This posting addresses a lingering issue of the Hun attribution. Actually, the issue is quite settled, and lingering comes from from the residue left by early studies. The problem had three components:

Identification of the Huns with very close, but not absolutely convincing “Hsiung-nu” or “Xiong Nu/Xiongnu” recorded in the Chinese annals;
Identification of the Huns' language, primarily in respect to the Eastern Huns and Western Huns, but also in respect to other places where Huns left their tracks;
Identification of the Huns with descendent Türkic and non-Türkic tribes, in linguistical and ethnological aspects.

Because the modern scientific studies coincided with the creation of the modern nationalism and its ugly embodiments under totalitarian regimes, the subject became embroiled in the internal nation-building of the 20th c., and as a consequence the scientific deliberations of the past became propaganda tools for the present, in spite of the ongoing advances in the Turkological studies. In respect to the Eastern Huns' language, E.G. Pulleyblank summed it up in his 1963 work “The most prevalent view nowadays, at least in the west, is probably that the Huns (E.Pulleyblank: Hsiung-nu) were ancestors of the Turks.” With the exception of China and Iran, that view is shared by the rest of the world scientific community, which also includes Russia (and former USSR), even though its political interests were completely identical with the concerns of China and Iran, because its Türkic minority was also dominated by a state-promoted titular ethnicity.

The subject of the Western Huns' language never had a pitch of the Eastern Huns' language, mostly because of the solid historical tradition going back one and a half millennia, a mass of evidence provided by Late Antique and Early Middle Ages sources, and obvious corroborating traces left in a number of disciplines. However, some temptations remained even there, until linguistically put to bed by O.Pritsak's 1982 work on onomasticon of the European Huns. O.Pritsak contained extensive bibliography, which included the names of such luminaries as G. Doerfer, L. Ligeti, G. Németh, P. Poucha, and O. Pritsak.

The subject of the identification of the Eastern Huns and Western Huns, extensively debated in the previous century, and in conclusion firmly established, was closed in 1948 by an unexpected source: a direct evidence discovered in the so-called “Sogdian Ancient Letters”, published by W.B.Henning. W.B.Henning not only closed the issue of Western Huns = Eastern Huns, but also closed the side issues of White Huns, Hunas, Khionites etc. That evidence was uniformly accepted as clenching evidence presented by a third party scholar, and was never directly questioned or challenged. An extract from that work follows.

The subject of the identification of the Huns with the Türkic and non-Türkic tribes is mostly guided by acceptance of many direct indications contained in the Chinese annals. That material is a literary cornerstone of the modern science and is universally accepted, save for finer points of deciphering and interpretation. The volume of the direct indications coming from different times and different sources is unambiguous, illuminating both the genetical descendency and linguistical affiliation of many descending tribes. L.P Potapov summarized the view in his 1969 work “A majority of the modern researchers, especially Soviet researchers, came to the opinion that the Huns were a political association of nomadic tribes, ethnically non-uniform, but with a prevalence in it of the Türkic-speaking tribes.” With few concentrated exceptions, this view not only did not diminish since then, but gain another extensive body of corroborating evidence and recognition with the later development of the modern genetical studies.

As a side issue, in the article “Khwarezmian Language”, on page 431 [495], W.B. Henning makes a loaded statement: “As regards the position of Khwarezmian within the circle of the Iranian languages, the closeness of its relation to Sogdian on the one hand and Ossetic on the other has perhaps been overstressed; some of the most striking resemblances to Sogdian may be due to loans. Some of the sound changes that so far have been neglected may be mentioned here, especially those that show the connection of Khwarezmian with the languages adjoining it towards the south and south-east.” On page 452 [516] W.B. Henning notes “...Khwarezmian strikingly resembles Pashto...”

Chinese hieroglyphic spelling and Arabic script are skipped, with [] indicating placeholders. Page numbers are shown at the beginning of the page NNN (BSOAS 1948) and at the end of the page [NNN] (Reprint). Page breaks in continuous text are indicated by //. Most of diacritics is dropped, to avoid font conflicts. The subheadings in bold blue, bold highlighting, and (Pinyin transcriptions) are added for the posting.

601

Dating

The Sogdian " Ancient Letters ", no doubt one of the most important of Sir Aurel Stein's many finds, have been attributed to the middle of the second century of our era, on the strength of archaeological evidence (Serindia, ii, 671 sqq.). Their editor, H. Reichelt, expressed a mild doubt (Die Soghdischen Handschri/tenresle des Britischen Museums, ii, p. 6), and so did Pelliot in his review of Reichelt's edition (T'oung Pao, xxviii, 1931, 457-463). If the date originally proposed by Sir Aurel Stein (between A.D. 105 and 137, or in 153) could be substantiated, the Letters, which are on paper, would have to be regarded as the oldest paper documents in existence.¹

For some time now I have held the view that the Letters were in fact written at the beginning of the fourth century, to be precise in 312 and 313. I owe thanks to Professor Haloun for explaining the nature of the archaeological evidence, which apparently conflicts with this opinion (i.e. of earlier date). The Sogdian Letters were found together with a large number (about seven hundred) of Chinese documents, of which a seventh part can be dated (see E. Chavannes, Les documents chinois). Those that are written on slips of wood are distributed in this way :

The majority of these wooden slips, those that belong to the time before the alleged invention of paper in A.D. 105, are irrelevant for the purpose of dating the Sogdian paper documents. Only two are later than A.D. 105, and of those two one only is satisfactorily dated, the one of A.D. 137. To argue that the Sogdian Letters must belong to a year in which occupation of the site is attested by the presence of a Chinese document, or to a year earlier than that, is perilous.²

1 Their claim to such eminence has already been widely admitted.
2 It is not as if the document of A.D. 137 were one of a series; it is isolated by a long gap (forty-three years - nearly a life-time) from the last preceding.
[315]
602

Moreover, this argument is deprived of whatever force one may like to accord it by the fact that Chinese paper documents, too, some (three) from the second (?) century, but most of them (eleven) from T'ang times, probably the eighth century, were found in the same area. As in these circumstances archeology cannot help us to reach a conclusion, we shall have to rely on the internal evidence derivable from the Letters themselves.

From paleography we can learn scarcely more than that the Letters are " pretty old ". The cursive ("Uigur") type of Sogdian script, familiar from the inscriptions, documents, and Manichaean books of the eighth, ninth, and tenth centuries, must have reached its final form by A.D. 600; the calligraphic handwriting used in Buddhist books perpetuated the stage attained by the cursive script in about A.D. 500. Somewhat earlier is the writing employed for the legends of the coins ascribed to the " Bukhar-khudahs " ; these legends, which were continued almost unchanged down to the end of the eighth century, were introduced probably in the middle of the fifth. The cursive script of the Letters is older again. In the absence of comparable material we have to confine ourselves to saying that the Letters, to judge by the handwriting, may belong to any period between A.D. 105 and A.D. 400.

The contents of the Letters, on the whole, do little to help us narrow down these limits. We do not know when Sogdians began to establish colonies along the caravan routes leading to China and within China itself, but it is likely that they did so long before the invention of paper; these colonies continued to exist until the tenth century at least. No doubt the agents of the "merchant-princes" of Sogdiana ¹ and the other colonists of Tunhuang or Ku-tsang sent letters home to Samarkand and Bukhara throughout the ages, letters in which they complained of postal difficulties (almost the chief content of the Letters) and the troubled times, listed the latest commodity prices and the current exchange value of silver, gave news of their families, and gossiped about their friends. The antiquated language, in comparison with the other Sogdian material; the preservation of several ideograms lost to later Sogdian ²; the absence of references to Buddhism and Manichaeism, and on the other hand the belief in the Old Iranian religion ³ and in particular the reverence // for the goddess Nanai ¹; the absence as yet of Chinese family names ² ; the apparent existence of cultural relations with the Indians of Shan-shan ³ - all these points go to confirm the estimate (second to fourth centuries) arrived at by the study of the handwriting, but are utterly useless for the purpose of determining the precise date (On Nanai: the name of this Goddess, reputed Iranian because Iranian-speaking people revered her, is a variation of the Türkic word for “mother”, Nana ~ Ani, much like the name for the pra-father is a Türkic word for “man”, adam, which in Türkic, unlike its popular borrowings in Hebrew, Arabic, etc., does not have any more sacral connotations than the word for “horse”. However, the Goddess name Nanai and its variations is used as a proof of Indo-European attribution, even though every Indo-European-lingual kid knows a right Indo-European word for mother, and it is not a Türkic Nana).

1 Cf. Barthold, Turkestan, p. 181.
2 Reichelt's list fp. 2) can be added to : 'MYh " mother ", iii, 12 ; *YMTw " when ", several times in Letter No. ii, regarded as an Iranian word by Reichelt (e.g. 21/2 [*a]( iii srS ' YMTtv MN cnt\_ compared with 6/7 'st iii srS ' YKZ Y MN entry s'r; in spite of the complementary -w it should correspond to k&, which does not occur in the long Letter ii; were it not for this notable absence of kS, one would of course think of c'n'icw; ka8 > kai is = ' YMT also in Pahiavi); KL " all " in iv, 3 (Reichelt iS), 8 (Reichelt m8), distinct from fcS {in KL the tails of the letters are crossed); on MR'Y see below, p. 611.
3 Proved by the personal names and the fiynpt-, i, 10 (cf. J3SOS., VIII, 583 sq.), who may well have been a priest at a local Nanai temple (Letter No. i was written in Tun-huang). Heathenish personal names, however, were only slowly abandoned by the Sogdians when they became Buddhists or Maninhaeans; the monks only of either Church had to take religious names. See e.g. the colophon of P 8 {Buddhists}; the Mahrnamag and Sogdica, 6-7 (Manichaeans).
[316]
603

It is a piece of great good luck that Nanai-vandak, the author of Letter No. ii, went beyond the narrow limits to which the other writers confined themselves and provided his correspondent, Nanai-δvar - one of the "merchant-princes" of Samarkand, a Sogdian Maës - with a brief account of contemporaneous events in China as viewed by a foreigner living in an outlying province of the Empire. In spite of Reichelt's here imperfect translation Pelliot at once recognized the gist of the passage and its importance: a great catastrophe had occurred which compelled the Βγpwr — Chinese Emperor to abandon Srγ = Lo-yang, his capital, which was burnt down. "Il s'agirait d'une de ces destructions comme Lo-yang en subit par exemple en 190; cette annee-la, Tong Tcho brula Lo-yang, et transfera 1'Empereur a Singanfou." However, he concluded, " il serait toutefois premature de vouloir preciser qu'il s'agit bien de cet evenement" (It would be one of those destructions as Loyang suffered in 190; this year, Tong Tcho burned down Lo-yang, and Emperor was moved to Singanfou." "however, it would be premature to state that the events are over") (loc. cit., p. 459).

The statement that Saraγ/Lo-yang was burnt down is of the greatest importance. By itself it is sufficient to invalidate all arguments proffered in favor of dating the Letters in the first half of the second century; for although Lo-yang may have been destroyed more than once, it did not suffer in the period A.D. 105 to 153. As a matter of fact Lo-yang suffered such a fate three times only (between A.D. 105 and T'ang), and each time its destruction presaged the fall of a dynasty: in 190 (Later Han), in 311 (Western Chin [Jin]), in 530-532-534-535 (Northern Wei). These therefore are the only dates to which the Letters can be referred. We can unhesitatingly exclude the events of the sixth century from consideration, on the general grounds given above. This leaves us with 190 and 311. The choice is made easy: in 311 Lo-yang, and with it the power // of Chin [Jin], is destroyed by the Hsiung-nu (Huns): in the Sogdian Letter the Huns are mentioned as the adversaries of the Chinese. As there can thus hardly be a doubt that the Letters were written in or shortly after 311, reference is made in what follows to events pertaining to that period only.¹

1 Sogdica, 7 ; JBAS., 1944, 137 ; SSOAS., XI, 737. Mnuk is said of Anahita, not only in Paikuli, but also in the inscription translated by Sprengling, AJSLL., Ivii (1940), 219 (in both instances the ideogram). Cf. further Korn-Steindorff, Sassanidische 8iegelsteine, plate vi, No. 1621, " Altindische Gemme," inscribed b'nwlcy; the figure (a lady in a somewhat extravagant posture) may represent Anahita/Nanai.
a Two such are known so far, "n = An and y'n = Xan, aee BSOAS., XI, 736. They clearly correspond to ^J- and J|| respectively. Cf. Boodberg, Harvard Journal of Asiatic Studies, i, 1936, 291, n. 32. Proftssor Haloun points out that in T'ang times the Chinese names must be understood to mean "man from Bukhara (An) " and " man from Samarkand (K'ang] ".
3 "The Indians and Sogdians" are mentioned in one breath in ii, 37 (in Lo-yang). Lou-Ian occurs once. The woman's name C't(')ysk, i, 1, ia found also in the Niya documents, Oatisa {-ga, -«ae)- Indian loan-words in the Letters were probably borrowed from the Indians of Lou-Ian. So e.g. Si/fcA = lekha, which was not used in Sogdiana (see the Mugh documents, 24, 25, 38, where n'mk instead). In prst(k) Reichelt recognized prasilia, cf. Niya 721, 6 mepoga prasia i. Here mepoga {left untranslated by Burrow) is prob, Pers. maipuxte (Syr. maipuxta, Ar. maifuxtaj, -buxtaj), and masu potga {see Burrow's note on 225) ia the same, half-translated; wiaipuxte is a sickly-sweet decoction of grape-juice or must with spices added. Cf. Athenseus, 31d-e (i, 57); Yule-Cordier, Marco Polo, i, 84, 153, and notes ; Kempfer, Amoen. Exol., 380 (The reference to the Indians, and Indian borrowing in the Sogdian, without specific attribution to a particular Indian language is interesting for the early 3rd c., when Indian can be defined in hundreds ways; the easiest explanation would be the Kuchean or Tarim dialects, later mislabeled Tokharian).
[317]
604

Letter

It may be convenient to reproduce here the relevant lines of Nanai-vandak's report (5-23, 29-38, and the date 61-3), even though Reichelt's text is on the whole satisfactory :

(Follows transcription of the letter)

1 I am indebted to Professor Haloun, who was ever ready to give of hia time to help with the problems presented by the Sogdian Letters, for supplying precise dates and data for the years before and after 311.
2 [Restored], (incomplete) or (uncertain) letters.—Heading and precise meaning of xwt'ynfi/ xwt'yzfl are unknown. Reichelt's translation is no doubt approximately correct; the derivation proposed by him ia unacceptable.
3 Cf. ZDMG., 90, 1936, 197 aq.
4 The only other restoration to be considered here is [cy]n'nw.
5 Or z'yz'y(c)h. The third and seventh letters (possibly -'-) are doubtful. The word may be incomplete at the beginning.
6 A scrap of paper which belongs somewhere else is stuck over the first half of the -t- and the lower part of L1 in line 17.
7 The lines must be slightly rearranged.
8 kn'x(?i)B. t kn'x(w)8. ?
9 Or jit. Perhaps ['Sn]».
10 Cf. [§ff\npnwh vi, 4 (ao to be restored).
11 Reichelt has km'yS instead. The third letter is certainly n/z, the fifth probably -n.
[318]
605

1 So probably, cf. V-s'c, ywtin-s'c (and Av. daeno.sac-},
1 In 'yncy(y) we can now recognize an older form of Man.Sogd. 'ycy (BBB., 96, b 78; the etymology there proposed must be withdrawn.). Reichelt's translation (" woman ") not merely makes strange sense, but involves a grammatical error (Snc'i in the place of Sn^cft).
3 Or " from him ". The correct translation ia in Reichelt's glossary, not however in his text. Instead of cn one can also say -nc (in euclisis), see line 34. This -»• is an enclitic pronoun of the 3rd pera. sing, and pi, ; it is often devoid of meaning (as datitms eihicus). Cf. here cn " from it " 31; -mnc " by me from them or him " 32; kwy&t-nQ, &ui£S'w;(-w34, kt'mw ZYn 10, kt'rw ZYnll in sentences in all of which the serf-interest ia involved, from the point of view of the subject. When it is doubled, the first is an ethical dative, while the second has the force of a full pronoun ; cf. here 'WZYn-n , . . wfs'y'nt " or . . . took (for themselves) their . . ." 18.
* With some hesitation I have accepted Reichelt's explanation of r&nk as from draiiQ-. One could refer to Skt. dranga " frontier*watch station " (ace. to Stein), Niya dratyya which, however, is " office " ace. to Burrow, SSOS., vii, 509 sq.
6 Cf. Parthian 'tnofyt in the inscr. of Shapur, lines 5 and 12 (in Greek *ai'o>, in Pahlavi 'trswhiy, BSOS., IX, 836). Sogd. w'c-, wyt- " emit, let " ; cf. P 13, 23.
• The second letter of the word I have restored as wy[knt\ " destroyed " is not quite certain. Should one read w\ and restore a finite form of the ate m occurring in line 15 ? Hence," pillaged" ?
7 " Moreover then the [Emperor was taken prisoner and led into captivity] by the [Hu]ns " or some such words may well have stood in this line.
[319]
606

1 The restoration of ]'r'nt £r[ as [w]Ywi Jfcr[t] is merely in the nature of a trial balloon. Unfortunately [wfr'nt does not occur elsewhere. It might belong to the stem of Av. vanta-, varaiBya (and posa.paiti-,pairi-varaya-, Sogd. ptvf(y}rfk, BBB., 78, Sogdica 33 sq. ?). One could compare vi'r'nt krt- with Pahl. wUy'c krty " pillaged " in the inscr. of Shapur, Una 14 (the explanation given in BSOS., IX, 836, has now been fully confirmed by the Parthian and Greek translations discovered since}. The meaning of these words is " to appropriate and tako away movable goods as well as prisoners from enemy territory " ; they take the name of the ravaged country as direct object; they cannot mean " to occupy an enemy country ". " Pillage " and " despoil " are fairly adequate equivalents (on the understanding that human beings are included in the spoils). Contrary to the views put forward by Sprengling, AJSLL,t Ivii, 1940, 368 sq., countries are not, as a rule, " carried into captivity."
2 A place-name. The reading ia very uncertain. See above, p. 604, n. 5.
3 my = me, from Old Iran. imai. Not to be confused with later Sogdian my, 'my, BSOAS., XI, 476, line 19, 736.
4 'zyy myS (Reichelt 'nyy) = zzyi med (as Pers. di-ruz), A classic example of Sogdian -t from Olr. -ah.
5 Sogdian ^epa6, originally = " own, belonging ", presumably conveys the idea of vassalage here.
6 wy'pryt- (vi -\- apa -f- ri£-) scarcely differed much in meaning from pr-yl- (apa + ri£) " left behind, remaining ".
7 Cinastan here seems to mean the region around Khumdan, or at least chiefly that region, but not " China ". That the original sense of that term {" the lands of J<| Ch'in ") should have been preserved till so late a date is surprising.—-The remark added above the line is in the nature of a self-correction, "... could drive the Huns out of Khumdan—-nay, even out of Ciitastan ".
8 My first translation was "... or the other nations possessed themselves of it (= 'Khumdan) ". It may, however, be better to keep " the remnant Chinese " as the subject also of the second half of the clause, and to give full value to the ending of n'pw which is probably meant to indicate the accusative case. The chief difficulty lies in the meaning of n'p- which ordinarily ia " tribe, nation, people ". Tor instance, the Naf-namak " list of nafs " (Sogdica, 8) enumerates names of nations (e.g. parstk " Persian "), not countries (pars " Persia "}. Still, it must be admitted that by translating n'p- as " country " (as was suggested by Reichelt) the sense is considerably improved here (as well as in line 10).
9 Or " freemen ", lit. " son(s} of a noble or freeman ", azat-pidrak refiecting Olr. *azata-puBraka-. Another hitherto unrecognized compound with azdta- is "ztk'r " noble{s)" Fragm. iio, 25 (Reichelt, i, 59).—The passage indicates unexpectedly great strength in the Sogdian colonies in China. If there were one hundred freemen alone in Tun-huang—to whose number one has to add that of their families and slaves—the total Sogdian population of that town can have been scarcely below one thousand souls.
10 It cannot be made out whether Nanai-vandak himself returned with his wife or one of his agents. In the passage not translated here there are several verbal forms in the first person plural; their existence could be adduced as arguing for the assumption, that Nanai-vandak undertook those journeys. Those sentences, however, seem to contain a general complaint of the bad times—of the windy and rainy weather in Reichelt's translation—and do not necessarily refc-r to the travels mentioned in lines 21-3. On balance it is more likely that the traveler of these lines is the one referred to in lines 6/7. Nanai-vandak would hardly have said " it is three years since a Sogdian came from inside ", if he had been that Sogdian himself. The question is important for determining the place where N. wrote the letter.—Lines 23, 2nd half, to 29, 1st half, have been omitted here chiefly because I do not understand them ; also, they do not seem to contribute to the solving of our problem.
[320]
607

1 The implication is : " what happened to the China trade ".
2 par = Av. para " guilt ", Pashto por " debt ". It occurs also as p'rh. So in iii, 35, where Reiehelt read w>VA ; " through the guilt of Farnxund, or through his debts, we have become the servants of Chinamen, I as well as my mother." Further in i, 9, " I live in the deepest distress, without clothes, without food. I try to obtain a loan (xps'm p'rh, R. xwVwi p'nh], but every one refuses to give me one " (nmt, R. " Filz ", is finite verb, from nm- " to agree, engage to do something, submit (lit. bend)", with L' "to refuse", cf. U nm'w i, 6, "I shall refuse," R. " PrUgel" ; nm't S'rt iii, 8 ; nm[t] iii, 11 ; further nm', JBSOAS., XI, 472, line 8).
3 Possibly = Buddh. zrmh, VJ., 399 ; zrm'w'nt, ibid., 213.
4 Probably = " the China branch of your business is completely ruined ; your (former) wealth from it = the China trade is no more ". Scarcely = " you would have no profit from my giving you a fuller account of recent Chinese history ".
5 The name means " slave of (Av.) ASis-va-yuhi ". The name of the goddess is spelt 'rtyw, 'rluy in the later Sogdian, see Orieiiialia, viii (1939), 92.
6 A gap makes it impossible to decide which of the following two meanings should be attri¬buted to the sentence. (I) " A. travelled with a caravan up to Kaean, but when the caravan went on, he had to stay there = in Kacan for six months, because he was ill or imprisoned." (2) " A. was on the road there = with the caravan for six months—an abnormally long time for the distance from K. to Saraγ; it demonstrates the state China was in four years ago."
7 A. and his party, or the caravan of line 35 with which A. travelled ?
[321]
608

Geographical reconstruction

Before discussing the substance of Nanai-vandak's report, we must try to identify the place-names mentioned by him. Nothing much need be said nowadays about Srγ — Saraγ = Lo-yang (near mod. Ho-nan-fu) and 'xumt'n = 'Khumdan (with a Sogdiau prothetic vowel), otherwise Khumdan¹ = Ch'ang-an (mod. Hsi-an-fu). Neither name has so far been explained satisfactorily.² (It was not too graceful for W. B. Henning to skip Saraγ, a Hunnic and Türkic word for royal palace and capital. The Türkic Sarai = palace, like in Sarai-Berke and dozens of other royal capitals' names is obvious. The Sogdian writer used Hunnic word for the capital, and that word was Saraγ. The Türkic word existed 250 years before the Türks were reportedly invented) I am all the better pleased at being able to quote Professor Haloun's views on the origin of one of them :

"Khumdan, I suspect, represents 咸陽 Hsien-yang (Xian-yang), the name of the capital of the Ch'in (Jin) empire since 350 B.C., which lay immediately to the north of Ch'ang-an across the Wei river. Hsien-yang, Middle Chinese yom-lan, was *g'em-diah in Ancient Chinese, which appears to correspond sufficiently closely with xum-dan (as to the final -n [y], cf. Srw'n—Qpoava,—-Tun-huan and the names discussed below under Kc'n and Kmzyn). One would have to assume that this appellation was continued in use and applied to Ch'ang-an when that town supplanted Hsien-yang as the capital (of the Han Empire) in 200 B.C. - but was not the name of Ch'in [Gynstn] also so retained, as the name of the whole country, after the fall of the Ch'in dynasty ?. Did these names, probably both together, reach the Sogdians through intermediaries or result from direct intercourse ? The latter, and with it perhaps not inconsiderable trade, is strongly suggested on the one hand by the influx into China of Indo-Iranian concepts in the course of the fourth and third centuries before our era (see Conrady, ZDMG., 1906 ; Maspero, La Chine antique, 607 sqq.), on the other by the use of Chinese nickel alloys in the Greco-Bactrian coinage of the early second century (Tarn, p. 87)." (A simpler and more linguistic-related "perhaps" is that Sogdian colonies were located in the Hun country, and they used lingua franca terminology of their surroundings)

Together with Saraγ a place 'nkp' is named, line 13. It is evident from the context that this must have been a town comparable in importance to Lo-yang (the capital of the Former Chin, dynasty) itself. From lines 15/6 we learn that 'nkp' and another locality whose name cannot be deciphered (n'yn'ymh 1) were considerable places reached by the Hsiung-nu (Huns) in their campaign of devastation. The passage seems to imply that 'nkp' was destroyed on that occasion, but it is not explicitly said. At any rate, 'nkp', like Lo-yang, " is no more " (line 13). So far the name has not been identified, chiefly, I suppose, because of the misleading spelling which inevitably suggests *Ank-pa, *Ang-pa or the like. Pelliot, loc. cit., 458, operated with *Angpa and could not explain it. I asked several eminent Sinologists what Angpa could be: it is natural that they could not give an answer to a question that had been put so utterly wrongly!

1 Cf. BSOAS., XI, 726.
2 See e.g. Schaader, Iranica, 48 sq. The spelling γwmδ'n, quoted by him from the Sogdian laser, of Karabalgasun (p. 45} does not in fact exist. In his edition of that inscription Hanson gave wiwmδt (p. 21, fragm. 7, line 6), but being at a loss to account for such a word suggested timidly (pp. 36 sq.) that one could perhaps read ywwiδ'ji instead, was sehr fraglich ist. It is out of the question by reason of the apparently assured -S-.—Schseiler quotes neither the Kumitdana of the Fan-yll-tsa-ming, fol. 39fc (Bagchi, Deux lex., i, 1929, 78, 295) nor, of course, the 'xwmt'n of our Letter.
[322]
609

 To find the corresponding Chinese name one has to remember the Sogdian transliteration γ'y 'nkwyn (Reichelt, ii, 70, 34} of 開开 K'ai-yüan (Kai-yuan), Middle Chinese k'ai-ngi^wai, which shows that the Sogdians inconveniently transliterated a foreign initial guttural nasal by 'nk-; no doubt they were unable or too lazy to produce such a sound and so said əŋ or əŋg or even əng instead. As soon as one realizes that 'nkp' represents ŋapa, or rather, as the final -a is likely to be a Sogdian ending, ŋap-, one sees that 鄴 Yeh is meant. Yeh (Ye) (Yecheng 鄴城) (near mod. Chang-te-fu, in Ho-nan, to the north of the Huang-ho) was one of the chief towns of Northern China; several dynasties used it as their capital city (so the Wei, 220-264, as one of several capitals ; the Later Chao, a Hsiung-nu (Hun) dynasty, from 330; the Eastern Wei, a remnant of the T'o-pa, 534-549). The Middle Chinese pronunciation of 鄴 was [ ] (ngi^wo + k'iap) = *ngiap [ŋ(i)ap].

China itself not only appears as cynstn = Chinastan (see p. 606, n. 7), but is also referred to as "inside", cf. Reichelt, p. 13, n. 9; whether the whole of China or merely a certain part of it is included in this term is not clear. Professor Haloun kindly suggests that it may perhaps imitate the Chinese [ ] nei " inside ", an elliptic designation of the Chinese home-country current in contemporaneous sources.

Turning now to the West, we have a firm point in Srw'n = Th^urwan = Ptolemy's θpoαva = Tun-huang. The town next in importance to Tun-huang, from the point of view of the Sogdian merchants—judging merely by the numbers of times the various towns are mentioned in the Letters—was Kc'n, Kc"n. Reichelt thought it might be Kao-ch'ang in the Turfan oasis; to Pelliot this identification seemed plus douteux, sans etre impossible (loc. cit., 460, cf. also T'oung Pao, xxviii, 1931,140); in my view it cannot be maintained. All the passages in the Sogdian Letters point to its situation to the east of Tun-huang, somewhere on the great trade route leading from that town to the Chinese capitals. Letter IsTo. v shows clearly that Kc'n was a great trading centre where the Sogdians disposed of the wares they had carried through the desert from the West and bought the silks and other articles they wished to take to Samarkand ; Kao-ch'ang can scarcely be so described! The Sogdian Letters were found together a little to the west of Tun-huang, on the road to Lou-Ian. Some were written in Tun-huang, one at least (v) in Kc'n. Evidently the letters the Sogdiana in Western China wrote to their home country were gradually brought to the westernmost town, i.e. Tun-huang, where they accumulated until a caravan left for Samarkand. It is unlikely in the extreme that a man sitting in Kao-ch'ang would send a letter to Tun-huang to catch the post to Samarkand—five degrees of longitude to the east.

For all these reasons I have long suspected that Kc'n is [] Ku-tsang = Liang-chon-fu; the Sogdian letter -c- stands for -ch- and -c- (" ts "); as to the finals, cf. δrw]'n : Tun-hu]ang, or y'n : k'ang (above, p. 603, n. 2). At the beginning of the fourth century Ku-tsang was the leading Chinese town to the west of the Huang-ho. It was the residence of Chang Kuei, the governor of the province of Liang (301-314), which at that time included Tun-huang (as well as the oasis of Turfan). Presently it was to become the capital of several // Liang dynasties (Former Liang, 317-376, founded by Chang Kuei's son Shih; Later Liang, 386-403; etc.).
[323]
610

Every Sogdian entering China had to pass through it. Moreover it is known, from Chinese sources that it was a town frequented by Sogdians. In a passage of the Pei-skik, recently treated by O. Maenchen-Helfen, Byzantian, xvii, 1944-5, 226 sqq., we read "Merchants of Su-te (= Soγd) used to go in great numbers to the region of Liang, but on the capture of Ku-tsang (439) all of them were taken prisoners. At the beginning of the period Wen-ch'eng (452-465) the king of Su-te sent an envoy to ask for their ransom, which was granted by cabinet order."

This identification is confirmed by the itinerary Chinanckath (Kao-ch'ang)-Khumdan which was contained in Jaihani's lost work and is known to us from Gardezi and—though not in the form of an itinerary—the Hududu l'Alam (tr. Minorsky, 84-5, 229-230). Here the stage named immediately after Kan-chou is []- (Gard.J, []- (Hud.)—the second spelling, which is the only correct one, is identical with the Sogdian spelling. As the first important town reached after leaving Kan-chou is Ku-tsang, its identity with "[]- cannot seriously be doubted.

Reichelt found not only Kao-ch'ang, but the neighboring oasis of Ha-mi/ Qomul as well mentioned in our Letter, in the guise of km'yδ, line 23. This was approved by Pelliot, loc. cit., 460. One asks oneself in vain why these Sogdian merchants, supposedly engaged in the China trade, should have gone to such outlying starved desert places where nothing worth mentioning was ever produced. Here, however, Keichelt's reading was wrong: it is clearly kmny./kmzy., see above, p. 604, n. 11; the presumed resemblance to Qomul therewith disappears.

There is little doubt that, as in the case of Κc'n, this place, too, should be looked for on the caravan route linking Tun-huang with the interior of China. I put the problem to Professor Haloun, who was able to identify the name without a moment's hesitation : it represents & fc£ Chin-ch'eng, Middle Chin, kidm-zlay, Chin-ch'eng was the great fortress town that covered the crossing of the Huang-ho. Situated on the right bank of the river (near mod. Lan-chou-fu), it was the first important town of China proper that the traveller from the West reached. A foreign merchant could scarcely find a place better suited to his requirements. A re-examination of the photograph shows that the last, somewhat shadowy letter is in fact -n. Hence, we read Kmzy(n), which a Sogdian may have pronounced Kemzin (note that here again the final Chinese y [" ng "] appears as -n).

The place mentioned in line 8, Kwr'ynk, was probably another town on the road to China, presumably to the east of Kemzin. Nanai-vandak through¬out complains of being cut off from " inside " ; he states that he cannot get into touch with his agent in Kwr'ynk; hence, one might argue, Kwr'ynk lay " inside " or even beyond. Eeichelt, on the contrary, identified the name with that of Krorayina/Lou-lan.
[324]
611

Yet it is unlikely that traffic to Lou-Ian was // subjected to more than ordinary difficulties at that time; indeed, from iii, 7, one learns that during the three years preceding the dispatch of the Letters the road to the West (on which Lou-Ian was the first considerable stage) was open not less than five times. Moreover, Lou-Ian is spelt kr'wr'n — Erordn in vi, 5. Surely the Sogdians must have adopted a standard spelling for the names of the more important places ? Had they allowed themselves such vagaries as writing kw'ynk and JtrWn indiscriminately, the unfortunate addressees of these Letters—to say nothing of us—would have had no chance of understanding what their correspondents were talking about. As a vehicle of communicating foreign, above all Chinese, names the Sogdian script may well claim to be the worst imaginable, even without the complication of vacillating orthography.

Another place-name, probably, is cyrδsw"n in the date. MR'Y, the ideogram for "master, overlord, prince" (presumably representing xwt'w1), is elsewhere preceded by a place-name (or an adjective derived from a place-name). So in the Mugh letters the official style of King Dewastic is sywSyk MLK' sm'rknBc MR'Y Syw'γyc " Sogdian King, Prince cf Samarkand, D." If we may rely on analogy and so regard cyr§sw"n as the name of a town, one lying in the region from which Nanai-vandak sent his letter (between Tun-nuang and the Huang-ho), can we help thinking of the old name of Su-chou, t@ ^ Chiu-ch'iian, Middle Chin. Tsisu-dz'i^dn 1 This over-vowelled monster may have been heard as *Ci-Bswan by the Sogdians who then may have yielded to temptation and inserted an unetymological -r- (as they did frequently, see BSOS., IX, 570); the resulting *cir-$swan seems to be adequately expressed by the Sogdian letters cyr&sw"n (the letter -c- = the sound -c- [" ts "] as in Kc'n).

Who then was this " Prince " or " Overlord " of Su-chou by whose years Nanai-vandak dated Ms letter ? Only one answer seems possible : the above-mentioned governor-general of the province of Liang, Chang Kuei. He had been appointed in 301 and died in 314 (on the 8th July); his 13th year thus coincides with 313. While this date appears to agree well with the other data derivable from our. Letter, it cannot be denied that the view presented here is scarcely more than a pis oiler. No objection can be raised to the assumption that foreigners Jiving in Su-chou or Tun-huang dated by the years of Chang Kuei; on 6th June, 313, even a Chinese may have been excused for not knowing who the legal Emperor- was ; in the eyes of foreigners coming from the West Chang Kuei, though a loyal servant of the Chin, must have seemed a ruler of great authority—he certainly was the leading representative of Chinese might to the west of the Huang-ho. What, however, is strange is that he should be called " the Overlord of Su-chou "—why Su-chou a and not, e.g., Ku-tsang ?

1 Cf. Schffider, Ung. Jahrb., xv, 575, n. 5. Schieder thinks cyr$su>"n is a personal name, but gives no reason for thia view.
2 Professor Haloun (who disapproves of my explanation) points out that & prefect of Chiu-ch'iian, iJH ^ Chang Chen by name, IB mentioned as having been involved in intrigues against Chang Kuei in 308. He was probably removed from his office at that time, and so whoever was prefect of Su-chou in 311 or a year or so later cannot have been in his 13th year of office.
[325]
612

Presumably because Nanai-vandak resided in Su-chou; he lived certainly to the west of the Huang-ho, but neither in Tun-huang nor in Ku-tsang. The Hugh letters show that the Sogdians were in the habit of calling their ruler the MR' Y of a town (in addition, it is true, to giving them a wider title); may not the Sogdians of Su-chou have done the same ? ¹

Historical background

Having dealt with the place-names we can now briefly review the relevant events of Chinese history to which the Sogdian Letter appears to refer. At the beginning of the fourth century we find the southern branch of the Hsiung-nu (Huns), admitted into Chinese territory conmie federes, comme troupes au service de Vempereur ², in occupation chiefly of mod. Shan-si, with their centre first in T'ai-yüan, then in P'ing-yang. Taking advantage of the conditions of chronic famine and revolt that characterized the rule of the Chin (Jin) emperor Huai-ti {beg. 307), the Hsiung-nu (Huns) shang-yu Liu Yuan proclaimed himself Chinese Emperor in 308. His son 劉聰 Liu Ts'ung (Liu Cong), who succeeded him in 310, fut VAttila de la Chine ³.

A minor effect of the troubled conditions, one that was not primarily connected with the activities of the Hsiung-nu (Huns), was the destruction of Yeh in 307. In the 5th month of that year (= 17th June-16th July) insurgents led by one 汲桑 Chi Sang (Ji Sang) captured the town, burnt its famous palaces—they burned for over a week—killed more than ten thousand people, and, after plundering the city, withdrew to the south-east. Even at that moment it could well be said-that "Yeh is no more".

The famine was, even for China, abnormally severe. ⁴ La saison ayant ete tres mauvaise dans le territoire de Loh-ydng, les denrees devinrent si rares que le peuple, pendant cette affreuse disette., en mnt jusqu'a manger de la chair humaine, et que les mandarins, ne pouvant plus y subsister, en sortirent presque tous pour aller ailleurs. L'Empereurfut lui-meme sur le point d'abandonner la mile.⁵

The final catastrophe took place in 311. The capital (Lo-yang), which had been deserted by the army, was invested by the Hsiung-nu (Huns). The Emperor had wanted to leave it, but had missed the opportunity. On the 13th July Lo-yang was stormed; the Emperor, Huai-ti, hoping to establish himself in Ch'ang-an, tries to flee, but is immediately captured and carried off to P'ing-yang.

1 No good alternative explanation is at hand. Mr'y is conceivably an Iranian word for " number " (it does not occur), and cyrSs- could be the (unatteated) Sogdian word for " 14 " (this would leave -w"n unaccounted for). However, to operate with " the 13th year of the 14th number = period (?) " ia useless. The nearest 13th year of the sexagesimal cycle ia 316, but the number of the cycle in the moat widely used system of chronology is fifty (and in none fourteen). Cyr$sw'>n tits no known name of a ruler or nien-ltao.
2 R. Grousset, L'Empire ties Steppes, 90. Cf. Sogdian Letter, Hue 16.
3 Grousset, loc. cit.
4 Cf. Sogdian Letter, lines 37-$.
5 A. des Michels, Hist. Geogr. des Seize Royaumes, notes du comm., p. iv. Cf. Sogd. Lett., line 11. The verb, 'pr'St, means " fled ", not " weggegaugen ", cf. iii, 33, " Farnxund abaconded : the Chinese sought him (xvryzy'id), but did not find Mm (fiyr'yitt)." Chr. Sogd. -pry&y [read prily ?], S.T., ii; pryz, George, 284 ; Yaghnobi apirez, apiraiz, GrlrPMl, i, 2, 335, all = " flee ". From apa-raz- (Skt. rah-, OPers. Tad-, MPers. rattan, NPers. rastav).
[326]
613

The palaces are burnt, the town is sacked and completely destroyed. It remained deserted until the Yuan Wei (T'o-pa) decided to make it their capital again (city walls built in 493, rebuilding of the palaces completed in 502}.

Yeh, or what remained of that town (which was an important fortress), was at that time commanded by Liu Yin, a Chin general. The Hsiung-nu (Huns), under Shih Leh (Shi Le), attacked him in the 6th month (beg. 20th July) of 312, but failed to take the town. It was captured and sacked only in the 4th month (12th May-9th June) of the next year, by Shih Hu (Shi Hu), who by his cruelty and blood-thirstiness evoked the disgust of even Hsiung-nu (Hun) generals.

Immediately after the destruction of Lo-yang the Hsiung-nu (Huns) initiated a great attack on Ch'ang-an (311, 8th month = 31st August-28th September), which was in the hands of Prince Mu, the Governor and C.-in-C. of all N.W. provinces. Having taken T'ung-kuan, the main fortress guarding the entry into the Wei valley, and defeated Mu's troops there, they advance to Hsia-kua. The general commanding the garrison troops of Ch'ang-an, Pei-kung Ch'un, who leads the auxiliaries sent by Chang Kuei from Liang-chou, leaves the town and surrenders. Prince Mu, thus abandoned by his soldiers, follows his example, and Ch'ang-an falls into Hsiung-nu (Huns) hands.

In the 10th month (29th October-26th November) the prefects of three adjoining prefectures collect an army and defeat the Hsiung-nu (Huns) before the gates of Ch'ang-an, but they succeed in holding the city. In 312, 3rd month (23rd April-21st May), Chang Kuei rushes troops to the East for its recapture. In the 4th mouth (22nd May-19th June) the allied Chinese relief troops at last force the Hsiung-nu (Huns), under Liu Yau, to abandon Ch'ang-an. They retire to their dominions to the east of the Huang-ho, not without removing 80,000 people from the city.

Prince Yeh (a nephew of Huai-ti), who had been proclaimed Regent after the fall of Lo-yang, now enters Ch'ang-an. He assumes the title of "Heir Presumptive" in the 9th month (17th October-15th November), 312. After the death of Huai-ti, which occurred in the 2nd month (14th March-llth April) of 313, Prince Yeh, the future Min-ti, ascends the throne in Ch'ang-an (4th month = 12th May-9th June). That month is marked by a minor attack on the city by Liu Yau, who again appears before it in the 10th month (5th November-3rd December), enters and burns the suburbs, but is defeated in the following month and withdraws. There is no point in pursuing this sad history any further here.

Nanai-vandak's report agrees with these events in all important details: the enemies are the Hun; the war is concerned chiefly with the three cities of Saraγ, aKhumdan, and 'Ngapa; Saraγ is completely ruined; the terrible famine ; the Emperor's flight, and possibly his capture (line 14 1). Taken one by one, these details are recurring features of Chinese history; but in their totality the points of agreement are too numerous to allow of any doubt: the report refers to 311. We may, however, go a little further and try to determine the precise date.
[327]
614

The Sogdian month Tγm'yc (= Persian Dai)¹ corresponds to 7th June-6th July in 311, to 6th June-5th July in 312, 313, 314. The year 311 is too early; for Lo-yang was taken only on 13th July. The year 314 is too late; for at that time Min-ti had occupied the throne for over a year, so that Nanai-vandak could not well talk of Huai-ti as the " last Emperor " ; Min-ti was strongly supported by Chang Kuei (Zhang Gui) in whose territory Nanai-vandak lived. We are thus left with either 312 or 313.

Here the choice is far more difficult. It depends above all on the assessment of the sources on which Nanai-vandak relied and the speed with which news reached his ears. On the assumption that he lived in Su-chou, there were eight hundred miles of war-torn and completely disorganized territory between himself and Lo-yang. From his own agents he had heard nothing for three years. Even though Chang Kuei (Zhang Gui), the mainstay of the Chin, was doubtless well informed of the latest events, accurate knowledge was scarcely accessible to a foreigner in Su-chou who had been deprived of his normal news supply. Even taking into account the influx of numberless refugees, the time-lag between any event and the moment it became known to Nanai-vandak (who presumably endeavored to check rumors) was in all likelihood scarcely less than a full year.

His information corresponds fairly exactly to the true position existing in June, 312, when the fate of Ch'ang-an (Chan'an) was in the balance. Allowing for the time-lag, we obtain June, 313, as the date of the letter. In that month Nanai-vandak can hardly have known yet of Min-ti's accession; he should perhaps have heard that Ch'ang-an (Chan'an) had been recaptured: the fight for that city, however, went on (with interruptions} until it was again taken by the Hsiung-nu (Huns) in 316. It cannot be made out whether he had heard of the death of Huai-ti. His information on the fate of ^əNgap^a provides no conclusive date—there seems to be a reminiscence of the events of 307— but apparently he knew that the city was attacked by the Hsiung-nu; that attack took place after the end of the Tym'yc of 312. Of its final capture, which occurred in a month that overlapped—by four days—with the Tym'yc of 313, he can scarcely have been aware ; unless one were to assume that his sources of information were extraordinarily accurate and speedy, which assumption would not square with his hesitation in the case of Ch'ang-an.

On all counts June, 313, seems the best date. It agrees well also with the repeated statement that all traffic with the heart of China was interrupted "three years ago", hardly long before the capture of Lo-yang. Three and a half years ago the famine had reached its height in the capital (lines 35-6); the wealthy foreign merchants, who no doubt knew their way about the " black market ", are not likely to have been the first victims. Is it then an accident that the June of 313 falls within the thirteenth year of Chang Kuei, the ruler of all Chinese lands to the west of the Huang-ho where Nanai-vandak // lived, and that a thirteenth year is mentioned in the date affixed to his letter ?

1 The first day of Tym'yc is the 271st day of the Sogdian year; the first day of the Sogdian year fell on 10th September in 310 and 311, on 9th September in 312-15.
[329]
615

As all the Letters no doubt belong to the same year, or to two successive years—the one dated in the 10th month probably belongs to the Chinese year preceding 313—we can tentatively assign the following dates to them: ¹

Hsiung-nu, Xiongnu, Xwn, and Huns

It remains to say a few words on the names used in the Sogdian Letter for the Chinese and Hsiung-nu. It is obvious that as cynsin = Chinastan is “the land of 秦 Ch'in”, so cyn = Ch'in is “the people of Ch'in” = “the Chinese”; it occurs five times in Letter No. iii ², but was not recognized by Reichelt.

Of far greater interest is xwn = Hsiung-nu; the first letter, by origin the Aramaic heth, serves in Sogdian, a language devoid of the sound h, to represent not merely the indigenous x (χ) - voiceless guttural continuant - but also any kind of foreign h-sound (occasionally also thee Chinese k'). Thus xwn can be read as Hun or Hūn or Xun or Xūn. In recent years there has been some considerable reaction, led by O. Maenchen-Helfin ³, against the firmly established but possibly naive belief in the identity -  in whatever terms conceived - of the Hsiung-nu of the Far east with the Hunni of Europe (with the Indian Hūna coming in as a weak third); much doubt has been thrown on the identity of even the names. Yet here we find the name that is indistinguishable from that of the Hūna, Ουννοι, Hunni, Arm. Hon-k', Saka Huna, Khwarezmian هون  Hūn ⁴, employed not of nomads of vague definition, but actually of the genuine Far-Eastern Hsiung-nu. And, what is more remarkable still, this name, unlike that found in the Saka Lehrgedicht ⁵, was in use well before the time when either the European Huns or the tribes that became known as Hûna to the Indians made their first appearance in history (i.e. AD 312-313).

2. Some of the passages have been translated in this article. ...
3. “Huns and Hsiung-nu” and “The Legend of the Origin of the Huns” (with full references), Byzantion, xvii, 1944-5, 222-251.
4. Found only in [ ]هون نجل hūn-zādek “slave” (lit.”a Hun's son” or “a Hun-boy”) and [ ]هولن “slave-girl” (lit. “a she-Hun”; formed with Khwar. femin. suffix -ān), the equivalents of Arabic 'abd and 'jariyah, Persian bande and kanizak respectively (fonts and orthography may be somewhat off).
 5. xvi, 9, Huna Cimgga may mean “Hsiung-nu and Chinese” (or Chinese' Huns, vs. free Huns), but this cannot be proved (cf. St. Konow, NTS., xi, 1938, 35).[329]