Jataka Gathas and Jataka Commentary

·期刊原文

Jataka Gathas and Jataka Commentary

By M. Winternitz

The Indian Historical Quarterly

Vol.IV MARCH, 1928 No.1

p. 1

It is of the utmost importance to know how far
the Jatakas can be used for historical purposes, more
especially for the history of Indian literary types,
and for the history of social life and institutions
in ancient India. H.Oldenberg(l) has used the Jatakas
in support of his famous, though now no longer
accepted, "Akhyana-theory", claiming them as proving
the existence, from the Vedic period onwards, of a
type of narrative poetry, composed in a mixture of
prose and verse, of which the verses only were
committed to memory and handed down, while the prose
story was left to be narrated by every reciter in his
own words.(2) G. Buhler(3), R.Fick(4), T.W. Rhys
Davids(5) and Mrs. Rhys
___________________

1. The Prose and Verse Type of Narrative and the
Jatakas (translated from the German).--Journal of the
Pali Text Society, 1910-1912, pp. 19ff.

2. Cf. my History of Indian Literature, English
translation, vol.I, pp.101ff.

3. On the Origin of the Indian Brahma Alphabet, 2nd
ed., Strassburg, 1898, p.16ff.

4. Die soziale Gliederung im nordostlichen Indien zu
Buddhas Zeit, Kiel 1897, translated by Shishir Kumar
Maitra(The Social Organisation in North-East India in
Buddha's Time, Calcutta 1920).

5. Buddhist India, London 1903, PP.201ff.

p. 2

Davids(1) were of opinion that the Jatakas such as we
have them give a picture of Indian life in the days
of Buddha, that is, in the sixth and fifth century
B.C., or at least at the time of the redaction of the
canon in the third century B.C. Since then, however,
it has become the almost general opinion of scholars
that only the Jataka-Gathas can claim canonical
authority, and be regarded as documents of the third,
or even the fifth century B.C., while the Jataka
Commentary, as we have it, can claim no higher
antiquity than the fifth or sixth century A.D.,
though in its prose parts also it contains old
traditions which in many cases may go back to the
same early period as the Gathas. More over, it was
generally believed that the original canonical
Jataka, consisting of Gathas only, was preserved to
us in the Phayre Ms. of Jataka verses.(2)

Professor J. Hertel,(3) it is true, has suspected
long ago that this MS. may be only an extract from
the commentary, such as there are certain Pancatantra
Mss. which contain only the verses without the prose
tales, but have been merely copied from complete Mss.
of the Pancatantra. Lately Friedrich Weller(4) has
examined not only the Phayre Ms. but also two
Mandalay Mss. of Jataka Gathas, and has come to the
conclusion that all these MSS. bave been
___________________

1. Notes on Early Economic Conditions in Northern
india, JRAS., 1901, 859ff.

2. Part of a MS. of the whole Tipitaka presented by
the king of Burma, dated Sakkaraj 1202 and 1203 (A.D.
1841-42), belonging to the Phayre Collection of the
India Office Library (see Oldenberg, JPTS., 1882,
P.60).

3. Wiener Zeitschrift fur die Kunde des Morgenlandes,
23, 1909, 279f.;24, 123; Zeitschrift der Deutschen
Morgenland, Gesellschaft 64, 1910, p.58,

4. Zeitschrift fur Indologie und Iranistik, 4, 1926,
PP.46ff. There is also a MS. of the Jataka Gathas in
the Academy of Leningrad, and Professor Serge
d'Oldenburg told Dr. Weller that in his opinion this
MS., too, was copied from a Ms., of the Jataka
Commentary.

p. 3

extracted and copied from the Jataka Commentary, and
by no means represent the ancient Verse-Jataka of the
canon.

Dr. Weller's chief argument is based on the
curious (redactor's or copyist's?) note at the end
of the Maha-Supina- Jataka(No.77):

Parinibbute pana Bhagavati Samgitikaraka
usabharukkhadini tini padani Atthakatham aropetva
labuniti adini Padani ekam gatham katva
Ekanipatapalim aropesun ti (Fausboll's ed., vol.I,
p.345).

This note refers to the traditional belief of the
Sinhalese Buddhists that at the council held after
the Buddha's departure not only the redaction of the
canonical texts (Pali), but also that of the
commentaries (Atthakathas) took place. It is
well-known that in the Jataka Commentary each Jataka
begins by quoting the first words of the first Gatha.
Now our Jataka No.77 begins with the words: labuni
sidantiti which is the beginning of the fourth line
in our Jataka Commentary, while the first line begins
with: usabha rukkha, What the note wishes to explain
is, why the Jataka begins with labuni sidanti, and
why it was placed in the Ekanipata. It says: "Now
when the Lord had passed away, the arrangers of the
Council put the three lines beginning with usabha
rukkha into the Atthakatha, made of the five lines
beginning with Iabuni one stanza, and received it
into the canonical text of Ekanipata". Whatever the
exact meaning of this note may be,(1) it is clear
that the writer
_____________________

1. Different translations have been given. T. W. Rhys
Davids (Buddhist Birth Stories, London 1880,
p.LXXVII, note 3) translates: "Those who held the
Council after the death of the Blessed One placed the
lines beginning usabha rukkka in the Commentary, and
then, making the other lines beginning Iabuni into
one verse, they put (the Jataka) into the Eka-nipata
(the chapter including all those Jatakas which have
only one verse)". Fausbol1 (Jataka edition, vol. VII,
p.iii) translates: "When Bhagavat was dead the
Council-holders put the three padas usabha rukkha
etc. into the Atthakatha, and made lapuni and the
other padas into one gatha and put it into the verses

p. 4

of the note found the three lines beginning with
usabha rukkha only in the Atthakatha. Now all the
three MSS. of Jataka verses contain these lines.
Hence Dr. Weller concludes that these Mss. must be
copied from an Atthakatha Ms., and not frorn a MS. of
the original Verse-Jataka. The conclusion would be
quite justified, if we only knew who was the writer
of the note: was it the redactor of the Atthakatha,
or some later copyist? Besides, the eight lines:

Usabha rukkha gaviyo gava ca
asso kamso sigali ca kumbho
pokkharani ca apakacandanam
labuni sidanti sila plavanti
mandukiyo kanhasappe gilanti
kakam suvanna parivarayanti
tasa vaka elakanam bhaya hi
vipariyaso vattati nayidhamatthi(1)

give, by means of catch-words, the contents of the
Maha Supina-Jataka. This is the story of a king who
was terrified by sixteen dreams, which were
interpreted by the (Pali) of the Ekanipata."
R.Chalmers (Jataka transl. ed. by E.B. Cowell, vol.I,
p.194) translates: "But after the passing of the
Blessed One, the Editors of the Great Redaction put
the three first lines into the Commentary, and making
the lines from 'And gourds that sank' into one stanza
(therewith), put the whole story into the First
Book", but adds: "I am not at all sure that this is
the correct translation of this difficult and
corrupt passage." Weller (l.c. p.51) translates:
''Nach Buddhas Tode nahmen die Konzilteilnchmer die
drei Verszeilen, die mit usabharukkha beginnen, in
die Atthakatha auf, machten aus den funf Verszeilen,
die mit labuni beginnen, einen Vers und nahmen ihn?)
in den Text des Ekanipata auf".
_______________________

1.This last line was missing in Fausboll's Mss., but
is warranted by the Veyyakarana, and by the Mss.
examined by Dr. Weller. Lapuni in Fausboll's edition
is a bad reading for labuni (sanskrit alibuni,
'pumpltins'). It is strange that the new edition of
the Jataka in Siamese characters, issued by their
Majesties Queen Aunt and Queen Suddhasinninath of
Slam in 1925, also has only seven niseda of eight
lines. This edition reads lavuni.

p. 5

Brahmins as foreboding great calamities, for the
prevention of which animal sacrifices with the
slaughter of numerous beasts and birds were
necessary, while the wise Bodhisatta interpreted them
as having no reference at all to the king himself,
but to some distant future when weak and unrighteous
kings would rule: Now the first three lines refer to
eleven of the sixteen dreams, while the five last
lines only indicate the last five dreams. If really
the canonical Jataka only contained the lines
beginning with labuni, the original Jataka would only
have related a story of five dreams, and the
Jataka-Atthakatha would have given an enlarged
version of an older story. This is, of course,
possible.(1) But we cannot be quite certain, as the
words labuni sidantiti at the beginning of the Jataka
may also be a mistake of the MSS. Of the
Jatakatthavannana.

The other facts which Dr. Weller mentions as
proving the MSS. in question to be copied from Mss.
of the Atthakatha, are:(1) that in some places the
word ti (iti) after the verse proves that some prose
text preceded it; (2) that one of the three MSS.
contains some prose passages(2) ; and (3) that the
MSS. also contain Samodhanagathas. The''connexion"
(samodhana) between the persons of the "story of the
past'" with those of the "story of the present" is
generally given in prose, that is, as part of the
commentary, only exceptionally also in verse. Here
it would be necessary to know whether in all or only
in some of these cases the verses are
__________________

1. In the Tibetan and Chinese versions of this
Jatakaa story (see S. d'Oldenburg, JRAS., 1893,
pp.5o9ff.) there are not sixteen, but only ten
dreams.

2. But all these prose passages are such as have a
word-for- word commentary; they are found in the
Kunala-Jataka(No.536) which is so different in style
from all the other Jatakas that Oldenberg ascribes to
it quite an exceptional position (JPTS.,1910-1912,
p.26 n.3. and in Jataka No.202, where the words
appamano buddho appamano dhammo appamano samgho etc.
(Fausboll ed., vol. II, p.147) are a kind of spell.

p. 6

found also in the three MSS. It is this Samodhana by
which a secular story is turned into a Jataka, and I
can see no reason why Samodhanagathas should not
occur also in a Verse-Jataka.

However, it must be admitted that our hope and
belief that the original Verse-Jataka is still extant
in MSS., has been shaken by Dr. Weller's arguments,
though a critical edition of the Jataka Gathas from
the four MSS., which give the Gathas only, or at
least a complete collation of these MSS. with
Fausboll's edition would be necessary in order to
establish all the facts of the case.

But Dr. Weller ought not to have doubted that a
Verse-Jataka ever existed at all. For it seems to me
that the very note at the end of No. 77, to which he
himself attaches so great an importance, proves that
there was a Verse-Jataka in which No. 77 began with
labuni, and a Jataka-Atthakatha containing all the
verses beginning with usbha rukkha. Apart from this,
however, there are good reasons for assuming that the
canonical Jataka was a Verse-Jataka.

In the Commentary(1) me often find references to
the "Pali" as distinguished from the Atthakatha,
where Pali cannot mean anything else but the
Verse-Jataka. Thus, in Jataka No. 142 (Fausboll, vol.
I, p. 488) we read Kalim papeti in the Gatha. The
commentator says that in the Pali they write phalam
papeti, but that this is not found in the Attha-
katha, and does not make good sense. In No. 255
(Fausboll, vol. II, pp. 293f.) the word agiddhita
occurs in the Gatha. The Commentary says: Paliyam
pana agiddhima ti likhitam

1 Mostly in the Veyyakara, the word-for-word
commentary to the gathas, but sometimes also in other
parts of the Commentary. As a matter of fact, we have
no means of distinguishing between the different
parts of the Jatakatthavannana. The distinction made
by Fausboll in his edition by printing the
Paccuppannavatthu in smaller type, is quite
arbitrary, and has no chronological meaning. It is
possible that the Veyyakarana may be later, but it
has never been proved.

p. 7

tato ayam atthakathapatho va sundarataro, that is to
say, he prefers the reading of the Atthakatha to that
of the Pali. In No. 479 (Fausboll, vol. IV, p. 236)
our text has two Gathas, of which the second only is
a good sloka, while the first cannot be called a
verse at all. The Commentator says that in the Pali
only the second verse is to be found. In No. 539
(Fausboll, vol.VI, p.36, G. 126) the commentator says
that the last; pada janna so yadi hapaye is only
found in the Pali, and not in the Atthakathas
(atthakathasu n'atthi). In No. 547 (Fausboll, vol.
VI, p. 547) the commentator says that in the Pali the
Gathas end with medini samakampatha, while in the
Atthakatha one more gatha follows.

In No. 505 ( Fausboll, vol. IV, p. 447, 1.3) we
read: Itopara uttanasambandhagatha Palinayen'eva
veditabba. "The following Gathas, as their
connection is clear, must be understood according to
the Pali only(1)." Quite similarly in No. 537
(Fausboll, vol. V, p. 460) the gathas 2-4 are
introduced by the words:Itoparam uttanasambandhani
vacanapativacanani Palivasen'eva veditabbani. "In the
following the speeches and counter-speeches, whose
connexion is clear, are to be understood according to
the Pali."(2) In both these passages only the
cononical verse-text can be meant. The same applies
to No. 533 (Fausboll, vol. V, p. 341), where 23
Gathas are given without any prose, and the
commentator says: Imsam gathanam sambandho Palivasen'
eva veditabbo ("the connexion of these Gathas must be
understood by the Pali itself").

In quite a number of other passages, in the
Veyyakarana, various readings are quoted as occurring
"in the Pali"
__________________________

I W.H.D.Rouse(Jataka Transl. ed. Cowell, vol. IV,
p277) translates wrongly: "The connexion of the
following verses is clear; they are arranged in due
succession."

2 Or, as H.T. Francis (Jaataka Transl. ed., Cowell,
vol. V, p. 249) translates more freely: "The verses
that follow are of obvious connexion and are to be
understood as uttered by alternate speakers in
accordance with the scripture context."

p. 8

(paliyam), or sometimes (vol. IV,p. 134; vol. VI, p.
274), "in the Pall manuscripts" (palipotthakesu).
Occasionally the commentator uses "Pali" also in the
sense of "language of the canonical text". Thus, in
the Maha-Ummagga-Jataka (No. 556, Fausboll, vol. VI,
p. 353), the Bodhisatta is said to have taught the
four counsellors the Gathas "in the Pali (language)."
And in No. 522 (Fausboll, vol. V, p. 147) the
commentator explains gambhirapanham by saying:
atthato ca palito ca gambhiram, "deep both in meaning
and in Pali (words)." In the commentaries of
Buddhaghosa and others, and in the Visuddhimagga the
Pali often occurs to introduce canonical quotations,
where it simply means "canon", "canonical text",
"scripture," much like sruti in Brahmanical texts. Of
course, in the Jataktthavannana also Pali means
"canonical text," but wherever the word occurs, it
refers to Gathas only. This shows that the canonical
Jataka was a Verse.Jataka, and handed down in
different MSS. from those of the Jataka-Atthakatha,
which consists of Gathas and prose.

E. Senart(l) has already shown that it would have
no meaning to call some verses osanagatha, '"final
stanzas," or to refer to them as being placed at the
end, especially when in our Jataka long prose
passages follow after this "end," if the author of
the commentary had not referred to the last stanza of
a Jataka consisting of stanzas only. The same scholar
has also pointed out(2) already, that Jatakas with
more than one Gatha occur in the Eka-Nipata. And this
seems to me the strongest proof of an original
Verse-Jataka, that the number of verses in the
different sections of our Jataka Commentary does not
tally with the titles of these sections.

It is well-known that the Book of Jatakas, like
the Thera- and Theri-Gathas and many other works of
Indian literature,
________________________

1 Journal Asiatique, 1901, ser. 9, t. XVII, pp. 397ff.

2 L. c., pp. 402f.

p. 9

is divided into sections according to the number of
Gathas belonging to one Jataka, the Eka-Nipata
containing one, the Duka-Nipata two, the Tika-Nipata
three gathas each, and so on, up to the
terasa-Nipata, "The Section of Thirteen (Gathas)".
The XIVth section is called Pakinnaka-Nipata or
"Section of (Jatakas with a ) Mixed (number of
gathas)" Sections XV to XXI, Visati-Nipata,
Timsa-Nipata, etc., contain Jatakas with a number of
Gathas in the twenties thirties, forties, fifties,
sixties, seventies, and eighties, the last section
(XXII) being the Maha-Nipata or "Large Section," i.e,
the Section with a large number of Gathas. Now what
do we actually find in our Jataka Commentary?

In the Eka-Nipata 14 of the 150 Jatakas have more
than one Gatha(1): Nos. 1, 25, 31, 35 with 4, Nos. 4,
40, 62 with 5, Nos. 67, 70, 78, 132, 150 with 2
Gathas each, No. 12 with 10, and No. 96 with 11
Gathss. In each of these Jatakas only one Gatha is
given with a v. and the Gatha MSS. contain only this
one Gatha.

In the Duka-Nipata 6 of the 100 Jatakas have more
than two Gathas: No. 159 with 4 Gs. (but the third
and fourth are only repetitions), 203 with 5, 211
with 3, 220 and 240 with 7 each, and 243 with 10 Gs.
(7 of which are quoted fromn the Vimanavatthu, and 1
is an Udana). In Nos. 203 and and 220 all the Gathas
have a v., and are in the A.

In the Tika-Nipata 6 Of the 50 Jatakas have more
than three Gathas: Nos. 257, 276, 285, 296 with 4
each, No. 269 with 9 long stanzas in the P., besides
3 Gs. in A., No. 284 with 10.Gs. in P. (between 2 and
3 there are 7 Gs. introduced by the words: imam
dhammam desesi).

In the Panca-Nipata 7 of the 25 Jatakas have more
than
_______________________

1 The Gathas (G.) in the whole of our Jataka
Commentary occur sometimes in the Paccuppannavatthu
(P.), sometimes(most frequently) in the Atitavatthu
(A.), sometimes partly in the one and partly in the
other. In the Veyyakarana (V.) also sometimes verses
are quoted. These have not been counted.

p. 10

five Gs., generally all with v.: No. 352 with 8, No.
354 with 10, Nos. 358, 371, 375 with 6 Gs. each, Nos.
372 and 374 with 7 Gs. each.

In the Cha-Nipata 9 of the 20 Jatakas have more
than six Gs., generally all with. v.: Nos. 376, 383,
389, 390, 391, 392 with 7 Gs. each, Nos. 380 and 385
with 8 Gs. each, and No. 382 with 17 Gs. in A. (all
with v.).

In the Satta-Nipata there are 21 Jatakas, 6 of
which have more than seven Gs.: No. 400 with 11 Gs.,
Nos. 402, 408 and 410 with 9 Gs. each, No. 405 with 8
Gs., and No. 415 with 12 Gs. (all in A. and with v.).

The Attha-Nipata contains 10 Jatakas, 7 of which
have more than eight Gs.: Nos. 417, 419, 420, 421,
and 423 have 9 Gs. (with v.), No. 422 has 15, and No.
425 has 11 Gs., all in A. and with v.

The Nava-Nipata has 12 Jatakas, 4 of which (Nos.
428, 429, 430, and 432) have more than nine Gs. The
Dasa-Nipata has 16 Jatakas, 5 of which (Nos. 440,
443, 447, 448, and 454) have more than ten Gs. The
Ekadasa-Nipata has 9 Jatakas, 5 of which (Nos. 455,
456, 458, 461, and 463) have more than eleven Gs. The
Dvadasa-Nipata has 9 Jatakas, two of which (No. 466
with 13 Gs., and No. 472 with 14Gs. in A., and one
Samodhana G. at the end) have more than twelve Gs.
The Terasa-Nipata has 10 Jatakas, 5 of which (Nos.
477, 479, 480, 482, and 483) have more than thirteen
Gs. (generally with V.)

Section XIV, the Pakinnaka-Nipata, consists of 13
Jatakas, in which the number of Gs. varies from 15 to
47. This Section with a mixed number of Gs. would
have no meaning at all, if the other Sections had
not, been originally intended to contain exactly as
many Gs. as are indicated by the title of the
Section. For why should not all the Jatakas with 14
to 19 Gs. be in the Pakinnaka-Nipata? Why is No. 382
with 17 Gs. in the Cha-Nipata, or No. 422 with 15,
Gs. in the Attha-Nipata? And why are Nos. 483, 489,
492, 493, 494, and 496 (with 20 to 26 Gs.) in the
Pakinnaka, and not in the

p. 11

Visati-Nipata? Why is No, 495 with 47 Gs. not in the
Cattalisa-Nipata?

The Visati-Nipata consists of 14 Jatakas which
ought to have from 20 to 29 Gs., but No, 499 has 31,
506 has 44, and 507 has 30Gs. The Timsa-Nipata has 10
Jatakas, 3 of which (Nos.514 to 516) have more than
39 Gs. The Cattalisa-Nipata has 5 Jatakas, 2 of which
(Nos. 524, 525) have 51 Gs. The Pannasa-Nipata has 3
Jatakas, one of which (No. 527). has 67Gs. The
Sattati-Nipata has 2 Jatakas which ought to have 70
to 79 Gs., but actually No. 531 has 92, and 532 has
93 Gs. In the Asiti-Nipata with 5 Jatakas, we find 3
(Nos. 534, 635, 637) which have more than 90 Gs.
There is no reason why No. 534 with 103 Gs. and No.
537 with 123 Gs. should not be included in the
Maha-Nipata, where we find No. 538 with 120 Gs.

It is clear that this arrangement of the Jatakas
according to the number of Gathas cannot be based on
our Jataka Commentary. The probability is that there
was an ancient and canonical Verse-Jataka, which was
thus divided in 22 Sections, and that this division
was kept- up in the Commentary, though in so very
many cases the number of Gathas no longer tallies
with the titles of the Sections.

It is much to be regretted that circumstances
prevented Dr. Weller from collating the Gathas Of all
the three (resp. four) MSS. with Fausboll's edition
of the Commentary. Only when this work` will have
been accomplished, we shall be able to see whether in
the cases, where the Commentary notes a difference of
reading between the "Pali" and the Affhakatha, the
Gatha MSS. agree with the one or with the other. And
then only will it be possible to see whether in these
Gatha MSS. the number of Gathas agrees with the
titles of the Nipatas or with the number of Gathas
found in the Commentary.

At the present state of our knowledge we are
bound to say that not only the prose but, also the
Gathas of the Jatakas contain much that did not
belong to the original

p. 12

canonical Jataka collection. Fausboll(Jataka edition,
vol. VI, Preliminary Remarks) tells us that for the
Maha-Nipata the Burmese Ms. offers a much enlarged
text, that in fact it differs so much from the text
offered by the other two MSS. that he would "advise
some scholar to give a separate edition of the
Maha-Nipata according to the Burmese redaction."
This, too, shows how uncertain our text of the Jataka
Book is, and how careful we have to be when using the
Jatakas for the purpose of historical research.

Nevertheless, it remains true that, on the whole,
the Gathas have a much stronger claim to be regarded
as canonical than the prose of the Jatakas. According
to the Buddhist tradition of Ceylon the original Pali
Atthakathas were translated into Sinhalese, and
afterwards re-translated into Pali by Buddhaghosa,
and others. The Jatakaa-Atthavannana also is
according to this tradition only a translation into
Pali, and a recast of a Sinhalese version of the
original Pali Jataka. Atthakatha. In the course of
this work of translating and re-translating, however,
the Gathas remained in their original Pali. If we
accept this tradition,(1) the Gathas are of course
more original than the prose. In any case, the prose
was
______________________

1 E. W. Burlingame (Journal of the American Oriental
Society, 38, 1918, pp. 267f.) has declared this
tradition to be "unreliable and misleading." For(I)
Buddhaghosa and the compiler of the Dhammapada
Commentary drew, independently of each other, from
common Pali originals; (2) the reader or compiler of
the Jataka Commentary copied both Stories of the
Present and Stories of the Past from canonical books
and from Buddhaghosa's commentaries, and (3)
Dhammapala drew in similar manner from Buddhaghosa's
commentaries and from the Dhammapada Commentary (of.
Burlingame, Buddhist Legends from Dhammapada
Commentary, Harvard Oriental Series, vol. 28, pp.
48ff., 56f.) But it is difficult to understand how
and why such a tradition about the Sinhalese
translations and retranslations should have arisen
without any historical background. On the other hand,
when we belive the tradition, it is only natural

p. 13

always more exposed to changes and enlargements.
Frequently it is nothing but the poor performance of
some inferior writer, especially when, as is often
the case, no prose is required at all. It is in the
prose only that allusions occur to Ceylon, and not
infrequently it is at variance with the Gathas. The
language of the Gathas, too, is more archaic than
that of the prose.

It is true that in some Jatakas, Gathas and prose
form a homogeneous whole. In many others, however,
the proses are nothing but useless commentatorial
accessories. Therefore the Jatakas cannot be taken as
examples of the ancient Indian Akhyana in the sense
of the prose-and-verse type of narrative, as
Oldenberg understood it. Not one, but several
literary types are represented in the Jataka
Collection. There are some Jatakas which were prose
stories with only one or two or a few verses
containing either the moral or the gist of the tale.
In these cases it is likely enough that the
commentary has preserved more or less of the old
prose stories. Another type of Jatakas is that of the
Campu in which the story itself is related
alternately in prose and verse, in which case the
commentary is often an expansion of the original
prose text. But there are other Jatakas which
originally consisted of Gathas only: some of them
ballads in dialogue form, others ballads in a mixture
of dialogue verses and narrative stanzas, others
again epics or fragments, and some even mere strings
of moral maxims to think that the monks who
translated from Sinhalese into Pali would take
canonical and even uncanonical Pali texts, wherever
they were available, from the original Pali works,
and not take the trouble of translating them from
Sinhalese. When we meet with the same stories in
different commentaries, it is not necessary to assume
that the one has copied from the other. It is more
probable that they were copied from the same
pre-existing sources. Many Jatakas, especially the
longer ones, probably existed as independent texts,
before they were included in a Commentary or
Collection.

p. 14

on some topic. In all these cases the entire prose
belongs to the commentary(1).

From all this it follows that when using any part
of the Jataka Book for historical purposes, we shall
always have to ask ourselves first, to which stratum
of the text that part belongs.
_____________________

1 Cf. my article `Jataka' in Hastings'Encyclopedia of
Religion and Ethics, vol. vii, p.492.