The Tokharians and Buddhism 1

Introduction: On the Tokharians and the Yuezhi

From the end of the nineteenth century to the beginning of the twentieth, a great number of manuscripts in Indo-European languages were discovered in northwest China (mainly in Xinjiang and Dunhuang, Gansu). It has been revealed that the languages in which these manuscripts were written include Gāndhārī, Pahlavī, Sogdian, Parthian, Khotanese, Tumshuqese, etc. Also found were texts in another ancient Indo-European language, different from the Indo-Iranian language listed above and written in the Brāhmī scipt. Two dialects of this language, A and B, have been identified. 2 Based on the colophons of Maitrisimit, a famous Buddhist play written in Uighur, F. W. K. Müller, Ε. Sieg, and W. Siegling named this ancient language “Tokharian”, in their works. One of these Uighur colophons, no.48, reads:

Nakridiš ulušta toγmiš Aryac in tri bodisvt kši aċari Äntkäk tilint in...

Toχri tilincä yaratmïs Il-baliqda toγmïš Prtanyarakšit kši acari Toχri

tilintin Türktilicä ävirmiš Maitri … [si] mit nom bitig. 3

W. B. Henning has translated this paragraph into English as follows:

The sacred book Maitreya-Samiti which the Bodhisattva guru ācārya Āryacandra, who was born in the country of Nagaradeśa4, had composed5 in the Twγry language from the Indian language, and which the guru ācārya Prajñarakita, who was born in Il-baliq6 translated from the Twγry language into the Turkish language.

During the decades that followed, many scholars hotly debated the nomenclature of this language and a series of related historical, geographical and ethnological issues, and especially its relationship to the Yuezhi and Kushan people.7 Most of them hold that the Tokharian dialects A and B are actually Agnean and Kuchean.8 However, many questions about this theory still need to be resolved, and “Tokharian” as a useful term should not be dismissed.

The extant Tokharian documents date from the period between the sixth and the eigth centuries. However, Tokharian itself is an ancient Indo-European language belonging to the Centum branch, more closely related to Celtic, German, Italian, and Greek than to other languages. 9 This means that an Indo-European people rather than those speaking Eastern Iranian (the Satem branch) entered modern Chinese territory at a very early time. The British scholar T. Burrow, who studied the Kharosthī documents unearthed in Niya, Loulan and Shanshan, pointed out long ago that many grammatical phenomena and the vocabulary of Niya vernacular were close to Tokharian. 10 Therefore, the residents of the Shanshan state were speaking a Tokharian language which was somewhat different from the later Agnean and Kushan. That is to say, there existed a third Tokharian dialect, and the Tokharian entry into the Tarim Basin can be traced back to the second and third centuries.

Furthermore, there have been some very important archaeological discoveries in Xinjiang in recent years which may provide new clues to the origin of the Tokharians. For example, in 1979 the Institute of Archaeology at the Xinjiang Academy of Social Science excavated forty-two ancient tombs in the lower Kongque River valley, seventy kilometers west of the presently dry Lake Lop-nor. These tombs, which constitute an important site of the Gumugou Culture, date from the Bronze Age, approximately 3000 years ago. The anthropometric studies of the human skulls collected from these tombs have shown that the Gumugou people possessed primitive Caucasoid features and that their physical characteristics had certain similarities to the Nordic or northern European type. 11 Moreover, a large number or mummies has recently been found in Xinjiang. These mummies, of which the oldest date from 4000 BC, also show Caucasoid features. May we surmise from these facts that, as early as three or four thousand years ago, the Caucasian residents of the Tarim Basin were already in certain ways related to the Tokharian people who came later?

The Yuezhi月支people recorded in the Chinese histories might be related to

the Tokharians. Since the 1970s several scholars have proposed that the Yuezhi were a branch of the Tokharians. Detailed arguments can be found in articles by B. Henning, A. K. Narain, Lin Meicun 林梅村, and myself. 12

It is commonly accepted that the “Yuzhi” 禺知 people mentioned in the Mu tianzi zhuan 穆天子传[Biography of Mu, the Son of the Heaven], the “Yuzhi”, 禺氏 people in the “Wanghui”, 王会 chapter on Yi Zhoushu 逸周书, as well as in the “Guoxu”, 国畜, “Kuidu”, 揆度, “Qingzhong jia”, 轻重甲, and “Qingzhong yi”, 轻重乙 chapters of Guanzi 管子[Book of Guanzi], the “Yuezhi”, people in the “Yiyin chaoxian”, chapter of Yi zhoushu, and the “Niuzhi”, 牛氏 people in the “Dishu”, 地数 chapter of Guanzi, are all the same as the Yuezhi people. During the Qin and Han Dynasties, the Yuezhi were one of the three major ethnic groups (the other two were the Eastern Hu and Xiongnu) to the north of China, living between Dunhuang and the Qilian Mountains, “residing wherever there were water and grass”, So they must have been active in the vast area from the Tarim Basin to the Ordos Grassland. The power of the Yuezhi was weakened after they were defeated by the Modu Chanyu of the Xiongnu. After their king was killed by another Xiongnu leader, Laoshang, the Yuezhi were divided into two groups, one called Greater Yuezhi and another called Lesser Yuezhi. The former moved westwards, conquered Bactria, and established a kingdom in south Central Asia, leaving a remarkable chapter in world history.

It is after the westward migration of the Tokharian-Yuezhi people that the term “Tokharian”, began to appear in the documents of various languages. According to Strabo’s Geoguaphy (xi. 8.2), the four nomadic peoples who took Bactria from the Greeks were the Asii, Gasiani, Tochari, and Sacarauli. Trogus, on the other hand, records that “the Scythian tribes, the Saraucae and Asiani, conquered Bactria and Sogdiana”, and that “the Asiani [became] the kings of the Tochari, and the Saraucae, were destroyed”. We believe that one or a few of the four people who were mentioned in the Greek sources as having conquered Bactria must have been the Yuezhi. The Yuezhi was a tribal federation dominated by the Tokharians. Yet in the course of their westward migration, they also absorbed various Eastern-Iranian speaking-Saka tribes. 13 According to “Xiyu zhuan”, 西域传[Account of the Western Regions] in both the Han shu 汉书[History of the Han Dynasty] and Hou Han shu 后汉书 [History of the Later Han Dynasty], the Greater Yuezhi were later broken into “five divisions under five xihou 翕侯 leaders”, of which the Kushan division was the most powerful. In the early first century, the Kushan xihou Kujula Kadphises unified the five divisions, broke away from the control of the Hellenized Bactrian dynasty, and established the Kushan Empire.

All the different Tokharian groups mentioned above were influential in the transmission of Buddhism across Central Asia to China. In the following section I will explore this point, relying principally on Chinese sources.

1. The Tokharians, the Yuezhi and the Transmission of Buddhism to China

Buddhism spread to northwest India and its neighbouring countries very early. According to the Aśokan inscriptions, Indian envoys reached Parthia, Bactria, Egypt, and Greece, We know for sure that as early as the mid-third century BC, Buddhism flourished in Qandahar in southern Afghanistan. In the early second century BC, the Bactrians, who were ruled by the Greeks, invaded northwestern India, but later Bactria itself became divided. Menander (or Menandros, rendered as Milinda in Pāli), the king of the Hellenistic city state whose centre was Sāgala (modern Siālkot in Pakistan ), is well known for his discourse with Nāgasena, a prestigious monk from Jibin (presenday Peshawar, Pakistan), and was allegedly converted to Buddhism. This discourse was recorded and compiled into the Milindapañhā in Pali and translated into Chinese as the Naxian biqiu jing 那先比丘经 [Sūtra of Bhiksu Nāgasena]. 14 After the Tokharians, namely the Yuezhi, conquered Bactria in the middle of the second century during their westward migration, they inherited Buddhism, which had already taken root there.

At the latest the Greater Yuezhi had converted to Buddhism by the first century ВС. The country expanded rapidly after Kujula Kadphises established the Kushan Dynasty, and within one hundred years the Yuezhi had invaded Parthia, taken Gaofu (today’s Kabul in Afghanistan), and destroyed Puda (today’s Gwadar in Pakistan) and Kashmir. From the first century AD, the famous Gandharan art began to a appear. In the early second century, the king of the Kushans, Vima Kadphises, known in the Chinese sources as Yan’gaozhen 阎膏珍, further expanded the country by occupying the Indus River region in Pakistan. Then the Kadphises royal house was replaced by the -ska family. The founder of this new royal house was the historically renowned Kaniṣka. 15

The exact date of Kaniṣka’s accession to the Kushan throne has not been confirmed, and the entire chronology of the Kushan empire has also been the subject of heated controversy. 16 According to our present understanding, Kaniṣka’s accession probably occurred sometime between AD 78 and 144, with the year AD 128 being the most likely specific date. Since Kaniṣka employed a policy of supporting and sponsoring various religions, Buddhism was able to develop rapidly. The famous Fourth Council of Buddhism (actually a conference of the Sarvāstivāda school) was summoned during Kaniṣka’s reign. He built Buddhist temples and stūpas throughout the kingdom. The Jaurya (Queli 雀离) Stūpa, which he built at his capital Puruṣapura (today’s Peshawar), was reportedly seen by the Northern Wei emissary Song Yun 宋云 and a pilgrim Huisheng 惠生 who passed by here on their way to India in search of Buddhist scriptures in the early sixth century. 17 Research has shown that Queli and Zhaohuli 昭怙厘18, the name of another Buddhist temple in Kucha reported by Xuanzang 玄奘 in the first chapter of his Da Tang xiyu ji 大唐西域记 [Accounts of the Western Regions], must be the same Tokharian word. 19

The most important Kushan Buddhist site excavated in former Soviet Central Asia is Kara-tepe in ancient Termez. The archaeological find include stone stature, sculptures, Kushan coins, and inscriptions in the Kharoṣṭhī and Brāhmī scrpts. There are also inscriptions in local Васtrian, written in a cursive style of Greek script. 20

Zhang Qian’s 张骞 journey to the Western Regions during the Western Han period marked the official opening of the Silk Road, which connected inland China with Central Asia.lt has long been a hotly debated issue when Buddhism was transmitted from India to China. Nevertheless, one thing is known for sure: the Tokharian-Yuezhi people played a key role in this transmission. In a passage from Yu Huan’s 鱼豢 Weilue 魏略[A Brief History of the Wei] quoted by Pei Songzhi 裴松之 in his commentary to the Dong Yi zhuan 东夷传 [Account of the Eastern Aliens] chapter in the Weizhi 魏志[History of the Wei] on the Sanguo zhi 三国志[History of the Three Kingdoms], there is a clear records: In the first year of the Yuanshou Reign of the Han Emperor Aidi 哀帝, Jing Lu 景卢, a student at the Grand Academy, received the dictation of the Futujing 浮屠经 from Yicun, an envoy sent to China by the king of the Greater Yuezhi. It was he who had reestablished [Buddhism in China]. All the terms such as pusai 蒲塞, sangmen 桑门, bowen 佰闻, shuwen 疏问, boshuxian 白疏, biqiu 比丘, and chenmen 晨门 appearing in this sūtra, are titles of [the Buddha’s] disciples.

This event is also reported in the following works: Liu Xiaobiao’s 刘孝标 commentary to the “Wenxue 文学[Literature]” chapter of Shishuo xinyu 世说新语 [New Words and Sayings of the World] “Shi Lao zhi 释老志 [Treatise on Buddhism and Daoism]” in “the Weishu 魏书[History of the Northern Wei]” “Jingji zhi 经籍志[Bibliographical Treatise]”, “Suishu 隋书 [History of the Sui]”, the fifth chapter of Falin’s 法林 Bianzheng lun 辩正论[Treatise on Defending the Right], Zhang Shoujie’s 张守节 commentary to the “Dayuan liezhuan 大宛列传[Account of Ferghana]”, the Shiji 史记[Records of the Historian], the 193rd chapter of the Tongdian 通典 [The Comprehensive Codex], the Buddha’s Sūtra [futu jing 浮屠经] of the Jin and Song dynasties cited in the 196th chapter of Tongzhi 通志[Comprehensive Accounts], and Jin zhongjing [The Middle Sūtra of the Jin] quoted in the second chaper of the Guangchuan huaba 广川画跋 [Guangchuan’s Postscripts to Paintings]. However, Jing Lu’s name is written as Qin Jingxian 秦景宪 in the Weishu, and in Bianzheng Lun we find another version of the story about Qin Jing going to the Yuezhi country, whose king ordered his son to teach [Qin] the Futu jing, which is similar to the account in the Jin zhongjing.

After the Greater Yuezhi migrated westwards to Bactria, they quickly assimilated themselves to the local culture. Therefore, it is highly possible that Buddhism was prevalent there in the late first century ВС, and that a Greater Yuezhi envoy to China at that time orally transmitted a Buddhist scripture to a Chinese student. 21 Tang Yongtong has correctly pointed out that the Greater Yuezhi’s invasion of Bactria was an important event in the history of Buddhist transmission to China, that the Greater Yuezhi converted to Buddhism during the Western Han period, and that Buddhism probably came to China from Bactria. Therefore the beginning of Buddhist translation should be traced back to the late Western Han.22 The scripture(s) referred to as Futu jing said to have been translated in this period might have been a scripture describing Buddha’s life, similar to the later sūtras like the Benqi jing 本起经[Sūtra on the Buddha’s Origin]23, and the Benxing jing 本行经[Sūtra on the Buddha’s Deeds]24, Later on, quite a few Buddhist monke from the Greater Yuezhi began to arrive in China for missionary and translation work.

There is a well known legend in which it is told that, in the seventh year of the Yongping 永平七年 reign, i.e. AD64, the Emperor Mingdi 明帝 dreamed of the Buddha and then sent envoys to the Western Regions in search of the Buddhist teachings. This highly fictional story has many different versions. Its earliest version is found in the preface to Sishier zhang jing 四十二章经[Sūtra in Forty-two Sections]:

One night in the past, the Emperor Han Mingdi dreamed of a deity, who had a golden hued body and rays like the light of the sun emanating from his neck, flying in front of the palace. This made the emperor ecstatic and pleasd. The next day the Emperor asked his ministers: “Who was that person?” The learned Fu Yi answered: “I have heard that in India there is a person who has obtained the Way, called the Buddha. He can easily rise and fly. He is most likely the deity you dreamed of.” Upon hearing this, the emperor understood and immediately sent twelve people, including the envoy Zhang Qian, the Court Gentleman Qin Jing, and an erudite student Wang Zun to the Greater Yuezhi. They copied the Sūtra in Forty-two Sections and placed it in fourteen stone cases. [The emperor] established stūpas and temples [for the sūtra].Thus the Dharma was widely spread, and Buddhist temples were set up everywhere.

Later various elements were added to the story, such as that when Zhan Qian and Qin Jing arrived in the Western Regions they met a monk called Zhu Moteng 竺摩腾, i. e. Kāśyapa Mātanga, from whom they copied the scripture in question, then returned to Luoyang, where it was kept in the fourteenth stone chamber of Lantai 籣台 or Orchid Tower. 25 All of these stories concerning the earliest transmission of Buddhism to China involved the Greater Yuezhi. Despite their obvious fictional elements, they clearly indicate that it was the Yuezhi who were closely linked with the early Buddhist translations in China.

Here we cannot discuss problems such as the authenticity, transhation and natrue of the Sūtra in Forty-two Sections in detail. However, its close relation with Dharmapada [Faju jing 法句经]has to be pointed out. 26 The Gandhārī version of this scripture written in Kharoṣṭhī script discovered in Khotan was thoroughly examined by J. Brough in the early 1960s. 27 Kharoṣṭhī was one of the official scripts used by the Kushan Empire, and the grammar and vocabulary in this Kharoṣṭhī Buddhist scripture resemble those of the Kharoṣṭhī inscriptions of the Kushan Empire. Hence a careful comparison between the Gandhārī Dharmapada and the Chinese Sūtra in Forty two Sections would be most helpful.

Professor Ji Xianlin has already argued that the languages of ancient Central Asia and Xinjiang, such as the various Iranian and Tokharian languages, influenced the Chinese translation of Buddhist scriptures. 28 As early as 1947, he demonstrated that the Chinese word fo is not a direct translation from the Sanskrit buddha, but probably of Tokharian origin, such as pät-in Agnean and pud- [or pūd] in Kuchean. Yet, according to Bernhard Karlgren’s reconstruction, the ancient prounciation of the Chinese character fo 佛 b’iwət / b’iuət begins with a voiced consonant, while in Tokharian it always begins with an unvoiced consonant. In 1970, the German scholar F. Bernhard supported Ji’s hypothesis, maintaining that fo was a transcription of *but in a Tokharian dialect that predates the A and В dialects (cf. pudñäktc in the В and ptāñkät in the A dialect). 29 E. G. Pulleyblank also regards the original from of fo to be but. 30 In 1979, a small bronze stature of a sitting Buddha, inscribed with one line of Kharoṣṭhī letters on the bottom, was found at a site in the ancient Chinese capital of Chang’an. Acсording to Lin Meicun, it is dated to no later than the end of the fourth century, and it was evidently produced by the Yuezhi immigrants from Kushan who had been moving to China in increasingly great numbers since the mid-second century. 31 The insсription on the bottom of this statue contains a word meaning Buddha, written as buca. The transformation from t into с is a known feature of Tokharian, also seen in the oldest stratum of Tokharian used in Kharoṣṭhī documents from Loulan. Therefore, buca is a Tokharian term used by the Yuezhi people. This evidence further confirms Ji’s hypothesis.

2. The Yuezhi Buddhist Translators in China

It is possible to know a great deal about the situation of Buddhism in the Greater Yuezhi kingdom through the Buddhist scriptures which were brought from that country to the East and there translated into Chinese.

Most of the people who came from the Western Regions to China and adopted the Chinese surname from the Western Regions to China and adopted the Chinese surname Zhi支 during the second to fifth century were more or less related to the Yuezhi. One of them, Lokakṣema (Zhi Loujiachen 支娄迦谶, sometimes abbreviated to Zhi Chen支谶), was the most famous Buddhist translator during the Later Han period.

He was originally a Kushan śrāmāna and arrived at Luoyang in the late years of the Emperor Han Huandi’s reign. In AD 178 and 179, he translated more than ten Buddhist sūtras from Central Asian languages into Chinese, including the Astasāhasrikā-prajñāpārаmitā sūtra, the Śūramgama samādhi sūtra, the Pratyutpanna buddha sammukhāvasthita samādhi sūtra, the Ajātaśatru kankṛtyavinodana, and the Ratnakuta. Among the sūtras translated by Lokakṣema the most noteworthy is that belonging to the prajñāpāramitā class of scriptures which laid the foundation for the early development of Mahāyāna Buddhism in China. The fact that his translation of the Astasāhasrikāprajnāpāramitā sūtra, also called the Xiaopin boruo 小品般若[Small Prajñāpāramitā], had already been re-translated twice by the time of Kumārajīva cleary shows its great influence. The Mādhyamika school of Mahāyāna might have evolved from the Mahāsāmghika tradition, which originated in southern India and had been transmitted to the north by the time of Kaniṣka. Chinese Buddhists regard Aśvaghoṣa as the first advocator of Mahāyāna Buddhism, and he was said to have been highly respected by King Kaniṣka. During the Eastern Han period, Mahāyāna scriptures had already become popular in the Kushan Empire. By the end of the Eastern Han, Mahāyāna sūtras, including the Prajñāpāramitā and the Vaipulya classes of scriptures, had made their way to China.Therefore, it is not surprising at all for us to see that the early Mahāyāna Buddhist system in China was established by the Yuezhi Lokakṣema, rather than by some one of another nationality.

It is known that a Yuezhi monk, Zhi Yao 支曜, engaged in Buddhist translation at Luoyang in AD 185. The Chengju guangming jing 成具光明经[Sūtra on the Completion of Brightness], is the only extant translation that can be definitely identified as being made by Zhi Yao, also belongs to the Mahāyāna tradition.

One of Zhi Chan’s known students was Zhi Lianag 支亮 (also styled Jiming 纪明). It is uncertain whether he was an upāsaka or śrāmāṇa, and some scholars even suggest that Zhi Liang and Zhi Yao were actually one and the same person. 32 In Chinese both liang and yao mean “light” or “brightness”; they were probably used to translate same Sanskrit word prabhāsaka.

Another Yuezhi monk Zhi Qian支谦(alsо names Yue 越 and styled Gongming恭明) translated as many as thirty-six Buddhist sūtras in forty-eight chapters between AD 222 and AD 253. His grandfather, Fadu 法度，the leader of the several of hundreds Greater Yuezhi people who migrated to China during the reign of the Emperor Han Lingdi, was appointed Court Gentleman by the Han court. Zhi Qian studied with Zhi Liang and thus became the second generation disciple of Lokakṣema. He is said to have studied Buddhist texts from the age of ten and various Central Asian languages from the age of thirteen. He is said to have mastered six languages and was well read in the Chinese classics. Sun Quan 孙权，the ruler of the Wu Kingdom, was deeply impressed by Zhi Qian’s explanation of Buddhist scriptures and gave him the title of Boshi 博士, i.e. Erudite Scholar, with the responsibility of working with Wei Zhao韦昭 and other scholars to counsel and instruct the crown prince. 33 The scriptures that he translated covered a wide spectrum, inchiding both Mahāyāna and Hīnауānа texts. His most important translations include the Vimalakīrtinirdeśa [Weimojie jing 维摩诘经] in two chapters, the Astasāhasrikāprajñāpāramitā sūtra [Da mingdu wuji jing 大明度无极经]34 in four chapters, a biography of the Buddha, the Taizi ruiying jing 太子瑞应经[The Scripture on the Auspicious Deportment of the Prince]35, etc. He also collated Wei Zhinan’s 维祗难 translation of the Dharmapada. Zhi Qian inherited Lokaksema’s philosophical system and tried to make his translations smooth and readable. For example, when he was translating the Anantamukhasādhaka-dhāranī36, he succeeded both in maintaining the original eight-syllable format and correctly translating the meaning, instead of just transcribing the sounds. He proved himself to be a literary master well versed in rhymes and cadence, as shown in his composition of the Zan pusa lianju fanbai 赞菩萨连句梵呗[Hymn of Linked Verse in Praise of the Bodhisattva]. The scriptural commentary he made for his own translation of the Śalistambhaka sūtra is the earliest example of this kind of Buddhist literature in China.

Yet another Buddhist translator with the surname “Zhi” was Zhi Jiangjie liang 支疆接粱 (Kālasivi?), who also might have come from the country of the Yuezhi. While residing in Jiaozhou 交州 (present-day Hanoi in Vietnam) in either AD 255 or AD 256, he translated the Saddharmapuṇḍarīka sūtra [Zheng fahua jing 正法华经]37 in six chapters.

However the most eminent translator during the Western Jin period was Dharmarakṣa (Zhu Fahu 竺法护), whose ancestors had lived in Dunhuang 敦煌 for generations. Although he was of the Yuezhi nationality, when Dharmarakṣa became a monk at the age of eight under an Indian monk Zhu Gaozuo 竺高座, he adopted his teacher’s surname. When he was young, Dharmarakşa travelled with his teacher to many countries in the Western Regions and learned several Central Asian languages and scripts. Following this he returned to China with a large number of Buddhist texts. In AD 266 he travelled from Dunhuang to Chang’an and Luoyang, and later crossed the Yangzi River. During his travels he is said never to have stopped teaching and translating. He translated some one hundred and fifty Hinayāna and Mahāyāna sūtras38, virtually covering all important texts circulating in the Western Regions. Thus, he greatly expanded the possibilities for the further development of Mahāyāna Buddhism in China. Among the eighty-six translations attributed to Dharmarakṣa that have survived up to the present are the Pañcavimśatikasāhasrikāprajñāpāramitā sūtra39 in ten chapters, the Saddharmapuṇḍarīka sūtra40 in ten chapters, the Daśabhūmika sūtra41 in five chapters, the Lalitavistara in eight chapters, etc. Dharmarakṣa was often assisted by men like upāsaka Nie Chengyuan 聂承远 and his son Nie Daozhen 聂道真, who not only took the responsibility of writing down Dharmarakṣa’s oral recitation and checking the translation, but also translated some texts by themselves. Besides, they recorded information about the original texts and the place of translation, which constituted the earliest Chinese Buddhist catalogue commonly called the Nie Daozhen lu 聂道真录[Nie Daozhen’s Catalogue]. 42

Although their ethnic attributes are not specified in scriptural catalogues, Zhi Fadu 支法度 and Zhi Daogen 支道根, two other Buddhist translators active during the fourth century, were most likely directly or indirectly related to the Yuezhi.

According to Biqiuni zhuan比丘尼传 [Biographies of Nuns]43, the monk Sengjian僧建 obtained the Mahāsaṃghikakarmavācana and the Prātimokṣa for nuns in the Yuezhi country between AD 335-342, and translated them at Luoyang. This fact indicates that the Bhikṣūṇī Prātimokṣa was in circulation in the Yuezhi. There was also a monk by the name of Zhi Shihin 支施论 (fl. late 4th cent.), who translated some Vaipulya scriptures, including the Susthitamati [devapūtra sūtra] pariprachā44, the Shang jinguangshou jing 上金光首经 [Scripture of the Supreme, Golden Light Uṣṇiṣa?]45 and the Śūraṃgama samādhi sūtra.

After the Former Qin Kingdom (359-394) unified North China and re-established direct communication with the Western Regions, a Tokharian monk called Dharmanandhī (Tanmonanti 昙摩难提) arrived in China and translated the Madhyamāgama and the Ekottarāgama sometime during the Jianyuan reign period, i. e. AD 364-389. These are the earliest translations of major Āgamas. The two eminent Chinese monks, Daoan 道安 (d.385) and Fahe 法和 (f l. 4th cent.), examined these Āgamas, while the former wrote a preface for the Chinese version of the Ekottarāgama.

In AD 433, the monk Daotai道泰obtained the Sanskrit version of the Mahāvibhāṣā in more than one hundred thousand gāthās from the area west of the Pamirs. Four years later, this sūtra was translated into Chinese at Liangzhou梁州by Buddhavarman, who also said to be of Tokharian descent. 46 It is well known that the Mahāvibhāṣā was quite popular among the Yuezhi.

In summary, Yuezhi monks translated a great number of Buddhist scriptures into Chinese, most of which seem to have been Mahāyāna sūtras, including the Avataṃsaka, Vaipulya, Prajñāpāramitā, Saddharmapuṇḍarīka and Nirvāṇa. These translations greatly accelerated the development of Chinese Buddhist doctrine and philosophy. As for the original languages in which these scriptures were written, no thorough examination has been made so far. It seems that most of them were written in some form of Sanskrit or Buddhist Hybrid Sanskrit, though some probably contained elements of various Central Asian languages including Tokharian. The question as to whether many of the early Chinese Buddhist sūtras were translated from Central Asian languages is still an important subject that needs further study.

3. Kumārajīva and Kuchean Buddhism

Kucha was a state established by the Tokharians on the northern edge of the Tarim Basin, and it is not clear when Buddhism first spread to this area. An account in the Ayuwang taizi huaimu yinyuan jing阿育王太子坏目因缘经[Scripture on the Causes and Conditions of Prince Aśoka]47 which says that Kucha was among the lands Asoka gave to his son Fayi 法益 is obviously a fable and should not be taken at face value. However, according to Chinese sources, as early as the third century some Buddhist monks from Kucha arrived in the Chinese heartland to translate and teach. It is for example recorded that a Kuchean prince referred to as Bo Yan 白延 took part in the translation of the Śūraṃgamasamādhi sūtra together with Zhi Shilun. It is also said that Bo Yan was good at both Chinese and foreign languages, well read in a variety of classics, and that he mastered both Buddhism and Confucianism. Other Kuchean Buddhists active in Chinese heartland during the Western Jin period were the layman Shan Yuanxin单元信 and srīmitra (Bo Shilimiduoluo帛尸梨密多罗), a member of the Kuchean royal house. Another famous monk, Fotudeng 佛图澄, who arrived at Luoyang in AD 310 and whose original surname was Bo帛, was also a Kuchean. After the Later Zhao regime was established, he became a confidant of the Zhao rulers such as Shi Le 石勒 and Shi Hu 石虎. He advised them to be lenient, and made every effort to spread Buddhism among the common people. Although he is not credited with having translated any Buddhist sūtras, he worked in northern China for many years and had a great impact on the subsequent development of Chinese Buddhism.

During the fourth century Buddhism became increasingly popular in Kucha, and the number of Buddhist monks in that county reached more than ten thousand. In the capital alone no less than one thousand temples and stūpas were established, and Buddhist statues were worshiped in the royal palace as well as in the temples. Some temples were magnificent and extensive, including the famous Queli Temple located at Subasi to the north of the seat of taday’s Kucha County, whose remains have been found by archaeologists. 48 At that time the most famous monk within the Hīnayāna Buddhist clergy around Kucha was Buddhakṣema (Fotushemi 佛图舌弥). He was in charge of many temples, including three large ones for the nuns, some of whom were princesses of the royal houses of Eastern Central Asian kingdoms, and who had come to Kucha to learn Buddhism. The Kuchean Vinaya was said to be very strict and even attracted monks from the Eastern Jin, who travelled the thousands of miles to request Vinaya texts from Buddhakṣema. Thus we can conclude that Kucha had become one of the most important Buddhist centres at that time. The earlier Buddhist caves at Qizil were also constructed during this period. In these caves many Sanskrit Buddhist manuscripts have been found dating from the second to fourth centuries. The majority of these belong to the Hīnayāna.

The most famous Kuchean monk was undoubtedly Kumārajīva (344-с. 413), whose dates are variously given. According to Sengzhao’s 僧肇 Jiumoluoshi fashi lei 鸠摩罗什法师诔[Memoir of the Dharma Master Kumārajīva]49, kumārajīva’s father, Kumārayāna, was an Indian. He resigned from the post of prime minister, became a monk, and then travelled across the Pamirs to Kucha where he was warmly welcomed by the reigning king. He was appointed to the position as court teacher, and eventually married the king’s sister, Jivā. When Kumārajīva was seven years old, he left home along with his mother and went to study the scriptures of the Abhidharma with Buddhakṣema. At the age of nine, Kumārajīva travelled with his mother across the Indus River to Kashmir, and further to Yuezhi (Gandhārā?), Kashgar and other places before they arrived in Yarkand. The Buddhist sūtras Kumārajīva studied prior to his twelfth year were Hīnayāna texts, especially those of the Sarvāstivāda School, which was popular in Kashmir. However, after he met the prince Sūryasoma of Yarkand in Kashgar, he turned his interest to the Mahāyāna. In addition to Hīnayāna and Mahāyāna Buddhism, Kumārajīva also studied the four Vedas and the pañсavidyā. After he returned to Kucha via Aksu, he became a bhiksu connected to the royal palace until he reached the age of twenty.

In the course of time Kumārajīva’s reputation reached China, where Daoan suggested in a letter to FuJian 苻坚, the ruler of the Former Qin, that Kumārajīva be invited to China. In AD 385 Fu Jian sent some troops under general Lü Guang 吕光 to Kucha and forcefully brought Kumārajīva with them back to Liangzhou. As it happened Fu Jian was assassinated soon after, and Lü Guang established his own regime in the Liangzhou area, the Northern Liang 北凉, where Kumārajīva stayed for more than ten years. In AD 401 Liangzhou fell to Yao Xing 姚兴, the founder of the state of Later Qin 后秦, who invited Kumārajīva to Chang’an and gave him the title of “national preceptor” 国师. After that time, Kumārajīva began to translate sutras with the assistance of hundreds of monks.

Among the hundreds of rolls of Buddhist texts translated by Kumārajīva in Chang’an were the Mahāprajñāpāramitā sūtra, the Saddharmapuṇḍariīka sūtra, the Vimalakīrtinirdeśa sūtra, the Amitābha sūtra, the Vajracchedikā-prajñāpāramitā sūtra, etc. Most of them were Mahāyāna scriptures and re-translations. Kumārajīva also introduced the Mādhyamika school of Indian Buddhism systematically to China and translated representative works of this school including the Mādhyamika śāstra, the Śata śāstra, the Dvādaśanikāya śāstra, the Mahāprajñāpāramitā śāstra, and Satyasiddhi śāstra. Kumārajīva started a new epoch in the history of Buddhist translation in China because he was successful in both correctly rendering the original meaning and expressing it in elegant Chinese. That is the reason why Sengyou僧祐(445-518), in the first chapter of Chu sanzang ji ji makes a distinction between Kumārajīva’s “new” translations and the “old” ones made by all his predecessors.

As a master of Buddhist translation, Kumarājīva authored only a few works himself, including the Shixiang lun 实相论 [Treatise on the Marks of Reality]. This work, which is said to have systematically expressed his philosophy, has unfortunately long been lost. His correspondence with Huiyuan 慧远 (344-416) was collected by later scholars and preserved in a book titled Dasheng dayi zhang 大乘大义章[Essays on the Essence of Mahāyāna] 50 in three chapters. Most recently, an ancient manuscript of Kumarājīva’s Dasheng Pusa rudao sanzhong guan大乘菩萨入道三种观[Three Contemplations of the Enlightened Mahāyāna Bodhisattva], has been found in Nagoya, Japan. 51 Its authenticity, how ever, needs further examination.

Concluding Remarks

Until the fifth and sixth centuries, Buddhism was still flourishing in Kucha. It was during this period that most of the Kuchean votive caves were built. Many Buddhist scriptures in Tokharian В (Kuchean) as well as temple registers and accounts of begging for alms dating from this period have been discovered. As seen in the wall-paintings in the caves as well as in the excavated scriptures, Hīnayāna Buddhism was still dominant in that area. During the AD 720s, Xuanzang passed through Kucha en route from China to India. In his Da Tang xiyu ji 大唐西域记 [Record of the Western Regions] he reported that there were more than one hundred Buddhist temples and no less than five thousand Hīnayāna monks and nuns. He also visited the two Zhaohuli Temples in the east and west, namely the great Queli Temple mentioned above. From the mid-seventh to the late eighth century, many Chinese people migrated to Kucha. Because of the cultural exchanges between the Chinese and Kucheans, some Buddhist caves mixed the art styles of both. From the second half of the ninth century, the Uighurs gradually replaced the Tibetans as the controllers of Kucha. The Uighurs also converted to Buddhism and tried hard to resist the eastward spread of Islam. The Turks had long since entered Kucha. Gradually they became dominant in the local population during later periods and eventually assimilated the native Kuchean population, while the Kuchean language was eventually replaced by Uighur. By the thirteenth century, the Kuchean people had converted to Islam. The Buddhist culture of the region as well as the Tokharian-speaking Kucheans themselves gradually disappeared from Central Asia.

However, the extinct Tokharians and their relation with Buddhism have been discovered by modern archaeology. All the Tokharian documents have been written in a form of slanted Brāhmī, which is referred to as Northern Turkestan Brāhmī by L. Sander. The Buddhist literature written in ancient Kuchean and Agnean consists mainly of such works as the Udānavarga and its commentary the Udānālaṃkāra52, the Prātimokṣa53, the Karmarvācanā, Karmavibhaṅga, the Pratītyasamutpāda, the Abhidharmakośa, the Catuṣparisat sūtra, the story of Nanda and his wife Sundarī, Mātṛceṭa’s Buddhastotra, etc. Also found were the Puṇyavanta-jātaka, a variety of avadāna stories taken from the Araṇemi Jātaka and so forth. Most of these stories are also found in the Avadānaśataka, the Divyāvadāna, the Jātakamālā and in the Avadānakalpalatā. In the early twentieth century, the German expedition led by Grünwedel and Von le Coq found some fragments of the Maitreyasamiti at Šoršuq near Karashahr (Yanqi). In the winter of 1974, a further forty-four sheets, altogether eighty-eight pages of the same work, were found in an ash pit near the north temple at the Siksim site, also in the vicinity of Karashar. 54 Other Buddhist texts related to the Maitreya cult included the Maitreyāvadānavyākarana, whose contents are in large part the same as those of Maitreyasamiti, but also have some significant differences. In addition to the above findings, there are also manuscripts and cave inscriptions related to Buddhism. Besides Buddhist literature, there are medical, legal, economic and Manichaean documents. 55

中文提要

吐火罗人与佛教

在佛教由印度传入中国的过程中，吐火罗人作为居间者，起了十分重要的作用。本文首先利用各国学者在历史、考古、语言等领域的新的研究成果，论述了吐火罗人与月氏的关系，肯定大月氏是吐火罗人建立的国家，进而主要依据汉文资料，详细说明月氏和贵霜在佛教传播中所扮演的角色，列举了出自月氏的译经者对佛经翻译事业所做的贡献。

焉耆、龟兹是吐火罗人在今新疆绿洲建立的国家，佛教兴盛，在西域佛教史上也占有重要地位。如佛典翻译大师鸠摩罗什，父籍天竺，生于龟兹，母为龟兹王妹，初学小乘，后习大乘中观学派。罗什所译经论，译文流畅，多具文采，对后世影响很大。本文对此也做了较详细的考述。

19世纪末至20世纪，甲、乙两种吐火罗语佛典和其他文书在新疆大量出土，这是历史比较语言学和宗教史研究上的大事。本文介绍了各国学者对吐火罗语文献的整理、刊布和研究状况，并在注释中列举了主要文献，供读者参考。其中特别引人注目的，是近年中国学者季羡林对甲种吐火罗语残卷《弥勒会见记》的考释和研究。

编补

关于本文论述的问题，请参看Mariko Namba Walter, Tokharian Buddhism in Kucha: Buddhism of Indo-European Centum Speakers in Chinese Turkestan before the 10th Century C. E., Sino-Platonic Papers, 85, 1998,30pp。

（原载Studies in Central & East Asian Religions, Vol.9, 1996)