en 126 avant notre 

ère. Après le départ de ce dernier, la ville de Lan-chi, capitale des Ta-hia, tomba au pouvoir 

des Grands Yué-tschi qui s’établirent définitivement dans la Bactriane … 

Here the Ruzhi 月氏 are termed “Indo-Scythians” — an epithet which shall reapear 

regularly from now on. The mistaken appellation “Skythai” for the 月氏 dates 

p 

b ack to Strabo, for whom nine tenth of the Asian Continent were yet unknown. In his 

t ime, the 月氏 were known to have come from regions just beyond the Jaxartes. The 

G ræco-Roman historian, therefore, took it for granted that the Ruzhi 月氏 were just 

another branch of the Sakas — called Scythians by the earliest Greek historians like 

H erodotos. When the easternmost Saka tribe, the Sakaraukai/Sacaraucae, finally 

r eached India in the first century BCE, it was natural to name these genuine Scythians 

“ Indo-Scythians.” 

The Ruzhi 月氏 , however, have never been Scythians — let alone Indo-Scythians. 

Two thousand years after Strabo we know that the 月氏 originated, not from regions 

near the Jaxartes, but thousands of kilometers further east from regions north and 

w est of the Yellow River where they were neighboring the proto-Huns and the archaic 

Chinese. The Ruzhi 月氏 came, not from Central Asia, but from the Far East and ori- 

were, not of Indo-European, but of Mongoloid stock (see below, p. 71). They 

ginally 

surely looked a great deal different from any of the Scythian tribes of our classical 

sources with whom the 月氏 only shared the pastoral way of life. 

The appellation “Indo-Scythians” for the Ruzhi 月氏 is a gross misnomer. It can be 

traced back to our classical Western and Eastern sources and the painfully difficult 

and time-consuming process towards their correct interpretation in modern times. 

In the Periplus, composed around the middle of the first century CE, the Indus Val- 

from the Kabul River down to the Erythræan Sea, is still simply called Skythia. In 

ley, 

t he early first century CE this part of India was in the hands of the foreign Parthians 

w ho had inherited it from the equally foreign Sakas or Scythians. The name “Skythia,” 

then, for a country formerly occupied by the Sakaraukai, a branch of the nomadic Scy- 

makes good thians, sense. 

(CASSON 1989: 73–74; 77) Periplus 38–39; 41 

After this region ... there next comes the Met¦ d taÚthn t¾n cèran ... kdšcetai 

seaboard of Skythia, which lies directly to paraqal£ssia mšrh tÁj Skuq…aj par' aÙtÕn 

the north; it is very flat and through it keimšnhj tÕn boršan, tapein¦ l…an, x ïn 

flows the Sinthos River, mightiest of the potamÕj S…nqoj, mšgistoj tîn kat¦ t¾n 

rivers along the Erythraean Sea ... 

'Eruqr¦n q£lassan potamîn ... 

The river has seven mouths, narrow and Èpt¦ d oátoj Ð potamÕj œcei stÒmata, lept¦ 

full of shallows; none are navigable except d taàta kaˆ tenagèdh, kaˆ t¦ mn ¥lla di£ - 

the one in the middle. 

ploun oÙk œcei, mÒnon d tÕ mšson, f' oá kaˆ 

At it, on the coast, stands the port of trade tÕ paraqal£ssion mpÒriÒn stin Barbari- 

— 10 —

of Barbarikon. 

There is a small islet in front of it; and be- 

it, inland, is the metropolis of Skythia 

hind 

itself, Minnagar. 

The throne is in the hands of Parthians, 

w ho are constantly chasing each other off 

it. 

Vessels moor at Barbarikon, but all the 

cargoes are taken up the river to the king 

at the metropolis ... 

The part inland, which borders on Skythia, 

is called Abêria, the part along the coast 

Syrastrênê. 

kÒn. 

PrÒkeitai d aÙtoà nhs…on mikrÒn, kaˆ kat¦ 

nètou mesÒgeioj ¹ mhtrÒpolij aÙtÁj tÁj Sku- 

q…aj Minnag£r: 

basileÚetai d ØpÕ P£rqwn, sunecîj ¢ll»louj 

kdiwkÒntwn. 

T¦ mn oân plo‹a kat¦ t¾n Barbarik¾n diorm…zontai, 

t¦ d fort…a p£nta e„j t¾n mhtrÒ- 

¢nafšretai di¦ toà potamoà tù ba- 

polin 

sile‹ ... 

TaÚthj t¦ mn mesÒgeia tÍ Skuq…v sunor…- 

zonta 'Abhr…a kale‹tai, t¦ d paraqal£ssia 

Su[n]rastr»nh ... 

More than a century after the unknown author of the Periplus, but writing in the 

s ame Alexandria in Egypt, it is the erudite Klaudios Ptolemaios who, in his Geography 

of about 170 CE, introduces the appellation “Indoskythia.” He uses the name for the 

very same region — the lands on both banks of the Indus River from where the latter 

receives the waters of the Kabul River down to the ocean. Up north, Ptolemaios had 

named two other geographic regions “Skythia”: the “Skythia this side of the Imaon 

Mountains” and the “Skythia beyond the Imaon Mountains.” This may have been the 

simple reason why he wanted to give the Skythia in India a more dictinct name and so 

change the name in the Periplus, Skythia, to “Indoskythia.” As far as these classical 

writers were concerned, the name Skythia/Indoskythia had a lot to do with the Sakas 

or Scythians — and nothing with the Ruzhi 月氏 . The latter arrived on the scene some 

time after the Sakas: in any case after the name “Skythia” had already been applied 

to the Panjab and the lower reaches of the Indus River. 

(MCCRINDLE 1885: 136) 

India within the (river) Ganges ... 

And furhter, all the country along the rest of the 

course of the Indus is called by the general name 

of Indo-Skythia. 

Of this the insular portion formed by the bifurcation 

of the river towards its mouth is Patalênê, and 

and the region above this is Abiria, and the region 

about the mouths of the Indus and Gulf of Kanthi is 

Syrastrênê ... 

MCCRINDLE, in his translation of Ptolemaios’ Indian chapters, 1885: 136–139, writes a 

short comment on the name “Indoskythia”: it shows that the greatest misunderstandings 

started in modern times: 

Geographia 7.1.55 

TÁj ntÕj G£ggou 'IndikÁj qšsij ... 

P£lin ¹ mn par¦ tÕ loipÕn mšroj 

toà 'Indoà p©sa kale‹tai koinîj mn 

'Indoskuq…a, 

taÚthj d ¹ mn par¦ tÕn diamerismÕn 

tîn stom£twn Patalhn», kaˆ ¹ Øper- 

keimšnh aÙtÁj 'Abir…a, ¹ d perˆ t¦ 

stÒmata toà 'Indoà kaˆ ¹ perˆ tÕn 

K£nqi kÒlpon Surastrhn» ... 

Indo-Scythia, a vast region which comprised all the countries traversed by the Indus, 

from where it is joined by the river of Kâbul onward to the ocean ... 

The period at which the Skythians first appeared in the valley which was destined to 

bear their name for several centuries has been ascertained with precision from the Chi- 

horde of Tibetan extraction called Yuei- 

nese sources. We thence gather that a wandering chi or Ye-tha in the 2nd century B.C. left Tangut, their native country, and, advancing west- 

the pasture-lands of Zungaria. Here they 

ward found for themselves a new home amid 

had been settled for about thirty years when 

the invasion of a new horde compelled them 

to migrate to the Steppes which lay to the north of the Jaxartes. In these new seats they 

halted for only two years, and in the year 128 

B.C. they crossed over to the southern bank 

of the Jaxartes where they made themselves masters of the rich provinces between that 

river and the Oxus, which had lately before belonged to the Grecian kings of Baktriana. 

This new conquest did not long satisfy their ambition, and they continued to advance 

— 11 —

southwards till they had overrun in succession Eastern Baktriana, the basin of the Kôphês, 

the basin of the Etymander with Arakhôsia, and finally the valley of the Indus and Syras- 

into several tribes, whereof the most pow- 

trênê. This great horde of the Yetha was divided erful was that called in the Chinese annals Kwei-shwang. It acquired the supremacy over 

the other tribes, and gave its name to the kingdom of the Kushâns ... These Kushâns of the 

Panjâb and the Indus are no others than the 

Indo-Skythians of the Greeks. In the ›Râjatara„gi‡î‹ 

they are called Sâka and Turushka ( Turks) ... 

This is one example of how the early translations of Shiji 123 — at that time availthe 

desk of an English scholar: the broad 

able in French and Russian only — reached 

outline is there, but the details are in shambles. The geography in the Chinese narra- 

This is strange because the Chinese his- 

tive is better understood than the chronology. 

torians are extremely careful and efficient 

in their methods of dating important facts 

and events. But it is there all given more si nico — that is the greatest barrier. In this 

short exposé the Eastern Ruzhi 月氏 merge with the Central Asian Sakas. In this way 

the 月氏 become the conquerors of Sakastana (Arachosia) and later the Indo-Scythians 

of the Panjab. It took time to correct these early confused misconceptions. 

MCCRINDLE, I like to note here, has one rare observation to offer: he states that the 

Ruzhi 月氏 conquered, not Bactria, but Eastern Bactriana — or Ta-hia/Daxia 大夏 . 

SPECHT’s equation Ta-hia = Bactriana is of course a great improvement over RÉMU- 

SAT’s first guess Ta-hia = Massagètes or “Grands Gètes” (Goths). Short two years later 

MCCRINDLE comes close to hitting upon a perfect Ta-hia = Eastern Bactriana — if only 

he had been able to read Zhang Qian’s report in the Shiji himself. Instead, from now 

on Ta-hia = Bactria will be repeated by just about every author. However, in order to 

understand the complex story of the Ta-hia (Daxia 大夏 ) or Tochari properly, the 

equation with Bactria is not good enough: indeed it is still misleading. It suggests that 

the Ruzhi 月氏 conquered the whole of Bactria which — as this study will show — 

they were unable to do for a long time (see below, p. 56). Ta-hia 大夏 cannot be the 

Chinese transcription of the name Bactria. Ta-hia ( Daxia), the Chinese Standard Histous, 

was later called Tu-ho-lo (Tuhuoluo) 吐火羅 = 

ries (e.g. the New Tangshu) tell 

Tocharistan. 

NEUMANN, 1837: 181, translated: 

Tu ho lo … vor Al ters war dies das Land der Ta hia — 

in the Chinese original text very clearly given as: 

吐火羅 … 古大夏地 or: Tu-ho-lo ... (is) the country of the old Ta-hia. 

Tu-ho-lo, it was soon universally recognized, is the Chinese name for To-cha-ra. It 

is not Bactria, but only the easternmost part of it, the country later called Toxårestån 

(and also Taxårestån) by Arab authors. That this very important clarification has constantly 

been overlooked has greatly helped to confuse the issue. But what is of interest 

for us here, is indeed Specht’s statement that Zhang Qian arrived at the Oxus River in 

the year 126. Coming from a Sinologist, this is disappointing. It is unfounded and not 

much more than a guess. 

In another posthumously published work, the Non-sinologist VON G UTSCHMID, 1888: 

59– 62, explains his own understanding of the Chinese sources: 

Es stünde schlimm um unser Wissen von dem Untergange jenes in ferne Lande versprengten 

Bruchtheils des griechischen Volkes, wenn nicht die Politik der chinesischen Regierung 

ein sehr lebhaftes Interesse an den Bewegungen der innerasiatischen Nomaden 

genommen hätte: diesem Interesse verdanken wir den Bericht eines chinesischen Agenten 

… Nach diesen Quellen wohnten die Yue-tshi, ein den Tibetanern verwandtes Nomadenvolk, 

ehedem zwischen Tun-hwang (d.h. Sha-tscheu) und dem Ki-lien-shan und wurden 

hier, wie alle ihre Nachbarvölker, 177 von dem türkischen Volke der Hiung-nu unterjocht. 

Eine Erneuerung des Kampfes zwischen 167–161 bekam ihnen übel: Lao-shang, der 

Shen-yu oder Gross-chan der Hiung-nu, erschlug ihren König Tshang-lun [the name of this 

— 12 —

king is not known — GUTSCHMID is quoting in the wrong place the mistaken translation of 

BROSSET 1828: 424 for the name of the chanyu Mo-du] und machte sich aus seinem 

Hirnschädel eine Trinkschale; sein Volk aber trat die Wanderung nach Westen an … 

Die sogenannten Grossen Yue-tshi zogen in das später von den Usun benannte Land 

(das Land am See Issyk-kul). Hier trafen sie ein anderes Nomadenvolk, die Sse, und schlugen 

ihren König, der mit seinem Volk zur Flucht nach Süden genöthigt ward … 

Die Grossen Yue-tshi liessen sich darauf im Lande der Sse nieder, erfreuten sich aber 

des Besitzes nur kurze Zeit: der Kun-mo oder König der Usun, eines Volkes, das westlich 

von den Hiung-nu gewohnt hatte, schlug die Grossen Yue-tshi und nöthigte sie, weiter nach 

Westen zu wandern. 

Die Zeit der Vertreibung der Yue-tshi aus dem Lande am Issyk-kul lässt sich genau datieren; 

dem chinesischen Agenten wurde während seiner Internierung bei den Hiung-nu 

(138–129) die Geschichte des Gründers des Reichs der Usun mitgetheilt: derselbe sei beim 

Tode des Shen-yu der Hiung-nu in ein fernes Land gegangen, habe sich in diesem niedergelassen 

und von da an dem Shen-yu den Gehorsam aufgesagt (Sse-ma-tsien im Nouv. 

Journ. Asiat. II, 429). Der einzige Shen-yu aber, der in dieser Zeit gestorben ist, war Laoshang, 

der 160 starb (WYLIE im Journ. of the Anthrop. Inst. III, 421), so dass also die Vertreibung 

der Grossen Yue-tshi in dieses oder das folgende Jahr zu setzen ist. War der Aufent- 

derselben am Issyk-kul ein so kurzer, so begreift es sich, wie er dem ältesten Bericht- 

halt 

erstatter ganz verborgen hat bleiben können … 

Die Grossen Yue-tshi, sagt der jüngere Bericht, wandten sich nun nach Westen, wo sie 

sich Ta-hia (d.i. Baktrien) unterwarfen; auch aus den Worten der älteren Quelle “geschlagen 

von den Hiung-nu hätten sie sich über Gross-Wan (Ferghana) hinaus entfernt, das 

Volk von Ta-hia geschlagen und sich unterworfen und alsbald ihr königliches Lager nördlich 

vom Flusse Wei (d.i. Oxus) aufgeschlagen,” folgt durchaus nicht nothwendig, dass der 

Einbruch über Gross-Wan erfolgt ist (noch weniger ein langer Aufenthalt daselbst, wie er 

angenommen zu werden pflegt). Vielmehr scheinen die chinesischen Berichte darauf zu 

führen, dass die Grossen Yue-tshi schon 159 direct in Sogdiana eingedrungen sind, also gerade 

in der Zeit der inneren Kriege, welche die Macht des Eukratides untergruben. Vielleicht 

ist die Eroberung eine allmähliche gewesen, da ja Baktrien im Jahre 140 noch als 

unabhängig vorkommt. 

Als die Yue-tshi schon in ihrer neuen Heimath sich niedergelassen hatten, schickte der 

Kai ser von China einen Agenten in der Person des Tshang-kien zu ihnen, in der Absicht, sie 

zur Rückkehr in ihre alte Heimath zu bewegen … Tshang-kien fiel den Hiung-nu in die 

Hände, entkam aber 129 nach Gross-Wan und ward von da durch das Land Khang-kiu (am 

mittleren Sir-Darja) zu den Yue-tshi geleitet. Diese aber fühlten sich in dem fruchtbaren, 

räuberischen Einfällen wenig ausgesetzten, von einer friedlichen Bevölkerung bewohnten 

Lande, das sie in Besitz genommen hatten, zu wohl, als dass sie auf die chinesischen Anträge 

eingegangen wären. Vergeblich begab sich Tshang-kien nach Ta-hia; er musste nach 

1-jährigem Aufenthalt (128–127) unverrichteter Sache heimkehren und hatte auf der Rückreise 

noch das Missgeschick, den Hiung-nu ein zweites Mal in die Hände zu fallen; erst 126 

langte er wieder in China an. 

Auf diesen Mann gehen fast ausschliesslich die lehrreichen Schilderungen von Land und 

Leuten zurück, welche die chinesischen Historiker uns liefern. Die Schilderungen sind so 

charakteristisch, dass sie die empfindliche Schwäche der chinesischen Berichterstattung, 

die aus ihrer Unfähigkeit, die Laute fremder Sprachen gehörig wiederzugeben, entspringende 

Willkür in ihrer geographischen Nomenclatur — die damals noch viel schlimmer 

war als in späteren Zeiten —, fast völlig wieder gut machen ... Namensanklänge haben 

hier mehr geschadet als genützt; selbst richtige Gleichungen hat man oft aus falschen 

Gründen gemacht, wie “Ta-hia = Baktrien” von den Dahen (die nie in Baktrien gewohnt 

haben) ... 

Here we have an Orientalist who had to read the Chinese sources in translation. Yet, 

he shows a clear understanding of their contents. The only really important informa- 

VON GUTSCHMID was lacking is that “Ta-hia” was not simply Bactria, but only tion its 

— 13 —

e astern part — the country later called Tocharestan. With this in mind he would have 

grasped that in the year 140 BCE not necessarily all of Bactria was still independent 

u nder Greek kings, but only the country around the capital Bactra in the West. The 

eastern part of Bactria had already fallen into the hands of those nomads who now — 

v ery shortly before Zhang Qian reached Daxia-Tochara — had lost this part of fertile, 

civilized, well populated Bactria, i.e. Tocharestan, to the superior 月氏 . These first 

nomad conquerors cannot have been the Tocharians — for the Tocharians were still 

there: they are described as well settled on the land and as good traders, but weak 

fighters — the first wave of nomad conquerors, of which the Shiji knows nothing beause 

Zhang Qian had missed these early invaders by a very short period of time, had 

c 

alr eady swept across Daxia. In this first déluge the Greek armies and the last Greek 

sovereigns had disappeared from Tocharestan. Terrified by the reappearance of the 

Ruzhi 月氏 , the faster and hardier horseback archers from an unknown world — the 

F ar Eastern Oikumene — the first-wave conquerors had disappeared, too, and had left 

behind a country which was now without a king. 

The victorious 月氏 quickly filled that vacuum. But it was all still very new. The 

uzhi 月氏 had barely erected their provisional seat of government as a tent city on 

R 

t he near side of the Oxus River when the envoy of Han emperor Wu appeared before 

t heir leader — who was the son of that unfortunate king whom the Xiongnu had slain 

more than thirty years previously. The mysterious first nomad conquerors of Tochara 

can hardly have been any other people than the one which the 月氏 had carried before 

them ever since the lands on the upper Ili River: that particular tribe of the Saka conederacy 

which has been variously called Sakarauloi / Sakaraukai, Sarancae / Saraucae, 

f 

[ Saka-] Aigloi / [Saka-] Augaloi, Sagarauloi, Sacaraucae, or Sakaurakai Skythai in the 

W estern, and simply Sai-wang (older Sak-wang) 塞王 in the Eastern historical sources. 

Chinese 塞王 has in the past often been misunderstood to mean “the king (s) of the 

Sai/Sak” — with consequences that turned out to be very misleading. This reading and 

translation was a capital blunder (see below, pp. 42, 43). 

FRANKE, 1904: 54–55, explains: 

Die verschiedenen Varianten für den Namen des Volkes, die sich bei den westlichen Autoren 

finden ... legen den Gedanken nahe, daß ›wang‹ einen Bestandteil des Namens bildete, 

also ›Saka-wang‹, und daß dadurch ein besonderer Stamm der Saka bezeichnet werden 

sollte. 

F.W.K. MÜLLER, 1918: 577 2 , strongly underlines this reasoning: 

塞 , jetzt zwar im Norden ›Sai‹ gesprochen, lautet aber noch in Canton ›sak‹. ›Sak‹ war 

die ältere Aussprache, wie die buddhistische Transkription für Upâsaka lehrt: U-pa-sakka 

優婆塞迦 . Dass ›Sai-wang‹ ein Name sein müsse, hat FRANKE mit Recht hervorgehoben. 

Seine Darlegung wäre noch schlagender gewesen, wenn er den Originaltext hinzugefügt 

hätte: 

昔匈奴破大月氏 In alter 

大月氏西君大夏 

而塞王南君罽賓 

Zeit besiegten die Hiung-nu die großen Yüe-tšï, 

die großen Yüe-tšï machten sich im Westen zu Herren von Tai-Hia, 

und die Sak-wang machten sich im Süden zu Herren von Ki-pin. 

Da in den beiden ersten Sätzen keine Rede von Königen ist, wird auch im dritten Satze 

王 nicht König bedeuten, sondern zum Namen gehören ... 

“Le grand déchiffreur berlinois” (MEILLET on MÜLLER) makes an intelligent stateent 

here. Of the Chinese name Saiwang/Sakwang 塞王 the first part, 塞 , is clearly a 

m 

transcription of Sak(a-), whereas the second part, 王 , meaning “king” and read wang, 

is rather strange in at least two respects. It does not recall the second part of the Western 

name –raukai (*rawaka, “swift”) and it is a very common character in Chinese — 

— 14 —

whereas in transcribing foreign names the Chinese show a marked tendency to use 

rare or even obsolete characters. The 王 in 塞王 might therefore be a scribal error 

which happened early and was not corrected by later scribes because they had no way 

to check in all the many cases of little-known foreign names. I find that De Groot, 1926: 

25, has discussed the problem at greater length: 

Das Zeichen 塞 lautet sowohl ›sik‹ wie ›sak‹, und daß dies lange vorher der Fall war, 

zeigt uns die Behauptung des Jen Ši-ku (HS 61, Bl. 4), daß es nur eine andere Schreibung 

für 釋 ›Sik‹ ist, Buddhas Stammname Sakja, der in der Tat in China immer durch dieses 

Zeichen oder durch 釋伽 ›Sak(Sik)-kia‹ wiedergegeben worden ist. Was haben wir uns nun 

bei dem Zeichen 王 ›ong‹ zu denken? Zunächst befremdet es, daß für die Transkription 

eines ausländischen Volksnamens gerade ein so alltägliches Zeichen, das einfach “König” 

bedeutet, gewählt und dadurch die Tür für Mißverständnisse weit geöffnet wurde; denn 

ein jeder mußte seitdem aus Sak-ong ohne Bedenken “König der Sak” lesen, was der Textschreiber 

gewiß nicht gewollt haben kann. Man ahnt somit, daß hier ein Schreibfehler vorliegt 

und ursprünglich das ähnliche 圭 ›ke‹ gestanden haben kann, das dann später durch 

kluge Gelehrte, die in dem Text das betreffende Volk auch bloß als ›Sak‹ erwähnt fanden, 

für einen Fehler für ›Sak-ong‹, “König der Sak” gehalten und dementsprechend “verbessert” 

wurde. Die Zeichen 圭 sowie 跬 , 閨 und 奎 , worin es als phonetisches Element steht, 

lauten ›ke‹; 佳 aber lautet ›ka‹, und 罣 , 卦 und 挂 werden ›koa‹ ausgesprochen. Der chinesischen 

Transkription zufolge kann also das in Frage stehende Volk ›Sak-ke‹ oder ›Sik-ke‹, 

›Sak-ka‹ oder ›Sik-ka, ›Sak-koa‹ oder ›Sik-koa‹ geheißen haben. 

De Groot makes it certain here that the translation “the kings of the Sai” for Saiwang 

塞王 must be a mistake — as explained in 1904 by Franke and 1918 by F.W.K. 

Müller. But when he goes on to suggest that we should read 塞王 simply as Sak–ka we 

cannot follow him. For in the Hanshu we find both Sai 塞 and Saiwang 塞王 (below, 

pp. 42, 43). These two hanzi 漢字 , or Chinese characters, stand for the general designation 

“Saka” 塞 on the one hand and for the more specific tribal name “Sakaraukai” 

塞王 on the other. Chinese Saiwang 塞王 is the equivalent of Latin Sacaraucae 

and Greek Sakaraàkai (Sakaraukai) — and must be explained accordingly. 

In 1979: 207, BAILEY adduces a brilliant solution to this problem: 

›Khotanese Texts‹ 2.77.6 ... Here in a dyadic phrase, ›bðrøka-‹ is the Turkish ›buiruq‹ “officer, 

commandant” from the verb ›buyur-‹ “to command,” hence supporting a similar 

source “command” for the parallel Saka ›røka-‹ from older *›rauka-‹. This word ›rauka-‹ is 

attested in the Saka title Latin ›Sacaraucae‹, ›Sa(ca)raucae‹, with the Greek *›Sakaraukoi‹ 

[sic] ... corresponding to the Chinese phrase ›sai uang‹ ... 

In 1994: 409, HARMATTA, without naming BAILEY, elaborated: 

The Sakas who invaded Bactria appear in the sources under different names, namely, 

Indian ›³aka-muru‡Ÿa-‹ka kings,” respectively, in so far as ›muru‡Ÿa-‹ can be regarded as the Saka title for “lord, 

Chinese ›Sai-wang‹, Greek *›Sakaraukai‹, Latin *›Saraucae‹. 

Of these both Indian ›³aka-muru‡Ÿa-‹ and Chinese ›Sai-wang‹ mean “Saka king” and “Sa- 

king” and Chinese ›wang‹ as the translation of it. As both the Chinese and the Graeco-Lasame 

peoples as conquerors of Bactria, we have to regard the ›Sa- 

karaukai‹ as identical with the ›³aka-muru‡Ÿa-‹ and the ›Sai-wang‹ respectively. Accord- 

ingly, the element ›-rauk-‹ in the name ›Sakaraukai‹ must have the same meaning as Saka 

tin sources mention the 

›muru‡Ÿa-‹ and Chinese ›wang‹. In fact, the word can be compared to Khotanese Saka 

›røkya-‹ “commander, lord,” going back to *›raukya-‹. Saka ›muru‡Ÿa-‹, too, has an equivalent 

in Khotanese Saka: ›rrund-‹ “possessing power, lo rd, king.” As it is proved by Saka 

›muru‡Ÿa-‹, both Khotanese terms ›rrund-‹ and ›røkya-‹ derive from the root *›mrav-/*›mru‹ 

“to declare, to order” as *›mrav-ant-‹ and *›mrav-aka-‹/*›mrau-ka-‹ respectively. Old Iranian 

*›mr-‹ was reduced to ›r-‹ in Khotanese Saka, while in the language of the Sakas of 

Gandhåra the initial ›mr-‹ was preserved. 

— 15 —

With these scholarly explanations we understand from our context that the meaning 

of the ethnic name Sacaraucae/Sakaraukai is not “Saka Kings,” but rather “King 

Sakas,” i.e. the easternmost ancient nomads from beyond the Jaxartes were, in fact, 

called the “Royal Sakas” — of which Chinese Sai–wang 塞王 is the exact translation: 

the “Kingly Sakas.” We will see (below, pp. 42, 43) that HULSEWÉ/LOEWE, in the year 

1979 translating important phrases on the Saiwang in Hanshu 96, did not grasp this in- 

genuine meaning of Saiwang 塞王 — to the correct interpretation of which 

trinsic, 

BAILEY, in the same year, shows us the way. 

In the next text, however, the Saka-raukai or Sak–wang nomads do not figure at all. 

In 1895: LXX–LXXII, CHAVANNES offered a fresh interpretation — but exclusively on the 

b 

asis of the Shiji where the Saiwang/Sakwang 塞王 are not yet mentioned: 

Ce n’était pas seulement par des colonnes militaires envoyées en pays ennemi que 

l’empereur Ou avait cherché à détruire la puissance des Hiong-nou; il eut recours aussi 

aux moyens diplomatiques et tenta de nouer des alliances avec les peuples qui pouvaient 

être disposés à faire cause commune avec lui. 

Parmi ces nations étrangères, aucune ne devait être plus hostile aux Hiong-nou que les 

Ta Yue-tche; battus une première fois par le chen-yu Mo-tou vers l’année 176 avant notre 

ère, ils avaient été complètement défaits par le chen-yu Lao-chang en l’an 165 avant J.-C.; 

leur roi avait été tué et, de son crâne, suivant la coutume barbare, le chef turk s’était fait 

une coupe à boire ... 

Après ce désastre, les Ta Yue-tche cherchèrent leur salut dans la fuite; ils se retirèrent 

d’abord dans la vallée de l’Ili, mais ils ne tardèrent pas à en être délogés par les Ou-suen 

et, recommençant un nouvel exode, ils se portèrent vers l’ouest; puis ils tournèrent au sud, 

franchirent l’Iaxartes et envahirent la Sogdiane qui appartenait alors au royaume grécobactrien; 

cet état, connu des Chinois sous le nom de Ta-hia, se trouvait déjà affaibli par les 

attaques du roi parthe Mithridate I er (174–136 av. J.-C.): il fut incapable de résister aux envahisseurs; 

les Ta Yue-tche purent refouler la population Ta-hia au sud de l’Oxus et s’établir 

eux-mêmes au nord de ce fleuve; ils ne devaient pas tarder à le traverser pour pénétrer 

en Bactriane ... 

L’empereur Ou ne savait sans doute pas que les Ta Yue-tche avaient dû fuir jusqu’en 

Sogdiane et il les croyait encore établis dans la vallée de l’Ili lorsqu’il projeta de contracter 

une alliance avec eux contre l’ennemi commun. Il chargea de cette mission, prédestinée à 

l’insuccès, un certain Tchang K’ien. 

Tchang K’ien partit en l’an 138 avant J.-C., avec une escorte d’une centaine de personnes; 

il sortit de Chine par la frontière du nord-ouest et fut presque aussitôt arrêté par 

les Hiong-nou qui l’envoyèrent au chen-yu Kiun-tch’en (qui avait succédé en 161 avant J.-C. 

à son père, le chen-yu Lao-chang). Tchang K’ien feignit d’accepter de bonne grâce sa détention; 

il se maria, eut des enfants et resta une dizaine d’années chez les barbares; on 

cessa de le surveiller de près; il en profita pour s’enfuir un beau jour avec ses compagnons. 

Se dirigeant vers l’ouest, il arriva d’abord dans le Ferganah, siège du royaume de Tayuan; 

il y fut bien accueilli et le roi lui donna des guides qui le menèrent dans le pays de 

K’ang-kiu, au nord du Syr-daria; aller de Ta-yuan dans le K’ang-kiu serait aujourd’hui passer 

de Kokand à Tachkend. Les gens de K’ang-kiu conduisirent Tchang K’ien dans le pays 

des Ta Yue-tche; il dut donc franchir de nouveau le Syr-daria pour arriver dans les contrées 

situées entre ce fleuve et l’Amou-daria, à l’ouest du Ferganah. 

Parvenu au terme de son voyage, Tchang K’ien ne tarda pas à reconnaître qu’il n’en tirerait 

aucun avantage diplomatique; les Ta Yue-tche se trouvaient bien dans leur nouvelle 

patrie; ils avaient oublié leur haine mortelle contre les Hiong-nou; ils ne se souciaient 

guère des Chinois, trop éloignés maintenant pour qu’une alliance avec eux fût profitable. 

Tchang K’ien passa un an (probablement l’année 128) chez les Ta Yue-tche et les suivit, 

peut-être dans une campagne qu’ils faisaient contre l’état de Ta-hia, jusqu’aux confins de 

ce royaume; mais il ne put rien obtenir et dut partir après s’être heurté à une fin de nonrecevoir 

absolue. 

— 16 —

Dans son voyage de retour, il fut de nouveau fait prisonnier par les Hiong-nou et resta 

dans leur pays plus d’une année; mais en 126 avant notre ère, le chen-yu Kiun-tch’en mourut; 

son frère cadet, I-tché-sié, et son fils se disputèrent le pouvoir; I-tché-sié finit par l’emporter 

et prit le titre de chen-yu; Tchang K’ien profita de ces troubles pour s’évader; il rentra 

en Chine avec sa femme turke et un seul de ces cent compagnons. Si le but particulier 

que s’était proposé Tchang K’ien n’avait pas été atteint, son expédition eut cependant des 

résultats considérables en ouvrant aux Chinois tout un monde nouveau. 

There are a good number of new points of view here. Some can be fully welcomed, 

others must be questioned as they served to mislead later authors. CHAVANNES leaves 

out the Saiwang completely; and he, too, equates Ta-hia squarely with the kingdom of 

the Græco-Bactrians. Yet, to repeat myself: “Ta-hia” cannot be the Chinese transcription 

of “Bactria” — as CHAVANNES would have readily admitted, I trust. In 1903, the 

g reat Sinologist published an excellent translation of the (New) Tangshu’s chapter 221B. 

One paragraph of it starts with the statement: 

Notice sur le T’ou-ho-lo (Tokharestan). Le T’ou-ho-lo ... c’est l’ancien territoire du (royaume 

de) Ta-hia. 

And in a footnote CHAVANNES adds: 

Lors de la mission de Tchang K’ien en 128 av. J.-C., le royaume de Ta-hia se trouvait au 

sud de l’Oxus. 

This may well be taken as a late correction of what he had said about Ta-hia in 

1895. With this important new information before him, how would he have rewritten 

his above exposé ? In any case, in 1907: 187, CHAVANNES stated in a footnote and without 

any further discussion: 

‹Ta-hia› (= Tokharestan). 

With better insight into what was to be found in the Chinese Standard Histories, 

the French Sinologist quietly corrected himself in time. 

The year in which the Ruzhi 月氏 finally decided to migrate west — a date of great 

historical importance, of course — CHAVANNES gives as 165 BCE. This date will be 

copied by many later authors. However, I have been unable to find a convincing discus- 

of this date. The Shiji and Hanshu are rather vague in their descriptions of the 

sion 

decisive fourth and last clash between the Xiongnu and the 月氏 . Following Shiji 110, 

the first one is the incident when Modu, the crown prince of the Xiongnu, was sent as a 

hostage to the 月氏 and his father, chanyu Touman, suddenly attacked the latter some 

t ime before 209 BCE; the second clash occurred a short time after 209 BCE (in which 

y ear Modu killed his father and became the new chanyu of the Xiongnu) when Modu 

started military campaigns to greatly enlarge his Empire of the Steppe by falling upon 

his neighboring nomad peoples, first the Dong Hu 東胡 in the East and then the 月氏 

in the West; the third is that bloody war in which the Xiongnu subdued the 月氏 , the 

Wusun and the whole of the oasis states in the Tarim Basin. After this great victory 

M odu wrote his historic letter to the Han court announcing boldly that all the peoples 

t hat draw the bow have now become one family. We are told in Shiji 110 that the 

e nvoy bearing the letter arrived at (the Han border town?) Xinwang 薪望 in July 176 

BCE. Far too often, this date has erroneously been given by later authors as the start of 

the great 月氏 exodus. For the dating of the last clash, however — in which the king of 

the 月氏 was killed by the Xiongnu, now under the command of Modu’s son, the Lao- 

chanyu (r. 174–161), and after which the 月氏 finally resolved to escape from 

shang 

Xiongnu domination — we have only indirect clues, as far as I know. 

But CHAVANNES, 1907: 189, has this footnote: 

C’est en 165 av. J.-C. que les Ta Yue-tche, vaincus par les Hiong-nou, commencèrent vers 

l’Occident le grand exode qui devait les amener du Kan-su dans la vallée de l’Ili, et, de là, 

jusque sur les bords de l’Oxus. Par suite d’une inadvertance que je déplore, j’ai indiqué 

— 17 —

l’année 140 av. J.-C., au lieu de l’année 165, dans une note de mes «Documents sur les Toukiue 

occidentaux» (1903, p. 134, n. 1). 

He provides no discussion of this dating, neither 1903 nor 1907. But 165 BCE, as the 

c rucial year in which the 月氏 started to migrate west, should not be too far off as 

S hiji 110 tells us that one year before, in 166 BCE, the Xiongnu chanyu in person led an 

army of some 140,000 horsemen deep into Chinese territory burning the Huizhong palce 

回中宮 and sending out advance parties which came as close to Chang’an as the 

a 

pal 

ace of Sweet Springs 甘泉 at Yong 雍 . Laoshang ante portas. It was the worst irruption 

the Han Chinese suffered at the hands of the vastly superior Xiongnu. The victorious, 

seemingly irresistible Xiongnu armies may well have topped there military 

exploit by also invading the territory of the 月氏 kingdom, in this or the following year. 

(WATSON 1993: 145) 

The Chan-yu remained within the borders of the 

empire [literally: within the border defences] for a 

little over a month and then withdrew. The Han 

(forces) pursued him beyond the border defences 

but returned without having been able to kill 

(any of the enemy). 

The Xiong-nu grew more arrogant day by day, 

crossing the border every year, killing many of 

the inhabitants, and stealing a great number of 

their animals ... 

Shiji 110. 2901 

單于留塞內月餘乃去 

漢逐出塞即還不能有 

所殺 

匈奴日已驕歲入邊殺 

略人民畜產甚多 … 

From these accounts we may guess that chanyu Laoshang had thus prepared the 

way for the Xiongnu to invade the 月氏 next. In the Hexi 河西 (modern Gansu) Corridor, 

where their ordos or royal camp was to be found near modern Zhangye 張掖 — 

originally called Zhaowu 昭武 by the Chinese, as we know from Weishu 102 and later 

Standard Histories —, and with Longxi 隴西 near the Western end of the Great Wall 

as the border town, the 月氏 were the immediate Western neighbors of Han China. In 

the north they had a long common border with the Xiongnu. Under these geographic 

conditions, it must have been rather easy for the latter to invade the wide-open homelands 

of the 月氏 in a similarly grand style in this or the next year. The 月氏 , obviously 

ill-prepared and thus taken by surprise, were beaten once again. This time their 

king was killed. And adding insult to injury, the Xiongnu chanyu had a ceremonial 

drinking cup made out of the skull of the 月氏 king. This was too much to bear for the 

once so proud nomads who had despised the Xiongnu. Deeply shocked by this defeat, 

the 月氏 reached the decision to dodge Xiongnu domination by escape. 

The only route open to them was the vacant country of their old western neighbors, 

the Wusun, between Dunhuang and the Lake Lopnor. When this small nation of nomads 

had been beaten by the 月氏 and when their king had been killed, the Wusun 

had decamped and fled to the Xiongnu. From what we read in the Shiji and Hanshu, 

we may infer that this happened in the time of the Xiongnu chanyu Modu, in any case 

before 176 BCE as the terminus ante quem. More than a decade later, presumably in 

about 165 BCE, the 月氏 commenced their historic trek from the Hexi Corridor in a 

northwesterly direction, past Lake Barkol between the Tianshan and Bogdashan mountain 

ranges, along the northern foothills of the Tianshan into the basin of Dzungaria 

and finally across the Borohoroshan into the upper Ili River valley. It was an exodus 

fateful for the history of Central and South Asia over the next few centuries. It was — 

eine Sternstunde der Menschheit. 

As for the chronology of Zhang Qian’s mission, which is the first topic of this compila 

tion of relevant primary and secondary sources: whereas later discussions are 

often far off the mark, CHAVANNES is coming very close to the truth here. A small cor- 

of his text would have sufficed to make it perfect: he should, in fact, rection have 

— 18 —

written “très tôt en 126” instead of simply “en 126.” For Shiji 110, the chapter on the 

Xiongnu, also informs us that Chanyu Junchen died in the winter following the secon 

d year “yuan-shuo” (one of the reign periods of Han Emperor Wu) 元朔二年其 

後冬 . 

We know that during the first century of the Former Han dynasty, down to the year 

104 BCE, the Chinese civil year began with the three winter months. The above information, 

therefore, translates into the Western calendar as: between 31 October 127 and 

26 January 126 BCE (Julian). This means that the Xiongnu Emperor Junchen died in 

the last two months of 127 or in the first four weeks of 126. Hence, Zhang Qian returned 

to Changan very early in 126. Had his journey lasted one year, he would have started 

in 127; if it had lasted three years, he would have left in 129; since his journey in fact 

lasted thirteen years, it becomes clear that Zhang Qian left China in 139 BCE — probably 

also early in that year. However, in the next text this topic becomes confused 

again, although the author is a close friend of CHAVANNES’. 

In 1897: 12–13 LÉVI writes: 

Seu-ma Ts’ien, qui composait ses Mémoires historiques environ cent ans avant l’ère 

chrétienne, y a inséré, au chapitre CXXIII, une longue relation des voyages de Tchang-k’ien; 

ses informations sur les Yue-tchi et les Ta-hia concordent presque littéralement avec la no- 

identique; les deux historiens ont fidèle- 

tion de l’Histoire des Han et attestent une origine 

ment reproduit le récit de Tchang-k’ien. 

“Les Ta-hia,” dit Seu-ma Ts’ien, “n’avaient pas 

de souverain; chaque cité, chaque ville 

élisait son chef. Les soldats étaient faibles et làches à la bataille, bons seulement à faire 

du commerce. Les Yue-tchi vinrent de l’ouest, les attaquèrent, les défirent et établirent leur 

souveraineté.” 

La soumission des Ta-hia était donc un fait accompli dès le voyage de Tchang-k’ien, 

vers 125 avant J.-C. La biographie de Tchang-k’ien contenue dans l’Histoire des premiers 

Han confirme ces données et les précise davantage ... Le rapport de Tchan-k’ien à l’empereur 

marque encore plus clairement l’enchaînement des faits. Expulsés de leur territoire 

par les Hioung-nou (165 av. J.-C.), les Yue-tchi avaient envahi le pays des Ou-suenn, leurs 

voisins de l’ouest, et tué leur roi Nan-teou-mi; puis, continuant leur marche vers l’ouest, ils 

avaient attaqué le roi des Se (Çakas), et les Se s’étaient enfuis bien loin au sud, abandonnant 

leurs terres aux Yue-tchi. Mais le fils de Nan-teou-mi, Koenn-mouo, resté orphelin dès 

le berceau, avait été nourri miraculeusement par une louve, puis recueilli par le roi des 

Hioung-nou; devenu grand, Koenn-mouo attaqua les Yue-tchi, qui s’enfuirent vers l’Ouest et 

allèrent s’établir sur le territoire des Ta-hia. L’intervention de Koenn-mouo exige au moins 

vingt ans d’intervalle entre la défaite des Ou-suenn et la soumission des Ta-hia; le premier 

événement se passe peu de temps après l’an 165; le second tombe donc vers l’an 140 et précède 

d’assez longtemps l’arrivée de Tchang-k’ien chez les Yue-tchi ... 

LÉVI got the story of the young prince of the Wusun (Usun; Ou-suenn) all confused 

which GUTSCHMID before him had understood so much better. The 月氏 had attacked 

their Western neighbors, the Wusun, in the time of Xiongnu Chanyu Modu (r. 209–174). 

It was he and his son Laoshang chanyu (r. 174–161) who had reared the young prince. 

At the time Laoshang died in late 161 BCE, the Wusun prince was old enough to attack 

the 月氏 and avenge his father. The same prince was an “old man” when Zhang Qian 

went on his second mission to the West, c. 118–115 BCE, to negotiate a treaty with the 

Wusun. With LÉVI the Kunmo (Koenn-mouo) would have been just over forty at that 

time — still very far from being “an old man.” 

In 1900: 533–534 BOYER writes: 

Les données à examiner avant tout sont naturellement celles contenues dans le 123 e 

chapitre du Sse-ki. On ne saurait, en effet, contester aux informations de Seu-ma Ts’ien la 

plus haute valeur, puisque, né vers le milieu du second siècle avant J.-C., il fut contemporain 

de Tchang-k’ien, dont il utilisa du reste la relation, qu’il cite dans ce même chapitre. 

— 19 —

En l’an 165 avant J.-C., les Yue-tchi habitent entre le pays de Toenn-hoang et les monts 

K’i-lien (la chaîne du T’ien-chan) dans le Turkestan chinois. A cette époque, vaincus par les 

Hioung-nou, qui tuent leur roi, ils émigrent pour la pluspart vers l’ouest, s’emparent du territoire 

des Se qui fuient au sud, sont de là chassés encore par le Ou-suenn Koenn-mouo, et, 

marchant toujours à l’ouest, arrivent au pays des Ta-hia (Bactriane), qu’ils soumettent. La 

question est justement de savoir tout d’abord ce que fut cette conquête, et nous allons y venir. 

Vers 125 avant J.-C., Tchang-k’ien visite les Ta Yue-tchi, installés dans leur nouvelle 

patrie. Le lecteur se souvient qu’il était chargé par l’empereur Ou-ti (140–86 avant J.-C.) 

d’amener ce peuple à servir d’appui à la Chine contre les Hioung-nou. Parti vers 135 avant 

J.-C., il avait d’abord été retenu dix ans captif chez ces derniers, lors de son passage sur 

leur territoire, et à son retour, ayant subi une autre année de captivité chez le même peuple, 

il rentra en Chine après treize ans d’absence, vers 122 avant J.-C. 

The thirteen years of absence of Zhang Qian are correctly reproduced by most authors. 

But to anchor this span of time firmly within the reign of Han Emperor Wu 

seems to be beyond the capabilities of quite a few. One wonders why. The Chinese text, 

as we have seen above from BROSSET’s translation, ties the return of Zhang Qian to the 

death of one particular Xiongnu chanyu — whose name Sima Qian neglects to menion 

in this particular place, Shiji 123. But he had given a full account of the Xiongnu 

t 

history, down to his own time, in Shiji 110: From this we know that it was Chanyu Jun- 

He died, as stated above, in the winter which followed year 2 (of Han Emperor 

chen. 

Wu’s reign period) “yuan-shuo” 元朔 . Evidently, the intricacies of the Chinese reign 

periods (a practice still in use in Japan which country I reached a first time in year 46 

“ showa” 昭和 or 1971) and the somewhat complicated Chinese calendar are the main 

stumbling block: a careful year by year concordance book between the Chinese and 

the Western calendars was still wanting. Before the great calendar reform of 104 BCE 

— in which Sima Qian participated — the civil Chinese year started with the three winer 

months (lunar months 10–12); after that date, it started with the three months of 

t 

spring (lunar months 1–3). The old (now unofficial) Chinese calendar still does. 

In 1903: 18–19, V. A. SMITH published a first English summary of these events: 

The early Chinese historians derived their knowledge of the migrations of the Yueh-chi 

chiefly from the reports of Chang-k’ien (Tchang-k’ien), who visited the Yueh-chi territory in 

or about B.C. 125. This officer was despatched in or about B.C. 135 by the emperor Wu-ti 

(Ou-ti, flor. B.C. 140–86) on a mission to the Yueh-chi, in order to obtain their assistance 

against the Hiung-nu, who constantly harried the Chinese frontiers. 

The envoy was intercepted by the Hiung-nu, who detained him for ten years, so that he 

did not arrive at the Yueh-chi chieftain’s camp until about B.C. 125. Returning from his mission, 

Chang-k’ien was unlucky enough to be again intercepted by the Hiung-nu, who detained 

him yet another year. When at length he returned to China in about B.C. 122, he had 

been absent from his native land for thirteen years, and was thus well qualified to bring 

back accurate information about the foreign nations whom he visited. 

The story of the travels of Chang-k’ien was recorded by his contemporary Ssu-ma- 

Ch’ien, the Chinese “Father of History” (born c. B.C. 145), in chapter 123 of his classical work 

the Sse-ki, or “Historical Record” ... 

As far as the chronology of Zhang Qian’s mission was concerned, RÉMUSAT’S authority 

was still unbroken — more than sixty years after the publication of his 佛國記 . 

However, one grave geographical misunderstanding in the Western translations 

goes back to the original text: the historians of the Ban family had been able to follow 

the migrations of the 月氏 as far as the lands of the Saiwang 塞王 or Sakaraukai — 

after which they lost track and then rediscovered their former Western neighbors, firm- 

established in Daxia 大夏 or Tochara. In Hanshu 96, Ban Gu then simply connect- 

ly 

ed the two points with a straight line and let the 月氏 hit upon and subjugate the Da- 

more or less directly after leaving the Ili River region, glossing over the fact that 

there was a wide gap between the two regions in space and between the two events 

xia 

in 

— 20 —

tim 

g 

e. Before him, Sima Qian, in Shiji 123, had known even less. He had the 月氏 mirate 

from the Hexi 河西 Corridor directly to Daxia 大夏 : 

... they then moved far away, beyond (Da) 

Yuan, and in the west attacked and subju- 

gated Daxia. 

乃遠去過宛西擊大夏而 

臣之 

The final encounter between the two nomadic nations, the Xiongnu 匈奴 and the 

uzhi 月氏 , arch rivals in East Asia, occurred sometime around 165 BCE. In the sum- 

R 

mer of the year 129 Zhang Qian found the 月氏 newly established in Central Asia. He 

was their guest, in their make-shift royal camp of tents north of the Oxus River (mo- 

Amu Darya), for more than a year. We must realize here that Sima Qian com- 

pressed a wide stretch of space and time into one short sentence. What really happen- 

dern 

ed during this generation and a half and during this migration over a distance of some 

3.500 km (as the crow flies), the Chinese historians were only able to find out and reconstruct 

very, very slowly. 

On the chronology of Zhang Qian’s mission FRANKE, 1930: 337–338, writes: 

Dafür stellte sich jetzt ein anderes Moment ein, das ... schicksalbestimmend für die 

nächsten Jahrhunderte werden sollte. Die Gesandtschaft, die im Jahre 138 abgesandt war, 

um mit den Yüe-tschi ein Bündnis gegen die Hiung-nu zu schließen, und die man, da 

nichts mehr von ihr verlautete, aus dem Gedächtnis verloren hatte, kehrte im Winter 126 zu 

125 nach Tsch’ang-ngan zurück. Und was hatte sie nicht erlebt und erfahren! 

Zwar ein Bündnis mit den Yüe-tschi hatte sie nicht erreichen können, aber sie hatte 

neue Welten entdeckt, Welten, die bis dahin unfaßbar gewesen waren und die nun das verblüffte 

Staunen der Zeitgenossen daheim erregten. An der Spitze der Gesandtschaft stand 

ein einfacher Mann aus Han-tschung (im südwestlichen Schen-si), der einen kleinen Posten 

im Palaste bekleidete. Er führte den Namen Tschang K’ien, der seitdem zu den berühmtesten 

der ganzen chinesischen Geschichte gehört. 

Dem Auftrage des Kaisers folgend, machte er sich zusammen mit einem Manne der Hu- 

Völ ker Namens Kan-fu und einer Begleitung von etwa hundert Mann auf den Weg nach 

dem Lande der Yüe-tschi. Da Tschang K’ien durch Kan-su zog, muß man annehmen, daß 

er das gesuchte Volk noch in seinen alten Wohnsitzen am Nan-schan wähnte. Bei den 

Hiung-nu aber, durch deren Gebiet die Gesandtschaft in jedem Falle hindurch mußte, wurde 

sie festgehalten und zum Schan-yü (vermutlich im Norden) geführt. Hier erfuhr 

Tschang, daß die Yüe-tschi “im Norden von den Hiung-nu” wohnten ( was tatsächlich nicht 

richtig war), und daß seine Weiterreise nicht gestattet werden könnte. Über zehn Jahre 

blieben die Chinesen in Gefangenschaft bei den Hunnen, dann gelang es ihnen zu entfliehen. 

Tschang K’ien ging, vermutlich am Südhang des T’ien-schan entlang, den Spuren der 

Yüe-tschi nach, und fand sie schließlich, nachdem ihm Leute von Ta-yuan (Ferghana) und 

K’ang-kü (die Kirgisen-Steppen nördlich vom Syr darja oder Jaxartes) das Geleit gegeben, 

in den Ländern am oberen Oxus (Amu darja) ... 

Tschang K’ien blieb ein Jahr im Lande, dann trat er, ohne seinen Zweck erreicht zu haben, 

die Rückreise an, allem Anschein nach auf der Südseite des Tarim-Beckens, um hier 

das Land der Tibeter oder Tanguten (K’iang) zu erreichen. Er geriet jedoch abermals in 

die Gefangenschaft der Hiung-nu, und erst nachdem er über ein Jahr festgehalten war, 

konnte er sich die beim Tode des Schan-yü ausgebrochenen Unruhen zu Nutze machen 

und fliehen. Nach dreizehnjähriger Abwesenheit langte er in Tsch’ang-ngan wieder an. 

Von seinen Begleitern war ihm nur sein treuer Kan-fu noch geblieben, alle anderen hatte 

er verloren. 

Tschang K’iens Reise ist eine Leistung, der in der Geschichte wenig Gleichartiges an die 

Seite zu stellen ist, nicht zum wenigsten was die Wirkung angeht. Er hatte nicht nur die 

Oasen-Staaten am Rande des Tarim-Beckens durchreist, sondern er hatte auch jenseits 

der Wüsten andere große Staaten mit zahlreicher Bevölkerung, mit großen, blühenden 

Städten, mit lebhaftem Handelsverkehr, mit verfeinerter Kultur und einer eigenen Schrift 

— 21 —

und Literatur angetroffen, von anderen großen und mächtigen Reichen, wie Indien, Parthien, 

Babylonien, dem Lande der Alanen u.a. hatte er gehört, kurzum Tschang K’ien hatte 

eine fremde Welt aufgefunden, er war in den Bannkreis der griechisch-indisch-persischen 

Kultur geraten. 

This is an exceptionally clear-sighted and eloquent presentation of Zhang Qian’s 

mission in search of the 月氏 . The imperial envoy is not called a general here, but the 

low-ranking palace employee (›lang‹ 郎 ) he really was at the time. 

A few minor errors in FRANKE’s exposé may be noted here. It should be clear that 

the Han Chinese did not expect the Ruzhi 月氏 to live in their old seats still. They had 

been told, strangely late in fact, that the 月氏 had decided to move far to the West — 

not for better grazing grounds, as stated by a number of later authors, but to be safe 

from further attacks by the Xiongnu. To search for the 月氏 , however, Zhang Qian and 

his men with their large baggage train had to pass the Hexi (Gansu) Corridor: this was 

the traditional and only line of communication from China to all countries further 

West. It was Zhang Qian’s bad luck that the Hexi Corridor, the former lands of the 

Ru zhi 月氏 , had been annexed by the Xiongnu in the decades between c. 165 and 139 

BCE. Wh en the mission passed Longxi 隴西 — the last Chinese town near the end of 

the Great Wall —, they entered enemy land and were quickly spotted and stopped. 

Another point is that Zhang Qian, living more than ten years among the Xiongnu, 

had more than enough time to find out where in fact the Ruzhi 月氏 were living now. 

N ot north of the Xiongnu, as the Chanyu had told him; not in the region between the Ili, 

C hu and Naryn rivers (any more) — where Zhang Qian might have been told to search 

for them; but much further West still, somewhere on the age-old caravan route which 

passed Shule 疏勒 (Kashgar) and the Congling 蔥嶺 mountains (the Pamirs). 

In Shiji 123 we are told that Zhang Qian lived a full decade or more amongst the 

X iongnu 匈奴中 ; this could be misunderstood to mean that he was living in the center 

o f the Xiongnu empire. So Hanshu 61 (see above, sentence (14) of the original text) 

pointedly tells us that he was living in the West of the Xiongnu realm 匈奴西 . With 

t his we may assume that Zhang Qian found excuses to roam about in the region of the 

Tarim Basin which he may have entered via the northern route skirting the southern 

slopes of the Tian shan 天山 mountain ranges. But the 月氏 had not passed this way. 

Their trek must have passed to the north of the Tian shan, i.e. the 月氏 must have 

trekked across the empty lands of what is called Dzungaria today — the western-most 

end of the Gobi desert — before they reached the upper Ili river and there came into 

conflict with the Sakaraukai or Saiwang 塞王 , the easternmost branch of the large Sa- 

ka Federation. 

This people led a nomad way of life just like the Ruzhi 月氏 . But they were Skythai 

and thus belonged to the Indo-European world of Central Asia. When the Far Eastern 

and thus mongoloid 月氏 clashed with these Sakaraukai/Saiwang and drove them 

w est, they opened the door to the Western Oikumene, to themselves as well as to 

Z hang Qian who, otherwise, would not have crossed this decisive — and until then 

nearly insurmountable — dividing line between two separate worlds. 

With this it should be clear that — after his escape — Zhang Qian was not traveling 

with his original one hundred men strong delegation across the Tarim Basin to get 

himself familiar with the area. He would not have gone far at that time, just as ten 

years before. Instead, he must have pretended to do something useful for the Xiongnu. 

And only when he was close enough to where he wanted to go, i.e. when he had 

reached the southwestern-most corner of the Tarim Basin, i.e. the area of Shule/Kashar, 

did he finally drop his disguise to make his escape from the Xiongnu — in just the 

g 

c ompany of his trustworthy Xiongnu servant Gan-fu 甘父 with whom he crossed the 

Congling 蔥嶺 (Pamirs) and via the Terek pass (3870 m) reached Da Yuan 大苑 

( Ferghana) within one month or so. His large mission must in fact have disappeared 

early in those ten years amongst the Xiongnu, not at a later time. 

— 22 —

The last but most important point to be discussed here concerns the chronology of 

Z hang Qian’s mission. FRANKE states that Zhang Qian returned to Chang’an “in the 

winter of 126 to 125.” As reproduced above, Shiji 123 says that Zhang Qian was able to 

e scape a second time from Xiongnu captivity and finally return to Han China after 

troubles had broken out at the Xiongnu court following the death of the chanyu 單于 . 

And from Shiji 110, also briefly quoted above, we know that the chanyu in question was 

Junchen 軍臣 : 

(WATSON 1993: 150) 

This (took place in) the second year of the 

Han (era) ›yuan–shuo‹ (127 B.C.). 

The following winter the Xiong–nu Shan–yu 

Jun–chen died. 

Shiji 110. 2906–2907 

是歲漢之元朔二年也 

其後冬匈奴軍臣單于死 

For FRANKE to translate this date into the Western calendar as “winter 126/125” is 

disappointing because it shows that FRANKE was not familiar with the (admittedly complex) 

Chinese calendar. As a Sinologist he had no excuse for this mistake, because in 

1910 the Chinese Jesuit, Father PIERRE HOANG 黃伯祿 (d. 1909), had published his 

Concordance des Chronologies Néoméniques Chinoise et Européenne (basing him- 

self on an older work in Chinese: the Li dai chang shu ji yao 歷代長術輯要 by 

WANG YUEZHEN 汪曰楨 , prepared 1836–1862 and published 1877). Now, from Shiji 123 

and Shiji 110 it is clear that Chanyu Junchen died in the winter of the third year yuanshuo 

元朔 — or in the winter of 127/126 BCE — which is given in HOANG’s Concordance 

in the following way: 

126 

Av.J.-C. 

Cycle de 

la lune 

12 

13 

14 

15 

16 

17 

18 

19 

20 

21 

22 

23 

* 

Lune 

1 er jour 

10 

11 

12 

j.7 

1 

2 

3 

4 

5 

6 

7 

8 

9 

9 

西漢 

Si Han 

武帝 

Ou Ti 

元朔 

Yuen-chouo 

3 

乙卯 

Y-mao 

Mois 

Solaire 

10 

11 

12 

1 

1 

2 

3 

4 

5 

6 

7 

8 

9 

10 

384 

Jour du 

mois 

28 

26 

26 

1 

24 

23 

24 

23 

23 

21 

21 

19 

18 

17 

Cycle du 

jour 

10 

39 

9 

15 

38 

8 

37 

7 

37 

6 

36 

5 

35 

4 

— 23 —

This chart is a translation and elaboration of what WANG YUEZHEN had written on 

the same year Han Wu Di ›yuan-shuo‹ 3 in an extremely abbreviated way, giving only 

the essentials of columns 2 and 5 of HOANG : 

乙卯 [ 漢武帝元朔 ] 三殷 [ 歷 ] 十癸酉 [10] 十一壬寅 [39] 正辛丑 

[38] 三庚子 [37] 六己巳 [6] 八戊辰 [5] 後九丁卯 [4] 朔 . 

HOANG tells us that the third year of the reign period ›yuan-shuo‹ 元朔 of Han Em- 

peror Wu consisted of 13 lunar months: 12 regular lunar months plus on e intercalary 

lunar month added to the end; seven of these thirteen months were long months of 30 

days and six were short months of 29 day 

s; thus this year contained 384 days. But 

what is really important for our context is the fact that this Chinese civil year started 

with a tenth lunar month. This is the first month of the winter season. The Chinese 

calendar knowing four season of exactly three months each, the winter season consists 

of the three lunar months ten to twelve, the spring season of the three months one to 

three, the summer season of the three months four to six, and the autumn season of 

the three months seven to nine. Intercalary months, to be added every second or third 

year, are regarded as a duplicate of the month preceding it. In our year the intercalary 

month is a second ninth month. 

To translate this year into the Western calendar, it suffices to know that the first 

day of this Chinese year corresponds to October 28, 127 BCE (given in columns 3 and 4 

in HOANG’s chart: 10/28). The first month of spring, HOANG’s chart gives as January 24, 

126 BCE. With this we know that the crucial “winter which followed year 2 ›yuanshu 

o‹,” as given in Shiji 110, lasted from October 28, 127 to January 23, 126 BCE. 

However, for reasons not altogether clear to me, HOANG gives all Western dates BCE 

in the Gregorian calendar. Hence, to express the above dates in the Julian calendar we 

have to add three days in each case. The year ›yuan-shuo‹ 3 thus began October 31, 127 

BCE (Julian) or on JD 167 5340. It is likely, then, that chanyu Junchen died in late 127, 

and it is clear that Zhang Qian returned to Chang’an (very early) in 126 BCE. 

Interpreting the Chinese calendar correctly, FRANKE should have written: ... kehrte 

im Winter 127 zu 126 nach Tsch’ang-ngan zurück; or better still: im Frühjahr 126. To 

give the great scholar due credit here, I like to add that he corrected himself this way 

in a later paper (1934: 269). 

When most, if not all, of Zhang Qian’s mission seemed well established, HALOUN 

1937: 246–252 succeeds in casting grave doubts over the whole topic again: 

Die beiden Abschnitte der Üe-tṣï–Wanderung genau zu datieren, ist recht schwierig. Unsere 

hauptsächlichen chinesischen Que llen , die Tṣa ṅ Tś’ien–Biographie und die Westländer- 

Monographie des Xan-ṣu 

(= Hans hu 61 + 96) bie ten nur ungenaue, z.T. widerspruchsvolle 

Angaben und scheinen verschiedene Kombinationen mit den sonst zur Verfügung stehenden 

Stellen zuzulassen. Der gesa mte Quellen stoff ist zuletzt von japanischen Forschern 

untersucht worden ... 

In seinem an neuen Fragestellungen und eigenständigen Ergebnissen so fruchtbaren 

Aufsatz über die Expedition des Tṣaṅ Tś’ien stellt Kuwabara die These auf, die Wanderung 

der Üe-tṣï von Kansu nach dem Ili s ei zwischen 172 und 

161, ihre Abwanderung von da nach 

dem Âmû-daryâ-Gebiet erst zwisch en 139 und 

129 erfolgt ... Eine mehr als ungefähre Zeitangabe 

für den ersten Wanderzug läßt sich m. E. den Quellen nicht abzwingen ... 

Es muß demnach bei dem etwa s rohen Datum 174–160 verbleiben, ja selbst eine kurze 

Spanne danach scheint u.U. nicht ausgeschlo ssen. In Hinsicht auf den zweiten Wanderungsabschnitt, 

den Zug vom T’ien- ṣa n nach Baktrien, ist die n eue, von Kuwabara und Yasuma 

vorgeschlagene Datierung 139 b is 129, durch die vo n Tṣaṅ Tś’ien in hunnischer Gefangenschaft 

verbrachte Zeit be grenzt , m.E. überzeuge nd, und l äßt sich, wie ich glaube, bei 

gleicher Begründung auf die Jahre zwischen 

135 und 129 einengen. Diese ausschließlich 

durch sorgfältige Interpretation der chinesischen Quellen gewonnenen Zahlen fügen sich 

vollkommen den aus den westlichen Quellen ableitbaren ein. 

— 24 —

Der Vorstoß der Nomaden erfolgte in einer einzigen auf Baktrien geradezu gerichteten 

Welle; die Annahme, daß die Üe-tṣï den Griechen bereits um 160 Sogdiana entrissen hätten, 

ist aufzugeben. 

Tṣaṅ Tś’ien, ausgesandt, um ein Bündnis zwischen den Chinesen und den Üe-tṣï abzuschließen, 

soll sich, nach den Angaben seiner Biographie, 129 bis Anfang 128 bei ihnen auf- 

haben (wenn wir als Rückkehrjahr 126 annehmen; auch 125 könnte mit guten 

Gründen in Erwägung gezogen werden, und der Aufenthalt in Baktrien läge dann entspre- 

gehalten 

chend je ein Jahr später). Die Daten sind nicht völlig gewiß, können aber wohl nur in geringen 

Grenzen schwanken. 

In a footnote HALOUN adds: 

In den Bericht sind z.T. sagenhafte Züge eingedrungen; kaum historisch kann das für 

die Entsendung des Tṣaṅ angegebene Jahr 139 — bzw. 138 — sein, vielmehr gehört die gerade 

dreizehnjährige Irrfahrt in der Fremde mehr als wahrscheinlich einem wohlbekannten 

Ausstattungsstück der Legende an; die allgemeine politische Lage spricht gegen die 

Einleitung der Mission vor frühestens 133; vor einer Überschätzung und kritiklosen Auswertung 

von Ṣï-tśi Kap.123 (Ferghâna) muß grundsätzlich gewarnt werden. Er bezeichnet die 

den Üe-tṣï unterworfene Landschaft Baktrien mit besonderem Namen, Ta-śia 大夏 , der nun 

zu den übrigen zu identifizierenden zuwächst. 

Über einen der vier Völkernamen ist man sich seit langem einig. Die Sai (Sǝ), mittelchinesisch 

und altchinesisch Sǝk, sind fraglos die Saken. Dürfen wir die “Sakarauken”, zumindest 

in der Sache, als einen ihrer Stämme auffassen, so geht es nach Zeit und Wanderungsrichtung 

doch wohl nicht an, sie mit den von den Üe-tṣï um 170–60 aus dem alten 

Sakenland vertriebenen Sai-uaṅ zu verselbigen, wie immer dieser bis jetzt ebenfalls noch 

unklare Name auszudeuten ist ... 

If we did not know better, we might indeed feel greatly disturbed by such seemingly 

convincing statements. When HALOUN decreed die allgemeine politische Lage spricht 

gegen die Einleitung der Mission vor frühestens 133, he sounded like a great authority 

on the subject — but he was utterly wrong as we know today. As it is, we realize here 

how dangerous forceful, but unproven, reasoning may be. Better than just naming 

sources it is to really go down ad fontes and quote verbatim the most important ones 

we have. I am trying to do just that in the present compilation. 

With the next author we return to the profoundly researched arguments of a great 

authority on the history of the Greeks. In 1938: 279–280, TARN writes: 

We must now turn to Chang-k’ien. He was sent in 138 by the Han emperor, Wu-ti, as his 

envoy to the Yueh-chi to solicit their alliance against the common enemy, the Hiung-nu. 

Where the Yueh-chi were at the time, and what route Chang-k’ien took, are not recorded, 

but it is generally supposed that he followed the northern route through Chinese Turkestan 

to Kashgar; thence he would have taken the route by Irkishtam and the Terek pass to Ferghana, 

which probably shows that he expected to find the Yueh-chi still north of the Jaxartes. 

On his way through Chinese Turkestan he was captured by the Hiung-nu and kept in 

more or less honourable captivity for some ten years; finally he escaped with his attendants 

and proceeded on his mission as though nothing had happened, a fact which illustrates 

the man’s force of character. He reached Ta-yuan (Ferghana); the Saca government 

passed him through to the K’ang-kiu and they in turn to the Yueh-chi, then camped between 

Samarcand and the Oxus ... 

Chang-k’ien says that the K’ang-kiu to the south of the river were “under the political influence 

of” (i.e. subject to) the Yueh-chi as those north of the river were to the Hiung-nu. 

They may possibly have extended to the Samarkand country, though if they did it was probably 

later. But the reason that the Ta-yuan entrusted Chang-k’ien to their safe-conduct, 

which would mean for him a considerable détour to the westward, more probably was, not 

merely that they were vassals of the Yueh-chi, but that Samarkand was still maintaining 

itself in some sort of quasi-independence and blocking the direct road ... 

— 25 —

Chang-k’ien failed to obtain the alliance of the Yueh-chi, who told him that they were 

tired of fighting and trekking and only wanted a peaceful life in the rich country which 

they had at last secured, and returned to China by the more difficult southern route from 

Badakshan over the Pamirs and so through Chinese Turkestan; he was again captured by 

the Hiung-nu, but after a year’s captivity he reached China in 126. In 115 he was sent on a 

mission to the Wu-sun, then apparently about Lake Issyk Kul and from there sent out subordinate 

envoys to visit the Western Countries up to and including Parthia, a country he 

himself never saw. He died in 114, a year after his return to China ... 

TARN’s chronology of Zhang Qian is correct for the year of return, but off one year 

for the departure from Chang’an. He does not state the year in which, according to his 

reckoning, the Chinese ambassador reached the Ruzhi 月氏 ordos on the north side of 

t he Oxus, but further down he says: 128. As for the onward route of Zhang Qian, TARN’s 

e xplanations are immaculate except that Zhang Qian was not able to gather his 100 

men mission together again after more than ten years with the Xiongnu. As discussed 

a bove, his escape could only work because he was in the company of no one but his 

trusted Xiongnu servant Gan Fu: just these two men continued the mission, not really 

“ as if nothing had happened.” As for the return journey: it is very unlikely that Zhang 

Qian took the direct, extremely difficult route by crossing the Pamirs from Badakhhan 

to Yarkand. He had no knowledge of this route whatsoever, and it would not have 

s 

h elped him to evade the Xiongnu. More likely it is that Zhang Qian returned the way he 

h ad come, i.e. via Samarkand and Ferghana — here he may have picked up the seeds 

of the grape which we know he imported into China — and along the age-old caravan 

route over the Pamirs back to Kashgar. From there he must have traveled the Souhern 

route around the Taklamakan, close to the mountain ranges there, inhabited by 

t 

the proto-Tibetans, to avoid the Xiongnu. That he was captured all the same shows 

how 

well the Xiongnu were in control of the whole region of the Tarim Basin at the 

time. 

One final point to discuss would be the position of Kangju 康居 . This was at first 

just a small kingdom “some two thousand li (800 km) northwest of Da Yuan” according 

to Zhang Qian’s Report. The Chinese envoy was searching for the Ruzhi 月氏 , not the 

Kangju. When the king of the Da Yuan 大苑 sent him to the Kangju, this then is a first 

hint at the fact that the 月氏 , established in the lands around Sa 

markand, had become 

the new Kangju after they had subjugated this small kingdom and in the process had 

extended the borders of it a considerable distance to the southwest, i.e. across the 

Jax artes into Sogdiana. What sounds like a mere guess is corroborated in some of the 

later Chinese Standard Histories: the Weishu 魏書 , Beishi 北史 , Suishu 隨書 , and 

Tangshu 唐書 (see below, pp. 27–29). 

As will become clearer later on in their history, one very curious fact about the Ruhi 

月氏 is that they were always trying to hide behind the Western, i.e. Central Asian, 

p eoples they conquered. In the Hanshu of Ban Gu, completed more than two hundred 

z 

years after Zhang Qian’s Report, the Kangju 康居 are suddenly a much bigger and 

m uch more powerful country, now extending across the Jaxartes as far as the Oxus 

a nd thus including all of former Greek Sogdiana with its capital in the area of 

S amarkand as described in Tangshu 221B (see below, pp. 28–29). This first Ruzhi 

p owerbase had hardly more than the name in common with the Kangju in the Shiji, 

w here Sima Qian is quoting Zhang Qian. But in the same Shiji 123 we are told — now 

bas ed on later sources — that in the year 101 BCE Kanjgu forces were lurking in the 

background, ready to spring on the Chinese armies of Li Guangli 李廣利 , the “Ershi 

g eneral” 貳師將軍 , who at that time was besieging the capital of the Da Yuan 大苑 in 

order to obtain the coveted “heavenly horses” 天馬 . Those “Kangju” were already the 

R uzhi 月氏 hiding behind this Central Asian name. The Shiji has a vague notion of 

these two different Kangju by stating in chapter 123 : 

— 26 —

In the south it is controlled by the Ruzhi, in 

the east it is controlled by the Xiongnu. 

南羈事月氏東羈事匈奴 

This can hardly apply to the original small Kangju kingdom, but reflects the fact 

that the 月氏 , after chasing the Sakaraukai/Saiwang 塞王 from the region, had foundd 

a strong kingdom in Sogdiana and, for reasons we can only guess, became known 

e 

there, not as (new) Sogdians, but as (new) Kangju: the Ruzhi 月氏 must have brought 

this name with them from across the Jaxartes. 

(WATSON 1993: 249) 

» If the Han (soldiers) do not kill us, 

we will promptly bring out (all) the fine horses 

so that you may take (as many) as you please, 

and will (supply) food for the Han army. 

(But) if you refuse to accept (these terms) we 

will slaughter (all) the fine horses. 

Moreover, rescue (troops) will soon be coming 

(to aid us) from Kang–ju. 

And when they arrive the Han army will have 

to fight (both) us within (the city) and the 

Kang–ju on the outside. 

The Han army (had better) consider the matter 

well (and decide) which course to take! « 

At this time scouts from the Kang–ju were keeping 

a watch on the Han troops, but since the 

Han troops were still in good condition, (the 

Kang–ju forces) did not dare to advance 

(against them). 

Shiji 123. 3177 

漢毋攻我 

我盡出善馬恣所取而給 

漢軍食 

即不聽我盡殺善馬而康 

居之救且至 

至我居內康居居外與漢 

軍戰 

漢軍熟計之何從 

是時康居候視漢兵漢兵 

尚盛不敢進 

Centuries later we will be told in the Chinese Standard Histories, i.e. for the first 

time in the Weishu, that the kingdom of the Kangju 康居 had been ruled by certain Ruzhi 

月氏 kings since the time of the Han. This seems to indicate that the 月氏 — after 

being evicted by the Wusun 烏孫 from the former homelands of the Sakaraukai/Saiwang 

塞王 , the region between the upper Ili and Chu Rivers and around Lake 

Issyk Köl —, had continued their migration westward and in the process had subjugated 

the small nomad state of the genuine Kangju and had extended it across the 

Jaxartes into Sogdiana. 

The kingdom of the Kang 康 is the former (kingdom 

of the) Kangju. 康居 . 

(The people) move about and have no land of their 

own. 

Ever since the advent of the Han (dynasty), generation 

followed upon generation without break. 

Their king’s original family name is ›Wen‹. 

He is a Ruzhi 月氏 . 

Anciently, (the Ruzhi) resided north of the Qilian 

mountain range in the town (i.e. ›ordos‹ = royal 

camp) of Zhaowu 昭武 (modern Zhangye). 

When they were beaten by the Xiongnu, they 

crossed the Congling mountains to the West and 

soon took possession of this land (of the Kang). 

They divided it amongst a number of individual 

members of the king’s (family) and that is why left 

and right (east and west) of the (main) state of 

Kang there are several (lesser) states and all 

Weishu 102. 2281 

康國者康居之後也 

遷徙無常不恒故地 

自漢以來相承不絕 

其王本姓溫 

月氏人也 

舊居祁連山北昭武城 

因被匈奴所破西踰蔥 

嶺 

遂有其國 

枝庶各分王故康國左 

— 27 —

(these rulers’) family name is Zhaowu to demon- 

strate (that they all do) not forget their roots (land 

of origin) ... 

右諸國並以昭武為 

示不忘本也 … 

姓 

The Weishu contains Chinese history from the late fourth to the mid-sixth century. 

The new facts on the migration of the 月氏 are therefore reported some four to five 

centuries after the last piece of information contained in Ban Gu’s Hanshu. As we can 

see here, the Chinese historians needed much time to find out what really had been 

happening in the Far West — reflecting the fact that in the intervening centuries com- 

off and reestablished a couple of times. 

munications with the West were cut 

At long last we read that the 月氏 did not at all migrate directly from the region be- 

reaches of the Oxus and into Daxia — 

tween the Ili and Chu all the way to the upper 

as the Hanshu wanted us to believe. The 月氏 were driven from the lands between the 

rivers Ili and Chu by the Wusun 烏孫 shortly 

before Laoshang chanyu of the Xiongnu 

匈奴 had died in late 161 BCE and they arrived at the Oxus River only months before 

Zhang Qian reached the region in the summer of 129. The Hanshu had kept us won- 

long span of one full generation. With 

dering what had happened to the 月氏 in this the additional information contained in the Weishu, we are finally able to fill out this 

blank in their history. 

The Ruzhi 月氏 subjugated the Kangju 康居 , established themselves as their kings 

and extended (or moved) this Kangju kingdom southwestward as far as Samarkand 

(Sogdiana). There, i.e. in Sogdiana, the 月氏 must have collided a second time with the 

Sakaraukai/Saiwang 塞王 , who were now dri 

ven south — as to the west, i.e. west of 

the middle reaches of the Oxus, the mighty empire of the Parthians 安息 was already 

blocking the way. 

And here, in former Greek Sogdiana, the 月氏 — for the first time as far as we can 

tell — established lesser principalities or viceroy-ships around their central or main 

kingdom. All these viceroys were members of the king’s family. After the Weishu 魏書 , 

or “Book of the Wei (Dynasty),” had reported this important new information for the 

first time, the next in the long line of Chinese Standard Histories, the Beishi 北史 (in 

chapter 97) and the Suishu 隨書 (in chapter 83), simply repeated the new facts. We 

have to wait another two hundred years to be told how many sub-kings or viceroys 

there had been in the realm of the new Kangju — and what their names were. The 

(New) Tangshu 唐書 finally has this additional information. 

(CHAVANNES 1903: 132–134) 

(Le pays de) K’ang 康 est appelé aussi Sa–mo–kien 薩 

末鞬 , ou encore Sa–mo–kien 颯秣建 (Samarkand); 

c’est le pays qu’on appelai t Si–wan–kin 悉萬斤 sous 

les Yuen Wei 元魏 . 

Du côté du sud il est à cent cinquante ›li‹ de Che 史 

(Kesch); 

du côté du nord-ouest, il est à plus de cent ›li‹ du Ts’ao 

曹 occidental (Ischtîkhan); 

au sud-est, il est à cent ›li‹ de Mi 米 (Mâïmargh); 

au nord, à cinquante li du Ts’ao central (Kaboûdhan); 

il est au sud de la rivière Na–mi 那密 (Zara fchan). 

Il a trente grandes villes et trois cents petites places. 

Le nom de famille du prince est ›Wen‹ 溫 . 

C’étaient à l’origine des Yue–tche 月氏 qui résidaient 

autrefois dans la ville de Tchao–ou 昭武 , au nord des 

(monts) K’i–lien 祁連 . Ayant été battus par les Tou– 

kiue (ici: les Hiong–nou), ils se retirèrent graduelle- 

(Xin) Tangshu 221B. 6243 

康者曰薩末鞬曰 

颯秣建 

元魏所謂悉萬斤者 

其南距史百五十里 

西北距西曹百餘里 

東南屬米百里 

北中曹五十里 

在那密水南城 

大城三十小堡三百 

君姓溫月氏人 

始居祁連北昭武城 

— 28 —

ment vers le sud en s’appuyant sur les (monts) Ts’ong 

ling 蔥嶺 

et entrèrent ainsi en possession de ce territoire. 

Les principautés qui sont détachées comme des rameaux 

s’appellent 

— Ngan 安 (Boukhârâ), 

— Ts’ao 曹 (Kaboûdhan), 

— Che 石 (Taschkend), 

— Mi 米 (Mâïmargh), 

— Ho 何 (Koschânyah), 

— Ho–siun 火尋 (Khârizm), 

— Meou–ti 戊地 (le Fa–ti 伐地 de Hiuen–tsang, à l’ouest 

de Boukhârâ), 

— Che 史 (Kesch). 

On les nomme communément les neuf familles (les 

huit plus Samarkand, la métropole des autres). 

Tous sont de la famille ›Tchao–ou‹. 

– 

為突厥所破稍南依 

蔥嶺 

即有其地 

枝庶分王 

曰安曰曹 

曰石曰米 

曰何曰火尋 

曰戊地曰史 

世謂九姓 

皆氏昭武 

With this late information we may safely assume: When the Da Yuan 大苑 sent 

Zhang Qian on to the “Kangju” they did not usher him north to the former small 

Kangju kingdom 康居國 — under Xiongnu 匈奴 domination already —, but due west 

to Samarkand were the Da Yuan knew the ›ordos‹ of the Ruzhi 月氏 to be. That it just 

had been moved south of the Hissar Mountains to the Oxus and thereby into the lands 

of the Daxia 大夏 was apparently not yet known in Ferghana. In the summer of 129 

BCE, these were very new developments. 

2. ARE WE ENTITLED TO EQUATE ›DAXIA‹ WITH TOCHARA ? 

At this junction of Western studies of the relevant Chinese sources, two improved 

translations of Shiji 123 were published in New York and Berlin. Below, I reproduce 

the beginning of these new translations, i.e. as far as the mission of Zhang Qian is narrated 

in this important chapter of the Shiji. For easy comparison I quote the translations 

once more together with the Chinese original. It will be clear from the first sen- 

tences that these are vastly improved renditions. 

(H IRTH 1917: 93–94) 

(DE G ROOT 1926: 9–10) 

Our first knowledge of Tayüan 

Die Spur von Ta Wan ist durch 

(Ferghana) dates from 

Chang K’ien. 

Tšang K’i¥n entdeckt worden. 

Tšang K’i¥n war ein Mann aus 

Chang K’ien was a native of 

Han-chung (Shen-si prov.). 

Han-tšung. 

In der Periode Ki¥n-juan (140– 

During the period of K’ién- 

134) war er Palastbeamter. 

yüan (140–134 B.C.) he was a In diesem Zeitraum verhörte der 

›lang‹. Sohn des Himmels Leute, die 

At that time the Son of Heaten. 

sich Hung-no unterworfen hat- 

ven made inquiries among 

Sie teilten ihm alle mit, 

those Hiung-nu who had surrendered 

daß Hung-no den König von 

(as prisoners) and Goat-si geschlagen und 

aus 

they all reported that the dessen Schädel ein Trinkgefäß 

Hiung-nu had overcome the gemacht habe; 

king of the Yüé-chï and made die Goat-si seien dann geflohen 

a drinking-vessel out of his und hegten eine dauernde 

Shiji 123. 3157–3159 

大宛之跡見 

自張騫 

張騫漢中人 

建元中為郎 

是時天子問 

匈奴降者皆 

言匈奴破月 

氏王以其頭 

為飲器月氏 

遁逃而常怨 

— 29 —

skull. 

The Yüé-chï had decamped 

and were hiding somewhere, 

all the time scheming how to 

take revenge on the Hiungjoin 

nu, but had no ally to 

them in striking a blow. 

The Chinese, wishing to de- 

clare war on and wipe out 

the Tartars, upon hearing this 

report, desired to communi- 

of the 

cate with the Yüé-chï. 

But, the road having to pass 

through the territory 

Hiung-nu, the Emperor sought 

out men whom he could send. 

Chang K’ién, being a ›lang‹, 

responded to the call and en- 

Yüé– 

listed in a mission to the 

chï; he took with him one Kan 

Fu, a Tartar, formerly a slave 

of the T’ang–i family, and set 

out from Lung–si (Kan–su). 

Crossing the territory of the 

Hiung–nu, the Hiung–nu 

made him a prisoner and 

sent him to the ›Shan–yü‹ 

who detained him, saying: 

»The Yüé–chï are to the north 

of us; how can China send 

ambassadors to them? 

If I wished to send ambassadors 

to Yüé (Kiang–si and 

Ch’ö–kiang), would China be 

willing to submit to us?« 

He held Chang K’ién for more 

than ten years and gave him 

a wife, by whom he had a 

son. 

All this time Chang K’ien had 

kept possession of the Emperor’s 

token of authority. 

And, when in the course of 

time he was allowed greater 

liberty, he, watching his opportunity, 

succeeded in making 

his escape with his men 

in the direction of the Yüé-chï. 

Having marched several tens 

of days to the west, he arrived 

in Ta-yüan. 

The people of this country, 

having heard of the wealth 

and fertility of China, had 

Rachgier gegen Hung-no, hät- 

sie 

ten aber niemand, mit dem 

sich verbinden könnten, 

um es 

anzugreifen. 

Han war damals gerade im Be- 

der Hu zu beschäftigen, und 

griff, sich mit der Vernichtung 

faßte auf Grund dieser Mittei- 

ei- 

lungen den Entschluß, zur Anknüpfung 

von Beziehungen 

ne Gesandtschaft (nach Goatsi) 

zu entsenden. 

Der Weg dorthin mußte quer 

durch Hung-no führen. Es wurden 

nun für eine Aussendung 

dahin geeignete Personen aufund 

K’i¥n, der als Palastbeam- 

gerufen, 

ter sich meldete, wurde nach 

Goat-si geschickt. Zusammen 

mit einem Manne aus T’ang-ji’, 

einem ehemaligen Hu’schen 

Sklaven Kam-hu (Kan-fu)verließ 

er Lung-si 

und zog durch Hung-no. Aber 

die Hung-no faßten sie und 

führten sie zum Tan-hu. Dieser 

behielt sie bei sich und sprach: 

»Zu Goat-si liegen wir im Nor- 

den; dürfte Han also dorthin 

reisen? 

Falls ich eine Gesandtschaft 

nach Ju¥’ schicken wollte (zu 

dem Han im Norden liegt), würde 

mir dann Han das erlauben?« 

Mehr als zehn Jahre hielt er 

K’ien bei sich; er gab ihm eine 

Frau, und er hatte Söhne von 

ihr. 

Das Diplom von Han, welches 

er führte, ließ er nicht verlorenvon 

gehen. 

Während er so im Zentrum 

Hung-no verweilte, erwarb er 

sich mehr und mehr Freiheit 

und benutzte diese, um mit seinen 

Angehörigen die Flucht zu 

ergreifen. Der Richtung nach 

Goat-si folgend, floh er westwärts 

und erreichte nach mehrmals 

zehn Tagen Ta-wan. 

In Ta-wan hatte man schon von 

den Reichtümern und Schätzen 

仇匈奴無與 

共擊之 

漢方欲事滅 

胡聞此言因 

欲通使 

道必更匈奴 

中乃募能使 

者 

騫以郎應募 

使月氏與堂 

邑氏 ( 故 ) 

胡奴甘父俱 

出隴西 

經匈奴匈奴 

得之傳詣單 

于 

單于留之曰 

月氏在吾北 

漢何以得往 

使吾欲使越 

漢肯聽我乎 

留騫十餘歲 

與妻有子 

然騫持漢節 

不失 

居匈奴中 

益 

寬騫因與其 

屬亡鄉月氏 

西走數十日 

至大宛 

大宛聞漢之 

饒財欲通不 

得見騫喜問 

— 30 —

tried in vain to communicate 

with it. 

When, therefore, they saw 

Chang K’ién, they asked joyfully: 

»Where do you wish to 

go?« 

Chang K’ién replied: 

»I was sent by (the Emperor 

of) China to the Yüé-chï, and 

was made prisoner by the 

Hiung-nu. 

I have now escaped them and 

would ask that your king 

have some one conduct me to 

the country of the Yüé-chï. 

And if I should succeed in 

reaching that country, on my 

return to China, my king will 

reward yours with untold 

treasures.« 

The Ta-yüan believed his account 

and gave him safe-conduct 

on postal roads to K’angkü 

(Soghdiana); 

and K’ang-kü sent him on to 

the Ta-yüé-chï. 

The king of the Ta-yüé-chï 

having been killed by the Hu, 

the people set up the crown 

prince in his stead. 

They had since conquered 

Ta-hia (Bactria) and occupied 

that country. 

The latter being rich and fertile 

and little troubled with 

robbers, they had determined 

to enjoy a peaceful life; 

moreover, since they considered 

themselves too far 

away from China, they had 

no longer the intention to 

take revenge on the Hu 

(Hiung-nu). 

Chang K’ien went through the 

country of the Yüé-chï to Tahia 

(Bactria), yet, after all, he 

did not carry his point with 

the Yüé-chï. 

After having remained there 

fully a year, he returned, 

skirting the Nan-shan. 

He wished to return through 

the country of the K’iang 

(Tangutans), but was again 

von Han gehört und Beziehungen 

anknüpfen wollen, jedoch 

war es nicht dazu gekommen. 

Als man nun K’i¥n erblickte, 

freute man sich und fragte ihn: 

»Wo willst du hin?« 

Die Antwort lautete: 

»Ich wurde von Han nach Goatsi 

entsandt, jedoch durch Hungno 

wurde mir der Weg verlegt. 

Jetzt bin ich auf der Flucht, o 

König, laß mich von Deinen 

Leuten begleiten, auf daß ich 

wirklich mein Ziel erreichen 

kann. 

Sobald ich dann wieder in Han 

zurück bin, werden die Schätze, 

welche es Dir, König, schenkt, 

alle Beschreibung übertreffen.« 

Ta-wan war einverstanden; es 

ließ (Tšang) K’ien weiterreisen 

und bot Leute auf, die ihm von 

der einen Station nach der anderen 

bis nach K’ang-ki das Ge- 

leit gaben. 

Und dann ließ ihn K’ang-ki 

nach Groß-Goat-si bringen. 

Als der König von Groß-Goat-si 

durch die Hu umgebracht war, 

hatte es seinen ältesten Sohn 

zum König erhoben; 

dann hatte es Ta-ha (Tochara) 

unterworfen und sich dort ansässig 

gemacht. 

Dieses Land war fruchtbar und 

reich; man wurde da selten von 

Feinden angegriffen und suchte 

nur ein ruhiges und freudiges 

Dasein zu führen; 

man betrachtete Han als weit 

entlegen und hegte gegen die 

Hu fast keine Rachsucht mehr. 

Von Goat-si reiste (Tšang) K’ien 

nach Ta-ha, ohne daß es ihm 

gelungen war, eine Entscheidung 

von Goat-si zu bekommen. 

Nachdem er länger als ein Jahr 

sich da aufgehalten hatte, trat 

er den Rückweg an. Er reiste 

den Nan-šan (Südgebirge) entlang, 

mit der Absicht, über 

K’iong (D{a-k’iong} heimzureisen; 

aber zum zweiten Male 

曰 

若欲何之 

騫曰 

為漢使月氏 

而為匈奴所 

閉道 

今亡唯王使 

人導送我 

誠得至反漢 

漢之賂遺王 

財物不可勝 

言 

大宛以為然 

遣騫為發導 

繹抵康居 

傳致大 

康居 

月氏 

大月氏王已 

為胡所殺立 

其太子為王 

既臣大夏而 

居 

地肥饒少寇 

志安樂又自 

以遠漢殊無 

報胡之心 

騫從月氏至 

大夏竟不能 

得月氏要領 

留歲餘還並 

南山欲從羌 

中歸復為匈 

奴所得 

— 31 —

made a prisoner by the 

Hiung-nu, 

who detained him for more 

than a year, when the Shanyü 

died and the “left” Luk-li 

prince attacked the rightful 

heir and usurped the throne, 

thus throwing the country 

into a state of confusion. 

At this time Chang K’ien, with 

his Tartar wife and T’ang-i 

Fu (Kan Fu), escaped and returned 

to China. 

(The Emperor of) China appointed 

Chang K’ién a ›T’ai- 

chung-ta-fu‹ and gave T’ang-i 

Fu the title ›Föng-shï-kün‹. 

wurde er von Hung-no entdeckt 

und länger als ein Jahr festgehalten. 

Da starb der Tan-hu; der linke 

Kok-li-König erschlug dessen ältesten 

Sohn und setzte sich 

selbst auf den Thron; 

Wirren herrschten im Reich; 

(Tšang) K’ien wandte sich mit 

seiner Hu-Frau und mit T’ang- 

ji’ (Kam-)hu zur Flucht und 

kam nach Han zurück. 

Han ernannte darauf (Tšang) 

K’i¥n zum ›Wesir des innersten 

Palastes‹, und T’ang-ji’ (Kam-) 

hu wurde ›Fürst, der vom Kaiser 

eine Sendung empfing‹. 

留歲餘單于 

死左谷蠡王 

攻其太子自 

立國內亂騫 

與胡妻及堂 

邑父俱亡歸 

漢 

漢拜騫為太 

中大夫堂邑 

父為奉使君 

By now, there were translations of Shiji 123 into four European languages: into 

French by B ROSSET (1828); into Russian by BICHURIN (1851); into E nglish by HIRTH (1917); 

and into German by DE GROOT (1926). Within the span of about one hundred years, Si- 

nology in Europe in general and the study of the Chinese Standard Histories in par- 

ticular had made great strides forward. Yet the problems of the Chinese language, the 

Chinese script and the cumbersome Chinese ways to transribe foreign names proved 

formidable obstacles. It seemed that a century of intensive research was not enough to 

overcome the difficulties invo lved. One example is that HIRTH, who came from Ger- 

many and taught at Columbia University in New York, still equated Daxia 大夏 with 

Bactria, thus copying the mist ake of many of his forerunners. The mistaken and miscan 

be traced back to RÉMUSAT 1825: 116: 

leading identification, in fact, 

Les contrées que Tchhang-kian avoit visitées par lui-même étoient 

— le Ta-wan ou pays de Schasch, 

— le pays des Ta-youeï-chi ou la Transoxane, 

— le Ta-hia ou la Bactriane, 

— et Kang-kiu ou la Sogdiane. 

Mais il rapporta des relations détaillées au sujet de cinq ou six autres états voisins. 

1846: 231, JULIEN, the successor of RÉMUSAT at the Collège de France, confirms: 

Je donnerai des notices historiques sur divers peuples de l’Asie qui ont joué un rôle imels 

les auteurs chinois 

portant dans cette partie du monde, et pour la connaissance desqu 

nous offrent seuls des renseignements solides et étendus. Je me co ntenterai de citer, pour 

le moment, 

— les Ta-hia ou Bactriens, 

— les ‘Asi ou Parthes, 

— les habitants du Khang-khiu ou Sogdiens, 

— les Yen-tsai ou Gètes, 

— les Youeï-tchi, de race indo-scythe, qui ont occupé successivem ent la Transoxiane, la 

Bactriane et le Caboul; 

— les Ou-sun, race blonde aux yeux bleus, appelée par quelques aut eurs, indo-germanique, 

etc. ... 

It proved to be very unfortunate, that after RÉMUSAT and 

JULIEN, the erroneous 

identification Ta-hia = Bactria continued to be copied and recopied by five to six gen- 

erations of scholars down to our own times (Vaissière 2002: 31). 

One year after HIRTH’s translation, in 1918: 572, F.W.K. MÜLLER quoted one short 

paragraph of Shiji 123 in translation and the original text which contained the line: 

— 32 —

西擊大夏 Im Westen schlugen sie die Tai-Hia (*Dai-Ha = Baktrer). 

He thus gave the false equation a semi-scientific touch. 

In contrast to MÜLLER, the famous Dutch Sinologist DE GROO 

T, who had come from 

Leiden to Berlin, introduces here the new and much improved equation Daxia = To- 

chara in his translation of Shiji 123. With this, he confirms what MARQUART 1901: 204 

had written: 

Aus dieser historischen Gegenüberstellung der griechisch-römis 

chen und chinesischen 

Berichte ergibt sich mit logischer Notwendigkeit die Gleichung: Ta -hia = Tochari. Ich 

treffe also in dieser Identifikatio n zufällig mit Kingsmill (1882: 74–79) 

zusammen; man wird 

aber hoffentlich zugeben, dass meine auf historisch-kritischem Wege gewonnenen Erklä- 

lediglich auf scheinbare Namensanklänge hin gemachten 

rungen nichts mit den wilden, 

Identifikationen j enes Sinologen gemein haben ... 

Shortly before DE GROOT, in 1920: 1617, the well-known Geographer HERRMANN, writ- 

ing on the Sacaraucae, included the crucial equation in the following explanations: 

Schon Chang-k’ien nennt uns 

im eigentlichen Baktrien das Volk 

der Ta-ha; es sind dies 

die Tocharer, die dort kürzlich die griechische Herrschaft abgelöst hatten. Und wie wir aus 

den Han-Annalen schließen können, folgten den Sacaraucae nach einer Reihe von Jahren 

die Kurschi (Guát-si), als sie von den Å-sun aus ihren neuen Sitzen verjagt wurden. Wenn 

wir dabei die griechischen Nachrichten über die letzten Könige von Baktrien in Betracht 

ziehen, so gehen wir wohl kaum fehl in der Annahme, daß die Sacaraucae zwischen 160 

und 150, die Tocharer um 135, endlich die Kurschi um 130 v.Chr. eingewandert sind. Mit diesen 

Kenntnissen ausgerüstet, sind wir endlich in der Lage, die abendländischen Berichte 

kritisch zu verwerten. Trogus Pompeius hat den Einbruch der Sacaraucae im 41. Buch behandelt: 

»Deinde quo regnante Scythicae gentes Saraucae et Asiani Bactra occupavere et Sogdianos.« 

Leider hat der jämmerliche Auszug bei Iustin. über dieses Ereignis keine Silbe aufbewahrt. 

Daß, wie Marquart (1901: 205) behauptet, die Sacaraucae mit Bactra, die Asiani mit 

Sogdiani inhaltlich zusammengehören, darf man aus dem Text nicht ohne weiteres herden 

Sacaraucae allein die Asiani genannt 

auslesen. Wesentlich ist hier nur, daß neben 

werden. Über beide Völker handelt auch das nächste Buch: 

»Additae his res Scythicae. Reges Thocarorum Asiani interitusque Saraucarum.« 

Hier lernen wir also noch ein drittes Skythenvolk kennen, die Thocari, die bei Iustin. 

XLII 2, 2 als ›Tochari‹ vorkommen; v. Gutschmid (1888: 70) und Marquart haben wohl Recht, 

wenn sie die letztere Notiz in dem Sinne auslegen, daß die Könige der Tocharer Asiani- 

schen Stammes seien. Denn hierzu stimmt genau die Angabe der Han-Annalen, daß die 

Guát-si, die Chang-k’ien noch im Norden des Oxus antraf, sich bald darauf des Landes der 

Ta-ha (Tocharer) bemächtigten. Somit können die Asiani nur die Kurschi (Guát-si) sein. 

N’s Å-sun are the Wusun and the Kurschi (Guát-si) our Ruzhi 月氏 . His 

HERRMAN 

last statement is of importance because it is the result of straightforward logical reasoning. 

It is also important to notice here that HERRMANN identifies the Ta-ha 大夏 of 

Z hang Qian and the Tochari of Trogus and says that they migrated into Bactria from 

somewhere else, sometime between the Sakaraukai and the Ruzhi 月氏 . Below, we will 

see that HERRMANN soon changed his mind on this point revealing a curious 

dilemma — which will be of great help for solving the whole question of who really 

t ook Bactria from the Bactrian Greeks. Sometimes the solution of complex problems 

can be found by asking the right questions. 

HERRMANN continues: 

Während Trogus nur von dem Einbruch zweier Skythenstämme spricht und erst bei 

späterer Gelegenheit mit ihnen zusammen die Thocari erwähnt, zählt Strabon insgesamt 

vier Völker auf, welche, vom Sakenland jenseits des Iaxartes ausgehend, den Hellenen 

Baktrien entrissen haben sollen ... 

— 33 —

In welcher Weise läßt sich nun diese Darstellung mit der des Trogus vereinigen ? ... 

Warum erwähnt Trogus die Tocharer nicht von vornherein bei Gelegenheit des Einbruchs 

in Baktrien ? 

Two years later, in 1922: 459 and more in passing, RAPSON provides us with a new 

and crucial observation on the Daxia: 

The report of Chang-kien, a Chinese envoy who visited the Yueh-chi in 126 B.C., is still 

extant. These nomads were then settled in Sogdiana, and the report speaks in somewhat 

contemptuous terms of their southern neighbours, the Ta-hia, by whom are apparently 

meant the native population of Bactria: they were a nation of shopkeepers, living in towns 

each governed by its magistrate, and caring nothing for the delight or the glory of battle ... 

This simple observation is also taken up by HERRMANN who adduces two more ar- 

and plenty of evidence for reversing his earlier statements on the Daxia. In 

guments 

the same year, 1922: 209–211 and now contra MARQUART, he writes: 

In diesen Zusammenhang gehört auch die Beurteilung des Völkernamens Ta-hsia. Wir 

haben gesehen, daß ein Volksstamm dieses Namens in der Geschichte Chinas nur einmal, 

nämlich im Jahre 1084 v.Chr. unter den westlichen Grenzvölkern aufgetreten ist, um dann 

für immer zu verschwinden. Nur durch die chinesischen Karten und Legenden hat sich der 

Name, wie wir an einigen Beispielen darlegen konnten, bin in die Han-Zeit fortgepflanzt ... 

Es ist daher gar nicht verwunderlich, daß schließlich CHANG CH’IEN genau denselben Namen 

auf das größte Kulturvolk des Westens, die Baktrer, übertragen hat, in der offenbaren 

Meinung, hier das uralte Westvolk endlich wiedergefunden zu haben. Wenn also, wie es in 

der Tat scheint, lediglich eine Namensübertragung vorliegt, dann haben wir keine Veranlassung 

mehr, zwischen den beiden Ta-hsia von den Jahren 1084 und 127 v.Chr. einen ethnographischen 

Zusammenhang zu konstruieren, wie es O. FRANKE in einer besonderen Abhandlung 

getan hat (OZ 1919–20, S. 125 ff.), so daß wir gegen seine Ergebnisse schon oben 

schwere Bedenken äußern mußten. Ebenso ist es ein Verstoß gegen die Methoden der historischen 

Geographie, daß man den Namen Ta-hsia in das vielerörterte Problem der Tocharer 

und Yüeh-chih hineingezogen hat. Während auf der einen Seite Tocharer und Yüehchih 

miteinander identifiziert werden, machen andere Gelehrte, namentlich J. MARQUART 

(1901), O. FRANKE (1920) und STEN KONOW (1920) die Ta-hsia zu Vorfahren der Tocharer, da diese 

schon vor den Yüeh-chih in Baktrien eingewandert und dann von den letzteren unterworfen 

sein sollen. Ausschlaggebend ist für sie der Namensanklang des rekonstruierten 

Lautes Ta-ha an Tocharoi. 

Aber war denn, wie SCHLEGEL, Marquarts Gewährsmann, behauptet hat, die alte Aussprache 

wirklich Ta-ha ? Neuerdings haben F.W.K. MÜLLER und unabhängig von ihm B. 

KARLGREN festgestellt, daß der alte Laut eher Tai-ha gewesen sein müsse (nach einer persönlichen 

Mitteilung KARLGRENs ist für 大 der alte Laut ›d’âi‹ das Normale, während ›d’â‹ 

nur bisweilen in der Poesie vorkommt). Wenn auch eine sichere Entscheidung in dieser 

phonetischen Frage vorläufig nicht möglich ist, die etymologische Verbindung mit Tocharoi 

ist mindestens sehr anfechtbar. Sie wird geradezu illusorisch, wenn wir drei Momente in 

Betracht ziehen, über die man bisher achtlos hinweggegangen ist. 

Das erste Moment besteht darin, daß die Chinesen, obgleich ihnen die Namen Ta-hsia 

und Tu-ho-lo (für Tocharoi und Tokhåra) durchaus geläufig waren, selber niemals auf den 

Gedanken gekommen sind, sie miteinander zu identifizieren. Dieser Fall wiegt um so 

schwerer, weil sie bei ihren nur selten unterbrochenen Beziehungen zu Baktrien immer 

wieder auf den alten Namen Ta-hsia zurückgekommen sind. Besonders bezeichnend ist 

hierfür die von CHAVANNES übersetzte Angabe der Tang-Annalen (618–906 n.Chr.): 

Le T’ou-ho-lo est appelé parfois T’ou-ho-lo ou Tou-ho-lo. C’est l’ancien territoire du (royaume 

de) Ta-hia. 

Hier werden also alle möglichen Transkriptionen für Tokhåra, Tukhåra geliefert; dagegen 

wird Ta-hsia nur aus rein geographischen Gründen hinzugefügt; an eine lautliche 

Übereinstimmung haben also die Chinesen niemals gedacht. 

— 34 —

Wenn auch diese Tatsache an sich nicht beweiskräftig ist, so gewinnt sie doch an Tragweite, 

sobald wir die beiden anderen Momente sprechen lassen. Das eine ergibt sich aus 

dem Bericht des Entdeckers CHANG CH’IEN. Während er von den Yüeh-chih hervorhebt, daß 

sie als Nomadenvolk von Osten her in das Oxusland eingedrungen seien, um sich an dessen 

Nordufer festzusetzen, betrachtet er die Ta-hsia als die seßhafte Bevölkerung Baktriens, 

die kriegerisch schwach, aber im Handel und Gewerbe äußerst tüchtig sei. Es ist ohne 

weiteres klar, daß ein solches Urteil nicht einem Volke gelten kann, das erst vor kurzem 

eingewandert ist, um das griechisch-baktrische Reich zu stürzen. Mit den Ta-hsia sind also 

zweifellos die alteingesessenen Bewohner gemeint. 

Dann können aber die Tocharer nicht mit den Ta-hsia, sondern nur mit den Yüeh-chih 

identisch sein, die, wie wir wissen, bald nach CHANG CH’IENs Expedition die Ta-hsia voll- 

unterwarfen und damit Herren von ganz Baktrien wurden. Daß diese Lösung die 

ständig 

einzig mögliche ist, wird uns durch die Nachricht bewiesen, die sich auf die frühere Heimat 

der Yüeh-chih bzw. Tocharer beziehen. 

Über die ältesten Sitze der Yüeh-chih sind die Angaben der chinesischen Annalen so 

klar gefaßt, daß die europäische Forschung zu einem völlig gesicherten Ergebnis gekommen 

ist: Die Yüeh-chih wohnten, als sie den Chinesen zuerst bekannt wurden, zwischen 

Tun-huang und Kan-chou, und als die Hauptmasse nach Westen auswanderte (um 160 

v.Chr.), behaupteten sich die Kleinen Yüeh-chih im Gebirge südlich davon; einige werden 

besonders in Huang-chung, dem heutigen Hsi-ning-fu, bezeugt. Was anderseits die Herkunft 

der Tocharer betrifft, die im Chinesischen erst seit dem 4. Jahrh. n.Chr. als Tou-ch’ialo 

und bald darauf als Tu-ho-lo bezeugt werden, so enthalten sich die offiziellen Annalen 

jeder weiteren Äußerung ... 

No doubt, Zhang Qian’s Chinese transcription Daxia 大夏 for the foreign name To- 

cha(ra) had been an unfortunate choice because this was the Chinese name for a total- 

different foreign people, first north and later west of China. This fact is also alluded 

ly 

t o by PELLIOT below (p. 37). In preparation of his mission to search for the Ruzhi 月氏 , 

Z hang Qian, a literate man, may have gone through what written sources on the 月氏 

t here were in his time. Sima Tan, the father of Sima Qian, had just been appointed 

court astronomer / astrologer by Han Emperor Wu (in 140 BCE). In charge of all maters 

of the calendar and in consequence of the imperial archives, Sima Tan with Zhang 

t 

Qian may have gone through such old books as the Yizhoushu 逸周書 where in chap- 

59 a number of foreign nations are mentioned bringing “tribute” (gifts) to the Chi- 

ter 

nese court: from due north 正北 , there were, among others, the Daxia 大夏 , the 

Xiongnu 匈奴 , and the Ruzhi 月氏 . 

With this chapter in the back of his mind, Zhang Qian may have believed for a moent 

that he had the Far Eastern Daxia before him in Eastern Bactria. He was, of 

m 

course, badly mistaken, and so the compilers of later Chinese Standard Histories 

changed the two-character transcription 大夏 into a variety of better-fitting three-charcter 

transcriptions (see below, pp. 37–38) to end the confusion which Zhang Qian’s 

a 

mistake had created. But it was too late: East Asian and Western authors in our times 

jumped on the coincidence and thus helped proliferate and worsen the confusion. 

HERRMANN’s main and only valid reason for his about-face was his new and brilli- 

observation, hinted at by RAPSON: Zhang Qian’s Daxia 大夏 were die alteingeses- 

ant 

sene Bevölkerung Baktriens — the autochthonous population of Bactria which, by the 

t ime of Zhang Qian, had become a melting pot of races, cultures and languages. After 

the Iranians and the Greeks it was the Sakas who left their traces in this local popula- 

This was of paramount importance. It proved that the Daxia = Tocharians had 

tion. 

not come from anywhere else. With this shrewd new insight into Zhang Qian’s text, 

HERRMANN was obliged to believe that the Daxia 大夏 could no longer be equated with 

the Tochari of Trogus and especially not with the TÒcaroi (Tochari) of Strabo’s list 

because the latter were said to be Scythian nomads and recent invaders of Bactria. 

This was HERRMANN’S dilemma. 

FRANKE, in 1930: 338, a text quoted above already, just confirms his earlier findings: 

— 35 —

Tschang K’ien ging, vermutlich am Südhang des T’ien-schan entlang, den Spuren der 

Yüe-tschi nach, und fand sie schließlich, nachdem ihm Leute von Ta-yuan (Ferghana) und 

K’ang-kü (die Kirgisen-Steppen nördlich vom Syr darja oder Jaxartes) das Geleit gegeben, 

in den Ländern am oberen Oxus (Amu darja). Hier waren sie gemeinsam mit anderen Völkerstämmen 

in das griechische Diadochen-Reich Baktrien eingefallen und hatten sich ihre 

gegenwärtigen Wohnsitze erobert. Sie führten in dem fruchtbaren und hochkultivierten 

Lande zusammen mit den Tocharern (Ta-hia) ein behagliches Dasein und waren zu großer 

Blüte gelangt. Es ist leicht zu verstehen, daß die Yüe-tschi unter diesen Umständen 

keine Neigung mehr hatten, gegen die Hiung-nu einen Rachekrieg zu beginnen und mit 

dem weit entfernten Reiche der Han ein Bündnis zu schließen. 

The early German Sinologist is one of the few who, unwaveringly, equate Ta-hia or 

Daxia 大夏 with the Tocharians — not the Bactrians. 

However, the complex issue of the identity of the Tochari kept vexing and torment- 

the greatest minds. KONOW is another example. In 1920: 231–233, he writes: 

ing 

Ich habe im vorhergehenden die Entdeckung meines Freundes SIEG, SbAW, 1918, S. 560 

ff., absichtlich nicht erwähnt, obgleich sie anscheinend alles, was ich bis jetzt über diese 

Fragen geschrieben habe, über den Haufen wirft. SIEG hat bekanntlich nachgewiesen, daß 

die indogermanische Sprache des nordöstlichen Turkistan, welche die Uiguren als tocri, 

d.h. doch wohl sicher tocharisch, bezeichnen, in den Texten selbst årši genannt wird, und 

daß dies Wort årši auch das Reich und dessen Bewohner bezeichnet. Man wird wohl ohne 

weiteres F.W.K. MÜLLER beistimmen, wenn er, idem, S. 566 ff., dies årši mit dem 'Asioi des 

Strabo und dem Asiani des Trogus zusammenbringt, indem ja die Asiani nach Trogus die 

Könige der Tochari waren oder wurden. Wir würden somit mit MÜLLER zu dem Ergebnis 

kommen, daß das Volk selbst tocri, Tocharer, die Herrscherschicht årši Asii, genannt wurde. 

Wenn weiter MÜLLER’s Annahme, die auch FRANKE, OZ 6, S. 83 ff., für wahrscheinlich hält, 

daß das chinesische Yüe-tschi eine Wiedergabe eben des Wortes årši ist, das Richtige treffen 

sollte, wäre somit die Frage gelöst ... 

Die Frage nach dem Verhältnis zwischen Årši und tocri hängt selbstverständlich mit 

einer anderen zusammen: wie steht es mit der allgemein angenommenen Gleichsetzung 

der Yüe-tschi mit den Tocharern ? 

In seinem ›×rånšahr‹, S. 200 ff., hat MARQUART aus einer Prüfung der chinesischen und 

klassischen Nachrichten den Schluß gezogen, daß es sich um zwei verschiedene Völker 

handelt. Auf den Inhalt der chinesischen Nachrichten habe ich schon oben hingewiesen. 

Auf ihrer Wanderung gegen Westen schlugen die Yüe-tschi die Ta-hia und unterjochten 

sie ... Nach dem Schi-ki waren sie mehr ein Handels- als ein Kriegervolk, weshalb sie von 

den Ta Yüe-tschi unterjocht wurden. Ähnlich ist die Darstellung im Ts’ien Han-schu ... 

Damit hat MARQUART, wie ich glaube mit Recht, die Bemerkung des Trogus reges Tocharorum 

Asiani zusammengestellt, und den Schluß gezogen, daß die Ta-hia mit den Tocharern, 

die Yüe-tschi mit den Asiani identisch sein müssen. Daraus folgt aber mit Notwendigkeit, 

daß die Yüeh-tschi und die Tocharer von Haus aus verschiedene Völker waren. Dazu 

stimmen auch die anderen Nachrichten, die uns zugänglich sind. Nach den chinesischen 

Berichten saßen die Ta-hia schon in Baktrien, als die Yüeh-tschi im 2. Jahrhundert v.Chr. 

das Land eroberten ... 

Thirteen years later, now also reversing his own earlier convictions, KONOW writes, 

1933: 463: 

Marquart hat auch nachzuweisen versucht, daß die beiden Formen denselben Namen 

wiedergeben ... 

Heute werden wohl wenige Gelehrte dieser Ansicht sein. Nach freundlicher Mitteilung 

Karlgrens wurde Ta-hia im 2. Jahrh. v.Chr. d’ât-g’â gesprochen, und eine solche Form kann 

unmöglich dem klassischen Tochari zugrunde liegen. Dagegen ist es wohl möglich, daß sich 

die beiden Bezeichnungen sachlich decken. 

To this PELLIOT answers one year later, 1934: 27–40: 

— 36 —

M. Sten Konow écarte le rapprochement fait par Marquart entre le nom de 大夏 Ta-hia 

et celui du Tokharestan, d’autant que M. Karlgren lui a donné *D’ât-g’å pour la prononciation 

de Ta-hia au second siècle avant notre ère, mais admet que les deux noms se recouvrent 

néanmoins en fait et sont synonymes ... 

Géographiquement, le Ta-Hia répond au Tokharestan, le pays des Tukhâras, et Marquart, 

on l’a vu, a proposé d’identifier phonétiquement Ta-Hia et Tukhâra ... je crois volontiers 

que les Chinois de la fin du II siècle avant notre ère ont utilisé, pour transcrire le nom 

e 

des Tukhâras, le nom de Ta-Hia ou «Grand Hia» déjà connu dans le domaine chinois. 

With this, KONOW soon reverses his opinion once again, and, in 1934: 6, states what 

he had written initially, namely in 1920: 

It seems to be generally admitted that the Saraucae, for which other texts have Sakaraukae, 

Sakarauloi, etc., correspond to the Saiwang, the Asiani or Asioi, as they are also 

called, to the Yüe-chi, and the Tocharians to the Ta-hia. There cannot well be any question 

about identifying the name Ta-hia, which according to Professor Karlgren was pronounced 

›d’ât-g’a‹ in the second century B.C., with Tochara, or Yüe-chi, old ›gwat-ti‹ or ›gat-ti‹, with 

Asioi. 

Here, all problems seemed well solved. HERRMANN — who, like KONOW, had subcribed 

to the equation Ta-ha = Tocharer in 1920 — had just two years later been 

s 

forced to reconsider the evidence and in consequence had changed his mind, too. But 

unlike KONOW he repeated his objections in 1937: 1633–1634: 

Als eine Umschreibung von Tokhâra hat man vielfach das chines. ›Ta-hsia‹ (›Ta-hia‹) in 

Baktrien ansehen wollen, vgl. Marquart (1901: 204), O. Franke (1920: 125ff.), Sten Konow (1920: 

233). Das ist aber ein Irrtum. Denn abgesehen davon, daß die alte Aussprache etwa ›tai-ha‹ 

war und sich somit von Tokhâra noch mehr entfernte als der heutige Laut, hat unser Gewährsmann 

Tschang K’ien (um 127 v.Chr.) mit Ta-hsia nicht ein Eroberervolk, sondern die 

alteingesessene Bevölkerung Baktriens bezeichnen wollen; er beging nur den Fehler, daß 

er hierauf den Namen eines ganz anderen Volkes Ta-hsia übertrug, das ihm in der Literatur 

als kulturell hochstehendes Fremdvolk angegeben war. 

TARN, in his famous book of 1938, observes the same important fact (see below, 

p. 67). With that we may go on to discuss the logical implications of this crucial obser- 

and suggest a solution to the dilemma: to accept that the Daxia 大夏 were the 

vation 

town-dwelling, autochthonous population of Eastern Bactria and still stick to the 

equation Daxia = Tochara. 

In China proper, WANG GUOWEI 王國維 , in his Study on the Western Hu 西胡考 , 

ontained in chapter 13 of his famous collection Guantang jilin 觀堂集林 of 1 

ed Marquart ( 馬括德 ) as his main authority when he explained, p. 15 b: 

c 923, 

nam 

My investigations show that the name Du–huo– 

luo [Tochara] originally derives from Da–xia. 

考睹貨邏之名源出大夏 

With this statement the great Chinese scholar joined the rather few enlightened au- 

who got the crucial equation Daxia = Tocha(ra) right. And with this clarification 

thors 

firmly established, we are in a position to rectify what BAILEY, 1947: 151, had written: 

Since the Chinese equated Buddhist Sanskrit Tuḫkhāra with their own name Üe-ṭṣï ... the 

equation of Τόχαροι and Üe-ṭṣï seems certain enough. 

BAILEY’s identification is based on a fallacious assumption. In Weishu 102 and in 

Beishi 97, the Chinese, in fact, transliterate “Buddhist Sanskrit Tuḫkhāra,” not with Üe- 

ṭṣï 

月氏 , but with — 

吐呼羅 (T’u–hu–lo, in Wade-Giles, Tuhuluo, in pinyin), and with 

吐火羅 (T’u–ho–lo; Tuhuoluo) in Suishu 83 and in Tangshu 221; the latter also gives 

吐豁羅 (T’u–ho–lo; Tuhuoluo), 

睹貨邏 (Tu–ho–lo; Duhuoluo), and 

— 37 —

吐呼羅 (T’u–hu–lo; Tuhuluo) as variants. In the last of these four Chinese Standard 

Histories, the (New) Tangshu, it is said — as I have quoted above: 

Tuhuoluo ... is the old Daxia territory. 

(Xin) Tangshu 221B. 6252 

吐火羅 … 古大夏地 

Hence, the Chinese equate Buddhist Sanskrit Tuḫkhāra, not with Üe-ṭṣï 月氏 , but — 

in the time of Sima Qian — with the name Daxia 大夏 . 

From the Shiji and the Hanshu we know that the Daxia 大夏 were subjugated by 

the Ruzhi 月氏 and hence cannot be identical with the latter. BAILEY’s informant on 

the Chinese sources is HALOUN (see above, pp. 24–25). The above erroneous equation 

Τόχαροι = Üe-ṭṣï must thus be his. BAILEY knew his limitations and frankly admitted: 

Indeed, it is evident that no scholar is equipped to control all the sources. We find scholars 

who have done admirable work in one branch stand helpless before essential documents 

in another. One scholar may know to the full Indian materials but be unable to 

handle the Chinese, or another may know the Chinese sources but have slight knowledge 

of Iranian. Admitting then in advance, as an excuse for the present excursus, that we are 

all inadequately equipped in some respect of Central Asiatic studies ... 

In 1985: 126, BAILEY corrects himself, or rather, he corrects his informant HALOUN: 

The Chinese writings of Bud. Skt. tukhåra- are 吐火羅 K 1129, 117, 569 t’u-xuo-lo 

< t’uo-xuâ-lâ and 兜佉羅 K 1017, 491, 569 tou-k’ie-lo < tÿu-k’ia-lâ. Hüan Tsang (A.D. 644) 

mentioned an old country of this people at a ruined site 都貨羅故國 tu-huo-lo ku kuo 

“old city of the T’u-huo-lo” at ancient Så¾a, modern Endere, west of ²er¾en (CHAVANNES 1903: 

155, 221) ... 

In our own times, the topic of Tochara has finally witnessed substantial progress. 

In 1994: 173–178, ENOKI / KOSHELENKO / HAIDARY write: 

As ›Ta-hsia‹ is an exact transcription of ›Tochara‹ (which was the central part of the 

Bactrian kingdom), if the Yüeh-chih were the Tocharians, the conquest of Ta-hsia by the 

Yüeh-chih means the conquest of the country of Tochara by the Tocharians, which seems 

rather strange. The evidence of Sz°-ma Ch’ien shows that Ta-hsia cannot be the Bactrian 

kingdom, but was the country of Tochara divided into several small political units at the 

time of the Yüeh-chih invasion. In other words the Græco-Bactrian kingdom had already 

been destroyed or divided when the Yüeh-chih arrived. Therefore, there is no need to accept 

the identification of the Tocharas with the Yüeh-chih ... 

If the explanation given above is correct, the country of Ta-hsia, which was conquered 

by the Yüeh-chih, cannot have been the Bactrian kingdom, which had already been destroyed 

before the arrival of the Yüeh-chih ... 

According to W. W. TARN, 1938: 272–73, Bactria 

was up to about 141 B.C. under the control 

of Heliocles, who is believed to be the last king of the Bactr ian kingdom. S o the invasion 

may have taken place in that year or some time later and must have been before the coming 

of the Yüeh-chih who occupied the Sogdiana-Bactria region between 136 and 129 (or 128) 

B.C. Strabo tells us that the Bactrian kingdom was destroyed by the Tocharians and three 

other peoples, and, according to Sz°-ma Ch’ien, the country which the Yüeh-chih conquered 

was Ta-hsia. As ›Ta-hsia‹ is believed to be a transcription of ›Tochara‹, and if these two 

statements are accepted, it cannot have been the Yüeh-chih who conquered the Bactrian 

kingdom. 

To these clear and logical statements one only needs to add some minor corrections 

or comments. For one: Ta-hsia/Daxia 大夏 was not the central, but the eastern 

part of Bactria. And when the 月氏 arrived on the scene, the Græco-Bactrian kingdom 

was already divided into two separate parts: Tochara, or eastern Bactria, which was 

ruled by the Sakaraukai or Saiwang 塞王 , and western Bactria, around the capital 

Bactra, which was still ruled by the la 

st Græco-Bactrian kings. When the Ruzhi 月氏 

— 38 —

then evicted the Sakaraukai/Saiwang from Tochara, the latter put an end to Greek rule 

in the city of Bactra (modern Balkh). 

With this, it was only the Sakas (S akaraukai) who took Bactria from the Greeks — 

in two distinct stages. The Ruzhi 月氏 , it seems, never clashed with the Bactrian 

Greeks directly. Instead, they always fought it out with their arch enemy, the Sakaraukai/Saiwang 

塞王 : in the plains of the Ili River, in Sogdiana, in Daxia. 

It is, therefore, of the utmost importance to realize that the Ruzhi 月氏 conquered 

and occupied, not the old Greek kingdom of Bactria, but only Daxia 大夏 , the lands 

which in later times were called Tokharestan (\ukhåristån in TABARÐ; it included 

provinces on both shores of the Amu Darya). This will greatly help to understand that 

the Græco-Bactrian kings lost Bactria proper in two stages and in two irruptions, but 

to just one nomadic people: the Sakaraukai/Saiwang. The Bactrian Greeks may have 

felt that these Sakas were being pushed by a still more powerful nomadic nation — but 

t hey themselves were fighting against, and loosing their kingdom to, this particular 

tribe of the Sakas, the Sakaraukai. When it becomes apparent that the Ruzhi 月氏 , in 

the time of Zhang Qian’s visit, had occupied only Tochara/Daxia 大夏 , we begin to 

wonder whether the 月氏 conquered the whole of Bactria in two successive stages as 

well — as I have been inclined to believe so far —, and in both stages not from the 

Bactrian Greeks, but from the Sakaraukai/Saiwang 塞王 . At first sight, this is corrobo- 

by our Western sources where Trogus first says that the Asiani occupied Sogdi- 

rated 

ana and the Sa(ca)raucae Bactra — taken by some writers to mean Bactria, but it is 

the city of Bactra where the Sacaraucae/Sakaraukai will maintain themselves for a long 

time to come (sse below). 

Prolog 41 (Stage One): 

Scythicae gentes, Saraucae et Asiani, Bactra occupavere et Sogdianos. 

So Trogus. He then goes on to say: Following this, the Asiani became the kings of the 

Tochari (in Eastern Bactria) — because the Daxia, as Zhang Qian had noticed and 

written down, had no overlords any more. 

Prolog 42 (Stage Two): 

Reges Tocharorum Asiani ... 

The development indicated here by Trogus — i.e. the foreign Asiani becoming the 

kings of the local Tochari —, finally proved to be fatal for the Sa(ca)raucae. 

Prolog 42 (Stage Three): 

... interitusque Saraucarum. 

The ultimate destruction if the Sa(ca)raucae, we are by now inclined to assume, 

must have happened in the Western part of Bactria. 

In the first stage, Bactria became divided between the Græco-Bactrians and the Sa- 

in about 145 BCE. In the second stage — which Zhang Qian witnessed in 

caraucae, 

129–128 BCE and wrote about —, Bactria became divided between the Sacaraucae/Saiang 

塞王 and the Asiani/Ruzhi 月氏 . The third and final stage, we are now tempted 

w 

t o think, witnessed the final occupation of all of former Greek Bactria by the 月氏 , a 

f ew years after the Chinese diplomat had returned to Han China. In this last stage, the 

Bactrian Greeks played no active part any more. We envision this final stage as a fierce 

showdown between two decidedly aggressive nomadic peoples of two greatly different 

worlds — and the lightly armed, very fast mounted archers of the Far East prevail over 

the heavily armed and slower riders of Central Asia. 

BERNARD, 1987: 758–767, is the first to recognize this final showdown in the wall friezes 

of a royal summer palace of the Ruzhi 月氏 at Khalchayan, in the valley of the 

Surkhan Darya, a short distance north from Termez on the Oxus: 

C’est dans le troisième quart du II e siècle av. n.è. que le royaume gréco-bactrien, c’està-dire 

l’État grec qui contrôlait depuis l’expédition d’Alexandre la vallée du moyen Oxus, 

— 39 —

entre l’Hindukush au sud et les monts du Hissar au nord, disparaît, submergé par des invasions 

nomades. Cette conquête, commencée par la rive droite et l’extrémité orientale de 

la rive gauche, où la ville grecque d’Aï Khanoum tombe vers 145 av. n.è., s’achève vers 130 

av. n.è. avec le règne d’Hélioclès, le dernier roi grec de la Bactriane proprement dite ... 

Ces nomades nous ont d’abord été connus par quelques allusions des textes classiques, 

principalement une phrase de Strabon ... Les sources chinoises, le Si-Ki, le Han Shou et le 

Hou Han Shou parlent, elles, du peuple des Yüeh-chih, à l’exclusion de toute autre ... 

En 129 l’ambassadeur chinois Chang K’ien les trouva installés sur la rive droite de 

l’Oxus, mais ayant déjà visiblement imposé leur autorité sur la Bactriane méridionale ... 

Quelque temps après le passage de Chang K’ien ils avaient donc mis un point final à la 

conquête de la Bactriane grecque ... 

Il est cependant possible d’y voir maintenant un peu plus clair grâce à deux découvertes 

faites, l’une à Xal¾ajan sur la rive droite de l’Oxus, dans la vallée du Surkhan-darya, il y 

aura bientôt trente ans ... Ces trouvailles nous offrent aujourd’hui, chacune à sa manière, 

une vision incomparablement plus riche et plus diversifiée de la culture des deux peuples 

que nous considérons avoir été les acteurs principaux de la conquête de la Bactriane grecque, 

les Yüeh-chih au centre et à l’est, les Saces ou Sacarauques à l’ouest ... 

Nous partirons du décor figuré ornant un pavillon royal à Xal¾ajan ... dont la construction, 

aux alentours de notre ère, est attribuée à un chef nomade des tribus yüeh-chih. Cette 

attribution repose sur la ressemblance que présente le type physique très particulier, mongoloïde, 

des principaux acteurs des scènes représentées à Xal¾ajan avec l’effigie monétaire 

d’un certain Héraos, qui régna à cette époque dans la région ... 

A partir d’une vingtaine de fragments importants recueillis sur le sol, M me Puga¾enkova, 

qui a fouillé et publié ce monument, a restitué une troupe de cavaliers passant au galop 

ve rs la droite. Elle y voit le retour victorieux d’une troupe d’archers montés yüeh-chih accompagnés 

d’un escadron allié de cavaliers bactriens. L’identification des premiers ne fait 

aucun doute. Ils appartiennent bien à la même ethnie que la famille princière; ils présentent 

les mêmes traits mongoloïdes avec les yeux étirés et obliques, les sourcils remontants 

vers les tempes, le front fuyant, le crâne aplati à l’arrière, les longs cheveux raides 

re jetés vers l’arrière et serrés dans un bandeau, la pilosité du visage limitée à de longs favoris 

et à des moustaches ... L’identification des seconds pose en revanche un problème. 

me 

M Puga¾enkova a proposé d’y voir des représentants de la noblesse bactrienne pour 

deux raisons: d’abord parce qu’ils sont d’un type physique tout à fait différent, de caractèr 

e nettement europoïde, avec, en plus des moustaches, une barbe très fournie; en second 

lieu parce qu’ils portent un lourd armement de cataphractaire ... 

Les études menées ces dernières années par un chercheur soviétique, M. V. Gorelik, 

tendent à montrer que l’armement cataphractaire aurait été mis au point chez des peuples 

nomades entre la mer d’Aral, les Tien-shan et le Pamir ... Diverses indications ... incitent 

également à reconnaître dans les cataphractaires de Xal¾ajan d’autres nomades 

auxquels les Yüeh-chih se seraient heurtés dans leur conquête de la Bactriane ... et qui se 

rattachent à cette nébuleuse de tribus nomades qui se jetèrent sur les provinces occidentales 

du royaume gréco-bactrien dès la deuxième moitié du II e siècle av. n.è. ... à cheval 

revêtus d’une armure de cataphractaire; celle-ci est identique à l’équipement des adversaires 

des Yüeh-chih à Xal¾ajan : cuirasse à longue jupe bardée de grands plaques métalliques 

quadrangulaires, haut protège-cou à bord évasé, casque moulant le crâne et couvrant 

la nuque et les oreilles ... Les peuples nomades qui jouèrent un rôle prépondérant, 

d’après ce que nous en disent les sources classiques, dans les événements qu’on voit se dérouler 

sur la frontière orientale de l’empire parthe à partir de 130 av. n.è., à savoir les 

Saces et les Sacarauques ... 

BERNARD is speaking here about the Western classical sources in which the Sakas- 

Sakaraukai are mentioned prominently amongst those conquering nomadic hordes 

who took Bactria from the Bactrian Greeks. The same nomads are mentioned in the 

Chinese classical sources and are there named Saiwang. We have seen (above, pp. 15– 

— 40 —

16), that these easternmost Scythians were thus called the “Royal Sakas” — the exact 

translation of Chinese Sak–wang 塞王 . 

Zhang Qian, however, seems to have heard absolutely nothing of any Sakas or Sa- 

And so the Shiji does not mention this nomadic people at all. Only the later 

karaukai. 

Hanshu does: in chapter 96 we read the ethnic names Sai 塞 and Saiwang 塞王 . In 

this we have the Chinese transcriptions of the general name Saka and the more speci- 

name Sakaraukai. Now, the Hanshu is partially an edition, and partially a continu- 

ation, of the Shiji: In some places the Hanshu account parallels the Shiji and, in other 

fic 

places, the Hanshu contains additional materials. This is so because the Hanshu was 

not completely rewritten. At first it was meant to be a simple continuation of the Shiji. 

T hen Ban Biao — or more likely his son Ban Gu — decided to incorporate all those 

par ts of the Shiji, which covered the first half of the Former Han Dynasty, before he 

himself continued with the second half — so that their final work would include the 

full period of the Former Han. Hence, Ban Gu called his family’s grandiose master- 

simply Hanshu 漢書 or “Book of the (Former) Han.” 

piece 

Obviously out of respect for the admired work of Sima Qian, the parts included 

f rom the Shiji into the Hanshu were carefully edited and corrected, but otherwise as 

litt le changed as possible. For this reason, the Hanshu is sometimes Sima Qian and 

sometimes Ban Gu. In other words, the Hanshu combines, or is composed of, two he- 

parts — at times the two diverging parts may lead to contradictions or in- 

terogeneous 

c onsistencies. This is clearly so when we want to piece together what the Hanshu 

contains about the fall of the Greek kingdom of Bactria and the conquering nomadic 

peo ples involved in it. In Hanshu 96 we read in the sections on the Ruzhi 月氏 , the 

Wusun 烏孫 , and the Jibin 罽賓 : 

HULSEWÉ/LOEWE 1979: 119–121 

The state of the Ta Yüeh–chih. 

The seat of (the king’s) government is at the 

town of Chien–chih. 

And it is distant by 11,600›li‹ from Ch’ang–an. 

It is not subject to the Protector General. 

... 

Originally (the people) dwelt between Tun– 

huang and Ch’i–lien. 

Then the time came when the ›Shan–yü‹ Mao 

Tun [Moduk, 209–174] attacked and defeated 

the Yüeh–chih, and the ›Shan–yü‹ Lao–shang 

[ Kiyuk, 174–161] killed (the king of) the Yüeh– 

chih, making his skull into a drinking vessel. 

The Yüeh–chih thereupon went far 

away, passing Ta Yüan and proceeding 

west to attack and subjugate Ta Hsia. 

The principal city was established north of the 

Kuei River (or Oxus) to form the king’s court. 

The remaining small group (of the Yüeh-chih) 

who were unable to leave sought protection 

among the Ch’iang tribes of the Southern Moun- 

and were termed the Hsiao Yüeh-chih tains ... 

Hanshu 96A. 3890–3891 

大月氏國 

治監氏城 

去長安萬一千六百里 

不屬都護 … 

本居敦煌祁連間 

至冒頓單于攻破月氏 

而老上單于殺月氏以 

其頭為飲器 

月氏乃遠去過大宛西 

擊大夏而臣之 

都媯水北為王庭 

其餘小眾不能去者保 

南山羌號小月氏 … 

To this passage we have a parallel text in Shiji 123. So what we read here is, in fact, 

Sima Qian quoting Zhang Qian’s Report, later copied and edited by Ban Gu for his Xiyu 

zhuan 西域傳 , the Hanshu chapter on the foreign peoples of the ›xiyu‹ 西域 or 

“Western Regions.“ We are told here that the Ruzhi 月氏 , beaten (a fourth time) by the 

Xiongnu, in one great trek went from the Hexi Corridor to the lands of the Daxia 大夏 

on the upper Oxus River (modern Amu Darya), beyond the kingdom of the Da Yuan 大 

— 41 —

宛 (Ferghana). For us to understand the full meaning of “The principal (= capital) city 

was established north of the Kuei River,” I have to quote XU SONG here, a Chinese 

scholar of the 19th century who wrote a commentary on the whole of Hanshu 96 (more 

will be said about him further down, p. 44). His interpretation — hinted at, but rejected 

by HULSEWÉ/ LOEWE, 1979: 121 — of the new principal city, established by the 月氏 , is 

of prime importance in our context. It proves that the new Royal camp or ordos of the 

still nomadic 月氏 , north of the Oxus, Gui Shui 媯水 in Chinese, replaced the old capital 

of Daxia 大夏 , located somewhere south of that river, in this way becoming the 

new capital. Hence, this new capital was of course located inside Daxia which — at 

least from now on — extended clearly from the Hindukush in the south to the Hissar 

Mountains in the north. In the past, too many scholars have speculated that this new 

ordos of the 月氏 had been outside the boundaries of Daxia, which induced them to 

greatly mistaken conclusions. 

(Xu Song’s) Supplementary comment: 

The Shiji says: “ Daxia is located over two 

thousand ›li‹ southwest of Da Yuan, south 

of the Gui River ” — consequently the capital 

of (earlier) Daxia times (was located) 

south of the river. 

(But) the Greater Ruzhi moved the seat of 

government to the north of the river. 

Hanshu Xiyuzhuan Buzhu, 27a 

補曰 

史記云大夏在大苑西南二 

千餘里媯水南蓋大夏時都 

水南 

大月氏徙治水北也 

It becomes evident here that the Ruzhi 月氏 , by the time of Zhang Qian’s arrival at 

their court, 129 BCE, had decided to moved their principal seat of government from Sa- 

had been for about a generation (see below) —, not 

markand in Sogdiana — where it 

to the old capital of Daxia, but to a convenient place on the near, or northern, bank of 

the Oxus River, possibly close to modern Termez (Termed, Tarmita, Dami 呾蜜 ) — at 

all times the most strategic point for crossing the river. 

HULSEWÉ/LOEWE 1979: 143–145 

The state of Wu–sun. 

The ›Greater K’un–mi‹’s seat of government is 

at the town of Ch’ih–ku. 

And it is distant by 8,900 ›li‹ from Ch’ang–an. 

... 

Originally it was the land of the Sai. 

When the Ta Yüeh–chih turned west, defeated 

and expelled the king of the Sai [recte: the Sai- 

Yüeh–chih, the Ta Yüeh–chih migrated 

wang], the latter moved south and crossed over 

the Suspended Crossing and the Ta Yüeh–chih 

took up residence in his [their] lands. 

Later, when the ›K’un–mo‹ [or king] of 

Wu–sun attacked and defeated the Ta 

to the west and subjugated the Ta 

Hsia, and the ›K’un–mo‹ of Wu–sun took 

up his residence here. 

It is said: » For this reason, among the people of 

the Wu–sun there are (elements of) the Sai race 

and the Ta Yüeh–chih race« ... 

Hanshu 96B. 3901 

烏孫國 

大昆彌治赤谷城 

去長安八千九百里 … 

本塞地也 

大月氏西破走塞王塞 

王南越縣度大月氏居 

其地 

後烏孫昆莫擊破大月 

氏大月氏徙西臣大夏 

而烏孫昆莫居之 

故烏孫民有塞種大月 

氏種云 … 

Zhang Qian’s Report contained a description of the Wusun kingdom, which we 

know was located between the Ili and Chu Rivers and in the region around Lake Issyk 

— 42 —

Köl (Ïssïγ Köl = “Hot Lake” — so named because it does not freeze over in winter), after 

at the latest 161 BCE. We find it in Shiji 123. But Ban Gu has not copied the above text 

from Sima Qian. He has written his own, much longer and greatly improved, i.e. 

updated, account of the Wusun — the small nomadic nation which had once been the 

Western neighbor of the Ruzhi 月氏 , in the half-dessert between Dunhuang and the 

Salt Lake 鹽澤 or Lop Nor. We are told here that the new lands of the Wusun were the 

old lands of the Sai 塞 , more specific: the Saiwang 塞王 or “Royal Sakas.” In this text 

we hear about the Sakaraukai from the Chinese side for the first time. 

At least a century and a half after Sima Qian, Ban Gu has been able to collect plenty 

of new information and at long last provides us with one new and very valuable 

detail on the trek of the 月氏 : on the lush grasslands of the upper Ili River valley, 

parts of modern Kazakhstan and Chinese Xinjiang, the 月氏 had clashed with the 

Saiwang, had driven them away from their excellent pasture grounds and had settled 

there themselves. With this, the first trek o f the 月氏 had come to an end on the upper 

Ili and around Lake Issyk Köl — so we realize from this text. 

Shortly after this, however, the Ruzhi 

月氏 were in turn attacked by the Wusun. 

They themselves were now beaten and driven away. Thus, the 月氏 started a second 

trek, this time seemingly all the way to the Oxus River and the lands of the Daxia 大 

夏 , or Tochari, whom they were able to subjugate. 

And the Wusun settled in the old lands of the Saiwang — without penetrating any 

further west. The Saiwang, we learn from 

the Chinese sources, had in the meantime 

escaped south, crossing a strange and very difficult mountain passage, i.e. a narrow 

gorge in the Himalayas (not the Hindukush, as WYLIE had thought). 

HULSEWÉ/LOEWE 1979: 104 

The state of Chi–pin. 

The king’s seat of government is at the town 

of Hsün–hsien. 

And it is distant by 12,200 ›li‹ from Ch’ang–an. 

It is not subject to the Protector General ... 

When, formerly, the Hsiung–nu con- 

the latter 

qu ered the 

Ta Yüeh–chih, 

moved west and established themselves 

as master of Ta Hsia. 

(It was in these circumstances that) 

the king of the Sai [recte: the Saiwang] 

moved south and established himself 

[themselves] as master of Chi–pin. 

The Sai tribes split and separated and repeatedly 

formed several states ... 

Hanshu 96A. 3884 

罽賓國 

王治循鮮城 

去長安萬二千二百里 


昔匈奴破大月氏大月氏 

西君大夏 


塞種分散往往為數國 … 

Jibin (Chi–pin) 罽賓 is also a name not mentioned in the Shiji. From the lay of the 

land, southeast of the Ruzhi 月氏 in Daxia, it could be close to a land called Shendu 

身毒 in Shiji 123 which is also described as being situated southeast of the Daxia. 

Hence, we have to look for the kingdom of Jibin in Northwestern India. Zhang Qian 

and the Shiji know nothing about Jibin and ve 

ry little about Shendu: not much more 

than just the name — and the fact that trade is going on between Daxia and Shendu. 

Ban Gu, however, has a long story to tell about Jibin. For the purpose of our context, I 

have quoted the above few sentences from the section on Jibin because they are of im- 

bit of information for us here: the 

Saiwang went south to establish themselves in Jibin, i.e. somewhere in Northwestern 

India. And one wonders why the Ruzhi 月氏 are mentioned here. Their east-west 

portance here. Seemingly, there is just one new 

trek has nothing to do with the north-south trek of the Saiwang — so we think. Yet, the 

two sentences about the 月氏 on the one hand and on the Saiwang on the other are 

clearly coupled with the simple connector “and” 而 . As this connection made no sense 

to them, the Translators were at a loss and eloquently tried to evade the problem by 

— 43 —

stating: “It was under these circumstances that...” But that is not what the Chinese 

text says — and it makes still no more sense than a simple “and ...”. 

In the previous paragraph on the Wusun, the Ruzhi 月氏 and the Saiwang 塞王 

were also brought together in two cause-and-effect sentences. Here, however, the logical 

connection is clear: the 月氏 arrive and the Saiwang are driven away. The 月氏 then 

occupy the evacuated land. Concerning the two similar sentences in the paragraph on 

the Jibin, is there a way to bring the Ruzhi 月氏 and the Saiwang 塞王 together in a 

way which is logical and makes sense? There is a very clear one when we take the 

crucial sentences in the paragraph on the Wusun as a model. 

This ingenious answer to the question has been provided by a nineteenth century 

Chinese scholar, quoted briefly above: XU SONG 徐松 (1781–1848). A man of letters, he 

was exiled for some offence and spent six years in Xinjiang (Sinkiang) 新疆 and used 

his time there for intensive studies of the history and topography of the region. One 

outcome was a new edition of Ban Gu’s Hanshu 96 to which XU SONG added his own 

and new comments. These proved to be of great value. We must be grateful that his 

work was published, if only posthumously, in 光緒癸巳 , or 1893. The title of the 

booklet in two parts is Hanshu Xiyuzhuan Buzhu 漢書西域傳補注 . Some eight 

years ago, I looked it up in the Berlin State Library and had copies made of some 

pages. I noticed that XU SONG’s comments to Hanshu 96 were later incorporated into 

the famous edition of the Hanshu by WANG XIANQIAN 王先謙 (1842–1918), the Hanshu 

with Supplementary Notes 漢書補注 , published 1900 (reprint 1983). 

Hanshu 96A.3884 

Previously, when the 

Xiongnu had demol- 

ished the Ruzhi, the 

Ruzhi (going) west 

(established them- 

selves as) the rulers 

of the Daxia. 

In consequence 

to 

this the Saiwang 

(going) south (estab- 

lished themselves 

as) the rulers of the 

Jibin. 

Supplementary note: 

(by Xu Song [1781–1848], 

Hanshu Xiyuzhuan Buzhu, 

1893: 20 b; reprinted in 

Wang Xianqian [1842–1918], 

Hanshu Buzhu, 1900: 96A. 

23 b) 

The Saiwang had 

been the kings of 

the Daxia. 

昔匈奴 

破大月 

氏大月 

氏西君 

大夏 

而塞王 

南君罽 

賓 

補曰 

塞王大 

夏之王 

也 

— 44 —

HULSEWÉ/LOEWE, whose English translation I have quoted a few pages further up, 

do not list the monograph of Xu Song in their extensive bibliography, 1979: 240–256, 

but in the innumerous notes to their translation, Xu Song figures prominently. XU’s 

brilliant and simple remark, which he inserted after the two sentences connected by 

“and” 而 , escaped HULSEWÉ and LOEWE. Or rather, the eminent Sinologists must have 

read it, but did not grasp the paramount importance of the small note which can be 

found 1893: 20b in XU SONG’s, and 1900: 23b in WANG XIANQIAN’s edition of Hanshu 

chapter 96. 

KONOW, 1934: 9–10, published a short discussion of this at first sight so difficult Chinese 

text and its intelligent interpretation by XU SONG. It was unfortunate, however, 

that his paper saw the light of day, not in Europe, but in India. KONOW explained that 

he had asked his friend KARLGREN for a translation of some important Chinese texts, 

including the one on Jibin (Ki-pin). I quote it here with KONOW’s remarks: 

»The Ki-pin kingdom, its king rules in the city of Sün-sien. It is distant from Ch’ang-an 

12,200 li ... Anciently, when the Hiung-nu beat the Ta Yüe-chi, the Ta Yüe-chi went west and 

›chiefed‹ (became rulers of) the Ta-hia, and the Sai-wang went south and ›chiefed‹ Ki-pin. 

The Sai tribes were scattered and constituted several kingdoms in various directions ...« 

Professor Karlgren here adds an important explanatory note: 

“Two interpretations of this passage are possible. The first is that at the time when the 

Yüe-chi went to capture Ta-hia, the Sai, from a more easterly region, went south to capture 

Ki-pin. But then we fail to see the logical connection. Why should the Ts’ien Han-shu here, 

under Ki-pin, mention the Yüe-chi movement? A movement from the north of the Oxus to 

the Ta-hia south of the river could not have caused a movement of the Saiwang from some 

more easterly country to Ki-pin. It would only be a coincidence in time which would 

explain the entry about the Yüe-chi movement in this place. 

The other explanation is more reasonable. The Sai had already spread in various 

directions, and the Sai-wang were chieftains in Ta-hia, when the Yüe-chi movement 

into the Ta-hia country came on. Hence the logical exposé: »When the Hiung-nu beat the 

Ta Yüe-chi, the Ta Yüe-chi went west and became rulers of the Ta-hia, and (the former 

rulers of the Ta-hia, who were now expelled, i.e) the Sai-Wang, the Sai kings, went 

south and became rulers over Ki-pin.« It is then but logical that the narrator adds: »The 

Sai tribes were already m uch scattered. « This explains why the Sai-wang sat as 

chieftains in Ta-hia. This latter explanation has been proposed by the learned Sü Song 

(about A.D. 1800), who in his commentary says: »The Sai-wang were the kings of Ta-hia.«” 

This was a very important revelation by KARLGREN and KONOW which went almost 

unnoticed. With the emendation by XU SONG, the two seemingly unrelated sentences 

suddenly make sense in a very unexpected, though 

definitely convincing way. With this 

emendation, the translati on of the Chinese text can be corrected accordingly: 

(HULSEWÉ/LOEWE 1979: 104) 

The state of Chi–pin. 

The king’s seat of government is at the town 

of Hsün–hsien. 

And it is distant by 12.200 › li‹ from Ch’ang–an. 

It is not subject to the Protector 

General ... 

When, formerly, the Hsiung–nu conquered the 

Ta Yüeh–chih, the latter moved west and es- 

of Ta Hsia. 

And consequently the Saiwang (who had 

tablished themselves as masters 

been the kings of the Daxia and were now expelled) 

moved south and established 

themselves as masters of Chi–pin. 

The Sai tribes split and separated and re- 

peatedly formed several states ... 


罽賓國 

王治循鮮城 

去長安萬二千二百里 


昔匈奴破大月氏大月氏 

西君大夏 


塞種分散往往為數國 … 

— 45 —

The Sakaraukai/Saiwang 塞王 , evicted by the Ruzhi 月氏 from their traditional pasture 

grounds in the upper Ili Valley some time after 165 BCE, established themselves 

west of the Jaxartes in Sogdiana — briefly hinted at by Strabo who calls them simply 

Sakai there (see further down, p. 79). Until that time, Sogdiana may still have been in 

the hands of the Bactrian Greeks. But soon afterwards, i.e. some time after 163 BCE, 

the Ruzhi 月氏 , evicted from the Ili River regions by the Wusun 烏孫 , now subjugated 

the Kangju 康居 further west and soon spilled across the Jaxartes themselves. There 

they ran into the Sakaraukai/Saiwang 塞王 once again and drove them south and out 

of Sogdiana. The Saiwang, forced to cross the Hissar Mountains, conquered the Daxia 

(Tochari) on the upper Oxus River and became their rulers for half a generation. They 

were the elusive nomads who stormed and burnt down Ai Khanum which at that time 

was called Eucratidia. With the fall of this major Greek citadel, the road was open for 

them to occupy the whole of Tochara, the eastern part of Bactria. Archaeological evidence 

from the careful excavations of the old city of Alexandria / Eucratidia / Ai Khanum 

tells us that this stronghold fell in about the year 145 BCE. 

1985: 97–102, BERNARD writes : 

Le dernier règne attesté dans le monnayage d’argent et de bronze recueilli dans la fou- 

ille d’Aï Khanoum est celui d’Eucratide I ... On peut supposer que l’assassinat d’Eucratide 

par l’un de ses fils (Hélioclès ou Platon) aurait été suivi d’une période d’incertitudes et de 

troubles dont auraient profité les envahisseurs nomades à qui nous attribuons l’incendie 

qui détruisit le palais et qui marque l’abandon de la ville par sa population grecque. Par 

une chance rare un document épigraphique découvert dans la fouille nous permet de dater 

assez exactement cet événement ... Dans la couche de destruction de la trésorerie du palais 

ont été découverts plusieurs fragments d’un gobelet en céramique à fond pointu qui, 

comme nous l’apprennent les deux inscriptions à l’encre qu’il porte, avait servi de bouchons 

à un vase contenant de l’huile d’olive. L’une d’entre elles était ainsi rédigée: 

”Etouj kd'. [- - -] ... 

L’opération est datée à la première ligne: œtouj kd' = année 24. L’ordre des chiffres, dizaines 

d’abord, unités ensuite, exclut qu’il manque à droite un chiffre de centaines ... La 

date est ainsi bien assurée: 24 et non 124. 

A quel roi rapporter cette année régnale ou cette ère ? ... Étant donné le contexte archéologique 

du bol inscrit, tout proche de la chute de la ville grecque, nous ne pouvons rapporter 

la date de 24 ni à l’ère séleucide (312 av. J.-C.), ni même à une ère supposée de Diodote 

(vers 250–240 av. J.-C.), laquelle serait encore trop haute. L’absence de monnaies d’Hélio- 

dans la fouille nous assure, d’autre part, que l’abandon de la ville par ces colons grecs 

clès 

se produisit sous le règne d’Eucratide ou immédiatement après ... L’année régnale ou l’ère 

à laquelle se rapporte la date de 24 ne peut donc être que celle d’Eucratide I qui avait, lui, 

les meilleures raisons du monde de marquer par un nouveau comput le caractère exceptionnel 

de son règne. Sa réputation de grand souverain et de grand capitaine était suffisam- 

ment connue des anciens pour que Trogue Pompée traçât un parallèle entre sa carrière et 

celle de Mithridate I. Apollodore d’Artémita le fait régner sur mille villes et, au dire de Strabon 

(XI, 11, 2), il avait donné son nom à l’une d’elles ... 

La date communément admise pour le début du règne d’Eucratide est 170 environ, d’al’ 

accession au trône de Mithridate I. La date 

de 24 sur le tesson inscrit de la trésorerie d’Aï Khanoum se place ainsi en 146 au plus tôt, 

en 142 au plus tard, si nous acceptons une marge maximale de 5 ans. On peut adopter une 

près la concomitance attestée par Justin avec 

date moyenne de 145 av. J.-C. Le règne d’Eucratide auquel on attribuait généralement jussensiblement 

plus long qu’on ne l’avait pen- 

qu’ici une quinzaine d’années aurait donc été 

sé, puisqu’il aurait duré quelque 25 ans. 

At the time, the French scholars based their reconstruction of the fall of the Græco- 

1961: 264–289) — where Sima Qian, excerpting 

Zhang Qian’s Report, knows nothing as 

Bactrian kingdom mainly on BURTON WATSON’s translation of Shiji 123 (first published 

yet of any Saka involvement in the nomadic incursions into Bactria. 

— 46 —

1990: 96–97, BOPEARACHCHI writes: 

The Chinese imperial annals (the ›Shih-chi‹ and the ›Han Shu‹) provide us with texts 

based on a report allegedly made by a certain Chang K’ien, an envoy of the Han emperor 

Wu Ti, to the western provinces between 138 and 126 BC. He tells us about the arrival in 

Central Asia of the Yüeh-chi in the second half of the second century BC. One could derive 

from this Chinese source a picture of a thrust which took place progressively in two stages. 

The numismatic data provided by the Qunduz and Ai Khanum hoards would thus corroborate 

this picture. 

In the first stage the Yüeh-chi nomads must have taken the territories situated north of 

the Oxus, i.e. Sogdiana, [together with] the region of Ai Khanum at the eastern extremity of 

the plain of Bactria, on the left bank of the river, and the second stage of this move must 

have already been completed at the time of the visit by the Chinese ambassador Chang 

K’ien in these regions in 129–128 BC. P. Bernard was able to date exactly the catastrophe 

which brought the existence of the Greek city of Ai Khanum to an end (or in other words 

the first invasion of Sogdiana by the Yüeh-chi), thanks to an inscription, found in a destruction 

stratum, giving a clear ›terminus post quem‹, the twenty-forth year in the reign of Eucratides 

(= 148–7 BC). It is likely that the destruction of the Greek city of Ai Khanum and 

the first stage of the Yüeh-chi invasion and also the death of Eucratides I took place more 

or less at the same time, that is around 145 BC. 

The assured dating of c. 145 BCE for the destruction of Eucratidia (Ai Khanum) by 

the Sakaraukai/Saiwang 塞王 is a most fortunate outcome of the French excavations 

i n Afghanistan (at that time the kingdom of Zahir Shah where I myself participated in a 

study and work program in the summer of 1965). After this date, Græco-Bactrian kings 

still maintained themselves in Western Bactria, i.e. in the capital Bactra and environs. 

This was Stage One of the Fall of Greek Bactria. 

The Ruzhi 月氏 , or Asiani, now consolidated their position in Sogdiana — as Trogus 

indicates in his Prolog to chapter 41. This done, they, too, crossed the Hissar 

M ountains to the south and conquered the land north of the Oxus River. After this, the 

Ruzhi 月氏 attacked, not the Bactrian Greeks in Western, but their arch enemy, the 

S aiwang, in Eastern Bactria, i.e. in Daxia/Tochara. As there was simply no other route 

o f escape left open for them, the Sakaraukai/Saiwang 塞王 then fell upon the Bactrian 

Greeks in the remaining part of Bactria, around the capital Bactra. The last Græcoactrian 

king was unable to withstand the Saka onslaught: he either was killed in the 

B 

d esperate fights to save Bactra, or in the end resolved to evacuate all of Bactriana and 

w ith his armies withdrew beyond the Hindukush barrier into the Kabul Valley. He was 

H eliocles, the eldest son of Eucratides. He killed his father in such a cruel, abominable 

way that we cannot understand his motives unless it all happened, not before, but imediately 

after the fall of Eucratidia and all of Eastern Bactria — which Eucratides 

m 

had failed to save by returning in time from his ambitious Indian campaigns. 

When in the west the Sakaraukai/Saiwang 塞王 stormed Bactra, the Ruzhi 月氏 in 

t he east occupied the lands, which the Sakaraukai had possessed for a rather short 

t ime: Tochara, or Daxia 大夏 . The Asiani, or Ruzhi 月氏 , became the new kings of 

t he Tochari and their make-shift royal camp on the near, or north, side of the Oxus — 

p ossibly not far from modern Termez — became their new political center and for 

some time replaced the old capital of Tochara, where the evicted Saka kings had 

r esided. In his superb Catalogue Raisonné, 1991: 74, BOPEARACHCHI gives the reign of 

H eliocles I as c. 145–130 BCE. He lost Bactra to the Sakas. This was Stage Two of the 

Fall of Greek Bactria. 

When Zhang Qian, with only his faithful Xiongnu servant left of his once large party, 

finally reached Central Asia, the king of Ferghana sent him to Samarkand. That was 

where he knew the king of the Ruzhi 月氏 to reside. But the war against the Sakaraukai/Saiwang 

塞王 in Daxia 大夏 just over, the 月氏 king’s court had been moved from 

Samarkand in Sogdiana to the newly conquered lands south of the famous “Iron 

— 47 —

tes” 鐵門 in the Hissar Mountains. 

Ga 

The court was established provisionally on the 

near side of the Oxus and there became the new capital of Daxia — as Xu Song has 

h 

elped us to understand the crucial sentence in the chapter on the Greater Ruzhi 大月 

氏 , Hanshu 96 A. 3891. Here, Zhang Qian was received by the new king of the 月氏 — 

h is name is not mentioned in the Chinese sources, but he was the son of the king slain 

b y the Xiongnu in about 165 BCE — and the Chinese envoy was later shown around 

t he old capital of Tochara and its flourishing markets. In this context it is of particular 

i mportance to know that Zhang Qian arrived on the shores of the Oxus River in the 

summer of 129 BCE — and not one to three full years later. 

The final destruction of the Sakaraukai in Western Bactria by the Ruzhi 月氏 — we 

are inclined to believe — could now only be a matter of years. It should have happened 

some time after Zhang Qian had returned to Han China. The Shiji knows nothing 

about it. But it seems only too logic that the 月氏 were now in a position to take the 

whole of Greek Bactria and in that way extinguish the Sakas who would then vanish 

f rom view, i.e. become 月氏 subjects like shortly before them the Tochari further to 

t he east. Pompeius Trogus had the story in his Historiae Philippicae, for in the Prolog 

to chapter 42 he promises to tell us the full story. Junianus Justinus’ Epitome of Trous, 

however, left us not a single syllable of it. This, I imagined, was Stage Three of the 

g 

Fall of Greek Bactria. 

However, there are strong indications that the actual historical developments in 

this final stage were more complicated than the straight-forward scenario sketched 

above. The first objections come from Zhang Qian himself. The Chinese envoy of Han 

Emperor Wu spent more than one full year in Daxia — summer 129 to late summer or 

early fall 128. However, he knows nothing of a recently-established kingdom of the 

Saiwang in neighboring Western Bactria. In his Report, and consequently in Shiji 123, 

the very name Saiwang 塞王 is not mentioned at all. For Zhang Qian the western 

neighbors of the Ruzhi 月氏 are the Parthians: their mighty kingdom is the largest far 

and wide. He calls it Anxi 安息 — the Chinese transcription of the name Ar(sa)cids. 

(WATSON 1993: 233–235) 

(27) (Zhang) Qian in person visited the lands of the 

Da Yuan, the Great Yue–zhi, the Da–xia, and the 

Kang–ju, and in addition he gathered reports on five 

or six (other) large states in their neighborhood. 

All of this information he related to the Son of Heaven 

(on his return). (The substance of) his report 

was (as follows): ... 

(59) (The capital of the) An–xi (Parthians) is located 

some several thousand ›li‹ west of (the capital of) 

the Great Yue–zhi. 

(60) Their people is settled on the land, cultivating 

the fields and planting rice and wheat. 

(They also make) wine out of grapes. 

(61) (They have) walled cities like the Da Yuan. 

(62) To them belong several hundred cities, small 

and big. 

In area (this country measures) several thousand ›li‹ 

square which makes it an extremely large kingdom. 

(63) It borders the Gui (Oxus) River. 

It has bazars and the inhabitants (who are) merchants 

use carts and boats to travel to neighboring 

countries, sometimes (journeying) several thousand 

›li‹. 

(64) With silver they make coins and the coins bear 

Shiji 123. 3160–3163 

騫身所至者大宛大月 

氏大夏康居而傳聞其 

旁大國五六 

具為天子言之曰 … 

安息在大月氏西可數 

千里 

其俗土著耕田田稻麥 

蒲陶酒 

城邑如大宛 

其屬小大數百城地方 

數千里最為大國 

臨媯水 

有市民商賈用車及船 

行旁國或數千里 

— 48 —

the face of their king. 

(When) the king dies, (the coins are) invariably 

changed and (new) coins (issued) with the face of 

his successor. 

(65) (The people) write horizontally on (strips of) 

leather and this way keep records. 

以銀為錢錢如其王面 

王死輒更錢效王面焉 

畫革旁行以為書記 

What a surprising wealth of detailed information on a country which Zhang Qian 

did not see in person. One likely explanation should be that the Chinese explorer met, 

not only Indo-Greek merchants in the marketplace of Lanshi 藍市 , the old capital of 

Daxia 大夏 (Tochara), but also those from Parthia whom he interviewed at length. And 

it is interesting to note that directly after Anxi 安息 (Parthia), Zhang Qian’s Report in 

Shiji 123 goes on to speak of Daxia/Tochara in a chapter which I shall quote further 

down (pp. 72–73). The absence of any information on a Saiwang state 塞王國 , now expected 

to be found between Parthia and Tochara, is telling. In all probability, no such 

independent Saka kingdom existed in Zhang Qian’s time, 129–128 BCE. 

In addition to the Chinese sources, we have two classical Western texts which help 

us to understand what happened to the remnants of Bactria just before it was evacuated 

by the last Greek kings. One is found in Justinus’ Epitome of Trogus’ History, and 

the other in Strabo’s Geography. 

(WATSON 1886: 276) 

Almost at the same time that 

Mithridates ascended the 

throne among the Parthians, 

Eucratides began to reign 

among the Bactrians; both of 

them being great men. 

But the fortune of the Parthians, 

being the more successful, 

raised them, under this 

prince, to the highest degree 

of power; 

while the Bactrians, harassed 

with various wars, lost not 

only their dominions, but 

their liberty; for having suffered 

from contentions with 

the Sogdians, the Drangians, 

and the Indians, they were at 

last overcome, as if exhausted, 

by the weaker Parthians. 

(SEEL 1972: 443–444) 

Ungefähr zur gleichen Zeit wie 

im Partherland Mithridates 

kommt in Baktrien Eukratides 

zur Regierung – beides bedeutende 

Männer. 

Aber das Glück, das den Parthern 

mehr gewogen war, brachte 

sie unter diesem 

auf den Gipfel ihrer Macht. 

Die Baktrianer dagegen, in manund 

herge- 

cherlei Kriegen hinworfen, 

verloren nicht allein ihre 

Herrschaft, sondern auch ihre 

Freiheit, denn sie erschöpften 

sich durch Kriege mit den Sogdi- 

mit Dran- 

anern und Arachosiern, 

gern, Areern und Indern und wur- 

den zuletzt, gleichsam ausgeblu- 

Parthern 

tet, von den an sich schwächeren 

überwältigt. 

Herrscher duce i 

Hist. Phil. Epit. 41.6.1–3 

Eodem ferme tempore, 

sicut in Parthis Mithridates, 

ita in Bactris Eucratides, 

magni uterque viri, 

regna ineunt. 

Sed Parthorum fortuna 

felicior ad summum hoc 

mperii fastigium 

eos perduxit. 

Bactriani autem per varia 

bella iactati non regnum 

tantum, verum etiam 

libertatem amiserunt, 

siquidem Sogdianorum 

et Arachotorum et Dran- 

garum et Areorum Indorumque 

bellis fatigati ad 

postremum ab invalidioribus 

Parthis velut exsan- 

gues oppressi sunt. 

Here we are told that the Bactrian Greeks had fought too many wars in too many 

directions. Finally bled out, they became victims of the actually inferior Parthians. 

After Eastern Bactria (Tochara) had been lost to the Sakaraukai/Saiwang 塞王 — who 

are obvio usly called Sogdiani in Justinus’ text — what exactly happened in Western 

Bactria in the crucial years 145–130 BCE ? If, indeed, the Parthians won the ultimate 

victory over the last Greek kings in Bactria, what was the fate of the Sakas after they 

were expelled for good from Daxia 大夏 (Tochara) by th e Ruzhi 月氏 — around the year 

130 BCE ? Have we been mistaken to believe that they went west and made an end to 

the Græco-Bactrian kings in their last stronghold, Bactra ? The next text, by Strabo, 

may indicate an answer to these questions. 

— 49 —

(JONES 1928: 275) 

Now at the outset Arsaces was 

weak, being continually at 

war with those who had been 

deprived by him of their territory, 

both he himself and his 

successors. 

But later they grew so strong, 

always taking the neighbouring 

territory, through successes 

in warfare, that finally 

they established themselves 

as lords of the whole of the 

country inside the Euphrates. 

And they also took a part of 

Bactriana, having forced the 

Scythians, and still earlier 

Eucratides and his followers, 

to yield to them; and at the 

present time they rule over so 

much land and so many 

tribes that in the size of their 

empire they have become, in a 

way, rivals of the Romans ... 

Geographika, 11.9.2 

Kat' ¢rc¦j mn oân ¢sqen¾j 

Ãn diapolemîn prÕj 

toÝj ¢faireqšntaj t¾n cè- 

ran kaˆ ¢utÕj kaˆ oƒ diade- 

x£menoi ke‹non: 

œpeiq' oÛtwj ‡scusan ¢fairoÚmenoi 

t¾n plhs…on ¢eˆ 

di¦ t¦j n to‹j polšmoij 

katorqèseij, éste teleutîntej 

¡p£shj tÁj ntÕj 

EÙfr£tou kÚrioi katšsthsan. 

'Afe…lonto d kaˆ tÁj 

BaktrianÁj mšroj bias£- 

menoi toÝj SkÚqaj kaˆ œti 

prÒteron toÝj perˆ EÙkrat…dan, 

kaˆ nàn p£rcousi 

tosaÚthj tÁj gÁj 

kaˆ tosoÚtwn qnîn, éste 

¢ nt…paloi to‹j `Rwma…oij 

trÒpon tin¦ gegÒnasi katÕ 

mšgeqoj tÁj ¢r- 

t¦ 

cÁj ... 

(RADT 2004: 353) 

Anfänglich waren er (Arsakes) 

selber sowohl als seine Nachfolger 

schwach wegen der Krie- 

die 

ge, die sie ständig gegen 

ihres Landes Beraubten zu füh- 

über soviel Land 

ren hatten. 

Dann aber wurden sie dadurch, 

dass sie dank ihrer Erfolge in 

den Kriegen sich das jeweils 

benachbarte Land nahmen, so 

stark, dass sie schließlich Herr 

über das ganze Gebiet diesseits 

des Euphrates wurden. 

Sie nahmen sich auch einen 

Teil der Baktriane, indem sie 

die Skythen – und vorher schon 

Eukratides und die Seinen – 

überwältigten; und haben jetzt 

die Herrschaft 

und soviele Völker, dass sie es, 

was die Größe ihres Reiches betrifft, 

in gewissem Sinne mit den 

Römern aufnehmen können ... 

From these statements, we are inclined to assume that the Sakas or Scythians have 

not been as successful in Wes tern Bactria as they had, before, in Tochara/Dax 

ia. They 

may have been able to take Bactra and oust the last Greek sovereigns from there, only 

to be soon overcome by the powerful Parthians who took a part of Bactriana, having 

forced the Scythians. Or, I thought for a while, they voluntarily enlisted Parthian help 

to withstand yet another attack by the dreadful Ruzhi 月氏 . Bu 

t Strabo’s text is rather 

clear here: the Sakas are crushed between two great powers. 

VON GUTSCHMID, cited abo 

ve a first time, comes very close to the truth when, 1888: 

70–71, he writes: 

Die Identität der Tocharer und der Grossen Yue–tshi unterliegt keinem Zweifel ... Wäh- 

östlich die Tocharer, 

rend aber die classischen Berichte zwei Hauptvölker kennen, mehr mehr westlich die Sakarauken, kennen die chinesischen auf dem 

Boden des ehemaligen 

Hellenenreiches nur ein einziges, die Tocharer, und neben ihnen in Margiana die Parther. 

Sollen also die Ersteren nicht Lügen gestraft werden, so bleibt nichts übrig als anzuneh- 

fällt, die Occu- 

men, dass vor dem Jahre 128, in welches die Anwesenheit des Tshang–kien 

pation eines Theiles des Hellenenreiches durch die Sakarauken und jene Thatsache liegt, 

für die wir ein Zeugniss des Strabo (XI, p.515) besitzen, nämlich die Wegnahme eines 

Theils Baktriens durch die Parther nach Bewältigung der Scythen. Wir wissen nunmehr, 

dass dieses Volk die Sakarauken, das ihnen entrissene Land Margiana war; eine Drachme 

des Phraates II. mit der Aufschrift Margian» (GARDNER, 1877: 33) gibt für das Letztere eine 

urkundliche Bestätigung. Dass den sorgfältigen chinesischen Berichterstattern ein so namhaftes 

Volk wie die Sakarauken unbekannt geblieben sein sollte, ist undenkbar. 

VON GUTSCHMID believes in an ethnic identity of the Ruzhi 月氏 and the Tochari 

which is untenable as this study strives to show. Also, he does not know XU SONG’s 

crucial comment which was published five years after his own book. But his reasoning, 

that the Chinese writers — Zhang Qian himself and his epitomator Sima Qian — 

would have mentioned the Sakaraukai, if they had been the sovereign rulers of Western 

Bactria in the time of Zhang Qian’s visit as they had reigned in Daxia/Tochara 

shortly before for some time, is of importance and very valid: with the exception that 

— 50 —

(JONES 1928: 281) 

Their cities were Bactra (al- 

so called Zariaspa, through 

which flows a river bearing 

the same name and emptying 

into the Oxus), and Darapsa, 

and several others; 

among these was Eucratidia, 

which was named after 

its ruler. 

The Greeks took possession 

of it and divided it into satrapies, 

of which the satrapy 

of Turiva and that of Aspionus 

were taken away from 

Eucratides by the Parthians. 

And they also held Sogdiana, 

situated above Bactriana 

towards the east between 

the Oxus River, which forms 

the boundary between the 

Bactrians and the Sogdians, 

and the Iaxartes River; 

and the Iaxartes forms also 

the boundary between the 

Sogdians and the nomads. 

the part of Bactria, which the Parthians took over from the Sakaraukai before Zhang 

Qian’s arrival on the Oxus, was not Margiana. This region had been lost to the 

Parthians earlier, i.e. in the time of Eucratides, as the following passage in Strabo’s account 

of Bactria shows. 

(RADT 2004: 357) Geographika, 11.11.2 

An Städten hatten sie Baktra, PÒleij d' econ t£ te B£ktra, 

das auch Zariaspa genannt wird 

¼nper kaˆ Zari£span 

( hindurch strömt ein gleichnami- 

kaloàsin, ¿n diarre‹ Ðm- 

Fluss, der in den Oxos münènumoj 

ger 

potamÕj mb£l- 

det), Darapsa und mehrere anraya 

lwn e„j tÕn ’Wxon, kaˆ D£- 

dere; 

kaˆ ¥llaj ple…ouj : 

dazu gehört auch das nach dem toÚtwn d' Ãn kaˆ ¹ EÙkrat…deia, 

einstigen Herrscher benannte Eu- 

toà ¥rxantoj 

kratideia. 

pènumoj. 

Die Griechen, die das Land in Oƒ d' katascÒntej aÙt¾n 

Besitz nahmen, haben es auch “Ellhnej kaˆ e„j satrape…aj 

in Satrapien eingeteilt (von de- 

diVr»kasin, ïn t»n 

die des Aspiones und Turi- 

nen 

te 'Aspiènou kaˆ t¾n Tou- 

ua dem Eukratides von den Par- 

rioÚan ¢fÇrhnto EÙkra- 

entrissen wurde). 

thern 

t…dhn oƒ Parqua‹oi. 

Sie setzten sich auch in den Be- 

”Escon d kaˆ t¾n Sog- 

von Sogdiane, das nach Osdian¾n 

sitz 

Øperkeimšnhn prÕj 

ten über Baktriane hinaus liegt, ›w tÁj BaktrianÁj metaxÝ 

zwischen dem Oxos-Fluss – der toà te ”Wxou potamoà, Öj 

das Land der Baktrier und das Ðr…zei t»n te tîn Baktr…- 

der Sogdier voneinander trennt – wn kaˆ t¾n tîn Sogd…wn, 

und dem Iaxartes; 

kaˆ toà 'Iax£rtou: 

dieser bildet die Grenze zwischen 

oátoj d kaˆ toÝj Sogmadenm£daj. 

den Sogdiern und den Nod…ouj 

Ðr…zei kaˆ toÝj no- 

The Satrapies of Aspiones and Turiva are otherwise completely unknown. GROS- 

KURD, 1831: 410, following DU THEIL, suggests amending Turiua or Turiva into Tapuria. 

This is a region, ment ioned by Polybios (10. 46: ...Tapour…an...), which had been fought 

ove r by Antiochos and the Bactrian king Euthydemos near the Areios River, i.e. in the 

area between Parthia, Bactria and Aria — in other words: in Margiana. Bactria in 

those times extended as far west as the Areios or Ochos River (Herî Rûd), a short dis- 

west of the Margos (Murghâb) River which flows through Margianê (Merv). The 

tance 

Ochos had been the border to Parthia. With this geographic situation in mind, VON GUT- 

S CHMID suggests that the Sakas, when evicted a second time by the Ruzhi 月氏 , now 

from Sogdiana, continue their exodus straight on in a westerly direction, i.e. by crossing 

the Oxus. But Zhang Qian, in Shiji 123, says that Parthia in his time in the east borders 

on the Oxus River 臨媯水 (see above, p. 48). This must have been so after the Bactrian 

G reeks had lost the two satrapies mentioned by Strabo. And this, then, is the reason 

why the Sakas/Sakaraukai/Saiwang turn south when chased out of Sogdiana by the 

Ruzhi 月氏 . Hence, the part of Bactria, which the Parthians take over from the Saka- 

is definitely not Margiana. Instead, it is the very heart of Bactria: the capital 

raukai, 

Ba 

ctra itself. In an preemptive strike the Parthians effectively prevent the Ruzhi 月氏 

from subjugating the Sakas in Bactra — and thus the two greatest powers in the area 

now become the main contenders for hegemony in Central Asia. 

And this is how the Sakaraukai/Sacaraucae meet their destruction — the interitus 

Sa(ca)raucarum of Trogus’ Prolog 42. Hence, it are the awe-inspiring Parthians who 

prevent the Ruzhi 月氏 from conquering the whole of former Greek Bactria. It goes 

— 51 —

without saying that the Parthians and the Ruzhi 月氏 become bitter foes — as many 

centuries later the Græco-Roman historian Ammianus will tell us (below, p. 65). 

That the Sakaraukai/Saiwang 塞王 maintained themselves in the capital Bactra 

and Western Bactria for some time as Parthian vassals, is amply proved by a recent 

find, also not yet known to VON G UTSCHMID : the small, but rich Saka necropolis on 

Tillya-tepe, about 100 km west of Bactra. The tombs contained mainly Part 

hian 

coins — and one gold coin of 

the Roman Emperor Tiberius, minted in Lugdunum (modern 

Lyon) 16–21 CE. This late date shows that the Ruzhi 月氏 were unable to con- 

date, or for more than 

quer the whole of Bactria before 30 CE as the earliest possible a century and a half after they had subjugated Daxia 大夏 (Tochara). The tombs of Sa- 

of NARAIN who, in 

ka nobles near Bactra disprove the seemingly convincing assertions 1957: 140, writes: 

The prominence which is given to the Ta–hsia in the Shih–chi 

is not found in the Ch’ien 

Han Shu. It therefore seems evident that Bactria proper south of the Oxus river must have 

come under the complete political subjugation of the Yüeh–chih either after the Shih–chi 

was written or at a time quite near its completion, when the news had not reached Ssu– 

ma–ch’ien, but definitely long before the composition of the Ch’ ien Han Shu. Shih–chi was 

completed in 99 B.C., and therefore, in round numbers, we may 

say that the occupation 

took place about 100 B.C. 

NARAIN’s is the same old mistake we had to deal with above and which dies hard. 

Implicitly, he believes that Ta–hsia (Daxia) stands for Bactria 

proper — which it does 

not. Instead, the Chinese transcription stands, not for Bactria, but for the Western 

name Tochara = Eastern Bactria, later known under the name Tocharistan. A comparison 

of Shiji 123 with Hanshu 96 simply proves that Daxia 大夏 was in the hands of 

the Ruzhi 月氏 about 100 BCE. Above, we have seen that a correct interpretation of 

Shiji 123 shows that this wa 

s already the case at the time of Zhang Qian’s visit to 

Daxia/Tochara, i.e. a full gen eration before 100 BCE. We do not have to wait for the 

much later Hanshu — for it h as nothing new to tell us in this respect. Saka Tillya-tepe, 

however, proves that Western Bactria cannot have been subjugated by the 月氏 before 

the times of the Later Han (26–220 CE), and so we have to turn to the Later Hanshu to 

check, i.e. to the Hou Hanshu 後漢書 . There, we are told that Bactra had been finally 

conquered . The feat was accomplished by one of the five well-known Ruzhi viceroys or 

xihou 翎侯 . In Weishu 102 they are said to have all belonged to the Zhaowu 昭武 family 

of the reigning monarch (see above, p. 28). He overthrew the other four viceroys, proclaimed 

himself king — in clear rebellion against the legitimate Ruzhi 月氏 king — 

and went on to oust the latter, we may assume. 

(DE GROOT 1926: 101) 

... 

Als ›Goat–si‹ von ›Hung–nª‹ 

vernichtet war, wanderte es 

nach ›Ta–ha‹ (Tochara) aus 

(PULLEYBLANK 1968: 247-248) 

The Great Yüeh–chih country 

has its capital at the 

city of Lan–shih (Khulm). 

To the west it is 49 days’ 

march to An–hsi (Arsak 

Parthia). 

To the east it is 6,537 ›li‹ 

to the seat of the Senior 

Administrator (Chang– 

shih) and 16,370 ›li‹ to Lo– 

yang. 

It has 100,000 households, 

400,000 mouths and over 

100,000 trained soldiers. 

Formerly the (Great) 

Yüeh–chih, on being over- 

thrown by the Hsiung–nu, 

Hou Hanshu 88. 2920–2921 

大月氏國居藍氏城 

西接安息四十九日 

行 

東去長史所居六千 

五百三十七里去洛 

陽萬六千三百七十 

里 

戶十萬口四十萬勝 

兵十餘萬人 

— 52 —

und teilte dort sein Reich in 

fünf Jabgu, nämlich 

›Hiu–bit‹, ›Šang–bi‹, 

›Kui–song‹, ›It–tok‹ 

und 都密 ›Tª–bit‹. 

Mehr als ein Jahrhundert 

später bekriegte und vernichtete 

丘就卻 ›K'u–tsiu– 

k'iok‹ (Kud{øla Kadphises), 

der Jabgu von ›Kui–song‹, 

die vier anderen Jabgu und 

erhob sich selbst zum König. 

Sein Reich nannte ihn Kö- 

von ›Kui–song‹. 

nig 

Er griff ›An–sik‹ an, nahm 

vom Lande ›Ko–hu‹ Besitz, 

vernichtete 濮達 ›P 'ak–tat‹ 

( Baktar, Baktra, Baktria) 

und ›Ke–pin‹ und setzte sich 

in den vollständigen Besitz 

aller dieser Reiche. 

Als ›K'u–tsiu–k'iok‹ über 

achtzig Jahre alt war und 

starb, trat sein Sohn 閻膏 

珍 ›J¥m–ko–tin‹ (Wima Kadphises) 

an seiner Stelle als 

König auf; dieser vernichtete 

auch noch 天竺 ›T'i¥n– 

tok‹ (Indien) und setzte dort 

als Verwalter und Befehlshaber 

einen Heerführer ein. 

Seither befand sich Groß- 

Goat–si im allerhöchsten 

Stadium von Reichtum und 

Blüte. Die Reiche nannten 

es allgemein das Königreich 

›Kui–song‹. 

Aber ›Han‹ nannte es bei 

seinem ursprünglichen alten 

Namen Groß-Goat–si. 

migrated to Ta–hsia (Bactria). 

They divided the 

country among five ›hsi– 

hou‹ (›yabgu‹) — 

Hsiu–mi, Shuang–mi, 

Kuei–shuang (Kushan), 

Hsi–tun and Tu–mi 

(Tarmita). 

Over one hundred years 

later the Kuei–shuang 

›hsi–hou‹, Ch’iu–chiu–chü 

(Kujøla Kadphises), attacked 

and overthrew the 

(other) four ›hsi–hou‹ and 

set himself up as king. 

The country was named 

Kuei–shuang. 

The king attacked An–hsi 

(Parthia) and took the territory 

of Kao–fu (Kabul). 

He also overthrew P ’u–ta 

(Pu•kalåvatð) and Chi–pin 

(Kashmir) and completely 

annexed these countries. 

Ch’iu–chiu–chü died when 

over eighty years of age. 

His son Yen–kao–chen (Vima 

Kadphises) succeeded 

him as king. 

He went on to overthrow 

T’ien–chu (India) and sent 

a general to rule over it. 

From this time on the 

Yüeh–chih were extremely 

wealthy and prosperous. 

All the nations call them 

the Kuei–shuang (Kushan) 

kings. 

But Han, keeping to the 

original name, calls th 

the Great Yüeh–chih. 

初月氏為匈奴所滅 

遂遷於大夏分其國 

為 

休密雙靡貴霜漈盻 

頓都密 

凡五部漈翎侯 

後百餘歲貴霜翎侯 

丘就卻攻滅四翎侯 

自立為王 

國號貴霜 ( 王 ) 

侵安息取高附地 

又滅濮達罽賓悉有 

其國 

丘就卻年八十餘死 

子閻膏珍代為王 

復滅天竺置將一人 

監領之 

月氏自此之後最為 

富盛 

諸國稱之皆曰貴霜 

王 

em 漢本其故號言大月 

氏云 

This text has extraordin arily important facts to teach us. At the outset of the Later 

Han, i.e. after the year 26 C E, the capital of the Ruzhi is still in Tochara: the “Lan–shi 

city” 藍氏城 of the Hou H anshu is obviously identical with 

the “Lan–shi city” 藍市城 

of the Shiji, there called the capital of Daxia (see bel ow, p. 72) — whether it is to be 

identified with modern Khulm (Tashqurghan), as PULLEYBLANK thinks, or modern Kunthree 

prominent Bactrian cities), 

duz (i.e. Darapsa, mention ed by Strabo as one of only 

as I believe, is of secondary importance. Of prime imp 

ortance is, that the Ruzhi 月氏 

of this time in the west border on Parthia, (the capi 

tal of which is) 49 days journey 

from Lanshi. There is no mention 

of any independent Saka kingdom in that direction. 

The Hou Hanshu goes on to report the establishment of the five ›xihou‹ (still read 

yap–hau in modern Cantonese, this is the Chinese transcription of the title ›yabgu‹) — 

a historic event which happened, no t in La ter Han, bu 

t in Former Han times. Strictly 

speaking, this report does not belong into the Hou Hanshu. But it is obvious here that 

Fan Ye 范嘩 (398–446), the compiler of the third Chinese Standard History, the Hou 

— 53 —

usurper Kadphises I origin 

ally started his conquests south 

of the Hindukush Mountains, as BOPEARACHCHI has shown from strictly numismatic 

evidence (1997: 208), or else from north of the Hindukush — even north of the Oxus Ri- 

ver —, as GRENET has shown in a very recent article on the grounds of much new evi- 

topic just mentioned. It is in 

dence (2007: passim), is part of the same hot and complex any case too big for the modest scope of this paper and calls for a lengthy discussion 

in a separate study. 

But why did Zhang Qian not describe the Sakarau 

kai or Saiwang 塞王 as reigning 

in Bactra ? This city was just a few dozen miles away 

from the provisional Daxia/To- 

chara capital on the banks of the Oxus where he discussed politics with the new Ruzhi 

月氏 king in the latter’s royal tent. As VON 

G UTSCHMID has pointed out correctly: it is 

impossible that Zhang Qian overlooked the Sakaraukai/Saiwang 塞王 . Hence, I think 

he did, in fact, mention the Saiwang — but in a more indirect way. Zhang Qian’s de- 

tailed Report on the Parthians, in Shiji 123, also contains the following paragraph — 

Hanshu, wants to correct Ban Gu’s Hanshu where the 

story is told in a different way. 

He then continues the story to its climax. This, in fact, is a highly controversial topic. 

What interests us here, is the fact that more than a hundred years later — we are 

back in Later Han times now — king Kujula Kadphises finally dares to attack Parthia, 

i.e. he conquers three Parthian possessions: Gaofu 高附 , Puta 濮達 , and Jibin 罽賓 . 

We may assume, in this order. 

Puta is Bactra, as only DE GROOT correctly states in his translation of the passage. 

He reads the two Chinese characters 濮達 as ›P ’ak–tat‹ = Baktar. As far as I can tell, 

this is the first time the name Bac tra is mentioned in the Chinese Standard Histories. 

This is a strong argument in favor of a very late conquest of Saka / Parthian-occupied 

Bactra by the Ruzhi 月氏 — which conquest, in fa ct, we find depicted in the wall 

friezes of Khalchayan, the 月氏 summer palace, up t he Surkhan River valley a short 

distance from modern Termez (see above, p. 39–40). One large figure there, with the 

well-known face of the Heraios coins, represents the founder of the Kushan Dynasty — 

and the trophy of a dead Saka warrior’s heavy plated armor at his feet is telling (see 

Puga¾enkova 1971: fig. 61). 

Whether the Ruzhi 月氏 

(WATSON 1993: 235) 

(71) The people are very numerous and of- 

Although the (ruler of) An–xi gives orders to 

ten have (are ruled by) petty princes. 

these dependent states, he regards them as 

foreign countries. 

Shiji 123. 3163 

人眾甚多往往有小君長 

而安息役屬之以為外國 

Zhang Qian could have 

heard of such Parthian vassal states as Characene, Elymais, 

or Persis. But more likely it is that he is speaking here of the state of affairs in close-by 

Bactra, ruled by petty Saka princes — whose descendents, five or six generations later, 

were burried on Tillya-tepe. For Zhang Qian, the Saiwang thus were local princes within 

the mighty Parthian kingdom 最大 ( 安息 ) 國 — which could rival the Roman Empire 

— and thus the Saiwang were just not worth mentioning by name. 

In this context, VON GUTSCHMID writes, 1888: 56–57: 

Diesen kleinen Königreichen gegenüber begnügten sich die Parther mit der Anerkennung 

ihrer Suprematie ... Waren die Parther in guter Verfassung, so zogen sie die Zügel 

straffer an, waren sie aber durch innere Unruhen geschwächt, so war ihre Suprematie ein 

leerer Name, und Alles gieng darunter und darüber; das Verhältnis blieb immer ein sehr 

prekäres, und die auf einer derartigen Basis beruhende Machtstellung der Parther war 

entfernt nicht die gleiche wie später die des Sassanidenreiches. 

When in the first century BCE, or in the later time of the Former Han, the one-time 

petty princes of Western Bactria had become great rulers in their own right in Northwestern 

India or Jibin 罽賓 , their historic status had changed dramatically. And so the 

— 54 —

Chinese, politically deeply involved in India, reported on the Saiwang 塞王 for the first 

time and recapitulated their earlier history in Hanshu 96 — to all historians’ delight. 

Without this historical source, we would not know of the short spell of Saka rule over 

the Tochari. However, on the role of the Sakas in Bactra — and Sakastana — the Chinese 

historians of later times remain practically silent. I presume that it escaped their 

attention because it was outside their geographic horizon. We find only very vague 

hints on the Saiwang of Sakastana in the Hanshu. 

One more quote from Trogus in Justinus’ Epitome, less puzzling in the light of the 

present findings and thus open for new interpretations, I like to reproduce here. 

(WATSON 1886: 277–278) 

After the death of Mithridates, 

king of the Parthians, Phraates 

his son was made king, 

who, having proceeded to make 

war upon Syria, in revenge for 

the attempts of Antiochus on 

the Parthian dominions, was 

recalled, by hostilities on the 

part of the Scythians, to defend 

his own country. 

For the Scythians, having been 

induced, by the offer of pay, to 

assist the Parthians against 

Antiochus, king of Syria, and 

not having arrived till the war 

was ended, were disappointed 

of the expected remuneration, 

and reproached with having 

brought their aid too late ... 

being offended at the haughty 

reply which they received, 

they began to ravage the 

country of the Parthians. 

Phraates, in consequence, 

marching against them ... 

took with him to the war a body 

of Greeks, who had been 

made prisoners in the war 

against Antiochus, and whom 

he had treated with great pride 

and severity ... 

As soon therefore as they saw 

the Persians giving ground, 

they went over to the enemy, 

and executed that revenge for 

their captivity, which they had 

long desired, by a sanguinary 

destruction of the Parthian ar- 

and of king Phraates him- 

my 

self. 

In his stead Artabanus, his 

uncle, was made king. 

The Scythians, content with 

(SEEL 1972: 445–446) 

Nach dem Tode des Mithridates, 

des Partherkönigs, wird sein 

Sohn Phrahates zum König eingesetzt; 

als dieser zur Rache für den von 

Antiochos versuchten Angriff auf 

das Partherreich einen Krieg gegen 

Syrien zu unternehmen beschlossen 

hatte, wird er durch 

Unruhen bei den Skythen zurückgerufen, 

um dort seine eigenen 

Interessen zu wahren. 

Die Skythen nämlich waren zur 

Unterstützung der Parther gegen 

den Syrerkönig Antiochos um 

Lohn aufgeboten worden, trafen 

aber erst ein, als der Krieg bereits 

zu Ende war, und sollten 

unter dem Vorwurf, sie seien bei 

der Hilfeleistung allzu saumselig 

gewesen, um ihren Lohn geprellt 

werden ... 

Darüber beleidigt, fingen sie an, 

das Gebiet der Parther selbst zu 

verheeren. 

Als nun Phrahates gegen diese 

aufbrach ... 

führte (er) ein Heer von Griechen, 

das er im Kriege des Antiochos 

gefangen und inzwischen so hoffärtig 

wie grausam behandelt 

hatte, mit sich in den Kampf ... 

Als sie daher sahen, daß die 

Front der Parther ins Wanken 

geriet, gingen sie mit ihren Waffen 

zum Feinde über und vollstreckten 

die langersehnte Rache 

für ihre Gefangenschaft 

durch ein Blutbad am Partherheer 

und am König Phrahates 

selbst. 

An seiner Stelle wird Artabanus, 

ein Onkel väterlicherseits, als 

Hist. Phil. Epit.42.1.1-2.5 

Post necem Mithridatis, 

Parthorum regis, Phrahates 

filius rex statuitur, 

qui cum inferre bellum 

in ultionem temptati ab 

Antiocho Parthici regni 

Syriae statuisse, Scytharum 

motibus ad 

sua defendenda revocatur. 

Namque Scythae in auxilium 

Parthorum adversus 

Antiochum, Syriae 

regem, mercede 

sollicitati cum confecto 

iam bello supervenissent 

et calumnia tardius 

lati auxilii mercede 

fraudarentur ... 

Superbo responso ofadversus 

eos proficisce- 

fensi fines Parthorum 

vastare coeperunt. 

Igitur Phrahates, cum 

retur ... 

exercitum Graecorum, 

quem bello Antiochi 

captum superbe crudeliterque 

tractaverat, in 

bellum secum ducit ... 

Itaque cum inclinatam 

Parthorum aciem vidissent, 

arma ad hostes 

transtulere et diu cupitam 

captivitatis ultionem 

exercitus Parthici 

et ipsius Phrahatis regis 

cruenta caede exse- 

cuti sunt. 

In huius locum Artabanus, 

patruus eius, rex 

substituitur. 

— 55 —

Scythae autem contenti 

victoria depopulata Parthia 

in patriam revertuntur. 

Sed et Artabanus bello 

Tocharis inlato in brac- 

chio vulneratus statim 

decedit. 

Huic Mithridates filius 

succedit, cui res gestae 

Magni cognomen dedere 

... 

Sed et cum Scythis 

prospere aliquotiens dimicavit 

ultorque iniuriae 

parentum fuit ... 

their victory, and with having 

laid waste Parthia, returned 

home. 

Artabanus, making war upon 

the Thogarii, received a wound 

in the arm, of which he immediately 

died. 

He was succeeded by his son 

Mithridates, to whom his 

achievements procured the 

surname of Great ... 

He fought successfully, too, several 

times, against the Scythians, 

and avenged the injuries 

received from them by his forefathers 

... 

König eingesetzt, 

Die Skythen jedoch kehrten zufrieden 

mit ihrem Sieg und nach 

Verwüstung des Partherlandes 

in ihre Heimat zurück. 

Aber auch Artabanus, der bei einem 

Krieg gegen die Tocharer 

am Arm verwundet wurde, starb 

sogleich. 

Ihm folgte sein Sohn Mithridates, 

dem seine Taten den Beinamen 

› der Große‹ verschafften ... 

Aber auch mit den Skythen focht 

er einige Male mit Glück und 

wurde so zum Rächer des den 

Ahnen angetanen Schadens ... 

Here we read about developments which take place soon after Zhang Qian’s depar- 

conquered by the 

ture. The Sakaraukai/Saiwang 塞王 , Sakas or Scythians for short, Parthians to prevent the Ruzhi 

月氏 from doing the same, had thus been forced to ac- 

II, the Arsakes 

knowledge Parthian suzerainty. They are soon enlisted by king Phraates 

Theopator Euergetes of the coi ns, as mercenaries in the latter’s crucial war against the 

Seleucids of Syria. The Sakas, still unsettled and restless in the narrow confines of 

their new “patria,” or fatherland, gladly jump to the chance to 

venture beyond it — 

but create grave problems for their Parthian overlords when the latter try to cheat 

them. With the help of Greek arms, they defeat the Parthians and king Phraates is 

slain in battle, about 127 BCE. 

In the end, the Sakas will in turn be defeated by king 

Mithridates II, ultor iniuriae p arentum, who decides to solve the Saka problem by 

settling the nomads for good in Drangiana — the country which eventually will become 

known by the name Sakastana (modern Sistan). 

All this we understand so much better when we realize that the later Saka-Parthian 

pell-mell evolved from Bactra. After 130 BCE, the Sakaraukai/Saiwang 塞王 , we may 

note down now, reigned in Western Bactria as petty princes and Parthian vassals for 

150 years and more. 

In between all of this, we are astonished to read in Justinus’ Epitome that Phraates’ 

successor, king Artabanos, the Arsakes Theopator Nikator of the coins, resolves to de- 

this would have 

stroy the dangerous newcomers and attacks the “Tochari.” Obviously, been impossible had an independent Saka state separated the two nations. Of course, 

these Tochari are no longer the timid people of Daxia 大夏 (Tochara) Zhang Qian in 

his Report had so described an d stated that he had found them with no king nor good 

soldiers. These new Tochari are now firmly governed by an East Asian nomad aristoa 

second shock. Arta- 

cracy and are boasting a fearful army. The Parthians are in for 

banos, too, is killed in battle — wounded by a poisonous arrow, he dies on the spot. 

This occurred in about 124 BCE. 

The incident proves that the Parthians had a direct common border with the Asiafrom 

north to south. 

ni/Tochari 月氏 — which ran right across former Greek Bactria 

Both sides, it seems, were unable to dislodge the other for another century and a half 

or more. In this long time the new Ruzhi 月氏 kingdom, north 

of the Hindukush, is 

confined to Daxia/Tochara. It is in this century and a half that the 月氏 become 

known by the name Tochari — it is the Far Eastern Ruzhi 月氏 who make this 

Central Asian name great and famous : as the designation of a country, of a people, of 

a language. 

When we can prove that the Tochari of Trogus are the Daxia 大夏 of Zhang Qian, it 

then becomes apparent that these Tochari were the indigenous population of Tochara, 

— 56 —

well settled on the land since centuries. They spoke their own language, Tocharian. 

But as the Tochari were constantly ruled by foreign invaders ever since the Achae- 

and finally the Ru- 

menids, these foreigners — Persians, Macedonians, Greeks, Sakas, zhi 月氏 — necessarily left their mark on the local language to varying degrees. Such 

new insights suggest new answers to question we find in the following text. 

In 1995: 439–441, RINGE writes: 

Fragmentary manuscripts, found at various sites along the northern arm of the silk road 

in the Tarim basin and dating from perhaps the 7th through the 10th 

centuries of the com- 

called “A” and “B.” 

mon era, preserve documents written in two Tocharian languages, Though it is clear that Tocharian A and B are separate languages (not dialects of a single 

language), it is also clear that t hey are very closely related; and it follows that they must 

still have been a single language (called “Proto-Tocharian”) until about a millennium or so 

before the date of our earliest documents. Historical and comparative analysis of the Tomany 

details remain 

charian language reveals the rough outlines of their prehistory, but 

unrecoverable ... 

The earlier Iranian loanwords in Tocharian seem to have been borrowed from lan- 

at an “Old Iranian” stage of development, but it is not easy to specify which Iran- 

guages 

ian language they were borrowed from; they resemble Avestan words, but that may be 

simply because Avestan is the most archaic Iranian language of which we have any record. 

The loanwords of the next oldest stratum are startlingly similar to Ossetic, the Iranian 

language now spoken in the Caucasus. That is especially intriguing because it is reasonably 

clear that Ossetic is descended from one of the Iranian languages, collectively called 

“Scythian,” that were spoken by nomads north of the Black Sea and in the steppes to the 

east during the last few centuries BCE. This suggests that the Tocharians could have been 

affiliated with one or another of the steppe confederations dominated by Iranian tribes at 

a relatively early period of their prehistory. Unfortunately that is all that can be said; neither 

the nature of such an affiliation nor the time and place at which it may have existed 

can be specified with any certainty. 

We are on firmer ground with the next clearly identifiable stratum of Iranian loanwords, 

which were borrowed from Bactrian. Bactrian was the Iranian language of the Kushåna 

kingdom, which flourished in Afghanistan and neighboring areas beginning in the first 

century of the common era. But while the fact of a relationship between the Kushåna kingdom 

and the Tocharians is tolerably clear, the nature of that relationship is (again) unrecoverable. 

If the Tocharians were Kushåna subjects, they must then have been living well 

to the southwest of where we find them historically; but it seems much more likely that 

they were outlying allies, or even trading partners, and in that case they could well have 

borrowed elements of Kushåna language and culture from a considerable distance. Later 

strata of loanwords — Sogdian, Khotanese and Sanskrit — can easily have entered the Tocharian 

languages after the Tocharians reached their historical home ... 

One year later, in 1996: 633, SIMS-WILLIAMS writes on the Kushan language : 

Durant les premiers siècles de notre ère, le bactrien aurait pu légitimement être compté 

au rang des langues les plus importantes du monde. En tant que langue des rois kouchans, 

il était certainement largement compris sur tout le territoire d’un vaste empire incluant 

l’Afghanistan actuel, le Nord de l’Inde et une partie de l’Asie centrale. Même après la chute 

de l’empire kouchan, le bactrien continua d’être écrit pendant au moins six siècles, comme 

en témoignent les inscriptions du IX e siècle trouvées dans la vallée de Tochi au Pakistan, 

ainsi que des fragments de manuscrits bouddhistes et manichéens trouvés dans la lointaine 

oasis de Turfan située à l’ouest de la Chine. Ainsi, sa carrière comme langue de culture 

a duré près de mille ans. 

These brilliant remarks teach us that the language of the Kushans, or, more generally 

speaking, of the Ruzhi 月氏 , played a paramount role in the spread of Buddhism 

— similar, in fact, to the importance of the language of the Romans in the spread 

— 57 —

of Christianity. After the initial Sanskrit, the imperial language of the Ruzhi 月氏 apparently 

became the Vulgata of Buddhism in the centuries and regions outlined by 

SIMS-WILLIAMS. The only amendment, I would venture to make here, is to the name of 

this language : with the above new findings of the Ruzhi 月氏 being confined to Tochara/Tachara/Daxia 

大夏 for a decisive one century and a half, the correct name is, not 

Trogus’ Reges Toc 

harorum Asiani = 月氏既臣大夏而居 of Sima Qian. 

(The kings of the Tochari were of the tribe of the Asii = the Ruzhi had since subjugated the 

Daxia and were then living there). 

The parallel text in Hanshu 61 has the same sentence a bit more to the point: 

Trogus’ Reges Tocharorum Asiani = 月氏既臣大夏而君之 of Ban Gu. 

(The kings of the Tochari were of the tribe of the Asii = the Ruzhi had since subjugated the 

Daxia and were then ruling them). 

In fact, the closest parallel to the three-word-phrase by Trogus I find in six charac- 

of Hanshu 96A. 3884 by Ban Gu (quoted above already): 

ters 

ASIANI reges TOCHARORUM 

RUZHI (in the West) rulers of DAXIA 

月氏 

( 西 ) 君大夏 

When, as I have been at pains to show, the Daxia 大夏 of the Chinese sources are 

the Tochari / TÒcaroi of the Western sources, and when Trogus and Strabo in the West 

a s well as Sima Qian and Ban Gu in the East are writing about the same historical 

e vents, which took place at the same time and in the same geographical area, then it 

follows that the rulers of the Daxia/Tochari were : 

the ASIANI (Trogus) = the ACIOI (Strabo) = the 月氏 (Zhang Qian). 

In 1922, the well-known Russian Orientalist BARTHOLD is the first Western scholar 

to state, originally in his mother tongue, later translated into English: 

Bactrian, but Tocharian. Under this name, we find this language mentioned in faraway 

Turfan texts — long after the last Kushan emperors had vanished. 

In this study, special emphasis is laid on bringing together both Eastern and Wes- 

classical sources for comparison. These sources complement each other greatly 

tern 

and together they help us to reconstruct the end of the Greek kingdom of Bactriana, 

tÁj sump£shj 'ArianÁj prÒschma, or the pride of all of Ariana, as Apollodoros of Aremita 

calls it. His lost Partik£, or Parthian History, must have been the source for 

t 

b oth Strabo and Trogus — contra Tarn (see below, p. 74). If this assumption is correct, 

we get the following important equation from the texts of Trogus 42 and Shiji 123: 

( MINORSKIJ 1956: 4) BARTHOLD 1922: 5 

In the middle of the second century B.C. the В середине II в. до Р.Х. произошло завоева- 

northern, and later the southern provinces ние северных, впоследствии и южных, облаистории, 

о котором говорят как западные 

of the Graeco-Bactrian kingdom were con- 

by Central Asian nomads, who in атскими кочевниками, завоевавшими потом 

стей греко-бактрийского царства средне-ази- 

quered 

due course subdued several Indian provin- 

and became known in Greek literature ми в греческой литературе под общим назва- 

также некоторые области Индии и известны- 

ces 

under the general appellation of Scythians. нием скифов. 

In the same century relations between Cen- 

Asia and China were established for Средней Азией и Kитаем; падение греко-бак- 

В том же II веке начались сношения между 

tral 

the first time. The fall of the Graeco-Bac- 

kingdom is the first event of world 

трийского царства – первое событие мировой 

trian 

history recorded both in Western (Greek) (греческие), так и дальневосточные (китайские) 

and Far-Eastern (Chinese) sources. 

источники. 

— 58 —

Well said. It is, however, not a Greek, but a Roman source which we have to combine 

with the Chinese texts: Trogus with Sima Qian / Ban Gu. The fact that Greek and 

Latin as well as Chinese historians had recorded the fall of the Greek kingdom in Bactria 

has been known in the West since two generations or longer. It is therefore a 

curious fact that no one so far ever tried to combine the statement by Trogus with that 

by Ban Gu in order to get the very simple and very obvious equation Asiani = 月氏 . 

Actually, I gladly correct myself: there is one who did. HERZFELD, 1931: 26–27, writes: 

Die beiden abendländischen Nachrichten über das Ende des graeco-bactrischen Reichs 

sind 1. Justin, Trog. Pomp. prol. XLI: 

qua re pugnantes [recte: quo regnante] Scythicae gentes Saraucae et Asiani Bactra 

occupavere et Sogdianos. 

Dazu prol. XLII: 

res Scythicae: reges Thocarorum Asiani interitusque Saraucarum. 

Dies Scythae bedeutet natürlich nicht sak-Sakå, sondern asiatische Nomaden. Nach 

de m Parallelismus der Stelle besetzen die Saraucae Bactra, die Asiani Sogdiana. 

Sogdiana haben aber in eben jenen Jahren nach ²ang-k’ien die ýuÿt-šï besetzt. 

Danach würden Asiani und ýuÿt-šï gleich sein. 

Die 2. Stelle ist Strabon XI, C 511: 

malista de gnwrimoi gegonasi twn nomadwn oƒ touj Èllhnaj afelomenoi thn Baktrianhn, 

Asioi kai Pasianoi kai Tocaroi kai Sakarau kai / loi, Ðrmhqentej apo thj peraiaj tou Iaxartou 

thj kat¦ Sakaj kai Sogdianouj, ¹n kateicon Sakai. 

Dieser Gruppe von Nomaden entsprechen bei Ptolemaios VII, 12 die Paskai, Iatioi, Toca 

roi und Augaloi ... Diese Zusammenhänge erheben die von F.W.K. MÜLLER 1918: 566 ff. 

angedeutete Vermutung zur Wahrscheinlichkeit, daß chin. ýuÿt-šï, das eine Aussprache wie 

uår-śi zuläßt, die Selbstbe zeichnung der Tocrï: årÝi wiedergibt, vgl. SIEG, 1918: 560 ff. Auf 

jeden Fall machen diese Zusammenh änge eine Trennung der Tocharer von den ýuÿt-šï unmöglich. 

Justin’s “reges Thocarorum Asiani“ besagt, daß das Königsgeschlecht der 

Asiani-Asioi-Iatioi-ýuÿt-šï über die Tocharer herrschte ... Hält man sich streng an den 

Wortlaut der kurzen griechischen und chinesischen Nachrichten, so ist der Sachverhalt 

ganz klar. Noch vor 160 müssen die ýuÿt-šï-årÝi-Asioi vor den wu-sun aus Farghåna 

weichen, aus dem sie zuvor die sak vertrieben haben. Sie besetzen das südliche Sogdiana, 

wo ²ang-k’ien sie i. J. 127 trifft. Die mitgewanderten Saraucae sitzen in dem unterworfenen 

Bactria = tai-hia. Das bedeutet das Ende des graeco-bactrischen Reichs ... 

As early as 1931, HERZFELD has the equation ýuÿt–šï = årÝi = Asioi. This name 

“årÝi” seems to come closest to how the Ruzhi 月氏 may have called themselves: “die 

Selbstbezeichnung der Tocrï” (here to be understood as: the self-appellation of the 

kings of the Tochari). The first and the third name are the Western and the Eastern 

transcription of the second: a(r)si i årÝi (a)ruzhi 月氏 . In this, the original årÝi 

is closer to both asii and ruzhi 月氏 — than asii is to ruzhi 月氏 . 

Why was this ingenious equation not accepted ? One raison may be that HERZ- 

gross mistakes: 

FELD’s intelligent discovery is lost in too many — the Asiani are not a dynasty, but a people; 

— they were not displaced from Ferghana, but from the Ili River; 

— they did not occupy southern Sogdiana, but the whole of Sogdiana; 

— Zhang Qian did not reach the Asiani in 127, but in 129 BCE; 

— the Sa(ca)raucae did not migrate together with the Asiani, but always ahead of them; 

— and finally: ›tai-hia‹ (Daxia) is not Bactria, but Tochara. 

Another reason surely was that HERZFELD is unable to substantiate the Chinese 

sources by quoting the relevant passages verbatim — original text with translations —, 

i.e. as emphatically as he does quote the Latin and Greek sources. And this, of course, 

was so because, from BROSSET in 1828 to WATSON in 1993, all Western translations of 

Shiji 123 were publisheed without the Chinese text. With one sole exception, I hasten to 

add: HIRTH, in 1917, added a handwritten edition of Shiji 123 to his pioneering transla- 

— 59 —

tion. The whole would have been perfect had he arranged translation and text in parallel 

columns — and the hanzi, or Chinese characters, two times larger. Any Chinese 

text should be given as much space as is allotted to the translation: this usually means 

that a Chinese character, representing a full word, should be printed about twice as 

large as a letter of the Western alphabet. 

Just like in HERZFELD’s long article, we can find innumerous instances in other 

Western literature where Greek and Latin historical sources are very carefully com- 

— resulting for instance in the equation Asiani = ”Asioi. But amongst this good 

pared 

number of authors, there is not one who includes the Chinese historical sources ad- 

i.e. Sima Qian and Ban Gu as prominently as Trogus and Strabo. 

equately, 

BAILEY, 1936: 912, inserts the following telling note in his text: 

Speculations on these ”Asioi Asiani are at present of little use. We cannot be sure from 

the Greek and Latin texts whether the Asiani were kings of the Tochari before or after their 

settlement in Tokhåristån. 

Here BAILEY subscribes to the equation Asiani = ”Asioi; he also knows that the 

Asiani settled in Tocharistan — and not in Bactria: so one is inclined to infer from 

h is simple statement. But a few pages before, 1936: 887, the author writes: 

Tokhåristån ... the capital was Balkh ... It was therefore the old Bactria. The evidence is 

furnished by Arabic, Armenian, Tibetan and Chinese sources. It is now not disputed. 

If BAILEY had known the Chinese sources as well as he did the Greek and Latin 

tex ts, he would have been assured by Zhang Qian apud Sima Qian that the Ruzhi 月氏 

no doubt became the rulers of the Daxia 大夏 after they had settled in the country of 

t hese Daxia, and surely not before. And if the same eminent Iranist had been familiar 

w ith the intricacies of the ways in which the ancient Chinese transcribed foreign names, 

h e quickly would have jumped to the conclusion that the Ruzhi 月氏 were the Asiani, 

or ”Asioi, and the country of the Daxia 大夏 Tochara, the later Tocharistan. He might 

then have pondered the question whether Lanshi city 藍市城 , said to be the capital of 

Daxia, could really be Bactra, or Balkh, as the latter was clearly located outside Tocha- 

Phonetically it is in any case closer to (Da)rapsa, mentioned by Strabo 

ra/Tokhåristån. 

to be one of the three largest and best-known cities of Bactriana (see below, p. 78). 

The earliest Histories of Han China, the Shiji and the Hanshu, exist since about two 

thousand years. They have been translated into Western languages since close to two 

hundred years. To make proper use of these texts from a Western point of view — 

does this still involve bridging a gap too wide for a single individual ? 

With the above clarifications, the next important step will be to understand that it 

was the policy of the Ruzhi 月氏 = Asii / Asiani to hide behind the peoples they conquered 

— or should I say: it was there policy to identify themselves with the peoples 

they conquered. As these subjugated nations, one after the other, were superior not 

only in numbers, but first and foremost in the sophistication of their culture, the 月氏 

were shrewd enough not to enforce their own coarse cultural background upon their 

subjects. In the contrary: they chose to amalgamate with them and take over what sophistication 

they encountered in the West. In one word: the Far Eastern 月氏 , who 

had been neighbors of ancient China from times immemorial, in Central Asia quickly 

became the new, or pseudo-, Tochari. This was their rather intelligent way to slip into 

a Western identity. This transformation has not only baffled and confused modern research 

on the Ruzhi 月氏 for a long time, this ingenious policy has in fact mislead peoples 

of their own times as well. 

After their long, intermittent stop-and-go migration, at the beginning of which the 

Ruzhi 月氏 had no clear notion where it would end, this mongoloid nation of cattle 

and horse breeders — as well as accomplished traders — finally showed up in a totally 

new world. They had crossed from one self-centered Oikumenê into another. Before 

the 月氏 , no nation in antiquity has done this. To be sure, no nation has achieved this 

— 60 —

in such an abrupt way. In the Western Oikumenê the 月氏 surely looked different, if 

not downright odd, and they spoke a language unintelligible for anyone — except maybe 

Zhang Qian who must have been fluent, not only in Chinese, but also in the 

language of the Xiongnu 匈奴 , the one-time close neighbors of the 月氏 . The language 

of the Xiongnu may have been close enough to that of the 月氏 for the two peoples to 

understand each other fairly well. The original language of the 月氏 , then, was everything 

but Indo-European. 

When in 1971, at age 31, I stepped from a Soviet ship onto Japanese soil in Yokohama 

横浜 , I became a six-year-old illiterate boy on the spot: it was a weird experience 

which I shall never forget. For the Ruzhi 月氏 , their culture shock first in Sogdiana, 

then in Bactria, must have been immensely greater. But they liked very much what 

they saw in civilized Central Asia in general, and south of the Hissar Mountains in paricular 

— as Zhang Qian noticed to his great chagrin. And of course, the 月氏 liked the 

t 

fact that they were now the masters of all these developed regions in this New World. 

When they had finally conquered Daxia 大夏 (Tochara) in 130/129 BCE and were about 

to settle in that land, they considered it greatly superior to any place of the various 

lands they had lived in since evacuating their old pasture grounds in the Hexi Corridor 

(modern Gansu) around 165 BCE. Daxia/Tochara was populous, well developed, had 

ple nty of land under the plough; it also had great cities with markets supplying the 

most luxurious goods. In the upheavals of the past decades, the Daxia/Tochari had lost 

the last two of their former rulers: first the Greeks, their colonial masters since Alexander’s 

conquests (some two hundred years previous), and after them the Sakaraukai/Saiwang, 

who had been their kings for less than the span of a generation. 

The Ruzhi 月氏 were now quickly filling the regal void in Daxia/Tochara. It is the 

月氏 who made the name Tochari great and famous — deliberately becoming Tochari 

after the conquest of Tochara. The strangers from the Far East were wise enough to 

adopt the local language — they could hardly do anything else —, and this local 

language was not Greek, but the language of the Tochari. Hence, the proper name for 

that language is Tocharian. Together with the Græco-Bactrian kingdom the name Bactria 

had disappeared. Only later Greek authors like Ammianus Marcellinus (or before 

him the unknown author of the Periplus), out of habit, go on to call the people of Tochara 

under their Asiani/Ruzhi 月氏 kings Bactrians. As a matter of fact, these authors 

were perfectly correct, for when the unnamed Greek merchant wrote down his 

trader’s guidebook in the Alexandria of Egypt, c. 50 CE, the Ruzhi 月氏 had just recently 

— finally — conquered all of former Greek Bactria — and the feat had earned 

them the epithet “warlike.” A good one hundred years later, however, and in the same 

Alexandria, Ptolemy splashes the name TÒcaroi in a number of variants all over his 

Northwest Indian and Central Asian map. The Asiani/ÅrÝi/Ruzhi have become Tochari/TÒcaroi/Tuchåra; 

they are called wealthy and powerful by any author in East and 

West from now on. 

The implications of this should be clear. It shows that HENNING, for one, is mistaken 

when in 1960: 47 he writes: 

... the genuine TÒcaroi, who, coming from Kan-su, conquered Bactria in the second 

century B.C. and caused that country to be renamed. The invaders, as is often the case, 

adopted the native language in the course of time. 

With his second statement, HENNING is clearly contradicting his first. Either the Ruzhi 

月氏 were the true TÒcaroi and brought their Tocharian language with them 

which they then enforced upon the people they conquered north and south of the 

upper Oxus. Or they adopted the native language, i.e. the language of the people they 

conquered, but then they themselves cannot have been the genuine TÒcaroi. For our 

lone trusted eyewitness, the Chinese ambassador Zhang Qian, tells us in his Report, 

epitomized by Sima Tan and Sima Qian in Shiji 123, that the Ruzhi 月氏 conquered 

Daxia 大夏 — which is the Chinese transcription, not of Bactria, but of Tochara. This 

— 61 —

all-important correction of a long-standing mistake is the result of this careful study of 

all relevant texts in the Chinese Standard Histories. Had HENNING been aware of this 

correction, he would have grasped himself that something was essentially wrong with 

his above statements. 

Incidently: as we will see below (p. 91), the oldest extant codex of Strabo’s Geography 

has, not TÒcaroi, but TACAROI. This seems to be the older version of this ethnicon. 

It is closely paralleled by the Chinese transcription Daxia 大夏 . The later name 

TÒcaroi is given in the Chinese texts as Tuhuoluo (Tu-ho-lo) 吐火羅 and a number 

of variants. I wonder: do we hear about the Tachari/Tochari from any classical writers 

before 129 BCE ? Such early sources would be of great help to clarify who the genuine 

Tochari were and where they lived in the third and second century BCE. Before 129 

BCE, the Tochari must not be confounded with the Ruzhi 月氏 . 

Above, I have quoted one important difference in one particular paragraph of Hanshu 

61 when compared with the parallel, i.e. the original, sentence in Shiji 123. It may 

be helpful, therefore, to reproduce the opening paragraphs of both chapters here, 

marking the instances where they differ by blank spaces or bold characters. 

Shiji 123. 3157–3159 

(1) 大宛之跡見自張騫 

(2) 張騫漢中人 

(3) 建元中為郎 

(4) 

(9) 

是時天子問匈奴降者 

皆言匈奴破月氏王以 


(5) 月氏遁逃而常怨仇匈 

奴無與共擊之 

(6) 漢方欲事滅胡聞此言 

因欲通使 

(7) 道必更匈奴中乃募能 

使者 

(8) 騫以郎應募使月氏與 

堂邑氏 ( 故 ) 胡奴甘 

父俱出隴西 

經匈奴匈奴得之傳詣 

單于 

(10) 單于留之曰 

(11) 月氏在吾北漢何以得 

往使吾欲使越漢肯聽 

我乎 

(12) 留騫十餘歲與妻有子 

Hanshu 61. 2687–2689 

(1) 

(2) 張騫漢中人也 

(3) 建元中為郎 

(4) 時匈奴降者 

言匈奴破月氏王以 


(5) 月氏遁而怨匈 

奴無與共擊之 

(6) 漢方欲事滅胡聞此言 

欲通使 

(7) 道必更匈奴中乃募能 

使者 

(8) 騫以郎應募使月氏與 

堂邑氏 

奴甘 

父俱出隴西 

(9) 徑匈奴匈奴得之傳詣 

單于 

(10) 單于曰 

(11) 月氏在吾北漢何以得 

往使吾欲使越漢肯聽 

我乎 

(12) 留騫十餘歲予妻有子 

— 62 —

(13) 然騫持漢節不失 

(14) 居匈奴中益寬騫因與 

其屬亡鄉月氏西走數 

十日至大宛 

(15) 大宛聞漢之饒財欲通 

不得見騫喜問曰 

(16) 若欲何之 

(17) 騫曰 

(18) 為漢使月氏而為匈奴 

所閉道 

(19) 今亡唯王使人導送我 

(20) 誠得至反漢漢之賂遺 

王財物不可勝言 

(21) 大宛以為然遣騫為發 

導繹抵康居 

(22) 康居傳致大月氏 

(23) 大月氏王已為胡所殺 

立其太子為王既臣大 

夏而居 

(24) 地肥饒少寇志安樂又 

自以遠漢殊無報胡 

之心 

(25) 騫從月氏至大夏竟不 

能得月氏要領 

(26) 留歲餘還並南山欲從 

羌中歸復為匈奴所得 

(27) 留歲餘單于死左谷蠡 

王攻其太子自立國內 

堂邑父 

(28) 

亂騫與胡妻及 

俱亡歸漢 

漢拜騫為太中大夫堂 

邑父為奉使君 

(13) 然騫持漢節不失 

(14) 居匈奴西騫因與 

其屬亡鄉月氏西走數 

十日至大宛 

(15) 大宛聞漢之饒財欲通 

不得見騫喜問 

(16) 欲何之 

(17) 騫曰 

(18) 為漢使月氏而為匈奴 

所閉道 

(19) 今亡唯王使人道送我 

(20) 誠得至反漢漢之賂遺 

王財物不可勝言 

(21) 大宛以為然遣騫為發 

譯道抵康居 

(22) 

康居傳致大月氏 

(23) 大月氏王已為胡所殺 

立其夫人為王既臣大 

夏而君之 

(24) 地肥饒少寇志安樂又 

自以遠遠漢殊無報胡 

之心 

(25) 騫從月氏至大夏竟不 

能得月氏要領 

(26) 留歲餘還並南山欲從 

羌中歸復為匈奴所得 

(27) 留歲餘單于死 

(28) 

國內 

堂邑父 

亂騫與胡妻及 

俱亡歸漢 

拜騫太中大夫堂 

邑父為奉使君 

— 63 —

Ban Gu is showing great respect for his a dmired predecessor. However, he is far 

from copying the older text verbatim as has often been claimed. Instead, Ban Gu is 

carefully editing Sima Qian — or Sima Tan 

who originally may have written this imthe 

quoted text Ban Gu : 

porta nt chapter, later to be extended by his son. In 

— deletes 12 words (Chinese characters) which he considers unnecessary; 

— replaces 5 words by similar ones; 

— adds 2 words which do not change the meaning; 

— drops data in four instances (considered irrelevant, but of value to us); 

— and provides us in three instances with valuable emendations. 

The correction in phrase (14) has been discussed above (p. 22); the one in (21) is of little 

importance; the first one in (23) is only half the truth as Ban Gu says that, when the 

king of the Ruzhi 月氏 had been slain by the Xiongnu, his widow the queen took 

over — whereas Sima Qian had said that the crown prince followed his father. The full 

truth is likely to be that the queen took over 

in 165 BCE from her dead husband and 

that her son, the crown prince, had become king some time before the 月氏 subjugat- 

The second correction in phrase (23) is the im portant one which I have cited above 

ed th e Daxia in 130/129 BCE — a full generation later. 

( p. 58). Here Ban Gu replaces the one word 居 by the two words 君之 and thereby 

is at pains here to clarify what Sima Qian ha d written. This clarification is important. 

For Sima Qian’s text translates 

... and (the 月氏 then) lived there (i.e. in Daxia) 居 , 

whereas Ban Gu gives us a valuable interpre 

tation of this somewhat vague statement 

by Sima Qian in that his text pointedly reads: 

... and (the 月氏 then) ruled over . 

them 君之 

大夏 (Tochara), the 月氏 were not 

In other words, after they had conquered Daxia 

s imply living amongst the Tochari (Shiji), bu t in fact ruled over them (Hanshu). This 

is ex actly what in imperial Rome Trogus — writ ing after Sima Qian, but long before 

Ban Gu — had formulated in just three terse words: 

... reges Tocharorum Asiani. The (Far Eastern) Asiani ( = 月氏 became) the kings of the 

(Central Asian) 

Tochari. 

This is how Trogus explained in his “Summa 

ry” (prologus) what he was going to 

tell his readers at greater length in chapter 42 o f his History — but the Epitome of 

Justinus has left us not one word of it. That of Trogus’ great work only his Prologi 

have come down to us is a tremendous loss, inde ed. As I have said already, it may just 

have been out of habit that the unknown Gree 

k author of the Periplus gives preference 

to the appellation Bactria, the time-honored nam e for the region between the Hissar 

and the Hindukush, when speaking of the Asiani/Ruzhi 月氏 . Towards the end of his 

l ifetime, in about the mid-first century CE, th is author writes down what he had seen, 

heard and experienced in a long life as a merchan t in a profitable, ocean-going trade in 

luxury goods — an invaluable source of first-hand 

information for us: 

( FABRICIUS 1883: 89) 

Es wohnen aber bei Barygaza 

in dem Binnenlande mehrere 

Völker — das der Arattier, der 

Arachusier, Gandaräer und 

das von Poklais, in welchem 

Bukephalos Alexandreia liegt. 

Und oberhalb dieser ist das 

sehr kriegerische Volk der Bak- 

trianen, die unter einem eigenen 

Könige stehen. 

(CASSON 1989: 81) Periplus 47 

Inland behind 

Barygaza 'Ep…keitai d kat¦ [nè]tou 

there are numerous peo- 

tÁj Barug£zhj mesogeaj 

ples: the Aratrioi, Arachusi- 

the peo- 

oi, Gandaraioi, and 

ple…ona œqnh, tÒ te tîn 'Aratr…wn 

kaˆ ['A]racous[…]wn kaˆ 

ples of Proklais, in whose 

Gandara…wn kaˆ tÁj Prokl[a]doj, 

area Bukephalos Alexan- 

a very 

dreia is located. 

And beyond these is 

warlike people, the Bactrians, 

under a king ... 

n oŒj ¹ Boukšfaarea 

loj 'Alex£ndreia. 

Kaˆ toÚtwn p£nw macimètaton 

œqnoj Baktrianîn, ØpÕ 

basilša Ôntwn ‡dion (tÒpon). 

— 64 —

Here, we seem to be getting a fleeting glimpse of the later 月氏 viceroy-ships — 

now reduced from eight to five — which had just been united by one of these five viceroys. 

He had subsequently established himself as king and had thus founded the second 

known Ruzhi 月氏 dynasty, that of the Kushan 貴霜 : Kujula Kadphises 丘就卻 . 

For this event, the terminus post quem is the end of the Former Han Dynasty, or the 

year 26 CE. This must be so, because the Hanshu knows much about these five 

viceroys, there called xihou 翎侯 , but nothing about their unification into one. The 

story of this unification — culminating in the epoqual creation of a new Ruzhi 月氏 

(rather early) in 

kingdom — is first narrated in the Hou Hanshu and thus happened 

the time of the Later Han. 

Writing in Latin, the Greek historian Ammianus Marcellinus, in looking back, 

reports on the ancient Græco-Bactrians and the more recent Ruzhi 月氏 : 

(ROLFE 1940: 379–381) 

The lands next to these the 

Bactriani possess, a nation 

formerly warlike and very 

powerful, and always at odds 

with the Persians, until they 

reduced all the peoples about 

them to submission and incorporated 

them under their 

own name. 

In ancient times they were 

ruled by kings who were for- 

even to Arsaces. 

midable 

Many parts of this land, like 

Margiana, are widely separated 

from the coast, but rich 

in vegetation; and the herds 

which graze on their plains 

and mountains are thickset, 

with strong limbs, as appears 

from the camels brought from 

there by Mithridates and 

seen for the first time by the 

Romans at the siege of Cyzicus 

[74 BCE]. 

Several peoples are subject 

to these same Bactrians, notably 

the Tochari. 

Res Gestae 23. 6. 55–57 

Proximos his limites possident 

Bactriani, natio an- 

tehac bellatrix et potentissima. 

Persisque semper 

infesta, antequam circumsitos 

populos omnes ad 

dicionem gentilitatemque 

traheret nominis sui, 

Quam rexere veteribus 

saeculis etiam Arsaci formidabiles 

reges. 

Eius pleraeque partes ita 

ut Margiana procul a litoribus 

sunt disparatae, sed 

humi gignentium fertiles, 

et pecus, quod illic per 

campestria loca vescitur 

et montana, membris est 

magnis compactum et validis, 

ut indicio sunt cameli 

a Mithridate exinde 

perducti et primitus in obsidione 

Cyzicena visi Romanis. 

Gentes isdem Bactrianis 

oboediunt plures, quas exsuperant 

Tochari. 

(VEH 1974: 415–416) 

Die angrenzenden Gebiete besitzen 

die Baktrianer, früher eine 

kriegerische und sehr mächtige 

Nation, dabei immer mit 

den Persern verfeindet, bevor 

diese [Baktrianer] alle umliegenden 

Völkerschaften unter ihre 

Herrschaft brachten und ihrem 

Namen einverleibten. 

In alten Zeiten wurde Baktrien 

von Königen regiert, die sogar 

einem Arsakes Furcht einjagten. 

Der Hauptteil des Gebietes ist 

ebenso wie Margiana weit von 

den Küsten entfernt, der Boden 

jedoch reich an Ertrag, und das 

Vieh, das dort auf Ebenen und in 

Bergen weidet, besitzt stattliche 

und kräftige Glieder, wie die Kamele 

bestätigen, welche Mithridates 

von hier kommen ließ und 

die Römer bei der Belagerung 

von Kyzikos zum ersten Male zu 

sehen erhielten. 

Eben diesen Baktrianern sind 

zahlreiche Völkerschaften untertan, 

insbesondere die Tocharer. 

Here, we see the Asii/Asiani = 月氏 , who had so long been a nation of nomads, well 

settled in their new lands, old Bactria. They are called Bactrians, warlike and very 

powerful. In the fourth centur y CE, Ammianus reminds us that in a ncient times — 

outside the scope of his History — the (Græco-) Bactrian kings had been feared by the 

neighboring Parthian Arsacids and that now, much closer 

to his own times, the Ruzhi 

月氏 , or new Bactrians, had for centuries been the kings of the Tocharians. 

Less than a century before the advent of the 月氏 , the description warlike and 

powerful would have been appropriate for the Bactrian Greeks under a Euthydemos, 

Demetrios, Eucratides, called “wicked but valiant Yavanas” in the Indian Yugapurå‡a 

(22–23) — or even under a Menandros, called “a good kin g of the Bactrians” by Plu- 

these valiant, forlorn, far-away Bactrian Greeks had 

tarchos (Moralia 281 D–E). But 

been deprived of their great and prosperous kingdom by 

nomadic hordes, storming 

into their civilized world from the wide steppes of the no rth. In the last years of the 

— 65 —

third century BCE, Euthydemos had warned Antiochos III, the Great, that these nomadic 

peoples, called Skythai by the Greeks and Sakai by the Achaemenids according 

to Herodotos, were a real danger to his kingdom (Polybios 11.34. 4–5). But when these 

feared Sakai nomads finally fell upon the Bactrian Greeks, in the time of Eucratides, it 

was so because they were pushed by a still more powerful nomadic people that had 

appeared out of nowhere and had carried all before them. Without the Chinese 

sources we would be at a total loss as to where this nation of nomads had come from. 

In another, to the Western classic historians as yet unknown world, they had been the 

neighbors of archaic Chinese kingdoms from times immemorial. Inside the Eastern Oikumenê 

the Ruzhi 月氏 , under a number of names, had always been well known. 

In the West, the once odd-looking strangers from another world had been at great 

pains to assimilate — but also to dominate. They had become the rulers of a good numof 

Central and South Asian peoples. In the intervening cen turies, the Asii/As iani = 

ber 

月氏 had not just conquered, first Tochara and much later the 

rest of Bactria, but also 

the Kabul and Indus Valleys and had eventually extended their empire far beyond 

these limits to the Gangetic Plains and the Erythraean Sea. Now, in the fourth century 

CE and by the Greek historian Ammianus, they are not called either by their original 

Eastern name Asii/ÅrÝi/Ruzhi, nor are they named new Tocharians, nor new Indo- 

at long last they had 

Scythians. Instead, they are here called the new Bactrians. When 

conquered the whole of Bactria, the 月氏 , for the first time, had felt at home in the 

West. Uprooted from the Hexi, evicted from the Ili, establishing themselves on the river 

Polytimetos (modern Zarafshan) or the wider region of Samarkand in Sogdiana, the 

Ruzhi 月氏 had finally settled down and become civilized 

Westerners in the famous 

lands between the Hissar Mountains and the Hindukush. Here they became the new 

Bactrians, worthy successors to the old Bactrians, the Greeks. 

Later still, in the seventh 

or eighth century, we get another faint echo, now of the 

original ethnonyms Asii/Asiani = 月氏 and Tochari = 大夏 — in the same, already fawe 

are told that the ÅrÝi 

miliar, relationship of rulers to ruled. In the Old-Uighur texts 

had been the kings of the Toχrϊ. As these Toχrϊ had readily been equated with the To- 

F.W.K. MÜLLER, 1918: 579, was among the first to do 

chari, their kings, the ÅrÝi, were to be identified with the Asii/Asiani. 

so: 

Zu diesem Erklärungsversuch des Namens årÝi durch ”ACIOI, Asiani würde auch die No- 

bedeuten, daß das 

tiz des Trogus »reges Thocarorum Asiani« gut passen, denn sie würde 

Volk »Tocharer«, die Herrscherschicht »årÝi« hieß. 

KONOW, 1932: 2–3, assisted: 

We know now that another Indo-European, but not Iranian, language was called Tokharian 

by the Uigurs ... It is true that that language is not called Tokharian in the published 

texts, but ÅrÝi, and the late F.W.K. MÜLLER was certainly right in maintaining that ÅrÝi 

must be the same name as ”Asioi and Asiani, the »reges Thocarorum« according to Trogus. 

It might therefore be maintained ... that the Uigur designation is due to the fact that the 

texts written in the language had come from the country of the Tokharians. The ÅrÝi may 

have come as conquerors from elsewhere. 

And again, in 1934: 7: 

It therefore seems probable that ÅrÝi-Toχri was the language of the Asioi-Yüe–chi who 

conquered the Tocharians-Ta–hia in the second century B.C. 

In his epoch-making book, 1938: 283–296, TARN has yet more to say about Zhang 

Qian and some of his findings are well worth quoting and discussing here: 

I have said that Chang-k’ien is quite clear that the conquest of the Ta-hia (Bactria proper) 

was the work of the Yueh-chi. But almost every modern writer known to me attributes 

that conquest to “Sacas” driven southward by the Yueh-chi, who are supposed to have 

— 66 —

occupied the country until the Yue-chi expelled or subdued them. Chang-k’ien, who was 

there, knows nothing about this ... 

Certainly Strabo says that the Sacas occupied Bactria (XI, 511), but the most cursory perusal 

of the context shows that throughout the whole section he is talking, not of the second 

century B.C., but of ... the seventh century ... 

An attempt has indeed been made since the theory was started to found the supposed 

Saca conquest of Greek Bactria a little more plausible by citing a passage in Trogus, but 

as a matter of Latin Trogus’ text will not bear the interpretation put upon it (prol. XLI: “Saraucae 

et Asiani Bactra occupavere et Sogdianos” ... to mean that the Saraucae occupied 

Bactria and the Asiani Sogdiana: a Latin writer who meant this would have said so, Trogus’ 

sentence, from its form, can only be a perfectly general statement). There is in fact no 

reason of any kind for thinking that Chang-k’ien was mistaken (the movements of the Saiwang 

have been already dealt with); and whatever happened to outlying parts of the Bactrian 

kingdom, the supposed Saca conquest of Greek Bactria proper is a myth. 

It is time to consider the Greek writers. Apollodorus attributes the conquest of the Bactrian 

kingdom to four nomad peoples, Asii, Pasiani, Tochari, and Sacarauli; “Trogus’ 

source” formally attributes it to two, Asiani and Saraucae (Saraucae for MSS Sarancae is 

certain), though subsequently he mentions the Tochari. Taking “Trogus’ source” first, one 

of his two names must represent Chang-k’ien’s Yueh-chi; and as the Saraucae (Sacaraucae), 

of whom something is known, are out of the question, the Yueh-chi are the Asiani. The 

form Asiani is an (Iranian) adjectival form of Apollodorus’ Asii, which is the substantial 

form; the Asii therefore are the Yueh-chi, whether (as some have supposed) the two words 

be identical, or not. From 1918 to 1936 it was further believed that a Central Asian text gave 

the name of a people Arsi who spoke toχrï (Tocharian) and who were the Greek Asii; the 

Arsi were accordingly supposed to be the Yueh-chi and much has been written about them. 

It has now been argued, with an impressive wealth of evidence, that Central Asian texts 

know no such people as the Arsi (Bailey 1936: 883, 905 sqq.) ... 

One must now turn to Chang-k’ien’s description of the country just after the conquest. 

First, what is the meaning of his name for Bactria proper, Ta-hia ? Before coming to what 

I think is the true view, two older explanations, which will die hard, must be noticed. The 

one most widely spread is that the Tocharoi were not the Yueh-chi at all, but were the Tahia, 

a theory which has worked utter confusion in the story ... There is not, and never has 

been, one scrap of evidence for the identification Tochari = Ta-hia except this alleged phonetic 

equivalence, and, even were that plausible, all it would prove would be, not that the 

Ta-hia were the Tochari, but that philology, though a good servant to the historian, can be 

a bad master. The matter is simple. The conquest of Bactria, we have seen, lies between 

141 and 128, and was almost certainly c. 130. The well-informed Apollodorus, in whose lifetime 

the event took place, said that the Tochari at the time were nomads. Chang-k’ien, 

who saw the Ta-hia in 128, said that the Ta-hia were communities of unwarlike traders living 

in walled towns. A conquering horde of nomads does not, in two or three years time, 

turn into communities of unwarlike traders living in walled towns; there is nothing else 

which need be said, except to regret the waste of labour and learning lavished on erecting 

theories upon such a basis ... 

To these bold statements by TARN, my personal first comment is: this contradiction 

— between the Tochari as conquering nomads in Strabo, and as timid town dwel- 

in Zhang Qian — is a valid and valuable observation. However, it does not exclude 

lers 

the equation Tochari = Daxia. It only means: the Tochari do not belong into the list of 

Strabo. 

An early critic of TARN’s book was BACHHOFER who, in 1941: 242–246, in a longer article, 

had one chapter, entitled “The Saka Conquest of Bactra.” With this, it is immeiately 

clear in what important point BACHHOFER is going to contradict his esteemed 

d 

colleague when he writes: 

— 67 —

Though they have often been published and constantly quoted, it will be best for the following 

discussion to have the records on the conquest of Bactria and the subsequent 

events at hand. They come from four sources, two of them Western and the others Chinese. 

The Chinese reports were recently translated anew by Karlgren and I am using his 

versions. 

BACHHOFER then quotes KARLGREN’s translations of the relevant passages in Shiji 

123 and Hanshu 96A which I have cited above. Of Strabo, he gives the passage which 

i ncludes our famous list with the names of four nomadic peoples. After these quotes 

BACHHOFER continues: 

Of Trogus original work only the prologues to the various books are left. Trogus wrote at 

the time of Augustus. His informant about the conquest of Bactria had written a history of 

events in the east at about the time of Apollodoros, shortly after 87 B.C. (Tarn 1938: 48). It is 

necessary to give the pertinent prologues in full, and not the usual excerpts (ed. O. Seel, 

Leipzig 1935). 

Prologus Libri XLI: Uno et quadragesimo volumine continentur res Parthicae et Bactrianae. 

In Parthicis ut est constitutum imperium per Arsacem regem. Successores deinde eius 

Artabanus et Tigranes cognomine Deus, a quo subacta est Media et Mesopotamia. Dictusque 

in excessu Arabicae situs. In Bactrianis autem rebus, ut a Diodoto rege consitutum est: 

deinde quo regnante Scythicae gentes Saraucae et Asiani Bactra occupavere et Sogdianos. 

Indicae quoque res additae, gestae per Apollodotum et Menandrum, reges eorum. 

Prologus Libri XLII: Secundo et quadragesimo volumine continentur Parthicae res. Ut 

praefectus Parthis a Phrate Himerus Mesenis bellum intulit et in Babylonios et Seleucenses 

saeviit: utque Phrati successit Mithridates cognomine magnus, qui Armeniis bellum intulit. 

Ut varia complurium regum in Parthis successione imperium accepit Orodes, qui 

Crassum delevit, et Syriam per filium Pacorem occupavit. Illi succedit Phrates, qui cum Antonio 

bellum habuit, et cum Tiridate. Additae his res Scythicae. Reges Thocarorum Asiani, 

interitusque Sacaraucarum. 

Thus the Chinese records credit the Yüeh-chi with the conquest of Bactria; Strabo four 

nomadic peoples, the Asii, Pasiani, Tochari and Sacarauli; Trogus two such peoples, the 

Asiani and the Sacaraucae. It was seen long ago that the Asii and Asiani, and the Sacarauli 

and Sacaraucae were identical. It might be possible to reduce Strabo’s list further, for 

there is much to recommend Haloun’s suggestion (1937: 244) to read h asianoi instead of 

pasianoi. 

Any attempt to correlate the western and Chinese accounts depends, of course, on the 

possibility of identifying the Yüeh-chi with one of those three, or possibly four, peoples 

mentioned by Apollodorus and Trogus’ source. The Sacaraucae (³aka Rawaka) were the 

³akas beyond Sogd (para Sugdam) of the trilingual gold tablet of Darius I. (Tarn, p. 291). 

They were probably identical with the Sai Wang of the Chinese sources, as proposed by A. 

Herrmann (Reallexikon, s.v. Sacaraucae). In any case, they cannot have been the Yüeh-chi. 

I need not recount the war of theories waged for decades over whether or not the Yüeh-chi 

were the Tocharians. This was done by Haloun ... it can be no longer doubted that the 

Yüeh-chi of the Chinese were the Tocharians of the western records. There is still the 

puzzling fact that Chang Ch’ien whose report is the source of all Chinese informations, credits 

the Tocharians alone with the conquest of Bactria, whereas Strabo, i.e. Apollodoros, 

makes them share this feat with two, or possibly three, other peoples, and Trogus’ source 

does not mention them at all in this connection. 

To evaluate these accounts properly, the circumstances under which their authors received 

their information must b e considered. Chang Ch’ien came from the east and north; 

he had, in fact, travelled in the wake of the Tocharians from China to Bactria. He came up 

in the rear, i.e. in the direction from which the attack on Bactria was launched, and, more 

important still, he arrived when everything was over. Of what had happened at the front 

he remained totally ignorant. He had not heard that the Greeks had been the rulers of 

Bactria. The only hint that the Tocharians did not succeed the Greeks immediately is the 

— 68 —

passage in the section about Ki-pin in the Ch’ien Han Shu, which cannot possibly be interpreted 

otherwise than in terms of cause and effect: the Tocharians drove the Sai Wang 

from Bactria to Ki-pin — see Karlgren’s discussion of the problem (in KONOW 1934: 10). 

With the help of his friend, the Sinologist KARLGREN, KONOW has been the first Wes- 

scholar to offer us this greatly improved understanding of the crucial two phrases 

tern 

in Hanshu 96A, section on Jibin. As I have found out, the same KONOW, some 18 years 

previous, or in 1916: 811, had written something very different on the same Chinese text: 

Ich glaube es im vorhergehenden wahrscheinlich gemacht zu haben ... daß die Ku•anas, 

d.h. die Yüe-tschi in Indien nicht im Gegensatz zu den ³akas auftraten, sondern sich vielmehr 

als ihre Erben benahmen. Der Schluß liegt jedenfalls sehr nahe, daß die beiden 

Stämme verwandt waren, und daß wir die ³aka-Eroberungen und die der Ku•anas als einen 

zusammenhängenden Fortgang ansehen müssen. Die Berichte über frühere feindliche 

Beziehungen zwischen den beiden haben auch keine allzu gute Grundlage. In den Han- 

Annalen heißt es einfach: 

»Vor alters, da die Hiung-nu die Ta Yüeh-tschi besiegt hatten, gingen die Ta Yüeh-tschi 

nach Westen und machten sich zu Herren von Ta-hia, die Sai-wang aber gingen nach 

Süden und machten sich zu Herren von Ki-pin.« 

KONOW then states in a footnote that he is quoting FRANKE 1904: 46. In this way he 

actually puts the blame on FRANKE: the accomplished German Sinologist had failed to 

understand that these two or rather three sentences only make sense when one grasps 

their logical connection. This connection is provided by the fact that the first phrase is 

t he cause for the second, and the second the cause for the third. FRANKE, however, had 

t ranslated “but ...” and in this way he had ruined the logical sequence of cause and 

effect. The third phrase thus became disconnected and meaningless. Only when we 

g rasp that we must combine all three sentences into one grand logical statement do we 

get a chance to understand the meaning of this typically terse ancient Chinese prose. It 

is condensing Eastern, Central and South Asian history of the second and first 

centuries BCE into just three short, interconnected phrases — to say the same in 

modern Chinese would require twice as many words. FRANKE’s translation, mea- 

corrected, would run like ningfully this: 

( FRANKE 1904: 46) 

( Das Königreich Ki-pin) 

... 

Vor alters, als die Hiung-nu 匈奴 die Ta 

Yüe-chi besiegt hatten, 

( da entflohen) die Ta Yüe-chi 月氏 nach 

Westen und machten sich zu Herren von Ta- 

(Tochara), 

hia 

woraufhin die Sai-wang 塞王 nach Sü- 

(entflohen) und sich zu Herren von Ki- 

den 

pin (Taxila) machten. 


罽賓國 

… 

昔匈奴破大月氏 

大月氏西君大夏 


Before the above short quote from Hanshu 96A, FRANKE, in his much-quoted book 

of 1904, translates an equally short quote from another Hanshu chapter. To under- 

the story of the Saiwang as told in the early Chinese sources, it is of importance 

stand 

to read these two closely related excerpts of two different Hanshu chapters side by side. 

I r eproduce FRANKE’s translation of the three or four sentences in Hanshu 61, adding 

the Chinese original. 

( FRANKE 1904: 46) 

Die Yüe-chi waren von den Hiung-nu besiegt 

worden (und) hatten im Westen die 塞王 Sai- 

wang angegriffen. 

Hanshu 61. 2692 

時月氏已為匈奴所破 

西擊塞王 

— 69 —

Die Sai-wang gingen nach Süden und wander- 

weit fort. 

ten 

Die Yüe-chi (aber) wohnten in ihrem Lande ... 

塞王南走遠徙 

月氏居其地 … 

Here the Hanshu remains rather vague in one important point: when attacked and 

beaten by the Ruzhi 月氏 , the Saiwang 塞王 preferred to migrate “... to the south, far 

away...” 南走遠徙 . This shows that the Chinese were clearly at a loss as far as some 

parts of Saiwang history were concerned. The Chinese historians had caught short 

glimpses of that history and in one way or another had to piece in the missing periods. 

R ead together with the imperfectly translated passage from Hansu 96A/Kipin, the 

impression for a few generations of scholars has been that the Saiwang moved in one 

l ong, uninterrupted trek from their original pasture grounds in the upper Ili River 

v alley all the way to the Northwest of India, independent of what the Ruzhi 月氏 had 

done after their initial clash with the Saiwang 塞王 . As today we know a good deal 

more of Saka history, we are safe to assume that these Scythian nomads must have left 

the valley of the upper Ili River sometime around 160 BCE, to arrive in the Panjab not 

before 60 BCE. Thus the trek would have lasted an impossible one hundred years. We 

have a great blank space here. What had really happened in this long time of at least 

three generations ? The clue to this was not to be found in new discoveries. It was solely 

to be found in a better understanding of what we have had before us all the time in 

the Chinese historical sources. 

KARLGREN’S closer look at Shiji 123 (paralleled in Hanshu 61) and Hanshu 96A on 

Jibin (Ki-pin), in what these books had to say about the migrations of the Ruzhi 月氏 

and the Saiwang 塞王 , led him to an intelligent re-appraisal and a careful new translation 

of one crucial sentence in the Hanshu chapter on Jibin. When combined with Xu 

Song’s short, but crucial 19th century comment — 

“The Saiwang had been the kings of the Daxia” 

塞王大夏之王也 

the hidden meaning of that sentence suddenly jumps to the eye. The Ruzhi 月氏 — 

who had evicted the Saiwang 塞王 first out of the upper Ili River valley, later out of 

Sogdiana, i.e. the region of Samarkand — now chase the Saiwang also out of the 

country of the Da xia 大夏 where the Saiwang had ruled for some time (less than a gen- 

In our context, we are now in a position to note down an important fact: The Chi- 

eration, as we know today). 

nese sources, i.e. Ban Gu and later compilers 

of the Standard Histories, also knew, not 

just one sole people — as VON GUTSCHMID, TARN and many later writers erroneously 

believed —, but two different peoples involved in wresting Bactria from the Greeks. 

The Central Asian Saiwang 塞王 took Eastern Bactria from the Greeks, and the East 

Asian Ruzhi 月氏 then took Daxia 大夏 (Eastern Bactria) from the Saiwang. The 

latter were forced to occupy Western Bactria 

, the area around the capital Bactra. And 

soon afterwards we expect the 月氏 to have taken the rest of Bactriana in one last 

great sweep — annihilating the Saiwang, as pronounced by Trogus. However, I believe 

that this was not so. 

When in the summer of 129 Zhang Qian arrived in Daxia 大夏 , or Tochara, from 

the north and the rear, the fighting there was over since possibly just a few months. 

The Ruzhi 月氏 had finally and firmly taken possession of Daxia by moving their 

ordos from Samarkand to the northern, or right, bank of the Oxus River. The Saiwang 

塞王 had disappeared — maybe without putting up much of a fight. They had left behind 

a Daxia without an army and without a king. A few months later, Zhang Qian 

heard nothing of the Greeks and, more surprisingly, he heard nothing of the Saiwang. 

Hence, the Shiji, too, knows nothing of the Saiwang. Zhang Qian heard, and saw, that 

the Daxia were good traders, but bad soldiers. 

From this alone we should have been 

able to deduce that the Ruzhi 月氏 did not conquer their new possessions from the 

gs of the Daxia — 

Daxia 大夏 , but from those who had been the kin the Saiwang 塞王 . 

— 70 —

Most of this we could have learned from KONOW’S elucidating Indian paper as early 

as 1934. In it, he quietly revoked his earlier utterances of 1916. Yet, too few European 

scholars read his new article. BACHHOFER, however, had studied it carefully and had 

fully grasped the important new facts coming from the new interpretation of a crucial 

passage in Hanshu 96A — the one on the kingdom of Jibin (Northwest India). BACH- 

HOFER’s reasoning is valuable and very much to the point when he declares : 

The situation, and the outlook were radically different for a Greek historian. To him this 

loss of Bactria was a catastrophe of the first rank, and his main interest must have been to 

find out to whom it was lost. The informants of Strabo and Trogus agree that the conquerors 

were the Scythian tribes of the Asiani and the Sacaraucae. They disagree only insofar, 

as the one mentions also the Tocharians and the other does not. But about the rôle of 

those two peoples they are both very definite, and there is no ground to doubt the correctness 

of their statement (as Tarn does, who says emphatically that “the supposed Saca conquest 

of Greek Bactria is a myth” — p. 284). It can be assumed that these writers, unlike 

Chang Ch’ien, were best advised about the first wave of assault which broke over Bactria. 

But even then there was room for being better or worse informed: when Trogus’ source 

does not count the Tocharians among those who wrested Bactria from the Greeks, this 

proves only that Trogus’ source was better apprised than Apollodoros with his sweeping 

statement ... 

BACHHOFER still believes in TARN’s fictitious “ Trogus’ source.” We see here that he, 

like so many before him, fails to understand that the Asiani and the Sacaraucae were 

not both Scythians, but belonged to two totally separate worlds which shortly before 

had known nothing about each other. Strabo had expressed himself to the fact that 

both peoples were Scythian tribes, but the Greek geographer was absolutely ignorant 

of the Far Eastern Oikumenê and Han China — except that he mentions, without further 

comment, the mšcri Shrîn kaˆ FrÚnwn (below, p. 78) which I take to mean the 

Tar im Basin beyond the Pamirs: to Strabo, then, the eastern end of the world was peo- 

pled by the Silk Traders and their Xiongnu overlords. When the Bactrian Greeks ex- 

tended their kingdom to Ferghana, as Strabo seems to indicate, the Sêres and Phrynoi 

had become their neighbors — for a brief period of time. These peoples had thus constituted 

the eastern end of the Hellenistic Oikumenê. 

But from the westernmost Tarim Basin the original homelands of the Asii/Asiani = 

月氏 in the Hexi 河西 Corridor (modern Gansu 甘肅 province) — around their old 

ordos of Zhaowu 昭武 (modern Zhangye 張掖 ), the Qog£ra (Thogara) of Ptolemaios 

6.16.8 — were still thousands of kilometers further east. In some of the oldest Chinese 

books such as the Guanzi 管子 , the Mu Tianzi zhuan 穆天子傳 or the Yizhoushu 

逸周書 , and under a number of different names such as Yuzhi 禺氏 , Niuzhi 牛氏 , 

Wuzhi 烏氏 , the ancestors of the Ruzhi 月氏 constituted one of the eight different 

tribes of the Xi Rong 西戎 or “Western Barbarians.” In the late seventh century BCE 

they finally moved into the Hexi 河西 , or area “West of the (Yellow) River,” — coming 

from the north and from still further east. The early Ruzhi 月氏 never had anything to 

do with the Western Oikumenê prior to the start of their migrations towards Central 

Asia in 165 BCE. 

Trogus, following Apollodoros’ book, had strictly separated the Asiani (Ruzhi 月氏 ) 

from the Sa(ca)raucae (Saiwang 塞王 ). But Strabo, with the same second hand information, 

had lumped the Far Eastern Asioi (Ruzhi 月氏 ) together with the easternmost 

Scythian tribes he knew: the Sakai/Sakaraukai of Central Asia (Saiwang 塞王 ). Thus, 

Strabo started the erroneous belief that the Asioi/Ruzhi 月氏 were just one more tribe 

of the Skythai/Sakai. This unfortunate confusion was carried over into modern studies 

on these nomadic peoples by SPECHT and LÉVI — and is still felt in our own time. 

With BACHHOFER we go back one step in the identification of the Ruzhi 月氏 . As a 

people from East Asia and of mongoloid stock, they were radically different from the 

Tocharians: first of all in their looks, and then in their language; and in a great num- 

— 71 —

er of other respects, too. To be sure, Zhang Qian would not have confounded the two 

peoples for one minute. His descriptions are clear. 

The great confusion only began some time after Zhang Qian had departed. The Ruzhi 

月氏 made every effort to became (the new) Tocharians. Above (p. 56) we have 

seen that Justinus 42.2.2 speaks of the Tochari who are attacked by the Parthian king 

Artabanos I. Of these AMANTINI, 1981: 547 3 , comments: 

Verso il 123 a.C. I Tocarii sono gli Yue-Tchi. 

The Ruzhi 月氏 strove to present themselves as worthy rulers of a people with a 

v astly superior culture which, for better or worse, had become their subjects. That the 

月氏 were different from every other people in Central Asia is clear from the fact that 

they alone — their close cousins, the Wusun, had remained behind in the region of the 

u pper Ili and Chu Rivers and around Lake Issyk Köl and in fact never crossed the 

Jaxartes — had arrived from a totally different, hitherto utterly unknown Oikumenê, 

or Inhabited World. In other words, before 160 BCE their is no way of mistaking the 

R uzhi 月氏 for a Central Asian people. Whereas, in the generations that followed, it 

b ecame soon nearly impossible to tell the two apart: first, the true and the new Kangju, 

and soon afterwards, the true and the new Daxia (Tochari). Here is the place to insert 

Zhang Qian’s description of the Daxia 大夏 or Tochari, as found it in Shiji 123; again in 

English and Chinese. 

(WATSON 1993: 235–236) 

(The capital of) the Da–xia (Bactria) is located 

over two thousand ›li‹ south-west 

of (the capital of) the Da Yuan, south of 

the Gui River (Oxus). 

Their people are settled on the land; they 

have cities and houses and they resemble 

the Da Yuan in their customs. 

(They have) no great ruler (but only) a 

good number of petty chiefs established 

in the cities and villages. 

Their soldiers are poor in the use of arms 

and afraid of battle. 

(But they are) clever at commerce. 

When the Great Yue–zhi had moved west 

attacking and conquering them, they 

completely subjugated and dominated 

the Da–xia. 

The Da–xia population is large, (numbering) 

some million or more (persons). 

Their capital is called the city of Lan–shi 

(Bactra) and it has a market where all 

sorts of goods are bought and sold. 

To their southeast is the kingdom of 

Shen–du (India). 

(Zhang) Qian reported: 

“ When ( I ) your servant was in Da–xia 

I saw bamboo canes from Qiong and 

cloth (made in the province) of Shu. 

I asked (the people): 

›Where did you get these (articles) ?‹ 

The people of the kingdom of the Da–xia 

replied: 

›Our merchants go (to buy them) in the 

markets of Shen–du.‹ 

Shen–du (they told me) lies southeast of 

the Da–xia (at a distance of) some sever- 

Shiji 123. 3164–3166 

大夏在大宛西南二千餘里 

媯水南 

其俗土著有城屋與大宛同 

俗 

無大 ( 王 ) [ 君 ] 長往往 

城邑置小長 

其兵弱畏戰 

善賈市 

及大月氏西徙攻敗之皆臣 

畜大夏 

大夏民多可百餘萬 

其都曰藍市城有市販賈諸 

物 

其東南有身毒國 

騫曰 

臣在大夏時見邛竹杖蜀布 

問曰 

安得此 

大夏國人曰 

— 72 —

al thousand ›li‹. 

Its people are settled on the land and 

greatly resemble the Da–xia. 

But (the region) is said to be flat, damp 

and extremely hot. 

Its inhabitants ride elephants (when they 

go) into battle. 

Their kingdom (is located) on a great 

river. 

According to (your servant Zhang) Qian’s 

calculations (the distance from the capital 

of) the Da–xia to the Han (capital 

Chang–an) is twelve thousand ›li‹, and 

its location is southwest of the Han (capital) 

...” 

吾賈人往市之身毒 

身毒在大夏東南可數千里 

其俗土著大與大夏同而卑 

溼暑熱云 

其人民乘象以戰 

其國臨大水焉 

以騫度之大夏去漢萬二千 

里居漢西南 … 

This is a good deal of information about the Daxia 大夏 , a people whose very name 

was well-nigh unknown until then. There is precise data on the location, the size and 

the political constitution of this country, as well as a rough description of the manners 

of its inhabitants, said to be similar to those of the people of Da Yuan 大苑 (Ferghana) 

and Shendu 身毒 (Northwest India). 

Hence the Daxia 大夏 = Tochari are everything but nomads. They are settled, 

good merchants and not at all warlike lik e the Saiwang 塞王 or the still more fearful 

Ruzhi 月氏 . With a population of about o ne million or more we know that we have to 

do here with the indigenous population of 

Eastern Bactria. Zhang Qian tells us that the 

月氏 attacked and conquered the Daxia. This is, in fact, not a really correct view of 

what had happened. He should have stated 

that the 月氏 attacked the Saiwang, the fo- 

The Saiwang then fled and left the coun- 

reign rulers of the peaceful local Tocharians. try to the 月氏 to occupy. We may infer from Zhang Qian’s description that the Sai- 

the 月氏 crossed the Hissar Mountains 

wang did not put up much resistance when 

and (for a third time) attacked the Saiwang, now in Eastern Bactria, which these no- 

some fifteen years previous. This was the 

mads had wrested from the Bactrian Greeks 

span of time the 月氏 had needed to consolidate their position up north in Samarkand 

and the whole of Sog diana, becoming the new, or pseudo-, Kangju 康居 after evicting 

the Saiwang from the region. Out of that power base they broke through the mountain 

barrier north of the Oxus and continued chasing their only and favorite enemy, the Saiwang 

塞王 , or “Royal Sakas.” The conquest of Daxia 大夏 by the Ruzhi 月氏 must 

have been what in Japanese would be ca 

lled asameshi mae 朝飯前 — an easy job. 

When Zhang Qian is shown around an unspecified town in Daxia — which I take to be 

the old capital, i.e. Lanshi 藍市 (the Darapsa of Strabo; see above, p. 51), a short dis- 

as usual: the markets are bristling with life 

tance beyond the Oxus —, it is all business 

and even the long-distance trade in luxury 

goods is working admirably well. 

3. HOW ARE WE TO UNDERSTAND THE FOUR NAMES IN STRABO’S LIST ? 

With KARLGREN, KONOW, and BACHHOFER we know now that the Chinese sources, 

too, speak of two nomadic peoples taking part in the conquest of Bactria. This is 

paralleled by what Trogus tells us: that, at first, the Asiani conquered Sogdiana and 

the Sa(ca)raucae Bactra (not Bactria); that following this the Asiani subjugated the Tobecame 

their kings; and that, f inally, the Sa(ca)raucae were destroyed, pre- 

chari and 

sumably in Bactria. 

Hence, it is only Strabo’s list which has, not two, but four names: the additional 

two names being those of the Pasianoi and the Tocharoi. Strabo names the Greek hisare 

able to show that the same Apollodoros torian Apollodoros as his source. If we 

and 

— 73 —

his Parthian History also was the main source for Trogus — and not a separate and 

very shadowy “ Trogus’ source” — then we 

shall be able to solve the enigma of Strabo’s 

longer list. In this context I must quote TARN’s famous book once more. In his Appen- 

(1938: 515), he dix 21 on “The Greek names of the Tochari” writes: 

Apollodorus, c. 100 B.C., has TÒcaroi (Tocharoi); this form was popularized by Strabo 

and has passed into common use as this people’s name. Ptolemy, VI.11.6, has this form in 

connection with Bactria, and also, VI.12.4, a form T£coroi (Tachoroi), with metathesis of 

the vowels, in connection with Sogdiana. In this form the aspirate comes in the second syl- 

is also found in the Sanscrit Tukhåra, and 

lable, not the first. This placing of the aspirate 

again in the name of Bactria (in various la nguages) from the fourth to the eighth century 

A.D., Tocharistan (presumably taken from TÒcaroi), and in forms derived from Tochariof 

Tocharistan, and Hsüan Tsiang’s Tu-ho-lo 

stan, like tocrï (tocarï or togarï) which is found later in Central Asian documents as the 

name of the Saca speech of the Kushans 

(Tuocuâlâ, Bailey) in the seventh century A.D. 

What form was used by the other Greek historian, “ Trogus’ source,” c. 85 B.C., can only 

be deduced, but certainly it was not TÒcaroi. The MSS of Trogus Prol. XLII give Thocarorum, 

Thodarorum, Thoclarorum, Toclarorum, to which the best MSS add, in Justin XLII.2.2, 

the form Thogariis. It is possible therefore, but by no means certain, that this Greek historian’s 

form had ›th‹ and was Thocaroi or Thagoroi; that is, that the aspirate came in the 

first syllable, not the second. This position of the aspirate is found later in the much-quoted 

forms Qagoàroi (Thagouroi) and Qog£ra (Thogara) in Ptolemy for a people and town in 

Kan-su on the way to Sera Metropolis, forms derived from agents of Maes Titianus and 

therefore second century A.D. 

This is an impressive array of names for one and the same — obviously rather suddenly 

famous and important — people and country. Yet TARN had overlooked NÖLDEKE 

who, 1879: 118, comments: 

›Tochåristån‹ ist das Land östlich von Balch. TÒcaroi hat Dionysius Per. 752 und Ptol. 6, 

11; so ist (nach einer freundlichen Mittheilung Rühls) bei Justin 42, 2, 2 die beste Lesart ›Tochariis‹; 

vrgl. Plin. 6, 17 § 55 (wo ›Focari‹ überliefert ist). 

Dagegen hat Strabo 511 T£caroi und Ptol. 6, 12 T£coroi, wie das Monument von Singanfu 

(T[a]cvr[i]st[å]n) schreibt. Jaq. III, 518 verlangt ›Tachåristån‹; auch ›Tach±ristån‹ 

kommt vor (eb.). Die genaue Aussprache dieses Volksnamens bleibt also unsicher. 

Thus, the oldest extant form of this crucial name is not included in TARN’s elaborate 

list: since 1956 we know it for sure that our codex antiquissimus of Strabo has 

the form TAXAPOI (Tachari) — immaculately legible (see below, p. 91). 

Here is the place to quote the one author who became the greatest critic of TARN’s 

greatest work: ALTHEIM. In 1947: 10–12, he writes: 

Wer war der hellenistische Geschichtsschreiber, der Trogus als Quelle diente ? Als Name 

bietet sich Apollodoros von Artemita an (einer griechischen Stadt östlich des Tigris, bei der 

sich die von Seleukeia kommende Straße nach Egbatana und Susa gabelte). Er schrieb in 

mindestens vier Büchern parthische Geschichte, memnhmšnoj kaˆ tîn Baktrian¾n ¢posths£ 

ntwn Èll»nwn, wie Strabon sagt (XV, p. 686). Ein Grieche, wohnhaft im Partherreiche; 

Verfasser einer parthischen Geschichte, die daneben noch die baktrische umfaßte ... 

Dennoch hat man Apollodoros von dem gleichzeitigen Geschichtsschreiber, den Trogus 

benutzte, trennen wollen. Man hat diesen als “ Trogus’ source” verselbständigen und so 

zwei gleichzeitige Autoren, die über den gleichen Gegenstand schrieben, erhalten. Und 

man hat dafür einen, wie man glaubte, durchschlagenden Grund angeführt. 

Strabon hat Apollodoros im großen Maßstab benutzt: auch dort, wo er ihn nicht ausdrücklich 

nennt. Die Erzählung, die er im 11. Buche gibt, geht zu einem Teil auf diesen Geschichtsschreiber 

zurück. Nach ihm hatten vier nomadische Stämme das baktrische Königreich 

erobert: die Asier, die Pasianer, die Tocharer und die Sakarauler. Dagegen kennt der 

Trogusprolog zum 41. Buch nur zwei Stämme: die Asianer und die Sarauker. Auf diese Ver- 

— 74 —

schiedenheit und auf sie allein gründet sich die Trennung von Apollodoros und von “ Trogus’ 

source.” 

Bei genauerem Zusehen ändert sich das Bild. Die Tocharer nennt zwar der Trogusprolog 

nicht, aber sie erscheinen in Justinus’ Auszug, waren also im vollständigen Werk mitenthalten. 

Weiter ist sowohl die Form Sak£rauloi bei Strabon wie ›Saraucae‹ bei Trogus 

korrupt. Zugrunde liegt ein ›Saka Ravaka‹ (*ravaka- “rasch beweglich”), das nur ›Sacaraucae‹ 

nom.pl. ergeben konnte und in den ›Sacaraucae‹ des Orosius (1.2.43) bewahrt ist. Also 

hat Strabon den Namen im ersten Teil richtig erhalten (vermutlich gibt überliefertes Sak£rauloi 

kaˆ den Versuch einer Verbesserung in Sak£raukai), Trogus dagegen den zweiten 

(mit Haplographie von ›Sacaraucae‹ zu ›Saraucae‹). Es liegt auf der Hand, daß beide 

Fassungen sich nicht ausschließen, sondern sich ergänzen und zusammen das Richtige ergeben 

... 

Ein Unterschied zwischen Apollodoros und “ Trogus’ source” besteht also nicht. Beider 

Überlieferung über den Nomadeneinbruch läßt sich auf den gleichen Bestand zurückführen. 

Damit fällt der einzige Grund, der sich gegen die Gleichsetzung von Apollodoros und 

“ Trogus’ source” hat anführen lassen. Die ›Partik£‹ des Mannes von Artemita lagen Trogus 

bei der Abfassung seines 41. und 42. Buches vor. 

To this criticism TARN, in 1951: 522, replied: 

Altheim’s book is based on a belief that my “ Trogus’ source” i s Apollodorus, which I 

consider impossible, if only for the simple fact that Apollodorus calls a certain people Tocaro…, 

while the MSS. of Trogus-Justin give five different versions of the name but never 

Tochari. 

TARN’s last argument in defense of a separate “ Trogus’ source” is not valid. Our repeatedly 

quoted Reges Tocharorum Asiani (SEEL 1956: 180; 1972: 324) confirms that 

Trogus, too, had the variant Tochari. 

In 1952: 2304–2308, KLOTZ warns: 

Auch Trogus selbst hatte bereits Irrtümer in den Eigennamen, die sich aus Lesefehlern 

nach griechischen Vorlagen erklären: ›Vesosis‹ Iustinus I, I, 6 wird durch Iordanes 6, 27 als 

Irrtum des Trogus erwiesen: CECΩCIC ist OE- verlesen, ebenso statt IANΔΥCIC I, 1, 6 ›Tanausis‹ 

... Die Benutzung des aus Strabo bekannten Apollodorus von Artemita hat Altheim 

Weltgeschichte Asiens I 1947, 2 erwiesen. 

ALTHEIM’s great contribution is to show that the source for both Strabo and Tr ogus 

was the lost book by the Greek historian Apollodoros of Artemita. There existed no 

other work on Parthian history to quote from at the time of the two classic historians, 

Trogus and Strabo, i.e. in the time of Augustus. It is surprising, therefore, that ALT- 

HEIM, with his intelligent insight into the real meaning of “ Trogus’ source,” did not proceed 

to the logical conclusion that not only the otherwise completely unknown “Pasianoi” 

ought to be deleted from Strabo’s list, but also the “ Tocharoi.” 

If it can be shown — as I believe ALTHEIM did — that for the same historic events 

o ccurring in the same geographic regions in both Strabo’s and Trogus’ work the one 

a nd only source is Apollodoros, then TARN’S whole argument for a separate and very 

mysterious “ Trogus’ source” falls to the ground. If this is so, we then know that Trogus 

and Strabo should inform us, not in a contradictory, but in a corresponding way on 

what happened to Greek Bactria in the second half of the second century BCE. 

Whereas Strabo lists four conquering nations, Trogus only lists two. In the name 

“ Pasianoi” was quickly discover a hidden “Asianoi” and a number of authors were inlined 

to suspect a scribal error in one way or another; ALTHEIM, as we will see pre- 

c 

sen tly, mentions the very earliest of these, the French scholar VAILLANT. The greater 

p roblem is Strabo’s second additional name: the “ Tocharoi.” Trogus does not have 

t his name in his list of conquering nations. Instead, he has the name “Tochari” as a 

n ation conquered. Who, then, is correct ? Many experts would not take it upon them to 

decide. Instead, they preferred to leave the contradiction as it was. 

— 75 —

ALTHEIM — in a book on the same subject, but published some twenty years after 

h is first work — is more precise and seemingly conclusive when in 1970: 369 he writes: 

Es bleiben die ”Asioi und Pasiano… auf der einen Seite, die ›Asiani‹ auf der anderen. 

VAILLANT’s Änderung von Pasiano… in À 'Asiano… ist auf den ersten Blick bestechend. Aber in 

dem Zusammenhang ”Asioi kaˆ Pasiano…, angesichts der Änderung des letzten Namens in 

À 'Asiano…, bleibt ka… unberücksichtigt. Man müßte denn eine zweite Korruptel annehmen, 

indem nach der Fehlschreibung Pasiano… die Hinzufügung der Kopula [ka…] unvermeidlich 

wurde ... 

›Asianus‹ ist also eine regelrechte Weiterbildung von ”Asianoj, die sich bezeichnenderweise 

bei Trogus findet. Er hat Apollodoros selbst nicht eingesehen, sondern vermutlich 

über Timagenes benutzt, während Strabo’s ”Asioi auf unmittelbare Kenntnis Apollodoros’ 

zurückgehen. Auch Ptolemaeus (Geogr. 6, 14, 10) kennt eine jüngere, durch ein Suffix weitergebildete 

Form: 'Asiîtai. 

Damit hat sich, was Apollodoros und “ Trogus’ source” zu trennen schien, verflüchtigt. 

Die Trogusprologi und Iustinus haben ungenau ausgezogen, offenkundige Korrupteln und 

eine ebenso offenkundig jüngere Form an die Stelle der ursprünglichen treten lassen. Alles 

sind Ergebnisse, die sich bei einer mehrfach gebrochenen Überlieferung zwangsläufig einstellen 

mußten. Die volle Angabe hat Strabon bewahrt, der als einziger Apollodoros selbst 

eingesehen hat. 

One page before, 1970: 368, ALTHEIM rewrites and extends his earlier statement of 

1947: 11, quoted above (p. 75), in an important way: 

Die Tocharer nennt zwar der Trogusprolog des 41. Buches nicht, sie erscheinen jedoch 

bei Iustinus (42, 2, 2) und im Prolog des 42. Buches, waren also im vollständigen Werk mitgenannt. 

Man lernt, daß ebenso die Trogusprologe wie Iustinus die Namen der nomadischen 

Stämme, obwohl sie vollständig in ihrer Vorlage enthalten waren, nur teilweise weitergegeben 

haben. 

ALTHEIM resolves the discrepancy between Trogus and Strabo — our main problem 

here — by the assumption that the four names in Strabo’s list of conquering nomad 

peoples are correct, whereas Trogus, and after him Justinus, for reasons not altoge- 

clear, skipped two of these names: the “Pasianoi” and the “Tocharoi.” 

ther 

Fortunately, besides Strabo and Trogus, we have a third and completely independent 

historical source on the same events — as BARTHOLD was the first Western author 

to state: the Chinese Standard Histories. For a long time it was believed (and some 

modern writers still do: see above, p. 40) that the Chinese sources present us a third 

variation, namely just one nomadic nation conquering Bactria: the Ruzhi 月氏 . 

Above, I have endeavored to show that this believe is mistaken. The Hanshu, compiled 

in the later first century CE by Ban Biao 班彪 and Ban Gu 班固 , father and son, 

confirms, not the four names of Strabo, but the two of Trogus. With the elucidating 

comment by XU SONG 徐松 , published posthumously in 1893, and the subsequent 

translation and publication of his intelligent discovery — of what the Chinese 

historians really meant to say — by KARLGREN and KONOW in 1934, or some forty years 

later, we finally realize: 

The Chinese say what Trogus says. The Saiwang 塞王 = Sa(ca)raucae and the Ruzhi 

月氏 = Asiani were the only two nomadic peoples who fell upon Bactria and destroyed 

the Græco-Bactrian kingdom there — in that order. And both Zhang Qian and 

Trogus, the former edited and published by Sima Tan 司馬談 and Sima Qian 司馬遷 , 

tell us that in the process the Daxia 大夏 = Tochari were subjugated by the Ruzhi 月氏 

= Asiani. Zhang Qian, moreover, provides us with the unique and valuable information 

that the Daxia 大夏 = Tochari were, in fact, no other people but the autochthonous, 

local population of Eastern Bactria — left behind after first the Bactrian Greeks and 

then the Saiwang 塞王 = Sacaraucae had been annihilated there. 

— 76 —

Strabo’s list with the four ethnic names, then, can be singled out as the one which 

must have become corrupt in the process of transmission. The “Pasianoi” and the “Tocharoi” 

should not be on that list. The important question, still to be solved here, is: 

By what mistaken way did the two names get into Strabo’s list ? So far as the “Tochaoi” 

are concerned, this question has not yet been asked by any author. In other words: 

r 

the “Tocharoi” in Strabo’s list have always been taken for granted. 

First HERRMANN’s and then TARN’S observation — that Strabo’s conquering Tochacould 

not be the settled Daxia 大夏 — has been a very intel 

ligent one. It 

roi nomads 

p rovided us with a clear point of departure: the problem definition. Both authors took 

Strabo’s authority for granted and hence could not accept the equation Daxia 大夏 = 

Tochari. However, with the above closer look at the Chinese historical sources, I hopeully 

showed that the equation Daxia 大夏 = Tochari is valid. If Trogus’ Tochari are 

f 

Zhang Qian’s Daxia 大夏 , we know at long last that this hitherto unknown ethnic 

name belonged to a people firmly settled on the upper Oxus River since a good number 

of centuries at the time of Zhang Qian’s visit. The envoy of Chinese Emperor Wu is our 

o nly Zeitzeuge, or eyewitness. The authenticity of his Report can be trusted. How, then, 

d id the name “Tocharoi” get into Strabo’s list — and thereby cause such havoc with all 

related research ? 

Zhang Qian’s Daxia 大夏 and Trogus’ Tochari appear to be the lowest strata, the 

autochthonous, or indigenous, inhabitants of Eastern Bactria. They are well settled in 

that region and must have been there since a long time: first under the Persians; then, 

for a full century, under the Bactrian Greeks; thereafter briefly under the kings of the 

Saiwang 塞王 . Finally they fall under the sway of the fearful kings of the 月氏 — who 

a re shrewd enough to adopt the name and language of the Tochari. Together with the 

administration and armies of the last Græco-Bactrian kings, the names “Baktriana” 

and “Baktrioi” disappear and rather suddenly there is this new name: Daxia = Tocha- 

It will now be quoted by every author in East and West (in that order), writing on 

ra. 

the history of Central Asia. I wonder: are the Tochari mentioned prominently anywhere 

before 129 BCE ? 

My impression is that the Ruzhi 月氏 picked up a local, little-known name to make 

it their own. Only thereafter it became great and famous. It seems that the 月氏 , Far 

Eastern strangers that they were, deliberately decided to use a Central Asian — for 

them: Western — name for their newly founded empire on the Oxus River, an empire 

which already included Sogdiana between the Hissar Mountains in the south, the Oxus 

in the west, and the Jaxartes in the east. It was to include much more in the future. 

But for an initial one century and a half, it did not include the whole of former Greek 

Bactria — especially not the age-old capital Bactra. The Ruzhi 月氏 , in search for a 

suitable name, must have referred to their new acquisition as “Tochara” when talking 

to Zhang Qian. He is the first author to reduce the new name to writing — one or two 

generations before Apollodoros of Artemita. 

Of course, it was bound to create confusion when the conqueror decided to use the 

name of the people conquered. But the 月氏 exceedingly liked what they had inherited 

from the Bactrian Greeks. It must have been a kind of paradise for them, a Schlaraffenland. 

And so they decided to become Tocharians — but also to become their new 

kings: reges Tocharorum Asiani. 

Apollodoros, too, had been enthusiastic about Bactria in Greek times. He even had 

a few remarks on the Baktrians as they appeared before the conquest of Alexander. 

But nowhere do I find the name of the Tochari. Strabo, quoting Apollodoros, writes: 

(JONES 1928: 279–281) 

And much of it produces everything 

except oil. 

The Greeks who caused 

(Radt 2004: 356–358) 

Es ist ein ausgedehntes Land, 

das alles außer Öl erzeugt, und 

die Griechen, die es selb- 

Bactria to revolt grew so ständig gemacht haben, sind 

Geographika XI. 11. 1–3 

Poll¾ d/ stˆ kaˆ p£mforoj 

pl¾n la…ou, tosoàton 

d' ‡scusan oƒ ¢post» santej 

“Ellhnej aÙt¾n di¦ 

— 77 —

durch die Trefflichkeit des Landes 

so stark geworden, dass 

sie die Herrschaft über Ariane 

und die Inder besaßen, wie 

Apollodor von Artemita sagt 

(FGrHist 779F7), und mehr Völ- 

ker unterworfen haben als 

Alexander ... 

Er sagt, Baktriane sei überhaupt 

das Prachtstück der 

ganzen Ariane; 

so haben sie ihre Herrschaft 

sogar bis zu den Serern und 

den Phaunern (Phrynern) ausgedehnt. 

An Städten hatten sie Baktra, 

das auch Zariaspa genannt 

wird (hindurch strömt ein 

gleichnamiger Fluss, der in den 

Oxos mündet), Darapsa und 

mehrere andere; 

dazu gehört auch das nach 

dem einstigen Herrscher benannte 

Eukratideia ... 

In alter Zeit unterschieden die 

Sogdianer und die Baktrianer 

sich in ihrer Lebensweise und 

ihren Sitten nicht sehr von den 

Nomaden; doch waren die Sit- 

ten der Baktrianer etwas zivilisierter 

... 

powerful on account of the 

fertility of the country that 

they became masters, not 

only of Ariana, but also of 

India, as Apollodorus of Artemita 

says; and more tribes 

were subdued by them than 

by Alexander ... 

In short, Apollodorus says 

that Bactriana is the ornament 

of Ariana as a whole; 

and, more than that, they 

extended their empire even 

as far as the Seres and the 

Phryni. 

Their cities were Bactra (also 

called Zariaspa, through 

which flows a river bearing 

the same name and emptying 

into the Oxus), and Darapsa, 

and several others. 

Among these was Eucratidia, 

which was named after 

its ruler ... 

Now in early times the Sogdians 

and Bactrians did not 

differ much from the nomads 

in their modes of life 

and customs, although the 

Bactrians were a little more 

civilised ... 

t¾n ¢ret¾n tÁj cèraj 

éste tÁj te ’ArianÁj pekr£toun 

kaˆ tîn ’Indîn, 

éj fhsin ’ApollÒdwroj 

Ð ’ArtamithnÒj, kaˆ ple…w 

œqnh katestršyanto 

½per ’Alšxandroj ... 

Kaq' Ólou dš fhsin ke‹- 

noj tÁj sump£shj 'ArianÁj 

prÒschma enai t¾n 

Baktrian»n : 

kaˆ d¾ kaˆ mšcri Shrîn 

kaˆ FaÚnwn (FrÚnwn) xšteinon 

t¾n ¢rc»n. 

PÒleij d/ econ t£ te 

B£ktra, ¼nper kaˆ Zari- 

£span kaloàsin ¼n dia- 

rre‹ Ðmènumoj potamÕj 

mb£llwn e„j tÕn ’Wxon, 

kaˆ D£raya kaˆ ¥llaj 

ple…ouj : 

toÚtwn d' Ãn kaˆ ¹ EÙkrat…deia 

toà ¥rxantoj 

pènumoj ... 

TÕ mn oân palaiÕn oÙ 

polÝ dišferon to‹j b…oij 

kaˆ to‹j œqesi tîn nom£dwn 

o† te Sogdianoˆ 

kaˆ oƒ Baktriano…: mikrÕn 

d' Ómwj ¹merètera Ãn t¦ 

tîn Baktrianîn ... 

Here, the inhabitants of the regions on the upper Oxus River, between the Hissar 

and the Hindukush Mountains, are called Baktriano… (Bactriani), their capital is 

named B£ktra (Bactra) and the province Baktrian» (Bactriana = Bactria). These are 

the names we are so very familiar with. The name TÒcaroi (Tochari) is mentioned nowhere. 

There is a good chance that Strabo has this name only once: in his ominous 

list. But later classical authors from Pliny to Ptolemy have this name everywhere, especially 

Ptolemy. In MANNERT, 1820: 455–467, we read: 

In Sogdiana sitzen nach Ptolemaeus die Paesikae (Paisika…) an den Oxianischen Bergen, 

also nördlich von Samarkand. An dem nördlichen Laufe des Jaxartes die Jatii (Plin. VI.16 

und 17 nennt auch Dacii und Parsicae) und Tachori ('I£tioi kaˆ T£coroi); also westlich und 

östlich um Kodgend. Die Tochari hatte Ptolemaeus schon in Bactriana als ein großes Volk 

angegeben; vermutlich sind beide nicht verschieden, und in Sogdiana ist der eigentliche 

Ursitz des Volkes zu suchen ... 

Baktrien, Hyrkanien und dem größten Theile nach auch Sogdiana erkannten von jener 

Zeit die Oberherrschaft der Persischen, vielleicht auch schon der Assyrischen Monarchen. 

Alexander fand bey seinem Eintritt in diese Gegenden eine Menge ansehnlicher, gut bevöl- 

kerter S tädte, welche in der Nachbarschaft roher Völker sich erhielten und blüheten; und 

die gleichzeitigen Schriftsteller rühmen den reichen Anbau der glücklichern Striche. Bey- 

des scheint ein hohes Alter von Kultur bey den Einwohnern voraus zu setzten, von welcher 

sich aber beym Mangel aller Nachrichten nichts Weiteres sagen 

läßt ... Diese Völker theil- 

am Nieder-Oxus ten die Schriftsteller nach der Lage in zwey Stämme: in die westlichern ... 

— 78 —

zu den Erstern rechneten sie außer mehrern kleinern Völkern vorzüglich die Parnae ... zu 

den östlichern die Asii, Pasiani etc., und vorzüglich die Tachari, Tachori, oder Tochari ... 

Auch die nördlichen Provinzen Baktriana, Sogdiana gehörten 

nicht den Parthern; diesen 

brachten die Einfälle der nördlichen Barbaren, der Asii, Pasiani und Tachari, Zweige der 

Sakae, den Untergang (Strabo XI.511). Sie fielen ein, drängten die ältern Bewohner des 

Landes gegen das südöstliche Gebirge, und blieben von nun 

an i n dem Besitze des 

größern Theils. Von dieser Periode werden erst die Angaben des Ptolemaeus richtig, der 

das beträchtliche Volk der To chari durch die ganze nördliche Länge von Baktria setzt, zu- 

(auch Dionys. Perieg. 

gleich aber durch die nochmalige Anführung derselben am Jaxartes v. 752 stellt die Toc£roi über 

den Jaxartes) auf die ältesten Sitze hinweiset, aus welchen 

sie gegen Süden wanderten. 

In the time of Ptolemy, when the Ruzhi 月氏 had created a powerful Empire under 

the dynasty of the Kushana, they appear in the classical (Western) sources under their 

new (Western) name TÒcaroi (Tocharoi), and this name is now applied to all the re- 

passed on their way to Bactria 

and India. The 月氏 have 

gions through which they had 

become Westerners in name, language and appearance. Their Far Eastern origin now 

seems to be a faint memory even to themselves. Their oldest habitat is now considered 

to have been just beyond the Jaxartes. With that they had to be Scythians — like all 

the other nomadic peoples from these northerly regions — as Strabo tells us. Here is 

the place to reproduce Strabo’s crucial passage in full. 

(JONES 1928: 259–261) 

On the left and opposite 

these peoples are 

situated the Scythian 

or nomadic tribes, 

which cover the whole 

of the northern side. 

Now the greater part 

of the Scythians, beginning 

at the Caspian 

Sea, are called Däae, 

but those who are 

situated more to the 

east than these are 

named Massagetae 

and Sacae, whereas 

all the rest are given 

the general name of 

Scythians, 

though 

each people is given a 

separate name of its 

own. 

They are all for the 

most part nomads. 

But the best known of 

the nomads are those 

who took away Bactri- 

from the Greeks, I 

ana 

mean the Asii, Pasia- 

Tochari, and Saca- 

ni, 

rauli, who originally 

came from the country 

on the other side 

(LASSERRE 1975: 83) 

A l’opposé de ces peuples, 

à main gauche, 

on a les peuples scy- 

les nomades, 

thes et 

qui couvrent tout le 

côté septentrional. 

La plus grande partie 

des Scythes, en comnt 

à la Mer Cas- 

mença 

pienne, sont connus 

sous le nom de Dahae, 

tandis qu’on appelle 

Massagètes et Saces 

ceux qui vivent plus à 

l’est et qu’on désigne 

tous les autres sous le 

nom général de Scythes, 

bien qu’ils aient 

chacun des noms particuliers. 

Ils ont tous une population 

en grande majorité 

nomade. 

Les plus connus d’entre 

eux sont ceux qui 

enlevèrent aux Grecs 

la Bactriane, à savoir 

les Asiens, les Pasiens, 

les Tokhariens et les 

Sacarauques, qui 

étaient partis des territoires 

situés au delà 

(RADT 2004: 341–343) 

Zur Linken ziehen 

sich diesen gegen- 

über skythische 

und 

nomadische Völker 

hin, die die ganze 

nördliche Seite bevöl- 

kern. 

Die meisten Skythen 

werden, angefangen 

beim Kaspischen 

Meer, Daer genannt, 

die weiter östlich als 

diese wohnenden 

nennt man Massageten 

und Saken, die 

übrigen bezeichnet 

man allgemein 

als 

Skythen und einzelne 

mit ihrem besonderen 

Namen. 

Alle sind größtenteils 

Nomaden. 

Am bekanntesten sind 

von den Nomaden diejenigen 

geworden, die 

den Griechen Baktrien 

entrissen haben: 

die Asier, die Pasianer, 

die Tocharer und 

die Sakarauker, die 

von dem anderen 

Ufer des Iaxartes her 

Geographika IX.8.2 

'En ¢rister´ d 

toÚtoij ¢ntipar£keitai 

Skuqik¦ œqnh 

kaˆ nomadik£, 

¤pasan kplhroàntai 

t¾n bÒreion 

pleur£n. 

Oƒ mn d¾ ple…ouj 

tîn Skuqîn ¢pÕ 

t¾j Kasp…aj qal£tthj 

¢rx£menoi 

D£ai prosagoreÚontai, 

toÝj d 

proseóouj toÚtwn 

m©llon Massagštaj 

kaˆ S£kaj Ñnom£zousi, 

toÝj d' 

¥llouj koinù mn 

SkÚqaj ÑnÒmati 

kaloàsin, „d…v d' 

æj ˜k£stouj : 

¤pantej d' æj pˆ 

tÕ polÝ nom£dej. 

M£lista d gnèrimoi 

gegÒnasi tîn 

nom£dwn oƒ toÝj 

“Ellhnaj ¢felÒmenoi 

t¾n Baktrian»n, 

”Asioi kaˆ 

Pasianoˆ kaˆ TÒcaroi 

kaˆ Sakaraàkai, 

Ðrmhqšn- 

— 79 —

of the Iaxartes River 

that adjoins that of 

the Sacae and the Sog- 

and was oc- 

diani 

cupied by the Sacae. 

de l’Iaxarte, à la hauteur 

des Saces et des ken und Sogdianern 

kamen, das den Sa- 

Sogdiens, et relevant gegenüber liegt und 

alors de l’autorité des im Besitz der Saken 

Saces. 

gewesen war. 

tej ¢pÕ tÁj pera…- 

aj toà 'Iax£rtou 

tÁj kat¦ S£kaj 

kaˆ SogdianoÚj, ¿n 

kate‹con S£kai. 

It is revealing to compare the three translations. The English rendering, being the 

o ldest, does not yet know the improved reading Sakaraàkai (Sakaraukai) in place of 

the older and corrupt Sak£rauloi (Sakarauloi). The French “... les Asiens, les Pasi- 

does not reflect the difference between ”Asioi and Pasiano…, i.e. between Asii 

ens...” 

a nd Asiani, which is of importance as we shall see later. And all three translations — 

together with the Greek original which I copy here from the edition of R ADT 2004 — 

ignore the fact that our oldest codex has TAXAPOI (Tacharoi) were later manuscripts 

have TÒcaroi (Tocharoi), under which form this all-important ethnonym is best 

known today. It is interesting to note here that MANNERT 1820 does in fact know the 

variant Tachari, as well as NÖLDEKE 1879 (T£caroi) and VON GUTSCHMID 1888 (Tacharer), 

whereas TARN 1938 (also 1951, 1984 and 1997) does not. 

LASSERRE, 1975: 83, adds a footnote to his edition and translation of this passage, 

which is of interest for us: 

La conquète de la Bactriane par les nomades eut lieu en 130 environs selon Tarn, 277 et 

294 (entre 133 et 129 selon P. Daffinà, ›L’immigrazione dei Sakå nella Drangiana‹, Rome, 

1967, 45), qui a reconn u dans ce passage le premier d’une série d’extr aits d’Apollodore d’Artémita. 

Traduisant ka 

te‹con par ›innegehabt hatten‹, Altheim-Stiehl, 609, confirment la 

thèse de Daffinà, 45–82, selon laquelle l’invasion aurait comporté deux vagues successives, 

bien distinguées, par le chroniqueur chinois [du] Ts’ien Han-Chou: celle 

des Saces, partis 

de l’actuel Ouzbékistan pour s’établir finalement dans l’actuel Sðstån, et celle des quatre 

autres peuples nommés ici, partis du même endroit pour enlever la Bactriane à Phraatès II. 

1970: 609 ALTHEIM writes: 

Die Sai [ 塞 ], mittelc hinesisch sәk , sind die Saken. Die Yüe-chi wohnten, als die Sai vor 

ihnen gewichen waren, im alten Sakenlande; dies fiel in eine Zeit, die 

dem Einbruch der 

Yüe-chi in Ta-hia voranging. Man hat nicht bemerkt, daß Apollodoros von Artemita das- 

griechische Bak- 

selbe berichtet. Er sagt von den skythischen Nomadenstämmen, die das 

trien eroberten: 

Ðrmhqšntej ¢pÕ tÁj pera…aj toà 'Iax£rtou tÁj kat¦ S£kaj kaˆ SogdianoÚj, 

¿n kate‹con S£kai. 

Das Imperfekt kate‹con bezeichnet die vorvergangene Handlung als dauernd: “aufbrechend 

von dem jenseitigen (nördlichen) Ufer des Iaxartes, das in der Gegend der Saken 

und Sogdianer (liegt), das die Saken (für längere Zeit) innegehabt hatten.” Demzufolge 

hatten auf dem Nordufer des Iaxartes Saken gesessen, deren Gebiet dann, vor dem Über- 

Damit bestätigt 

schreiten des Flusses, durch jene Nomadenstämme eingenommen wurde. sich, daß die Sai den Yüe-chi vorangezogen waren. 

That much is correct: Apollodoros knows of two nomadic waves. The nomads who 

took Bactria from the Greeks came from somewhere beyond, i.e. east of, the Jaxartes. 

In those regions — of modern Kirgizstan and Kazakhstan, not Uzbek 

istan — the Sakas 

had lived a long time. They are now found living amongst the Sogdians: their crossing 

the Jaxartes from East to West, therefore, constituted the first wave 

of nomadic irrup- 

burst forth 

tion, obviously into Greek Sogdiana only. Then, more nomadic peoples 

from those regions beyond the Jaxartes where the Sakas had long used to live: this 

constitutes the second 

wave of nomadic irruption, now into Sogdiana and later into 

Bactria. 

Strabo, excerpting Apollodoros, relates that in the second wave came four nomadic 

peoples: the Asioi, Pasianoi, Tocharoi and Sakaraukai. With this it seemed an intelli- 

apart from gent conclusion that the Sakai of the first wave had to be strictly told the 

— 80 —

Sakaraukai of the second wave — as above LASSERRE suggests in his footnote. We 

know, however, that the Sakaraukai are one tribe of the great Saka Federation: they are 

called Saka + *rauk-, or “Royal Sakas” — the exact translation into Chinese being 

Saiwang 塞王 . With this we have to interpret Strabo in a slightly different way: that in 

the first wave had come a Saka tribe, namely the Sakaraukai, and in the second wave 

then came the remaining three nomad nations, the Asioi, Pasianoi and Tocharoi. 

Trogus, also excerpting Apollodoros as we know by now, simply relates — in the 

surviving “Table of Contents” to his lost book — that the Sa[ca]raucae and the Asiani 

broke into Bactra and Sogdiana, respectively. This, too, seems to indicate that the Sacaraucae/Sakaraukai 

came in a first wave, but that in a second wave there appeared on 

the scene, not three, but only one more nomadic people, the Asiani/Asioi. With Trogus 

we see that the Sacaraucae have advanced from Sogdiana into Bactra (as he pointedly 

writes, not into Bactria — which for some time broke apart into Tochara and Bactra): 

this is the state of affairs somewhat later in time than the one Strabo presents us. 

This much — or rather this little — we can deduce from our (Western) classical 

sources. They bequeath us a discrepancy or a problem which, as we have seen above, 

has been left unsolved for a long time. L ASSERRE, in his footnote 1975: 83–84 following 

ALTHEIM, goes on to explain that Trogus, and after him Justinus, by negligence left out 

one or two names in the list of conquering nomads. 

Trogue Pompée ap. Justin., Prol. XLII, compte seulement deux nations dans la seconde 

vague, les ›Sacaraucae‹ et les ›Asiani‹, qu’il appelle pourtant ›reges Tocharorum‹. Si les 

›Asiani‹ sont les ”Asioi (Daffinà, Altheim-Stiehl), les Pasiano… manquent, probablement par 

omission de l’épitomateur, et pourraient n’avoir été qu’une fraction ethnique de l’un des 

deux prédominants, ce qui expliquerait la divergence entre les témoins grec et latin d’Apollodore. 

Rarely has an effort been made to compare the Western with the Eastern sources in 

extenso. The main reason surely is that this requires a working knowledge of a dozen 

living, or dead, languages and their different scripts. One way around the problem is to 

assemble and collate here, word by word, the best expertises from both worlds: East 

and West. 

It has always been claimed that the bloody clashes between the Xiongnu 匈奴 and 

he Ruzhi 月氏 , deep in East Asia and in the second century BCE, have been decisive 

t 

f or the history of Central and South Asia in the next five hundred years or so. This is 

tru e, but it is not the complete truth. It has been overlooked that a third nomadic 

nat ion played a crucial part in shaping this chapter of Asian history. For when the Ruzhi 

月氏 , after their last crushing defeat at the hands of the Xiongnu, decided to escape 

Xiongnu domination, their trek to the West ended somewhere in the regions of the up- 

Ili River — where they dislodged the Sakaraukai/Sakwang 塞王 . As far as the Ru- 

per 

z hi 月氏 were concerned, they would have remained in those regions east of the 

Jaxartes and would have continued roaming about the region with their flocks and 

trading their horses for Chinese silk and other luxury goods for a good profit. Central 

Asia would have experienced only one nomadic irruption — that of the Sakaraukai — 

which would have overrun Sogdiana to end Greek domination there. But the Greek 

kingdom in Bactria proper might have survived for several centuries longer. 

That this was not the course of history in Central Asia in the second century BCE is 

due to a small nomadic people of whom Zhang Qian has left us the very first precious 

accounts: the Wusun 烏孫 or “Grandsons of the Raven.” 

(WATSON 1993: 237–238) 

This same year (122 B.C.) 

the Han sent the swift cavalry 

(of General Huo Qu– 

bing who succeeded in) defeating 

several ten thou- 

Shiji 123. 3167–68 

是歲漢遣 

驃騎破匈 

奴西 ( 城 ) 

(HULSEWÉ 1979: 213–217) 

In this year [121 B.C.] 

the general of cavalry 

on the alert (P’iao–ch’i) 

defeated the Hsiung– 

nu on their west side, 

Hanshu 61. 2691-2 

是歲驃騎 

將軍破匈 

奴西邊殺 

— 81 —

sand men of the Xiong–nu 

in the Western Regions 

(penetrating) as far as the 

Qi–lian Mountains. 

The following year (121 

B.C.) the Hun–ye king led 

his people and surrendered 

to the Han (Chinese). 

And (in the whole region) 

from Jin-cheng and He-xi 

west along the Southern 

Mountains all the way to 

the Salt Swamp the Xiong– 

nu completely disappeared. 

The Xinong–nu occasionally 

had scouts appear (in 

the region), but even they 

were rare. 

Two years after this (119 

B.C.) the Han (armies) 

attacked the Shan–yu and 

( sent him) running to the 

north of the desert. 

After this the Son of Heaven 

a number of times 

questioned (Zhang) Qian 

about Da–xia and the 

other states (of the west). 

(Zhang) Qian, who had lost 

his marquisate, availed 

himself (of this chance) 

and replied with this report: 

“ When your servant 

(Zhang Qian) was living 

among the Xiong–nu he 

heard about the king of the 

Wu–sun (people), whose 

title was Kun–mo. 

The Kun–mo’s father was 

(the ruler) of a small state 

on the western border of 

the Xiong–nu (territory), 

THE XIONG–NU attacked 

and killed his father, and 

the Kun–mo, who had just 

been born, was cast out in 

the wilderness (to die). 

(But) the ravens, bearing 

meat in their beaks, flew 

over the place (where he 

was), and the wolves came 

[ 域 ] 數萬 

人至祁連 

山 

其明年渾 

邪王率其 

民降漢 

而金城河 

西西並南 

山至鹽澤 

空無匈奴 

匈奴時有 

候者到而 

希矣 

其後二年 

漢擊走單 

于於幕北 

是後天子 

數問騫大 

夏之屬 

騫既失侯 

因言曰 

臣居匈奴 

中聞烏孫 

王號昆莫 

昆莫之父 

匈奴西邊 

小國也 

匈奴攻殺 

其父而昆 

莫生棄於 

野 

烏嗛肉蜚 

killing men by the ten 

thousand and reaching 

the Ch’i–lien Mountain. 

That autumn the K’un– 

yeh king surrendered 

to the Han with his 

community. 

(The area) west of 

Chin–ch’eng (commandery) 

and the (Yellow) 

River and along 

the southern hills as 

far as the Salt Marsh 

was empty and without 

Hsiung–nu; 

occasional patrols of 

the Hsiung–nu went 

there, but only rarely. 

Two years later Han 

attacked and drove 

the Shan–yü to the 

north of the desert (119 

B.C.). 

The Son of Heaven frequently 

asked (Chang) 

Ch’ien about the states 

such as Ta Hsia. 

Since he had already 

lost his noble rank, 

(Chang) Ch’ien took 

the opportunity to report 

as follows: 

“ When I was living 

among the Hsiung–nu 

I heard of Wu–sun; 

the king was entitled 

K’un–mo, and the 

K’un–mo’s father (was 

named) Nan–tou–mi; 

originally (Wu–sun) 

had lived with the Ta 

Yüeh–chih between 

Ch’i–lien (mountains) 

and Tun–huang; 

and they had been a 

small state. 

THE TA YÜEH–CHIH 

attacked and killed 

Nan–tou–mi, seizing 

his lands; and his 

people fled to the 

Hsiung–nu. 

An infant K’un–mo had 

數萬人至 

祁連山 

其秋渾邪 

王率眾降 

漢 

而金城河 

西 ( 西 ) 並 

南山至鹽 

澤空無匈 

奴 

匈奴時有 

候者到而 

希矣 

後二年漢 

擊走單于 

於幕北 

天子數問 

騫大夏之 

屬 

騫既失侯 

因曰 

臣居匈奴 

中聞烏孫 

王號昆莫 

昆莫父難 

兜靡 

本與大月 

氏俱在祁 

連焞煌間 

小國也 

大月氏攻 

殺難兜靡 

— 82 —

and suckled him (so that 

he was able to survive). 

(When) the Shan–yu 

(heard of this he) was 

filled with wonder and, 

believing that (the Kun– 

mo) was a god, he took 

him in and reared him. 

When he had grown to 

manhood, (the Shan–yu) 

sent (the Kun–mo) to command 

a band of troops and 

he several times won merit 

(in battle). 

The Shan–yu gathered his 

father’s people together 

again, restored them to the 

Kun–mo and ordered him 

to be the senior guard in 

the Western Regions. 

The Kun–mo took over and 

looked after his people and 

led them in attacks on the 

small settlements in the 

neighborhood. 

(Soon) his skilled archers 

其上狼往 

乳之 

單于 

怪以 

為神而收 

長之 

及壯使將 

兵數有功 

單于復以 

其父之民 

予昆莫令 

長守於西 

( 城 )[ 域 ] 

昆莫收養 

其民攻旁 

小邑 

recently been born, 

and the Pu–chiu Hsihou, 

who was his 

guardian, took him in 

his arms and ran 

away. 

He laid him in the 

grass and searched for 

food for him; and on 

coming back he saw a 

wolf suckling the child; 

furthermore there were 

ravens holding meat in 

their beaks and hovering 

at (the child’s) side. 

Believing this to be suto 

the K’un–mo 

pernatural, he then 

carried (the child) 

back to the Hsiung–nu, 

and the Shan–yü loved 

and reared him. 

When he had come of 

age, (the Shan–yü) delivered 

his father’s people; 

He had him lead 

troops, and on several 

occasions he did so meritoriously. 

At the time the Yüehchih 

had already been 

defeated by the 

Hsiung–nu; 

Making for the west 

they attacked the Saiwang. 

The Saiwang moved a 

considerable distance 

to the South and the 

Yüeh–chih then occu- 

pied their lands. 

Once the K’un–mo had 

grown to adulthood, he 

asked permission of 

the Shan–yü to avenge 

his father’s wrongs. 

Going west he at- 

tacked and defeated 

the Ta Yüeh–chih, who 

again fled west, moving 

into the lands of 

Ta Hsia. 

The K’un–mo despoil- 

ed the population of 

奪其地人 

民亡走匈 

奴 

子昆莫新 

生傅父布 

就翎侯抱 

亡置草中 

為求食還 

見狼乳之 

又烏銜肉 

翔其旁以 

為神遂持 

歸匈奴單 

于愛養之 

及壯以其 

父民眾與 

昆莫使將 

兵數有功 

時月氏已 

為匈奴所 

塞 

破西擊 

王 

塞王南走 

遠徙月氏 

居其地 

昆莫既 

健 

自請單于 

報父怨遂 

西攻破大 

月氏 

大月氏復 

西走徙大 

— 83 —

(numbered) several ten 

thousand, trained in ag- 

gressive warfare. 

When the (›Lao–Shang‹) 

Shan–yu died (161 B.C.), the 

Kun–mo in fact led his people 

in a trek far away (declaring 

himself an) independent 

(ruler), he refused 

(any longer to journey to) 

the court meetings of the 

Xiong–nu ... 

控弦數萬 

習攻戰 

單于死昆 

莫乃率其 

眾遠徙中 

立不肯朝 

會匈奴 

Ta Hsia [of the Ta 

Yüeh-chih], and then 

remained there in occupation. 

His forces gradually 

grew stronger 

and at the death of the 

Shan–yü he was no 

longer willing to attend 

at the court of the 

Hsiung–nu and serve 

them ... 

夏地 

昆莫略其 

眾因留居 

兵稍彊 

會單于死 

不肯復朝 

事匈奴 

It may well be the first time that these two “ parallel” texts, excerpts of Zhang Qian’s 

biography in the Shiji and the Hanshu, are re produced side by side — together with 

their English translations — in Western literature. We have an excellent example here 

how the Hanshu is not copying, but carefully 

editing the nearly two hundred years 

older Shiji with corrections and additions which are of the greatest importance (I have 

made it clear where the two Chinese texts differ). 

A seemingly minor correction is that the Hanshu has “West of the He (= Yellow Ri- 

tells us that 

ver)” 河西 where the Shiji writes “West of Hexi” 河西西 . The full sentence an area, formerly occupied by the Ruzhi 月氏 , then annexed by the Xiongnu 匈奴 , is 

now empty. The ancient nam e of this area is Hexi 河西 (“West of the River”). Hence, it 

cannot be located west to itself. Instead, the 

Hanshu wants to tell us here that the 

former homelands of the Ruzhi 月氏 had a common border with Han China at (the 

city of) Jincheng 金城 — where the Great Wall 長城 ended at that time — and at the 

(Yellow) River 河 . In fact, we are told elsewhere that the old territory of the 月氏 extended 

even a little beyond that river. Above, we have seen that Longxi 

隴西 was still 

the border town at the start of Zhang Qian’s m ission, even after the 月氏 had since 

long left their old lands. 

HULSEWÉ/LOEWE misunderstood the correction in the Hanshu text 

and, 1979: 213, 

translated 金城河西 ( 西 ) as “ (The area west of) Chin-ch’eng and Ho-hsi there has never been a Chinese com- 

(command- 

eries) ...” The Translators ignore the fact that 

m andery by the name of Hexi (Ho-hsi). Instead, Hexi 河西 has been the ancient name 

of the whole of modern Gansu Corridor. The c orrect translation of the four characters 

金城河西 should be “ West of Jincheng (city) and the (Y ellow) River ...” This gives us 

the crucial eastern limit of the former homelands of the Ruzhi 月氏 — which is miss- 

in the often repeated, but unfortunately very abbreviated formula “... between the 

ing 

Qilian (Mountains) and Dunhuang ...” 祁連焞煌間 (also appearing in our above Han- 

former Ruzhi 

shu text) — which just gives us the southern and western limits of the 月氏 country. The northern limit is always left out: in that direction 

the 月氏 had a 

common border with the Xiongnu which was not too well defined, it seems. 

In other words, when the 

Shiji and the Hanshu report that the oblong corridor of 

la nd between the Yellow River and the Salt Swamp 鹽澤 or Lopnor was empty because 

the Xiongnu had first annexed this corridor and later evacuated it again, then we may 

realize that we have here the clearest definition of the former lands of 

the Ruzhi 月氏 

a nd their western neighbors, t he Wusun 烏孫 — the latter, therefore, should have lived 

between Dunhuang 焞煌 and the Lopnor. 

The Wusun are the main topic of the above two excerpts from the Shiji and the 

Hanshu. These two books provide us with the earliest known bits of 

history of this 

small nomadic nation. As far as I know, the Wusun are not mentioned in any Chinese 

text older than the Shiji. It is curious that the Shiji and the Hanshu tell us the genesis 

of the Wusun in two sharply 

different versions. In the Shiji Zhang Qian says that the 

Wusun, at an unspecified time, had been attacked and routed by the Xiongnu. But in 

— 84 —

月氏 wh 

the Hanshu the same Zhang Qian tells us that it had been the Ruzhi o fell 

upon the Wusun and killed their king. In the older book, the Wusun live west of the 

Xiongnu, but in the later one, they lived amongst the 月氏 . Which book 

is reporting the 

true facts ? All told, the Hanshu is much better informed about the Wusun than the 

Shiji. With this, we should be safe to assume that the Hanshu is not correcting the Shi- 

ji here, but Zhang Qian — who had told his story in a way whic h suited his aims. If the 

Wusun had been maltreated by the Xiongnu, they should bear the Xiongnu a grudge, 

and the Han should be able to win the Wusun as allies against the Xiongnu. Zhang 

Qian, in about 119 BCE, wished to be sent out again as envoy, now to 

the Wusun. He 

told his story accordingly — and it worked. Some two hundred years 

later, the Han- 

it quietly 

overwrites the Shiji without any discussion. In every respect, the later story makes bet- 

ter 

shu, for the sake of historical truth, lets Zhang Qian report the true story: 

sense. 

It is in the Hanshu that we first hear of the Saiwang 塞王 , the Sakaraukai or Sacaraucae 

of our Western sources. The Ruzhi 月氏 , trekking west, somewhere out there in 

the west clash with the Saiwang, and the Saiwang were forced to move “far to the 

south.” But what is more important: the Hanshu goes on to tell us why, where and 

when the Wusun 烏孫 — the “Grandsons of the Raven” — now attacked the 月氏 . 

This is the decisive piece of information. The 月氏 , newly settled in the old lands of the 

Saiwang, are not permitted to remain where their first trek had ended. Beaten by the 

Wusun — who were commanded by the young kunmo 昆莫 , or king, Lie–jiao–mi 獵驕 

靡 , a great leader —, it is now the 月氏 who are forced to move, and this time we are 

told where to: namely into the country of the Daxia 大夏 , or far to the southeast. And 

it is the Wusun who now settle for good in the old lands of the Saiwang, freshly evacuated 

by the 月氏 . When soon afterwards the Xiongnu chanyu dies (Laoshang, in late 

161 BCE), the young Wusun kunmo feels well settled in his new lands and, in consequence, 

strong and powerful enough to shun the Xiongnu court meetings — usually 

taken as a sign of submission. 

The Wusun never crossed the Jaxartes. It was, therefore, a most irritating blunder 

when HULSE WÉ, 1979: 217, translated 昆莫略其眾因留居 as “The K’un–mo despoiled 

the population of Ta Hsia, and then remained there in occupation” — the literal 

meaning being “The Kunmo worsted their population and remained to settle there for 

good.” This “their” 其 refers back to the last sentence: “The Ta Yüeh–chih again fled 

west, moving into the lands of Ta Hsia” 大月氏復西走徙大夏地 . Grammar as well 

as common sense tell us here that the main two antagonists are the Ruzhi 月氏 and 

the Wusun 烏孫 ; so that “their” must refer, not to the far-away Daxia 大夏 , but to the 

Ruzhi 月氏 , newly settled in the old lands of the Saiwang. The printed wrong translation 

made one or two generations of Non-sinologists believe that the Wusun went west 

as far as the Daxia and remained in their lands. They did not. Instead, the Wusun 

never crossed the Jaxartes. 

The Chinese sources, in extenso reproduced and discussed here, thus prove: the 

only nomadic peoples crossing the Jaxartes from east to west in the second century 

BCE are the Saiwang 塞王 and the Ruzhi 月氏 — one after the other, in this order and 

within a few years time. It all happens in the lifetime of Chanyu Laoshang (174–161), or 

to be more precise: the successive two nomadic irruptions from east of the Jaxartes 

into Sogdiana happen after 165 and before 160 BCE, i.e. within less than five years. For 

the next half generation, the fights between the different invaders are confined to Sogdiana. 

Then, another two waves of nomadic irruptions inundate Bactria: in the first, of 

c. 145, the Græco-Bactrian kings are forced to give up the eastern parts of their kingdom, 

called Daxia 大夏 (Tochara) by Zhang Qian, and in the second, of c. 130 BCE, the 

Greeks, bled out, are finally annihilated in the capital Bactra itself. 

With this amount of historical facts well established by the Chinese sources, we 

may return to the Western classical counterparts. For Trogus, the starting point of his 

narrative on the fall of the Græco-Bactrian kingdom is the conquest of Sogdiana by the 

— 85 —

Ruzhi 月氏 , and of Bactra by the Sacaraucae. Trogus reports nothing of the short-time 

occupation of the 月氏 in the upper Ili region and of the Sacaraucae-Saiwang in Sogdiana. 

In consequence, the origins of these two nomadic nations are unknown to him. 

But we have only his prologi, a kind of “ Table of Contents.” His full text on the 

subject — where he may have explained the historical facts in great detail — unfortunately 

is lost. 

Of Strabo we have the main text of his book in beautiful preservation, and so it is 

not surprising that he seems to know more. He tells us that the center of nomadic unrest 

must be looked for beyond the Jaxartes River (modern Syr Darya). From that region 

at the eastern end of the world — originally the homelands of the Sakai, now 

established in Sogdiana — four different nomadic peoples burst forth: the Asioi, the 

Pasianoi, the Tocharoi and the Sakaraukai. The last-mentioned Sakaraukai are one of 

those specific tribes which are also called by their general name Skythai or Sakai, as 

Strabo tells us. Hence, they are the Sakai who are now found living amongst the Sogdians. 

This means, they crossed the Jaxartes first and thus constituted the first wave of 

nomadic irruptions into Sogdiana and later Bactria. In the second wave, therefore, only 

three different peoples crossed the Jaxartes at once: the Asioi, the Pasianoi and the 

Tocharoi. 

Compared with Trogus, Strabo’s starting point takes us one step further back in 

time and space: the Sakaraukai are found in Sogdiana and the other three peoples are 

still located beyond the Jaxartes. Interpreting just Strabo’s text, we would be safe to 

assume, that the Asioi, Pasianoi, and Tocharoi had come from somewhere else and had 

pushed the Sakaraukai out of their original homes and across the Jaxartes. If so, it 

would be logic to assume that the three peoples had not come from the west, but from 

somewhere further north; or further east — for Strabo the latter was terra incognita. 

In 1852: 352, LASSEN reminds us: 

Den Chinesischen Geschichtschreibern, die nicht nur die Geschichte ihres eigenen Landes 

vollständiger und genauer geschrieben haben, als es von irgend einem andern Asiatischen 

Volke geschehen ist, sondern auch die Verhandlungen der fremden Völker mit ihren 

Herrschern und ihre Geschichte, wenn diese eine Beziehung zu der ihres eigenen Volkes 

hatte, treu und sorgfältig aufgezeichnet haben, verdankt es die Nachwelt allein, noch eine 

Kunde von der Völkerwanderung zu besitzen, deren Hauptereignisse hier dargelegt werden 

müssen, weil ihre gewaltige Strömung zuletzt auch Indien erreichte. 

It is a fortunate fact that our terse Western sources on this fateful migration of a 

number of nomad nations — two short prologi by Trogus, four consecutive phrases by 

Strabo — are replenished by our Eastern sources. Here, we have to compile a wide 

range of relevant passages from at least seven of the first seventeen Chinese Standard 

Histories 正史 , or from the Shiji 史記 , the Hanshu 漢書 , the Hou Hanshu 後漢書 , the 

Weishu 魏書 , the Beishi 北史 , the Suishu 隨書 , and the Tangshu 唐書 . Here, firstclass 

information is sometimes found in the most unexpected chapters of these bulky 

books. Scattered bits and pieces must be collated, exploited, evaluated. In this paper, I 

am trying to do just this in a more comprehensive way than has been done so far. 

The final result here is: the Chinese sources, in particular Shiji 123, Hanshu 61, 

and Hanshu 96 greatly extend and confirm what we gather from Trogus and Strabo. 

But at times, these Eastern sources may also contain crucial corrections. The most important 

correction in our context may well be that those nomadic nations which destroyed 

the Græco-Bactrian kingdom north of the Hindukush in the second century 

BCE were, not four, but only two in number: the Sakaraukai or Saiwang 塞王 and the 

A[r]sii or Ruzhi (Yuezhi) 月氏 . This, in fact, is also confirmed by Trogus. 

Strabo’s Pasianoi are a phantom people, not mentioned in Trogus and thus not in 

the lost book of Apollodoros — they are also unknown in the Chinese sources. 

Whereas Strabo’s Tocharoi are mentioned by Trogus and well known by the Shiji, 

Hanshu etc. — but not as a conquering people of nomads, but as a people conquered 

— 86 —

and well settled. With this, we are back to the above question: how did the two names 

Pasianoi and Tocharoi get onto Strabo’s list ? 

Since the very beginning of modern Strabo studies, it has been seen that in the unknown 

name Pasianoi we have the well-known name p-ASIANOI. At a time when Latin 

was still the lingua franca of all learned men in Europe, the French scholar VAILLANT 

speculated that the ethnic name Pasianoi should be amended in such a way that it 

could be understood as the Greek version of Trogus’ Asiani. 

In 1725: 61, VAILLANT writes: 

ARSACES VIII. Artabanus hujus nominis secundus, in avi sui Artabani memoriam ita 

nuncupatus videtur. Hic Arsacis Mithridatis frater minor, & Arsacis Phriapatii filius fuit ultimus. 

Post mortem Phrahatis II, ex fratre Mithridate filii, a Scythis in prælio interempti, 

Rex a Parthis, qui in veram & antiquam Parthiam se receperant, in hoc turbido Parthici 

Imperii statu constituitur; anno ante Chr. 126. V. C. 628. Seleucid. 186. Arsac. 130. 

(Ann. 131.) Scythæ, post insignem victoriam de Parthis relatam, ea contenti, provinciis 

illorum depopulatis, in patriam revertuntur. Justin. lib. 42 cap. 1. 

Interim alii Scythæ, dicti Nomades, Græcos Bactrianæ, quibus jam libertas oblata fuerat 

a Parthis, internecione delent. Strabo lib. 11. pag. 511. De his Scythis Nomadibus maxime innotuerunt, 

qui Græcis Bactrianam ademerunt — Asii PasianÒi, emenda ¾ AsianÒi, vel Asiani, 

& Tochari, ac Sacarauli, vel Sacauraci. 

In “De (his Scythis) Nomadibus maxime innotuerunt ...” we have the Latin translation 

of Strabo’s crucial phrase: M£lista d gnèrimoi gegÒnasi tîn nom£dwn ... And 

we notice in passing that VAILLANT has a Strabo edition before him with the defective 

Sacarauli (Sak£rauloi) in place of the correct Sacaraucae (Sakaraàkai), confirmed by 

the Vatican palimpsest, discovered a century or so later. 

VAILLANT is suggesting that in times, when the Strabo text was being copied in lower-case 

or minuscular script, an original ¾ AsianÒi had become corrupted into PasianÒi. 

The meaning of Greek À 'Asiao…, Latin “vel Asiani,” would be in English: “or (else 

c alled) the Asiani.” With this, the ominous Pasianoi would disappear. Strabo’s list 

would include only three names: the Asioi (or Asianoi), the Tocharoi and the Sakarau- 

However, a corruption ¾ P seemed somewhat far-fetched. In upper-case Greek 

kai. 

letters (majuscular script), the same scribal error would become more plausible. And 

so the French Abbé LONGUERUE, in 1732: 14, rewrites his compatriot in this way: 

Anno A.C. 127. V.C. 627. Seleuc. 185. Arsac. 129. Scythæ contenti victoria depopulata Parthia 

(provinciis Parthorum imperio subditis) in patriam revertuntur. Interim alii Scythæ 

Græcos Bactrianæ, quibus jam libertas adempta fuerat a Parthis, internecione delent. 

Strabo lib. XI. pag, 511. de Scythis Nomadibus maxime innotuerunt, qui Græcis Bactrianam 

ademerunt Asii Græce Pas…anoi (emenda —H 'Asianoˆ vel Asiani) & Tochari & Sacarauli 

(vel Sacauraci). 

With this, the corruption H P would be much easier to accept; and it would have 

occurred earlier: in the first few centuries of transmission when such Greek manuscripts 

were written in capital letters and in the scriptio continua, i.e. wíthout extra 

spacing between the words. The scholarly Abbé could have explained the corruption actually 

as ΗΑCΙΑΝΟΙ becoming misspelled as ΠΑCΙΑΝΟΙ — in this way, the scribal error 

would have made a tiny difference in writing, but a fearful one in meaning, indeed. 

This would be an elegant way to reduce Straho’s list from four to three names. But 

would it be typical for Strabo to mention one name in two different versions ? One authority 

on the subject, LASSEN, does not think so. In 1852: 360, he comments in a note: 

Longuerue in seinen Annal. Arsac. p. 14, und Vaillant, de Arsacid. imper. I, p. 61, haben À 

'Asianoˆ als Emendation vorgeschlagen; s. die Notiz zu der Stelle in der Ausgabe von 

Tzschucke IV, p. 474. Dieser Vorschlag scheint jedoch nicht annehmbar, da Strabon kaum 

die Verschiedenheit der Namen erwähnt haben wird. 

On the same page, LASSEN proposes his own ingenious way to solve the problem: 

— 87 —

Von diesen Wanderhirten sind besonders jene bekannt geworden, welche den Hellenen 

Baktrien entrissen, die Asier, Pasianer, Tocharer und Sakarauler, ausgezogen vom jenseitigen 

Ufer des Iaxartes neben den Saken und Sogdianern, wo [gleichfalls] Saken sassen. 

In a note GROSKURD adds: 

Den umständlichsten Bericht über den Skytheneinbruch hat uns Strabon aufbewahrt. 

Nach ihm waren unter den Nomaden in N. Sogdiana’s diejenigen die berühmtesten geworden, 

welche den Hellenen Baktrien weggenommen hatten, nämlich die Asier, die Pasianer, 

die Tocharer und Sakarauler. Sie waren ausgezogen aus dem Lande jenseits des Jaxartes 

und dem Theile Sogdiana’s, welchen die Saker besassen. Ausser dieser Stelle finden sich 

nur zwei kurze Notizen aus dem Werke des Trogus Pompeius ... Die Verschiedenheit dieser 

Angaben betreffen theils die Zahl der Völker, theils ihre Namen. Strabon führt vier auf, 

Trogus Pompeius dagegen nur drei; seine Sarancae müssen die Sakarauler des erstern 

seyn. Da die Pasianer sonst nirgends vorkommen, möchte es kaum zweifelhaft seyn, dass 

in seinen Text dieser Name aus einer Randglosse, in welcher bemerkt worden war, dass 

die Asier von andern Asianer genannt wurden, durch die Abschreiber eingedrungen ist. 

LASSEN understands Strabo’s text clearly in that way that the nomadic nations who 

took Bactria from the Greeks had come: 

(1) from beyond the Jaxartes (east and north of the river); 

(2) from that part of Sogdiana which was occupied by the Sakas (west of the river). 

Strabo, surely basing himself on Apollodoros, then adds the information that the 

o riginal homes of the Sakai — or, more specific, the Sakaraukai — had been beyond, 

or east and north of, the Jaxartes. With this we know that the first stage of this migra- 

of nations had been the displacement of the Sakai/Sakaraukai from their original 

tion 

seats beyond the Jaxartes into Sogdiana, i.e. to west of the river. And we know that 

those peoples — who had pushed the Sakaraukai west across the river and were now 

living beyond the Jaxartes in the old seats of the Sakaraukai — must originally have 

come from somewhere else. This was an important realization. For shortly before 

LASSEN, GROSKURD, a great interpreter of Strabo, still finds the half-sentence ... ¿n 

kate‹con S£kai difficult and “dark” when, in 1831: 397, he translates the whole sentence 

this way: 

Im Text steht blos: neben den Saken und Sogdianern, wo Saken sassen; ein sehr unver- 

ständiger und dunkler Ausdruck. Zwar wohnten allerdings auch jenseits des Iaxartes Saken. 

Dieses muss aber bestimmter und als Gegensatz ausgedrückt werden. Wahrscheinlich 

fehlt vor kate‹con nur kaˆ aÙt¾n, gleichfalls. 

In 1967: 52, DAFFINÀ, still uncertain how to understand Strabo here, writes: 

Strabone, attingendo probabilmente ad Apollodoro di Artemita, dice che la Battriana fu 

sottratta ai Greci da quattro popoli nomadi: gli ”Asoi, i Pasiano…, i TÒcaroi, e i 

Sakaraàkai, partiti dall’ opposta sponda dello Iaxart±s. Il testo in questo punto è confuso e 

non si capisce bene a quale lato del fiume Strabone intenda riferirsi; probabilmente al lato 

destro, cioè settentrionale, ma è ovvio, in ogni caso, che i quattro invasori non si diedero 

convegno in uno stesso luogo alla stessa ora e che quella di Apollodoro-Strabone è soltanto 

una notizia semplificata e succinta dei loro movimento. 

DAFFINÀ, beyond his problems to understand Strabo forthwith, makes one impor- 

statement: “Obviously the four invaders did not converge in the same place at the 

tant 

same time.” This means that the last-mentioned Sakaraukai crossed the Jaxartes first, 

while the other three peoples occupied the old seats of the Sakaraukai on the far side 

of the river. This is what Strabo says. With LASSEN 1852, and more recently ALT- 

HEIM/STIEHL 1970 (see above, p. 80), we know how this “dark” half-sentence must be un- 

... where the Sakai had lived a long time (be- ... ¿n kate‹con S£kai. 

derstood: 

fore they were driven west into Sogdiana). 

— 88 —

Now, what we have to deduce so painfully from a bare minimum of words in our 

Western sources, we find fully and unmistakably explained in the Chinese Histories. 

Originally, the Sai 塞 or Saiwang 塞王 had their traditional pasture grounds in the 

pper Ili River valley and environs. The East Asian, or mongoloid, Ruzhi 月氏 , a final 

ime bloodily beaten by the Xiongnu 匈奴 in about 165 BCE, decide to escape Xiongnu 

u 

t 

d omination and start to migrate west. Somewhere out there, obviously in the valleys of 

the upper Ili and its tributaries, they clash with the Central Asian, or Indo-European, 

Saiwang, forcing them to escape across the Jaxartes. Meanwhile the new Kunmo 昆莫 , 

or king, of the Wusun 烏孫 — who had been a new-born child when the 月氏 had at- 

and beaten the Wusun, killing their old king Nan-tou-mi 難兜靡 — has grown 

tacked 

up and now asks the chanyu or emperor of the Xiongnu, Laoshang 老上 , the permis- 

sion to avenge his father. Before the death of this chanyu, occurring late in 161 BCE, he 

attacks the 月氏 , triumphs over them, and it is now the turn of the 月氏 to cross the 

Jaxartes into Sogdiana. There, they clash a second time with the Saiwang. A second 

time the Saiwang have to move, this time to the south. 

It all fits well enough with what Trogus and Strabo tell us of the Sacaraucae/Sakaraukai 

and the Asiani/Asioi. In the more coherent story of the Chinese sources, however, 

there is nowhere any mention of the Pasianoi — or of the Tocharoi as an invading 

nomad nation. This makes it certain that these phantom or real peoples have no 

place in the incursions of nomadic nations which eventually destroyed Greek Bactria. 

Strabo should have told us: First the Sakaraukai crossed the Jaxartes from East 

to West and shortly thereafter the Asioi. For Trogus has only these two invading peoples. 

Our two Western authors of Augustean time copy from the same lost book of 

Apollodoros of Artemita. With this, the last vexing question to solve here is: How did 

the superfluous two ethnic names Pasianoi and Tocharoi get onto Strabo’s list ? By logical 

deduction, we can be sure now that they have not been there when Strabo wrote 

his Geography. 

We have seen that LASSEN considers it unlikely that Strabo should have bothered to 

m ention a particular name in two different versions. He prefers to think that the name 

Pasianoi was originally just a marginal note — explaining that the Asioi were elseere 

called Asianoi: h asianoi — and this note later slipped into the main text as pa- 

wh 

sianoi. Thev erysame could have happened to the name Tocharoi in Strabo’s list. GROS- 

KURD’s careful Index to Strabo’s Geography shows, 1834: 439, that this name, too, ap- 

in the whole huge work only once, i.e. on this page 511 (the pagination of Casau- 

pears 

b onus’ edition of 1620), and so may also have been just a note in the margins — to the 

e ffect that in another work this name was found mentioned in connection with the fall 

of Greek Bactria — which a later copyist adopted into the main text, together with the 

corrupt Pasianoi. This sounds all very convincing. However, our oldest Strabo codex, 

the Vatican palimpsest, squarely disproves such reasoning. 

Daffinà, 1967: 52–53, remarks already : 

Nella prima metà del Settecento il Vaillant congetturò, difatti, che ”Asioi kaˆ Pasianoˆ 

fosse corrompimento di un originario ¥sioi À ¢siano…. Lasciando stare che la lettura tradizionale 

viene ora confermata dal palinsesto vaticano e perciò ha il consenso dei codici tutti, 

il restauro consigliato dal Vaillant, ancorché accettato e difeso fino a non molti anni fa, 

non è tanto paleograficamente impossibile, quanto intrinsecamente ingiustificato. In effetto 

esso è stato sostenuto soprattutto perché dei Pasianoi non si sapeva bene che fare. 

The Vatican palimpsest, very fragmentary though it is, shows Strabo’s most important 

phrase on Casaubonus’ page 511 in an immaculate state of preservation. The list of 

conquering nomad nations has all four traditional names. On the discovery of the palimpsest’s 

69 (of once 462) folia, ALY writes, 1929: 3–5 : 

In dem Kasten, in dem bislang die Überreste des Vat. Gr. 2306 aufbewahrt wurden, befindaß 

er diese Handschrift am 16. März 1844 im 

det sich ein Zettel von der Hand Angelo Mais, 

römischen Kunsthandel erworben hat. Sie enthält von zweiter Hand Stücke des Penta- 

— 89 —

teuch; darunter standen von einer Hand des 6. Jahrh. “o piu antica” Teile von Strabons 

Geographie. Des außerordentlichen Wertes dieses Fundes wurde man sich jedoch erst be- 

wußt, als der Basilianer Pater Cozza-Luzi in der Bibliothek des griechischen Klosters Grotta 

Ferrata in den Albaner Bergen unter Pergamentresten von Handschriftenkustoden drei 

ebenfalls doppelt beschriebene Blätter fand, die offensichtlich aus derselben Strabonhandschrift 

stammten und nach der oberen Schrift zu urteilen entweder derselben Handschrift 

angehörten, die Mai erworben hatte, oder einer sehr ähnlichen. Seit dem Jahre 1875 hat 

sich (Guiseppe) Cozza-Luzi eingehend mit der Entzifferung der Strabonreste beschäftigt 

und nach einer kürzeren Voranzeige von 1875 seine Ergebnisse in sieben Teilen von 1884– 

1898 veröffentlicht ... Schon die Anordnung der Schrift in drei Kolumnen beweist, daß wir 

es mit einer Handschrift von beträchtlichem Alter zu tun haben ... 

Alle diese Merkmale zusammengenommen empfehlen als Datierung den Anfang des 6. 

Jahrhunderts, wenn nicht gar die zweite Hälfte des 5. Jahrhunderts. Die Schrift ist außerordentlich 

regelmäßig und fest. In Verbindung mit dem sehr feinen Pergament zeigt sie an, 

daß wir keine übliche Handelsware vor uns haben, sondern eine gute sorgfältige Abschrift. 

Sie hat einst den ganzen Strabon enthalten. Außer einem Blatte des 1. Buches stammen 

alle Blätter aus dem 8.–17. Buche ... 

Nachdem ein paar Blätter des Strabon in Grotta Ferrata gefunden sind, kann man mit 

einiger Wahrscheinlichkeit sagen, daß auch V 1 und V 2 dort gewesen sind. Man begreift wenigstens 

dann, wie V 2 in den römischen Kunsthandel gekommen ist. Die Bibliothek von 

Grotta Ferrata stammt zum großen Teil aus Süditalien ... 

Andererseits beweist die wesentlich spätere Hand von V 2 , daß nicht etwa V 1 von fern 

her importiert sein kann, sondern daß der Strabon dort irgendwo in Kalabrien aufgelöst 

wurde und sein Pergament zu mehreren anderen Handschriften verwendet wurde. Da 

liegt es nahe, an das Bistum Rossano zu denken, in dessen unmittelbarer Nachbarschaft 

im Laufe des 6. Jahrhunderts die ersten Eremitenzellen entstanden waren ... 

Ordnen sich aber die Tatsachen soweit sinnvoll zusammen, so dürfte es unzweifelhaft 

sein, daß auch der etwas ältere Strabonkodex nicht in Rossano geschrieben ist. Ich möchte 

auf seine Schrift ein Wort Mercatis anwenden, das er bei einer anderen Gelegenheit gesagt 

hat: “non si scrive cosi bene nella provinzia.” Und das Konstantinopel Justinians ist es, 

das uns die ersten Zeugnisse einer Bekanntschaft mit Strabon in dem Lexikon des Stephanos 

liefert. 

ALY published the Strabon palimpsest — in transcription and facsimile — 1956 in 

Rome, and in the language which had been spoken in the eternal city for a thousand 

years. From his book’s introduction, p. V–IX, we get a few additional details: 

In 69 foliis membranaceis duorum Bibliothecae Apostolicae Vaticanae codicum, quos antiquo 

more palimpsestos vel rescriptos vocamus, sub scriptura mediaevali vetustissima 

scriptura Graeca uncialibus litteris composita latet, qua commentarios geographicos Strabonis 

tradi iam dudum cognitum est ... 

1 

Is nunc in Bibliotheca Vaticana sub nº 2306 asservatur (V ). Eum antea in conventu ordi- 

nisi litterae abundantes modulo minore exarantur, interdum paulo angustiores scribuntur. 

nis S. Basilii Cryptensi, qui hodie Grottaferrata audit, fuisse tribus foliis comprobatur, 

quae ad hunc diem vitreis munita in Bibliotheca conventus in memoriam antiquae possessionis 

spectantur (V c ). Jam Cardinalis Pentateuchum quidem inesse, Strabonis autem 

›Geographica‹ litteris unicialibus conscripta subesse vidit ... 

Numeri versum non variantur. Scriptura aequalis et elegantissima. Spatia litterarum 

praeter terminationes versuum locis diversissimis operis non mutantur; in finibus versum, 

Quamquam linea subsidiaria ad capita litterarum finienda nusquam conspicitur, tamen 

altitudinis litterarum summa est aequalitas, quasi pertica emensa esset. Attamen, qui 

scripsit, munere suo non tarde aut nimis curiose functus est, nam lineolae, quibus in finibus 

versuum littera N significatur ... 

— 90 —

ALY 1956: 69 Vat. Gr. 2061 A, fol. 281, r. I, lin. 12–25 (V 1 40.54) 

In his chapter De contextus virtutibus vitiisque palimpsesti, ALY, 1956: 196, adds 

the following commentarius criticus to the famous sentence of Strabo (11.8.2; Cas.511). 

The meaning of “L” is: all later codices. 

281 r I 19/20 = 8,2 p. 462,11 C A K A P A Υ Κ Α Ι : 

T A X A P O I 

L sak£rauloi kaˆ; kaˆ del. Groskurd. 

L quoque exhibent. 

22 = 8,2 p. 462,12 T O Υ T Ι Α Ξ A P T O Υ : L toà „ax£rtou. 

24 = 8,2 p. 462,13 C O Γ Δ O A N O Y C : L sogdianoÚj. 

We see here that the Strabo Palimpsest Vat. Gr. 2061 A (V 1 ) — folio 281, recto, column 

I, lines 12–25 — very clearly shows the much-quoted sentence we are concerned 

with in this paper: 

Vat.Gr. 2061 A (V 1 ), 281, r I, 12–25: 

MAΛICTA / ΔEΓNωPIMOIΓEΓO / 

NACINTωNNOMA / ΔωNOITOΥCEΛΛH / 

NACAΦEΛOMENOI / THNBAKTPIANHN / 

ACIOIKAIΠACIANOI / KAITAXAPOIKAICA / 

KAPAΥKAIOPMHΘEN / TECAΠOTHCΠEPAI / 

ACTOΥTIAΞAPTOΥ / THCKATACAKACKAI / 

COΓΔOANOΥCHN / KATEIΧONCAKAI / 

Transcription: 

M£lista / d gnèrimoi gegÒ- / 

nasin tîn nom£- / dwn oƒ toÝj “Ellh- / 

naj ¢felÒmenoi / t¾n Baktrian»n / 

”Asioi kaˆ Pasianoˆ / kaˆ T£caroi kaˆ Sa- / 

karaàkai, Ðrmhqšn- / tej ¢pÕ tÁj pera…- / 

aj toà Tiax£rtou / tÁj kat¦ S£kaj kaˆ / 

SogdoanoÚj, ¿n / kate‹con S£kai. 

When we consider the fact that Strabo’s Geography remained practically unknown 

in the first few centuries of its existence, we may well have here, in this palimpsest of 

the fifth century CE, the well-preserved text of the editio princeps of the first century 

C E. If this assumption is correct, Strabo himself wrote Tiaxartes in place of Iaxartes 

and Sogdoanoi in place of Sogdianoi — mistakes which were corrected in later codices. 

We also see here that the correct Western name of the one tribe of the great Saka 

Federation, located originally beyond the Jaxartes, was Sakaraukai < *saka rauka-. 

— 91 —

fective Sakarauloi back to Sakaraukai. A few modern authors understood this correctly. 

But GROSKURD in the 19th century mistakenly deleted kai and kept Sakarauloi 

— in fact, he should have deleted -loi instead, to get Sakaraukai. Trogus originally 

must have had Sacaraucae — which name later copyists corrupted into Sarancae. 

The Chinese equivalent to Saka is Sai 塞 (ancient pronunciation sak) and for Saka-raukai 

it is Sai–wang (Sak–wang) 塞王 . 

Later Strabo codices had shown Sakarauloi kai, constituting an effort to correct a de- 

But of the greatest interest in the palimpsest’s rendering of our terse, desperately 

important sentence is that Strabo originally wrote, not Tocharoi, but Tacharoi. This, 

apparently, was no mistake, for all later codices invariably also show Tacharoi, as ALY 

states in his commentarius criticus, quoted above. It is very curious, therefore, that all 

modern editions (and all translations) of Strabo’s Geography just as unanimously 

seem to have Tocharoi: MEINEKE 1877: 718; JONES 1928: 260; LASSERRE 1975: 83; RADT 

2004: 340 are the ones I have before me here. The only explanation for this “falsification” 

I can think of is that the accepted name of later times was Tocharoi and Tocharistan. 

Today, therefore, we must take note again of the fact that the earliest Western representation 

of this important ethnic name was Greek TAXAPOI (Tachari). In this way, 

as the newly published Vatican Palimpsest shows, Strabo copied the name from Apollodoros’ 

book. If so, the lost book of the Roman historian Trogus, too, should have 

shown an original Reges Tacharorum Asiani. Only later on, this name changed to Tochari/TOXAPOI. 

How very helpful it should be if Trogus’ magnum opus would surface 

somewhere — or at least the opusculum of Apollodoros. 

Under these circumstances, it is of particular interest to note that this change of the 

first vowel from –a– to –o– is closely paralleled in the Chinese sources. The Shiji and 

Hanshu (i.e. Zhang Qian’s Report) have Ta–hia/Daxia (Wade-Giles/Pinyin) 大夏 . In the 

later Weishu, Beishi, Suishu and Tangshu, this name is changed to T’u–ho–lo/Tuhuoluo 

吐火羅 and a few close variants: Tangshu 221 introduces the spellings Tu-ho-lo/ 

Duhuoluo 吐豁羅 , 睹貨邏 , and T’u-hu-lo/Tuhuluo 吐呼羅 . 

In this, we have a firm additional proof that Daxia 大夏 is the Chinese transcription 

of Tacha( ra); and the la t er Tuhuoluo 吐火羅 is the (improved) Chinese tran- 

scriptio n of Tochara. The Chinese sources even give us a terminus ante quem for the 

distinct change of the initial vowel in this nam e: the Weishu, first of the Chinese 

Standard Histories to introduce the new transcription T’u–ho–lo/ Tuhuoluo 吐火羅 , 

narrates the history of the Wei Dynasty, i.e. history of the late fourth to the mid-sixth 

century CE. 

And in lines 18–20 of column I, recto of folio 281, we have Strabo’s ominous list of 

four names: ACIOIKAIΠACIANOI / KAITAXAPOIKAICA / KAPAΥKAI — ”Asioi kaˆ Pasianoˆ 

kaˆ T£caroi kaˆ Sakaraàkai ... All four are said to be the names of those noriana 

from the Greeks: tîn nom£dwn oƒ toÝj “Ellhnaj 

madic peoples who took Bakt 

¢felÒmenoi t¾n Baktrian»n. And they all are said to have broken loose, like an ava- 

lanche, from the far side of the Jaxartes: Ðrmhqšntej ¢pÕ tÁj pera…aj toà Tiax£rtou. 

With only this one sentence by Strabo 

and the summaries of two of his books 

(chapters) by Trogus, we would be convinced that the Asii, Pasiani, Tachari, and Saca- 

Jaxartes, had inundated Sogdiana and 

raucae had stormed forth from beyond the 

Bactria like a tidal wave, and had flushed the Greeks out in one great sweep. In the 

end, the Asii would rule over the Tachari, and the Sacaraucae would be utterly annihilated. 

Not the slightest shadow of doubt would be cast over this scenario. 

The one remaining question would be: what had happened to the Pasiani ? 

Fortunately, we have no less then seventeen bulky Chinese history books to sift 

through to find out what really happened. The information thus gathered is a hundred 

times that collected from our meager Western sources — in quantity and quality. 

Those history books were composed by men of letters mostly in Chang’an and Luo- 

two great capitals near the Yellow River and thus very far away from Central Asia. 

yang, 

But these far-away historians — Sima Tan, Sima Qian, Ban Biao, Ban Gu and many 

— 92 —

others — could boast to have had an eyewitness on the scene, the great first Chinese 

explorer to reach the Oxus River, in the summer of 129 BCE: their man in Daxia (Tachara), 

Zhang Qian. 

Yet, even with Zhang Qian’s Report to start from, it took the Chinese some seven to 

eight centuries to piece the full story together on the Nomadensturm or nomadic irruption 

which swept away the Greek kingdom of Sogdiana and Bactria, north of the Hindukush 

Mountains. The first detailed information we collect from chapters 110 and 123 

of the Shiji and chapters 61, 94, and 96 of the Hanshu; the last additions and amendments 

we find in Tangshu 221. We are told that four nomadic peoples were involved to 

set this early historic Völkerwanderung into motion: 

— the Xiongnu 匈奴 ; 

— the Ruzhi (Yuezhi) 月氏 ; 

— the Wusun 烏孫 ; 

— the Saiwang 塞王 . 

The Xiongnu and the Wusun never crossed the Jaxartes. They remained practically 

unknown to classic Western historians like Trogus and Strabo. And the Ruzhi 月氏 

would have roamed around the lush pasture grounds of the upper Ili Valley forever, if 

the Wusun kunmo or king had not developed an ardent desire to avenge his father — 

who had been killed by the 月氏 — and in the process had driven the 月氏 across the 

Jaxartes, shortly after the Saiwang 塞王 , evicted from the Ili by the Ruzhi 月氏 . 

Thus, the Chinese historians tell us that only two nomadic peoples broke loose 

like an avalanche from the far side of the Jaxartes — and not together, but in two 

distinct waves: 

— the Sacaraucae/Sakaraukai/Saiwang 塞王 (pushed by the Ruzhi 月氏 ) 

— and the Asiani/Asioi/Ruzhi 月氏 (pushed by the Wusun 烏孫 ). 

These clarifications we owe to the Chinese history books. They contradict Strabo in 

two crucial points: in the number of peoples involved in wresting Bactria from the 

Greeks, and in the number of stages it took to drive the Greeks south across the Hindukush. 

Why, then, did Strabo include the Pasiani and the Tachari in his list ? They do not 

belong there. The Vatican Palimpsest, quoted above, suggests that the list in question 

included the four names from the very beginning. But we know that both Trogus and 

Strabo copied the Bactrian matters from Apollodoros of Artemita. Since Trogus names 

just the Sacaraucae and the Asiani as conquerors of Sogdiana and Bactria, Strabo 

after him must have named just these two peoples as well. 

There is one solution to this dilemma. Strabo, when he died, left his Geography in 

an unpublished manuscript form. It was published some time after his death by an 

unknown editor who was at great pains to get the manuscript ready for publication. 

ALY, who studied Strabo’s live and work with exceptional care, writes in PAULY’s RE, 

1931: 76–154: 

Strabon von Amaseia, stoischer Philosoph, Historiker und Geograph ... 

Das Werk des Apollodoros, das hauptsächlich das östliche Asien nördlich des Tauros behandelt 

hatte und vor 36 v.Chr. entstanden sein wird, scheint Strabon selbständig benutzt 

zu haben ... Diese vielen und umfangreichen Entlehnungen aus Autoren der verschiedenen 

Zeitalter müßten bei Strabon den Eindruck einer einheitlichen Komposition seines 

Werkes empfindlich stören, wenn diese nicht ohnehin fehlte ... Diese Unausgeglichenheit 

und die häufigen losen und unzusammenhängenden Anfügungen kurzer Notizen an das 

Ende besser ausgearbeiteter Abschnitte hat man dadurch zu erklären gesucht, daß Strabon 

sein Werk nicht endgültig überarbeitet habe. Daß letzteres tatsächlich nicht der Fall 

war, pflegt man wohl zutreffend aus der Nichterwähnung Strabons bei Plinius u. a. zu 

schließen ... 

Während die ƒstorik¦ Øpomn»mata (Strabo’s earlier work) im 1. Jahrhundert gelegentlich 

benutzt werden (sicher von Josephus und Plutarch) beginnt die Überlieferungsgeschichte 

der Geographie mit einer großen Lücke. Sie ist nachweislich nicht von Ptolemaios 

— 93 —

enutzt ... Das Bild ändert sich erst, und zwar vollständig, als im 6. Jahrhundert Stephanos 

von Byzanz sein Lexikon Iustinian widmet, in dem er Strabon reichlich benutzt ... 

Aus spätestens der gleichen Zeit stammt der vatikanische Palimpsest, der sich über 

Grotta Ferrata nach Unteritalien, wahrscheinlich nach Rossano in Kalabrien zurückverfolgen 

läßt ... Von der von A. Mai 1844 entdeckten und seit 1875 von Cozza-Luzi und dem Verfasser 

zum großen Teil entzifferten Hs., die einstmals aus ca. 460 Folien bestanden hat 

und in 3 Kolumnen geschrieben ist, sind bisher 68 Folien bekannt geworden, jetzt Vat. Gr. 

2061 A (V 1 ) und 2306 (V 2 ); 3 Blätter noch in Grotta Ferrata (V c ). Abgeschrieben ist Strabon 

in Unteritalien nicht. V dürfte mit der Eroberung Italiens dorthin gekommen sein ... 

Allgemeiner Zustand des Textes: Schon der Palimpsest ist nicht fehlerlos. Seine Fehler 

beschränken sich jedoch auf ausgelassene Zeilen und oberflächliche Unzialverlesungen, 

zumal in den zahlreichen Eigennamen ... 

And again in 1957: 17–21: 

Wenn man daran geht, Irrtümer aufzudecken, dann ist die Einsicht, wie sie entstanden 

sind, der erste Schritt zu ihrer Widerlegung ... 

Was hat Strabon gewollt und was hat er auf Grund der sein Schaffen bedingenden Voraussetzungen 

ausführen können ? Dazu gehört natürlich die Frage nach der Vollendung 

und Publikation seines Alterswerkes ... 

Wie oft ist ihm eingefallen, in einen fertigen Zusammenhang Dinge hineinzufügen, die 

er anderswo nicht unterbringen konnte. Dann werden die Rückbeziehungen der Pronomina 

zweideutig. Zusammenfassend kann man von dem Werke wie von dem Imperium, das 

es darstellte, sagen, dass es ›laborat mole sua‹. Der Verfasser war hochbetagt. Veröffentwird 

oft in 

licht ist das Werk wahrscheinlich erst nach seinem Tode, und der Herausgeber 

Verzweiflung über Nachträge letzter Hand gewesen sein, die sich in den Text nicht einfügen 

liessen. Alles das muss, soweit es beweisbar ist, dem Leser bei der Deutung und Beurteilung 

des Textes gegenwärtig sein. 

We see here that, in thirty years of Strabo studies, ALY developed the firm conviction 

that the old gentleman left an unpublished manuscript at his late death. His earlier 

work, the Ìstorik¦ Øpomn»mata or “Historical Memoirs” — curiously, this title is 

a literal translation of Sima Qian’s book title Shiji 史記 — in 47 books, narrating history 

mainly between 144 and 27 BCE, had been published possibly soon after the latter 

date, in any case in his life time. But his later work, the Gewgrafik£ or “Geographical 

Matters,” became his Alterswerk. When Strabo died high in his eighties, he left a 

manuscript which was finished, but to which the author had continued adding notes in 

the margins until close to his death. If this was so, it becomes easy to see what 

happened in the context of Geography 11.8.2. When Strabo at first had read Apollodoros 

of Artemita’s book, his “list” of conquering nomadic nations contained the same 

two names which before him Trogus had copied from Apollodoros. But some time 

later, when Strabo got hold of a copy of Trogus’ work — likely, this happened shortly 

before his death — he wrote down in the margins, next to 11.8.2, two names not found 

in his own text: 

— ACIANOI (Asiani) as a Latin variant to his own Greek ACIOI (Asii), 

— TAXAPOI (Tachari) as a name wholly new to him — it is mentioned nowhere else in 

his Geography. His manuscript in this particular place may have looked like this: 

HACIANOI 

TAXAPOI 

MAΛICTAΔEΓNωPIMOIΓEΓONACIN 

TωNNOMAΔωNOITOΥCEΛΛHNAC 

AΦEΛOMENOITHNBAKTPIANHN 

ACIOIKAICAKAPAΥKAI 

OPMHΘENTECAΠOTHCΠEPAIAC 

TOΥTIAΞAPTOΥTHCKATACAKAC 

KAICOΓΔOANOΥCHNKATEIΧONCAKAI 

— 94 —

These were just two of an untold number of similar additions or corrections found 

in the margins of his manuscript when Strabo died — before he himself found the 

time to write the final draft for publication. This tough and time-consuming work was 

d one by a later editor who must have been an educated man. All the same, he must 

have had a hard time to decipher the handwriting of the more than octogenarian philo- 

especially when it came to the many unfamiliar names. This editor, we see, 

sopher, 

misread HACIANOI (“or Asiani”) as ΠACIANOI and, unsure what to do with this and 

t he other strange name, finally resolved to include them in Strabo’s “list” of conquering 

nomads. Are we going to blame him for the disastrous effects of his bona fide blunder ? 

I should like to note here that something similar happened to the Hanshu. In the 

capital of the Later Han, Luoyang, the book’s main author, Ban Gu 班固 , died in prisn 

in the sixty-first year of his age (92 CE) before his great work had been finished. The 

o 

Chinese emperor then ordered his widowed younger sister Ban Zhao 班昭 by edict to 

continue and complete the work (see Hou Hanshu 114). This outstanding woman did 

this for the rest of her life, or for some 20 years. SWANN 1932: 69: It is even possible 

that she revised and reedited the entire work. 

Strabo’s Geography was less fortunate than Ban Gu’s Hanshu in that it was nei- 

composed nor finished in the capital — let alone under imperial patronage — as 

ther 

the “Book of the Han” was. The Geography saw the light of day far from Rome and 

without any higher assistance. The exact time and circumstances of its publication we 

c an only guess from the contents of the work itself — as ALY did. Pliny in Rome did 

n ot excerpt from it in the first century CE, Ptolemy in Alexandria did not make use of it 

i n the second. The first to quote extensively from Strabo’s Geography was Stephanos 

i n Constantinopolis in the sixth century CE. The exceptionally early Strabo codex, now 

c alled the Vatican Palimpsest, had already been written by that time and probably in 

t hat same city. Its European sections, extensively used, subsequently perished. But the 

I nner Asian chapters, arousing little interest, survived and were later erased — the 

well-prepared folios were actually written over twice. The regained lowest text, in an 

elegant oblique uncial handwriting and an early arrangement in three narrow columns, 

should be very close to that of the original publication, early in the first century CE. 

In our context, this means that an obvious corruption in sentence 11.8.2 of Strabo’s 

Geography can be traced back to the editio princeps where an unknown editor was 

faced with the problem of working an abundance of marginal notes into the main 

text — most of the time succeeding rather smoothly, we may assume, but in a number 

of instances committing smaller or greater mistakes. 

That in Strabo’s “list” of four conquering nomadic peoples the name ΠACIANOI 

(Pasiani) was a corruption of HACIANOI (À 'Asiano…: “or Asiani”) — and thus could be 

deleted from the list — was recognized early in the 18th century by VAILLANT and LON- 

GUERUE. That the TOCAROI (Tochari), the earliest version of this name being TACA- 

ROI (Tachari), did not belong onto that list either, was half guessed by ALTHEIM in the 

20th century, but he was unable or unwilling to substantiate this line of thinking (see 

above, p. 76). 

Our firm Gewährsmann or informant here, and sole eyewitness known by name, is 

Zhang Qian. In his lost first-hand Report, epitomized by Sima Qian in Shiji 123, he describes 

the Daxia 大夏 as the native population of a country of the same name, well 

settled there since long — and he describes them as recently subjugated by the Ruzhi 

月氏 . To this, Ban G u adds in Hanshu 96A that, just before, the Daxia had been ruled 

briefly by the Saiwa ng 塞王 — as early as 1893 

pointed out by XU SONG 徐松 , reprinted 

in 1900 by WANG 

XIANQIAN 王先謙 , but overlooked by HULSEWÉ/ LOEWE in 1979. 

With this, the wh ole problem centers around the decisive question whether or not 

we are entitled to eq uate Chinese D axia 大夏 with Western Tacha(ra), i.e. the Daxia 

of Zhang Qian with the TACAROI (Tachari) of Strabo, the Tochari of Trogus, and the 

provincial name Tachar in present-day northeastern Afghanistan. If this question can 

be answered in the positive, the Tachari / Tochari have been the inhabitants of Tachara / 

— 95 —

Tochara or Eastern Bactria and they have been well settled there since long. They have 

never been nomads and they have not come from beyond the Jaxartes a short time 

before. Like the phantom Pasiani, they can be deleted from Strabo’s “list” of conquering 

nomads. This leaves just two names on Strabo’s vexed list: the Sacaraucae/Sakaraukai/Saiwang 

塞王 and the Asiani/Asioi/Ruzhi 月氏 — confirmed by Trogus, following 

Apollodoros, and also confirmed by the Shiji / Hanshu. The second of these nomadic 

nations kept chasing the first — intermittently, i.e. in a number of stages — from 

the Ili River in the second century BCE to the Indus in the first century CE. In this process, 

a first climax is the final subjugation of the Sakas in Bactra by the self-styled 

Ru 

zhi 月氏 king Kujula Kadphises 丘就卻 , a good time after the Lugdunum aurei of 

Tiberius went on circulation (see above, p. 52), and before the publication of the Periplus 

(above, p. 64), i.e. some time between 30 and 60 CE. 

PAUL BERNARD, first to recognize the historic showdown between the Sakas and the 

Ruzhi 月氏 in the wall friezes of Khalchayan — unearthed and pieced together by the 

renowned Russian archaeologist GALINA PUGACHENKOVA a short distance north of the 

Oxus River and published by her 1965–1971 —, writes as quoted above (p. 40): 

Ces nomades nous ont d’abord été connus par quelques allusions des textes classiques, 

principalement une phrase de Strabon ... Les sources chinoises, le Si-Ki, le Han Shou et le 

Hou Han Shou parlent, elles, du peuple des Yüeh-chih, à l’exclusion de toute autre. 

The second statement, we have seen above, is now to be modified. But the famous 

French Archaeologist — who excavated Ai Khanum, located in modern Afghan Tachar 

province, which thus preserves the ancient name Tachara / Daxia 大夏 to the present 

day — continues : 

Ces trouvailles nous offrent aujourd’hui une vision incomparablement plus riche et plus 

diversifiée de la culture des deux peuples que nous considérons avoir été les acteurs principaux 

de la conquête de la Bactriane grecque, les Yüeh-chih au centre et à l’est, les Saces 

ou Sacarauques à l’ouest. 

Here, BERNARD is intuitively stating a historic fact of importance: as established in 

this study, our age-old Eastern and Western sources unanimously confirm that only 

two nomadic nations were involved in ending Greek rule north of the Hindukush — 

the Royal Sakas/Saiwang 塞王 and the A(r)sii/ÅrÝi/(A)ruzhi 月氏 . 

Berlin, May 2008 


— 96 —

Bibliography 

ALONSO-NÚÑEZ 1989 ........... 

ALTHEIM 1947–48 ................ 

ALTHEIM / STIEHL 1970 ......... 

ALY 1968 ............................. 

ALY 1957 ............................. 

ALY 1956 ............................. 

ALY 1933 ............................. 

ALY 1931 ............................. 

ALY 1929 ............................ 

AMANTINI 1981 .................... 

AMMIANUS (VEH) ................ 

AMMIANUS (SEYFAHRT) ....... 

AMMIANUS (ROLFE) ............ 

BACHHOFER 1941 ................ 

BAILEY 1985 ....................... 

BAILEY 1979 ....................... 

BAILEY 1952 ....................... 

BAILEY 1947 ....................... 

BAILEY 1936 ....................... 

BANERJI 1908 .................... 

BARTHOLD 1956 ................. 

ALONSO-NÚÑEZ, J. M.: ›The Roman Universal Historian Pompeius Trogus on India, 

Parthia, Bactria and Armenia‹, pp. 125–155 in: Persica. Uitgave van het Genootschap 

Nederland-Iran / Publication de la Société Néerlando-Iranienne, Nr. XIII, 1988–1989 

ALTHEIM, FRANZ: Weltgeschichte Asiens im griechischen Zeitalter. 2 vols., 

Halle (Saale) 1947/1948 

ALTHEIM, FRANZ / STIEHL, RUTH (edd.): Geschichte Mittelasiens im Altertum, 

Berlin 1970 

A LY , W OLFGANG : Strabonis Geographica, recensuit Wolfgang Aly. Volumen primum in 

quo continentur: Praemonenda de nova Geographicorum editione quae vivus 

impressit WOLFGANG ALY. Libri I–II (Prolegomena Strabonis) quos ab editore prelo datos 

iteratis curis perpoliverunt ERNST KIRSTEN et FRIEDRICH LAPP. Praefatiunculam scripsit 

HILDEBRAND HOMMEL, Bonn 1968 

——— : Strabons Geographika in 17 Büchern. Text, Übersetzung und erläuternde 

Anmerkungen von WOLFGANG ALY. Band 4: Strabon von Amaseia. Untersuchungen 

über Text, Aufbau und Quellen der Geographika, (Antiquitas. Reihe 1: Abhandlungen 

zur Alten Geschichte, Band 5), Bonn 1957 

——— : De Strabonis Codice Rescripto, cuius reliquiae in Codicibus Vaticanis 

Vat. Gr. 2306 et 2061 A servatae sunt. Scripsit WOLFGANGUS ALY. Corollarium adiecit 

FRANCISCUS SBORDONE. Accedunt Tabulae XXXVI (Studi e Testi 188), 

Città del Vaticano 1956 

——— : ›Neue Beiträge zur Strabon-Überlieferung‹, von WOLF ALY aus Freiburg i. Br, 

vorgelegt von REGENBOGEN, pp. 3–32 in: Sitzungsberichte der Heidelberger Akademie 

der Wissenschaften, philosophisch-historische Klasse, Band XXII, Jahrgang 1931/32, 

Heidelberg 1933 

——— : ›Strabon von Amaseia, stoischer Philosoph Historiker und Geograph‹, 

coll. 76–155 in: Paulys Realencyclopädie der classischen Altertumswissenschaft, 

2. Reihe, 7. Halbband, Stuttgart 1931 

——— : ›Der Strabon-Palimpsest Vat. Gr. 2061 A‹, herausgegeben von WOLF 

A LY, vorgelegt 

von OTTO GRADENWITZ, pp. 3–42 in: Sitzungsberichte der Heidelberger Akademie der 

Wissenschaften, philosophisch-historische Klasse, Band XIX, Jahrgang 1928/29, 

Heidelberg 1929 

AMANTINI, LUIGI SANTI: Giustino. Storie Filippiche. Epitome da Pompeo Trogo, 

Milano 1981 

VEH, OTTO: Ammianus Marcellinus. Das römische Weltreich vor dem Untergang. 

Sämtliche erhaltene Bücher übersetzt von OTTO VEH, eingeleitet und erläutert von 

GERHARD WIRTH (Die Bibliothek der Alten Welt), 

Zürich / München 1974 

SEYFAHRT, WOLFGANG: Ammianus Marcellinus. Römische Geschichte (Ammiani 

Marcellini Rerum Gestarum). Lateinisch und deutsch und mit einem Kommentar 

versehen von WOLFGANG SEYFAHRT, 3. Teil: Buch 22–25 (Schriften und Quellen der Alten 

Welt, 21 / 3), Berlin 1970 

ROLFE, JOHN C.: Ammianus Marcellinus, with an English Translation by JOHN C. ROLFE 

(Ammiani Marcellini Rerum Gestarum Libri qui supersunt), 3 vols., 

Vol. II: Book XX–XXVI, Cambridge (Mass.) 1940, 2 1948, 

3 1956 

BACHHOFER, LUDWIG: ›On Greeks and ³akas in India‹, pp. 223–250 in: 

Journal of the American Oriental Society (New Haven) 61, 1941 

BAILEY, HAROLD W.: Indo-Scythian Studies. Being Khotanese Texts, Volume VII, 

Cambridge 1985 

——— : ›North Iranian Problems‹, pp. 207–210 in: 

Bulletin of the School of Oriental and African Studies (London), XLII/2, 1979 

——— : ›Six Indo-Iranian Notes‹, pp. 55–64 in: Transactions of the Philological Society 

(Oxford), 1952 

——— : ›Recent Work in Tokharian‹, pp. 126–153 in: Transactions of the Philological 

Society (Oxford), 1947 

——— : ›Ttaugara‹, pp. 883–921 in: Bulletin of the School of Oriental Studies (London), 

Vol. VIII, Part 4, 1935–1937 

BANERJI, R. D.: ›The Scythian Period of Indian History‹, pp. 25–75 in: The Indian 

Antiquary (Bombay), Vol. XXXVII, 1908 

BARTHOLD, WILHELM (VASILIJ V.): Four Studies on the History of Central Asia. Vol. I. 

Translated from the Russian by VLADIMIR FEDOROVIC and TATYANA MINORSKY 

(Russian Translation Project Series of the American Council of Learned Societies, 21), 

Leiden 1956 

— 97 —

BARTHOLD 1928 ................. 

BARTHOLD 1922 ................. 

BAYER 1738 ....................... 

Beishi ............................... 

BERNARD 1991 ................... 

BERNARD 1987 ................... 

BERNARD 1985 ................... 

BERNARD 1973 ................... 

BI¾URIN 1851 ..................... 

BIVAR 1983 ........................ 

BOPEARACHCHI 1997 .......... 




BOYER 1900 ...................... 

BRENTJES 1996 ................ 

BROSSET 1887 .................. 

BROSSET 1828 .................. 

CHARPENTIER 1917 ............ 

CHAVANNES 1907 .............. 

CHAVANNES 1905 .............. 

CHAVANNES 1903 .............. 

CHAVANNES 1895–1905 ...... 

CURIEL/FUSSMAN 1965 ...... 

DAFFINÀ 1967 ................... 

DE LA VAISSIÈRE 2002 ...... 

DE GROOT 1921 /1926.......... 

——— : Turkestan down to the Mongol Invasion, by W. BARTHOLD, Second Edition, 

translated from the original Russian and revised by the Author with the assistance of 

H . A. R. G IBB, London 1928 

——— : Istoriya Turkestana (History of Turkestan, in Russian), Tashkent 1922 

BAYER, GOTTLIEB SIEGFRIED: Historia Regni Graecorum Bactriani in qua simul Graeca- 

in India Coloniarum vetus memoria explicatur auctore Theophilo Sigefrido 

rum 

Bayero, St. Petersburg 1738 

L¯ YÁN–SHÒU (LI YEN-SHOU) 李延壽 (7th cent.): B®i–sh¯ (Pei-shih) 北史 (›History of the 

Northern Dynasties‹), 100 ch., woodblock edition, 32 vols. in 6 cassettes, Taipei 1956; 

punctuated edition, Taipei 1980 

BERNARD, PAUL: ›Sarianidi, V. I.: Xram i nekropol' Tilljatepe. Moscou, 1989 (Le sanctu- 

aire et la nécropole de Tilljatepe, (en russe)‹. Book review, pp. 160–164 in: 

Studia Iranica (Paris) 20/1, 1991 

——— : ›Les nomades conquérants de l’empire gréco-bactrien. Réflexions sur leur 

identité ethnique et culturelle‹, pp. 758–768 in: 

Comptes rendus de l’Académie des Inscriptions et Belles-Lettres (Paris), 

Novembre-Décembre 1987 

——— : Fouilles d'Aï Khanoum. IV: Les monnaies hors trésors. Questions d'histoire 

gréco-bactrienne (Mémoires de la Délégation Archéologique Française en Afghanistan 

XXVIII), Paris 1985 

——— : Fouilles d'Aï Khanoum. I (Campagnes 1965, 1966, 1967, 1968): Texte et Figures 

(Mémoires de la Délégation Archéologique Française en Afghanistan XXI), 

Paris 1973 

BI¾URIN, NIKITA YAKOVLEVI¾ (“IAKINF”) — Áè÷ óðèí, Í. ß. (Èàêèíô): Collection of 

Information on the Peoples who inhabited Central Asia in Ancient Times — Ñîáðàíèå 

ñâåäåíèé î íàðîäàõ, îáèòàâøèõ â Ñðåäíåé Àçèè â äðåâíèå âðåìåíà, 

3 vols. + map on three large sheets, St. Petersburg 1851. 

Republished, with a new introduction by A. N. BERNŠTAM — À. Í. Áåðíøòàì, 

in 3 vols., Moscow 1950–1953 

BIVAR, ADRIAN DAVID H.: ›The History of Eastern Iran‹, pp. 181–231 in: 

EHSAN YARSHATER (ed.): The Cambridge History of Iran. Volume 3.1: The Seleucid, 

Parthian and Sasanian Periods, Cambridge 1983 

BOPEARACHCHI, OSMUND: ›The Posthumous Coinage of Hermaios and the Conquest of 

Gandhara by the Kushans‹, pp. 189–213 in: R. ALLCHIN / B. ALLCHIN / N. KREITMAN / 

in Context. East-West Exchanges at the 

E. ERRINGTON (edd.): Gandharan Art 

Crossroads of Asia, 1997 

——— : ›The Euthydemus' Imitations and the Date of Sogdian Independence‹, pp. 1–21 in: 

Silk Road Art and Archaeology (Kamakura, Japan) 2, 1992 

——— : Monnaies gréco-bactriennes et indo-grecques. Catalogue raisonné (Biblio- 

Issues of Later Indo-Greek Kings‹, pp.79–103 in: 

thèque Nationale), Paris 1991 

——— : ›Graeco-Bactrian 

Numismatic Chronicle (London), Vol. 150, 1990 

BOYER, A.-M.: ›L'époque de Kani•ka‹, pp. 526–579 in: Journal Asiatique (Paris) XV/3, Mai- 

of the Sakas (and other tribes of the Central Asian 

Juin 1900 

BRENTJES, BURCHARD: Arms 

steppes), Varanasi 1996 

BROSSET, LAURENT (ed.): Bibliographie analytique des ouvrages de Monsieur Marie- 

Félicité Brosset, Membre de l'Académie Impériale des Sciences de Saint-Pétersbourg: 

1824–1879, St. Petersburg 1887 

BROSSET, MARIE-FÉLICITÉ: ›Relation du pays de “Ta ouan”; traduite du chinois par 

M. Brosset jeune‹, pp. 418–450 in: Journal asiatique (Paris), Tome II, 1828 

CHARPENTIER, JARL: ›Die ethnographische Stellung der Tocharer‹, pp. 347–388 in: 

Zeitschrift der Deutschen Morgenländischen Gesellschaft, 71. Band, Leipzig 1917 

CHAVANNES, ÉDOUARD: ›Les pays d'Occident d'après le Heou Han chou‹, 

pp. 149–234 in: T'oung Pao (Leiden), série II/VIII, 1907 

——— : ›Les pays d’Occident d’après le Wei lio‹, 

pp. 519-571 in: T'oung Pao (Leiden), série II/VI, 1905 

——— : Documents sur les Tou–kiue (Turcs) occidentaux (Recueillis et commentés 

suivi de ›Notes Additionnelles‹), Paris / St. Petersburg 1903, New York 1969 (reprint) 

——— : Les Mémoires Historiques de SE–MA TS'IEN. Traduits et annotés par Édouard 

Chavannes, 5 vols., Paris 1895–1905, 1967 (reprint, with one additional vol.) 

CURIEL, RAOUL / FUSSMAN, GÉRARD: Le trésor monétaire de Qunduz (Mémoires de la Dé- 

dei Sakå nella Drangiana 

légation Archéologique Française en Afghanistan XX), Paris 1965 

DAFFINÀ, PAOLO: L’immigrazione 

(Istituto Italiano per il Medio ed Estremo Oriente, Reports and Memoirs IX), Rom 1967 

DE LA VAISSIÈRE, ÉTIENNE: Histoire des Marchands Sogdiens (Bibliothèque de l’Insitut 

des Hautes Études Chinoises, vol. XXXII), Paris 2002 

DE GROOT, JOHANN JAKOB MARIA / FRANKE, OTTO: Chinesische Urkunden zur Geschichte 

Asiens. Erster Teil: Die Hunnen der vorchristlichen Zeit. 

Zweiter Teil: Die Westlande Chinas in der vorchristlichen Zeit, Berlin 1921 /1926 

— 98 —

DEGUIGNES 1759 .............. 

DEGUIGNES 1756 .............. 

DONG ZUOBIN 1960 ........... 

DORN’EICH 2002 ............... 

DORN’EICH 2000 ............... 

DORN’EICH 1999 ............... 

DORN’EICH 1998 ............... 

DUBS 1938–55 .................. 

ENOKI 1999 ...................... 

FÃ XIÃN ........................... 

First Four Histories ........ 

FRANKE 1934 .................... 

FRANKE 1930 .................... 

FRANKE 1920 .................... 

FRANKE 1918 .................... 

FRANKE 1904 .................... 

FUSSMAN 1991 ................... 

FUSSMAN 1980 .................. 

GARDNER 1886 ................. 

GARDNER 1877 ................. 

GRENET 2007 ................... 

GUTSCHMID 1888 ............... 

HALOUN 1937 .................... 

HALOUN 1926 .................... 

Hanshu ............................ 

HARMATTA 1994 ................ 

DEGUIGNES, JOSEPH: ›Recherches sur quelques événements qui concernent l'histoire des 

Rois Grecs de la Bactriane, et particulièrement la destruction de leur Royaume par les 

Scythes ...‹, pp. 17–33 in: Histoire de l'Académie Royale des inscriptions et belles-lettres 

(Paris) XXV/2, 1759 

——— : Histoire générale des Huns, des Turcs, des Mogols, et des autres Tartares 

Occidentaux ... Ouvrage tiré des Livres Chinois, et des Manuscrits Orientaux de la 

Bibliothèque du Roi, 2 vols., Paris 1756 

DNG ZUÒ–BÐN (TUNG TSO–PIN) 董作賓 (1895–1963): Chronological Tables of Chinese 

History 中國年歷總譜 , 2 vols., Hongkong 1960 

DORN’EICH, CHRISTOF MICHAEL: HS 96. Ban Gu ∙ Annalen der Älteren Han ∙ Die Westländer 

班固漢書卷九十六上下西域 (Hanshu chapter 96A+B. Original text and 

Western translations by BICHURIN / SCHOTT, CHAVANNES, DE GROOT, FRANKE, HULSEWÉ / 

LOEWE, SPECHT, WYLIE), computer typescript, Berlin 2002 

——— : Neuzeitliche Quellen zur Geschichte der Griechen und Kuschanas in Zentral- 

1738–1999, computer typescript, 

asien und Indien. Auszüge aus der Fachliteratur 

Freiburg / Berlin 2000 

——— : Chinesische Quellen zur Geschichte der Yuè–zhð und Kuschanas (Excerpts 

from the Shiji, Hanshu, Hou Hanshu, Sanguozhi, and Beishi. Original text and Western 

translations by BROSSET, CHAVANNES. DE GROOT, HIRTH, HULSEWÉ/LOEWE, SPECHT, 

WATSON, WYLIE, ZÜRCHER), computer typescript, Berlin 1999 

——— : Orient ‘65. Weltwanderjahre eines Architekten, Band 2: Afghanistan- und 

Indientagebuch, computer typescript, Berlin 1998 

DUBS, HOMER HASENPFLUG: The History of the Former Han Dynasty by Pan Ku. 

3 vols., Baltimore 1938, 1944, 1955 

ENOKI, K. / KOSHELENKO, G. A. / HAIDARY, Z.: ›The Yüeh–chih and their migrations‹, 

pp. 171–189 in: History of Civilizations of Central Asia. Vol. II, Paris 1994, Delhi 1999 

FÃ XIÃN 法顯 : A Record of Buddhist Countries. By FA–HSIEN. Translated from the 

Chinese by LI YUNG–HSI (San Shih Buddhist Institute), 

Peking 1957 

Zhonghua Shuju 中華書局 edition of the First Four (Chinese Standard) Histories 

前四史 in 4 vols.: Vol. I: Sh¯–jì 史記 ; Vol. II: Hàn–shø 漢書 ; Vol. III: Hòu Hàn–shø 

人名地名索引 , 

後漢書 ; Vol. IV: Sån–guó–zhì 三國志 with Index 

Peking 1 1997 

FRANKE, OTTO: ›Grundsätzliches zur Wiedergabe fremder Länder- und Ortsnamen im 

Chinesischen‹, pp. 244–280 in: Sitzungsberichte der Preussischen Akademie der 

Wissenschaften. Philosophisch-Historische Klasse (Berlin), 1934 

——— : Geschichte des Chinesischen Reiches. Eine Darstellung seiner Entstehung, 

seines Wesens und seiner Entwicklung bis zur neuesten Zeit. I. Band: Das Altertum 

und das Werden des Konfuzianischen Staates, Berlin / Leipzig 1930 

——— : ›Das alte Ta–hia der Chinesen. Ein Beitrag zur Tocharer-Frage‹, 

pp. 117–136 in: Ostasiatische Zeitschrift (Berlin), 8. Jahrgang, 1919/1920 

——— : ›Einige Bemerkungen zu F.W.K. Müllers “Tocrï und Kuišan (Küšän)”‹, 

pp. 83–86 in: Ostasiatische Zeitschrift (Berlin), 6. Band, 1918 

——— : Beiträge aus chinesischen Quellen zur Kenntnis der Türkvölker und Skythen 

Zentralasiens (aus dem Anhang zu den Abhandlungen der Königlich Preussischen 

Akademie der Wissenschaften vom Jahre 1904), Berlin 1904 

FUSSMAN, GÉRARD: ›Le Périple et l’histoire politique de l’Inde‹, pp. 31–38 in: 

Journal asiatique (Paris), Tome CCLXXIX / 1–2, 1991 

——— : ›Nouvelles inscriptions Ýaka: ère d’Eucratide, ère d’Azès, ère vikrama, ère de 

Kani•ka‹, pp. 1–43, pl. I–IV in: BEFEO (Paris), LXVII, 1980 

GARDNER, PERCY: The coins of the Greek and Scythic Kings of Bactria and India in the 

British Museum (Catalogue of the Indian coins in the British Museum), London 1886, 

Chicago 1966 (reprint) 

——— : The Parthian Coinage. With Eight Plates (The International Numismata 

Orientalia), London 1877 

GRENET, FRANTZ: ›Nouvelles données sur la localisation des cinq “ Yaghus” des Yuezhi. 

L’arrière plan politique de l’itinéraire des marchands de Maès Titianos‹, 

pp. 325–341 in: Journal asiatique, Tome 294, Numéro 2, 2006, Paris 2007 

GUTSCHMID, ALFRED VON: Geschichte Irans und seiner Nachbarländer von Alexander 

dem Grossen bis zum Untergang der Arsaciden, von ALFRED VON GUTSCHMID mit einem 

Vorwort von TH. NÖLDEKE, Tübingen 1888 

HALOUN, GUSTAV: ›Zur Üe-t•ï-Frage‹, pp. 234–318 in: Zeitschrift der Deutschen 

Morgenländischen Gesellschaft (Leipzig) 91/2, 1937 

——— : Seit wann kannten die Chinesen die Tocharer oder Indogermanen 

überhaupt?, 1. Teil, Leipzig 1926 

BÅN GÙ (PAN KU) 班固 (32–92): Hàn–shø 漢書 (›Annals of the [Former] Han Dynasty‹), 

100 ch. (main commentator YAN SHI–GU 顏師古 , 581–645), woodblock edition, 40 vols. in 

6 cassettes, Taipei 1956; punctuated edition, 20 vols., Peking 1 1962, 2 1975 

HARMATTA, JÁNOS: ›Languages and scripts in Graeco-Bactria and the Saka kingdoms‹, 

pp. 397–416 in: History of Civilizations of Central Asia, Vol. II, Paris 1994, Delhi 1999 

— 99 —

HENNING 1960 ................... 

HENNING 1949 ................... 

HENNING 1938 ................... 

HERRMANN 1937 ................ 

HERRMANN 1922 ................ 

HERRMANN 1920 ................. 

HERZFELD 1931/1932 ........... 

HIRTH 1917 ........................ 

HOANG 1910 ....................... 

Hou Hanshu...................... 

HULSEWÉ 1979 .................. 

HUMBOLDT 1843 ................ 

JULIEN 1846 ..................... 

JUNGE 1939 ...................... 

JUSTINUS (RUEHL) ............. 

JUSTINUS (SEEL) ............... 

KAHARMAN 2000 ................ 

KARLGREN 1957 ................. 

KARLGREN 1940 ................. 

KARLGREN 1923 ................. 

KARLGREN 1915–1926 ......... 

KINGSMILL 1882 ................ 

KLAPROTH 1826 ................ 

KLOTZ 1952 ...................... 

KONOW 1934 ..................... 

KONOW 1933 ...................... 

KONOW 1929 ...................... 

HENNING, WALTER BRUNO: ›The Bactrian Inscription‹, pp. 47–55 in: 

Bulletin of the School of Oriental and African Studies (London) XXIII / 1, 1960 

——— : ›The Name of the “Tokharian” Language‹, pp. 158–162 in: 

Asia Major (New Series), Vol.1, Part I, London 1949 

——— : ›Argi and the “Tokharians”‹, pp. 545–571 in: 

Bulletin of the School of Oriental Studies (London) IX/3, 1938 

HERRMANN, ALBERT: ›Tocharoi‹, coll. 1632–1641 in: Paulys Realencyclopädie der 

classischen Altertumswissenschaft, 2. Reihe, 12. Halbband, Stuttgart 1937 

——— : ›Die Westländer in der chinesischen Kartographie‹, pp. 89–386 in 

SVEN HEDIN: Southern Tibet, vol. VIII, part II, 

Stockholm 1922, Delhi 1991 (reprint) 

——— : ›Sacaraucae‹, coll. 1611–1620 in: Paulys Realencyclopädie der classischen 

Altertumswissenschaft, 2. Reihe, 2. Halbband, Stuttgart 1920 

HERZFELD, ERNST: ›Sakastån. Geschichtliche Untersuchungen zu den Ausgrabungen am 

Køh ð Khwådja‹, pp. 1–116 in: Archaeologische Mitteilungen aus Iran (Berlin) IV/1+2, 

Oktober 1931/Januar 1932 

HIRTH, FRIEDRICH: ›The Story of Chang K'ién, China's Pioneer in Western Asia. Text and 

Translation of Chapter 123 of Ssï–Ma Ts'ién's Shï–Ki‹, pp. 89–152 in: Journal of the 

American Oriental Society (New Haven) XXXVII/2, September 1917 

HOANG, PIERRE (BÓ–LÙ) 黃伯祿 (d. 1909): Concordance des chronologies néoméniques 

chinoise et européenne, par Le Rév. Père P. HOANG, du clergé de Nanking 

(Variétés sinologiques No. 29), Chang–hai 1910 

FÀN YÈ (FAN YEH) 范嘩 (398–446): Hòu Hàn–shø 後漢書 (›Annals of the Later Han 

Dynasty‹), 120 ch., woodblock edition, 40 vols. in 6 cassettes, Taipei 1956; 

punctuated edition, 18 vols., Peking 1965 

HULSEWÉ, ANTHONY FRANÇOIS PAULUS / LOEWE, MICHAEL A. N.: China in Central Asia. 

The early stage: 125 B.C.–A.D. 23. An annotated translation of chapters 61 and 96 of the 

History of the Former Han Dynasty, Leyden 1979 

HUMBOLDT, ALEXANDER VON: Asie Centrale. Recherches sur les chaînes de montagnes 

et la climatologie comparée, par A. DE HUMBOLDT, Tome premier, Paris 1843 

JULIEN, STANISLAS: ›Notices sur les pays et les peuples étrangers, tirées des géographes et 

des historiens chinois‹, pp. 228–252 in: 

Journal asiatique (Paris), 4 e série, tome VIII, 1846 

JUNGE, JULIUS: ›Saka-Studien. Der Ferne Osten im Weltbild der Antike‹, pp. 1–115 in: 

Klio. Beiträge zur alten Geschichte, Beiheft 41 (Neue Folge, Heft 28), 

Leipzig 1939 

RUEHL, FRANZ: JUSTINUS MARCUS JUNIANUS: M. Iuniani Iustini Epitoma Historiarum 

Philippicarum Pompei Trogi. Ex recensione FRANCISCI RVEHL. Accedunt prologi in 

Pompeium Trogum ab ALFREDO DE GUTSCHMID recensisti, Berlin 1885 

SEEL, OTTO: M. IVNIANI IVSTINI Epitoma Historiarum Philippicarum POMPEI TROGI. 

Accedunt Prologi in Pompeium Trogum post FRANCISCUM RVEHL iterum edidit OTTO 

SEEL, Stuttgart 1972 

KAHARMAN MUHAN 卡哈尔曼穆汗 : ›A Study on the Names of Saka, Hun, Rouzhi and 

Tielei Tribes‹ 塞、匈奴、月氏、铁勒四名称考 , pp. 24–30 in: 

The Western Regions Studies 西域研究 (Urumqi), No. 4, 2000 

KARLGREN, BERNHARD: Grammata Serica Recensa, 

Stockholm 1957 

——— : Grammata Serica. Script and Phonetics in Chinese and Sino-Japanese, 

Stockholm 1940 

——— : Analytical Dictionary of Chinese and Sino-Japanese, 

Paris 1923, New York 1974 (reprint) 

——— : Études sur la phonologie chinoise (Archives d’Études Orientales, Vol.15, 1–4), 

Leiden/Stockholm 1915– 1926 

KINGSMILL, THOMAS W.: ›The Intercourse of China with Eastern Turkestan and the Adja- 

Century B.C.‹, pp. 74–104 in: The Journal of the Royal 

cent Countries in the Second 

Asiatic Society of Great Britain and Ireland (London) XIV/1, 1882 

KLAPROTH, JULIUS: Tableaux Historiques de l’Asie, depuis la monarchie de Cyrus 

jusqu’à nos jours; accompagnés de recherches historiques et ethnographiques sur 

cette partie du monde; ouvrage dédié à MM. Guillaume et Alexandre de Humboldt, 

par J. KLAPROTH, Paris 1826 

KLOTZ, ALFRED: ›Pompeius Trogus‹, coll. 2300–2313 in: Paulys Realencyclopädie der 

classischen Altertumswissenschaften (Stuttgart), 42. Halbband, 1952 

KONOW, STEN: ›Notes on Indo-Scythian Chronology‹, pp. 1–46 in: 

Journal of Indian History, Vol. XII, Part I, Serial Nr.34, April 1933, 

Madras 1934 

——— : ›War “Tocharisch” die Sprache der Tocharer?‹, pp. 455–466 in: 

Asia Major (Leipzig), IX (Otto-Franke-Festschrift), 1933 

——— : Kharoshƒhî Inscriptions. With the Exception of Those of AÝoka 

(Corpus Inscriptionum Indicarum, vol. II, part 1), 

Calcutta 1929 

— 100 —

KONOW 1920 ...................... 

KOSHELENKO/SARIANIDI 

1992 

KRUGLIKOVA 1977 .............. 

KUWABARA 1916 ................. 

LASSEN 2 1867–74 ............... 

LASSEN 1847–61 ................ 

LASSEN 1838 ..................... 

LEBEDYNSKY 2003 ............. 

LEUZE 1935 ....................... 

LÉVI 1896–97 .................... 

LIU 2001 ........................... 

LONGUERUE 1732 .............. 

LULOFS 1929 ..................... 

LYONNET 1998 ................... 

MÃ DUÅN-LÍN .................... 

MÄNCHEN-HELFEN 1945 ..... 

MÄNCHEN-HELFEN 1938 ..... 

MANNERT 2 1820 .................. 

MARICQ 1958 ..................... 

MARQUART 1901 ................. 

MCGOVERN 1939 ................ 

MITCHINER 1975-76 ............ 

MÜLLER 1918 ..................... 

MUKHERJEE 1970 .............. 

NARAIN 1957 ...................... 

NEUMANN 1837 .................. 

NEUMANN 1833 .................. 

——— : ›Beitrag zur Kenntnis der Indoskythen‹, pp. 220–237 in: 

Ostasiatische Zeitschrift (Berlin), 8. Jahrgang, 1919 / 1920 

KOSHELENKO, G. A. / SARIANIDI, V. I.: ›Les monnaies de la nécropole de Tillia-tepe 

(Afghanistan)‹ (Traduction du texte par P. BERNARD), pp. 21–32 + pl. 1 in: 

Studia Iranica (Paris) 21/1, 1992 

KRUGLIKOVA, IRINA: ›Les fouilles de la mission archéologique soviéto-afghane sur le site 

gréco-kushan de Dilberdjin en Bactriane (Afghanistan), par M me IRINA KRUGLIKOVA‹, 

pp. 407–427 in: Comptes rendus de l’Académie des Inscriptions et Belles-Lettres 

(Paris), Avril-Juin 1977 

KUWA’BARA JITZU’ZÔ 桑原隲藏 : ›Chô Ken no ensei‹ 張騫の遠征 (›The Expedition of 

Zhang Qian‹, in Japanese), pp. – in: 

Zoku shiteki kenkyû 續史的研究 (Tokyo), Feb. 1916 

LASSEN, CHRISTIAN: Indische Alterthumskunde, 

Bonn/Leipzig 1847/1852/1858/1861 (4 vols.); Leipzig 2 1867/1874 (2 vols.) 

——— : Zur Geschichte der Griechischen und Indoskythischen Könige in Baktrien, 

Kabul und Indien durch Entzifferung der Altkabulischen Legenden auf ihren Münzen, 

Bonn 1838 

LEBEDYNSKY, IAROSLAV: Les Nomades. Les peuples nomades de la steppe des origines 

aux invasions mongoles (IX e siècle av. J.-C. – XIII e siècle apr. J.-C.), 

Paris 2003 

LEUZE, OSCAR: Die Satrapieneinteilung in Syrien und im Zweistromlande von 520-320 

(Schriften der Königsberger Gelehrten Gesellschaft. Geisteswissenschaftliche Klasse, 

XI/4), Halle 1935 

LÉVI, SYLVAIN: ›Notes sur les Indo-Scythes‹, pp. 444–484 in: Journal Asiatique (Paris), 

Neuvième Série VIII/3, Novembre-Décembre 1896; pp. 5-42: IX/1, Janvier-Février 1897; 

›Note additionelle sur les Indo-Scythes‹, pp. 26–531: X/3, Novembre-Décembre 1897 

LIU XIN–RU 刘欣如 : ›Migration and Settlement of the Yuezhi-Kushan: Interaction and 

Interdependence of Nomadic and Sedentary Societies‹, pp. 261–292 in: Journal of World 

History (Hawai’i), Vol. 12, Nr. 2, Fall 2001 

LONGUERUE, LOUIS DUFOUR DE: Annales Arsacidarum, auctore Lvdovico dv Fovr de 

Longverve, abbate S. Joannis de Jardo ad Melod. et septem fontium in Therascia, 

Argentorati 1732 

LULOFS, H. J.: Strabo over Scythen. Specimen van antieke Anthropogeographie 

(Geographische en Geologische Mededeelingen), Utrecht 1929 

LYONNET, BERTILLE: ›Les Grecs, les Nomades et l’indépendance de la Sogdiane, d’après 

l’occupation comparée d’Aï Khanoum et de Marakanda au cours des derniers siècles 

avant notre ère‹, pp. 141–159 in: Bulletin of the Asia Institute. O. BOPEARACHCHI, 

C. ALTMAN BROMBERG, F. GRENET (edd.): Alexander’s Legacy in the East. Studies in 

Honor of Paul Bernard. New Series / Vol.12, 1998 

MÃ DUÅN-LÍN 馬端臨 : Wén Xiàn Tõng K¬o 文獻通考 (Encyclopedia of Historical 

Documents, in Chinese), edited by WÁNG YÚN-W÷ 王雲五 , Shanghai 1936 

MÄNCHEN-HELFEN, OTTO: ›The Yüeh-chih Problem Re-examined‹, pp. 71–81 in: Journal of 

the American Oriental Society (Baltimore) 65/2, April-June 1945 

——— : ›Zur Üe-t•ï-Frage‹, pp. 958–959 in: Anthropos (Wien) XXXIII /5–6, 1938 

MANNERT, CONRAD: Geographie der Griechen und Römer (10 vols.). 

Theil IV: Der Norden der Erde von der Weichsel bis nach China, Nürnberg 1795, 2 1820 

MARICQ, ANDRÉ: ›Inscriptions de Surkh-Kotal (Baghlån). La grande inscription de 

Kani•ka et l'étéo-tokharien, l'ancienne langue de la Bactriane‹, 

pp. 345–440 in: Journal Asiatique (Paris), CCXLVI/4, 1958 

MARQUART, JOSEPH: ×rånšahr nach der Geographie des Ps. Moses Xorenac'i. Mit histo- 

und topographischen Excursen (Ab- 

risch-kritischem Kommentar und historischen 

handlungen der königlichen Gesellschaft der Wissenschaften zu Göttingen. Philologisch- 

Historische Klasse, Neue Folge III/2), Berlin 1901 

MCGOVERN, WILLIAM MONTGOMERY: The Early Empires of Central Asia. A Study of 

the Scythians and the Huns and the part they played in world history. 

With special reference to the Chinese Sources, Chapel Hill (North Carolina) 1939 

MITCHINER, MICHAEL: Indo-Greek and Indo-Scythian Coinage (9 vols.), 

London 1975/1976 

MÜLLER, F. W. K.: ›Tocrï und Kuišan (Küšän)‹, pp. 566–586 in: Sitzungsberichte der 

Königlich Preussischen Akademie der Wissenschaften (Berlin), 1918 

MUKHERJEE, BRATÐNDRA NÅTH: ›The Wu–Sun and the Yüeh–Chih‹, 

pp. 127–129 in: Journal of the Asiatic Society (Calcutta), XII/1–4, 1970 

(published 1972) 

NARAIN, ABODH KISHOR: The Indo-Greeks, Oxford 1957 

NEUMANN, CARL FRIEDRICH: Asiatische Studien. Erster Theil, 

Leipzig 1837 

——— : ›Pilgerfahrten Buddhistischer Priester von China nach Indien. Aus dem 

Chinesischen übersetzt, mit einer Einleitung und mit Anmerkungen versehen von 

D. CARL FRIEDRICH NEUMANN, Erste Abtheilung‹, pp. 114–177 in: 

Zeitschrift für die historische Theologie, 3. Band, 1. Stück, Leipzig 1833 

— 101 —

NÖLDEKE 1879 .................. 

OROSIUS (ZANGEMEISTER) 

PELLIOT 1936 .................... 

PELLIOT 1934 .................... 

Periplus (CASSON) ............ 

Periplus (FABRICIUS) ........ 

PETITOT/BERNARD 1975 ..... 

POKORA 1978 .................... 

POKORA 1962 .................... 

PLINIUS (BRODERSEN) ....... 

POSCH 1995 ...................... 

PTOLEMAIOS (RONCA) ....... 

PTOLEMAIOS (MCCRINDLE) 

PTOLEMAIOS (NOBBE) ....... 

PUGA¾ENKOVA 1971 ........... 

PUGA¾ENKOVA 1966 ........... 

POUGATCHENKOVA 1965 .... 

PULLEYBLANK 1995 ........... 

PULLEYBLANK 1991 ........... 

PULLEYBLANK 1970 ........... 

PULLEYBLANK 1962 ........... 

RAPSON 1922 .................... 

RASCHKE 1978 .................. 

RÉMUSAT 1836 .................. 

NÖLDEKE, THEODOR: Geschichte der Perser und Araber zur Zeit der Sasaniden. 

Aus der arabischen Chronik des Tabari. Übersetzt und mit ausführlichen 

Erläuterungen und Ergänzungen versehen von TH. NÖLDEKE, 

Leyden 1879, Graz 1973 (reprint) 

OROSIUS, PAULUS: Pauli Orosii historiarum adversum paganos libri VII. 

Ex recognitione CAROLI ZANGEMEISTER, 

Leipzig 1889 

PELLIOT, PAUL: ›A propos du “Tokharien”‹, pp. 259–284 in: 

T’oung pao (Leiden), Vol. XXXII, 1936 

——— : ›Tokharien et Koutchéen‹, pp. 23–106 in: Journal asiatique, 224/1, 

Paris 1934 

Periplus Maris Erythraei. Text with Introduction, Translation, and Commentary by 

LIONEL CASSON (Greek Original : reprint from HJALMAR FRISK: Le Périple de la mer 

Érythrée, Göteborg 1927), 

Princeton 1989 

Periplus des Erythräischen Meeres Per…plouj tÁj 'Eruqr©j Qal£sshj. Von einem 

Unbekannten. Griechisch und deutsch, mit kritischen und erklärenden Anmerkungen 

nebst vollständigem Wörterverzeichnis herausgegeben von B. FABRICIUS 

(i.e. HEINRICH THEODOR DITTRICH), Leipzig 1883 

PETITOT-BIEHLER, CLAIRE-YVONNE / BERNARD, PAUL: › Trésor de monnaies grecques et 

gréco-bacteriennes trouvé à Aï Khanoum (Afghanistan) / Note sur la signification historique 

de la trouvaille‹, pp. 23–69 + pl. I–VI in: 

Revue Numismatique (Paris) XVII, 1975 

POKORA, TIMOTEUS: ›Pan Ku and Recent Translations from the “Han Shu”‹, 

pp. 451–460 in: Journal of the American Oriental Society (New Haven) 98/4, 

October-December 1978 

——— : ›The Present State of the Translations from the Shih chi‹, pp. 154–173 in: 

Oriens Extremus (Wiesbaden) 9 / 2, Dezember 1962 

PLINIUS SECUNDUS, GAIUS: Naturalis Historiae Libri XXXVII. 

Liber VI, lateinisch–deutsch. Herausgegeben und übersetzt von KAI BRODERSEN, 

Darmstadt 1996 

POSCH, WALTER: Baktrien zwischen Griechen und Kuschan. Untersuchungen zu 

kulturellen und historischen Problemen einer Übergangsphase. Mit einem 

textkritischen Exkurs zum Shiji 123, Wiesbaden 1995 

PTOLEMAIOS, KLAUDIOS: Geographie 6, 9–21. Ostiran und Zentralasien. Teil I. 

Gewgrafik¾ Ùf»ghsij — ›Geografische Anleitung‹, griechisch und deutsch, 

herausgegeben von ITALO RONCA (Istituto Italiano per il Medio ed Estremo Oriente, 

Centro Studi e Scavi Archeologici in Asia: Reports and Memoirs XV/1), Rom 1971 

MCCRINDLE, JOHN WATSON: Ancient India as described by Ptolemy. Being a translation 

of the chapters which describe India and Central and Eastern Asia in the treatise on 

geography written by KLAUDIOS PTOLEMAIOS, the celebrated astronomer, with 

introduction, commentary and index, London 1885, Amsterdam 1974 (reprint) 

NOBBE, CARL FRIEDRICH AUGUST: Claudii Ptolemaei Geographia I– III (Lib. I–VIII). Edidit 

C.F.A. NOBBE, Leipzig 1843–45; 

Hildesheim 1966 (cum introductione a AUBREY DILLER) 

PUGA¾ENKOVA, GALINA ANATOL’EVNA — Ïóãà÷åíêîâà, Ã. À.: The Sculpture of 

Khalchayan — Ñêóëüïòóðà Õàë÷àÿíà (in Russian, with English summary), 

Moscow 1971 

——— : Khalchayan. On the Problem of Early North Bactrian Culture — Õàë÷àÿí. 

Ê ïðîáëåìå õóäîæåñòâåííîé êóëüòóðû Ñåâåðíîé Áàêòðèè, 

Tashkent 1966 

——— : ›La sculpture de Khaltchayan‹, pp. 116–127 + pl. XXXII–XXXV in: 

Iranica Antiqua (Leiden) V/2, 1965 

PULLEYBLANK, EDWIN GEORGE: ›Why Tocharians?‹, pp. 415–430 in: The Journal of 

Indo-European Studies (Washington, DC), Vol. 23 / 3+4, 1995 

——— : Lexicon of Reconstructed Pronunciation in Early Middle Chinese, Late Middle 

Chinese, and Early Mandarin, Vancouver 1991 

——— : ›The Wu–sun and Sakas and the Yüeh-chih migration‹, pp. 154-160 in: 

Bulletin of the School of Oriental and African Studies (London) XXXIII, 1970 

in: 

——— : ›The Consonantal System of Old Chinese‹, pp. 58–144; 206–265 

Asia Major (London) IX / 1, 1962 

RAPSON, E. J.: The Cambridge History of India. Volume I: Ancient India, 


RASCHKE, MANFRED G.: ›New Studies in Roman Commerce with the East‹, pp. 604–681 

(Text), 681–1076 (Anm.), 1076–1361 (Biblio., Addendas, Index) in: HILDEGARD TEMPORINI / 

WOLFGANG HAASE (edd.) : Aufstieg und Niedergang der römischen Welt. Geschichte und 

Kultur Roms im Spiegel der neueren Forschung, Vol. II 9,2, Berlin / New York 1978 

RÉMUSAT, JEAN PIERRE ABEL: 佛國記 Fo¥ Kou¥ Ki ou Relations des Royaumes Boud- 

Exécuté à la fin 

dhiques: Voyage dans la Tartarie, dans l'Afghanistan et dans l'Inde. 

du IV e siècle, par Chy F£ HIAN. Traduit du Chinois et commenté par M. ABEL RÉMUSAT, 

— 102 —

RÉMUSAT 1829 ................... 

RÉMUSAT 1825 ................... 

RICHTHOFEN 1877 .............. 

RINGE 1995 ....................... 

RITTER 2 1837 ..................... 

RTVELADZE 1994 ............... 

SALLET 1880 ..................... 

Sanguozhi ....................... 

SARIANIDI 1985 .................. 

SCHIPPMANN 1980 .............. 

SCHLEGEL 1900 ................. 

SCHOTT 1841 ..................... 

SCHWARTZBERG 2 1992 ....... 

SEEL 1956 ........................ 

Seidenstraße 2007 .......... 

SELLWOOD 1983 ................ 

Shiji ................................ 

Shiji Lexicon 1991 ............ 

SHIRATORI 1941-44 ............ 

SHIRATORI 1902 ................ 

SIEG 1918 ......................... 

SIEG / SIEGLING 1908 .......... 

SIMS-WILLIAMS 1996 ......... 

SIMS-WILLIAMS / CRIBB 1996 

SMITH 1903 ...................... 

SPECHT 1897 .................... 

SPECHT 1890 .................... 

SPECHT 1883 .................... 

ouvrage posthume recu, complété, et augmenté d'éclairissements nouveaux par 

MM. KLAPROTH et LANDRESSE, Paris 1836 

——— : Nouveaux Mélanges Asiatiques. Ou Recueil de Morceaux de Critique et de 

Mémoires, Vol. 1, Paris 1829 

——— : Mémoires sur plusieurs questions relatives à la géographie de l’ Asie Centrale, 

par M. ABEL-RÉMUSAT, Paris 1825 

RICHTHOFEN, FERDINAND FREIHERR VON: China. Band I, 

Berlin 1877, Graz 1971 (reprint) 

RINGE, DONALD: ›Tocharians in Xinjiang. The Linguistic Evidence‹, pp. 439–444 in: 

The Journal of Indo-European Studies (Washington, DC), Vol.23 / 3+4, 1995 

RITTER, CARL: Die Erdkunde im Verhältniß zur Natur und zur Geschichte des 

Menschen oder allgemeine vergleichende Geographie. 

Siebenter Theil. Drittes Buch. West-Asien, Berlin 2 1837 

RTVELADZE, EDUARD VASILEVICH: ›Coins of the Yuezhi rulers of northern Bactria‹, 

pp. 81–96 in: Silk Road Art and Archaeology (Kamakura, Japan) 3, 1993 / 94 

SALLET, ALFRED VON: ›Die Nachfolger Alexanders des Grossen in Baktrien und Indien‹, 

pp. 165-231, 271-411 + pl. IV-X in: Zeitschrift für Numismatik (Berlin) VI, 1879; 

pp. 296-307 in: VII, 1880 

CHÉN SHÒU (CH'EN SHOU) 陳壽 (233–297): Sån–guó–zhí (San–kuo–chih) 三國志 

(›History of the Three Kingdoms‹, in Chinese), 65 ch., woodblock edition, 20 vols. in 

3 cassettes, Taipei 1955; punctuated edition, 5 vols., Peking 1 1959, 6 1975 

SARIANIDI, VIKTOR IWANOWITSCH: Baktrisches Gold. Aus den Ausgrabungen der 

Nekropole von Tillya-Tepe in Nordafghanistan, Leningrad 1985 

SCHIPPMANN, KLAUS: Grundzüge der parthischen Geschichte (Grundzüge 39) , Darmstadt 

1980 

SCHLEGEL, GUSTAVE: ›The Secret of the Chinese Method of Transcribing Foreign 

Sounds‹, pp. 1–32 in: T'oung Pao (Leiden), Série II/I, 1900 

SCHOTT, WILHELM: ›Pater Hyacinth’s Beschreibung der Djungarei und des östlichen 

Turkestan‹, pp. 164–177 in: Archiv der wissenschaftlichen Kunde von Russland (Berlin), 

1. Band, 1841 

SCHWARTZBERG, JOSEPH E. (ed.): A Historical Atlas of South Asia, 

New York 1 1978, 2 1992 

SEEL, OTTO: Pompei Trogi fragmenta, Leipzig 1956 

Ursprünge der Seidenstraße. Sensationelle Neufunde aus Xinjiang, China. 

Ausstellungskatalog, herausgegeben von ALFRIED WIECZOREK und CHRISTOPH LIND, 

Stuttgart 2007 

SELLWOOD, DAVID: ›Parthian Coins‹, pp. 279–298 + pl. 1–9 in: EHSAN YARSHATER (ed.): The 

Cambridge History of Iran. Volume 3.1: The Seleucid, Parthian and Sasanian Periods, 


SÐMÃ QIÅN (SSU-MA CH’IEN) 司馬遷 (145–85): Sh¯jì (Shih-chi) 史記 (Memoirs of the 

Historian), 130 ch., woodblock edition, 48 vols. in 6 cassettes, Taipei 1966; 

punctuated edition, 6 vols., Peking 1959 

CÅNG XIØ–LIÁNG 倉修良 (chief ed.): Shiji Cidian 史記辭典 (Lexicon to the Shiji, in 

Chinese), Jinan 1991 

SHIRA’TORI, KURA’KICHI 白鳥庫吉 : Sai–iki shi ken–kyø 西域史研究 (Studies on the 

History of the Western Regions, in Japanese), 2 vols., Tokyo 1941/1944 

——— : ›Über den Wu–sun-Stamm in Centralasien‹, pp. 103–140 in: Keleti Szemle 

közlemények az ural-altajinép-és nyelvtudomány köréböl (Revue Orientale pour les 

études ouralo-altaïques), Tome III, Budapest 1902 

SIEG, EMIL: ›Ein einheimischer Name für Tocrï‹, pp. 560–565 in: Sitzungsberichte der 

Königlich Preussischen Akademie der Wissenschaften (Berlin), 1918 

SIEG, EMIL / SIEGLING, WILHELM: ›Tocharisch, die Sprache der Indoskythen. Vorläufige 

Bemerkungen über eine bisher unbekannte indogermanische Literatursprache‹, 

pp. 915–934 in: Sitzungsberichte der Königlich Preussischen Akademie der 

Wissenschaften (Berlin), 1908 / 1 

SIMS-WILLIAMS, NICHOLAS: ›Nouveaux documents sur l’histoire et la langue de la 

Bactriane‹ (Texte traduit par FRANTZ GRENET), pp. 633–654 in: 

Comptes rendus, Académie des Inscriptions et Belles-Lettres (Paris), 1996 

SIMS-WILLIAMS, NICHOLAS / CRIBB, JOE: ›A New Bactrian Inscription of Kanishka the 

Great‹, pp. 75–142 in: Silk Road Art and Archaeology (Kamakura, Japan) 4, 1995 /1996 

SMITH, VINCENT A.: ›The Kushån, or Indo-Scythian, Period of Indian History, B.C. 165 to 

A.D. 320‹, pp. 1–64 in: The Journal of the Royal Asiatic Society of Great Britain and 

Ireland (London), January 1903 

SPECHT, ÉDOUARD: ›Les Indo-Scythes et l'époque du règne de Kanichka d'après les 

sources chinoises‹, pp. 152–193 in: Journal Asiatique (Paris), Neuvième Série X/1, Juillet- 

Août 1897 

——— : ›Note sur les Yué-tchi‹, pp. 180-185 in: Journal asiatique (Paris), Huitième Série 

XV/2, Février-Mars 1890 

——— : ›Études sur l'Asie Centrale d'après les historiens chinois‹, pp. 317–350 in: 

Journal asiatique (Paris), Huitième Série II/3, Oct.-Nov.-Déc. 1883 

— 103 —

STAËL-HOLSTEIN 1914 ....... 

STRABO (RADT) ................ 

STRABO (LASSERRE) .......... 

STRABO (JONES) ............... 

STRABO (MEINEKE) ........... 

STRABO (GROSKURD) ........ 

Suishu ............................ 

Tangshu ......................... 

TARN 1938–51 .................... 

THIERRY 2005 ................... 

THOMAS 1906 ................... 

TSCHERIKOWER 1927 ......... 

VAILLANT 1725 .................. 

VAN WINDEKENS 1941 ........ 

VASCONCELLOS 1883 ......... 

VIVIEN DE SAINT-MARTIN 

1850 

WANG GUOWEI 1923 .......... 

WANG XIANQIAN 1900 ........ 

WANG YUEZHEN 1877 ........ 

WATSON 1993 ................... 

WATSON 1969 ................... 

WATSON 1961 ................... 

WATSON 1956 ................... 

WATSON 1886 .................. 

Weishu .......................... 

WYLIE 1881-82 .................. 

XU SONG 1893 .................. 

STAËL-HOLSTEIN, A. VON: ›Koþano und Yüeh-shih‹, pp. 643–650 in: Sitzungsberichte der 

Königlich Preussischen Akademie der Wissenschaften (Berlin) XX–XXI, 1914 

RADT, STEFAN: Strabons Geographika. Band 3, Buch IX–XIII: Text und Übersetzung, 

Göttingen 2004 

LASSERRE, FRANÇOIS: Strabon, Géographie. Tome VIII (Livre XI), Texte établi et traduit 

par FRANÇOIS LASSERRE (Collection des Universités de France), Paris 1975 

JONES, HORACE LEONARD: The Geography of Strabo, with an English translation by 

HORACE LEONARD JONES, in 8 vols., V: Book X–XII, Cambridge (Mass.) 1928 

STRABO: Strabonis Geographica. Recognovit AUGUSTUS MEINEKE, 3 vols., 

Leipzig 1866–1877 

GROSKURD, CHRISTOPH GOTTLIEB: Strabons Erdbeschreibung in siebenzehn Büchern. 

Nach berichtigtem griechischen Texte unter Begleitung kritischer erklärender 

Anmerkungen verdeutscht von CHRISTOPH GOTTLIEB GROSKURD. 4 vols., 

Berlin 1831–1834 

WÈI ZH×NG 魏徵 (580–643): Suíshø 隨書 (›Historty of the Sui-Dynasty‹), 85 ch., 

woodblock edition, 20 vols. In 4 cassettes, Taipei 1956; 

punctuated edition, , Peking 

OØ–YÁNG XIØ 欧阳修 (1007–1072) / SÒNG QÍ 宋祁 (998–1061) : (New) Táng–shø ( 新 ) 唐 

書 (›New History of the Tang-Dynasty‹), 225 ch., woodblock edition, 40 vols. in 7 cas- 

settes, Taipei 1956; punctuated edition, 20 vols., Peking 1975 

TARN, WILLIAM WOODTHORPE: The Greeks in Bactria and India, 

Cambridge 1 1938, 2 1951 (reprint, with Addenda), 

Chicago 3 1985 (reprint, with new introduction and bibliography), 4 1997 (additional bibl.) 

THIERRY, FRANÇOIS: ›Yuezhi et Kouchans. Pièges et dangers des sources chinoises‹, 

pp. 421–539 in BOPEARACHCHI, OSMUND / BOUSSAC, MARIE-FRANÇOISE (edd.): Afghanistan. 

Ancien carrefour entre l’Est et l’Ouest. Actes du colloque international, organisé par 

CHRISTIAN LANDES et OSMUND BOPEARACHCHI au Musée archéologique Henri-Prades- 

Lattes du 5 au 7 mai 2003, (Indicopleustoi. Archaeologies of the Indian Ocean, 3: 

Afghanistan), Turnhout 2005 

THOMAS, F. W.: ›Sakastana‹, pp. 181–216 in: The Journal of the Royal Asiatic Society of 

Great Britain & Ireland (London), January 1906 

TSCHERIKOWER, VIKTOR (ABÎGEDÔR): Die hellenistischen Städtegründungen von 

Alexander dem Grossen bis auf die Römerzeit (Philologus, Supplementband XIX, 

Heft 1), Leipzig 1927, New York 1973 (reprint) 

VAILLANT, JEAN FOY: Arsacidarum Imperium, sive Regum Parthorum Historia. Ad fidem 

Numismatum accommodata. Per J. FOY VAILLANT. Tomus primus, 

Paris 1725 

VAN WINDEKENS, A. J.: ›Huns Blancs et Årçi. Essai sur les appellations du “tokharien”‹, 

pp. 161–186 in: Le Muséon, 54/1–4, Louvain/Leuven 1941 

VASCONCELLOS-ABREU, GUILHERME DE: ›De l'origine probable des Toukhares et de leurs 

migrations à travers l'Asie‹, pp. 165–188 in: Le Muséon, 2, Louvain/Leuven 1883 

VIVIEN DE SAINT-MARTIN, LOUIS: Études de Géographie Ancienne et d’Ethnographie 

asiatique, par M. VIVIEN DE SAINT-MARTIN, Tome premier, Paris 1850 

WÁNG GUÓWÉI 王國維 : Guåntáng jílín 觀堂集林 (Collection of a Wealth of Studies, 

in Chinese), 4 vols., Shanghai 1923, Peking 1959 (reprint) 

Hanshu with 

WÁNG, XIÅNQIÅN 王先謙 (1842–1918): Hàn–shø b÷–zhù 漢書補注 (The 

Supplementary Notes, in Chinese), Changsha 1900, Peking 1983 (reprint) 

WÅNG, YU×ZH×N 汪曰楨 : (Lìdài) chángshù jíyào ( 歷代 ) 長術輯要 (Abrégé de la 

chronologie néoménique des dynasties successives), 1877, 

Taipei 1965 (reprint in: Sìbù bèiyào, z¯bù 四部備要子部 ) 

WATSON, BURTON: Records of the Grand Historian. By SIMA QIAN, translated by BURTON 

WATSON. Han Dynasty I/II (revised edition), 2 vols., New York/Hongkong 1993 

——— : Records of the Historian. Chapters from the Shih chi of SSU–MA CH’IEN, 

translated by BURTON WATSON, New York/London 1969 

——— : Records of the Grand Historian of China. Translated from the Shih chi of SSU– 

MA CH'IEN by BURTON WATSON, 2 vols., New York/London 1961 

——— : Ssu–ma Ch’ien: The Historian and His Work. BURTON DEWITT WATSON. 

Submitted in partial fulfillment of the requirements for the degree of Doctor of 

Philosophy, in the Faculty of Philosophy, of Columbia University, New York 1956 

WATSON, JOHN SELBY: Justin, Cornelius Nepos, and Eutropius, literally translated, with 

notes and a general index, London 1853, 1886 (reprint) 

WÈI SHÕU 魏收 (507–572): Wèishø 魏書 (›History of the Toba-Wei Dynasty‹), 114 ch., 

woodblock edition, 50 vols. in 7 cassettes, Taipei 1956; 

punctuated edition, , Peking 

WYLIE, ALEXANDER: ›Notes on the Western Regions‹ (Translated from the “Tsë¥n Han 

Shoo”, Book 96, Part 1 & 2), pp. 20–73 in: The Journal of the Anthropological Institute 

of Great Britain and Irland (London) X, 1881; pp. 83–115: XI, 1882 

XU SONG 徐松 : Supplementary Notes on the Western Regions chapter of the Hanshu 

(Hanshu, ch. 96) 漢書西域傳補注 , in two parts, 

no place of publication, Preface dated 光緒癸巳 = 1893 

— 104 —

XUÁN ZÀNG (BEAL) ........... 

YARDLEY 2003 .................. 

YASUMA 1932 .................... 

Yizhoushu ....................... 

YULE 1873 ........................ 

ZEJMAL' 1992 ................... 

ZEJMAL' 1983 ................... 

XUÁN ZÀNG (HIUEN TSANG) 玄奘 : Si–yu–ki 西域記 . Buddhist records of the Western 

World (Translated from the Chinese of HIUEN TSIANG, A.D. 629–644, by SAMUEL BEAL. 

2 vols., Trübners Oriental series 45), London 1884, Delhi 1969 (reprint) 

YARDLEY, JOHN C.: Justin and Pompeius Trogus: A Study of the Language of Justin’s 

Epitome of Trogus (Phoenix. Journal of the Classical Association of Canada/Revue de la 

Société canadienne des édudes classiques, Suppl.Vol./Tome suppl. XLI), 

Toronto 2003 

YASU’MA YA’ICHI’RÕ 安馬彌一郎 : ›Gesshi no seihô-idô ni tsuite‹ 月氏の西方動に 

就いて (›On the Western Migration of the Ruzhi‹, in Japanese), pp. 660–666 in: 

Shigaku zasshi 史学雑誌 (Tokyo) 43, 1932 

A Concordance to the Yizhoushu 逸周書逐字索引 (The ICS Ancient Chinese Text 

Concordance Series 前秦兩漢古逐字索引叢刊 , in Chinese), 

Hongkong 1992 

YULE, HENRY: ›Notes on Hwen Thsang's Account of the Principalities of Tokháristán, in 

which some Previous Geographical Identifications are Reconsidered‹, pp. 92-120 in: The 

Journal of the Royal Asiatic Society of Great Britain and Ireland (London) VI, 1873 

ZEJMAL' (ZEIMAL', ZEYMAL'), EVGENIJ VLADISLAVOVI¾: ›Eastern (Chinese) Turkestan on the 

Silk Road, First millennium A.D.: Numismatic evidence‹, pp. 137–177 in: 

Silk Road Art and Archaeology (Kamakura, Japan) 2, 1991 / 1992 

——— : ›The Political History of Transoxiana‹, pp. 232–262 + pl. 19–24 in 

3.1: The Seleucid, 

EHSAN YARSHATER (ed.): The Cambridge History of Iran. Volume 

Parthian and Sasanian Periods, Cambridge 1983 

— 105 —

— 106 —

Fig.1 Letter envelope in memory of Zhang Qian's mission. 

Fig.2 Map Hanzhong to Chenggu, with location of Zhang Qiuan's grave. 

— 107 —

Fig.3 Zhang Qian receives his official emblem, the staff of an envoy, from Emperor Wu. 

Fig.4 The king of the Da-Yuan holds a reception for Zhang Qian at his court in Fergana. 

— 108 —

Fig.5 Zhang Qian in audience with the queen of the Arsi (Yuezhi) on the shores of the Oxus. 

Fig.6 Zhang Qian writing his report to Emperor Wu after his return from Bactria. 

— 109 —

Fig.7 Map showing Central and East Asia, with the empires of the Yuezhi, Xiongnu and Han. 

Fig.8 Modern Chinese newspaper article on Zhang Qian, the founder of the Silk Road. 

— 110 —

張騫 Zhang Qian. The Secret Envoy of Han Emperor Wu in Search of the Arsi (Yuezhi) and the Fall of the Græco-Bactrian Kingdom. (Annotated Compilation of Eastern and Western Sources)

Create successful ePaper yourself

Delete template?

Save as template?