Assimilation and Transformation of Esoteric Buddhism in TIbet and China
by Carmen Meinert
Historical Background Understanding the formation of Buddhism during a time that Tibetan historiography narrates as the ’time of fragmentation‘ (Tib. sil bu’i dus), it is worthwhile to also look at the fringes of the Tibetan cultural sphere and at comparable developments in neighbouring cultures. As a Sinologist and Tibetologist, I would like to offer perspectives from the Central Asian oasis of Dunhuang in particular and on a comparison with similar
developments in Song China around the turn of the first millennium with regard to assimilation strategies of precarious esoteric rituals advocating violence in the Tibetan and Chinese context. However, before I come to my actual topic let me briefly recall the broader political context in Inner Asia in rather tumultuous times after the demise of the Tibetan Empire. This is helpful, or even crucial, in order to understand the ’time of fragmentation‘ in Tibet, with its impact on cultural and religious life, not as a single phenomena but within the wider frame of the destabilisation and reorganisation of large parts of Inner and East Asia as well. Within less than 60 years the three major empires, namely the Tibetan Empire, the Uighur Khaganate (840), and the Chinese Tang Dynasty (906) collapsed, left a power vacuum, and triggered numerous migrations.1 One of the immediate results, which affected all
neighbouring peoples as well, was the disruption of an interdependent network of long-distance trade and transmission of Buddhist doctrines along the main centres of the Silk Roads in Eastern Turkestan, at least up to the early/middle 10th century. Control over territories and particularly over the main trade routes was driven by economic interests to such an extent that even travelling monks and texts became part of tribute. This point will be exampled further down. As the Tibetans withdrew from their conquered areas in the Hexi corridor and Eastern Turkestan in the two decades following the collapse of the Tibetan
1 The various migrations triggered by these events affected large areas of Central Eurasia. They are well summarised in Elverskog 2010: 120-123.
Empire, some of the trading centres became independent: In Dunhuang, local Han Chinese rulers, the Zhang clan, established the Guiyijun regime (ca. 848890ies), while Khotan was ruled by an unbroken Khotanese royal lineage between 851 and 1006, when it was conquered by the Qarakhanids after more than twenty years of warfare. Uighur immigrants that had fled the Kyrgyz invasion (in 840) eventually established Buddhist kingdoms along the northern part of the Silk Roads: the Qočo Uighurs (866-1209) in Turfan, Bešbaliq (Beiting) and Kuča; and, in present-day Gansu province, the Yugour or Ganzhou Uighurs (866-
1028). The latter extended their control even over Dunhuang since the 890s as well, a fact that is mirrored in marriage alliances between Ganzhou Uighurs and members of the Cao family, the ruling clan of Dunhuang since the 920s. Tibetans that had settled in Eastern Turkestan during the years of Tibetan dominion fled south-eastward to the Kokonor region2 and eventually established the Tsongkha Kingdom—with Jiaosiluo (9971065) as the major figure and Tibetan ruler who was able to play a role in the long struggle among Uighurs, Tanguts, and the Song court in the first half of the 11th century.3 As for
the Chinese Empire, it took half a century to establish a unified empire again in 960, the Song court, which however, continued a state of warfare along its northern borders with the Tanguts, Khitan, and eventually the Jurchen, far into the 12th century. The Jurchen eventually conquered and ruled the northern parts of China (1115-1234). Thus, with the collapse the three major empires, the Tibetan, the Uighur, and Tang China, most of Inner and East Asia entered into at least two centuries characterised by uncertainties, shifting alliances, and numerous hostilities. However surprisingly, particularly the second part of the tenth century seems to have been fruitful times for religious exchanges again—despite the difficulty of travelling along the main
trading routes. This becomes apparent when looking at the corpus of manuscripts from the Mogao caves in Dunhuang with a greater part of Tibetan, Chinese, Khotanese and other texts dating from exactly around this times—as recent palaeographic research of Sam van Schaik, Tom Davis, and Jacob Dalton has shown on the basis of the analysis of some Tibetan texts.
4 This brings me closer to my actual topic, namely to questions of how violence in an Esoteric Buddhist ritual context was transmitted at these very tumultuous times I have just sketched here in the Tibetan and Chinese cultural sphere. Since the texts presented in this article were translated at around the
2 At the time the population of Liangzhou, the centre of the Hexi corridor, was largely Tibetan. Cf. Iwasaki 2003: 106. 3 Petech 1983: 177; Horlemann 2003. 4 Dalton; Davis; van Schaik 2007: 18; cf. also Rong 1999-2000: 247–75 and Rong 2002: 8– 28, especially pp. 18–19.
same time, it is worthwhile to compare the different contexts that allowed for the translation work. Such a comparison might even help to fill in some of the missing gaps we have with regard to the ’time of fragmentation‘ in Tibetan historiography. Within the ritual manuals of Tantric or Esoteric Buddhism as it was transmitted from the 7th and 8th century onward from India to Tibet and China, we find so called ābhicāraka rites. These are rites performed to
avert and destroy hostile forces and obstacles on the path of liberation. It is important to notice that such rites, which demand the performance of wrathful activities (Tib. drag po ’phrin las, Chin. xiangfuye 降伏业), originally had not only an apotropaeic function—that is, to avert harm— but a soteriological one: they also aim at achieving salvation, the very attainment of Buddhahood. As such, wrathful activity is only one skillful means among
the four enlightened activities, of which the other three are the activities of pacifying (Tib. zhi ba’i ’phrin las, Chin. xizai ye息灭业), enriching (Tib. rgyas pa’i ’phrin las, Chin. zengyi ye 增益业), and magnetising (Tib. dbang gi ’phrin las, Chin. huairou ye 怀柔业). From the point of view of the history of religion, the development of Tantric Buddhism in Tibet may be seen as an unbroken success story of more then 1300 years. In the Chinese cultural setting, however, Esoteric Buddhism (Chin. mijiao 密教) as a school in its own right was a rather short-lived phenomenon. Although it basically flourished
in the Tang Dynasty (in the 8th and 9th centuries), some elements, nonetheless, found their way into society far beyond this time.5 In order to analyse the different modes of adaptation of violence in the ritual context of Esoteric Buddhism in two different cultural settings, namely in the Tibetan and the Chinese around the turn of the first millennium, the following questions will be dealt with: What was transmitted and assimilated and what not? How was it transmitted? And in the case of China, which transformations are traceable of those elements of the teachings that were not transmitted? In a nutshell, my concern is threefold: (1) To introduce the history of the transmission of two central and comparable cycles of 18 esoteric texts, each of which include ābhicāraka rites, in order to demonstrate the beginning stages of
5 The transmissions and developments of Esoteric/Tantric Buddhism in China were very complex, not only with regard to negotiating processes between the religious and political fields, but also with regard to intra-religious contacts. This complexity is dealt with at length for the first time in the recently published handbook on Esoteric Buddhism in East Asia (Orzech; Sørensen; Payne 2011). Moreover, in recent years Shen Weirong has shown the countinuity of transmission of Tantric materials to China through the Tanguts and Mongols up to the Manchu Qing Dynasty; cf. Shen 2011, Shen 2010.
assimilation and segregation or delimitation of Esoteric teachings in Tibet and China respectively: The Māyājāla cycle (Tib. sGyu ’phrul dra ba) in the Tibetan and the Vajraśekhara cycle (Chin. Jingang ding jing 金刚顶) in the Chinese tradition; (2) To provide examples taken from the Guhyasamājatantra (Tib. gSang ba’i snying po’i mdo, Chin. Mimi dajiaowang jing 祕密大教王經), a text which is shared in both cycles. Here, I will contextualise the use of violence, particularly with regard to a ritual that became known in the Tibetan tradition as ’liberation through killing‘ (Tib. sgrol ba); And, (3) to summarise the effects that the different assimilation strategies had in Tibet and China with regard to the handling of Esoteric knowledge.
I. History of Transmission of Two Esoteric Text Cycles: Vajraśekhara in China and Māyājāla in Tibet The ’latest ritual techniques‘ of Esoteric Buddhism were transmitted from India to Tibet and China particularly in the 10th century. Here, the 18 Tantras of the Māyājāla and the Vajraśehkhara cycle seem to build a bridge in the transmission process to such different cultural settings as Tibet and China.6 These cycles probably formed the earliest canon of Esoteric Buddhist literature that as such was widely known in Inner and East Asia (maybe even from around the 7th/8th centuries), although some of the texts included varied in the two cycles.7 One of the texts indisputably transmitted in both cycles is the Guhyasamājatantra, the Tantra of the Secret
Gathering. On the basis of this text (and some related materials) I will provide examples, against the background of the Tibetan and Chinese cultural settings, of the assimilation and separation—or delimitation—respectively, of ābhicāraka rites in both cycles—rites to avert and destroy hostile forces. In the rNying ma School, the old or early transmission of Tantric Buddhism to Tibet, we find a complete collection of 18 texts in the so-called Māyājāla cycle that have been transmitted until the present time in the rNying ma rgyud ’bum, the Tantra Collection of the Old School. However, as we know texts included in this collection became the target of criticism during the 11th century, when the New Schools, during the so-called second phase of translation of Buddhist scriptures in Tibet, condemned those texts due to their partly questionable contents. The first Tibetan translation of the Guhyasamājatantra nonetheless had a huge impact, especially with regard to the volumes of commentarial literature it eventually generated. The earliest Tibetan translation of the Guhyasamājatantra
6 Kenneth Eastman (1981, 1983) first brought attention to the similarities between these two cycles in the Chinese and Tibetan tradition. 7 For a list of all text included in both cycles, see Eastman 1981.
now available to us was prepared probably in the second half of the 10th century in the Inner Asian oasis of Dunhuang. It is transmitted in two copies from the collection of Dunhuang manuscripts, IOL Tib J 438 and PT 5—the former a very neat translation in seventeen chapters,8 the latter a summary of the pith of the text in just 129 lines. Since the Dunhuang manuscripts were sealed for almost a thousand years, until the year 1900, they were spared the censorship or corrections of later generations. With regard to China, the Vajraśehkara cycle was transmitted first already during the Tang Dynasty in the middle of the 8th century by the Indian monks Vajrabodhi (Jingang zhi 金刚智, 661–732) and Amoghavajra (Bukong 不空, 705–774). The arrival of these Indian monks in China marked the beginning of what later Chinese historiography labelled as the relatively short-lived Esoteric School (Chin. mijiao 密教). In the Chinese Buddhist canon, it is said that when the Indian monks travelled to China by boat, the Sanskrit manuscripts of the 18 Tantras of the Vajraśehkhara cycle had to be thrown overboard to avoid the sinking of the ship. However, as we known from contemporary research, the situation was probably quiet different, as can be seen in this quote from Kenneth Eastman:
It is probable that the winds about to sink Vajrabodhi’s boat were sociopolitical winds blowing from insult and the perception of impropriety, and that the storm was quelled only after the canon of the complete Vajraśekhara collection and the attempts to translate it were, so to speak, thrown overboard.9
Thus, it were the stormy socio-political winds that prevented a full translation of all texts at the time. Therefore, the only knowledge of these 18 texts in 8th century China is found in a short summary in the Chinese Buddhist canon, in a text called Jingang ding jing yujia shibahui zhigui, or short Shibahui zhigui 十八会指归 [Indications of the Goals of the Eighteen Assemblies].10 The alreadymentioned Guhyasamājatantra is summarised here in merely nine lines. Therefore, the earliest and only translation of the Guhyasamājatantra included in the Chinese Buddhist canon dates from the Northern Song Dynasty, namely from 1002.11 It covers eighteen chapters and was prepared by the Indian monkscholar Dānapāla (Shihu 施护, ? –1018) on the order of Taizong, the second
8 Parts of IOL Tib J 438, that are relevant for the phur ba rite, have been translated in Cantwell; Mayer 2008: 166–180. 9 Eastman 1981: 32. 10 T. 18, no. 869, 287a.28–b.7. For a translation of this text, see Giebel 1995: 107–201. 11 T. 18, no. 885.
emperor of the Song Dynasty, who also aimed at a revival of Esoteric Buddhism. Here we have an interesting parallel to the Tibetan situation, when Esoteric Buddhist texts were translated a second time! Although this Chinese version is not a literal word-for-word translation, it is a very valuable source for my study concerning the treatment of the notion of violence in Esoteric Buddhism in China. It is important to mention that this translation probably had very little impact on the development of the Esoteric tradition in China. No commentarial literature at all has developed around it.
II. Use of Violence in the Ritual Context in the Guhyasamājatantra The Guhyasamājatantra is a rich source for my study because of its controversial nature. Although the goal of the text is the attainment of Buddhahood within the present lifetime, its explicit accounts of sex and violent acts again and again aroused strong criticism. The text finds original, pure nature in those very deeds that are generally rejected from an ethical viewpoint. In other words, even violence is used in the ritual context for soteriological means, namely for the attainment of awakening. The nature of meditation practice and the true meaning of the Dharma according to the Guhyasamājatantra are vividly described in the opening section of chapter five. Unless otherwise indicated, my English translations are based on the Tibetan Dunhuang manuscript IOL Tib. J 438, as it is closer to the Sanskrit critical edition (published by Yukei Matsunaga in 1978) than the Chinese version. In the relevant passage we find the Buddhist teachings turned completely upside down. Let me quote the essential passage:
The [three] types [of poison] passion, hatred and ignorance, arise from non-conceptuality. [Those who practice this] will accomplish the best, the highest of the supreme vehicle. Those born of low caste or who do despised work, those who are intent upon killing, [if they practice this] highest vehicle, they will accomplish this supreme vehicle. Sentient beings who commit sins with immediate retribution, and who practice [this teaching] will accomplish it. [However,] those who denigrate the master in their hearts, even if they practice [this teaching], will not accomplish it. [But] those who kill, those who are fond of lying, those who are attached to the wealth of others, those who will continually engage in sexual congress, those who take excrements and urine as food, these will become recipients who practice these [teachings].
12 12 IOL Tib J 438, fol. 12b.3–13a.1: [12b.3] /rnam par myi rtog las ’byung ba/ /’dod chags zhe sdang gti mug rigs/ / [12b.4] leg pa’i mchog ni bla myed pa/ /mchog tu gyurd pa grub par byed/ /rigs ngan smye sha can la stsogs/ /gsod pa don du gnyer pa rnams/ /theg pa bla na myed pa yi/ /theg mchog ’di la grub par ’gyur/ / [12b.5] mtshams myed byed pa la stsogs pa’i/
To sum up, those forms of conduct that are usually regarded as unethical become the prerequisites for practicing the teachings of the Guhyasamājatantra: Those who are intent upon killing, will accomplish this supreme vehicle! How is that possible? The key to understanding this utterly radical viewpoint is the very notion of ’non-conceptuality‘ (Tib. rnam par mi rtog pa, Chin. wu fenbie 无分别). How should we view the term ’nonconceptuality‘? Let me refer to a definition in another Chinese Dunhuang manuscript that dates to around the same time as the Tibetan manuscript of the Guhyasamājatantra, namely the Āryāvikalpapraveśanāmadhāraṇī [[[Sublime]] dhāraṇī of Entering into Non-Conceptuality]. Let me quote the gist of it:
[…], when the bodhisattva mahāsattva has completely given up all aspects of conceptual signs through not taking them to mind, he vigorously exerts himself in the basic space of non-conceptuality. […]. [Then he] is able at all times and in all ways to work for the welfare of all sentient beings and is also able to continue spontaneously and without interruption the deeds of a Buddha.13
Understanding the notion of non-conceptuality in this way means that, whoever completely transcends conceptuality, rests in the basic space of nonconceptuality. Now, according to the above-quoted passage from chapter five of the Guhyasamājatantra, such an individual will perceive that even the three root poisons of passion, hatred and ignorance arise from non-conceptuality. Non-conceptuality is obviously the main reference point, and is, so to speak, a vessel within which even outrageous conduct such as killing is transformed into the path of awakening. When we compare the Tibetan passage with the equivalent Chinese passage prepared by Dānapāla in Song China, some differences are noticeable. The Chinese translation reads:
/sems can dag gis bsgrubs na ’grub/ /snying nas slob dpon na smos pa dag/ /bsgrubs kyang grub par yongs myi ’gyur/ /srog gcod pa’i sems can dang/ /brdzun du smra la dad pa dang/ [13a.1] gzhan gyi nor la chags pa dang/ /rtag du ’dod pa spyod pa dang/ /bshang gci zas sa za ba dang/ /de dag bsgrub pa’i snong du ’gyur/. See also the Chinese equivalent in T. 18, no. 885, 473c.29-474a.15: 諸大士當知彼一切法離諸疑惑。真實出生若貪若瞋。及癡法 等此三平等。了此法性。是即無上大菩提性。The Sanskrit equivalent is in Matsunaga 1978: 15, verse 1-6. 13 For the Chinese Dunhuang manuscript cf. Jiang 23: line 47-57; and for the Tibetan text cf. “’Phags pa rnam par mi rtog par ’jug pa zhes bya ba’i gzungs”. TT. 32, no. 810, 3b.1-8.
Dharmas such as passion, hatred and ignorance that arise form ultimate reality are all equal. [...] Those who constantly let arise the mind of killing and harming sentient beings […] will obtain accomplishment, […].14
In comparison with the Tibetan translation two points are noteworthy in the Chinese translation: Firstly, here we have one of the very few passages in the Chinese translation that does explicitly use the term ’killing‘ (Chin. sha 杀) as a valid means to accomplishment.15 As we shall see later, the explicit use of this term is usually avoided in the Chinese text. Secondly, the translator Dānapāla does not use the standard term ’nonconceptuality‘, in Chinese wu
fenbie 无分别, to translate the Sanskrit term nirvikalpa, but instead uses the term ultimate reality (Chin. zhenshi 真实). Dānapāla was certainly aware of the term ’non-conceptuality‘, as he also prepared a Chinese version of the just-quoted Āryāvikalpapraveśanāmadhāraṇī. There he uses the standard Chinese translation wu fenbie for Sanskrit avikalpa.16 From one point of view, namely the traditional stance of a Buddhist doxographer, an argument can be made that ’non-conceptuality‘ and ’ultimate reality‘ have the same fundamental meaning. But in our context, namely how cultural forces have shaped the transmission of these rites, this explanation is inappropriate. For us this passage is highly illuminating. I would put forth the working hypothesis that
the Āryāvikalpapraveśanāmadhāraṇī played a crucial role in the transmission of various Buddhist traditions in Inner Asia in the 9th and 10th centuries. Among the corpus of Dunhuang manuscripts are one complete Chinese manuscript (Jiang 23) and two Tibetan fragments of this text (IOL Tib J 51 and 52). Many Tibetan Dunhuang manuscripts, whether they set forth Chinese Chan Buddhism, Tibetan rDzogs chen or, in the case of the Guhyasamājatantra, Mahāyoga teachings, center on the theme of ’nonconceptuality‘. Moreover, in these traditions even the legitimisation of a violent action such as killing is grounded in non-conceptuality and takes nonconceptuality as its main reference point. In this respect, it is remarkable that it was none other than Dānapāla who in Song China was responsible for the translation of both, the Guhyasamājatantra and the Āryāvikalpapraveśanāmadhāraṇī. We may assume that he was well aware of the Buddhist discourses in Dunhuang. In fact, he was detained for months by the rulers of
14 T. 18, no. 885, 473c.29–474a.5: 真實出生若貪若瞋。及癡法等此三平等。 [...]常起殺 害諸眾生心。[...]得成就。 15 The term sha 杀 appears only three times in the Chinese translation (T. 885, 474a.4, 474a.8, 498c.13). 16 Cf. T. 15, no. 654.
Dunhuang on his way from India to China in the year 979!17 Nonetheless, neither text is a literal word-for-word translation, and each obviously has been prepared under the strong censorship of Song China. I will come back to this issue further below. Let us first return to the Guhyasamājatantra, where it is said that the killing of a victim leads to spiritual accomplishment for the slayer. The slain faces an equally unexpected destiny, namely, he is reborn as a bodhisattva in the pure realm of Buddha Akṣobhya, as we can see in this passage from chapter nine. The Tibetan text reads:
And the Chinese translation has:
The Chinese translation is very cryptic here, and seems to distort the original sense; the term killing is omitted. Even more interesting is a passage in chapter fourteen that specifically describes the killing rite mentioned above and known in the Tibetan tradition as sgrol ba, ’liberation through killing‘. This is a crucial passage that illustrates how, in dependence upon cultural sensibilities, the meaning shifts as it moves from one cultural context into another. Let me first provide my translation of the Tibetan before comparing the different versions.
One shall especially meditate on killing the body, speech and mind of all sentient beings in the ten directions. Towards enemies [perform killing] without hesitation. [The Tantric practitioner] wears clothes that are soaked in the moisture of blood and [[[bodily]]] secretions or in the moisture of excrement and urine and stamps with the foot on the
17 Sen 2003: 121; Song huiyao jigao 200 (daoshi 2): 7892a. 18 IOL Tib J 438, fol. 22b.1–2: /rdo rje gsang ba ’di lta bus/ /sems can thams cad bsad na ni/ /myi bskyod sangs rgyas zhing dag du/ /rgyal ba’i sras rnams skye bar ’gyur/ /’di ni zhe sdang gi rigs kyi dam [22b.2] tshig de kho na’o/. 19 T. 18. no. 885, 477b.12–15: 眾祕密金剛 能破壞一切 阿閦金剛生 住諸佛境界。
This passage hints at the various parts of the rite, which are:
(4) the killed one is transferred to a pure realm (which is not explicitly mentioned in this passage). It is not clear from this passage whether the term liṅgaṃ refers only to an effigy of the slain that should be vicariously destroyed, and thereby would
bring about the death of the victim and his liberation, or whether it actually refers to the killed person himself. Judging from the 11th century critique on practices such as ’liberation through killing‘ (sgrol ba) by Lha bLa ma Ye shes ’od,21 it is not unlikely that there might have been strong reasons for
banishing a rite, that, so to speak, got out of control and was actually acted out as a human sacrifice. There are a few other Tibetan Dunhuang manuscripts related to the Guhyasamājatantra that give even more detailed accounts of this rite, such as a passage in PT 840. The description here is dramatic and points to the connection between the heart and conceptual thought that, again, should be eradicated.
[After killing the evil one], collect the mind and then transfer it. The heart needs to be fed [into the mouth of the deity] because the heart is the abode of conceptual thought. In order to purify [the thought] into the state of wrath, it must be transformed into a goddess or into light. Then recite the mantra A TSITA KHA RAM GI and the body, speech and mind of the Great Glorious Ones, the male and female Wrathful Ones, the Powerful Lady and the accomplished brothers and sisters dissolve. Then initiation is attained.22
20 IOL Tib J 438, fol. 53b.1–2: /phyogs bcu’i sems can thams cad kyi/ /lus dang ngag sems gsad par ni/ /khyad bar dag du bsgom bar bya/ /dgra rnams la ni gdon myi za/ /khrag dang chu’i gsher pa ’am/ [53b.2] bshang gcis gsher par byas pa’i gos/ /bgos nas ling ga rkang pas mnan/ /khro bo’i rgyal pos (=por)
sbyar bar bya/ /brgya rtsa brgyad ni yongs rdzogs na/ /sangs rgyas dag kyang nges par ’jig/. 21 Karmay 1980: 150–162. 22 PT 840, line 47–49: sems  bsdu zhing spar/ /tsi ta bstabs pa ni rnam par rtog pa thams cas kyi rten yin bas/ /khro bo’i ngang du sbyang ba  ’i phyir/ /lha mo ’am ’od du bsgyur te/ /A TSI TA KHA RAM GI zhes brjod de/ /dpal chen rnams dang khro bo  dang khro mo rnams dang/ /dbang phyug ma dang grub pa lcam dral gyi sku gsung thugs thim nas/ /dbang thob par bya//. See also the recent translation of PT 840 by Jacob Dalton (2011: 212).
But let us turn once again to the Chinese translation of the Guhyasmājatantra. In the equivalent passage to the quotation from chapter fourteen, it is striking that all ’dangerous‘ parts are carefully and completely cut out or turned around with a different meaning. The Chinese translation reads as follows:
Or again, [he?] expounds: [in order that (?)] all Wrathful Kings subdue hatred and evil and perform all auspicious activities, recite [their] mantras. Then, those who are to be subjugated are all those extremely evil sentient beings who denigrate the spiritual master and who denigrate the Mahāyāna teachings, do various evil deeds, and follow erroneous knowledge. They destroy the seed of Buddhahood and cannot [even] with diligence seek the path to awakening. Moreover, toward the activities of body, speech and mind of all sentient beings in the ten directions let arise such [[[deeds]]] as destructive thoughts and evilmindedness. Then with this teaching you will be able to subjugate [them]. If you perform the teachings of subjugation, at each place (?) that you visualise the wrathful image, time and place [need to] correspond according to the teaching ceremony (?). Concerning [the visualisation] of the wrathful image, first recite his mantra 108 times. All evil ones will certainly be summoned and thoroughly caused to be subjugated.23
When the three versions in the Sanskrit, Tibetan, and Chinese are compared, two methods may be detected concerning how the meaning of Esoteric Buddhist texts in China was altered consciously and systematically during the translation process from Sanskrit into Chinese, namely, by means of omittance and by means of paraphrase. In comparing the three versions, the following four points should be emphasized: (1) The Chinese text omits the term killing, which leads to the distortion of the original sense already mentioned above;24 (2) The Sanskrit version talks about “killing the body, speech and mind of all Buddhas”, whereas the Tibetan and
23 T. 18, no. 885, 492b.15–22:復次宣說一切忿怒王調伏怨惡作諸吉祥事大明持誦行。 當所調伏者。即彼一切極惡眾生。所謂謗阿闍梨及謗大乘。作諸魔事隨順邪明。 壞佛種性不能勤求佛菩提道。又於十方一切眾生身語心業。起破壞想生怨惡心如 是等輩。即以此法而可調伏。若作調伏法者。當於一處想忿怒。依法儀軌時處 相應。於忿怒前誦彼大明一百八遍。一切惡者決定句召。悉令調伏。 24 Cf. also the quoted passage of chapter nine of the Guhyasamājatantra on p. 349 above.
Chinese texts refer to sentient beings;25 (3) The next verse in the Sanskrit text, “who are evil-minded enemies”, is left out completely in the Chinese version and is replaced by a different passage. Here, subjugation is used instead of killing and, moreover, the Chinese translation refers to it as a visualisation process (Chin. xiang 想);26 And, (4) most importantly, the concrete instruction on the killing procedure is again omitted completely in the Chinese version.
III. Assimilation and Transformation of Esoteric Knowledge in China and Tibet In the two very different cultural settings of Tibet and China we can observe equally different assimilation strategies of Esoteric teachings with regard to precarious Esoteric teachings. The situation in Dunhuang, and eventually Tibet, appears more unambiguous and may be summarised on the basis of the little historical information we have. The situation in China, however, is very well documented so that we can trace back a rather complex socio-political context that influenced the process of the translation work and led to a certain
separation from and discontinuation of Esoteric teachings. I will outline both historical contexts. In Tibet the latest esoteric ritual techniques from India were completely transmitted, and all major Tantric cycles translated. There are even more Tibetan Dunhuang manuscripts—partly published in the research of Cathy Cantwell and Rob Mayer27— that give very elaborate descriptions of the killing ritual, including descriptions of the vows that need to be taken as well as the results that could be attained. However, as already briefly mentioned, these practices also met resistance in Tibet. In the 11th century, the Tibetan scholar Lha bLa ma Ye shes ’od wrote in an edict in Western Tibet that rites such as
25 For the Sanskrit text see Matsunaga 1978: 67: daśadiksarvabuddhānāṃ kāyavākcittaghātanam/. “One shall meditate with the correct procedure on killing the body, speech and mind of all Buddhas in the ten directions, […].” For the Tibetan text see IOL Tib J 438, fol. 53a.4: /phyogs bcu’i sems can thams cad kyi/ /lus dang ngag sems gsad par ni/ […]. “One shall especially meditate on killing the body, speech and mind of all sentient beings in the ten directions.”
For the Chinese text see T. 18, no. 885, 492b.15–16: 於十方一切眾生身語心業。“Toward the activities of body, speech and mind of all sentient beings in the ten directions […].” 26 T. 18, no. 885, 492b.16-17: 起破壞想生怨惡心如等輩。即以此法而可調伏。 若作調伏法者。當於一處想忿怒。依法儀軌時處相應。“[…] let arise such [[[deeds]]] as destructive thoughts and evil mindedness. Then with this teaching you will be able to subjugate [them]. If you perform the teachings of subjugation, at each place (?) that you visualise the wrathful image, time and place [need to] correspond according to the teaching ceremony.” 27 Cantwell; Mayer 2008; cf. also Dalton 2011.
’liberation through killing‘ (Tib. sgrol ba) should be prohibited.28 There is hardly any historical evidence to prove that such a rite became out of control and might have been even practiced as a human sacrifice. Maybe the comparison with the situation in China helps to understand the process that led to certain texts being eventually excluded from the canon in Tibet. Nonetheless, the rNying ma pa tradition integrated this rite ’liberation through killing‘ in the so-called ’ancillary rites‘ (Tib. smad las) of the Mahāyoga teachings, as Cathy Cantwell has already shown.29 The rite, in fact, not only functions as a remedy to avert harm, but particularly to achieve salvation; the community of practitioners destroys symbolically in themselves the
obstacles on the path of awakening. The teaching was sublimated, and so to speak internalised: it aims at the destruction of all ego-clinging within the practitioner itself. It is in this way that the transmission of wrathful activity and ritual violence survived in the Tibetan tradition up to the very present. The assimilation of even such contentious teachings was integrated into the frame of salvation, so to speak, through the strategy of ’trial and error‘. Why, on the other side, were the teachings of the Guhyasamājatantra translated in such a distorted way in Song China? We can find answers to this question when we look at the socio-political context at the turn of the first millennium, the time of the translation of the Guhyasamājatantra. In 980 the
translator Dānapāla (d. 1018) from Uḍḍiyāna was invited by emperor Taizong to the Song Chinese capital of Kaifeng to the newly established ’Institute for the Translation of the Buddhist Scriptures‘ (Chin. Yijing yuan 译院). He was one of about the first 80 Indian monks who arrived in China in the late 10th to the 12th centuries. We know that he travelled together with his Kaśmīri cousin Devaśāntika (Tianxizai 天息災, d. 1000) through Eastern Turkestan. Although travel seemed to have been possible again in 979, a Chinese historical source, Song huiyao jigao 宋会要辑稿 [Collection of Important Documents of
the Song Dynasty], relates that the two of them were nonetheless detained for months in Dunhuang by the local ruler, Cao Yanlu. Although further information about the reasons of their detention could not be found, we know from a number of Khotanese Dunhuang manuscripts that it was a common practice to regard travellers—and even monks—just like commodities, as a kind of tribute.30 Sometimes monks travelling with merchants, delegations or other lay people were allowed to continue their way after some time, yet often only when they had been stripped of any belongings they carried with them. We can only speculate how Dānapāla and Devaśāntika spent their days in
Dunhuang. It is not unlikely that their knowledge was used during translations in Dunhuang as well (by the way, this was not an uncommon practice in China: one of the greatest translators in China, the Kučean monk Kumārajīva (344413), made a virtue out of necessity when he was held captive at the court in Chang’an in the 4th century). In any event, Dānapāla and Devaśāntika managed to flee and arrived in the Chinese capital Kaifeng in 980. Well-trained and versed in several Indian and Inner Asian languages, Dānapāla became renowned as the most productive of the Indian translators in Song China.31 What may
have been the reason for his negligent imprecision in translating the Guhyasamājatantra? The unique feature of Song translation projects is that they seem to have been part of a politically motivated project of centralisation and cultural integration.32 In fact, the translation and printing of Buddhist scriptures was quite consciously not intended for the Buddhist community in China. Rather, such activities were used by the Song court as instruments of civil policy, internally to demonstrate its emphasis on literary learning as a modality for integrating the formerly divided country, and externally as a
tool in diplomatic relations with neighbouring countries along the northern border.33 With the rise of independent states of the powerful Khitan and Tangut tribes in the 10th and 11th centuries, establishing the Liao (916-1125) and Xixia dynasties (1038-1227) and endorsing Buddhism as a state religion, Buddhist monks and texts became increasingly entangled in foreign affairs. The Tanguts even ordered six copies of the Chinese Buddhist canon after its first printing in 983. Buddhism became an instrument of diplomacy. In fact, the translation institute in Kaifeng functioned under the bureau in charge of diplomatic affairs (Chin. honglu si 鴻臚寺)34 and Dānapāla at the time of the translation of the Guhyasamājatantra functioned as the Acting Chief Minister of State Ceremonies (Chin. shi hong lu qing
試鴻臚卿).35 It is thus clear that the translators were under orders from the censors of the Song court to produce texts that were ’tactful‘, helpful in diplomatic relations, and in no way harmed or contradicted the self-image of the Chinese court. And, any secret instructions they might contain needed to be avoided since they could—in case they were misused—destroy China itself. Thus these texts were definitely not meant to meet solely the interests and needs of the practising Buddhist community in Song China.
31 Sen 2003: 112, 120–22. 32 Loc. cit.: 114. 33 Loc. cit.: 133. 34 Loc. cit.: 112, 118. 35 Cf. T. 18, no. 885, 469c.22.
However, this is only one side of the coin that lead to the discontinuation of Esoteric Buddhist teachings. The other side of the coin might give an even more intriguing answer to the question of whether violence in a ritual esoteric context might have been processed in a different way. As a result of the neglect in Imperial translation projects, the centre of gravity of Chinese Esoteric Buddhism shifted from court to country. In fact, in 10th and 11th
centuries Song China there arose a competition about esoteric knowledge in popular beliefs among Buddhist, Daoist and shamanists alike. Some rites found their way into exorcism, which came to reflect the exuberance, unpredictability and variations of village culture.36 Tantric Buddhism was popularised and, for example, found its way into demonpossession cults, in which surrogates were used to perform exorcism. In such cults a spirit-medium would identify himself with a demon, similar to the Tantric practitioner who identifies himself with a deity.37 As such, we find in the Daoist compilation from the 14th
century, the Daofa huiyua 道法會元 [Compilation of Daoist Rites], exorcist ritual practices in the tradition of the Thunder Rites (Chin. leifa 雷法), as they were mainly practiced in South China during the Song Dynasty.38 With the tradition of the so-called Thunder Rites there was a remarkable turn noticeable in religious Daoism: lay practitioners appeared who roamed through the villages of South China and practiced exorcist rites. In their practices, they assimilated the main characteristics of Esoteric Buddhist rituals, such as the identification with a divinity through the process of visualisation in
order to attain the power of the deity! Those lay exorcists were called fashi 法师, here with the meaning ’master of ceremonies‘ —although they were neither fully ordained Daoist priests (Chin. daoshi 道师) nor spirit-mediums (Chin. wu 巫), as they were found in each village. However, it is not only the uneducated part of society that turns to those fashi, but also highly-educated literati and officers. And it is just this very intellectual elite,
magistrate-cum-exorcist, in the South of China that attempt to learn exorcist practices to gain control over regional cults in the villages. What we can see here is a play, or, to be more precise, a struggle for power between magistrate-cum-exorcist and spirit-medium in the villages. The real motivation, however, behind this ’ritual armament‘ is rather profane: to avert the magic powers of the spirit-medium in areas of other ethnicities, in
36 Davis 2001: 152. 37 Loc. cit.: 122. 38 Skar 2000: 420–423. After the first publication of this article in 2013, I have published another article exemplifying how Tantric Buddhism was fully appropriated into Daoist Thunder Rites practice during the Song Dynasty. Cf. Meinert 2013.
this particular case of the Yao people. Because they are regarded as a potential danger for the Chinese Empire! Thus, Chinese magistrates formed a fatal alliance with demonic forces in order to attain only worldly powers. One of the rites that was practiced is called the ’rite to decapitate the hun soul‘ (Chin. zhanhun fa 斩魂法).39 Here, the life force of the victim is gathered in a ’talisman to chase fake spirit-mediums‘ (Chin. zhui xiewu fu 追邪巫符)40—a representation that very much resembles the effigy of a liṅgaṃ, that is, effigies of victims that are used in the Tibetan killing rite sgrol ba. Those
fashi who in Song China practiced the Daoist ‘rite to decapitate the hun soul’ did aim to solely avert harm, yet they did not pursue any higher aims of salvation. On the basis of the materials presented, the following points with regard to assimilation strategies of Esoteric Buddhist knowledge in the Tibetan and Chinese cultural setting may be concluded. (1) As for the Tibetan context, we may summarise that the use of violence in a ritual context was
fully assimilated during the early period of translation. I would argue that the location and historical circumstances in Dunhuang in the second half of the 10th century were in favour of such translation projects, which were far away from any central power and thus far from strong censorship. I suggest that it is worth while to research further the sociopolitical contexts at the fringes of the Tibetan cultural sphere to better understand how Buddhism was shaped during the intermediate phase (bar dar). Although there might have been cases of misuse of these questionable teachings, including wrathful
activity, they were nonetheless integrated in the larger concept of salvation and, in fact, interpreted as another powerful and skilful means to deconstruct clinging to the illusory ego. Wrathful activity was thus a means for attaining salvation. The potential misuse of the ‘rite of liberation through killing’, however—it was banned by Ye shes ’od in the 11th century in Western Tibet—does not appear unlikely any more, if we compare the situation
with similar developments in China in the 11th century. Here the situation of misuse of exorcist rites is very well documented. (2) The situation in China for coping with Esoteric Buddhism was quite different. Here, the Chinese attempt to control the transmission of Esoteric Buddhist texts in the Song Dynasty from the very beginning led to a discontinuation of the teachings and brought the opposite effect: There was no controlled and authentic transmission of questionable esoteric rites that
39 The short title of the whole ritual text is Shoushe xiewu fa 收攝邪巫法 [[[Rites]] for Seizing Perverse Sorcerers] and the full title is Beiyin Fengdu Taixuan zhi mo heilü shoushe xiewu fa/ 北陰酆都太玄制魔黑律收攝邪巫法. It is found in Daofa huiyuan 道法會元 264. 40 This rite is also discussed by Judith Boltz (1993: 276–277), who also provides an image of the talisman.
Boltz, Judith Magee. 1993. “Not by the Seal of Office Alone: New Weapons inBattles with the Supernatural”. In: Ebrey, Patricia Buckley; Gregory, Peter N. (ed.). Religion and Society in T’ang and Sung China. Honolulu: University of Hawaii Press: 241-305. Cantwell, Cathy. 1997. “To Meditate upon Consciousness as vajra: Ritual ‘Killing and Liberation’ in the rNying-ma-pa Tradition”. In: Krasser, H. et al. (ed.). Tibetan Studies. Proceedings of the 7th Seminar of the International Association for Tibetan Studies, Graz 1995. Wien: Verlag der Österreichischen Akademie der Wissenschaften: vol. 1, 107-118. Cantwell, Cathy; Robert Mayer. 2008. Early Tibetan Documents on Phur pa fromDunhuang. Wien: Verlag der Österreichischen Akademie der Wissenschaften. Dalton, Jacob P. 2011. The Taming of the Demons: Violence and Liberation in Tibetan Buddhism. New Haven/London: Yale University Press. Dalton, Jacob; van Schaik, Sam. 2006. Tibetan Tantric Manuscripts from Dunhuang: A Descriptive Catalogue of the Stein Collection at the British Library. Leiden: Brill. Dalton, Jacob; Davis, Tom; van Schaik, Sam. 2007. “Beyond Anonymity: Paleographic Analyses of the Dunhuang Manuscripts”. Journal of the International Association of Tibetan Studies 3: 1–23. Available online at: http:// www.thlib.org/collections/ texts/jiats/#!jiats=/03/dalton/ (accessed February 25, 2012). “Daofa huiyuan 道法會元 264 [Compilation of Daoist Rites]”. In: Zhengtong Daozang 正統道 1220 [[[Daoist]] Canon of the [[[Ming]]] Zhengtong Era [Compiled in 1445]]. Zhengyi bu 正乙部. Available online at: http://www.ctcwri.idv.tw/ CTCWRI-MTS/CMT07正乙部/CH07XX/CH07029-3 道法會元(三)/CH0702 9-3XX/CH07029-3-264 卷二百六十四北陰酆都太玄制魔黑律收攝邪巫法. htm (accessed February 27, 2012). Davis, Edward L. 2001. Society and the Supernatural in Song China. Honolulu: University of Hawai’i Press. Eastman, Kenneth W. 1981. “The Eighteen Tantras of the Vajraśehkhara/Māyājāla”. Unpublished paper presented to the 26th International Conference of Orientalists in Japan, Tōkyo, May 8th 1981. ___1983. “Mahāyoga Texts at Tun-huang”. Bulletin of the Institute of Buddhist Cultural Studies 22: 42–60.
Ebrey, Patricia Buckley; Gregory, Peter N. (ed.). 1993. Religion and Society in T’ang and Sung China. Honolulu: University of Hawaii Press. Elverskog, Johan. 2010. Buddhism and Islam on the Silk Road. Philadelphia/Oxford: University of Pennsylvania Press. “Foshuo yiqie rulai jingang sanye zuishang mimi dajiaowang jing 佛說一切如來金剛三業最上祕密大教王經 [The Supreme Scripture, the King of Great Teachings Taught by the Buddha to all Tathāgatas in the Adamantine samādhi (= Guhya-samajātantra)]”. Translated by Dānapāla in 1002 (Shihu 施护). T. 18, no. 885. “Foshuo ru wu fenbie famen jing 佛說入無分別法門經 [[[Sūtra]] of Entering into the Dharma Gate of Non-Conceptuality Taught by the Buddha]”. Translated by Dānapāla. T. 15, no. 654, 805–806. Giebel, Rolf W. 1995. “The Chin-kang-ting ching yü-ch’ieh shi-pa-hui chih-kuei: An Annotated Translation”. Journal of Naritasan Institute for Buddhist Studies 18: 107–201. Hansen, Valerie. 2005 [published in 2010]. “The Tribute Trade with Khotan in Light of Materials Found in the Dunhuang Library Cave”. Bulletin of the Asia Institute 19: 37-46. Horlemann, Bianca. 2003. Aufstieg und Niedergang der Tsong-kha-Stammeskonföderation im 11./12. Jahrhundert an der Schnittstelle von Tibet, China und Zentralasien. Frankfurt: Peter Lang. IOL Tib J 51 and 52. rNam par mi rtog pa ’jug pa zhes bya ba’i gzungs [The dhāraṇī of Entering into
Non-Conceptuality]. Dunhuang manuscript fragments preserved in the British Library in London. IOL Tib J 438. A Dunhuang manuscript copy of the Guhyasamājatantra in seventeen chapters from the British Library in London, available at http://idp.bl.uk/database/ oo_scroll_h.a4d?uid=2102561696;recnum=5290;index=1 (accessed February 25, 2012). Iwasaki, Tsutomu. 2003. “The Tibetan Tribes of Ho-hsi and Buddhism during the Northern Sung Period”. In: McKay, Alex (ed.). History of Tibet. Richmond: Routledge Curzon: vol. 2, 106-122 (originally in: Acta Asiatica 64 (1993): 1734). Jiang 姜 23. Ru wu fenbie zongchi jing 入無分別总持經
[The dhāraṇī of Entering into Non-Conceptuality]. Dunhuang manuscript preserved in the Beijing NationalLibrary. “Jingang ding jing yujia shibahui zhigui 金刚顶瑜伽十八会指归 [Indications of the Goals of the Eighteen Assemblies of Yoga of the Vajraśehkhara]”. Compiled by Amoghavarja 不空. T. 18, no. 869.
Karmay, Samten G. 1980. “The Ordinance of Lha Bla-ma Ye-shes-’od”. In: M. Aris, Michael; Suu Kyi, Aung San (ed.). Tibetan Studies in Honour of Hugh Richardson. Warminster: Aris and Phillips Ltd.: 150–162. Kurz, Johannes L. 2003. Das Kompilationsprojekt Song Taizongs (reg. 976–997). Bern: Peter Lang. Martin Ahern, Emily. 1981. Chinese Ritual and Politics. Cambridge: Cambridge University Press. Matsunaga, Yukei. 1978. The Guhyasamāja Tantra. A New Critical Edition. Osaka: Toho Shuppan, Inc. Meinert, Carmen. 2013. "Buddhist Traces in Song Daoism: A Case from Thunder Rite (Leifa) Daoism". In: eJECAR
electronic Journal of East and Central Asian Religions 1, Henrik Sørensen (Hrsg.), Online-Publikation, 39-52. Doi: http://journals.ed.ac.uk/ejecar/article/view/735 Orzech, Charles D.; Sørensen, Henrik H.; Payne, Richard K. (ed.). 2011. Esoteric Buddhism and the Tantras in East Asia. Leiden/Bosten: Brill. Petech, Luciano. 1983. “Tibetan Relations with Sung China and with the Mongols”. In: Morris Rossabi (ed.). China Among Equals: the Middle Kingdom and its Neighbors, 10th-14th Centuries. Berkeley/Los Angeles: University of California Press: 174–203. “’Phags pa rnam par mi rtog par ’jug pa zhes bya ba’i gzungs [The Sublime dhāraṇī of Entering into Non-Conceptuality]”. Translated by Ye shes sde, Dānaśīla and dKa’ ba dpal brtsegs. TT. 32, no. 810, 3b.1–8. PT 5. A Dunhuang manuscript fragment of the Guhyasamājatantra in 129 lines from the Bibliothèque National de France in Paris. Rong, Xinjiang. (1999-2000). “The Nature of the Dunhuang Library Cave and the Reasons for Its Sealing.” Cahiers d’Extrême-Asie 11: 247–75. ___ 荣新江. (2002). “Zailun Dunhuang zangjing dong de baozang—Sanjie si yu cangjing dong 再论敦煌洞的宝—三界寺与洞 [Further Discussion of the Treasure in the Library
Cave of Dunhuang—Three Stages Monastery and the Library Cave]. In: Rong Xinjiang. Dunhuang xue xinlun 敦敦敦敦敦 [New Contributions to Dunhuang Studies]. Lanzhou: Gansu jiaoyu shuban she: 8–28. Sen, Tansen. 2003. Buddhism, Diplomacy, and Trade: The Realignment of Sino-Indian Relations, 600–1400. Honolulu: University of Hawai’i Press/ Association for Asian Studies. Shen, Weirong. 2010. “Reconstructing the History of Buddhism in Central Eurasia (11th-14th Centuries): an Interdisciplinary and Multilingual Approach to the Khara Khoto Texts”. In: Chayet, Anne; Scherrer-Schuab, Christina;
Robin Françosie; Achard, Jean-Luc (ed.). Edition, éditions: l’écrit au Tibet: évolution et devenir. Munich: Indus Verlag, 1-26. Shen, Weirong. 2011. “Tibetan Buddhism in Mongol-Yuan China (1206-1368)”. In: Orzech, Charles D.; Sørensen, Henrik H.; Payne, Richard K. (ed.) Esoteric Buddhism and the Tantras in East Asia. Leiden/Boston: Brill, 539-549. Skar, Lowell. 2000. “Ritual Movements, Deity Cults, and the Transformation of Daoism in Song and Yuan Times”. In: Livia Kohn (ed.). Daoism Handbook. Leiden: Brill, 413–463. Song huiyao jigao 宋会要辑稿
[Collection of Important Documents of the Song Dynasty]. Compiled by Xu Song 徐松 (1781-1848) et al. Beijing: Zhonghua shuju, 1987. Wayman, Alex. 1977. Yoga of the Guhyasamājatantra. The Arcane Lore of Forty Verses. A Buddhist Tantra Commentary. Delhi: Motilal Banarsidass. Worthy, Jr., Edmund-Henry. 1983. The Funding of Sung China, 950–1000: Integrative Changes in Military and Political Institutions, Ph.D. dissertation, UMI: Ann Arbor