A Sanskrit Fragment of The Ajatasatru-kaukrtya-Vinodana-sutra
A Sanskrit Fragment of The Ajatasatru-kaukrtya-Vinodana-sutra
A Sanskrit Fragment of The Ajatasatru-kaukrtya-Vinodana-sutra
B a n d 35
SÜRYACANDRÄYA
Edited by
Paul Harrison and Gregory Schopen
I N D I C A E T T I B E T I C A V E R L A G • S W I S T T A L - O D E N D O R F 1998
SÜRYACANDRÄYA
Edited by
Paul Harrison and Gregory SchopenQ*»
Contents
Heinz BECHERT
Siglinde DIETZ
Helmut E I M E R
The dKar chag to the "Supplementary" (kha skoti) Volume added
to the Narthang Kanjur 23
Ronald E . EMMERICK
More Verses from the Manjusrinairätmyävatärasütra 33
Richard GOMBRICH
Michael HAHN
Yoshiro IMAEDA
J.W. DE J O N G
Lewis R. LANCASTER
Marek MEJOR
K.R. NORMAN
Lambert SCHMEHAUSEN
Gregory SCHOPEN
Tadeusz SKORUPSKI
Albrecht WEZLER
Klaus WILLE
Peter ZIEME
Although it is c e r t a i n l y not a well-known text, there are good reasons for regarding
the Ajätasatru-kaukrtya-vinodanä-sütra (AjKV) as a M a h ä y ä n a sütra of some
importance in India. In the first place, the AjKV is cited or mentioned in a number
of significant commentarial works. For example, it is quoted on five occasions, in
one passage at great length, in the Sütra-samuccaya (SS), whose attribution to
Nägärjuna (2nd C e n t u r y C.E.?) is upheld by many scholars, though it is not
1
without problems. I n the SS the AjKV is referred to as the Ma skyes dgra 7 le u
(Ajätasatru-pahvarta) or the Ma skyes dgra 7 mdo (Ajätasatru-sütra), with or
without the prefix 'Phags pa (Ärya). There is also a brief citation in the Dazhidu
lun JK"sf§Ltm or *Mahä-prajnä-päramitä-upadesa (T.1509, 340c), but the at-
tribution to Nägärjuna of this work is generally not accepted. The citation itself is
also somewhat problematic, since the title given—Fang bo jing Wl^-ML—suggests
that the reference is not to the AjKV as we know it, but to a smaller sütra which
appears to have been one of the "building blocks" from which the larger text was
2
constructed, and which survives independently in Chinese translation (T.629).
Although the AjKV appears not to have been used by Säntideva, when we come
down t o the 8th C e n t u r y w e find it referred to in Haribhadra's Abhisamayälam-
kärälokä ( W O G I H A R A 1973: 22) as the Ajätasatru-soka-vinodana-sütra, but unfor-
t u n a t e l y n o citations from i t are given. It i s similarly referred t o i n the second
Bhävanä-krama o f Kamalasila (c. 740-795) (see G O S H I M A 1983: 23) and the Rim
gyis jug pa 7* sgom don o f Vimalamitra (fl. late 8th C e n t u r y ) . The number of
3
known references i s thus not large, but the distribution i s interesting, suggesting
as i t does that the AjKV enjoyeö a slight resurgence i n p o p u l a r i t y with Indian
1
The attribution of the SS to Nägärjuna is maintained, for example, by PÄSÄDIKA (1997:
493-494) and LINDTNER (1982: 11, 172-178). For the Tibetan text of the SS see PÄSÄDIKA (1989).
An English translation by the same author was serialized in "Linh-Son"publication d'etudes
bouddhologiques in Issues Nos. 2-20 (1978-1982). The relevant citations from the AjKV in text
and translation are to be found on pp. 21 (= No. 4 [1978], pp. 26-27), 94 (= No. 11 [1980], pp.
37-38), 97 (= ibid, p. 39), 146-154 (= No. 15 [1981], pp. 27-32), and 182 (= No. 19 [1982], p.
54).
2
C f . LAMOTTE(1970: xxxv).
3
See Tibetan Tripitaka Peking Edition, Dbu ma A 402b2 (Vol. 102, p. 173). One notes
that the passage in which the reference to the AjKV occurs—and thus the reference itself—is an
almost verbatim repetition of KamalasTla's text. The overail relationship between these two
compilations may well merit closer attention.
68 Jens-Uwe HARTMANN & Paul HARRISON
Buddhist scholars in the 8th Century. This is, one suspects, because its treatment
of certain themes S t r u c k a chord with other developments taking place in Buddhism
at that time, but to be sure o f this one would have to study more carefully the
4
works in which the text is mentioned.
The second reason for regarding the AjKVas an important document is that it
was among the first M a h ä y ä n a sütras translated into Chinese by Lokaksema in
the late 2nd Century: his version survives as T.626, the Azheshi watig jing H K l t ä
There are i n fact three complete Chinese translations o f the text, the second
having been made by Dharmaraksa in the late 3rd Century—T.627, the Wenshuzhili
puchao sanmeijing ~$C $F.~5L^JH^Hll^S—and the third by Fatian
}
in the
lOth—T.628, the Weicengyou zhengfa jing 7 ^ Ü W l E Ü f L I n scope and basic
structure all three renditions are similar, although Fatian's is an outstanding
example of Chinese bowdlerization (see H A R R I S O N 1993: 152-156), and is peculiar
in other respects as well. I n matters of detail it is so wildly divergent from all
other versions that it can hardly be considered a translation of the AjKVat all, and
is best seen as a free adaptation o f the text, o f little Utility for comparative
purposes, unlike the two older versions. Not only does the antiquity of Lokaksema's
translation make the AjKV historically important, then, but it may also be said
that from the point o f view o f content i t is one o f the jewels o f his collected
works. Rieh in narrative incident, packed with interesting doctrinal elements, and
also complex and demanding i n its more philosophical or theoretical passages,
the AjKV is perhaps the most sophisticated and evolved of the M a h ä y ä n a sütras
translated into Chinese by the Indo-Scythian master. It is, in short, an eloquent
witness to the level o f development which M a h ä y ä n a Buddhism had attained by
the middle of the 2nd Century.
Why then has this text been so neglected? Partly because until now, in order
to study it, we have had to rely on the three Chinese versions and the Tibetan
translation, the 'Phags pa ma skyes dgra 7 'gyod pa bsal ba zhes bya ba theg pa
chen po 7 mdo, which was revised by Manjusrigarbha and Ratnaraksita at the
5
beginning of the 9th Century. None of the commentarial citations and references
reviewed above yields a Single piece o f the original Sanskrit—or perhaps we
should say Indic—text. Recently, however, this S i t u a t i o n has changed. In 1994,
4
That the AjKV had a following in Tibet during the early period is also suggested by its
citation in an unidentified siddhänta in the Stein collection, for which see de LA VALLEE POUSSIN
(1962: 226, Cat. No. 704). Note, however, that the whole paragraph referring to fol. 13 of Cat.
No. 704 has been misplaced: it actually belongs to Cat. No. 705 (cf. the listing for the AjKV'm
the index, where the correct number is given). The short passage quoted is the same as the last
of the five citations in the SS listed above, and could, therefore, have been taken from it. The
AjKVis also the probable source of the much-used image of the lion-eub, on which see JACKSON
(1992).
5
The translation is listed in the IDan (or IHan) kar ma (No. 257) as Ma skyes dgra'i
'gyod pa bsal ba, 1,500 slokas or 5 bam po, so we know it was circulating in Tibet by the
beginning of the 9th Century. None of the available editions of the Kanjur gives any hint as to
the identity of the original translators. The colophons of the Tabo and Newark versions give the
revisers' names as 3äkyaprabha (instead of Manjusrigarbha) and Ratnaraksita.
A Sanskrit Fragment of the Ajatasatru-kaukrtya-vinodanä-sutra 69
b
Sam FOGG, Manuscripts from the Himalavas and the Indian Subcontinent: Catalogue 17,
London, 1996, pp. 46-47.
70 Jens-Uwe HARTMANN & Paul HARRISON
7
We wish to thank Jens BRAARVIG for his help in getting access to the fragment. and Mi-
Martin SCH0YEN for his permission to publish it.
A Sanskrit Fragment of the Ajätasatru-kaukrtya-vinodanä-sütra 71
Transliteration Conventions
Parentheses or round brackets ( ) signify restoration in a gap, Square brackets [ 1
damaged aksaras, pointed brackets < > an addition by us, and curved brackets { }
a deletion by us. A cross + denotes a destroyed aksara, two dots .. denote an
illegible aksara, one dot an illegible part of an aksara. O Stands for the punch
hole, /// marks the point where the fragment breaks off, * denotes the viräma and
I denotes the punctuation mark in the ms. (resembling a horizontal comma).
recto
1 tathaiveyam parsat p a s y ä m i I äha [ k ] . + + O nas tvam mahäräja tarn
kaukrtyam pasyasi I äha yathaiva mamjusnr iyam parsat* pürve caksu .. + + ///
2 mahäräja tathägatena änantaryakärinah O anamtaram narakagatih tat kirn
tvam mamjusnr narakam gamisyasi I äha tat kirn ca mamjusri ta[thä| ///
3 ayam n i r v v ä n a g ä m i I äha no hidam mahä O räja äha tathäbhisambuddhe
mamjusri sarvadharmaih tad apy aham dharmam na samanupasyämi + + ///
4 d h a r m a d h ä t u g a t i y a [na] ca dharmadhätur apäyagä O rni I na svargagämi I
na nirvänagämi I abhltä mamjusrih sar[va]dharmä dharmadhätugatl ///
verso
1 änamtaryagati mamjusri dharmadhätuh änamtarya O täyäm etad adhivaca-
nam I dharmadhätuprakrtikäny änamtaryäni yä änamtaryaprakrtih tat pra .r ///
8
2 yam na yamti [na] svargam I na nirvänam h O mamjusnr äha tarn sästärasya
tvam mahäräja vacanam vilomayisyasi I räjäha näham mam .. ///
9
3 tmyakoti I [bhjütakotl I da[s]itä yä [nairä] O tmyatä na tatra kä cit satvatä
I asamtä mamjusri satvasya na tatra kas cid yo bhisam[skaret]. .. ///
1 0
4 tyamtavi[no]di[ta] .. tvä[m] mamjusri I prahi[n]. O mahäräja k ä m k s ä I
äha tadatyamtaprahinatvän mamjusri I äha tat katham te mahärä[j]. + ///
8
This visarga is probably used as a punctuation mark.
9
The v in the ligature tva is open at the bottom and therefore looks almost like tta. Since
the v is similarly open in nirvänagämi in line 3, the word is transliterated as satvatä.
10
An elision mark appears here, with äh.. being added below the line.
72 Jens-Uwe HARTMANN & Paul HARRISON
11
For ease of reading the square brackets have been removed from our transliteration of
the Sanskrit. Except for obvious errors the Sanskrit has not been corrected, but its salient
linguistic peculiarities are addressed in the footnotes to the text or to the translation o f the
Tibetan version. A füll study of the language of this manuscript is best postponed until all its
surviving leaves have been edited.
12
For the time being we would like to acknowledge the assistance of Dr Valrae REYNOLDS
in supplying a microfilm of the relevant volume of the Newark Kanjur (E) and Dr Ulrich PAGEL
for making available a xerox of the London Manuscript Kanjur (L). The collation of the Tabo
version (A) is on the basis of photographs taken by Dr Cristina SCHERRER-SCHAUB and Paul
HARRISON during the 1993 expedition to Tabo Monastery.
A Sanskrit Fragment of the Ajatasatni-kaukrtya-vinodanä-sütra 73
13
differences between the available witnesses at this point. The resulting edition,
therefore, although lacking a complete apparatus, is adequate for the comparative
study of our manuscript fragment. It w i l l be apparent that there are significant
differences between the Tibetan and the Sanskrit. For this reason, a füll restoration
of those parts missing in the Sanskrit is not attempted. Some conjectures are,
however, presented in the notes to the English translation of the Tibetan version,
with reference also to the Chinese translations, in which the passage can be found
in the following places:
T.626 (Lokaksema), 4 0 2 b l 4 - c l 4
T.627 (Dharmaraksa), 423b27-424a5
T.628 (Fatian), 444b4-c8
The relevant sections of the first two of these translations (T.626 & T.627) are
also presented in the following pages. The text reproduced is based on the Taishö,
but does not follow it exactly, being edited on the basis of the variant readings
provided in the Taishö's critical apparatus, and also freely repunctuated. The
sigla used in the footnotes are as follows:
K: i ^ f f i ^ , Korean Edition of 1151 (base text of TaishÖ)
S: 5fc2|s:, Song Edition of 1239
Y: J C ^ , Yuan Edition of 1290
M: M i n g Edition of 1601
G: K f t ^ H * ^ * ( Ä 5fcfc) " O l d Song Edition" (1104-1148) belonging to
the Library o f the Japanese Imperial Household
N : TE^f^WMM^f: The T e m p y ö mss. (729-) and the Chinese mss. of the
Sui (581-617) and Tang (618-907) Dynasties belonging to the Imperial
Treasure House Shösö-in i n Nara, collectively called the Shögo-zö ( N for
Nara is used instead of S for Sheng to avoid confusion with the Song Edition).
The first characters in each case (jwj, 7f£, etc.) are those which appear in the
Taisfws own apparatus, with the exception of the last edition, for which the
siglum is W.. Finally, the order in which all the different versions are given does
not necessariiy reflect the chronological sequence of recensions of the AjKV.
n
Note, however, that our passage falls right at the beginning of what is marked as bam po
5 in E, L, N and Q, but as bam po 4 in S and T. A and F mark no bam po division at this point,
although A has two double shads with a space between, suggesting some kind of division.
74 Jens-Uwe HARTMANN & Paul HARRISON
wumffm *- m m °
taczwmm * mw
/ / / tathaiveyam parsat
pasyämi I
äha <l>
2
k.+ + nas tvam mahäräja
ÜB o tarn kaukrtyam pasyasi 1
(1) £ G M S Y : o m . KN. 7
äha <l>
aiüf##aHij fiffM
(2) PBI GMSY: M KN.
yathaiva mamjusrir iyam
(3) JÜ KN: om. GMSY.
( 4 ) Note that the Taishö (and parsat pürve caksu(sä) ///
thus presumably all witnesses
collated) repeats the foregoing
sentences at this point, with (1) f|5 GKNSY: M. (1) Note that the fragment
only minor changes in word- (2) Variant forms of this has iyam parsat, while correct
ing, as follows: SCWtSWiä name are given in the text at Sanskrit would require imäm
this point: A l f GS, Ü M Y , parsadam (cf. BHSG 15.15 &
'Sitf K N . These editions are 21.9).
m - ttwttim"This dit- not consistent throughout. The (2) It is difficult to fill the
tography is to be deleted from form rÜüf will be used here. gap; katham would be expect-
the text. (3) X KN: X S GMSY. ed according to the Tibetan ji
(4) KN: GMSY. Itar, but does not help in ex-
:
(5) St GMSY: f K N . plaining the -nas after the gap.
(6) ÖP GMSY: M K N .
(7) i f f l GKMSY: fflg N.
A Sanskrit Fragment of the Ajätasatru-kaukrtya-vinodanä-sütra 75
0
M
1
/// (desitä) mahäräja ta-
thägatena änantaryakä-
i t ° rinah anamtaram naraka-
gatih <l>
tat kirn tvam mamjusnr
narakam gamisyasi I
O äha <l>
tat kirn ca mamjusri ta-
thä(gatena) ///
'jam dpal gyis smras pa I Manjusri said, "Great king, since the
rgyal po chen po de bzhin gshegs Realized One (tathägata) has said that
pas mtshams med pa byed pa n i de l
a person who commits the "immedi-
1
ma thag tu sems can dmyal bar 'gro'o~ ates" w i l l go immediately to hell, w i l l
2
zhes gsungs na I ci rgyal po chen po you, great king, go to hell?"
khyod sems can dmyal bar 'gro 'am I When he had said that, K i n g
de skad smras pa dang I 'jam dpal Ajätasatru replied as follows to Prince
3
gzhon nur gyur pa la rgyal po ma skyes Manjusri, "Manjusri, did the Realized
dgras 'di skad ces gsol to II One awaken fully to any dharma say-
'jam dpal ci de bzhin gshegs pas ing, T h i s one goes to a bad rebirth,
3
' d i ni ngan song du 'gro ba'o II ' d i ni this one goes to heaven, and this one
mtho ris su 'gro ba'o II ' d i ni mya goes to n i r v ä n a . ' ? "
4 5
ngan las 'da' bar 'gro b a ' o zhes I He said, "No, great king."
chos gang yang mngon par rdzogs par
sangs rgyas sam I
smras pa I (1) The five änantatya, or "(offenses in-
rgyal po chen po de ma yin no II volving) immediate (retribution)" are matri-
cide, patricide, killing an arhat, provoking
dissension in the Sahgha, or causing the Tath-
(1) ni FLNQST: ni I A , om. E; (2) 'gro'o ägata's blood to flow. See BHSD, s.v.
AELST: 'gro FNQ; (3) 'gro ba'o AFLNQS: (2) In this sentence the Sanskrit and
'gro'o ET; (4) 'da' bar ALNQST: 'das par Tibetan versions deviate: in the Tibetan text
EF; (5) 'gro ba'o FLNQST: 'gro ba'o I A , Manjusri is asking the question ("Will you,
'gro'o II E. Great King, go to hell?") which fits the
following reply of Ajätasatru, while in the
Sanskrit Ajätasatru appears to be asking
Manjusri the same question. However, this
makes no sense: it is likely that manjusri was
written in error for mahäräja, and that we
should emend the text accordingly. Note that
in T.626 this is split into two questions: "Have
you heard that the Buddha has said...?" and
"Do you know, king, that you will go to hell?"
T.627 is closer to the Tibetan and to the
Sanskrit.
(3) In the Sanskrit there is nothing which
corresponds to the stock phrase in Tibetan de
skad smras pa dang / jam dpal gzhon nur
gyur pa la rgyal po ma skyes dgras 'di skad
ces gsol to.
78 Jens-Uwe HARTMANN & Paul HARRISON
ig ° äha <l>
mr-mmm « m m tathäbhisambuddhe mam-
1 2
jusri sarvadharmaih tad
apy aham dharmam na
samanupasyämi ///
/// dharmadhätugatiya na
3
ca dharmadhätur apäya-
gämi I na svargagämi I na
nirvänagämi I
(1) S J I G K M S Y : M A N .
(1) The Sanskrit text here
(tathäbhisambuddhe ... sarva-
dharmaih) is problematic in
terms of case usage. There are
several possibilities, but inter-
pretation as a poorly Sanskri-
tized "instrumental absolute"
(perhaps from abhisambud-
dhehi sarvadharmehi) seems
most plausible, since instru-
mental for locative is well at-
tested in Buddhist Hybrid San-
skrit (cf. BHSG, 7.30-7.34; cf.
also BHSD, s.v. abhisambu-
dhyate). The correct Classical
Sanskrit equivalent would thus
be tathäbhisambuddhesu ..sa-
rvadharmesu. Tibetan chos
thams cad la also suggests the
locative, as does T.627.
(2) Tad apy ... dharmam
should read tarn apy ...dha-
rmam, unless this is a case of
a neuter modifier with a mas-
culine noun, cf. BHSG 6.14.
The expression is found fre-
quently enough, e.g. in Chap.
1 of the Astasähasrikä-prajnä-
päramitä-sütra{tam apy aham
bhagavan dharmam na sam-
anupasyämi yaduta prajnä-
päramitä näma, etc., VAIDYA'S
ed., p. 3).
(3) Read sünyatägatiya na
ca sünyatä ? See note to trans-
lation.
A Sanskrit Fragment of the Ajätasatru-kaukrtya-vinodanä-sütra 79
/// ä n a m t a r y a g a t i mam-
j u s r i d h a r m a d h ä t u h än-
A 1
° § t $ £ ^ t £ & a m t a r y a t ä y ä m etad adhi-
vacanam I dharmadhätu-
# ° w&<fcM&* ° e prakrtikäny änamtaryäni
frlJi# * tt^AMW * yä änamtaryaprakrtih tat-
pra(k)r(ti)///
/// ( a p ä ) y a m na yamti na
(1) A G M S Y : y K N . svargam I na nirvänam Ii'
(1) ÜS GMNSY: W. K.
(1) Here II represents the vi-
sarga used as a punctuation
mark.
A Sanskrit Fragment ofthe Ajatasatru-kaukrtya-vinodanä-sütra 81
A i ^ umm¥°
^B«
III (a)tyamtavinoditatvam
<l>
mamjusri {I} äh(a) <l>
HB - p r a h i n ( ä te) m a h ä r ä j a
kämksä I
äha <l>
tadatyamtaprahinatvän
<l>
3E g ° mamjusri {I} äha <l>
tat katham te mahäräj(a)
M ° ///
E M °
HB °
(1) Ü G M S Y : Ä K N .
(2) It is possible that the text
is corrupt at this point.
Abbreviations
AjKV Ajätasatru-kaukrtya-vinodanä
BHSD Franklin EDGERTON, Buddhist Hybrid Sanskrit Dictionaiy, New Häven, 1953.
BHSG Franklin EDGERTON, Buddhist Hybrid Sanskrit Grammar, New Häven, 1953.
T. or Taishö TAKAKUSU Junjiro & WATANABE Kaigyoku, eds., Taishö shinshü daizökyö,
100 vols., Tokyo, 1924-1935.
Works Cited
CONZE, Edward
1975 The Perfection of Wisdom in Eight Thousand Lines & Its Verse Summary, 2nd
ed., Bolinas.
GOSHIMA Kiyotaka
1983 The Tibetan Text of the Second Bhävanäkrama, Kyoto.
HARRISON, Paul
1993 "The Earliest Chinese Translations o f M a h ä y ä n a Sütras: Some Notes on the
Works o f Lokaksema," Buddhist Studies Review, 10, 2, pp. 135-177.
JACKSON, David
1992 "Birds in the Egg and Newborn Lion Cubs: Metaphors for the Potentialities and
,,
L i m i t a t i o n s o f " A l l - a t - o n c e " Enlightenment, in IHARA S h ö r e n and
YAMAGUCHI Z u i h ö , eds., Tibetan Studies: Proceedings of the 5th Seminar
of the International Association for Tibetan Studies Narita 1989, Narita,
V o l . I , pp. 95-114.
LAMOTTE, Etienne
1970 Le Traite de la grande vertu de sagesse, Tome I I I , Louvain.
LINDTNER, Christian