Studies on Early Hungarian and Pontic History (Routledge Revivals) 9780367195588, 9780429203169, 0367195585

Published in 1999, Professor C.A. Macartney was one of the foremost 20th-century authorities on the history of the Danub

99 12 71MB

English Pages 629 [675]

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Cover
Series Title
Half Title
Title
Copyright
Contents
Acknowledgements
Preface
Carlile Aylmer Macartney, 1895-1978
A Select List of C.A. Macartney's Works
I The End of the Huns Byzantinisch-Neugriechische Jahrbücher 10 (Berlin, 1934)
II On the Greek Sources for the History of the Turks in the Sixth Century Bulletin of the School of Oriental and African Studies 11 (London, 1944)
III The Attack on 'Valandar' Byzantinisch-Neugriechische Jahrbücher 8 (Berlin, 1930)
IV On the Black Bulgars Byzantinisch-Neugriechische Jahrbücher 8 (Berlin, 1930)
V The Petchenegs Slavonic and East European Review 8 (London, 1930)
VI The Eastern Auxiliaries of the Magyars Journal of the Royal Asiatic Society (1969) (Cambridge, 1969)
VII The Lives of St Gerard Studies on the Earliest Hungarian Historical Sources 1, Études sur l'Europe Centre-Orientale 18 (Budapest, 1938)
VIII The Composition of the Zágráb and Várad Chronicles and their Relationship to the longer Narrative Chronicles Studies on the Earliest Hungarian Historical Sources 2, Études sur l'Europe Centre-Orientale 18 (Budapest, 1938)
IX The Relations between the Narrative Chronicles and other Historical Texts Studies on the Early Hungarian Historical Sources 3 part 1, Études sur l'Europe Centre-Orientale 21 (Budapest, 1940)
X The Attila Saga, the Hun Chronicle, and T Studies on the Early Hungarian Historical Sources 3 part 2, Études sur l'Europe Centre-Orientale 21 (Budapest, 1940)
XI The Hungarian Texts Relating to the Life of St Stephen Studies on the Early Hungarian Historical Sources 3 [recte 4], Études sur l'Europe Centre-Orientale 21a (Budapest, 1940 [1942])
XII The Interpolations of the Chronicon Posoniense and the Genealogy of Almus in the Chronicon Budense Studies on the Early Hungarian Historical Sources 4 [recte 5], Études sur l'Europe Centre-Orientale 21a (Budapest, 1940 [1942])
XIII Unrecognised Components of the Chronicon Budense Studies on the Earliest Hungarian Historical Sources 6 (Oxford, 1951)
XIV The Origin, Structure and Meaning of the Hun Chronicle Studies on the Earliest Hungarian Historical Sources 7 (Oxford, 1951)
XV The Hungarian National Chronicle Medievalia et Humanistica 16 (Denton, Texas 1964)
XVI Dlugosz et de le Chronicon Budense Revue d'histoire comparée (Budapest, 1946)
XVII The First Historians of Hungary Hungarian Quarterly (Budapest, 1938)
Index
Recommend Papers

Studies on Early Hungarian and Pontic History (Routledge Revivals)
 9780367195588, 9780429203169, 0367195585

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Also in the Variorum Collected Studies Series

PETER F. SUGAR

Nationalism, Politics and Religion

FRANCIS J. THOMSON

The Reception of Byzantine Culture in Medieval Russia

ERIK FUGEDI

Kings, Bishops, Nobles and Burghers in Medieval Hungary

NIKOLAI TODOROV

Society, the City, and Industry in the Balkans, 15th—19th Centuries

ZSIGMOND PAL PACH

Hungary and the European Economy in Early Modem Times

PETER GUNST

Agrarian Development and Social Change in Eastern Europe, 14th -19th Centuries

JERZY TOPOLSKI

The Manorial Economy in Early-Modern East-Central Europe: Origins, Development and Consequences

ANTONI MACZAK

Money, Prices and Power in Poland, 16th—17th Centuries

GUSTAVE ALEF

Rulers and Nobles in 15th-Century Muscovy

SAMUEL H. BARON

Explorations in Muscovite History

THOMAS S. NOONAN

The Islamic World, Russia and the Vikings, 750-900

JERZY KLOCZOWSKI

La Pologne dans l’Eglise m^dievale

WLADIMIR VODOFF

Princes et principautes russes, Xe-XIIe siecle

ALAIN DUCELLIER

L’Albanie entre Byzance et Venise (Xe-XVe s.)

VARIORUM COLLECTED STUDIES SERIES

Studies on Early Hungarian and Pontic History

C. A. M A C A R T N E Y

C.A. Macartney

Studies on Early Hungarian and Pontic History

edited by Lorant Czigany and Laszlo Peter

First published 1999 by Ashgate Publishing Reissued 2019 by Routledge 2 Park Square, Milton Park, Abingdon, Ox.on, OX! 4 4RN 52 VanderbiltAvenue, New York, NY 10017

Routledge is an imprint ofthe Taylor & Francis Group, an informa business Copyright© Lasz16 Peter 1999 All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Publisher's Note The publisher has gone to great lengths to ensure the quality of this reprint but points out that some imperfections in the original copies may be apparent. Disclaimer The publisher has made every effort to trace copyright holders and welcomes correspondence from those they have been unable to contact. A Library of Congress record exists under LC control number: ISBN 13: 978-0-367-19558-8 (hbk) ISBN 13: 978-0-429-20316-9 (ebk)

CONTENTS Acknowledgements Preface

vii ix-xiii

Carlile Aylmer Macartney, 1895- 1978, G. N. H. Seton-Watson

xiv-xxxv

Proceedings o f the British Academy 67 (London, 1981)

A Select List of C.A. Macartney’s Works I II

The End of the Huns

Byzantinisch-Neugriechische Jahrbiicher 10 (Berlin, 1934)

IV V VI

VII

1-9

On the Greek Sources for the History of the Turks in the Sixth Century

10-19

The Attack on ‘Valandar’

20-31

On the Black Bulgars

32-40

The Petchenegs

41-54

The Eastern Auxiliaries of the Magyars

55-64

The Lives of St Gerard

65-99

Bulletin o f the School o f Oriental and African Studies 11 (London, 1944)

III

xxxvii-xliii

Byzantinisch-Neugriechische Jahrbiicher 8 (Berlin, 1930) Byzantinisch-Neugriechische Jahrbiicher 8 (Berlin, 1930) Slavonic and East European Review 8 (London, 1930)

Journal o f the Royal Asiatic Society (1969) (Cambridge, 1969)

Studies on the Earliest Hungarian Historical Sources 1, Etudes sur VEurope Centre-Orientale 18 (Budapest, 1938)

VI

VIII

The Composition of the Zagrab and Varad Chronicles and their Relationship to the longer Narrative Chronicles Studies on the Earliest Hungarian Historical Sources 2, Etudes sur l ’Europe Centre-Orientale 18 (Budapest, 1938)

100-117

IX

The Relations between the Narrative Chronicles and other Historical Texts Studies on the Early Hungarian Historical Sources 3 part 1, Etudes sur VEurope Centre-Orientale 21 (Budapest, 1940)

118-218

X

The Attila Saga, the Hun Chronicle, and T Studies on the Early Hungarian Historical Sources 3 part 2, Etudes sur VEurope Centre-Orientale 21 (Budapest, 1940)

219-350

XI

The Hungarian Texts Relating to the Life of St Stephen Studies on the Early Hungarian Historical Sources 3 [recte 4], Etudes sur VEurope Centre-Orientale 21a (Budapest, 1940 [1942])

351-374

XII

The Interpolations of the Chronicon Posoniense and the Genealogy of Almus in the Chronicon Budense Studies on the Early Hungarian Historical Sources 4 [recte 5], Etudes sur VEurope Centre-Orientale 21a (Budapest, 1940 [1942])

XIII

Unrecognised Components of the Chronicon Budense Studies on the Earliest Hungarian Historical Sources 6 (Oxford, 1951)

XIV

The Origin, Structure and Meaning of the Hun Chronicle 449-560 Studies on the Earliest Hungarian Historical Sources 7 (Oxford, 1951)

XV

The Hungarian National Chronicle Medievalia et Humanistica 16 (Denton, Texas 1964)

561-568

XVI

Dlugosz et de le Chronicon Budense Revue d'histoire comparee (Budapest, 1946)

569-586

XVII

The First Historians of Hungary Hungarian Quarterly (Budapest, 1938)

587-598

Index

375-378

379-448

599-629

ACKNOWLEDGEMENTS Grateful acknowledgement is made to the following publishers, editors and organisations for their kind permission to reproduce essays included in this volume: The British Academy, London (Macartney biography); ByzantinischNeugriechische Jahrbiicher, Amsterdam (I, III, IV); SOAS, London (II); the Modem Humanities Research Association and W.S. Maney & Sons Ltd. (V); Cambridge University Press (VI); Ostmitteleuropaische Bibliothek published by E. Lukinich (VII, VII, IX, X, XI, XII); Basil Blackwell, Oxford (XIII. XIV), Medievalia et Humanistica, Denton, Texas (XV); Les Presses Universitaires de France (XVI), Hungarian Quaterly, Budapest (XVII). The editors and publishers would like to express their grateful acknowledgement of financial support towards the publication of this volume from The Warden and Fellows of All Souls College. The Editors would also like to thank Dr Martyn Rady for all the help they have received from him.

PREFACE Carlile Aylmer Macartney (1895- 1978) was a most distinguished scholar on the Danubian basin; he combined knowledge based on meticulous research with an interest in public issues on which, once he made up his mind, he never hesitated to take sides. Not an expert with narrow interests, Macartney’s scholarship was capacious. As Hugh Seton-Watson pointed out in his biography, his expertise encompassed Austria, Hungary, Central Europe and South-East Europe as well as the history of these regions. He wrote books and extensive studies on Austrian politics after 1918; on contemporary Hungarian society; many works on the national question and on the borders of the post First World War settlements; on Maria Theresa; on the Hungarian nobility in the eighteenth century; on Eastern Europe between the two World Wars; and The Habsburg Empire 1790-1918, a comprehensive narrative history which he published at the age of 73, and which is arguably his magnum opus - unless that was the two volume October Fifteenth: A History of Modem Hungary 1929-1945, published in 1957. Macartney’s studies were not, however, confined to the Modem Ages. Soon after the start of his career he acquired an absorbing interest in the earliest periods of Hungarian and Pontic history and attained expertise in demanding subjects like the origins and migrations of the Magyar tribes, the histories of other nomadic steppe-peoples, such as the Petchenegs, Kavars or Cumans, the early history of the Bulgars and the Turks, and, above all, the medieval history of the Kingdom of Hungary. Many of these subjects involved controversial questions that had not been seriously tackled in English or French. Macartney worked his way through the considerable amount of earlier works written in Latin and German in the eighteenth and the nineteenth centuries. The new literature on the subject was, however, accessible only to those proficient in Hungarian, the language in which scholarly studies were mostly written after 1900. It was this new Hungarian (as well as the older Latin and German) literature that Macartney reassessed in his studies by a thorough systematic examination of practically the whole corpus of written sources relating to early Hungarian history. With the fresh eye of an outsider, he developed a new approach by questioning traditionally held assumptions, transmitted from generation to generation about the provenance and the textual relationships of the early Hungarian chronicles and surviving fragments.

X

How he embarked on this enterprise, he explained in the Preface of the concise summary of his findings, published under the title The Medieval Hungarian Historians, A Critical and Analytical Guide (1953), from which is worth quoting in extenso: The various historical and biographical works in which the Hungarians of the Middle Ages recorded their own origins and early doings are less numerous and less important than their counterparts from France, Italy or Germany. Nevertheless, they constitute a not inconsiderable body of literature, which is of great value for the history, not only of Hungary and the Magyar people, but of the whole of South-Eastern Europe. Naturally, however, before they can be safely used as historical sources, they require much editing and interpretation. In the case of each of them, its date and degree of trustworthiness require to be examined, and where - as is the case with the great majority of them - they consist of various component parts, put together by a later hand, then this investigation has to extend to the parts as well as the whole. Where a relationship is discernible between more than one of them, then the nature of that relationship must be investigated. Which is the original? Which the copy? Or, if both derive from a lost common source, which has preserved that source the more faithfully? (p.v)

Having appraised the old Latin and German works, Macartney pointed out the inaccessibility of the new Hungarian literature to most outsiders and made the following confession: Some years ago I conceived the idea that it would be worthwhile for a nonMagyar who happens to read Magyar to go through this mass of material, which enshrines the results achieved by so many learned and subtle brains, and to present its fruits to the non-Magyar reader in a short form which would nevertheless give him all that he needed to know for safe and profitable utilisation of the texts. Imbued with a proper respect for the august scholars of modem Hungary, I imagined that my work would be purely that of a summarizer, interpreter and translator, and did not anticipate that it would occupy me for more than a few months. On looking, however, more closely into the works of Homan and Domanovszky in particular (it is on their results that the whole framework really depends), I found with mingled interest and horror, that I could not possibly carry out the work in the way in which I had proposed it to myself. The labours of these scholars were always ingenious, and, on many points of detail, illuminating; but their main conclusions seemed to me to be, with hardly an exception, demonstrably incorrect. I was obliged therefore to undertake a whole series of highly controversial and argumentative studies in which by elaborate reasoning I sought to demolish their conclusions and to establish my own. (pp. vii-viii; and see also below pp. 119-123)

XI

These studies, he summed up, set out his answers to the various problems presented: ‘I cannot expect to be right in every point of detail, and there are some in which I must frankly admit defeat’ (p. ix). He regarded The Medieval Hungarian Historians the study as I originally planned it, but resting on a somewhat different basis from that which I had originally expected. It summarizes not so much the results achieved by my Magyar contemporaries and colleagues, as those results in so far as I have been able to accept them, or my own results, where I have been obliged to differ from other critics. I have not attempted to repeat here all the lengthy and weary polemics in which I indulged in my Studies. I take my results as achieved ip. viii).

Achieve them he certainly did, yet he was disappointed too. Whereas his works on the modem periods received proper attention and his authority as the most knowledgeable expert on Hungarian affairs was securely established, his seven part Studies in the Earliest Hungarian Historical Sources (1938- 1951) were never accorded the scholarly interest they have undoubtedly deserved. For Macartney re-examined - indeed in some cases he was the first to examine - the composition and the sources of the larger chronicles, the complex relationships between the major and minor narrative texts and made detailed textual comparison with the corresponding Western sources. The shorter chronicles and fragments shed new light on the provenance of many passages in the larger chronicles which were composite versions. He carefully lifted from the text one layer of interpolations after another, separating passages of different provenance. Sometimes he used the method of subtraction, by peeling off successive outer layers of the finished product. Other times he went in reverse, starting with the components of a nucleus and explaining how it had grown incrementally through editions and changes. Imaginative and inventive though his reconstructions frequently were, he remained modest and critical about his own conclusions. Never claiming more than what he had in fact established, he was not afraid of admitting discrepancies, gaps, and failures in attempting to solve textual puzzles. As his work advanced he refined and sometimes changed his interpretations (always spelling out where and what). He firmly adhered, however, to the main conclusions he had reached early on. The central contention which runs like a red line right through the Studies was the discarding of an assumption that Hungarian scholars traditionally take. They believe that the large number of overlapping identical passages in the later chronicles can be explained by a lost, single comprehensive Gesta Hungarorum, an osgeszta which, according to a leading medievalist, Balint Homan, was composed at the court of St Ladislas around 1080 and which preserved an official version of ancient ancestral memories {see below

xii e.g. pp. 541, 544, 561). The case for the existence of a proto-gesta rested on the principle of the highest common factor: the common elements extracted from the surviving chronicles. This hypothetical ‘Gesta’ was the main source on which most other chronicles and fragments were based, consciously correcting, adopting, enlarging, or reducing it to serve changing political purposes. Impressed though he was with the high quality of his Hungarian colleagues’ work, Macartney nevertheless thought that they were all hypnotised by their ‘Gesta’ theory {see below p. 502 n. 2). Of course - he argued - some early writings may well have been lost, but there was no need to make the a priori assumption of an all-embracing single gesta, ‘which no one yet has ever seen, or ever will’ (p. 451). The provenance of the elements as well as the composition of the larger chronicles - Anonymus, Simon de Keza, the Buda Chronicle - and even those of the shorter narrative chronicles were complex. The borrowings by copyists were astonishingly extensive and mutually contaminating. It was a mistake to attribute ‘each complex of confusions to a single brain’. The surviving chronicles were ‘products of copyists, each of whom added his quota of error by not quite understanding what he read (see below pp. 505- 506, a point which, of course, applies to medieval texts in general). Further, by casting his net wider in reconstructing who used what source for which passages, Macartney demonstrated that the Hungarian texts were, in the growth of alternative national traditions dependent to a far greater extent on Western sources than had been assumed. The hypothesis of the eleventh-century ‘Gesta’ is closely tied to the question of whether a genuine tradition connected either the Huns to the Hungarians or Arpad’s dynasty with that of Attila. From the thirteenth century onwards Hungarians believed that they were descendants of the Huns. Scholarship in the nineteenth century discarded this tradition: the Huns were Turkic, the Hungarians, Finno-Ugrian, became the dominant view. Hungarians had discovered the Hun-Hungarian identity, it was now argued, in Western sources well after their arrival in the Danubian region. Writers in the West merely observed that the Hungarians were like the Huns rather than asserting that the two peoples were of the same stock. In the early decades of this century, however, scholars looked afresh at the question of the Turkic elements in the ethnic composition of Hungarians. This gave Homan and others the chance to reopen the vexed question of the connections with the Huns. No serious scholar, let alone Homan, a formidable medievalist, argued either that the Hungarians and the Huns were ethnically related or that most of what the Hungarians had known about the Huns did not in fact come from Western sources. They did argue, however, that the conquering Magyars, or at least Arpad’s family, brought with them a claim to connections with the Huns (or at least with Attila). And it was the lost hypothetical ‘Gesta’ that happened to provide the corroborating (if not the core of the) evidence.

Xlll As one would expect, Macartney’s frontal attack upon the ‘Gesta’ hypothesis was directed at the elements of the Hun tradition (e.g. pp. 332, 362). He argued that the Hungarian chroniclers could easily find in the putative immediate foreign (Western) sources practically all the elements of the ‘Hun tradition’. For instance, it is just conceivable that Geoffrey of Monmouth’s English legend into which the story of the rape of wives crept, might even have been the source of the Hun connection found in the Hungarian chronicles (p. 544). On the other hand, those early foreign writers who came into contact with the Magyars did not know of the Hun-Hungarian connection (pp. 342, 546). The rejection of a comprehensive eleventh-century ‘Gesta’ revealed that the Hungarians had preserved alternative national traditions. Hungarian historians did not take up Macartney’s points in the Studies. Some of the reasons for this neglect were obvious. Hungarian scholars were cut off from their Western colleagues after the Second World War. Before the Iron Curtain closed Macartney’s contacts with Hungarian scholars had been well established. A well-known, popular figure in the country with many friends, he was elected external member of the Hungarian Academy of Sciences in 1946. His repartee to a leading scholar, Gyula Nemeth, who quipped that he was learning English from Macartney’s The Magyars in the Ninth Century (published in 1930) was, ‘I hope you learn early Hungarian history from it as well’. In fact, Nemeth devoted a fair amount in his university lectures in Budapest in 1947 to Macartney’s views. But after 1947 a deafening silence set in for decades. Nor did Macartney’s findings receive critical attention by scholars in the West, who still lacked the linguistic equipment to deal with the historical sources relating to the early history of the Danube region. Another reason for this neglect was that Macartney’s detailed source analyses were scattered in many foreign, now inaccessible periodicals. Hence the present publication, in which the author’s works that illuminate one of the dark comers of Europe’s past are collected. They form a soifrce for Macartney’s The Medieval Hungarian Historians and are indispensable to the student of the Age of Migration and the early history of the Kingdom of Hungary. The works are placed in a thematic order, leaving together, however, the largest portion of this collection, the seven part Studies (pp. 65- 560). Their republication may induce scholars to take the subject further. Since 1945 most Hungarian medievalists have abandoned the search for Hunnish connections, yet most of them still hold fast to the eleventh-century lost ‘Gesta’ theory. The availability of Macartney’s Studies might lead to their revisiting the knotty question of the hypothetical ‘Gesta Hungarorum’ which, with computer-technology to help in untangling relationships among a large number of texts, could either confirm or produce better answers than have so far been found on the genesis of the early Hungarian historical sources. London, December 1998

LORANT CZIGANY & LASZLO PETER

CARLILE AYLMER MACARTNEY1 1895-1978 A ylmer M a c a r t n e y was born on 24 Ja n u a ry 1895. His father was a barrister, and his family of N orthern Irish, and more distant Scottish, origin. His m other’s nam e before m arriage was Louisa G ardiner. M acartney won a scholarship at W inchester, and was in College from 1909 to 1914. He won the school prize for Greek verse and the K enneth Freem an classics prize, as well as the G oddard Exhibition, the school’s aw ard for outstanding classical scholarship. In his last year he held the school office of Prefect of Library. He won an open scholarship at T rinity College, Cam bridge, in 1914. H e did not go up to Cam bridge in October, as w ar had m eanwhile broken out in Europe. He joined the arm y as a private. M any years later he used to say th at the m aster in charge of the Officer T raining Corps at W inchester had disliked him, and refused to recom m end him for an im m ediate commission; and th at he had rem ained eternally grateful to this m an, since he would otherwise almost certainly have been killed. Seven out of fifteen scholars of his year were in fact killed, but M acartney survived. He was commissioned in Novem ber 1914, served in the H am pshire Regim ent for two years, and was then transferred to Royal Field Artillery in ja n u a ry 1917. H e was wounded at Potisje near Ypres in Ju ly 1917. After two m onths in hospital in England, a training course and another period in hospital in Ireland, he returned to France in Ju n e 1918 and served in an anti-aircraft battery until the end of hostilities. After the w ar he returned to take up his scholarship at Trinity, but did not stay to complete his degree. C a r li le

1 In my research for this memoir, I received valuable help from the following persons, to whom I should like to record my gratitude: Mr G. C. W. Dicker, Keeper of Old Wykehamist Records, Winchester College; Miss Alex Ward, Head of Army Historical Branch, Ministry of Defence; Miss Dorothy Hamerton, then Librarian at Chatham House and in charge of its archives; Dr Laszlo Peter, of the School of Slavonic and East European Studies in the University of London; and most of all Miss Elizabeth Barker, the historian of wartime British diplomacy in Central Europe, for her unfailingly generous advice and assistance in tracing papers in the Public Record Office, and for her personal memories of C. A. Macartney’s activities. To all of these my grateful acknowledgements.

412

XV

In 1919-20 he spent some time in C entral Europe as a jo u rn alist, and it is from then that his study of the languages, politics, and cultures of th at region began. In 1921 he was appointed viceconsul in Vienna, a position which he held for four years. In this period his knowledge of the intricacies of D anubian life deepened, and he became personally acquainted with m any of the persons and problems throw n up by the convulsions of the perished H absburg M onarchy. In 1923 he m arried Nedella, daughter of Colonel M am archev of the Bulgarian army. After leaving Vienna, he worked for the Survey o f International Affairs, produced at the new Royal Institute of International Affairs at C hatham House, under the direction of his friend and elder fellow-Wykehamist Arnold Toynbee. M acartney wrote the second volume of the 1925 Survey. In 1926 he joined for two years the editorial staff of the Encyclopaedia Britannica. In 1928 he moved to the League of Nations U nion, and worked in its intelligence departm ent for eight years, partly in Geneva and partly in London, becoming their chief specialist on the national minorities of the new m ulti-lingual states of Central and Eastern Europe, and indeed an expert in this field unsurpassed in any country of Europe or beyond. D uring these years his first books appeared: The Social Revolution in Austria in 1926; The Magyars in the Ninth Century in 1931; Hungary (a general history in the ‘M odern W orld5 series) in 1934; and National States and National Minorities in the same year. M acartney was eager to obtain an academic appointm ent, and applied for the Sir Ernest Cassel C hair of International Relations at the London School of Economics. His referees were G ilbert M urray, Arnold Toynbee, and M onty Rendall, and he also had supporting testimonials from Sir Jam es Headlam -M orley, J . L. Garvin, and R. W. Seton-W atson; but his application was unsuccessful. In Novem ber 1931 Toynbee recom mended him for the Stevenson C hair at LSE, but this too was unsuccessful. T he scholar appointed to this chair was C. K. (later Sir Charles) W ebster. M acartney’s impressive scholarly achievements, however, received recognition with his appointm ent to a Fellowship at All Souls, in 1936. His association with the college lasted for more than forty years, and was undoubtedly a blessing for him, since it enabled him to study and write in a university setting, w ithout being compelled to do regular teaching or academic adm inistration, for which tasks he was probably not tem peram entally suited. H e m ade his home at Boars Hill, within easy reach of All Souls, and could travel to and from London when required. In 1937 he published another m ajor book: Hungary and Her Successors.

XVI

4 X3

During the Second W orld W ar he was a leading m em ber of the enlarged Research D epartm ent of the Foreign Office, directed by Toynbee, located in Oxford, and consisting largely of persons formerly closely associated with C hatham House, together with the small bu t extremely efficient secretarial infra-structure of th at institution. M acartney’s task was to supply expert advice to the Foreign Office in connection with C entral Europe in general, and H ungary in particular. Possibly m ore im portant was his contribution to the H ungarian Section of the BBC Overseas Service. His regular broadcasts, in H ungarian with a strong English accent, were immensely popular in H ungary, where he became known more widely than ever before as M akartni Elem er (the latter H ungarian Christian nam e being pronounced approxim ately as his own Christian nam e of Aylmer). I t was a tribute both to his reputation and to a certain surviving liberalism in w artim e H ungary th at his series of volumes of Studies in the earliest Hungarian historical sources, which had begun to appear in Budapest in 1938, could continue in th at city despite the state of w ar between the two countries, parts 4 and 5 being published in 1942 (but with the false date of 1940 printed on them in order to satisfy censorship!). T he series was completed by parts 6 and 7 in Oxford in 1951. After the w ar M acartney returned to All Souls, and his literary output once more revived. His Medieval Hungarian Historians (I 953) summarized his conclusions on the seven-part series of 1938-5 k In 1957 he published a two-volume work curiously entitled October Fifteenth. T he title refers to the date of A dm iral H orthy’s abortive attem pt to break away from H itler in 1944; b u t the book is a detailed history of internal and foreign policies in H ungary from 1929 to 1944. W hile working on this book, M acartney held the M ontague Burton C hair of International Relations at Edinburgh University, a part-tim e post whose lecturing duties were not arduous, but perm itted him to rem ain in Oxford and to travel northw ards from time to time. A short history, entitled simply Hungary, appeared in 1962. In 1968 M acartney published his last m ajor work, The Habsburg Empire iyg o -ig i8, a detailed narrative of Austria and H ungary in the nineteenth century and up to the collapse of the M onarchy, with special emphasis on the history of institutions. H aving hitherto specialized in the early M iddle Ages and the tw entieth century, M acartney applied himself in his last years to the Early M odern period. The result was two slighter but valuable works, Maria Theresa and the House o f Austria (1969) and a collection of

414

XVII

well-chosen documents with explanatory notes, The Habsburg and Hohenzollem Dynasties in the Seventeenth and Eighteenth Centuries (1970), this latter appearing in his seventy-sixth year. M acartney was an individual scholar, in the tradition of E uropean learning in the age, long since expired, when scholars sought both to learn and to diffuse the results of their learning, and when universities had not become extensions of state bureaucracy. H e was also a participant in international politics, with opinions passionately held and with a sense of duty to exert political influence. This account of his life will therefore have two themes, which cannot be effectively unscram bled from each other: his m ajor historical works and his involvement in the field of international affairs and foreign policy. His first book, The Social Revolution in Austria, was published in 1926. N ot quite 300 pages long, it was divided about half-and-half betw een a sum m ary of A ustria’s recent history and a discussion of the m ain social classes, institutions, legislation, and unsolved problems. It began w ith a short penetrating essay on the H absburg M onarchy, and then described in m uch greater detail the period of defeat and attem pted revolution, and the emergence of a dem ocratic republic. There followed two chapters on the Social Dem ocratic Party, the conflicting groups within its leadership, the institution and operation of works councils in factories, and socialist achievements in housing in V ienna. T hen came a chapter each on the peasants and the Church, on the m iddle classes, and on the relations between Jews and Germans, in the broader economic and cultural spheres as well as in political life. This book showed M acartney to be a shrewd observer as well as a promising historian, com m anding the clear straightforw ard style, with no words wasted and with no needless erudite obscurities, and the ability to put scenes and incidents vividly before the reader’s im agination, which were to m ark most of his published works. T he book was at the same time a personal statem ent, with irony and occasional venom. These were, however, rather im partially distributed. Financial speculators, left socialist dem agogues, anti-socialist R om an Catholic priests, and Galician Jewish im m igrants m ight each in tu rn resent some of his ju dgements. His next book, The Magyars in the Ninth Century, published in 1930, was a study of a little explored corner of early medieval history. Using Byzantine and Latin sources, as well as passages translated for him from Arabic and Persian authors, besides secondary works by G erm an and H ungarian historians, he

XV111

415

endeavoured to distinguish from each other the M agyars and the other peoples of the steppe—Cumans, Kavars, Petchenegs, and m any others—of whom very little was known, and to reach conclusions about the origins of the ninth-century invaders of Central Europe who founded the H ungarian state. In 1934 came Hungary, a volume in Ernest Benn’s M odern W orld series edited by H. A. L. Fisher. It followed approxim ately the p attern of his Austria book, with about two-fifths devoted to history and the rest to analysis of the Constitution, the Church, the m ain social classes and the problems of foreign policy. The differences between aristocratic m agnates and gentry, and the role played by the latter in political leadership and in adm inistration, were clearly explained. So also were the problems of the peasantry, which comprised landless labourers on vast estates, dw arf holders with too little land, and prosperous farmers found chiefly in the west and south-west. T he inadequacy of such land reform legislation as had occurred was also m ade abundantly clear. The position of Jews in business was another im portant theme: H ungarian Jews emerged in a more favourable light than had Austrian Jews in the earlier book. A fault of the book was the very small space devoted to education and to the role of the intellectual, as opposed to the business, section of the middle class, whether Jew ish or M agyar. It is convenient to m ention at this point M acartney’s later works in the history of H ungary, an im portant p a rt of E uropean history and one in which he attained high standards of scholarship, but a field of study of less widespread interest than his publications, from the 1930s on, in international politics. O utstanding was his series o f‘Studies in the earliest H ungarian historical sources’, of which eight successive parts appeared between 1938 and 1952. The firste five were published in Budapest, the last in O x fo rd . In this o b scu re and specialized field th ere was and is certainly no one in Britain, and few it any in H ungary capable of a critical judgem ent of M acartney’s work, which rem ained an object of adm iration mingled with awe. His last book on a H ungarian theme was a short comprehensive history of that country, published in 1962 and bearing the same simple title as his 1934 book, Hungary, with which it should not be confused. His fourth book, National States and National Minorities, of 1934, was a m ajor work of scholarship, the most original, and possibly in the long term the most valuable, of all his writings. It was a result both of very wide historical reading and of experience of the activities of the League of Nations in Geneva and of its most influential champions in England. H e traced the emergence of

416

XIX

the concept of nationality from medieval western Christendom; the intricate pattern of conquests, and coexistence of successive dom inant and subject peoples, in C entral and Eastern Europe; and the concepts of individual nationality within a m ulti-national empire, as they were embodied in the millet system of the O ttom an Em pire, or discussed by political thinkers in A ustria-H ungary. H e then examined the nationalist movements th at grew in the D anube basin and the Balkan peninsula in the nineteenth century and up to 1914; the adoption by the Allies of the aim of national self-determination in the G reat W ar; and the redraw ing of frontiers and creation of new states which followed the war. Almost half the book was devoted to a detailed analysis of the purposes and the operation of the M inority Treaties, placed under the supervision of the League, which nine states were obliged to sign. His conclusions were pessimistic. W here a com m unity which shared a national consciousness was broken by a frontier (and he recognized th at this was sometimes unavoidable), a nationally conscious group was torn away from its com m unity and placed in a state where another nationally conscious com m unity was dom inant. (This situation was different from one in which a whole nationally conscious com m unity was placed under the rule of another, though writers on nationalism often confused, and still confuse the two.) M acartney’s conclusion was th at ‘a national state and national minorities are incom patibles5. As M acartney saw it, there were three possible solutions to the problem . O ne was to change the frontiers so th at they should greatly reduce the num bers left on the ‘wrong sides5. This remedy, which became generally known as Revisionism, had severe lim itations, and was not applicable at all to certain regions. T he second was an exchange of populations. This was in fact adopted in several cases, the most im portant being between Greece and Turkey, where tremendous hardships were caused, but the two governments, recently at w ar with each other, m ade concerted efforts to m itigate them. Both these solutions were discussed at length in M acartney’s book, but he himself preferred a third, which was to replace nationality as the basis of legitimacy of governm ent by some higher principle. Essentially, he believed th at the problems of national minorities could only be solved by the abolition of national states. He quoted with approval Lord A cton’s condem nation of m odern nationalism: ‘A state which is incom petent to satisfy different races condemns itself; a state which labours to neutralise, to absorb or to expel them, destroys its own vitality; a state which does not include them is destitute of

XX

4i 7

the chief basis of self-government.5 Looking at the world of 1934, he could not see examples which offered m uch im m ediate help or hope to victims of national discrim ination in C entral Europe. T he two cases which came nearest to this were the Soviet U nion and the U nited Kingdom . In the first case, while lacking enthusiasm for the Bolshevik form of governm ent, he nevertheless took seriously the claim of the Soviet authorities to have introduced genuinely equal treatm ent of all nationalities. In the second case, as a result of a long history, it had become true of Britain th at ‘the state can fairly be said to be equally the state of all nationalities inhabiting it5. But as M acartney would not have wished to see a re-enactm ent of the Bolshevik Revolution in the D anube valley, and it was useless to advise the D anubians to repeat six hundred years of British history, there was little comfort to be derived from either example. In the next years M acartney concentrated on the H ungarian minorities, and in 1937 published Hungary and Her Successors. This book, like its predecessor exceeding 500 pages, exam ined in detail the status of the H ungarian minorities in lands which had been ceded by the peace treaties to Austria, Czechoslovakia, R om ania, and Yugoslavia. In stating his conclusions he had in m ind the situation at the time of the collapse of the old H ungary and the situation as it was in the mid-1930s. A t the time of writing it was clear to him th at H ungarians both in the residual H ungarian state and in the m inority areas of neighbouring states bitterly resented the situation created by the peace treaties, and th at this resentm ent was by no means confined, as opponents of H ungary especially in Czechoslovakia liked to argue, to the landow ning upper class. It was equally clear to him th at the non-H ungarian nations which had formerly lived under H ungary had no desire to be reunited with H ungary: in his view this had been true of Rom anians and Serbs already in 1918 but only partially of Slovaks, but in 1937 was equally true of all three. He therefore rejected the ‘m axim um 5 revision advocated by nationalists in Budapest—th at is, the restoration of the old frontiers. H e did, however, favour a ‘m inim um 5 revision, which he specified in detail. He believed that it would be not only an act of justice to the H ungarians but a measure beneficial to the peace of Europe if H ungary were to recover from Czechoslovakia the Grosser Schiitt island between the two branches of the D anube east of Bratislava, and some portion of the plain north of the Danube; and from R om ania the plains lying west of the Bihar m ountains. T he second of these changes would restore 400,000 H ungarians to H ungary

418

XXI

while placing only 40,000 Rom anians under H ungarian rule. A part from this he believed th at R uthenia, formerly the n o rth eastern corner of H ungary and now the eastern province of Czechoslovakia, should be restored to H ungary. T he Ruthenes were not Czechs or H ungarians, and would be a m inority in either state. T heir national identity was by no means clear (U krainian, or Russian, or a distinct R uthene people), bu t in M acartney’s view their economic situation would improve if they were included in H ungary. M acartney was thus advocating adoption, within limits, of the first of the three solutions which he had set forth in his earlier book. H e rem ained, however, convinced th at the third solution was still the best. H e reiterated his opinion th at in C entral Europe the ‘national state5 (H ungarian from 1867 to 1918, the three Successor States since 1918) had proved a failure. He still held out as an ideal the state which stands above nationalism, and offers equal respect to all cultural communities existing within it. H e believed th at ultim ately the best candidate for such a state would have been a reconstituted H ungary. His reasons were th at the H ungarian state before 1918 formed, w ithin its natural frontiers, an adm irable geographical and economic unit, and th at there was an earlier H ungarian tradition, long antedating the age of nationalism , which fitted Lord Acton’s criteria. It was summed up in the advice given to his son by St. Stephen, the pagan chief who becam e the first crowned Christian king of H ungary and was later canonized by Rome: Regnum unius linguae uniusque moris imbecille etfragile est. A year after the appearance of the book, the M unich surrender brought about the m utilation of Czechoslovakia, in consequence of which H ungary regained from Slovakia substantially m ore territory than M acartney had proposed, but did not acquire R uthenia. T he latter was given far-reaching autonom y within the rum p republic, and was designed by H itler to be a centre for U krainian nationalist activities directed against the Soviet U nion and Poland. W hile the new C zechoslovak-H ungarian frontier was still being negotiated, M acartney published, on 26 O ctober 1938, in The Times a 2,500-word article, in which he set forth in precise and unem otional language the H ungarian case. Essentially the article was a concentrated sum m ary of the facts and argum ents of his book; but it was rem arkable for almost totally ignoring the international situation which had m ade revision of frontiers possible. Its opening paragraph contained the following words: This is one of the rare occasions on which it may prove possible to get a settlement founded on acknowledged principles ofjustice; to prove that revision need not necessarily strike a blow at peace, but may lay the

XXII

419

foundations for a more enduring order. For in so far as a settlement

affording honourable satisfaction to both parties can now be reached, a new era of cooperation in the Danube valley may open. This extraordinary complacency infuriated those who felt th at the destruction of the old dem ocratic Czechoslovakia was in itself a tragedy, and still more th at H itler’s diplom atic success had created a solid base of G erm an dom ination in C entral Europe from which it would be able to threaten the rest of the continent and the world. Among these was M acartney’s older colleague R. W. Seton-W atson, who in previous years had greatly adm ired M acartney’s work, and encouraged and tried to help his career. A painful exchange of letters took place between them, after which their relations, though outw ardly correct, rem ained irreparably strained. Five m onths later came the complete dism em berm ent of rum p Czechoslovakia, and this time H ungary, w ithout the approval of Hitler, forcibly annexed R uthenia. This won M acartney’s approval, for the reasons stated in his earlier writings, and also because it pu t an end to the U krainian schemes. As H ungary had acted against G erm an wishes in this case, though in accordance with them after M unich, it showed, he believed, th at her rulers placed H ungarian interests first, and had not become vassals of Hitler. At the outbreak of w ar in Septem ber 1939 M acartney joined the Research D epartm ent of the Foreign Office located in Oxford. In Novem ber 1939 the British M inister in H ungary, M r (later Sir) Owen O ’M alley requested th at M acartney should come to H ungary. T he Foreign Office, and Arnold Toynbee as his im m ediate chief, agreed to this, and he spent February and M arch 1940 there. He m et a great num ber of H ungarians, including his old friend C ount Paul Teleki, a distinguished academ ic geographer as well as an elder statesm an, who had become Prim e M inister for the second time in February 1939. M acartney’s findings were summarized in a long m em orandum forw arded by O ’M alley to the Foreign Office on 29 M arch 1940. Teleki and the Regent, Adm iral H orthy, were essentially pro-W estern. Teleki had preserved a considerable measure of political liberty, m aking H ungary an oasis in the D anubian fascist desert. This was recognized by H ungarian Jews and socialists, who knew th at they could only lose by any further changes. However, the ability of H itler to promise more gains of territory, at the expense of R om ania and Yugoslavia, which Britain could not do, m ade it difficult for any H ungarian governm ent to resist G erm an pressures.

420

XX111

T he V ienna Aw ard of 30 Ju ly 1940, dictated by R ibbentrop and Giano to the governments of H ungary and R om ania, divided Transylvania between the two, leaving hundreds of thousands of each nation on the ‘w rong5 side of the line, but representing very large gains for H ungary. I t was discussed in a m em orandum by M acartney for the Foreign Office of 9 Septem ber 1940. His recom m endation for policy was th at Britain should rem ain uncom m itted to either the R om anian or the H ungarian side. ‘Keep the carrot dangling . . . Do nothing irrevocable.5 He outlined various possible future settlements: an independent Principality of T ra n sylvania; a restoration of the T rianon frontier; cession to H ungary only of the western plainland region; either of two earlier official H ungarian proposals which differed from the V ienna Award; a treaty compelling the R om anian governm ent to grant the H ungarians autonom y within a united Transylvania; a transfer of populations; or the retention of the existing partition. T he first, though attractive, he considered irrealizable; the second unjust; population exchange so brutal and so expensive as to be unacceptable; promises of autonom y worthless; and the various H ungarian proposals of the past open to serious objections. Paradoxically, he ended up m oderately in favour of w hat the Axis dictators had imposed: the present line ‘. . . not entirely to be sneered at as an attem pt to secure a not very inequitable ethnographic lin e . . . m ay well be regarded as im perm anent, but so m ust any solution5. In April 1941 G erm an pressure, and the strongly pro-G erm an com m and of the H ungarian arm y, brought H ungary into the w ar against Britain’s ally Yugoslavia, and this was rew arded by annexation of territory from th at country. U nable either to consent to this policy or to prevent it, C ount Teleki shot himself. O n 27 Ju n e H ungary went to w ar with the Soviet U nion. T he rulers of the now m uch enlarged H ungary showed little sign of following the advice of St. Stephen or of Lord Acton. They displayed a national intolerance towards their new non-H ungarian subjects not less than that of the Succession States between the wars. In Ja n u a ry 1942 reprisals against Yugoslav partisan activities culm inated in a massacre in Novi Sad (Ujvidek), the m ain city of the annexed region of Yugoslavia, on the Danube, in which it was estimated th at some 2,250 Serbs and 700 Jews lost their lives. Nevertheless pro-British and dem ocratic forces not only existed, b u t were tolerated by the governm ent to a m uch greater extent than anywhere in Europe outside neutral Sweden and Switzerland. Liberal and socialist ideas could be expressed in print, and

XXIV

421

p art of the press supported the concept of an Independence F ront—which in the circumstances could only m ean united action by opponents of the T hird Reich. Miklos K allay, who became Prime M inister in M arch 1942, was himself bitterly anti-G erm an, protected the dem ocratic elements, and sought for an opportunity to w ithdraw from the w ar against Russia. His emissaries established contact with the British authorities during the sum m er of 1943. T he surrender of Italy aroused hopes th a t the Allies would be able to move into Yugoslavia from Italy* in which case he would have brought H ungary over to their side. However, the inability of the Allies to occupy m ore th an southern Italy, and the fate m eted out by a vengeful Mussolini and his G erm an patrons to those Italians, including his own son-in-law, who had abandoned him, destroyed such hopes. In M arch 1944 G erm an forces occupied H ungary, and forced H orthy to accept a puppet governm ent, recruited from reliably G erm anophile politicians of the upper and middle classes, which provided further support to the w ar effort against Russia and obligingly sent hundreds of thousands of H ungarian Jews to their deaths in the G erm an exterm ination camps. In O ctober 1944 R egent H orthy m ade a last attem pt to break loose from Germ any, ill planned and quickly crushed. H ungary was placed under the jo in t rule of G erm an forces and of a semi-criminal rabble of H ungarian fascists, and became the m ain battleground between the southern forces of the G erm an and Russian armies. At the end of the fighting Budapest was in ruins, the economy collapsed, and the people at the mercy of Russian soldiers encouraged by their com m anders to wreak vengeance for the sufferings of Russia on any one who fell into their hands. D uring these two years M acartney continued to produce learned comments on the H ungarian situation, with diminishing effect. He m ust have been generally informed at the time of K allay’s peace efforts, but took no p a rt in the negotiations and contacts, which were handled through the channels of the Special O perations Executive under the direction of the Foreign Office, in consultation with the U nited States and Soviet governments. As the victory of the Allies approached, M acartney hoped th at Britain would use its influence to secure a settlem ent in C entral Europe that would be ju st to all peoples concerned. In particular, since both R om ania and H ungary had been allies of H itler, he hoped th at the T ransylvanian problem would be considered w ithout prejudice either way. It is arguable th at his influence m ay have contributed to the use, in the terms of armistice given by the

422

XXV

Allies to R om ania in Septem ber 1944, of the qualified form ula th at R om ania would be given ‘Transylvania, or the greater p art ofit, subject to confirm ation at the peace settlem ent’. In practice, it was the Soviet governm ent alone which decided the frontier, and Stalin preferred simply to restore to R om ania w hat she had had before, while keeping for himself the eastern h alf of M oldavia, or Bessarabia. A t the peace conference with the lesser enemy states, held in Paris in the summ er of 1946, neither the British nor the Am erican governm ent raised any objection. M acartney’s six years as a governm ent specialist on C entral European affairs m ay thus be said to have achieved nothing, even if he did provide m uch well-digested inform ation. M ore im portan t were his broadcasts, which were widely listened to, highly popular, and m aintained or increased the sym pathy for Britain felt by m any Hungarians. They were, however, an object of fierce attack by the Czechoslovak exiles. A feature of this quarrel was the hostility between M acartney and the exiled C ount M ichael Karolyi, the ill-fated radical Prime M inister of 1918-19, who was excluded from the BBC broadcasts, essentially on the ground th at his m em ory and his nam e were anathem a to the effective political class in H ungary. K arolyi’s claim to be heard was espoused by the Czechs, and this further strengthened M acartney’s objection to him. T he inability of these two m en to do each other justice, K arolyi a somewhat quixotic aristocrat with a noble record of struggle on behalf of social reform and political liberty, M acartney a devoted friend of H ungary w ith deep understanding of its social and political issues, must be a source of some surprise and great regret to an observer of m odern H ungary. M acartney’s broadcasts, however, also raised a serious question on which there was genuine disagreem ent am ong British officials concerned with H ungary. O ne view was th at by m aintaining indirect com m unication with the rulers and the wider political class of H ungary, by showing understanding for these m en’s difficulties while firmly insisting that the alliance with G erm any would prove, and was already proving, disastrous for H ungary and should be abandoned, the broadcasts were encouraging those who were trying to minimize H ungary’s m ilitary effort in Russia, and who were trying not w ithout success to preserve liberal decencies in H ungarian political life. Thus, it could be claimed, the broadcasts were prom oting Allied interests. T he opposite view was that, adm irably intentioned though they were, the broadcasts were encouraging H orthy, Kallay, and their friends to believe th at they could go on fighting the Soviet Union w ithout incurring

423 any real displeasure from the W estern powers. I t was not denied that M acartney emphasized the solidity of the W estern-Soviet alliance and urged the H ungarians to break with H itler. But the mildness of his tone, their personal knowledge of his affection for H ungary and of the sort of H ungarians who were his friends, and the fact th at the British governm ent used him as its spokesman, caused them to ignore his warnings and to believe th at they could get away with a policy of fighting against one ally and seeking the protection of the others. T he broadcasts were thus, it was argued, more harm ful th an beneficial to Allied interests. In August 1943 M acartney’s broadcasts were stopped. I t seems likely th at this was due rath er to the prevalence of the second over the first of these two British views th an to successful intrigues by Czechs. M acartney, however, was deeply h u rt by the decision, which he rem em bered with bitterness in later years. After the w ar M acartney returned to the calm of All Souls and Boars Hill, and took up again his literary labours. T he last thirty years of his life produced two m ajor works and several others of lesser but substantial value. T he first was October Fifteenth, published in two volumes in 1957 by Edinburgh University Press, at a time when M acartney held his part-tim e chair there. T he title is taken, as already noted above, from the date in 1944 when Regent H orthy m ade his unsuccessful attem pt to break away from Germ any, and was instead arrested and deported by his former allies, being rescued from comfortable incarceration in a G erm an castle by the liberating Am erican arm y. I t is an extremely detailed expose of H ungarian history between 1929 and 1945. I t is based on his own experience, spread over m ore than thirty years, on conversations with leading persons (including visits to both H orthy and K allay in their im m ediate post-war captivity), on massive study of the H ungarian press, on such official documents as had been pu b lished or were available up to the time of writing, and on more interesting docum ents given to him by various H ungarians who had been able to acquire them , as well as on private diaries and on published mem oir literature. T he rath er large portion devoted to foreign policy partly duplicates w hat has been done by others, and may be faulted in detail as greater quantities of official documents are studied by historians. But the narrative of internal politics, and the presentation of the personalities and motives of all the prom inent, and m any less prom inent, figures is of incom parable value. Individuals come to life in these pages as they are portrayed by an author who combines insight into character with XXVI

424

XXVll

rem arkable literary skill. No other writer, H ungarian or foreign, has done this job, or is or ever will be in a position to do it. It is safe to say th a t this book will be a source of knowledge for all who m ay ever wish to study this particular place and period in history. It is also arguable th at any one who gives himself time to peruse atten tively these 1,000 pages (an enjoyable b u t protracted experience) will emerge with a better upderstanding of the tragic predicam ent of small nations and of persons charged with small bits of power and responsibility in m id-tw entieth-century Europe, and indeed of the hum an political condition tout court. T he other m ajor work was a history of A ustria-H ungary from the death of Joseph II to the abdication of Charles, entitled The Habsburg Empire iyg o -ig i8, published in 1968 and running to nearly 900 pages. T he emphasis is on constitutional developments, and these are viewed m ainly from the centre, which means V ienna rather than Budapest, let alone the lesser centres such as Prague or Lemberg. Even H ungary, of which he knew so m uch, recedes into the background, and his interest returns to Austria, where his concern with C entral Europe had begun nearly fifty years earlier. H absburg rulers and their bureaucrats are the leading figures, and M acartney shows in his handling of them the em pathy and the clarity of expression th at m ark so m uch of his work. His grasp extends to the intricacies of the financial bureaucracy, which do not m ake easy reading, and he reveals thorough knowledge of peasant problems and of their interconnection with national conflicts. In the Preface he expresses mild disdain for the ‘tribal histories’ produced in profusion in recent years by members of the lesser ‘nationalities’: one sees w hat he means, but this is to do less than justice to a good deal of valuable scholarship th at has seen the light in neighbouring states, despite difficult intellectual conditions. M acartney’s distaste for nationalism as legitimacy was unabated, but he was no longer concerned to find a new legitimacy: this book was devoted only to the past. T he statem ent of the late Professor Pribram , quoted in this Preface, th at a historian of the M onarchy needed fourteen languages, was exaggerated: h alf th at num ber from within its borders, plus French and very m arginally Russian, would suffice, and of these M acartney had sufficient com m and. T he difficulty of the subject is only partly linguistic. It is rath er th at the essence of this em pire’s history is th at it produced a culture, with its own unity, to which all the arts, literature, physical and hum an landscapes, and a way of fife contributed, but which was more than any of these. L iterature can more easily be treated in words

XXV111

425

than the other aspects. There was such a thing as ‘A ustrian’ literature, if we m ay use th at adjective to cover the whole Kulturraum (no English equivalent). However, this ‘A ustrian’ literature is marvellously elusive. M ost of the best of it was expressed in one language—G erm an—yet m uch th at was essential rem ained em bedded in other languages from which it could not be fully extracted. At the same time the fact th at G erm an was the language of a wider great literature in no way perm eated by an ‘A ustrian’ spirit m ade any clear dem arcation impossible. M acartney in his Preface stated th a t he felt compelled to leave out ‘all Kulturgesckichte proper, as distinct from literary etc. activities which had their im portance for the developm ent of national movements’. This is a perfectly reasonable decision, b u t it does m ean th at the innerm ost essence of the subject is missing. But could any m ortal grasp th at essence? No one yet, certainly in English, and probably in any other language, has better described the anatom y of the m any-lim bed and m any-headed creature than M acartney, and it seems unlikely th at any one will do a better jo b for some time yet. M acartney was elected a Fellow of the British Academ y in 1965. He was also m ade a Corresponding M em ber of the A ustrian, and was aw arded the G rand Decoration of H onour, with gold, of Austria in 1974. H e revisited H ungary several times after the war, the first being in 1945 and the last at a historical colloquium in Budapest in 1971. He was elected a Corresponding M em ber of the H ungarian Academy in 1947, but two years later, at the height of the Stalinist terror under M atyas Rakosi, his m em bership was term inated. Though H ungarian historians working under the communist regime held him in high regard, and he had good personal relations with them, they were not able to obtain a reversal of this insulting and unjust action, doubtless imposed on the Academy by the party leadership; and this was a source of lasting bitterness to him. He also m ade one lecture tour in the U nited States, and was elected a Freem an of the City of Cleveland (Ohio), the centre of the largest population of Americans of H ungarian descent. M ention must be m ade in conclusion of his activities on behalf of H ungarian exiles in Britain. H e was one of a small group of British academics who visited Austria imm ediately after the H ungarian Revolution of O ctober 1956 to see w hat could be done for students from H ungarian universities who had escaped from the Soviet invasion. He took a personal interest in those who came to Britain and were enabled to continue their studies in this

426

XXIX

country. He was also a founder, and for m any years President, of the A nglo-H ungarian Fellowship, which held regular meetings in the hospitality of the Polish H earth in South Kensington, at which lectures were given and problems discussed which interested cultured H ungarians uprooted from their hom eland. In M acartney’s lifetime two world wars started in Central Europe; not only did crises in th at region trigger off two hum an catastrophes, b ut am ong the deeper causes of both disasters were forces a t work in those lands, which were little understood in the north and west of Europe at the beginning of this century, though they became more fam iliar as they spread to Russia, the Muslim lands, and the whole world beyond Europe and N orth America. T he tragedy of tw entieth-century C entral Europe had its social, national, constitutional, and international aspects. All four strikingly foreshadow developments in Asia and Africa in the m idcentury. T he social scene was m arked by the antagonism , sometimes latent and sometimes brutally evident, betw een pre-industrial upper classes, with privileges and m entalities th at had disappeared in most of northern and western Europe, and a peasantry with too little land and prim itive m ethods of agriculture, growing poorer as its num bers increased and its needs were neglected by rulers overwhelmingly interested in cities and industry. T h a t p a rt of the peasantry which became uprooted from the village and swelled the population of the new u rb an agglom erations, added to the confusion and the unrest. In m ulti-lingual and multi-confessional C entral Europe social discontents became inseparable from national. From subcultures based on m inority languages emerged educated elites no longer willing to be treated as members of second-class communities, ruled by persons who considered themselves to belong to the only, official, nation in the state. M ere cultural diversity was slowly transform ed into national consciousness; leaders appeared who claimed th at their peoples too were nations, and as such dem anded their own sovereign states; and more and more members of these peoples followed their lead. Increasingly, the old political legitimacy was underm ined, and the attem pts to replace it had small success. T he political challengers of the old order spoke of democracy, b u t w hether th at was a suitable word to describe w hat they got, is rather doubtful. Dem ocratic constitutions were duly enacted, b ut old methods of governm ent rem ained. M ajority rule was exploited by dem agogues, or obstructed by officials, or both processes occurred

XXX

427

together. Universal suffrage, distorted in this way, could not so much liberate the peoples as offer the masses to the demagogues as a bludgeon with which to beat their rivals; and after years of confusion new tyrants emerged. Each new state was nom inally sovereign, but each was constrained not only by unsettled frontier disputes with small neighbours but also by the rivalries of neighbouring G reat Powers. T he relationship between small and large states was not simple. T he big bullied the small when they had the chance, but the small used the big for their own ends too. M acartney was well aw are of all four sets of problems. His first book gave precedence to the social factor. T he policies and leadership of the Austrian social dem ocrats gave him grounds for cautious optim ism —which the tragic events of the mid-1930s proved wrong. H e also well understood the yawning social gulf in H ungary between great landowners and landless workers or dw arf holders. He was never—as his critics at times portrayed him —indifferent to the need for a land reform; but he did not see why the fact th at R om ania and Czechoslovakia had carried out land reforms should in itself justify their retention of lands inhabited by H ungarians. As the years passed, he gave greater attention to the problems of nationalism, which in fact assumed first place in the politics of the 1930s. He continued to regard nationalism as a bad basis of legitimacy for a state. H e would always have preferred to see forms of governm ent based on some higher principle, under which citizens would have had freedom to develop their own national cultures. But no such state appeared in Central Europe after the collapse of the H absburg M onarchy, and the H absburg M onarchy had not been such a state. H e preferred democracy, at least in the sense in which th a t word was habitually used in the West in his lifetime, to dictatorial forms of governm ent, but he was sceptical of its prospects in Central Europe. H ungary’s governm ent in the 1920s was a kind of Whiggish constitutionalism, with a parliam ent com parable to the British before 1832, and with a large measure of civil liberties, at least for city dwellers. Yugoslavia’s attem pts at dem ocracy broke down by 1929. The agony of R om anian dem ocracy was more protracted, but it was never a very healthy organism. Czechoslovakia came m uch nearer to a W estern dem ocratic model, and M acartney gave it credit for this, yet rem ained sceptical of a state in which the dom inant nation formed less than h alf the population. As for H ungary, in the 1930s the old oligarchic landow ning

428

XXXI

elite lost a great deal of its power, and some dem ocratization certainly occurred, in the sense of draw ing m uch wider social forces into political life, but this did not make for m ore of the W estern type of democracy. M acartney’s scepticism was increased, but it is a fair criticism of him th at he modified it in the case of H ungary by giving it the benefit of more doubts than he was willing to concede elsewhere. Preoccupied as he became with the need to repair injustices done to national minorities by the Versailles peace settlem ent, M acartney saw the growing rift between the W estern and C entral G reat Powers in Europe as a struggle between Have-Nots and Haves. H e saw th at H ungarian revisionism inevitably received more support from Mussolini and H itler than from French or British leaders; but the fact th at H ungary drifted towards a status of satellite of the Axis did not diminish his sym pathy for the H ungarian cause, either before or after a state of w ar existed between Britain and Germany. H e would not adm it that, since the T hird Reich represented a threat to all civilized values, any person or governm ent which followed its lead was thereby betraying hum anity. M any years later he m ade to the present writer, in the most friendly m anner, the comm ent th at R. W. Seton-W atson’s mistake had been to think th at if the national discontents which beset the H absburg M onarchy could be removed by national self-determ ination C entral Europe would have a happy future; and th at Seton-W atson’s son had m ade an essentially analogous mistake in thinking that if the social oppressions and discontents of the same region in the 1 9 3 0 s could be removed by socialism, there could be a happy future. H e was right in both cases, but he could not him self produce a better cure. T he H ungary whose wrongs he tried to right proved as fiercely nationalist in its aims and its practice as the Successor States had been to their H ungarian subjects. T he peace setdem ent of 1 9 1 8 - 2 0 , inspired by W ilsonian principles of selfdeterm ination, was followed not by a com m unity of free peoples but by a Kleinstaaterei of m utually hostile neighbours. T he revolution of 1 9 1 7 - 2 1 in Russia, m ade in the nam e of socialism, created not a brotherhood of workers’ republics, but a totalitarian empire that crushed all national cultures, including the Russian, even if it used Russians to th at end; and its extension to the D anubian lands artificially froze some national antagonisms and exacerbated others. R estoration of the old super-national or pre-national legitimacies of m onarchy or of the universal church m ight be greatly preferable, but their chances seemed negligible.

XXX11

429

T he tw entieth century appeared stuck with the facts of nationalism, in Europe and further afield. T he disintegration of the overseas colonial empires reinforced this conclusion. Indeed their fate placed the fate of A ustria-H ungary and its Succession States in a different perspective: how far M acartney was aw are of this is not clear from his writings. As for the rem aining land-based colonial empire, from the Elbe to K am chatka, no doubt in time it would be one with Nineveh and Tyre, bu t the time-scale was not predictable. No way was visible in M acartney’s lifetime of preventing the national consciousness which arises from economic and cultural developm ent from turning into political nationalism . Economic prosperity and social transform ations could, and sometimes did, make nationalism milder, but they did not seem able to cure the illness, or to prevent recurrent outbreaks. T he Kleinstaaterei of the D anube basin between the world wars was repeated, with strikingly similar ugly features, by the Kleinstaaterei in other continents which followed the Second W orld W ar. Yet if a proliferation of small sovereignties was dangerous, attem pts to suppress them , and attem pts within m ulti-national states to suppress national cultures or to deny national consciousness, proved no less dangerous. The lessons of A ustria-H ungary and of the Successor States were clear, but they were not being learnt. T here was one im portant problem of these decades in C entral Europe, of which M acartney was well aware, which deserves a few more words—the Jewish problem. Jews and M agyars in the nineteenth and tw entieth centuries were interwoven in a web of love-hate relationships which cannot easily be unravelled. W hen in the first decades of the nineteenth century Jews began to pour into H ungary from the formerly Polish territories which had been annexed by Russia and Austria, they encountered the same grass-roots hostility which had characterized all pre-m odern Christian societies. However, the dom inant land-owning classes soon found th at Jews could be useful to them in the task which they set themselves of transform ing H ungary into a strong m odern national state. Jews revealed talents for industrial entrepreneurship, and for the m odern intellectual professions, which M agyar gentry-folk lacked, and did not particularly wish to acquire. As these skills were clearly useful to a m odern state, the H ungarian political elite encouraged Jews to develop them, and m ade their task easier for them. This was especially true after 1 8 6 7 , when H ungarian governments acquired, under the H absburg Crown, almost complete internal

430

XXX111

sovereignty. For their p a rt Jews eagerly em braced the M agyar language, became M agyar patriots—even to the extent of espousing and prom oting the official policy of pressing Slovaks, Rom anians, and others to turn themselves into M agyars—and increasingly inter-m arried with M agyars and sought assimilation. T he flourishing new urban economy and literary-scientific culture of H ungary at the turn of the century was thus largely the product of an unusually fruitful synthesis, within the growing educated elite, of M agyar and Jew . M eanwhile at the lower levels of the H ungarian social pyram id latent hostility to Jews rem ained. In politics H ungarian Jews tended to be liberals or socialists. This did not worry the M agyar rulers. M ost of them subscribed to some sort of liberalism, and they were not m uch worried by socialism am ong industrial workers. T he two forces in which they did see potential danger were nationalism am ong the nonM agyars and agrarian revolt among dw arf-holder peasants and agricultural labourers. H ungarian Jews had virtually no contact with, or sym pathy for, either of these forces. Things changed dram atically in 1 9 1 9 . Socialism, especially its left wing, communism, briefly trium phed when defeat in war overthrew the old regime in H ungary, and m any of the socialist and communist leaders were Jews. O verthrow of the comm unist regime brought a wave of anti-semitism in H ungary. A lthough the re-establishment of an orderly oligarchical form of governm ent, with elements of liberalism, under C ount Stephen Bethlen in the 1 9 2 0 s, brought improvements, the old honeymoon period between M agyars and Jews was past. In the 1 9 3 0 s new motives for anti-semitism arose with the world economic depression: aspirants for jobs in business and the professions, and persons throw n out of such jobs, saw the Jews, who were num erous in those fields, as rivals or exploiters. As the hold of the old oligarchy relaxed, and middle-class or even plebeian elements were draw n into political life, pressure for restrictions on Jews grew. It reached its climax when the Sztojay governm ent of M arch 1 9 4 4 —a governm ent, it must be noted, not of wild fascist fanatics but of highly respectable middle-class politicians—sent hundreds of thousands of Jews to H itler’s exterm ination camps. The destruction of most H ungarian Jews (those of Budapest were spared deportation), was not the work of the old reactionary oligarchy, but was a result of the relative dem ocratization of the 1 9 3 0 s. These processes M acartney closely observed and understood. He knew th at the social and cultural forces involved were too

XXXIV

431

complex for simple denunciation. H e avoided the usual pejorative adjectives of both left and right, but he sometimes spoke and wrote bitingly about H ungarian Jews, as well as about H ungarian Christians, and his expressions in m atters in which he felt emotionally involved were not always judicious; but though perhaps it is true that on the whole he disliked Jews, it is not fair to call him, as he sometimes was called, an anti-Semite, and absurd to call him a fascist. This was recognized by the Israeli organizers of a conference held in H aifa in April 1 9 7 2 , on the tragically sensitive subject of relations between Jews and their host peoples in Eastern Europe in m odern times, who invited M acartney to read a paper on H ungary. Both in his formal presentation and in informal discussions his sincerity and straightforwardness impressed conference participants. M acartney had a rather retiring nature, was at times truculent, and could bear grudges for a long time. His career did not follow the norm al academic cursns honorum, bu t intellectual excellence was something which he adm ired in others and which he himself attained. N either the bland pieties of the academic world, to which the words ‘liberal values’ were conventionally, though often misleadingly, attached, nor the bureaucratic mechanisms of the institutions which are still called universities because no more suitable nam e has yet been found for them, held m uch attraction for him. Yet he was in a broad sense a teacher, and exercised influence not only through his books, and the m any lectures which he gave both to academ ic and to wider audiences, b u t also through the help and advice which he generously provided for individuals who sought it. His basic opinions, in so far as one can judge in so reticent a person, were far from the public ethos of the 1 9 7 0 s. An im portant p a rt of his code was loyalty to friends, and this he unhesitatingly extended to those, in H ungary or elsewhere, who fell victim to political catastrophe, including those who m ight be said to have brought their fate upon themselves. O ne example is an old friend from M acartney’s Vienna days in the 1 9 2 0 s, General H indy, who in Ja n u ary -F e b ru a ry 1 9 4 5 was the jo int com m ander, with the Germans, of the garrison of Budapest when besieged by the Soviet Army. By holding out to the bitter end H indy could be said to have caused the destruction of most of the capital and the loss of m any thousands not only of Russian m ilitary but of H ungarian civilian lives. Taken prisoner by the Russians, he was tried as a w ar criminal and executed. M acartney dedicated his book October Fifteenth ‘to the honoured m em ory of Ivan Hindy, General,

432

XXXV

| August 2 6 th 19 4 6 *. T he rights and wrongs of that particular case must be left to future historians, or to a higher authority still, but M acartney’s gesture shows the man. Victrix causa deis placuit, sed victa Catoni. Popularity of person or opinions, intellectual fashions, and the polished image did not stand high on his scale of priorities, but he valued honest praise and resented its absence. These aspects of his character no doubt have various origins, all unknow n to the present writer; b u t it is perhaps not fanciful to attribute them in p art to th at peculiar com bination of hum ility and arrogance, of respect for the tru th and for those who seek it with contem pt for all kinds of self-advertisement, which formed the W ykeham ical ethos of his day. H e will often, as a schoolboy, have heard the words: ‘grant th at we, whose lot is cast in so goodly a heritage, m ay strive together the m ore abundantly to share with others w hat we so richly enjoy; and as we have entered into the labours of other men, so to labour th at other men m ay enter into ours’. W hatever m ay have been his inw ard vision, his outw ard perform ance—fallible judgem ents and perverse actions included—does not greatly diverge from th at prayer. He laboured mightily, and left behind him works which others will use, and using, rem em ber him. G. H. N. S e t

o n -W a t s o n

A SELECT LIST OF C. A. MACARTNEY’S WORKS 1915

Poems, The Little Books o f Georgian Verse, etc., no. 2, selected by S.G. Ford (Erskine Macdonald: London, 1915), 40 pages. 1926

The Social Revolution in Austria, includes bibliography and index (Cambridge University Press, 1926), xi + 288 pages 1928

‘Austria since 1918’, The Slavonic (and East European) Review 1 (1928), pp. 288-303. Survey of International Affairs [for] 1925 2, by C.A. Macartney et al. (Oxford University Press, 1928), x + 486 pages. 1929

‘The Armed Formations in Austria’, Journal of the Royal Institute of International Affairs 8 (1929), pp. 617-632. ‘Hungary since 1918’, The Slavonic (and East European) Review 7 (1929), pp. 577-594. Minorities. The Story of Racial, Religious and Linguistic Minorities in Europe (League of Nations Union: London, 1929), 39 pages. ‘The Petchenegs’, Slavonic (and East European) Review 8 (1929), pp. 342-355. 1930

‘The Attack on “Valandar”’, Byzantinisch-Neugriechische Jahrbiicher 8 (1929/1930), pp. 159-170. ‘On the Black Bulgars’, Byzantinisch-Neugriechische Jahrbiicher 8 (1929/ 1930), pp. 150-158. The Magyars in the Ninth Century, includes map (Cambridge University Press, 1930), 241 pages. Minorities. The Story of Racial, Religious and Linguistic Minorities in Europe, Foreword by Lord Dickinson (League of the Nations Union: London, 1930), second edition, 39 pages.

xxxviii

A SELECT LIST OF C.A. MACARTNEY’S WORKS

‘A Select List of References on Minorities’, Journal of the Royal Institute of International Affairs 9 (1930), pp. 819-825. 1931

‘Minorities: a problem of Eastern Europe’, Foreign Affairs 9 (1931), pp. 674-682. Refugees. The Work of the League (League of Nations Union: London, 1931), 127 pages. 1932

World Labour Problems, 1932(-I938). A Record of International Labour Organisations 1 parts, parts 1-5 by C. A. Macartney (League of Nations Union: London, 1932-38). 1933

What the League has Done; 1920-1932, in collaboration with Maurice Fanshawe (League of Nations Union: London, 1933), seventh edition, 110 pages. 1934

‘The End of the Huns’, Byzantinisch-Neugriechische Jahrbiicher 10 (1934), pp. 106-114. Hours of Work and Employment (League of Nations Union: London, 1934), 94 pages. Hungary, with a foreword by H.A.L. Fisher, includes map (E. Benn: London, 1934), part of the series The Modem World, 376 pages. National States and National Minorities (Royal Institute of International Affairs: London, 1934), ix + 553 pages. 1935

Social and Economic Planning, an account of a Conference held by the Industrial Advisory Committee of the League of Nations Union, February 19-21, London, 1935 (League of Nations Union: London, 1935), 95 pages. 1936

What the League has done. 1920-1936, in collaboration with Maurice Fanshawe (League of Nations Union: London, 1936), ninth edition, 93 pages. World Planning. The I.L.O. and the New Economic Order, an Account of a Conference held by the Industrial Committee of the League of Nations Union, February 18-20, London, 1936 (League of Nations Union: London, 1936), 92 pages. Magyarorszag, Hungarian version of Hungary, 1934, translated by Sandor

A SELECT LIST OF C.A. MACARTNEY’S WORKS

xxxix

Fest and Miklos Kallay (Revai: Budapest, n.d. [1936]), part of Vildgkdnyvtar, 255 pages. 1937

Hungary and Her Successors. The Treaty of Trianon and its Consequences, 1919-1937, includes maps (Oxford University Press, 1937), ix + 504 pages. 1938

‘The First Historians of Hungary’, The Hungarian Quarterly (1938, part II), pp. 630-641. ‘Grievances of Minorities: the case of Hungary’, a letter, The Times (October 7, 1938). ‘Hungary and Czechoslovakia’, Contemporary Review 154 (1938), pp. 677-683. ‘Hungary and the Present Crisis’, International Affairs 17 (1938), pp. 749-768. ‘Magyar and Slovak. The rival claims: an opportunity for just revision’, The Times (October 26, 1938). Studies of the Earliest Hungarian Historical Sources part I: The Lives of St Gerard; part II: The Composition of the Zagrab and Varad Chronicles and their relationship to the longer narrative Chronicles, Etudes sur TEurope Centre-Orientale - Ostmitteleuropaische Bibliothek 18, ed. by E. Lukinich, 1938, pp. 456-507, (Sarkany: Budapest, 1938), 52 pages. 1939

The Danubian Basin, includes maps, Oxford Pamphlets on World-Affairs 10 (1939), 32 pages. 1940

1m cuenca del Danubio (Oxford University Press, 1940).

The Danube Basin, reprinted with corrections, Oxford Pamphlets on WorldAffairs 10 (1940), 32 pages. ‘Pascua Romanorum’, Szazadok 74 (1940), pp. 1-11. Os problemas da bacia Danubiana (Marinha: Lisbon, 1940). Studies on the Early Hungarian Historical Sources III, part I: The relations between the narrative Chronicles and other historical texts; part II: The Attila Saga, the Hun Chronicle, and T., Etudes sur VEurope CentreOrientale - Ostmitteleuropaische Bibliothek 21, ed. by E. Lukinich (Sarkany: Budapest, 1940), 228 pages. 1941

The Danube Basin, new edition, includes maps, reprinted Oxford Pamphlets on World-Affairs 10 (1941), 31 pages.

xl

A SELECT LIST OF C.A. MACARTNEY’S WORKS

A. W.A. Leeper: A History of Medieval Austria, ed. by R.W. Seton-Watson and C.A. Macartney (Oxford University Press, 1941), vi + 420 pages. 1942

Problems of the Danube Basin, includes map, Current Problems 12, ed. by E. Barker, (Cambridge University Press, 1942), 160 pages. Studies on the Early Hungarian Historical Sources III—IV: [recte part IV]: The Hungarian Texts Relating to the life of St Stephen; IV [recte part V]: The interpolations of the Chronicon Posoniense and the Genealogy of Almus on [!] the Chronicon Budense, Etudes sur VEurope CentreOrientale - Ostmitteleuropaische Bibliothek 21a, ed. by E. Lukinich (Sarkany: Budapest, 1940), 32 pages. N.B.: The correct numbering and date of publication are supplied in Studies, etc., vi-vii, 1951. 1943

A Dunamedence problemai, Hungarian version of Problems of the Danube Basin, 1942 (Keresztes: Budapest, 1943), 140 pages. 1944

‘On the Greek Sources for the History of the Turks in the Sixth Century’, Bulletin of the School of Oriental and African Studies 9 (1944), pp. 266-275. Problems of the Danube Basin, second impression, includes map, Current Problems 12, ed. by E. Barker (1944), 160 pages. 1946

‘Dlugosz et le Chronicon Budense’, Revue d'histoire comparee (1946), pp. 301-316. 1951

Studies on the Earliest Hungarian Historical Sources VI: Unrecognised components of the Chronicon Budense; VII: The Origin, Structure and Meaning of the Hun Chronicle, (B. Blackwell: Oxford, 1951), 184 pages. 1953

Chapter 7 ‘Hungary’ , in The European Nobility in the Eighteenth Century. Studies of the Nobilities of the Major States in the pre-Reform Era, ed. A. Goodwin, includes bibliography (A. and C. Black: London, 1953), pp. 118-135. The Medieval Hungarian Historians: A Critical and Analytical Guide (Cambridge University Press, 1953), xv + 190 pages.

A SELECT LIST OF C.A. MACARTNEY’S WORKS

xli

1954

Foreword to Miklos Kallay: Hungarian Premier (Columbia University Press, 1954), xxxvii + 518 pages. 1957

Contributions to the Encyclopaedia Britannica 24 vols (Chicago, etc., 1957). ‘Eastern Europe’, The New Cambridge Modem History 1, ed. by G. R. Potter (Cambridge University Press, 1957), pp. 368-394. ‘The Habsburg Dominions’, The New Cambridge Modem History 7, ed. by J. D. Lindsay (Cambridge University Press, 1957), pp. 391-415. October Fifteenth. A History of Modem Hungary, 1929-1945, includes maps, Edinburgh University Publications. History, Philosophy and Economics 6 (Edinburgh University Press, 1957), part 1: xvi + 493 pages; part 2: 519 pages. N.B.: early issues of vol. 1 published in 1956. 1959

‘Francis Rdkoczi and the Hungarian Rising of 1707-1711’, History Today 9 (1959), pp. 565-571. 1960

‘The Austrian Empire and Its Problems, 1848-1867’, The New Cambridge Modem History 10, ed. by J.P.T. Bury (Cambridge University Press, 1960), pp. 522-551. ‘Der Vertrag von Trianon: Werden, Wesen und Folgen’, Der Donauraum 2 (1960), pp. 14-19. 1961

‘Hungary’s Declaration of War on the U.S.S.R. in 1941’, Studies in Diplomatic History and Historiography in Honour of G.P. Gooch, ed. by A.A.Sarkissian (Longman: London, 1961), pp. 153-165. October Fifteenth. A History of Modern Hungary, 1929-1945, with a preface and notes on sources to the second edition (Edinburgh University Press, 1961), part 1: xix + 508 pages; part 2: 535 pages. 1962

Hungary. A Short History, includes plates and maps, Edinburgh University Publications. History, Politics and Economics 13 (Edinburgh University Press, 1962), ix + 262 pages. Independent Eastern Europe, in collaboration with A.W. Palmer, includes maps (Macmillan: London, 1962), viii + 499 pages. 1963

‘Britain and Eastern Europe’, Studies for a New Central Europe 1 (1963), pp. 2-4.

xlii

A SELECT LIST OF C.A. MACARTNEY’S WORKS

1964

‘The Hungarian National Chronicle’, Medievalia et Humanistica, XVI. (1964), pp. 3-10. 1965

‘The Austrian Monarchy, 1792-1847’, The New Cambridge Modem History 9, ed. by C.W. Crawley (Cambridge University Press, 1965), pp. 395-411. Hungary and Her Successors, reprinted from the 1937 edition (Oxford University Press, 1965), 504 pages. 1966

Contributions to the Chambers's Encyclopaedia, new, revised edition (Pergamon Press: Oxford, 1966). Independent Eastern Europe, in collaboration with A.W. Palmer, includes maps (Macmillan: London, 1966), vii + 499 pages. ‘Ungarns Weg aus dem Zweiten Weltkrieg’, Vierteljahrshefte fur Zeitgeschichte 14 (1966), pp. 79-103. 1967

‘Das ungarische Nationalitatengesetz vom Jahre 1868’, Der osterreichischungarische Ausgleich von 1867. Vorgeschichte und Wirkungen, ed. by P. Berger (Verlag Herold: Vienna, 1967), pp. 219-230. 1968

The Habsburg Empire, 1790-1918, includes maps (Weidenfeld and Nicholson: London, 1968), xiv + 8 8 6 pages. Hungary and Her Successors, reprinted from the 1937 edition (Oxford University Press, 1968), xxi + 504 pages. The Magyars in the Ninth Century, reprinted with author’s note for Cambridge University Press Library Edition reprint, includes map (Cambridge University Press, 1968), 241 pages. National States and National Minorities, reprinted from the 1937 edition (Russell and Russell: New York, 1968). 1969

‘The Eastern Auxiliaries of the Magyars’, Journal of the Royal Asiatic Society (1969), pp. 49-58. Maria Theresa and the House o f Austria (English Universities Press: London 1969), vii + 168 pages.

A SELECT LIST OF C.A. MACARTNEY’S WORKS

xliii

1970

‘Britain and Eastern Europe’, Toward a New Central Europe, ed. by Francis S. Wagner, Problems Behind the Iron Curtain Series 7 (Danubian Press: Florida, 1970), pp. 52-58. ‘The Compromise of 1867’, Studies in Diplomatic History, ed. by R. Hatton and M.S. Anderson (Longman: London, 1970), pp. 287-300. The Habsburg and Hohenzollem Dynasties in the Seventeenth and Eighteenth Centuries, ed. by C. A. Macartney, Documentary History of Western Civilisation, ed. by E.C. Black and L.W. Levy (Macmillan: London, 1970), vii + 379 pages. 1971

Geschichte Ungarns, (Kohlhammer: Stuttgart, 1971). The Habsburg Empire, 1790-1918, reprinted with corrections (Weidenfeld and Nicholson: London, 1971), xviii + 908 pages. 1974

‘Hungarian Foreign Policy During the Inter-War Period with special reference to the Jewish Question’, Jews and non-Jews in Eastern Europe, 1918-1945, ed. by B. Vago and G.L. Mosse (Wiley: New York, 1974), pp. 125-136. 1977

‘1848 in the Habsburg Monarchy’, European Studies Review 1 (1977), pp. 285-309 1978

The House of Austria. The Later Phase, 1790-1918, includes maps (Edinburgh University Press, 1978), 306 pages. A condensed version of The Habsburg Empire (1968). 1993

Teleki Pal miniszterelnoksege, 1939-1941, Hungarian version of October Fifteenth (1957), vol. 1 chapters xiii-xix, translated by Geza Cserenyey, with an introduction and postscript by Istvan Csicsery-Ronay (Occidental Press: Budapest, 1993), pp. xii + 290.

I

The End o f the Huns

The dreadful fall of the Avars, after the pride of their prime*, was proverbial among the peoples of a n tiq u ity ; but it is h a rd ly less astonishing than the end of the Hunnish Em pire. F or a m om ent Attila seemed to hold the whole civilised world at his mercy; then he died, and his empire crum bled so utterly th at the very place where his sons ruled after him Is uncertain. Only a single com paratively consecutive account of the events in question has come down to us, in the shape of Jo m a n d e s confused work, and this lends itself to m any interpretations. Study of this work, and of the relev an t passages of the other contem porary authors, has convinced me th a t the m eaning usually, if not universally attributed to one p articular p hrase of Jornandes has led to a misconception of the entire situation after Attila’s death which obscures the history at that time, not only of the Huns, b ut also of the Ostrogoths. Jo rn a n d e s’ story m ay be sum m arised as follows: c 50. On A ttila’s death, his four sons attem pt to divide his em pire between them *. The subject nations rebel, and a g re a t all against all battle takes place on the Nedao, in Pannonia, ending with a victory for the Gepids. A ttila’s eldest son, Ellac, is killed, and his brothers flee «juxta litus Pontici m aris, ubi prius Gothos sedisse descripsim us. C esserunt itaque Hunni, quibus cedere p u tab atu r universitas». H ereupon the Gepids, «Hunnorum sibi sedes viribus vindicantes, totius Daciae fines velut victores potiti», conclude peace with Rome. .

n On the Greek Sources for the History o f the Turks in the Sixth Century

I

N the following pages an attem pt will be made to resolve certain obscurities in the references made by Byzantine writers to the Turki nations of Central and Western Asia in th at period—the sixth century—when the two civilizations first came into contact. I t is hoped thereby to throw additional light on the situation of the various Turki tribes at th at date. The central question is th a t of the identification of the Avars ('A fidptis) of the Byzantines with the Juan-Juan or Jwen-Jwen of the Chinese—an identification commonly accepted by scholars since the days of Chavannes 1 and admitted by most modern authorities, including M arquart 2 and Nemeth 3 ; but questioned by Parker on geographical grounds, and by other authorities by reason of a small but important and quite definite chronological discrepancy. The Juan-Juan were the most im portant race in Central Asia from about a . d . 300 or 315 until the middle of the sixth century. Their centre appears to have been somewhere near Karakorum . 4 After a . d . 546, according to Chinese sources, the nation known as the Turks, who had formerly been subjects of the Juan-Juan, revolted against their masters, utterly defeating their Khagan, Anakwe. His successor carried on the struggle for a few years, but had to take refuge with the Chinese, and the Turks were strong enough to insist on the massacre of the survivors of the nation in a . d . 555.5 The Greek references to the Avars begin with a passage from Priscus, quoted in excerpt by Suidas (s.v. " A f t a p i s ) and by Constantine Porphyrogenetos (Exc. de Leg. Gent., ed. de Boor I. 586),6 according to which the Saraguri, Urogi, and Onoguri were ejected from their homes between a . d . 461 and 465 by an attack from the Savirs, who were fleeing before the Avars, who had themselves been driven from their homes by nations inhabiting the shores of the Ocean, and emigrated on account of excessive mists from the Ocean, and a great plague of gryphons. The first-named nations sent embassies to the Empire, and we find them afterwards on the Volga and north of the Black Sea ; the Savirs, north of the Caucasus. Then silence falls on the Avars for 100 years ; after which we come to the well-known series of events beginning in a . d . 558. For these the sources are fairly numerous. 1 H is to ire generate dee H u n s , ii, p . 334. 2 E r a n S a h r, p . 63. 2 A H o n fo g la ld M a g y a r sag K ia la h u ld s a , p p . 100 flf. 4 P a r k e r , A T h o u s a n d Y e a r s o f the T a r ta r s , 2 n d e d ., p . 118. 4 I b id ., p . 121. N e m e th , o p . c it., p . 101, p u ts t h e d a t e a t a .d . 5 5 2 . • G . M o ra v c s ik , Z u r Oeschichte der O noguren, U ngarische Ja h rb ilch er, B d . x , H e f t 1 /2 , p p . 5 5 fif., p ro v e s c o n c lu s iv e ly t h a t t h e e x t r a c t fro m C o n s ta n tin e is s im p ly a n a b b r e v i a ti o n o f t h a t fro m S u id a s .

11 O N T H E G R E E K S O U R C E S F O R T H E H IS T O R Y O F T H E T U R K S

267

(1 ) Menander tells ns th at a nation bearing the name of Avars arrived at the territory of the Alans about a . d . 558 and sent envoys to Constantinople. Justinian sent them to make war against the Hunnish tribes north of the Caucasus. Thereafter they attack the Slavonic Antae also ; and then send a second embassy to ask for land within the frontiers of the Empire .1 (2) Thereafter, according to Theophanes, “ the Hermichiones, the nation living this side of the barbarians next to the Ocean ” sent to Constantinople asking Justinian not to give the Avars land . 2 (3) Theophanes Byzantinus, quoted by Photius, describes this nation as “ the Turks living east of the Tanais, formerly called Massagetse, whom the Persians in their own tongue call Kermichiones ” . 3 The name of their king was 'A gk ^Xt or 9AaKrjXros*. Theophanes dates the embassy at July, a . d . 563. (4) Corippus refers to the same embassy, calling the king Scultor. 4 (5) Next comes another fragment from Menander, as follows : “ Silzibul, the leader of the Turks, having heard of the flight of the Avars, and how they had gone off after ravaging the territory of the Turks, said in his boastful barbarian way th a t ‘ they are not birds to escape the sword of the Turks by flying through the air, nor fish to evade us by taking refuge in the depths of the sea, but they roam about the surface of the earth ; and when I have finished my war with the Epthalites, I shall attack the Avars and they shall not escape my assault \ With these boastful words Silzibul is said to have started on his campaign against the Epthalites . ” 5 (6) In 568 the Turks sent an embassy (their first) to Constantinople. By this time the Epthalites had been subdued. The Greeks also asked : “ tell us how many of the Avars escaped from the dominion of the Turks (rrjg t c j v TovpKcov d(/>rjvLaa€v cm/cparetas*) and if any are still with you (nap9 vpiiv). The answer was : some are still our subjects, and those who ran away somewhere number, I suppose, about 2 0 ,0 0 0 .® (7 ) The embassy proved a success, and a return embassy was sent to Central Asia in the same year. The Turks are here described as “ the Turks formerly called Sakai ” .7 On this occasion no reference was made to “ Avars ” . (8 ) In Tiberius’ second year, however, yet another embassy was sent. This time the Roman envoy, Valentinus, came in for a bad time from his host, the Turkish Khagan Turxanthus, who accused the Greeks of allying themselves with “ his slaves the Varchonites (meaning the Avars) ” (rols 8 c 817 OvapxviTais t o l s rjpcrdpoLs SovXois — ihrjXov 8 c t o v s *Afiapovs) who were 1 * 3 4 6 • 7

E x c . de L eg. Q ent. f r . 1-8 (d e B o o r, p p . 442 ff.). T h e o p h . C hro n ., e d . d e B o o r, p . 239. P h o t. B ib . c o d . 6 4 K«piuxt'a>v. I n J u s t . y iii, 390 ff. M e n a n d e r, E x c . de S e n t., p p . 3 6 4 -5 . E x c . de L eg. O ent.t fr. 7, p . 45 2 . E x c . de L eg. R o m ., fr. 7, p . 192 : ra>v T o v p k w v ra>v Z

qkwv

KaXovfxIvwv

to

n a X a i,

12 268 Turkish subjects, and whom he would come and destroy as soon as he wished.1 (9) Finally, Theophylactus Simocatta reports 2 a letter purporting to have been written to Mauricius in a . d . 598 by the Turkish Khagan “ boasting of his victories According to this the Khagan first conquered the “ Abdelites ”, then allied himself with “ Stembischada ” , and then subdued the Avars, whose survivors fled to the Taugas (the Chinese) and the Mukrit (Mu-ko, a Tonguz race north ofiChina ? ) 3 ; then he attacked the Ogor. There follows a long digression. The Ogor, it is explained, were a powerful nation living to the east, near the River Til which the Turks call “ black ” . Their oldest princes, from whom some of their tribes took their names, were called “ Var ” and “ Hunni ” ( O v a p /c a t X o v v v l ). In Justinian’s day some of these tribes fled to Europe and took the name of Avar ; these were the European Avars, who should more properly be called pseudo-Avars. To this day some of them bear the names of Var and Hunni. Then the ELhagan subdued the Colch and afterwards proved victorious in a civil war, etc. The above (first) description of the Avars’ end tallies so well with th at given by the Chinese sources of the end of the Juan-Juan th at it is hard indeed to resist the conclusion th at the two nations are identical. Two objections have, as we said, been raised :— Parker declares th a t the frontier of the Juan-Juan did not reach far enough west to justify the identification. The empire of their first great ruler, Zaron, is said to have extended no further west than Harashar—at the most Tarbagatai or Kuldsa .4 He is said, indeed, to have annexed the remnants of the Hiung-Nu north and north-west of his kingdom ; but even so Parker does not believe th at his empire reached as far west as the Issyk-Kul, and says that “ there is not the faintest trace of any Jwen-Jwen intercourse with any western people, except the Epthalites and the people north-west of Ili ” .5 On the other hand, the Annals of Wei (c. C II fol. 15) say th at “ the kingdom of the Ta Yiie-Chi (Kusan Huns) . . . was bordered on the north by the JwenJwen, and they were many times exposed to their attacks. They therefore moved westward and established themselves in the city of Po-lo ” . 6 The reference is to a date about a . d . 450, and the move of the Kusan was from the Oxus valley, with the centre at Merv, to the district between the Caspian and the Aral Sea, Po-lo being probably Balxan or B a X a a p ., east of the present Bay of Krasnovodsk . 7 This passage shows the Juan-Juan well to the west of Lake Balkash in 1 1 3 4 * 6 7

I b id ., fr. 14, p . 205. T h e o p h . S im . V I I ec 7 -9 . C h a v a n n e s , D o c u m e n ts s u r les T o u e -K io u e O c cid e n ta u x , p . 228. A T h o u s a n d T e a r s o f the T a r ta r s , p . 117. I b id ., p . 121. C it. S p e c h t, E tu d e s s u r V A sie C en tra ls, p p . 11, 12. M a r q u a r t, E r a n id h r , p . 5 5 .

13 O N T H E G R E E K S O U R C E S F O R T H E H IS T O R Y O F T H E T U R K S

269

the middle of the fifth century, and it is worth noting th at the migration of the Kusan before them, reported by the Chinese, almost coincides in time with the flight of the Savirs before the Avars related by Priscus. In the latter case the Avars were themselves fugitives before peoples living eastward of them again, mists and dragons ; while the Juan-Juan had shortly before been heavily defeated by the Chinese. 1 The coincidence, if it is one, seems very remarkable. Moreover, it is not absolutely necessary to assume from Prisons’ story a permanent establishment of the Avars at all far to the W e st; there are instances enough of nomadic races fleeing hundreds of miles after a defeat such as the Juan-Juan had then suffered (which, it must be emphasized, was a military defeat a t the hands of Chinese generals, not an attack from another nomadic race bent on seizing their feeding-grounds) and returning afterwards to their own homes. Parker’s geographical scruples are thus superfluous.2 The chronological difficulty is th at the Chinese sources place the destruction of the Juan-Juan much earlier than Theophylactus puts the fall of the ‘‘ A vars” , and in any case before the destruction of the Epthalites, which probably did not take place until at least after a . d . 562.3 Chavannes’ attem pt to get round the difficulty by suggesting th at the “ Abdelites ” are not identical with the Epthalites 4 is wholly unconvincing. This difficulty also, however, can be resolved if we remember two things. Firstly, it is im portant to realize th a t the whole of Theophylactus’ story cannot 1 N e m e th , o p . c it., p . 110, q u o te s B ic u rin , S o b r a n ie j S v e d e n i fo r a g r e a t c a m p a ig n w a g e d b y th e C h in e se a g a in s t th e J u a n - J u a n in a . d . 4 5 8 . * P ro fe s s o r N e m e th , in d e e d , a rg u e s (o p . c it., p p . 107, 186 f.) t h a t t h e w h o le o p e r a tio n to o k p la c e m u c h f u r th e r e a s t t h a n h a s u s u a lly b e e n s u p p o s e d : t h a t th e O g u r tr i b e s h a d p r e v io u s ly b e e n liv in g o n th e I r ti s c h , th e S a v ir s b e y o n d t h e m a g a in , a n d t h a t th e r e s u lt o f t h e A v a r s ’ a t t a c k w as t h a t t h e S a v irs m o v e d o n ly a s f a r a s t h e I r ti s c h , t h e O g u rs to th e V o lg a. I t is t r u e t h a t w e fin d l a t e r tr a c e s o f a S ib ir k in g d o m o n th e I r ti s c h ; tr u e a lso t h a t th e “ O c e a n ” r e f e r r e d to b y P ris c u s is c e rta in ly n o t t h e C a s p ia n b u t r a t h e r th e d i s t a n t o c e a n s u r r o u n d in g th e w o rld , a n d th e u lt im a te c a u se s o f th e g r e a t m o v e m e n t a r e c o n c e iv e d a s ly in g v e r y f a r to th e E a s t in d e e d . O n t h e o th e r h a n d , th e S a v ir s a p p e a r , n o t lo n g a f te r w a r d s , n o r th - e a s t w a r d o f th e C a u c a s u s, a n d to e x p la in th is , N 6 m e th h a s to a s s u m e a f u r t h e r m ig r a tio n , o f w h ic h w e h a v e n o tr a c e , o f a ll o r p a r t o f th e n a tio n . I t s e e m s to m e m u c h m o re p la u s ib le to ta k e t h e t r a d iti o n a l v ie w , w h ic h b rin g s t h e S a v irs on to th e V o lg a a n d th e S a r a g u r i, O n o g u ri, a n d O g o rs w e s tw a r d o f it, i.e . to th e D o n d is tr ic t. T h e r e is , h o w e v e r, n o re a s o n to s u p p o s e t h a t th e w h o le o f th e S a v ir s w e re d r iv e n w e s tw a rd ; p a r t , p e r h a p s th e b u lk o f th e n a tio n , m a y w ell h a v e r e m a in e d b e h in d o n th e I r ti s c h . S im ila rly , n o t a ll o f th e O g u r tr i b e s w ill h a v e b e e n d riv e n w e s tw a r d b y th e S a v irs . O f th e th r e e n a tio n s m e n tio n e d b y P ris c u s , o n ly th e O n o g u rs p la y a n y th in g o f a p a r t in l a t e r h is to r y . T h e S a r a g u r i v a n is h f r o m th e sc e n e a l m o s t im m e d ia te ly ; th e “ U ro g i ” a r e n o t h e a r d o f a g a in a t a ll in a n y o rig in a l s o u rc e (sin ce th e li s t g iv e n b y th e s o -c a lle d Z a c h a r ia s R h e t o r is a m e re c o m p ila tio n fro m P r is c u s , P ro c o p iu s , M e n a n d e r, e tc .), a n d e v e n i f w e a s s u m e th e ir i d e n tit y w ith th e O g o rs o r U ig u rs , th e s e , to o , a r e v e r y u n im p o r ta n t a s n e ig h b o u rs o f th e E m p ir e d u r in g th e n e x t c e n tu r y . C le a rly , th e re fo re , th e s e “ O g u rs ” w e re o n ly a s m a ll f r a g m e n t o f th e w h o le i m p o r ta n t a n d n u m e ro u s n a t io n . 3 T h e a c c o u n ts o f th is a r e n o t d a t e d , b u t a c c o rd in g to T a b a r i, C h o sro e s u n d e r to o k h is c a m p a ig n a g a in s t t h e E p t h a li te s a f t e r h is p e a c e w ith R o m e (a .d . 5 62) a n d a f t e r in te r v e n i n g c a m p a ig n s a g a in s t th e K h a z a r s a n d th e A ra b s ( T a b a r i, t r . N o ld e k e , p p . 1 6 6 -7 ). 4 D o cu m en ts, p p . 25 0 f.

14 270 possibly, as some early critics assumed, be a textual reproduction of the Khagan’s letter of 598. To prove this, it is only necessary to point out th at the Khagan ascribes to himself the victory won over the Epthalites forty years previously, and says th at he accomplished this in alliance with “ Stembischada ” ; but Stembischada was not his ally but his predecessor, being identical with Menander’s Silzabul or Dilzabul.1 In fact, only the exordium and possibly the end of the letter are genuine quotations. The rest of the passage is really a long digression on the Eastern peoples, in which Theophylactus appears chiefly to be quoting from memory, his main source (direct or indirect) being, quite clearly, Menander. I t is very important to realize this. Now, Menander, although he mentions th at Turxanthus describes the European Avars as “ Varchonites ” , invariably refers to them himself as “ Avars ” , and this was clearly the name under which the nation was known in Constantinople. If, therefore, it transpires to be true th at these “ pseudoAvars ” were not identical with the real Avars, and if the references to “ Avars ” which appear to occur too late can be ascribed to the false, and not the true Avars, then our difficulties vanish. Regarding the former point there is nothing a priori impossible, or even improbable, in Theophylactus’ story. The name itself, as Nemeth points out ,2 simply means “ rebellious, disobedient ” , and is precisely such an appellation as would easily be assumed by a fragment of a nation. The references in Menander and Theophanes need, then, present no difficulties, since all of them refer to the “ pseudo-Avars ” . Theophylactus’ description of the flight of the “ Avars ” to the Taugas and Mukrit obviously, it is true, refers to the true Avars and not to the pseudo-Avars ; and he also dates this campaign after the Epthalite War. But we have shown th at he is only summarizing, and roughly enough, and may easily have been misled as to date by the passage in Menander (No. 5 above) which indicates th at the Turks would take on the “ Avars ” after the Epthalites. Menander was speaking here of the pseudoAvars, but Theophylactus misread him, or remembered him inaccurately. The true Avars, then, are really identical with the Juan-Juan, but our information concerning them is sadly reduced. I t now consists o f : the passage from Priscus3 ; the story in Theophylactus of their flight to Taugas and Mukrit, and his remark th at they were “ held to be the most active people among the Scythians ” ,4 and, presumably, th at their supreme ruler was called Khagan (since their imitators took the title with the name). The true reading in the second passage quoted above is probably /tap/Sapcov, 1 I b id ., p . 229. * O p. c it., p . 105. * T h e se a re , o f c o u rs e , th e t r u e a n d n o t th e fa lse A v a r s ; th e p o in t is m iss e d b y F e h 6 r , B id g a risc h -U n g a risc h e B e zie h u n g en , p. 26. 4 T h e o p h . S im ., iii, 7, p . 284, e d . B o n n . A4yerai yap 4v r o ts r o ts Z k v 6 lk o Is t o t u >v *A fia p w v vn ctva i tvrpcxtaraTO V 6vXov.

15 O N T H E G R E E K S O U R C E S F O R T H E H IS T O R Y O F T H E T U R K S

271

the whole passage being most likely a reminiscence of Priscus, and through him of Herodotus. The remainder of our references all refer to the “ pseudo-Avars ” ; but before we can turn to them we must consider briefly the third nation playing a role in our sources : the Kermichiones. Who were these people ? They cannot have been the Turks—Toue-Kioue, since their embassy reached Constantinople while the Avars were still negotiating with Rome for settlement inside the frontier—probably, therefore, as early as a . d . 558, whereas the true Turks appeared there first in 568 ; further, their ruler’s name was *AcjKrjXr or Scultor, while the Khagan of the Turks at th at time was Silzabul, Dizabul, or Istami. Neither can they have been the JuanJuan, as Marquart suggests ; nor the Epthalites, who were well known to the Byzantines and would not have been described in this way. Moreover, the Epthalites were known as White Huns , 1 and Mr. Bailey has pointed out th at the word Karmir xydn, meaning Red Chyon, occurs in a Pahlavi text in juxtaposition with Spet xyon, or White Chyon. The name Chyon, originally th at of some other race, was “ transferred later to the Huns owing to the similarity of sound ” . 2 The nation can hardly be other than th at which appears in the fourth century, under the name of Chionitae, in the steppes on the north-west frontier of Persia. These Chionites were probably a branch of the Huns, the other branch of which afterwards appeared in E urope; the latter appear to have attacked and conquered the Alans, then living between the Ural and the Volga, about a . d . 360,3 while the first mention of the Chionites is dated a . d . 356.4 In the fifth century, the name is replaced by th a t of the Kui§an or of the Kidarite Huns, who are clearly identical with the Kusan. The Kusan had, as we said, moved westward from the Oxus valley about a . d . 450. Most of the Kusan, however, emigrated to India after the defeat inflicted upon them by Peroz in a . d . 468.5 Some writers have assumed th at the Epthalites, who had crossed the Oxus and settled in the old Bactria about a . d . 425, now absorbed the remnants of both Kusan and Chionites 6 ; but there is no evidence to support this. The power of the Epthalites never extended, so far as we can tell, to the northern as distinct from the eastern frontiers of Persia, whereas the Kidarite Huns were so far west as to constitute a danger to the Roman Empire, as well as to Persia . 7 Meanwhile, the Huns continued to be active on the north at least up to the end of the fifth century .8 1 P ro c o p iu s D .B . P ersico , p . 10 : t o O vw to v rutv ' Eda\iTu>v 26vos, o vo n ep Xevtcous ovofidCovot. a B u lle tin o f the S chool o f O rien ta l S tu d ie s , v o l. v i, p a r t 4 , p p . 9 4 5 -6 . * F r o m a C h in e se s o u rc e q u o te d b y H i r t h , Ueber W o lg a -H u n n en u n d H iu n g - N u (S . B . der k . 6. A h . der W ise, z u M U n ch en , p h il. h is t. K l., 1899, p . 248). 4 A m m ia n u s M a rc e llin u s , x v i, 9 , 2, 4. 5 P ris c u s , fr. 4 1 , cf. M a r q u a r t, E r a n ta h r , p . 58. 8 S ee v o n W e se n d o n k , K u s a n , C h io n ite n u n d H e p th a lite n (K lio 26, p p . 366 ff.). 7 P ris c u s , fr. 33, 37 ; P s . J o s h . S ty lite s c. 9. a S ty lite s ,lo o . c it.,r e f e r s s p e c ific a lly t o th e “ H u n s c a lle d K u s h u n a w e C e r ta in ly m a n y o f t h e c a m p a ig n s a g a in s t th e H u n s d e s c rib e d b y h im , b y Z a c h a r ia s R h e to r , e tc ., m u s t r e f e r to th e

16 272 The statement th at they were known as “ Turks ” is very interesting, and is not, I am convinced, confined to Theophanes. I believe Menander to be referring twice to the Kermichiones under the name of Turk, where it has usually been supposed th at he is speaking of the Toue-Kioue. These passages occur in the fragment quoted above as No. 5, the words “ having heard of the flight of the Avars and how they had gone off after ravaging the territory of the Turks And again in No. 6 “ how many of the Avars escaped from the dominion of the Turks I t is true th at we thus find the name “ Turk ” used almost in consecutive breaths to describe different peoples ; but it must be remembered th at we probably have here, not Menander’s original text, but only summaries. I t seems certain from the embassy of 563 th at the pseudoAvars had been the subjects, not of the Toue-Kioue b u t of the Kermichiones 1 ; also the wording in passage No. 5 is strange indeed if the reference were to Silzabul’s own territories, while his cheerful and unconcerned attitude would contrast oddly with that of Turxanthus at the later period. I t will be noted, in fact, that the Toue Kioue establish a claim over the pseudo-Avars only after their conquests in Western Asia, not before. The ambiguity, moreover, is probably less than we suppose. I t will be noted th at Theophanes describes the Kermichiones as the “ Turks, formerly called Massagetae ” , while Menander calls the Toue-Kioue “ Turks, formerly called Sakae ” . The distinction is probably deliberate, and affords at the same time a valuable clue to the positions of the two nations. The Kermichiones lived in the homes of the old Massagetae, viz. on the Jaxartes and the A ra l; the Toue Kioue further east, in the homes of the old Sacae. There is another passage in Theophylactus where the name appears to be used in the same way. A pseudo-Avar dignitary, having got one of the wives of his Khagan into trouble, decided “ to flee to his ancestral tribe. These are Huns, living in the East, neighbours of the Persians, and to many more familiar under the name of Turks ” (ov$ kcl I TovpKovs anoKaXtlv t t o XXo l s yvuipLfJLwrepov)}

The name “ Turk ” appears to have been used a t an earlier date still for the peoples living north-west of Persia. Tabari uses it of the Ku§an Huns in the fourth century, and both he and Dlnawarl tell of an inroad by “ Turks ” in the reign of King Bahram V (420-438).3 In view of the Byzantine usage which we now see to have been frequent, it is difficult to dismiss these references n o r th e r n fr o n tie r , n o t to th e E p t h a li te s , s in c e th e P e r s ia n s c o u ld n o t h a v e a p p e a le d to R o m e f o r s u b sid ie s a g a in s t t h e la t t e r . S o m e o f th e H u n s in q u e s tio n m u s t, in d e e d , h a v e b e e n n e i th e r in E p t h a li te n o r in K u S a n -C h io n ite t e r r ito r y , b u t n o r th o f th e C a u c a s u s ; b u t in v iew o f th e v a g u e n e ss o f o u r s o u rc e s i t is im p o ss ib le to a s s ig n e a c h re fe re n c e to it s p r o p e r s u b je c t. 1 Cf. T h e o p h a n e s , a n d in p a r t ic u l a r C o rip p u s loc. c it. Q u o s c o n t r a in g r a to s d e fe n d im u s , a r m a p a r a m u s ? O b s ta m u s d o m in is , p ro fu g is d a m u s o s tia s e rv is ? * T h e o p h . S im ., i, 8, 5. 3 M a r q u a r t, E r a n ia h r , p p . 5 0 -2 . T a b a r i, tr . N o ld e k e , p . 99.

17 O N T H E G R E E K S O U R C E S F O R T H E H IS T O R Y O F T H E T U R K S

273

as purely anachronistic; but it may be necessary to revise our estimates of the first appearance of the name. The Kermichionites, then, are identical with the Chionites of two centuries previous, and in the middle of the sixth century clearly ruled over a considerable empire north of Persia, in the Jaxartes-Aral district. They must have played a part in the history of the time and age much larger than has generally been recognized.1 I t is quite clear from our various accounts of the embassies—Corippus is explicit on the point—th at the pseudo-Avars had been subjects of the Kermichiones. For the rest we have only one statement of the first importance on their origin : the remark of the Turkish Khagan, preserved by Menander, that their real name was “ Varchonites ” (Ovapx^vZrai). Unhappily, this does not get us much further. Marquart explains the name as “ denizens of the Orkhon ” 2 ; but this is only conjecture. Nemeth is able to adduce a very similar Mongol tribal name ,3 but this does not show us where our race was residing in the fifth century. More important is his information th at the pseudoAvars belonged linguistically to the same branch of the Turkish family as the Huns and the Toue-Kioue, and not to the Ogur or Bulgar branch which was living at the time on the Volga and the Black Sea . 4 Theophylactus appears to tell us more^ but the appearance is deceptive. I t is most important to keep firmly in mind that in the passage in question, viz. in the alleged letter from the Turkish Khagan (excluding its beginning and end), Theophylactus is not an original source, but is only directly or indirectly epitomizing Menander, and distorting him in the process. His statement th at the pseudo-Avars were originally a federation of tribes known respectively as “ Var ” and “ Chunni ” is clearly a mistaken rationalization of the writer’s own : an attem pt to be intelligent about the name Varchonite. The “ Vars ” are the Avars ( = pseudo-Avars), the Chunni the Bulgars, who, as we know, were almost as numerous in the Avar kingdom in Hungary as the Avars themselves, undoubtedly included Hunnish eleihents, and were frequently described by the Greeks as Huns. This point is confirmed by Theophylactus’ statement th at the Kotzageri, who “ were also of the Var and Chunni ” , fled from the Turks to the Avars . 5 Now, the Kotzageri were, in fact, a component of the Avar kingdom in the West, and almost as numerous as the Avars themselves, whom they considered rather as their equals than their superiors6; but they were never “ Varchonites ” , but Bulgars by origin. Incidentally, the figure of 10,000 in Theophylactus’ 1 N e m e th , fo r e x a m p le , in h is a d m ir a b le w o rk , ig n o re s th e m a lto g e th e r ; h e a lso g o es s lig h tly a s tr a y in h is c o n c lu s io n s b y a c c e p tin g lite r a lly (a p p a r e n tly ) T h e o p h y la c tu s ’ in v e n tio n s o n th e o rig in s o f th e p s e u d o -A v a rs . * E r a n s a h r, p . 52. 8 * 6 *

O p . c it., p . 103. I b id . T h e o p h . S im ., v ii, c. 8. M e n a n d e r, f r. 3, e xc. d t L eg . G ent., p . 4 4 3 .

18 274 account is also taken from Menander, 1 and most likely the hordes of Tarniach and Zabender could be equally well identified if we had Menander’s work complete .2 We may, therefore, disregard the story about the Var and Chunni; and we must proceed no less cautiously with the statement th at the pseudo-Avars were an Ogor tribe. I t may be remarked, to begin with, th at even Theophylactus does not say th at the pseudo-Avars themselves lived on the “ Til ” ; but only th at they were an Ogor tribe, and th at the Ogors lived there. This, again, we can almost certainly trace back to Menander, who mentions the Volga under its familiar name of 'ArrjX (in the form of *ArrLXa) and shows Ogors living on it .3 He does not, it is true, use the epithet “ black ” in the fragments of his work remaining to us ; but then we do not possess all his writings. The Volga was, of course, in the East to a Greek, so there is no need to boggle at the phrase irpos r fy avaroXrjv or, with Marquart, to identify the TLX with the Togla. The fact, however, remains th at Theophylactus does tell us th at the pseudoAvars were an Ogur tribe (which, linguistically and anthropologically , 4 is not correct). Further, both he and Menander omit all mention of the Kermichiones, whose tributaries the “ Avars ” had certainly been. I t is, moreover, remarkable th at we find in our sources no mention of an Ogor state before the arrival of the Turks, while we do find Kermichiones ; on the other hand, the destruction of the Kermichiones is nowhere mentioned, whereas Theophylactus mentions the destruction of the Ogors as the next great achievement of the Turks after the Epthalite war ; Menander shows the Khagan boasting of the destruction of the powerful Onogurs, and mentions Ogurs as Turkish subjects on the Volga. I t would really appear as though the empire known to the Persians as Kermichiones must have been identical with th at known to the Turks as Ogor— a supposition which does not, of course, imply th at it was racially homogeneous. I t is perfectly possible th at Ogur = Bulgar tribes had gained the upper hand after their westward move in the fifth century, but the Persians had retained the old name. Of this empire the pseudo-Avars must have been one of the eastern components. It is interesting to note th at according to Corippus, who is very specific on the point, they had come to the Caucasus through Persia . 6 Menander 1 I b id ., p. 4 5 8 : *ccu 8r) TrapeKeXevoaro 8c*a ycXtabas t w v Korpiyovpujv Xcyopevcov Ovwtuv Siafirjvau rov Ldov Trorapov. 2 I s h o u ld n o t b e s u rp ris e d i f ZapevBtp tu r n e d o u t to b e id e n tic a l w ith th e Z a fk p y d v , th e C u tig u r le a d e r m e n tio n e d b y A g a th ia s (H is t. V -1 0 , e tc .) a n d M e n a n d e r, a n d r a p v ia x to h a v e s o m e c o n n e c tio n w ith M e n a n d e r’s rapyLrios o r rapytos. 8 E x c . de L eg. G ent., fr. 8, p . 453. 4 T h e A v a r sk u lls e x a m in e d in H u n g a r y sh o w a h ig h p r o p o r tio n o f p u r e ty M o n g o l ty p e s . 6 L o c . c it. 11.277 ff. Q u ern P e rs a e ti m u e r e fe ra e , g e n ib u s q u e m i n a n tis A d m o v e re m a n u s , p a c e m e t m e r u e r e p r e c a n d o N i f a c e re n t, f r u s t r a celsis B a b y lo n ia m u ris C in c ta f o r e t, d o m in o s A v a re s n u n e fe r r e p a r a t a . R u p im u s E u p h r a te m , g e lid o s s u p e r a v im u s a m n e s . . . .

19 O N T H E G R E E K S O U R C E S F O R T H E H IS T O R Y O F T H E T U R K S

275

also says th at they only reacned the Caucasus “ after long wanderings ” (7roAAd TTaLpavoaTTjcjavTGs). That they took the southern route, being bought off, very likely, by the Persians, is rendered the more probable by the fact th at they attacked the various Hunnish tribes after and not before reaching the territory of the Alans (so Menander, whom Theophylactus only summarizes inaccurately). Our sources cannot tell us anything more of the “ pseudo-Avars ” in their early homes, except only the statem ent th at the fugitives numbered some 20,000, while others of the nation remained behind and became subject to the Toue-Kioue, the latter statem ent being confirmed by the later story of the fugitive Avar dignitary.

m

The Attack on ‘Valandar’

T h e sto ry w hich M as’u d i r e la te s in h is ^M eadow s of G o ld » ’) Of a n a tta c k b y fo u r n o m a d n a tio n s of th e n o r th on a G re e k f ro n tie r fo rtre s s n a m e d «V a la n d a r# is o n e fo r w hich n o s a t i s fa c to ry e x p la n a tio n h a s y e t b e e n fo u n d . T h e fo llo w in g e s sa y c o n s titu te s an a tte m p t to lin k it u p w ith a n e p iso d e in th e r e la tio n s b e tw e en G reece a n d B u lg a ria in th e te n th c e n tu ry . T h e f ir s t p ro b le m to be c o n s id e re d c o n c e rn s th e n a m e s o f th e n a tio n s in v o lv ed . T h e m ss. h a v e m a n y v a ria tio n s , b u t th e y V

a p p e a r to r e p r e s e n t a p p ro x im a te ly th e fo llo w in g n a m e s: B a g n a ; .

v

v

B a g g it'd ; B ag n ak ; a n d a n a tio n k n o w n a s tn e N fik a rd a . T w o of th e se n a m e s a re c o m p a ra tiv e ly e a sy to e lu c id a te . V

V

T h e B agniik a re o b v io u sly the P e tc h e n e g s. T h e B a g g ird (B a s h k irs) is th e n a m e u n d e r w hich M as’u d i, a n d m a n y o th e r O rie n ta l w r ite r s 1) h a b itu a lly d e sc rib e th e M a g y ars. The N fik ard a h a v e h ith e rto b a ffle d c o m m e n ta to rs, b u t I h a v e b e e n p u t on th e r ig h t tra c k th ro u g h th e w o rk k n o w n a s th e «A brege d es M e rv e ille s» 3). T h is w ork, w hich is u n d o u b te d ly based , in p a r t a t least, on lo st w ritin g s of M as’udi, n o ta b ly h is «Book of D iffe ren t S o rts of K n o w le d g e * 4) c o n ta in s a sectio n o n a p eo p le w ho a re d e sc rib e d in M. C a r ra de V a u x ’ ed itio n a s E l Ih ta ra d a h . T he section r u n s : «E1 I h ta r a d a h . T h e se a r e ‘) A translation of the passage is given by M arquart, Streifziige p. 60 fL 2) Abulfeda, Id risi, D um asqi and n u m ero u s later w riters use this form , and practically all of them , in consequence, fall into the same confusion as Mas’udi and give us two nations—B ashkirs and U u n k ar—living in the plains of Hungary. For a convenient list ofrefe rences, see Nem eth., Magna Hungaria , Leipzig and Vienna 1929; also my own work, The M agyars in the Ninth Century, Cam bridge, 1930 pp. 33 ff. ’) L’Abrege des M erveilles ; ouvrage attrib u e a Mas’udi ; M. C arra de Vaux, Paris 1898. F o r discussion of the relationship between this work and th at of Mas’udi, see Mr. C arra de Vaux’ introduction and my M agyars in theNinth Century, pp. 152—5. 4) Macartney op. cit. p. 152.

21 160

th e d e s c e n d a n ts of A m ir so n of J a p lie t, a n d o c c u p y a v a s t k in g d o m b e tw e e n th e R u m is a n d th e F r a n k s . T h e ir k in g is p o w e rf u l; th e y h a v e m a n y cities. M ost of th em a r e C h ristia n s to d a y , b u t som e of th em h a v e n o relig io n . T h e y h a v e to fig h t a g a in s t th e F r a n k s a n d th e S la v s w h o s u rr o u n d a n d re p u ls e them . T h e ir d r e s s is th e sam e as t h a t of th e R um is> . N ow , th e d e s c rip tio n of th is n a tio n is on th e w hole w ell a p p lic a b le to the M a g y a rs (w ith som e in ac c u ra c ies, in d e e d , b u t n o t too m a n y fo r M as’udi) in th e ir p r e s e n t h o m e s in H u n g a r y ; a n d w h en we fin d th a t a v a r ia lectio of th e n a m e r u n s E l A nk ira d a h , we see t h a t th e n a m e is in fa c t n o n e o th e r th a n th e n a m e « H u n g a ria n » . All la te r fo rm s, fo u n d in Dima&qi a n d o th e rs , g o back, in fact, to th is o rig in a l fo rm , a n d it is n o t h a r d to see th a t th e N fik a rd a h of th e m ss. of th e «M eadow s of Gold» h a s m e re ly tra n s p o s e d th e in itia l v o w el a n d c o n s o n a n t, so t h a t w e sh o u ld re a d h e re U n k a r d a h o r A n k a r d a h = H u n g a ria n s . V

T h e B ag n a a re co m m only ta k e n a s a v a r ia n t on th e n a m e o f P e t c h e n e g 1) b u t it is d iffic u lt to se e h o w o r w h y tw o fo rm s s h o u ld h a v e c re p t in. A n o th e r p o s s ib ility is s u g g e s te d if we com p a r e th e o th e r p a s s a g e s in w hich M as’u d i s p e a k s o f th e se «Vala n d a r h o rd e s* , a s h e te rm s them . In th e L iv re d ’A v e rtis s e m e n t pp. 244, 5 M as’ud i s a y s t h a t m o st o f th e E u r o p e a n p ro v in c e s of th e B y z a n tin e E m p ire a r e o c c u p ie d b y «the B u rg a r a n d b y h o rd e s o f n o m a d T u rk s c a lled V

V

v

V a l a n d a r . . . T h e y c o n sist o f the B a g n a k , th e B a g n a , B a g g ird a n d N u k a r d a ( A n k a rd a h )..........» I b id pp. 247,8. In o u r b o o k «of th e D iffe re n t S o rts of K n o w led g e e t c * . . , we h a v e s p o k e n . . . of th e c o u n trie s w hich v

w e re in v a d e d b y th e B u r g a r , th e B ag n ak T u rk s a n d th e o th e r V alandar® . Ib id p. 194 sp e a k s ef «the B u rg a r, w h o a re a b ra n c h of th e v

S la v s, a n d th e B ag n ak , w ho a re T u rk s. T h e G re e k s h a v e p e o p led w ith th em s e v e ra l of th e ir f o rtre s s e s to w a rd s th e fro n tie rs o f S y ria , a n d h a v e m a d e of them a r a m p a r t a g a in s t th e B or') So M arquart, Streifziige pp. 67

22 161 V

giin a n d o th e r h o stile n a tio n s w h ich s u r r o u n d th e ir E m p ire * . N ow in th e second a t le a s t of th e s e th re e p a s s a g e s , it m u st s tr ik e th e r e a d e r th a t th e « B u rg a r» a re a p p a r e n tly in c lu d e d a m o n g th e V a la n d a r n a tio n s ; a n d in th e th ir d p a s s a g e th e y a re clo sely co u p led w ith th e P e tc h e n e g s . In th e f ir s t th e y a r e d is tin g u ish e d from the V a la n d a r n a tio n s , b u t y e t a s so c ia te d w ith th em . T h is re p e a te d a sso c ia tio n b e tw e e n B u lg a rs a n d V a la n d a r n a tio n s is a s trik in g one, a n d will, I b e lie v e, le a d u s to th e s o lu tio n of the m y ste ry . T h e n a m e u n d e r w h ich M as’u d i k n e w V

V

th e D a n u b e B u lg a rs w as B orgfin a n d the n a m e B ag n ii is n o th in g m o re th a n a c o rru p tio n of th is nam e, u n d e r th e in flu e n ce V

of th e su c ce e d in g B agnuk, th u s g iv in g us th e D a n u b e B u lg a rs for th e la s t of th e se n a tio n s. W h y , th en , do we g e t th e a d d itio n of the o th e r n am e, B u rg a r ? T h e a n sw e r is to b e fo u n d in th e p a s s a g e in w h ich M as’udi tr e a ts o f the V olga B u lg a r s ') . T h is p a s s a g e c o n sists, a c tu a lly , of th re e lo n g p a ra g r a p h s . T h e f ir s t a n d th e th ir d a re b a s e d on Ib n F o z la n ’s d e sc rip tio n of V olga B u l g a r i a 2), u n d e r th e n a m e of B u r g a r ; th e seco n d p a r a g r a p h is an in te rp o la tio n a n d d e s c rib e s th e ,d e e d s of th e M a g y ars, w ho h a v e b e e n c o n fu se d w ith the B u lg a rs . T h e o p e n in g s e n te n c e s of th is p a s s a g e r u n : T h is k in g m a k e s r a id s a g a in s t the te r r ito r ie s of C o n s ta n tin o p le w ith som e 50,000 r id e r s a n d m ore, a n d lets his b r ig a n d h o rd e s ro am o v e r it to th e la n d of R om e, A n d alu s* ), th e la n d v

of B organ, th a t of th e G a lle g o s 4) a n d F ra n k s . F ro m h e re to C o n s ta n tin o p le is n e a rly a tw o m o n th ’s c o n tin u o u s jo u rn e y th ro u g h c u ltiv a te d te rrito rie s a n d d eserts» . T h e n follow s a s to ry a b o u t a m e e tin g b e tw e e n th e s e B u rg a r a n d «M oslims from th e d istric t of T a r s u s on th e S y ria n m ilita ry fro n tie r* , a n d the p a s s a g e e n d s : J) T ranslated by M arquart, Streifziige pp. 149 ff. J) Mas’udi’s accounts both of the Volga B ulgars and of the Bashkirs n o rth of the Aral Sea are undoubtedly based directly on Ib n F ozlan’s w ork, not confirm atory of it. ’) = Andalusia. 4) = Galicia, Spain.

23 162

♦T he B u rg a r a re a m ight}' people, u n a p p ro a c h a b le a n d of g r e a t c o u ra g e , to w hom th e n e ig h b o u r in g p e o p le s a re s u b je c te d . O ne h o rse m a n from a m o n g th em [from th o se w h o a c cep ted Islam w ith th a t king] can d e ty 100 a n d 200 u n b e lie v e rs . T h e in h a b ita n ts of C o n s ta n tin o p le to d a y p ro te c t th e m se lv e s a g a in s t them o n ly th ro u g h th e ir w alls, a n d e v e n so all th o se w ho live in th a t clim ate p ro te c t th e m se lv e s o n ly th ro u g h w alls a n d fo rtre s s e s W hen, now , M as’udi r e f e r s to th e B u r g a r a n d th e V aland ria n s (L iv re d ’A. p. 244). to th e B u rg ar, the Bagnfik, etc. Ib id 247', a n d to «the B u rg a rs w h o a re a b ra n c h of th e S la v s a n d th e B agnak (Ibid p. 194) he is in e v e ry case th in k in g of the V olga B u lg a rs. T h e r e m a rk th a t th e B u lg a rs a r e «a b ra n c h of th e Slavs* d o es n o t re fe r to th e a s sim ila tio n of th e D a n u b e B u lg a rs b y th e ir S la v o n ic su b je c ts, b u t to Ib n F o z la n ’s d e s c r ip tion of th e k in g of th e V o lg a B u lg a rs a s th e « ru le r of th e Slavs*; a n d th a t in sp ite of M as’u d i's ow n c o rre c tio n th a t th ey a re «a T u rk is h people*. S uch sm a ll in c o n siste n c ie s d id n o t w o rry o u r a u th o r. B u t w hile M as’udi d e s c rib e s the V olga B u lg a rs as takingp a r t in th e ra id s on C o n sta n tin o p le , th is w as, of co u rse, n o t a c tu a lly th e case. T he n a tio n w hich did do so w as th a t of the D a n u b e B u lg a rs , a lr e a d y in c lu d e d by M a s’u d i in h is list of th e V

V

V a la n d a r h o rd e s u n d e r th e n a m e B o rg an , c o rr u p te d in to B a g n a . An in k lin g , h o w ev er, from a n o th e r so u rc e m u st h a v e re a c h e d him th a t B u lg a rs w ere c o n c e rn e d in th e se r a id s a n d w a r s ; a n d v

as h e did n o t know th a t th e B o rg an w ere id e n tic a l w ith th e D a n u b e B u lg a rs , h e d r a g g e d in the o n ly B u lg a rs w hom he k n e w (those of the V olga) a n d a d d e d th em to h is list in the L iv re d ’ A v e rtisse m e n t. T he fo u r V a la n d a r n a tio n s are, th en , th e P e tc h e n e g s, th e D a n u b e B u lg a rs a n d th e M a g y ars, th e la s t n a m e d a p p e a r in g u n d e r th e tw o n a m e s of B a s h k irs a n d H u n g a r ia n s . W e a re now r a th e r b e tte r e q u ip p e d to d e a l w ith th e s to r y in th e «M eadow s of Gold*. I t is c le a r th a t m o re th a n o ne s to ry h a s b e e n b le n d e d h e re in to a d ish a rm o n io u s w hole. It is, h o w ev er, a s to n is h in g how little elim in a tio n will m a k e th e s to ry c o h e re n t. S ince th e s e n te n c e s :

24 163

•T h e y e x te n d th e ir p r e d a to ry e x c u rs io n s a s f a r a s R o m e w h ich is in th e d ire c tio n of A n d a lu s. T h e y a re v ic to rio u s o v e r all th e n a tio n s w ho liv e t h e r e *, a n d th e c o rre s p o n d in g . • T h e y m a d e p r e d a to ry e x c u rs io n s all o v e r th e s e c o u n trie s, a n d a s f a r a s th e c o u n try of (th e S la v s a n d R om e. A t p r e s e n t th e ir in v a s io n e x te n d s e v e n to) th e f r o n tie r s of S p ain , F r a n c e a n d th e G allegos. T h e p r e d a to r y in c u rs io n s of th e a b o v e - m e n tio n e d T u rk is h n a tio n s c o n tin u e to th is d a y to in fe s t C o n sta n tin o p le , a n d th e a b o v e - m e n tio n e d k in g d o m s# , o c c u r e ls e w h e re (in th e r e p o r t of th e M a g y a rs e m b o d ie d in th a t on th e V olga B ulgars), w h ere th e y a re c o n n e c te d so le ly w ith V

the M a g y a rs (the B ^ g g ird ), we h a v e a r ig h t to s e t th em a sid e as fo rm in g o n e s e p a r a te in g re d ie n t o f th e p r e s e n t s to ry of an in v a s io n b y fo u r n a tio n s. But we c a n g o f u r th e r th a n this. W e can d e ta c h c e rta in s e n te n c e s as b e lo n g in g p r o p e r ly to th e «A nk a rd a h * a lo n e T h e se a re th e s e n te n c e s im m e d ia te ly fo llo w in g th e m e n tio n of th e ir n a m e : • T h e ir k in g s h a d so v e re ig n p o w e r 1), th e y h a d w a rs w ith th e B y z a n tin e s a fte r th e y e a r 320 o r in th a t y e a r* *), a n d p e r h a p s also th e p h ra s e w hich I h a v e p u t in b ra c k e ts above: « ..........th e S la v s a n d R o m e 8). A t p r e s e n t th e ir in v a s io n e x te n d s e v e n t o ..........» I t a p p e a r s to m e lik ely th a t th e fo rm e r, a t le a s t of th e se p h r a s e s (w hich refe r, o f course, to th e g r e a t H u n g a r ia n a tta c k on C o n s ta n tin o p le in 934 A. D.) w as ta k e n fro m th e o rig in a l a c c o u n t in th e «B ook of D iffe re n t S o rts of K n o w le d g e * on w hich th e p a r a g r a p h on th e A n k ira d a h in th e «A br6ge d e s M erveilles> is also b a sed . B ut s tr ip a w a y th e se a c co u n ts of M a g y a r a c tiv itie s, su b l) cf. the A b re g e ; «their king is powerful*. 4) Ib n al A thir, who retells the story after Mas’udi, fixes the date as 322 ( = 934 A. D.) (M arquart, Streifziige p. 64). *) The Abrege several tim es m entions hostilities betwen the H u n g arian s, Slavs and Rumi. T his p hrase m ay, how ever, belong to the passage V

on the B aggird.

25 164

tra c t tw o from th e n u m b e r o f n a tio n s e n g a g e d , a n d w h a t h a v e w e left V Q u ite a c o h e re n t a n d c o n s is te n t s to ry , u n d a te d , in w hich tw o n a tio n s o n ly a re e n g a g e d . T he re a l h e ro e s of the V a la n d a r s to r y a re u n d o u b te d ly th e P e tch e n e g s. It w a s th e P e tc h e n e g s , a n d th e y alone, w ho liv ed o n the fro n tie r ,of th e K h a z a rs , a n d th e A l a n s 1). It is th e k in g of th e P e tc h e n e g s w ho is r e p r e s e n te d as ta k in g c o m m a n d a t V ala n d a r ; h is n a tio n is th e «b r a v e s t of th e fo u r* . T h e P e tc h e n e g s a lo n e w e re e x p e lle d fro m th e A ral S e a b y th e G uz, Ivharlu k h , K aim ak, e t c .2). F in a lly th e s to r y t h a t M oslem m e rc h a n ts from K h a z a ria , A lania, etc. w e re m o b ilised fo r th e cam paign,, seem s to echo th e a c co u n t of th e c o n v e rs io n of th e P e tc h e n e g s to Islam , p re s e rv e d b y A1 - B eltri in a p a s s a g e a lm o st c e rta in ly d e riv e d from M as’ u d i’ s lo st w o r k 3). I t m a y b e a d d e d th a t the w hole s to ry seem s to b e to ld fro m a P e tc h e n e g p o in t of view . O ne m ay c o n je c tu re th a t its a u th o r w as o n e of th o se m e rc h a n ts w ho w ere also re s p o n s ib le fo r th e v e ry o p tim istic a c c o u n t of th e P e tc h e n e g c o n v e rsio n to Isla m . The P e tch e n e g s, th e n , p la y e d th e le a d in g p a r t in th e Va la n d a r s to r y ; a n d o ne o th e r n a tio n w as c o n c e rn e d in it. W e h a v e se en re a s o n to c o n je c tu re th a t th is n a tio n w as th e «B orv

gan», o r D a n u b e B u lg a rs ; a n d in p o in t of fa c t if tw o n a tio n s w ere c o n c ern e d in th is ra id , o n e b e in g th e P e tc h e n e g s , the o th e r can on ly h a v e b e e n th e D a n u b e B u lg a rs. F o r th e y alone w ere im m ed ia te n e ig h b o u rs of th e P e tc h e n e g s 4) ; a n d a G re e k fro n tie r fo rtre ss, on th e b o r d e r s of o ne of th e n o m a d n a tio n s, e ig h t d a y s ’ m arc h fro m C o n s ta n tin o p le , c a n n o t p o ssib ly h a v e b e e n a n y w h e re else th a n on th e B u lg a ria n fro n tie r. E x a c tly w hich fo rtre s s is m e a n t can h a rd ly be s a id w ith full certainty*

*) Constantine Porphyrogenetos, De A dm inistranclo Im perio (D. A. I.) ch. 42. 2) cf. D. A. I. ch. 37. 3) See my The M agyars in the N inth C entury, pp. 7. ff. where I show that the lost w ork of Mas’udi was m ost probably one of a l—B ekri’s main sources. 4) The M agyars were also n eighbours of the Petchenegs, but separated from them a four d a y s’ journey (D. A. I. ch- 42): and in 949 A. D. had not been at w ar with the Petchenegs since S96 (Ibid ch. S,37;.

26 165

b u t th e s u g g e stio n a d v a n c e d b y M a rq u a rt of D e v e lto s ') se em s to m ee t e v e ry re q u ire m e n t. In a n y case, if n o t D e v e lto s, it m u st h a v e b e e n som e fo rtre s s in th e im m ed ia te n e ig h b o u rh o o d . N ow , th e re is o ne e v e n t in G re c o -B u lg a ria n h is to ry to w hich I b e lie v e th a t M as’u d i’s s to r 3r can b e a p p lie d ; a n d if it can be so a p p lie d , som e little lig h t will b e th ro w n on a c o m p lic a te d e p iso d e of B y z a n tin e h isto ry . T h is is th e G re c o -B u lg a r w a r of 913 —7. T h e sto ry of th is w ar, as told by m o st of th e G re e k c h ro n ic le rs 2) is th a t S ym eon, th e a m b itio u s C z a r of B u lg a ria , s u d d e n ly in v a d e d th e B y z a n tin e E m p ire in fo rce in 913, a n d p itc h e d h is c a m p b e fo re th e w alls of C o n sta n tin o p le . S e e in g th e s tr e n g th of th e w alls, h o w e v e r, he h e s ita te d to a tta c k , a n d a p e a ce w as co n clu d ed , la r g e ly th ro u g h th e m e d ia tio n of N ic h o la s, th e v e n e ra b le p a tr ia r c h of C o n s ta n tin o p le an d r e g e n t fo r th e y o u n g C o n s ta n tin e —a m an w hose in flu e n c e o v e r S ym eon, w h o h a d b e e n b r o u g h t u p in C o n sta n tin o p le , a n d w ith a view to a clerical c a re e r, m u st h a v e b e e n c o n sid e ra b le . In th e fo llo w ing y e a r, h o w ev er, the q u e e n -m o th e r h a v in g re fu s e d to c a rr y o u t p a r t of the co n d itio n s, a n d m a rry the y o u n g E m p e ro r to o n e of S y m e o n ’s d a u g h te rs , S y m eo n once m o re in v a d e d T h ra ce , a n d a c o n su lta tio n h a v in g b e e n h eld as to how b e s t to a v e r t th e d a n g e r, J o h n B ogas v o lu n te e re d , in r e tu r n fo r the d ig n ity of p a tric ia n , to lea d th e P e tc h e n e g s a g a in s t th e B u lg a rs . He w as d e s p a tc h e d on his e r r a n d w ith g ifts; a tr e a ty w as c o n c lu d e d , a n d B ogas, a fte r re c e iv in g fro m the P e tc h e n e g s h o s ta g e s a n d a p ro m ise of assistan ce, r e tu r n e d to C o n sta n tin o p le . M eanw hile S ym eon c o n tin u e d h is a d v a n ce , and su c ce e d e d , b y th e tre a c h e ry of its c o m m a n d er, in c a p tu r in g A d ria n o p le . T h e G re e k forces, h o w ev er, do n o t a p p e a r to h a v e b e e n d e s p a tc h e d a g a in s t him a t th e tim e, n o r w e re th e P e tc h e n e g s e a lle d u p o n . I t w as o n ly in 917 th a t a g r e a t c o m b in e d e ffo rt w a s m ad e. A h u rrie d p e a ce w as c o n c lu d e d w ith th e S a ra c e n s, a n d th e E a s te r n L e g io n s w ere tra n s f e r r e d to E u ro p e , th e w h o le of ') D’O hsson in his «Voyage d ’Abou Kassim» gives a va ria lectio Veli•den, which approaches Develtos very nearly. 5) Theophan, cont. pp. 3S7—9. C edrenus, etc.

27 166

th e la n d fo rc e s b e in g p laced u n d e r th e co m m a n d of L eo P h o cas, w hile R o m a n u s L e c a p e n u s, th e D r u n g a r iu s of th e F le e t, w as s e n t to th e D a n u b e to o c c u p y th e r iv e r m o u th s a n d f e r r y th e P e tc h e n e g s o v er. U n f o r tu n a te ly fo r th e G re e k s, p e rso n a l: fe u d s p r o v e d s tr o n g e r th a n th e g e n e ra l a d v a n ta g e . P h o c a s w as th e en e m y of L e c ap e n u s, w h o d e te s te d B o g as. T h e P e tc h e n e g s d u ly a rriv e d , b u t a v io le n t q u a r r e l b ro k e o u t b e tw e e n L e c a p e n u s a n d B ogas, a n d th e P e tc h e n e g s «s e e in g th em f ig h t in g a n d s q u a b b lin g w ith each o th e r, r e tu r n e d hom e*. P h o c a s. d e p riv e d of th is d iv e rsio n , s u ffe re d a te r r ib le d e fe a t a t th e h a n d s of th e B u lg a rs on th e A ch elo u s, n e a r A nhialo, w ith «a s h e d d in g of blood su ch as h a d n o t b e e n seen fo r ages* (A ug, 20, 917). T h e la te r c o u rs e of th e w a r w as w a g e d w ith v a r jd n g fo rces. A d v a n c in g on C o n sta n tin o p le , S y m eo n w as c h e c k e d a n d d e fe a te d . It is n o t e x p lic itly s ta te d th a t A d ria n o p le w as reco v e re d , b u t th is m u st h a v e b e e n the c a se , fo r la te r, w h ile th e w a r w as b e in g p u rs u e d a fte r R o m a n u s L e c a p e n u s h a d b e e n p ro c la im e d E m p e ro r, S ym eon a g a in c a p tu r e d A d ria n o p le ; b u t th e p o p u la tio n re v o ltin g a n d e x p e llin g th e g a r r is o n w hich he h a d left th e re , it q u ic k ly fell o n ce m o re in to th e h a n d s o f the G re e k s. It w as on ly a fte r this, a n d a fte r S y m e o n h a d once m ore led his fo rc e s to th e g a te s of C o n s ta n tin o p le , th a t p e a c e w as c o n c lu d e d , th e p a tr ia r c h N ic h o la s a g a in p la y in g a p a r t in the n e g o tia tio n s . A s e rie s of le tte rs from th e P a tria rc h N ich o las to th e B u lg a ria n C z a r h a v e b e e n p r e s e r v e d 1), w hich th ro w a s lig h tly d iff e r e n t lig h t u p o n the w hole su b je c t. T h e le tte rs, a lth o u g h o c c u p y in g an in o rd in a te a m o u n t of space, c o v e r o n ly a s h o r t p e rio d . T h e y o b v io u sly b eg in in th e p e rio d w hich e la p s e d b e tw e e n th e conclusion of p e a ce in 913, a n d th e fre s h o u tb u r s t of h o stilitie s in 914. T he P a tria rc h b e g in s b y tra n s m ittin g to th e C z a r a p a in fu l r u m o u r w hich h a d com e to h is e a r s th a t S ym eon w as a lre a d y p r e p a r in g to r e p u d ia te th e re c e n t t r e a ty t a n d re n e w h is a tta c k s on C o n s ta n tin o p le 2). H e b e g s S y m e o n ') E dited by Migne, Patrologia Graeca, vol. Cl J) op. cit. col 48.

28 167

to sen d to D e v e lto s h is a s s u r a n c e th a t h e in te n d s to o b s e rv e the tr e a ty (n oiip ivtov). W h e n the n e x t le tte r is w ritte n , th e w a r h a s o b v io u sly b e e n ren e w ed , a n d th e P a tria rc h h e a p s on th e C z a r his p a te r n a l r e p r o a c h e s ') B u t th e m ost in te r e s tin g le tte r — the seco n d w ritte n u n d e r th e c h a n g e d c o n d itio n s —is o ne e x c u sin g the a c tio n of th e G re e k s in a w ay w hich th ro w s a c e r tain now lig h t on w h at, to ju d g e fro m the ta le s of th e c h r o n iclers, h a d b een a b a re fa c e d b re a c h of fa ith b y S ym eon. F o r it is c le a r from th is le tte r th a t th e h o s tilitie s h a d a c tu a lly b e e n com m enced b y th e G re e k s, on th e p le a th a t th e B u lg a rs w ere p r e p a r in g th e m se lv e s to ta k e th e o ffen siv e, (ex actly a sim ila r situ a tio n , in fact, to th a t w hich led to th e o u tb re a k o f the S e cond B a lk a n W a r in 1913, ju s t a th o u s a n d y e a rs la te r). T he P a tria rc h fra n k ly a d m its t h a t his p e o p le h a d been a t fault, a n d an a rm y h a d b een collected to m arc h a g a in s t B ui g a ria , a lth o u g h w ith o u t N ic h o la s’ k n o w le d g e 5) B ut th is h a d been d o n e b y th e m ilita ry c o m m a n d e rs in M acedonia a n d T h race, on th e p le a th a t S y m eo n w as a b o u t to b r e a k the T re a ty . «A nd th e a d d e d f u rth e r, (you k n o w B oga, w ho h a s been a p p o in te d C o m m a n d e r in C h e rso n . T h is C o m m a n d e r in C h e rso n n e v e r s to p s a s s u r in g us t h a t th e B u lg a rs a re m a k in g e v e ry e ffo rt to g e t h o ld of the P e tc h e n e g s a n d a n y o th e r n a tio n s in th o se p a r ts fo r a ra id a n d w a r a g a in s t th e G r e e k s . . . A nd h e s a y s t h a t a b o u t six te e n e m b a s s ie s h a v e com e from th e P e tc h e n e g s u p to d a te , s a y in g t h a t e m b a s s ie s h a v e b e e n s e n t them from th e B u lg a rs , n o t once o r tw ice b u t m a n y tim es, a s k in g them to u n ite w ith t h e m ; a n d th e B u lg a rs a re so keen on th is th a t th e y w ish to c o n tra c t a m a rria g e w ith a ch ild of th e ir ow n ro y a l fam ily a n d se al th e a llia n c e b e tw e e n th e P e tc h e n e g s a n d th e m s e lv e s )». D isq u ie te d by th is, the g e n e ra ls h a d a tte m p ted to seal th e p e a ce b e tw e e n G reece a n d B u lg a ria b y fre sh c o m p a c ts ; a n d o n ly w h en this fa ile d h a d th e y g o t to g e th e r a d e fe n siv e a rm y . I t w as th e s e re p o rts , th e P a tr ia r c h w e n t on, w hich h a d im pelled him to w rite to S y m eo n « b e fo re th e o u tb r e a k of th is la') Ibid. col. 57. !) Ibid col. 71 e v e x a xfjc ngoc Boul/yapou; exotpaxeia? f| xoaautrj xivr^oi?.

y £Yov£

29 168

m e n ta b le w ar*. H e r e p e a ts th a t b o th th e G re e k m o b ilisa tio n a n d th e e m b a ssy to th e P e tc h e n e g s h a d b een p u re ly d e fe n s iv e m e a su re s. «I k n o w w h a t y o u will s a y , a n d rig h tly * , th e e m b a rra s s e d P a tr ia r c h g o es on. «If th e y w ere on ly m a k in g a d e m o n s tra tio n , w h y d id th e y in v a d e B u lg a ria n soil a n d co m m it h o stile acts b e fo re th e o u tb re a k of w ar* ? B ut e v e n h e re th e ir a c tio n co u ld b e d e fe n d e d , etc., etc. N one of th e P a tria rc h ’s f u r th e r le tte r s , w hich c o n tin u e o v e r p ra c tic a lly th e w hole p e rio d of th e w ar, m a k e a n y f u r th e r d i re c t re fe re n c e to P e tc h e n e g p a rtic ip a tio n , In te re s tin g , h o w e v e r, is th e e m p h a sis w hich N icholas lays, n o t on ly o n th e h o r r o r s of f ra tric id a l w a r b e tw e e n tw o C h ristia n n a tio n s , b u t on the evil c o n seq u e n c e s for b o th of th em if e ith e r sh o u ld s u ffe r too h e a v i l y '). O n ly a fte r p e a ce n e g o tia tio n s h a d a lr e a d y b e e n be g u n , a n d w ere n o t p ro c e e d in g too well, N ich o las u rg e s S y m eo n to b e re a s o n a b le , on th e g ro u n d th a t th e E m p e r o r 2) e ith e r h a d a rr a n g e d , o r w as a b o u t to a rra n g e , a c o a litio n of R u s s ia n s , P e tc h e n e g s, A lans a n d M a g y a rs a g a in s t B u l g a r i a ’). W e th e r (as seem s p ro b a b le ) th is w as m ere b lu ff o r n o t 4), n o th in g m o re cam e of the sc h e m e ; a n d n o t v e ry lo n g a fte rw a rd s , p e a c e w as sig n e d . T h e P a tria rc h ’s le tte rs p ro v e th a t P h o c a s ’ n e g o tia tio n s w ith th e P e tc h e n e g s m u st a c tu a lly h a v e b e g u n n o t la te r th a n 914, a n d Z o n a r a s 5) is in c o rre c t in p la c in g th em in 917— u n less, i n d eed , his eordXr) is to b e ta k e n in th e s e n s e of a p lu p e rfe c t. B ut m o re in te re s tin g still is the P a tr ia r c h ’s a d m issio n th a t th e G reek fo rc e s h a d co m m itted a n a c t of w a r a g a in s t B u lg a ria , in tim e of peace. ') Ibid. col. 110, etc. *) i. e. Lecapenus. This letter was then w ritten after 910. ’) op. cit. Col. 152. The M agyars are described here as oi lx tijs Auaew; Toupxoi; w here the story is repeated (col. 153> as ToOpxoi sim ply. J) An attem p t was, how ever, actually m ade to arrange an alliance with the M agyars. Constantine tells us (D. A. I. ch. 32) th a t d u rin g the latter p a rt of the w ar Michael, Prince of Zachlum ia, told Sym eon that P e te r of Serbia had been persuaded by the G reek E m p e ro r to attack Hulgaria, in collaboration with the M agyars. The attack, how ever, does not ap pear to have come off. ") Z onaras III 464 ed. Bonn.

30 169

The c o in cid en ce o f th is actio n w ith th a t a s c rib e d b y M as’ u d i to th e g a rris o n of V a la n d a r is m o re th a n r e m a rk a b le , a n d it is v e ry d ifficu lt to Im a g in e t h a t b o th p a s s a g e s do n o t re fe r to th e sam e ev en t. I t co u ld h a rd ly h a v e o c c u rre d d u r in g the lo n g a n d p e a ce fu l re ig n of S y m e o n ’s su c ce sso r, P e te r n o r is su ch a ra id lik e ly to h a v e been u n d e r ta k e n in a tim e of re a l p eace b e tw e en G reece and B u lg a ria . If, h o w e v e n , S y m e o n w as still th re a te n in g th e G reeks, b u t d iv e r te d by a d is p u te w ith th e P e tch e n e g s, this w o uld g iv e the c o m m a n d e rs on th e f ro n tie r an o p p o rtu n ity w hich th e y could n o t re s is t. If. then, we ta k e M as’u d i’s s to r y a s r e f e r r in g to th e e v e n ts of 913 —4, we h a v e a n e x p la n a tio n fo r f u r th e r o b s c u ritie s in th e G reek n a rra tiv e . F o r w hy, if B o g as w e n t to C h e rso n as e a rly as 914, did his a llia n c e w ith th e P e tc h e n e g s b e a r fru it o n ly in 917? It is re a s o n a b le to su p p o se th a t M as’u d i’s n a r r a ti v e re a lly fills a g a p here. The P e tc h e n e g s w ere a c tu a lly o n b a d te rm s w ith the B u lg a rs o v e r som e co m m ercial q u e s tio n . B o g as w as s e n t off h u r rie d ly to m ak e m is c h ie f ; b u t the tw o n a tio n s p a tc h e d u p th e ir q u a rr e l, on th e p re c ip ita te actio n of th e g a rris o n of V a la n d a r, a n d the n e g o tia tio n s th e r e fo r e cam e to n o th in g a t th e tim e. T h e R u s s ia n C hronicle, it m a y b e r e m a r k e d , seem s to sh o w th e P e tc h e n e g s p a s s in g th ro u g h th e S te p p e in 9 1 5 ;') th e c h ro n o lo g y m ay b e a t fa u lt, b u t it c a n n o t h a v e b e e n ta k e n fro m G re e k so u rc e s, w hich W ould h a v e p u t the d a te a t 917. T h e a p p lic a tio n fo r fre e p a s s a g e th ro u g h R u s s ia n te r r ito ry will th e n h a v e b e e n m a d e in o r d e r to m a k e w a r a g a in s t th e G reks, a n d not, as g e n e ra lly a ssu m e d , on b e h a lf of th em . A fter this, th e r e la tio n s b e tw e e n B u lg a rs a n d P e tc h e n e g s w ere frie n d ly fo r a tim e, a n d th e re w as ta lk of a r o y a l m a r ria g e . It w as th re e y e a rs b e fo re su fficie n t fre s h d isu n io n h a d b een c re a te d to allo w of a G reco —P e tc h e n e g a llia n c e , a n d ev e n th en , it w as in e ffe c tu a l. T he n a m e V a la n d a r, V a la n d ria n s , re m a in s in e x p lic a b le to ‘i C hronique ditc de N estor, ed. P aris I 51. «ln 915 the Petchenegs •appear fo r the first time in R ussia ; they m ake peace with Ig o r, and gain th e banks of the Danube In 916—20 Ig o r m akes w ar on the Petchenegs*.

31 170

me. M a rq u a rt s u g g e sts t h a t V a la n d a r m a y h a v e b e e u an old B u lg a ria n n am e fo r D e v e lto s. I p r e f e r to th in k a P e tc h e n e g nam e, in view of the p r o b a b le so u rc e of th e s t o r y ; b u t th e id ea is a fru itfu l one. I t is, h o w e v e r, in c o n c e iv a b le th a t th e G re e k s could re a lly h a v e te rm e d th e P e tc h e n e g s , B u lg a rs a n d M a g y a rs «V alan d rian s» . If, h o w e v e r, th e s to r y of V a la n d e r cam e to M as'udi from a P e tc h e n e g s o u rc e , it d o e s seem p o ssib le th a t th o se P e tc h e n e g s (fo r th e w hole n a tie n w as s u re ly n o t e n g a g e d ) w ho took p a rt in th e a tta c k on th e f o r t r e s s m ig h t h a v e b o rn e its n a m e fo r a w hile a s a s o rt of b a t t l e —h o n o u r. O r is it p o ssib le to c o n n e ct th e n a m e w ith th a t of a n y of th e P e tc h e n e g hordes, e. g. th e BeXenapvi? m e n tio n e d b y C e d r e n u s ?

IV On the Black Bulgars

A nation of «Black Bulgars* is m entioned under that name only four times in the whole of m ediaeval lite r a tu r e ; twice in the oldest R ussian Chronicle and twice in the De A dm inistrando Im perio of Constantine P orphyrogenetos. The letter (the au th en ticity of which is, however, much disputed) alleged to have been sent from the R abbi Chisdai to the Khagan of the Khazars, has a reference to Bulgars which is often taken in connection witli the Black Bulgars. The term is not found before the ninth, or after the tenth century A. D. The references are as follows: R u ssia n Chronicle. A single version of this Chronicle re ports under the date 864 th at «Black Bulgars slew a son of Askold in b a ttle * ,). A second reference occurs in the treaty concluded in 945 between Prince Igor of Kiev and the Byzantine Empire. In this notable document, the R ussian prince prom ises in the future

sto Sanctorum Gerua.sii et Protasii feria tertia,

Post hunc autem usurpavit sibi coronam Stephanus filius Bele ceci, de quo supra, per menses quinque.

Post istum Stephanus frater suus coronam usurpat mensibus V et diebus V

tandemque devincitur

-ubi multi nobiles Hun- In quo praelio plures regni nobiles occiden-garie corruerunt. tur,

107 42 K

B

W

Post hec expulsus est idem vero de regno et expulsus de regno. expulsus, demum venit Zemphlyn in Zemlu, Obiit in castro Zem- ubi et len A. D. MCXXIII suam. tertio Idus Aprilis, feria quinta. Cuius corpus quiescit.

finivit

Albe Albae requiescit

obiit

in

vitam A. D. MCLXXIII tertio Idus Aprilis

cuius corpus quiescit

A lbe

Idem vero rex Stephanus filius Geyse migravit ad Dominum A. D. MCLXXIII, quarto Nonas Martii, feria prima. Cuius corpus Strigonii quiescit. Postea regnavit Bela frater eius qui fures et latrones persecutus est et petitionibus loqui traxit originem, ut Romana habet curia et imperii.

Sed post hunc regna- Post hunc regnavit vit Bela Graecus, quern Bela tertius filius GeyBecha et Gregor apud sa, de quo supra. imperatorem Graecorum diutis tenuerunt His quidem fures et latrones persecutus est petitionibusque loqui traxit originem, ut Romana habet curia et imperii.

annis XXIII, mense I, Qui coronatus est Idibus Januarii feria pridiebus XIX. Obiit auma. Regnavit autem tem A. D. MCXC nono annis XXIII, mense Kalendas Maii, uno, diebus XIX. Obdormivit autem in D omino anno eiusdem MCXCDC Kalendas Maii, feria tertia. corpus Albo Cuius corpus in Au- Albae jacet tumulatus cuius quiescit. burn ecclesia tumulatur.

108 43

B

K

f

W Hie reliquit filios; quatuor Primus fuit dux Henricus, qui successit ei in regno; secundus fuit dux Andreas, postea rex, pater regis Bele quarti; tertius fuit dux Salomon et quartus fuit dux Stephanus

B has one factgiven by neither ,W nor Z: the name and qualities ofEmeries wife. Z givesthe date of Ladislaus death as May 2 1211; W and B agree an May 7, 1201.

Huic successit Andreas Sed post hunc regna- Post hoc regnavit rex filius Bele tertii vit Andreas Andreas, filius regis Bele tertii (ut supraf (see below) — XXX, mensibus tribus diebus XXVI. qui Andreas coronatus est vicesimo septimo die post obitum regis Ladislai, quarto Kalendas Junii in Penthecostes.

_

__

Cuius uxor fuit domina Gertrudis de Alamania, de qua genuit Bel&m, Colomannum, Andream et Beatam Elyzabeth.



(See below)

(Story of Bank)



Post hec Andreas Terram Sanctam visitavit ad mandatum pape. . . Et ibi in Terra Saneta super exercitum Christianorum contra soldanum Babilonie

rex potens et illustris. Iste etiam Terram Sanctam visitavit, ubi per omnes principes Christianorum capitaneus ordinatur et exercitum soldani Ba-



109 44

B

K

capitaneus et dux preficitur et mox victor •efficitur gloriosus. Mansit autem tribus mensibus.

biloniae cum Hungaris et Zaculis effugavit et honore multiplici cum gente sua per Assirios et alias nationes praevenitur indeque cum summa gloria revertitur in Hungariam.

W

Hie multa meritoria opera exercuit in edificandis et dotandis ecclesiis secularibus et regularibus

tFollows a long paragraph on presentations •of holy relics made by Andrew to various churches)

et introducendis religiosis diversis victoriaque habita ad sepulchrum Domini. (Marriage of St. Elisabeth) Migravit autem rex Andreas ad Dominum A . D. MCCXXXV tricesimo -sui

anno

regni

O biitA . D. MCCXXXV, undecimo Kalendas Octobris (see above)

Cuius corpus in monasterio de Egrus feliciter requiescit.

Cuius corpus in monasterio suo Egres requiescit

( s e e above)

Hie reliquit filios tres. Quorum primus fuit dux Bela, tandem rex, qui ei successit. Secundus dux Colomanus. Tertius dux Andreas (et unam filiam, beatam scilicet Elizabeth: not in Z)

110 45 B

K

W

Rex Bela post eum filius eius coronatus est pridie Idus Octobris feria prima, qua cantatur, etc.

Post hunc autem regnavit Bela filius eius, apud Fratres Minores Strigonii tumulatur.

Huic Andree regi suecessit filius eius rex Bela quartus et regnavit annis XXXV, m ensibus septem.

(Description mony)

of

cere-

Tempore autem istius Istius quidem in die- Huius tempore, Bele regis A. D. bus MCCXL anno scilicet MCLI Mangali sive

Mond Lisviae

Tartari

Tartari

cum quinquies nis armatorum

Tartari

cente- de tribus partibus re- cum multitudine gni in Hungariam piosa adeunt cum quinquies centenis millibus armatorum

co-

habentes adhuc centuriones et decuriones ad milia XL. regnum Hungariae invaserunt (See below, Hungaria ideo, etc.) Contra quos Bela rex iuxta flumen Seo prelians vincitur, in quo prelio fere extinguitur militia regni Hungarie universa.

regnum Hungariae in^ vaserunt et flebiliter devasterunt. Quibus in Soio rex praefatus contraveniens a Mong Lis devincitur A. D. MCCVLI. Ubi fere tota regni militia est deleta

Ipso vero Bela rege ipso Bela coram ad mare fugam faciente ad mare fugiente Tartari usque ibi ipsum crudelitur insecuntur.

eis

Ill

46 Manserunt enim ipsi Tartari in regno Hungarie tribus annis. Hungari, ideo multo plures post exitum illorum fame perierunt, etc. Post hec autem rex Bela reversus est de maritimis partibus et ducem Austriae Fredericum virum bellicosum ante Novam Civitatem gens occidit in prelio Hungarorum et transfixit per maxillam, etc.

Manserunt enim ipsl Tartari tribus annis continuis in regno (See above, et flebiliter devasterunt.)

Quo quidem de mari revertente per ducem Fridericum de Austria bello impetitur. Quern ante Civitatem Novam Hungari cum lancea in maxilla transfixum peremerunt

Obiit autem A. D. MCCXXV Nonas Maii feria Vl-a, in fesio Inventionis Sancte Crucis in insula Budensi et sepultum est

Obiit autem A. D. MCCLXXV Non. Maii.

corpus eius Strigonii in ecclesia Fratrum Minorum constructa, etc.

Cuius corpus requie scit in ecclesia Fratrum Minorum Strigonii.

Vir virtutibus plenus, etc.

Hie habuit duos filios. Primus fuit dux Stephanus, postea rex. Secundus fuit dux Bela. Post ipsum A. D. 1270 Postea regnavit Ste- Post hec regnavit rex cepit regnare iilius phanus rex filius ejus Stephanus -ejus Stephanus super totam Hungarian!

112 47 (Sec below)



de quo supra annis 2, mensibus 3, diebus 5

— Qui

Ottocarem,



Hie fuit in wultu austerus

etc. Qui Boemie regem etc.



Praeterea Budin civi- Iste etiam civitatem tatem Bulgarorum Budyn suo dominio subjugavit



(Campaign against Ot- (Campaign against Ottocar of Bohemia) tocar)

■et Bulgaros superans



Bulgaris in praelio superatis (not in Z)

regem eorum conpulit dominumque Bulgaro- ducem eorum sibi subsibi deservire rum eo vivente sibi jugavit

compulit deservire

Regnavit autem duobus annis



(Not in Z)

(See above)

et mortuus est in anno Migrans tandem ex hoc et obiit A. D. 1278 id. tertio regni sui in ma- seculo Aug. gna insula et in insula vocata beate Virginis in coenobio monialium requiescit cumulatus

et sepultus est in ecclesia Beate Virginis in insula Budensi in loco Bulgariam.

cujus corpus requiescit in insula Budensi in ecclesia Beate virginis (in loco Beginarum.)

This comparison is instructive. A certain common substratum of course exists, due to the fact that all the texts are constructed in the same way; as lists of kings, padded out with facts. We must necessarily assume that these lists will to some extent coincide. When, however, we turn to anything, in any text, beyond this bare minimum for which we must allow, we find the following result:

113 48 1. 2. 3. 4. 5.

Some few passages peculiar to B alone One or two common to Z and W only Several common to W and B only Several common to B and K only Only one, and that a very famous one (the making of Bulgaria tributary) found in both W and K.

Generally speaking, B's text equals precisely that of K plus, that of the part of W which is independent of Z. Thus for Coloman, K has his nickname and his campaigns; W, that he was bishop of Varad; B, all three. For Andrew IL K gives his pilgrimage, W his pious acts and place of burial, R the two- For Bela IV, K gives the numbers of the Tatars, the battle on the Sayo, Bela's flight, etc: W gives the one other detail that they stopped in Hungary for 3 years. B has exactly the two combined. So accurate is this addition sum that we find for Stephen V that B gives with K Stephen's capture of the Vidin, and with W that of the „Bulgars’\ although immediately after, with both, sources, he gives the subjugation of the Bulgarian king. Hence we may draw the following conclusions: ,W cannot have been summarising from B; for it is beyond* reason to suppose that he would have excerpted precisely those facts which K had not thought fit to mention. Neither can K have been summarising from B, since he would, not have omitted precisely what W had recorded. Therefore B's text has been composed by combining a text practically identical with K's with some other source connected with W. This cannot have been W in its present form, since W has certain passages common to Z which B has not taken over. Moreover, even where B and W agree in sense, B is often the fuller. Therefore there must exist another text, lost to us, and used independently by B and by W. B combined it with the archetype copied out also by K; ,W combined it (in abbreviated form); with Z. If we call this text *,W, we get the following genealogy

114 49

Other material

Archetype = K

!_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 1

l

B

*W

Z

1 1_ _ _ _ I I

W

As to the exact form of *W, we can only guess, but guess with considerable confidence, that it was similar to that of Z, of the Knauz and Toldy Codices, viz: a list of kings, padded out with a few details, and going back to the time of the Conquest. Broadly speaking, wherever W differs from Z, this is due to the influence of *W. Let us now apply these conclusion to the other passages where B's text may have originated in similar fashion. I begin10 with the opening of K 43, B 63, to which I add for comparison the corresponding passage in W: — K

B W Porro Toxun genuit Iiic (sc. Toxun) habuit Geycham et Michae- duos filios, quorum lem, prius fuit dux Geysa, pater be&ti Stephani regis, Secundua vero fuit Michael. Dux Michael vero genuit i$te Michael habuit Calvum Ladizlaum et duos filios, ducem viVazul, delicet Wazul et ducem Ladislaum Calvum. Iste dux Vazul habuit tres filios. Hofum primus fuit dux Andreas, postea rex. Secundus fuit dux Bela, demum rex, ut infra dicetur. Tertius fuit dux Leuente. Dux autem Ladislaus Calvus, de quo supra, habuit filium, Bonuzlo vocabatur. Modo re-

10 In one earlier passage the date of the Conquest (K 25, B 26) B may have altered K‘s date under the influence of W.

115 50 Anno vero dominicae incarnationis 967 Geicha dux divino praemonitus oraculo genuit sanctum regem Stephanum; Michael vero, frater Geichae genuit Wazul et Zar Ladizlaum.

Gcycha vero divino pracmonitus oraculo anno Dominicae incarnationis 969,

Quemadmodum in legenda Beati Stephani scriptum est, genuit Sanctum Stephanum regem ex Sarolth filia Gyula.

deamus. Dux Geysa, filius Toxu dc quo supra, anno ab incarnatione Dei 963 nuto divino cepit cogitare de ritibus paganismis destruendis et de cultu divino ampliando, seque fecit baptizari et tandem in hujusmodi sancto proposito mortuus est, prout plenius hec in legenda beati regis Stephani filii sui continentur, qui scilicet sanctus rex Stephanus natus est anno Domini 969.

This is a particularly fine example of interpolation. Not only has B put in from *W the little line ffporro Toxun genuit Geycham et Michaelem", but he most also have altered the date of St. Stephen's birth under the influence of the same source, as our comparison of the text shows. For in fact, no version of the Legend of St. Stephen gives the date of his birth at all; much less does it place it at 969. W does not ascribe that statement to the Legend; he gives the date independently, while referring to the Legend (quite correctly) for a description of Geza’s death. B. however, has telescoped W’s text for the purposes of interpolation, thus arriving at his erroneous statement. In the passages relating to the death and burial of St. Stephen, B’s text again exactly equals K-j-W. B70

K45

W

Sepultus est autem in Sepultus est Albae in Cuius corpus Albe in basilica Albensi ecclesia Beate Virginis ecclesia maiori venegloriosae ratur, quam ipse fundavit et uberrime dotavit

116 51 quam ipse in honorem sanctissime genetricis Dei semper virginis Marie construxerat, ubi multa signa et miracula fiunt intervenientibus m entis eiusdem sanctissimi regis Stephani ad laudem et gloriam Domini nostri Jesu Christi, qui est benedictus in secula seculorum Amen. Con- Confestim igitur Huniestim quoque totius garorum citara in lucythara Hungariae ver- ctum est conversa, etc. sus est in luctum, etc.

Hie quantis a Deo gratiis fuit insignitus et quanta opera exercuerit meritoria vide in sua legenda.

Immediately after, we have another very fine instance. K calls Peter „Petrum Venetum filium sororis suae'*. W has „Petrus Theutonicus, frater Resla Regine consortis Stephani, de sorore sancti Stephani genitus". B has „Petrus Alamanum, vel potius Venetum, fratrem regine*', and afterwards works out a complicated genealogy justifying W's statement, as follows:

1)

Gertrude = William — 2 ) Stephen’s (“sister of (Brother of Sigismund sister theEmperor“) of Burgundy)

Stephen — Gisella

Stephen = Gisella Peter Incidentally, this is incorrect; Peter’s father was the Doge Otto Orseolo.11 The Life of St. Stephen makes Peter Stephen's sister’s son, and some have supposed this variant in B’s text to have come from his trying to reconcile the information in the Vita and the archetype used also by K .12 There is, however, no evidence that B used the Vita, in any of its forms, at all. The only passages in his text 11 H 6 m a n, Magyar Tortenet I, 242. 12 Legenda Major c. 15; Hartvic c. 21.

117 52 which indicate such use are those which we have quoted and in the first two of these, as we saw, he certainly worked, not directly from the Vita, but through the intermediary of *W. This is equally certainly the case h6re. The epithet „AlamanumMcomes not from the Vita (which in no way suggests a German origin for Peter) but from *W, and B characteristically combines what he found in the two sources into the comprehensive but not very intelligent phrase „Alamanum vel potius Venetum”. It is *W which has imported from the Vita the alternative relationship, which B tries in so complex a fashion to justify.

A small point arises in connection with Andrew I. B and W both mention that Tihany abbey is ,,juxta lacum Balaton’, which K omits. Similarly, it is B and W only who refer the reader to the Legend (B 140, gesta) of St. Ladislaus. In these passages it is, however, difficult to show the composition of B as clearly as we have done. The passages already quoted should, however, suffice to prove our thesis: to wit, that there once existed the source, now lost, which we have termed *W, which on the one hand was combined with Z to produce the present Varad Chronicle and on the other hand, was unknown to K6 zai, but was used by the author of B as one of his sources. Another of his sources was the archetype used by K, and that, in the passages which we have considered, must closely have resembled K*s present text; since in most of them, B's text consists of little more than a combination of K’s and W ’s. In other passages the divergence is greater. With these I do not deal in the present essay; but I venture to maintain that the proof here adduced that B has, in one important case, interpolated a text resembling K's, should allow a reasonable presumption that he does so in other cases also. In other words, it is probably that in many cases (not necessarily in all), the common archetype of K and B resembled K much more closely than B.

IX The Relations between the Narrative Chronicles and other Historical Texts

he author wishes to express his most sincere and grateful thanks to the many Hungarian friends who have helped him with advice on difficult points which have arisen in connection with the problems discussed in these pages; to the printers, who have wrestled valiantly with a difficult handwriting in a foreign tongue; and above all, to his friend, Professor Imre Lukinich, of the Royal Hungarian Academy of Sciences, without whose never-wearying help this work could not have appeared in its present form. Oxford, February 1940.

119

FOREWORD A foreigner should perhaps make some apology for venturing into a field which so long and 6 0 naturally has remained the almost exclusive preserve of Hungarian scholars. It was, indeed the extreme shortage of critical literature in any other language than Magyar on the Hungarian Chronicles 1 that first induced the present writer to commence his task, with no more presumptuous an aim than that of interpreting to a wider public the results achieved by such scholars as Homan, Domanovszky, Pais and Madzsar — results which, owing to the language difficulty, are not so widely known outside Hungary as they should be. Only gradually, as his work proceeded, did he find himself forced to modify its character under pressure of an inescapable conviction that brilliant as arc the results achieved by these scholars, not all of them can be regarded as final. From an interpretation pure and simple, his work thus became in part an argument directed against not all, but part of those very views which he had originally thought to make more widely known. He wishes, however, to emphasise the obvious fact that even where he lias had occasion to disagree with his predecessors, yet he has done so in full consciousness of the fact that without their scholarship and acumen he could not even have begun his own work. The ultimate object of this essay is to discover what are the various elements of which the Hungarian national Chronicles are composed. This involves, in the first instance, consideration of the mutual relationships (where such exist) between the different 1 The latest at all detailed work on the subject by a non-Huntfarhn — Kaindl’s "Studicn zu den ungarischen G eschichtsquellen” — is already more than a generation old. This work, to which frequent reference will be made under the title “Kaindl, Studien”, appeared in the Archiv fur oe. Gcschiclile 1894 — 1902.

120 2 texts, and particularly, of course, the longer texts, which contain many components. And here one preliminary word of warning must be uttered at the very start. The fact that all our more important texts or groups of texts contain a considerable amount of identical material, and that other writers, quoting phrases which appear to be drawn from this common store-house, give as their source the “Gesta Hungarorum”, has given rise to the very natural supposition that all our main historical texts arc based on a lost “Gesta Hungarorum”. The pursuit of this “Gesta ’, the determination of its date and its exact contents, have become a main preoccupation of Hungarian scholars engaged in this field; in the course whereof they have tended to lose the ground from under their feet. The leader in this quest has been Professor Homan, the most distinguished of Hungarian mediaevalists, who in 1925 published on the subject a study" which unites astounding learning with unexampled acumen. Working, broadly speaking, on the principle of the highest common factor, he compared all relevant texts; extracted from them all material found in more than one; discarded all not common to more than one; and compiled of the residue an only original “Gesta Hungarorum”, which he presented as the common ancestor of all our existing texts. According to this theory, the “Gesta” was compiled in 1092. An extract was made from it almost immediately, and distributed to the monasteries founded by St. Ladislaus; from this extract derive, indirectly, the shorter narrative chronicles (the Zagrab, Varad, Knauz, etc. codices). The whole Gesta was brought up to date in 1127, and on this version the Anonymous Notary worked, padding from foreign sources and from “oral tradition of the Upper Theiss district”. Further additions were made afterwards to the original text, and in the 13th century another writer (Kezai) took this work, turned the early part of it into the Hun Chronicle by the process of substituting the name of Hun for that of Hungarian, and padding with matter genuinely relating to Huns but gathered from book sources, while abbreviating correspondingly the beginnings of the Hungarian history. The results of this work are to be found in the so-called Codex Sambucus * B. Homan. A Szent Laszlo-kori Gesta Ungarorum es XII—XIII. szazadi lcszarmazoi (Bp. 1925; quoted hereafter as Human, Gesta).

121 3 of B, which is really K’s own work; the texts passing under K’s name being only extracts therefrom made by a later scribe. Of all the native texts relating to Hungarian history, the only ones to which Homan allows any originality are some of the minor Lives of the Saints; but the Legend of St. Ladislaus, the historically important passages of the Vita S. Gerardi, and probably parts of the Legenda Major of St. Stephen are, in his view, all derived directly from the 1092 “Gesta” ; and it is to this work that the foreign and other sources which quote a “Gesta Hungarorum” also refer/ One or two later scholars have suggested certain minor modifications of this theory. Thus M adzsar 4 assigns to the “Gesta” a date about a generation later than that given by Homan. Doma* novszky, as we shall see, queries his view as to the relationship between K’s work and the Codex Sambucus. But no one has questioned the essentials of the theory: the existence of a single “Gesta”, comprising all the most important of the material, up to a certain date, upon which our existing texts afterwards drew, And yet they can and must be questioned. That a text or texts existed bearing the name of “Gesta” on which many of the later texts drew, is one thing; that there was a ever a single “Gesta”, on the scale imagined by Homan, is quite another. The principle of the highest common factor is one which ought to be used with great caution in making deductions as to the nature and extent of lost works. If two texts, a and b, contain common passages, and neither can have borrowed from the other, we can, and indeed must, assume a common source or sources from which both derive. But this need not be a single, all-embracing x. In certain circumstances, the hypothesis of two mutually independent sources, x and y, is equally plausible. Translated into concrete terms; if both Anon, and Kezai contain identical descriptions of Scythia, and also identical passages on St. Stephen, it does not follow that they both drew on a common source containing both these things; they may both have used a description of Scythia and of a — quite separate — Life of St. Stephen. We are on still quaggier ground when we have to deal with 3 texts, a, b and c, of which a and b contain certain common 3 The foreign sources which quote or appear to draw on this material ;are Odo de D eogilo, Godfrey of Viterbo, A lbericus Trium Fontium, ThomaS of Spalato, the Polish—Hungarian Chronicle and the Cracow Anonymus. 4 I. Madzsar, a II. Geza-kori nevtelen (Bp. 1920; quoted henceforward as Madzsar, Nevtelen).

122 4 information, clearly from a common source x, and b and c further common information not in a. In such cases it is highly dangerous to assume that x contained also the information common to b and c. It may have done so; and if a is clearly addicted to epitomising and to omissions, the assumption may not be unreasonable. But where it is equally possible, and analysis indicates, that b interpolates more than a omits, it is much more likely that we have to deal here with two independent originaL sources, rather than with one only. Apart from being logically unnecessary, the hypothesis of a single, all-embracing "Gesta” is a priori highly improbable. .When one considers that the story of the Magyars* early wanderings was (on Homan's own hypothesis) long preserved only by means of oral tradition, the unlikelihood at once becomes patent that a single writer should have arisen who gathered all the surviving traditions into a single Authorised Version — the one and only Gesta — before which all else gave way. It is not only a priori improbable; it is certainly incorrect. The pages of K and B themselves, and for that matter of Anon.r are full of references to alternative traditions and writings. K and B, for evample, explicitly refer to and declare to be false a tradition that Andrew, B6 la and Levente were the sons of Wazul.5 The Zagrab and Varad Chronicles, however, contain that very tradition. K 23 and 26 refers to and in part quotes an alternative tradition of the Conquest: that Marot and not Svatopluk was ruling in Hungary when the Magyars entered (re-entered) it; Anon, gives one Menumorout as ruling there, and does not mention Sviatopluk. K 40 rejects a legend concerning the death of Lei which B 60 gives. K 42 and B 62 record without qualms a legend which Anon, rejects on the explicit ground that he "has not found it in any written source", but knows it only as a folk-legend. Homan, it appears, regards the variants in the different texts as having all arisen after the composition of the Gesta; but tjhis is quite an arbitrary supposition. In the case, for example, of the shorter narrative Chronicles, which contain little more than the most elementary facts of princely genealogies and burialplaces, it is at least equally probable that several different versions should have been committed to paper independently of one another. We cannot suppose that the knowledge of these facts would be confined to a single quarter. Similarly, two versions of • K. 55; B. 86.

123 5 a popular legend, such as that of Achtum, or two different lists of the Seven Heroes who led the Magyars into Hungary, may perfectly well have come down through the generations, without passing through the bottle-neck of a single official transcription.

As to the long argument from the work of Albericus Trium Fontium with which Homan begins his “Gesta", and on which he lays so much stress: one may accept with gratitude his brilliantly reasoned reconstruction of Alb.'s later information, and of the “Somogyvar" list of kings; but as regards the older notes used by Alb., Homan proves no more than that Alb. used another short text like the various shorter narrative Chronicles. This resembles Z, Kn., etc., no more and no less than they resemble and differ from each other, and its existence affords no sort of additional proof of derivation from a single “Gesta", much less a “Gesta" in the form which Homan conceives it to have taken, i. e., a text .substantially identical with the Buda Chronicle up to 1092; and it does not at all suggest that there was any sort of break in the material precisely in St. Ladislaus’ reign. For in fact, the few details in it over and above the list of kings pure and simple end, not in 1091, but with the election of Aba. Homan himself admits 6 that “the influence of the Chronicle can only be recognised beyond question for the period up to 1041.“ “Only detailed examination,' he writes, “can decide the origin of the information for the following period.’’ But we never get this examination. We are told 7 that the reason why none of the fuller information appears in Alb. is that his source was “an extract/' But this only carries us one step back. Why did the extract confine itself to a list of kings for this interesting period, while giving fuller information for the preceding years ? 8 It has been necessary to dwell on these points, because the authority with which Homan's great learning invests him is such that practically all Hungarian scholars accept his hypotheses, particularly that of the “Gesta", and all their subsequent work is conditioned by this fact. They thus become subconsciously unwilling to accept the conclusions to which an unprejudiced examination of the texts themselves would inevitably lead them.

6 Homan, Gesta, p. 13. 7 Id. p. 29. 8 Kaimdl, Studien VII. pp. 8 (436) ff. works out a very similar argument, which also (although on quite different grounds) shows a break in the Hungarian tradition at the end of the 11th century; but his reasoning is no more cogent than H6cnan,s.

124 It is to such an examination that we must now turn, eschewing the easy method of a priori assumption and undertaking rather the laborious task of examining the different texts in detail. From the historical point of view, the central problem is clearly that of the relationship between the various members of the group of longer narrative Chronicles, viz; K6 zai (K), the various texts of the Chronicon Budense (B) and the Chronicon Pictum Vindobonense (V) . 9 I propose therefore to begin with this group, and among them, first with the relations between the longer texts, V and B.

V and B. As is well known, these two groups of Chronicles are practically identical, except that V contains a large amount of additional material. Domanovszky's studies10 have established the relationship between them beyond any reasonable doubt. The author of V had before him a text of B, which he used as his main 9 The details regarding the different MSS, and the mutual relationships^ betw een the MSS of each text, are admirably set out by Domanovszky in his introductions to the several texts in Scriptores Rerum Hungaricarum (Scr. R. H.) Vol. I. I shall not consider here the “Chronicon Posoniense” or the “Chronicon Monacense" printed in volume II. of the Scr. R. H., since although each, particularly the former, contains a few private additions, essentially they are no more than copies (in places, mere extracts) of B and V respectively, unimportant for the purposes of this enquiry. 19 Particularily A Budai Kronika and Die Interpolation der wiener ungarischen Bilderkronik (Ungarische Rundschau, 1908.); quoted henceforward as Dom., B. K. and Dom., Interpolatkmen respectively.

125 7 source; but also adding to it, and in certain respects altering it. The prologue is obviously his own, and was composed by him, as he says himself, in the year 1358. Here he expressly says that he used and checked many different sources.10* The Biblical introduction which follows (cc3, 4) was composed independently by V on the basis of a number of sources which he himself names: the Bible, Josephus, St. Jerome, and a certain St. Sigilbertus, Bishop of Antioch, author of a Chronicle of Oriental Nations, of whom, and of his work, unhappily, nothing more is known. At the same time, V must have had before him a text closely resembling that of B and K (which are identical at this point), since he polemises at some length against the theory, which BK maintain, that Nemroth, father of Hunor and Magor, was the son of Tana. In fact, however, he repeats this very genealogy in c28.10b After this point, V seldom leaves out or alters anything in B, but he supplements him very greatly. The additions are small up to the accession of Andrew I (BV. 89), but here, at a point at which the texts of B and K also diverge abruptly, V begins to have a great deal of additional material. He remains about twice as full as B up to the disappearance of Salamon (BV 136), after which point B’s own text suddenly dwindles away. Hereafter, up to the reign of G&za II (V 167), almost the whole material is peculiar to V. At this point the texts again converge and it is only in the last chapter of all (BV 2 1 2 ) that V again has a fuller text. Domanovszky proves conclusively that every difference between V and B (with the possible exception of a small passage here and there, where something may have fallen in error out of our present text of B) is due to interpolation of B’s text by V.11 It is important to observe that this process goes on even where B is shortest. Thus in the reign of Coloman K and B give in practically identical form a short account (K 64, B 152) which is obviously intended to cover the whole reign. V reproduces this at the end of his own independent and much longer narrative. But B has also a little passage of his own, not in K, to the effect that 10» p. 1 collect*, ex diversis chronicis veteribus, earundem veritates. ascribendo et falstta/tem omnino refutando. p. 2. diversis sanctorum doctorum ystoriis. 19b Cf, Dom. B. K, p. 735 ff, 741; K&zai Simon Mester Kronikaja (Bp 1905; quoted hereafter as K. S. M. K., pp. 24—30J. 11 So also, but less fully, Kaindl, Studien XII. 74. (440).

126 8

Coloman had the little Bela blinded “extractum 12 de matris gremio”, by the advice of certain councillors. V incorporates this in his story, in c l51, where he writes that the King took council with the Councillors in question, “quorum consilio ducem et filium ejus de gremio matris sue extractum excecaverat.” Yet this contradicts his own narrative of the previous chapter, which says that Bela, at the time of his blinding, was “infantulus”. The text of B which V interpolated was either the Codex Acephalus (A) itself, or a text containing the same material as A. It is important to be clear on this point. Dom. writes 18 that Cod. A of B has “preserved two of the interpolations’’. These are V 92, and a summary of Stephen ITs reign, printed in the Scr. R. H. under B 153. We can prove nothing as regards the former, which is a complete whole in itself, introduced in the text as an alternative story ;14 but the latter is not a summary of V; a glance at the texts will show that V has combined A ’s words with other material of his own. Similarly, at the end of the Chronicle, the surviving text of V has taken in Cod. A’s account of the death of Charles and the coronation of Louis the Great; we return to this point.15 At the same time, a peculiar relationship certainly exists between parts of V’s interpolations and the text of B to which they are added. The chapters in question are those which begin (so far as the interpolations are concerned) just alter the coronation of Andrew I (V 89) and end approximately with the disappearance of Salamon (BV 136). The exact extent of these passages, and the exact nature of the relationship between the texts, must be left till later: but although in most cases it is clear that V interpolated, and it is probable that he did so in all cases, he often seems to have been doing so from the identical text of which B is a summary. An excellent example is in c 117. V has certainly interpolated B's text here: B is, on the other hand, as surely only ?i summary of the text which V replaces. V’s interpolation in c 126 is actually necessary to the sense, which B spoils, as we shall see, by attempting to use another text also. In c 89 V does not seem to have used B’s summary. Domanovszky18 describes all these passages as “fragments of 12 The text of B. is corrupt here, and reads "tractatura". 18 Dom., Interpolationen p. 797. 14 It begins with the words “dicunt alii". 15 See below p. 34. 16 Dom., B. K. p. 810.

127 9 an organic whole” which, except for a few minor details, is ‘‘systematically concerned with justifying B£la and his sons G 6 za and Ladislaus at the expense of Andrew and his son Salamon.” Similarly, Homan writes of them :17 “The 1051—1091 interpolations fit harmoniously into the original, shorter texts known from the Buda Chronicle. Not only are there no essential differences of outlook between the interpolated ‘Gesta' and the interpolations: there are not even differences of nuance. Sometimes it looks as if V's fuller text were nothing else than an artistic, more detailed paraphrase of the less colourful, dryer — although this is its most picturesque part — text of B . . . At other times it recounts facts which are missing from the latter, but fit exactly into the list of facts given in the shorter text. Sometimes V tells the same story twice, in the same way, but once with more detail. In other places it is patent that the text of B is an extract from that interpolated by V — an ♦extract which in one place has turned the sense upside down.” We shall return later to consider more closely the exact relaiionship between the two versions of this important source. V and K. We must leave until a later stage further consideration of tfie origins of V's interpolations, except for one group of them, which must be taken now. In general, V's text agrees with B, not K. Where the texts differ widely, which is nearly always due to the presence in B of additional material, V always has that additional material. On the other hand, he has very few of K's independent passages. There are, however, certain passages in which V agrees with K against B. Most of these consist of simple additions to B’s text of minor passages which are in K and not in B; but in a few cases where K and B differ, V has discarded B's text in favour of K's. The most important of these passages, all of which occur in the early parts of V's text, are: the following: 1. 2. 3.

BV. 2 1 , K 2 1 , the SzSkely in Transylvania. BV. 2 2 , K 2 2 , additional note on Chaba. BV. 23, K 23, the text of the note on Sviatopluk. 4. Ibid., the alternative (Morot) version. 5. BV. 46—7, K84—5, the family of Oliver and Ratold (B lias a different version.) 17 H6man, Gesta, pp. 69—70.

128 10 6.

7. 8. 9. 10. 11.

BV. 52, K91, additional note on Martindorf family. BV. 73, K 48, note on Godfrey, Markgraf of Austria. BV. 90, K 57, note on previous betrothal of Sophia. Ibid., inebriated state of German troops. BV. 96, K59, personal note on Bela I. BV. 108, K 60, discord between Salamon and his cousins

In some of these passages V might, theoretically, be following a lost text of B which agreed rather with K against all extant MSS of B. This seems, however, unlikely, since in the main, V. agrees exactly with B against K where the texts cover the same ground. Where he agrees with K, this nearly always takes the form of small additions to the text and the correspondence with K is then nearly always exact. Much the most reasonable conclusion is therefore that V normally used a text of B, but also collated with a MS which up to a point (we shall see shortly up to what point) agreed with K. This leads us on to consideration of the complicated question (round which controversy has raged for a century) of the relationship between B and K. B and K.

The older generation of scholars saw this problem in terms of the existing texts only. For them, therefore, it resolved itself into the question: which was the original? Was K a summary of B, or was B an expansion of K? A later generation perceived, and has come generally to accept, another possibility: that both works drew on a common source. But if this is correct, we still have to consider which adhered more closely to the original. Was this original something like B's text, summarised by K; or was it like K's, and expanded by B? I do not propose to go here into the back history of the controversy,18, much of which has been rendered useless by the mistaken identification of the Codex Sambucus with K’s own work. It will be sufficient here, having stated the problem, to examine only such theories and arguments as — rightly or wrongly — have not yet been exploded. The strongest arguments for believing K's to be the original 18 153— 180.

For a summary

of the

early

views,

see Dom., K. S. M. K. pp..

129 11

work lie in the opening chapters: the "Hun Chronicle” and the early part of the Hungarian history which follows it. These are very closely interconnected, and it is clear from the sentences of the beginning of each, that the hand which composed the Hun Chronicle is also the hand which reduced the opening pages of the Hungarian history to their present form .19 Not only do all these chapters form a whole, but they have the appearance of being signed, as it were, by K 6 zai. The text begins with a short dedication to Ladislaus IV (c 1 ), in which the author names himself by name. This is followed immediately by a prologue (c 2 ) which again is connected by cross-references to the Hun Chronicle proper which begins with c 3, and to the Hungarian history which follows.20 There are, moreover, certain astrological allusions occuring both in the prologue, in the Him Chronicle and in the chapter on Ladislaus IV, which is admittedly the work of Kezai himself.21 There are also a considerable number of italidsms in the Him Chronicle, one of which occurs also in the prologue,22 and finally, one or two phrases which seem characteristic .28 On these grounds many of the older enquirers supposed that the Chronicle must necessarily be Kezai‘s own work, and one modem scholar, to whom the collation of the stylistic peculiarities is due, still maintains this view.25 M. Madzsar, however, the scholar in question, is almost alone in his views today. They have been practically swept off the field by Professor Domanovszky. In those of his studies which are 12 Sec below, p. 101. 20 See especially the crossreference from c. 2. (volens itaque veritatem imitari, sic improsperos ut felices interseram etc.) to c. 24. (digestis igitur hunorum natalibus praeliis felicibus et sinistris, etc.) 21 c. 2 fin. rerum omnium, quae sub lunari circulo esse habent., c. 3 U6que lunarem circulum, c. 4 Velut Martis filius, cujus quidem constellatio conception! nativitatique eius die in audacie et celeris virtutibus naturalibus subministrat. 22 Scartabellae (in Prologue) partita — terida — passagium — potestafi = podesta — dimorare — conquestrare — cambul — sdbterrare — repansere — arnesium — antiqualia — lucta — strata — manganum — Vecca Venesia — guerra. 23 Elevatis baneriis c 8; vexilis erectis c 26; erecto banerio c 74 (Ladislaus IV). 24 The latest was Kaindl, Studien pp. 34 (400) ff.t where see the arguments in favour of this view. 25 I Madzsar, A Hun Krdnika Szerzdje (TortSneti Szemle 1920.) pp. 75 ff.

130 12

specially devoted to K£zai,2C Dom, compared the texts of K and B throughout, checking them, in the numerous cases where such control is possible, against the common source from which both are derived (e. g., the Annales Altahenses) or against a third writer (Anonymus) who has used the same material. He had no difficulty in showing that in some cases K's text is the purer, or the nearer to the original: in others, that of B. Therefore, he argued, neither could have borrowed from the other, but both alike derive from a common original. An almost identical proof of this point had incidentally, been given before him by Kaindl.27 The greater number of the variants which Dom. and Kaindl take for comparison are comparatively small points, and in such cases it must be pointed out that the argument is not conclusive. For our text of K is derived from late and corrupt MSS.2S The Codex Sambucus, which precedes a text of B with K's dedication and prologue, gives the latter in a much purer form than any of the ostensible MSS of K, even giving a complete final sentence omitted by all the latter. Therefore the whims, errors or individual inspirations of various copysts might perfectly well account for all K's minor aberrations. As for B, its earliest MS dates from the 14th century. It was admittedly not composed until 60 years after K wrote, and in that time there would be time and to spare for corruptions, additions and omissions in K's MS. Nevertheless, there are certain passages, in relation to which Dom.’s arguments are convincing. Two of these occur early in the Chronicle. They relate respectively to the death of Lei (K 40, B 60) and to the list “De Nobilibus Advenis”. B, in the middle of his description of Scythia (c 6 ) has these words: Scytia enim comprehensione una cingitur, sed in tria regna dividitur principando, scilicet in Bascardiam, Benciam et Magoriam. Habet quoque provincias C et VIII, que dudum per filios Hunor et Magor et centum et octo progenies, que egressi fuerant cx femoribus eorundem, de paludibus Meotydis intrantes Scytiam sunt divise. 26 K. S. M. K. and the later K6zai es a Hun Kronika (Bp. 1933) and A Hiinkronika szerkesztoje (Bp. 1934.). 27 Studien VII. and VIII. passim. 28 The original MS., which was still in existence late in the 18tih century, in lost. We have a printed version made from it at that time; also an MS. cop ied from the older MS. in 1701, and two copies made from that again.

131 13 K puts these sentences at the end of his description, adding the following words: Centum enim et octo generationes pura tenet Hungaria et non plures. Aliae autem, si quae ipsis sunt conjunctae, ad venae sunt vel ex captivis oriundi, quum ex Hunor et Mogor in palude Meotida centum et octo progenies absque omni missitalia fuere generatae. Quorum ergo advenarum generatio in fine huius libri apponetur seriatim. At the end of his history he recommences: Cum pura Hungaria plures tribus vel progenies non habeat quam generationes centum et octo — and proceeds to give his list of “noble strangers/* B, on the other hand, places the list after the settlement of the 7 Captains in Pannonia, when he asks: Cum igitur codices quidam contineant, quod isti capitanei septem Pannoniam intraverint et Hungaria ex ipsis solis edita sit ac plantata — unde ergo venit generatio Acus, Bor, Aba, aliorumque nobilium Hungarorum, cum isti non hospites, sed de Scythia descenderunt? He answers this riddle, and then goes on to give a list of “hospites.** It is very difficult to suppose that B worked here on the basis of K*s text. The sentence on the division of Scythia into three parts obviously belongs rather to the earlier than to the later part of the chapter, both by its nature, and by the analogy with Anon., whose own references to Dentumoger (to which the Dencia and Magoria obviously correspond) come early in the chapter. If B worked on K*s text, he must have performed a very elaborate piece of editing: he not only took the List of Strangers from the end of the book and put it back into the middle: he also cut out K*s sentence saying that it was to be found at the end, and then transferred the preceding sentence to an earlier point in his chapter. Further, K 33 has a sentence: Isti quidem capitanei loca descensumque, ut superius est dictum, sibi eligerunt, similiter et generationes aliae ubi eis placuit eligentes. B 35 has a corresponding sentence, and this he follows up by his list pf further settlers. K looks as though he were leading up to a similar list, but omits it. Thus it looks as though B’s order were the original; and he could not easily have got it out of a text resembling K*s.

132 14 The list itself leads us to a similar conclusion. The first name given by B (c 38) — Deodatus, founder of the monastery of Tata — is omitted by K, who, however, has an «cho of the story. B says that Deodatus was one of the two men — St. Adalbert of Prague being the other — who baptised St. Stephen. K expands himself on the story of Geza and Stephen, incorporating some of the material in B 37 and re-writing with the help of Hartvig's Life of St. Stephen. Thus B 37, 38 really correspond to K 76 — the first chapter of his appendix: K's omission of Deodatus* name may be due to the fact, mentioned by B, that Deodatus had no children and therefore “his generation is not found in Pannonia.” It is obvious that in this case B could not possibly have reached his result from K. K must have had before him a text which, if not exactly that of B, must at least have resembled it closely. When we look further into the list we find that it contains 11 families which are given by both K and B, in approximately, although not exactly, the same order. The resemblance would be closer still, but that the author of B quite clearly missed out K's first name (Tibold de Fanberg, K. 77) and scrawled it in, probably tn the margin, towards the end of his list, with the result of making his text hopelessly corrupt in that passage (B51).‘J* Nevertheless, there are differences, greater than can be attributed to mere copyists' errors, between the two lists. K has three families not mentioned by B: Altman (K83), Rodoan and Bogath (K. 89) and Lodan (K. 90). It has been observed that the properties of all these three families were in the neighbourhood of K£za, the family of Bogat-Radvan actually owning Keza itself.80 It is therefore moderately certain that K interpolated these three names. He also has a note on Tibold de Fanberg (c 27) which is not in B; a passage, not in B, on Hunt and Paznan (c78), and a longer and different account of the family of Martinsdorf (c 91). B on the other hand has a much fuller, although a different account of Hunt and Paznan (c 41). In c 44 he has one family (Kyquinus 38 K. 77 has: tunc eo tempore ad ducem Gcicham venit comes Tiboldus de Fanberg, qui Grauu Tibold est vocatus, unde de eadem progenia Grauu quidem usque hodie simpliciier nominantur. Ex illo illi de Bobochi orti detrivantur. Jsti etendm sunt Vandenchumlant. B. 51 has: eorum vero qui dicuntur de Bobocha de Samberg descend it in Pannoniam.

80 Domanovszky, Sc. R. H. I. 137.

133 15 and Renaldus — Kokenyes-Radnot) not mentioned by K; incidentally, this passage contains a serious historical error .31 The latter part of B 45 also contains a passage not in K, and it is interesting to note that this contains a reference to Count Vyd, the villain of a series of later passages which are also omitted by K. B 46 differs from K's version in K 84 of the same family. Here again B could hardly have been working directly from K; there would be no possible reason for him to omit precisely the three families from K's own district. And it is almost equally hard to suppose that K used a text originally identical with that of B. There would be no reasonable explanation for his omissions and alterations. The variants in the list are only reasonably to be (explained if we assume that both K and B worked on a common, older list, and that each made independent additions to it. The second passage decisive against the authorship of K occurs in B. 60, K. 40. Here B has preserved a semi-legendary tale of how the hero Lei, having been captured by the Emperor Conrad at the Battle of the Lechfeld, and offered his choice of deaths, asked that his horn might be brought to him and when this was done, smote the Emperor with it on the forehead, saying: “Thou shalt go before me and serve me hereafter." K refers to this legend, and deliberately rejects it, giving instead an alternative version of his own. Here, therefore, once again, B cannot have been working on K's text. And since K cannot have been working on B's, since he wrote some 60 years •earlier than B, the only possible conclusion is that both worked on an older common source. The same irresistable conclusion is forced on us in connection with certain other passages also, notably where the Annales Altahenses, or a text resembling them, has been used; but to this point we shall return in our detailed analysis of the two texts. The instances already quoted suffice for the time to prove the main contention, that K and B derive from a common archetype. There remain the difficulties of K's apparent signature of the H. C., and of the verbal echoes and reminiscences between the parts which appear to be undeniably K's work, and those 31 Queen Margaret is given as wife of "King BeLa son of Zar Ladislaus" = Bela I. Actually she was the wife of B£la III.

134 16

which, it appears, must be ascribed to his archetype. Homan3' has attempted to reconcile the conclusions of Dom. with the objections of Madzsar by means of an ingenious compromise theory to the effect that K is indeed the author of the H. C., and of the whole Chronicle, up to the reign of Ladislaus IV, as in Bo. K’s original text, however, has not survived. An extract of it was made, and this it is that has come down to us under the name of K. A different writer, then, of Andrew I ll ’s day, took K’s text in its original form, and rewrote the section dealing with Ladislaus IV. This, subsequently brought up to date once again, is what has come down to us as the Codex Sambucus of B. Up to a point, this theory is beguiling, since it does to some extent reconcile Dom.'s results with Madzsar’s But apart from the fact that Homan's argumentation, whereby he seeks to show in B’s account of Ladislaus IV’s reign traces that the writer was acquainted with, and polemised against, K£zai’s earlier text, is, to put it respectfully, far-fetched and unconvincing (besides which, we have another explanation for the phenomena which he noted32*) — even apart from this, there are serious objections to his theory. 1 . It is, as Dom. points out,33 intrinsically improbable that the Codex Sambucus represents K’s work. The dedication and prologue, on which the theory is founded, are written on a separate sheet of paper, with a clear break between them and the next part of the text. Moreover, they are headed “prohemium quoddam de gestis et factis Hungarorum regum.” But this is not how an author describes his own work. 2 . An unknown adapter would be just as capable as K himself of changing, abbreviating or expanding an original. It isr however, unlikely that he would do so in precisely the same way as K, by inserting in the list “De Nobilibus Advenis” names of personal interest to K£zai. It is surely stretching the arm of coincidence too far to ask us to believe that precisely those three names were inserted by someone else, and not by K6 zai. 3. There is a third argument which has not, I believe, yet been advanced, but seems to me convincing. There are certain passages in K’s text which can be set down beyond any doubt as his own work. These include, at least, his personal dedication, 32 Homan, Geeta, pp. 63— 4. 3*a See below, p. 31. 33 Scr. R. H. I pp 220—9.

135 17 his glorification of his master, Ladislaus IV, and the three names in the list “De Nobilibus Advenis.” Now, as we remarked, there are a number of passages where V, who otherwise agrees closely with B, appears to have collated his text with a text exactly resembling that of K. If that had been K's own, we should have expected to find V making at least some use of these special passages of K's own. But in fact, he shows no knowledge whatever of them. The only reasonable conclusion is therefore that he was collating, not from K, but from the archetype to which K made his own private additions. More striking still is the case of Dandolo’s Chronicle. Dandolo has a great deal of Hungarian information, drawn from various sources. One of these is the Chronicle of Paulinus Minorita,34 and this, in its turn, undoubtedly drew on some version of the Hungarian National Chronicle. A careful comparison of the relevant texts will, however, reveal the following result: a) Dandolo (Paulinus) certainly knew some form of the Hun Chronicle, or at least, a form of the Hungarian legend which attributed the original Descent from Scythia to Attila. b ) In his history of the latter part of the 1 1 th, all the 12 th, and the first three quarters of the 13th centuries, he followed a text which was practically identical with that of K, not that of B. There are one or two cases in which a word or a phrase is nearer to B’s text than to K’s ;85 but these small variants can easily be explained by the fact, to which we have already drawn attention, that our MSS of K are late and corrupt; or of course, that K did not copy his original faithfully. Generally speaking, Dandolo and K have the identical omissions, as compared with B, and even the identical errors. There can therefore be no question of their having summarised independently a text resembling that of B. c) From the above facts, we should deduce the fact that the Italian source had drawn on K, but for the fact that Dandolo’s use of it stops abruptly with Stephen V, giving for Ladislaus IV 34 For proof that Dandolo worked through the medium of Paulinus, see S. Eckhardt, A Pannonhalmi Hun Tdrtenet Keletkez£se (Szazadok, 1929, pp 465—491, 605—632). Paulinus’ work, unfortunately, has never been printed in full. 35 So Dand. IX. 1. 46. Sed juvenio virgo cum spon&a sua virgine poscea miraculis coruscans praeven tu s e st m orte; IX. 7. 10. Hie co n stitu it omnia fora fieri sabbato.

136 18

only his accession, but no details of his reign.:)a He omits, therefore, precisely that part of the text which is certainly K’s own work. Therefore, once again, the natural conclusion is that he has worked, not on K, but on K’s archetype. We shall return later to expand the further conclusion: that the archetype closely resembled K and not, as Dom. imagines, B. A corollary is, of course, that the Him Chronicle cannot be K’s work. These various considerations afford definite proof that Homan’s compromise theory cannot be accepted: we must agree with Dom. that both K and B worked, for the whole period from the beginning up to the reign of Ladislaus IV, on a common archetype, which each then altered, expanded or abridged in various ways. The difficulty of the verbal echoes and repetitions is not so formidable as Madzsar assumes. Nearly all of them vanish if we can suppose, as Dom. does in his later work,37 that only c. 1 (the dedication proper) and not also c. 2 (the prologue) is Kezai's personal work. The advocates of K’s original authorship object that he could not have passed off another's work as his own by 38 Space precludes me from giving in full, in parallel columns, relevant passages from Dandolo. These are found in Dand V. 3. 6., V. 3. 7., V. 4. l . f V. 5. 1. — (Hun Chronicle), VIII. 10. 10,. VIII, 10. 10., VIII. 10. 14., IX. 1. 46., IX. 5. 3., IX. 7. 9., IX. 7. 10., IX. 7. 21., IX. 8. 6., IX. 9. 8., IX. 10. 11., IX. 11. 16., IX. 15. 4., X. 1. 3„ X, 3. 11. The exact correspondence with K, as against. B. is very striking. Thus Dand. IX. 8. 6., K 63 both give the invaders of Hungary as “Bessi", against. B’s “Cuni". K, as is known, omits the reign of Stephen II altogether. Dand. IX. 11. 16. has the death of Coloman, and goes on: “Bela igitur in regno successit“. The last passage of agreement is Dan. X. 7. 19., which runs as follows: Bela rex Hungariae lancea in maxilla percussus a subditis obiit; filius autem ejus Stephanus in regno succedit. Hie Othocarum regem Bohemiae ante flumen Recham venientem potenter repulit, civitatem Budam sibi subjecit et similiter dominium Bulgarorum: CL K 72—3 Quern ante civitatem Novam Hungari cum lancea in maxilla transfixum peremerunt. Postea regn tvit Stephanus rex filius ejus, qui Boemiae regem nomine Otocarum ante fluvium Rebcha contra venientem cum Boemis videlicet, Australibus, Stiriensibus, Brandbergensibus et caeteri® mixtis gentibus expulit virtuose. Iste etiam civitatem Budam suo domino subjugavit, dorainiumque Bulgarorum eo vivente sibi compulit obedire. Then Dan. X. 9. 3. has: Post Stephanum regem Hungariae Ladislaus obtinuit regnum. But of K6zai’« narrative of Ladislaus' reign, he has not a word. 37 See his “K6zai 6s a Hun Kronika". In his earlier work, when he had not yet been faced with Madzsar's objections, he regarded the authorship of c. 2 as undecided. the

137 19 a method so primitive as adding three or four lines of dedication i but wrongly. We have an almost perfect analogy in the work of Miigeln, whose “Chronicle of the Huns”, up to the end of St. Ladislaus’s reign* is a mere verbatim translation of one of the MSS of the B group; and yet Miigeln, whom we have no reason to regard as a dishonest man* does not find it against his conscience to take the preface to the H. C. (KB/3), add to it his own name and a dedication to Rudolph of Austria, and proceed as though what followed were his own original work. As we shall show, K did at least as much original work as^ Miigeln, and is just as well entitled to his own dedication. And if a phrase occurs in ihe chapter on Ladislaus IV which echoes similar expressions in earlier chapters, this is, after all, not to be wondered at: in copying out so much of his original, it is only to be expected that K would have steeped himself in its phraseology. We must therefore reject Homan’s and all other rival theories and agree with Dom. (and Kaindl) that K and B alike worked on a common original. We shall henceforward assume the existence of this original, describing it by the symbol X. This need not, however, commit us to accepting Dom.’s view as to the relative degree of freedom or the reverse with which the two copyists treated the said original. This is the question to which we must now turn. K, B and X. Unlike Kaindl38, who believed K to have reproduced X fairly faithfully, Dom. holds the text of X, up to the reign of Stephen V (in which he places its composition) to be practically identical with that of B, while K’s is merely a lazy and slipshod abridgement.59 K, he says, was a man of great, and progressively increasing laziness. He began copying “with great zeal,” but even as early as the description of Scythia “his Magyar straw-flame flickered out” (Dom.'s phrasing, not mine), “he got bored with the trade of scriptor. . . The further he advanced, the more his ennui mastered him. He breathed more freely only when his source, too, 38 Studien VIII. 23 (225) and VII and VIII. pa ssim . K. S. M. K. p. 127 K6zai k^tsegteleniil nagyon m eg ro vid ite tte , kivonatolta . . . aimak a szdvegnek pedig, amelyet K. kivonatolt, nagyon k o ze l k e lle tt alln ia a bovebb szovegezlshez.

138 20

became meagre under Coloman, but soon he began to abbreviate even this scanty information.’*40 Dom.’s views have been widely — almost universally accepted. Homan, for example, although he suggests the modification ta which we have referred, yet accepts without question Dom.’s main conclusion as to the relationship between the two texts as we possess them; he believes, that is, that the text generally known as K is only an abridgement of that of B .41 This appears now tobe the general, orthodox view held by all, or nearly all Hungarian scholars. It is therefore with regret, but with conviction, that I am obliged to differ from it absolutely. The passage which I have quoted is excellent invective, but as serious criticism it is singularly deficient. Progressive laziness is not a common phenomenon; a man is, as a rule, either lazy or industrious, all through. Certainly, he is not likely to work in such fits and starts as Dom. attributes to the wretched K£zai. For the first half of K’s text is almost as full as B’s own, and shows sign of careful work. There are places where it lacks material contained in B, but others where it contains information not to be found elsewhere, and indicative of a certain critical sense and even of a certain degree of research. Even in the shortest part of the text, such passages occur, while at the end of his work, this idle man summons up the energy to compose a brand new account of his own of his master’s reign, and to add fresh names to the list “de nobilibus advenis.” Moreover, if K had been merely lazy, one would have expected him to summarise the text before him at least in most of its points. He does not do so. His work is no summary. Admittedly, he is a more laconic writer than B; where the two are working on a common source, he generally falthough not always) reproduces it more shortly than B. Even then, however, he does not leave out anything of importance. But this is precisely what he does do in the later part of his work. Here there is no summary: whole wads and dollops of material are left out wholesale. And what is really surprising, is that this is very carefully done. For example — we shall come back to these points later in greater detail — B has certain 40 Id. p. 133. 41 Homan, Gesta p. 63. Domanovszky . . . vizsgalatai vegleg eldontottGk a Kezai probUmat.

139 21

information about the family of the Gyula and Sarolta. This occurs in three places: B 30, B 63 and B 65. Now, K omits all three passages. Again B has a very long epic (to which we have already referred) in which one of the figures is a certain Vyd. K has no trace whatever of this epic: he omits it complete. But B also mentions Vyd in his list “de nobilibus advenis”, K omitting liim here also. Which is more likely: that K remembered that he had cut out mention of this man earlier, and therefore thought that he ought not to mention him later, or that B inserted both passages? We will not, however, argue general probabilities any further, but will proceed to our examination of the texts themselves. This will prove the picture to be very different from that drawn by Dom. Our first step will take us in a somewhat unexpected direction: towards the shorter narrative Chronicles.42

The longer and the shorter narrative Chronicles, There exist, as is well known, a considerable number of short Hungarian narrative Chronicles, all of which are cast in much the same form. They give, as a rule, little beyond a list of rulers of Hungary, with something on the family relationships of each monarch, the dates of his accession and death, the length of his reign and his place of burial. Sometimes, however, they also include a few details of more general historic interest, and many of them have also here and there a note on local affairs 1 The chief of these texts are the following: 1 . The Zagrab and Vdrad Chronicles (Z and W) contained in the books of the Statutes of Zagrab and Nagyv&rad respectively. 2 . The three related texts named after Knauz, Comides and Toldy respectively, with the Munich and Vienna MSS akin to them. 3. The Leutschau Chronicle (L. C.). Of these, W, as Szentpdtery has proved ,43 is undoubtedly a copy of Z (or of a text from which Z was itself copied), collated 42 In an essay published last year in this series I pointed out som e of the facts set forth in the following section; but at that time I had only perceived the situation in regard to B., X. and W.f not to Co. and L. C also. I am, therefore, to my regret, obliged to go over the ground again here. 48 See his introduction to the texts in Scr. R. H. I. 197 ff.

140 22

with and corrected and supplemented from a second text which, without prejudice to its nature, we may call *W. The main source of Co., T, and Kn., and of the less important Vienna and Munich MSS, is again a single codex, now lost, which we may call *Kn. Co., however, supplemented this source, fairly extensively, from a second source, which may be called *Co.44 The Leutschau Chronicle, which is in German, contains much particular information on the affairs of the Zips Germans, but also certain details of national interest. The latter again, are clearly based on some other source, which we shall call *L. C. Under the combined influences of Homan's theory as to the 1 1 th century “Gesta” and Domanovszky’s as to the substantial identity of B with the 13th century X,45 the following theory of the origin of these texts has come to find general acceptance: The groundwork, i. e., the common basis of Z and W, and *Kn., are held to derive from some version of the “Gesta”. As W where it differs from Z, Co. where it differs from Kn. etc. show many points of resemblance with B, their editors believe that their authors collated their original texts with some version or other of B: i. e., both *W and *Co. are simply B.40 As for L. C.f its editor dismisses the Hungarian source to which it twice refers, once under the name of “ungerische cronica” and once under that of ”dy cronica martiniana” as being in each case a version of the Chronicon Pictum Vindobonense.47 But to take now the former of these assumptions: we have already seen that there is no theoretical necessity, nor even justification, for deducing the existence of a comprehensive “G^sta.” We now come to the practical difficulties attendant on the theory. For if Z, *Kn., but also K and B itself are all based on the same text, how is it that they do not resemble one another more closely? What is the origin of the very considerable difficulties which exist, after all, between them? 44 See the introduction by E. Bartoniek, Scr. R. H. II. 323 ff. Also Domanovszky, a Pozsony Kronika 6s a kisebb latin nyelvti kr6nik£k (Bp. 1908); described henceforward as Dom., P. K. 48 Even before these theories were promulgated, however, Kaindl, Studien XI. pp. 69 ff. took the shorter Chronicles as being extracts from the longer. 46 For W: Szentpetery, Scr. R. H. I. 200; for Co., Dom.. P. K„ p 44; E, Bartoniek, Scr. R. H. II. 323 ff. 47 B. Pukinszky, Scr. R. H. II. 279. 48 Z. W. c. 6.

141 23 Homan dismisses these airily enough: they crept in somehow, much in the way in which Lucretius' first errant atom diverged. And yet this is a very bold assumption. To take first the ZW group: as regards some of their 11 th century history (where they refer to a “Cronica Hungarorum") they seem to be summarising a text resembling B's; but for the earliest period, the differences between these two texts and BK are so considerable as to make it almost impossible to believe that these two texts derived from a written text which was also the source of X. Thus Andrew, Bela and Levente are made the sons of Wazul — a statement against which X protests strongly (K 55, B 87); the name of Zar Ladislaus' son (Bonuzlo) is given; and it is specifically stated that Emeric was Stephen's only son; also that he was a virgin. It therefore seems much more likely that ZW represent an alternative tradition, which never stood in any direct relationship with X beyond the derivation of both from the common national tradition .49 In the case of the group deriving from *Kn, it is certain that they have to some extent borrowed from B. At the very, end of each text comes a passage which agrees almost verbatim with the postscript added by Hess, the printer of B, to his text, this being an addition made by him after B had concluded his own work. Further, all the texts begin with a brief reference to the Hun Chronicle, in which they agree exactly with B, rather than K, giving, as B does, the name of Keue twice among the 7 Hun Captains (K has Keve and Rua), and giving the order of the 7 Magyar Captains to agree with B, not with K. But these resemblances do not extend to the whole text. The MSS are late, and it appears that a late copyist of *Kn collated his text roughly with B. In the body of the text, the Kn. group resemble the ZW group far rather than K or B. Zoltan, son of Arpad, has an integral part in their story; Emeric appears as Stephen's only son, and his virginity is specifically mentioned. The date of St. Stephen's coronation is given, and the fact that this ceremony was performed by St. Adalbert. Thus here again it seems that we have an independent version of the national tradition preserved. We cannot, as regards this particular point, draw many 49 The two text® appeal in their c 1 to a “relatio Hungarorum in scriptis ab olim redacta”, but there is no earthly reason for supposing this to have been the text from which X derived; in view of the differences noted above, it was surely rather one of the alternative traditions to which K55 B87 refer.

142 24 conclusions one way or another from the L. C., which has very little national information on early Hungarian history; but an unprejudiced view of the texts whence any conclusions can be drawn should, I think, lead us inescapably to the conclusion that the various chronicles, including X, represent variants, not on a single “Gesta”, but on a single national tradition which, concerned as it was with the barest minimum of facts relating to the rulers of the land, was necessarily and naturally fairly uniform .50 There remains the question of the resemblances to B found in W where it differs from Z, in Co. where it differs from Kn., T„ etc. and in L. C. Here the same difficulty arises. These texts are supposed to be extracts from an X substantially identical with B. But exactly the same origin is attributed to K. But in that case, the various texts, including K, ought surely to resemble one another fairly closely. We can reasonably allow for a small number of variants, but we cannot expect such variants to cover a high proportion of the different texts; nor can we suggest any valid reason why each excerpter should have been interested in a different part of B’s text. The weakness of the advocates of the orthodox view is that they have compared the various texts only with B, and not with one another. If, however, we undertake this further comparison, we reach results which at once force us to adopt an entirely different view of their relationship. Thus if we disregard the cases, already enumerated, where W differs from B and K alike (all of which are to be attributed to his component Z, and thus due, as we showed, to the use by Z of an alternative tradition), and concentrate on those cases in which W agrees with B but differs from K, (these being the points on which he also differs from Z), we find the following list in the period up to St. Ladislaus’ death only: 1. The date of the Conquest (B 26); W, 88—; B, 888; K. 872. 2. Peter’s relationship to Stephen and Gisella; W, frater Giselle; K45, filium sororis suae (sc. Gisellae); B 70, fra ter reginae, and also son of Stephen's sister. 3. The description of Peter: W, Theutonicus; K 45, Venetum; B 70, Alamanum vel potius Venetum. 50 It is true that Z. W. on the one hand and Kn., etc. on the other, have many points of resemblance, and it is not impossible that all alike were influenced by the “Somogy Chronicle", the existence of which H. induces in his ‘‘Gesta"; but even so, there axe enough variants between them to force us to postulate, both for them and for the Hungarian source of Albericus, alternative traditions also.

143 25 The following points are in both B and W, but not in K: 1 . The words “jure hereditario possidendam” (B28J. 2 . The election of Almus (B 26, 28). 3. Toxun begat Geyza and Michael (B63). 4. The reference to the Legend of St. Stephen in connection with his birth (B63). 5. The day of Gaza’s death (B 130). 6 . His title of “rex chris tianissimus” (B 130). A similar comparison of Co., K and B over the same period gives the following number of points in Co. and B only, and not in K: 1 . The date of Attila’s coronation {B 1 0 , Cod. D only). 2 . The end of the genealogy in B26: Almus genuit Arpad, Arpad genuit Zoltan, Zoltan genuit Toxun. 3. Toxun genuit Geyssam et Michaelem (B63). 4. St. Stephen's mother Sarolt, daughter of Gyula (B63). 5. That St. Stephen founded the church where he was buried t B 7 °). 6.

The name of Aba’s monastery, Sar (B 76; K has “juxta quendam ecclesiam”). 7. The place of Peter’s death, Zamur (B85). 8 . That Peter founded Pecs Cathedral in the 3rd year of his reign (B 85). 9. The date of Andrew I’s coronation, A. D. 1147 (B 8 6 ). 1 0 . Andrew I’s alleged name of “Albus“ (B 8 8 ). 1 1 . His marriage to a Russian wife (id). 1 2 . The patron saint of his monastery at Tihany (B 8 8 , 93). 13. The town, Zirc, where he died (B93). 14. The birth of Bala’s sons, Andrew and Ladislaus (B 80, and 8 8 ). 15. A sentence to the effect that there were two pagan reactions in Hungary in the 10 th century (B95). 16. The description of B6 la I, “piissimus rex’’ (B96). 17. The patron saint of his monastery, St. Salvator (B96). 18. Salamon buried in Pola, and his wife and mother in Agmund (B136). 19. Salamon had no children (B 1 0 0 ). 2 0 . Geza I’s epithets, “catholicus, devotissimus et christianissimus princeps“ (B 130). 2 1 . Geza I begat Coloman and Almus (B 1 0 0 ). 2 2 . His burial place, the Church of the Blessed Virgin (B130). 23. The length of Ladislaus I’s reign (K gives this differently) ^ind his place of burial. These almost exactly, are also the points on which Co. differs from Kn., etc .51 31 See the enumeration of those points in Dom. P. K.; or the parallel texts in Scr. R. H. II. 327 ff.

144 26 Admittedly, one or two of these points occur in both W and Co., and cannot fairly be counted twice. When, however, we consider that the whole length of Co.'s text for the period from which we have taken these extracts consists only of a page of print, covering material for which the possibilities of variation are, after all, very limited (no text, for example, could give the names of the kings differently), can we really believe that Co. managed in this meagre space so to epitomise a text also used by K — and used, for most of this part, to produce a fairly ample text — as to include no less than 23 facts which K managed to omit? The comparison with W is only less striking, but leads to the same conclusions. This conclusion is, of course, that contrary to the general theory which holds the field today, B combined a text of X which resembled K*s with *W and *Co. B is not a source tor the latter, but they for him. And we shall find the same relationship to apply in the case of the L. C. The texts for this earlier period are too ample for us to set them side by side in their entirety, and show how frequently that of B consists of a neat addition sum of the others (the fact is also often obscured by his use of yet other sources). I will however, give here one or two passages where it can be proved conclusively that B has reached his result by combination and interpolation, not the other texts by summarising. Perhaps the best of all is the following passage from K 43, B 63, W 2 : K B W Anno verodominicae Geycha vero divino Dux Geysa, filius incarnationis 967 praemonitus oraculo Toxu, de quo supra, Geicha dux divino anno Dominicae in- anno ab incamatiopraemonitus oraculo carnationis 969, ne Dei 963 nuto digenuit sanctum regvino cepit cogitare em Stephanum; Michde ritibus paganisael vero, frater Geimis destruendis et chae, genuit Wazul de cultu divino ampet Zar Ladizlaum. liando, seque fecit quemadmodum in leg- baptizari et tandem enda Beati Stephani in huiusmodi sancto scriptum est, genuit proposito mortuus Sanctum Stephanum est, prout plenius regem. in legenda beati regis Stephani filii sui continentur, qui scilicet sanctus rex Stephanus natus est anno Domini 969

145 27 Thus B’s text, as it stands, ascribes to the Legend of St. Stephen the statement that he was bom in 969. Now, this is not correct; no version of the Legend gives the date of his birth a t all, much less ascribe it to the year 969. B has obviously got his result through combining X. (here = K) with *W, who both give^ the date 969 and refers to the Legend, quite correctly, for an account of Gaza’s end, B then achieving his confusion by telescoping *W. So, too, in the accounts of Peter’s relationship to Gisella:: K 45 makes him son of Gisella’s sister. Co. is silent, but X o. must have had its own version since T. and Kn. both have "Petrus Alamanus filius Wilhelmi ex sorore Sancto Stephani progenitus.,, W 4 has “frater Resla (sic) regine, consortis sancti regis Stephani, de sorore sancti Stephani genitus." The last 5 words are not in Z, and it looks as though *W had taken them from the Vita S. Stephani (Hartvic c. 2 1 ), which correctly makes Peter son of Stephen’s sister. B works out an amazingly complicated genealogy, as follows: Peter’s father, William, brother of Sigismund of Burgundy, married, firstly, Gertrud, sister of the Emperor, by whom he had Gisella, and secondly, Stephen’s sister, by whom he had Peter. As Stephen also married Gisella, Peter was at once his sister’s son and his wife’s brother (i. e., half brother). This result could not possibly have been obtained from real life, since Peter’s father was not any William, but the Doge Otto Orseolo, while Sigismund of Burgundy died about 500 years earlier. Nor could the result have been got from any of the sources by itself, but only from a valliant attempt to combine all of them, plus the Him Chronicle. Again, K at this juncture calls Peter "Petrus Veneticus"; W, "Petrus Theutonicus", and Co., "Petrus Alamanus,’’ while B again combines all three, writing in B 70 "Petrus Alamanus vel potius Venetus", and in B91, "Petrus Theutonicus vel verius Venetus." There is no reason whatever to suppose that Andrew I ever bore the name of "Albus.” We can see how Co. made the mistake, writing "coronatus est rex Andreas Albus" for "Alba" = at Sz6 kesfeh£rvir; but B takes over the word in a phrase (iste quidem rex Albus Andreas et catholicus est vocatus) in which the mistake could not have arisen. B gives the births of B£la Vs sons twice: once in B80, where

146 28 he is describing Bela’s life in Poland, and a second time in B 8 8 where, as the context shows, he is using Co. Co. writes of the pagan revolts: “Postquam autem sanctus rex 5tephanus convertit Hungariam ad Christum bis in paganismum "versa est: primo tempore regis Andreae Albi, secundo vero tempore regis Belyn, sicut nunc scriptum est.” Whatever these last words mean exactly, the phrase is far more in place in Co. than where it recurs in B, at the end of B 95, after the description of how the second rising was crushed. The following chapters of the texts are set out in diagrammatic form (Appendix II) and the reader is requested to study this table carefully .42 This comparison speaks for itself. Obviously, there is a certain substratum which must be common to any Chronicle: the names of the kings, and a few bare details about the lengths of their reigns and their places of burial (which every Chronicler is always anxious to give) 33 Once, however, we have deducted this, and turn to what the various Chronicles contain over and above this bare minimum, we find the following remarkable result: a) There are one or two words or phrases only found in one of the texts above; one matter to which both W and Co. Tefer (Andrew’s skill as a collector of relics); and one or two where W and K coincide; for these we shall find a special •explanation. b) Against these, we have to set numerous passages common to B and K only; others common to B and W only; others again, common to B and Co. only; yet others, to B and L. C. only. For most of the reigns, we have an exact equation of B = K + W + Co. + L. C. Thus for Coloman, K has his nickname and his campaigns, W tells us that he was bishop of Varad, Co. says that his brothers died young, and gives (under the heading of Bela II) the genealogy of that monarch, L. C. says that many evils happened in his 58 I have omitted from them those parts of the L. C. which are obviously of local origin and interest. M Although K's source for this period is weak on this point, he regularly gives the burial-places of the earlier kings. His failure to do so here can therefore only be due to a weakness in his source at this point, and' not to any lack of personal interest in the question.

147 29 reign. B has all four, in spite of the fact that the respectful and pious version given by Co. that Coloman’s brothers were “prevented by death from ascending the throne** contradicts what he has from other sources. For Andrew II, K has his journey to the Holy Land, W, his benefactions to Churches, Co., his wife, L. C. the Bank murder and the marriage of St. Elizabeth, B has all combined. For B6 la IV, K gives the number of the Tartars, the battle on the Soio, Bela's flight, etc. W has the one additional detail that the Tartars stopped 3 years in Hungary. L. C. mentions: the famine which followed their invasion. B exactly combines the three accounts. He then gives the Austrian war, only in K, and Bela's virtues and epitaph, only in Co. So accurate in places is the addition sum, that where K says that Stephen V took Viddin, and W says that he defeated the Bulgars in battle, B gives both. Later, Co. gives Stephen’s son, Ladislaus, and his daughter, Maria. W says that he had three daughters, one of whom was Maria. B follows Co.’s wording, but substitutes for his “filiam" the words “inter alias filias imam.”' Finally, for Ladislaus the Cumanian, where K is out of the picture, B equals most exactly, almost to a word, the sum of W. Co. and L. C. while the three shorter texts have hardly one single word in common. Even in the account of Ladislaus' murder, W has the place, L. C. the words “dy her alleczeit libet", B combining both accounts. Here there can be no question of differing tastes or interests (of which, in any case, the texts give little evidence). It is quite out of the question that K in summarising an X = B should have taken precisely what W, Co. and L. C. did not think fit to mention; W. just what did not interest K, Co and L. C.; Co., exactly what K, W and L. C. kindly left him; L. C. anything left over by the other three. No reasonable person, after setting these five texts side by side, can continue to believe that the four shorter texts are so many independent extracts from the longer. The only possible solution to the problem of their composition is that B combined the originals of the texts used severally, one each, by the other four. I say “the originals", for obviously B did not take the four texts which have come down to us. It was X he used, and not K, X being substantially identical with K, but in places slightly shorter (since K made his own little additions), in places, probably, a shade longer. Similarly, it was not W that he used, but rather *W, the text which the author of our present W used to combine with Z; not the present Codex Comides, but *Co, the source which

148 30 the author of our Co. combined with *Kn;5i and not the Leutschau Chronicle, but the text on which the author of that Chronicle worked: the “Chronica Martiniana”. Each of these originals was certainly longer than the extracts which the shorter texts have presented; so that it is more than probable that one or the other contained most of the few passages in B which cannot be traced to them with absolute certainty. Tthe true relationship between our texts is thus that shown in the following diagram.

We are now able to see more clearly the composition of the -various shorter Chronicles. Z, as we said, was clearly based originally on some short alternative version (not that of X) of the early history of the Magyars. It was continued in the conventional form of a List of Kings, to which various small additions were made. The author knew, and put in short references to, ihe Lives of SS Stephen, Emeric and Ladislaus, and also to a work which he describes (c 6 ) as the “Cronica Hungarorum”, which was in all probability that text which we call the “Cycle”, i."«. the 1 1 th century work on the rivalry between Salamon, B6 la and Ladislaus. He did not, however, attempt to write history in any real sense of the term, but simply padded out his skeleton with an occasional note, added perhaps from memory. *W was a similar text, to which similar small additions were made. One of these is important for us. When writing of Ladislaus IV, the author mentions, consecutively, those two events also related by K, and does so in language which, laconic as it is, 54 Practically all the passages where Co. differs from Kn., T etc. are also those where it agrees with B. but not with K. The ohief exception — the reference to Coloman's nickname — must therefore fallen out of the copy of *Kn. whence the other texts are derived.

149 31

is yet reminiscent of K’s.1 B, as we saw, takes these phrases over into his text, whence Homan, not understanding the true position, deduced that K was identical with X, and that a writer of Andrew Ill's reign took the text and wrote it over in a sense unfavourable to Ladislaus IV, without, however, wholly eliminating all traces of the earlier, laudatory account." The real position must be quite different. These notes of *W‘s, and also his brief note on Stephen V's victory in Bulgaria, which he also has in common with K, must have been derived by him, directly or indirectly, from K. He need not have seen K's own text. K. was, after all, a Court historian, and echoes of his work can easly have penetrated to Nagyvarad. This explanation removes any need to postulate a writer of Andrew III.'s day — a hypothesis which Domanovszky and others have found hard to accept, for various good reasons. The influence of *W on B cannot be traced with certainty after the reign of Ladislaus IV., although there is, as we saw, some reason to suppose that his text contained the genealogy of Charles Robert. His additions to Z stop with the death of Andrew III. His text must therefore have been composed soon after that event.

*Kn. appears to have run as far as the death of Sigismund in 1435. Its basis was again a List of Kings, which, as regards the early history, stood nearerto Z than to X and may have been influenced by Z's own source. The 15th century author collated very roughly with a text of B itself, only, however, to the extent of adding a line on the Hun history, etc. As regards the body of the text, he made no changes or additions from B.

*Co. is based on a similar skeleton List of Kings, but one which represents a very much more original tradition; since there are, as we saw, very many small points of difference from all the other versions. B certainly uses this source for the wording of Andrew Ill's coronation, parentage, nomenclature and death (B 187) ; traces of it are found also in B 1894 and perhaps in B 191.5 It is difficult to be certain of any later influence. It is true that #Co., like B but unlike Kn., gives Charles Robert's marriage with Elizabeth of Poland, and also gives the name of his son, Charles; but this is probably due to the use of a common source. By far the most interesting and important of our subsidiary sources is *L. C. As we saw, the author refers early in the text to a “ungerische cronica", and later to a “cronica Martiniana.” 1 W 20 Qui prias Ottocharo rege Bohemie in prelio interempto et Cumanis Dei adiutorio viriliter in bello campestri superatis. 2 Homan, Gesta pp. 63— 4. 3 Scr. R. H. I. 225—6. 4 In the words (in both texts) “quem Hungari Ladizlaum vocaverunt". 8 In the words “ex puLsus fuit de regno Hungarie” (Co.); "de regno expeilitur Hungarorum" (B).

150 32 It is by no means certain that these are identical, or at least, that the German compiler is not quoting one of them at first remove. For there are three hands to be traced in this work. The German author's own work begins with his annalistic notes at the end of the 14th or beginning of the 15th century, and the same writer probably filled in the gap between the end of his main written source and the beginning of his annalistic notes. He also made a few additions and adaptations to the earlier part of the text. His main written source, however, was the Cronica Martiniana which he mentions ad ann. 1235. This must have been a work of first-class importance for the 14th century history of Hungary. It is by far the most important source for B’s last chapters. To it B undoubtedly owes not only the passages which we have already adduced, but also the end of B 186, the Papal message in B 187, and the greater part of the material up to where the annalistic notes begin in B 197. The campaign in Wallachia certainly comes from the same source, as the comparison with L. C. shows, and although no traces of them have survived in L. C/s short summary, we may fairly attribute B 206—7 and 211 to the same source, on the grounds of similarity of style. etc .6 These chapters represent what the author of the Cronica has put in from his own, personal knowledge of contemporary events. As, however, we have seen, he wrote up the past also; Domanovszky is perfectly right in attributing to some of the chapters dealing with the opening years of the 14th century a retrospective character, and in seeing a connection between them and some of B’s final chapters 7 although, as we now see, the uniformity which he notes applies not to the whole text, but only to those parts of it contributed by the Cronica Martiniana. What the author of that work clearly did was to take one of the usual skeleton Lists of Kings, and to bring it up to date, also padding it out, for the earlier period over which his own memory could not help him, either from memory, or from some source or sources not known to us. This explains the late and inaccurate character of the 12th and 13th century chapters in B which are derived from this source. Thus the story of Bank which it gives differs widely from the usual accounts of this famous murder ,8 and grave objections have also been raised to the account of the Arad massacre, the last sentence of which 6 Cf. the names of the persons mentioned, and in general, the author's habit, shared by no other, of specifying by name the non-royal persons chiefly engaged; also the following phrases: B 174 (Bank murder): Sed proh! dolor, humani generis inimico procul dubio suadente, etc.; B 205 (Felician murder): diabolus mis it in cor cuduedam militis .. . proh! d o lo r ... 7 Scr. R. H. I. 225—6. * Huber, Die Ermordung der Konigin Gertrud von Ungam; Archiv fur oe. Geschiohte, LXV, 1184.

151 33 appears to be unhistorical.9 It is probably this skeleton which L. C. quotes (himself probably quoting the Cronica Martiniana) as the “ungerisdhe cronica.” 10 Finally, the insight which we have now gained into the composition of B’s last chapters enables and compels us to submit to some small degree of revision the conclusions reached by various scholars, especially Domanovszky, 11 on the date and final method of composition of that Chronicle. I must, however, deal with this question very shortly. All the texts of V break off in the middle of a sentence in c 209, except V 5, which finishes the chapter, gives c 2 1 0 , omits c 2 1 1 , and gives c 2 1 2 in the words of the Cod. A of B, with additional material. Of the texts of B, R stops at the end of c 2 1 1 , S and A give different accounts of the death of Charles Robert and coronation of Louis. B and D give further annalistic notes, exactly similar to those in B 197—205 and 2 1 0 , ad ann. 1335, 1338—1342; and end with John de Kukiilld’s Life of Louis the Great. Domanovszky sees quite correctly that the additional material in B/D constitutes an addition to the original text, but believing as he does that cc206 ff. constitute B’s original work, instead of only a source used by B (who was a compiler pure and simple), he fails to draw the necessary conclusions. Moreover he, like others before him, is led astray by the fact that S, A and V 5 end with the same events; he does not even apply to V and A the rule which he applies in all other cases where he has investigated their relationship; but holds A 2 1 2 to be a summary of V 2 1 2 . In reality, however, the relationship is exactly the same here as elsewhere; and we must remark that, as the cases of cc 92 and 153 show, it is the Cod. A of B (or copy of it) on which V worked. Here again V has taken A’s text, and has added to it further material from another source. This material was used also by Dlugoss12 and perhaps by Kiikiillei also ,13 but did not come from L. C., whose work was not known to Dlugoss. 9 Huber, Mitt. dee In st fur oe. Geschichtsforschung, IV. 134. 10 I remark in passing that his source appears also to have been used by the author of that curious work, the Chronicon Rythmicum Austriacum, (M. G. H., SS. pp 350 ff), who gives an account of the Bank murder irt language distinctly reminiscent of B’s (11.260 ff), the marriage of St. Elizabeth (11.255 ff), and B6la IV s war against Ottokar of Bohemia (11.609 ff). He also has a curious reference to a massacre of guests alleged to have occurred on the accession of B6la IV, reminiscent of the massacre under Bela II. As the author is notoriously inaccurate (his editor, op. cit. p. 349 suggests that he often even writes from memory) he may easily have confused the two Belas. 11 Scr. R. H. I. 219. ff.; also B. K., and recently, A magyar kiraly kronika XIV. szdzadi folytatdsa (Misc. in honour of A. Berzeviczy, Bp. 1934, pp. 31—34). 12 Hist. Pol. IX. pp. 1063—4.

152 34 We thus have Cod. R ending with c211, and A, S and B/D continuing it in three different ways. Now, since the annalistic notes used by B/D from 1335 on are of exactly the same type as those used by all the texts up to 1332, there is no possible reason why R, A and S should have omitted them had they been in the original common text; nor why A and S should have given different versions of the death of Charles Robert and the coronation of his successor. We are therefore bound to conclude that the Cronica Martiniana stopped in 1333 (c 2 1 1 ), and that in that year, or the following one, B, who, we repeat, was a compiler pure and simple, although a compiler on the grand scale, carried through his work, using the various sources which we have enumerated: X, *W, *Co. *L. C, the further sources to which we shall draw attention when examining more closely the relations between B, K and X, and also one further source, which must have been equally independent of B, to wit, a series of annalistic notes, probably kept by a Minorite Friar of Buda ,14 which B used up to the year in which he himself worked. The true end is as R has it, after c 2 1 1 . The scribes of A and S respectively, writing after the coronation of Louis the Great, each put in his own little tail-piece to bring the story up to date, while V., as usual, elaborated on A. Meanwhile, the annalistic notes were continued, and the common source of B/D made use of them up to 1342 when making his own copy of the MS.

The Lives oi SS. Stephen and Emeric. This is the most convenient, although perhaps not the most logical place to discuss the question which of the narrative Chronicles used the Lives or Legends of St. Stephen or St. Emeric. Dom. assumes 1 simply that the original text of X = B, knew and used them both; and for proof, he quotes, as regards Stephen, the passage quoted above on St. Stephens’ birth and as regards Emeric, the following (K 43, B 63): 13 Ktikullei also has the names of the dignitaries present at the coronation; but Dlugoss has them also, and in the same form as V. 14 Marczali, Ungaros Geeohichtsquellen pp. 52 ff., Kaindl Studien XI. 59 (425) f. and Domanovszky op. cit. are agreed on attributing the end of B to a Minorite Friar of Buda. They find indications of this in numerous chapters, •which we now see to have come to B from three different sources. The annalistic notes clearly come from a minorite source, and it is easy to suppose the compiler, B himself, to have been an inmate of the monastery ^whence the annalistic notes originated. For that matter, the author of *Co. may have been a minorite friar also. The Cronica Martiniana, on the other band, must, as its name indicates, have been composed in Pannonhalma.

153 35 K A t rex Stephanus plures quidem filios genuit, sed super omnes unum genuit Emricum nomine, Deo et hominibus honorabilem.

B At rex Stephanus plures quidem genuit filios, sed inter" alios habuit unnm filium nomine Emericum Deo amabilem et hominibus honorabilem, cuius memoria in benedictione e s t Hie autem per inscrutabile divini consilii judicium raptus est de medio "ne malicia mutaret intellectual ejus et ne fictio deciperet animum ejus“f quemadmodum de inmatura morte scribitur in libro Sapientiae. Ut ergo acta ne agamus et exposita ne exponamus, quot et quantis virtutibus floruerit, et quam fervens in Dei servitio sanctus confessor Christi Emericus dux fuerit, scribere supersedimus. Quisquis enim hoc sciere voluerit, ex legenda eiusdem beatissimi confessoris plenam sanctissime conversations eius notitiam habere poterit

Dom. draws from these passages the truly astonishing deduction that "the source which used the Legends" (viz. of SS. Stephen and Emeric) lay before K6zai, whereas as we see, the reference to the Legends is, precisely, in B alone, and not in K. Further, the later references in K to Emeric — the frustration by his untimely death of his father’s intention to hand over the government to him are, precisely, not in the Legend. Surely, if ever there was a clear case of interpolation, it is here. To this we may perhaps add the fact that while K 44 speaks only of Stephen’s intention to hand over the kingdom "Emrico duci, suo filio" (later, dux and Emricus), B adds a phrase "sanctissimis moribus divinitus instructo", which summarises the first chapter of the Legend (a second reference to the same chapter is probably to be found in the statement that Emeric was adorned with the "Catholic and political virtues"), and refers to him as "Beatus Emericus**, "beatus dux Emericus." Are we really to believe that Kgzai painfully cut out, not merely the direct reference to the Legend, but all the indirect references to it also? 1 Dom. K. S. M. K. pp 97, 98.

154 36 We must therefore conclude that X was composed without reference to the legend of St, Emeric, while B interpolated the text from that legend. The apparent references to the legend of St. Stephen again, occur to B only, and not in K. They consist of the passage quoted above, and of one other in B 70, refering to miracles practised at Stephen's tomb.2 Here again Dom. would have us believe that. X used the Legend, and that K omitted the reference to it. But in the first place, it is really impossible to suggest any valid grounds why K should have gone out of his way to eliminate from his text precisely those passages which derive from the Legend of the great national saint and hero; it will hardly be suggested that he had an anti-Stephen complex, on top of the anti-Transylvanian complex with which Dom. credits him.3 Moreover, as it happens, K actually himself uses the Legenda (Leg. Maior 3, Hartvic c 3) in his list “De Nobilibus Advenis" (c76), in a passage in which his text differs widely from that of B. Kaindl4 draws the remarkable conclusion that this passage is not K’s own work, whereas just the contrary deduction clearly emerges; that in this passage, he is taking the trouble to do original work. In the narrative part of the Chronicle, however, he does not trouble to take similar pains; nor do those passages which he has in common with B show any knowledge of any of the Legends of St. Stephen. It is true that they refer to certain facts which are also mentioned in the Legenda: G6 za’s vision, the birth of Emeric; also, like the Legenda, they credit Stephen with other children besides Emeric, while certain of the shorter Chronicles make Emeric Stephen's only son. But G£za, Stephen and Emeric were not nobodies; they were the greatest national figures of their age, and it cannot be maintained that a few of the most 2 Kaindl Studien VIII. 31 (233) reaches the same conclusions, but his arguments are not very convincing. 3 Ubi multa signa et miracula Hunt intervenientibus meritis eiusdem sanctissimi regis Stephani ad laudem et gloriam Domini nostri Jesu Christi, qui esi benedictus in saecula saeculorum, Amen. ,a See below, p. 39. 4 Studien X. 56^—57 (422—3). Kaindl* 6 conclusion, that the references to the Legend are interpolated by B. (Studien VIII. 30—232) is sound, but his arguments in favour of it seem to me inconclusive. In any case, he has failed to recognise the role of *W in B’s text.

155 37

striking facts or legends about them could not have got recorded without the use of the Legend, which incidentally, although it records the same facts, does so in entirely different language. The existence of this small minority of common facts is far less remarkable than the total difference between the vast majority of facts recorded in KB on the one hand, and the Legenda on the other — a difference of which any one may convince himself by reading the two documents together.8 The only logical conclusion is therefore that X contained no references to the legend of S i Stephen; that he here copied X faithfully, and B interpolated the references. I doubt however, whether he did so from the legend itself. We have seen that in the first passage, B was not really using the Legenda. but abbreviating *W. This second passage also corresponds to words in W: Hie quantis a Deo gratiis fuit insignitus et quanta opera exercuerit meritoria vide in sua Legenda. Alternatively, it may have come from the Legend of St. Emeric, which writes (c 8 ) that “ quotidiana fiebant miracula" at St. Stephen’s tomb. After dealing with these general further sources used by B, use of which appears in various places in the counse of his work it will be most convenient to take the remainder of the text in order.

The Hun Chronicle and the first period of Magyar history. The texts of the H. C. in K and in B are, as we said, almost identical. Such small differences as exist are to be explained by the different individualities of the two copyists. There is just one considerable passage in K (the end of hi$ c. 1 0 ) which is not in B. As V also omits it, one would naturally suppose it to be K’s own; but as we shall see, it is almost certainly to be attributed to the original.6®Apart from this, the only material 6 The Legenda has only one other point of contact with the Chronicle; they both mention S t Gerard (K 56, B 83, Leg. Maj. 13, Hartvic 14) But the Legend cannot have been the Chronicle's source, since the latter contains a phrase “de Rosacio" which is not in the Legend; moreover, the references to Bel are different: the Legend describes the endowment of the Monastery, the Chronicle, Gerard's life there as a hermit. This preoeded the endowment. ® I have quoted V's text here: that of B is hopelessly corrupt. 6* See below, p. 158.

156 38 difference in the two texts is that already noted: that K transfers the list “de nobilibus advenisMto the end of his work, and alters his c. 6 accordingly. We have already seen reason to suppose that in this particular case, K was probably the innovator, while B kept to the original. No sooner, however, do we come to the join between the H. C. and the Magyar history proper, than we find important differences: In c 23 K has “de fluvio HungM, while B has “de Erdelew et de fluvio Hung.**6 A t the end of the same chapter, K has a note of his own on Morot. B 26 gives an account of the vision of Almus' mother, whence that hero derived his name,7 his birth and his genealogy as fa t as Noah. The accounts of the Descent into Hungary differ, B's being much the longer. K26, B 27 begin in much the same way; but K begins with a sentence of his own referring back to the slaying of Sviatopluk (cf. B 54), and ends with a curious fragment of his own. B has a sentence on the 108 tribes who descended from Scythia. K 27 gives Arpdd's descent as “de genere Turul“, and say 9 that, in virtue of his special position, he was the first leader to cross the “Ruthene Alps'* and to fix his camp on the Ung. B 2 S (the corresponding chapter) leaves out the reference to the Turul family; notes that Arpdd's father, Almus, “in patria Erdelw occisus est, non enim potuit in Pannoniam introire", and then goes on with a long and vivid account of the symbolic purchase of Hungary by the Magyars from Sviatopluk; their attack on him, and his death. The end of the chapter returns to wording similar to K's, although rather less laconic. These are important differences, not only of form, but also of substance, particularly since they appear to raise the historically interesting question of the route by which the Magyars entered Hungary. It is therefore comprehensible that the controversy over these passages should have been particularly acute. We may begin with the two longer passages found in K alone: the ends of K 23 and K 26 respectively. The latter is almost certainly K's own work. That it is some-one's interpolation, will not be denied, in view of its almost complete lack of connection with the context. It is most likely K's own, since the words with 7 Alom in Magyar = sleep or dream; dlmos is the adjective

157 39 which the authenticity of the Marot story are denied — “usque hodie fabulose Marot ipsum fuisse asseverant” — are almost exactly those which he uses in another passage when making a similar protest .8 The end of K23, on the other hand, is also given by V and should therefore, if our theory of the relationship between V, K and X is correct, belong to the original X. It fits, moreover, well enough into the context. We have then to find an explanation why B omitted it; but this should not be difficult, in view of what we have to say on B's text. Both Pauler 9 and Kaindl10 regarded B's Transylvanian story as a late interpolation; principally on the ground that K's story, which agrees also with that of Anon., quite clearly makes the Magyars enter Hungary from the North, and B 23 has the same $tory. Dom. on the other hand, asserts confidently that the original text throughout is that of B. K put in the Hung-Hungari etymology in K 25, and he tried here to eliminate all signs of Transylvania; but he betrayed what he was doing by leaving the tell-tale words “et alia sex castra“. These, says Dom., show that in the original K had before him the words “septem castra“, which, again, necessarily implied that the passage referred to Transylvania. It could not refer to anything else, because in c 43 K uses the words “septem castraM to mean Transylvania .11 For explanation as to why K should have spent all this trouble and ingenuity (surely remarkable in a man so foolish and so lazy as he maintains K to have been), he can only suggest that K “dislikes speaking of Transylvania / 1 He offers, however, no explanation why K should have possessed this unique trait, not shared by any other Magyar in life or literature .12 Now, this argument rests on the fundamental assumption that if 7 camps are anywhere mentioned, the reference must be to Transylvania, which is sometimes described in the Hungarian documents and chronicles (including, as Dom. rightly 8 K 40 Quidam vero ipsos a liter damnatas fabulose asseverant. This appears to be an alternative version of the Conquest; or of the attack of the Magyars upon Slovakia (“Great Moravia”) and might be extrem ely valuable if we knew more of it; since it may preserve the truth of the old reproach afterwards levelled at Arnulph, that he “opened the gates“ (i. e., the passage across the Danube at Pest) and let through the abominable people of Gog and Magog. 9 A Magyar nemzet tortenele II 772—3. 10 Studien VII pp 46 (248—9). 11 K. S. M. K. pp 77—8. 12 Incidentally, K does mention Transylvania in hie account of the battle of Kerles (K 63), where B does not.

158 40 points out, a passage in K himself) under the name of “Septem Castra." But to deduce that because this name existed, and could be used to denote Transylvania, therefore it is impossible ever to speak of 7 camps, if one wants to, without meaning Transylvania, is completely illogical and unwarranted. Least of all are we entitled to limit in this respect the freedom of speech of the old Hungarian chroniclers; for the name “Septem Castra“ is not even the original Hungarian name for Transylvania, which is Erdeel, or Erd-alja. The name Septem Castra is undoubtedly a translation from the German “Siebenbiirgen“, and that name again, whatever its origin, did not come into use before the settlement of Transylvania by the Saxons in the 1 2 th century, and was not in general use until the 13th; even up to the 15th, it was most usually used to describe, not the country as a whole, but the German part thereof (the Sachsenboden), and more particularly, the district round Hermannstadt.18 But the 7 Captains, at least, the Hetumoger, did not derive their existence from the German Siebenbiirgen: they are an integral part of the very oldest Magyar national tradition; they figure in X, in Anon., and in the oldest Magyar national tradition of all — that preserved by Constantine Porphyrogenetos.11 And not only the Captains but — and this is the all-important point — the camps also. We draw attention elsewhere 15 to the curious feature in Magyar tradition that in almost every case where it mentions the early Captains, it draws attention also to the foundation of camps by them or by their descendants. This appears both in K’s and B's lists (K 27—33, B 28—34), but even more clearly, as I show elsewhere, in Anon. Anon, is certainly guiltless of the charge of making the Magyars enter Hungary via Transylvania; but he makes all his paladins (or their descendants) — viz., 7 Magyar and 7 “C un’ leaders — found “castra." This takes place, as a rule, after the campaigns which correspond to our Svatopluk story; but some camps are founded immediately after the entry into Hungary. I am willing to believe that B's wording “ibique terreis castris septem preparatis .. . permanserunt" might in fact represent the original text of X, K having made a rather clumsy altera13 Teutsch, Gesch. dcr Siebenbiirger Sachsen, 1925), I. 18. 14 De Administrando Imperio c. 38. 15 See below, pp. 136 if.

4th

ed.

(Hermannstadt,

159 41 lion. But I think it perfectly clear that the phrase, as written by X, had no reference to Transylvania. B has fallen into precisely the same trap as Dom., centuries after him; he believed that “Septem CastraMmust refer to Transylvania, and he interpolated the text accordingly .16 For if we look closely, none of the references to Transylvania are in the least necessary to the story, which indeed is far better without them. The first (c 23) is obviously clumsy; whatever the spot from which the Magyars conducted their negotiations with Sviatopluk, it was certainly one spot, and not two. As a matter of fact, Miigeln's German Chronicle has here a slightly different text: “da komen de ungem von Erdelew und von dem wasset Ung genant.“ This, although it stands alone* is almost certainly correct, since whenever any of our authors mentions the Ung or Ungvar it is nearly always in connection with the special point that this was the place whence the Magyars derived their name .17 Thus there is strong presumption that the original of c 23 ran “de fluvio Hung vocati”, in the light of which the absurdity of B4s addition becomes even more apparent. In the second passage (B 26), to remove the references to Transylvania, even taking the rest of the text from “exinde montes” to “permanserunt” as the original X, only leaves the story the same as K‘s and as Anon.'s; and there is a phrase here which shows that the Ung district is really that meant: the phrase “in confinium regni tiungarie." Anon, uses this more than once, and in each case with reference to the mountain frontier above the Ungvar district, i. e., the south side of the Carpathians ,18 and the H. C. itself, in c 2 1 , uses it in the same sense.19 It is never used for Transylvania. As for the remaining passage (B 28): whatever Almus* fatherland may have been, no one will maintain that it was Transylvania. To say that he died “in patria Erdelew“ is simply 16 This is roughly, Kaindl’s argument also; I have had to expand it, owing to the necessity of answering Domanovszky's later counter"arguments. 17 Cf. Anon, cc 2, 13, etc. 18 c 11 where the frontiers of Kean are said to reach usque ad conhnium Ruthenorum et Polonorum; c. 12, ad confinium H ung. . . ad confinxum Ruthenorum. 19 In KB/21 the Szekelys, having heard of the Magyars' return, go to meet them in Rutheniae limbus (K), in Rutheniafa (B). In spite of B's loose wording (K's text is certainly correct here), this cannot mean South Russia, Since the meeting l s shown as having taken place only after the Magyars crossed the Carpathians; as Anon, also gives it (c 50).

160 42 nonsense, and we should have to strike out the word "Erdelew’* if only to save the credit of our author. Thus all the mentions of Transylvania can and must go as simple interpolations by B. But B has three more considerable passages not in K: the Almus genealogy (B26), the passage about the eagles in the same chapter (deinde transierunt. .. descenderent in Hungariam) and the long Sviatopluk story in B 28.. As we have shown the references to Transylvania to be in depen-, dent additions, without organic connection with the story, it does not follow from our deletion of them that we should let the whole Sviatopluk story go as well. Nevertheless, we must dismiss all these three passages also as interpolations, although on different grounds. It is obvious enough that the first and third of these have been interpolated by somebody. While K's text in cc 24, 25 runs on quite smoothly and logically, the sudden digression to the birth of Almus which separates the corresponding passages in B altogether breaks the thread of the narrative. It has no connection with what precedes or follows it, and its chronology is, of course, quite incompatible with that of the H. C. The H. C. makes Edumen, son of Chaba, and grandson of Attila, already return to Hungary; while according to his note, Ed's brother was the greatgreat-grandfather of the man who returned .20 Furthermore, both K and B, who in this case agrees exactly with K, most explicitly make Arpad the first Captain, who leads the Magyars across the Carpathians. In that passage, they ignore Almus altogether, except to mention him as having been Arpad's father. When, therefore, B explains that Arpad was the leader because Almus had died on the way, it is fairly obvious that he is inserting this note also, in order to explain away the contradiction in which his insertion of the previous remarks on Almus might appear to have involved him. Two-fold interpolation by B' is surely infinitely more probable than two-fold omission by K. Even if K had thought that he saw reason to omit the chapter on the vision of Almus' mother, this would be no reason whatever for him deliberately to leave out the harmless little remark about his death. The long Sviatopluk episode in B 28 is no less clearly out *° According to the strict text of the H. CM the genealogy would be wrong too, but the sentence in KB 23 Sed vero in Scytia remansit apud patrem is certainly an interpolation, the ex « to enim, etc.f referring to Edumen: and not to Ed. This, particular difficulty is therefore only apparent.

161 43 of place where it stands. Its abrupt intrusion completely breaks the thread of the clear and reasonable note on Arp&d’s ancestry. Moreover, the Sviatopluk story has already been told, in summary, in KB/23, and according to K’s text (K 26, tandemque Sviatopluk interempto), is already over and done with. It is true that we need not suppose the author of the H. C. to have been a born story teller. In the confusion of making the join between Him and Magyar history, he might tell the same story twice. But we must observe that the story in KB/23 is not quite the same as that in B 28. The former makes Sviatopluk meet his end in a fortress West of the Danube, the latter, by drowning in the Danube as he fled before the Magyars. Thus it is certain that either B, or X before him, inserted this second version of the Sviatopluk legend; and it is far more likely that it should have been B. If X had the two versions before him, it is difficult to suppose that he would have adopted the clumsy method of first summarising the one, and then inserting the other, at length but out of place, later; while such interpolation would be the more natural course for B, (who on principle never leaves things out) when confronted with a text in the form of KB/23. By this means, too, we avoid the necessity of supposing K to have cut his text here — a task which if he did perform it, he performed with some skill, cutting out precisely the irrelevant matter. One further point arises in this connection: K has the words “tandemque Z. interempto, quemadmodum superius est narratum" at the beginning of K26, L e., before the division into armies, B has in the corresponding passage, “cumque in eis castris permanerent, irruptionem circumquaque jacentium dominorum forinidantes deliberate communi consilio” — they elected seven captains. The words “postquam autem Zuatopolug per Ungaros, ut superius dictum est, necatus” come much later, only in c 54, and after the division into armies, the list of captains, and even after the list “De Nobilibus Advenis." Now, firstly, it is in itself far more likely that B should have postponed the phrase on Sviatopluk’s death, than that K should have brought it forward, for it is a very rare thing for a copyist to take in a phrase in advance which belongs several pages ahead in his text. Secondly, while there is no particular reason why K should have made this alteration, there is good reason for B to have done so, since he is giving a full narrative of the death of Sviatopluk a f t e r the point at which the phrase occurs in K. And thirdly, I think that K’s

162 44 order is historically the more. nearly correct. In B 28 itself, the campaign against Sviatopluk is clearly imagined as being carried out by the Magyars as an united body. More convincing still is the parallel from Anon, who, as we shall see later, has preserved a parallel version of the tradition of the Conquest. Anon., it is true, has his 7 Captains who elect Almus their leader in Scythia; but as we shall see, these are not necessarily the same as the 7 leaders in Hungary. Actually, he has the same order of events as K. The Conquest is completed, and only after this — in Anon's narrative, actually after the death of Arpad and the succession! of Zoltan — do we find the phrase “alios autem const!tuerunt ductores exercitus, cum quibus diversa regna vastarent, quorum nomina hec fuerunt: Lelu filius Bolsu, Bulsuu vir sanguinis films Bogat, Bonton filius Culpun“. This is the list which is the true parallel to X’s list of the Seven Captains. If B’s phrase “irruptionem circumquaque jacentium dominorum formidantes” belongs to the original, as I suspect, then it almost certainly refers to the later period, after the conquest of Pannonia is completed. The whole narrative, including B’s own alternative account in B 28, obviously regards the stage before the campaign against Sviatopluk as only temporary, a matter of spying out, to be followed at once by a new advance. The time for fortifications and defence would come later. There is still one difficulty in K’s text: that the death of Sviatopluk (K26) comes between the foundation of the camps and the election of the generals. Thus K also appears to imagine camps to have been built at the first stage. I believe this to be due to the difficulties into which X got himself in connection with the end of the H. C. The remaining passage which must be regarded as an interpolation by B consists of the lines in B 26 “deinde transierunt. . . descenderent in H ungarian!W herever and however the Magyars crossed the mountains into Hungary, it is at least certain that they crossed them only once, not twice. It requires therefore no great perspicacity to see that these sentences represent an alternative, parallel version to those which follow them. The second version is certainly that which corresponds to K s lines in K 23, as the parallel phrases show: the 7 camps, the phrase “confinium regni Hungarie", and certain phrases which recur also in the H. C. (invitis gentibus memoratis; cf K 8 , invitis gentibus praefatis; pro uxoribus et rebus suis conservandis; cf. KB/8 , cum uxoribus etenim, tabemaculis et bigis descender ant de eorum terra), B’s sentences

163 45 "deinde transieruiit , . . et equoe” on the other hand, with their legendary character and pious style, recall the tones of the Sviatopluk episode in the following chapter. It appears to me that all these three fragments — deducting, of course, the latter part of Almus' genealogy, beyond the point to which K also takes it — the rest is pure thirteenth century fiction, compiled in the clumsiest and most transparent fashion*2 — all the rest belong to a single whole, being an alternative version of the early Magyar tradition, That they are at least as old and respectable as X's own version, there is no reason whatever to doubt; the Almus legend recurs in Aiiom, (c3), the Sviatopluk legend itself, in a slightly different form, in die H. C. itself, and (adapted) again in Anon. Only — it happens to be an alternative story, which B combined with X’s version of the same events. In the chapters which follow, up to the reign of St. Stephen, the texts agree closely enough, and where they differ, by no means bear out the suggestion that K was a lazy or careless writer. He certainly omits the legend of the%death of Lei (K40, B60), but he does so, not out of idleness, but out of a critical sense. He supplies another version, taken from certain ‘Tibri cromcarum" which, alas! have never been identified. Some further material in his cc 39—41 seems to be taken from the same source. Whether he left out, or B put in, the curious note on the Seven Disgraced Magyars (B 35, 36) is a matter on which it is absolutely impossible to judge.23 His note on Verbulchu (K33J appears to be his own. Each text has, as we said, variations of its own on the list “De Nobilibus Advenis", B adding (B 53) some words from the lost source of the L. C. Now, however, we come again to passages which require very careful consideration; and it will be necessary to skip, for the time, over the lists of the Seven Captains and to turn straight to those chapters which deal with the life of St. Stephen. The latter half of K43 runs as follows: Sanctus namque rex Stephanus coronatus et tandem duce Cuppan interfecto, Iula avunculo suo cum uxore et duobus filiis 22 The line Ed—Chaba—Attila—Bendekuz is taken from the H. C. itself; the other names are taken at random from Hun and Hungarian history. 28 Kaindl VIII. 51 (253) argues that this passage could not have been in. the “Gesta" because this represents Toxun’s reign as peaceful. This is not convincing; but the passage seems to me to have come from one of the manT y variants of the national tradition, probably omitted by X.

164 46 de Septem Castris in Hung&riam adducto, et adjuncto Septem Castris Pamnoniae, post hasp cum Kean Buigarorum duce et Sclavorum praeliatus est. Quo devicto de ipsius thesauro Beatae Virginis ecclesiam de Alba dicare non omisit, quam fundasse perhibetur. These laconic words correspond to the whole oi cc 64—6 inclusive in B; and to more: for Miigeln has also a chapter an the coronation of S i Stephen which must have been in the original.24 Obviously, K s text here represents an extreme condensation of a text which must have been much the same as B's (including the chapter from Miigeln); for the few facts which it gives are precisely those given in B's text — one to each chapter. Moreover, certain other phrases in K refer back to the missing m aterial .25 It even seems certain that K's text is a summary of one practically identical with that of B, for the following curious reason: K says that B waged war against the Gyula, and after that, against the Kean of the Bulgars and Slavs. "Quo devicto, de ipsius thesauro Beatae Virginis ecclesiam de Alba dicare non omisit, quam fundasse perhibetur." A hasty reading of B 6 6 would in fact give this impression, because that chapter begins with the story of the campaign against the Kean, and goes on with the foundation of the church at SzeIcesfehervar. I believe, indeed, the impression to be still general in Hungary today (Homan adopts it in his "Gesta"). Nevertheless, if we read the chapter carefully, we see that the foundation of the church must have been from the Gyula's wealth, for the following reasons: 1 . B says explicitly that the church was many 24 M. c 18. We oam hardly suspect M of having drawn on a separate source, i. e., one of the legends of St. Stephen, for be never does so in any other passage in this earlier part of his text. Further, both K, and B in his corresponding words “Porro Beat us Stephanus, postquam regie celsitudinis coronam divinitus est adeptu s. . clearly refer to the missing description. 35 K 78 (List of Noble Strangers): Hunt et Paznan. . . sanctum regem ~Stephanum in flumen Goron Theutonico more gladio militari accmxerunt. B 41 (list) Hunt et Paznan, qui Sanctum regem Stephanum in flumine Goron Theutonico more accinxerunt. B 46 (narrative) Sanctus autem Stephanus. . . ad amnem Goron primitus accinctus est gladio ibiqu e. . . duos principes Hunt el Paznan constituit. K 79 (list) Vecelinus. Hie Cupan ducem in Simigio interfecit. B 40 (list) Vecellinus, qui cum Santcto Siephano in Symigio Cupan ducem inter fecit. B 64 (narrative) In eodem autem proelio Wecelinus comes inter' ie c i l Cupan duoem.

165 47 times burned down, because the Gyula*s money had been ill-gotten; he had the Gyula and his family baptised in it, obviously because this was thought to be appropriate; 3. the Kean's riches were used to found another church, that in 0-Buda (B 67). The first lines of B 6 6 , down either to “devicit et occidit" or to “pretiosis lapidibus" (almost certainly the latter) must therefore have got misplaced in the text of B; and they seem also to be misplaced in K's summary; which seems to show that the text summarised in K, was identical, a t least in arrangement, with that of B. Nevertheless, B says specifically that he proposes at this point to insert material omitted by other writers.28 Who, then, were those other writers? We are agreed that B worked on the text of X, and not of K; therefore he must surely be referring to X here. These words can have no meaning, except that X here left out (or rather, summarised in a line or two), material which B reinserted. And we find very clear indications in more than one other passage elsewhere. K 29 has only a line or two on the Gyula, the “Third Captain1'. B 30 replaces this by a long story of how the Captain Gyula discovered a Roman city in Transylvania. He had a beautiful daughter called Sarolta, who became S t Stephen's mother. Then the Gyula molested the Magyars in Hungary and was taken prisoner by St. Stephen — but not the first Gyula, but the third. Anon, has 3 references' to the same story. In c 24 he says that Transylvania was discovered by Gyula’s grandfather, and his descendants would have continued to hold it, had not Gyula and his sons, Biua and Buena, refused to accept Christianity, and molested the Magyars. In c 27 he repeats this, with a full genealogy of the family; and in c 26 uses practically the same words as B 65 to describe Transylvania. The passage in c24 summaries that in c27; but the latter, with that in c 26, practically add up to B's story in cc 30 and 65 combined, the two being clearly parts of one whole.27 But there is no possible reason why K should have omitted the description 2.

36 Nos enim ea potius, que ab aliis scriptoribus praetermissa sunt, breviter ac summatim s crib ere intendirous. 37 B's reference to Sarolta in c63 may come from the same source; or it may be interpolated from *Co.

166 48 from the List of Captains, merely because he had abbreviated the story of St. Stephen's exploits; clearly, B, in replacing the material of his c 65, took that part of it which referred to the earlier Gyula, the Captain, and put it where he thought that it belonged. In doing so, he left an awkward join in his text of B 65, which reads like a duplication ;28 but if we imagine the note from B 30 inserted in c 65 between the wards "gubernaeula possidebat” and “anno itaque", we get a smoothly running text, as well as one almost identical with that of Anon.29 There is one further passage where B has again re-inserted matter from these ohapters, and where it is beyond reason to suppose that K carefully omitted this, because he had left it out in his chapters on St. Stephen. In c 77, in describing Peter’s coronation, B speaks of the “sacris insignibus sancti regis Stephani" — a clear reference back to the account in c 6 6 of two “racionalia" sent by the Pope, for coronation purposes, to St. Stephen. B 77 is based an the Annales ALtafaenses, but that particular phrase does not come from the Annals .30 It may be objected that it is a remarkable coincidence that B should be able to lay his hand on the material which X omitted. I do not think so. We must get it right out of our heads that there was only one “Gesta". X is only one history of quite a number. What is unique about his work is that he produced the Hun. Chronicle, and a certain version of later events, which K used as practically his only source, and B as his main source for the early period. There certainly existed other versions of the early national history, no less important. I have not included B 67 in the above. Grave doubts appear to exist31 as to the veracity of the account that St. Stephen founded the Church of O-Buda at all; also the summary in K takes no account of it and it contains words (morose - carefully, suido, bysso), not found elsewhere in X or B. It may well be an interpolation by B from an entirely different source. .. . bellum gesoil contra prooavunculum suum nomine Gyul&m, qui tunc temporis totius Ultra Siluam regni gubernaeula possidebat. Anno itaque Domini M il beatus rex Stephanus cepit Gyulam ducem cum uxore et duobus filiis suis et in Hungariam transmisit. 38 The date may perhaps come from another source again: the Annales Altahenses; see below, p. 49. 30 Possibly also the reference to the frequent fires in the church may be an addition by B; cf. B 203, describing one such tire. 31 J. Karacsony, Peter kiraly 6s az obudai pr6postsig; Sz&zadok, 1897. pp. 291 ff.

167 49

The use oi the German sources. No part of the Hungarian Chronicle has been discussed more fully than that in which it stands in a certain relationship to various German sources, and in particular to the Annales Altahenses; since not only Hungarian scholars, but German also, have interested themselves in it, the latter hoping to find some light cast upon the composition of the A. A. K 43, B 68 both mention an event — the defeat of the Gyula by St. Stephen — which the A. A. also record (ad ann. 1003). B alone gives the date, and that a year different from the A. A. After this, although the A. A. have one or two remarks about Hungarian affairs, there are no signs that the Hungarian Chronicles have in any way borrowed from the German source, until 1041; with the single exception that B 68 gives under the heading "Incidentia de aliquibus” a note on certain natural phenomena which is undoubtedly taken from the A. A. ad ann. 1020, 1021.81a Ad ann. 1041, however, the A. A. begin a long narrative, clearly from a single, special source, on Hungarian affairs. From now on until 1046 the resemblances between the A. A. and the Chronicles, particularly B, are extremely close; and there is no doubt that, while the original source of the Annals' own information was very likely, at least in part, a Hungarian one, the Hungarian Chronicles have themselves used, as one of their sources, a German text. Many phrases which only a German would use are taken over verbatim; in some cases with the outlook, e. g. where gratification is expressed at a German success. Every imaginable combination has been suggested to explain the mutual relations of K, B and the A. A. over these years. Some of them may at once be dismissed, as having been effectively disproved. Into these we need not go, contenting ourselves with recapitulating the facts on which there can now be no disagreement more. 1 . It is certain that for that narrative from 1041 on K and B draw on a common source, parts of which seem to represent a truely national tradition, while other parts are rather Germanic in outlook, and are certainly in some way related to the Annales Altahenses being drawn either from the Annales themselves, or from some derivatory or source thereof. 2 . In the passages derived from this source, and common It may be noted that the early part of the A. A. is particularly rich in information on fires, floods, earthquakes, etc.

168 50 to both K and B, B is usually the nearer to the text of the A. A. but sometimes K is the nearer. In other cases it seems that K's text could only have been reached through the medium of a text closely resembling B's. We are therefore forced to agree that we are here once more in the presence of our old friend X, from whom K and B alike derive. It does not, however, follow that X was B's sole source. Domanovszky concludes that B's text more faithfully represents that of X, and that K's is an extract therefrom, on the following grounds: 1 . Because the fuller text (B’s) has borrowed more from and is regularly (Dom. says “always", but that is not strictly accurate) nearer to the original text of the Annals. 2 . Because K has no independent borrowing of importance from the Annals. 3. “Because in other passages also K follows the order of events in the Chronicle." 4. Because certain corrupt passages in K's text are only to be explained through the medium of the fuller text .32 None of these arguments however, in the least excludes the possibility of an original use by X of some German source, followed by an independent collation of the text by B. with the A. A. themselves; some of them, indeed, and notably nos. 1 . and 2 . are rather arguments in favour of such a collation .83 We must now examine the case for or against this. Dom. is not the first to contend that in certain passages of the present part of the Chronicles, those passages, to wit, where K and B agree fairly closely, neither copied from the other, but both worked on the common third source X. The same conclusion was reached by Kaindl34 and even before him, by Rademacher.35 But both these writers, as also Zeissberg,86 were struck by certain facts which led them to postulate a further recollation of the text by B. They differed, indeed, from each other, and I cannot agree exactly with any one of them, on the exact relationship between the various texts; but the main outline of their argument 31 Dom., K. S. M. K. p. 104. 33 I confess that I do not understand what no. 3 means. 34 Studien: VIII. passim . 36 Zur Kritik der ungarischen GeschLahtsquellem; Forschungen zur deutschen Geschichte, XXV (1885) pp. 385 ff. 36 Zur Kritik der Annalen von Altaich; Zeatschrift fur die oesterreichischen Gymnaftien, 1875 pp. 497 ff.

169 51 seems to me unassailable, and I set it out herewith, modified by certain considerations of my own. As we have already pointed out, K’s text undeniably contains two elements. There are passages of pure Hungarian national tradition; and there are passages where that has been blended with a German source which stands in some close relationship with the A. A. Examples of the former are K45 on the death of St. Stephen, K 46 on the beginnings of Peter’s reign, the opening of K47 on the first conspiracy against Peter, the end of K49 on St. G erards sermon, or the end of K50 on the flight, death and subsequent disinterment of Aba. Now, all these passages are given almost as fully, and in almost identical words, by K as by B .37 On the other hand, in the passages connected with the A. A., K is not merely a little, but very much shorter than B. Thus he has only 3 lines (K 51J to correspond with the whole of B 77, in which B practically copies out the Annals. He telescopes (at the end of K 48) the campaigns of 1042 and 1043, whereas B gives the latter, again, with very many close verbal reminiscences of the A. A. He omits entirely B 6 8 , which comes straight from the A. A., and also such passages as that on the dust-storm in B 76 which is, again, taken straight from the A. A., and introduced by B with the words “tradunt *etiam Teutonici.” Even where the two texts relate the same events, then if those are events also recorded by the Annals, B has little addi~ tions apparently taken from the Annals, and his text is in genera^ closer to that of the A. A. than is K’s. One may take for example the case of Buda. The A. A. mention him ad ann. 1041 as P eters councillor and evil genius, whom the Magyars insist on punishing. B mentions him no less than four times: in the corresponding passage in words almost identical with those of the A. A .38 and three times more (in B 70) in almost identical terms. K never mentions the name .40 87 On the proof of re-collation Kaindl, in particular, is excellent (Studien VIII. 10 = 212 ff). As, however, w ill seen, I cannot accept his view that X ever used the A. A. directly. 38 A. A.: quendam illi fidelecn, nomine Budonem, horum omnium malorum auctorem, cuius omnia fecerat consilio; B 72: scelerati6«imum Budam barbalum, omnium malarum intentoienx, cuius consilio Petrus Ungari&ra afflixerat. 40 K 44 = B 69 has habito consilio infidelium; K 47 = B 72, ex quibus unum.

170 52 Now, it is almost incredible that K should copy out, quite faithfully, the text of X where this follows the national tradition, and shorten it so drastically where it happens to be drawing on a foreign source. To believe this of him, we should have to attribute to him, not only a very perfervid Magyar national sentiment (in which case one would expect him to go further, and not to admit any foreign phrases or sentiments at all), but also an extreme and almost unbelievable sharpness of eye. How in the world should he be able to distinguish so accurately the two elements in the text before him? But much more than this: if, disregarding B's text, we take only that of K, we find, not only that it has only a very limited number of passages in which it agrees at all closely, verbally, with the A. A.: but the course of events described also differs very substantially. K 44 makes Gisella the authoress of the blinding of Wazul, while Stephen advises the sons of Zar Ladislaus to flee; in the A. A., Stephen has Wazul blinded and the others exiled. In K 45 Gisella makes Peter Stephen's successor; in the A. A. he is Stephen's own choice. In K 46 the cause of the dissension between Peter and his subjects is the king's partiality for Germans and Italians, and also his lascivious conduct; in the A. A., his ill-treatment of Gisella. In K47, the Magyars seek for someone to liberate them. They elect Aba, who marches against Peter. Peter flees, whereupon the Magyars take his chief supporters, including one who, from the description, must be Buda, and torture them to death. In the A. A., the Magyars ask for Buda to be delivered up to them. Peter objects, but they take and torture him notwithstanding. Thereupon Peter flees in terror, whereupon the Magyars, seeing that he is gone, elect Aba in his place. k 48 ff.: In the third year of A bas reign Peter advances (descendit, the regular word always used for “comes down from Germany") with the Emperor Henry and a great army. Aba sends messengers to Henry and receives a hostile reply. Seeing that Henry means to invade Hungary and restore Peter, Aba invades Austria. The Emperor at Cologne, hearing that Aba is “molesting the Germans without reason" decides to invade Hungary in revenge. He reaches the frontier, but the messengers “of King Aba and of the Magyars" promise to give him every satisfaction, except

171 53

that of taking Peter back. The Emperor accepts gifts and receives a promise that the German prisoners shall be restored, and returns to Besan^on. Aba, however, now thinking himself safe, commits sundry acts of tyranny against the Magyars. They conspire against him, and some conspirators escape and appeal to the Emperor. He then marches into Hungary, defeats Aba after a fierce battle in which innumerable Germans are killed; (Aba also perishing in his flight) and restores Peter. According to the A. A., Henry at once (i. e., immediately on Aba's succession) receives Peter graciously. Aba sends his ambassadors to him at Strasbourg to ask whether the Emperor means peace or war. Henry replies that this depends on Aba's own conduct. Nevertheless, Aba gets in his own blow first, and invades Austria. Thereupon Henry marches into North-Western Hungary. The inhabitants offer to submit unconditionally; only they will not have Peter back as king. Henry thereupon installs a nephew of St. Stephen's in nine of the local towns, and retires. The next year (1043) he prepares for another expedition. Aba sends ambassadors offering to restore the prisoners whom he has taken, and to pay ransom and reparations. Henry refuses to treat until he has reached the frontier, where he is met by another ■embassy, offering even better terms, which he accepts, and concludes peace with Aba. In 1044 there is a conspiracy against Aba in Hungary, which is betrayed. Some of the conspirators escape and appeal to Henry. He hesitates, on account of his recent treaty, but at last yields and marches against Aba. After a great battle, in which very few Germans, but innumerable Magyars are killed, he triumphantly xestores Peter to the throne. K52, 5 3 describe Peter's renewed acts of tyranny against the Magyars, their decision to bring back the sons of Zar Laszlo, and the rebellion against Peter, who is captured outside FehSrvar and blinded. The A. A. have a whole episode ad ann, 1045 recounting a visit by the Emperor to Hungary, where Peter does him homage. The conspiracy is recounted in language very different from K's, ad ann. 1046. It will be seen that K's story contains as much national tradition, as German. The latter is used only as a subsidiary source, the result of which is, that it is distorted to no small extent. Nevertheless, the story which emerges is on the whole connected and reasonable, although not to be reconciled exactly with the

172 54 account of the Annals, and except in one or two passages only not even resembling it very closely. Those passages in B which do not correspond with K, on the other hand, are absolutely straightforward summaries of the A. A In all of them, I find only one alteration of the sense of the A. A.: where it is stated (B 77) that the Emperor granted the Magyars “Hungarica scita” ; the A. A. writing “scitaTeutonica.“ The A. A.'s seems to be the correct version, since another chronicler in the corresponding passage, writes “Ungarios petentes lege Baioarica donavit"41 Nevertheless, it is possible to suppose that B’s alteration was in this case less the result of deliberate policy than of genuine misunderstanding, particularly as the preceding word in his text is “Ungaris.” Otherwise these passages, m sharp contrast to the others, not only adhere closely to the wording of the A. A., but make no attempt to modify its story, either as regards the substance or the spirit. I am bound therefore to agree with the three German scholars mentioned above that by far the most natural explanation for the relationship between the two texts is that K’s represents a combination, by X, of national with German tradition, in the course of which the latter has been very considerably modified; while B’s text shows traces of a further direct re-collation of the German text. There are certain cases in which this seems particularly clear: where we find very clear traces of the re-collation and interpolation. We have already mentioned the case of the name of Buda; it is far more likely that B should have inserted the name each time, than that K should regularly have gone to the pains of leaving it out; particularly as in c 69 B seems almost to have lost his head, and Buda figures no less than three times: as Stephen’s own messenger, as Gisella’ wicked councillor, and as the father of Sebus, Gisella's messenger. Kaindl raises another neat point: in connection with the campaign of 1042, K has ’’invasit Austriam”, which is sense; the A. A. have “terrain Baioarium“, which is sense also, since Austria at that time formed part of Bavaria, and is regularly so described by the Annals ;48 but B has “Austriam et Bavariam’V 41 Herimannus Augiensis ad. aim. 1044. 42 Studien VIII p. 28 (230), 43 The A. A. only use the term "Austria” twic, and that in their early summary part (ad ann. 1012, marchione Henrico in Austria; ad ann. 1018, Henricus marchio Austriae periit).

173 55 which is nonsense, and can only be explained on the assumption that he is combining two texts. Further, K, and B in agreement with him, make Peter’s first “descent” upon Hungary, in the company of the Emperor, take place in the third year of Aba’s reign, which would be 1043 or 1044. This is, of course, too late, since the A. A. show that Peter got into touch at once with the Emperor (1041), and Aba’s first embassy, the hostile or ambiguous reception of which led him to invade Austria, must be dated 1042. Nevertheless K, by telescoping the campaigns of 1042 and 1043, succeeds in making a story which is not indeed accurate, but reads all right. B, on the other hand, corrects this story by inserting, at the beginning of B 75, an account of the campaign of 1043, beginning with the words “sequenti anno.” This does indeed make the account historically more accurate; but he fails to notice that he ought in that case to have altered the earlier phrase that the trouble began in the third year of Aba’s reign. In connection with the same passages, there is one phrase which proves the interpolation through re-collation especially clearly. K, as we said, telescoped the campaigns of 1042 and 1043. It is. not very cleverly done. In actual fact, if we may go by the Annals, Henry in 1042 never came near Aba. He invaded only a corner of the country; it was the emissaries of the people who met him, and the point at issue was whether they would have Peter back as king, which they refused. In consequence, a nephew of St. Stephen’s was put in. These negotiations were conducted exclusively with the local population. Aba obviously took no part in them: Henry could not have debated with him the question who would be king of Hungary. With the local population, on the other hand, there could be no question either of presents or of the return of prisoners; it was not they who had taken prisoners, or who owed reparation to Gisella. These were the subjects discussed in 1043. K, combining the two, makes the ambassadors come “from King Aba and from the Magyars” and not only discuss the question of the king, but also those of gifts and the return of prisoners. So K’s text ends up that Henry, after receiving gifts and a promise that the prisoners would be returned “consilio inductus ducis Loteringiae et plus allectus muneribus” returned to Besani* II et genuit unum {ilium nomine Ladieleum et Hliam Mariam, qua* fuit data in matrimouium Carolo CUmdo regi Sicilia*, de qua genuit Carolum Martollum. Qui quidem Carolus Martellus genuk Corobertum, poet: Carolum Robertum, quern post** Hungari vocaront Carolum.

Martallum vocabulo. Carolus w o Martallua a fills, imp* tratoris Rodolphi Clemencit nomine genuit filium, qnom piimo in to m sua vocoron al Corobortum, quasi Carolam Robertum. hxHungaria is ta n oblato Robarto ipeum Hungari Carolam vocarerunt.) at mortuus aat in anno tertio regni aui in Magna insula at aapultns ast in acclaaia Baata Virginia in inaula Budanst in loco baginar-

migrant tandem a s boc aeculo at in insula vocata Baataa Virginia in ooenobio monialium raquiasoit tumulatus.

Cui suocasait Ladislaus fil- Post hunc ragoat Ladislaus ioa ajus at coronatus ast tartius filius ajus.

at obiit A. D. 1278 id. Aug. cujus oorpua requieedt in insula Budaosi in sodasia Baata Virginia in loco Baginsrum. Hie babuitunum filium, at ducam luidialaum tandam ragam. Huio succaasit Bliua ajus.

Obiit rax Stapbanus options prinoapa a. D. MCCLXXIV. anno tartio ragni aui. Sopultus oat in aodaaia Baata Virginia in insula Budansi in loco Beginarum. Succaasit filius ajus Ladislana at coronatus ast oodaa Ladislaus anno, quo mortuua ast patar Anno dni MCLXXV1II hi ajus, ridalicat a. D. 1280 I adialsns sain run konig

aodam anno, quo mortuus (From this point bis own — ast patar afus, vidaHost account. It is, bowsvtr. Qui prius Otbocaro raga 1272. Qui rax Ladislaus a. notiosablo that it dosoribas Bohamia ia prsallo iatarD. 1276 only tba two ayants ralated ampto imparators adjuvants by W.). — circa Moreviam occidit ragam Otbocurum supradictum in prsalio. at Cumaaoa dai adiutorio Postsa cum a. D. 1282 ririlitar in bello campastro Oldamir dux Cumania conouparitla ate. gragato axardtu Cumanorum (So# below) droa lacua Hood vooatum solans boatilitar ragntun inradars Hungarorum, nt suo dcaaiaio sdbjugarat, rax Ladialaus at fortis Josos pro gante sun at rogno pugnaturua annssiit. . . rictoriam obtinoit dhrino fratus anxilio. Huius tempore Tartari inTandam poataa pauci do atinetu Cumanorum, qui ipsia qui avasarant, avmaarant, at aa in Tartarian! ad Tartaroa fngisutao quoreoaperant, aacnnda rica rum instinctu Tartari a. D. anno adlkat a. D. 1285 1285 aacuada rica in HunHungarians invaserunt at uagarian! adveaeruat at usque qoa Paatb misarabilitar omPssth universa miaorabilitsr nia cromararunt* coabuasarant lata rax lata saim rax Ladislaus rsgis da filiam Karoli da Apulia in batrak. coaiugium 1 Va text. B baa at Bui garoa enparans. * Saa W i te x t Stephanu s. . . hsbuit filias tras, ax quibus una rocabatnr Maria. * In Wa taxt this paaaaga comas altar Ladislaus' death.

dor hatis ayn otrait mit dam Ottokaro konig can Boaon and dinshtg yn, abir Rndolpbns Rondacbar konig dar bnlff mit seinam gro*san folk, dam kooig Ladis-

is filiam oonjugso

B hababat. Sad spreto^ thoro conjugal) filiabns adhaaait Cemanorum, Eduam, Cup* •loch et Mandulam ntciU* ac alia* quamplurcs io con* cubiaaa hababat, qoarum amora cor ajui eat da* pravatum at a tuia baronibus at ragni nobilibus odio bababatur. Hinc iaaupar contra earn, qttia Comaoice at non eathoMoa convereabator, Philippas Firmanue aadia apoatolica legato* adinvenit, qui barbaa radara, crinaa detruncara oontra moraa Httngaricoa at pillaos Ctunanicoa. qsoraa u n i in Hangaria iaa in consuetudine habab* ator, abicara damandabat, regam attam anatbamath ▼inculo fariaoa, ut pagaaoa odkat, ritnm Cbriatianornm dibfarat at tboro vivarat oootagalL Sad nicbil in raga proficient rupatriarit. Poat bac in brari ipaa rax a. D. 1290 faria m e. prox. aata feetum S. Margaratba Virginia at martyria prope oaatrom Kanaazag ab ipaia Cumania, quibua addseearat, aat mbsrabilHsr intarlactna Tempore eeim iatina ragia Ladtalai oapit Hangaria a ana magnilioa gloria reflect) in caaua flabilas at event n^ Caparant in aa taatina baila oonanrgara, caaitataa confriagi, villa par conbnationaa ad niohikan radigi, pax at caaoordia panitna cooctdcari, divites deficera at nobilaa maticari pra iaopia pauper* talk. Illo taapora biga, actboat dnarum rotarum vehbnlua a ragni incoba onorn a ragia Lndizlai dnoabatur, quia propUr coat inna apoba animalia vahicnla trabentla dafaoarant in ragno, aad homines oora pacorum bigia paadabant In quo raga, quia aina barada ^gcaaait, dafadt aaman maemHnum aanctornm ragum (Codd. A, B, D). Ragaavit autam 18 annia at dim. (Cod. A) (Data above) Idaoqua rax Ladblaus aapnltna aaaa dicitnr Chans dini (Codd. B, D)

K

W

Co.

L. C.

Sad aprato jugo conjugali adhaaait filiabua Cumanorum Ayduam, Cupcbach at Man* dulam vocatia ac alias quampluraa in concubina* hababat

Diaar konig wart durcb dan lagatum ganant FBrmanua in daa pobiat bannan galagit, dorum daz das bar nicht cristanlicb in dar aa labit, anndar bar batta zuschaifaa mit dar knn tocbtar,









iaa dam par ipaoa Ctmmnos at aic fuit occiaua propa caatrum Kowaaxag aat a Cum anii. intarlactna Tampers iatiu* Ladialai ragia inoapH Hangaria raflacti a aua magnificantia in caaua Habile.

Dornoch wart har dorch dy kun dy har allacxost libat dQnrticlicb gatiigit. In dam Jorx MCCXC.

In dar selbin cxeii eoUtund gros krig in daa Unda, a bo daz dax lolka alao arm wart, daz aa nicht batta pharda noefa fy do mit aa batta mogan arbitan anndar ty muaten selbyoder aicb anapaaea, and dy wagm ezin, «i»wi dornma dy karran mit erwan radan konig Laalaa wagan ganaaL



in quo raga, quia aina htrade dacaaait, dafadt aaman mascul inum aanct* orum ragum. circa XII annum ragni nd, Ragaavit rax Ladiebue an* anno sc. MCCLXXXX nb 17. Obiit anno 1291 Sepokua aaaa dicitur Chamo* dim

To save apace, I do not produce the following chapters; but the addition aum is just as accurate for B 186, 187, which ia exactly compoeed from Co. + L C., with aoma additions (Andrew lll'e genealogy aad bis wsr against Austria) the souros of which cannot be traced. The composition of the last chapter* of B b dbcuaad on pp. 31—34.

XI The Hungarian Texts Relating to the Life o f St Stephen

In the following pages I discuss the sources, dates and mutual relationships of the various old Hungarian texts which have as their sole or principle subject, the life of St. Stephen. With this essay I conclude my series of special investigations into the Hungarian Chronicles and similar texts of the Arpad period;1 for the remaining texts with which I have not dealt (e. g. the Carmen M iserable, the Legenda' SS. Zoerardi et Benedicti) present no problems calling for special investigation of an argumentative character. All that I think necessary to say on these texts, together with the results of my more controversial articles will, I hope, appear shortly in a comprehensive work on the subject now in course of preparation. There are five texts to which the description given above obviously applies: the Legenda Minor and the Legenda Maior of St. Stephen; Hartvic’s Life of St. Stephen; the PolishHungarian Chronicle, and the Codex Ossolinski. For reasons which will appear, I add a note on the Annales Posonienses. The first question is that of the respective dates and the mutual relationship between the two Legends of St. Stephen, and Hartvic’s Life. Marczali’s old theory2 that the Vita was the original, and the two Legendae excerpts from it, has long been abandoned. A part from the fact that excerpts were not made in this fashion in the Middle Ages, there is another quite definitive proof. Long ago, Kaindl pointed out3 that there must have 1 Studies on the earliest Hungarian historical sources; I, and II. Bp., 1938; III, Bp. 1940. Pascua Romanorum, Szazadok, Bp.. March— April, 1940. - Ungarns Geschichtsquellen ira Zeitalter der Arpaden (Berlin, 1882), p. 18. :J Studien zu den ungarischen Geschichtsquellen I (Vienna, 1894).

352 6

existed a text of Hartvic earlier than, those which today bear his name. For comparison with the P. H. shows that the excerpts from Hartvic contained in that text agree with our text of Hartvic where the latter follows the Leg. Ma.f and in many cases where it adds independent information, but that they show no trace of the Leg. Mi. It is true that Kaindl worked on certain false premises which vitiate some of his arguments. He believed, for example, that the Pest Codex of Hartvic (BI) was the original whence the remaining texts derive. Miss Bartoniek has had no difficulty in disproving this4 and in showing that the transcriber of BI re-collated his text of Hartvic with the Leg. Ma., putting back certain passages of the latter omitted by Hartvic. The fact, however, which Kaindl noted remains a fact, and has not been disproved. It is incomprehensible to me how Miss Bartoniek should reject Kaindl’s view with the simple words that it “is rejected by Ketrzynski and not accepted by Homan either”,5 Deer, in his introduction to the P. H. in the 6ame volume, denying it equally flatly and with equal absence of proof.6 It remains, as I have said, true that the text of Hartvic used in the P. H. is a text compounded of the Leg. Ma. and some of the further material contained in the present Hartvic; but showing no trace of the Leg. Mi., nor of certain other, obviously inter-connected portions of the additional material. We shall return later to the relationship between the two texts of “Hartvic” ; meanwhile, it is clear that since the Leg. Minor is not a constituent of the shorter text, it equally cannot derive from it. We return now to the respective dating of the two Legendae, and to the relationship between them. As is well known, our MSS. of the two Legendae all plainly derive from a single lost Archetype which contained both texts; but in all the texts,. and therefore, obviously, in their archetype the L. Ma. ends abruptly with Stephen’s deathbed appointment of his successor. Its unfinished condition makes the dating of the Leg. Ma. more difficult, but we have a reasonably certain terminus ad quem. Hartvic was composed J Scr. R. H. II. 373. 3 Ibid. H Op. cit. p. 291. K aindl’s point is, however, recognised and accepted by P. David. Les Sources de l’histoire de Pologne (Paris, 1934), p. 116.

353 7 during the reign of Coloman (1095— 1116), to whom it is dedicated; certainly after 1108, and probably after 1112.7 If (as we shall see to be probable) the Leg. Ma. is a component of Hartvic, it must have been written before that date. The termiiMS a quo cannot be fixed so easily. From the fact that the Legenda does not contain any mention of the canonisation of St. Stephen, which took place in 1083, nor - of the necessary preliminaries to that event (miracles performed at Stephen’s tomb, etc.) some critics have concluded that it was composed actually before the canonisation. It is, however, sufficiently obvious that the end of the text is lacking, and more than probable that the final chapters of Hartvic contain the missing end of the Legend. In that case, we must put the date of its composition at or after 1083, and probably after the death of St. Ladislaus in 1095 ;8 but on this point we cannot be quite certain. The date of the Leg. Mi. is usually assumed to be some time in the reign of Coloman. It cannot — so critics believe — have been later, because it was used by Hartvic as an ingredient of his Vita; nor earlier, because it refers to Coloman s predecessor, Ladislas, who died in 1095, as “pic memorie". As a m atter of fact, the former argument, as we have seen, is fallacious unless we suppose that the original Hartvic is the text which now bears that name, and that an earlier composition existed which was that used by the P. H. — a possibility, but, as we shall see, not an easy one. So far as that is concerned, the Leg. Mi. might be of almost any date in the twelfth century. But there are great difficulties in dating it so late. The author says in his first chapter that he collected his material “from the reliable and truthful narrations of Stephen’s con temp o ra rie s " ,a n d strongly implies that his sources were oral.10 7 In Hartvic c. 19 reference is made to a journey to Constantinople made by Duke Almus in 1108 or 1109; and as the Legend of St. Emeric, which is generally held to have been composed in 1112, also refers to the journey, it is usually assumed that Hartvic drew on the Legend. This is, of course, possible, but it is by no means necessary, since the incident must have been fairly widely known, and there is no verbatim agreement between the two accounts. 8 Hartvic in c. 24 refers to St. Ladislas as already dead, but it is not certain that he took this passage from the Leg. Ma. * c. 1: sicut fideli et veraci relatione tunc temporis viventium acccpimus. “Tunc temporis” must, of course, mean St. Stephen’s own life-time.

354 8

Now, St. Stephen died in 1038; it is therefore very difficult to suppose that the work was composed long after his canonisation. There are two objections to an early date: the one is the reference to Ladislas as already dead. The other is the fact that in several cases the L. Mi. and the L. Ma. cover the same ground. This applies to the suppression of Koppany’s insurrection;11 certain religious foundations of S. Stephens,12 and the Petcheneg irruption into Transylvania.13 If, therefore, as Miss Bartoniek (for example) concludes,14 the Leg. Min. used the Leg. Ma. as a source, it must, of course be the later text of the two. If, however, we look at the texts more closely, it is very difficult to suppose that the Leg. Ma. is the original and the Leg. Min. the derivative. In the cases of parallelism, the L. Mi. is usually fuller, and much more archaic. In the first, it knows the name of the City of Veszprem and something of its tactical importance in the campaign, together with the details of the duration to Pannonhalma. In the third, the uses the true Magyar form “Bisseni” for Petcheneg, instead of the pseudo-learned “Bessi”, and has such details as the mention of the “tribunus plebis in ulterioribus morans”. Others of his anecdotes, e. g. those in Chs. 6 and 7 bear every appearance of antiquity, and are precisely of the type which a later age — which stressed more and more Stephen the Saint, rather than Stephen the warrior — would be inclined to omit. I therefore have no hesitation in saying that the Leg. Min. is the source for the Leg. Ma., not conversely, and in dating the Leg. Mi. at very shortly after Stephen's canonisation. The two words “pie memorie", after Ladislas* name, are an addition such

2 « A ndreas, B61a et Levente in Poloniae regno exsulantibus. D lu gosz, 198.

577 309

DLUGOSZ E T L E CHRONICON BU D EN SE

Pour la periode Pierre — Aba jusqu’h la restauration de Pierre (Chr. Budense 71—78, Kezai 46—51, Dlugosz 222—3, 232—3) les trois textes concordent de nouveau. L’ordre des evenements chez Dlugosz differe de nouveau a peine de l’ordre dans X : il fait preceder la vengeance qu'il tira de Buda, de la fuite de Pierre, et le Chr. Budense se conforme encore a X . Tout s’en inspirant pour rddiger son texte de X , il le complete par des informations puisnes dans XDL. Il y a en realite quelques points dans Dlugosz que le Chr. Budense n’a pas reproduits, tel le consentement donne par les hauts personnages laics et ecclesiastiques du pays h l’dlection d ’Aba et le couronnement de celui-ci & «A lb a ». L’absence de ces points du texte du Chr. Budense suffirait k elle seule & prouver que Dlugosz n’a pu copier cette chronique. Cependant, alors que, pour des raisons q u i nous sont restSes inconnues, le Chr. Budense a omis de prendre tous les details offerts par la variante xDl, il a d ’autre p art incorpork & sa matiere to u t ce que sa source pouvait lui offrir relativement k la narration de X . Ainsi le chap. 71 du Chr. Budense n'est q u ’une reproduction du chap. 46 de K6zai, avec une addition de XDL, relative au jeu de mots de Pierre sur le nom de la Hongrie. L’addition est presentee d’une m aniere assez obscure a la fin du chapitre. Dans le chap. 72 le Chr. Budense su it de nouveau X (chap. 47 de Kezai), mais il em prunte k XD L la punition de Buda et de ses fils ( X ne cite pas le nom). Quelques-unes des additions sont assez meticuleuses, le Chr. Budense se m ettant en peine d’incorporer les expressions des deux p rin cip als sources, meme lk oil elles tra ite n t du m em e cas. A in si: K tzai

Chr. Budense Omnes autem constitutiones et exactiones Aba quas Petrus Rex secundum consuetudinem Suam statuerat, Aba Rex in irritum revocavit

Dlugosz

eaquae Petrus

Aba . . . constitutiones regis gratificando

statuerat in

ungarus

irritum revocans

irrita fecit

Dlugosz passe rapidem ent sur la periode suivante, que le Chr. Budense em prunte surtout a X en la complktant par les Annales Altahenses. Dans l’histoire de la conspiration cohtre Aba X (49), il rapporte les offenses d ’Aba, Dlugosz (232) la date du Q uadragesim a; le Chr. Budense contient tous les deux (75). Dans son ch. 76, apres avoir suivi X (ch. 50 de Kezai) le Chr. Budense intercale soudain une phrase qu i

578 310 rem on te a D lugosz. L ’original, ZD L . a v a it ic o u p sur du m a teriel non u tilise par D lu gosz. K eza i place la b ataille su iv a n te a Menfo, D lu gosz a G yor, le C h r. B u d e n s e a place «M enfew ju x ta jaurinum ». X consacre seu lem en t une p artie bien cou rte a u x aventu res des trois freres en P ologn e (ch. 52 de K ezai). L e C h r. B u d e n s e en d it beaucoup p lu s (la derniere m o itie du ch. 78, 79, 80). D lu gosz donne une h isto ire tres am p le de la guerre c iv ile polono-pom eranienne, m a is to u te differente de celle de K eza i ou du C h r. B u d e n s e ; cependant il ajou te : « A liis p lacet B elam ungariae ducem cu m tyran n o pom eraniae congressum sin gulari eum certam ine v ic isse et ob id in generum regis ad scitu m e sse . » La seule p ossib ility, c ’est que D lu gosz resu m e ici sa source. On ne sa it pourquoi il n ’a pas trou ve l ’h istoire assez interessa n te pour la donner en entier. Il se p eu t a u ssi que l ’original *DL a it ete tres court e t q u e le r e cit de 1’even em en t a it ete co m p lete par l ’au teu r d es in terp olation s d e la C h r. I llu m in e e d e V ie n n e rep ro d u ces, com m e nou s le savon s (q u oiq u e non entierem ent), par le C h r. B u d e n s e .1 D ans la seconde con sp iration contre Pierre, la com p araison des a d d ition s red evien t facile. L es ch. 2 3 4 — 5 de D lu g o sz, quoique, com m e nou s l ’avon s dejh m en tion n e, rep rod u isan t les ad d itio n s de X de secon d e source ch. 55, 56 de K ezai) concordent entierem ent avec les ch. 5 3 — 4 (de K ezai, e t le reste du ch. 56, il a m em e q u elq u es phrases to u t id en tiq u es avec celles d e K eza i, bien q u ’il en differe sur certain s p oin ts essen tiels p. ex . lorsqu’il d it q u e A ndre e t L ev en te furent les prem iers a v en ir en H ongrie, ta n d is que B ela resta en P ologne. K eza i parle to u t le tem p s « d es trois freres ». D a n s le te x te de X , une phrase a ete deplacee avec d es resu ltats d esa streu x pour les com binaisons h isto riq u es,2 de sorte q u e la narration de D lu g o sz est non seu lem en t la p lu s com plete, m a is encore la p lu s com prehensible. L e C h r. B u d e n s e , q u i est ab solu m en t concordant dans ses chap. 8 1 — 6, a encore tro is n otices propres : la citation syllab atim des paroles de st Gerard (a supp oser une re-interpolation de la L e g e n d a M a io r S . G e r a r d i) , q u elq u es 1 Ch. 186 de D lugosz : «Mieslaus rex . . . Andream, Belam et Leventem . . . quos Hungaria, u t s u p r a d e m o n s t r a v i m u s fugientes susceperat . . . > 2 La phrase e st la su ivante chez K£zai (ch. 5 3 ): « Qui cum in P est advenissent, ab-sconce u t poterant >, etc. Les autres narrations m ontrent tou t k fait clairement que le massacre des Allem ands et des Italiens et la reprise du paganisme 6taient les conditions faites k Andr6 et k L evente lors de leur arriv^e en H ongrie. De plus, la joie feinte, cachant une tristesse veritable, n ^ ta it pas le sentim ent 6prouv6 par Pierre loj-squ'il enten dit la nouvelle du retour de ses freres (pourquoi serait-il joyeu x?) m ais d'Andrd en entendant ses demandes im pies.

579

311

DLUGOSZ E T L E CHRONICON B U D E N S E

n ou v ea u x d eta ils sur des 6v6ques m artyrises, q u e D lu gosz aura o m is en copiant, XD L ; e t la n ote relative a la sorciere fa ite prisonniere par B ela Ie r ; pour ceu x -ci, il s ’en rem et a une autre source. D u reste, sa narration e st u n sim p le resum e d ’ad d itio n s de X e t XD L , com m e l ’exem p le su iv a n t du Chr. B u d en se (81) le p r o u v e : C h r. B u d e n se

K iz a i

p. 336 11. 11— 22 — 23— 25 p. 177 11. 22—3 25— 27 — 28— 30 » 11— 14 p. 336 1.30—337 1.4 — p. 337 11. 4— 6 » 23— 5 7— 13 —

D lu g o sz

col. 233 11. 41—9 » 49—51 > col. 234

52— 4

» 56—62 1.5

Ici encore le C h r. B u d e n se , dan s son in gen u ite, reproduit l ’ordre des ev en em en ts de X avec la con fu sion de K ezai, due a la p h rase m a l placee. D lu g o sz n ’a rien q u i repondrait a u x p reten d u es con q u etes d ’A ndre au com m en cem en t de K ez a i (p. 57) — e t q u i son t une p artie de la liste an n otee d es rois dans X . D ’autre part, il y a une lacu ne dans X (K ezai) la oh il s ’a g it de la m ort d e L ev en te et de l ’in v ita tio n adressee a B ela par A ndre, le C h r. B u d e n s e prend to u t cela dans XD L , d e sorte q u e les d e u x te x te s (le Chr. B u d en se e t D lu gosz) son t id en tiq u es sur ce point, excep tion fa ite de la phrase, p u isee dan s X par le C h r. B u d e n s e sur les con q u etes d ’A ndre e t u n e au tre au chap. 88, faisan t encore d efau t chez D lu g o s z : « H ec ig itu r prim a regni huius fu it d iv isio sem inarium discordiae e t guerrarum in ter duces et reges H u n garie. » J e consid ere cette phrase com m e une a d d itio n , au co m m en cem en t du recit fetendu, donne to u t au lon g par les interp olation s d e la C h r. I llu m in e e d e V ie n n e e t resu m e par le C h r. B u d e n s e sur « la discorde e t des guerres entre les rois e t les d u es de H o n g rie ». L e passage su iv a n t (Chr. B u d en se 89, D lu g o sz 237) contie n t l ’exem p le excep tion n el m en tion n e plu s h a u t oh * D L conn a it un p assage connu egalem en t des in terp olateu rs de la C h r. I llu m in e e d e V ie n n e ; m a is, to u t d e s u ite apres, les ta to n n em en ts recom m encent d e p lu s b elle. L e ch . 57 de K 6zai e st le parallele du ch. 2 3 7 — 8 d e D lu gosz, e t le C h r. B u d e n s e com b in e les d eu x dans son ch. 90. A in si les prem ieres 13 lig n e s du ch. 90 du C h r B u d e n s e concordent a v ec D lu g o sz, excep tion fa ite des nom s Zala e t Z elice (D ans X n ous tro u v o n s le nom d e B o d oh ot. D lu g o sz de nouveau rapporte le m essa g e d es b ogu s k G ebhart. X , e t non D lu g osz, co n tien t la n arration rela tiv e a u x A llem ands q u i se cachent sous leurs 6cus, p ar terre, ta n d is q u e X con-

580

312 tin u e d irectem en t avec l ’histoire d e la trah ison tram ee contre Salom on e t « S op h ia#). L e C h r. B u d e n s e en fa it au tan t, m ais D lu g o sz la place, plus ju stem en t, k une d ate posterieure. D e s lors, X com m ence & palir. K ezai raconte la longue et fa n ta stiq u e legend e de B orsony, h la place d e la fin du ch. 9 0 d u C h r. B u d e n s e , et il om et le recit du m ariage veritab le (co m m en cem en t du ch. 91) que, d ’autre p art, le C h r. B u d e n s e raconte d ’une m aniere differente d e celle de D lu gosz. N ou s revien d ron s encore sur ce point d on t je crois avoir trou ve l ’ex p lica tio n dan s l ’u tilisation par le Chr. B u d en se d ’un te x te d e X m eilleu r q u e n ’eta it celu i d on t d isp o sa it K 6zai. D lu gosz o m et les n o tes d es regnes d ’£ tie n n e h Andre (ch. 58 d e K 6zai, lis te an n otee d es rois), bien que, p. 2 5 0 — 1, il raconte le couronn em en t d e S alom on de la m em e fagon que le C h r. B u d e n s e (m a is non a v ec les interpolations d e la C h r. I llu m in e e d e V ie n n e dans ce ch a p itre e t le su ivan t), d e m §m e q u e la fu ite e t le retou r d e B ela (ch. 93). Il en trem ele son recit d ’une q u a n tite con sid erab le d ’in form ation s ad d ition n elles, q u i fo n t d6faut dans le C h r. B u d e n s e e t oil le r61e d e B oleslas d e P ologn e e st p e in t so u s u n jour trop clair et d es couleurs exagerees, ta n d is q u e la source e s t assu rem en t quelq u e source polonaise contem p o ra in e d e D lu gosz. Ici le t e x te de X d e v ien t (te l q u e K 6zai n o u s l ’a gard6) s i d efectu eu x, s i p lein de lacu n es m an ifestos q u ’il n ’e s t p lu s p ossib le de faire des c o m p a r is o n s relativ e s a u x a d d itio n s. L es pages pr6c6dentes o n t to u te fo is servi h prouver notre prem iere a ffir m a tio n : ici, com m e autre part, D lu g o sz n ’a p as u tilisd le C h r. B u d e n s e . L e d it Chronicon e t D lu g o sz on t, au contraire, p u ise leur m atiere k une source com m u n e perdue, nom m ee par nous *DL.

IV Q ue dire encore au su jet du XDZ? Il ne p arait 6tre id en tiq u e n i au x Co n i au XL . C . D e m & ne, je ne p u is decouvrir des ressem blances entre C O e t L. C. e t D lu gosz relativem en t k cette periode, excep te le cas u nique d e la genealogie altern ative de P ie r r e ; ic i encore D lu gosz p arait clairem en t d istin gu er des sources, car apres avoir term m e sa narration dans des term es ressem blan t k ceu x em p lo y es par X , il continu e de la m aniere s u iv a n t e : «A n n a les au tem H ungarorum eum (sc. P etru m ) non e x B u rgu n d ioru m genere fu isse testa n tu r », etc. C’e st h l ’avenem en t d e Colom an que l ’u sage fa it de x C o d ev ien t clair (331,) ce q u i n o u s fa it regarder com m e probable q u e sa n o te sur la m ort d e s t L a d isla s et probablem ent a u ssi l ’inform ation relativ e k l ’a cq u isition de la Croatie (cf. p. 132 du C h r. B u d e n s e )

581 DLUGOSZ E T L E CHRONICON B U D E N S E

313

derivent de la meme source, quoique la derniere n’ait pas survecu en Co. Du reste, x D l ne peut etre regardee comme narration continue que jusqu’k la disparition de Salomon et au triomphe final complet de st Ladislas. Outre ceux dejk discutes, il y a encore huit passages de Dlugosz que nous pouvons tranquillement attribuer & cette source. Ce sont les suivants: p. 259 (seconde apostasie des Hongrois, C h r. B u d e n s e ch. 95); p. 266 (Salomon demande le secours de l’Empereur, C h r. B u d e n s e ch. 97 commencement); p. 268—9 (restauration de Salomon, fuite de Geza, C h r. B u d e n s e ch. 97 moitie); p. 270—1 (retour de Geza, reconciliation des cousins, C h r . B u d e n s e ch. 97 fin); p. 283 (commencement des rivalites entre les cousins; sans details, sinon que Salomon 6tait victorieux d’abord et qu’ensuite Ladislas preta son appui k Otto de Moravie; Salomon vaincu s’enfuit h Pozsony; Ladislas fonde la cathGdrale de Vac, Salomon en appelle a l’Empereur, mais G6za corrompt les conseillers de celui-ci; couronnement de Gfeza, les 6veques essayent de reconcilier les cousins, mais sans succes); p. 288 (mort de Geza, couronnement de Ladislas, C h r. B u d e n s e ch. 130—1); p. 299 (Salomon ourdit une conspiration contre Ladislas; pris et remis en liberte, il s’enfuit chez les Allemands, puis chez les Bulgares; il disparait; enterre h Pola; C h r. B u d e n s e ch. 133—6, ib id . Salomon sans descendants, ch. 100). Tous ces passages, quoiqu’ils omettent beaucoup de ce que le C h r. B u d e n s e et encore plus la C h r . I llu m in e e d e V ie n n e ont trouve digne d’etre conserve, n’en forment pas moins avec ceux qui les precedent, une narration continue qui, commen?ant par Vazul et la succession disputee des heritiers de st Etienne, raconte l’histoire de ces disputes jusqu’a la disparition de Salomon, qui peut etre regardee comme marquant sa fin definitive. Et avec la meme force qui me fait trouver la theorie de M. Homan d’un seul « Gesta Ungarorum j> compose au temps de st Ladislas absurde et exageree, j’incline vers la conception que l’histoire de son avenement au trone a ete mise en ecrit a sa cour; x D l peut etre un effort de ce genre, recousu avec le temps a quelque histoire plus ancienne. Une confirmation de cette theorie se trouverait peut-etre dans le C h ro n ic o n P o lo n o - H u n g a r ic u m , bien deprecie et qui, malgre ses absurdites et ses deformations, semble base sur une narration de cette sorte, contenant beaucoup plus d etements communs avec 1’x D l qu’avec aucun autre des textes connus. Ainsi p&s de trace de ces parties du C h r . B u d e n s e dont Dlugosz ne savait rien non plus, tandis qu'il a plusieurs points de communs avec lui. De cette sorte le C h r . P o lo n o -H u n g ., comme Dlugosz (243), veut que les Hongrois, apres l’assassinat de Pierre, se soient venges «sur la reine » (il n’est que naturel

582 314 que le C h r. P o lo n o -H u n g . la decrive comme la femme de Pierre, tandis que \ ’x D l en fait la belle-soeur de celui-ci). Le C h r. P o lo n o -H u n g . represente la Duchesse de Pologne Damborovce comme la Protectrice des trois freres, tandis que Dlugosz (268) nomme aussi Dobrogniewe la protectrice du roi Geza. L’autre probleme interessant est celui de la parente de 1’x D l et de l’X. Comme nous l’avons vu, entre les premiers chapitres des deux histoires la ressemblance est vraiment grande, et meme la redaction est souvent identique. Meme plus tard, bien que l’identite redactionnelle fflt moins rapprochee, il y a des points de ressemblance evidents. Tous les deux commettent, par ex., la grave erreur de supposer que le nom de la fiancee de Salomon etait Sophie (au lieu de Judith); tous les deux semblent placer la mort de Bela avant la rentree de Salomon dans le cortege de l’empereur. Les relations des deux recits sont done evidents. II ne s’agit pas de versions alternatives du meme original, tous les deux ont ete copies et proviennent de sources communes. V Le pas suivant dans la voie de nos recherches nous ramene a Paulinus. Je compte, dans un autre essai, pouvoir approfondir les relations existant entre Paulinus et les textes hongrois et prouver que les conclusions tirees par M. A. Eckhardt, le seul savant qui jusqu’a present se soit occupe de la question, sont absolument fausses. M. Eckhardt a cru que Kezai avait servi de source a Paulinus; je demontrerai que non seulement Kezai n’etait pas la source a laquelle a puise Paulinus, mais encore que X ne Petait pas. Cette source etait une compilation anterieure h X meme, peut-etre un premier brouillon compose ar l’auteur de X, et meme que X pourrait bien etre maitre ^ezai, quoique le texte en cause ne soit certainement pas celui qui aujourd’hui porte le nom de Kezai. II n’est pas douteux qu’avant la composition de X — source commune de Kezai et du C h r. B u d e n s e , il n’ait existe un pre-X, source commune de Paulinus et de X. Dans l’essai mentionne, je continue mes preuves jusqu’au regne de st Etienne seulement, ayant reserve l’examen de la periode suivante au present essai. Faisons remarquer comme excuse preliminaire que le texte de Paulinus (quelques fragments exceptes), n’etant pas encore imprime et que les difficultes d’examiner des manuscrits continentaux (il n’y a pas de manuscrit de Paulinus dans les lies Britanniques) etant formidables dans les circonstances

E

583 DLUGOSZ ET LE CIIRONICON B U D E N S E

315

actuelles, je n'ai pu examiner le manuscrit entier de Paulinus. La matiere des XIIe et XIII® siecles semble toutefois relativement peu importante, de sorte que le copiste de Paulinus, Dandolo, a pu nous en donner une idee suffisante. II nous offre un important te r m in u s a d q u e m : son texte correspondant a celui de Kezai va jusqu'a l'avenement de Ladislas IV. Du reste, a partir de l'avenement du roi Coloman, le texte se borne k la maigre «liste annotee des rois » qui forme Tessentiel de l'ouvrage de Kezai depuis Coloman jusqu'& Ladislas IV. La periode qui nous interesse ici est celle allant du regne de st Etienne a celui de st Ladislas. Pour la succession d’Etienne et la conversion des Hongrois, Paulinus emploie un langage bien different de X, chacun des ecrivains ayant rejet e le recit de l'autre. Ce qui n'empeche pas Paulinus de se rapprocher de Kezai danS son recit relatif k st Etienne: « H ie devicto B ulgarorum duce a tq u e Slavorum de eius th esau ris hedificavit atq u e d o tav it ecclesiam sancte m arie de a lb a et tan d em m ultis nationibus subiugatis renunciare solio e t filio henrico conjunctus gubernare regnum . Sed princeps virgo cum sposa sue virgine p rae m atu ra coruscans m orte praev en tu s est. p Puis viennent les passages suivants : « M ortuo au tem rege P e tru s fra te r regine, vel secundum aliquos, ex sorore nepos reg n av it. Quai honores et officia om nia Teotonicis e t latin is d a re nisus, e t ty ra n n u m se magis quam regem exhibens, anno te rtio regni sui baronum et episcoporum consilio resp u itu r, et a b a sororius regis stephani in stitu itu r. « P e tru s au tem ad H enricum Cesarem in B avariam auxilium p e titu ru s confugit, et dum red iret anno te rtio cum cesare, A ba invasit A ustriam et usque ad flum en Trem se spoliavit e t reversus est. P ostea m isit exercitum ad spoliandam C arintiam , e t cum red iret exercitus, im petu principis A ustrie dim is spolia. H aec audiens im p erato r de colonia exiens, h a b ita deliberatione ab aquilonari p a rte D anubii venit a tq u e receptis captivis a tq u e dam pnis resarcitis Cesar in B urgundiam red iit. Post hoc O rpad rex insolescere in ta n tu m u t Ungari pro cesare m itte re n t et P e tru m superiorem reciperent. E t cum iteru m P e tru s et fautores eurum iniquum agerent, n itu n tu r Ungari P e tro deposito ad filios Z arladislai regnum transferee. Illis ergo vocatis eorum p a rte c la m atu r u t occidantur universi theotonici e t latin i. Sic U ngari p er loca om nia incipiunt rebellare et captus tan d em rex exoculatur e t A ndreas m aior n a tu rex iniungitur e t Noricos Boemos ac Polonos sibi fecit censuales. H enricum quoque Cesarem se obsidentem de g ratia cum solis equis om nibus aliis dim issis abire perm isit e t filiam eius accepit filio suo Salomon. A ndreas rex Confectus senectute in iu n x it filium Salomon assentiente B ela fra tre regis cum filiis suis. P o stea in te r Salomon ac L adislaum e t Gezam o rta discordia. Salomon tandem ex regno fugit. » Dandolo ajo u te : « Maximis regnum conquassatur angustiis p (IX . 7. 1) e t fo u m it encore les notes s u iv a n te s :

584 316 IX . 7. 10. «Salomon ta n d e m rex a fra trib u s expellitur, G eyza a u te m regnavit. T unc populus ille quasi fidem relinquit, p o stea re d it. H ie c o n stitu it om nia fo ra fiere sa b b a te e t B isantia cunere fecit. » IX . 8. 6. « Bessi U ngaris infestissim i spoliata U ngaria fugiebant, sed ab U ngaris ita oppressi d ic u n tu r u t ne unus quidem effugerit. s>

Un examen des differents passages nous a permis de constater que quelques-uns appartiennent a la liste annotee des rois. II n’est pas possible de determiner si les autres passages attribues par nous k cette source, tels les ch. 56, 57 et le commencement du ch. 58 de K6zai figuraient dans le p r e - X , omis plus tard par Paulinus ou si e’est X qui les a ajoutes au p r e - X . II reste bien une certaine narration continue, mais elle est encore plus breve que celle de Kezai. Ce n’est meme pas un resume: il correspond exactement h cette partie de l’histoire qui traite des etroites relations hungaro-allemandes, bien connue par consequent des chroniqueurs allemands qui n’ont pas manque d’en donner la description maintes et maintes fois, — les guerres d’Aba avec l’empereur, les aventures qui le mettaient en contact avec celui-ci. D’autre part, manque complet de donnees relatives k l’histoire de la'Hongrie: rien sur la trag6die de Vazul, les trois freres en Pologne, les relations entre Andr6 et B61a. Continuant notre analyse, nous avons constate que la narration continue de X doit 6tre divisee en deux partie: une histoire des relations germano-hongroises, empruntee a quelque chronique allemande et une histoire nationale hongroise du conflit de la succession, parallele au recit de x D l . La meilleure preuve du caractere compose de la narration sur ce point, se trouve au ch. 53 de Kezai, passage auquel nous nous sommes dej& referes. X rendu confus par l’histoire des tueries des Allemands et des Latins, finit par s’y embrouiller completement. L’histoire de la Hongrie de X a du certainement parler de l’aveuglement de Vazul, de la fuite des freres et dire quelque chose de leurs aventures en Pologne et de leur vie apres leur retour. Nous possedons en verite tres peu de donnees relatives aux phases ulterieures de cette histoire, si peu que nous serions meme tentes d’admettre que l’ecrivain n’a pas continue son histoire au delci du ch. 53, tout au plus 57 de Kezai. Cette hypothese se trouve toutefois contredite par les premieres lignes du ch. 61 de Kezai, oh celui-ci decrit d’une fagon assez detaillee un episode relativement bref de la lutte qui suivit. De plus, en retranchant (apres les trois premieres lignes) le ch. 57 du recit, nous nous trouvons de nouveau en face d’un recit detaille dont Paulinus n’offre aucun pendant, tandis

585 DLUGOSZ E T L E CHRONICON B U D E N S E

317

que, si nous l’incorporons, la question se pose pourquoi le texte de Kezai n’a rien qui corresponde au ch. 88 du C h r. B u d e n s e . La seule explication raisonnable serait d’admettre que le manuscrit du recit utilise par X ou Kezai etait defectueux, avec des fragments restes $a et lk integres et lisibles. Le fait est que les chapitres de Kezai sont pleins de phrases qui indiquent les lacunes du manuscrit. Ainsi la deuxieme phrase du ch. 90 du C hr. B u d e n s e . A la fin du mkme chapitre la legende entiere, telle que Kkzai la donne, ne se rapporte ni pour le sens ni pour le style k ce qui precede; il y manque quelque chose comme la fin du ch. 90 du C h r. B u d e n s e . II y a encore une large lacune. Si, comme le texte de Dandolo le prouverait, la phrase de Kezai relative au couronnement de Salomon est empruntee a la liste annotee des rois, toute la matiere correspondant aux ch. 91, 93 du C h r. B u d e n s e fait defaut. Une comparaison avec Dandolo induit k penser que le tour du ch. 59 de Kezai est celui de la liste annotee des rois, bien qu’un fragment du texte plus complet ait pu survivre. Dans cette partie, il y a une lacune evidente Ik ou l’histoire entiere du ch. 97 du C h r. B u d e n s e aurait dh se placer. Apres ce point, le manuscrit de X oil Kezai semble n’avoir rien conserve, k 1’exception peut-ktre du commencement du ch. 61 et de la fin du ch. 60 de Kkzai. Nous avons sans aucun doute affaire ici k un MS entierement dkfectueux, dont seuls les fragments indiques ont survecu. Est-ce la copie de X sur laquelle le C h r. B u d e n s e a travaillk et qui etait, elle aussi, defectueuse — ou bien seulement celle de Kezai? Pour resoudre ce probleme, il nous faut encore recourir k des comparaisons. Nous savons que le C h r. B u d e n s e a utilise X, mais non Kezai; chaque fois done que nous rencontrerons dans Kezai un passage quine se trouve pas dans le C h r. B u d e n se , il faudra supposer que ce passage est Pceuvre de Kezai lui-meme. Il est done facile de distinguer la fin du ch. 57 et le ch. 61 de Kezai imprimes en caracteres italiques dans le vol. S c r ip to r e s R e r u m H u n g a r ic a r u m avec la note «solus Kezai refert ». Il est evident que le «solus Kezai refert», la fin du ch. 57, et tout le ch. 61 k partir des mots in veste monachali auraient du etre imprimes en italique, ou rien du tout. Cette partie derive evidemment d’une legende monastique et n’a rien k voir avec la «narration continue ». Il y a encore un passage qu’on aurait dh imprimer en italique: la partie moyenne du ch. 60 de Kezai. Je l’ai dejk prouve autre part,1 le C h r. B u d e n s e connaissait les deux ou trois dernieres phrases de ce chapitre et il les amalgama dans 1 S tu d ie s , t. I l l , p. 72.

586 318

son ch. 121, avec des resultats desastreux pour la continuity de sa narration. D’autre part, l’histoire que l’empereur aurait, dans ce stade de la guerre, accorde son secours a Salomon, voire penetre lui-meme jusqu’a Vac, n’a aucun pendant dans le C h r. B u d e n s e . Elle est basee clairement sur une confusion de la bataille de Mogyorod et l’expedition posterieure de l’empereur et, a mon avis, une erreur de Kezai, basee sur sa memoire. II y a neanmoins dans le C h r. B u d e n s e un certain nombre de details non contenus dans Dlugosz, et qu’il est impossible d’attribuer soit aux interpolations V, ou au resume du C h r. B u d e n s e , soit & toute autre source etrangere du C h ro n ico n . II s’agit en particulier du ch. 90 et du commencement du ch. 91, peut-etre aussi du commencement du ch. 94. II semble raisonnable de supposer que ces passages ont du moins figure dans la copie de X, omis par Kezai, seulement parce qu’ils manquaient ou etaient illisibles dans la copie. Concluons done que la narration de X a forme un autre recit parallele a x D l de la lutte dynastique commen?ant par l’aveuglement de Vazul/et terminee par le triomphe de Ladislas. Comme x D l, il a ete mis en ecrit probablement au temps du regne de Ladislas, inspire des memes traditions et matieres que x D l et representant a peu pres le meme melange de faits et de fictions historiques. Les deux recits —- X et x D l — furent completes plus tard par la matiere identique ou pratiquement identique d’une chronique allemande, en relation probablement avec un resume des A n n a le s A lta h e n s e s . Le resume employe par X devait etre le plus court des deux.

xvn The First Historians o f Hungary

A LMOST every people, before it becomes generally /% acquainted with the art of writing, celebrates the r 1"% griefs and glories of its past in some kind of heroic JL jLsong which constitutes at once its first literature and its oldest historical records. The rule is universal enough: it is the character and the fate of these productions which is so variable. Rare indeed it is to find a straightforward and reliable narrative. Is the people dull and brutish, or is it subjected to excessive vicissitudes, its memories become short and its records perish altogether; on the other hand, too prolonged oral transmission or too great imaginative genius may evolve a work like the Iliad or the andient Irish cycles, which are read with delight as literature, but present to the his* torian more problems than enlightenment. In any case, there appear to be two absolute pre-requisites for the creation and survival of any considerable literature of the nature which we describe. The first is a lively and fertile national genius : the second, conditions which do not preclude some continuity of tradition. With the former the Magyars seem always to have been amply endowed, so that had this been the only condition of a rich epic and historic literature, their products might have vied with those of any nation. We have ample proof that long after their conquest of Hungary — indeed, well into the Middle Ages — there existed among them a large class of minstrels and balladmakers who frequented and entertained, not only the Court, but also the humbler centres of social life, by composing or reciting heroic epics on the national grandeur. If these have perished, leaving hardly a trace behind, it is due to the lack of continuity. Up to the end of the ninth century the nation

588

FIRST HISTORIANS

631

constantly changed its abode. Local associations could not survive; all reminiscence was perforce exclusively personal, and as the wars and forays conducted by each generation in the featureless steppes can have presented few individual distinguishing features, it was natural that the fame of the personal exploits of each generation should be quickly overshadowed by, or confounded with, those of its successor. No less important was the later breach in the nation's cultural tradition. Christianity, and with it the institution of writing — on any but the most primitive scale — came to the Magyars from outside. Its monkish bearers were conscious revolution# aries, whose aim it was to substitute an entirely new life for that to which their converts had been accustomed ; and among the victims of their zeal, to be eradicated with the remaining relics of an unhallowed past, were the ballads which celebrated with too much gusto a prowess hardly compatible with the more respectable Christian virtues. For these reasons, the native Hungarian records of the national life in the pagan era are disappointingly meagre. Nothing approaching the character of a collection of ancient sagas has survived, like the Irish or the Icelandic, the Niebe# lungenlied or the Romaunt de la Rose. All we get of this earliest national literature is an occasional tantalising reference by a later monkish chronicler, with here and there a quotation, sometimes of considerable length but always rewritten, condensed, adapted to the changed outlook of a later age and, more often than not, "corrected" by reference to western authors. Nor were the monks of Hungary able altogether to replace what they destroyed. Life was too precarious, the centres of learning too sparse to make possible the growth of a medieval monastic literature comparable to that of Germany or Italy. It is probable that nothing which could fairly claim the title of a historical work in the western sense of the word was composed in Hungary until late in the eleventh century A. D., and even then the first authors — who may well themselves have been of foreign origin — drew a considerable proportion of their material from western sources. Nevertheless, the oldest Hungarian Chronicles are by no means to be despised, either as history or as literature. Hungary,

589 632

after all, was converted to Christianity at the beginning of the eleventh century, and the succeeding generations en* joyed a considerable religious and intellectual life, several of the products of which have survived. Many of the nations of Central and Eastern Europe might be happy if their earliest records were as full or as interesting. The most important material is contained in a single long narrative Chronicle, which has come down to us in various forms and under different names. The oldest, and also the shortest version is the Chronicle of Simon de Keza, written about 12&3. Later, but fuller, is the Buda Chronicle, while more material still is found in the magnificent illuminated codex — one of the glories of Europe — known as the Chronicon Pictum Vindobonense. The exact relationship between these three texts is still under dispute, but it seems clear that all three are founded on a lost original, which each expanded, or epitomised, in different respects. This original was itselfprobably composed only in the thirteenth century, and is a compilation from many different sources, of varying interest and value. The beginning (which is practically identical in all three versions) presents, unhappily, particular difficulties, since it is dominated by an unfortunate confusion between the Magyars and the Huns. The author, indeed, distinguishes between the “first entry into Europe” under Attila and the second under Arpad, but the distinction is unclear, and it is obvious that the “Hun Chronicle” consists very largely of Magyar history, taken from the original record of the Magyar invasion in the ninth century, and roughly adapted to fit the story of Attila, yoo years earlier. This Magyar material is supplemented out of various Latin writers on the Huns, w'ith perhaps, one or two native traditions relating to Attila himself. It may safely be said that the Chronicle brings us nothing whatever that is new or valuable on the subject of the Huns. Its importance lies only in the possible indication which some (in my opinion, wrongly) believe it to offer that the Magyars in the steppes preserved a genuine national tradition of a historical connection between themselves and the Huns, and in the fragments of distorted Magyar history which it contains. If these be sorted out and combined with the succeeding

590

633 FIRST HISTORIANS chapters, we get the Magyars' own account of their own early origins and of the conquest of Hungary. One must, however, admit the result to be disappointing. The references to the Magyars' life before the Conquest are brief and obscure, and begin only a generation or so back; foreign sources tell us far more. The most interesting passage is a fine, but purely legendary account of the Conquest itself. The Magyars send a messenger to Sviatopluk of Moravia, then the ruler of the land, begging of him a handful of grass, another of earth, and a horn filled with the waters of the Danube, as samples, which he readily grants them, believing them to be rustici or peasant colonists. They then send him a white horse, saddled and bridled with fine gold, and again ask in return grass, earth and water. The incautious duke having told them to take "as much as they want,'' they maintain that they have lawfully purchased the whole country; and when he protests against this form of dealing, they fall upon him and slay him. The history of the tenth century is compiled partly from western sources, chiefly Regino's Chronicle, partly from ancient traditions. The combination is none too skilful, and gives rise to many chronological and other difficulties. The national traditions, however, although all too few, are romantic enough. The longer Chronicles, for example, preserve the legend of how the hero Lei, taken captive by the Emperor Conrad at the Battle of the Lechfeld in 9JJ, asks as a last favour that his famous horn be brought to him. When this is done, he smites the Emperor with it on the brow and slays him at one blow, saying : "Thou shalt go before me and serve me in the next world." The legend, as ICezai rightly remarks, is unhistorical, but as a record of the pagan beliefs of the early Magyars, it is as accurate as it is interesting. The heroic cycle whence this and other stories are derived appears to have run "up to the days of Toxun,'' the last purely pagan prince, who ruled in the middle of the tenth century. Of Toxun's son Geza, the father of St. Stephen, we hear practically nothing and of St. Stephen himself comparat* ively little, while the account of the confused and miserable decade which followed is drawn mainly from western sources (which are, indeed, exceptionally full for this period and are

591 634

probably largely based, in the last instance, on information from Hungary). It is different when we come to the latter half of the eleventh century. This, in history, was largely filled with internecine strife between the two brothers, Andrew I and B6la 1, and between Andrew's son Salamon and Bela's sons, Geza II and Ladislaus I. Kezai relates all this very briefly, but the longer Chronicles go into much detail. Their narrative covering this period is certainly a complete and homogeneous whole, by a separate hand. It is not wholly unexceptionable as history : it is episodic, in places legendary, and strongly biassed in favour of Bela and his sons, but it is certainly almost contemporary, containing a wealth of invaluable detail, and is told with a vigour and skill which Mallory himself might envy. I give here two samples. In the first Andrew*, feeling death approach him, decides to have his infant son, Salamon (whom he has previously betrothed to Judith, the infant daughter of the Emperor Henry III) crowned in his own lifetime, and breaks the news to Bela. We must note that in those days the Crown did not normally pass by law of primo# geniture, but rather to the oldest male member of the royal family. Bela was therefore his brother's lawful successor. "But since carnal love and family affection are wont to pervert justice, love for his son overcame justice in King Andrew, and in the twelfth year of his reign, when he felt him# self growing old, breaking his pledged word — which kings should not do — he had his son Salamon, a child ofy years old, anointed and crowned king of all Hungary. And he pretended that he did this to avert the destruction of the realm, saying that the Emperor would not have given his daughter to Sala* mon if he had not had him crowned. But when at Salamon's coronation they sang 'thou shalt be lord over thy brethren,' and the interpreter translated to Bela that the child Salamon was being made lord over him, he was grievously wrath. "Others say that Salamon was anointed king with the consent of Bela and his sons Geza and Ladislaus, and of all the nobles of the realm, but afterwards the apple of discord was set between them. For secret councillors, such as are rife in our time, whispered to the king that Salamon would not be able to reign unless his brother Bela were slain. Andconvers* ely they suggested to Bela that now the time was favourable

592

FIRST HISTORIANS

635

for him to seize the realm when Salamon was but a child, and his father stricken in years and sick. And so the King and the duke came to Varkun. But the King knew that his son could not rule after his death unless the duke consented. He called to council two henchmen of his, and took council with them, saying: 'I will make trial of the duke and set the two choices before him, whether he will have the crown or the duchy/ And the King had set before him the crown on a red cushion, and next to it a sword, signifying the dukedom. 'If the duke will take the duchy, and live in peace, let him have it; but if the crown, then do you twain rise up straight# way, and with that same sword smite off Duke Bela's head/ And they promised to do so. But while they thus took council together, Nicolaus, the captain of the guard who kept the guard before the palace door, heard it all. And when they had called the duke to the King and he entered by the door, the master of the guard said to him swiftly, 'if thou wouldst live, take the sword/ but he could not say more. But when the duke entered, he saw the crown lying with the sword before the King, and marvelled. And as soon as he had sat down, the King, who was lying in bed, raised himself, and sitting in his bed said, 'Duke, I have crowned my son, not from covetousness, but to save the realm from destruction at the Emperor's hands. But thou mayst choose freely: wouldst thou have the kingdom, take the crown, if the dukedom take the sword. The other thou shalt leave to my son; but the crown were thine by right/ At once the duke understood Nicolaus' words, and said: 'Let thy son, who is anointed, take the crown, and give me the dukedom.' And straightway he received the sword. And the King did a rare thing : he bowed himself down at the duke's feet. For he thought that the duke had given his son the crown in singleness of mind, as his own brother Levente had given it to him.1 But the duke acted thus out of fear." The second extract concerns a later episode — the repulse by Salamon and his cousins of a horde of Eastern invaders. "Then the Blessed Duke Ladislaus saw one of the paynims who was carrying on his crupper a fair Magyar maiden. And the blessed Duke Ladislaus thought that the maiden was the 1 A t h i r d b r o th e r , L e v e n te , h a d d ie d a p a g a n , m a k in g n o c l a im t o t h e t h r o n e .

593 636

daughter of the Bishop of Varad,1 and although he was sore wounded, yet he pursued the paynim swiftly on his horse Bay. But when he caught him up he could not smite him with his lance, because neither could his own horse gallop faster, nor did the other’s horse stay, but always there was the space of a man’s arm between the lance and the paynim’s back. Then the holy Duke Ladislaus cried out to the maiden and said: 'Fair sister, take the paynim by the girdle and throw thyself to the ground.' And she did so. And when the Blessed Saint Ladislaus had pinned him to the earth with his lance, he would have slain him. But the maiden prayed him earnestly not to slay him. Whence we may see that there is no faith in woman; for perchance she wanted him freed because she lusted after him. But the holy Duke struggled long with him and cut his muscle through and slew him. But she was not the bishop’s daughter.” After the reign of St. Ladislaus, Kezai’s Chronicle degener* ates into little more than a list — not always accurate — of kings, with their dates of accession and places of burial. The Buda Chronicle is little fuller; but the Chronicon Pictum continues with a very detailed and spirited narrative. This, again, is episodic, and relates chiefly to the affairs of the royal family, but what it gives is told in lively fashion, and most of the information is probably accurate enough. Urn happily, the narrative breaks off suddenly in i iyi ;2 and there follows a long century during which no one in Hungary appears to have troubled to make even an elementary record of passing events. Things improve only at the end of the thirteenth century, when K6zai offers us a laudatory and not very ingenuous account of the reign of his master, Ladislaus the Cumanian, while the fuller Chronicles carry on the story in considerable detail up to the coronation of Louis the Great. Our information on the early period is to some extent supplemented by the Lives and Legends of the various Saints, of whom the royal dynasty alone produced three (with a doubtful fourth)8 in the eleventh century. These do not as 1 The 1 The

later Coloman I. German “Chronicle of the Huns/' by Mtigeln, which is simply a free transla* tion of the Buda Chronicle, carries these extracts on in abbreviated form up to 1167. * St. Stephen, St. Emeric, St. I^adislas and Salamon, who was at one time revered as a Saint in Pola, but has not found very general acceptance.

594

FIRST HISTORIANS

637

a rule contain as much historical material as we mighthave hoped. One must indeed admire the single-minded religious fervour of the medieval monk, to whom a whole war or invasion is interesting only as an opportunity for his hero to display some remarkable magnanimity, or his corpse to work a miracle. A less devout age, however, feels that it would gladly exchange twenty pages of holy reflections for a few names or dates. The Legends of St. Emeric and St. Ladislas are almost purely pious compilations. The three Lives of St. Stephen contain more that is of interest to the historian, but still fall grievously short of what we might have hoped from a biography of so famous a figure. There is, however, one "Life” which is a historical document of first-rate importance — the "Vita Maior” of St. Gerard, the Italian missionary who was invited by St. Stephen to help convert the land to Christianity and was martyred in the pagan revolt of 10J4 on the hill of Buda which still bears his name. This work as it has come down to us is, indeed, a compilation to which the finishing touches were probably not put until the fourteenth century. A large part of it is, however, undoubtedly very old indeed (in all probability, almost contemporary with the Saint's own life) and not only its pictures of the conversion of the heathen Magyars, and of the early organisation of the Church, but also its mysterious account of the war waged by St. Stephen against a pagan adversary named Achtum occupying the territory later known as the Banat, are of the highest historical interest and importance. Finally, we come to the famous Anonymous Notary of King Bela, whose statue stands in the Varosliget of Budapest, its hidden face appropriately symbolising the mystery which surrounds the man. Everything about him is uncertain. He begins his work by describing himself as "P, called master, sometime notary to the most glorious King Bela of blessed memory” — thus ingeniously concealing not only his own name, but also his date ; since there were no less than four Kings Bela of Hungary. The classical method for any young Magyar historian to make his debut is that of propounding a new theory as to the identity of the man, or at least the king, and the ink which has flowed on the subject would turn the

595

638

Danube blue?black. When I state here my strong personal preference for the reign of Bela III ( 1172 — 1 1 9 6 ) it is in full knowledge that a large and influential party would go to the stake for Bela II ( 11^1 —1 14 1 ). Strictly speaking, however, the question is as unimportant, and probably in the long run hardly more soluble than that of which song the syrens sang. Anonymus' work purports to be a description of the origin of the Magyars and the Conquest of Hungary, down to the days of Geza I. The author claims to have drawn exclusively on written sources, rejecting "lying peasants' fables and the prolix songs of minstrels" and to set forth the true facts for the first time. The Magyar nation had every reason to accept his claim gratefully, and not too critically : for his narrative forms to the patriot a refreshing contrast to the usual utterances on the subject of the Magyars which adorn the contemporary western Chronicles, and run something after this fashion : "In this year the bestial nation of Hungries, eaters of foxes and wolves, drinkers of blood, more cruel than any wild ani? mal, again despoiled our land, slaughtering, raping and pillage ing." Anonymus on the other hand represents the Magyars as a little band of disciplined paladins, men of war indeed, but of knightly prowess and chivalrous demeanour : to strengthen which impression he endows his chief figures with character? istics, physical and moral, borrowed from Dares Phrygius' "Fall of Troy" and from the Legend of Alexander the Great. The Magyars, under their seven chiefs, enter Hungary to possess it by hereditary right, in virtue of the descent of Almus, father of Arpad, from Attiia, the first great Magyar. They come from the East, and on their way do battle outside various Russian cities — Kiev, Vladimir and Halicz, whose princes submit to them and lavish costly gifts upon them On the way they collect seven "Cumans" with their retinues, who follow them into Hungary as their allies. On crossing the Carpathians, they first pitch camp at Mun? kacs and Ungvar, whence they derive their name of "Hung? arians." The local inhabitants promptly submit to them, as to their rightful lords "with great deference and fear." Then begins the conquest of the country, the different parts of which, after Attila's death, had fallen under the rule of

596

FIRST HISTORIANS

639

various sovereigns : a certain Salanus, grandson of the Great Khan, Duke of Bulgaria, between the Danube and the Tisza ; Menumorot, grandson of Marot, between the Maros and the Szamos : Glad, ancestor of Achtum, between the Maros and the Lower Danube. All these, with "Romans" in Pannonia, a Duke of Bohemians and Slavs in Western Slovakia, a Bulg* arian relative of Salanus' near Belgrade, and a Duke of the Vlachs named Gelou in Transylvania, are in turn met and defeated in glorious fashion. The exploits are usually carried out by one or more of the seven Chieftains, or by one of the seven Cumans, who after performing their deeds, return to Arpad and receive praise and entertainment, followed by the gift of an estate. As each enemy is defeated, Arpad himself moves steadily south and west until he has taken up his own headquarters south of the present Pest, and the whole of Hung* ary is conquered. To set a crown to the achievement, other heroes career through the Balkans, conquering everybody. There follow some chapters which correspond closely with those of the narrative Chronicles dealing with tenth century history, and based largely on western sources. The work com eludes, or rather, breaks off— since it was certainly the author's intention to carry it further — with a chapter describe ing the arrival of foreign settlers in the days of Toxun. As we said, Anonymus, when first his work (after lying for centuries unheeded in the Hofbibliothek at Vienna) became generally known, was enormously popular in Hungary. Impli* cit faith was attached to his lightest word, his narrative was repeated almost verbatim in history books, and paintings representing the scenes which he describes adorned the walls of Hungary's public buildings. Later, however, unkindly frank investigators (mostly foreign) pointed out a whole series of borrowings, the effect of which was greatly to shake faith in his reliability. It was not merely that where he purports to be recounting native Magyar tradition he is often in fact using the words of a western Chronicler, or even of a clas* sical author: he lifts personal descriptions, battle^scenes, accounts of revels wholesale from his reading. Far worse than that: actual historical events which, according to him, took place at the Conquest of Hungary, or near the end of the ninth century, can be shewn to have occurred indeed, but at a far

5 97 640

later date and in quite other circumstances. Public opinion was forced to revise its estimate, which it did the more willingly when it awoke to the fact that in including Gelou, Prince of the Vlachs, in the list of enemies conquered by the Magyars on their entry into Hungary, Anonymus had innocently given his support to the Rumanian thesis in the Great Vlach Control versy. Thus poor Anonymus fell out of favour again, except among the Rumanians, but he was by now too popular a national figure to be buried. The present generation is endeavouring more scientifically to sift the wheat from the chaff in his story. For wheat there undoubtedly is. He is quite capable, for example, of preserving some tradition or quotation in a purer form than we have it elsewhere; the only trouble being that he inserts it in the wrong context, and possibly changes some of the personal names. If we can nevertheless trace it to its proper source, it may prove valuable, not in the setting given it by Anonymus, but in that where it properly belongs. Thus he gives us a fuller narrative of the tenth century raids than do the narrative Chronicles themselves, although transferring part of the material to the Conquest; he preserves details of the Achtum legend omitted by the author of the Life of St. Ger# ard, which may serve to explain the identity of that notable prince : his seven Cumans were certainly not Cumans, and as surely did not throw themselves at the feet of Almus, father of Arpad, outside Kiev, swearing in Biblical language to follow him faithfully unto the last generation — yet rightly under# stood, the episode of the Cumans throws quite a new light on the ethnographical composition of the invaders whom Arpad led into Hungary. Anonymus is not to be swallowed whole; he is a bran pie out of which, with care, one can pick the most precious dainties. Unhappily, the process of selection and interpretation is one so delicate and so uncertain that it does not seem likely that the Anonymus controversy will find its end for many a long day. But with all his shortcomings, his inaccuracy, his airiness, his unscrupulous handling of his sources, the pride and affec# tion with which Hungary regards him are not entirely mis# placed. He is a national figure who has contributed in his own way to giving Hungary her national soul. When we think of

598

FIRST HISTORIANS 641 Kim, and of Kis fellows, Kezai, the anonymous authors of the longer Chronicles, the historians and biographers of the Saints and Kings, we may echo the Anonymous Notary's own words, with which he concludes his preface : "Then happy is Hungary, to whom are given sundry gifts, for at all times she can rejoice in the gift of her chronicler, because she has the beginning of the genealogy of her kings and nobles, from which kings be honour and glory to the eternal King and to Mary His holy Mother, through whose grace may the kings of Hungary and her nobles hold their realm in prosperity now and evermore !"

IN D E X This index was prepared with the assistance of Ms Judit Nyerges of the International Association of Hungarian Studies, Budapest. The only abbreviation introduced is H which stands for H u n g a r y or H u n g a r i a n . All other abbreviations used are by the late Professor C. A. Macartney. The d r a m a tis p e r s o n a e are not given a uniform description. Variance in spelling arising from misprints are ignored in modem names, but inconsistent usage by medieval authors is, wherever possible, identified. Explanations given in brackets are Macartney’s and follow the practice established in his T h e M e d ie v a l H u n g a ria n H is to r ia n s (1953).

A, s e e C o d e x A c e p h a lu s A. A., s e e A n n a le s A lta h e n s e s A. P., s e e A n n a le s P o s o n ie n s e s *A. P., 202, 208 Aba Sdmuel, King of H., 57, 65, 71, 93, 101, 123, 169-71, 173, 178, 187,214, 223, 309, 314, 320, 395, 396,458, 572, 574, 577,583, 584 family of, 131, 303, 308, 377, 378, 422, 428, 436, 437, 542 Gerard’s sermon against, 70, 76, 78-80, 182 Abah, fortress of, 427 Abaujvdr (Castrum Novum), 183, 396 Abdelites, 12, 13 Abel, son of St. Stephen, 360 Abou Kassim, 26 Abudiacum s e e Abah, fortress of, Abulfeda, 20 Accio, 431 Achaia, 478, 495, 506 Achelous, 27 Achtum (Ajtony, Ohtum), legend of, 57, 66-9, 77-99, 123, 234, 235, 270-73, 278, 317, 318, 323,328, 594, 596, 597 Acilles (= Attila), 229 A c ta S a n c to r u m , 65, 69, 72, 493 Acus, s e e Akus Adalbert, St., Bishop of Prague, 132, 141, 361 Adelhaid (Athleida), alleged mother of St. Stephen, 222, 224, 368, 369, 371, 373 Adelheid, wife of Stephen II, 204 Ademar, 330 Admunt (Admont, Agmund), 143, 189-91, 437

Adrianople, 26, 27,47 Adriatic Sea, 497, 558, 559 Adtila (Atila), s e e Attila, city of Aecio, 428 Aethiopes (Nigri Saraceni), s e e Saracens Aethiopia (India Minor), s e e Ethiopia Aetius, 495, 496,498, 499, 506, 508, 511, 525, 558 Africa, 465, 522, 544, 545 Agathias, 18 Agetius, 557 Agmund, s e e Admunt Agnellus, Abbot of Ravenna, 543 Agrippina, s e e Cologne Ajtony, s e e Achtum Akus (Acus), 131, 303, 309 Aladar, 312, 479, 482, 491, 513, 539 Alania, 25, 502, 546 Alans (As), 1,11,15, 19, 25, 29, 35, 43, 62, 237, 240,246, 489, 492,496, 501^1, 544, 556, 558 Alaric (Alaricus) the Goth, 312,496, 525, 541 Alba (= Andrew? Aba?), 371, 372 Alba (Szdkesfehdrvdr), 55,66,70, 83, 104-7, 115, 145, 164, 171, 181, 184, 189, 198, 199, 208, 344, 345, 359, 360, 366, 373, 391,420,515, 537, 562, 577 Albania, 51, 489, 545, 546 Albano de Bosnia, 463 Al-Bekri, 25, 34,41,42 Albericus Trium Fontium, 101, 121, 123, 142, 197,213, 330, 372 Albero, 428 Albert of Habsburg, 419 Albertus (= Lampertus), 371

600

INDEX

Albos Russos (= Massagetae?), 484 Albus (= Andrew I ?), 143 Aldrian (= Aladar?), 312 Alemannia, 290 Alexander Magnus, Alexander of Macedon, s e e Alexander the Great Alexander the Great (and G e s ta of), 233, 242, 243, 261, 271, 273, 317, 518, 521, 595 Alexius Comnenus, 41, 48, 49, 59 Alfold, s e e Great Plain Alfoldi Andrds, 3 Alfred the Great, King of England, 231, 531 Al-Khazar, 37 Allemands, 578, 579, 581, 584 Almus, Dux of H., 55, 56, 143, 156, 159, 160, 162, 163,220, 221, 228, 233, 249-251, 253, 256, 257, 260, 261, 283, 284, 305, 318, 332, 342, 343, 375-8, 384, 534, 535, 547, 549, 550, 555, 595, 597 election of, 249 genealogy of, 160, 163, 375-378, 390 Almus dux, son of G6za I, King of H., 103, 105, 143, 195-197, 201, 202, 204, 205, 344, 345, 372, 402, 428,431, 434, 444-8, 566 his journey to Constantinople, 353 Almus’ mother, s e e Emese Alpine passes, 498 Alplozuri (Alpzuri), 238, 503 Alpra, Cuman dux, 425 Alps, 364, 386,511 Altino, 463, 559 fall of, 315 Altman family, 132 Aluta river, 56 Amanus, Bishop of Metz, 557 Amir, son of Japhet, 21 Ammianus Marcellinus, 3, 15, 456 Anakwe, khagan, 10, 14 Anastasius, Abbot of Pdcsvdrad, 66, 80 Andalusia, 22, 24 Andr6, s e e Andrew Andreas, dux, son of Andrew II, 109 Andreas Albus, s e e Andrew I Andrew, brother of Ladislas IV, 405 Andrew, brother of Louis I, 384 Andrew, dux, 404 Andrew, son of Stephen V, 410 Andrew, son of Yuri Dolgoruki, 265 Andrew I, King of H., 71, 101, 114, 117, 122, 125-7, 141, 143, 145, 146, 174, 175, 177-181, 183-185, 187, 206, 207, 213, 222, 304, 319, 320, 394, 396,406, 419, 430, 437,439,440, 565, 567, 572, 574-6, 578-580, 583-4, 591

coronation of, 184, 213, 215 death of, 217 his victory over the Poles and Bohemians, 304 Andrew II, King of H., 60, 102, 108-10, 113, 147, 259, 346, 347, 403, 404,416, 423, 425, 428, 431, 432, 434, 443, 566, 571 Andrew III, King of H. 102, 134, 149, 193, 346, 350, 405, 410-12, 414,416, 419, 420, 562, 566 Angelus Accialus Florentinus (and his V ita C a r o li M a g n i ), 343, 479, 564 Angisciros ,2 Anhialo, 27 Ankardah (Ankiradah, Unkaradah), 21, 24 Anna Comnena, 47, 52 A n n a le s a p r im o c h r is tia n o d u c e M ie s c h o n e , 409 A n n a le s A d m u n te n se s, 190 A n n a le s A lta h e n s e s (A. A.), 83, 130, 133, 166-77, 179, 184-6, 193, 198, 214, 216, 217, 304, 340, 373, 382, 392, 394, 398, 423, 428,434, 436, 437, 439, 528, 565, 566, 574, 575, 577, 586 A n n a le s B e r tin ia n i, 340 A n n a le s B o io r u m , s e e Aventinus A n n a le s C o lm a r ie n s e s , 433 A n n a le s C r a c o v ie n s e s , 369 A n n a le s F uldenseS y 340, 341, 426, 455, 485, 486, 554 A n n a le s H ild esh eim en seS y 81, 83, 87 A n n a le s K am en zen seS y 369-371 A n n a le s P o lon oru n iy 409,410 A n n a le s P o s o n ie n s e s (A. P.), 81, 194, 198, 199, 200-203, 205, 213, 351, 369, 373, 374, 445, 566 A n n a le s R e in h a r d sb r u n n e n se s , 528 A n n a le s S a n g a lle n s e s M a io r e s , 293 A n n a le s S ilesia c iy 2 2 2 A n n a le s V a r a d e n s e s y 566

316 Annalista Saxo, 186 A n n a ls o f W ei, 12 Anne, daughter of Bdla IV, 413 Anon., s e e Anonymus A n n a le s V e n e tic i B r e v e s ,

A n o n y m i D e s c r ip tio E u r o p a e O r ie n ta lis , s e e

Cracow Anonymus Anonymus, notary of King B61a, 55-8, 77, 83-90, 92-4, 97, 98, 120-22, 130, 131, 157-9, 162, 163, 165, 166, 201, 204, 220, 224, 226-44, 246, 248-80, 282-90, 294-99, 301-3, 307-9, 317-35, 343, 365, 369, 377, 389, 395,464,465,468,47(M, 481, 505, 514, 523-9, 534-7,542,545, 547-9, 551-4, 565, 594-8

INDEX

Anterior, King of Troy, 464, 559 Anthemius, 427,429 Antonius Durus ,232 Antonius Florentinus, 506, 516 Apa-Ogmand, 329 Apostolic Crown, s e e Holy Crown Apulia, 220, 226, 349, 512, 558 Aquila (= Attila) 224, 227, 364, 391, 517, 521, s e e a ls o Attila Aquileia (Aquilegia), and its destruction, 226, 227, 312, 315, 316, 342, 361,462-5, 468, 494, 495,497-9, 512, 516, 541, 559 Aquilonis mare, 245, 556 Aquincum, 528 Arabs, 13 Arad massacre, 150 Aral Sea, 12, 16, 22, 25 Archelaida, a place, 310 Archelaides, 310 Arc(h)elaus, king of the Goths, 312, 541 Archelaus Tetrarcha, 310 Arderic (Ardaricus) the Gepid, 7, 312, 559 Arelate kingdom, 557 Argentina, s e e Strasbourg Armenian sources, 202 Am, s e e Amspeckius Arnold of Liibeck (and his C h r o n ic a S la v o r u m ), 266, 267, 333, 528 Amspeckius, Vitus, 418, 424, 425,427, 431-5 Amulph, duke of the Bavarians, 157, 287, 324, 457, 484-6, 532 his wars, 483 Arpdd, Dux of H., 54-7, 85, 94-6,98, 106, 141, 143, 156, 160, 161,221,228, 233, 249, 250, 252-7, 264, 269, 273-5, 282, 286, 331-3, 337, 342, 364, 365, 391,423, 426, 455, 515, 526, 534, 536, 537, 546, 547, 549, 550, 551, 555, 560, 564, 589, 595-7 ancestry of, 160 death of, 220 Arpdd, dux son of G£za II, King of H., 105, 345 Arpdd, House of, 88, 204, 309, 567 Arpds, a place, 52 Arsia river, 496 Arthur (Arturus), King, legend of, 449, 519-522, 563, 564 Aschericus, abbot, Archbishop of Kalocsa, 81, 358 Asen of Bulgaria, 266 Asia Central, 8, 10, 11,338 Minor, 501 Western, 10, 16

601

Askold, 32 Asparuch, 34 Assyria, Assyrians, 109, 346,465 Attaliotes, 53 Attila (Athyle, Atila, Atilla, Atthila, Atthyle, Atyla, Atyle, Et(h)ela, Et(h)ele), 1-3, 5-9, 18, 57, 94, 95, 135, 143, 160, 163, 178, 194,220-33, 241, 244,249-52, 257, 273-5, 278-82, 284, 285, 288, 307, 308,310,312,315,316, 332, 336-43, 360, 361, 364, 365, 367, 370, 373, 377, 378, 385, 389-91,424,427, 429, 449,450,454-7, 462-5, 467, 470, 471, 476, 478-82,487, 490, 492-500, 502, 504-10, 512, 513, 517-20, 522-5, 527-9, 533, 534, 536, 537, 540, 541, 543, 544, 546-9, 556-9, 563, 564, 589, 595 fla g e llu m D e i , 333, 368,463, 557, 559 his palace, 275 his sword, 333 his victory over Romans and Germans, 276, 284, 306 Italian campaigns of, 475, 495, s e e a ls o Aquila Attila, city of, s e e Ecilburg Attila, legend/saga of, 219, 229-39, 242, 252, 262, 263, 273, 274, 276-8, 280-84, 287, 288, 302, 306, 311, 313, 314, 317, 318, 322-43, 362, 365, 369, 373, 383, 429, 469, 523-7, 545, 547, 549 Romance of, 468, 475 the Polish saga of, 516, 545 Vandal and Lombard elements in, 457 Attila’s sons, 2,4, 5 Augsburg, 407, 427, 565 battle of, 231 Aureliani, s e e Orleans Australes, s e e Austrians Austria, 59, 172, 226, 319, 365,404,405, 411, 414, 424,428,437, 498, 561, 571, 583 Andrew Ill’s war against, 350 invaded by Aba, 171, 173 Austrians, Australes, 136, 348 Av., s e e Aventinus Avarius, ruler of the Avars, 498 Avar-Magyar identity, the question of, 343 Avars, 1, 6, 10-19, 57, 58, 236, 262, 263, 281, 282, 291, 294, 297, 324, 336, 339-41, 362, 363,422,454-7,479, 482, 483, 485, 491, 498, 513, 528, 531, 532, 535, 542, 546, 548, 555, 563, 564 European, 12, 14, s e e a ls o pseudo-Avars Avars, King of the (= Attila!), 342 Avarum solitudines, 297

602

INDEX

Aventinus (and his A n n a le s B o io r u m ), 175, 176, 299, 300, 310, 311,418, 421, 422, 425-33, 484-6, 491, 492, 504, 513, 539, 549, 552, 554 Avitochol, Bulgarian prince, 336, 339 Avitus, 7 Aydua, s e e Edua Aymas, 560 B, s e e C h r o n ic o n B u d e n s e B/D, s e e B and D Babenberg, Ernst von, 59 Babylonia, 18, 346 Bdcs, Archbishopric of, 358 Bactria, 15 Baggird (= Bashkirs), 20, 21, 24 Bagnak (= Petchenegs?), 20-23 Bagrtak Turks, 21 Bahrain V, King of Persia, 16 Baian o r Batbaian, 34 Bailey, Sir Harold, 15, 62 B a io a r ic a le x , 172 Balaton lake, 56, 63, 117 Balkanpasse, 267 Balkans, 46, 48, 250, 330, 506, 507, 596 Eastern, 267 Balkash lake, 12 Balxan (Po-lo?), 12 Bamberg, 241,242 Banat (Bdndt), 4, 91, 98, 270, 594 Bdnd, 151 Bandtke (Bandtkie), Jerzy Samuel, 369 Bdnhid, 486,515,553 battle of, 483 Bank (Bang) banus, 147, 150, 346, 404, 416, 423, 425, 428, 431,433 Baranya, county of, 56 Barbarossa, s e e Frederick Barbarossa Bardores, 2 Barsacia, s e e Bascardia Bartholomaeus Anglicus (Glanville), 532 Bartoniek Emma, 140, 352, 354, 359, 360 Bascardia (Barsacia?, Basciria, Baskiria, Pascatir), 130, 243, 245-8, 550 Basel, 467, 557 Bashkirs, 20, 22, 23, 245, 247, s e e a ls o Baggird Basiana, 2 Basil II, emperor of the Greeks, 84, 87 Basilea, s e e Basel Basilius, St, 267 Baskird Ihura, 501 Bastarcos, s e e Bascardia Baszko, Godyslaw, 364, 516-18, 522 Bator, son of Opus, 566

Bavaria, 172, 177, 290, 424, 439, 498, 583 Bavarian, Bavar, confused with Avar, 456, 498 Bazarab, 405 Beatrix, queen consort of Matthias I King of H., 388, 476 Beazley, C. R., 246 Becha, 107 Bega, 252 Begon, nickname of Bdla I, 102 Behem (= Bohemia), 348 Beke, son of Thomas, 384, 414 B & 65 hermitage of, 66 monastery of, 155 Bdla, grandson of Aquila, 222 Bdla, of the Attila saga, 558 Bela, son of B61a IV, 404 Bdla, son of St Stephen, 360 Bdla Cecus, s e e Bdla II, King of H. Bdla Graecus, s e e Bdla III B£la I, King of H., 101-3, 114, 122, 127, 128, 133, 141, 143, 145, 146, 148, 175, 177-80, 185, 187, 188, 206, 207, 210, 212, 214-17, 222, 363, 371, 372, 396, 400, 406, 428, 437, 565-568, 572, 574, 575, 578-80, 582-^, 591,592 in Poland, 184 Bela II, King of H., 102, 103, 106, 136, 146, 151, 189, 195, 200, 203, 204, 265, 266, 268, 302, 344, 345, 402,403,407, 408, 428, 445, 595 Bdla III, King of H., 105, 107, 108, 126, 133, 203, 265, 266, 268, 302, 345-7, 364, 382, 403, 409, 571,595 Bdla IV, King of H., 60,61, 102, 108-11, 113, 147, 151, 193, 347, 348,401,404,413, 416, 418,423,425, 428, 431, 432 Belar, 240, 241, 337, 464, 544, 546 Belgrade, 49, 91,95, 267, 496, 498, 573, 596 Belial, his children, 267, 330 Bclitzky Jdnos, 91 Beliud, 89, 272 Belyn, nickname of Bdla I, 102, 146 Bemen (= Bohemia), 349 Bencia (= Dencia), 130 Bendefy Ldszld, 246, 248 Bendekuz, 163, 378 Benedict, 544 Benedict, St, s e e V ita S S Z o e r a r d i e t B e n e d ic ti Bereka (Bereke, Berela, Bereska, Berzeba, Berzeke), 240, 290, 546, Bemoldus (Bertholdus), 189,428 Bertholdus dux, 291, s e e a ls o Pertold Besan^on, 171, 173,465

INDEX

Besenyd, s e e Petchenegs Beseny