133

Further Remarks on the New Manuscript

of the Dirghagama

Jens-Uwe Hartmann
Three years ago, Sadakata Akira published a short paper on two
folios of a manuscript belonging to the recent finds from the area of
"Greater Gandhra", to use Richard Salomon's term, and reported to
have been found in Northern Pakistan.' Since the middle of the nineties
of the last century, bundles of this manuscript had begun to reach the
Western rare book market and, luckily, from the very beginning scholars
were gran ted access to the originals and to reproductions in various
forms. Already the first bundle gave rise to the supposition that it
belonged to a manuscript of the Dirghgama,2 the "Collection of Long
(Discourses of the Buddha)", in a Sanskrit version, and in 2000 abrief
survey of that first bundle was published. 3
Due to its size (about 50 X lOcm) and its state of preservation the
manuscript stands out among the scrolls, leaves and thousands of
fragments of Buddhist manuscripts in Indian languages which have
reached Europe, Japan and the USA within the last five years from the
Northwestern corner of the former Indian Buddhist world and have
been made available for scholarly study. One of the later bundles inSadakata 1999.
2 In autumn 1998 Matsuda Kazunobu and Klaus Wille recognized independently of each other that the manuscript apparently contained stras
of the Dirghgama, Matsuda when viewing the original folios in London,
and Wille with the help of photographs.
3 Hartmann 2000.
1

-98-

New Manuscript of the Dirghgama (Hartmannl

New Manuscript of the Dirghgama (Hartmannl

c1uded the last part of the manuscript with a final colophon preserving

preceding section. The folio conc1udes with two very brief colophons,

Dirghagama as the title of the whole work. This conveniently confirmed

one for the section and one for the entire manuscript.

134

135

the previous supposition and underlined the importance of the text.

The text area of the verso side is interrupted four tim es , once m

Since only a small part of the contents and structure of this Sanskrit

the left third by the space left empty for the punch hole and three times

version of the "Collection of Long Discourses" have so far beenrecovered

by roundels consisting of four concentric circ1es, the interior of each

from the Sanskrit fragments found in Central Asia, the principal results

being decorated with a different flower motiL While the other folios

from a study of the newly arrived folios will be communicated in the

mostly contain eight lines, the text remaining for the verso side of this

following pages.

one comprises only five and a half lines. Regrettably, the upper layer
of birch bark has peeled off in three places, which results in several

1. The final folio

Regrettably, the manuscript as it has reached the West is still by
no means complete, and due to the present political developments

gaps in the text.

1.1. Transliteration of folio 454 verso'
1

[a] .. [rod]d[nam* 11 ] 0 ssvato py ahetus ca antak amars

(autumn 2001) and the vagaries of the art and book market it remains

tath . sa:rp.jiii caivpy asa:rp.jiii ca 0 tath uccheda [n }irvrtalf

unpredictable whether all its surviving parts will ever be known or

unmajjanti ca sa:rp. .... p[r]atitya 0 spa[r]s[o] bhi

made accessible to inter es ted scholars. So far, about 160 folios-com-

k~;u[I)] 11011 hetvathabrahmailam* 11 11 uddnam* suka jivaka

plete or in fragments-have been ascertained; they belong mainly to

o rj ca vasi~thah ks[y]apena [ca] + [hm]. [j] ++ 0krtv

the seetions with the folio numbers 72-116, 264-330 and 385-454 and

ca

amount to roughly one third of the whole manuscript, since the final
folio must bear the page number 454 (see below). This last folio is of

virgo bha 0 vati samudditalf 11011 trdaI).91 piriga 0 l .... [dv J.

of key words which, unlike an antaroddna, lists the text titles of the

4 The text was read together with my friend Klaus Wille, Gttingen;
Lance Cousins and Somadeva Vasudeva, Oxford, contributed some corrections of our readings and various very useful suggestions; Paul Harrison,
Christchurch, corrected the English of this paper. I wish to thank all of
them. Moreover, I am grateful to the present owner of the manuscript for
putting excellent black and white photographs at my disposal and granting
permission for their publication. - Parentheses or round brackets ( l
signify restoration in a gap, square brackets [ ] damaged ak:pras or
uncertain readings, pointed brackets < > an addition without gap, three
oblique dashes/// mark the point where the fragment breaks off; a cross
+ denotes a destroyed ak~ara, two dots .. denote an illegible ak~ara, one
dot denotes an illegib1e part of an ak~ara, the asterisk * denotes the virma;
o stands for the punch hole and 0 for the circular decorations.

-97-

-96-

particular importance for our understanding of the structure of the

text, and therefore it will be introduced first. On the recto side it contains text of the final part of the Brahmajlastra, the Sanskrit parallel to the Brahmajlasutta of the Dighanikya. This points to a
considerable difference in structure, since this sutta opens the Pli
collection, while it falls last in the present Sanskrit version, as becomes
clear now from the new manuscript. The verso side of the folio begins
with an antaroddiina, i.e., a verse-like list of key words from the preceding stra; then follows a longer uddiina, i.e., another verse-like list

[ca] lohityabha .i
4

+ k .. e .i Ca]

0 mahalli pr
@ h kasyapa

s caiva @ brahmajalam a
[pJ. [t]. s samaptal).

pta @ s ca dirgha

11

nopama

0
11

11

++++

[v].

11

Middle Indic language ( s). Beyond such normal devia tions from the
usage of classical Sanskrit -

@ .......... silaskandhani-

sama

0 + mal).

137

be explained by the transformation of texts originally composed in (a)

.. @ marffisa

[s ta]c ca bhi @ k~u~u bha~ate .

ko bhavati sa .. " ..... [i]~tha

New Manuscript of the Dirghagama (Hartmann)

New Manuscript of the Dirghagama (Hartmann)

136

which was never intended, to be sure -

the verses are apparently defective in several places, as will become clear
from a closer look at the section summarized in them.

The antaroddna is included in the separate Tibetan translation

of the stra edited by Friedrich Weller (Weller 1934), forming 223

1.2. Reconstruction

of his edition, and also in the second Tibetan translation which appears

For the sake of a more lucid presentation, uncertain and problematic

as a citation of the complete stra in Samathadeva's Abhidharmako-

readings are not signified by square brackets in the reconstruction. The

satikopayika.' This second version of the antaroddna is shorter than

verses are provisionally numbered in order to facilitate reference in the

the first and deviates in two places. Without a more detailed discussion

discussion of each key word.

of such phenomena, the Tibetan translations of each key word and the

a(nta)roddanam

number of the respective section of the stra they refer to, as already

(1)
(2)

11

sasvato 'py ahetus ca antaka amaras tatha

will be given here.

unmajjanti ca sarp(kli~tal).)

ssuato 'py : rtag dan gcig pu (Weller 1934) : rtag dan phyogs gcig

hetva <r >thabrahmajalam

11

uddanam
(3)

pointed out by Weller in his translation (Weller 1935-36,p.38 [= p.679]),

sarpjni caivapy asarpjni ca tatha uccheda nirvrtal). < 11 >

pratitya sparso bhik~uIJii

11

(Upayika). In the Sanskrit text, there is no basis for gcig pu "zum

11

Teile (ewig)" (Weller), or phyogs gcig, but the pada is metrically

11

suka jivaka raja ca

v<a>si~thal:).

kasyapena ca <I>

deficient. Weller refers to 26ff. and 57ff.

(bra)hm(a)jaOena) krtva ca v<a>rgo bhavati samuddital).

(4)
(5)

trdarffi pirigala(treyo)

dv(e) ca lohityabha(~)i(tau) <I>

k .. e .i a( tha) mal)91Sas

tac ca bhik~u~u bha~ate

mahalli pr(~thapalas ca)

brahmajalam anopama

........ " Silaskandhanip(a)t(a)s samaptal).

11

11

11

ahetus ca: rgyu med dan (both translations); Weller refers to 90

ff.

antak: mtha' ldan : mtha' yod; Weller refers to 108ff.

11

amars tath : de bzin lha mi spon : de bzin 'chi med dan; Weller refers

v. ko bhavati sa " .... <I>

(vas)i~thal). kasyapas caiva

(ga)mal).

11

11

samaptas ca dirgha-

to 135ff. The separate Tibetan translation apparently renders

the full expression amarauik$epa6 which has to be abbreviated in
the Sanskrit text for metrical reasons.

11

saT(l-fiii caiupy asaT(l-fiii ca : 'du ses 'du ses med pa dan : 'du ses bcas
1.3. Commentary

dan 'du ses med

The uddnas display the metrical licence and the usual liberties

with regard to sandhi and word forms which are, at least partly, to

-95-

11

'du ses yod min med min; Weller refers to

The uddiina is found in TP vol. 118, p. 169/2/1-2

For this term cf. Hartmann 1989, p. 49 ad VI.
-94-

fol. 177al-2.

138

New Manuscript of the Dlrghgama (Hartmann)

New Manuscript of the Dlrghgama (Hartmann)

159ff. Only the Upyik lists all the four possibilities.

tath uccheda: chad dan : chad; Weller refers to 183ff.

sutra.

rj ca: a reference to the sutra corresponding to the Smannaphala-

nirurtal} : mya nan 'das pa dan: mya nan 'das dan; Weller refers to
190ff.

sutta (DN 2).

uasi;;thal}: probably to be corrected to usi.thal}; it denotes the parallel

unmajjanti ca saT[L(kli!l}) : dmyugs dan kun nas rmons pa dan:

gyens pa dan

139

11

yan dag rmons pa zes bya dan; Weller refers to

to the Tevijjasutta (DN 13).

kSyapena ca: this is the key word for the text corresponding to the

207ff. A fragment of the Brahmajlasutra from Central Asia

Kassapasihandasutta (DN 8).

suggests theform sT[Lklesylf (cf. Hartmann 1992, Hs. 16 R4),

(bra)hm(a)j(lena) krtu ca: the parallel to the Brahmajlasutta

but the a of saT[L in the present manuscript is clearly short, and

(DN 1), the first text of the Pli and the last one of the Sanskrit

therefore saT[1( klitl}) is reconstructed.

version; the reconstruction follows corresponding lines in the

pratitya: brten dan: brten nas byU7i. dan; Weller refers to 212ff.

uddnas of the Ekottarikgama, cf. Tripathi 1995: 20-22 ( 16.0,

sparo : reg pa; Weller refers to 214ff.

22.56,35.0 and 36.0)7.

bhik;;uTJ: dge slon dan: dge bas so; Weller connects the word with

u<a>rgo bhauati samuddital}: "(with this) the sechon is summarized"

220, adding a queshon mark, but this passage appears to be the

only possible point of reference. In the Upyik, the antaroddna

(samuddita probably m.c. for samudita, cf. BHSD s.v.).

trdaryjipiT}gal(treyo) du(e) ca lohityabh()i(tau): with this half

ends with this key word.

verse another key word list begins. Apparently the Silaskandha

hetua<r> thabrahmajlam: according to Tibetan rgyu dan don chos dra

sechon of the DA began with the TridaI).<;lisutra, and there is reason

ba ste, tha ma tshans pa'i dra ba'o (Weller 223) one would

to assume that now all the sutras in that sechon are summarized

expect averse line like hetuarthadharmajlaT[L brahmajlaT[L ca

in the form of an uddna. If this assumption is correct, several key

pascimam. Weller refers to 221.

words must have been omitted, as will be shown below.

The half verse is apparently quoted by Samathadeva in his

The antaroddna is followed by the uddna verses listing a key

Abhidharmakosatikopyik in order to elucidate the placement of

word (in most cases probably the short htle) for each stra in order

the first Lohityastra. In this commentary, which is preserved only

to establish their sequence.

in its Tibetan translation, he explains: phun po las I dbyug (dbyu

suka: this denotes the parallel to the Subha-sutta of the Dighanikya

D) gu gsum dan gon bu gsum

(DN no. 10).

11

(P

I )

Zu he (hi D) ta yi (ta'i P)

mam bsad giiis I ( I deest P) zes bya ba'i lu he (hi D) ta'i mdo

jiuaka: Jivaka Kumrabhrta appears as the interlocutor of the Buddha;

without parallel in the Dighanikya; the correspondence is found
in the Majjhimanikya (MN I 368-371). Cf. SHT VI 1525, line R3
which also seems to preserve the transition from this to the next

-93-

I owe these references to my friend Fumio Enomoto.

-92-

140

New Manuscript of the Dlrghagama (Hartmann)

New Manuscript of the Dlrghagama (Hartmann)

141

dmi po la, etc. 8 Since it follows from other quotations that dbyug

sponding to the Poghapdasutta (DN 9), although it becomes

gu gsum serves as a translation of TridaI)!;lin, here it must denote

the TridaI)!;listra. Therefore the reference phun po las surely has

obvious from the manuscript itself that three more stras, namely
SroI)8.tWya-, Ktatl).g.ya- and Amb~thastra, are arranged

to be taken as an abbreviation of tshul khrims kyi phwi po las,

between Mahalli- and

(sila)skandhikiiyiim or the like. Neither for the TridaI)!;listra nor

defective here. Since the reading of the last preserved

Pr~thaplastra.

The uddiina is evidently

ak~ara

is

for the Pingaltreyastra (the key word is reconstructed according

unambiguous, the name must be reconstructed as PT( !2.thapiila) in

to the name as it is preserved in the stra itself) is a parallel text

accordance with the form attested in the text of the stra and in

attested in the Pli canon, and only one of the two Lohityastras

another uddiina. One could ass urne that here we are confronted

appears to have found its way into the DN (Lohiccasutta, no. 12).

with an unnamed Piwoddna, i.e., a summary of several uddiinas

k .. e .i: according to the sequence of the stras in the manuscript, a key

which selects the first key word of each (cL Panglung 1980), but

word for the text corresponding to the Kevaddhasutta (DN 11) is

there are two reasons against this: first, the fact that PT!2.thapiila

ak~ara

remains

appears as the second key word in the corresponding uddiina (cL

seem to point in the direction of the expected word, but the recon-

below) and, second, the fact that in the f ollowing line three con-

struction of kaiuarti is definitely exduded.

secutive stras are enumerated.

to be expected, e.g., kaiuarti, and the preserved

a( tha) mar:uJisas: the introductory part of the stra suggests a corre-

Part of the text omitted here can be reconstructed from other

spondence with the Jliyasutta (DN 7); in the Pli version two

passages. On folio 410 of the manuscript, an uddiina is found in

Amb~thastra,

wandering ascetics appear, namely Mal).<;lissa and Jliya, but in the

the lines rl-2 between the Ktatwya- and the

Sanskrit text only Ma:ry;lisa.

which is repeated at the end of the section on folio 430r8. In both

tac ca bhik!2w}u bhii!2ate: "and (then the Buddha) speaks this (stra

also) among the monks". This must be a reference to the second

cases the text is only partly preserved, but the sequence becomes
dear.

version of the Ma:ry;lisastra, inwhich the Buddha reports his

encounter with the wandering ascetic and his disco ur se with hirn

Folio [4]10r 1-2:

to the monks. Due to this repetition, the stra is summarized in

auaT[L!2t ha l} PT!2.t ha / / /

the manuscript by means of the usual abbreviations in roughly two

'" [ta]ma iina[ndo] bhik!2usiistaril} II

Folio 430r8:

lines (fol. 391 v6-8).

mahalli: this denotes the parallel to the Mahlisutta (DN 6).

[a] T[L[ b] ii!2.tha PT!2thapiilas ca

PT( !2thapiilas ca): apparen tly, this key word ref ers to the text corre-

sruta<T[L> ma(ha) Cl] (Z) [0

TP vol. 118, p. 265f. = vol. thu, fol. 119a8-b1; TD vol. nu, fol. 74al-2;
Honj Yoshifumi already saw lu he ta as corresponding to Lokecca, cf.
Honj 1985, p. 782.
8

-91-

kiiTal}O-uiidi ca pudgaliil}

nya] tama anando <bhik!2u> siistii-

ni II

This attests to the sequence of Amb~tha, Pr~thapla, KraI)8.vdin,

Pudgala (on four kinds of persons) and Sruta. The rest of the line
-90-

142

New Manuscript of the Dlrghgama (Hartmann)

New Manuscript of the Dlrghgama (Hartmann)

ak~aras

143

as vargo is impossible. Apart from

must refer to the following two stras, the first on another old

reading the first two

brahmin and the second arepetition thereof. It seems odd that

that, one would expect such a line as the last pda of an uddna

nearly half averse is used for two texts, but no other explanation

verse, but hardly as pda b which it appears to be in the present

suggests itself at the moment. Obviously, at least pda d (nando

case. In the face of the obvious break in the sequence of texts

bhik!?usstri?) refers to a kind of duplicate of the preceding stra

indicated by the following key words - at least four stras remain

at the end of which it is said (fol. 430r7):

11

ya[ th] jin;w vrddho

unmentioned -

one suspects that part of the text is omitted in the

mahallaka evam anyatamo brhmalJlJ yu!?mn nando sst

verse. It seems difficult to find another explanation for the inter-

bhik!?ur;JiTf/- bhate

ruption of the sequence, and the reconstruction of the line is at

11 ;

it is not clear which text anyatama belongs

present better left open.

to, since both of them have this word in their introduction.

The sequence of stras between Mahallin and Amb~tha can

(vs) i!?thalJ: evidently the enumeration jumps to the text corresponding

also be confirmed from another uddna, but so far only from one

to the Tevijjasutta (DN 13), reconstructed here with reference to

preserved in a Central Asian Sanskrit text. The fragmentary folio

the name as preserved in verse 3b.

c of manuscript SHT V 1290 of the German Turfan finds contains

ksyapas caiva: again this key word denotes the text corresponding

the end of the Ktatl).9-ya- and the beginning of the Amb~tastra.

to the Kassapasihandasutta (DN 8). A comparison of verses 3

In line R8 (p. 210) of fragment c, parts of the uddna are preserved:

and 5 demonstrates clearly that different wordings are (or, perhaps

[pha] .. s01JtlJthyas ca kutatv-?hyena pascimlJ 1'///. At the

better, can be) used for those uddnas summarizing a section and

beginning of the line traces of two ak~aras are preserved, the first

those summarizing a larger part or a whole work.

of which was read as [pha] by the editors of the volume. However,

brahmajZam anopam: the first word refers of course to the Brahma-

the graphically very similar [ha] is equally possible, and with a

jlastra with its parallel in DN 1. The second word is most likely

high degree of prob ability the following syllable contains an 1.

'0

to be taken as an attribute - in the senseof anupamam, "matchZess"

Therefore it is very tempting to reconstruct the also metrically

with lengthening for metrical reasons, cf. BHSD s.v. anopama -

fitting title Mahalli which would result in the following half sloka:

since Samathadeva who quotes the Brahmajlastra as the last

(ma) haZ( Zi) sOlJtlJ-thyas ca kutatlJ-thyena pascimlJ.

v. ko bhavati sa ...... : the textual remains recall pda 3d v <a> rgo
bhavati samudditalJ, ,,( with this) the seetion is summarized", but

text in the Silaskandha section,ll confirms that no other stra can

possibly follow .

.. .. .. .. .. silaskandhanip()t(a)s samptalJ: possibly to bereconstructed

as 11 dirghgame silaskandhanip()t(a)s samptalJ "the Silaskandha

It is difficult to decide if this sign is to be understood as a number

or as a punctuation mark; if it is to be taken as a numerical symbol, it
may possibly indicate a numbering of the uddnas within the Silaskandha
section.
10 CL the facsimile in SRT V, plate 85.
9

-89-

seetion (in the Dirghgama) is completed". This colophon concludes

the last seetion of the Dirghgama. Previously, its Sanskrit name
11

Honj 1985, p. 783.

-88-

144

New Manuscript of the Dlrghagama (Hartmann)

was known in the form of SHaskandhik only from a quotation in
the

Abhidharmakosabh~ya

New Manuscript of the Dlrghagama (Hartmann)

145

folios 410-416) are inserted in a later section and probably numbered

which is explained in the Vykhy as

as folios 442-444. This can only be explained if one assurnes that the

silaskaT[Ulhik nma nipta!J, "the section named Silaskandhik"; 12

leaves were first copied, then mixed up, wrongly rearranged and only

the corresponding Pli term is Silakkhandhavagga.

then finally numbered. 15 The last folios of the manuscript display the

samptas ca dirgh(ga)ma!J: "the Collection of Long (Discourses) is

following situation: on folio 451r2 the Ksyapastra begins, which

also completed". The terms for section, nipta in Sanskrit and

should cover a number of leaves; the next folio 452, however, contains

vagga in Pli, and also the terms for the collection as a whole are

text from the Brahmajlastra, but not from its beginning; it sets in

different: in the Sanskrit version gama is chosen, while the Pli

with the section corresponding to 151 in Weller. The next folio 453r

prefers nikya (although gama in this context is by no means

connects directly with 452v; it has not yet been separated from the final

unknown to the Pli tradition).

folio 454, and therefore no photographs of folios 543v and 454r are
available to uso The remaining text of the Brahmajlastra ( 206-223

2. Lass

01 text

in the manuscript 13

in Weller's editon) fits quite weIl into this gap, and we assurne that
the final part of the stra is completely preserved. In the first line of

In my earlier paper I observed that text seemed to be missing in

the manuscript, without being able to explain that omission. The
apparently missing text was a whole stra, the Prasdaniyastra (Pli

the verso side of folio 454 the antaroddna begins ( 223 in Weller).
The various gaps in the uddnas have already been referred to above.
It appears that the manuscript was copied in a way which aimed

Psdikasutta), which could not be found in the place indicated by the

at reproducing its exemplar folio for folio. This is indicated by the

uddna and by the parallel in the Central Asian manuscripts.!4 Since

sometimes fairly copious spacing between the

then, considerable textual gaps have been noticed in the last part of the

of a page or by the filling up of the last line with strokes. Both serve

manuscript, and therefore it is now reasonable to assurne that most

as a means for reproducing justified margins in order to conform to

ak~aras

in the last line

or all of these losses occurred in the transmission process of the text.

the format of the exemplar.!6 Such phenomena will have to be studied

Some of the missing passages may be recovered in rather unexpected

very carefully throughout this manuscript, but also in comparison with

places, once the manuscript is better studied and more folios have

other similar manuscripts. At the moment there is reason to suspect

become available. Lance Cousins kindly drew my attention to the fact

that some folios had already been lost in one of our manuscript's

that three leaves of the Amb~thastra (which basically covers the

exemplars, and without attempting to recover them from other manuscripts in the process of copying, the folio numbers were simply applied

Abhidh-k-bh(P) 255,13 with Abhidh-k-vy 420,18.

13 This paper was first read on March 29, 2001, at the 28th Deutscher
Orientalistentag, and I wish to thank the participants for the stimulating
discussion of this point.
14 Hartmann 2000, p. 364.

15 Cousins refers to the same phenomenon reported by Andrew Skilton

in arecent paper, cf. Skilton 2000, p. 1l.
16 For similar observations on Khotanese manuscripts in Central Asia
cf. Sander 1988, pp. 547f.

-87-

-86-

12

146

New Manuscript of the Dirghgama (Hartmann)

New Manuscript of the Dirghgama (Hartmann)

147

to the remaining leaves. This does not yet explain the fragmentary text

4 . Lohitya TI

?-386rl

of the uddnas, but here, too, it will be necessary to take all the evidence

5. Kaivartin

386rl-390vl

Kevaddha, DN 11

from the available folios into consideration before we can reach a better

6. MaI).gJ.sa I

390vl-391v6

Jliyasutta, DN 7

understanding of the textual developments within this particular man-

7 . MaI).gJ.sa TI

391 v6-8

without correspondence in the DN

uscript.

8. Mahallin

391 v8-396v6

Mahli, DN 6

3. List of the sutras in the Silaskandha section

without correspondence in the DN

9. SroI).atWya

396v6-40lrl

SOI).ada$, DN 4

10. Kutatwya

40lr2-409v8

Kutadanta, DN 5

11. Amb'i'tha

410r2-416r3

Ambattha, DN 3

12. Pr'i'thapla

416r3-423(?)v7Potthapda, DN 9

enough information for an attempt at reconstructing the sequence of

13. KraI).avdin '8

424r4-424v3

without correspondence in the DN

sutras within the Silaskandha section with a fairly high degree of cer-

14. Pudgala

424v3-426vl

without correspondence in the DN

tainty. In the following list, the first and/or the last folio of each

15. Sruta

426vl-427v5

without correspondence in the DN

The uddnas and the available folios of the manuscript provide

sutra is mentioned, if preserved and determinable, and a reference to

16. Mahalla

427v6-430r7

without correspondence in the DN

the Pli parallel in the DN is added. The remark that no corresponding

17. Anyatama

430r7

without correspondence in the DN

text is traced in the DN only means that no sutta appears in the DN

18. Suka

430r8-433r2

Subha, DN 10

with the same or a similar title and with corresponding interlocutor(s).

19. Jivaka

433r2-?20

without correspondence in the DN

The exact relation between the so far unknown Sanskrit texts and the

20. Rj

?-447(?)v2

Smannaphala, DN 2

'9

suttas of the Pli Nikyas will have to be darified in the course of an

(three folios [442-444J of the Amb'i'tha and one [445 J of the

edition and dose study of each sutra. According to the present state

Brahmajla are inserted here)

of knowledge the Silaskandha section contains altogether 23 texts as

21. Vsi'i'tha

447(?)v2-451rl Tevijja, DN 13

opposed to 13 suttas in the Silakkhandhavagga of the DN -

22. Ksyapa

451r2-v8

texts of the Pli find a correspondence in the Sanskrit -

all the 13

and 9 in that

Collection of Long Discourses which is only preserved in Chinese translation and generally ascribed to the school of the Dharmaguptakas. '7
1 . TridaIJ.9in fol.

?-367r3

2 . Pingal treya

367r4-?

without correspondence in the DN

3. Lohitya I

?-?

Lohicca, DN 12

17

without correspondence in the DN

However, cf. Boucher 2000, pp. 67f.

-85-

Kassapasihanda, DN 8: cL above
for the discussion of the omission of text

23. Brahmajla

452rl-454r

Brahmajla, DN 1: cL above for

the discussion of the omission of text

18 The following key words are omitted in the uddiina, but cf. the
commentary on the key wordpr(i?thapiilas) above.
19 The reading of this and the following key word are not fully sure,
cf. the remark on the key wordpr(i?thapiilas ca) above.
20 The transition from this text to the next is preserved, but the folio
is damaged at the left margin, and therefore the folio number is not preserved.

-84-

148

New Manuscript of the Dlrghgama (Hartmann)

New Manuscript of the Dlrghgama (Hartmann)

Abbreviations and Literature

Abhidh-k-bh (P)

schen Texte aus den Turfan-Funden, Beiheft 2), 37-67.

Abhidharmakosabh'jyam 0/ Vasubandhu, ed. P.

1992

Pradhan, rev. 2nd ed. Aruna Haldar, Patna 1975.

Abhidh-k-vy Sphu!iirth Abhidharmakosauykhy by Yasomitra, ed.

2000

Franklin Edgerton, Buddhist Hybrid Sanskrit Dictionary, New

stine Chojnacki, J.-U. Hartmann u. Volker M. Tschannerl, Swisttal-Odendorf 2000 (Indica et Tibetica, 37), 359-367.

The Digha Nikya, ed. T.W. Rhys Davids, J. Estlin Carpenter, 3

Vols., London 1890-1911 (Pali Text Society).

SHT V - VI

Honj Yoshifumi
1985

by Samathadeva"), Indogaku Bukkygaku Kenkyu 33, 783-779.

Sanskrithandschri/ten aus den Tur/an/unden, Teil 5: ed.

Panglung, Jampa Losang

Stuttgart 1985, 1989 (Verzeichnis der orientalischen Handschriften

1980

"Preliminary Remarks on the Uddnas in the Vinaya of the

MUlasarvstivdin", TibetanStudies in Honour 0/ Hugh Richard-

The Sde-dge Mtshal-par Bka'-'gyur, A Facsimile Edition 0/ the

son, ed. Michael Aris and Aung San Suu Kyi, Warminster, 226-

18th century redaction 0/ Si-tu Chos-kyi-'byun-gnaspreparedunder

232.

the direction 0/ HH. the 16th Rgyal-dban Karma-pa, 103 vols.,

Sadakata Akira

Delhi 1976-1978; Sde-dge Bstan-'gyur Series, Published as Part 0/

1999

"Girugitto syahon: Tenson-ky danpen no kaidoku" ["The Gilgit

the Dgons-rdzogs 0/ HH the Sixteenth Rgyal-dban Karma-pa, 213

Manuscript: Deciphering the Mahagovindasutra Fragment" ] ,

vols., New Delhi 1982-1986 (reprinted together as The Tibetan

DaihOrin, January 1999, 30-35.

Tripitaka, Taipei Edition, vol.1fL, Taipei 1991fL).

TP

"Upyik shoden no J-agon" ("The Dirghgama Transmitted

Lore Sander und Ernst Waldschmidt, Teil 6: ed. Klaus Wille,

in Deutschland, X, 5-6).
TD

"Bemerkungen zu einer neuen Handschrift des Dirghgama",

ViuidharatnakaraTJ4aka. Festgabe fr Adelheid Mette, ed. Chri-

Haven 1953.
DN

Untersuchungen zum Dirghgama der Sarvstivdins, Gttingen

1992 (still unpublished Habilitation thesis).

Unrai Wogihara, Tokyo 1932-1936.

BHSD

149

The Tibetan Tripitaka, Peking Edition (repr.), ed. Daisetz T.

Sander, Lore
1988

Suzuki, 168 vols., Tokyo, Kyoto 1955-1961.

"Auftraggeber, Schreiber und Schreibeigenheiten im Spiegel

khotansakischer Handschriften in formaler Brhmi", Studia

Indogermanica et Slavica. Festgabe fr Werner Thomas zum 65.

Boucher, Daniel
2000

Geburtstag, ed. Peter Kosta a.o., Mnchen (SpeciminaPhilologiae

Review of Richard Salomon, Ancient Buddhist Scrolls From

Gandhra, Washington 1999, in Sino-Platonic Papers 98, 58-71.

Hartmann, Jens-Uwe
1989

Slavicae, Supplementband 26), 533-549.

Skilton, Andrew
2000

"The Letter of the Law and the Lore of Letters: The Role of

"Fragmente aus dem Dirghgama der Sarvstivdins", Sanskrit-

Textual Criticism in the Transmission of Buddhist Scripture",

Texte aus dem buddhistischen Kanon: Neuentdeckungen u. Neu-

Contemporary Buddhism 1.1, 9-34.

editionen, 1, Gttingen 1989 (Sanskrit-Wrterbuch der buddhisti-83-

Tripathi, Chandrabhal
-82-

150

New Manuscript of the Dlrghagama (Hartmann)

1995 Ekottariigama-Fragmente der Gilgit-Handschrift, Reinbek (Studien zur Indologie und Iranistik, Monographie 2).
Weller, Friedrich

1934 Brahmajiilasutra, Tibetischer und Mongolischer Text, Leipzig.

1935-36

"Das tibetische Brahmajlastra", Zeitschrift fr Indologie

und Iranistik 10, 1-61

Kleine Schriften, 2 Bde., ed. Wilhelm Rau,

Wiesbaden 1987 (Glasenapp-Stiftung, 26), 642-702.

Professor,
Institut fr Indologie und Iranistik
der Ludwig-Maximilians-Uniuersitt Mnchen
Guest Professor,
International College
for Aduanced Buddhist Studies

-81-