Remarks On Trantristic Hermeneutics

Author(s): Ernst Steinkellner
Publisher: Ernst Steinkellner
Catalog link: https://jainqq.org/explore/269287/1

Page #1
REMARKS ON TANTRISTIC HERMENEUTICS The means which earlier Buddhist, especially mahayanistic exegesis has applied in its interpretation of the authoritative revelation (agamah, lun) have been explained by E. Lamotte. Here we find a unison-as characteristic for every living religion-of change or deepening of beliefs and development of exegetical methods which enable one to understand the resulting differences in accordance with tradition. mChan 'grel P PKr Religious contents, beliefs, are subject to history. They stagnate and die as soon as they become unresponsive to new needs and questions arising in the societies harbouring them, but they also lose their religious value I when they deviate from tradition by such changes. Hermeneutics2 works against such deviation. Its methods have been established in order to separate, via an act of interpretation, the various forms of religious contents PKrTip Prad Prad. abhis. prak. Prad, uddy. Prad. tika Man goal VJIST bSad sbyar = = = BY T ERNST STEINKELLNER (Vienna) 1= Pradipoddyolanabhisamdhi prakasika (Bhavyakirti). P 2658a,b. Pradipoddyotanodyotah (Sri Karunasripada). P 2655. Pradipoddyotanatika (Aryadova). P 2659. =dPal ysan ba 'dus pa'i bsad pa'i rgyud Ye ses rdo rje kun las blus pa'i rgya cher bsad pa, rGyud bsad thabs kyi man nag ysal bur bstan pa (Tson kha pa). P 6198. Vajrajnanasamuccatantram. P 84. - Abbreviations rGyud thams cad kyi rgyal po dpal gsan ba 'dus pa'i ryya cher bsad pa sgron ma gsal ba'i tshiy don ji bzin 'byed pa'i mchan gyi yan 'grel (Tson kha pa). (Title of Sendai-catalogue No. 5282) P 6166. Peking Edition. Ed. D. T. Suzuki, Tokyo-Kyoto 1955-1961. Pancakramah (Nagarjuna): Etudes et textes tantriques, Pancakrama. Ed. L. do la Vallee Poussin, Gand-Louvain 1896. Pancakramatippani (Parahitaraksita): cf. P'Kr. Pradipoddyotana (Candrakirti): cf. n. 6. sGron ma rab tu gsal bar byed pa'i bsad sbyar, mtha' drug rab tu ysal bar byed pa (Bu ston). Collected Works. Ed. Lokesh Chandra, Now Delhi 1967. Vol. Ta, 3. f. 1-271a. La critique d'interpretation dans le bouddhisme: Annuaire de l'Institut de philologie et d'Histoire Orientales et Slaves 9 (1949), pp. 341-361. L.e. the study of those methodological principles which rule over the interpretation and explanation of revelatory texts. 445

Page #2
within a tradition in such a way that the unity with revelation is preserved in each case and that freedom is created at the same time for a rational foundation of the differences. While the possibility of change in this way guarantees the survival of a religious tradition, it is hermeneutics which guarantees the continuity of this tradition as such. Hermeneutics in this tradition-maintaining function is to be found at its most valuable in Buddhism, at first in connection with the development of Mahayana-as the natural consequence of the notorious breaks of the new tradition with the older ones and, on the other hand, of the living vigour of the Buddhist communities.2a In none of the other great religions do we find the appearance of "new revelations" as natural as we do in Buddhism. The reasons for this are too manifold to be considered here. I only want to point out that it is always the hermeneutic categories of the interpreters which show the connection of the "new revelation" with the existing one and account for a meaning and purpose of the evident differences and thereby do away with the developmental ruptures in an ever new synthesis of the gaping differences. In the history of Buddhism there exist above all two great masses of "new revelation", the Mahayanasutras and the Tantras, which both have brought about the need to formulate corresponding hermeneutic categories. Or shall we say, which have arisen in dialectical unity with the rudimental concepts of new hermeneutic categories? In spite of the great historical import of hermeneutics especially in the case of these changes within the Buddhist tradition, this theme has been rather neglected, except for Lamotte's description of mahayanistic hermeneutics. A history of Buddhist hermeneutics, because of its traditionestablishing and tradition-maintaining function, is therefore an urgent desideratum, above all in the history of Buddhism. Compared with such an enormous task the philological problem I want to present here may seem trifling and to some even artificial and forced. I dare to present it, not because I can hope to elaborate on a set of interpretational rules generally applicable, but rather because I want to share it as a problem with others in the field, and because I think that each respective single text studied with this in mind may yield a richer understanding of certain tantric texts than we have usually been satisfied with up to now. 2 Rudiments of a hermeneutical theory are extant of course in older Buddhism, mainly provoked by the need to deal with the Pudgalavada. They are summarized e.g. in the Sutra-quotation on the "four points of reference" (pratisaranam) (cf. Mahavyut patti SS 74). 446

Page #3
When older hermeneutics prescribes that we have to view certain revelatory texts in their implicit meaning (neyurthah, dran don) and others in their explicit meaning (niturthuh, nes don), we have no difficulty in doing so, for we can clearly recognize the hermeneutic tools and can understand the texts subjected to them in their own context. The problem arises when we are summoned by the exegetes to understand one and the same text with different meaninys. This is the case e.g. with certain statements of evident madhyamaka-character in the Tantras. If a commentator holds the view that the text should be understood with a fourfold meaning, 4 we have to ask ourselves how far we can follow this invitation in a historical interpretation. It is not possible to follow these prescriptions of the tantristico commentators in a historical critique at full length, for they could be applicol as far as possible to the whole suitable material of the older tradition, even where it is clear that it is non-tantric. But can we also resist these prescriptions in those cases where a text of seemingly non-tantric character is to be found in a tantric context ? And if we accept the tantristic interpretation of such texts : have the words been uttered by the creators of the Tantra with these meanings in mind? Or has a non-tantric text been incorporated into the Tantra, because to its creator(s) it seemed capable of having the tantric meanings or rather, because it carried these meanings in his understanding? However one may judge these alternatives and however important definite answers for a history of the formation of the Tantras would be, the question at what moment a not evidently tantric text becomes tantric can only .be answered pragmatically, i.e. with the help of the context. If it is also to be found outside the tantric context, it can be interpreted in its own right, having a non-tantric meaning, too. Within the context of a Tantra, however, we have to assume that it also hears the tantric meanings. That is, I think, that without the existence of text-critical arguments outside of a Tantra we are for pragmatical reasons not entitled to neglect the tantric meaning as called for by the commentators when we interpret such seemingly non-tantric texts. Only when the developmental succession and the coexistence of coherent tantristic conceptual structures will have been clarified and thereby the dependent hermeneutical instruments will be localized historically it will also be possible to interprete the Tantras them 3 E.g. the Aksayamatinirrlesasutrum quoted in Prusannapada, ed. by La Vallee Poussin, St.-Petersbourg'1913, p. 43, 4-9 (cf. Lamotte, loc. cit. p. 352). - Cf. below 453. s I use the words "tantric" in the sense of "related to the Tantras" and "tantristic" in the sense of "related to the systematic or religious traditions based on the Tantras". 447

Page #4
selves more critically in each single case. As long as these developments cannot be judged we have to proceed for the time being from those literary strata which give us exact exegetical advice from a certain historical mo. ment onwards in order to understand the meaning of the tantric revelation. But what is tantric meaning in such a milieu ? There is no answer to this question generally valid for all tantric revelations. It is well known that the various Tantra-traditions are considered as self-contained corpora and it is, therefore, necessary to make a beginning with a particular Tantratradition where we find a clue at all. The most elaborate system of tantristic hermeneutics I have found so far has been developed by the so-called "Arya-school" ('Phags lugs pa) of commentators on the Guhyasamujatantram. This school produced a considerable amount of literature from about the eighth to the twelfth century A.D. which has been held in high esteem especially by the more scholarly inclined Tibetan tantristic exegetes like Bu ston and Tson kha pa and has thereby exerted great influence on the interpretational techniques of the dGe lugs pa scholars. A summary of the exegetic categories and rules of this school has been yiven by Candrakirti in the 53 introductory verses of his Pradipoddyotani. According to Candrakirti's words? he only summarizes the precepts of the "Explanatory Tantras" (ryakhyatuntra-, bsad pa'i rgyud). Matsunaga Yukei(r) has questioned the "traditional" identification with the Vajrajnanasamuccayatantrar' as the source for these categories and has assumed 6 P 2650 (sGron mo ysal bur byed pa zes bya ba'i rgya cher bsad pa). A manuscript of the Sanskrit text was found by R. Sankrtyayana in "Sa-lu" (Zva lu); cf. his report: Sanskrit Palm-Leaf M88. in Tibet: J BORS 31 (1935) p. 37 (No. 112). The photographs of the Ms are kept at the K. P. Jayaswal Research Institute in Patna. They were available to me through the kind help of Prof. H. Bechert, Gottingen, who has provided me with a micro-film of the copy kept at the Indologisches Seminar der Universitat Gottingen and to whom I would like to acknowledge my thanks on this occasion. Prof. J. W. de Jong, Canberra, has prepared a transcription of the entire Mis and, since my own copy made from the micro-film proved rather difficult to read. I am extremely grateful to Prof. de Jong for having sent me a copy of his transcription of those parts of the text I needed most. I also want to thank Watanabe Shigeaki, Tokyo, for having pro. cured copies of the Cone and Derge editions of the Srijnanavajrasamuccayah for me. The text of these introductory verses covers ff. lb-3b4 of the Sanskrit Ms and ff. 1-4b1 of the Peking edition. ? Prad vr. 130, 400, 52a, 53d. 8 A Doubt to Authority oj the Guhyasamaja-Akhyana-tantras: IBK 12 (1964), pp. 840 f. E.g. Bu ston, Sad sbyar f. 21a5, 28a4, 2965; Tson kha pa, mChan 'grel f. 582, 9a7, 11a5. But this is not even the opinion of the older, Indian commentaries: the corresponding identifications they give are the "Vajramala and others" (Sri Karunasri 448

Page #5
that at least the second part of this Explanatory Tantra depends on the Pradipoddyolanii. 10 As the Derendrapariprcchu,11 one of the other Explanatorv Tantras, has not been translated into Tibetan and is therefore no longer extant, we cannot definitely estimate the degree of originality of Candrakirti in composing his exegetical system. It seems to me, however, that although we can possibly trace most of the components of his system"? to either the Explanatory Tantras or the Sadhanas13 of the school, the system as a whole might indeed have been conceived by Candrakirti. And where Candrakirti's description does not provide complete information, e.g. with regard to the names of all the categories, we may use, as Matsunaga demonstrated,14 the second part of the Vajrajnanasamuccuyatantram as a corroborative instance. Candrakirti's summary is in any case an important turning point in the development of the exegesis and hermeneutics of the Cuhyasamaja-tradition. Starting from it the formation of its different components will have to be investigated in the future, since his summary itself represents already the final culmination. . Tantristic exegesis is the main theme of Candrakirti's introduction. After a concise summary of the "five stages" (pancakramah, rim lna) (vv. 2-6) and the motives of his following elaboration (vv. 7-10) he gives pada's Prart. uddy. f. 1868, Bhavyakirti's Prid. abhis. prak. f. 11264) or the "Caturde. vipariprccha and others" (Aryadeva's Prad. tika f. 14a8), thus corroborating Matsunaga's doubts. 10 loc. cit. pp. 838 f. and 836. 11 According to A. Wayman, (The Buddhist Tantrus. London 1973, p. 14) this text is cited in the Pradi poddyotron, but I do not know where. sritkotikam vyakhyanam and caturvilham akhyanam (except for akstrarthah!) arts Extant in the first part of the Vajrajnanasamuccilyatantram (cf. Mutsunaga, loc. cit. p. 838). The nine topics of the first two "preparations", viz, u polyhatuh (comprising : samjna, nimittim, karta, prama, pratyojanam) and nyayah (comprising: santanah, nudanam, niruktis, hetuh) ure rooted in the general Indian tradition of a methodology valid for the composition of sastras and Buddhist "anthropology" respectively. The fifth (dvividhabhedah) and sixth (pancapudgalah) "preparations" are related to the Cureer of the tantristic adepts and although I do not know of an older list than Can. (rakirti's it seems very likely that the pancapurlyalu-list is earlier, too. Unfortunately I cannot give more information at this time for luck of more specific research. The ser. enth (saty dvayavinirnayah) is only of summarizing function (cf. Prad v. 51). 13 The Pindikytasadhanam (P 2661; = Pindikramasadhanani) contains the utpattikramah, the Prncakraniah (P 2667) the utpannakramah of the school. For the Sanskrit texts cf. PKr. * loc. cit. pp. 8:39 f. 29 449

Page #6
a thoroughly systematic survey of a group of exegetical tools collectively called "seven preparations" (suptalumkarah, rgyan bdun). 15 These "seven preparations" provide the formal means which are used in interpreting the revelation of the Guhyusumijutantrum "with its extensive objects condensed in a small textual space". 16 Matsunaga has given a surveyexplanation of these "seven preparations".!? As a whole they are those exegetical topics which help to organize the contents of the Tantra and 15 According to Bhavyakiti (Prod.bhis. prok. f. 10114-102:15) lampkaruh, ryyin is either a reality which makros perfect, complete (olnos po syrub par byerl par gyur p, f. 10155) - and in this case it is a sprcific ornament of word or of meaning" (syru ilun vlon rgym, f. 10165 f.)-, or the means or function which makes perfect (hycu pa syrub par byeil pui, f. 1011,5). The difference between these two meanings of the word is taken to l the difference of what is to be expresses (brjod prar byr bri, *racyri-) from whut is to be revealed (usual bar byri ba, *cyangya-) (f. 102:1-4). This sober explanation of the trini's meaning is kept fairly general and would allow the translation ornament" as used for the term within the context of the science of poetics. Navertheless I think we have to refrain from translating rlamkarah in our context with ornament", for it evidently means more than a beautifying -lement and, although it beurs essentially the same meaning us in the context of poetics, the . usual translation's emphasis on the esthetic aspects of the conceptual spectrum of the word is too narrow. This is especially clear when we rocall its sixth item which consists of the five kinds of men (pulynlu-, ynni cag) un recipients of the tantric revelations (dealt with in Prud vv. 41-43). Therefore this attrmpt of another translation for our context. My comprehension of the meaning of alamkaruh in our context is this: the seven alimkaru. are those el-mnts which the Tantra is presumed to be using in order to fulfil its revelatory purpose at its best on every level of recipiency. The Tantra is "ready, complete, maslo adequatr, arranged, prepare?" by means of these alim kara.. The possibility of the Tantra's full value bring present includes, of course, a consideration of the recipient's abilities, too. N iated by their being taken into consideration on the side of the Tantra the reipients become a particular category of formation of the Tantru itself. Naturally, then, it is impossible to understand and explain the meaning of the Tantra without taking re-course to these rupkuril. Thus we have to go back to the broadler meaning of the term as it "magical-religious expedient" (cf. e.g. J. Gonda, The Jeuning oj the Iloril lumkari. Selected Studies II, Leiden 1975. 265). I, therefore, assumne a double meaning of the term, both expressed by the proposed translation preparation" (German: "Zurichtung"): a) "preparations" of the rexegitical object, the Tantra, which prepare its words in a way that they muy be understood in its full mening by the apt persons. b) "preparations" us categories of exegetical study when the preparations" of the Tantra are being taken into consideration and used to analyze it. It is this lutter ineaning of the term we are concerned with in our paper. Of course I still translate the word by "ornament" in the compounds sublu-, arthalamkarah. 16 gral payranthe prabhutarthe samaje (Prud v. Sab). 17 Shichi shoku ni (suite: IBK 11 (1963), pp. 470-476. 450

Page #7
to formulate its interpretation in a philological and systematical way. That is with the help of these tools the tantric revelation can be shaped into a system of tantristic religion. 17 There are two of these "preparations" of special interest for our theme: the third one, which is called "explanation by six ends/alternatives" (satkotikam vyakhyanam,*rgyas bsad mtha' rnam pa drug), and the fourth, which is called "fourfold explanation" (caturridhum akhyanam, bsad pa rnam ba bzi).18 Both "preparations" are called "explanations" (akhyanam, vyakhya, vyakhyanam, bsad pa, rgyas bsad pa) which would characterize both as hermeneutic instruments. On closer examination, however, we find that they differ substantially. While the set of the "fourfold explanation" is truly hermeneutic, as will be shown later, the third "preparation", "explanation by six alternatives" consists actually of different kinds of speech. It is defined as "an ascertainment by means of words of six alternatives, which is given by the methods of the Yogatantra and elucidates the sealed meaning. That is: this category presents six alternatives of words or speech as used by the Tantra to propound its object.20 It contains these alternatives in three pairs which are already known from pre-tantristic exegesis,21 although I have not come across the set as such in non-tantric Mahayana-literature so far. The pairs are:22 "implicit or hinted meaning" (neyartha-, dran [ba'i]don) and "explicit or evident meaning" (nitartha-, nes [pa'i] don); "intentional language" (sandhyaya bhasitam, dgons [pas] bsad) and "non-intentional language" (no sandhya, nasandhya, dgons [pa] min [pas bad]);23 "literal or standard (words)" 17 Of course we have to keep in mind that these tools are applied only to this particular Tantra, the Guhyasamaja, and cannot be taken as a set of tools generally applicable to any tantric revelation. 18 Prad v. 1ted; later the third is simply called "six alternatives" (satkoti-, mtha' drug, v. 24b), the fourth "explanation" (vyakhya, bsud pa, v. 31b). For other variants of the terms cf. Matsunaga, loc. cit. p. 471. 19 punas trtiyo 'lankaras satkotipadaniscayah | yogatantranayoddisto mudritarthaprabodhanah | Prad v. 24. 20 Cf. the expression rjod byed in Bu ston, bu sbyar f. 21a6. 21 Cf. Lamotte, loc. cit. passim. 22 I give the terms as they appear in Prad vv. 25-30. 23 On the latter two terms ef. Lamotte, loc. cit. pp. 354 ff. and p. 354, n. 2 (for earlier literature); A. Bharati; Intentional language in the Tantras: JAOS 81 (1961), pp. 261-270; A. Wayman, Concerning samdha-bhasa (samdhi-bhasa) sandhya bhasa. Melanges d'indianisme a la memoire de Louis Renou, Paris 1968, pp. 789-796; Twilight language and a Tantric song, in The Buddhist Tantras. London 1973, pp. 128-135; G. R. Elder, Problems of language in the Buddhist Tantras: History of Religion 15 (1976), pp. 231-250. 29* 451

Page #8
(yatharula-, ji bzin sgra) and "non-literal or coined (words)" (na rutaaruta-, [ji bzin sgra] ma yin, ji bzin min pa). While the second and third pairs contain terms which indicate differen sorts of speech24 in the revelatory texts, it is only the first pair of alternative which suggests a hermeneutical import. For these two differentiate the texts according to the way they bear their meaning and prescribe a corre sponding interpretation of the relevant texts. But this seems to be strictly true only for their non-tantristic usage in Mahayanistic and Abhidharmi exegesis. It is still to be investigated whether their meaning and application remained exactly the same in tantristic exegesis as well. So far I have no definite answer for this question which requires a much more extensive historical examination than I could venture on at this time. I think, how ever, that within tantristic exegesis-or at least within our list of the "si alternatives"-these two terms are not primarily hermeneutic, i.e. pre scriptional for an interpretation, but rather refer to different kinds of speech like the other terms of the list. The fact that they are "ornaments of mean ing" (arthalamkarah, don gyi rgyan) does not impede this assumption.2 To give a rough conclusion: the "six alternatives" seems to represent first, possibly heterogeneous list of categories within tantristic exegesis which derives from earlier, non-tantristic exegesis and is intended to give a structurized survey of the actually extant kinds of tantric speech. That this and the next "preparation", are both called "explanation" has the following reason: the third "preparation", "explanation by six alternatives" is "explanation" because the Tantra "explains" by means of using six There is evidently some difference of opinion among the commentators as to which of these "preparations" are ornaments of words and which of meaning. Tibetan exegete: (Bu ston, bad shyar f. 24b1f.; cf. also Tson kha pa. Man gsal f. 207b3f.) say that the third pair must be taken as both, sabda- und arthalamkarah, while Bhavyakirti (Prad abhis. prak. f. 107b5f.) considers the first pair and "non-intentional language" as arthalam karah, and the third pair and "intentional language" as sablalamkarah. 25 Because in their case it is nonetheless the choice of different words which serves as a basis for the different meanings. In other words, they are ornaments of meanings but are definitely also different kinds of speech. If, however, due to their earlier, mala yanistic hermeneutic function one accepts a hermeneutic import of the two terms, too the question of their particular hermeneutic function and its relation to the hermeneu tic categories of the list of the "fourfold explanation" must be answered. This probleir has been investigated e.g. by Tson kha pa in his commentary on the Jnanavajrasa muccayah (Man gsal f. 206b2-20763), where he can show at least one case, wherealthough in different texts-one and the same tantric text has been interpreted once with the help of the "fourfold explanation" and in the other instance with the help of the "implicit and explicit meaning". 452

Page #9
different kinds of speech, while the fourth, "fourfold explanation", is "explanation" because the Tantra "explains" by referring to four different meanings. And both of them can again be taken dialectically as "tantric explanations and ils "tantristic explanations" as well. I may add a systematical argument to strengthen this interpretation of the whole list being non-hermeneutic: there is no attempt to be found which relates the terms of this list to the very core of the tantric development-structure, i.e. the "five stages" (pancakramah, rim lyn), which relation turns out to be the main motive for a distinction of tantric meaninys. The contrary is true of the fourth "preparation", the "fourfold explanation": its structure is directly related to the "five stages" and it can thus be considered as the genuine set of tantristic hermeneutics. That it is later than the "explanation by six alternatives", too, is evident from the even the "five stages" itself does not appear in the oldest parts of the relerant literature, namely the Julatantram itself and the original part of the oldest of the Explanatory Tantras, the Vajramalu.26 It is, therefore, basically this set of four different explanations of one and the same text which constitutes the categories of hermeneutics used in this particular school of the Guhyasamaju-tradition and which can give us a clue in our search for tantric meaning. This list contains the following terms:27 1. "literal meaning" (aksurarthah, tshig gi don) 2. "common meaning" (sumastungam, spyi'i don) 3. "the hidden or pregnant (meaning)" (garbhi, sbas pa) 4. "the ultimate (meaning)" (kolikam, mthar thug pa )28 26 CF. Matsunaga, at Doubt to futhority..., pp. 843f. 27 As uppearing in Praed v. 31. The following are the main Tibetan variants of these terms: 1. yi ge'i don (Praul v. 33, V.JPST f. 294a 4), 3. sbas pa (Prod v. 35, VINST ibid.), sbris pa'i don (Prad. (ohis. prak. f. 108b3, 5), 4. mtha' (Prad v. 36), mthar thug (:thugs P) pa'i don (Prad. abhis. prk. f. 109a2). 28 As the terms are of a somewhat strunge appearance I muy venture an explanation. aksurarthah is evidently to be taken as "the ineaning of the words", where aksurais short for aksarisamudaya., i.e. paulu-. It is, however, not only the meaning of the single words but also the meaning of the whole statement. With the broadest meaning of aksara- as "words individually and taken together" I take the term, therefore, as "literal meaning in general". samastangam, where anga- is translated into Tibetan by don (= artha-), is a difficult term. Except for some remarks of Bhavyakirti I have not found anything that could be of help. Bhavyakirti explains anya- with reference to the two motives which are given in Candrakirti's definition of this category (Prad v. 34): 1. the dispelling of 453

Page #10
The arrangement of these four terms is in accordance with the career of a tantristic vogi. The first term, however, refers to the secular level of understanding. It is therefore only an ornament of words (sabdalamkarah), while the others are ornaments of meaning (arthalamkorah).29 Thus we may refer only to the latter as categories of "tantric meaning" strictly speaking, although all four form a well-conceived and logically arranged set of tantristic hermeneutics. Candrakirti's definitions 30 relate these four categories and their eight remorse (kroukrtyarinirylli.) with those who adhere to the Sutras and 2. the connecting with the stage of generation (utpattikramatyoga-) (cf. Prad. ubhis. prak. f. 108a7). Thus samastangn could be taken as it bahuvrihi: "with common inombers, parts" said of arthah, the whole expression being understood as "meaning, where both (rele. vant) inembers are commonly given". This bahuvrihi would then have been substantivized secondarily to si mistangan with the same sense. That this meaning has both members, further, ineunis nothing else but that this meaning is of relevance for those who adhere to the Sutras as well as for those who practice the stage of generation. Therefore I translate the term as "common meaning". garbhi is an attribute of arthuh, mostly used, however, by itself. Cancrakirti (Prad. 35) define's garbha. as the first three stages of the stage of perfection (simpronnakramni.) (cf. also Prull. abhis. prak. f. 108b6f.). garbhin., therefore, is that meaning which refers to these three stages. kolikum is explained by Candrakirti himself (Pruul u. 36). He says the word kolnih. meant "encl, limit" (puryantavacaka-). The word is in other contexts not known to bear this meaning, apprars, however, among others in a list of coined words" (uruta.) given by Tson kha pa in his commentary on Prud v. 30 (cf. mChan 'grel f. 8a4), kolikam like garbhi is mostly used by itself as a substantivized attribute of arthah. In Prad s. 36, however, it is clarly an attribute (kolika[h]proktah). 29 Cf. Bhavyakirti, Prad, abhis. prak. f. 108a3 and 109a2. This may be the reason for the conspicuous lacking of the aksurarthuh in the first part of the Vajrajnanasa. muccuyntaentrom, too (cf. P 84, f. 29052-293b8), which has already been noted by Matsunaga (lexe, cit., $39). If this cannot be accounted for by another reason, which I clo not se, upare entitled to take Candrakirti indeed for the one who has put together the complete list of the "fourfold explanation" by incorporating with the inclusion of the "literal meaning" the realm external to the religious career, thus truly starting a * system of tantristic hermeneutics of enduring influence. 30 Prad vv. 33-36 (= f. 256-3a2): bahyrsastrarilo nike sblumatra parayanah/ tadyut pretti prikasayri aksirartho nigadyritel/ "The many knowers of the non-Buddhist Sastras (consider) the words alone as the last '''sort. In order to explain the analysis of the words) the literal meaniny is statel." sutrantadau prapannunam kaukrtyarinivsttitah/ ut pattikramiyogac cu samastangam prakasitam// "In order to dispell the remorse with those who keep to the Sutras etc. and to connect with the stage of generation the common meaning is proclained." 151

Page #11
subdivisions to the levels of understanding gained in the progress of a tantristic yogi's career. These meanings are being borne by the words and sentences of the Tantra and are successively available to the practitioner: 1. The "literal meaning of the teachings is meant for those secular, nonBuddhist scholars, who cling to the words alone, in order to explain the wordanalysis (ryutpattih, bhod pa), because-as Bu ston says31 -- if the analysis is not given, they will not become candidates (vincyah, gdul bya). 2. The "common meaning" is of relevance for two levels of progress : firstly for those who adhere to the Sutras, i.e. Hinayana- and MahayanaBuddhists, with the particular purpose to dispell their scrupulous uneasiness or remorse (kaukrtyam, 'gyod). And secondly for those who adhere to the yogic practices of the "stage of generation" (utpattikramah, bskyed pa'i rim), i.e.- according to Bu ston2 -- those who follow the Kriya-, Carva- and Yoga-Tantras. The other two terms are reserved for vogis in the "stage of completion" (sampannakramah, rdzogs pa'i rim) and comprise together five categories which are directly related to the "five stages" (pancakrumah):33 3. The pregnant meaning" has three functions related to the first three stages : fa) Willumination of the nature of passion" (ragadharmaprakasanam, 'dod chags chos rab ston) refers to the first step of the completion, namely the "stage of diamond muttering" (vajraja pakramah, rdo rje bzlas pa'i rim pa). (6) "perception of the truth of conventional existence" (samurteh satya sambodha[h], kun rdzob bden pa rtogs) refers to the third 34 step, the garbhiti yarbham asyisti ragaudharmaprakasanah/ samurtes satyasyomborlha jnanatrayavicetanam// "The pregnant is that (meaning) which has the garbha, (i.e.): illumination of the nature of passion, full cognition of the truth of conventional existence, discerning cognition of the three knowledges." pruibhasvaraprukasus ca yuganaddhaprabodhanam/ drivirhah kolika proktah kolah paryantavacakah// "The illumination of the Clear Light and the awakening towards coincidence is called the twofold ultimate (meaning). kolah is a word for end." 31 bSud sbyar f. 25b6: deltar mi bsad na de rnams gdul byar mi 'gyur roll 32 bSal sbyar f. 26a2f. 33 The relations are shown by Bhavyakirti whom I follow here. Cf. Prad. abhis. prak. f. 10862-109al. 34 According to the sequence in PKr. So far I cannot explain the change of place. 455

Page #12
"stage of establishing oneself" (stadhisthanakramah, bdag la byin gyis rlob pa'i rim pie)35 (c) "discernment of the three knowledges" 31 (nunutraya ricetunam, ye ses gsum po rnam brlugs pr) refers to the second step, the stage of purification of mind" (ciltuvisuddhikramah, sems rnam par day pri'i rim pr). 4. The ultimate meaning" finally has the two functions related to the fourth and fifth step: (d) "illumination of the Clear Light" (prabhusuraprakasah, 'od gsul ba rab ston) refers to the fourth step, the "stage of total enlightenment" (ubhisumbodhikroh, mnon par byan chub pa'i rim p.).37 (c) "awakening towards coincidence" (yngrinuddhuprabohrenam, cun du 'jug jurtos pu) belonging to those on the last level of the "stage of coincidence" (yuganuddhukramah, cun du "jug pri rim pa). This well-arranged Istem of meanings of the tantric revelation3s gives 35 It is artually the prodicate in the definition of this kriemuh in Phr [V v. 10ab). 36 The threknowl.lgs are defined in Pkr III vv. 4 f., 7 ff., 15 ff., 23 ff., 37ab; Matsunaga prefers to use the form trynjnarim (loc. cit., passim), but I find only (vi)jna. natryc (-5. Pkr II v. fbie, III v. tib, v. 36c). Cf. A. Waymun, Notes on the Sanskrit term jnanre: J4OS 75 (1955). pp. 253-268. 37 Cf. PK?ip 44,2 f.: arresuryoprulesakam iti prublasiverier. Earl erre cuiturthabhisumborlh ih. 38 Cf. table. The tomy with asterix are supplemented on the basis of Bhavyakirti's comments (Prarl, abhis. prok. f. 10%b2-109al). "Fourjolil erpla Variety Levels of understanding nation" 1. aksarurthah 1. cyrut pritti prukasa- . 1. bahyrisastravidah tshig gidon, bkort pris rab bstan pa phyi rol bstan bcos nikhas pa yi ge'i don 2. salinastanyam 2. kurtyniniurttitah 2. sutrantadau proprinuh spyi'i don 'yyorl pre zlog par byerl pa | nuo 80:lit rub augs pa 3. utprilikram yogit 3. *ul pattikrinh (kriya, bskyl pr'i rim van lian pa carya, yogatartir followers) bskyeol pri'i rim po 3. garbhi 4. ragadharmaprokusana- 4. *sumpinnakrivih: shrs pot . Pol chags chos rab ston rdzos pa'i rim pre a. *vajrija prikrumah rdo rje balus pa'i rim pri 5. semesteh sittynsambordhah 6. *srudhisthannkruunuh kun riisob bilen pa rtogs pa bdag la byin gyis rlob pa'i rim pre 6. jnanatrayaricetanam c. *cittavisuddhikrumah ye ses osumn po mam brlags pa sems rnam par dag pa'i rim pri 456

Page #13
the impression of being conceived by a certain author with the background of a longer development of its constituent parts. And-for a workinghypothesis - I would consider Candrakirti as the one who has finally put these hermeneutical pieces together. In any case we have to assume that the final development of these tantristic hermeneutics took place around 800 A.D.39 And starting from that period this hermeneutical system was obligatory, at least for all Indian and later Tibetan scholars with an exegetical interest in the Guhyasamijacycle. In the case mentioned at the beginning, of a seemingly non-tantric statement within the Tantra we must, therefore, be aware of the fact that it has-if possible - more than one or two meanings, namely four: one which can be taken as literal, while the others are tantric. List of terms a. Silnskrit 6. Tibetan aksararthah arutaalamkarah upoughatuh karta kolikam kaukrtyam garbhi caturvidhar akhyanam jnanutrayam jnanatrayavicetanam dvividhabhedah travajnanam na rutanasandhya nidanam nimittam niruktih (Variants of the same term are included) kun rdzob bden pa rtogs bkod pa dgons pa min pas bsad dgons pas bsad dgons min dgons bsad 'gyod rgyan rgyan bdun rgyas bsad mtha'rnam pa drug nes don nes pa'i don ji bzin sgra ji bzin sgra ma yin ji bzin min pa mtha' mtha' drug nitarthah neyarthah mthar thug pa mthar thug pa'i don 4. kolikam 7. prabhasvaraprakasah d. *abhisambodhikramah mthur thug 'ord ysal ba rab ston mnon par byar chub pa'i rim pa 8. yugunadihaprabodhanam e. *yuganuddhakramah zun du 'jug pa rtogs pa zun du 'jug pa'i rim pa 39 I follow A. Wayman (The Builhist Tantras. London 1973, p. 14) in giving the first half of the ninth century to Candrakirti. 457

Page #14
no sandhya nyayah pancakrarnah poncapudgalah prabhasvaraprakasah prama prayojnurm yuganaddhaprabodhanan ragndharmaprakasanavijnanatrayain vyakhya vyutpattih sat kotikam vyakhyanam sut kotih samnvTteh satynsambodhah samjni antyadvayavinirunyah santannh sandhyayn bhasitam saptalamkarah Hamnstangam (Iran (lon drar ba'i don 'dod chags chos rab ston spyi'i don abas abas pat sbas pa'i don tshig gi don zu du 'jug pa rtogs pa 'od gsal ba rab ston yi ge'i don ye ses gaum po rnam brtags pa rim lna bsad pa bsad pa mnamn pa bai