Report On Kanjur Of Ta Pho

Author(s): Ernst Steinkellner
Publisher: Ernst Steinkellner
Catalog link: https://jainqq.org/explore/269418/1

Page #1
A Report on the 'Kanjur' of Ta pho by ERNST STEINKELLNER The first description of the manuscript collection in Ta pho (1) was given by A.H. Francke (1914: 37-43) in the report on his visit to Tabo July 29th to 30th 1909. According to Francke's remarks (ibid.: 40) he found two piles of manuscripts consisting of 'loose and disarranged leaves'. Francke noticed the high quality of the handwriting and identified the Prajnaparamita contents. He assigned the 'manuscript' to Rin chen bzan po's own times on the basis of its orthography (2) and deduced the period of the 'manuscript' from its orthography's relation to that of the inscription which commemorates the renovation under Byan chub 'od (ibid.: 41) (3). Finally, he clearly recognized the great significance of these manuscript materials for the history of textual transmission in Tibet: 'The value of such a manuscript for critical purposes is enormous. Works like the Prajna-paramita have up to the present been known only from modern manuscripts or wood-prints. Here, on the other hand, we obtain a text, as it was known in the translator's own days' ("). G. Tucci and E. Ghersi visited Tabo during July 18th to 21st 1933 (cf. KlimburgSalter 1990: 158 ff.) and a great part of Indo-Tibetica III.I (Tucci 1935: 21-115, pls. I-LXII) is devoted to a description of the Ta pho monastery. A whole section, S 22 'La biblioteca di Tabo', is the result of Tucci's experienced investigation of these materials (ibid.: 86-89, pls. XLIV-XLV). In this chapter, Tucci devotes attention to the possible history of the library's devastation and to the difficulty of distinguishing between ancient original manuscripts (1) The proper old spelling of modern Tabo and the etymology of the name are unclear. Several differently etymologized spellings such as lta, rta, sta can be found, and the variations po/pholbo are also attested in inscriptions and manuscripts. For the sake of convenience, I follow the proposal of KlimburgSalter (1987: fn. 9). For the present village, I keep to the spelling Tabo. (2) The characteristic features underlined by Francke are: the da drag after r, I, n, and they subscribed to m before i and e. Strangely, Francke found the inverted i to be absent in Ta pho (ibid.: 92). () This inscription was first published and translated in Tucci 1935: 195-204. A new edition and translation by Steinkellner and Chr. Luczanits will appear in the volume mentioned in fn. 29. (4) Francke seems to have taken the 'twelve books of the Prajna-paramita' (?) as a single unit. His enthusiastic words must be related, nowadays, to our knowledge of a long tradition of copying in these areas. But, on principle, his feelings of the importance of what he saw, were quite in place and deserve our admiration [1] 115

Page #2
and later copies, and he emphasizes the great importance of these various copies for a critical edition of the Tibetan translation of the Prajriaparamita, in its different versions, and of other canonical texts: 'perche esse ne hanno fedelmente e con cura minuziosa tramandato il testo come era uscito dalle mani dei primi traduttori' (ibid.: 88), and finally, gives a list of the most noteworthy remains of the library ("). Since, in fact, he can hardly have had a closer look at more than random samples of these manuscripts and fragments, some of Tucci's remarks can no longer be accepted with regard to the material as a whole, for instance his statement: 'I manoscritti sono tutti in grafia antica' (ibid.: 87). In fact, the writing being exemplified with the four specimens of plates XLIV-XLV, is not typical of about half of the collection. And only a few, not many (ibid.: 88) of the manuscripts and copies are also covered with interlinear glosses. Tucci thought that the core of the library consists of manuscripts and copies of West Tibetan origin and contains translations made by Rin chen bzan po or his collaborators and their schools; that the work of copying was continued in Ta pho for a considerable time and strongly supported by the lay people who would donate the costs of copying; and that this tradition had a lasting effect, producing a sort of conservativism, even after the general cultural assimilation with the main centres of central Tibet had taken place. As possible times for the devastation of the temple, Tucci proposed the raid by the Dogra general Zorawar Singh's soldiery in connection with his crushing the rebellion in Zans dkar 1839 or the Dogra expedition against Tibet of 1849 (cf. Petech 1977: 143 ff.; Datta 1973: 116, 131 ff.). After the devastation of the temple and the library 'i pietosi fedeli raccolsero gli sparsi resti dell'antica biblioteca e li deposero senz'ordine nelle scaffalature provvisorie' (ibid.: 87). The most recent mention (6) of the library in Ta pho is found in Klimburg-Salter (1982: 158, pl. 53), based on her visit in 1978 (7). After their visit to Tabo in 1989, both Klimburg-Salter and Tauscher reported the library to be still extant, and I therefore proposed that the next mission to Tabo should also include a team of philologists to investigate the library in particular. From August 30th to September 12th 1991, Tabo was again visited by a team of scholars from Italy, Germany, and Austria, this time within the framework of a formal research cooperation between the Istituto Italiano per il Medio ed Estremo Oriente (IsMEO) and the Institute for Tibetan and Buddhist Studies, University of (5) Tucci mentions the following texts: Abhisamayalankaraloka, Vinayasangraha, Pancalvimsati]sahasrikaprajnaparamita, Satasahasrikaprajnaparamita, Astasahasrikaprajnaparamita, Bodhicaryavatara, Saddharmapundarika, Satyadvayavatara, and Samksiptamandalasutravrtti. (6) D.L. Snellgrove (1957: 183-88), after his journeys to the Himalayan regions in 1953 and 1954 gives a description of Ta pho and its main temples, and does not mention the library at all. (7) On this occasion I would like to express my gratitude to D. Klimburg-Salter for having first informed me of the library and also for having raised my curiosity and interest in the various inscriptions in Ta pho. 116 [2]

Page #3
Vienna (ITBS) (8). The team, which aimed at investigating the Ta pho library consisted of Dr Elena De Rossi Filibeck (Rome), Dr Jampa Losang Panglung (Munich), Dr Helmut Tauscher (Vienna), and myself. About eight days of work were altogether available for an inspection of the Ta pho library. We were greatly helped in this work, both physically and in scholarly ways, by the kind and knowledgeable abbot of Ta pho, the Geshe bSod nams dBan 'dus, and a number of monks, among whom should be mentioned in particular Tshe ldan bZan po, Phrin las rDo rje, and bSod nams sTobs rgyas, the painter. Without their interest and help we would not even have been able to achieve as much as we now can report below. After the first day of work, when we opened fourteen bundles and found their contents to be an unbelievable mixture of numerous manuscripts mainly of Prajnaparamita and Sutra texts by numerous hands, from very different periods with very different papers and formats, many seemingly rather old, but many also considerably later, we decided first to scan through the whole collection of altogether sixty bundles. It took us four tiring days. The impression was overwhelming. We had seen a veritable treasure (deg): altogether 38,000 folios, presumably written between the 11th and 17th-18th centuries. Because of the incredible disorder of this treasure, no immediate systematic listing or photography was possible, and only curiosa, such as remains of gser yig manuscripts, illustrated inceptive folios, and miniatures, were photographed in passing. Time being precious, any attempt at arranging the abundant materials in order to define and document individual units was out of the question, so we decided to confine the subsequent work to certain tasks which could help us to form an idea of the collection's character, its shape, and its genesis. Microfilm documentation also had to be limited to one specific manuscript, that of the Pancavimsatisahasrika which we took out of the bundles (cf. below p. [16]), and to various folios or groups of () This cooperation was made possible through the intense interest of the president of ISMEO, Prof. Gh. Gnoli, and the strong scholarly assistance of Prof. L. Petech and Prof. M. Taddei. For previous cooperations between IsMEO and ITBS, cf. Klimburg-Salter 1990: 145-71. And for a report on the whole expedition, the work of only a part of which in terms of time and people is reported here, compare the report of Klimburg-Salter above. The participation of the Austrian members and Dr Panglung was supported by a grant from the Fonds zur Forderung der wissenschaftlichen Forschung in Osterreich and that of the Italian members through IsMEO and the Consiglio Nazionale delle Ricerche which is gratefully acknowledged. Tiger Tops Mountain Travel India, New Delhi, contributed to the field expedition expenses. () This 'Kanjur' consists of manuscript-folios exclusively. A very few random folios of block-prints were occasionally found, only one of which is worth mentioning: six or seven folios of a print of Tson kha pa's Lam rim chen mo were met with in widely separated places stressing the general impression that the presence of printed material in these bundles is purely accidental. These leaves are now collected in bundle V. Cf. the note by D. Jackson on these folios in the appendix. After photographing, the leaves had to be added to those bundles that were slightly smaller. The bundles are referred to by Roman numerals only. [3] 117

Page #4
folios not typical of the collection in general and therefore more easily discernible in their heterogeneous nature, i.e. some folios of Vinaya commentaries (10) and of various scholastic bsTan 'gyur texts (11). Finally we made a selection of samples of the more ancient writing styles available (12) and of the volume and folio signatures (13). The results of this inspection are summarized below. There is no doubt that this is only a first attempt at assessing the collection's great value for the textual history of canonical literature, particularly of the Prajnaparamita class, and for Tibetan codicology (14). Only a few remarks are possible at the moment; future work will have to make full use of the materials themselves. The following report is mainly descriptive, and it regards the general condition of the library and a number of specific features we were able to discern. Some remarks on the possible genesis of the library have been made, based on observations which remain necessarily hypothetical until a more detailed study of the whole collection can be undertaken. 1. The Library and Its Repository The library consists of 60 huge and heavy bundles of c. 70 cm (length) x 20-25 cm (height) x 21-22 cm (width). The bundles are wrapped in bright orange cotton covers (dpe ras) which were donated by his Holiness the Dalai Lama after a visit to Tabo in 1976. Each bundle nearly fills the space of one of the bookcase openings. Before the cotton wrappings were given, the bookcase units contained either one large or two smaller bundles in 1978 (Klimburg-Salter 1982: pl. 53), the older covers giving the impression of being fairly new, too. It may therefore be assumed that these earlier covers were donated together with the construction of the bookcases in the early 1940s (cf. below pp. [5] f.). Before 1933, the folios do not seem to have had any covers -- at least not a conspicuous number ot them (15). The present two bookcases are laterally attached to the central 'altar' which, early in this century (cf. below p. [5]), was constructed as a frame permitting the view from the assembly hall of the fourfold Vairocana located at the hall's west end (16). Each (19) Now collected in LII. (11) Now collected in V. (12) Now collected in XLVIII. (13) Now collected in XLVIII. (14) This discipline - aside from research in the Dunhuang manuscripts - is hardly existent in Tibetology so far. But it may well become established in the future on the basis of the enormous richness of the Ta pho manuscript treasure. (15) '[...] sono ammassati senza ordine fogli sparsi [...]' (Tucci 1935: 86). '(...) pile of manuscripts about five feet high [...] many hundreds of loose and disarranged leaves [...]' (Francke 1914: 40). (16) This Vairocana is not of one body with four heads, but of four complete bodies attached to a pole at their back (cf. Tucci 1935: 68 ff. and pl. XXII). 118 [4]

Page #5
case is 280 cm high, 150 cm wide, and c. 80 cm deep. The sideboards, shelf fronts, and vertical separations are brightly decorated in carved painted wood and form a kind of grid. Both bookcases are symmetrically structured by six rows of five openings each, 26 x 22 cm in size and of vertically rectangular shape (Figs. 1, 2). The backside to these elaborately decorated bookcases is a most simple one: the cases are open (17), and the bundles are not supported by full boards but only by two strong wooden poles to each row (Fig. 3). Klimburg-Salter (1982: fig. 47) shows the relation of the back of the book cases and Vairocana seen from South side. A comparison of Francke's plate XVa and Tucci's plate XXI shows that the 'altar' facing for the East image of Vairocana must have been created between 1909 and 1933 (18). The lateral extension of the wooden and richly decorated altar facing into two bookcases on both sides cannot be seen on Tucci's plate XXI, but these cases are clearly present -- at least one is - in Klimburg-Salter's figure 53 which was photographed in 1978. Since Tucci speaks only of a 'rozzo scaffale lungo la parete sinistra', i.e. to the left of the entrance, one would assume that the two cases on both sides of the central altar facing did not yet exist in 1933. However, while not given on Tucci's figure 2, the map of the 'Du khan, in figure 1, the map of the whole monastery, the outline of which now would be represented by the altar facing proportionally including the two bookcases are definitely visible in the sketch of this 'Du khan (Tucci's gTsug lag khan). But it may be that this sketch gives only the central 'altar' facing and is not supposed to include any, as yet nonexistent, bookcases. Otherwise it would seem very strange that Tucci did not mention these bookcases * at all. The 'two piles' of manuscripts which Francke saw in 1909, must have been moved onto the roughly made shelf before Tucci's visit. When Klimburg-Salter visited in 1978, a lecture-throne (19) had been added to the altar facing and bookcases on both sides of the older central facing had been attached (cf. Klimburg-Salter 1982: 158a, fig. 53). This arrangement remains the same today (20). This review based on the remarks and photographs of earlier visitors matches well with bSod nams sTobs rgyas' recollections, now a man of fifty and the eldest (17) Should Tabo become open to international tourism in the near future, this treasure will be in immediate danger. Although wrapped in bundles, the lower ones are within reach of anyone passing towards the 'cella', and the mere presence of the monks may well turn out not to be a sufficient protection. It will, therefore, be of utmost importance for the undisturbed existence of this collection in the future, to protect it by adding at least a glass front to the facing and closing the backside of the cases. (18) On this occasion, the two statues assumed to be of Rin chen bzan po (Snellgrove 1957: 148) which were still placed in front of Vairocana at the time of Francke's visit, were also integrated into the new structure as well as the other accompanying items (cf. Tucci 1935: pl. XXI). (19) It may already have been there in 1933, for the dotted lines in front of the Vairocana-circle in Tucci's fig. 1 may well represent such a throne. (20) The gridlike facing of the bookcases is not conspicuous on the latter photograph, but this seems to be due to the protruding position of the bundles. 151 119

Page #6
Fig. 1 Fig. 2

Page #7
Fig. 3 [7] 121

Page #8
local member of the present monastic community. He entered Ta pho monastery at the age of twelve in 1953. According to him, 'in the times of my teacher' the manuscripts were simply piled up, partly with, partly without wrappings. And bSod nams sTobs rgyas thinks that the present bookcases may have been constructed around the time of his birth, i.e. around 1941. 2. The Folios The library consists of a collection of more or less fragmentary remains of manuscripts, it is true. Nevertheless it would be quite inappropriate to speak of 'manuscripts' here and in the present state of the library. To refer to this treasure's 'manuscripts' will not be possible before their remains can actually be reconstituted out of the unbelievably thorough state of disorder which is the most prominent characteristic of the present collection (21). Only in a very few cases do we find more than five to ten folios of the same manuscript together, no such unit exceeding a small finger's breadth -- and there hardly remains even a correct page sequence within such units. Normally, singles, doubles, and triples are the manuscript-units which are dispersed among the 60 bundles (22). I therefore have to limit these remarks to a description of the folios as the manuscript unit for the time being. The number of folios does not vary much in the bundles. Sample counting shows around 640 folios in the average bundle. This means that the Ta pho library contains about 38,000 (!!) folios altogether. These folios survive in a well-preserved state. The climatic conditions of the upper Spiti valley are still very favourable to the preservation of paper and ink (23): therefore, the difference in age between the folios cannot be inferred well from signs of their aging only. What now is well preserved has, however, not always been treated well. Besides their general state of incompleteness and disorder, there are many leaves stained with dirt, manure, and droppings, revealing a history of improper or disrespectful treatment, and many other leaves show more or less severe marks of burning, revealing a probable history of hostility and violence directed towards the sacred writings (24). (21) It must be noted here that the folios are all placed in the proper way, with the verso side upwards and the signatures to the left throughout the collection. This is evidently due to the feeling of propriety on the part of the monks who gathered the surviving folios after each catastrophe. (22) E.g., the folios from the Pancavimsatika-manuscript which we extracted from the bundles (cf. below p. [16]) were to be found dispersed in fifty-eight of these bundles. (2) A general climatic change is said to take place throughout the Himalayan range with a considerable increase of precipitation. In addition, the Government's environmental efforts to stop desertification in the upper Spiti valley seem to be attracting moisture. During our stay, stray monsoon clouds often emptied their load onto the valley, an event considered abnormal by the Tabo people. (24) On the possible conclusions from this state, see below p. [17]. 122 [8]

Page #9
The folio's variation in size and paper is also great. The largest are 69 x 22 (60.5 x 13) with ten lines, or 68 x 21 (58 x 13) (25) with nine lines, the smallest 43.5 * 10.5 (40 X 8) with eight, or 45.5 x 8.5 (42.5 * 5) with five lines. No less diverse is the quality of the papers of these folios (26). The leaves include thin, finely fibred papers as well as heavy, coarsely mottled ones, supple as well as stiff, granular absorbent as well as smooth papers and even a polished kind; sometimes, moreover, two papers are joined together. Binding circles, mostly pairs, are frequent. From among thirty folios which we selected as specimens of the writing, only nine had no such circles. 3. The 'Kanjur . As to its contents, the collection is, with only slight exceptions (27), exactly appropriate to its function as the ritual presence of the Buddha-word in the 'Du khan. In this function and, moreover, because practically all traditional parts of a real bka' 'gyur are actually represented in some way -- including works of 'Dul ba, Ser phyin, mDo man, Phal chen, dKon brtsegs, rGyud - it can truly be called a bka' 'gyur. But this bKa' 'gyur is only a functional one, not a literal one. Its components originated in different times and are contained in many different manuscripts. Only after closer inspection will it be possible to discern whether or not smaller original units had been incorporated. My preliminary impression is that most of the older manuscripts contain text of the yum kind, i.e. the Prajnaparamita in its long ('Bum, Satasahasrika), middle (Ni khri, Pancavimsatisahasrika) and short (brGyad ston pa, Astasahasrika) versions. And I would assume that these yum texts are indeed the origin and now historical core of the whole collection, as was already supposed by Francke and Tucci. Thus these masses of old manuscript remains from many parts of Ka''gyur literary sections still function - and must already have been functioning for a long time - ritually as a bka' 'gyur in the sense of the later canonical collection. This would not exclude the possibility that the origins of this collection consisted perhaps of such texts as the yum which were in a very prominent way considered to be the Buddha (25) The first number gives the length and width of the paper, the second that of the part covered with writing. (26) An experienced specialist will be able to base a study of Tibetan paper and its varieties on this paper hoard. But I cannot even give a description in appropriate technical words here. Nevertheless I should say that the thin, finely fibred and slightly absorbent papers seem to be the oldest ones. It is they which also display the orthographic and palaeographic features discussed below (cf. pp. [10] f.). (27) The one major deviation consists of around 100 folios from different, mainly scholastic texts of the bs Tan 'gyur and some medical texts. The insignificance of these deviations shows the mere fortuitousness of the presence of bs Tan 'gyur texts, comparable to the accidental presence of folios from wood prints. [9] 123

Page #10
word at a period long before the canonical collections and the establishment of their sectional structures and diversity in the 14th century (28). 4. Palaeography and Orthography Nearly all folios are written in a clear and mostly beautiful dbu can script (Figs. 4-7). These leaves, from the oldest manuscripts onwards, testify to very high standards in writing and calligraphic training. Only a few leaves fall below the average high quality. The folios show hardly any differences of aging relative to one another, and the palaeographic and orthographic differences are the only ones of immediate use for establishing a chronology. Since we have long inscriptions on the temple's walls with palaeographic and orthographic peculiarities of the same kind as in some manuscript remains, the shape and form of the texts contained in these manuscripts can indeed be attributed to the same time as the wall inscriptions. This gives us an absolute date, for the temple was founded in all probability in 996 A.D. and renovated in 1042 A.D. as traditionally accepted (cf. Klimburg-Salter 1987: 685), or in 1008 A.D. and renovated in 1054 as recently proposed by L. Petech (29). And in terms of these fragments' significance for textual history, it is not crucial whether the presently extant folios remain from the actual originals of the collection's beginnings, or 'only' from faithful copies made in -- not too much - later times (cf. Tucci 1935: 87); though naturally this is true only as long as they are faithful copies. The palaeographic (1. inverted i; 2. s-P, S-t, s-ts-ligatures of a horizontal kind) as well as orthographic and phonetic peculiarities (3. putting a tsheg before sad; 4. haplography in case of the same end- and beginning-consonants; 5. palatalisation of m before i and e by means of subscript y; 6. occurrence of da drag; 7. exaggerated use of 'a chun) (30) will finally serve as a clue for the relative chronology of these and other manuscripts of this kind. These features occur all together or in various combinations. It is possible that the occurrence of all peculiarities together means the greatest antiquity, and that later one feature after the other fades away. But persistent conservativism on the one hand and later archaizing mannerisms (31) on the (28) For more details on the collection's contents, cf. below, pp. [14] f. (29) In a lecture 'Jungste Forschungen zur Fruhgeschichte des Konigreichs Guge (Westtibet)' given at our institute on March 19, 1991. A final version of this paper will be included in a volume published from the Tucci Archives, IsMEO, on the medieval Buddhist remains in Spiti-Kunawar commemorating the 100th anniversary of the birth of Giuseppe Tucci. (30) Cf. Taube 1980: 11 ff.; Eimer 1991: 251. Of course, the peculiarities mentioned are only the most conspicuous cases. A complete survey will most probably reveal additional features. (1) There are, e.g., folios with relatively 'modern' writing which show an occasional da drag. 124 [10]

Page #11
other hand do not allow the application of such simple schemes with complete certainty. Additional categories will also have to be found and applied in order to gain a chronology of the writing. The Ta pho library may well be just large enough that something of the sort could be discovered after closer, perhaps even statistical, investigation. For the time being, however, I'm inclined to take the occurrence of all or of a majority of the above peculiarities, together with the sort of paper, which I think to be the oldest (cf. above, fn. 26), as an indication for a dating to the earliest period of the monastery's, or the area's, writing offices, i.e. from the 11th century A.D. Miniatures are rather rare, and none have been found in manuscripts assumed to belong to the earliest period. Sometimes the space reserved for a miniature is still empty. All miniatures were photographed. 5. A Note on Volume Signatures In his remarks on manuscript finds from Guge, H. Eimer notes for the first time a peculiar kind of 'signature' ('Kennsiglum') on the folios besides the pagination (Eimer 1991: 254). This signature consists in most cases of two letters, one atop the other, the lower sometimes slightly shifted to the right. This kind of signature seems to be a characteristic of West Tibetan manuscript writing, for it is used for very many of the folios in the Ta pho library. Because of the total dispersion of all coherent fragments, it was not possible to identify the meaning of all the letters used in this system. We are able to interpret, however, the most frequent ones on the basis of several sequences of folios counted over the hundred positions. The following letters were found to occur in the lower position: the three na, na, ma (relatively often), na (only once), da (twice), ya (once), wazur (once) (32). Eimer interprets the upper signs as indicating a volume, fascicle or a text within a collection, and thinks that the lower sign indicates, more likely subdivisions of the text rather than hundred units in folio-numeration. But our material clearly shows that only the latter alternative seems to be the case. However, it was possible to determine only the signs for 100, 200 and 300 during the time available. In some manuscripts (Fig. 5), additional markings by one, two, three vertical strokes or one, two, three small crosses, always made by other hands and mostly in addition to two letter signatures are to be found (33). These additional markings are (32) There are also several signatures with three letters in vertical order, the third one being always ria. In these three letter signatures, we find in addition colour differences: kha-ria-ra, kha-na-na, kha-ma-na (each once), with the first two letters in red, the last in black; tha-ma-na (once) all red and one kha-ma-yu (?) in red. The meaning of this kind of signatures is still unclear (see Fig. 4). (3) Fig. 5 comprises some irregularities. Only the second and third examples from the left are typical for the rest of the collection. The others are still enigmatic. The manuscript of the Lalitavistara which was inspected by Klimburg-Salter in Gondhla, Lahul (cf. her report above), shows even a linguistic [11] 125

Page #12
ku- ) tstshaa tstshaa mi kcu-gsum- gnyis gn-de-lgs-ps-s-ngnyegs-dng- <> + gi-shun / 7 | che.k +++ m3s-ni + byng-chub 'di-l-by- l-b-'u r-l- ln-nug gzugs Page #13
shes-p-ni- rnm-p-th? mej-yi-m saams-2 1 pr-dog-z su-myed-de-- mkhyend-- su-byr-me mu-rgy-sde-p- 'jug- 3pr-dg- pn-dg- rnm-p? p-nyid- m thord / / / cd-mkhyend-p-nyid-nm-pr-dg-p- $ byr-med-s-s-m-yin-th-mye-dd-do / [mznd-kyis- p'i-snyoms-pr-'jug-p-rnm-pr-dg-p-dd / g 32 rnm-pr-dg-ps / tshor-b-rnm-pr-dg-p- / chos thms-cd-mkhyend-p-nyid-rnm-pr-dg-p-'di-l- pr-dg-ps]] rnm-p-thms-cd-mkhyend-p-nyid-n- dg-p-'di-l-gnyis-su-med-de-gnyis-su-byr-med- nyid-rnm-pr-dg-p-ste / de-ltn-n-maaphrd-kyis- m-yin-th-mi-dd-do / / mthrd-kyis-gns-p'i- mthond-kyis-gns-p'i-snyoms-pr-'jug-p- p'nyon lc'i-'dr Fig. 6 MO brgy2:shm-n- \:]]nyn-ky-gr-du-yod-p- spynd-p-'di-ls / de-dg-ls kun-ns-nyon-mongs-p-'m-]] '5 ]]de'-g'-gu-b'chos-sde-dge m-'m / khwaam-the-bumun-p- / spyd-ngo / / 's-'grel-de-rnm- de-dumugs-su-spyd-p'i-shub- ng- stostn-p- Fig. 7 [13] 127

Page #14
evidently of interpretational character. That means it was felt necessary to explain by these markings the meaning of these signs, which at a time later than the writing of the original manuscript had already become unusual or obsolete. These simple graphic and explanatory additions reveal the meanings: subscript na = 100, subscript ma = 200, subscript na = 300. But the real proof of these meanings is provided by cases of subsequent folios from the same manuscript which include a folio with a hundred number (34). The subscripts, therefore, mean different hundreds as stated above, and are used to count the numbers from 'hundred and l' to the next hundred. And the system observed in the Ta pho fragments seems to be the same as that in the Tsaparang remains (cf. Eimer 1991: fn. 34) (35). If future investigation of more presumably old West Tibetan manuscripts should strengthen my assumption that these subscripts were no longer understood at a later time and had to be interpreted by additional signs, the presence of this kind of signature in manuscripts may turn out to be an additional characteristic significant for their age. 6. A Note on the Contents of the Ta pho Kanjur' The great diversity and number of manuscripts, the total dispersion and thorough mingling, the conspicuous lacking of manuscript beginnings and ends, the short time for inspection available and the thereby enforced speed of our work render the following notes highly accidental and preliminary. More definite information will become available after the photos of the specimens have been studied closely. A final survey of the manuscripts and texts in this treasure will not be possible before the necessary work of determining, separating, and reuniting the manuscripts has been accomplished. explanation of the signature ka na. Between the volume signature ka-na and the page number rgu bcu tham pa another hand added gya dgu, evidently to interpret ka-na. (34) 1) In LII we find a preceding folio: 'kha-na (followed by) one cross (and number) go rgyu (-99)'; middle folio: 'kha-na (followed by) two crosses (and number) gris brgya tham pa ( = 200)'; following folio: 'kha-ma (followed by) two crosses (and number) gcig ( - 1)'. 2) or without additional markings; 'ja-na 99' is followed by 'ja-na 200' and 'ja-ma l', and later 'ja-ra 1'. Similar from LIV without markings: 3) 'ga-na 200' is followed by folio 'ga-ma 1', and 4) 'ga-ma 300'. 5) From XLVI we get: 'na-ma 99' is followed by 'ra ma 300' and 'ra-ra l'. 6) Figure 6 shows a preceding folio: 'cha-ma (and number) go rgu ( = 99)'; middle folio: 'cha-ma (and number) sum rgya tham pa ( = 300)'; following folio: 'cha-na (and number) gcig ( = 1)'. 7) Figure 7 shows a preceding folio: 'kha (and number) go dgu ( = 99)'; middle folio: 'kha (and number) brgya tham pa ( = 100)'; following folio 'kha-na (and number) gcig ( = 1)'. (35) The sequence of the letters na-ma-ra in use here for indicating units of hundreds within a volume, has a parallel, at least for the first two letters, in a method of pagination used in some Dunhuang manuscripts. There, the subscribed letters na and ma are attested to be used as differentiating the single letters of the alphabet to make for second and third series of counting pages in place of the normal pagination by ciphers. Cf. Imaeda 1981: 7 f. 128 [14]

Page #15
The list of texts given by Tucci (36), now however, can already be somewhat augmented. We noted the following additional Sutras and Tantras: Angulimaliyasutra, Astamandalakasutra, Kasyapaparivarta, Khar sil gyi mdo, Gandavyuha, Guhyasamajatantra, Purnapramukhavadana, Pradipadaniyasutra, Prasenajidgatha, Phal po che bzun bar 'gyur ba'i gzuns, Mahaparinirvanasutra, Ratnakuta, Lalitavistara, Srimaladevisimhanadasutra, Samadhiraja, Sumagadhavadana, Suvarnaprabhasa, Suvikrantadevaputrapariprccha, Sutrasamuccaya. In terms of quantity, the texts represented in these manuscripts show a rather uneven distribution. Our general impression is that c. 20-30% of the whole collection consists of old manuscript remains, i.e. material from the earliest period of monastic activities in the area, the 11th and 12th centuries. Nearly all of these remains contain texts of the yum kind, i.e. 'Bum, Ni khri and brGyad ston pa. Also in the more modern layers of the collection, the preponderance of yum texts is obvious. Roughly about 50-60% of the whole collection consists of manuscripts with yum texts. The remaining portion consists mainly of Sutra texts of various kind that could be taken to represent the mDo man section of a later-day bKa' 'gyur. Other kinds of texts are extant only in comparably insignificant numbers, and the impression that they entered the library as remnants to be disposed of is particularly prominent with regard to these remains of manuscripts which normally comprise only one to five folios. Not more than 102 folios could be preliminarily identified as containing Vinaya texts (37). A first survey by Dr Panglung resulted in separating 14 different bKa' 'gyur texts and 20 different commentaries from the bs Tan 'gyur or of possibly Tibetan provenance. A few leaves of medical texts and a now already identified dictionary (38) corroborate the impression that this sort of material indicates that such texts were not a major part of the original holdings in the library. These folios must have been added to the collection after having been already broken up and rendered useless. This impression is finally strengthened by the extant fragments of manuscripts with texts from the bs Tan 'gyur. Again, we have no more than c. 100-110 folios remaining from such manuscripts (39). (36) Tucci 1935: 89 (cf. above, fn. 5) and the Mahayanasangraha from plate XLIVa. From among these texts, we did not find at this first glance fragments of the Bodhicaryavatara, Satyadvayavatara and Samksiptamandalasutravrtti. (37) These Vinaya folios are stored now in LII. (38) These folios are examined more closely by Dr Panglung in this volume. In a paper delivered at the 6th Seminar of the Int. Association for Tibetan Studies at Fagernes, 21-28.8.1992, Dr Panglung was able to identify this 'dictionary' as a fragment of a new version of the sGra shyor bam po gris pa. (99) These folios are now stored in V. A first tentative identification yields: an Abhidharma commentary, a Bodhicaryavatara commentary, some Pramana texts, a Madhyamaka text, a few leaves of Siksasamuccaya, Abhidharmasamuccaya, Abhisamayalankaraloka and Uttaratantra. I shall publish a detailed list of these possibly bs Tan 'gyur materials in due time. Dr Tauscher presented his investigation of some fragments with the texts of Dharmakirti's Sambandhapariksa, -vrtti and tika at the 6th Seminar of the International Association for Tibetan Studies at Fagernes, 21-28.8.1992. [15] 129

Page #16
7. A Fragmentary Manuscript of the Pancavimsatika This manuscript was chosen for extraction from the bundles because its various characteristics made it easily recognizable: middle weight and soft, absorbent paper of light colour, with binding circles and holes, a rather large format of 66 x 21 cm (60.5 x 16 cm) with 13 lines. There is no beginning and no end, but a number of chapter endings are present. This manuscript was dispersed among 58 bundles within the whole 60-bundle library, with sometimes no more than one or two folios found in a single bundle. It is now wrapped in a cover and has been added to LII. The reason for taking out this manuscript was -- aside from the ease in identifying its leaves -- the need to procure a basis for answering the following questions: is this material, which dates from a period even before the beginning of serious bka''gyur editorial efforts and long before the late available editions proper, really up to our expectations? Is this an example of the Pancavimsatika in Tibetan translation as it was transmitted in the 11th to 12th centuries? And how much and in what ways does it differ from the editions available so far? Answers to these questions and more detailed information will be given after a more thoroughgoing study by E. De Rossi Filibeck (cf. her contribution to this volume). 8. Some Hypothetical Remarks on the Genesis of the Library and the Causes for Its Present Condition The more or less homogeneous character of the library's oldest parts, i.e. the fact that they mainly contain Prajnaparamita texts, and the great number of manuscripts with the same content (40) both suggest that the two assumptions made by Tucci are very likely correct, namely: firstly that the original nucleus of the library consists of texts, the translations of which had been made in Western Tibet by Rin chen bzan po and his school (Tucci 1935: 88). In Rin chen bzan po's biography, the information is conveyed that the teacher had seven copies of yum and three copies of mdo man (41) given to each temple he founded (42). The nucleus of the library, therefore, consisted (40) In bundle I, e.g., there are 30 different manuscripts of the Astasahasrika with the volume signature 'red ka' and 7 with black ka'. (41) Here to be understood only as a generic term (comparable to the information on the deposit of a complete Tripitaka to Rad nis, cf. Tucci 1933: 70 and below, fn. 43). (42) In the biography of Rin chen bzan po by his disciple Guge Khri tan pa Ye ses dpal which is extant in two versions (Collected Biographical Materials about Lo chen Rin-chen-bzan-po and His Subsequent Reembodiments, ed. Rdo-rje-tshe-brtan, Delhi 1977, texts nos. 3 and 6) it is reported that Rin chen bzan po provided three sets of Sutra collections (mdo man cha gsum) and seven sets of the Satasahasrika Prajnaparamita ('bum cha bdun) to the 21 smaller places (yul chun) he had founded (ibid.: 96, 2-4; 262, 3-6; cf. also ibid., text no. 5: 179, 3-5). I owe these references to Dr H. Tauscher. Snellgrove's translation of the passage in text no. 3 above (Snellgrove & Skorupski 1980: 92; for the 130 [16]

Page #17
most probably of the manuscripts with yum texts and additional sutras translated at the same time (43). And secondly, that from the very beginning these manuscripts continued to be copied, both for reasons of dispersion and renewal as well as of gaining merit (Tucci 1935: 87). Monks and soon lay people, too, must have joined their different but mutually supportive interests in giving growth to a considerable production of manuscripts with a very good average quality. This kind of accretion may have continued for centuries. But it was hardly the only kind that took place. It is evident (cf. below, fn. 47) that at least one attempt at burning the manuscripts happened in the monastery itself. But aside from traces of burning, wear, and age, there are a considerable number of folios with substantial quantities of dried dirt, manure, etc. (44), which are not to be expected in a monastic library. For even after the devastation of the monastery itself, the library would not have been left lying around outside, but would have been brought back into the temple as soon as possible. Therefore most of these dirty leaves in very bad condition, and possibly also a number of the burned ones, did not originally belong to the Ta pho library. They must have belonged to either other temples in the area (45) or private houses. When they were burned, destroyed or became useless in their respective places, they were deposited as sacred waste in the only collection of manuscripts (or later manuscript remains) that seems to have survived the vicissitudes of centuries in the valley (46). Although it is very likely that many of the scorched folios were burned elsewhere, the single fact we can prove, is that the last burning took place in the monastery text cf. 107, 3 f.) is wrong. What is said, is: 'In these 21 places, he provided after the lapse of one year three sets of the Sutra collection (and) seven sets of the Satasahasrika' (yul chun ni su rtsa gcig tu lo gcig 'khyud 'khor lo mdo man cha gsum 'bum cha dbun[...]zabs tog mdzad do. Variants: lo gcig no. 6: lo no. 3; 'khyud no. 3: khyud no. 6; 'khor no. 3: khor no. 6; man no. 3: mans no. 6; gsum no. 6: gsum gsum no. 3; tog : rtog nos. 3, 6). (43) Interesting in this connection is the mention in Rin chen bzan po's biography of the equipping of the temple at Rad nis, his birthplace, with a basic library: 'Regarding the number of extant volumes (poti), there was a complete Tripitaka (sde snod gsum ka). (And) there were all full and medium-length (versions of the) Prajnaparamita, (and) in addition to the complete 18 originals (yum) and copies (sras) (of these) two sets of the Astadasasahasrika, five of the Astasahasrika. In short, in all there were 468 volumes (poti)' (ibid., no. 3, 103, 1-4; in Snellgrove's ed. 107, 31-108, 2 and transl. 94). Aside from the fact that the note on 'a complete Tripitaka' must be taken as having been made in general terms as referring to a collection of Vinaya and Sutra texts, the more detailed indication of Prajnaparamita holdings are a trustworthy testimony as to the main character of Rad nis' first library. And that the first library at Ta pho was equipped in a very similar way can safely be inferred. (44) For reasons of respect, the monks and we removed the grosser dirt while handling the folios, so that now the general picture is much better. (45) All other nearby monasteries and temples have been devastated and ruined at one time in their history because of their strategic possibilities. Only Ta pho, situated at the bottom of the valley, was never a threat and therefore not so much in danger either (cf. Tucci 1935: 86 f.). (46) According to the abbot bSod nams dBan 'dus, a 'Kanjur' of the Ta pho kind consisting of c. 10 bundles still exists in the Sel mkhar monastery south of Sumdo. [17] 131

Page #18
itself (47). In X, I found a folio (48) of the Pancavimsatika manuscript that we collected from the bundles among a number of other different single folios - and of a much later type, with new orthography, and all possessing exactly matching burned edges. This folio has, therefore, been burned after being already mingled with other folios of quite different origins. And such mingling can only have taken place in the monastery. That this group of auto-da-fe victims consisted of well-mingled very old and very late folios would further indicate that this incident happened at a late stage of the library's history. For the general and thorough disorder of the manuscript remains in this library, there seems to be only one possible explanation: repeated devastation. Tucci (1935: 86 f.) thought only of the visit of Zorawar Singh's army. But as shown above, the last burning -- and that definitely seems to be due to a Dogra invasion (49) -- was done to a collection already in total disorder. In other words, I think that the thoroughness of the folio mixture in the Ta pho library is the result of several attempts at burning and destroying. The manuscripts and leaves or bundles were probably always thrown out to be burned on the spot (5) or outside. When all was over, they were collected and restored as far as possible to the temple and to their ritual functions. And there must have been always remains from other places at such times to add to those from the monastery itself. There was never need, however, even if sufficiently trained monks were available, to put them back in order, for their ritual function was not diminished. Even in as 'remote' a valley as this lower Spiti area, there is no want of historical occasions for such catastrophes. The hostile Moslem neighbourhood to the South and (47) This proof cannot be re-examined, I am afraid. The crucial folio was added to the Pancavimsatika and is now in LII, while the adjacent folios remain in X. And I forgot to mark or separate them, nor did I make any notes. Anyway, because of the inscriptional and archaeological evidence for a fire inside the 'Du khan (cf. below, fn. 49), this is not the only basis for our conclusions. (48) The signature is lost. () An inscription on paper placed between figures 22 and 23 on the north wall immediately to the side of the book shelf refers to a devastation and restoration of the temple during the period of the Dogra wars. It says that 'after the appearence of [Zorawar] Singh's army (sen gyi dmag) on the 16th day of the 9th month of the fire-bird year (i.e. 1837 A.D.) and the subsequent demolishing (bzig) of the statues (sku) in this temple (sprul pa'i gtsug lag khan) the monks who had escaped ('byol sor) began a restoration ('byol gsos ?) there on the 5th day of the 4th month of the earth-dog year (i.e. 1838 A.D.)' (me bya zla ba dgu pa'i tshes bcu drug la sen gyi dmag thon nas sprul [: sphrull pa'i gtsug lag khan 'di yi sten sku mams bzig (: sig] nas de la 'byol son dge 'dun mams kyis sa khyi zla ba bzi pa'i tshes Ina 'byol gsos (: sos] kyi 'go btsug pa yin D. A fire is not mentioned in this inscription, but the whole wall at this spot is rebuilt and must have been broken down (oral communication from D. Klimburg-Salter) as a result of considerable destructive efforts that may have included an attempt at burning the library too. The incident may well have taken place during the time of the flight of mTshog sprul to Spiti, when 'he was hotly pursued' and finally reached Bashahr in October 1937 (Petech 1977: 142). (59) As in the case of Zorawar Singh's ravaging. 132 [18]

Page #19
West provided for centuries a sufficient background to explain the destruction particularly of Buddhist statues, temples, and books (51). 9. Outlook on Future Work The photographs taken in 1991 of selected materials will first be examined by the members of the team, and from this study a more concrete assessment of the library's philological value should result. After organizing provisions to be made in order to protect the library by encasing, the original manuscripts have to be reconstituted from the dispersed folios. Only then can a sizeable body of early manuscripts from the period before the canonical collection be catalogued and photographed, and thus become available for future philological studies (32). This library's probable but not yet actually confirmed usefulness for the prehistory of the bKa' 'gyur collections, for the history of single texts mainly of the Ser phyin and mDo man classes, for the history of the Tibetan translation idioms, the development of palaeography, orthography, manuscripts writing styles, manuscript organization and paper production and trade in Western Tibet now remains just a hypothetical assumption. But it is a valid one, and justifies the need for further scholarly work and for careful preservation of this valuable treasure. That an independent textual tradition existed in Western Tibetan areas until the late 17th century (cf. Eimer 1991: 245-47) was recently demonstrated by M. Hahn's investigation of texts from the manuscript Kanjur of Phudrag (Phug brag) in Ladakh (Hahn 1988). And this is again confirmed by Paul Harrison's recent edition of the Drumakinnararajapariprcchasutra, Tokyo 1992 (cf. particularly pp. XXXII-XXXVII). Moreover, manuscripts seemingly belonging to the ancient periods in style and orthography were found in the area since Francke started to look for such materials (53). More single manuscripts and even collections are still to be identified. (51) Ta pho itself seems to be mentioned only once specifically in the sources as the place near which 1839 the regent dNos grub bstan 'dzin was caught in his flight (cf. Datta 1973: 116; Petech 1977: 143). But Spiti was troubled or possibly troubled, e.g., during the expedition of Sultan Zain ul-Abidin before the middle of the 15th century (Petech 1977: 23), during Mirza Haidar's expedition to Tibet of 1533 (ibid.: 27), during its conquest by bKra sis rnam rgyal of Ladakh after the middle of the 16th century (ibid.: 30), during an invasion by armed parties from Kulu' in 1822 (ibid.: 131), and finally during the Dogra period (ibid.: 138 ff.). While devastation motivated in addition by religious hostility (cf. the note on Ghulam Khan in Francke 1977: 158) may be primarily connected with Moslem soldiery, there is no doubt that Spiti was also subject to invasions by soldiers, bandits and other violent people from Ladakh, e.g. during the tensions between Ladakh and Guge in the times of Sen ge rnams rgyal (ibid.: 41 ff.), when the Ladakhis also seem to have made use of Moslem mercenaries. (52) The recent development in Kanjur research has been surveyed by Helmut Eimer in his introductory article (Eimer 1992). (53) Cf. e.g., Francke 1914: 78, 119 f., (b) nos. 1, 18, 19. [19] 133

Page #20
The Ta pho 'Kanjur' is only a further example testifying to a real basis of our assumptions that there are in West Tibet the remains of a manuscript tradition which merit attention. For in all probability, these remains are to some extent of import as a new textual source, independent of the hitherto known transmissions of Tibetan canonical texts. Postscript A succinct version of this report was presented during the 6th Seminar of the International Association for Tibetan Studies at Fagernes, 21-28.8.1992, together with the papers by Dr Panglung and Dr Tauscher on specific fragments in the collection (cf. fn. 38 and 39). These papers and Dr De Rossi Filibeck's investigation of the Pancavimsatika-ms. are published in this volume. During the Seminar, Chr. Cuppers, P. Harrison, C. Scherrer-Schaub, and Lobsang Shastri expressed their intention to cooperate in the future. With their help it will be possible to reconstitute the original manuscripts soon. In a letter of July 17, 1992, the abbot of Ta pho, Geshe Sonam Wangdu, informed me of the fact that the bookcases for the 'Kanjur' are already under construction. The necessary financial support for this work has kindly been offered during the conference at Fagernes by Mr Tenzin N.G. Ronge of the Deutsch-Tibetische Kulturgesellschaft, Konigswinter, and is gratefully acknowledged. APPENDIX Note on a Leaf from a Xylograph Edition of the Lam rim chen mo Recently Located in Spiti by D. JACKSON The five or six folia from a xylograph of the Lam rim chen mo recently found in Spiti (Ta pho monastery) belong to a previously unidentified and possibly 'early' (pre-17th-century?) gTsan or mNa' ris edition. In this edition, each side contains seven lines of text, and the size of the printing blocks (excluding the characters indicating foliation, but including the thick lines at the ends) is, for example: fol. 322a: 50.3 x 6.4 cm, and fol. 322b: 50.6 x 6.1 cm. Margin notation: pa lam rim che ba sum brgya ner gnis. Fol. 322a, begins: rab kyi yul sun 'byin pa'i phyir roll. It ends: de nas blo gzan zig gis ran. Fol. 322b, begins: bzin med pa de nid and it ends: gal te tshig ran bzin med pa nid kyis dros po ran bain med. 134 [20]

Page #21
Though this block-print was at first provisionally identified as an old dGa' ldan print (dga' Idan par rnin), it would now appear to be another edition. The print of the Lam rim chen mo in the old dGa' ldan edition which I have inspected indeed has seven lines per side (characteristic of many 15th-century editions), but it has no margin notation, and the blocks (as well as individual characters) are somewhat smaller, being typically in the range: 46 x 5.8 to 47 x 6.0 cm in size. The edition found in Spiti differs from the better known and more recent editions in that the latter have six lines per side, whereas it is similar to them in that all of their margins state: 'pa lam rim che ba', i.e. vol. 13 (pa) [of Tson kha pa's collected works), namely the Lam rim chen mo. In the bkra sis lhun po (?) edition (see Yoshimizu's Naritasan catalogue, p. 167, no. 1397), the work consists of pa, fols. 1-491a. In the Lhasa Zol (?) edition (Yoshimizu, p. 209, no. 1564), the work has the identical margin notation, though it consists of pa, fol. 1-523a. I have checked these, as well as the obviously different Peking edition, and found them all to contain a different text on folio 322. Somebody more deeply involved in the study of the Lam rim chen mo should systematically describe all its known editions! REFERENCES Datta, C.L. (1973) Ladakh and Western Himalayan Politics 1819-1848. New Delhi. Eimer, H. (1991) Einige Bemerkungen zu Handschriftenfunden aus Guge/Westtibet, Zentralasiatische Studien, 22, 1989-1991, pp. 244-55. Eimer, H. (1992) Zum Stand der Kanjur-Forschung. In Ein Jahrzehnt Studien zur Uberlieferung des tibetischen Kanjur. Wien. Francke, A.H. (1914) Antiquities of Indian Tibet. Part I: Personal Narrative. Calcutta. Francke, A.H. (1977) (1907) A History of Ladakh. With Critical Introductions and Annotations by S.S. Gergan and F.M. Hassnain. New Delhi. Hahn, M. (1988) Bemerkungen zu zwei Texten aus dem Phudrag-Kanjur. In H. Eimer, ed., Indology and Tibetology, pp. 53-80. Bonn. Imaeda, Y. (1981) Histoire du cycle de la naissance et de la mort. Paris. Klimburg-Salter, D.E. (1982) The Silk Route and the Diamond Path. Esoteric Buddhist Art on the Trans Himalayan Trade Routes. UCLA Art Council, Los Angeles. Klimburg-Salter, D.E. (1987) Reformation and Renaissance: A Study of Indo-Tibetan Monasteries in the Eleventh Century. Orientalia Josephi Tucci Memoriae Dicata, ed. G. Gnoli & L. Lanciotti, SOR, LXI 2, pp. 683-702. Roma. [21] 135

Page #22
Klimburg-Salter, D.E. (1990) Tucci Himalayan Archives Report, 1: The 1989 Expedition to the Western Himalayas, and a Retrospective View of the 1933 Tucci Expedition. EW, 40, pp. 145-71. Snellgrove, D.L. (1957) Buddhist Himalaya. Travels and Studies in Quest of the Origins and Nature of Tibetan Religion. Oxford. Snellgrove, D.L. & T. Skorupski (1980) The Cultural Heritage of Ladakh. Vol. 2. Warminster. Taube, M. (1980) Die Tibetica der Berliner Turfan-Sammlung. Berlin. Tucci, G. (1933) Indo-Tibetica II. Rin c'en bzan po e la rinascita del Buddhismo nel Tibet intorno al Mille. Roma. Tucci, G. (1935) Indo-Tibetica III. I templi del Tibet Occidentale e il loro simbolismo artistico. Parte I. Spiti e Kunavar. Roma. 136 [22]