Articles by alphabetic order
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
 Ā Ī Ñ Ś Ū Ö Ō
1 2 3 4 5 6 7 8 9 0


A brief survey of the Tibetan translation of the Vinayasutra in the 'Dul ba'i mdo'i 'grel pa

From Chinese Buddhist Encyclopedia
Jump to navigation Jump to search



In the light of the Sanskrit manuscripts of the Vinayasutra Luo Hong February 23th, 2011

This is an improved version of an article published under the same title in Buddhist and Pali Studies in Honour of the Venerable Pro¬fessor Kakkapalliye Anuruddha (ed. Dhammajoti) Hong Kong, Centre of Buddhist Studies, The University of Hong Kong, 2009, 299-342.


Abstract


A comparison of the Tibetan translation of the Vinayasutra with the pratikas attested in all the four canonical commentaries in the light of the Sanskrit manuscripts by far known to us shows that there is a constant and consistent divergence between the Tibetan translation of the Vinaya- sutra, the pratikas in the Tibetan translation of the Vinayasutravrttyabhidhanasvavyakhyana and those in the Tibetan translation of the Vinaya- sutratika as one group (G 1)

and the pratikas in the Tibetan text of the Vinayasutravyakhyana and those in the Tibetan text of the Vinayasutravrtti as another group (G 2). The divergence between the two groups sometimes points to the discrepancy between the Sanskrit originals, however, in most cases, it just reflects the translators' different treat¬ment of the identical Sanskrit original. In this art¬icle, we confine ourself to the investigation of the different translations of the same Sanskrit original. The main aim here is to evaluate the textual mer¬its of the pratikas in the Tibetan text of the Vina- yasutravrtti. Judging from the stylistic peculiarit¬ies and the archaism attested, we are convinced that the pratikas embedded in this text could be dated back to an earlier period than that in which the translation in G1 was prepared. The transmission history of the Vinayasutravrtti has also been briefly discussed, the fact here detected and discussedmay shed new light on the relationship between Sar- vastivada and Mulasarvastivada.

A brief survey of the Tibetan translation of the Vinayasutra in the 'Did
2 ba'i mdo'i 'grel pa



Contents



1 Different translations of the Vinayasütra 10
2 Evaluation by Tibetan scholars 12
3 Divergences between G1 and G2 20
3.1 Concerning terminology 21
3.2 Concerning case-ending 25
3.3 Concerning conjugation 27
3.4 Concerning compound 30
3.5 Concerning syntax 32
4 Stylistic peculiarities of the translation in the 'Dul
ba'i mdo'i 'grel pa 35

4.1 VSVTib's faithful adherence to original 35
4.1.1 Retaining word order 36
4.1.2 No interpreting additional added . . . . 36
4.1.3 Mechanical translation of wordings . . 37
4.1.4 Expressive depiction of the taddhita
suffix 38


4.2 VSVTib's conciseness 38
4.2.1 Preferring monosyllabic wording . . . . 38
4.2.2 Ignoring upasarga 40
4.2.3 Ignoring tva and tal 43
4.2.4 Ignoring plurality 44
5 Archaism in the 'Dul ba'i mdo'i 'grel pa 47

5.1 Archaic expressions 47
5.1.1 VS-1.71 ... 'o cog 47
5.1.2 VS-1.91 sngon cad 47
5.1.3 VS-1.42 ltag ma 48
5.1.4 VS-1.41 spya nga brim 48
5.1.5 VS-1.120 rmed pa 49
5.2 Commonness sharing with P.T. 903 50

5.2.1 Indicative terminology 50 
5.2.2 Monosyllabic words 50
5.2.3 Plurality 51
5.3 Migration of equivalents 52
5.3.1 From VSVTib to G1 53
 

 
6 The transmission history of the 'Dul ba'i mdo'i 'grel pa 57
7 Conclusions
A brief survey of the Tibetan translation of the Vinayasutra in the 'Did

4 ba'i mdo'i 'grel pa
Abbreviations and Sigla

VS
The Vinayasutra.

VSMSA
Photocopy of a Sanskrit manuscript of VS deposited in China Tibetology Research Centre.
VSMSB

Microfilm of a Sanskrit manuscript of the VinayasUtra deposited in Research Institute of Sanskrit Manuscripts & Buddhist Literature at Peking University.
VSRS

Rahula Sankrtyayana ed., Vinayasütra of Bhadanta Gunaprabha. Singhi Jain Sastra Siksapitha. Singhi Jain Series 74, Bombay 1981.
VST
Transliteration of VSMSB by the Study Group at Taisho University.
VSS

The VinayasUtmvrttyabhidhanasvavyakhyana.

VSSMSA

The Sanskrit manuscript
VinayasUtmvrttyabhidhanasvavyakhyana
Bengali script.

VSSMSB

The Sanskrit manuscript
Vinayasutravrttyabhidhànasvavyàkhyàna

dBu med script.
VSSBG
Eds. P.V. Bapat and V.V.Gokhale, and Auto-Commentary on the Same by Gunaprabha. Chapter I-Pravrajya-vastu, Compared with the Tibetan Version. K.P.Jayaswal Research Institute, Patna: 1982.
VSST

Edition of VSSMSB prepared by the Study Group at Taisho University.
VSTib

'Dul ba'i mdo.
[D No.4117 wu 1v1-100r7; P No.5619 zu 1-109v8]
VSSTib

'Dul ba mdo'i 'grel pa mngon par brjod pa rang gi rnam par bshad pa.
[D No.4119 zhu 1v1-zu 274r7; P No.5621 'u 1-yu 342r8]
VSTTib
'Dul ba'i mdo'i rgya cher 'grel pa.
[D No.4120 'u 1v1-yu 388r7; P No.5622 ru 1-lu 464r8]
VSVyTib

'Dul ba'i mdo'i rnam par bshad pa.
[D No.4121 ru 1v1-263r7; P No.5623 shu 1-314r6]
VSVTib

'Dul ba'i mdo'i 'grel pa.
[D No.4122 lu 1v1-344r7; P No.5624 su 1-429r8]
G1
VSTib, VSSTib and VSTTib.

G2
VSVyTib and VSVTib.
 
A brief survey of the Tibetan translation of the Vinayasutra in the 'Did
6 ba'i mdo'i 'grel pa
Nyi 'od
mTsho sna ba Shes rab bzang po, 'Dul ba mdo rtsa'i rnam bshad nyi ma'i 'od zed legs bshad lung gi rgya mt- sho. (Beijing 1998)
Rab gsal
Bu ston, 'Dul ba mdo'i rnam par 'byed pa 'dul ba rgya mtsho'i snying po rab tu gsal bar byed pa. (Bu ston thams cad mkhyen pa'i bka' 'bum zha pa, lHa sa 1917)

Rin 'phreng
dGe 'dun grub pa, Legs par gsungs pa'i dam pa'i chos 'dul ba mtha' dag gi snying po'i don legs par bshad pa rin po che'i 'phreng ba. (Beijing 1999)
mGul rgyan

'Jam dbyangs bzhad pa ngag dbang brtson 'grus, 'Dul ba'i dka' gnas rnam par dpyad pa 'khrul spong blo gsal mgul rgyan cinda ma ni’iphreng mdzes skal bzang re ba kun skong. (sKu 'bum byams pa gling)
sPyi rnam

Bu ston, ’Du
lba spyi’irnam pargzhag pa ’dulba rin po che’i mdzes rgyan. (Bu ston thams cad mkhyen pa'i bka' 'bum zha pa, lHa sa 1917)
Dag
Dag yig gsarbsgrigs. Ed. Dag yig 'di'i tshomsgrig tshan chung, (Xining, 1979)
brDa

brDa dkrol gser kyi me long. Ed. bTsan lha ngag dbang tshul krims, (Beijing,1996)
BG

Bod rgya tshig mdzod chen mo. Ed. Zhang yisun et al, (Beijing, 1998) 1Vol.
Editorial signs
< >
contain aksara(s) added


contain aksaras or danda deleted


virama
0
virama lacking
avagraha


danda / shad
||
double danda / nyis shad


stringhole

A brief survey of the Tibetan translation of the Vinayasutra in the ’Dul ba'i mdo'i 'grel pa

The present paper is part of the first steps towards preparing a first critical edition of the Vinayasutra (henceforth VS). We are, from a point of view of textual criticism, aiming at evaluating the merits of the translation of this text as embedded in the form ofpratikas in the ’Dul ba’i mdo’i ’grelpa (henceforth VSVTib).

This evaluation is derived, on the one hand, from a com¬parison with the widely circulated version of the transla¬tion ofVS, as preserved in the ’Dul ba’i mdo’i ’grel pa mngon par brjod pa rang gi rnam par bshad pa (henceforth VSSTib) and the ’Dul ba’i mdo’i rgya cher ’grelpa (henceforth VSTTib) and one less known version preserved in the ’Dul ba’i mdo’i rnam par bshad pa (henceforth VSVyTib).

On the other hand, the evaluation is based upon the examin¬ation of the Sanskrit original of the first chapter of the VS, the Pravrajyavastu. The original is now available in two co¬dices: VSMSB which was brought to light by Rahula Sankrtya- yana, VSMSA which was recently identified by the present au¬thor in the library of China Tibetology Research Center.

A Chart is adopted for the sake of comparing the transla¬tion in VSVTib with Sanskrit texts and the Tibetan transla¬tions in VSTib and other canonical commentaries:


VS-1.***

VSMSA Transliteration of the sutra(s) in VSMSA
VSMSB Transliteration of the sutra(s) in VSMSB
VSSMSA Transliteration of the pratika(s) in VSSMSA
VSTib Translation of the sutra(s) in VSTib
VSSTib Translation of the sutra(s) in VSSTib
VSTTib Translation of the sutra(s) in VSTTib
VSVyTib Translation of the sutra(s) in VSVyTib
VSVTib Translation of the sutra(s) in VSVTib



Note


The number of the sutra is editorial. The pratika(s) in VSSMSA as given in VSSBG are provisionally consulted, they need to be reexamined. The pratika(s) in VSSMSB are for the time being not included. As for the Tibetan translations, Derge (D) and Peking (P) were compared. In most cases, they agree with each other, in the case of merely

orthographic divergences, the text of D is given. A blank in chart indicates the absence of the sutra(s) concerned in correspondent text.
A brief survey of the Tibetan translation of the Vinayasutra in the Dul 10 ba'i mdo'i 'grel pa


1 Different translations of the Vinaya- sütra


Already, the first sutra is a good example:
VS-1.1

VSMSA atha niryânavrttam || (VSMSAIM)
VSMSB | atha niryânavrttam | (VSMSBIM)
VSSMSA atha niryânavrttam || (VSSBGp.3)
VSTib nges par 'byung ba'i tshul khrims kyi
dbang du byas te ||

VSSTib de nas nges par 'byung ba'i tshul khrims kyi dbang du byas te ||
VSTTib nges par 'byung ba'i tshul khrims kyi
dbang du byas te ||
VSVyTib ngesparthegpa'itshuldugtogs paste
|
VSVTib gtogs pa nges par theg pa'i tshul te |


Note

Obviously, the translation in VSVTib is a completely ver¬batim repetition of the Sanskrit in Tibetan: gtogs pa as equi¬valent of atha and nges pa theg pa’i tshul, niryanavrt- tam. Such literal translation practice characteristically dis¬tinguishes VSVTib from the other commentaries. Transla¬tion in VSVyTib shows the same equivalents except for the re¬versed word order.


The dramatic divergences here attested between VSTib, VSSTib and VSTTib on the one hand (henceforth G1), and VSVyTib, VSVTib (henceforth G 2) can hardly be understood without consulting the Sanskrit text and the commentaries in which the translations are embedded.
The divergences need to be explained, for the only other oc-currence of atha in VS-1.448 confirms that this is by no means accidental. The translators of VSVyTib and VSVTib, it seems,


11
abide by certain standards, like their colleagues.
VS-1.448

VSMSA atha nisritapratipat || (VSMsAl0a5)
VSMSB | atha nihsritapratipat* | (VSMsB6a3)
VSSMSA atha nihsrita-pratipat || (VSSBGp.43)
VSTib gnas pas sgrub pa'i dbang du byas te
de nas de ma thag tu sngar song ste phyag bya'o ||
VSSTib gnas pas sgrub pa'i dbang du byas te
||
VSTTib gnas pas sgrub pa'i dbang du bya ste
||
VSVyTib gnas pa'i sgrub par gtogs pa'o ||
VSVTib gtogs pa na gnas pas nan tan du bya'o
||


Note

As in the case of VS-1.1, VSTib, VSSTib and VSTTib share the same translation of the sutra under consideration while VSVyTib and VSVTib, almost the same wordings which are ar¬ranged in divergent word order.


A brief survey of the Tibetan translation of the Vinayasutra in the 'Did 12 ba'i mdo'i 'grel pa


Have such divergences ever been fully recognized and dis-cussed? Not yet by modern scholars, but certainly in the ex-egeses penned by indigenous Tibetan commentators, e.g. by mTsho sna ba Shes rab bzang po (13th cent. C.E.), questioning his ’Dul ba mdo rtsa’i rnam bshad nyi ma’i ’od zed legs bshad lung gi rgya mtsho (hencforth Nyi 'od):

khrus ras bcang bar bya'o zhes pa la sogs te skra bzed bcang bar bya'o 'ur rdo bcang bar bya'o || zhes | bcang bar bya'o zhes 'dul ba nas lan mang du gsungs na | de thams cad ci nas kyang 'chang dgos ma bcangs na nyes par 'gyur ba sgrub pa'i bslab bya yin nam | 'od te ci bcangs kyang chog ma bcangs kyang

chog dgar 'jug pa gnang ba'i bslab pa yin zhe na bcang bar bya'o zhes gsungs pa thams cad spyir gtan la dbab pa ni | yo byad drug la sogs pa dgag dgos kyi rgyu mtshan dang ldan pa rnams ni ci nas kyang 'chang dgos pa sgrub pa'i bslab pa yin la | dgag dgos kyi rgyu mtshan dang mi ldan khrus ras bcang bar bya'o 'ur rdo bcang bar bya'o zhes pa la sogs pa | pa rnams ni bcang bar bya'o zhes gsungs kyang bcang du gnang ba yin te | TT ka las | khrus ras bcang bar bya’o

zhes bya ba ’di brjod de | byed pa’i tshig dang por bstan pa’i phyir rang dgar spyad pa yin par rig par bya’o zhes dang | 'grel chung las kyang | khrus ras bcangs kyang nyes medces bshad byed pa'i tshig dang por bstan pa'i phyir | zhes pa ni rgya dpe la bcang bar bya'o khrus ras zhes pa ste | de ni sgra log 'gyur du bsgyur ba'o || 'grel chung gi mdo rtsa ba las chongs shig khrus ras zhes sgra thad 'gyur du bsgyur snang ngo | do || 'o na bcang du gnang ngo zhes ma gsungs par bcang bar bya'o zhes sgrubtshigtusmos pa'i rgyu mtshan gang zhe na 'dod pa chung zhing chog shes pa la bcom ldan 'das kyis rnam grangs du mas bstod cing | de las


ldog pa la rnam grangs du mas smad pa dang | bcang bar bya'o zhes brjod na | yul dus gang zag 'ga' zhig la gnang yang thams cad la gnang ba ma yin nam snyam pa'i dvogs pa bsal ba'i phyir sgrub bcang bar bya'o zhes | tshig tu bstan to || (Nyi 'od, p.170)
In this passage, the exegete examines the divergent translations of VS-1.194 and presents his judgments:
VS-1.194

VSMSA dharayeta snatrasatakam ||
(VSMsA6a1)
VSMSB | dharayeta snatrasatakam |
(VSMsB3b6)
VSSMSA dharayet snatrasatakam || (VSSBGp.30)
VSTib khrus ras bcang bar bya'o ||
VSSTib khrus ras bcang bar bya'o ||
VSTTib khrus ras bcang bar bya'o ||
VSVyTib khrus ras bcang |
VSVTib


Note chong shig khrus ras |

By comparing the two translations in the light of the Sanskrit text (rgya dpe), he characterizes the one presented in G1 and VSVyTib as ‘translation with the [original] word-order be¬ing reversed (sgra log 'gyur du bsgyur ba)', while that at¬tested in VSVTib, ‘word-by-word translation (sgra thad 'gyur du bsgyur)'.

Moreover, he indicates two other sutras as points of reference, both of which we have identified:
A brief survey of the Tibetan translation of the Vinayasutra in the 'Did
14 ba'i mdo'i 'grel pa
VS-1.172

VSMSA dharayet kesapratigrahanam ||
(VSMsA5b3)
VSMSB | dharayet kesapratigrahanam |
(VSMsB3b3-4)
VSSMSA dharayet kesapratigrahanam ||
(VSSBGp.28)
VSTib skra bzed bcang par bya'o ||
VSSTib skra bzed bcang par bya'o ||
VSTTib skra bzed bcang par bya'o ||
VSVyTib sgra bzed chongs shig |
VSVTib 'chong shig skra bzed pa |


Note


VS-2.319
VSMSA dharayet ksampanam || (VSMsA20a1)
VSMSB | dharayet ksampanam | (VSMsBlIal)
VSSmsA
VSTib ur rdo bcang bar bya'o ||
VSSTib ur rdo bcang bar bya'o ||
VSTTib ur rdo bcang bar bya'o ||
VSVyTib 'ur rdo bcang |
VSVTib chongs shig 'ur rdo |


Note


Admittedly, the remarks here are safe and sound for al¬though it is something of a truism that during the course of the transmission of a certain text, variant readings are in¬evitably doomed turning up, in this particular case, how¬ever, no such discrepancy occurs, therefore, we feel some¬what relieved on behalf of our author that what has been dis¬cussed here is still of validity.

On the same topic remarks are also given by Bu ston in 'Dul ba spyi'i rnam par gzhag pa 'dul ba rin po che'i mdzes rgyan (henceforth sPyi rnam), a fact reflects that in special cases the divergent translations had drawn much the attention of indigen-


ous scholars and bifurcated views came about as a result:


gnang ba'i bslab pa ni | rgya cher 'grel pa las | gnang ba zhes bya ba ni | ci dgar spyod pa'i yul la bya ba gang yang rung ba zhi mi byed dam | byed na ltung ba zhes bya ba'i nyes pa med pa ste | zhes bshad pas | byas kyang nyes pa med | ma byas ky- ang nyes pa med pas ci dgar byed du gnang ba ste | khrus ras bcang bar bya'o || zhes pa lta bu ste | rgya cher 'grel du | byed pa'i tshig dang por bstan pa'i phyir rang dgar spyad ba yin par rig par bya'o || zhes gsungs te | de'i don | bya'o khrus ras bcang par | zhes rgya dpe'i skad dod la yod pas | bya'o zhes pa'i byed pa'i tshig dang por bstan pa thams cad | ci dgar gnang ba'i brda chad min kyang | bod kyi 'gyur la de'i brda chad cig snang ngo || (sPyi rnam, 7a3-6)


The doubt voiced here by Bu ston concerning the overgeneraliz-ation of the symbolizing function of the verb arranged at the be-ginning of one sutra is quiet reasonable since we find no men-tion of alternativeness thus suggested in the canonical com-mentaries on the two sutras further adduced in Nyi 'od.
Naturally, there are other divergences among the translations, e.g., different treatments of compounds pointed out in Nyi 'od, as on VS-1.42:
A brief survey of the Tibetan translation of the Vinayasutra in the Dul ba'i mdo'i 'grel pa

VS-1.42

VSMSA supatram ity anevamtvam bruyuh |
VSMSB sarvve || (VSMsA2a6)

| supatram ity anevam bruyuh sarve |
VSSMSA (VSMsB2a1)
supatramiti anevamtve bruyuh sarve ||
VSTib (VSSBGp.11)

thams cad kyis de lta bu nyid ma yin te
VSSTib lhung bzed bzang ngo zhes brjod par bya'o ||
thams cad kyis de lta bu nyid ma yin te


VSTTib lhung bzed bzang ngo zhes brjod par bya'o ||
thams cad kyis de lta bu nyid ma yin

VSVyTib te | lhung bzed bzang ngo zhes brjod par bya ba ||
de lta ma yin na lhung bzed bzang ngo
VSVTib zhes thams cad kyis brjod || de lta ma yin na thams cad kyis lhung


Note bzed bzang ngo zhes smros shig |

rang gi rnam bshad dang TT ka las de Ita bu nyid ma yin te zhes bshad do || 'drel chung las ni de lta bu nyid ma yin na zhes bshad de | ma yin na zhes pas skyon dang bral ba dang yon tan dang ldan pa gang yang go bar mi nus la | de lta bu nyid chung che skya ba | ma yin te zhes pas chung che skya ba'i | skyon dang bral ba dngos su bstan te tshad ldan dang kha dog bzang ba'i | yon tan dang ldan pa shugs la bstan pas sgra bzang po yin zhes gsung ngo || de lta bu nyid ma yin te | zhes pa sgra bzang shos 'di mdo rtsa ba'i gzhung du bkod tshod kyi dbang du byas pa'o || lag len byed tshod kyi dbang du byas pa ni | lung gzhi dang karma sa tam las de lta bu nyid ma yin na re re nas lhung bzed bzang ngo zhes brjod par bya'o zhes gsungs la | TTka las kyang yon tan dang


ldan pa yin na bzang ngo | zhes brjod par | bshad do ||
(Nyi 'od, pp.92-3)

Unlike the aforementioned case of VS-1.194 where Sanskrit text (rgya dpe) is clearly referred to, the subjects of the eval¬uation and remark here, however, are somewhat ambigu¬ous, because we do not know for sure whether our au¬thor aimed to evaluate just divergent Tibetan translations or di¬vergent Sanskrit originals behind, or both. In other words, we do not know whether he had consulted Sanskrit texts or not. It is highly possible he had not done so, since we find in extant Sanskrti manuscripts divergent readings: an- evamtvam, avevam and anevamtve, except for anevam, the rest two correpond well to de lta bu nyid ma yin te and de lta bu nyid ma yin na repectively, and this had not been made mention of. So that it seems that he had focused on evaluating

divergent translations that that attested in VSSTib and VSTTib is better than the one preserved in VSVTib be¬cause the later cannot make clear the status of the alms- bowl's being free from the deficiencies and being in posses¬sion of the good qualities, which, however, has been success¬fully conveyed by the translation in VSSTib and VSTTib either directly or indirectly. It is further remarked that the former is rendered from the point of view of establishing

stand¬ards in the root-text of VS, while the later, out of the consider¬ation of conforming to the need of the enforcement of the rule thus given. It needs to be confirmed that whether such intrinsic discrepancyis of an original nature or due to later development.
Besides such frank criticism as in the case of VS-1.42, we also meet with compromising efforts in later Tibetan scholars with the intention to heal the divergences attested among the trans-lations. This may be more often the case as long as the discrep-ancies are not incurable. The remark on the translations ofVS- 1.88 in Nyi 'od is just a case in point:


A brief survey of the Tibetan translation of the Vinayasutra in the 'Didba’i mdo’i ’grel pa
VS-1.88

VSMSA sraddhasilasrutatyagaprajnasampannatvam

VSMSB || (VSMsA3a5-6)

| sraddhasilasrutatyagaprajnasampan-

VSSMSA natvam | (VSMSB2M)


sraddha-sTla-sruta-tyaga-prajna-
VSTib sampannatvam || (VSSBGp.16) dad pa dang tshul khrims dang thos
VSSTib pa dang gtong ba dang shes rab phun sum tshogs pa nyid do ||
dad pa dang tshul khrims dang thos
VSTTib pa dang gtong ba dang shes rab phun sum tshogs pa nyid do ||
dad pa dang tshul khrims dang thos
VSVyTib pa dang gtong ba dang | shes rab phun sum tshogs pa nyid do ||
dad pa tshul khrims thos pa | gtong ba
VSVTib | shes pa rnams dang ldan pa'o || dad pa dang tshul khrims dang thos


Note pa dang gtong ba dang shes rab dang ldan pa yang ngo ||


lnga tshan 'di dang 'og ma ste lnga tshan 'di gnyis la phun sum tshogs pa gnyis zer ba'i rgyu mtshan ci zhe na | lnga tshan gnyis po chos can | phun sum tshogs pa'i lnga tshan yin te | dad pa la sogs pa'i yon tan tsam po dang ldan pa ma yin gyi yon tan phun sum tshogs pa'i chos lnga ldan yin pa'i phyir ro || dper na slob pa'i chos lnga ldan ma yin gyi mi slob pa'i chos lnga ldan yin pa de la mi slob pa'i lnga phrugs zhes brjod pa dang 'dra'o || (Nyi 'od, p.120)

And the same topic becomes the main concern of correspond¬ent passage of in dGe 'dun grub pa's Legs par gsungs pa'i dam pa'i chos 'dul ba mtha' dag gi snying po'i don legs par bshad
“”A brief survey””


pa rin po che’i ’phreng ba (henceforth Rin 'phreng), which be-trays a constant focus of attention on the divergent translations between G1 and G2 in the circle of indigenous scholars:
gsum pa phun sum tshogs gnyis kyi yon tan la gnyis las dang po ni | dkon mchog gsum la dang ba'i dad pa phun sum tshogs pa nyid dang | rang bzhin dang bcas pa'i kha na ma tho ba spong ba'i tshul khrims phun sum tshogs pa nyid dang | 'dul bashes pa'ithos pa phun sum tshogs pa nyid dang | sems byung ma chags pa'i ngo bo nyid kyi gtong ba phun sum tshogs pa nyid dang | chos rab tu rnam 'byed kyi shes rab phun sum tshogs pa nyid do | phun sum tshogs pa ni ldan pa yin la ldan pa la phun sum tshogs pa'i sgras ston pani de dag khyad pa can dang ldan pa ston pa yin te | lta ba phug sum tshogs pa bzhin no || (Rin 'phreng, p.40)

The explanation cannot be accepted without preservation since we are not informed by extant Sanskrit texts such divergence.
Now there is no need to doubt either the existence of divergent Tibetan translations of VS which can be roughly reduced into two groups or the recognition of their existence among Tibetan scholars. We should, however, bear in mind that one condi¬tion which is the very basis of their discussions had been pre-supposed by them: an assumed fact, that the divergence re¬flect mainly different treatment of the same, or almost the same Sanskrit original readings of the sutras which are embedded in different pustakas. The presupposition, however, cannot hold good universally, as we have discussed in the cae of VS-1.42, therefore evaluation on divergences attested among Tibetan translations should be carried out in the light of Sanskrit texts.
A brief survey of the Tibetan translation of the Vinayasutra in the 'Did


20 ba'i mdo'i 'grel pa

3 Divergences between G1 and G2


The divergences between translations preserved in G1 and G2 ranging from terminology to syntactic construction are mul¬tiple as well as profound. With the help of a newly iden¬tified Sanksrit manuscript of VS: VSMSA, which sheds a flash of light on the divergent readings in Sanskrit origin¬als, we find that in most cases, the divergences attested in ca¬nonical Tibetan texts can be defined as different treatments of identical Sanskrit original.
Doubtlessly, a systematical comparison of VSMSA with VSMSB would make possibleless conjectural our analysis and more re-liable our conclusion with regard to the discrepancies attested in canonical commentaries. For the sake of making more or less solid the basis of our comparison, one principle is adop¬ted in selecting the examples that only those where VSmsA and VSmsb agree with each other in the respects concerned are to be adduced. 
21


3.1 Concerning terminology


VS-1.638

VSMSA bhavati pitrvat* pitrasayatve nujnayo

VSMSB raja || (VSMsAl3b4-5)

| pitrvat pitrasayatve nujnayam raja |
VSSMSA (VSMsB7b6)

pitrvat pitrasayatve 'nujnayam raja ||
VSTib (VSSBGp.58)

rgyal po pha ma'i bsam pa can nyid
VSSTib kyis gnang ba ni pha ma dang 'dra'o || rgyal po pha ma'i bsam pa can nyid
VSTTib kyis gnang ba ni pha ma dang 'dra ba'o ||
rgyal po pha ma'i bsam pa can nyid
VSVyTib kyis gnang ba ni pha ma dang 'dra'o || pharbsams pa'i rgyal pos gnang ba ni
VSVTib pha bzhin no |
pha bzhin du phar bsam pa'i rgyal pos


Note gnang na yang 'gyur ro |

The equivalents in G2 of pitrvat and pitmsayatva are of par-ticular interests in that they reflect an inclination towards lit-erally rendering, and it is easy to see that their counter¬part in G1 is also grammatically supported. And we find ana¬logue example which can serve to inform us more about the technique utilized in VSVTib:

A brief survey of the Tibetan translation of the Vinayasutra in the 'Did
22 ba'i mdo'i 'grel pa

VS-1.120

VSMSA jivakapitrkaQm anujnatam tabhyam aduradesam pravrajyapeksam saptahan dharayet* || (VSMsA4a2)

VSMSB | jivatpitrkam ananujnatam* | tabhyam

aduradesam pravrajyapeksa saptaham dharayet* | (VSMsB2b5-6)

VSSMSA jivatpitrkam ananujnatam tab-


hyam aduradesam pravrajyapeksam saptaham dharayet || (VSSBGp.21)
VSTib rab tu 'byung bar 'dod pa pha ma gson
zhing de gnyis kyis ma gnang ba yul thag mi ring ba ni zhag bdun pa bsrel bar bya'o ||
VSSTib rab tu 'byung bar 'dod pa pha ma gson
zhing de gnyis kyis ma gnang ba yul thag mi ring ba ni zhag bdun bsral bar bya'o ||
VSTTib rab tu dbyung bar 'dod pa pha ma gson

zhing de gnyis kyis ma gnang ba yul thag mi ring ba ni | zhag bdun bsrel bar bya'o ||
VSVyTib rab tu 'byung bar 'dod pa'i pha ma
'tsho la de gnyis kyis kyang mi gtong zhing yul yang nyen zhag bdun gyi bar du gzhag |
VSVTib pha 'tsho la ma gnang ste de dag gi yul
mi ring na rab tu 'byung bar 'dod pa de zhag bdun du 'chol shig |


Note

Single reading for jivakapitrkam / jivatpitrkam'. pha 'tsho la is presented in VSVTib, in G1 is attested pha ma gson zhing, VSVyTib gives pha ma ’tsho la, we need not to feel doubt that the unfamiliarity with grammatical prescrip¬tion on the part of the translators of VSVTib might have in¬volved since in the following interpretation pha ’tsho la is paraphrased as pha ma gnyis ka'ang ma shi ste yod la.

Furthermore, judging from an extra note given in VSTnb, the optional elliptical form might even be more comprehensive so as to include the possible situation when only one of the parents is still alive:

de gnyis kyis zhes bya ba gnyis kyi tshig tu byas pa'i phyir pha ma gson zhing zhes bya ba 'dir yang pha ma gnyis gzung bar bya'o || gal te gcig gson na yang de'i tshe de kho na bzhin du bsgrub par bya'o ||

We would like to adduce one more instance where different treatment of terminology is attested: 


A brief survey of the Tibetan translation of the Vinayasutra in the 'Did
24 ba'i mdo'i 'grel pa


VS-1.105

VSMSA krt etat

VSMSB tirthyanaradhitaQditirthyantavarjyam

|| (VSMsA3b4)

| krt etat tirthyanaradhitaditirthyantam

VSSMSA varjam* | (VSMsB2b3)

krt etad tirthya-anaradhitadi-
VSTib tirthyanta-varja[na]m || (VSSBGp.

19)
bkol ba de ni mu stegs can dang sha
VSSTib kya la sogs pa'i 'jug gi mu stegs can bkol ba'o ||
bkol ba de ni mu stegs can dang
VSTTib shakya la sogs pa'i mjug gi mu stegs can bkol ba'o ||
bkol ba de ni mu stegs can dang

VSVyTib shakya la sogs pa'i mjug gi mu stegs can bkol ba la'i ||
de dag la mu stegs can dang | sakya la
VSVTib sogs pa mu stegs can zhes pa'i bar du ma gtogs pa khrid |
krid ces de dag la mur 'dug dang ma



Note rangs pa la sogs pa mur 'dug gi bar du ma gtogs |


krt is literally rendered in G1 as bkol, while in G2, transcribed into slightly different forms. That which is given in VSVTib: krid seems to be a better choice than khrid in VSVyTib. It is worth noting that according to Nyi 'od the transcription attested in VSVTib: krid had given rise to misunderstanding ofits being a Tibetan wording and subsequently leading to further division of the terminology of ’kol ba into two categories, i.e., bzhag bkol and bkri bkol, which is, as noted in the memorial verse de¬voted to summarizing the critical remarks, somewhat confus¬ing (cung zad ’khrul):
bzhi pa bkol ba'i sgra bshad pa ni | rab tu 'byung


ba'iphyirnyebar'ongspala|mdosngamalas mustegs
can zhes pa de logs shig tu | bkol zhing | sakya dang me
ba ral ba can zhes | mdo phyi ma las sakya dang me
ba ral ba can dang sems mgu bar ma gyur pa'i mu
stegs can dang gsum bkol te | mu stegs can la sogs pa bzhi po de dag logs shig tu bkol ba'i lhag ma
nyid yin pas na bkol ba zhes bya ste | TT ka las |

mu stegs can zhes bya ba'i sgra de dag bkol ba'i
lhag ma ni bkol ba zhes bya zhes bzhag bkol dang bkri
bkol zhes 'dir sgra bshad gnyis 'chad pa ni mi rigs te | bkol te
bzhag pa bzhi po la | bkol ba'i sgra bshad ma 'jug ste || bkol
ba'i ming can ma yin phyir | ming can ma yin 'grel pas grub

| bkol ba'i ming can drug po la | bkri bkol sgra bshad cung
zad 'khrul | krid ces rgya skad song bzhag la | bod skad yin
par bshad phyir ro | zhes smras so || krid ces 'grel chung las
bkol ba'i sgra don du byung ste | krid ces pa rgya skad yin la
| bkolbazhes pabod skadyin no || so || (Nyi 'od, p.146)


Now it is rather clear that when special terms are involved, G2 tends to give literal translation, while in G1 it is normal they are free translated, but this holds good only in general, for we have already an exceptional example where VSVTib alone per¬sists such practice: in VS-1.120.


3.2 Concerning case-ending

In general, case-endings in Sanskrit original find well their cor-respondents in Tibetan translation to which the translations of VS are not exceptional, but sometimes we come across example where VSVTib had adopted approach other than that utilized commonly in G1 and VSVyTib:
A brief survey of the Tibetan translation of the Vinayasutra in the ’Dul
26 ba'i mdo'i 'grel pa


VS-1.66

VSMSA vinïtasamvâsatâyâm || (VSMsA2b5)

VSMSB | vinïtasamvâsatâyâm | (VSMsB2a3)

VSSMSA vinïtasamvâsatâyâm || (VSSBGp.12)

VSTib dul bar gnas pa nyid dang ||
VSSTib dul bar gnas pa nyid dang ||
VSTTib dul bar gnas pa nyid dang ||
VSVyTib dul bar gnas pa yang ngo ||
VSVTib dul bar gnas par yang ngo ||


Note
The -r appended to dul bar gnas pa is in no doubt used to de-noting the Vaisayikasaptami in Sanskrit text, and it should be noted, only in VSVTib does such a signal for case-ending in Sanskrit occur, all the other four remain silent about this ter-mination in their translations of the sutra.
VS-1.12


VSMSA upâdhyâyam yâceta || (VSmsA1b5)

VSMSB | upâdhyâyam yâcet* | (VSmsB1b3)

VSSMSA upâdhyâyam yâceta || (VSSBGp.7)

VSTib mkhan por gsol bar bya'o ||
VSSTib mkhan por gsol bar bya'o ||
VSTTib mkhan por gsol bar bya'o ||
VSVyTib mkhan por gsol |
VSVTib mkhan pos gsol cig |


Note

It seems that in two ways is akathitadvitiya been identified, in G1 and VSVyTib, as in sampradanam, and in VSVTib, karanam and accordingly it is depicted respectively in the form of -r and -s.



3.3 Concerning conjugation


VS-1.11

VSMSA sacet* parisuddhyati sarvve j ©

VSMSB bruyuh || (VSMsAlb5)

| sacet* parisuddhyatiti sarve bruyuh |
VSSMSA (VSMsBlb3)
sa cetparisuddha iti sarve bruyuh ||
VSTib (VSSBGp.7)
thams cad kyis gal te yongs su dag
VSSTib na'o zhes brjod par bya'o || thams cad kyis gal te yongs su dag
VSTTib na'o || brjod par bya'o ||
thams cad kyis gal te yongs su dag
VSVyTib na'o zhes brjod par bya'o || desteyongssugtsangna'ozhesthams
VSVTib cad kyis smros shig |
de ste kun gtsang na'o zhes thams cad


Note kyis smros shig |


The handling of lih Optative in G2 is characteristically diver¬gent from that in G1 in that participle shig instead of auxili¬ary verb bya plus rdzogs tshig is appended to skul tshig of the main verb.
Another policy G2 adopted in the same case is to put forth merely skul tshig:


A brief survey of the Tibetan translation of the Vinayasutra in the Dul
28 ba'i mdo'i 'grel pa
VS-1.17

VSMSA padayor nnipatya pratigrhnit* ||
(VSmsA1b6)
VSMSB | padayor nipatya pratigrhnit* | (VSmsb1b3)
VSSMSA padayor nipatya pratigrhnita ||
(VSSBGp.7)
VSTib rkang pa gnyis la gtugs te blang bar bya'o ||
VSSTib rkang pa gnyis la gtugs te blang bar bya'o ||
VSTTib rkang pa gnyis la gtugs te blang bar bya'o ||
VSVyTib rkang pa la phyag byas te blang |
VSVTib

Note rkang pa la phyag byas te blangs |
VS-1.526
VSMSA gamane vilambitam udikset ||
(VSmsA11b2)
VSMSB | gamane vilambitam udikset* | (VSmsB6b5)
VSSMSA gamane vilambitam udikseta ||
(VSSBGp.47)
VSTib 'gro ba'i tshe thogs na bsdad parbya'o ||
'gro ba'i tshe thogs na bsdad parbya'o
||
'gro ba'i tshe thogs na bsdad parbya'o ||
'dong ba na 'phyi ba bsdad ||
VSSTib
VSTTib
VSVyTib
VSVTib


Note 'gro ba'i tshe lus pa la bsdad ||


We are taught by other example that VSVyrib seems to be prone to use only skul tshig while VSVTib persists in adopting shig or its variants:
29

VS-1.4
VSMSA prstväntaräyikam parisuddhäya

VSMSB purvvopädhyäyatvenävakäsam kur- yat* || (VSmsA1b2)

| prstväntaräyikam parisuddhäya pur-
VSSMSA vopadhyayatvenavakasam kuryat* | (VSmsb1b1-2)

prstvä äntaräyikam parisuddhäya
VSTib purvo[pädhyäyatvenä-vakä]sam kuryät | (VSSBGpp.5-6) dang po'i mkhan po nyid kyis bar
VSSTib chad dris nas yongs su dag pa la skabs dbye'o ||
dang po'i mkhan po nyid kyis bar
VSTTib chad dris nas yongs su dag pa la skabs dbye'o ||
dang po'i mkhan po nyid kyis bar
VSVyTib chad dris nas yongs su dag pa la skabs dbye'o ||
bar chad dris la yongs su gtsang na
VSVTib mkhan pos snga mar skabs dbye | bar chad dris la kun gtsang na sngon


Note gyi mkhan pos skabs phye shig |
 

A brief survey of the Tibetan translation of the Vinayasutra in the 'Did 30 ba'i mdo'i 'grel pa
3.4 Concerning compound
VS-1.89

VSMSA sTlasamadhiprajnavimuktitat*jnanadarsaneh
|| (VSMsA3a6)

VSMSB | sTlasamadhiprajnavimugtitajnan-
adarsane | (VSMSB2M)
VSSMSA sTla-samadhi-prajna-vimukti-
taj jnana-darsanaih || (VSSBGp.
16)

VSTib tshul khrims dang ting nge 'dzin dang
shes rab dang rnam par grol ba dang de shes pa dang mthong ba rnams kyis kyang ngo ||
VSSTib tshul khrims dang ting nge 'dzin dang
shes rab dang rnam par grol ba dang de shes pa dang mthong ba rnams ky- ang ngo ||
VSTTib tshul khrims dang ting nge 'dzin dang

shes rab dang rnam par grol ba dang de shes pa dang mthong ba rnams kyis kyang ngo ||
VSVyTib tshul khrims dang | ting nge 'dzin
dang | shes rab dang | rnam par grol ba dang | de shes pa mthong ba rnams so||
VSVTib tshul khrims dang ting nge 'dzin dang
shes rab dang rnam par grol ba dang de'i ye shes mthong ba yang ngo ||


Note


With regard to the exposition of the compound tajjnanadarsana there arises a divergence between VSVTib and those belonging to G1. VSVTib asserts that it is to be paraphrased as rnam par grol ba’i ye shes mthong ba, VSTTib, however, holds another view: rnam par grol ba shes pa dang mthong ba zhes bya ba’i tha tshig go, the same is the explanation presented in VSVyTib. The question whether VSVTib has deciphered the compound

correctly or not, would, due to the silence on further exposi¬tion of this compound in VSS, have remained unsettled, not for the remarks we met with in Rin 'phreng, which positively de¬nounce the glossy presented in VSVTib:

shes rab byed par | de'i ye shes mthong ba yang ngo zhes panimnyanpadangbsampa'idgra bcompa'i mya ngan las 'das pa'i rang bzhin zhi ba mthong ba'o zhes bshad pa'i phyir zer ba mi ’thad de | nyon mongs pa'i 'ching pas bcings pa rnams mislobpa'irnamgrolphunsumtshogs pa'galba'i phyir ro || (Rin 'phreng, p.41)
To prove this, one parable is added in Nyi 'od:

so so skye bo chos can | mi slob pa'i rnam par grol ba phun sum tshogs pa ma yin par thal | nyon mongs pa'i 'ching bas bcings pa yin pa'i phyir rnam par grol ba la sogs pa'i yon tan de rnams bdag la mi ldan na gzhan rgyud kyi yon tan sngo bgrang shes pa tsam gyis rnam grol phun sum tshogs par ga la 'gyur te sbrang pos phyug po'i nor la sngo bgrang shes kyang sbrang po de nor rdzas phun sum tshogs pa ma yin pa bzhin no | ro | (Nyi 'od, pp. 120-1)
Of course, this is not essentially the final decision, but at least we know now how the Tibetan scholars judged. 
A brief survey of the Tibetan translation of the Vinayasutra in the 'Did


32 ba'i mdo'i 'grel pa

3.5 Concerning syntax

na tiksena dantam jihvam karnnam cot*ghrset* || (VSMsAl0a3)

| na tiksnena dantam jihvam karnan cotghrset* | (VSMsB6a2)

na tiksnena dantam jihvam karnan ca utghrset || (VSSBGp.43)

mi rno bas so dang lce dang rna ba drung bzhar bya'o ||
mi rno bas so dang lce dang rna ba drung bzhar bya'o ||
mi rno bas so dang lce dang rna ba drung bzhar bya'o ||
rno bas so dang lce dang rna ba mi brud ||
rnon pos lce dang so dang rna ba mi brud ||

In this case VSVyTib agrees with VSVTib with regard to the grammatical role of na: independent negative governing pre-dicative, in G1 however, it is recognized as forming together with tiksa a nan-compound. Both can be right and there is no substantial divergence in the sense. It is impossible to judge from the reading in the manuscripts whether na is intended to be in combination with tiksa or not, and this might have been a problem confronting the translators.

 
VS-1.10

VSMSA sarvvasanni j patena va sannisanne

VSMSB nulayanam va || (VSMsAlb4-5)

| sarvasannipatena va samnisanne nu¬
VSSMSA layanam va | (VSMsBlb3) sarvasannipate va [samn-
VSTib isanne']nulayane va || (VSSBGp.
7)

thams cad 'dus pa la yang rung | gnas
VSSTib khang so so na 'khod pa la yang rung ngo ||
thams cad 'dus pa la yang rung | gnas
VSTTib khang so so na 'khod pa la yang rung ngo ||
thams cad 'dus pa la yang rung | gnas
VSVyTib khang so so na 'khod pa la yang rung ngo ||
thams cad 'dus pa'am gnas gnas na
VSVTib 'khod pa la'o ||
thams cad 'dus 'khod pa la'am gnas
Note gnas su'o ||


The translation preserved in VSVTib is at odds with the rest four versions in regard to the syntactical role of sannisanne, ’khod pa. In G1 and VSVyTib, it is understood as in direct connection with anulayanam, thus yields: ‘Either when assembles together or when stay in residential cells severally, [let propose the mo-tion].' Whereas, the translation in VSVTib says: ‘When all as-sembles sitting, or when in separated cells (gnas gnas su’o), [let put forward the motion.]'

Obviously, in VSVTib, sannisanne is taken as bearing intimate relation with sarvasannipate and anulayanam is considered a separate unit.
To this point, pertinent Chinese materials are in favor of that which is preserved in G1 and VSVyTib:

A brief survey of the Tibetan translation of the Vinayasutra in the 'Did
34 ba'i mdo'i 'grel pa

(<<WT24>>p.456.1)

‘Then, propose the motion to the community for [him], all of the community should assemble without absence, or make known by visiting [each] cell.'
Now we have, through the investigating of the instances listed above, come into a strong impression that there exist certain consistent principles which the translators follow constantly. And these principles, or we can call them stylistic peculiarit¬ies, as has been mentioned before, had already been pointed out by Tibetan scholars.



4 Stylistic peculiarities of the transla-tion in the 'Dul ba'i mdo'i 'grel pa


By stylistic peculiarities we refer to the divergent principles adopted in different translations of VS. The dichotomy at¬tested between G1 and G2 which we have discussed in last sec¬tion can be stylistically summarized as such that G2 dis¬tinguishes itself from G1 in its preferring literal transla¬tion to free one and its constant seeking after concise expres¬sions. The latter tendency reminds us the outstanding peculi¬arity of a sutra-text.

In VSVTib the faithful adherenc to original is carried out to such an extent that the translation becomes awkward, but this we should understand as deliberately designed for the sake of re-taining original style of VS, not symptom of inadvertence or in-competence. For this we need only to read the interpretaion in VSVTib where syntactic construction of the sutra concerned is neatly rearranged and terminologies duely explained. In this section, we aim to provide more specified descriptions of the outstanding features of the translation of VS in VSVTib.



4.1 VSVTib's faithful adherence to original



Thad 'gyur, as defined in Nyi 'od, means faithful adherence to Sanskrit original and it seems that the principle has been over- enthusiastically carried out in VSVTib, this we can infer from the following examples.

A brief survey of the Tibetan translation of the Vinayasutra in the 'Did

36 ba'i mdo'i 'grel pa

4.1.1 Retaining word order

VS-1.51
VSMSA samkuna caturamgulenaitat sadhu ||
(VSMsA2b2-3)
VSMSB | samkuna caturangulenaitat sadhu |
(VSMsB2a2)
VSSMSA samkuna catura[ngulena eta]t sadhu ||
(VSSBGp.12)
VSTib de ni thur ma sor bzhi pa bzang ngo ||
VSSTib de ni thur ma sor bzhi pas bzang ngo
||
VSTTib de ni thur ma sor bzhi pas bzang ngo
||
VSVyTib de ni shing bu sor bzhi pa na dge'o ||
VSVTib phur bu sor bzhi la de dge'o ||
Note
In VSVTib alone, the word order of Sanskrit text is retained, elsewhere, etat / de is put at the very beginning of the transla-tions, in the following cases, too, VSVTib adheres faithfully to the original: VS-1.5; 1.13; 1.27; 1.41; 1.43; 1.45; 1.51; 1.64; 1.72; 1.87; 1.102; 1.104; 1.108; 1.111; 1.126; 1.543.



4.1.2 No interpreting additional added




VS-1.485

VSMSA na tadviruddham* || (VSMsAl0b5-6)
VSMSB | na tadviruddhim* | (VSMSB6M)
VSSMSA na tadviruddham || (VSSBGp.45)
VSTib de dang zhe 'gras pa la mi bzhar ro ||
VSSTib de dang zhe 'gras pa la mi bzhar ro ||
VSTTib de dang zhe 'gras pa la mi bzhar ro ||
VSVyTib de dang mi mthun pa la ma yin ||
VSVTib de dang mi mdza' ba la ma yin ||


Note


In G1 ahyet / bzhar which is drawn from the last sutra has been
37

added to complete the whole sentence, however in VSVTib as well as in VSVyTib the conciseness of sutra-style is retained.
4.1.3 Mechanical translation of wordings
VS-1.21

VSMSA yacnantaram va || (VSMsA2a1)
VSMSB | yajnanantaQram va | (VSMsBlb4)
VSSMSA yancanantaram va || (VSSBGp.8)
VSTib gsol ba'i 'og tu yang rung ngo ||
VSSTib gsol ba'i 'og tu yang rung ngo ||
VSTTib gsol ba'i 'og tu'ang rung ngo ||
VSVyTib gsol ba'i rjes la yang ngo |
VSVTib gsol ma thag tu yang ngo |


Note


There is no fixed term for anantara in Mvy, but it is easy to recognise that the expression: ...ma thag tu adopted in VSVTib is more accurate than that in the rest texts: ...’i ’og tu.


Analogous examples are available in the following cases:


VS-1.29 tadyatha G1: dper na; VSVyTib: 'di Ita ste; VSVTib: de ’di lta ste
VS-1.45 asabdita G1, VSVyTib: ma bos; VSVTib: sgra maphy- ung
A brief survey of the Tibetan translation of the Vinayasutra in the 'Did
38 ba'i mdo'i 'grel pa

4.1.4 Expressive depiction of the taddhita suffix
VS-1.16
VSMSA upadhyayah kasayani vastrani dadyat*
|| (VSMsAlb6)

VSMSB | upadhyayah kasayani vastrani
dadyat* | (VSMsBlb3)
VSSMSA upadhyayah kasa[ya]ni vastrani
dadyat || (VSSBGp.7)
VSTib mkhan pos gos ngur smrig dag sbyin
par bya'o ||
VSSTib mkhan pos gos ngur smrig dag sbyin
par bya'o ||
VSTTib mkhan pos gos ngur smrig dag sbyin
par bya'o ||
VSVyTib mkhan pos chos gos ngur smrig sbyin
||
VSVTib mkhan pos ngur smrig can gyi gos
sbyin ||


Note

The suffix can in ngur smrig can gyi gos, normally expressing the idea of in possession of one object indicated by the word whereto it affixed, is uniquely attested in VSVTib to depicting taddhita suffix an in kasaya. And this uniqueness, in my view, is of particular interest, sinceitcan serveto introduce to us both the knowledge possessed by the translators of VSVTib in regard to a specific branch of Sanskrit grammar and the effort they had made to convey it in Tibetan.


4.2 VSVTib's conciseness

4.2.1 Preferring monosyllabic wording

It is a tendncy shared in G2:
39

VS-1.474
VSMSA upanat*poccanam || (VSMsAl0b3)
VSMSB upanapoccanam | (VSMsB6a6)
VSSMSA upanat-pronchanam || (VSSBGp.44)
VSTib mchil lham gnyis phyi'o ||
VSSTib mchil lham gnyis phyi'o ||
VSTTib mchil lham gnyis phyi'o ||
VSVyTib lham phyi ||
VSVTib lham phyi ||


Note

The equivalent to upanah in G1 is mchil lham, in G2, lham, we have several analogs:
VS-1.54 samaya G1: dus tshod; G2: dus
VS-1.612 alohini G1: zla mtshan (mi zag pa); G2: khrag (med pa)
VS-1.510 upasthdyakah G1: nad g-yog; G2: g-yog
And we have example where VSVTib alone abides by the rule:
VS-1.456


VSMSA saprayojanam cet* parisravanasya-pi || (VSmsA10a6-10b1)
VSMSB | saprayojanam <cet*> par-
isra<va>nasyapi | (VSMsB6a4)

VSSMSA sa-prayojanam cet parisravanasya api || (VSSBGp.44)
VSTib gal te dgos payodnachutshags kyang ngo ||
VSSTib gal te dgos payodnachutshags kyang ngo ||
VSTTib gal te dgos payodnachutshags kyang ngo ||
VSVyTib gal te dgos pa lta na chu tshags kyang ngo ||
VSVTib

Note de ste | dgos na tshags kyang ngo ||
A brief survey of the Tibetan translation of the Vinayasutra in the 'Did
40 ba'i mdo'i 'grel pa
4.2.2 Ignoring upasarga

According to the sGra sbyor bam gnyis it is not always essen¬tial to report each upasarga, and it seems, leaving out of con-sideration for the time being the chronology, this is a normal practice in G2, especially in VSVTib:
VS-1.9

VSMSA kaccit* parisudhyatiti prstva suddham
arocayet* || (VSMsAlb4)

VSMSB | kancit parisuddhyatiti prstva
suddham arocayet* | (VSMsBlb2-3)

VSSMSA kaccitparisuddha' iti prstva, suddham
arocaye[t] || (VSSBGp.7)
VSTib ci yongs su dag gam zhes dris la dag
na zhu bar bya'o ||
VSSTib ci yongs su dag gam zhes dris nas dag
na zhu bar bya'o ||
VSTTib ci yongs su dag gam zhes dris la dag
na zhu bar bya'o ||
VSVyTib ci yongs su gtsang ngam zhes dris la
gtsang na zhu'o ||
VSVTib ci gtsang nga'am zhes dris la gtsang
na 'dri ||


Note


VS-1.83
VSMSA pitakabhijnatvam || (VSMsA3a5)
VSMSB | pitakabhijnatvam | (VSMSB2M)
VSSMSA pitakabhijnatvam || (VSSBGp.15)
VSTib sde snod mngon par shes pa nyid do ||
VSSTib sde snod mngon par shes pa nyid do ||
VSTTib sde snod mngon par shes pa nyid do ||
VSVyTib sde snod mngon par shes pa'i ||
VSVTib sde snod shes pa ||


Note 


41
VS-1.85
VSMSA adhisilQlacittaprajnasiksanta ||
VSMSB (VSMsA3a5)
| adhisTlacittaprajnam siksanta |
VSSMSA (VSMSB2M)
adhisTlacittaprajnam siksatta ||
VSTib (VSSBGp.15)
lhag pa'i tshul khrims dang sems dang
VSSTib shes rab la slob pa nyid do || lhag pa'i tshul khrims dang sems dang
VSTTib shes rab la slob pa nyid do || lhag pa'i tshul khrims dang sems dang
VSVyTib shes rab la slob pa nyid do || lhag pa'i tshul khrims dang | sems
VSVTib dang | shes rab la slob pa'o || tshul khrims dang sems dang shes rab
Note bslab pa dang ngo |


We find no trace of pari, abhi, and adhi in VSVTib, in G1 and VSVyTib, to them are assigned yong su, mngon par and lhag pa respectively.
Butthedepressionofupasargais not without restraint, when it is necessary, VSVTib and VSVyTib depict it properly: 
A brief survey of the Tibetan translation of the Vinayasutra in the 'Did
42 ba'i mdo'i 'grel pa
VS-1.114
VSMSA unas ced anyasma{d}y upanisrit-
VSMSB yartham arpayet* || (VSMsA4a1)
| unas ced anyasmay upanisrityartham
VSSMSA arpayet* | (VSMsB2b5)
unas ced anyasmai upanisrityartham
VSTib arpayet || (VSSBGp.20)
gal te ma lon na gzhan la re zhig gnas
VSSTib pa'i don du gtad par bya'o || gal te ma lon na gzhan la re zhig gnas
VSTTib pa'i don du gtad par bya'o || gal te lo ma lon na gzhan la re zhig
VSVyTib gnas pa'i don du gtad par bya'o || gal te ma tshang na gzhan la nye gnas
VSVTib su 'chol ||
de ste ma tshang na gzhan la nye gnas
Note kyi phyir gtod cig |


In case of suppressing of upa in upanisriti, the meaning of the term can not be fully expressed, therefore it is retained, in G1 we find a free translation of upanisriti: re zhig gnas pa.
43


4.2.3 Ignoring tva and tal


VS-1.106
VSMSA ratn
ânâm varnnasya tïrthyânâm av-
arnnasya bhutasyoktâv akupyatvam ârâdhitacittatâ || (VSMsA3b4)
VSMSB | ratnânâm varnasya tïrthyânâm
avarnasya bhutasyoktâv akupyatvam ârâdhitacittatâ | (VSMsB2b3-4)
VSSMSA ratnânâm varnasya tïrthyânâmav- arnasya bhutasya uktau akupy¬atvam ârâdhitacittatâ || (VSSBGp.
19)

VSTib dkon mchog rnams kyi bsngags pa
dang mu stegs can rnams kyi bsngags pa ma yin pa yangdag parbrjodpa na mi g-yo ba nyid ni sems mgu bar gyur pa nyid do ||
VSSTib dkon mchog rnams kyi bsngags pa
dang| mustegscanrnamskyibsngags pa ma yin pa yang dag par brjodpa ni mi g-yo ba nyid ni sems mgu bar gyur pa nyid do ||
VSTTib dkon mchog rnams kyi bsngags pa
dang| mustegscanrnamskyibsngags pa ma yin pa yangdag parbrjodpa na mi g-yo ba nyid ni sems mgu bar gyur pa nyid do ||
VSVyTib dkon mchog bstod pa dang | mu stegs
smadpayangdagparsmras kyangma 'khrugs na sems mgur bar byas pa'o ||
VSVTib dkon mchog bsngags mur 'dug smad
par yang dag par smra ste ma 'khrugs na sems rangs ba'o ||


Note


In G1 nyid is employed to depict the suffix tva and tal however, in G2 they are not expressively given in the translation of the
A brief survey of the Tibetan translation of the Vinayasutra in the 'Did
44 ba'i mdo'i 'grel pa
sutra:

akupyatvam G1: mi g-yo ba nyid; G2: ma ’khrugs
drddhitacittatd G1: sems mgu bar gyurpa nyid; VSVyTib: sems mgur bar byas pa; VSVTib: sems rangs ba
4.2.4 Ignoring plurality
By and large, VSVTib is mute with plurality, for example:
VS-1.62

VSMSA pataniyan° dharmman || (VSMsA2b4)
VSMSB ©(pa)taniyan dharman* | (VSMsB2a3)
VSSMSA pataniyan dharman || (VSSBGp.12)
VSTib ltung bar 'gyur ba'i chos rnams dang
||
VSSTib ltung bar 'gyur ba'i chos rnams dang
||
VSTTib ltung bar 'gyur ba'i chos rnams dang
||
VSVyTib ltung ba'i chos rnams so ||
VSVTib ltung ba'i chos kyang ngo ||


Note


But it does not follow that the simplification is unsubject to any principles, for we have instance where when it is deemed intrinsically necessary the pluraty is expressed in VSVTib:
45
VS-1.104
VSMSA masi tirthya i©ti pravrajyartham
upasa kranta prcchet*|| (VSMsA3b3)
VSMSB | masi tirthya iti pravrajyartham upas-
amkrantam prcched upasampadakas ca | (VSMsB2b3)
VSSMSA ma'si tirthyah iti pravrajyartham upas-
amkrantam prcched, upasampadakas ca || (VSSBGp.19)
VSTib rab tu 'byung ba'i phyir nye bar 'ongs
pa la mu stegs can ma yin nam zhes dri bar bya zhing bsnyen par rdzogs par byed pa dag gis kyang bya'o ||
VSSTib rab tu 'byung ba'i phyir nye bar 'ongs
pa la mu stegs can ma yin nam zhes dri bar bya zhing bsnyen par rdzogs par byed pa dag gis kyang bya'o ||
VSTTib rab tu 'byung ba'i phyir nye bar 'ongs
pa la mu stegs can ma yin nam zhes dri bar bya zhing rdzogs par bsnyen par byed pa dag gis kyang ngo ||
VSVyTib rab tu 'byung bar 'das pa la khyod mu
stegs can ma yin nam zhes dri bsnyen par rdzogs par byed pa dag gis kyang ngo ||
VSVTib khyod mur 'dug pa ma yin nam zhes
rab tu 'byung bar 'ongs pa la dri | bsnyen par rdzogs pa dag gis kyang ngo |


Note


Here a device typical of composition in sutra style is concerned: parinama, which can serve to make grammatically proper the expressions drawn from afore in new syntactical construction, in the present case, in keeping with the shifting of the sub¬ject from upadhydyah to upasampddakdh, the main verb prc- chet is transformed into prccheyuh. In VSTTib as well as in
A brief survey of the Tibetan translation of the Vinayasutra in the ’Dul


46 ba'i mdo'i 'grel pa
VSVyTib the transformation is extensively discussed and justi-fied, in VSVTib this is also concerned, though in a much eco-nomical scale:
dri ba ni gsang ste ston pa'i slob dpon bskos nas 'drir gtang ba de nyid dge 'dun kun gyis dris par chud do |
Thus we realize that the translator follows perfectly the text, and the pratikas in VSVTib are therefore a production out of well-organized principles.
47


5 Archaism in the 'Dul ba'i mdo'i 'grel pa


The age of a text equals approximately to its value in editing, in this section we aim to report the archaism reflected in both the pratikas and their interpretation in VSVTib.

It is true that in VSVTib, there are no such telling features typical ofarchaismas thereversedgigu, themtha'rten'a, thedadrag, since they can easily be done away with, and doing so in any case would be the first choice ofthe revisers. And in ourcase I suppose it might be the only choice of the reviser, even so, we can still find those which have been left untouched, and they become our starting point:
5.1 Archaic expressions

5.1.1 VS-1.71 ... 'o cog

gnas la ma zhus par gnas pas bya ba mi bya | u pa de sa'am gnas bcas pa'i slob dpon la ma zhus par gnas pa'i slob ma las ci bya'o cog ma dris par mi bya'o || ma dris kyang byar rung ba rnams ni 'og nas 'chad do || (VSVTib)
This is a well-attested character of archaic composition.
5.1.2 VS-1.91 sngon cad

dran pa dang ldan pa yang ngo || sngon cad bshad cing bslabs pa'i rnams ma brjed par dran pa dang | de bslabs ba phyis mi brjed pa'o || snga ma gnyis dang 'di dang gsum gyi steng du mang tu thos pa dang tshul khrims dang ldan pa gnyis kyis bsnan te lnga phrugs gcig ste (|) (VSVTib) From sngon cad to sngo chad is a result of replacing ca by cha.

5.1.3 VS-1.42 ltag ma

de lta ma yin na thams cad kyis lhung bzed bz- ang ngo zhes smros shig | mdo ltag ma ltar dris nas mi che mi chung mi dkar na bzang ngo zhes dge 'dun gyis lan gsol cig | gral dpon gyis brtags te bz- ang ngo zhe na 'og ma rnams kyis nan tan du ma mthong yang bzang ngo zhes smras na'ang nyes med | (VSVTib)
According to the Li shi’i gur khang, ltag ma is an archaic syn-onym of gong ma, and we find that gong ma is attested in VSVyTib:
lhung bzed bstan nas dge 'dun gyis bya ba ni | de lta ma yin na lhung bzed bzang ngo zhes thams cad kyis brjod ces pa yin no || gong ma'i skyon gsum lta bu med na 'dus pa thams cad kyis brjod par bya'o || (VSVyTib)

ltag ma is also attested in VSVTib's interpretation of VS-1.101 and VS-1.86.
5.1.4 VS-1.41 spya nga brim

dge 'dun la lhung bzed kyang nye bar bstan cing ma ches sam ma chung ngam mi dkar ram zhes so || lhung bzad ni sngags kyis ma btab pa'i snga rol du dge 'dun la ma ches sam ma chung ngam mi dkar ram zhes bya zhing mthar gyis spya nga brim | (VSVTib)
The wording spya nga brim, which is not included in the Bod rgya tshig mdzod chen mo and Jaschke but obviously, semantic-ally synonymous to spyan sngar brim pa’o in VSVyTib, is re-gistered in the gSer gyi me longinthesenseof‘putbeforeone's eyes and see':
49

spya nga brims pa | sbyan sngar bsgrigs nas gzigs pa'i don te | <tun hong gter yig gna' bo'i bod yig shog dril> las | dmag myi mang pa bsags te | spya nga brims nas | dmag ru drug la | gtsigs gsal bar stsal nas | zhes pa lta bu'o || (gSer gyi me long, p.463)
5.1.5 VS-1.120 rmed pa

pha 'tsho la ma gnang ste de dag gi yul mi ring na rab tu 'byung bar 'dod pa de zhag bdun du 'chol shig | rab tu 'byung bar 'ongs pa de pha ma gnyis ka'ang ma shi ste yod la rab tu 'byung bar gnang ba'i cha'ang med la pha ma'i yul kyang mi ring na 'dri zhing rmed pa yod dam mi 'ong brtag pa'i phyir zhag bdun du bsdad te ma 'ongs na rab tu dbyung yul ring ba'i tshad ni pha mas nyams kyis mi lcogs shing mthus mi khugs pa tsam la bya'o || (VSVTib)
rmed pa is registered as an archaic wording in dictionaries. Bod rgya tshig mdzodchen mo (p.2155):
rmed pa | [rnying] 'dri ba dang | zhu ba dang | smra ba | Ш H, °
gSergyi me lung (p.671):

rmed pa | (1) 'dri ba'am smra ba'i don la 'jug ste | <ngag sgron gyi 'grel pa > las | ...dper na | <'dul ba lung gzhi > las | bcom ldan 'das rnams ni mkhyen bzhin du rmed par mdzad pa ste | mkhyen bzhin du rmed kyi | zhes pa lta bu'o || .
5.2 Commonness sharing with P.T. 903

The list of the archaic wordings is not exhaustive, but even we add more, it might still be dismissed as flimsy, therefore we need other evidence, and to this point, the commonness of the divergence between P.T.903 and its canonical correspondent and that between G1 and G2 can serve to add weight to our argument that the translation of VS in G2, especially that in VSVTib is rather old.
5.2.1 Indicative terminology

de la deng nI dge 'dun gyI gsol' sbyIn yang bco lnga pa : | (2.4) de la bye brag ni deng dge 'dun gyi gsol sbyong bco lnga pa lags la | (T)
Obviously, the word gso sbyin, not being registered in Mvy, yet commonly occurring in P.T.903, is an earlier form of gso sbyong, and it is attested in VSVTib:


yang dag par blangs te ma bshad pa 'ang kun shes par grub par chud do pa yang ngo || khrims rdzogs par yang dag par thob pas na | phas pham pa la sogs pa kha cig ni bshad de thos pa yin lhag malasogs pamangpo gzhandabshaddumizinte ma thos mod kyis phyis gso sbyin 'chad pa'i tshe yang thos par 'gyur phyis kun bslab cing shes par byar yod pas na | da bshad ba dang 'dra ste phyis lobs bar 'gyur ro zhes bsgo (VSVTib)

5.2.2 Monosyllabic words

lham phud la gus pa cI rIgs su byas te : | tshog tsog por 'dug la : (2.1)
mchil lham gnyis phud la gus pa ci rigs su byas nas tsog tsog por 'dug ste | (T)
Here we meet with lham for mchil lham, analogous example can be found in VS:
51

VS-1.474
VSMSA upanat*poccanam || (VSMsAl0b3)
VSMSB upanapoccanam | (VSMsB6a6)
VSSMSA upanat-pronchanam || (VSSBGp.44)
VSTib mchil lham gnyis phyi'o ||
VSSTib mchil lham gnyis phyi'o ||
VSTTib mchil lham gnyis phyi'o ||
VSVyTib lham phyi ||
VSVTib lham phyi ||


Note

5.2.3 Plurality
tshe dang ldan ba zhes bya ba nI : | | yang na gzhon ba | yang na rgan ba'o : || (1.4)
tshe dang ldan ba dag ces bya ba ni yang na gzhon ba yang na rgan ba'o || (T)
dag cannot be found in P.T.903, so in the same case translation preserved in VSVTib:
VS-1.62


VSMSA patanTyan0 dharmman || (VSMsA2b4)
VSMSB © (pa)tamyan dharman* |
(VSMsB2a3)

VSSMSA patamyan dharman || (VSSBGp.12)
VSTib ltung bar 'gyur ba'i chos rnams dang
||

VSSTib ltung bar 'gyur ba'i chos rnams dang
||
VSTTib ltung bar 'gyur ba'i chos rnams dang
||
VSVyTib ltung ba'i chos rnams so ||
VSVTib ltung ba'i chos kyang ngo ||


Note


From the view of textual criticism, textus ornatior is later than textus simplicior, and to the tendency reflected in item b and item c, it seems, this can also be analogously applied.


5.3 Migration of equivalents



Now it seems that we can say with more certainty that the trans-lation now in G2 is earlier, or even much more earlier than that in G1. Butwearenotataposition forthetimebeing to declare with any degree of certainty how far can we put back the abso- lutelimitofourtext,sincewehavenotbeingableto study care¬fully all the relevant materials. However, the following evid¬ence along with what we have just put forward can at least make certain the relative chronological order of the translations of the two groups.

A gross investigation shows thatthereexistinmanycasesanin- teresting transmission of the equivalents among the comment-aries. The wordings which are employed to explain the equi-valents attested as pratikas in G2, turn out to be the terminolo-gies utilized in the translation of the sutras in G1. This, coupled with the fact that those shifting substitutes are in general more comprehensible than what are replaced by them suggests that an effort aiming to remove inconvenience resulting from com-paratively difficult wordings has presumably intervened, forwe know it is normal in the course of transmission lectio difficilior is replaced by comparatively pellucid one. 
5.3.1 From VSVTib to G1

VS-1.32
VSMSA pathasya trir ekagathaparivarttanam ||
VSMSB (VSMsA2a3)

| pathasya trir ekagathaparivartanam |
VSSMSA (VSMsBlb5)
pathasya trirekagatha-parivartanam ||
VSTib (VSSBGp.10)

klog pa ni tshigs su bcad pa gcig lan
VSSTib gsum bzlas na'o ||
klog pa ni tshigs su bcad pa gcig lan
VSTTib gsum bzlas na'o ||
klog pa ni tshigs su bcad pa gcig lan
VSVyTib gsum bzlas na'o ||
klog pa ni tshigs bcad gcig lan gsum
VSVTib bslab pas so ||
lhogs shig pa ni tshigs su bcad pa gcig
Note lan gsum klags pas so ||


lhogs shig pa ni tshigs su bcad pa gcig lan gsum klags pas so || klog pa'i slob dpod ni gnyis kas slob dpod dang slob mar 'du shes bzhag la zhal nas gsungs pa'i tshigs su bcad pa gcig lobs par mkhan pos smras la slob mas bzlas te tsog tsog por 'dug la lan gsum du bzlas na slob dpon du 'gyur | (VSVTib)
 

VS-1.456
VSMSA saprayojana cet* parisravanasya-pi ||
(VSmsA10a6-10b1)

VSMSB | saprayojanam <cet*> par-
isra<va>nasyapi | (VSMsB6a4)
VSSMSA sa-prayojanam cet parisravanasya api || (VSSBGp.44)
VSTib gal te dgos payodnachutshags kyang ngo ||
VSSTib gal te dgos payodnachutshags kyang ngo ||
VSTTib gal te dgos payodnachutshags kyang ngo ||
VSVyTib gal te dgos pa lta na chu tshags kyang ngo ||
VSVTib


Note de ste | dgos na tshags kyang ngo ||

de la de ste | dgos na tshags kyang ngo || mkhan po bsod snyoms la 'gro ste zan slar gnas su mi za'am | slar gnas su mi 'khor ba sta bu la chu tshags dgos na'ang bkru ste brdzangs | (VSVTib)
Other examples are: VS-1.31; 1.59; 1.76; 1.79; 1.108; 1.112; 1.171.
5.3.2 From VSVyTib to G1
VS-1.586

VSMSA atracayah samghah || (VSMsAl2b6)
VSMSB © | atracayas sa ghah | (VSMSB7M)
VSSMSA atraca[ya]ssamghah || (VSSBGp.51)
VSTib 'dir dge 'dun gyis bsnan par bya'o ||
VSSTib 'dir dge 'dun gyis bsnan par bya'o ||
VSTTib 'dir dge 'dun gyis bsnun par bya'o ||
VSVyTib 'dir dge 'dun klag go |
VSVTib 'dir dge 'dun gnyis kyis so ||
Note 
55

'dir dge 'dun klag go zhes pa ni dge slong ma bnyen par rdzogs pa'i tshe dge slong gi dge 'dun gyis bsnan cing klag pa'o | (VSVyTib)
VS-1.509
VSMSA kalpate bhaisajyasya sa ghatah
VSMSB kevalasya glanasya paribhogah || (VSmsA11a4)
| kalpate bhaisajyasya samghatah |
VSSMSA | kevalasya glanasya paribhogah | (VSmsb6b3)
kalpate bhaisajyasya samghatah


VSTib kevalasya glanasya paribhoga || (VSSBGpp.46)
dge 'dun las byung ba'i sman ni nad
VSSTib pa 'ba' zhig gis spyad du rung ngo || dge 'dun las byung ba'i sman ni nad
VSTTib pa 'ba' zhig gis spyad du rung ngo || dge 'dun las byung ba'i sman ni nad
VSVyTib pa 'ba' zhig gis spyad du rung ngo || dge 'dun gyi sman nad pa nyi tshes
VSVTib spyad du rung ||
dge 'dun gyi sman la nad pa 'ba' zhig
Note gis spyad par brtags pa'o ||

dge 'dun gyi sman nad pa nyi tshes spyad du rung zhes pa ni nad pa 'ba' zhig gis dge 'dun gyi sman spyad du dbang bar ston pa'o || (VSVyTib)
VS-1.450

VSMSA varttapracchanam° || (VSMsAl0a5)
VSMSB | vartapracchanam | (VSMsB6a3)
VSSMSA varta-prcchanam || (VSSBGp.43)
VSTib bsnyun gsol lo ||
VSSTib bsnyun gsol lo ||
VSTTib bsnyun gsol lo ||
VSVyTib nad dri ||
VSVTib snyun gsol |


Note

nad dri zhes pa ni phyag byas nas snyun yang
gsol cig pa'o || (VSVyTib)

Other examples can be found in VS-1.9; 1.22; 1.78; 1.88; 1.452.
The fact that the phraseology used to paraphrase the pratikas in G2 sometimes is attested as formal expression in the transla¬tion of the sùtras in G1 is of particular interest and regular oc-currence. In my view, it points to the chronological priority of the texts in G2 to that in G1, since it is hardly imaginable that the revisers would try to complicate the wordings and expres¬sions, and not, on the contrary, to make it more felicitous and readable.



6 The transmission history of the 'Dul ba'i mdo'i 'grel pa


As the translation of VS in VSVTib bears many an archaic fea-ture, as it resembles P.T.903—a text not having under¬gone intensive revision—in many respects concerning the us¬age of wordings and expressions, as it might in the his¬tory of the transmission of the translation of VS have served as a donor source of equivalents, we have ventured to con¬clude that it is anearlyversionofthetranslationofthecompen- dium of the Mulasarvastivadavinaya, and had somehow dur¬ing the course of time escaped later retouching.


The tentative conclusion, however, is purely based upon the evidences gathered from textual investigation, it is hard to avoided being thus challenged: is it also sustainable from the historical point of view? To answer this, we would have to turn to the colophons, which are not by themselves free from incon-sistency, yet in many cases the only material we can rely upon. As sharp as the differences of the translations between G1 and VSVTib is the variant records contained in the colophons of these two groups, VSVTib, VSSTib and VSTTib, those forming G1, each has in its colophon a record of the joint-board in charge of the translating of the text, however neither in the colophon of VSVTib nor VSVyTib can we find any clue in regard to the translation activities.


Another source which should be consulted is the catalogues edited during the Great Revision, among them two have been handed down to us: the Pho brang stong thang lhan dkar gyi chos 'gyur ro cog gi dkar chag and the sNgon dus yar lungs 'phang thang ka med na bzhugs pa'i bka' bstan mdo phyogs gtso che ba'i dkar chag chos rgyal lo pan rnams kyis sgrigs pa | dkar chag 'phang thang ma. The following is what we are informed from these two texts:
'dul ba mdo | shlo ka nyis stong bdun brgya ste bam po dgu (D.sna tshogs.jo.304b5)

'dul ba'i mdo'i rgya cher 'grel pa | shlo ka nyi khri chig stong ste bam po bdun cu || || (D.sna tshogs.jo.305a2)
'dul ba mdo | bam po dgu 'dul ba'i mdo rgya cher 'grel pa | bam po bdun cu || (p.34)
Except the arrangement of the record, the basic informa¬tion concerning VS™ and VST™ agrees, and both cata¬logues make no mention of VSS™, naturally so, since it were translated much later. But why VSV™ had not been registered in the earliest two catalogues available? Since, as we have sup¬posed, they should have been rendered earlier than the stand¬ard text now in G1, and it seems th
at we cannot doubt the trust¬worthiness of the silence because the colophons of both, as we have mentioned above, echo well to this.
It might have been registered in the mChims phu ma, we may suppose, but it is unlikely that the editorial principles and compiling arrangement of the lost catalogue would differ much from its companion volumes, and the drafts of VSVy™ and VSV™ would be separated from those of VS™ and VST™, though we cannot definitely rule out the possibility that they were at that time deposited in bSam yas mchims phu.
Is it due to VSVTib's being compiled note of Tibetan scholar? For it is well-known that the authenticity of VSVy™ and VSVTib had long been challenged by Tibetan scholars. They are called tshom pa'i gzhi. And this might owe its origin to the remarks of Bu ston, who suspected that the two commentaries are by indigenous scholars:

'dul ba' mdo'i 'grel pa la | slob dpon de nyid kyis mdzad pa'i rang 'grel stong phrag bcu bzhi pa ste | tshigs bcad stong phrag bcu bzhi ni | 'di yi
59


tshad du shes bya ste | dpal ldan rgya chen thams cadkyis || rgyal srid byed dus 'dibyas so || zhes so|| 'di steng pa lo tstsha ba ga rod tshul khrims 'byung gnas kyis phyis bsgyur ba ste | 'gyur shin tu ngan no || slob dpon chos kyi bshes gnyen gyis mdzad pa'i rgya cher 'grel pa shlau ka nyi khri chig stong | bam po bdun cu pa 'di dang rtsa ba gnyis ji na mi tra dang klu'i rgyal mtshan gnyis bsgyur bas 'gyur shin tu bzang ngo || slob dpon shes rab byed pa'i 'grel pa dang | 'grel chung du grags pa 'di bod kyi zin bris su byas pa yin nam yang snyam mo || de nas rnam rgyal go cha dang | rnam dag gi 'grel pa gnyis yod par bshad do || (sPyi rnam, 57a2-5)


The suspicion is quiet reasonable since we find in VSVTib indeed clues—perhaps exactly the basis of Bu ston's remarks —which hint to its not being original work. Firstly, it is much more compact than VSSTib and VSTTib, for example, to VS- 1.1, each of the two exegeses assign two and a half fo¬lios, yet in VSVTib only less than two lines has been de¬voted, which give an impression of being a brief summary of the long and somewhat tedious remarks of the former two.


And in the interpretation of VS-1.292, a comparison of the measurement between India and Tibet further confirms the reasonability of Bu ston's suspicion:
kha'i mtha'i cha nas mtheb mo'i sor gcig gi lhag mar 'bras btsos pa yang yul dbus su gyur 'chang gi spyod pa'i pra ta gnyis man chad tshod ma dang sngor ngu bcas par shong ba de ni tshad ma'o || lhung bzed kyi kha nas mtshon gang man chad du 'bras btsos pa rgya gar gyi yul gyi dbus ma ga dha'i sphyod pa'i pra ta gnyis tshod ma dang sngor ngu bcas par shong ba na lhung bzad chen po'o || pra ta gcig tshod ma dang sngor ngu bcas pa shong ba ni chung ngu'o || lhung bzad 'bring po pra ta gnyis man chad gcig yan chad ci tsam shong yang bla | gyur 'chang zhes bya ba ni yul na sbyar zhing spyod pa bre phul lta bu ste pra ta la bya | pra ta ni rgya gar srang bcu drug gi lcid kyi 'bras skam shong bapratagciggitshoddo || rgya gar yul gyi srang gcig la bod kyi srang gcig dang zho drug yoddergyagaryulgyisrangbcudruglapratagcig srang sum cu rtsa gnyis la pra ta do ste lhung bzed chen po'itshadbodsrang nyi shurtsabrgyadlapra ta gang gi tshad de bdun phul gcig dang zho phyad dang drug shong ba ni pra ta chung ngu'i tshad de chen po ni chung ngu do'i tshad do || (VSVTib)


It can hardly be argued that the explanation here being out of a Tibetan hand. But it does not follow that VSVTib as a whole can be so labeled, since as has been analyzed, the pratikas in VSVTib in most cases correspond word by word to the two Sanskrit ma¬nuscripts now available to us. It might be that VSVTib is a mix¬ture of the translation of the sutras and the interpretation writ¬ten down by Tibetan scholars during the course of cooperat¬ive translating and therefore, might reflect a primitive stage (the stage of drafting) of the whole process. But this is still not a good excuse for its absence from the officially edited catalogues where we find such titles as bsgyur ba las zhu chen ma bgyis te gtan la ma phab pa.
Another attribution might be more acceptable, that the mute might reflect an informed official choice, in other words, trans-lations of the two commentaries might have been deliberately banned from distribution and circulation by the authoritative office, which anticipated the possible disorder in practice due to the intrinsic inconsistencies among the commentaries. One example:



VS-1.73
VSMSA aikannapancasadvyamaparyantad vi-
harato gamana || (VSMsA3a1)
VSMSB | ekannam pancasavyamaparyantad
viharato gamanam | (VSMsB2a4-5)
VSSMSA ekannapancasat-
vyamaparyantadviharato gamanam || (VSSBGp.13)
VSTib gtsug lag khang nas 'dom bzhi bcu
rtsa dgu'i mtha' tshunchaddu 'gro ba yang ngo ||
VSSTib gtsug lag khang nas 'dom bzhi bcu
rtsa dgu'i mtha' tshunchaddu 'gro ba yang ngo ||
VSTTib gtsug lag khang nas 'dom bzhi bcu
rtsa dgu'i mtha' tshunchaddu 'gro ba yang ngo ||
VSVyTib bshang gci dang sor shing bca' ba
dang | gtsug lag khang gi nyen kor dang bcas pa'i mchod rten la phyag bya ba dang | gtsug lag khang nas 'dom bzhi bcu rtsa dgu'i bar du 'gro ba rnams ma gtogs so ||


VSVTib gtsug lag khang nas 'dom bzhi bcu
rtsa dgu'i mthar 'gro ba yang ngo ||
Note
While it is held in VSSTib and VSTTib that to worship a caitya one should gain the permission of the Supporter beforehand, VSVTib says there is no fault even if one go for this without asking permission:

gtsug lag khang nas 'dom bzhi bcu rtsa dgu'i mthar 'gro ba yang ngo | | gtsug lag khang ra bas bskor na ra ba'i phyi sgo nas grong dang nags la sogs pas bsu ba med na | 'dom bzhi bcu rtsa dgu thsun chaddu 'chags padang mchod rten laphyag
'tshal bar 'gro na ma zhus par song yang nyes med | (VSVTib)
And this had incurred direct criticism from Nyi 'od:


mchod rten la phyag bya ba ni zhes pa'i TT ka las | gtsug lag khang gi nye bar spyod pa'i nyie 'khor na yod pa'i mchod rten la phyag 'tshal bar 'gro na'ang zhu mi dgos la | gtsug lag khang gi nang na yod pa'i mchod rten la phyag 'tshal ba zhu mi dgos zhes don de ltar bshad pas mchod rten de dag gtsug lag khang gi nye 'khor na yod pa dang bcas pa'i gtsug lag khang gi nang du mchod rten la zhar byung gi phyag bya ba ni zhu dgos pa la ma gtogs so zhes bstan to || 'chag pa'i bsam pa kho nas 'gro ba na lam gar mchod rten la phyag 'tshal bar byed na des nyes par mi 'gyur ro zhes TT ka las bshad do || 'grel chung las ni mchod rten la phyag 'tshal bar 'gro na ma zhus par song yang nyes med ces ched du dmigs nas phyag 'tshal ba yang zhu mi dgos par 'chad do || de mi ’thad de | mchod rten la ched du dmigs nas phyag 'tshal ba de chos can | zhu mi dgos pa ma yin par thal | bya ba chen po yin pa'i phyir ro || rtags dang khyab pa gzhung gis grub ste | TT ka las | bya ba gang sbyor ba chen pos brtsam pa ni lhung bzed dang chos gos kyi las la sogs pa ste | de ni ma zhus par mi bya'o zhes gsungs lhung bzad kyi las tsam yang bya ba chen po yin par bshad pas mchod rten la phyag 'tshal ba de bya ba chen po yin pa la lta ci smros | so || 'dom bzhi bcu rtsa dgu | zhes pa ni 'grel chung las gtsug lag khang rva bas bskor na rva ba'i phyi sgo nas grong dang nags la sogs pa sab sub med na 'dom bzhi bcu rtsa dgu tshun chaddu 'chag parsong na nyes med ces nags la sogs pa sab sub yod na mi gnang bar 'chad do || de ltar na sgos kyi dgag byas bcas pa bzhag par 'gyur bas ha cang thal ches so zhes gsungs so || (Nyi 'od, pp.106-7)
63


So that it would seem that we have reason to think that it is the divergence which is apt to give rise to confusion and contradiction in practice should be responsible for the ab¬sence of VSVTib in the two catalogues. For this we can offer an¬other example, in colophon of the Vinayottaragrantha it is re¬corded that it is due to the questionable though complete trans¬lation of the Prccha, there arose disorder in daily prac¬tice in the Community (lag len la ’khrul pa mang po by¬ung ba). Such a chaos was finally put to rest (lag len mi mthun pa thams cad kyi sgro ’dogs der chod do), after the turn¬ing up in bSam yas mchim phu of an old though incom¬plete translation of the same text, which goes back to the time of Klu'i rgyal mtshan.
The potential of confusion due to its unorthodox view might be the reason of VSVTib's unpopularity. And to a certain ex¬tent the source of the view can also be traced. In colophon of VSVTib the author is given as Yon don gyi 'od, not Yon don 'od, a fact which might be considered insignificant and be ignored not for another point that the religious section to which the au¬thor belongs is given as Sarvastivada (thams cad yod par smra ba not Mulasarvastivada (gzhi thams cad yod par smra ba, as is widely accepted as Gunaprabha's nikaya:


mal stan gyi dngos po ste phran tshegs su gtogs pa'o || rdzogs so || || 'dul ba'i mdo'i rtsa ba'i 'grel pa chung ba slob dpon mang du thos shing yon tan dang ldan pa yon tan gyi 'od thams cad yod par smra bas byas pa'o || || lan cig zhus || ||
Is itpossible that the so called Yod don 'od ma represents point of view other than that reflected in VSS and VSTTib? Let's fur-ther consult one concrete example where VSVTib is deemed de-viating from traditional authority


VS-1.76
VSMSA krte vasarayed iti || (VSMsA3a2)
VSMSB | krte vasarayed iti | (VSMsB2a5)
VSSMSA krte avasarayed || (VSSBGp.14)
VSTib byas na bzod par bya'o ||
VSSTib byas na bzod par bya'o ||
VSTTib byas na bzod par bya'o ||
VSVyTib chod na 'bral bar ro zhes pa |
VSVTib e ma'o dge 'dun gyis de la smad par


Note 'dod pa la smad pa'i las mi byed pa dang | byas pa 'chags pa yang ngo |


rang gi rnam bshad las kyang byas na bzod par bya'o zhes bya ba ni nan tur gyi las byas pa la e ma'o dge 'dun gyis 'di la bzod par bya zhes bya'o zhes bshad do || byas na bzod par bya'o zhes bla ma la dge 'dun gyis nan tur gyi las byas na slob mas khro ba mi bstan par bzod par bya 'o zhes kha che dad kha cig skyor lung || zer ba ni mi rigs te sngar lung khungs su drangs pa'i lung gzhi dang 'gal la | sngar drangs pa'i rang gi rnam bshad dang yang 'gal lo || byas na bzod par bya'o zhes pa ni bla ma la dge 'dun gyis nan tur gyi las byas na bla ma la slob mas gros btab ste bzod pa gsol du bzhug par bya zhes snga rabs pa sha sbas rnams gsung zhing | 'grel chung las kyang mkhan po bskrad de phyi logs na 'phyan par gyur na de ltar ma byed par dge 'dun la bzod par gsol cig par gros gdab zhes 'grel chung las bshad do || de ni mi rigs te dge 'dun la bzod par gsol cig ces gnas kyi bla ma la gros gdab dgos na skabs 'di nyid kyi lung gzhi las mkhan po dang slob dpod dag la nan tur gyi las byas pa de bzod par gsol zhing nan tur gyi las byas pa'i rgyu de yang spong ba dag la zhes gsungs pa dang 'gal lo || ... (Nyi 'od, p.108)


Here we are informed the existence of three kinds of interpret-ation of this sutra:
65

1. The pupil says to the Community: ”Please forbear him!” (VSS)
2. The pupil says to him: ”Please forbear the Community!” (Kha ched dad kha cig)
3. The pupil persuade him to beg forbearance from the Com-munity (snga rabs pa Sha sbas rnams and VSVTib)
Accordingly, we are taught:

1. There existed a certain section of Kasmir (Kha che dad kha cig skyor lung), which is at odds with the one reflected in VSS in regard to the interpretation of the wording avasarayed (bzod par bya'o).
2. The predecessors Sha sbas and so forth (snga rabs pa Sha sbas rnams), agree with VSVTib in asserting that the disciple should urge the delinquent authority to beg forbearance from the community.


3. Both are denounced by the author of Nyi 'od as unfoun¬ded for their being contradistinctive to the authoritative writ¬ings.
So it seems it is its heretical interpretations, which re¬fleet the view of a section other than that held by the Mulasar- vastivadin and which might give rise to chaos in daily prac¬tice should mainly be responsible for the absence of VSVTib in the two officially authorized catalogues.


Now we have reached that the drafts of VSVyTib and VSVTib have been repelled from the two official catalogues, yet they succeeded in getting through the screen and finally gained a place in Tanjur. How this had happened, in what way had they escaped the revisions? Besides intrinsic incon¬sistency between VSVTib and other commentaries, we sug¬gest that the condensed style of the genre in which the text is couched requires to reduce to the extreme minimum the consumption of the wordings, and the readers are expected to have a good knowledge of the supporting works which will provide essential information for them to supply the omis¬sions as well as to take the subtle hint given through the un¬usual arrangement of the word order and the sporadic im¬plantation of the sutras in their commentaries would be ac¬counted for the maintenance of the status quo, the manipula¬tions of which might lead to such a rearrangement as amount¬ing to the retranslation of the whole text and it is most likely such an uproarious chaos might have been deliberately avoided.


Lastly, we quote one more paragraph containing remarks on VSVTib's interpretation of VS-1.74 fromNyi 'od, since this pas-sage contains bundles of informations with regard to the main concern of the present section:


'gyodpabsal ba dang sdig lta spong ba'i skabs su|'dulba'isdesnodmishes nanignas kyi bla ma byar mi rung bas | gnas kyi bla ma sde snod 'dzin pa yin par nges pa de la | slob ma sde snod 'dzin pa yin par ma nges pa des | ltung ba las 'gyod pa bsal ba dang sdig lta spong ba 'gal zhe na | ’grel chung las mkhan pos sngon sde snod shes shing mkhas pa las rlung nad la sogs pas dbang po nyams shing mi dran par gyur pa de lta bu la bya’o || zhes so || khyad par du yang sdig lta spong ba la ni mngon pa’i sde snod gnas kyi bla mas mi mkhyen yang slob mas shes pa yang srid pas ’gal ba med do || (Nyi 'od, p.107)
The resorting here to the justification in VSVTib points to the fact that VSVTib had, after surviving heavy revision, been con¬stantly consulted by scholars despite its suspicious authenticity.


67
7 Conclusions



1. Among all the Tibetan translations of VS, the one embedded in VSVTib distinguished from the rest in the following aspects:
a. It faithfully adherents to Sanskrit original in word order and phraseology, and due to this, VSVTib is defined by mTsho sna ba Shes rab bzang po as a thad 'gyur of VS.

b. It adopts a policy of compact handling of such grammatical elements as plurality and taddhita suffix, which, echoes the extreme conciseness typical of sutra text.
2. Some of the t
erms employed in VSVTib to paraphrase the comparatively difficult equivalents attested in the pratikas had probably served as the source of the terms in the translation of thesutras preserved in G1, sometimes we find even a borrowing of one whole paragraph:
VS-1.171


VSMSA na civarena kesasmasrvavatarayet || (VSMsA5b3)
VSMSB bhajeta lekham malapakrstyai na civarena kesasmasrv avatarayet | (VSMsB3b3)
VSSMSA na civarena kesasmasrv avatarayet || (VSSBGp.28)
VSTib chos gos gyon bzhin du skra dang kha spu breg par mi bya'o ||
VSSTib chos gos gyon bzhin du skra dang kha spu breg par mi bya'o ||
VSTTib chos gos gyon bzhin du skra dang kha spu breg par mi bya'o ||
VSVyTib gos yod bzhin du skra dang kha spu mi breg |
VSVTib

Note gos dang skra kha spu mi breg |

gos dang skra kha spu mi breg | lus la chos gos gyon bzhin du skra dang khaspu mibreg | (VSVTib) It is easy to see that the paraphrase of the sutra here amounts nearly word by word to the translation presented in G1. The in-formative transmitting of the equivalents, along with the ana-logues VSVTib sharing with P.T.903 and its containing of words and phrases of an archaic nature, suggests, that the translation in VSVTib goes back to a stratum anterior to that to which other translations belong. It may be a text which had not undergone heavy revision in its transmission.


3. Since VSTib and VSTTib were formally registered in the lHan kar and ’Phang thang ma catalogues, the translation in VSVTib, however, due to its bearing of different views had been deliberately brushed aside since it might have given rise to chaos in practice. But it had not totally fallen into obli¬vion. The translation of the root-text along with the inter¬pretation had been constantly consulted by indigenous schol¬ars, although they bear some ambivalence towards its authen¬ticity and authority.


4. To sum up, from the point of view of the textual cri¬ticism, the translation in VSVTib could, in general, serve as a more valuable testimony than that in G1 for our crit¬ical edition of VS. For, in most cases it has faithfully pre¬served the terms chosen and the syntactic construction as¬sumed in the Sanskrit original. In addition, its credibil¬ity is also strengthened by the fact, that in all possibil¬ity, it may be older than any of the Sanskrit manuscripts now available to us.


5. It would seem that, with this case of literature, we have also an unparalleled sample at hand for the study of variant Tibetan translations of a Sanskrit text. The materials concerned are rich, due to the huge size and far-reaching influence of the work; the divergences attested among them are easy to be sorted out, because the genre, the complex culture surroundings had served 69 to magnify them; and the discrepancies among them are con-siderable and informative, because they belong to chronologic-ally separate stratums.



Source