Printer Friendly

Sogdian documents from Khotan, II: letters and miscellaneous fragments.


In the first part of this study (Bi and Sims-Williams 2010) we published four Sogdian economic documents from the Khotan area that were acquired, together with manuscripts in Chinese, Khotanese, and Tibetan, by the Museum of Renmin University of China, Beijing, in the spring of 2010. The documents edited in the present article, which completes the publication of this collection of Sogdian manuscripts, comprise one nearly complete letter (no. 5), five fragments of letters (nos. 6-10), and some tiny scraps of which too little is preserved for the contents to be clearly identified (nos. 11-13). (1) Unfortunately the findspots of the documents in this collection are not recorded, but, as argued in our previous article, it seems likely that they derive from sites in the Khotan region, such as Mazar Tagh, Dandan Uiliq, and Old Domoko, where the presence of Sogdians is already attested in around the eighth to ninth centuries. (2) As for their date, the only specific indication we have is a fragmentary Chinese date formula on the reverse of one of the documents (no. 12), which seems to indicate a date towards the end of the eighth century (see discussion below). In this connection it is worth noting that most dated Chinese documents in the Renmin University collection are dated in the eras of Dali, Jianzhong, and Zhenyuan, i.e., the second half of the eighth century, the last stage of the Tang empire's control of Khotan.

The letters in the Renmin University collection are in general rather poorly preserved and written in a type of cursive script that is often highly ambiguous. Luckily, we know quite a large number of Sogdian letters from other collections, some of them well preserved and clearly written, and these provide comparative material that makes it possible to interpret at least the more formulaic parts of the new letters. When it comes to the substance of the letters, however, this material gives us little help, with the result that the readings and interpretations offered below are in many cases hypothetical.

Following the edition of documents nos. 5-13, the present article contains an inventory giving the catalogue numbers and dimensions of all thirteen documents.


No. 5 (Fig. 1-3)

Catalogue No.: GXW 0114. An almost complete sheet of paper bearing
thirty-two lines of writing on one side only, with a large hole in
the center.


2 'nwty msy'tr pryw 'xsnkw p'scyk xyp[??]'wnt (a)





7 ZY MN m'xw py[??]'r ''ycw 'ntwxch ZYL' prm'y xwrt


9 'ysk (f) wytk kwn'y skwn w'nkw kt txsyc[??]ntk mn'

10 prm'rz "st rty 'z-w MN 'sm'xw L' m'ck' xwz-'m L'

11 ksymn (g) p'rZY xwty pr mn' (h) ps'ycyk p's'y 'k[??]ry 'z-w


13 L' kw twrkst' s'r rty L' twpytstn s'r rty tw'




17 "ycw 'sty rty prm'rz 'yt'k 'krty 'zw kw prw'n s'r xrtym








25 'sm'xw L' w[??]'y '('...r) kys' ... 'w L' xcy .... tym 'z-w


27 kwn'wtys w'nkw kt txsyc[??]ntk '(nsps)t (s) ZY 'zw 'z-c't



30 'nxwy tys prwy[??]u py[??]'r kt mn' w'xr ZY <m(n') (d) prn
     s(yk)r> (v) 'ny'n syxw'yw



(a) The end of the line is left blank. (b) Or just pr'[??]t? (c)
Sic. (d) Uncertain. (e) The first letter looks like w, but the
reading is certain from the context. (f) Or '[??]sk, less likely
'ysk, '[??]sn, etc. Hardly 'rsk. (g) ksym '? ksym'7 (h) Or m'n. (i)
Sic, mistake for -stn. (j) Or ny'tyl (k) Fairly clear, apart from
s. (l) The individual letters are uncertain, but the reading of the
word as a whole is very likely. (m) Only slight traces can be seen
of the first three letters, but the k is almost certain. (n) Or
rymx'? RBmx'? (o) The k is quite uncertain here. (p) The traces of
the end of this word are quite uncertain. (q) Or mn "Not tykws.
(s)-ns- quite uncertain. The following letter is either p or w. (t)
Or 'krt'ym? In either case, the penultimate letter is badly

(u) Sic, mistake for prywy[??]. (v) These words are squeezed in between lines 30 and 31; it is not clear to which line they belong. (w) Uncertain; only the initial t(?) and the final s are really visible.

A letter and respectful message to the noble lord, (my) great hope, (2) greater than every hope, the dear, excellent, revered master (3-4) Anyan. From his well-wisher (and) friend, who is without service (and) who has not arrived at service, (his) insignificant servant Takhsichvande, (5) a respectful message and much homage (to your) honor. If you are well, (6) we rejoice. We too, by God's help, are well up to the present day. (7) Please don't worry on our account.

(8-9) But you were ... unjust insofar as you were accusing(?) me, (your) insignificant servant, (saying): "Takhsich-vande (10) has taken my profits(?)." I do not ask for capital from you, nor "merchandise]?), but you yourself have cast upon me (what there is) to be cast. Now I (12) have been broken and have suffered harm. But I did not go to Sogd, (13) nor to Turkestan, nor to Tibet. (As for) your 14goods, wherever I brought (your) profits(?), the major (part of) your goods (15) has gone(?) out of (my?) hands, both (to) Khumdan and to (the) Uygur(s), also (to) Turkestan, (16-17) also to Sogd, and a profit(?) was not made (for you)(?). Whatever goods there are with me, a profit(?) was taken (for you)(?). I went to Parwan (18) and I sent a letter to you. I sent the letter from Parwan. 19(I) here, a broken [man], have sent le[tter after lett]er(?), and your honor (20) [has] not(?) [done] likewise]?)! With Wisak-fam and Remghan (21) ... evil(?) ... you were ... so that (22) Wisak-fam ... me ... [Wisak]-fam(?) was giving ..., and I (23) refused to take (it), and [he was] irritated and angry ... (24) he noticed your intention that you should do your own work. (25) (In that case) you would have nothing! ... decreased]?) ... is not ... I am still (26-27) alive, [and] it would have been proper that you should treat me(?) well, (saying): "Takhsich-vande has exerted himself]?) so that I should be well, 28and however much capital [he needs]]?) I will give (it to him)." But Wisak-fam (29) came here and said to me: "I was glad that (30) it entered (my?) mind, because he abandoned me and ... my(?) honor; I cut out]?) Anyan (31) because he did not notice]?) my many years love (for him), and now (32) he has not gone to you at all." [May you be] informed of all my news ...


Line 1. Another letter beginning with the phrase [??]ykh ZY ptskw 'nh "a letter and respectful message" is no. 7 below. The phrase also occurs in the first lines (though not as the first words) of L44 (3) and probably of no. 8 below.

Lines 1-2. ZY [TEXT NOT REPRODUCIBLE IN ASCII] RBch 'nwth MN wyspn'c 'nwty msy'tr pryw 'xsnkw p'scyk xyp[??]'wnt "to the noble lord, (my) great hope, greater than every hope, the dear, excellent, revered master." Most of these honorific phrases are well attested in the address formulae of letters such as those from Mt. Mug, (4) though p'scyk 'revered, honorable' seems to be a new variant in place of the synonymous [TEXT NOT REPRODUCIBLE IN ASCII], which occur frequently in Manichaean letters. (5) It is interesting to note that the ideogram ZY = 't 'and' is here misused for the preposition't 'to' (represented in the oldest Sogdian letters by the ideogram 'D).

Line 3-4. The phrase naming the sender is inset as a sign of humility, a feature that is attested in many Sogdian letters and has been attributed to Chinese influence. (6) Conversely, as a sign of respect, the name of the addressee is placed at the very beginning of a new line, in the otherwise blank space to the right of the phrase naming the sender, the end of the preceding line being left blank to make this possible. This feature seems also to be attested in nos. 6 and 7 below; variants of the same layout can be seen in no. 8 and in L44, (7) in both of which the end of line 4 is left blank, the name of the addressee (with the postposition s r 'to') stands alone on line 5, and the phrase naming the sender begins on line 6 (inset in the case of L44; this cannot be verified in the case of no. 8 below, which has a lacuna at the critical point).

Assuming that it is to be read 'ny'n rather than 'zy'n, the name of the addressee may consist of the Chinese family name An given to families which had originally come from Bukhara and attested elsewhere in Sogdian in the spelling "n, (8) and the Sogdian personal name y'n. (9) However, a purely Chinese or purely Sogdian name cannot easily be excluded. The sender, Takhsich-vande, bears a well-known Sogdian name, attested in the "Ancient Letters" and the Upper Indus inscriptions in the spelling txs'yc[??]ntk. (10) Both names occur again further on in this letter, 'ny'n in line 30, txsyc[??]ntk in lines 9 and 27.

MN [TEXT NOT REPRODUCIBLE IN ASCII] "from his well-wisher (and) friend, who is without service (and) who has not arrived at service, (his) insignificant servant." These self-deprecatory phrases referring to the sender have close parallels in Bezeklik Letters A (lines 26-29) and B (lines 18-23). (11) Similar wording can be restored in Otani 2314 and L118. (12) The form [TEXT NOT REPRODUCIBLE IN ASCII] is presumably abl. sg. of [TEXT NOT REPRODUCIBLE IN ASCII], as the word for 'well-wisher, friend' is spelled in Bezeklik Letter B, line 20. The f. form [TEXT NOT REPRODUCIBLE IN ASCII] is attested in a similar context in L44, line 6. (13)

Line 5. RBprn or RBfrn "(your) honor," lit. "great glory," is commonly used in the Mug and Bezeklik letters as a respectful way of referring to the addressee.

Lines 5-6. [TEXT NOT REPRODUCIBLE IN ASCII] "if you are well, we rejoice." This is a version of a standard formula attested for instance in the Mug letter Nov. 2, lines 5-6: [TEXT NOT REPRODUCIBLE IN ASCII] "If you, the lord, are in (high) honor (and) well, then, sir, I rejoice." (14) In the present variant of this formula, 2nd pi. forms are used to refer to a single addressee, a usage that is also attested below in no. 6, lines 3-4, and no. 8, line 7, and elsewhere. In the rest of this letter the 2nd pl. pronoun 'sm'xw is used several times to refer to the addressee, but the relevant verbs are all (with the exception of wny[??]'y, line 24) in the 2nd singular, e.g., prm'y 'please' (line 7), m't'ys 'you were' (line 8). Similarly in the case of the 1st person, the writer seems to refer to himself either by sg. forms such as mn' 'me' or plurals such as m'x 'us'. In some cases the latter could be understood as referring to the writer together with his family or associates, but at least in line 8 the context seems to imply that it refers to Takhsich-vande in particular.

Lines 6-7. In Sogdian letters the inquiry or expression of good wishes regarding the health of the addressee is sometimes followed by a phrase informing him of the good health (or alternatively, the miserable circumstances) of the writer and his associates, but there seems to be no close parallel for the present wording. For the phrase in line 7 one may compare Letter F, lines 17-18: (15) [TEXT NOT REPRODUCIBLE IN ASCII] "your wife (and) children are all well. Don't worry!"

Line 8. For the expression m'xw wysp'n ([??]nt)k "us (= me), (your) little (= insignificant) slave" one may compare m'xw wsp'n [??]ntyt[w] in Bezeklik Letter A, line 51. (16)

Line 9. 'ysk(?) wytk kwn'y skwn "you were (or are) making ..." The meaning of 'ysk(?) is unclear. It is tempting to read 'rsk 'envy, jealousy', but the second letter does not have the usual shape of an r. The following word is also problematic. Since the verb kwn- 'to make' or 'to put' requires an object, it does not seem likely that wytk is the pp. of wyn 'to see'. The noun wytq Tine, string, marker', which is known from a Christian Sogdian text, (17) seems more promising, although one would have to assume an idiomatic usage of the phrase 'ysk(?) wytk kwn- "to make/put a ... line," which seems to introduce a quotation and for which the context seems to imply a meaning such as 'to accuse' or 'to complain'.

Lines 10-11 seem to contain several financial terms: m'ck', prm'rz, and ksymn(?). The only one that is known from other Sogdian texts is m'ck' 'capital' (also in fine 28), which must represent [mack], a form intermediate between m'tskh [matsk] 'id.' and Manichaean Sogdian m'sk [mask] 'substance, matter, basis'. (18) prm'rz (also attested in lines 14, 16, and 17) looks like a verbal noun belonging to Manichaean Sogdian frmrz 'to remove'. (19) Here it is object of "s 'to take', while in line 17 it is the subject of 'yt'k 'krty 'was taken'. On the basis of the etymology suggested, one may guess that the basic meaning of prm'rz is 'deduction', which in the context might be understood as a technical term for a percentage deducted from the trading profits and handed over to the owner of the capital or stock in trade, in this case Anyan. As for ksymn or ksym', the pairing with m'ck' 'capital' suggests that it may refer to the merchandise supplied by Anyan.

Line 11. Several words in this line are ambiguous: mn' 'me/my' or m'n 'mind'? p's'y 'he/it/ you cast' or obl. of p's 'respect'? ps'ycyk fut. participle '(that) which is to be cast' or Tatter, later'? (20)

Lines 12-13. The names of the countries [TEXT NOT REPRODUCIBLE IN ASCII] 'Sogd', twrkst 'Turkestan' (for twrkstn as in line 15), and twpytstn 'Tibet' are all quite clear, though one may wonder what exactly 'Turkestan' (contrasted with y'yywr 'the Uygurs' in line 15) implies at this period. Cf. Yoshida and Moriyasu 1989: 14-15, who discuss the term 'Turkestan' in a Sogdian contract written in Qocho in 639 c.E., suggesting that it refers to the whole of the northern steppe region inhabited by Turkic peoples, from the Syr Darya to Mongolia.

Line 14. Here, and many times in the following lines, "ycw 'thing, matter, business' may be a rather vague term ('stuff') for the merchandise provided by Anyan as stock in trade for Takhsich-vande and his other agents.

Lines 14-15. MN [??]st' xrty 'sty "has gone out of (my?) hands" or MN [??]st' ny'ty 'sty "has been taken out of (my?) hands"?

Line 15. x(w)mt'n 'Khumdan', the Sogdian name for the Chinese capital city of Chang'an, was identified by Gustav Haloun (21) as a transcription of Chinese Xianyang, the name of the former capital, which lay a few kilometres northwest of Chang'an.--In this context, [TEXT NOT REPRODUCIBLE IN ASCII] can hardly be anything other than the name of the Uygurs, though it differs from the forms found elsewhere in Sogdian, 'wy[??]wr and yw[??]wr. (22) The obvious source is a Chinese form with initial velar articulation, e.g., Huihu [TEXT NOT REPRODUCIBLE IN ASCII], Late Middle Chinese * x[??]uaj-xhut (23) perhaps via a Khotanese form such as hve:hvu:ra. (24)

Lines 15-16. For the use of 'nyw 'other' as an adverb 'again, also' see Sims- Williams and Hamilton 1990: 26.

Lines 16-17. Here it becomes clear why the writer has mentioned many places to which he has not gone: since goods have been sent to these distant places in the hands of other agents, Takhsich-vande cannot be held responsible for the fact that Anyan has not received his share of the profits.

Lines 17-18. prw'n no doubt represents the older name of Aqsu, the various forms of which were studied long ago by Henning. (25) According to Henning, the form p'rm'n = barman is attested in unpublished documents (presumably in Sogdian script, as implied by his transcription with a). The present form, however, agrees with the Manichaean Sogdian adjective prw'nc in the Mahrnamag, line 77 (Muller 1913: 11), whose final -c is no doubt the same suffix as that of sm'rkn[??]c 'of Samarkand', etc. (Y.Y.), rather than a feminine ending as assumed by Henning.

Lines 19-20. These lines, with their complaint about the addressee's failure to keep in touch, seem to contain a version of a well-known cliche. (26) If '(nx)w(s)ty 'broken' is correctly read, it may be taken to refer to the writer (cf. mSy 'here') and be understood in the light of the phrase in lines 11-12: 'k[??]ry 'z-w mnxwyw "now I have been broken."

Line 20. At the end of the line we seem to have two personal names. The first occurs again in lines 22 (perhaps twice) and 28 and is probably to be read ws'kprn, although no individual attestation is entirely clear. Such a name is not attested elsewhere, but the personal name ws'k, lit. 'Peace', is found in the Mahrnamag, line 116 (Muller 1913: 14) and -prn 'glory, fortune' is very common as a second component of Sogdian names. The second name, which occurs nowhere else, could be read in several ways. As a mere possibility, we suggest rym[??]n, lit. 'Destroying impurity'. Wisak-farn and Remghan may be Anyan's agents, perhaps (in view of the mention of '[??]z-yk 'evil'(?) in line 21 and Wisak-farn's confession cited in lines 29-31) those blamed by Takhsich-vande for withholding Anyan's due share of their profits.

Line 24. tykw(y)s is a highly unusual, perhaps unique, spelling for tykws, the 3 sg. impf. of the verb tkws 'to look, observe'.--For the 2 pl. opt. form wny[??]'y 'you should do' cf. Gershevitch 1954: 115 ([section] 754) and Yoshida 2009a: 286.

Line 25. kys could in theory be a noun meaning 'teaching, (false) teaching, heresy'; cf. the derivative kysyk '(false) teacher, heretic', but in this context a form (3 sg. impf. or pres, inf.) of the verb kys 'to decrease' seems more likely. (27)

Lines 26-27. p[??]kh m't kt(t'm') syr'kwf ([??]r)ty kwn'wtys, lit. "it was the rule that you should have made me well-treated," i.e., "it was (= would have been) proper that you should treat me well." For the expression p[??]kh m 't cf. the Sogdian version of the title of Psalm 29, where p[??]kh [m't] seems to translate Syriac p" hw' "it was proper" (Sims-Williams 2014: 46). The use of the simple preterite m 't rather than a modal form ("it would have been ...") is perhaps surprising here, but the following irrealis kwn'wtys "you should have made" may have been felt adequate to indicate the hypothetical sense of the whole sentence.--kwn 'wtys is a unique 2 sg. irrealis in which the irrealis suffix - 'wt- is followed by a 2 sg. ending which is probably that of the opt. middle (-ys < *-aisa; see Yoshida 2009a: 282) rather than that of the intransitive pret. (-ys < 'ys 'you are', 2 sg. pres, indie, of the verb 'to be').

Line 27. '(nsps)t, 3 sg. intr. pret. of 'nspt 'to arouse oneself, exert oneself, act zealously', the past stem of which is indirectly attested by Manichaean Sogdian 'nspstky' 'zeal', an abstract noun derived from the past participle. (28)

Line 28. It is difficult to find an alternative reading to the adverb (w)y(t)w(r) 'so long, so far', but its function here is not clear.

Line 30. On 'nxw 'mind' see Sims-Williams and Durkin-Meisterernst 2012: 13-14.--The significance of the inserted phrase m(n') prn s(yk)r, apparently "he led (away?) my glory/ honor," is not clear.--syxw'yw, lit. 'I raised, took up, took away, removed'. Here it seems necessary to assume a special sense such as 'cut out, exclude (from the deal)'. Wisak- fam's statement quoted here seems to provide his excuse for witholding the share of profits due to Anyan.

Line 32. The words mn' w'xs s'tw (xw)ty [??]rt(p) [??] ('....y)s "[May you be] informed of all my news ..." seem to have partial parallels in the Bezeklik letters, e.g., Letter A, line 59: [TEXT NOT REPRODUCIBLE IN ASCII] [...] "may you be informed of all the news (and) [send] (your) command"; Letter B, lines 43-4: [TEXT NOT REPRODUCIBLE IN ASCII] "receive (our) trivial current news (so that) you yourself may be informed (and) [send] (your) command, lord." (29)

No. 6 (Fig. 4a)

Catalogue No.: GXW 0435. Two tattered pieces of paper which almost join, forming an irregular-shaped fragment without margins. One side bears Sogdian text written with a brush in a very casual style; the other is blank apart from a couple of unclear marks.

1 [          ]wcws tk(syna)[
2                (cn)n p[.....](sb )[           ]
3 [     ] x [ ] x c (rt)k[??] 'sm'x [trts'r pr prn zwk ZY]
4                (')yz-c'tt 'skw([??]')[ [TEXT NOT REPRODUCIBLE
                   IN ASCII] m'x kstrt']
6        ]........ [TEXT NOT REPRODUCIBLE IN ASCII] '(wn)'kw cyn"k [

(a) The s almost sure, as is the final n/'. (b) The last letter is
probably s or s. (c) One can see the bottom of k or p, then, after a
gap, the tail of a final n/'. (d) Reading suggested by Y.Y. (e) The
second letter is w or p. The first does not resemble any normal Sogdian
letter and is perhaps the result of a correction.

[To the noble lord ... (1) ...] ... the tejin (2-4) ..., from P[...]sh(?) [his servant]. If you [there] are [in high honor, healthy and] rested, [then, sir, we insignificant ones] (5) [re]joice. And, sir, we here [are all well up] to the [present day(?) (6)...] and, sir, that desire [to see you] ...


A fragment from near the beginning of a letter in a format similar to no. 5. The first word of line 3 should be the name of the addressee, whose titles and honorific epithets will have been contained in line 1 and the lost lines preceding it. The inset lines 2-4 will have contained the name(s) of the sender(s) and the beginning of the series of greeting formulae.

Line 1. tk(syn), cf. (tk)syn in no. 7, line 3, and t'ksy(n) in no. 11, line 1. These forms may represent the Chinese title tejin (Late Middle Chinese [TEXT NOT REPRODUCIBLE IN ASCII] tsin), lit. 'specially advanced', a supplementary (honorific) title under the Tang. It is worth noting that amongst the so far unpublished Chinese documents from Khotan in the Renmin University collection, three fragments involve a certain tejin who bears the titles General (jiangjun) and Commissioner (shi) or Commissioner-in-Chief (dashi); his surname is Mi, indicating that his family originally came from Maymurgh. It cannot be excluded that he is the same person as the tejin of these Sogdian documents, though this is no more than a plausible surmise. Pavel Lur'e has kindly drawn our attention to the occurrence of taksin in Islamic sources as the title of the ruler of a group of the Chigil, a Turkic people living to the south-west of the Issyk kul (see Minorsky 1970: 297-300).

Line 3. The traces of the name are compatible with a Sogdian name ending in - prn, but certainly do not demand such a reconstruction.

Lines 3-5. On this formula expressing good wishes for the health of the recipient, and in particular on the use of the 2nd pl. 'sm'x ... 'skw([??]') 'you are' to address a single addressee, see above on no. 5, lines 5-6.

Line 5. The proposed restoration is based on the equivalent passage in no. 5, line 6, which reads: [TEXT NOT REPRODUCIBLE IN ASCII] "we too, by God's help, are well up to the present day." Cf. also L44, line 11: m'x ms m[??]y nwr my[??] prm zwk ... "we too here [are] healthy up to the present day." (30) In the present text, the first letter of (n!w)[r] is admittedly far from being a clear n, but no other reading is any less problematic.

Line 6. cyn"k must be a variant of cyn'kh 'desire'; cf. no. 8, line 1. Other letters use an alternative formula containing the related adjective cyn'wt 'desirous', (31) the purpose of all such formulae being to express the writer's desire to see the addressee.

No. 7 (Fig. 4b)

Catalogue No.: GXW 0436. Fragment from the top right corner of a page with writing on one side only.

1 [??]ykh ZY ptskw['nh
2 s[ ]yta 'srwb    [
3 (tk)sync         [

(a) Or -[??]t. The missing second letter is a small one, e.g., 'or
n/z. (b) Or 'skw? Less likely -yw. (c) Almost certain.

A letter and humble address [to ...] ... the tejin(?) ...


Line 1. See above on no. 5, line 1.

Lines 2-3. If tksyn is a title as suggested above, the words in line 2 (which do not appear to be Sogdian) are perhaps his personal name. The phrase naming the writer(s) presumably began somewhere to the left of these lines, after the blank space.

No. 8 (Fig. 5)

Catalogue No.: GXW 0115. An irregular-shaped paper fragment from the top of a page with writing on one side only and traces of burning at some of the edges.

1 prw RBkw cyn'kh [                 [??]ykh Z]Y ptskw'nh
2 'kw [TEXT NOT REPRODUCIBLE IN ASCII] xwt'ynh [RBch 'nwth cnn
  wys](pn)'cyw 'nwty
3 [ms]y['t]r ZY cnn wyspn'cyw "[??]' 'yw (yw)[[??]w](yk)str [??]r'n
4 [ptst'](t) pt[??]yw-'ysc('nc)ha [??] ([??])'(mpnw)h        b
5             ]....[ ]syhc s'r                            b
6 [    ?ptskw'n]h cnn c'nks' syr ([??])w 'psy ZY wp'rs
7                ']s'r prw prn(h zwk) 'yc'tw (pr'[??])tyst'
8                '](M)a xy'std pr('y)w rty '(z)w s(y)r w([??]s'm)
9                ]s(yra )kw(n'a.............................)[ ](k)
10                                                           ] ZY
11                                                             ]....

(a) Uncertain. (b) Note the blank space at the end of lines 4 and
5. (c) Ends - snh, -syh etc. The preceding traces are quite
ambiguous. (d) Or [??]ynst, etc.; not 'ynst.

With great desire [to see her(?) I send(?) a letter an]d respectful message (2) to the noble lady, [the great hope, greater than eve]ry hope (3) and more necessary than everyone, the mighty 4[support], the honorable mistress (5) ...

(6) [... a respectful message] from the zhangshi, very many questions and inquiries. (7) [If] you have arrived at ... in (high) honor, healthy (and) rested, (8) together with ..., then I rejoice very much. (9)... do/did(?) well(?) ... contempt (10) ... and (11) ...


In its layout (and much of its phraseology) this document resembles L44. As pointed out above (comm, on lines 3-4, no. 5), the end of line 4 is left blank so that the name of the recipient can stand at the beginning of a new line as a sign of respect. The sender is named in line 6, but since the right edge of the page is missing at this point it is not clear whether this and the following lines are inset as in the case of L44.

Line 1. prw RBkw cyn 'kh "with great desire." Otani 2483 is a fragment of another letter that begins with the same words in the variant spelling pr RBkw cyn'. (32) For the partly restored phrase [[??]ykh Z]Y ptskw'nh "a letter and respectful message" see above on no. 5, line 1.

Lines 2-4. Similar sequences of laudatory phrases are attested in L27, L44, (33) and So 10920. (34) The phrase [TEXT NOT REPRODUCIBLE IN ASCII] xwt'ynh "noble lady," the exact fern, equivalent of [??] xwt'w "noble lord" in L27, line 2, is remarkable for the use of the previously unattested fem. form [??], lit. 'goddess'. Elsewhere [TEXT NOT REPRODUCIBLE IN ASCII]- is commonly inflected as a mase, stem even when used to refer to a female divinity or as a polite form of address to a woman. (35) It is not clear exactly how [??]([??])'(mpnw)h, lit. 'mistress of the house', is to be understood here. One meaning of this word is 'wife', and it is quite possible that the writer of the letter is in fact the husband of the addressee, but it seems equally possible that the word is used here as an honorific meaning 'mistress, lady'.

Line 6. c'nks' may represent the Chinese title zhangshi (Late Middle Chinese trian sr'), lit. 'senior scribe; Aide', which is attested as camssi in Khotanese. The final -' of the Sogdian form is perhaps an abl. ending (cf. "[??]' in line 3). During the Tang period the title is used for an official, usually of executive status but of varying rank, in the Prefectures and Area Commands. Judging from the Chinese and Khotanese documents from Khotan, the position of zhangshi of the local Area Command was sometimes held by a member of the royal family of the kingdom of Khotan. (36) In Turkish cangsi is well attested as a personal name or name-component. Several examples are cited in Clauson's dictionary, including one Cangsi who is referred to by Mahmud al-Kasghari as an "amir" of Khotan. (37)--syr [TEXT NOT REPRODUCIBLE IN ASCII] 'psy ZY wp 'rs "very many questions and inquiries (concerning your well-being)." No closely similar phrase is known from other Sogdian letters, but the general idea is clear from the context and partial parallels such as the following in Bezeklik Letter C, line 3: [TEXT NOT REPRODUCIBLE IN ASCII] m'n ps'ty "inquiring very much (about your state of) mind." (38)

Lines 7-8. These lines contain a variant of a formula we have already met in nos. 5 and 6. The choice of the verb (pr'[??])tyst' 'you have arrived (at)', as opposed to 'skw([??]') 'you are' in no. 5, lines 5-6, and no. 6, lines 3-5, may indicate that the addressee has just undertaken a journey.--xy'st? [??]ynst? This word seems to be unknown and may not be Sogdian. It is tempting to read '](M) 'ynst pr('y)w "together with the women," but this would amount to an emendation.

No. 9 (Fig. 6)

Catalogue No.: GXW 0117. Two pieces of paper that almost join, giving a fragment from the bottom of a page with writing on both sides. To judge from the width of the paper, (39) line 2 of the Recto may be complete.


1 [ ]......t([??]ya....) ([??]')w m('*)[b     ]s('r )L'C
3 [ ]......ksy s('r)ye m('x)f n[??]'nt kwnty-skwn
5    'njywi x[r']yn'[ya '](n)ywi prsyj ny'zk '[??]zyl

(a) Uncertain. (b) m'xl or mn'? (c) Or [??]n. d[??]' (or [??]n,
hardly [TEXT NOT REPRODUCIBLE IN ASCII]) seems to be a complete
word. (e) Less likely s'ny. (f) Less likely mn'. (g) An unclear
mark above the last two letters. (h) Sic, without final -'. (Above,
the word s'r shows through from the Verso.) (i) Or 'yw? 'Mistake
for prysyl? Or read p'sy? pnsy? (l) xzy'r? (m) Less likely 'yny.
There is a small blank space at the end of the line.

(1) ... there(?) ... the cow ... was not to ... (2) You say/said thus: "I [am] the owner of the cow" (3) ... now he is making ... with us(?) (4) ... message there ...: "Make inquiry about the cow" (5) ...


1 [??](ykh)a / ywmyt s'r (a)
2 [cnn?] / kmsrc (b)

(a) Reading suggested by Y.Y. (b) Or kmtkc etc.

(1) Letter(?) to Ghomet (2) [from?] Kamsarch.


Recto. The letter appears to involve a dispute about the ownership of a cow, but the details are quite unclear. Line 5 is incomprehensible: every word can be read or restored in more than one way, but no combination of readings seems to offer a connected sense. Since it is the last line on the page, with a small blank space at the end of the line, this line presumably concludes the letter. If prsy is a mistake for *prysy, the first four words could perhaps be translated "you should buy another (and) *send another," but this would leave the final words (ny'z 'need' or zy'r 'poison' + '[??]zy 'evil' or 'yny 'this') unexplained.

Verso. At one end of the sheet stands the solitary word [??]ykh(J) 'letter', written at 90[degrees] to the text on the Recto; below this another line may be lost. The two lines at the other end of the sheet are written at 180[degrees] to the text on the Recto and thus at 90[degrees] to the word [??]ykh. Yutaka Yoshida kindly draws our attention to the fragmentary letter Otani 6341, (40) the Verso of which bears an address formula laid out in the same remarkable fashion. (41)--Of the two personal names, [??]wmyt is well known from the Upper Indus inscriptions, (42) while kmsrc looks like an ethnic adj. in -c (from *Kamsar 'source of the river Kam'; cf. Minorsky 1970: 363 n. 1?).

No. 10 (Fig. 7a-b)

Catalogue No.: GXW 0433. A fragment of paper from the top right corner of a page with writing on both sides, formerly folded many times into a narrow strip. The address(?) on the Verso is written at 90[degrees] to the text on the Recto.


1 't [TEXT NOT REPRODUCIBLE IN ASCII] xw([??])[w (a)  ]....[
3 nm('nw) (c) rty x'nkh (d) p[??]kh [
4 zn' (e) [??]rnt (f) z-'**[(g) ] [??][
6 k**[

(a) Less likely xw(t)['w. (b) Or [??]wx-. (c) Doubtful, but hardly
nm'cw. (d) Mistake for x'n'kh. (e) Or z'n; less likely ZNH or ZYn.
(f) Less likely [??]r't. (g) z-'wr[? Less likely z-'tk[.

(1) To the noble lord(s) [...] (2) Zar-farn and Dhugh-pash, [from Rust ... your servant, who is] (3) looking forward [to seeing you]. And [I offer homage to you] in that manner [in which] (4) (people) offer (it) [to the gods(?) ...] (5) Dhugh-pash in ...


1 [..........z'rpr](n)         (cnn) [        ]
2 [ZY [TEXT NOT REPRODUCIBLE IN ASCII] ....]       rw(st)[        ]
  (blank space ?)
3             ](...s)k(...n)...[

(1-2) [To ... Zar-far]n(?) [and Dhugh-pash(?).--From Rust[...].--(3) ...


R2. (z')rprn and [TEXT NOT REPRODUCIBLE IN ASCII] are presumably the names of the addressees (or perhaps the senders, in which case the text must be differently reconstructed). Neither name is attested elsewhere, but at least z'rprn 'Thousand-glories' is comprehensible as a combination of two attested name-components, z'r- as in z'ryw[??] 'Thousand-fighter' (43) and the extremely common -prn.

R3. nm'n (variant nm'nw), which apparently means 'desiring, looking forward to', forms part of the special vocabulary of Sogdian letters, where it occurs in the expressions wyn nm'n "looking forward to seeing (you)" and 'sp's nm'n "looking forward to serving (you)." (44)

R3-4. The words rty x'nkh p[??]kh ... [??]rnt suggest a version of the formula of L44, lines 8-9: [rt[??]n ](w)'n'kh p[??]kh w'[??]ry[??] nm'cyw [??]r'ym c'nw ZKn zwrny-zwrny-cy-kt pwt'ysty s'r nm'cyw [??]r'nt "we offer homage [to you] in the same manner as (people) offer homage to the Buddhas of the various periods." (45) An older variant of this formula refers to 'the gods' ([TEXT NOT REPRODUCIBLE IN ASCII]) rather than 'the Buddhas'; (46) cf. also pr [TEXT NOT REPRODUCIBLE IN ASCII] nm'cyw "in the manner of a god, much homage" in Otani 6341. (47) Within this hypothesis, however, it is difficult to find a plausible interpretation of the word preceding [??]rnt, which looks like zn' 'knowledge', or of the words following it, the traces of which suggest a form of z-'wr [TEXT NOT REPRODUCIBLE IN ASCII]- 'to help'.

V2. If rw(st)[...] is the name of the sender, it is tempting to interpret it as 'Rustam', though we know of no other Sogdian who was named after the famous hero.

No. 11 (Fig. 8a)

Catalogue No.: GXW 0437. A paper fragment from the edge of a document, with many small needle holes as if it was formerly sewn onto another document or package. Verso blank.

1 ]('n) t'ksy(n)

2 ]c'kw

... the tejin ... document)?) ...


Line 1. For t'ksy(n) see above on no. 6, line 1.

Line 2. If c'kw is a complete word, it may represent the widely diffused Chinese term cel chai (Early Middle Chinese [TEXT NOT REPRODUCIBLE IN ASCII]) 'document', which is attested as a loanword in Bactrian [TEXT NOT REPRODUCIBLE IN ASCII] 'receipt', Choresmian ck 'deed (of divorce)', Middle and New Persian cak 'document, contract', Arabic sakk. According to Livsic 2008: 310 and Lurje 2011: 158, c'kw 'document' is already attested in Sogdian in the Mug document A-13, line 5, but Grenet and de la Vaissiere (2002: 187 n. 33) interpret the word as the name of a bridge, which is less awkward syntactically.

No. 12 (Fig. 8b-c)

Catalogue No.: GXW 0284. A paper fragment from the top edge of a document. The other side contains a date in Chinese: ["[...]6th year, intercalated 2nd month, 10th(?) (48) [day ...]."

1 ]* m's'k (a) ksk f
  (blank space)
2 ].. (blank)
... the old man(?) Kasakk(?) ...

(a) Or mns'c?


A personal name ksk is unattested in Sogdian but could be understood as a hypocoristic in -akk from a compound containing Old Iranian *kasu- 'little'; cf. personal names such as Avestan Kasupitu-, Vedic Kasu-, etc. (49) Sogdian k's'kk, read by Lurje as Kasak (with a long vowel in the first syllable), is probably a different name. (50) The preceding word could be read as m's'k 'old man' or as mns'c '(he) arranged'.

More informative than the Sogdian text is the Chinese text on the other side. The presence of this date formula indicates that the fragment was written during the period of Chinese rule in Khotan, which lasted from about the mid-seventh century to the end of the eighth century. Unfortunately the characters preceding the partially preserved 'six' are missing, leaving open the possibility of restoring 'sixteen' and giving no indication of the Chinese era (nianhao) concerned. Since the calendar tables compiled by Chen Yuan and other scholars do not include any sixth or sixteenth year with intercalated second month during the relevant period, it seems likely that the fragment was written at a time of poor communication between the Western Regions and the Tang government, probably some time between the rebellion of An Lushan in 755 and the Tibetan occupation of Khotan at the end of the eighth century. As scholars have noted, local calendars were used in Khotan when the official calendars issued annually by the central government were not able to reach the Western Regions; this was also the case in Dunhuang after regular connections with the Tang court were severed in 786. Though the intercalations in the local calendars of Khotan and Dunhuang differ from the official ones, they take place in the same years as in the official calendar and in most cases the timing of the intercalation differs by no more than one or two months (cf. Kumamoto 1996: 39; Zhang and Rong 2008: 240-55). On this basis we have three possible years to consider. One of these, the sixteenth year of Dali (781), (51) can be excluded immediately, since a Chinese document from Dandan Uilik (D.V.6 = Or.8210/S.5864) shows that this year did not have an intercalated second month in Khotan. Thus the options are reduced to two: the sixth year of Dali (771) with intercalated third month or the sixth year of Zhenyuan (790) with intercalated fourth month. (52)

No. 13 (Fig. 7c)

Catalogue No.: GXW 0125. A fragment of paper from the right edge of a page, apparently containing a list, with one word in each line. Verso blank.

1 ([??]*)[
2 (znt*)  [
3 [??]'y (a)  [
4 **[ ]     [
5 (ZY)[    ] [
(a) Or RNY? BRY? [??]ry?

Fragment   Catalogue no.   Height x width (53)

No. 1      GXW0116         19.5 x 22.0 cm
No. 2      GXW0438         5.5 x 22.5 cm
No. 3      GXW0432         10.5 x 23.3 cm
No. 4      GXW0434         17.0 x 20.0 cm
No. 5      GXW0114         42.0 x 27.2 cm
No. 6      GXW0435         8.5 x 18.5 cm
No. 7      GXW0436         5.0 x 9.5 cm
No. 8      GXW0115         21.5 x 28.0 cm
No. 9      GXW0117         7.5 x 28.0 cm
No. 10     GXW0433         8.0 x 12.5 cm
No. 11     GXW0437         7.0 x 6.8 cm
No. 12     GXW0284         4.3 x 11.0 cm
No. 13     GXW0125         5.5 x 4.3 cm


Bailey, H. W. 1985. Indo-Scythian Studies, Being Khotanese Texts, vol. VII. Cambridge: Cambridge Univ. Press.

Bi Bo and N. Sims-Williams. 2010. Sogdian Documents from Khotan, I: Four Economic Documents. JAOS 130/4: 497-508.

Clauson, G. L. M. 1972. An Etymological Dictionary of Pre-thirteenth-century Turkish. Oxford: Oxford Univ. Press.

Dankoff, R., and J. Kelly. 1982-85. Mahmud al-Kasyari, Compendium of the Turkic Dialects, I-III. Cambridge, Mass.: Printed at Harvard University Office of the University Publisher.

Dickens, M., and N. Sims-Williams. 2012. Christian Calendrical Fragments from Turfan. Living the Lunar Calendar, ed. J. Ben-Dov, W. Horowitz, and J. M. Steele. Pp. 269-95. Oxford: Oxbow.

Gershevitch, I. 1954. A Grammar of Manichean Sogdian. Oxford: Basil Blackwell.

--. 1979. No Old Persian spadmaida. Festschrift for Oswald Szemerenyi on the Occasion of His 65th Birthday, ed. B. Brogyanyi. Pp. 291-95. Amsterdam: J. Benjamin.

Gharib, B., 1995. Sogdian Dictionary: Sogdian--Persian--English. Tehran: Farhangan.

Grenet, F., and E. de la Vaissiere. 2002. The Last Days of Panjikent. Silk Road Art and Archaeology 8: 155-96.

Henning, W. B. 1938. Argi and the "Tokharians." Bulletin of the School of Oriental Studies 9/3: 545-71.

--. 1948. The Date of the Sogdian Ancient Letters. Bulletin of the School of Oriental and African Studies 12/3: 601-15.

Kasai, Y. 2014. The Chinese Phonetic Transcriptions of Old Turkish Words in the Chinese Sources from 6th-9th Century: Focused on the Original Word Transcribed as Tujue. Studies on the Inner Asian Languages 29: 57-135.

Kudara, K., W. Sundermann, and Y. Yoshida. 1997. Iranian Fragments from the Otani Collection: Text Volume. Kyoto: Hozokan.

Kumamoto, H. 1996. The Khotanese Documents from the Khotan Area, with an Appendix by T. Saito. Memoirs of the Research Department of the Toyo Bunko 54: 27-64.

Livsic, V. A. 2008. Sogdijskaja epigrafika Srednej Azii i Semirec'ja. St. Petersburg: Filologiceskij Fakul'tet Sankt-Peterburgskogo Gosudarstvennogo Universiteta.

Luije, P. B. 2011. Personal Names in Sogdian Texts. Iranisches Personennamenbuch, vol. II/8. Vienna: Verlag der Osterreichischen Akademie der Wissenschaften.

Mayrhofer, M. 1979. Die altiranischen Namen. Iranisches Personennamenbuch, vol. I. Vienna: Verlag der Osterreichischen Akademie der Wissenschaften.

Minorsky, V, 1970. Hudud al-calam, 2nd ed. London: Luzac.

Muller, F. W. K. 1913. Ein Doppelblatt aus einem manichaischen Hymnenbuch (Mahrnamag). Abhandlungen der Preussischen Akademie der Wissenschaften, Phil.-hist. Klasse, 1912, No. 5. Berlin: Verlag der Koniglichen Akademie der Wissenschaften.

Ragoza, A. N. 1980. Sogdijskie fragmenty central'noaziatskogo sobranija Instituta Vostokovedenija. Moscow: Izdatel'stvo "Nauka."

Rong Xinjiang. 2013. Eighteen Lectures on Dunhuang, tr. Imre Galambos. Leiden: Brill.

Sims-Williams, N. 1981. The Sogdian Fragments of Leningrad. Bulletin of the School of Oriental and African Studies 44: 231-40.

--. 1991. A Sogdian Greeting. In Corolla Iranica: Papers in Honour of Prof. Dr. David Neil MacKenzie, ed. R. E. Emmerick and D. Weber. Pp. 176-87. Frankfurt am Main: Peter Lang.

--. 2014. Biblical and Other Christian Sogdian Texts from the Turfan Collection. Berliner Turfantexte, vol. XXXII. Turnhout: Brepols.

--, and D. Durkin-Meisterernst. 2012. Dictionary of Manichaean Sogdian and Bactrian. Dictionary of Manichaean Texts, vol. III/2. Turnhout: Brepols.

--, and J. Hamilton. 1990. Documents turco-sogdiens du IXe-Xe siecle de Touen-houang. Corpus Inscriptionum Iranicarum, pt. II, vol. III/3. London: School of Oriental and African Studies.

Skjaervo, P. O. 2002. Khotanese Manuscripts from Chinese Turkestan in the British Library: A Complete Catalogue with Texts and Translations. London: The British Library.

Wen Xin. 2008. Zhonggu shiqi Yutian guo zhengzhi zhidu yanjiu [Study in the Political Institutions of Pre-Islamic Khotan], M.A. thesis, Peking University.

Yoshida Yutaka. 1991. Sogdian Miscellany III. In Corolla Iranica: Papers in Honour of Prof. Dr. David Neil MacKenzie, ed. R. E. Emmerick and D. Weber. Pp. 237-44. Frankfurt am Main: Peter Lang.

--. 2000. Sute wen kaoshi [Studies of Sogdian texts] and Baizikelike Moni jiao sute wen shu-xin de geshi [The form of the Manichaean Sogdian letters from Bezeklik], In Tulufan xinchu Moni jiao wenxian yanjiu [Studies in the Manichaean texts recently discovered at Turfan], ed. Liu Hongliang. Pp. 3-199, 250-79. Beijing: Wenwu chubanshe.

--. 2009. Turco-Sogdian Features. Exegisti monumento: Festschrift in Honour of Nicholas Sims-Williams, ed. W. Sundermann et al. Pp. 571-85. Wiesbaden: Harrassowitz Verlag.

--. 2009a. Minor Moods in Sogdian. East and West: Papers in Indo-European Studies, ed. K. Yoshida and Brent Vine. Pp. 281-93. Bremen: Hempen Verlag.

--. 2013. When Did Sogdians Begin to Write Vertically? Tokyo University Linguistic Papers 33: 375-94.

--. and Moriyasu Takao. 1989. A Sogdian Sale Contract of a Female Slave ... [in Japanese]. Studies on the Inner Asian Languages 4 (1988 [1989]): 1-50.

Zhang Guangda and Rong Xinjiang. 2008. Yutian shi congkao [Collected Inquiries on the History of Khotan], new ed. Beijing: Renmin University of China Press.



(1.) For the opportunity to publish these manuscripts we would once again like to thank the Museum and the working group of scholars and experts to whom the study of this collection has been entrusted. We are also extremely grateful to Yutaka Yoshida (= Y.Y.), who kindly read a draft of this article and sent us a large number of corrections and additional remarks, only a few of which we have been able to incorporate in the final version.

(2.) In this connection it is interesting to note a Khotanese document from Mazar Tagh (Skjaervp 2002: 132), to which Yutaka Yoshida kindly draws our attention, which contains the phrase suli bisa "in the Sogdian's house" or (as Yoshida prefers) "the Sogdian village."

(3.) Ragoza 1980: 35-36 (see also Sims-Williams 1981: 235-36).

(4.) LivSic 2008: 84-212.

(5.) See the references in Sims-Williams and Durkin-Meisterernst 2012: 133 and 151-52 respectively.

(6.) See Yoshida 2000: 253, 258.

(7.) For the latter see the photo in Ragoza 1980: 137 (PI. XXIII, bottom).

(8.) Sims-Williams and Hamilton 1990: 27, 28.

(9.) Lurje 2011: 453, no. 1490.

(10.) Lurje 2011: 398, no. 1274.

(11.) Yoshida 2000: 8-9,90.

(12.) Kudara, Sundermann, and Yoshida 1997: 72; Ragoza 1980: 76.

(13.) Sims-Williams 1981: 236.

(14.) Livsic 2008: 109-11.

(15.) Sims-Williams and Hamilton 1990: 52.

(16.) Yoshida 2000: 10.

(17.) Dickens and Sims-Williams 2012: 277.

(18.) See Sims-Williams and Durkin-Meisterernst 2012: 109b and references given there.

(19.) Sims-Williams and Durkin-Meisterernst 2012: 81.

(20.) Gharib 1995: 302-3.

(21.) Apud Henning 1948: 608.

(22.) Sims-Williams and Hamilton 1990: 69.

(23.) Kasai 2014: 132.

(24.) Bailey 1985: 42-43.

(25.) Henning 1938: 567-68.

(26.) Sims-Williams and Hamilton 1990: 55.

(27.) Sims-Williams and Durkin-Meisterernst 2012: 106a.

(28.) Gershevitch 1979: 292-93; Sims-Williams and Durkin-Meisterernst 2012: 12b.

(29.) Cf. Yoshida 2000: 65-66, who interprets these phrases a little differently.

(30.) Ragoza 1980: 36 (but reading zwk for nyk with Sims-Williams and Durkin- Meisterernst 2012: 94 s.v.jwk). As Yutaka Yoshida kindly points out, another example of the phrase nwr my[??] prm in the introductory formulae of a letter is to be found in Ch/So 14715b, line 5.

(31.) See the references in Sims-Williams and Durkin-Meisterernst 2012: 68.

(32.) Kudara, Sundermann, and Yoshida 1997: 75.

(33.) Ragoza 1980: 26, 36 (as corrected in Sims-Williams 1981: 235-36).

(34.) Yoshida 2000: 33.

(35.) See Sims-Williams and Durkin-Meisterernst 2012: 50.

(36.) See Wen 2008.

(37.) Clauson 1972: 426b. On the passage from KaSghari see Dankoff and Kelly 1982-1985, II: 343 fn.

(38.) According to Yoshida 2000: 86-87 and 2009: 574, such Sogdian expressions are calqued on the Uygur formula ukus ukus koyul ayitu, lit. "asking very much (about your) mind/heart."

(39.) In its present condition, no. 9 is 28 cm wide, as is no. 8. No. 5, which is obviously written on a complete sheet and which probably indicates the local standard dimensions of paper in Khotan during the period of these Sogdian documents, is 27.2 cm wide. Several of the unpublished Chinese documents from the Khotan area in the Renmin University Museum collection also are between 27 and 28.2 cm wide. In Dunhuang the size of the paper used for writing documents or copying Buddhist siitras was 26 x 39 or 26 x 52 cm; see Rong 2013: 484.

(40.) See Kudara, Sundermann, and Yoshida 1997: 106-7 with Pl. 33; Yoshida 1991: 241-43.

(41.) In order to obtain a layout in which the beginnings and ends of the lines are adjacent and identical in orientation we must suppose that the letter was folded several times into a long strip; this strip was then folded diagonally in the middle and the right-hand portion wound around the left-hand portion.

(42.) Lurje 2011: 196, no. 495.

(43.) Sims-Williams and Durkin-Meisteremst 2012: 230.

(44.) See Yoshida 2000: 48-49.

(45.) Ragoza 1980: 36, as corrected in Sims-Williams 1981: 235 and 1991: 182.

(46.) See Sims-Williams 1991: 182.

(47.) See Yoshida 1991: 241-42.

(48.) Since the characters following 'ten' are missing, any day between the 10th and 19th is possible.

(49.) Mayrhofer 1979, I: 57, no. 206.

(50.) Lurje 2011: 203, no. 520.

(51.) Officially the second year of Jianzhong, but the era Dali remained in use because Khotan had not yet learned of the change of eras at the Tang court.

(52.) On a small fragment from Domoko (Dom.0158 = Or.8212/1373), to which Dr. Luo Shuai has kindly drawn our attention, the one surviving character zhen is very likely the beginning of a date formula, since it clearly stands at the top of the page. The handwriting of this character, according to Prof. Zhu Yuqi, an expert on Chinese calligraphy, is very close to that of the Chinese characters on our no. 12 (and, coincidentally, several letters written at 90[degrees] to the character H look like the Sogdian word "[??]yt[ 'people'). Although the two fragments cannot be joined directly, it is still possible that they belong to the same document or were written by the same scribe.

(53.) Height and width are indicated on the basis that the Sogdian text is written horizontally, which is merely assumed for the sake of convenience and cannot be taken for granted (see Yoshida 2013).
COPYRIGHT 2015 American Oriental Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Bo, Bi; Sims-Williams, Nicholas
Publication:The Journal of the American Oriental Society
Article Type:Critical essay
Date:Apr 1, 2015
Previous Article:Writing and the recognition of customary law in premodern India and Java.
Next Article:Debates around Jixia: argument and intertextuality in Warring States writings associated with Qi.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters