Chapter

THE MAHAYAMSA

THE GREAT CHRONICLE OF CEYLONilalt tot jfcoctetg

THE MAHAVAMSA

THE GREAT CHRONICLE OF CEYLON

TRANSLATED INTO ENGLISH

WILHELM GEIGER, Pn.D.

PROFESSOR OF INDOGERMANIC PHILOLOGY AT ERLANGEN UNIVERSITY

ASSISTED BY MABEL HAYNES BODE, PH.J[)^

LECTURER ON PALI AT UNIVERSITY COLLEGE, LONDON

UNDER THE PATRONAGE OF THE GOVERNMENT OF CEYLON

3Lon&on

PUBLISHED FOR THE PALI TEXT SOCIETY

HENRY FROWDE

OXFORD UNIVERSITY PRESS, AMEN CORNER, E.C.

1912OXFORD

FEINTED BY HORACE HART AT THE UNIVERSITY PRESS

2 6 MY '24

OH 30EDITOR'S PREFACE

A FEW words are necessary to explain how the present work came to be written; and one or two points should be mentioned regarding the aims it is hoped to achieve. Early in 1908 the Government o£ Ceylon were contemplating' a new and revised edition of Tumour's translation of the Maha-vamsa, published in 1837 and reprinted in L. C. Wijesinha's Mahavamsa published in 1889, and were in correspondence on the subject with the Ceylon Branch of the Eoyal Asiatic Society. The Society appointed a numerous and influential Committee, and recommended myself as Editor for Europe.1 By their letter of July 18, 1908, the Government of Ceylon requested me to undertake that post. I took the opportunity at the Congress of Orientalists held at Copenhagen in August, and again at the Congress on the History of Religions held in September at Oxford, to consult my colleagues on the best plan for carrying- out the proposed revision. They agreed that the method most likely to lead to a satisfactory result within a reasonable time was to entrust the work to one competent critical scholar who eould^ if necessary, consult members of the Ceylon Committee, but who should be himself responsible for all the details of the work, I reported to Government accordingly, and recommended that Prof. Geiger, who had just completed his edition of the text, should be asked to undertake the task. The Government approved the plan, and asked me to make the necessary arrangements. Those arrangements have resulted in the publication of the present volume.

Professor Geiger has made a translation into German of his own revised critical edition published by the Pali Text Society

3 See the Jonm-al of the Ceylon Branch of the Eoyal Asiatic Society, ¥01 xxi, no. 61, pp. 40-42, 70, 86.vi Editors Preface

in 1908 ; and added the necessary introduction,, appendices, and notes. Mrs. Bode has translated the German into English ; and Professor Geiger has then revised the English translation.

The plan has been to produce a literal translation, as nearly as possible an absolutely correct reproduction o£ the statements recorded in the Chronicle. It is true there is considerable literary merit in the original poem, and that it may be possible hereafter to attempt a reproduction also, in English unrhymed verse, of the literary spirit of the poem. But a literal version would still be indispensable for historical purposes. For similar reasons it has been decided to retain in the translation certain technical terms used in the Buddhist Order. In a translation aiming at literary merit some English word more or less analogous in meaning might be used, regardless of the fact that such a word would involve implications not found in the original. Thus bhik&hu has often been rendered c priest' or e monk*. But a Wdkkhu claims no such priestly powers as are implied by the former term, and would yield no such obedience as is implied in the other; and to discuss all the similarities and differences between these three ideas would require a siaall treatise. There are other technical terms of the same kind. It is sufficient here to explain that when such terms are left, in the present translation, untranslated, it is because an accurate translation is not considered possible. Most of them are, like Mi&Mit, already intelligible to those who are likely to use this version. But they are shortly explained in foot-notes; and a list of them, with further interpretation, will be found at the end of the volume.

The Ceylon Government has defrayed the expense of this, as it did of the previously published translations of the Mahl-vamsa.

T. W. EHYS DAVIDS,TABLE OF CONTENTS

CHAPTER PAGE

INTRODUCTION ...... ix

Abbreviations ...... Ixiv

I. The Visit of the Tathagata .... 1

II. The Race of Mahasammata . . .10

III. The First Council..... 14

IV. The Second Council..... 19

V. The Third Council ..... 26

VI. The Coming of Vrjaya . . . . 51

VII. The Consecrating of Vijaya . . .55

VIII. The ConsecratiDg of Panduvasudeva . . 62

IX. The Consecrating of Abhaya ... 65

X. The Consecrating of Pandukabhaya . . 68

XI. The Consecrating of Devanampiyatissa . 77

XII. The Converting of Different Countries . 82

XIII. The Coming of Mahinda .... 88

XIV. The Entry into the Capital .... 91 XV. The Acceptance of the Mahavihara . . 97

XVI. The Acceptance of the Cetiyapabbata-vihara. 114

XVII. The Arrival of the Relics . . . . 116 XVIII. The Receiving of the Great Bodhi-tree . 122

XIX. The Coming of the Bodhi-tree . . .128 XX. The Nibbana of the Thera . . . .136

XXI. The Five Kings......142

XXII. The Birth of Prince Gamam . . .146

XXIII. The Levying of the Warriors . . .155

XXIV. The War of the Two Brothers . . . 164 XXV. The Victory of Dutthagamani . . . 170

XXVI. The Consecrating' of the Maricavatti-vihara . 179 XXVII. The Consecrating of the Lohapasada . . 182viil Table of Contents

CHAPTER, PAGE

XXVIII. The Obtaining of the Wherewithal to build

the Great Thupa..... 187

XXIX. The Beginning of the Great Thupa . . 191

XXX. The Making of the Kelic-Chamber . . 198

XXXI. The Enshrining of the Relics . . . 209

XXXII. The Entrance into the Tusita-Heaven . 220

XXXIII. The Ten Kings..... 228

XXXIV. The Eleven Kings..... 238

XXXV. The Twelve Kings..... 246

XXXVI. The Thirteen Kings ..... 256

XXXVII. King Mahasena..... 267

APPENDICES

A. The Dynasty of Mahasainmata .... 273

B. The Buddhist Sects...... 276

C. Campaigns of Pandukabhaya and Dutthagamani . 288

D. List of Pali Terms occurring in the Translation . 292

INDEXES

A. List of Geographical and Topographical Names . 298

B. List of Terms explained in the Notes . . . 299

ADDENDA ......... 300

MAPS

Ancient Ceylon ..... To face page 1

Anuradhapura..... n 137INTRODUCTION

§ 1. Literary questions concerning Dipavamsa and MLahavamsa.

THE LITERARY QUESTIONS connected with the Mahavamsa and the development of the historical tradition In Ceylon have been thoroughly discussed in my hook Dlpavamsa, and MaMvamsa.1 I believe that I have there demonstrated that the two Ceylonese Chronicles are based upon older materials and for this reason should claim our attention as sources of history.

Now, however,, R. O. FRANKE has taken a decided stand against my inferences.2 He disputes the existence of an older historical work as foundation of Dip. and Mah.

The former appears to him to be only a botched compilation of Pali quotations from the Jatakas and other canonical works. But the author of the Mah. has merely copied the Dip. and the same applies to Buddhaghosa and his historical introduction to the Samanta-Pasadika. I have however, I hope, succeeded in combating the doubts and objections raised by FEANKE.S

The defects of the Dip.j which naturally neither can nor should be disputed, concern the outer form, not the contents.

1 Dip. und Mah, *und die geschicJitliche V"berlieferung ^in Ceylon,

Leipzig-, 1905. Translated into English by E. M. COOMABASWAMY, Dip. and Mah., Colombo, 1908. Quotations in the following pages

follow the English edition. I may also refer here expressly to OLDEN-BERG'S remarks, Dtp., ed. Introd,, p. I foil. (1879), as the starting-point for my own.

8 Dtp. und Mah. in the Wiener Zeitsahr. /. d. Kunde des Morgenl. 21, pp. 203 foil.; 817 foil.

s N&ck einmal Dtp. und Mah.; Zeitschr, d. D. morgenl. GeseUsch. 63, p. 540 foil. I note that OLBEKBERO In the Archwf. Religionswissensch. 13. p. 614S agrees with my Inferences against FEAKKE.x Introduction

But that the author of the Dip. simply invented the contents of his chronicle is a thing impossible to believe.

Thus it is our task to trace the sources from which he drew his material. This is made possible for us by the Maha-vamsa-Tlka,, i. e. the native commentary on our chronicle which, under the title Vamsatthappakasini, was composed by an unknown author.

I will then here briefly sum up the principal results of my labours, referring, for confirmation in detail, to my earlier works.

1. In Ceylon there existed at the close of the fourth century A.D., that is, at the time in which the Dipavamsa was composed, an older work, a sort of chronicle, of the history of the island from its legendary beginnings onwards. The work constituted part of the Atthakatha, i. e. the old commentary-literature on the canonical writings of the Buddhists which Buddhaghosa took as a basis for his illuminating works. It was, like the Atthakatha, composed in Old-Sinhalese prose, probably mingled with verse in the Pali language.

2. This Attkakatha~MaJidmmsa existed, as did the Atthakatha generally, in different monasteries of the island, in various recensions which diverged only slightly from one another. Of particular importance for the further development of the tradition was the recension of the monks of the Mahavihara in Anuradhapura, upon which the author of the Hah. Tika drew for his material.

3. The chronicle must originally have come down only to the arrival of Mahinda in Ceylon. But it was continued later and indeed, to all appearance^ down to the reign of Mahasena (beginning of the fourth century A. p.), with which reign the Dipavamsa as well as the Mahavamsa comes to an end.

4. Of this work the DIPAVAMSA presents the first clumsy redaction in Pali verses.1 The MAHAVAMSA is then a new treatment of the same thing, distinguished from the Dip.

1 So far as language is concerned, the author** have been

indicated, for numerous verses? by FKAKKI ; herein lies the

merit of his work, although 1 cannot consent to Mi conclusions.Introduction xi

by greater skill in the employment of the Pali language, by more artistic composition and by a more liberal use of the material contained in the original work. While the authorship of the Dip. is not known the author of the Mahavamsa is known as Mahanama.1

5. It is also on the Dip. that BUDDHAGHOSA bases his historical introduction to the Samantapasadika;2 but he completes and adds to its information with statements which could only have been drawn directly from the Atthakatha.

6. The MAHAVAMSA-TIKA brings to the contents of the Dip. and Mah. further additions, taken from the original work. It was certainly not composed till between 1000 and 1250 A. D. But there can be no doubt that the Atthakatha-Mahavamsa lay before the author, as he also supposes it to be known to his readers and accessible to all.3 For this reason his statements as to the original work, its form and its contents, naturally acquire particular importance.

These conclusions are not in any way altered if I am now inclined to consider the relation between Mah. and Dip. as a closer one than in my first work. That the author of the former knew the latter and used it I have naturally never disputed. But I should now wish, in agreement with FLEET, to go much further and regard the Mah, as a conscious and intentional rearrangement of the Dip., as a sort of commentary to this latter. I also think now that the quotation of the ' Mahavamsa of the ancients' in the procemium of our Mah. refers precisely to the Dip. I have besides already indicated the possibility of this view in my Dtp. and MaL} p. 17. FLEET 4 then translates the well-known passage of the later Culavamsa (38. 59) datva sahassam dipetum Dipa-vamsam samadisi in very illuminating fashion: 'he (king* Dhatusena) bestowed a thousand (pieces of gold) and gave orders to write a dlpika on the Dfpavarasa/

1 See RHYS DAVIDS, Journ. Boy. As. Soc. 1905, p. 391.

2 Edited by H. OLBENBEEG, The Vinaya Pltakam, iii, p. 283 foil

3 1 have indicated in Z.DMG. 63, p. 549 foil., passages in the Mah, T. which undoubtedly bear this out.

4 JJR.AJS. 190% p. 5, n. 1.xii Introduction

The interpretation hitherto given: that this is an allusion to a public recitation of the Dip. must then be abandoned. But this dipika, which was composed by order of Dhatusena, is identified by FLEET with our Mahavamsa. Thus, at the same time, the date of its origin is more precisely fixed. Dhatusena reigned, according to calculations which are to be confirmed further on., at the beginning of the sixth century after Christ. About this time the Mahavamsa was composed.

§ 2. The Trustworthiness of the Ceylon Chronicles.

After these preliminary observations the Ceylonese Chronicles should now be judged particularly with respect to their value as HISTORICAL SOURCES, and the historical data drawn from them should be brought together.

In their character of historical sources the Dip. and Mah. have been very differently appreciated.

PRANKE goes the furthest in scepticism. If he did in the beginning at least admit the POSSIBILITY I that the author of the Dip. had some document or other before him, he has lately said most positively: * in the absence of any sources, the last-named work (i.e. the Dipavamsa) must be considered as standing unsupported on its own tottering feet/ 2 And therefore according to him no historical value can be conceded to the Dip. nor to the Mah. nor finally to the Snap. FRANKE'S scepticism, to which I shall return in discussing the history of the councils, ceases to be well founded as soon as we accept the thesis that the Ceylonese Chronicles are based on the Atthakatha. With this the tradition recedes several centuries, and the probability that it contains historical recollections is correspondingly reinforced, and that thesis must, as I have explained above, be considered as confirmed.

KERNS too expresses himself with great caution on the historical value of Dip. and Mah. He indeed says in his Manual cf Indian Buddhism, p. 9, £. . . the chronicles

1 Literarbche* Centrulttatt, 1906, No. 37, column 1275,1. 2.

2 Journal of the Pali Text Soc. 1908, p. 1.

s, German translation by Jacobi, ii, p. 288.Introduction xiii

Dipavamsa, Mahavamsa, and Sasanavamsa deserve a special notice on account of their being so highly important for the ecclesiastical history of Ceylon.1 But here, however, it is only admitted that the chronicles can be utilized as of value for the period from Devanampiyatissa onwards or perhaps only for a yet later time. For the most ancient times, when the history of continental India is also to be taken into consideration, KEEN is hardly inclined to accept them as authentic sources.

A very trenchant verdict is pronounced by V. A. SMITH in his Asoka on the Ceylonese Chronicles. He says in the plainest fashion: (in this work (i. e. in the Asoka) the Ceylonese chronology prior to B.C. 160 is absolutely and completely rejected, as being not merely of doubtful authority but positively false in its principal propositions/1

Perhaps V. A. SMITH has since modified his judgement. For he says now:2 c These Sinhalese stories the value of which has been sometimes overestimated, demand cautious criticism at least as much as do other records of popular and ecclesiastical tradition/ This sounds less cutting. The warning to handle critically, which the excellent historian considers necessary with regard to the Ceylonese Chronicles, is certainly justified. It applies to all historical documents, and I have no intention at all of disputing the justice of it.

The judgement pronounced by RHYS DAVIDS 3 on Dip. and Mah. sounds much more favourable. He says: f The Ceylon Chronicles would not suffer in comparison with the best of the Chronicles, even though so considerably later in date, written in England or in France/ He also lays stress on the fact that, as is self-evident, those Chronicles contain no pure history. But they represent the traditions of their time and permit us to draw retrospective conclusions as to earlier periods.

Lately H. C. NoEMAN4 has defended the Ceylonese Chronicles, with complete justice as it seems to me, against

1 AwJca, the Buddhist Emperor of India, p. 57.

1 Early History of India (2nd ed., 1908), p. 9.

* India, 1903, p. 274.

4 A Defense of the Chronicle* of the Southern Buddhists, J.R.A,S.

1908, p. 1 foilXIV

Introduction

undeserved distrust and exaggerated scepticism. I draw attention expressly to this essay because it naturally has many points of contact with my own researches.

If we next consider the two chronicles as a whole, without any prepossessions, it is not easy to understand whence this widespread doubt of their trustworthiness. The presentation of the subject, taken as a whole, may be called modest and simple, indeed dry. True, there is no lack of fables and marvellous tales. But they appear as outward decoration which can be easily omitted. Besides, we always meet with such stories of miracles in connexion with events of a quite clearly defined category, namely, when it is a question of celebrating the splendour and majesty of the Buddhist Order.

Mahinda arrives in Ceylon in marvellous fashion, flying through the air; miraculous phenomena accompany the 'Establishment of the Doctrine*, the arrival of the relics, the planting of the Bodhi-tree, and so forth. None of this can appear strange to us. The ornament with which tradition here decks out the victory of the Order and the true faith enfolds a deeper meaning. The facts in themselves are extraordinarily simple; but to the pious sentiment of the believer they seemed great; and fantasy glorifies them with the many-coloured lights of miracle and legend.

I do not conceal from myself that this judgement of the situation lays itself open to the reproach that our method is simply to eliminate from the tradition all the miraculous stories and consider what is left over as authentic history.1 But I think WINDISCH 2 has shown admirably how, in fact, in the Buddhist tradition, around a relative small nucleus all kinds of additions have collected in time, by which events, originally simple, are withdrawn gradually into the region

1 V. A. SMITH, Asoka, pp. 45-46 : * Most writers have been content to lop off the miraeles and to accept the residuum of the story as authentic history. Such a method of interpreting a legend does not seem to be consistent with sound principles of historical criticism.*

2 Mara und Buddha (Abhandl. d. pML-hut. CL der K. Sticks. Gesellsch. d. Fto, xv, 4, 1895), Buddha's Geburt (ib., xxvi, 2, 1907), Die position dfs Mahavagtu (*&,, xxvii, 14,. 1909).Introduction sv

of the marvellous. 'But we must not therefore pout away the child with the bath. Here, too, the task of Science is to lay bare the grain of truth; not only this, but she must seek the meaning and significance of the mythical crown of rays that has gathered round the nucleus. For the mythical is often the covering of deep thoughts/ l

We shall, of course, be obliged to begin by removing the mythical additions. But we need by no means take the residue as current coin. Here we are concerned to examine how far the tradition is established as trustworthy., by internal or external evidence,, and how far shaken as being untrustworthy.

If we pause first at internal evidence then the Ceylonese Chronicles will assuredly at once win approval in that they at least WISHED to write the truth. Certainly the writers could not go beyond the ideas determined by their age and their social position, and beheld the events of a past time in the mirror of a one-sided tradition. But they certainly did not intend to deceive hearers or readers. This is clear from the remarkably objective standpoint from which they judge even the mortal foes of the Aryan race. That certainly deserves to be emphasized. It is true not only of dominating personalities (such as, to all appearance, Elara was) but also of the two usurpers Sena and Guttika it is said, Dip. 18. 47 and Mah. 21. 11: raj jam dhammena karayum.

Besides, the obvious endeavour to make out a systematic chronology is such as to inspire confidence at the outset. Indeed, whole sections of the Dip. consist entirely of synchronistic connexions of the ecclesiastical tradition with profane history and of the history of India with that of Ceylon.

§ 3. External support of the Chronicles.

The above certainly are, in the first place, only general considerations, the value of which I myself would by no means estimate too highly. Meanwhile it is more important that the Ceylonese tradition has after all found support to a considerable extent from external testimony.

. * WINDISCH, Buddha's Geburt, p. 4.XVI

Introduction

1. First as to the LIST OF INDIAN KINGS BEFORE ASOKA,* the statements concerning Bimbisara and Ajatasattu as contemporaries of the Buddha agree with the canonical writings and, in respect of the names, with those of the Brahmanic tradition.

The Jaina-tradition has other names; this, however, does not affect the actual agreement. There can be no doubt that the nine Nandas as well as the two forerunners of Asoka: Candagutta and Bindusara, were altogether historical personages. Here also, in the number of years of Candagutta's reign the Ceylonese tradition agrees completely with the Indian. V. A. SMITH,£ too, does not hesitate to accept the number 24 as historical.

Besides the renowned counsellor of Candagutta, the brahman Canakka (Skt. Canakya) is known to the Ceylonese Chronicles. In respect of the length of Bindusara's reign their statements differ from those of the Puranas by three years, in respect of that of Asoka by only one year. The Ceylonese tradition concerning Indian history since the Buddha is, therefore, not unsupported.

2. The CONVERSION OF CEYLON is, according to Dip. and Mah., and finally, according to the unanimous tradition of the country ifcself, the work of Mahinda, a son of Asoka, and his sister Samghamitta. V. A. SMITH calls the stories relating to this in the Chronicles 'a tissue of absurdities'7.3 Asoka himself mentions Ceylon, as he explains, twice in Ms Inscriptions: in the Rock-Edict XIII, among the countries to which he despatched missionaries, and in Bock-Edict II, among those in which he provides for distribution of medicines.4 Since these Edicts belong to the thirteenth year

1 Cf. the tables to § 9.

2 Early History of India, pp. 115-118, Cf. also AsoJca? p. 95.

3 Asoka,. p. 45. OLDENBEEG also (ibid., p. 46) considers the tradition a pure invention.

4 Cf. the translations in V. A. SMITH'S Asoka, pp. 129-133 and

pp. 115-116. The expression cikisaka (=Skt. cikitsa, p.tikiccha)* which SEKAET translates nmtdes, is rendered by BUHLEE (see Z.D.M.O. 48, 1894, p.. 50) * hospitals'.Introduction

of Asoka's reign there appears to be an error in the Ceylonese tradition which puts the conversion. o£ Ceylon as far on as the eighteenth year. On the other hand Asoka, in the opinion of SMITH, would, if he had really handed over his son Mahinda and his daughter Samghamitta to the Church, and had brought about the conversion of the king of Ceylon, certainly not have neglected to bring it into notice. The name (Samghamitta' is, he thinks, from its very meaning, suspicious.

I discuss the arguments in the reverse order, The name Samghamitta is of course that which she herself assumed on entering the Order. That3 beside this name, under which she became a renowned saint of the Buddhist Church, the lay-name fell into complete oblivion can certainly not cause any surprise.

That Asoka makes no mention of Mahinda and Samghamitta in his Edicts is an argumentwm e silentio. That there is any cogency in such an argument V. A. SMITH will surely not maintain. It is indeed very difficult to say in what connexion the king would be obliged to speak of the matter. It can be perhaps expected chiefly in the so-called Minor Bock-Edict I, the Edict of Rupnath, Sahasram and Brahma-giri. But here the reason would again disappear if with FLEET1 we date this edict in the year 256 A.D. In this case, the sending of Mahinda would be about twenty years earlier than the edict, and would belong to past times.

I certainly do not wish to decide here for or against FLEET'S theory. But it is clear that we are standing on too uncertain ground to allow ourselves to proceed without hesitation from an argumentum e silentio.

Now, finally, what as to the mention of Missions to Ceylon in the Asoka Inscriptions earlier than the thirteenth year of the king's reign ?

I may observe that, at the outset, it is not absolutely certain whether by the Tambapanni of the Inscriptions Ceylon is really meant. Possibly the name may designate the

1 'The Conversion of Asoka,' J.RA.S. 1908, p. 486 foil.; * The Last Edict of Asoka; #., p. 811 foil.; 'The Last Words of Asoka/ /&., 1910, p. 1301 foil. . -..,...

bIntroduction

Tinnevelli district at the southern extremity of India, where the river Tamraparni flows into the sea.1 But, at the same time, if Tambapanni should be understood to mean Ceylon the authenticity of Dip. and Mah. is not affected in the

ESSENTIAL points.

Let us look at the positive contents of the tradition. "We are certain of: (1) the name Mahinda as the apostle of Ceylon. Nor is that disputed by V. A. SMITH. Here the Ceylonese narrative finds gratifying support from Hiuen-thsang 2 who mentions Mahendra by name expressly as the man by whom the true doctrine was spread abroad in the kingdom of Simhala. It is certain: (2) that this Mahendra was a near relative of king Asoka. The Chinese pilgrims call him the younger brother 3 of this latter, the Ceylon Chronicles call him his son. Here we have two conflicting reports, and it would be simply arbitrary to prefer the statement of the Chinese pilgrims to the Ceylonese tradition,

But at what result do we arrive if we put together these established facts and the mention of Ceylon- in the earlier Asoka Inscriptions? Simply and solely that which is self-evident, namely, that before Mahinda relations existed between continental India and Ceylon and efforts were made to transplant the Buddhist doctrine to Ceylon.

But with Mahinda this process comes to a successful end. We ean understand therefore that all the interest became concentrated in Ms person, and that tradition wrought together in dramatic fashion that which was a thing of slow continuous development. I consider that this would always and ia all circumstances have been the critical judgment on the

1 Imp. Gfa&tteer of India, s.v« C£ on this subject HULTSZCH,

J.RJL& 1910, p. 1810, n. 4,

1 ST. JULIEN? MJmoirvs sur leg contrfos occidentales, par Hiouen-ii, p. 140; BEAL, Si-yu-ki, Buddhist Meeonfa of the Western

World, trans! from the Chinese of Hitten-tfasang, ii, pp. 246-24?; T. WATTEBS, On Ymn Chwang, Ii 93, 230, 2$4.

3 Besides Hiuen-thsang we have mention by Fa-Man (see LEGGE, A of Buddhitftle Kingdoms by Fd-Jtun, p, 77) of a younger

brother of a monk, without, however, mention of

iiis luoie Q*r allusion to the mission to Ceylon.Introduction xjx

reports of our Chronicles as to the conversion o£ Ceylon. The fact, in essential respects, holds good, but it is a question of putting it in the right light.

Besides, a hint that Mahinda's mission was preceded by similar missions to Ceylon is to be found even in Dip. and Mah., when they relate that Asoka, sending to Devanampiyatissa, with presents for his second consecration as king, exhorted him to adhere to the doctrine of the Buddha.1

Certainly on chronological grounds this cannot be immediately connected with the notices of the conversion of Ceylon to be found in the inscriptions. But it shows us that, even from the point of view of the Chronicles of Ceylon, Buddhism was not quite unknown in that country already before Mahinda's time.

3. The HISTORY OF THE MISSIONS as related in Dip, and Mah.2 receives most striking confirmation in the inscriptions discovered. On the inner lid of the relic-urn which was found in Tope no. 2 of the Sanchi group there is this inscription: Sapurisa(sa) Majhimasa e(relics) of the pious man Majjhima'. On the outer lid is Sapurisa(sa) Kasapagotasa Hemavatacariyasa e (relics) of the pious man Kassapagotta (i.e. of the Kassapa clan), the teacher of the Himalaya'.3 Now Majjhima is, in fact, named in the Mah. as the teacher who converted the Himalaya region and Kassapagotto there appears as his companion in the Dip.4

Again in the superscription of a relic-casket from Tope no. 2 of the Sonari group the same Majjhima is mentioned.

On another urn from the same Tope we again find the name of Kassapagotta, this time with the epithet Kotiputta and again with the designation ' Teacher of the whole Himalaya'.

In a third urn-inscription Gotiputta (i. e. Kotiputta Kassapa-

1 Dip. 12. 5-6 ; Mali. 11. 34-35 ; Smp. 323 5~8.

2 Dip. 8. 1-13 ; Mah. 12. 1-54. Cf. also Smp. 31417-31825.

8 See CUNNINGHAM, The BWsa Topes, p. 287. Cf. RHYS DAVIDS, Buddhist India, pp. 299-301.

* Mah. 12. 6, 41; Dip. 8.10. Cf. Snip. 31719; Mahalodhivamsa (ed. STRONG) 1155, where also Kassapagotta is mentioned together with Majjhima. Cf. also Mah. Tika, 2227.

b2xx Introduction

gotta) appears in connexion with Dadabhisara. This is evidently the Dundubhissara of the Dip. and the Mahabodhi-vamsa who was also among those theras who won the Himalaya countries to the Buddha's doctrine.1

Finally the name of the thera who, according to tradition, presided over the third council under Asoka's rule, is also shown to be authentic by an inscription in a relic-casket from Tope no. 2 of the Sanchi group.2 There is no doubt that by the Sapurisasa Mogaliputasa is meant the Moggaliputta Tissa of the Ceylonese Chronicles.

4. Moreover, the narrative of the transplanting of a branch of the sacred Bodhi-tree from TJruvela to Ceylon finds interesting confirmation in the monuments.

At least GRUJSTWEDEL, in an ingenious and, to me, convincing way,3 points out that the sculptures of the lower and middle architraves of the East Gate of the Sanchi Tope are representations of that event. Since the Sanehi-sculptures belong to the second century B. c. the representation is distant from the event by roughly speaking, only 100 or at most 150 years.

§ 4. Errors in the Clironology of the Earliest Historical Period,

I consider that such objective confirmation of the Chronicles

proves at the very least this much : that their statements are not absolutely untenable and are at least worthy of being tested. Naturally they are not infallible and the longer the interval between, the time of the events and the time when they are related, the greater the possibility of an objective error, and 60 much the more will the influence of legend be noticeable.

As regards the oldest period from Vijaya to Devanampiya-tissa we feel a certain distrust of the tradition and traditional

? L 7., pp. 816-317.

s CUXKXHGHAX, I. /., p. 289.

s GRUXWEDEL, Buddhixt. In Indien, pp. 72-73. Cf. also RHYS

DAVIDS, India, p. 302.Introduction xxi

chronology from the very fact that Vijaya's arrival in Ceylon is dated on the day o£ the Buddha's death.1 This seems to be a Massed account. Besides, there are the round numbers for the length of the single reigns which have in themselves the appearance of a set scheme and involve^ moreover, a positive impossibility in respect of the last two kings of that period., PANDUKABHAYA and MUTASIVA..

According to our Chronicles2 Pandukabhaya was born shortly before the death of Panduvasudeva. Then followed the reign of Abhaya, twenty years, and an interregnum, of seventeen years. Then Pandukabhaya ascends the throne at the age of thirty-seven years. He reigns seventy years. That would bring his age to 107 years !

This, however, is not enough. Pandukabhaya*s successor is his son Mutasiva. He is born of Suvannapali whom Pandukabhaya had already married before the beginning of his reign. Mutasiva must then have been past the prime of manhood when he succeeded to the throne. In spite of this a reign of sixty years is attributed to him.

It seems to me that certain names and events in the tradition may indeed be maintained, but that the last reigns were lengthened in order to make Vijaya and the Buddha contemporaries.

That in respect of certain facts, the tradition is by no means without value for that first period of Ceylonese history, is shown, for instance, by the account of Pandukabhaya* s campaigns,3 which decidedly gives an impression of trustworthiness.

Also after Devanampiyatissa's reign we find matter for doubt.4 A reign of forty years is attributed to the king

1 Mah. 6. 47. In the Dip. 9. 21-22 it is stated, in a somewhat more general way, that at the time of the death of the Buddha (parinib-b an as am aye, not precisely on the day of the death) Vijaya landed in Ceylon. The same in Smp. 32020.

2 Dip. 11. 1, 4; Mah. 9. 28; 10. 106. See previously TUBNOUK, Mahdwanso, Introd., p. li.

3 Mah. 10.26 foil. See below, Appendix C, p. 288 foil.

4 Cf. also on this subject FLEET, J.E.A.S. 1909, p. 840.XXII

Introduction

mentioned, who is said to have been Mutasiva's second son, although he was no longer young when he ascended the throne. But to him succeeded three younger brothers, Uttiya,1 Mahasiva and Suratissa, each of whom reigned ten (= thirty) years. Nay, after the intervening rule of the two Damilas, Sena and Guttika, which lasted twelve years, a fourth brother, Asela, ascends the throne and also reigns ten years.

The reigns of the sons of Mutasiva, who himself occupied the throne for sixty years, would then cover a period of nioety-two years!

We see clearly that also in the period between Devanampiya-tissa and Dutthagamani there were still gaps in the tradition which were filled in with fictitious construction. For the line of Devanampiyatissa we have again the remarkable round numbers 40 +10 +10 +10 +10.

In the later periods we encounter no such difficulties and impossibilities. The chronology is credible, the numbers appear less artificial and more trustworthy.

But even in that first historical period one fact stands out clearly and distinctly from the wavering traditions concerning the times immediately before and after. That is the reign of Devanampiyatissa and the arrival of Mahinda in Ceylon. And with this we approach the general standpoint from which we have to judge the historical tradition as to the earliest and earlier times in our Chronicles.

§ 5. The Tear of the Buddha's Death.

We have to do with a monkish tradition. The starting-point of its chronological statements is the year of the Buddha/s death. For this tradition naturally not every event

nor every historical personage is important to an equal degree,

but chiefly in so far as they were of importance for the development of the Buddhist community. There are isolated

occurrences and personalities connected, even in early times,

1 The name of Uttiya and his consort is confirmed by an inscription In Periya-Puliyankulam (Northern Province). See ArcJmed&gic&l Sur-rty of Ceylon, Annual $@portf 1905 (xx. 1909 j, p. 45.Introduction

with a certain date which announced the time that had passed since the Buddha's death.1 As for the intervening period the traditions concerning it were far less well established and precise, especially from the chronological point of view.

Here fictions were made, building up and completing the tradition from which subsequently, with those fixed points as framework, the chronological system was developed that we find in the Dip. and Mah.7 as also in the Introduction to the Smp0 and again in the later historical literature of Ceylon. IE the Dip-,, the oldest source accessible to us, this system appears already complete. It is most certainly not a creation of the author of the Chronicle but only taken over, in all probability, from the Atthakatha.

One of the fixed dates, which was established at a specially early period, and which evidently forms the corner-stone of the whole system, is the number 218 for the consecration (abhiseka) of Asoka. The Dip, 6. 1, says :?

dve satani ca vassani attharasa vassani ca I sambuddhe parinibbute abhisitto Piyadassano ll

£ 218 years after the Sambuddha had passed into Nirvana Piyadassano (Asoka) was consecrated/ And the Mah. 5. 21 :?

Jinanibbanato paccha pura tassabhisekato Sattharasam vassasatadvayam evam vijaniyam.

'After the Nirvana of the Conqueror and before his (Asoka's) consecration there were 218 years; this should be known/

3 In the same way, to date the Mahavira in the Jaina tradition the number 155 is evidently decisive as being the sum total of the years between his death and the beginning of Candragupta's reign. See Hemacandra's Parisistaparvan, ed. JACOBI, viii. 339 ; Pref., p. 6. If we accept the year 321 B.C. for this last event we have as result 476 B.C. as the year of Mahavira's death. Certainly this is in contradiction with the Buddhist reckoning in so far as, according to Majjh. Nik. II. 24318 foil., the ' Nigantha Nataputta' (i.e. the Mahavira) must have died BEFORE the Buddha. OLDENBERG, Z.DM.Gt. 34, p. 749.XXIV

Introduction

THAT is TO SAY, THAT AFTER A LAPSE OF 218 YEABS, i.e.

SOMETIME IN THE YEAR 219 AFTER THE BuDDHA^S DEATH, THE CONSECRATION OF ASOKA TOOK PLACE.1

Since Asota had already reigned four years before he performed the abhiseka ceremony2 his accession falls 214 years after the Nirvana. According to the Ceylonese tradition the reign of Asoka was preceded by that of Bindusara, lasting twenty-eight, and that of Candagutta lasting twenty-four years (Mah. 5. 18; Dip. 5. 100). Thus Candagutta would have ascended the throne 214 ? (28 -f 24 years), L e. 162 years after the Nirvana.3 Now this event is one of the few in the earlier Indian history which we can date with some approach to certainty. It falls in the year 321 B.C. or within two years of this date/ allowing for error.

THUS THERE RESULTS AS THE PROBABLE YEAR OF THE

BUDDHA'S DEATH (321 + 162) = 483 B.C. As he died at the age of eighty years the year of his birth should be put at 563 B.C.

But we must emphatically state that this calculation too is hypothetical, that we are only able to give an approximate and not a perfectly exact result. Moreover, we shall see below that, in the Ceylon Chronicles themselves, there is a contradiction which we can hardly pass by.

First of all the whole calculation, as OLDEN BERG 5 has quite justly insisted, rests on the supposition that the date

1 Slightly different In the Snap., p. 29920, which puts the abhiseka in the year 218 (dvinnaxn vassasatanam npari at^harasarae vasacl. On the tradition on Asoka's age of the Northern Buddhists see § II.

2 Dip. 6. 21-22; cf. Smp. l.L Moreover, Mah. 5. 22 contains the

statement. NORMAN, J.ItA.S. 1908, p. 10, Is mlgtaken when he says that, according to the Mah., accession should be put at the year 218 A.B. and the abhiseka at 222.

a With this calculation cf. FLEET, J.R.A.S. 1906, pp. 984-986 and 1909, p. 1 folL, and particularly p. 28 foil. See also WICKEEMA-&INGHE, Eeyfanica, i, p. 142, n. 7.

4 V. A. SMITH, J.RJ..8.1901, pp. 831-834; of India,

pp.

s 19109 po 611.Introduction

218 for Asoka's abhiseka is authentic. It really seems to me that it is just on this very point that scepticism is least necessary. The date is supported by the best testimony and has nothing in it to call for suspicion. The interval of time is certainly not so great that the preserving, within the ecclesiastical world, of a definite tradition a;s to an event of such great importance should be improbable or indeed impossible.

On the other hand we must not forget that the date 321 for Candragupta^s accession, which forms a point of support for the hypothesis, is only approximately correct. A little shifting back or forward is therefore quite possible.

Finally, there is the supposition that the length of Canda-gutta's reign (twenty-four years) and Bindusara^s (twenty-eight) is established with certainty. Now it seems indeed that, with regard to the former, scepticism is quite out of place. Here the northern tradition is in agreement with the southern,1 which is certainly an important point. On the other hand there is a difference of three years in respect of Bindu-sara's reign. Here again there is a possibility that the date may be shifted.

Nevertheless it does seem that on the much-disputed question of the year of the Buddha's death there is a tendency toward unison. Marked differences of view are disappearing, the accepted dates are less far removed one from another.2

The chronology current in Ceylon, Burma, Siam starts out from the middle of the year 544 B.C.3 as the date of the Nirvana. That this date is wrong and contains an error of, roughly speaking, sixty years, is now, we may say, generally admitted. Moreover, FLEET4 has pointed out that this reckoning is by no means based on a continuous tradition

1 Of- below the tables to § 9.

2 For earlier views see FLEET, J.R.A.S. 1909, pp. 4-5; MABEL DUPE, Chronology of India, p. 7 ; KERN, Manual of Indian JBuddhism, p. 107, n. 6.

3 Not 543 ! See WICXREMASINGHE, EpigrapMa Zeylanica, i, p. 122, n. 7. The year of Buddha, 2444, began on May 13, 1900.

4 ' The Origin of the Buddhavarsha, the Ceylonese Beckoning from the Death of Buddha/ J.R.A.S. 1909, p. 323 foil., esp. 332.Introduction

from early times. It is rather a relatively late fabrication, which, probably does not go back further than the twelfth century A.D.1 How the error of sixty years came into the era certainly still needs explanation.

Again,, the date 477 B.C. as the year of the Buddha's death, which was accepted by MAX MULLEU and CUNNINGHAM, must be given up, It rests on the erroneous premise that the year of Candragupta's accession was 315 B.C.2

V. A. SMITH3 accepts 487 or 486 B.C. as the year of the Nirvana, GOPALA AIYER/ who starts from 269 as the year of Asoka's coronation, the year 486 B.C. Both attach, some importance, it would seem, to the so-called cdotted Record',5 which was continued in Canton up to the year 489 A. D. and marks each year, from the date of the Buddha onwards, with a dot. In the year 489 A. D. the number of dots amounted to 975, which would bring us to the year 486 B.C. as the starting-point.

I would not for my part attach too much importance to this ' dotted Record'. It is singularly improbable that in the course of time?it is a question of nearly a thousand years !? not a single error or oversight should have occurred. The essential, to my thinking, is that the difference between the various reckonings is already reduced by now to three or four years. But if V. A. SMITH, from his own standpoint, arrives at a result so closely approaching that to which the corrected Ceylon-Tradition brings us, he might well have been led to a somewhat milder judgment as to their trustworthiness and their value.

Finally, the whole difference comes down to this: whether, agreeing with the Puranas, we allow Bindusara a reign of twenty-five years, or, in agreement with the Mahavamsa, allow him twenty-eight years. In the former case we come to the

1 As it now appears (see below) in the eleventh cautery. s S.B.E., x, 2nd ed., 1008, pp. 43-47. 8 Early History of India, pp. 41-43.

4 The Date of Buddha; Ind. Ant, xxxvii, 1908, p. 341 foil. a See TAKAKUSU, J.R.AJS. 3896, p.436foil; 1897, p. IIS; FLEET, ik, 1909, p. 9,Introduction

xxvu

year 486 as the year of the Nirvana, in the latter case to 483 B.C. I£ we then take the 219th year after the Nirvana as the year o£ Asoka's abhiseka, there results in the former case 268/67 B.C., in the latter 265/64 B.C.

It would be of great importance to us if we might refer the date 256 at the end of the so-called ' Minor Rock-Edict I'l to the years elapsed from the Nirvana to the publication of the Edict. This opinion was formerly held, represented particularly by BOHLEK, and F.LEET.2

But recently the interpretation of that Edict was cleared up to a certain extent. The merit belongs to F. W. THOMAS. 3 He was the first to point out that the expressions vivuthena and vivasa (vivutha), which appear in connexion with the number 256, should be derived from vi-vas in the sense ' to be absent from home, to dwell far away'. Then in his second article he has ingeniously demonstrated that the number 256 does not denote years but nights, i, e. nights and days. In the Sahasram text he first discovered the wordlati = ratri in duve sapamnalatisata = Skt. dve satpancasaratrisate.

These discoveries were acknowledged both by FLEET and HuLTZSCH.4 But now opinions diverge. F. W. THOMAS takes it to mean that Asoka published the Edict when on a religious journey. The number would refer to the 256 changes of camp in the course of this tour of inspection.

But FLEET interprets vivutha and vivasa in another way. According to him the allusion is to the renunciation of the household life, to the life far from house and family. He takes it to mean that Asoka after a reign of thirty-seven years had renounced the throne and the world to spend the rest of his life in religious retreat. His dwelling was the mountain

1 The Edict is to be found in Rupnath, Sahasram, in Brahmagiri and elsewhere. V. A. SMITH, AsoJca, p. 138, n. 3.

2 Cf. BtiHLER, Epigraphia Indica,, iii. 138; FLEET, 'The last Edict of Asoka,' J.R.A.S. 1908, p. 811 foil.

3 Ind. Ant. xxxvii, 1908, pp. 22-23, and especially *Les vivasal.i d'Asoka', Journal Asiatique, May-June, 1910, p. 507 foil.

4 FLEET,/The Last Words of Asoka/ J.R.A.S. 1910, p. 1302 foil.; HULTZSCH, < A Third Note on the Rupnath Edict/ #>., p. 1308 foil.XV1U

Introduction

Tinnevelli district at the southern extremity o£ India, where the river Tamraparni flows into the sea.1 But, at the same time, if Tambapanni should be understood to mean Ceylon the authenticity of Dip. and Mah. is not affected in the

ESSENTIAL points.

Let us look at the positive contents of the tradition. We are certain of: (1) the name Mahinda as the apostle of Ceylon. Nor is that disputed by V. A. SMITH. Here the Ceylonese narrative finds gratifying support from Hiuen-thsang2 who mentions Mahendra by name expressly as the man by whom the true doctrine was spread abroad in the kingdom of Simhala. It is certain: (2) that this Mahendra was a near relative of king Asoka. The Chinese pilgrims call him the younger brother 3 of this latter, the Ceylon Chronicles call him his son. Here we have two conflicting reports, and it would be simply arbitrary to prefer the statement of the Chinese pilgrims to the Ceylonese tradition.

But at what result do we arrive if we put together these established facts and the mention of Ceylon in the earlier Asoka Inscriptions ? Simply and solely that which is self-evident, namely, that before Mahinda relations existed between continental India and Ceylon and efforts were made to transplant the Buddhist doctrine to Ceylon.

But with Mahinda this process comes to a successful end. We can understand therefore that all the interest became concentrated in his person, and that tradition wrought together in dramatic fashion that which was a thing of slow continuous development. I consider that this would always and in all circumstances have been the critical judgment on the

1 Imp. Gazetteer of India, s.v. Of. on this subject HULTSZCH, J.&A.S. 1910, p. 1310, n. 4.

8 ST. JiTLiEHj Memoires sur Us conMes occidentales, par Hionen-ii, p. 140; BEAL, Si-yu-ki, Buddhist Records of the Western Wortd, transl. from the Chinese of Hiuen-thsang, ii, pp. 246-247; T. WATTEBS, On Yuan Chwanff, ii. 93? 230, 284.

3 Besides Hiuen-thsang we have mention by Fl-hian (see LBGOE, A Rtwrd of Buddhifilc Kingdoms "by Fd-hien, p. 77) of a younger brother of Asota? who became a monk, without, however, mention of his name or allusion to the mission to Ceylon.Introduction x*£

reports of our Chronicles as to the conversion of Ceylon. The fact, in essential respects,, holds good, but it is a question of putting it in the right light.

Besides, a hint that Mahinda's mission was preceded by similar missions to Ceylon is to be found even in Dip. and Mah., when they relate that Asoka,, sending to Devanampiyatissa, with presents for his second consecration as king,, exhorted him to adhere to the doctrine of the Buddha.1

3. The HISTORY OF THE MISSIONS as related in Dip. and Mah.2 receives most striking confirmation in the inscriptions discovered. On the inner lid of the relic-urn which was found in Tope no. 2 of the Sanchi group there is this inscription: Sapurisa(sa) Majhimasa c(relics) of the pious man Majjhima', On the outer lid is Sapurisa(sa) Kasapagotasa Hemavatacariyasa ((relics) of the pious man Kassapagotta (i.e. of the Kassapa clan), the teacher of the Himalaya3.3 Now Majjhima is, in fact, named in the Mah. as the teacher who converted the Himalaya region and Kassapagotto thero appears as his companion, in the Dip.4

Again in the superscription of a relic-casket from Tope no. 2 of the Sonari group the same Majjhirna is mentioned.

On another urn from the same Tope we again find the name of Kassapagotta^ this time with the epithet Kotiputta and again with the designation ' Teacher of the whole Himalaya',

In a third urn-inscription Grotiputta (i. e. Kotiputta Kassapa-

1 Dip. 12. 5-6 ; Mah. 11. 34-35 ; Smp. 3235~8.

2 Dip. 8. 1-13; Mah. 12.1-54 Cf. also Smp. 31417~31825.

8 See CUNNINGHAM, The BUlsa Topes, p. 287. Cf. BHYS DAVIDS, Buddhist India, pp. 299-801.

4 Mah. 12, 6, 41; Dip. 8.10. Cf. Snip. 31719; MaMbodhivamsa (ed. STRONG-) 1155, where also Kassapagotta is mentioned together with Majjhima. Cf. also Mah. Tlka, 222r.

b2xx Introduction

Finally the name of the thera who, according to tradition, presided over the third council under Asoka's rule, is also shown to be authentic by an inscription in a relic-casket from Tope no. 2 of the Sanehi group.2 There is no doubt that by the Sapurisasa Mogaliputasa is meant the Moggaliputta Tissa of the Ceylonese Chronicles.

4. Moreover, the narrative of the transplanting of a branch of the sacred Bodhi-tree from Uruvela to Ceylon finds interesting confirmation in the monuments.

At least GRUNWEDEL, in an ingenious and, to me, convincing way,3 points out that the sculptures of the lower and middle architraves of the East Gate of the Sanehi Tope are representations of that event. Since the Sanchi-sculptures belong to the second century B. c. the representation is distant from the event by roughly speaking, only 100 or at most 150 years.

§ 4. Errors in the Chronology of the Earliest Historical Period.

I consider that such objective confirmation of the Chronicles proves at the very least this much: that their statements are not absolutely untenable and are at least worthy of being tested. Naturally they are not infallible and the longer the interval between the time of the events and the time when they are related, the greater the possibility of an objective error, and so much the more will the influence of legend be noticeable.

As regards the oldest period from Vijaya to Devanampiya-tissa we feel a certain distrust of the tradition and traditional

1 CUNNINGHAM, I. Ly pp. 316-317.

2 CUNNINGHAM, 1.1,, p. 289.

3 GRVIFWBVEI^ Buddhist. Kunst in Indien, pp. 72-73. Of. also RHYS DAVIBS, Buddhist India, p. 302.Introduction xxi

chronology from the very fact that Vijaya^s arrival in Ceylon is dated on the day of the Buddha's death.1 This seems to be a biassed account. Besides, there are the round numbers for the length of the single reigns which have in themselves the appearance of a set scheme and involve, moreover, a positive impossibility in respect of the last two kings of that period, PANDUKABHAYA and MUTASIVA.

According to our Chronicles2 Pandukabhaya was born shortly before the death of Panduvasudeva. Then followed the reign of Abhaya, twenty years, and an interregnum of seventeen years. Then Pandukabhaya ascends the throne at the age of thirty-seven years. He reigns seventy years. That would bring his age to 107 years !

This, however, is not enough. Pandukabhaya's successor is his son Mutasiva. He is born of Suvannapali whom Pandukabhaya had already married before the beginning of his reign. Mutasiva must then have been past the prime of manhood when he succeeded to the throne. In spite of this a reign of sixty years is attributed to him.

It seems to me that certain names and events in the tradition may indeed be maintained, but that the last reigns were lengthened in order to make Vijaya and the Buddha contemporaries.

That in respect of certain facts, the tradition is by no means without value for that first period of Ceylonese history, is shown, for instance, by the account of Pandukabhaya's campaigns,3 which decidedly gives an impression of trustworthiness.

Also after Devanampiyatissa>s reign we find matter for doubt.4 A reign of forty years is attributed to the king

s Hah. 6. 47. In the Dip. 9. 21-22 it is stated, in a somewhat more general way, that at the time of the death of the Buddha (parinib-b an as am aye, not precisely on the day of the death) Vijaya landed in Ceylon. The same in Smp. 32020.

s Dip. 11. 1, 4; Mah. 9. 28; 10. 106. See previously TUKNOUK, Mahdteanso, Introd., p. IL

s Mah. 10.28 foil. See below, Appendix C, p. 288 foil.

4 Of. also on this subject FLEET, J.RA.S. 1909, p. 340.XXII

Introduction

mentioned, who is said to have been Mutasiva's second son, although he was no longer young when he ascended the throne. But to him succeeded three younger brothers, Uttiya,1 Mahasiva and Suratissa, each of whom reigned ten (= thirty) years, Nay, after the intervening rule of the two Damilas, Sena and Guttika, which lasted twelve years, a fourth brother, Asela, ascends the throne and also reigns ten years.

The reigns of the sons of Mutasiva, who himself occupied the throne for sixty years, would then cover a period of ninety-two years!

In the later periods we encounter no such difficulties and impossibilities. The chronology is credible, the numbers appear less artificial and more trustworthy.

§5. The Year of tlie Buddha's Death.

We have to do with a monkish tradition. The starting-point of its chronological statements is the year of the Buddha's death. For this tradition naturally not every'event

nor every historical personage is important to an equal degree,

but chiefly in so far as they were of importance for the

development of the Buddhist community. There are isolated

occurrences and personalities connected, even in early times,

1 The name of Uttiya and Ms consort is confirmed "by an inscription In Periya-Puliyanknlam (Northern Province). See Arcfauofagicvl $ur~ ftf of Btport, 1905 (ix, 1909), p. 45.Introduction

Here fictions were made, building up and completing the tradition from which subsequently, with those fixed points as framework, the chronological system was developed that we find in the Dip. and Mah., as also in the Introduction to the Snip., and again in the later historical literature of Ceylon. In the Dip-j the oldest source accessible to us, this system appears already complete. It is most certainly not a creation of the author of the Chronicle but only taken over^ in all probability, from the Atthakatha.

One of the fixed dates,, which was established at a specially early period, and which evidently forms the corner-stone of the whole system^ is the number 218 for the consecration (abhiseka) of Asoka. The Dip, 6. 1, says :?

dve satani ca vassani attharasa vassani ca I sambuddhe parinibbute abhisitto Piyadassano II

c 218 years after the Sambuddha had passed into Nirvana Piyadassano (Asoka) was consecrated/ And the Mah. 5. 21 :?

Jinanibbanato paccha pura tassabhisekato Sattharasam vassasatadvayam evam vijaniyam.

' After the Nirvana of the Conqueror and before his (Asoka's) consecration there were 218 years; this should be known/

3 In the same way, to date the Mahavira in the Jaina tradition the number 155 is evidently decisive as being the sum total of the years between his death and the beginning of Candragupta's reign. See Hemacandra's Parisistaparvan, ed. JACOBI, viii. 389 ; Pref., p. 6. If we accept the year 321 B.C. for this last event we have as result 476 B.C. as the year of Mahavira1 s death. Certainly this is in contradiction with the Buddhist reckoning in so far as, according to Majjh. Nik. II. 24318 foil., the 'Nigantha Nataputta ' (i.e. the Mahavira) must have died BEFOKEthe Buddha. OLDENBERG, Z.DM.G. 34, p. 749." Introduction

THAT is TO SAY, THAT AFTER A LAPSE OP 218 YEARS, i.e.

SOMETIME IN THE YEAR 219 AFTER THE BUDDHA^S DEATH, THE CONSECRATION OP AsOKA TOOK PLACE.1

Since Asoka had already reigned four years before he performed the abhiseka ceremony2 his accession falls 214 years after the Nirvana. According to the Ceylonese tradition the reign of Asoka was preceded by that of Bindusara, lasting twenty-eight, and that of Candagutta lasting twenty-four years (Mah. 5. 18; Dip. 5. 100). Thus Candagutta would have ascended the throne 214 ? (28 + 24 years), i. e. 162 years after the Nirvana.3 Now this event is one of the few in. the earlier Indian history which we can date with some approach to certainty. It falls in the year 321 B.C. or within two years of this date,4 allowing for error.

THUS THERE RESULTS AS THE PROBABLE YEAR OP THE

BUDDHA'S DEATH (321 + 162) = 483 B.C. As he died at the age of eighty years the year of his birth should be put at 563 B.C.

But we must emphatically state that this calculation too Is hypothetical, that we are only able to give an approximate and not a perfectly exact result. Moreover, we shall see below that, in the Ceylon Chronicles themselves, there is a contradiction which we can hardly pass by.

First of all the whole calculation, as OiDENBERG5 has quite justly insisted, rests on the supposition that the date

1 Slightly different in the Smp., p. 29920, which puts the abhiseka in the year 218 (dvlnnam. vassasatanam npari attharasarae vasse). On the tradition on A&oka's age of the Northern Buddhists see § 11.

2 Dip. 6. 21-22; cf. Smp. LL Moreover, Mah. 5. 22 contains the same statement. NORMAN, J.RJL.S. 1908, p. 10, is mistaken when he says that, according to the Mah., accession should be put at the year 218 A.B. and the abhiseka at 222.

3 With this calculation cf. FLEET, J.R.A.S. 1906, pp. 984-986 and 1909, p. 1 foil., and particularly p. 28 foil. See also WICKBEMA-SIKGHE, Epiffraphia Zeylanica, i, p. 142, n. 7.

4 V. A. SMITH, J.RJL8.1901, pp. 831-8S4; Early Histoi-y of India, pp. 38-39.

8 Archit f&r ReliffiQH&eisjenackqft, 1910, p. ill.Introduction

218 for Asoka's abhiseka is authentic. It really seems to me that it is just on this very point that scepticism is least necessary. The date is supported by the best testimony and has nothing in it to call for suspicion. The interval of time is certainly not so great that the preserving, within the ecclesiastical world, of a definite tradition as to an event of such great importance should be improbable or indeed impossible.

On the other hand we must not forget that the date 321 for Candragupta's accession, which forms a point of support for the hypothesis, is only approximately correct. A little shifting back or forward is therefore quite possible.

Finally, there is the supposition that the length of Canda-gutta's reign (twenty-four years) and Bindusara's (twenty-eight) is established with certainty. Now it seems indeed that, with regard to the former, scepticism is quite out of place. Here the northern tradition is in agreement with the southern,1 which is certainly an important point. On the other hand there is a difference of three years in respect of Bindu-sara's reign. Here again there is a possibility that the date may be shifted.

The chronology current in Ceylon, Burma, Siam starts out from the middle of the year 544 B.C.3 as the date of the Nirvana. That this date is wrong and contains an error of, roughly speaking, sixty years, is now, we may say, generally admitted. Moreover, FLEET 4 has pointed out that this reckoning is by no means based on a continuous tradition

1 Cf. below the tables to § 9.

2 For earlier views see FLEET, J.R.A.S. 1909, pp. 4-5; MABEL DUFF, Chronology of India, p. 7 ; KERK, Manual of Indian Buddhism, p. 107, n. 6.

8 Not 543! See WICKREMASINGHE, Epigraphia Zeylanica, i, p. 122, n. 7. The year of Buddha, 2444, began on May 13, 1900.

* ' The Origin of the Buddhavarsha, the Ceylonese Reckoning from the Death of Buddha,' J.R.AS. 1909, p. 323 foil, esp. 332.Introduction

from early times. It is rather a relatively late fabrication, which probably does not go back further than the twelfth century A.D.1 How the error of sixty years came into the era certainly still needs explanation.

Again, the date 477 B.C. as the year of the Buddha's death, which was accepted by MA.X MtJLLEnand CUNNINGHAM, must be given up. It rests on the erroneous premise that the year of Candragupta's accession was 315 B.C.2

V. A, SMITH3 accepts 487 or 486 B.C. as the year of the Nirvana,, GOPALA AiYER,4 who starts from 269 as the year of Asoka's coronation, the year 486 B.C. Both attach some importance,, it would seem, to the so-called 'dotted Record',5 which was continued in Canton up to the year 489 A. D. and marks each year, from the date of the Buddha onwards, with a dot. In the year 489 A. B. the number of dots amounted to 975, which would bring us to the year 486 B.C. as the starting-point.

Finally, the whole difference comes down to this: whether, .agreeing with, the Puranas, we allow Bindusara a reign of twenty-five years, or, in agreement with the Mahavamsa, allow him twenty-eight years. In the former case we come to the

1 As it now appears (see below) in the eleventh centurjr.

2 S.B.E., x, 2nd ed., 1908, pp. 48-47.

* Earty History of India, pp. 41-43.

* *The Bate of Buddha; Ind. Ant. xKvii, 1908, p. 341 foil.

6 See TAKAKUSU-, J.R.A.S. 1890, p. 436 foil; 1897, p. 113; FLEET, ? p, ?,Introduction

year 486 as the year of the Nirvana^ in the latter case to 483 B.C. I£ we then take the 219th year after the Nirvana as the year of Asoka's abhiseka, there results in the former case.268/67 B.C., in the latter 265/64 B.C.

It would be of great importance to us if we might refer the date 256 at the end of the so-called c Minor Bock-Edict I'l to the years elapsed from the Nirvana to the publication of the Edict. This opinion was formerly held, represented particularly by BUHLEE and P.LEET.2

But recently the interpretation of that Edict was cleared up to a certain extent. The merit belongs to J\ W. THOMAS/* He was the first to point out that the expressions vivuthena and vivas a (vivutha), which appear in connexion with the number 256, should be derived from vi-vas in the sense ' to be absent from home, to dwell far away'. Then in his second article he has ingeniously demonstrated that the number 256 does not denote years but nights, i.e. nights and days. In the Sahasram text he first discovered the word lati=ra tri in duve sapamnalatisata = Skt. dve satpancasaratrisate.

These discoveries were acknowledged both by FLEET and HuLTZscn.4 But now opinions diverge. F. W, THOMAS takes it to mean that Asoka published the Edict when on a religious journey. The number would refer to the 256 changes of camp in the course of this tour of inspection.

But FLEET interprets vivutha and vivasa in another way. According to him the allusion is to the renunciation of the household lifes to the life far from house and family. He takes it to mean that Asoka after a reign of thirty-seven years had renounced the throne and the world to spend the rest of his life in religious retreat. His dwelling was the mountain

1 The Edict is to be found in Kupnath, Sahasram, in Brahmagiii and elsewhere. Y. A. SMITH, Asoka, p. 138, n. 3.

2 Of. BtJHLER, Epigraphia Indica, iii. 138; FLEET, 'The last Edict of Asoka/ J.B.A.S. 1908, p. 811 foil.

3 Ind. Ant. xxxvii, 1908, pp. 22-23, and especially 'Les vivasal.i d'A^oka', Journal Asiatigue^ May-June, 1910, p, 507 foil.

4 FLEET, 'The Last "Words of Asoka/ J.R.A.S. 1910, p. 1302 foil.; HULTZSCH, ' A Third Note on the Bupnath Edict,116., p. 1308 foil.Introduction

Suvarnagiri near Girivraja in Magadha.1 Hence In the passage which is preamble to the Edict in the Mysore versions Suvarnagiri is named, and not the capital Pataliptitra, as the place where the Edict, the 'last word of Asoka', was published.

Moreover, the number 256 has, according to FLEET, a special significance. It was not by chance that Asoka published the Edict on the 256th day of his life in retreat. At this very time the 256th year since the Nirvana came to an end. Asoka would thus have spent, for each year elapsed since the Buddha's death, one day in religious contemplation as a brahmacarl.

This is a very ingenious idea. But it would be hazardous for the present time to base further conclusions on this bold and seductive combination.

§ 6. Traces of an era in Ceylon reckoned, from 483 B. C.

Recently, however, the date 483 seems to have found further support. Here we must take into consideration an important observation of WiCKREMASiHGHE,2 which completes the proof adduced by FLEET and discussed above, of the late origin of the Ceylonese era, that starts from the year 544, Indications are to be found that in earlier times, and indeed down to the beginning of the eleventh century, an era persisted even in Ceylon which was reckoned from 483 B.C., as the year of the Buddha's death. From the middle of the eleventh century the new era took its rise, being reckoned from the year 544, and this is still in use.

In dealing with the question we have to date the immediate predecessors of king ParSkramabahu I, beginning* with Udaya III (1507 A.B.).3

As to ParSkramabahn I, we have information from inscrip-

1 Of. also on this, FLEET, * The Con¥eraon of Asoka,1 J.M.A.S. 1908, p. 486 foil

s See Epiffrttfkia %eyl&nic®9 i, p. 155 foil.

s The names are given in WiJESlf HA? The Mah&ra^sa* Part II, translated, pp. xxil-xxiiiIntroduction

tions, confirmed and completed by literary data, according to which he was crowned when 1696 years had elapsed since the Buddha's death, that is, in the year 1697 A.B. Eight years later, 1705 A.B., a second coronation apparently took place. In the fourth year afterwards, when 1708 years had gone by since the Nirvana, that is, in 1709 A. B., he held a Buddhist Synod.1 According to the Ceylonese era those are the years 1153, 1161, 1165 A. D. But this date for Parakramabahu is supported by an entirely independent source, namely a South-Indian inscription at the Temple of Tinivallsvara in Arpak-kama. Thus for the second half of the twelfth century the existence of the Ceylon era, reckoned from 544, is established with certainty.

Now according to the Culavamsa2 (56. 16 foil.) the six predecessors of Parakramabahu, from Parakrama Pandn onwards, reigned 107 years* Thus the accession of the last-named prince falls at 1590 A. B. or, according to the Ceylonese era, 1046 A.B. Moreover, this date is confirmed by the South-Indian Manimangalam inscription, which is dated in the same year,3

According to the latter, Parakrama Pandu was conquered and killed in this year by the Cola king Eajadhiraja I, It Is true the Culavamsa gives Parakrama Pandn a reign of two years, but we must rather take the accession and death of the king as falling in one and the same year, 1590 A. B. = 1046 A. D. Thus it is proved, at the same time, that the Ceylon-era also existed in the middle of the eleventh century.

But from a South-Indian inscription we can also fix a date for Udaya III among the predecessors of Parakrama Pandu, a date which throws a completely new light on the whole reckoning of eras.

1 See the Qalvihara-Inse. of Polonnaruwa, 11, 1-4 (ED. MULLEE, Ancient Inscr. of Ceylon, pp. 87, 120); Nikaya-sangraha, ed, WICK-REHASXNGHE, pp. 2G2S, 221 Cf. Epigr. ZeyL i, p. 123.

2 I designate thus the later continuation of the Mahavamsa from 37. 51 onwards.

5 HULTZSCH, South Indian Inscriptions^ iii, no. 28, p, 53; Epigr. L pp. 80, 155.Introduction

Since, according- to the Ctilavamsa,1 the time between the accession of Udaya III and that o£ Parakrama Pandu amounts to ninety-three years eight days, and, as we saw above, the latter ascended the throne in 1590 A. B., we have consequently for the accession of this former king the date 1497 A. B. But this year, according to the Tanjore inscription of king Bajendra Coladeva, must be about the year 1015 A. D.

The inscription2 gives an account of a military expedition to Ceylon. This invasion by Cola corresponds as to its details with one which, according to the Culavamsa 53. 40 foil., occurred under Udaya III at the beginning of his reign. KIELHORN has calculated the time of Coladeva's accession as between the end of 1011 and the middle of 1012 A. D.; the expedition falls between the fourth and sixth year of the reign, that is, between 1015 and 1018. These years must coincide with the years 1497 and 1498 A. B. Of the 1497 years ( ? 1015) remain 482, which fall within pre-Christian times. In other words : THE BUDDHA DIED 483 B. c.

So, with WICKBBMASINGHE (L 1.9 p. 157) we must state the matter thus. The author of that part of the Culavamsa which deals with the kings from Udaya III to Parakrama-bahu I lived at a time when the present era, reckoned from 544 B. c., was in use. He was acquainted with three well-established dates,, 1497,1590, and 1692 A. B., for the accession of Udaya III, Parakrama Pandu, and Parakramabahu I. But he did not know that the first of the three dates was based on quite a different era, reckoned from 483 B.C. The interval between Udaya III and Parakrama Pandu amounted,, in his view, to ninety-three years, but was in reality only thirty-one years (1015-1046 A. ».).

Certainly, considering the detail in which the events of the period from Udaya III to Parakrama Pandu are described by the Culavamsa, it is difficult to say at what point we should undertake to strike out the surplus of sixty-two years. The

1 See WIJESINHA, 7. /.» p. xiii.

3 HCLTZSCH, South Indian Inscr. ii, no. 9, pp. 90-93; KXBLHOBH, JSyigtraphia Indica} vii, p. 7 ? Epfgr. Zeyl. i, p. 79.Introduction

principal part must perhaps fall within the reign of Mahinda V and the interregnum that followed^ for which thirty-six years and twelve years are set down. But that the tradition regarding the period in question is not well established is easily explained by the unrest and confusion which prevailed at that time.

§ 7. The dates of Devanampiyatissa and Duttha-gamani.

The tradition according to which Asoka was consecrated king 218 years after the Nirvana certainly arose in India. The first envoys of Buddhism brought it to Ceylon with them,, and here A CHRONOLOGICAL CONNEXION WAS ESTABLISHED

BETWEEN THE REIGN OP ASOKA AND THAT OF DEVANAMPIYATISSA, under whom Buddhism made its entry into Ceylon.

That Devanampiyatissa and Asoka were really contemporaries we have no reason to doubt. On the one hand the Ceylonese tradition concerning the missions is supported by the discoveries in the Bhilsa-topes. On the other hand we know from Asoka's inscriptions that as a matter of fact an eager missionary-activity prevailed in his time.

According to the Dipavamsa DEVANAMPIYATISSA was consecrated king 236 years after the Buddha's death,1 i.e. in the 237th year. According to the Mah. 11. 40 the consecrating of Devanampiyatissa took place on the first day of the bright half of the ninth month, Maggasira (October?November).

Now since, according to Dip. 11. 14, the consecration of Tissa was later by a certain number of years?I shall discuss the passage further on?AND six MONTHS later?than, the abhiseka of Asoka, this latter event must have taken place

1 Dip. 17. 78 :

dve satani ca vassani chattimsa ca samvacchare sambuddhe parinibbute abhisitto Devanampiyo. Observe that the formula used is the same as in 6. 1 for dating Asoka's abhiseka. See above, p. xxiii. The date 236 is also to be found in the NiMya-samgraha, ed. WICKBEMASITOHE, p. 103, and it results in Dip. and Mah. as the sum total of the reigns of all the kings from Vijaya to Devanampiyatissa.xxxii Introduction

in the third month Jettha (April-May)/ and in fact,, as we know, in the 219th year after the Nirvana.

According to the tradition prevailing in Ceylon2 the Buddha died on the full-moon day of the second month of the year Vesakha (March-April), according to our reckoning : of the year 483 B. c. Thus on the same day 265 B. c. the year 218 A.B. would have come to an end. A month later,, roughly speaking, Asoka would be consecrated. In the month Vesakha, 247 B.C. the year 236 A.B. came to an end. In the autumn of the same year the first coronation of Devanampiyatissa took place. A second coronation3 of this king was celebrated in the following Vesakha (March-April), 246 B.C.

But there are certain statements which are not in agreement with this reckoning. In a passage in the Dip.4 it is said that Mahinda came to Ceylon 236 years after the Nirvana. And it is said expressly that this arrival took place on the full-moon day of the third month Jettha (April-May).5 But a new Buddha-year had begun in the preceding month. Thus if Tissa's first consecration falls in the 237th year A.B., then Mahinda's arrival falls in the 238th, that is, not 236 but 237 years had elapsed since the Nirvana,

This contradiction was discovered by FLEET 6 who made an ingenious attempt to explain it.

The full-moon day of Vesakha as the day of the Buddha's death is open to doubt. This day recurs only too frequently in the Buddha's life. On the other hand FLEET points out

1 On the names of the months in the Indian calendar see our transl., note to 1. 12.

2 Mah. 3. 2; Buddhaghosa in Sum. I. 610 and Smp. 2833> 4. Cf. Dip. 5. 1 foil, for-the same results.

8 Dip. 11. 39; Mah. 11.42.

4 Dip. 15. 71 :

dve vassasata honti chattimsa ca vassa tatha Mahindo nama namena jotayissati sasanam.

5 Dip. 12. 44; 17. 88 (thirty days after the second consecration !); Mah. 13. 18. At Dip. 11. 40 read tato masam atikkamma. See OLDENBERG, note on this passage.

6 'The Day on which Buddha died.1 J.JR.A.S. 1909, p. 1 ML; particularly 6, 11, 31.Introduction

that according- to a notice in Hiuen-thsang* the sect o£ the Sarvastivadins puts the date of the Nirvana, contrary to the usual statement, at the eighth day of the second half of the eighth month of the year, Kattika (Sept.-Oct.).1 Following this FLEET reckons the day of the Buddha's death as falling on October 13, 483 B.C.

If we take this day as our point of departure the above-mentioned contradiction disappears. The year 218 A.B. came then to an end on October 13, 265, and Asoka was not crowned in this year, but in the year 264 B.C. in the third month.2 The year 236 A.B. ends on October 13, 247 B.C., a month later in the year 237 A. B. Tissa was consecrated king;3 in the same year, five months later, there followed the second4 coronation, and yet one month later the arrival of Mahinda in Ceylon.

We have then the following dates :?

1. October 13, 265, end of the year 218 A.B.

2. April 25, 264, Asoka's abhiseka.

3. October 13, 247, end of the year 236 A.B.

4. November 6, 247, Tissa's first coronation.

5. April 16, 246,, Tissa's second coronation.

6. May 16, 246, Mahinda comes to Ceylon.

But here I must point out a difficulty which shows, to say the least, that our sources are not always exact in their calculation of time supposing- we do not accept a variation by even one year. The death of Mutasiva, and therefore also the first crowning of Devanampiyatissa, we find transferred to the seventeenth year of Asoka, in Snip. 321l, and, as it appears^ also in Dip. 11. 14.5

1 See BEAL, Buddhist Records of the Western World, ii, p. 33 ; STANISLAS JULIEN, Memoires, i, pp. 334-335.

2 The day, according to FLEET, is April 25. J.E.A.S. 1909, pp. 26 and 31.

8 According to FLEET, I. L, p. 32, on November 6.

4 According to FLEET, LL, on April 16,

5 The phrasing in the Smp. Asokadhainmarajassa sattara-same vasse idha Mutasivaraja kalam akasi Devan.ampiya-tisso rajjam papuni is not at all ambiguous. The Dip. expresses*

cXXXIV

Introduction

But now even if we set out from April 25, 264 (not 265) B.C. as the date of Asoka's abhiseka, the seventeenth year is already ended on the same day of 247. Then Tissa's coronation, as the dates 218 and 236 have already shown, falls, without any doubt, in the eighteenth (not seventeenth) year of Asoka.

But that notice in the Snip, is not an isolated example. At Mak 20. 1 the planting of the Bodhi-tree in Anura-dhapura is transferred to the eighteenth year of Asoka. This, too, does not agree with the reckoning elsewhere. There can be no doubt that that event falls in the nineteenth year of Asoka.1 Naturally, together with that chronological statement, other dates based upon it and given by the Maha-vamsa 20. 2 foil, are shifted also.

It suffices to point out these discrepancies. They are merely to show that caution is after all not out of place.

2. Further, there is an interesting date connected with the time of YATTAGIMANI. We have, namely, according to Mah. 33. 80-81, an interval of 217 years 10 months and 10 days between the founding of the Mahavihara by Devanampiya-tissa and that of the Abhayagiri-vihara by Vattagamani.2

The date of the consecration of the Mahavihara can be exactly ascertained by the Ceylon chronology. On the full-moon day of the month Jettha Mahinda came to Ceylon. This was, according to FLEET'S calculation,,3 May 16 (246 B.C.). A day later, on May 17, Mahinda came to the capital and

itself less clearly; however, by the words tamhi sattarase vasse chamase ca ana gate I can only understand that there were six months still to come to complete the seventeenth year.

1 We can hardly use the passage Dip. 12. 42-43 for chronology. But it seems to give the correct reckoning, the nineteenth year of Asoka, for Mahinda1s arrival in Ceylon.

3 The same date, possibly taken from the Mak, is to be found in the Nik Samgr., p. 11s6. The Mah. Tika, p. 115 (on Mah. 5. 11-13), gives as the date of the schism of the Dhammarucika of the Abhaya-giri the round number of 217 years after the founding of the faith in Ceylon.

8 J.R.A.S. 1909, p. 28. For the following cf. Mah. 14. 42 ; 15 11, 24.Introduction

spent the night in the Mahameghavana. This the king presented to Mahinda and his companions as an arama on the following day, May 18, 246 B.C. This then is the day of the founding of the Mahavihara. We are brought then to the end of March 28 B. c. for the founding of the Abhayagiri-vihara.

I now believe that we ought to attach special importance precisely to those dates which state generally the interval between two important events. The date number 218 in connexion with Vattagamani was also known in later times.

It is implied in the number 454 of Vattagamani in the Galvihara-Inscription of Polonnaruwa.1 For this has evidently arisen from the addition of 236 (the date of Devanampiya-tissa) to 218.

Moreover, there can be no doubt as to the statement in Mah. 33. 78 foil, that the founding of the Abhayagiri-vihara took place in the second half of the reign of Vattagamani. Therefore I do not hesitate to place the beginning of this second half of Vattagamanias reign at the end of the year 29 or the beginning of the year 28 B. c.

Of course this leads us into certain difficulties when we add up the figures of the individual reigns between Devanampiya-tissa and Vattagamani according to the readings accepted in my edition. From these figures it results that Vattagamani ascended the throne for the second time in the year 39 B. c. "We have then a difference, in round numbers, of about ten years.

This difficulty disappears if we read2 Mah. 21. 11, with the Singhalese MSS. (duve) dvavisavassani, not with the Burmese duve dvadasa vassani, to give thus to the Damilas Sena and Guttika twenty-two and not twelve years* reign. To be sure the Dip. (18. 47) has dvadasa vassani, which certainly must be taken into account. On the other hand the later Ceylonese literature (Thupavamsa, Pujavaliya, Raja-

1 ED. MULLEK, Ancient Inscriptions of Ceylon, p. 87 (Sara siya supsenses hawuruddak). See FLEET, J.R.A.S. 1909, p. 330.

2 In my edition I originally accepted the former reading, however in the ' Corrections ' (p. 368) I have given the preference to dvadasa.

c 2XXX VI

Introduction

valiya1) only gives the number 22. In any case at the time the Tliup. was composed, according to it, the date stood so in the Mali.

Naturally, to be consequent, we must also read Mah. 27. 6 in the prophecy concerning Dutthagamani, with the Sinhalese MSS, cha cattalisa satam '146' or cattalisa satam f 140 \ From the point of view of textual criticism the latter reading seems to me to be the safer; also I should be inclined to believe that in this connexion a round number would be more appropriate.

I confess that I only brought myself unwillingly to depart from the reading of the Burmese MSS. They contain elsewhere, without doubt, the better text. Perhaps we must conclude that, in regard to Sena and Guttika, the Burmese recension adopted the reading of the Dip. and that, in accordance with this, in Mah. 27. 6, also the number was altered to chattimsasatavassani to do away with the mistake thus caused in the addition total.

Taking as a basis the date 483 B. c. we can provisionally draw up a list of the kings according to Dip. and Mah.2

§ 8. List of the Ancient Kings of Ceylon.

Length < 3f Reign Buddh

No. Name Dip. Mali. Dip. Mab. Era 483B.C. Christian Era

Y. M. D. Y. M. B.

1 Viiavfl .... 942 7 74 S3 ? " *JQ 1?88 483 445

o 8 4 & Interregnum . Panduvisudeva Abbaya . . . Interregnum . Pandukabhaya . Mutasiva . . . 11.9 10.5 10.7 11.11 11.4 11.5 f"17 7fi> 8.5 9.25 10.52 10.105 10.106 11.4 2 __ __ so ------- 20 ------- 17 ------- 70 ------- 60 ------- 1 ----------- SO -------- 20 -------- J7 -------- 70 -------- 60 -------- 88-39 39-69 69-89 89-106 10B-176 176-23B 445-444 444-414 414-S94 394-377 377-307 807-247

236 ------- 236 --------

1 For,tlie passes see Dip. and Mali., p. 120.

2 See FLEET'S list, J.E.A.SL 1909, p. 350. The particular aim of tliis Introduction

obliges me, on my side, to draw up a table to enable the reader of the translation to a rapid survey.Introduction

XSLXVIl

Name Dip. Mah. Length of Reign Buddh. Era 483 B.C. Christian Era

Dip. Mah.

Y. M. D. Y. M. D.

Devanampiyatissa. TJttiya .... 17.92 17.93 18.45 18.46 18.47 18.48 18.49 18.54 20.7 20.8 20.9 20.12 20.13 20.14 20.15-17 20.19 20.22 20.24 2025 20.26-30 20.35 21-30 21.33 21.37 21.38 21.40 21.41 20.28 20.57 21.1 21.8 21.11 21.12 21.14 (27.6) 32.35, 57 33.4 33.19 33.28 33.29 33,37 33.56-61 33.102 34.1 34.13 34.15 34.18-27 34.30 34.37 34.69 35.1 35.9 35.12 35.14 35.27 40 ------- 10 ------- 10 ------- 10 ------- 12 ------- 10 ------- 44 ------- 40 ------- 10 ------- 10 ------- 10 ------- 22* -------- 10 ------- 44 ------- 236-276 276-286 286-296 296-806 306-328 3S8-338 388-382 382-406 406-424 424 424-433 433-439 439s 439-454 454-466 466-480 480-492 492-495 495-499 499-521 521-549 549-561 561-571 571-574 574-575 575 575-578 B.C. 247-207 207-197 197-187 157-177 177-155 155-145 145-101 101-77 77-59 59 59-50 50-44 44 44-29 29-17 17-3 3 B.C. -9 A. D. 9AJD.-12A.D. A.D. 12-16 16-38 38-66 66-78 78-88 88-91 91-92 92 92-95

Mahasiva .... Suratissa .... Sena . ) Guttika \ ' ' '

Elara ....

Dutthagamani . . Saddhatissa . . . Thulathana . . . Lanjatissa . . . Khali atanaga . . (Maharattaka) . Vattagamani . . Five Damilas Pulahattha (3 y.) Bahiya(2y.) . . Panayamava (7y.) Pilayamava (7 m.) Da'thika (2 y.) . . Vattagamani . . MahaeuliMahatissa Coranaga .... Tissa . . .

136 ------- 146 -------

24 ------- 18 ------- ? 1 30 96 ? 6 ------ ------ 1 24 ------- 18 ------- ? 1 10 9 ? 15 6 -------

57 7 11 57 1 25

? 3 ? ? 5 ? 14 7 ? 12 ------- 14 ? ? 12 ------- 3 -------

Siva . . . . \ Vatuka ... Darubhatikatissa V Niliya .... Anula ...'..) Kutaka,nnatissa . . Bhitika'bhaya . . Mahadathikamaha-_ -naga .... Amandagamani Kanirajanutissa . Culabhaya . Sivall . . Interregnum . .

60 ------ 60 3 ?

22 ------ 12 ------ 98 ? 3 ----- - 1 ------ 22 - ------ 28 ------- 12 ------- 98 ? 3 ------ ?1 __ __ ? 4 ?

1 According to the Burmese MSS. only 12 years. See p. xxxv.

2 See the same figure Nik. samgr. 1014.XXXVH1

Introduction

Length of Reign.

Buddh.

No. Name Dip. Mah. Dip. Mah. Era 483B.C. Christian Era

Y. M. D. Y. M. D.

40 Ilanaga .... 2U3 35.45 6 ------- 6 ------- 578-584 95-101

41 Candamukhasiva . 21.45 35.46 87 ? 87 ? 584-593 101-110

42 Yasalalakatissa . . 21.46 35.50 87 ? 78 ? 593-601 110-118

43 Subharaja . . . 21.48 35.56 6 ------- 6 _ ? 601-607 118-124

44 Vasabha .... 22.11 35.100 44 ------- 44 ------- 607-651 124-168

45' Vaiikanasikatissa . 22.12, 27 35.112 3 ------- 3 ------- 651-654 168-171

46 (3-i^abahiikagani ani 22.14, 28 35.115 22 ------- 22 ------- 654-676 171-193

47 Mahalfanaga . . 22.17, 29 35.123 6 ------- 6 ------- 676-682 193-199

180 2 ? 182 3 ?

48 Bhatikatissa. . 22.22, 3&! 56.1 24 ------- 24 ------- 682-706 199-223

49 Kanitthatissa . 22.25, 31 36J5 18 ------- 18 ------- 706-724 223-241

50 Khujjanaga . . 22.32 36.18 2 ------- 2 ------- 724-726 241-243

51 Kuncanaga . . 22.33 3619 ?j __ __ 1 , __ __ 726-727 243-244

52 Sirinaga I . . 22.36 36.23 19 ------- 19 ------- 7if-¥461 244-263

5.1 Voharikatlssa * . 22.45 36.27 22 _____ 22 ------- 746-768

54 Abhayanaga 1 . 22.38 36.51 22 ------- 8 ------- 768-776 285-293

55 Sirinaga II . . 22.46 36.54 2 ------- 2 ------- 776-778 293-295

56 Vijayakumara . 22.51 36.57 1 ------- 1 ------- 778-779 295-296

57 Samghatissa . . 22.52 36.64 4 ------- 4 ------- 779-788 296-300

58 Samghabodhi . 22.53 36.73 2 ------- . 2 ------- 783-785 300-302

59 G-othakabhaya . 22.60 36.98 13 ------- 13 ------- 785-798 302-315

60 Jetthatissa . . 22.65 36.132 10 ------- 10 ------- 798-808 315-325

01 Mahasena. . . 22.66 37.1 27 ------- 27 ------- 808-835 325-352

167 ------- 153 -------

Total sum . . 836 9 11 834 7 25

Of conrse the dates set down can only be regarded as having an approximate value. For the Chronicles, mostly, give the reign of each individual king rounded off in whole

years. Rajavali and Pujavali reckon the sum total at 844 years, 9 months 25 days, the Mkayasamgraha reckons the time up to Mahasena^s accession at 818, and thus the time

up to his death at 845 years.2

1 The Dip. places Abhayanaga before Voharikatissa. This appears to be the cause of the mistake in the figures. The same length of rpigB. is ascribed to Voharikatissa as to his predecessor, who is really his successor. According to Nik. samgr. 12® Voharikatissa ascended the throne 752 years, 4 months 10 days after the Buddha's death.

s Bajavali, eel B. GUNASBKABA, p. 4222; Ptjiv., ed. idem, p. 2IP; Nik. S., ed. WICKBEJCASINGHE, p. 14W.Introduction

From Devanampiyatissa to Mahasena's death 609 years elapsed, according to the later sources.1 But this only proves that the accession of the former should be dated 236 A.B. (609 + 236=845), but naturally nothing can be deduced from this statement to aid us in dating the Nirvana itself.

I will now supplement my list with the names and dates of the immediate successors of Mahasena :?2

62. Siri-Megbavanna 27 years 352-379 A. D.

63. Jetthatissa ' * 9 ? 379-388 ?

64. Buddhadasa 28 ? 388-416 ?

65. Upatissa 42 ? 416-458 ?

66. Mahanama 22 ? 458-480 ? 67-fSotthisena to) 9Q ,on KAQ

75. IPithiya J 29 - 480-509 ?

76. Dhatusena 18 ? 509-527 ,,

77. Kassapa 17 ? 527-544 ?

For this later period we now have an interesting Indian-Ceylonese synchronism which appears to confirm the reckoning having as point of departure 483 B. c.

SYLVAIN Llsvi3 has communicated the following passage from the account of the Chinese Wang Hiiien ts'e. The king of Cheu-tzeu (L e. Ceylon), by name Chi-mi-kia-po-mo (i. e. Sri-Meghavarman *)_, sent two bhiksus to India to the monastery erected by Asoka near the sacred tree of the Buddha in Bodh Gay a. They found no lodging here and subsequently told their king. He sent an embassy to the king then ruling over India^ San-meou-to-lo-kiu-to (i. e. Samu-dragupta), and sought permission to build on the sacred spot a monastery for Ceylonese pilgrims. Thus the synchronism of king Siri-Meghavaima, the successor of Mahasena, with Samudragupta is confirmed. The latter,, according to

1 See Epigr. ZeyL i, p. 143.

2 Cf. Clilav. 37.99,104,178,208,247 (according to the numbering of the Colombo edition of 1877: Mah. 37. 49, 54,128, 158, 197); 38. 1, 112; 39. 58. As to numbers 62, 64, 77, it is said that they died in the twenty-eighth (or twenty-ninth or eighteenth) year. So it is possible that the dates have again been made later by one year.

s Jbwm. As. 1900, pp. 316 foil., 401 foil.

4 The form of this name, as given by the Chinese narrator, results from a confusion between varna and varman.xl Introduction

V. SMITH/ reigned from 326 to (about) 375, the former, according to our reckoning, from 483 as the year of the Nirvana 352-379 A.D.

According to Chinese sources2 another embassy came from Ceylon to China, sent by king Kia-ehe, i. e. Kasyapa, in the year 527 A. D. Evidently this is a reference to Kassapa I whose reign, according to my list, did in fact begin about 527.

§ 9. The Indian Kings from Bimbisara to As oka.

In the table on the next page I have brought together the names of the kings from Bimbisara, the contemporary of the Buddha, to Asoka, according to the Ceylonese, the Burmese, the Nepalese, and the Jaina tradition. On this I will first make the following observations.

The BURMESE TRADITION 3 is undoubtedly dependent on the CEYLOKESE, as represented by Dip. and Mah. Buddhaghosa 4 is also in complete agreement with the Mah. He certainly ascribes a reign of eighteen instead of eight years to Anuruddha and Munda, but the sum total of the reigns of all the kings reckoned up by him at the conclusion is only correct if we alter that eighteen to eight.

The NEPALESE list o£ the Asokavadana5 comes perhaps midway between the Ceylonese and the Jaina tradition. It is specially remarkable that in this too appears the name of

1 Early History of India, p. 266 foil. (of. Ind. Ant. 1902, p. 257). See also FLEET, J.R.A.S. 1909, p. 348. s SYLVAIN Livi, /. /., p. 42; foil. Cf. now also E. R. AYBTON,

J.R.A.8. 1911, p. 1142, on a new fact which speaks in favour of the

reckoning from 483 B.C. On the other hand a difficulty presents itself with respect to the embassy of Mo-ho-nan (i.e. Mahanama) to China in the year 428 A.D. (SYLV. LEVI, pp. 412, 421). At the time there reigned in Ceylon not Mahanama but his elder brother Upatissa II. The former did not ascend the throne till 458 A.D.

8 See on this BIGANDET, TJie Life or Legend of Gaudama ike (I860), pp. 347, 361-363, 371-872, 374-375.

4 Snip, :i2ln foil Cf. also Sum. 15323 foil, where the kings from Bimldilm to N%adasa are enumerated,

6 Cf, BCKNOVF, Introduction £~359. It Is noteworthy that the name of Candragupta isTABLE OF INDIAN KINGS

IHpaViimsa Y&irs Mahavamsa Years Burmese trad. Years Asokavadana Years Jain a trad. Years Puranas Years

Bimbiafira 52 Bimbisara 52 Bimbisara Bimbisara £renika Sisunaga 40

(8,66 69) (2. 29-80)

A jataaattu l 82 Ajatassattu * 32 Ajatassattu 35 Ajatasatru Kunika Kakavarna 36

(8,60-61) (2. 81-82)

Udayabhadda 16 Udayabhadda 16 Udayabhadda 15 Ujayin or U day in 60 Ksemad barman 20

(5.97) (4.1) Udayibhadda

? Anuruddha j g Anuruddha ) Ksatraujas 40

? - Munda j Munda j Munda Bimbisai'a 28

(4.2-8)

Nftgadasa 24 Nfigadasaka 24 Kagadasaka 4 Ajatasatru 25

(11.11) (4.4)

Susunaga 10 Susunaga 18 Susunaga 82 Darbhaka or

(5. 98) (4,6) Darsaka or

Kfilftsoka ? Kalasoka 28 Kalasoka 28 Kakavarnin Harsaka 25

(4.44; 5, 25, 80) (47) Sahalin Udayin 33

Ten Sons of Ten Sons of Bhaddaseiia Tulakuci Nandivardhana 42'

Kalftsoka 22 KaJa«oka 22 and 8 Brothers 83 Mahamandala Mahanandin 43

(5, 99) (5. 14) Prasenajit

? - Nine Nandas 22 Uggasen an anda Nanda Nine Nandas Mahapadma 100

(6. 15) and 8 Brothers 21 and 8 Sons

Candagutta 24 Candagutta 24 Candagutta 24 Candragupta Candragupta 24

(5. 100) (6. 16-18)

BindusSra ? Bindusara 28 Bindusara 27 Bindusara Bindusara Bindusara 25

(5, 101) (6.18)

Asoka 87 Asoka 87 Asoka Asoka Asoka Ajsoka 36

(5. 101) (20. 1-6)

1 The Tibetan tradition appears to be very similar to the Ceylonese. According to it Ajatasatru reigned thirty-two years and Dharmaisoka fifty-four years, from the first to the latter were ten generations of kings, ROCKHILL, Life of the Buddha (1907), p. 233.xlii Introduction

Munda whom the Jainas do not know but who is mentioned in the Anguttara-Nikaya.1 Thus the Ceylonese tradition is in this point confirmed by the Northern tradition.

The JAINA list is based on the Parisistaparvan of Hema-candra.2 It is, I think, generally admitted 3 that in this list Srenika and Kunika correspond to the Bimbisara and Ajata-sattu of the Pali sources. On the other hand the names from Anuruddha-Munda downwards to the Nandas are missing. But among these names those of both Munda and Kalasoka are well established by other testimony, as we shall see presently.

The PUEANIC list has the series Bimbisara-Ajatasatru? Udayin (=Udayabhadda) in common with the Ceylonese. But the Puranas insert yet another king before the last-named, and the Ceylonese Chronicles place those three kings at the head of the whole list; the Puranas range the corresponding four kings in the second half of the list. Moreover, I cannot say that the Purana list inspires me with much confidence. The tradition as to individual names is very unstable in the different Puranas. The same is the case with the dates of the individual reigns, although the totals agree fairly well.4

The question then arises: which list merits the most confidence, the Ceylonese, the Jaina, or that of the Puranas ? JACOBI 5 is disposed to give the preference to the Jaina list. He adheres to the view that Kalasoka, 'the black Asoka/ and Kakavarnin (Kakavarna), 'the crow-coloured/ are one and the same person. That is certainly correct and is confirmed by the fact that Kalasoka in the Pali sources is named

1 A. III. 57M foil. OLDENBERG has already, Z.DM.G. 34 (1880), p. 752, stated this fact.

3 Ed. JACOBI (BibL Ind.), I. 22 foil.; VI. 22 foil., 231 foil.; VIII. 1 foil, 297 foil.; IX. 14 foil.

s JACOBI, The Kalpasutra of WiadmMhu (Abhandl fur die Kunde de* MorgenL vii. l)f Introduction, p, 2. The combination Srenika «= Bimbisara occurs ROCKHILL, Life of Buddha (1907), p. 67.

4 See MABEL BUFF, Tke Chronology of India, Table to p. 322.

8 J%? Kalpasutm, Inttod.; also Z. D. N. G. 34, pp. 185-186, Of. OLDENBERG, Z. D. if. G. 34, p. 750 foil; and further, JACOBI, Z. D.M. G. 35, p. mi MlIntroduction x^i *

as the successor of Susunaga and Kakavarna in the Puranas as the successor o£ Sisunaga.1 Here at least the Southern and the Northern tradition are in agreement.

JACOBI moreover believes Kakavarnin = Kalasoka to be identical with the TJdayin of the Jaina tradition, the Udaya-bhadda2 of the Southern Buddhist sources* The ground for his belief is that it is said of both Udayin and Kalasoka that they removed the royal residence from Rajagrha to Patali-putra. He believes that the Ceylonese tradition has made two kings out of one person, has inserted various new kings between them and has thus artificially filled up the gap of 100 years which, according to the Ceylonese view, had elapsed between the Nirvana and the Second Council. The list of kings as finally drawn up by JACOBI is this :?

Bimbisara (Srenika).

Ajatasatru (Kunika).

Munda (=Darsaka, Harsaka,, &c.).

Udayin (Kal&soka,, ICakavarnin),

Nanda dynasty.

I confess that, in agreement with OLDENBEUG/ I do not feel convinced by JACOBINS grounds for identifying Kalasoka with Udayin. The removal of the residence from Eajagrha to Pataliputra is attributed to Udayin by the Jainas,4 and by the Brahmans (in the Puranas), to Kalasoka in the Burmese tradition5 which, beyond a doubt, comes from Ceylon. Hiuen-thsang attributes it to king Asoka whose lifetime he places a hundred years after the Nirvana. He does in fact know only ONE Asoka whom he names Wu-yau, or, as rendered once phonetically, '0-shu-kia.6 But to all appearance he combined

1 The identification of Kalasoka with Kakavarna has not been taken into account by V. A. SMITH (J.R.A.S. 1901, p. 839 foil), who completely denies the existence of Kalasoka.

2 The name is written Udayibhadda, Mah. 4. 1, 2 in the Sinhalese MSS. The same in B. I. 5025 foil.

3 Z.DM.G. 34, p. 751 foil.

4 Parisistaparvan, VL 33 foil., 175 foil.

5 See RHYS DAVIDS, Buddhist Suttas (S. B. E. xi), Introd., p. xiii.

6 BBAL, Buddhist Records, ii, p. 85 foil.; ST. JULIEN, MJmoires, i, p. 414 foil.xliv Introduction

two different kings in one person. For if he attributes the founding of Pataliputra to an Asoka; this cannot possibly fit in with the historical Dharmasoka of the third century B.C. For we know that Pataliputra was already, under Candra-gupta, the capital of the country. Thus when Hiuen-thsang says that '0-shu-kial or Wu-yau founded the city of Pataliputra he repeated a tradition which originally referred not to the Asoka of the third century but to an earlier king,, who must have lived before Candragupta.

I shall return once more to this subject. Here I will only observe that Hiuen-thsang, in any case with respect to the removal of the royal residence, is against the tradition of the Jamas and nearer to the Burmese. We can say then that the removal is attributed by the Jainas and Brahmans to Udayin, by the Buddhists to Kalasoka.

Is really the only solution to conclude that the two names were one and the same person ? May it not be conjectured with equal or yet more probability that we have here simply a difference in the tradition among the Jainas and Brahmans on the one hand and the Buddhists on the other? Besides even in the Brahmanic tradition Kakavarna = Kalasoka and Udayin are again two different personages. Here then the same duplication must have occurred as in the Southern Buddhist tradition. It becomes therefore the more difficult to accept JACOBI'S hypothesis. It seems greatly preferable to conclude that the Jaina list is defective. In this list Munda too is missing, who seems to be sufficiently established by the Asokavadana and the mention in the Anguttara-Nikaya.

If finally the choice lies between the list of the Puranas and that of the Ceylonese Chronicles, which seems to be more probable and trustworthy., I do not hesitate to give the preference wholly and unreservedly to the latter.

In the Paranas, Nandivardhana and Mahanandin2 must

3 The former in SEAL, p. 90, the latter p. 85. Both names are thus used indifferently in connexion with the same event. This proves that we ought not to conclude, with OLDENBERG- (Vin. Pit. i, Introd., p. xxxiii, n. 1), that the two names represent a remembrance of two different Asokas.

1 It teems that Nandivardhana is to represent the ten sons of Kill-Introduction x*v

fill up some gap or other in the chronology. The reigns of these two together are put down at eighty-five years! But no deeds whatever are recorded.1

Again, in the Puranas yet another king, called Darsaka, fee., is inserted between Ajatasatru and Udayin. That is certainly an error. The Pali canon indubitably asserts/ that TJdayi-bhadda was the son of Ajatasattu and probably also his successor. Otherwise the reign of the father and son would extend over eighty-three years.

Moreover that the two generations of the Nan da,, namely Mahapadma and his eight sons, together reigned for a century is a statement that does not bear the stamp of probability.

The chief difference between the Puranas and the Ceylonese sources lies in the place taken by Kalasoka (Kakavarna) and his father. In the former they are placed at the head of the .whole dynasty, in the latter they are ranged after Bimbisara and Ajatasattu and their immediate successors. Thus, before all, the question is which of the two traditions we decide to accept and whether any reasons can be adduced for our decision.

Now we see that the tradition of Ceylon in its details always finds support from without. Its greater fullness of detail, generally speaking, as against the Jaina list finds a parallel in the Puranas.3 In this respect the Southern Buddhist and Brahmanic traditions support each other.

In all forms the tradition as to the series is well established :? nine Nandas?Candragupta?Bindusara?Asoka, The succession Bimbisara?Ajatasattu-?Udayabhadda is confirmed by the Jaina list and the Asoka vadana. Munda, entirely absent from the Jaina list and the Puranas, is named in the

soka. At least the Mahabodhivamsa (ed, STRONG, p. 98) includes a prince of this name among them. Mahanandin looks like a duplicate of Nandivardhana.

1 Even V. A. SMITH, Early History of India, p. 86, has to admit that they are mere (nominis umbrae'.

2 In the Samannaphala-suttanta, D. I. 5025 foil. The same according to the Tibetan tradition. EOCKHILL, Life of Buddha (1907), p. 91.

3 Also in Tibetan sources. See note to the Table.xlvi

Introduction

Buddhistic canon and in the Asokavadana. And in the way the Asokavadana puts Kakavarnin AFTER Udayin and Munda as the Ceylon Chronicles place their Kalasoka, not BEFORE them as the Puranas place their Kakavarna.

Thus the greater probability seems to be in favour of placing Kakavarna and with him naturally his father Sisunaga in the second half of the series of kings,, not in the first.

I believe then that with respect also to the series of Indian kings before Asoka, the Ceylonese tradition is more valuable than that of the Brahmans and Jainas. The last-named is certainly defective. But as to the Puranas I am compelled to think that when the dynasty before Candragupta had once received the name £aisunaga, then in order to exalt its greatness and antiquity, the eponymos and his immediate successors, including Bimbisara and his successors, were placed at the head of the whole series of kings. This would end in a reversal in the order of the first and second half.

At the present time greater stress is laid, and with justice, on the importance of research in Northern Buddhism.1 It is most important for the understanding of the development of Buddhism. Still I believe that if we wish to learn the origins of Buddhism, and especially the history of those origins, we shall have to draw chiefly upon the Pali sources.

The dates of the Indian kings according to the Southern Buddhist tradition are the following:?

(1) Bimbisara2

2. Ajatasattu

3. Udayabhadda

4. Animiddha)

5. Munda J

6. Nagadasaka

7. SusunSga

8. Kalasoka

9. Ten sons of Kalasoka

11. Nine Nandas

12. Candagutta

13. Bindusara

14. Asoka (a) before and (b) after the abhi seka

1 ?f. e.g. WALLBSEB, Z.D.M.G. 1910, p. 238, in a discussion of DE LA VALISE POUSSIN'S Bouddhteme.

* As to the chronological relation between Bimbisara and the

B.B. 60? B.B. 8 B.C. 543? B.C. 491

8-A.B. 24 491? 459

A.B. 24? A. B. 40 459? 443

? 40- ? 48 i 443? 435

48- 72 435? 411

72- 90 411? 393

90? 118 393- 365

118? 140 365- 343

140- 162 343? 321

162- 186 321? 297

186? 214 297? 269

214- 219 269- 264

219- 256 264? 227

Introduction

§ 10. The Acariyaparampara and Indian-Ceylonese synchronisms.

In the chronological system on which the Dip, and Mali, are based the succession of the great teachers from Upali down to Mahinda plays an important part. This acariya-parampara is of interest because in it there is a continuous synchronological connexion between the history of Ceylon and that of India. Here the system appears carried out in detail and completed.1

Of course the dates must not be considered altogether authentic. Besides, for the most part they fall within the most uncertain period of Indian-Ceylonese history, before the accession of Devanampiyatissa. They only show how in Ceylon the several names and events of tradition were fitted into the framework of the few well-established leading- dates.

It seems doubtful too that the theras mentioned, with the exception of Upali and M6ggaliputtatissa, were Vinaya-pamokkha if indeed this should be taken to mean one having recognized authority in the Church.

Sonaka did not even take part in the Second Council which took place in his time. The leading personages in this were Bevata, Sabbakami, Sambhuta Sanavasi and Yasa. Evidently it was only a question of proving that the ' Succession of TeachersJ of Mahinda could be traced back to Upali, the great authority in the Vinaya at the time of the Buddha.

The list is as follows:?

Buddha more precise statements are furnished by Dip. 8. 56 foil, and Mah. 2.28 foil. According to these the two met for the first time when the Buddha was thirty-five and Bimbisara thirty years of age, i.e. 528 B.C. This was the year 15 of Bimbisara's reign. After that Bimbisara reigned yet another thirty-seven years (till 491 B. a). He was succeeded by Ajatasattu. Eight years after his accession the Buddha died. 1 See NORMAN, J.R.A.S. 1908, pp. 5-6. The list of the patriarchs according to the Northern tradition is quite different. In this the succession is: (1) Zasyapa, who presided over the First Council; (2) Ananda; (8) Sanakavasa; (4) Upagupta, the president of the Second Council; (5) Daitika or Dhitika; (6) Kala, who was principally concerned in the conversion of Ceylon. See BEAL, * Succession of Buddhist Patriarchs' (Ind. Ant. ix, 1880, p. U8 foil.).xlviii Introduction

1. UPALI.1 (a) At the time of the Buddha's death (483 B.C.) he had completed forty-four years from his upasampada. So we should have for this last the date 527 B.C. Buddha's death, according to tradition,, coincides in time with the coming o£^ Vijaya to Ceylon and with the 8th year of Ajatasattu. Vijaya dies in the 14th year of Udayabhadda, i.e. 446 B.C., in the 16th year of the same king, i.e. 444 B.C., Panduvasudeva is crowned king in Ceylon.2

(b) Upali after the Buddha's death becomes Chief of the Vinaya and remains so for thirty years. The sum total of his years, reckoned from the upasampada, amounts to seventy-four. He dies therefore 453 B.C. after, as Dip. 4. 38 says, Udaya had reigned six years.

2. DASAKA.3 (a) He is ordained by Upali, when the latter has completed sixty years of his priesthood, or sixteen years after the Buddha's death, i.e. 467 B.C. This agrees with the statement that it happened in the year 24 of Ajatasattu and in the year 16 of Vijaya. According to Mah. 5. 106 he was then twelve years old, thus the year of his birth was 479 B. c.

(b) Dasaka is (after Upali) for fifty years Chief of the Vinaya, i. e. he dies 403 B. c.5 or according to the Dip., in the year 8 of Susuiiaga. In Ceylon meanwhile (Dip. 11. 10) Panduvasudeva has died in the year 21 of Nagadasaka, i.e. 414 B. c., and Abhaya has been crowned king.

3. SoNAKA.4 (a) He is ordained a priest by Dasaka when the latter has completed forty-five years from his upasampada, therefore 422 B.C. Thus according to Dip. 4. 41. But according to Dip. 5. 78 Dasaka had only been forty years a priest when Sonaka was ordained by him. This brings us to 427 B. c. Here therefore the tradition is uncertain. It also points to the year 10 of Nagadasa or the year 20 of Panduvasudeva as the year of Sonaka's ordination, i.e. 425 or 424 B.C.

1 Dip, 4. 34. 38; 5. 76, 95, 103.

2 Dip. 11. 8. The number of years of Vijaya's reign (38) brings us to 445 as the year of his death. The length ®of the interregnum is given Dip. 11. 9, Mah. 8. 5, as one year.

.3 Dip. 4. 27-28, 43; 5. 91, 95, 96, 98, 104. 4 Dip. 4. 41; 5f 78, 79,32, 95,96, 99, 105.Introduction xlix

(b) Sonaka is Chief of the Vinaya for forty-four years and a priest for sixty-six years. Since Dasaka died 403 B. c. Sonaka's death would fall in 359 B* c. This would bring us again to 425 as the year of ordination. The statement that Sonaka died in the year 6 of the reign of Asoka's sons points also to 359 B»C. as the year of his death. The most probable date of Sonaka's ordination is, however, 423 or 422 B. c.., as we shall see from Siggava^s chronology. According to Mah. 5. 115 Sonaka was fifteen years old when he met Dasaka. He was therefore born in 438 or 437 B. c. In Ceylonl the year 11 of the interregnum between Abhaya and Pandukabhaya corresponds to the year 10 of Kalasoka (=383 B.C.) and the year 58 of Pandukabhaya to the year 2 of Candagutta (= 319s. cf).

4. SIGGAVA ,2 (a) Sonaka confers ordination on Siggava forty years after his own upasampada. At that time Kalasoka had reigned ten years and half a month. In Ceylon eleven and a half years of the interregnum after Abhaya had elapsed. Thus we come to the year 383 (or 382) B. c. and to the year 423 (or 422) as the year of Sonata's upasampada.

(#) Siggava is a priest for seventy-six years and dies in the year 14 of Candagutta. This coincides with the year 307 B. c. There must be an error in the statement that he was head of the Church for fifty-five years. Since Sonaka's death may be reckoned with all probability as occurring in the year 359, Siggava, if he died in 307, can only have held this office fifty-two years.

The year of Siggava's birth, since he was eighteen years old at the time of his meeting with Sonaka (Mah. 5. 120), falls in the year 401 B.C.

5. (a) MoGGALiPUTTATissA^3 He is ordained by Siggava sixty-four years after the latter's upasampada, in the year 2

1 Dip, 5. 69, 81; 5, 80.

2 Dip* 4. 44-46 (cf, with this the note in OLDENBERG'S edition);.

5. 73, 95, 96, 106.,

3 Dip. 5. 69, 81, 95, 96, 101, 107, Belies of (Moggaliputta)tissa, attested by an inscription, have been found in the Sanchi-tope no* 2. See CusHiNGHAar, Ehika Topes, p. 289.

Introduction

of Candagutta and 58 of Pakundaka (i.e. Pandukabhaya)., therefore 319 B.C.

(I) He is Chief of the Vinaya for sixty-eight years after Siggava and dies eighty years after ordination, twenty-six years after Asokaj>sabhiseka( = 264 B.C.). The first two statements accord with 239 B.C., the last with 238 B.C. However, if we place the consecration of Asoka as early as the year 265, which results (see above, p. xxxii) from dating the Buddha's death on the full-moon day of Vesakha, then even according to this reckoning Moggaliputta's death should be placed at 239 B.C.

6. MAHINDA.1 (a) Mo^galiputta ordains Mahinda in the year 6 of Asoka, (reckoned from the abhiseka) or the year 48 of Mutasiva. This brings ns; in both cases, if we take the spring of 265 as that of Asoka's abhiseka, to the time between the spring of 259 and 258. Mahinda was born2 204 A.B. i.e. 279 B. c., thus he was ordained at the age of twenty.

Mahinda comes to Ceylon twelve and a half years after his ordination and eighteen years after Asoka's abhiseka,3 as we have already seen, in the spring 246 B.C.

(U) He dies in the year 8 of Uttiya's reign and on the 8th day of the bright half of the month Assayuja.4 The year of his death is therefore 199 B.C.

I. ACAKIYAPABAMPAIll

Priest Chief of Vinaya

1. Upali . . . 44B.B.? 30A.B. = 527 B.C.? 453B.C. from 1 A.B.

2. Dasaka . . 3GA.B.-94 , = 467 -403 ? ? 30 ?

3. Sonaka . . 00 ^ ? 124, = 423 -359 ?

4. Sig^ava , . 100,, ?176, = 383 -307 ? » 124 ?

5. Moggaliputta 164,, -244, = 319 -239 ?

6. Mahinda . . 224 ? -284, = 259 -199 ?

1 Dip. 5. 82. The time between the ordination of Moggaliputta

and that of Mahinda is here stated to be sixty-six years. It would

be correct to say sixty, as OLDENBERG has already observed.

2 Dip. 6. 20 foil; 7. 21-22; Hah. 5. 209. & Dip. 12. 42; Hah, 13,1,5.

4 Dip. 17. 93, 95; Hah. 20. 32-33.Introduction

II. CEYLONESE AND INDIAN SYNCHRONISMS

Year of Ceylon King Year of Indian King Year of Christian Era

Vijaya 1 = Ajatasattu 8 483 B.C.

16 = j<» 24 467

37 = Udayabhadda 14 446

Panduvasudeva 1 = »j 16 444

20^ Nagadasaka 10 425/4

Abhaya 1 = >» 21 414

Interregnum 11 = Kalasoka 10 383

Pandukabhaya 58 = Candagutta 2 319

Mutasiva 1 = 5? 14 307

48 = Asoka 6 259

§ 11. Tne Buddhist Councils.

According to the Southern Buddhist tradition three Councils, as is known, took place, the first immediately after the death of the Buddha, the second a hundred years later under Kalasoka, the third 236 years after the Nirvana in the reign of Dhammasoka.

There has heen repeated discussion,, especially in recent times, as to the authenticity or non-authenticity of the history of the Councils.1 I am not able, within the limits of this introduction,, to go into all the details. I will rather restrict myself, in the first place, to a resume of that which is recorded in the Pali sources as to the Councils. By way of comparison I will then indicate the most important statements of the Northern Buddhist tradition. Finally, I will endeavour to extract the historical kernel which, in my opinion, is contained in the Ceylonese tradition concerning those events.

1 I would refer chiefly to MINAYEFP, Eecherches sur le Bouddhisme, p. 13 foil.; OLDENBERG, ' Buddhistische Studien/Z.D.Jf.G1. 52, p. 613 foil.; KEEN, Manual of Indian Buddhism, p. 101 foil; T. A. SMITH J.R.A.S. 1901, p. 842 foil.; L. BE LA YALLEE Poussisr, 'Les premiers GoBciles (bouddMques),' Le MusSon, N.S. 6. 1905, p. 213 foil. (cf. 'The Buddhist Councils/ Ind. Ant. 1908, pp. 1 foil., 81 ML); B. 0. FRANKE, *Thc Buddhist Councils at Rajasraba and Vesall/ J.P.T.S. 1908, p. 1 foil; EHYS DAVIDS, Dialogues of the Buddha, iL 76, 77. The Chinese accounts of the First Council have been brought together by SUZUKI, * The first Buddhkt Council,* in theMonist^ sdv. 2, 1904, p. 25S folL

d2Hi Introduction

I can only incidentally, where it appears to me to be absolutely necessary, take up a position with regard to views of other inquirers, and must avoid many explanations which suggest themselves, in order not to overstep the space allotted to me.

First, with regard to the SOUTHERN BUDDHIST SOUUCES for the history of the Councils, the principal, both in age and importance, are Khandhaka XI and XII of the Cullavagga in the Vinaya-Pitaka l which deal with the First arid Second Council.

Then follow the Dip. and Mah. with accounts of the three Councils 2 and also the historical Introduction to Buddha-ghosa's Samantapasadika.3 Moreover, Buddhaghosa treats of the First Council, frequently with the same wording, in the Introduction to his Sumangalavilasinl.4 As secondary sources we may mention the Mahabodhivamsa5 and Sasanavamsa,6 and also in the Sinhalese language principally the Nikaya-Samgraha.7

The NOETHEBN BUDDHIST ACCOUNTS will be mentioned in treating of the several Councils.

The First Council.

The account in C.V. is this :

Malifikassapa, travelling with his disciples from Pava to Kusinara, hears of the death of the Buddha. The monks are profoundly grieved, but Subhadda comforts them with the frivolous utterance that they can now do what they will, and that they are freed from an irksome control.

Thereupon Mahakassapa proposes to undertake a samglti of the Dham ma and the Vinaya, that the doctrine may thus

1 OLDBNBERG, Vin. Pit. ii, p. 234 foil. CLS.JB.K xx, p. 370 foil

? GEIGEB, Z>F/>. and Mah. p. 108 foil. In the Dip. there is a double

account of each Council. 8 See OLBENBEEG, Vin. Pit. iii, p. 283 foil. 4 Ed. RHYB DAVIDS and CABPEHTEB, i. (P.T.S. 1886), p. 2 foil.

? Ed. STROM IP.T.S. 1891), p. 85 foil.

? Ed. M. BODE (P.T.S. 1897), p. 3 foil.

7 Ed. WlCKBEXASINGHB, 1890, pp., 8, 4S 8.Introduction *"*

be kept pure. To this end 500 monks are chosen, among whom, by the wish of the assembly, is Ananda,, though he is not yet an Arahant.

The Council takes place in Rajagaha and passes off in the manner described in the Mah.

Some points are to be added from the C.V. namely:

(1) Ananda relates that the Buddha had, in his presence, declared the community of monks empowered after his death to do away with the less important precepts,1 if they wished. Since they are not able to agree in deciding what is to be understood by this expression, they resolve not to do away with any precept.

(2) Certain reproaches are cast upon Ananda. Although he is not conscious of any fault he acknowledges himself guilty from respect for the Assembly.

(3) The thera Purana enters Rajagaha. He is called upon to take part in the work of the Assembly. He renders due acknowledgment to this work but prefers to hold by that which he himself has heard from the Master's lips.

(4) Ananda further relates how the Buddha, before his death, had also pronounced the monk Channa liable to the brahmadanda penance. The fulfilling of this duty is entrusted to Ananda. Channa is deeply troubled. With zealous endeavour he attains to arahantship, upon which the penance is remitted.2'

As regards the time at which the First Council was held, the Dip. 1. 24; 5.4 mentions the fourth month after the Master's death. This was the second Vassa-month, i.e. Savana, the fifth month of the year.3

This reckoning is based on the tradition according to which the Buddha died on the full-moon day of the month Vesakha.

Buddhaghosa and the Mah. agree with this statement.4 The latter certainly mentions the bright half of Asalha the

1 Khuddanukhuddakani sikkhapadani. SeeMahaparinib-banasutta,D.II. 154.

2 I omit the episode of Udena, C.Y. XL 13-14.

3 See M.V. III. 2. 2 (OLDENBEEG, Vin. Pit. i, p. 137). * Smp. 28582-S5, 28684; Sum. 610~20, 8"-"/Mah. 3.Iiv Introduction

fourth month of the yearl as the beginning of the Council, but adds that the first month was spent in preparations, thus the proceeding did not begin till the month Savana.

It is an obviously later addition which we find in the Sum., that not only the Vinaya and the Dhamma, in all their details, but also the Abhidhamma are established at the First Council.

The same is found in the later tradition.

Among the NORTHERN BUDDHIST SOURCES dealing with the First Council I mention the Mahavastu.2 Here, in agreement with the Southern tradition Kasyapa is given as the originator of the Council, the number of the bhiksus taking part in it is stated to be 500 and the place the Sapta-parna grotto near Eajagrha.

There is, besides, an account in the second volume of the Dulva, the Tibetan Vinaya of the Sarvastivadin sect.3 The fixing of the Canon took place, according to this source, in the following order ; (1) Dharma, by Ananda ; (2) Vinaya, by Upali; (3) Matrka (i.e. Abhidharma) by Mahakasyapa himself. It is worthy of remark that the Dulva puts the accusations brought against Ananda in the time before the beginning of the proceedings, thus before his attainment of arahantship.

Fa-Man and Hiuen-thsang4 also mention the First Council. Tire former gives the number of the bhiksus as 500, the latter as 1,000; the former speaks in a general way of 4 -a collection of sacred books', the latter expressly mentions also the redaction of the Abhidharma by Mahakasyapa.

The Second Council.

According* to C.V. XII. the Second Council takes place 100 years after the Buddha's death, and is brought about

by the dasa vatthuni5 of the Vajji monks of Vesall, which

1 The £031 moon of Asalha of the year 488 fell, .according to .f ACQBI'S reckoning (see FLEET, J.R.A.S. 1909, p. 20) on June 24.

;1 Ed SJBCABT, i, p. 69 foil

' See ROCKRILL, Lift of the, (1907), p. 148 foil.

4 BEAU JReeonto, i, pp. k-lxi; iit pp. 162-164; ItEGGEr

ll"0rtl* of Bwltikivtic Kingdoms* p. 85.

?" ON ton pvinU, according to the PSli-tradition, .see below inIntroduction *v

signified a relaxing of monastic discipline. In the further course of its narrative, too,, the C.V. agrees with the Mah. and the rest of the SOUTHERN BUDDHIST SOURCES. The contrast comes out distinctly between the city-dwelling monks of Vesalland the Arahants living in solitary retreat (a ran flak a, Vin. II. 2996).and of strict tendencies.

Yasa^s speech in presence of the Vesalian upasakas is given in full extent. The disciple of Revata, whom the Vajji monks bring over to their side (Mah. 4. 30) is called Uttara. It is also characteristic that the orthodox monks before they undertake the refutation of the heresies first assure themselves of the consent of SabbakamT, the Samghathera at that time.1

The number of those taking part in the Council is given unanimously as 700.2 The Dip. and the Mah. set the time of the Council in the eleventh year of the reign of Kalasoka (=383-382 B.C.), later documents put it in the tenth year.3 The locality is generally considered to be the Valikarama.4 Only the Dip, (5.29) mentions the Kutagarasala of the Mana vana monastery, I do not think we need attach any importance to this discrepancy, which probably takes its rise in some misunderstanding.

Still it is of importance that the Dip. 5. 30 foil, states, to complete the narrative, that the heretical monks held a separate Council, called Manasamgiti, and that they here

the Translation, note to 4. 9. See for further observations L, DE LA VALLEE POUSSIN, Le Muaeon, N.S. vi (1905), p. 276 foil.; Ind. Ant. 37 (1908), p. 88 foil.

1 C.V. XII. 2. 4-6 = Yin. II, p. 80S19 foil

2 C.V. XII. 2. 9 ( = Vin. II. 30785); Dip. 4. 52 ; Mah. 4. 62 ; Snip. 2947. But when the Dip. 5. 20 speaks of 1,200,000 who took part in the Council it does not contradict itself in this. By this naturally exaggerated number the Dip. means those who took part in the General Assembly. Mah. 4. 60 and Snip. 2949 give for this the same number.

3 Dip. 4.44,47; Mah. 4.8. Cf. Mahabodhiv. 966; Sasanav. 71 ~3 ; Nik. Samgr. 4n.

4'Mah. 4. 50,63 ; Smp. 9415; Mahabodhiv. 9620 ; Sasanav. 613; Nik. Samgr. 64. .Ivi Introduction

made out a different redaction of the Canonical Scriptures. With this may be compared the brief notice in Mah. 5. 3-43 according to which the heretical monks of the Second Council, under the name Mahasamghika, formed a separate sect, as the first branching-off from the orthodox doctrine.

In the NORTHERN TRADITION we have accounts of the second Council in the Dulva,1 from the Tibetan historian Taranatha2, from. Fa-Man and Hiuen-thsang.3

As according to the Southern sources so according to these accounts the ten points of the Vajji monks form the starting-point of the movement.

As to the date there is great uncertainty. In the same way, with respect to the place, the tradition wavers between Vaisall and P&talipufcra.4 Of the famous theras of the Second Council mentioned in the Southern scriptures we meet the following in the Northern:?Sarvakama = Sabbakami, Yasa=Yasa, Salha = Sslha, Sambhuta = Sambhuta Sana-vasl, Revata= Revata, Kuyyasobhita (?) = Khujjasobhita and Ajita = Ajita.

The TMrd Council.

With respect to the Third Council we must, in the first place; depend on SOUTHERN" BUDDHIST SOURCES since it has up to this time been accepted that the Northern Buddhist took no account of this Assembly of the Church. Our oldest source is the Dip. 7. 34-43, 44-59; then comes Smp. 30627 foil, then Mah. 5. 228 folK Respecting the course of events we may refer to the translat