Dīpavaṁsa
[The Chronicle of the Island]

Introduction

[1] The Dīpavaṁsa, an edition of which I here lay before the public, is a historical work composed in Ceylon by an unknown author. George Turnour, who first drew the attention of European scholars to the Dīpavaṁsa, See the Journal As. Soc. Bengal, vol. VI, p. 790, 1054; VII, p. 919 et seq. declared it to be identical with a version of the Mahāvaṁsa to which the Mahāvaṁsa Ṭīkā occasionally alludes, the version preserved in the Uttaravihāra monastery. This is certainly wrong. The arguments by which Turnour supports his theory are entirely founded on errors and inaccuracies. Turnour says, for instance, that the opening lines of the Dīpavaṃsa are quoted in that Ṭīkā as contained in the Uttaravihāra-Mahāvaṃsa. The Ṭīkā quotes those lines indeed (India Office MS. of the Mahāv. Ṭīkā fol. ko'), but without the slightest reference to the Uttaravihāra-Mahāvaṃsa. The only passage referring to the Mahāvaṃsa of the Uttaravihāra which I know, is quite sufficient to show that the work was different from the Dīpavaṃsa. In the Mahāvaṃsa Ṭīkā (India Off. MS. fol. gha') we are told, that the Uttaravihāra-Mahāvaṃsa contained a statement somewhat different from the usual one, with regard to the descendents of king Sīhassara: “Uttaravihāravāsīnaṃ pana Mahāvaṃsa Sīhassarassa rañño puttapaputtakā caturāsītisahassāni rājāno ahesuṃ, tesaṃ kaniṭṭhiko Bhaggusakko (corrected to Bhagusakko) nāma rājā. tassa puttapaputtakā dvāvīsatisahassāni rājāno ahesuṃ, tesaṃ kaniṭṭhiko Jayaseno ’ti vuttaṃ.” The Dīpavaṃsa (3, 43. 44) in the contrary agrees with the statements of our Mahāvaṃsa (p. 9) which are in contradiction to the Mahāvaṃsa of the Uttaravihāra. We must undertake, therefore, a research of our own as to the origin of the Dipavaṁsa and its position in the ancient literature of the Ceylonese.

The Mahāvaṁsa Ṭīkā, when commenting on the title [2] of the Mahāvaṁsa, contains the following important remarks: I give this passage according to the India. Office MS. (fol. ko'), together with the various readings of the Paris MS. (fol. ko') which M. Feer has kindly communicated to me. – Comp. Turnour’s Introd. to the Mahāvaṃsa, p. XXXI.

“Mahāvaṁsaṁ pavakkhāmīti mahantānaṁ vaṁso tantipaveniparamparā ’ti pi sayam eva vā mahantattā ubhayattha paridīpitattā vā Mahāvaṁso, tam Mahāvaṁsaṁ. tesam pi buddhabuddhasāvakādīnam pi guṇamahantānaṁ Mahāsammatādinam pi vā rājamahāmattānaṁ (rājamahantānaṁ?) pavenidīpitattā ca buddhāgamanadīpakārehi mahādhikārattā sayam eva mahantattaṁ veditabbaṁ.

anupamavaṁsaanuggahādinaṁ (°dīnaṁ Paris MS.)
sabbaṁ aññataṁ (aññātaṁ Par. MS.) kataṁ suppakāsitaṁ
apariyāgataṁ (ariyābhataṁ Par. MS.) uttamasabbhi vaṇṇitaṁ
suṇantu dīpatthutiyā sādhusakkatan ti (comp. Dīpav. 1, 5)

iminā Aṭṭhakathānayen’ ev’ assa mahantattaṁ paridīpitani (paridīpitattaṁ?) ca veditabbaṁ. – tenāhu porāṇā:

dīpāgamanaṁ buddhassa dhātuñ ca bodhiāgamaṁ (bodhiyāgamaṁ Par. MS.)
saṅgahattheravādañ ca dīpamhi sāsanāgamaṁ
narindāgamanaṁ vaṁsaṁ kittayissaṁ suṇātha me ’ti

(comp. Dīpav. 1, 1). imāya pana gāthāya nayena (gāthānayena Par. MS.) pi assa sakhyā (saṅkhyā Par. MS.) mahantattaparidīpitattaṁ ñeyyaṁ. evaṁ Mahāvaṁsan ti laddhanāmaṁ Mahāvihāravāsīnaṁ vācanamaggaṁ porāṇaṭṭhakathaṁ ettha Sīhalabhāsaṁ hitvā Māgadhikabhāsāya pavakkhāmīti adhippāyo.”

As two stanzas are quoted here, the first of which is said to be taken from an Aṭṭhakathā, the second from a work of the “Porāṇā” and as immediately afterwards mention is made of the Sinhalese “Porāṇaṭṭhakathā”, the contents of which are expressed in the Mahāvaṁsa in Pāli, there can be very little doubt that this Aṭṭhakathā and [3] that work of the Porāṇā are identical with each other and with the Porāṇaṭṭhakathā mentioned afterwards. If I am right in asserting the identity of the works alluded to under these various expressions, we learn from this passage the following important fact, viz.: there was an Aṭṭhakathā handed down and probably also composed in the Mahāvihāra, written in Sinhalese, the introductory stanzas of which allude to a historical account of Buddha’s visits to Ceylon, of the Councils, of Mahinda etc., intended to be given in that very Aṭṭhakathā, and these stanzas are almost identical with the opening lines of the Dīpavaṁsa. There must, therefore, most probably be some relation between this Aṭṭhakathā and the Dīpavaṁsa which will be of a sufficient interest to justify a further inquiry.

Let us see what results we may obtain as to the contents of this Aṭṭhakathā and as to its literary form? Let us ask particularly in what way the Pāli stanzas quoted from this Aṭṭhakathā were connected with its main substance which was composed in Sinhalese?

The Mahāvaṁsa Ṭīkā, after the passage quoted above, goes on to give some details about the way in which the author of the Mahāvaṁsa made use of this Sinhalese Aṭṭhakathā on which his own work is based. It is said in the Ṭīkā “ayaṁ hi ācariyo poraṇamhi Sīhala[ṭṭha]kathā-Mahāvaṁse ativitthārapunarutti-dosabhāvaṁ pahāya taṁ sukhaggahaṇādippayojanasahitaṁ katvā ’va kathesi.” The work in question is called here, as repeatedly afterwards, Sīhalaṭṭhakathā-Mahāvaṁsa. It is not difficult to account for this expression where the two at first sight contradictory elements of a (theological) commentary (Aṭṭhakathā) and of an extensive historical narrative (Mahāvaṁsa) are combined together. If we look at Buddhaghosa’s Aṭṭhakathā on the Vinaya, we find that the author has there prefixed to his explanation of the sacred texts a detailed historical account of the origin of the Tipiṭaka, its redaction in the three Councils, and its propagation to Ceylon by Mahinda and his companions. Buddhaghosa’s commentary is based, as is well [4] known, on that very Sinhalese Aṭṭhakathā of the Mahavihāra, which we are now discussing, and we may assume almost with certainty, that to this Aṭṭhakathā a similar historical introduction was prefixed which may have been even of a greater extent than that of Buddhaghosa. If our inference is correct, we have found in what way the title “Sīhalaṭṭhakathā-Mahāvaṁsa” could apply most properly to this work: it is a historical account (Mahāvaṁsa) which forms a constituent part of a theological commentary (Aṭṭhakathā), and which is composed in order to give to the dogmatical contents of the letter the indispensable historical foundation. In order to show how an Aṭṭhakathā could contain extensive and detailed historical statements, we may refer also to the quotations which the Mahāvaṃsa Ṭīkā gives from the Aṭṭhakathā of the Uttaravihāra, concerning the history of Susunāga (Turnour’s Introduction, XXXVII), of the nine Nanda kings (1. I. p. XXXVIII), of Candagutta (1. 1. p. XLII), of the descendants of Mahāsammata (1. I. p. XXXV). It we compare the last passage with the statements about the same subject given above (p. 1, n. 2) from the Mahāvaṃsa of the Uttaravihāra, we are lead to believe, that the Mahāvaṃsa also of the Uttaravihāra formed a part of the Aṭṭhakathā of that monastery in the same way as was the case in the Mahāvihāra.

We are not very likely to go too far astray, if we try to form an idea of the style of composition of this Sinhalese Aṭṭhakathā-Mahāvaṁsa according to the analogy of what we have before us in Buddhaghosa’s comment. According to this, the Aṭṭhakathā-Mahāvaṁsa appears to have been written, in the main, in prose; it was intermixed, however, with a considerable number of stanzas in order to emphasize the more important points of the narration and to raise them above the level of the rest. The Pāli verses quoted above, which are taken apparently from the introduction to the whole work, render it highly probable that all these stanzas were composed in Pāli. It was necessary, of course, to convey to the Sinhalese clergy of that time the understanding of the sacred Pāli texts by a commentary written in Sinhalese; if, however, in the course of such a commentary the most important and remarkable points were to be expressed in a metrical form, we may easily understand, that for such a purpose [5] the language of the sacred texts which contained so abundant a stock of models for metrical composition, appeared much more appropriate than the popular dialect of the Sinhalese.

A considerable number of verses ascribed to the “Porāṇā”, i. e. taken from the ancient Sinhalese Aṭṭhakathā, and quoted by Buddhaghosa or in the Mahāvaṁsa Ṭīkā, present the same close resemblance and almost identity with passages of the Dīpsvaṁsa, which we have observed in the two verses given above. In proof of this I give the following examples:

Samanta-Pāsādikā (MS. Orient. 1027 of the Brit. Museum) fol. ga’: tenāhu porāṇā:

Vedisagirimhi Rājagahe vasitvā tiṁsa rattiyo
kālo ’va gamanassā ’ti gacchāma dīpam uttamaṁ. |
paḷinā Jambudīpāto haṁsarājā va ambare
evam uppaṭitā therā nipatiṁsu naguttame. |
purato puraseṭṭhassa pabbate meghasannibhe
patiṭṭahahiṁsu kūṭamhi haṁsā va nagamuddhanīti.
    Comp. Dīpav. 12, 35-37.

Samanta-Pās. fol. kāh’ – kha: sā panāyaṁ (i. e. the second Council)

yehi therehi saṅgītā saṅgītesu ca vissutā
Sabbakāmi ca Sāḷho ca Revato Khuddhasobhito |
Yaso ca Sāṇasambhūto, ete saddhivihārikā
therā Ānandatherassa diṭṭhapubbā tathāgataṁ, |
Sumano Vāsabhagāmi ca ñeyyā saddhivihārikā.
dve ime Anuruddhassa diṭṭhapubbā tathāgataṁ. |
dutiyo pana saṅgīte yehi therehi saṅgaho
sabbe pi pannabhārā te katakiccā anāsavā ’ti.
    Comp. Dīpav. 4, 50-54.

Mahāv. Ṭīkā fol. khau: tenāhu porāṇā:

yakkhānaṁ buddho bhayajananaṁ akāsi,
te tajjitā taṁ saraṇaṁ akaṁsu buddhaṁ,
lokānukampo lokahite sadā rato
so cintayi attasukhaṁ acintamassa. |
imañ ca Laṅkāthala mānusānaṁ [6]
porāṇakappaṭṭhitam vutt[h]avāsaṁ
vasanti Laṅkāthala mānusā bahu
pubbe va Ojamaṇḍavaradīpe ’ti.
    Comp. Dipav. 1, 66. 73.

Mahāv. Ṭīkā fol. ṭāṁ: tenāhu porāṇā:

Suppatiṭṭhitabrahmā ca Nandiseno Sumaṇadeviyā
putto mātā pitā c’ eva gihibhūtā tayo janā ’ti.
    Comp. Dīpav. 19, 9.

Mahāv. Ṭīkā fol. ḍh’: tenāhu porāṇā:

Anulā nāma yā itthi sā hantvāna naruttame
Catumāsaṁ Tambapaṇṇimhi issariyaṁ anusāsitā ’ti.
    Comp. Dīpav. 20, 30.

These passages which we may almost with certainty pronounce to belong to the ancient Aṭṭhakathā-Mahāvaṁsa of the Mahāvihāra, will suffice to show, to what extent the author of the Dīpavaṁsa borrowed not only the materials of his own work, but also the mode of expression and even whole lines, word for word, from that Aṭṭhakathā. In fact, a great part of the Dīpavaṁsa has the appearance not of an independent, continual work, but of a composition of such single stanzas extracted from a work or works like that Aṭṭhakathā; many of the repetitions and omissions The most striking example of such repetitions is the account of the three Couneils, each of which is described twice, the author, no doubt, having had before him two different authorities. The case is similar in the following passages:

(12, 51-53:)

 samaṇo ’ti maṃ manussalokeyaṃ khattiya pucchasi bhūmipāla,
 samaṇā mayaṃ mahārāja dhammarājassa sāvakā
 tam eva anukampāya Jambudīpā idhāgatā. |
 āvudhaṃ nikkhipitvāna ekamantaṃ upāvisi,
 nisajja rājā sammodi bahuṃ atthūpasaṃhitaṃ. |
 sutvā therassa vacanaṃ nikkhipitvāna āvudhaṃ
 tato theram upagantvā sammoditvā c’ upāvisi.

(17, 27:)

 puṇṇakanarako nāma ahū pajjarako tadā
 tasmiṃ samaye manussānaṃ rogo pajjarako ahū.

An instructive example of the abrupt and fragmentary character of some parts of the Dīpavaṃsa is contained in the account of the conversion of young Moggallāna (5, 55-68), which would be almost unintelligible if we did not possess the same narrative in the Mahāvaṃsa. (pp. 28-83).

A careful consideration, however, of this passage shows that the boundary line between what is said and what is omitted does not present those signs of capricious irregularity which inevitably characterise gaps caused by a copyist’s carelessness. The omissions are governed by a certain principle. The important and interesting parts of the story are fully told; the less prominent events which form only the connecting links between the chief points of the narrative, are altogether omitted. I think that both these omissions and those repetitions are to be accounted for, in the main, by the extremely awkward method in which the author compiled his work, though I do not deny, of course, that many of the gaps which are found throughout the whole work, are to be referred to a different origin, viz. to the misfortnnes to which the tradition of the Dīpavaṃsa has, from then till now, been exposed.
which render some chapters of the Dīpavaṁsa almost illegible, we may account for not by the inadvertance of copyists, but by this peculiar method of compilation.

The results we have obtained regarding the connexion between the Dīpavaṁsa and the ancient Aṭṭhakathā of the Mahāvihāra, furnish us with a clue for gaining an insight into the relative position of the Dīpavaṁsa and the second important historical text of the Pāli literature, the Mahāvaṁsa. The two works are, indeed, in the main nothing but two versions of the same substance, both being based on the historical Introduction to the great Commentary of the Mahāvihāra. Each work represented, of course, their common subject in its own way, the Dīpavaṁsa following step by step and almost word for word the traces of the original, the Mahāvaṁsa proceeding with much greater independence and perfect literary mastership. The Dīpavaṁsa, as regards its style and its grammatical peculiarities, betrays the characteristics of an age in which the Sinhalese first tried to write in the dialect of the sacred texts brought over from India; there are passages in the Dīpavaṁsa which remind us of the first clumsy attempts of the ancient German tribes, to write Latin. The Mahāvaṁsa is composed very differently; its author masters the Pāli grammar and style with a perfect ease which cannot have been acquired but after many fruitless attempts, and which may he compared with the elegant mastership of Latin composition by which the Italian posts and scholars of the renaissance excelled. The turning point between the ancient and the modern epoch of Pāli literature in Ceylon, is marked, no doubt, by the great works of Buddhaghosa, which were not less important from a literary than from a theological point of view.

Both Mahāvaṁsa and Dīpavaṁsa finish their records at the same point, viz. with the death of king Mahāsena. This coincidence is, of course, nothing but a consequence of the two works being derived from the same source. We may find in this a new confirmation of our opinion that this source is the Aṭṭhakathā-Mahāvaṁsa of the Mahavihāra Monastery. The reign of that very king Mahāsena was a fatal time to this monastery. A hostile party succeeded in obtaining king Mahāsena’s sanction for destroying the Mahāvihāra; during a period of nine years the monastery remained deserted by its former inhabitants; afterwards, after long and violent ecclesiastical struggles, it was reconstructed. We may easily understand, therefore, why historical writers belonging to the Mahāvihāra fraternity should stop just at the epoch of Mahāsena’s reign, where the past destinies of their spiritual abode were divided from the present.

After these remarks about the relation of the Dīpavaṁsa to the ancient theological commentaries and to the Mahāvaṁsa, we now proceed to collect the data which throw a light on the question, at what time the Dīpavaṁsa was composed. Turnour infers its anterior origin, compared with the Mahāvaṁsa, from the fact of the first lines, as he says, of the Dīpavaṁsa being quoted in the Mahāvaṁsa Ṭīkā, the authorship of which he ascribes to Mahānāma, the author of the Mahāvaṁsa itself. But apart from Turnour’s opinion on the age of the Mahāvaṁsa Ṭīkā being totally wrong, we have seen, that those lines are quoted in the Mahāvaṁsa Ṭīkā not from the Dīpavaṁsa itself, but from the Aṭṭhakathā on which the Dīpavaṁsa is founded. So we lose the date on which Turnour’s opinion is based. What remains, are the following data:

1) The Dīpavaṁsa cannot have been written before A. D. 302, because its narrative extends till that year. [9]

2) Buddhaghosa was acquainted with a version of the Dīpavaṁsa which, however, differed in some details from that which we possess. Some lines from that version of the Dīpav. are quoted in the Samantapāsādikā. They partly agree with our text; partly they diifer in such a way that they cannot be reconciled with it. See my notes on Dīp. 11, 17; 12, 2. – The Dīpav. is also quoted in the Aṭṭhakathā on the Kathāvatthu; see the note on 5, 30.

3) The continuator of the Mahāvaṁsa (p. 257, ed. Turnour) tells us, that king Dhātusena (A. D. 459-477) ordered the Dīpavaṁsa to be recited in public at an annual festival held in honour of an image of Mahinda.

4) These data being given, it is only of a secondary importance, that the Mahāvaṁsa Ṭīkā, which was composed in much later times, mentions an Aṭṭhakathā on the Dīpavaṁsa. fol. ṇe (with reference to the ecclesiastical quarrels in Mahāsena’s reign): ye keci bhikkhū vā ussāpitanikāyantaraladdhikā vā tath’ eva amhākaṃ ācariyā akaṃsū ’ti yadeyyuṃ, tesam pi taṃ parikappitavipphanditam eva, tesaṃ abhilāpamattaṃ va ’ti Dīpavaṃsaṭṭhakathāyaṃ vuttaṃ.

The result is, that the Dīpavaṁsa – be it in that very version which we possess or in a similar one – was written between the beginning of the fourth and the first third of the fifth century. We do not know as yet the exact date of the composition of the Mahāvaṁsa, The arguments of Turnour who brings it under the reign of Dhātusena (Introd. p. LIV), are extremely weak. but if we compare the language and the style in which the two works are written, there will scarcely be any doubt as to the priority of the Dīpavaṁsa. The words, besides, by which Mahānāma characterizes the works of his predecessors:

porāṇehi kato p’ eso ativitthārito kvaci
atīva kvaci saṅkhitto anekapunaruttako,

apply so extremely well to those peculiarities of the Dīpavaṁsa of which we have spoken above, that they appear to have been written most probably with reference to this very work. [10]

I have made use in editing the text of the Dīpavaṁsa, of the following MSS.:

I. MSS. written in Burmese characters.

1) F: MS. belonging to Major Fryer who brought it to England from British Burmah. About the third part of the Dīpavaṁsa (6, sī -15, si) is wanting; instead of this the MS. contains a fragment of the Thūpavaṁsa. The MS. has been written Sakkarāj 1190 = A. D. 1828.

2) N: A Collation of the MS. presented by the late king of Burmah to the Colonial Library in Colombo. This MS. was collated by Gombadde Watte Dewa Aranolis with the MS. M (see below), for the use of Mr. Rhys Davids; its readings are indicated, for the most part in English characters, sometimes in Sinhalese writing, at the margin of M. If the reading of N is not expressly indicated by Dewa Aranolis, but if we are, from the silence of the collation, to draw the conclusion, that N agrees with M a conclusion which is, of course, by no means always a safe one, I designate such readings by n. If only a part of the single words is indicated, I include in brackets those parts which we are to supply from M.

II. MSS. written in Sinhalese characters.

3) G: MS. of the Paris National Library (collection Grimblot; fonds Pāli 365). Although this MS. is written in Sinhalese characters, its readings agree at a good many passages with the Burmese MSS. The text of G has been corrected from a MS. very similar to B.

4) A: MS. of the India Office (Pāli Collection no. 95).

5. 6) B. C: Copies of two MSS. of the Dadalle Wihāre, made for Mr. Rhys David, now in the Cambridge University Library (Add. 945. 946). In the five first Bhāṇavāras there are frequent corrections in C made from another MS. than that from which C has been copied. I designate these corrections by c. [11]

7) M: Copy made by Gombadde Watte Dewa Aranolis for Mr. Rhys Davids from a MS. of the Busse Wihāre. Written on paper. Now in the Cambridge University Library (Add. 944).

8) R: MS. of the Cambridge University Library (Add. 1255). This is a copy of a MS. belonging to Mr. James d’Alwis.

9) D: MS. in the possession of Mrs. Childers, London; it is a copy made for the late R. C. Childers.

10) E: MS. of the Paris National Library (Coll. Grimblot, fonds Pāli 366).

11) S: A copy written on paper, which the priest Subhūti of Vaskaduve was kind enough to send me. There are some good corrections, written with pencil, on the first leaves, which I desiguate by s.

I have used, besides, the following abbreviations:

X = all Burmese MSS.
Y = all Sinhalese MSS.
Z = the class of MSS. represented by CDEMRS.

All our MSS. are derived from the same original source which was very incorrectly written in Burmese characters, as we may infer from some of the blunders common to all of our MSS. See, for instance, 1,6.55; 4,45; 11,3; 22,18. Perhaps this was the MS. brought in 1812 from Siam to Ceylon by the Modliar George Nadoris. See Journ. As. Soc. Bengal VI, p. 790

The way in which the single MSS. are derived from their common source, will be shown by the following table:

  Burmese Archetype
   X   Y
  NFsG1  G1  W  Z
two elements combined AbcG2 CDEMRS

As to their critical value, the Burmese MSS. (X) deserve to be classed first; least is the value of Z, the [12] apparent correctness of which is owing not to real ancient tradition, but to arbitrary attempts of emendators.

It appeared desirable to print not the text corrected as far as possible, but the text of the codex archetypus, and to give in the footnotes my own emendations as well as those tried already in the MSS. In many passages I have refrained from correcting manifest grammatical blunders, errors in numbers of years etc., because 1 was afraid of correcting not the copyist but the author himself. Many passages also appeared to me too hopelessly corrupt for me to try to correct them. Of the various readings I could give, of course, only a selection, or the work would have increased to its threefold extent.

I cannot finish without having expressed my sincere thanks to the librarians and owners of MSS. who very liberally lent them to me or allowed me to collate them. My special thanks are due also to Dr. R. Rost who aided my undertaking from its beginning to its end with the greatest kindness in many various ways, and to my learned friends Dr. G. Bühler and Mr. Rhys Davids, but for whose kind and indefatigable assistance I should not have been able to add to the Pāli text of the Dīpavaṁsa a translation written in the language of a foreign country.

Berlin, September 1879.

The Editor.