² Indeed, mashe 麻射 can be traced back to the Khotanese māysa-, or in its locative form māśa, meaning “house” or “workshop”. The meaning of māysa- as “workshop” can be attested in the compound māśa-vīraa- which is rendered as māśa-worker (p. 102) or house-worker (p. 92) in Skjærvø, Prods Oktor, Khotanese Manuscripts from Chinese Turkestan in the British Library: a Complete Catalogue with Texts and Translations (London, 2002)Google Scholar. I have discussed this word in a previous article; see Qing, Duan 段晴, “Yutian yu gaoseng mai nu qiyue” 于闐語高僧買奴契約 [A Khotanese contract recording a monk's purchase of a slave], Journal of Dunhuang and Turfan Studies Vol. 11 (2009), pp. 11–27 Google Scholar, especially p. 19. The Chinese jialan 伽蓝 derives from the Sanskrit samghārama, “monastery”.

³ Xianlin, Ji 季羡林 et al. (annotated), Da Tang Xiyu ji jiao zhu 大唐西域记校注 [Records of the Western Regions composed during the Tang dynasty, with annotations] (Beijing, 1995 [reprint]), p. 1001 Google Scholar.

⁴ See Bailey, H. W., Khotanese Texts IV (hereafter KT IV, Cambridge, 1961)Google Scholar: Khotanese text of Hedin 15 and 16 on pp. 29–30; Chinese-Khotanese equivalents on p. 53; the equivalent Chinese part on p. 173.

⁵ Yutaka Yoshida considered buna as “racine ou base (d'un arbre, employé pour compter les mûriers)”; see his “Découvertes récentes en chine et au Japon: peintures manichéennes et documents sogdiens (VIII^e–XIII^e s.)”, Annuaire de la section des sciences historiques et philologique 142 (2009–2010; published 2011), pp. 57–59, especially p. 59.

⁶ Changru, Tang 唐长孺 (ed.-in-chief), Tulufan chutu wenshu (tulu ben) 1, 吐鲁番岀土文书 (图录本) 壹 [Documents unearthed at Turfan (Illustrations) 1] (Beijing, 1992), p. 6 Google Scholar. See also Changru, Tang, “Tulufan wenshu zhong suojian sizhi shougongye jishu zai Xiyu gedi de chuanbo” 吐鲁番文书中所见丝织手工业技术在西域各地的传播 [The spread of the silk handcraft industry throughout the Western Regions as seen in the Turfan documents], in Shan ju cun gao 山居存稿 (Beijing, Zhonghua shuju, 1989), pp. 388–398 Google Scholar, especially p. 395; and Qing, Duan, “‘Mulberry’ in Khotanese: a New Khotanese Loan Deed in the Hetian Museum”, Bulletin of the Asia Institute, new series, Vol. 19 (2005; published 2009): Iranian and Zoroastrian Studies in Honor of Prods Oktor Skjaervo, pp. 5–14 Google Scholar, esp. pp. 11–12.

⁷ Chunwen, Hao 郝春文, “Du Dunhuang wenxian zhaji (you er ze)” 读敦煌文献札记 (又二则) [Notes on Manuscripts from Dunhuang], in Zhang Guangda xiansheng bashi huadan zhushou lunwenji 張廣達先生八十華誕祝壽論文集 [Papers in celebration of the 80th birthday of Zhang Guangda] (Taipei, 2010), pp. 791–795, especially p. 792Google Scholar.

⁸ Guangda, Zhang 张广达 and Xinjiang, Rong 荣新江, Yutian congkao (zengding ben) 于阗史丛考 (增订本) [Studies in the history of Khotan, revised edition] (Beijing, 2008), p. 258 Google Scholar. A Judaeo-Persian letter refers to a piece of śaṃsī, which was also measured by the piece, rather than by length. In this respect, at least, the śaṃsī and thaunaka are worthy of comparison. In terms of value, Hedin 26 gives the value of one thaunaka as 150 mūra; the document Or.9268B, dated 10 years later, gives the value of one śaṃcī measuring 3 feet (Kh: chā) as 200 mūra. Perhaps it is possible that śaṃcī and thaunaka refer to the same thing?

⁹ Yoshida, Yutaka, “Notes on the Khotanese Secular Documents of the 8^th–9^th centuries”, in Macuch, Maria, Maggi, Mauro and Sundermann, Werner (eds), Iranian Languages and Texts from Iran and Turan, Ronald E. Emmerick Memorial Volume (Wiesbaden, 2007), pp. 463–473 Google Scholar.

¹⁰ Yoshida, “Notes on the Khotanese Secular Documents”, pp. 469–470.

¹¹ Prior to this, Bailey had understood pe'ma as “sheep's wool”, pe'mīnaa or pi'mīnaa, being constructed from pe'ma and the suffix īnaa (“made from”), thus pe'mīnaa-thau “woollen cloth”. Skjærvø, Catalogue, p. 577, had suggested it might be cotton cloth.

¹² This document, of unknown provenance, is dated to the 15^th year of an unidentified king and is entitled “The Head Tax Per Half Man of Gaysata and the Amounts of Summer Clothes to be Collected”, Skjærvø, Catalogue, p. 577. The document bears the shelf mark Godfrey 2 and is marked by Skjærvø as “Khot. missing frags. 2 [Musée Guimet, Paris]”.

¹³ These are recorded in the Song shi 宋史 [Song history] and Song huiyao jigao 宋会要辑稿 [Important documents of the Song], respectively, and are cited in Feng, Zhao 赵丰and Le, Wang王乐, Dunhuang sichou yu sichou zhi lu 敦煌丝绸与丝绸之路 [The Silk Road and textiles from Dunhuang], (Beijing, 2009), p. 219 Google Scholar.

¹⁴ For the relationship between these 2 documents, see Chunwen, Hao 郝春文, Tang Wudai Song chu Dunhuang seng ni de shehui shenghuo 唐五代宋初敦煌僧尼的社会生活 [The social life of Buddhist monks and nuns in Dunhuang during the Late Tang, Five Dynasties and the early Song], (Beijing, 1998), p. 129 Google Scholar.

¹⁵ For discussions on men-dri, see Emmerick, Ronald. E., Tibetan Texts Concerning Khotan (London, 1967), pp. 33, 145Google Scholar; and Yoshida, “Notes on the Khotanese Secular Documents”, p. 470.

¹⁶ One manuscript from the Hetian [Khotan] Museum (HTB000397), and two documents in the St Petersburg Collection, which have three separate object numbers (SI.P.96.11, SI.P.103.16 and SI.P.103.31), see Emmerick and Vorob'ëva-Desjatovskaja, Saka Documents, pp. 111, 141, 148. Unfortunately, in an earlier article (Duan Qing, “‘Mulberry’ in Khotanese”) I did not realise that SI.P. 96.11 and SI.P.103.16 were from the same manuscript.

¹⁷ Zhao Feng and Wang Le, Dunhuang sichou yu sichou zhi lu, pp. 219–222.

¹⁸ Wang, Helen, Money on the Silk Road, The Evidence to c. AD 800 (London, 2004), pp. 95–97 Google Scholar.

¹⁹ Chinese Buddhist Electronic Text Association (CBETA), T50, no. 2053, p. 252, a11.

²⁰ Viśya Sīhä's visit to Chang'an in 650 CE is recorded in several places in the Old Tang History (for example, in the annals section (benji·本紀); see Xu, Liu 劉昫 (887–946), Jiu Tangshu 舊唐書 (Beijing, 1975)Google Scholar.

²¹ See Duan Qing, “Yutian yu gaoseng mai nu qiyue”.

²² See Skjærvø, Catalogue, p. 557.

²³ The Khotanese word pīha-, which has always been understood as “price”, should probably be considered as a word for money, borrowed from Chinese bì 幣, Early Middle Chinese bjiaj^{h ;} , Late Middle Chinese pɦjiaj`; see Pulleyblank, E. G., Lexicon of Reconstructed Pronunciation in Early Middle Chinese and Middle Mandarin (Vancouver, 1991, p. 36)Google Scholar. When the horse groom paid “using his own special money”, it reflects that many different types of coins were in circulation in Khotan at this time.

²⁴ These are in the National Library of China and private collections, see Xinjiang, Rong 荣新江 and Xin, Wen 文欣, “Hetian xin chutu Hanyu – Yutian shuang yu mujian kaoshi” 和 田 新 出 汉 语 － 于 阗 语 双 语 木 简 考 释 [Newly discovered Chinese-Khotanese bilingual tallies], Journal of Dunhuang and Turfan Studies 11 (2008), pp. 45–70 Google Scholar, especially pp. 58–60.

²⁵ See Skjærvø, Catalogue, p. 8. For details of Skjærvø's English translation, see Emmerick, R. E. and Skjærvø, P. O., Studies in the Vocabulary of Khotanese III (Vienna, 19 97), p. 171 Google Scholar. For the translation of Khotanese hālai as “disabled”, see Qing, Duan, “Bisā- and Hālaa- in a New Chinese-Khotanese Bilingual Document”, Journal of Inner Asian Art and Archaeology 3 (2009), pp. 65–73, esp. p. 70Google Scholar.

²⁶ I will discuss this in more depth when I publish the series of Khotanese documents in the National Library in Beijing. For an explanation of this word, see the entry on Pājä in Skjærvø, P. O.'s “Legal Documents Concerning Ownership and Sale from Eighth-Century Khotan”, in Russell-Smith, L. and Ghose, M. (eds), From Nisa to Niya: New discoveries in Central and Inner Asian Art and Archaeology (London, forthcoming)Google Scholar.

²⁷ It is a suggestion of Skjærvø (Catalogue, p. 69) that śaṃcī refers to plain white silk (白绢 bai juan). The same term (šmsy/šamsī) also appears in one of the Judaeo-Persian letters from Dandan Uiliq: see Zhan, Zhang 张湛 and Guang, Shi 时光, “Yi jian xin faxian Youtai Bosiyu xin zha de duandai yu shidu” 一件新發現猶太波斯語信劄的斷代與釋讀 [A newly discovered Judea-Persian letter], Journal of Dunhuang and Turfan Studies 11 (2008), pp. 71–99, especially p. 94Google Scholar. Zhang and Shi, following Skjærvø's suggestion, believe that šamsī is a transliteration of the middle Chinese term dzǝm sɨ 蚕丝 (E. G. Pulleyblank, Lexicon, pp. 44, 292) which means literally “silkworm silk”. However, the context of pāḍā 12 indicates that śaṃcī was a kind of textile, perhaps a special product of Khotan, and that it was valued at the rate of 200 mūra for 3 feet (Kh: chā). Is it possible that the suggestion of Skjærvø was correct? And if so, can we see pāḍā 12 as evidence that juan-silk had become a means of payment in Khotan?

²⁸ Puñadatta appears as a ṣau-official in a wooden document (Hedin 33) that is addressed to a man called Vasirasaga, who was active in about 767.

²⁹ This document is now in the National Library of China.

³⁰ “Militia Unit of Frontier Mercenaries” is translated from the Chinese tuanjie fan bing 团结蕃兵. After the An Lushan Rebellion of 755–763, the mercenaries were not registered with the centrally organised military forces of the Tang dynasty, but were recruited as necessary for territorial defence or combat. In peace time they were summoned for training only when there were slack seasons in farming.

³¹ This form is significant and may be indicative of an early date. There seems to have been a revival of the practice of writing orders on wooden tablets when Khotan was under Tibetan rule, but these were very thin and crudely finished, and look a bit twisted and irregular.

³² Yoshida, Yutaka, “On the Taxation System of Pre-Islamic Khotan”, Acta Asiatica, Bulletin of the Institute of Eastern Culture 94 (2008), pp. 95–126, especially p. 98Google Scholar.

³³ Yoshida, “On the taxation system”, p. 110, deemed that the fee of 9 mūras was for the secretary (Kh. Ka'rä) and director (Kh. varāysai). I consider ka'rä-varāysai as a compound and translate ka'rä-varāysai va as “for sharing with the secretary”.

³⁴ For the original Khotanese and translation, see Bailey, H. W., Saka Documents Text Volume (published on behalf of Corpus Inscriptionum Iranicarum, London, 1968), pp. 1–3 Google Scholar; also Yoshida, “On the Taxation System”, p. 110.

³⁵ This follows Bailey's transcription but offers a translation partly borrowed from Skjærvø. For the original Khotanese, see KT IV, p. 44. For the revised translation, see Emmerick and Skjærvø, Studies in the Vocabulary of Khotanese III, p. 73.

³⁶ In this translation I have followed Bailey's transliteration, and borrowed from Skjaervo's translation.

³⁷ See note 57.

³⁸ For the original Khotanese, see KT IV, p. 45. This translation is partly borrowed from Skjærvø; see Emmerick & Skjærvø, Studies in the Vocabulary of Khotanese III, p. 73.

³⁹ P.3348 has the title “唐天宝四载豆卢军上河西支度使和籴正帐牒”, as given in Jinxiu, Li 李锦绣, Tangdai caizheng shigao 唐代财政史稿 [Research manuscript of Tang dynasty financial history] 5 vols (Beijing, 2007), Vol. 5, p. 158Google Scholar.

⁴⁰ Yoshida, “On the Taxation System of Pre-Islamic Khotan”, p. 113, discusses thaṃga in detail, and concludes that the Khotanese thaṃga corresponds to the Chinese 秤 cheng.

⁴¹ This translates literally as “white” cotton but is perhaps better understood as “plain” cotton? The corresponding Chinese text is “出奭毛 毺細氈, 工紡績”, “多衣絁紬白 氎”, see Ji Xianlin 季羡林 et al., Da Tang Xiyou ji jiao zhu, p. 1001.

⁴² The term kapāysaji thau “cotton cloth” appears in SI.P.103.27; see Yoshida, “On the Taxation System of Pre-Islamic Khotan”, p. 113.

⁴³ Skjærvø gives a revised reading and translation that differs from Bailey's (KT IV, p.45); see Emmerick and Skjærvø, Studies in the Vocabulary of Khotanese III, p. 173.

⁴⁴ Vidyadatta also officiated in a lawsuit document (pāḍā) (Or.9268B) that is dated to the 17^th year of an unnamed king.

⁴⁵ According to Bailey, “hvā cai ṣī śī’na” is a group of four Chinese signs. While ṣī could be Chinese史, śī’na could well represent Chinese 人. See his translation and commentary in KT IV, pp. 140–141. I suggest that the Khotanese ṣī might correspond to the Chinese 使, hence ṣī-śī’na 使人 literally “carrier”. Anyway, “hvā cai ṣī śī’na” represents the moneylender in this document, whereas the debtors had all signed at the end of this document.

⁴⁶ I have retranslated this passage following Bailey's transliteration; see Bailey, H. W., Saka Documents Text Volume (published on behalf of Corpus Inscriptionum Iranicarum, London, 1968), p. 13 Google Scholar; and KT IV, pp. 38, 140. For research on particular vocabulary, see Emmerick and Skjærvø, Studies in the Vocabulary of Khotanese III, p. 65.

⁴⁷ In the list of names of people who signed to accept prepayment in Hedin 26, there are two names – Upadattä and Viśi'ripuñi (KT IV, p. 38) – which occur also in the name list of BH01–15 (see Duan Qing, “Bisā- and Hālaa-”, p. 66), but which are ascribed to two different hamlets or villages. Upadattä's hamlet is not clear due to damage to the document, whereas Viśi'ripuñi (also written as Vaśi'ripuñi), whose Chinese name was 勿日本 wu ri ben, and who was noted as being disabled, came from Bayajvā biśa “hamlet of Bayajuas”(see Duan Qing, “Bisā- and Hālaa-”, pp. 66, 70). This means that the men of Hedin 26 defined as residents of Gaysāta canton came from different hamlets or villages, and not just from one settlement.

⁴⁸ The original text gives gvaṃcāṃdä (3 pl. pf. tr. m.) and gvaṣceṃ (1 sg. pf. tr.), from the root gvaṣc-. This Khotanese word always seems to appear in the context of interest payments. For this reason, I believe that the 300 and 500 refer to interest. I have followed Skjærvø's reading and offer new translations, retaining those areas where I agree with him or where the meaning is uncertain; see Skjærvø, Catalogue, p. 5.

⁴⁹ This document is in the St Petersburg collection. I have followed the transcription in Zhang Guangda and Rong Xinjiang, Yutian congkao, p. 276.

⁵⁰ There appear to be some errors in the calculation of these three numbers in the original document.

⁵¹ I have retranslated it here, following Skjærvø's transcription (Catalogue, p. 577), and for convenience, have replaced the Khotanese numerals with Arabic numerals.

⁵² Zhang Guangda and Rong Xinjiang, Yutian congkao, p. 261.

⁵³ For a thorough discussion of this bilingual document, see Kumamoto, Hiroshi “Sino-Hvatanica Petersburgensia (Part I)” Manuscripta Orientalia Vol. 7, No. 1 (March 2001), pp. 3–9 Google Scholar. The original text is linked to the PDF icon at http://www.gengo.l.u-tokyo.ac.jp/~hkum/works.html.

⁵⁴ The document is dated to the 17th day of the month of Rrāhaja in the 17^th year of Viśa’ Vāhaṃ's reign. This corresponded with the last month of the Chinese year. Calculations were made as to how many people were to pay tax in the month of cvātajä (which corresponded with the first month of the Chinese year). Skjærvø, Catalogue, pp. 7–8.

⁵⁵ The document is dated to 29^th day of the month of Braṃkhaysja in the 19^th year of Viśa’ Vāhaṃ's reign.

⁵⁶ Li Jinxiu, Tangdai caizheng shigao, p. 158.

⁵⁷ “Weaver” is rendering Khotanese kīrarā in the plural nominative. I regard the old translation “working men” as meaningless for every man paying tax was a working man. Hedin 1 is dealing with payment of silk, and the amount of silk of every man that was going to be settled. In this case a professional weaver better suits the context. Thus I regard kīrarā hvaṃḍi as “weavers” (literally “weaver-men”) which can be seen representing professional households engaging in silk textile production, similar to the zhi hu 织户 in central China.

⁵⁸ I am grateful to Zhang Zhan for bringing these conversions to my attention. The translation follows the readings by Bailey and Skjærvø; see Emmerick and Skjærvø, Studies in the Vocabulary of Khotanese III, p. 140.

⁵⁹ Tang huiyao 唐会要 [Important documents of the Tang], juan 83.

⁶⁰ Skjærvø,Catalogue, p. 3.

⁶¹ This way of understanding the Khotanese (i.e. as textiles “with a value of” x mūra, rather than textiles “and” mūra), was suggested by Skjærvø (private communication, May 2004); see Helen Wang, Money on the Silk Road, p. 106, n. 9.

⁶² Emmerick, R. E. and Vorob'ëva-Desjatovskaja, Margarita I., Saka Documents Text Volume III: the St Petersburg Collection (London, 1995), p. 149 Google Scholar.

⁶³ Details are given in footnote 16.

⁶⁴ This document is in the British Library collection.

⁶⁵ Skjærvø, Catalogue, p. 4.

⁶⁶ This document is in the St Petersburg collection.

⁶⁷ Yoshida, ‘On the Taxation System of Pre-Islamic Khotan’, p. 113, gives a very good translation of this Chinese sentence, and discusses the Chinese character cheng and its Khotanese equivalent thaṃga. For Emmerick's reading and translation, see Emmerick and Vorob'ëva-Desjatovskaja, Saka Documents, pp. 156–157.

⁶⁸ Xinjiang, Rong and Xin, Wen, “Newly Discovered Chinese-Khotanese Bilingual Tallies”, Journal of Inner Asian Art and Archaeology 3 (2008; published in 2009), pp. 99–118 Google Scholar, especially p. 108.

⁶⁹ See note 57.

⁷⁰ Yoshida, Yutaka, “Kōtan shutsudo 8–9 seiki no kōotan-go sezoku bunsho ni kansuru oboegaki” コータン出土8–9世紀のコータン語世俗文書に関する覚え書き, Kōbe-shi Gaikokugo daigaku gaikoku-gaku kenkyūjo kenkyū 38 (2006) p. 47, n.66 summarises discussions on the Six TownsGoogle Scholar.

⁷¹ My reading differs from Skjærvø’s; see the original document on http://idp.bl.uk/database/large.a4d?recnum=13825&imageRecnum=110659; and in Skjærvø's Catalogue, p. 98.

⁷² Li Jinxiu, Tangdai caizheng shigao, p. 158.