1!;::;: ' "'M-i-h!;!:', ■1>IC' UV'J"y Vti hr -^i'- 1 ' ; , ■ < ' '■ ■.'. i 1;';. ^ '* •>• ' * ! 1 ■ ', 1 ''!•i^'i;^i^ c-i ^> <^y' .\^ v^^' •^/>- .^''' '/ o. "^y- V^ -oo^- 'if ■^^ v^ .0^^. o •%^^ .^' -V ^^.-. , . ■* { <^^ \^°^. ,0 o. .#' ^^ == % 0^- .o '^- * ■' K ' ^> ' "' ^ ' V*^^ . ' ^ . ^ * ■' -^ " ' ^Q ^^ v^ WESTEEN EUEOPE IN THE FIFTH CENTURY AN AFTERMATH BY THE LATE E. A. FREEMAN, M.A., Hon. D.C.L. REGIUS PROFESSOR OF MODERN HISTORY IN THE UNIVERSITY OF OXFORD LONDON MACMILLAN AND CO., Limited NEW YORK: THE MACMILLAN COMPANY 1904 All rights reserved (:>nSi^ OS OXFORD : HORACE HART PRINTER TO THE UNIVERSITY PKEFACE When the ]ate Mr. E. A. Freeman set forth on his last visit to Spain his immediate interest was the completion of his History of Sicily. It was known, however, by his friends that he had left behind in an unfinished state the materials for a volume on Western Europe in the Fifth Century. Like many other historical students he was much interested in the few historical notices that have survived concerning the events in Britain during the fifth century. He desired to understand them, and as far as possible to fit them in with what we know of the general political development of Western Europe, and he felt that the only way of approaching this subject with any chance of permanent success was to make sure of the events that had happened in Graul. If we understood clearly what had occurred there we were at least in possession of information which would keep us from wrong ideas as to what might have happened in insular Britain. The incidents that are recorded are so brief and isolated that, taken by themselves, they fail to give us any idea of what was going on, but when we look at them in the light created by events in Gaul we perceive faint traces of a connexion between them ; it is the fading influence of the magic name, res- pvMica Bomana, and the efforts that were being made, secular and religious, to revive it for the salvation of the island. It was then for this purpose that he had given as professor two or three courses of lectures on this subject, and it is evident from such ^ Preface portions of his manuscript that remain that he had set out his work with the view to its publication. Some of the chapters he had completed, some were still fragmentary, and for each section he had pro- vided some notes or indications of notes, and in what Avas meant for an appendix he had discussed at greater length than was possible in the text one or two questions of especial importance. The manu- script of these lectures, just as it was found, was handed over to his friend, the late Professor York Powell, who very kindly undertook to see the volume through the press. This, however, he never accom- plished, and after his premature death the portions which he had worked off, a rough print of the rest, and such sheets of the manuscript as could be found, were returned to Mr. Freeman's executors. Professor York Powell had revised for the press sheets B to P, i.e. the first 224 pages. The rest was all in the rough, and called for arrangement, correction, and the verification of the references, an amount of work which his numerous engagements had probably made it impossible for him to accomplish. It is obvious, therefore, that the present volume suffers very much for lack of the author's final notes and arrangement, but it was felt that work so good, carried out on ground which had never before been so carefully considered, should not be allowed to remain un- published. It is now offered to the historical student, a mere earnest of what it would have been, and yet a fragment too valuable to be allowed to perish. T. SCOTT HOLMES. TABLE OF CONTENTS CHAP. PAGE I, The Invasion of Gaul .... 1 II. A Tyrant of the West .... 35 III. CONSTANTINE EmPEROR AND MaXIMUS TyRANT 81 IV. The Barbarian Invaders .... 130 V. West-Goths and Burgundians . . .171 VI. Wallia and the Settlement of Aquitaine . 226 VII. Theodoric the West-Goth and Aetius . 264 VIII. Chlodowig the Frank 288 APPENDIX 1. Aetius and Boniface . . .305 II. The Second Carausius . .371 INDEX 375 WESTERN EUROPE IN THE FIFTH CENTURY. I. [THE INVASION OF GAUL.] The movements within and without the Empire which, in the course of a few years at the beginning of the fifth century, altogether changed the face of Western Europe have never, as far as I know, been told in our own tongue, perhaps not in any other tongue, as a connected tale. The facts are recorded by Gibbon with his usual accuracy, clearness, and careful reference to authorities ; but they are scattered over several chapters and are never brought together in their relation to one another. To Gibbon, with Rome itself as his main subject, their importance lay chiefly in their purely Eoman aspect, as so many blows dealt to the power of Rome. To our latest English inquirer into these times they naturally come n the same way, important only as they bear on the destinies of Italy and her invaders. Mr. Hodgkin does not give, because he was not called upon to 4ve, a minute or a consecutive narrative any more aan Gibbon does. Of the German writers on the VolJcerwanderung, Dahn and Pallmann hardly touch B 2 Western Europe in the Fifth Century, [i. these particular years ; Wietersheim has a careful and critical examination of the facts and authorities ; but it hardly amounts to a narrative "^^ Of writers dealing specially with our own island, Lappenberg has a sketch, to the purpose as far as it goes, of the British side of the story, but he hardly attempts to connect it with the continental side. Mr. Green, in the Making of England, attempts no examination of authorities, and he gives a few words only to the continental side ; but it is clear that he had fully grasped the connexion between the two. Tillemont in a past age, Clinton in the age just before our own, have brought the authorities together with their usual painstaking research. And I venture to think that the time has not yet come when we can afford to cast away collectors whom no scrap of information in the original writers ever seems to escape. But Clinton does not attempt a narrative, and the narrative which the worthy Tillemont does attempt, though it is well to follow the example of Gibbon and Hodgkin in keeping it ever at our elbow, can hardly be looked on as sufficient according to the standard of modern criticism. Fauriel, in his Histoire de la Gaule Meridionale sous la domination des conquerants Germains, has used his authorities well, and he comes nearer than any other writer to giving a con- nected narrative of the events with which we are immediately concerned. Still his point of view, the point of /view of a countryman of Sidonius and Gregory, is distinctly South-Gaulish. It is no part * Dahn has since in his Urgeschichte come much nearer to a con- nected story. I.] The Invasion of Gaul, 3 of his business to take any special points to connect the continental with the insular story. As for myself, I must say that, while I have taken the deepest interest in attempting to put together a fuller and more connected narrative of the whole story than I have yet seen, and in the work which is the necessary condition of so doing, the minute examina- tion of the evidence of the original writers, I have a motive beyond. In much that I shall have to say from this Chair, I shall strive to guide you into Britain by way of Gaul, into England by way, if not of France, yet of the elements out of which France slowly grew. If I keep you long with the Goth and the Frank in their Gaulish realms, it will not be only because of the surpassing interest and instruction of their story in their Gaulish realms, but also because a full understanding of their position in their Gaulish realms is the best means to enable us by force of contrast to grasp the true position of the Angle and the Saxon in their British realms. I am leading you to Northumbrian Bseda by the guidance of Arvernian Gregory. If I am set in this Chair to strive to show that European history is one unbroken tale, I am set in it also to strive to show that Englishmen are Englishmen. I believe that the latest theories of all go once more to set aside that doctrine as an old wives' fable. Now I venture to think that the spritely youths who, I am told, blow their trumpet somewhat loudly to say that what they are pleased to call * the Teutonic theory' is exploded, have not given much of their time to any very deep study of Gregory of Tours. The plain truth, so despised of many, that B 2 4 Western Europe in the Fifth Century. [i. we are ourselves and not somebody else, is more easily grasped if we look first at the fortunes of those branches of our race which did not remain ourselves but did become somebody else, and see how utterly unlike those fortunes are to ours, I trust, before many terms are over, to set before you a dis- tinctly English story. As yet, I am dealing with our kinsfolk in foreign lands. The new theories will tell you that we were no more in our conquered island than they were in the conquered mainland. It is well then, before we examine what was the place that the Jute, the Angle, and the Saxon held in Britain, to understand thoroughly what was the place which the Burgundian, the Goth,, and the Frank held in Gaul. Of that inquiry the present course will bring us only to the threshold ; but it is a stage which cannot be left out. The main importance of these years lies in this, that in them the ground was made ready for the plantation of abiding Teutonic settlements in the three great lands of the West, in Gaul, Spain, and Britain. In Gaul, and still more in Spain, not only is the ground made ready, but the settlements actually begin ; in Britain the ground is made ready, but hardly more. In our meagre notices of Britain in these years Teutonic invaders are never distinctly mentioned. They have shown themselves at an earlier time as unsuccessful invaders ; they were soon to show themselves again as abiding settlers; but during the special years with which we are about to deal the Teuton shows himself in Britain at most as a passing plunderer of the coast; his future dwelling- L] The Invasion of Gaul. 5 place is making ready for him ; but lie does not as yet take any steps to secure possession. Yet even at this time our own people play no inconsiderable part in the story. It is not to be forgotten that there was a Saxony in Gaul before there was a Saxony in Britain; Bayeux was a Saxon city before Win- chester. Among all the invaders of Gaul the Saxon pirates of the coast are spoken of as the most dreaded, and the rovers of the Channel were not likely to keep themselves to its southern shore only, though it is only on its southern shore that they have found chroniclers of their doings. But beyond this, both at this time and in the generation when the Angle and the Saxon did begin to occupy the great island, it is of the highest moment to mark the connexion between the affairs of Britain and the affairs of the mainland. The Teutonic conquest of Britain, owing to the special circumstances both of the invaders and of the land invaded, took a wholly different shape from the Teutonic conquest of most parts of the main- land. But it was none the less part of the general Volherwanderung, and it was largely affected by the same causes as the Teutonic movements on the main- land. And. one side of the difference between the English conquest of Britain and the Frankish conquest of Gaul, namely the difference in the state of the invaded lands and their inhabitants, was largely owing to the events of these particular half-dozen years. At a first glance the events of these years may seem to offer us little more than a series of uninter- esting and almost unintelligible struggles for the crown of the declining Empire of Kome, or at any 6 Western Europe in the Fifth Century. [i. rate for the imperial dominion in the provinces beyond the Alps. Emperors or tyrants rise and fall, and, by a strange fate, men whose revolt at least shows them to have been men of some energy, are overthrown to the profit of an Emperor who at no time of his reign showed any energy whatever. Honorius cannot keep Rome from the barbarians ; but he can, by the hands at least of his generals, destroy every rival claimant of his diadem and can win back a large part of the provinces which they had usurped. We may safely say that Constantine, Gerontius, Jovinus, Heraclian, were any of them better fitted to reign than the son of Theodosius. But these men have a higher interest than comes from anything that connects them with Honorius. Their rise and fall are directly connected with some of the leading events in the history of the world ; their tale cannot be told without telling the tale of the separation of Gaul, Britain, and Spain from the Roman dominion ; the setting up and putting down of the rival tyrants cannot be recorded apart from the revolutions which at least opened the way for the growth of the leading nations of Western Europe. As usual, the history of these years has to be made out by piecing together a great number of authorities, none of which are of first-rate merit. We have an unusual wealth of accounts, such as they are, written by men who lived at the time ; but there is none who claims a high place as a narrator, still less is there any who could understand the full significance of his own days. Nor is there any who gave himself speci- ally to remark and to record that particular chain of I.] The Invasion of Gaul. 7 events with which we are specially concerned. All is fragmentary; one fact has to be found here and another there. The age, as one of the great turning- points of the world's history, needed a Polybios to grasp its full meaning ; we have not even an Ammi- anus to set down events in order and to make shrewd observations on them as he goes along. We can hardly doubt that the History of Olympiodoros, the Greek of Egypt, some scraps of whose many books are preserved to us by Ph6tios, would, if he had come down to us whole, have given us something more like a narrative, and that a narrative of some merit, than his followers. He has at any rate given us fragments of considerable importance, whose value has been fully set forth by Mr. Hodgkin. We seek in vain for some further knowledge and some further remains of the two writers quoted by Gregory of Tours, Sulpicius Alexander and Henatus Profuturus Frige - ridus. The collection of names borne by the last writer, with its Christian, its Koman, and its Teutonic elements, raises a certain curiosity about himself. Sulpicius may have concerned himself chiefly with the Franks, a people with whom we have at this moment less to do than with some others. From Orosius we have the complete work of a contem- porary; from Zosimos we have the nearly complete work of most probably a younger contemporary. Both the zealous Christian and the zealous pagan wrote with an object somewhat different from that of simply recording events as they happened, and the prejudices of both must be allowed for in measuring the value of their witness. Z6simos too, though 8 Western Europe in the Fifth Century. [i. a contemporary, one wlio was alive at tlie time and who wrote not very long after, can hardly be called an original writer. He seems to have written from the accounts of writers, some of whom could not have been much earlier than himself, but whom we may guess that he did not always understand. Though his account of these years seems complete, yet it is almost as fragmentary as those of 01ympiod6ros. It consists of pieces put together with very little regard to connexion or to chronological order, one most likely taken from one source and another from another. Yet some of the scraps of narrative thus embedded, whencesoever they may come, are of the highest moment. They preserve several of the most essential parts of our present story for which we should look in vain elsewhere. We have another narrative, full in some points,in the Ecclesiastical History of Sozomenos, also a writer contemporary, or nearly so. The writers of our own island in after times, British Gildas and Nennius, English Baeda, who in some measure follows Gildas, and the English Chronicler who in some measure follows Baeda, can of course tell us nothing of our times beyond such traditions, written or oral, as may have lingered on till their days. But it is always well to know how the events of a past age looked in the eyes of the descendants or successors of the men who were touched by them at the time. We are now in the age of the Annalists. And two of them, as being both contemporary and local, would, if they had written at greater length, have been the very best of all our authorities. Even as it is, the Aquitanian Prosper and the Spanish Idatius count I.] The Invasion of Gaul. 9 for as much as any of the more lengthy writers, and Idatius himself enlarges with some force when he comes to the sorrows of his own land. A British or an Armorican annalist, an annalist from the banks of the Rhine, would have been priceless indeed ; but for such we have to yearn in vain. Our nearest approach to such a help is found in that annalist on whom one side of the description of the Aquitanian annalist has so oddly been bestowed, and who commonly figures as Prosper Tiro. Whoever he was, and at whatever value we rate him in other matters, we are thankful for his few and short notices of that island world which the world of Eome seems largely to have forgotten. Above all, we are thankful to him for the one notice from outside, a notice seemingly contemporary, which has come down to us of the English Conquest of Britain. We get some help also from some writers in prose and verse whose object was not that of directly and simply recording events. We press into our service alike the pagan laureate and the Christian preacher. The stately hexameters of Claudian, the less famous elegiacs of the poet of Divine Providence, the long harangue of Salvian, the occasional notices of Jerome, all form part of our materials. Actions of Stilicho were, if not the true causes, at least the immediate occasions, of the events with which we are concerned; and where Stilicho acts, we presently hear the trumpet voice of the poet from whom we should never have learned that the devout Honorius was not a worshipper of Jupiter. Our most living picture of the invasion of Gaul itself comes from a poet of 10 Western Europe in the Fifth Century. [i. another kind, whom some have thought to be the annalist Prosper in yet another shape. Prosper or no Prosper, he is a contemporary witness, whose verses may be more safely accepted as true to fact than the sounding lines of Claudian. He is a man of Gaul who painted the sufferings of Gaul in which he himself had shared. His verse is written to point a moral, the moral of Divine Providence ; so is the prose of Salvian in his treatise of kindred title, where he gives his picture of the evils and sorrows of the time while discoursing of the government of God. We would fain believe that the Teuton was as virtuous and the Roman less vicious than the Roman preacher paints them; but we must doubtless apply the same rule to both, and take off something from the brightness of the one portrait and from the blackness of the other. Saint Jeiome we have to thank for a few fiery touches of the time, for a few geographical details, for a slightly puzzling list of nations, all which certainly add to our knowledge. Altogether our materials are far from scanty; many important periods are far worse off. We cannot venture to ask for a Polybios at every great turn of the world's history. We are inclined to lament that we have no such light as Ammianus throws on the century that goes before and Procopios on the century that follows. It is by a sound instinct as to the general march of events, though with some disregard to exact chrono- logy, that Baeda and the English Chronicler connect the separation of Britain from the Roman dominion with the Gothic taking of Rome. Rome was broken I.] The Invasion of Gaul. ii by the Goths, and since then no Eoman kings reigned over Britain *. It was not the actual taking of Kome, but it was that Gothic invasion of Italy of which the taking of Eome was the most striking * Bseda (i. 11) brings in his first date with some chronological solemnity. He had Orosius before him, but he leaves out and amplifies to suit his own purpose. His date stands thus ; " Anno ab incarnatione Domini quadrigentesimo septimo, tenente imperium Honorio Augusto filio Theodosii minore, loco ab Augusto quadragesimo quarto, ante biennium Romanse irruptionis quae per Alaricum regem Gothorum facta est, cum gentes Alanorum, Sue- vorum, Vandalorum, multseque cum his alias, protritis Francis, transito Hreno, totas per Gallias ssevirent." At the point of time thus marked, first Gratian and then Con- stantine are set up ; the history of Constantine follows, and then " Fracta est autem Roma a Gothis millesimo centesimo sexagesimo quarto suae conditionis, ex quo tempore Romani in Britannia reg- nare cesserunt, post annos ferme quadringentos septuaginta ex quo Gaius Julius Caesar eamdam insulam adiit." The English Chroniclers leave out all the former extract, and translate the second under the year 409 (in the late Canterbury version, 408). The fullest form is in the Peterborough version ; " Her waes tobrocen Romana burh fram Gotum, ymb XI hund wintra and X wintra Jjaes j^e eo getimbred waes. SicSSan ofer J)aet ne rixodan leng Romana cinigas on Brytene. Ealles hi tSaer rixodan IIII hund wintre and hund seofenti wintra sicStSan Gaius lulius ]3set land erost gesohte." The other versions have several small differences, and the word leng is in Peterborough only. And it may be noticed that, while Baeda does not imply that the Romans had ruled in the island ever since the landing of Gaius Julius, the Chroniclers do. Baeda leaves out the actual separation of Britain, as recorded by Zosiraos. Coming between the expedition and the taking of Rome, it got mixed up with both those events, and was lost between them. He also placed, like so many others, the taking of Rome in 409 instead of 410. 12 Western Europe in the Fifth Century. [i. incident, which led to that general breaking-up of the Eoman power in the West, of which the departure of the legions from Britain was that side which most directly concerned ourselves and our predecessors on British soil. As a matter of fact, Britain had really fallen away from the dominion of Kome before Eome was taken by Alaric. In truth, the actual taking of Rome, looked at as something having a practical effect on the course of events at the time, was of less importance than that it now seems to have or than it seemed to have in the eighth and ninth centuries. In more senses than one, Suis et ipsa Eoma viribus ruit. Eome had so thoroughly spread herself over the whole of her own world, the whole of that world had so thoroughly become Eome, that the direct impor- tance of the local Eome had come to be less than that of many other cities. Eome was neither a seat of government nor the guardian of an exposed frontier. Her actual capture and sack was a solemn and terror- striking incident, which gave endless opportunities for pointing a moral ; it was the sign that an old day was passing away and that a new day was coming ; it was a thing to be remembered in later days as no other event of those times was likely to be remem- bered; but at the moment it made little practical difference to any but those who immediately suffered by it. What really changed the face of Western Europe was not that Eome was taken but that Eome was threatened. It was the presence of Alaric in Italy, a presence of which the taking of Rome was as it were the formal witness, which opened the way for I.] The Invasion of Gaul. 13 the separation of the Western lands from the Empire and for the beginning of the powers of the modern world. Yet, at the moment when our immediate story begins, Alaric was not in Italy; he had entered the land and he had left it ; he had left it, as Eoman poets and official writers loudly proclaimed, a defeated man, chief of a people that Eome had crushed for ever*. He had entered Italy, it would seem, with Kadagaisus, as his ally. Such seems the express witness of such authorities as we havef. It may be that Alaric and Radagaisus entered Italy by distinct paths, and that the warfare of the Eoman armies in Ehsetia, which is described as happening at the same time as the coming of Alaric, may have been warfare directed against another Gothic leader who came in alliance with him \. The fight of PoUentia has been * The whole poem of Claudian on the sixth Consulship of Honorius is an expansion of this theme ; but it comes out most tersely in the inscription, if it he genuine, " Getarum nationem in omne sevum domitam." Hodgkin, i. 722-7. t Prosper distinctly couples Eadagaisus with Alaric ; " Stilichone et Aureliano coss. Gothi Italiam, Alarico et Eadagaiso ducihus, in- gressi." So Cassiodorus, changing the style to " Halarico et Eadagaiso regibus." X So Hodgkin, i. 711-33. The words of Claudian, De Bello Getico, 279, are; " Irrupere Getse, nostras dum Esetia vires Occupat, atque alio desudant Marte cohortes." This would certainly be more naturally taken of some movement in Ehsetia itself, quite distinct from the Gothic invasion. I believe there is no other reference to Eadagaisus as a partner of Alaric in this invasion. As for Ehsetia, we must not forget another obscure reference in Claudian, De Bello Getico, 414 ; 14 Western Europe in the Fifth Century. [i. variously described as a Eoman victory, a Gothic victory, and a drawn battle^'; it is certain that its practical effect was favourable to Eome. Alaric left Italy, and again, as in the last days of the fourth century, the Imperial power was undisputed through- out all the lands of the West. But that power was no longer what it had been even at the beginning of the last year of that century. When Stilicho entered on his second final consulship, whatever dangers seemed to threaten the dominions of the Western Emperor still came from the lands which were under the rule of his Eastern brother. The Eastern power of Eome, destined to live on unbroken for more than eight centuries, had been shaken by the coming of the Goth, and had needed the help of the West to rid itself for a while of his presence and his ravages. The Western division of the Empire, destined so soon to break in pieces, still seemed to be safely guarded by the arm of its consul. A few years before Stilicho had, we are told, restored the power of Eome on the Ehenish frontier almost by "Accurrit vicina manus quam Eaetia nuper Vindelicis auctam spoliis defensa probavit." Whatever this refers to, it can hardly be taken of a Gothic invasion under Eadagaisus. The question, however, though of some importance for the history of Italy, matters little for that of Gaul and Britain. * The question is discussed by Mr. Hodgkin, i. 722. It concerns us very little, as whatever was the military result of Alaric's invasion of Italy, it led to the withdrawal of the legions from the Rhine. I have followed, with Mr. Hodgkin (i. 734-6), the chronology of Pallmann (402 a.d.), which seems based on the sure witness of Prosper. T.] The Invasion of Gaul. 15 a look. Drusus and Trajan had been outdone. The Suevian and the Alaman obeyed the laws of Eome. The Frankish kings, with their long yellow hair as the badge of freedom and kingship, were set up and put down at Stilicho's bidding, and Francia — we long for a definition of its boundaries — would no more dream of casting forth the kings that Stilicho gave than Frovincia — ^we are almost tempted to use the later form of the name — would dream of casting out the immediate lieutenants of the Emperor. The Salian had betaken himself to the tilth of the ground ; the Sicambrian had beaten his sword into a pruning- hook ; the traveller crossed the border-stream or sailed along its waters, and asked which shore of Khine was that which Eome specially claimed as her own. Britain, delivered and guarded — waUed in, we are tempted to render it — at the word of the conqueror, had seen the Scot driven back to his own island ; she no longer feared the Pict, nor looked with dread lest every wind should bring the keels of the Saxon to her shores ''^ We wish that we had some further * Claudian's poem on the First Coosulship of Stilicho seems to be our only authority for these exploits. He specially enlarges on the speed of his patron's victories (i. 188-97) ; " Miramur rabidis hostem succumbere bellis Cum solo terrore ruant? Num classica Francis Intulimus ] Jacuere tamen. N'um Marte Suevos Contudimus, quis jura damus ? Quis credere possit ? Ante tubam nobis audax Germania servit. Cedant, Druse, tui, cedant, Trajane, labores. Vestra manus dubio quidquid discrimine gessit, Transcurrens egit Stilichon ; totidemque diehus Edomuit Rhenum, quot vos potuiatis in annis ; Qv£m ferro alloquiis, quern vos cum milite, solus." 16 Western Europe in the Fifth Century. [i. authority for this glowing picture than the laureate strains in which Claudian welcomed his patron's con- sulship ; but all cannot be imagination. Ten years Again (i. 218); " Tempore tarn parvo tot prselia sanguine nullo Perficis ; et Luna nuper nascente profectus Ante redis quam plena fuit." The Franks especially are subdued and subdued for ever (i. 203); " Ingentia quondam Nomina, crinigero Jlaventes vertice reges, Qui nee principibus, donis precibusque vocati, Paruerant, jussi properant, segnique verentur Oflfendisse mora." (i. 236.) " Provincia missos Expellet citius, fallax quam Francia reges Quos dederis. Acie nee jam pulsare rebelles Sed vinclis punire licet." (i. 220.) " Rhenumque minacem Cornibus infractis adeo mitescere cogis Ut Salius jam rura colat, flexosque Sygambrus In falcem curvet gladios, geminasque viator Cum videat ripas, quce sit Romana requiret." One longs for some otber account, even the driest entry in the Annals. Taken literally, the poet's words imply that Stilicho brought the Franks and other nations to submit, without striking a blow and even without the presence of an army. The account of Stilicho's doings in Britain is even vaguer (ii. 247); " Inde Caledonio velata Britannia monstro, Ferro picta genis, cujus vestigia verrit Cserulus, Oceanique sestum mentitur, amictus; Me quoque vicinis i)ereuntem gentibus, inquit, Munivit Stilichon, totam cum Scotus lernen Movit et infesto spumavit remige Tethys. Illius eflfectum curis ne tela timerem Scotica, ne Pictum tremerem, ne littore toto S[^ Prosjpicerem duhiis venturum Saxona ventis." I.] The Invasion of Gaul. 17 or more of quiet in Britain and on the German frontier seem to show that the successes of Stilicho in the first years of the two brothers, however they may have been tricked out by the poet's fancy, were real successes which did their work for a season. His Frankish successes especially seem to have been of real importance and to have had an effect on the events with which we are more immediately con- cerned. The Franks on the left bank of the Ehine, those who were settled within the borders of the Empire as its subjects, though sometimes turbulent subjects, the Salians presently to be so famous, appear in our story as discharging the duty of Eoman allies. But that such successes as those of Stilicho were needed to keep the professed subjects of the Empire, in their allegiance is the surest sign of the growing weakness of the Eoman power in the Western lands. It might be at any moment restored to its full geographical extent and to the outward form of its ancient authority. But the fabric of dominion needed constant propping, not to say rebuilding, and a time came when rebuilding was no longer possible. Before the fourth century was ended, before the year was ended to which Stilicho gave his name, Alaric was in Italy, and to withstand the presence of Alaric in Italy, the mainstay of the Eoman power in the Western lands out of Italy was taken sway. Whether Alaric won or lost the field of Pollentia, his coming indirectly tore away Britain from the Eoman dominion, and began the work of dismemberment in Gaul and Spain. For the Gothic invasion of Italy needed to be with- c 18 Western Europe in the Fifth Century. [i, stood with all the forces that the declining power of Kome could muster. If Pollentia was a Eoman victory, it was a victory that was won only by leaving the distant frontiers of Rome exposed to every invader. To meet Alaric came not only the troops which had lately defended Rheetia''^, but the troops that guarded the most distant outposts of Rome. The Rhine was left without its defenders ; the men who had kept watch against Chatti and Cherusci and the yellow Sicambri f — in these last at least we see the Ripuarian Franks — came to the defence of Italy ; so did even the legion which had guarded Roman Britain against the Pict and the Scot J. We are bidden to believe that, even when the legions were gone, the dread of the name of Stilicho was so great that it was enough to guard all these frontiers without material help. The over- throw of Alaric struck such fear into all hearts that no subject dared to revolt, no enemy to invade ; even proud Germany remained at peace, and did not risk the passage of the border-stream, although no * See above, p. 13, note %. t De Bello Getico, 420; "Quseque domant Chattos immansuetosque Cheruscos, Hue omnes vertere minas, tutumque remotis Excubiis Rhenum solo terrore relinquunt." X lb. 416; "Venit et extremis legio prsetenta Britannis, Quae Scoto dat fraena truci, ferroque notatas Perlegit exsangues Picto moriente figuras." As to the particular legion referred to, sixth or twentieth, see Hodgkin, i. 716. I.] The Invasion of Gaul. 19 soldier guarded its Roman bank *. And yet this daring flight of panegyric seems to have some ground of fact to start from. When Claudian wrote, things may well have been quiet on the German border ; for they seem to have remained so for more than two years longer. We have no record of any movements on the Ehine till the date, so minutely given, when, on the last day of the year 406 f, the great Teutonic invasion of Gaul began. It was an invasion, not an occupation. Those who now crossed the Ehine found no settled dwelling-place till they had crossed the Pyrenees as well. It was Spain, not Gaul, which the actual invaders of the moment tore away from the Empire. To Gaul the actual invasion was a frightful blow ; but, had nothing more come of it, it would have been only a passing blow. It was the working of this great movement on lands * De Bello Getico, 423 ; "Ullane posteritas credet? Germania quondam Ilia ferox populis, quae vix instantibus olim Principibus tota poterat cum mole teneri, Tarn sese placidse praestat Stilichonis habenis, Ut nee praesidiis nudato limine tentet Expositum calcare solum, nee transeat amnem, Ineustoditam metuens attingere ripam." He goes on with a comparison between Stilicho and Camillus his only equal ; "Vestris namque armis Alarici fracta quievit Et Brenni rabies." t Prosper Aq. ; "Arcadio VI et Probo Coss. Vandali et Alani, trajecto Eheno, Gallias pridie kal. Januarias ingressi." Clinton would read " Jun." for " Jan." placing it in the summer instead of in the winter. C 2 20 Western Europe in the Fifth Century. [i. beyond the bounds of Gaul which caused it to have any lasting effect on the state of Gaul itself. Our best authority speaks only of Vandals and Alans as having taken part in the invasion. Yet there can be no doubt that those other writers are quite correct who add the name of the Suevians to the list*. These three nations, Vandals, Alans, and Suevians, are those which we find a few years later establishing kingdoms in Spain. And of those we must remark that two only are strictly Teutonic nations. The Alans, though their history is so much mixed up with that of various branches of the Teu- tonic race, and though we may believe that they had become in some measure Teutonized, were in them- selves barbarians in the strictest sense of the word, aliens to Teutonic as well as to Eoman fellowship. Their invasion would of itself, under other circum- stances, have belono-ed to the same class as the later invasions of the Hun, the Avar, and the Magyar. * The chief of these is Orosius (vii. 39), whose account we shall have to examine presently. His list is " Gentes Halanorum et, ut dixi, Suevorum, muUceque cum his alice." So Zosimos (vi. 3), who tells the story rather out of place, to explain the causes of the move- ment in Britain which followed the invasion of Gaul, but which he tells before it. His words are iv rois irpoXa^oixn xpovois, eicrov rjbri TrjV vnaTov exovros 'ApKabiov Koi Upo^ov, BavbiXoi 2vr}0o7s Koi 'AXavols eavTovs avafii^avres tovtovs virep^avres tovs totvovs rois vnep "AXirecriv edvfo-iv fXvpTjvavTo. The Vandals are here made the kernel of the invasion, as they are also by Salvianus (Gub. D. vii. 12), who, in describing the Vandals, tells us how " excitata est in perniciem et dedecus nostrum gens ignavissima (cf. Livy, iii. 67 ; v. 28), quae de loco ad locum pergens, de urbe in urbem transiens, universa vastaret." I.] The Invasion of Gaul. 21 As it is, their migration is part of the Teutonic migration, a strange side of it, but one which we cannot separate from the other sides. It is an application on a great scale of the universal law that a great national migration always carries with it some who do not belong to the main stock of the invaders, but who are from some cause led to throw in their lot with them. In this way it may be per- fectly true, as we may be led to gather from the words of an ecclesiastical writer, that a crowd of other nations, Teutonic, Slavonic, Heruli, Gepidae, Sarmatians, Quadi, and many others *, had a share of some kind in the work. Detached bands of any of these nations or any others may have followed the lead of any of the chiefs of the movement. But, if so, they were lost in the general mass ; it was the three nations already spoken of. Vandals, Alans, and Suevians, that gave the movement its character ; it is these three that are distinctly visible in the story and in its results ; it is these three that made Gaul a highway to Spain, and that found in Spain an abiding place for a longer or shorter season. * Jerome in his letter (xcii. vol. iv. p. 748) to Agerucliia, gives his list ; " Innumerabiles et ferocissimse nationes universas Gallias occuparunt. Quidquid inter Alpes et Pyrenseum est, quod Oceano et Eheno includitur, Quadus, Wandalus, Sarmata, Halani, Gepi- des, Heruli, Burgundiones, Alemani, et, O lugenda respublica, Pannonii hostes vastarunt." The list reads very like a fancy one ; but there must be some special force in this mention of Pannonian enemies. Who are meant % The Huns % Fauriel (i. 39) seems to put the Gepidse of Jerome's list in the place which is held by the Suevians in most versions. In i. 42 he seems puzzled at hearing so little about them. 22 Western Europe in the Fifth Century. [i. As to the immediate occasion of the movement we are in the dark. It is hardly possible to reconcile the language of our authorities with the view that the Teutonic invaders of Gaul in this year were the remnants of the host with which the mysterious and terrible Eadagaisus, whether he had any share in the earlier invasion of Italy or not, certainly led into Italy the year before *. But whoever were the followers of Eadagaisus, it seems plain that they were utterly cut off in Italy by the generalship of Stilicho f. And all our accounts speak of the in- * This was the view of Gibbon and of the earlier writers to whom he refers. It does not seem to be adopted by any modern scholar (see Hodgkin, i. [733, 739, 824]), and it certainly is not suggested by the language of Prosper and the other writers. Gibbon rests on the phrase of Orosius, that the invading nations were stirred up by Stilicho. See note below on Orosius' charge. + Orosius heads his chapter (vii. 37) with the heading, " Ea- dagaisus hostis Italiam intravit et coesus est cum gente sua." And all his expressions are to the same effect ; " Eadagaisus solus . . . suos deseruit." The rest were worn out with hunger or taken prisoners. So Zosimos, vi. 26 ; uTrav to irokeynov navwXedpia di.e(f>- Bfiptv [6 ^TiXixcov] SxTTe fxrjbeva ax^^°^ ^'^ tovtchv TTepiarudrjvai, itXtiv eXa;^i(T- Tovs o(Tovs airos tji 'Pcopaicov irpoaedjjKe avp-fiaxi-a. One may be sure that there is exaggeration in all this ; but such phrases seem quite inconsistent with the notion that the Suevians, Vandals, and Alans who crossed the Ehine under their own kings were the remnant of this defeated host. The " luminous passage of Pi'osper's Chronicle," " In tres partes, per divisos principes, divisus [Eatlagaisi] exercitus," is not from the true Prosper, but from the chronicle so oddly called that of Prosper Tiro (see Hodgkin, i. 702-9 [founding himself on Holder-Egger, Neu. Arch. 1876]), and it goes on: "aliquam re- pugnandi Eomanis aperuit facultatem. Insigni triumpho exercitum tertise partis hostium, circumactis Hunnorum auxiliaribus Stilico I.] The Invasion of Gaul. 23 vaders of Gaul in this year as nations, nations crossing the Rhine by a fresh movement, not at all as the remnants of a defeated army. That the invasion was planned in concert with Badagaisus — if so, most likely in concert with Alaric — is perfectly possible * ; but it seems easier to suppose that the nations beyond the Rhine simply took advantage of the withdrawal of the legions which followed on Alaric's invasion of Italy. In any case the coming of these armed nations was not unexpected. Honorius, or those who were so busy at the work of legislation in his name, put forth more than one decree in which an attempt was made to provide for the defence of the provinces. But we hear nothing of any move- ments of the legions to the threatened frontier. We find instead, a touching appeal to the lovers of their country, the lovers of peace, to stand forth each man as his zeal and courage called him, and to do each man his duty in this hour of utmost need. The slaves, too, were called on to help ; in such a strait as the land was in it mattered more what a man could do than what was his state of life ; the slaves of the foreigners in the Boman service, and of those who usque ad internecionem delevit." It is hard to say what amount of value we should yield to this statement. Its exactness certainly looks as if it rested on some authority; yet it is hard to infer with Gibbon (Cap. xxx, note 4) that the "luminous passage" connects the history of Italy, Gaul, and Germany. The chronicler puts the invasion of Gaul two years later as a wholly distinct event. * So Faurielj i. 40. But the passage in Procopius, Bell. Vand. i. 3 (on which see note below), does not seem to bear on the matter. In the other passage which he refers to Jordanes, 31, the Vandals and Alans go into Gaul " ob metum Gottorum." 24 Western Europe in the Fifth Century. [i. ' were actually under arms, were specially bidden to go and fight by the side of their masters. The freeman was promised pay and part of that pay in advance ; the slave was promised a lesser pay, but accompanied by the precious gift of freedom *. Such an appeal from an Emperor who certainly had no thought of joining the muster sets us a-thinking ; * Clinton pointed out that some of the laws of Honorius, which had been thought to refer to the invasion of Italy by Radagaisus, really refer to the invasion of Gaul. They are dated in the consulship of Arcadius and Probus, that is 406. That which calls on the slaves is, what we should hardly have looked for, a little earlier than that which calls upon the freemen. Its date is Ravenna, xv. Kal. Mai., and it runs thus ; " Contra hostiles impetus non solas jubemus personas considerari, sed vires, et licet ingenuos araore patriae credamus incitari, servos etiam hujus auctoritate edicti exhortamus ut cum primum se bellicis sudoribus ofierant, prsemium libertatis, si apti ad militiam arma susceperint, pulveratici etiam nomine binos solidos acceptari : prsecipue sane eorum servos quos militia armata detentat, foederatorum nihilominus et dediti- ciorum, quoniam ipsos quoque una cum dominis constat bella tractare" (viii. Cod. Theod. Tit. xiii. De Tironibus, p. 387). Gothofred has a note on pulveraticv/ni, which here at least means the pay — earned by services amid the dust of warfare — which was to be the reward of the slave who turned a soldier. The other law (p. 388) is dated from Eavenna two days later ; " Provinciales pro imminentibus necessitatibus omnes invitamus edicto quos exigit ad militiam innata libertas. Ingenui igitur qui militise obtentu arma capiunt amore pacis et patrige, sciant se denos solidos paratis rebus de nostro percepturos serario, quibus tamen ternos ex summa supradicta jam nunc solidos prsebere mandabimus; nam optimos futures confidimus, quos virtus et utilitas publica necessitatibus obtulit." We are not told whether the owners of the liberated slaves were to be paid their value, which would seem to be only recoverable in the case of masters who were themselves under arms. T.] The Invasion of Gaul. 25 among things we notice that the meaning of the word country — fatria — has widened a good deal since a prince who moved from Rome to Capri was held to have forsaken his country "'^ The Roman name, now shared by all free inhabitants of the Empire, was held to have created a country and a nationality which, artificial as they might be, were deemed, at least officially, to be capable of calling up the feeling of patriotism in men's hearts. The barbarians then were making ready for the great migration, and the Romans were at least called upon to make ready to withstand them. But are we to believe that he who before all men united both characters, the greatest of living warriors, barbarian by descent, but beyond all men Roman by calling, had stirred up the nations which now poured into the Empire which he had twice saved % At least one contemporary writer tells us, and at least one later writer copies his tale, that the invaders of Gaul were led thither by the invitation of Stilicho f. He hoped, * Tacitus. + Orosius (vii. 38) after some other hard words against Stilicho, charging him among other things with sparing the Goths, goes on ; ' Prseterea gentes alias copiis viribusque intolerabiles, quibus nunc Galliarum Hispaniarumque provincise premuntur ; hoc est Halanorum, Suevorum, Wandalorumque, ipsoque simul motu impulsos Burgundionum ultro in arma soUicitans deterso semel Eomani nominis metu suscitavit." And in a following chapter (vii. 40) we read ; "Ante biennium Eomanse irruptionis [the taking of Eome by Alaric] excitatae per Stiliconem gentes, Halanorum, ut dixi, Suevorum, Wandalorum, multseque cum his alias, Francos proterunt, Ehenum transeunt, Gallias invadunt, directoque impetu Pyrenseum usque perveniunt, cujus obice ad tempus repulsse per 26 Western Europe in the Fifth Century. [i. we are told, that by raising a storm which he trusted to quell, but which none other could, he might be able to transfer the Empire from his son-in-law to his son*. The tale is the statement of an enemy, but, circumjacentes provincias refunduntur." Gibbon (c. xxx. note 46) on the phrase " excitatse a Stilichone gentes," says, " They must mean indirectly. He saved Italy at the expense of Gaul," By a somewhat forced construction this meaning might be put on the second passage of Orosius ; but the first distinctly asserts direct dealings with the invaders on the part of Stilicho. Of the Burgundians mentioned by Orosius we shall have to speak again. To the same effect is the chronicle called Prosper Tiro ; " xvii [Arcedii et Honorii]; Diversarum gentium rabies Gallias dila- cerare exorsa, immissa quam maxime Stiliconis indigne ferentis filio suo regnum negatum." It is not till three years later that he mentions the three nations spoken of by Orosius. The bitterest enemy of Stilicho is the poet Kutilius Namatianus ; but his verses (ii. 41 et seqq.) speak rather of Stilicho as letting the Goths into Italy than as doing anything with regard to Gaul. He tells us how Stilicho " Immisit Latise barbara tela neci : Visceribus nudis armatum condidit hostem, Illato cladis liberiore dolo. Ipse satellitibus pellitis Roma patebat Et captiva prius quam caperetur erat : Nee tantum Geticis grassatur proditer armis ; Ante Sibyllinae fata cremavit opis." Sdzomen (viii, 25 ; ix. 4) refers to the charges of treason against Stilicho, but does not speak of this particular charge, unless it lurks in the description of him as -navras ws il-niiv ^ap^dpovs re Koi 'Pdfiaiovs TTfidonevovs ex(ov. Nor has Philostorgius (xi. 3 ; xii. 1), though seemingly hostile to Stilicho, anything about Gaul. See also Dahn, K'dnige der Germanen, v. 42 ; Wietersheim, ii. 138. * The words of Orosius (vii. 38) are ; " Sperans miser sub hac necessitatis circumstantia, quia et extorquere imperium genero I.] The Invasion of Gaul. 27 even as the statement of an enemy, it is strange. Yet we can hardly doubt as to disbelieving it. It is not a statement of visible facts : it is a surmise or a mere invention, such as we are used to in all ae^es. In the eyes of Stilicho's enemies, any mischief that happened was necessarily Stilicho's work. In any case Stilicho and his legions did not this time fly to the defence of the Gaulish border; nor do we hear to what extent either the patriotic youth of Gaul or the able-bodied slaves of the barbarian mercenaries took up arms at their distant Emperor's bidding, to defend the peace of their country. Such fighting as was done seems to have been the work of defenders of the Empire of another kind. For Vandals, Alans, and Suevians at least did not enter the Gaulish provinces without finding an enemy to withstand them. Something was done in the way of diplomacy or bribery. One Alan leader, Goar by name, was persuaded to forsake the hostile enter- posset in filium, et gentes barbarse tam facile comprimi quam coramoveri valerent. Itaque ubi imperatori Honorio exercituique Eomano bsec tantorum scelerum scena patefacta est, commoto justissimo exercitu occisus est Stilico, qui, ut unum puerum purpura indueret, totius generis humani sanguinem dedit." Orosius bas a suspicious knowledge of the inner workings of Stilicho's mind, for which he is not so good a witness as he is for the crossing of the Rhine. The oddest thing of all is the confusion of Gregory of Tours (ii. 9), who misreads the second reference of Orosius into a campaign of Stilicho at the head of the barbarians ; " Horosius autem et ipse historiographus in septimo operis sui libro ita commemorat, Stilicho, congregatis gentibus, Francos proterit, Ehenum transit, Gallias pervagatur, et ad Pyrenseum usque perlabitur." 28 Western Europe in the Fifth Century. [i. prise, and to enter the service or alliance of Kome. And if the Komans of Gaul failed in their duty, the allies of Kome on the Gaulish border at the present stage of affairs did theirs manfully. The Franks, that is clearly the Kipuarian Franks on the right bank of the Khine, met the Vandals in battle. The Vandal king Godegisl and twenty thousand of his warriors were slain ; the whole Vandal host would have been cut to pieces if the Alan king Eespendial had not come to its help *. The Franks were overthrown by their joint forces, and the invaders seem to have met with no further resistance in passing the border stream or in spreading themselves where they would over the whole land. The districts first to be harried were naturally the lands which^ under Eoman dominion, still bore the German name f , * This comes from one of tlie lost writers who were made use of hy Gregory of Tours (ii. 9), him who bears the names of Eenatus Profuturus Frigeridus, which there, as Gibbon (c. xxx. note 28) re- marks, " denote a Christian, a Roman subject, and a semi-barbarian." " Interea Eespendial rex Alamannorum, Goare ad Eomanos trans- gresso, de Eheno agmen suorum convertit, Wandalis Francorum bello laborantibus, Godigyselo rege absumpto, acie vigenti ferme minibus ferro peremptis, cunctis Wandalorum ad internitionem delendis, nisi Alamannorum vis in tempore subvenisset." Wieter- sheim (ii. 158) is of course right in reading "Alanorum" for " Alamannorum." The mistake is more likely to be due to Gregory than to Eenatus. This is the explanation of the two words of Orosius, " Francos proterunt." Of Goar we shall hear again. Wietersheim points out the error of Procopius (Bell. Vad. i. 3) in making Godegisl lead the Vandals into Spain ; " Doch ist dieser Schriftsteller iiber Fiihereres anzuverlassig." t Salvianus, De Gal. vii. 12 ; "Primam a solo patrio effusa est I.] The Invasion of Gaul. 29 and whicli by that name might seem almost to invite the kindred invader. Thence they passed into the specially Belgian land, the Franks, it would seem, no longer withstanding them. Thence they passed into the flourishing land of Aquitaine, and step by step spread themselves over the whole of Gaul, through which they marched and harried as they thought good by the space of three years. Of the sufferings of the land we have more than one vivid picture from contemporary hands. Not the castles perched on the rocks, not the towns crowning the lofty hills, not the cities girded by their rivers — the poet of Divine Providence knew well how to hit off the characteristic features of Gaulish sites — could withstand the craft and the arms of the barbarians *. The head of all, the Imperial dwelling of Constantine and Yalentinian, Augusta of the Treveri, shorn now in common speech of its Imperial style, now under- went one of the many sieges and storms that it suffered in that age. All the usual horrors of a sack, fire and sword and leading into captivity, fell on the in Germaniam primam, nomine barbaram, ditione Romanam ; post cujus primum exitium arsit regio Belgarum, deinde opes Aquita- norum luxuriantiam, at post hsec corpus omne Galliarum, sed pauUatim id ipsum tamen, ut dum pars clade cseditur, pars exemplo emendaretur." He is here speaking of the Vandals. The words in italics seem quite inconsistent with the notion that these Vandals had formed part of the host of Radagaisus. * Carmen de Divina Providentia, 35 ; "Non castella petris, non oppida montibus altis Imposita, aut urbes amnibus sequoreis, Barbarici superare dolos atque arma furoris Evaluere omnes : ultima pertulimus." 30 Western Europe in the Fifth Century. [i. devoted city. The streets ran with blood and were heaped with dead bodies ; the buildings were black- ened with the flames. We are even told, in the usual style of exaggeration, that the whole city was burned. For it is certain that Trier was not left desolate without an inhabitant. It still remained a city ; and, when the storm had passed by, the first thought of its citizens, of the nobles who seem to have escaped the sack, was to send their prayer to the Emperors that the games of the circus might begin once more among them*. We are not told by which of the nations that shared in the invasion this present overthrow of Trier was wrought ; nor is any such distinction observed in the case of any of the other towns that are specially named. Mainz, Moguntiacum, was stormed and thousands of its people were slain, in the great church f . Venerable * Salvianus, vi. 15; " Excisa ter continuatis eversionibus summa urbe Gallorum, cum omnis civitas combusta est. . . . Exciclio unius urbi adfligebantur quoque alise civitates." The hon'ors are painted in full. Then we read; "Pauci nobiles qui excidio superfuerant, quasi pro summo deletse urbis remedio circenses ab imperatoribus postulabant." They are then soundly rebuked. The phrase "ab imperatoribus" would seem to point to the lawful Emperors, Honorius and his colleague, rather than to Constantine or to any other of the tyrants. If so, the petition can hardly have been made for some years, when things may have mended a little. It might be in 409, when Honorius acknow- ledged Constantine as a colleague. Salvian clearly exaggerates in his description. Trier was certainly not utterly destroyed, as is witnessed by the buildings earlier than this time, the basilica and parts of the metropolitan church among them, which still remain. t Jerome ad Ageruchiam, xci. s. a. 748 ; " Maguntiacum nobilis I.] The Invasion of Gaul, 31 as the present representative of that church is, it does not, like the great church of Trier, itself survive as a witness of those awful times. Vangiones, Worms, fell after a long siege ; we might even infer that for a while the city ceased to exist. Rheims, Amiens, Arras — the tribal name had already supplanted the name of the city — Nemetae and Argentoratum, cities to be more famous under their later names of Speyer and Strassburg, suffered the same havoc as Trier and Mainz. The Morini, most distant of mankind, did not escape in their home at Terouanne. Of the towns of northern Gaul no other, save Tournay, is named ; but the like havoc went on through the whole country. None escaped save a few of the towns of the Lyonnese and Narbonnese provinces, of Aquitaine and of Novempopulania, the later Gascony, One city alone of the south is specially mentioned ; Toulouse was in some way spared yet greater sufferings by its bishop Exsuperius, but the griefs which the city did undergo brought tears to the eyes of those who heard of them *. Heathens and heretics cared nought quondam civitas capta atque subversa est, et in ecclesia multa hominum millia trucidata." He does not speak of Trier. * lb. *' Vangiones longa obsidione deleti. Eemorum urbs prgepotens, Ambiani, Atrabatse, extremique bominum Morini, Tornacum, Nemetse, Argentoratum, translati in Germaniam. Aqui- tanise, Novemque Populorum, Lugdunensis et Narbonensis pro- vincise, prseter paucas urbes populata sunt cuncta. . . . Non possum absque lacrimis Tolosse facere mentionem, quae ut hucusque non fueret sancti episcopi Exsuperii merita prsestiterunt." In bis next letter to Julian, Jerome describes bis correspondent's losses and sufferings tbrougb an incursion of barbarians whicb seems to be tbe same. 32 Western Europe in the Fifth Century. [i. for sacred places, kings and persons, for the hallowed church and its vessels, for the devout widow, for the consecrated virgin, for the hermit who had withdrawn from the world to serve God in his solitary cave. Barbarians, we are told, cared not for age or sex ; they slew the innocent children with no more mercy than those whose death might be the just punishment for the sins of a longer life■^^ Those who escaped the sword escaped it only to pass into bondage. The sujfiferings of the clergy are told by one of their own body. They were scourged with whips, branded, loaded with chains. The poet himself had to march under the rod along the hard and dusty road among the wagons and weapons of the barbarians, while his aged bishop, torn from his burned city, led his people like the banished shepherd of a flock of wounded sheep f. As usual one plague followed on another ; * Carmen de Provideritia, 41 ; " Majores aniii ne forte et nequior setas Offenso tulerint quse meruere Deo : Quid pueri insontes, quid commisere puellse Nulla quibus dederat crimina vita brevis % Quare templa Dei licuit popularier igni ? Cur violata sacri vasa ministerii ? Non honor innuptas devotee virginitatis, Non texit viduas relligionis amor. Ipsi desertis qui vitam ducere in antris Luerant, laudantes nocte dieque Deum, Non aliam subiere necem quam quisque profanus ; Idem turbo bonos sustulit atque malos." t lb. 53 ; " Nulla sacerdotes reverentia nominis almi Diseruit miseri suppliciis populi ; I.] The Invasion of Gaul. 33 if leading into captivity was the fate of those whom the sword spared, the sharp hunger came in the end to slay them who escaped leading into captivity *. Three years of havoc like this wasted the land. No help could come for Kome or Eavenna. The some- thing which professed to be help came from another quarter, though in truth the help rather took the shape of adding the curse of civil war to the curse of barbarian invasion. The troops that still kept Britain for Eome passed over into Gaul. Britain was lost ; we can hardly say that Gaul was saved. The barbarians presently passed on to ravage another Roman land, and so Sic duris csesi flagris, sic igne perusti, Inclusse vinclis sic gemuere manus. Tu quoque pulvereus plaustra inter et arce Getaium Carpebas duram hoc sine fasce viam, Cum sacer ille senex plebem usta pulsus ab urbe, Cum pastor laceras duceret exsuisfeves." " Tu" seems to be the poet himself. He had said ; " Heu csede decenni Vandalicis gladiis sternimur et Geticis." The ten years would count from the invasion in 406 to the peace and restoration of Placidia in 416. It is possible therefore that the description of himself and his bishop may belong to a later time than 406-7, and that " Getse " should be taken in the strict sense of " Goths." Yet the picture would seem to refer to an incursion of altogether untamed barbarians rather than to the move- ments of the comparatively civilized West-Goths. * Jerome, u. s. " Quas [urbes] ut ipsas foris gladius intus vastat fames." One thinks of the picture drawn in our own tongue (Chron. 1086), " Da ]?a wreccse men laegen fordrifene full neah to deatSe, and sicScSan com se scearpa hunger and adyde hi mid ealle." D 34 Western Europe in the Fifth Century. much of Gaul as clave to the Eoman name was left to be torn in pieces by adventurer after adventurer who rose up to take his chance of winning the rule of Gaul or, if his luck carried him so far, the rule of tlie whole Western World. 11. [A TYRANT OF THE WEST.] We have thus far seen the more part of Gaul harried by invaders, both Teutonic and otherwise, from beyond the Ehine. From the sufferings of another land, from the doings, partly of strangers, partly of more distant kinsfolk, we must turn for a moment to look at the doings of our own people, we must turn for a somewhat longer time to look at the fates of the land which the doings of our own people were before long to make our own land. One annalist of this time, not to be sure the one highest in authority, but the one who seems to have kept his eye most steadily fixed on the matters which most immediately concern ourselves, speaks of Gaul at this time as a land ravaged, not only by Vandals and Alans, but also by Saxons *. Now fully to under- * The Pseudo-Pro sper, or whatever we are to call him, has under the sixteenth year of Honorius (409) this entry ; " Saxonum incursione devastatam Galliarum partem Wanali atque Alani vastavere ; quod reliquum fuerat Constantinus tyrannus obsidet." The entry is odd ; he had mentioned (see above, note, p. 26) the invasion of 406 in its right place, but without the mention of Vandals, Alans, or any nations by name. And one does not see why he specially places the harrying of Gaul by Vandals and Alans in the year in which they left Gaul. Still he could hardly have imagined a Saxon inroad, if none had taken place ; and he P 2 36 Western Europe in the Fifth Century, [ii. stand the course of things in the fifth century, it is ever needful to bear in mind that the events which led to the settlement of Angles and Saxons in the isle of Britain, and thereby to the growth of the Enolish nation in that isle, do not stand alone. Thev form part, we should never forget, no less than the settlements of Burgundians, Franks, and Goths, of the great tale of the Wandering of the Nations. But the story of the Angles and Saxons differs widely in every detail from the story of the Franks and Bur- gundians. How far is the difference marked by the distinction which has for many ages divided the Teutonic race into the two great branches of High and Low % It might be hard to say how far that distinction, a distinction which we may most truly describe as the parting off of the later High-Dutch forms from the elder forms common toGothand Saxon, had already gone in those days. The later Franks, the Eastern Franks, the Franks of the Carolingian age, appear as a High-Dutch people, at any rate as a people ruled by kings whose speech is High-Dutch. But the names borne by the kings of the Merowingian house distinctly keep the Nether-Dutch forms *, and the first settlements of the Franks, those at least coDBects it with the Vandal invasion, seemingly placing it a little earlier. And the entry of this year strangely connects itself with the entry of the year before, to which we shall presently come again. Most likely he is right in his facts and confused in his dates. * The names for instance beginning with Theod- take essentially the same shape that they would in Gothic or English. The thorn lives ever, in writing at least, in the French forms of Thierry and Thibault. So in Chilpe-ric we have the Low-Dutch form AeZp-. I II.] A Tyrant of the West. 37 which mark their first appearance in trustworthy- history, are found in those lands on both sides of Ehine which, wherever the speech of the people has been allowed to abide, are Nether-Dutch still. To Gregory of Tours the city which had been Argento- ratum was still Nether-Dutch Strateburg *. The difference is most likely merely one of chronology; when the first Frank ish conquests began, High- Dutch was not yet, and Chlodwig, no less than Alaric, spoke a tongue essentially the same as our own. But if the Saxon and the Frank were still all but one in point of language, their conditions and their relations to other men were widely different. There is already a wide gap between the Northern and the Southern German ; we might rather say between the German of the sea and the German of the land. The German of the land is already either an ally of the Empire, serving in its armies and loaded with its honours, or else he is an experienced invader of its continental provinces. Very often he flits to and fro between the two characters ; but in either or both he has become familiar with Eoman things ; even as an enemy he is not untouched with admiration and reverence for the state of things into the midst of which he forces his way. In the Gaulish wars of the fifth century the Frank steadily appears as the ally of Kome, till he finds it convenient to overthrow the last remnant of Eoman authority in Gaul, and that, it may be, in the character of the officer of a lawful Emperor overthrowing the rule of a tyrant f. * Greg, Tur. ix. 36, x. 19. t It is, to say the least, worth arguing whether Chlodwig did 38 Western Europe in the Fifth Century, [ii. All the incursions of the Franks, like those of the Goths, are made by land. Both Franks and Goths have been heard of on the water in earlier days * ; but on the water they wrought only sudden and passing exploits ; the historic life of both those nations was wholly a life by land. Altogether unlike them in this age are the northern Germans, the Germans of the sea, the men who have not been brought within the magic circle of Koman friendship and Koman enmity, who have yet to be taught that feeling towards the mighty past of the Empire and its still abiding present which was felt ahke by the heathen Frank and the Arian Goth. I said that they had not been brought within the magic circle of Eoman enmity any more than within that of Eoman friendship. They had indeed felt, alike in Gaul and on the coasts of Britain, the might of Eome when Kome was ruled by Valentinian f ; but they had simply been beaten back in isolated invasions ; the German of the sea had not gone through the same unbroken apprenticeship to Eoman ways which the German of the land found as much in his warfare against Eome as in his warfare under the Eoman banners. The main cause of the difference doubtless lay in the fact that he was the German of the sea. The Saxons of this age answer to the Danes and Northmen of the ninth and tenth centuries, in whom, not overthrow Syagrius in tte character of an officer of the Emperor Zeno. * See Aurelius Victor, Csesares 23, Zosimos, 1. t See the Saxon wars of Valentinian's reiga in Ammianus, xxvii. 8, xxviii. 2, 5. II.] A Tyrant of the West. 39 after an interval of some ages, the heathen and sea- faring Teutonic warrior again comes to life. But the Danes and Northmen come on us as a kind of second outburst of the great Wandering after a lull of centuries ; the Saxon expeditions and settlements of the fourth, fifth, and sixth centuries are as essen- tially a part of the first great movement as the marches of Alaric ; they form one special side of it, a side that has a character of its own, but both are alike parts of the same great drama. Specially must it be borne in mind that the Saxon inroads of the fifth century, just like the Scandinavian inroads of the ninth, touched both sides of the Channel alike, and that settlements were made on both sides alike. The Saxon of the fifth century seems to have been before all things a haunter of the Channel ; we do not hear of him now, as we do at a time a little earlier*, as threatening the shores of Northern Britain. It is the British sea, the sea which parted Gaul and Britain, which was his special home ; it was there that it was his sport to cleave the wave with his light barks clothed, not yet with iron, but with the skins of slaughtered oxen ; there it was that the men of Armorica were ever looking for hira as the sea-rover who was to bring desolation to their coasts t- Yet he did not always keep himself within * See the passages from Claudian collected in N. C. i. 11; " Maduerunt Saxone fuse Orcades," &c. t Sidonius, Pan. in Avitum, 369 ; " Quin et Aremoricus piratam Saxona tractus Sperabat, cui pelle salum sulcare Britannum Ludus, et assuto glaucum mare findere lembo. 40 Western Europe in the Fifth Century, [ii. the narrow seas; he could brave the strength of Ocean himself, and show himself as a sudden enemy on the western coast of Gaul * . In the eyes of the men on whose shores they showed themselves every Saxon was a sea-robber, a chief of sea-robbers ; plunder was the one trade which they all learned ; it was the one work which the leaders enforced, and which the followers undertook with gladness t. The Saxon was an enemy at once fiercer and more wary than all other enemies ; he was schooled in ship- wrecks ; no danger daunted him ; he fell suddenly on those who did not look for him ; he escaped in safety from those who was looking out for his coming \. The gods of his bloody creed called for the Francus Germanum primum Belgamque secundum Sternebat, Rhenumque, ferox Alamanne, bibebas Eomanis ripis." The advance of the Saxons by sea and of the other nations by land is here well marked. * See the letter of Sidonius to Nammatius (Ep. viii. 3, ed. Baret), where he gives a picture of the Saxon sea-rovers (cf. Hodgkin, Invaders of Italy, ii. 366). News is brought from Saintes that his friend was called upon " littoribus Oceani curvis inerrare contra Saxonum pandos myoparones." Much learning about the " myoparones " will be found at p. 506 and p. 98 of the old edition of Sidonius by Savaron. Some seem to make the " myoparones " mere coracles, such as seem to be implied in the word " pelle " in the extract in the last note. Others, who are surely right, make them much larger ships, doubtless the " lembi " of the same extract, the Illyrian X€>/3oi of which we read so often in Polybios. They are, I presume, the " keels " of our own story, t lb. ; " Saxonum . . . quorum quot remiges videris, totidem te cernere putes archipiratas ; ita simul omnes imperant, parent, decent, discunt latrocinari." % lb. ; " Hostis est omni hoste truculentior. Improvisus aggre- II.] A Tyrant of the West. 4i slaughter and torture of his captives; when he was about to turn his sails from the mainland to his own home, he deemed it a sacred duty to pick out one man out of every ten to perish by a cruel death as the thank-offering of their captor's piety *. So did our fathers look in the eyes of the man who has made the Gaul of the fifth century a living thing for all time. That the Saxon in these days never appears as the ally of Eome is hardly needful to say. While the Frank fights under the Imperial banner, the Saxon leagues himself with the Goth as the enemy of both t. It is important to notice that the Saxons of these days, inhabitants, one might almost say, of the British seas, not only harried on both sides of the ditur, prsevisus elabitur : spernit objectos, sternit incautos ; si sequatur, intercipit, si fugiat, evadit. Ad hoc exercent illos naufragia, non terrent. Est eis qusedam cum discriminibus pelagi non notitia solum sed familiaritas." * lb. ; '' Priusquam de continenti in patriam vela laxantes, hostico mordaces ancoras vado vellant,mos est remeaturis decimum quemque captorum per sequales et cruciarias poenas, plus ob hoc tristi quod superstitioso litu necare, superque collectam turbam periturorum mortis iniquitatem Eortis sequitate dispergere. Talibus eligunt votis, victimis solvunt ; et per hujusmodi non tarn sacrificia purgati quam sacrilegia polluti, religiosum putant csedis infaustse perpetratores, de capite captivo magis exigere tormenta quam pretia." Orosius also (vii. 32) paints our early picture ; " Saxonum gen- tem in oceani littoribus et paludibus inviis sitam, virtute atque agilitate terribilem, periculosam Eomanis finibus." t In the complicated alliances in Gaul between 463 and 468 (see Greg. Tur. ii. 18) we find Romans, Franks, and Britons on one side, Goths and Saxons on the other. 42 Western Europe in the Fifth Century, [ii. Channel, but establislied themselves onbothsides. The Saxon of Bayeux and the later Saxon of Winchester were colonists who went forth as parts of one general movement, and just as among the Scandinavians of a later time, the same keels often show themselves on both sides of the narrow seas. The results of the settlements have indeed been widely different in the two cases. The coming of the Saxons, along with the kindred Jutes and Angles, of whom, by those names at least, we hear nothing in Gaul, wholly changed the face of Britain. The face of Gaul was immeasureably less changed than the face of Britain, and, so far as it was changed, it was mainly the Frank who changed it. In the new Teutonic Britain which the events of this century called into being, the Saxon is one of the two great elements alongside of the kindred Angle. In Gaul we must always remember that the Saxon is a real element in the mixed population ; but he is a very subordinate element. His work was local and temporary. He kept a field ready for the coming of the Norman. The Scandinavian invaders of Gaul in the ninth century, the Scandinavian settlers in Gaul in the tenth, found a land already partly Teutonic to receive them ; the truest Normandy, Normandy west of Dives, is specially Norman because it is partly Saxon. One main reason which made the Saxon settlements in Britain so much greater and more lasting than those in Gaul was doubtless that in Britain the Saxons and their fellow- invaders by sea, Anglian and Jutish, had the field to themselves. They came in small parties, a few II.] A Tyrant of the West. 43 keels at a time ; but they had no Teutonic rivals in Britain, like Goths, Burgundians, and Franks in Gaul, coming by land, and naturally coming in far greater bodies. In Gaul therefore Saxon settlements were small and scattered, and they were gradually merged in the greater Teutonic elements in the country. In Britain they were also small and scattered, but there was nothing to interfere with their growth, except the resistance which they might meet from the Roman and Celtic elements in the land. From the Celtic element in Britain, Saxons and Angles did indeed meet a long and stubborn resistance, such as none of the Teutonic conquerors of Gaul met from any enemy. The Teutonic kingdoms in Gaul were founded in a moment ; all save one fell in a moment. The Teutonic kingdoms in Britain, so much smaller in extent, were the work of generations ; and they did not fail, but were merged into a single kindred whole. But the main difference of all is that in Britain the Teutonic conquerors displaced the Celtic andRoman inhabitants in a way that they never did in Gaul. They did, not only as in Gaul, form one element among the people of the land ; they became the people of the land itself. They made the land England in a sense in which Gaul never became Gothic or Burgundy, or even France. Some of the causes which led the way to the wide difference between Teutonic settlement on the northern and the southern sides of the Channel will meet us in the chain of events which we have just now reached. The same annalist who speaks thus casually of 44 Western Europe in the Fifth Century, [ii. Saxon harry ings in Gaul of which we have no other record, does not directly connect that harrying with anything that happened in Britain ; but he has a remarkable entry the year before, the meaning of which seems to be that the Roman power in Britain then practically came to an end *. This entry would seem to be meant as a short summary of a chain of stirring events of which we can, by the help of writers who record things more at length, put together something like a continuous narrative. Britain, the other world, was stirred, as Honorius himself was stirred, by the great movement of the Teutonic nations beyond the Ehine. We may perhaps venture to guess that the Saxon harrying of Gaul, so darkly hinted at, had already taken place, and that it had been accompanied by some Saxon harryings in Britain. However this may be, the legions in Britain, forsaken by their Emperors at Bavenna, feared lest the storm which was sweeping over Gaul should spread to Britain alsof. In such a case they took the law into their own hands. While the Germans and Alans were gathering, while Honorius was calling on the patriots of Gaul to arm, the army of Britain chose an Emperor, a tyrant, of their own, Marcus by name. The step was not new. * This is the famous entry in the so-called Prosper Tiro under the fifteenth year of Honorius (408) ; " Hac tempestati prse vale- tudine [al. viribus] Eomanorum vires funditus attenuatse Britannise." The reading is doubtful, and the phrases anyhow are odd ; but there seems no doubt as to the general meaning. + The well-known saying of Jerome, Ep. xlii. ad Ctesiphontem (vol. iv. p. 481), "Britannia fertilis provincia tyrannorum." IT.] A Tyrant of the West. 45 Britain was already known as a land fruitful in tyrants *. There Carausius and Allectus had reigned ; * Olympiodoros, p. 451 ; eV ravrais rats BperavviaK, TtpXv fj 'Ovoipiov TO e^dofxov vnaTevaai, es crTatnv Spfirjcrav to eV avrais crTpaTiwTiKov, MdpKov Tiva aveinov avroKpuropa. As Wietersheim (ii. 160) remarks, the date is fixed to 406, as the seventh consulship of Honorius was 407. We thus see the effect of the mere movement of these nations before they actually crossed the Rhine. Olympiodoros tells the story out of place to account for the elevation of Constantino, yet this date so carefully given must surely be trustworthy. It most likely marks the revolt as taking place late in the year 406, perhaps while the two Augusti were already consules designati. Zosimos in his later account (vi. 2 — he has two earlier references to Constantino) seems not to have noticed the force of npo, and places the elevation of Marcus in the seventh consulship, or 407 ; "Ert /3ao-t- \evovTos ApKabiov Koi vwdrcov ovTav 'Ovapiov to e^Sopov Koi Qeobocriov TO bivrepov, ol eu TJj Bperavvia (TTparevopfvoi (rraaiaaavrfs dvdyovai MdpKov en\ Tov ^acriXfiov Bpovov koi wj KparoiipTi rav avToBi irpaypdrcov fTTf idovTo. Then, in the same chapter, he goes on to describe Constantino's crossing into Gaul (which he had already recorded in v. 27) and a good deal of what he did there. Then (vi. 3) he goes back to record the migration of 406, and then comes to the election of Marcus a second time. The Vandals, Suevians, and Alans harry Gaul, koI woXvv epyacrdpevoi (fjovop im(J3o^oi Koi rots iv Bperavviais arpaTOTretois iyevovTO, crvvqvdyKaaav 8e, teei tov pi) KaTU cr(f)as npoeXdelv^ els rrjv tcov Tvpdvvaav opprjaai xeiporov'iav, MdpKov Xe'yca Koi Tpariavov Kul eVi rovrois KavrrTav- Tivov. So Sozomen (ix. 11), who takes a more general view; vtto fie TovTov TOV xpovov [the taking of Rome] noWav dviaTapevav Tvpdvvcav ev TTJ Trpos bvaiv apxfj, oi pev irpos dXkrjKoav TrinTOpres, ol 8e 7;apa86^a>s avWap^avopevoi, ov ttjv Tvxovaav enepapTvpovp 'Ovapia 6eop, iraXiv K<0P(TTaPTiP0P \eip0T0P0vdyrj, KavcrTavrivos Tore els to TOV avTOKpdTopos dva^i^d^fTM ovopa. So Zosimos (vi. 2) ; ayova-i Tpariavov fls piaov Kai oKovpyiba kcu uTfcfiavov ewiOevTeg i8opv(p6povv «s /3ao-iXea" Ti.] A Tyrant of the West. 47 Constantine only from his enemies ; their portrait is of course unfavourable ; yet he must have differed in some way from his two momentary predecessors ; he must at least have had some strength of character to do all that he did, and to bear up for several years against enemies of all kinds and from all quarters. The tale of his first acts is but darkly told, or rather the facts are fairly clear, but it is less easy to judge of causes and motives. Almost dva-apearrjaavres 8f Koi Tovra Tea-crapcnv vcrrepov firj(ri TrapaKvaavTes dvat- povari, KavaTavTivm irapabopres rfjv ^aaiXeiav, Sozomen has been quoted already in note, p. 45. Orosius (vii. 40) leaves out Marcus, but mentions Gratian ; " His [Halanis, Wandalis, et Suevis] per Gallias bacchantibus apud Britannias Gratianus municeps ejusdem insulse tyrannus creatur et occiditur. Hujus loco Constantinus ex infima militia, propter solam spem nominis sine merito virtutis eligitur." This looks as if Gratian was not a soldier, but a native or inhabitant of Britain of whatever class. It is possible that the four months of Gratian came wholly within the year 407, in which case the harrying of Gaul would have begun before his election. Prosper does not mention either Marcus or Gratian; but he gives the right date for the elevation of Constantine ; " Honorio vii et Theodosio ii Coss. Constantinus ex infima militia, ob solam speciem nominis in Britannia tyrannus exoritur, et in Gallias transit." Idatius does not trouble himself about British or Gaulish matters, though he has much to say when the invaders reach Spain. Marcellinus knows nothing about them till 411, when he says " Constantinus apud Gallias invasit imperium," and goes on with his later story. To these we may in a manner add Zosimos in his first account of Constantine, v. 27 (see below, note, p. 49). One is tempted to think that he knew nothing about Marcus and Gratian till he wrote the latter account. It would almost seem as if nothing was known in Italy of the movements in Britain till Constantine had actually landed in Gaul. 48 Western Europe in the Fifth Century, [it. the first act of the British tyrant was to forsake his island and to carry the legions across to Gaul. Of his motives for this step we are told nothing. We may read the tale in several ways. Some of the expressions used in describing the elevation of Marcus almost read like a formal secession from Eome and the establishment of a separate empire in Britain. But, if such notions were really held the year before, they certainly had no place in the policy of Con- stantine. It might rather seem that his object was to preserve the unity of the Empire, at any rate the unity of its provinces beyond the Alps. In this view it niio'ht be a wise course not to wait to be attacked in the island, but to cross to the mainland and to deal a blow at the enemy on what he was ' fast making his own ground. Britain might thus be saved by a campaign in Gaul. But if this was the motive, the thought of saving Britain must soon have passed away from the minds of Constantine and his soldiers. Whether they cared for such, an object or not, the course of things on the mainland soon made it hopeless for them to think of keeping up any relations with the great island. The crossing of Constantine into Gaul thus became the end of the Eoman power in Britain *. He landed at that Bononia of northern Gaul, once Gessoriacum, which, though not the starting-point * The words of Olympiodoros (p. 451) are noteworthy; ras BpeTTavias edaar, TTfpaiovrai afia twu airov eTit Bovaviai', The phrase of Zosimos (vi. 2) ttjv Bperravlav KaraXinatv is hardly SO expressive ; but of course neither was meant to convey the whole of its full meaning. IT.] A Tyrant of the West. 49 of Caesar, has been in all ages one of the chief points of passage between the island and the mainland *. He brought with him, it would seem, the whole of the Eoman force with which Britain had been held or defended. It was under the command of two generals, Justinian or Justin, and Neobigast, and it would seem that it was put under their command before the army left Britain f. Of their names, the one is clearly Eoman, the other clearly Frankisb, and we shall presently see that Constantine was on good terms with others of the Prankish allies or subjects of Eome. His stay at Boulogne was not long ; but it is hard to trace his course in the early stages of his advance. He presently gathered under his obedience whatever troops were to be found in Gaul, whether Frankish allies, legionaries who had been left behind by Stilicho, or patriots who had answered the summons of Honorius the year before J. The * 01ympiod6ros (p. 451) describes Boulogne as Boj/cai/t'a, ttoXi? QVTdi KoXovfievT], TrapadaXaaa-ia Koi TrpdiTi] iv tols rcav TiiKKiav opiois Keifievrj, The description of Zosimos (vi. 2) is more remarkable ; Trparr} Se avrt] irpos Ttj 6a\d(T(rrj Kflrai., Tepfiav'ias ovaa jroXty t^s kutco. One IS re- minded of bis description of Paris (iii. 9) as Tepp.av'i.as noklxvrj. t Olympiodoros at least says, 'lovarivov kcli Neo^iyaa-Trjv a-TpaTTjyovs 1rpo^aK6p.evos Koi tch Bperavvlas eaa-as, Trepaiovrai. Zosimos WOrds are, 6 8e 'lovcTTiviauov Koi Ne^toydoTJ/v apx^eiv rav iv KeArots arpaTicoTaiv eTrepaiadr). It would almost seem as if they crossed before their master. Nebiogast is clearly a Frankish name, like Arbogast and like the four sages of the Saliau Law, Wisogast, Bodogast [or Arogast], Saligast, and Windogast. A. Holder, Lex Salica (St. Gallen MS. 731 and Ed. Heroldina), Leipzig, 1880. X Olympiodoros, u. s. ; epda [ev Bovavi^ StarptS/^as aal o\ov rov TaKKov Ka\ 'Akvtovov (TTpaTia)Tr]v l8i(moirja-dp,€vos. The opposition between Gaul and Aquitaine is curious. So Zosimos ; navTa olKeiaadfievos ra E 50 Western Europe in the Fifth Century, [ii. authority of Honorius was represented in Gaul by the Prsefect Limenius and the general Chariobaudes*. The name of this last speaks for his barbarian birth ; we seem to see in him an English Herebald. Of their action at the moment of invasion we hear nothing. These names appear only at a later stage, when we are told that they had fled before the tyrant. But at what stage of his course they fled, and whether they oflered any armed resistance to the invader before they fled, on these points we are left whoUy in the dark. On the whole the chances are against any fighting between the followers of Constantine and any who remained loyal to Honorius. Our authorities are most confused ; but on the whole the story reads as if so much of Gaul as still obeyed any Boman prince at all submitted to Constantine without a blow. The mission of the new prince, the object which o-TpaTiVfjLara fifxP'- ^^^ ^AXTrecov ovra. He makes them stay at Boulogne only a short time, biarpi^as rjjiepas Tivds. Zosimos writes through the whole story as if he got his facts from Olympiodoros, but thought he could improve his language. * These two officers are mentioned by Zosimos in his casual way at a later stage, namely at the time when they are murdered ; v. 32. The soldiers (see p. 5) rponov nva irapa^opoi yeyovoTfs At/ue- viou Tf Tov iv To'is VTTep ras "AXireis 'ddveaiv ovra ttjs av\r)s virap^ov anoacfxiTTOvtn, koi afia rovra Xapio^avbrjp tov OTpaTtjyov t5>v eKeltre ray- fiarav' ervxov yap 8ia(jivy6vTfs top rvpavvop Koi v7ravTrjv ^ap^apav KaTaa(f)d^avTfs pepos, rois 8e (jievyovcrtv ova ene^fXdovTfS {fj yap hv airavras iravaiXedpia bie({>6(ipav) fi/eSaKav avrois dvaKTtja-apevois TTjv ^TTap Koi ^ap^dpav irXrjdos avvayayuvaiv d^iopd\ovs yivivdai, E 2 52 Western Europe in the Fifth Century, [ii. informant goes on to say that Constantino placed guards on the borders and secured the whole course of the Ehine. It is a zealous pagan who speaks ; his mind goes back to the days of the hero of his own creed, and he tells us, with some injustice both to the strong reign of Yalentinian and to the more recent exploits of Stilicho, that Constantine guarded the Ehine as it had never been guarded since the days of Julian *. On the other hand, the new Emperor or tyrant stands charged with doing the republic great damage by allowing himself to be many times cheated by the barbarians by treaties, vague, it would seem, in their terms, and not strictly kept f. This, we may be sure, refers to the barbarians who were already in Gaul, the Vandals, Suevians, and Alans. Some under- standing between them and Constantine, there must have been. For two years they and he carry on their operations in Gaul, each, it would seem, without any interruption from the other. And when the scene of action is moved from Gaul to Spain, each party carries on its operations there also with as little of mutual let or hindrance. It was most likely only by winking at their presence and at their doings that Constantine obtained possession, so far as Eoman troops and Eoraan administration were Zosiinos, VI. 3 ', 8ia rmra toIpvv tovtois tois tottoh (jiiiXaKas eyKare- CTTTjae KavcrravTwoiy as av fxr/ rrjV eli TaKariav dveifiivTjv €)(Oiev irapahov. (yKartcTTrjcre 8e koi tm 'Frjva naaav d(T(paKetav, eK tS>v '\ovki,avov ^a(TiKea>p naiBcov, but does not mention Julian by name. The monastic profession of Constans comes from Orosius ; " Con- stantem filium suum, proh dolor, ex monacbo Csesarem factum." The collection of Flavian names in the family of this private soldier is certainly remarkable. Most likely they were popular in Britain. Gregory of Tours (ii. 9) calls Constans Constantius; not so his authorities. t So far as it is safe to make any inferences from such a confused tale as that of Zosimos, I seem to see two messages. One (v. 27) comes when Honorius is at Eome and Stilicho at Eavenna — idrfKoiiTO ms KtavaraVTlvos eiri6ep.evos e'lr] tt} rvpavvibi Koi eK tt)s BpfTavviKTJs vfjo-ov Trepaim6e\s iv to'ls vnep ras "AXneis e6v€(ri TrapayevoiTo, to, ^aai\fa)s 58 Western Europe in the Fifth Century, [ii. Honorius, Emperor and Consul, was at an unusual place, namely in Eome itself. Stilicho was at Ravenna. At that moment tlie friend of Alaric, he was, we are told, making ready for an expedition beyond Hadria, to be carried on in fellowship with the Gothic king, an expedition the object of which was to transfer the cities of lUyricum from the obedience of Arcadius to that of Honorius. His schemes were thwarted by two rumours, by a false report of the death of Alaric and by the true report of the advance of Constantino. This last news was announced to Stilicho by letters from Honorius him- self*. It was not often, one would think, that the Augustus had news to tell to the Consular, news at least of a graver kind than the revolutions of the poultry-yard. Stilicho now gave up the thought of an Illyrian campaign, and hastened to consult — so we are told — his sovereign as to what was to be done. When Gaul had been attacked by a vast alliance of barbarians, nothing had been done beyond the issuing of proclamations in the province itself. iv Tails TToKeai npaTTcov. This seems to be still in 407. Then, under the next consuls (see v. 28), come deliberations of Honorius and Stilicho (v. 31) at Bologna, in which one consideration is, ^8r, Kav- (TTavTLVov Tov Tvpdvuov TTjv TdkaTiav naarav StaSpaynoi/ros Koi iv rfi ApeXdrat SiarpidovTos. These, in any writer who at all regarded order, would imply two stages. * Zosimos, V. 27. Stilicho is making ready atEavenna; but 8vo KoiKvfiaTa avve^T) irapenneaelv, ^Tjfij] re i>s 'Wdpixos TeQveats eijj hiabpapiovaa Koi €K TTJs 'Pa>firis 'Ovaplov ypafipara tuv jSatrtXecos duaSodevra, These con- tained the first piece of news in the last note, p. 57. The report about Alaric was doubtful, and was soon known to be false ; ra 8e irepX ttjs dvapp^a fas Kavaraprivov "Keyofieva irapa irduriv eKpoTfi. Kanke, iv. 1. 232. II.] A Tyrant of the West. 59 But tbe rise of a rival Emperor was a more serious matter. The deliberations of Stilicho and Honorius seem to have been carried on into the next year(408), the year of the consulship of Bassus and Philip, the year which saw the death of Arcadius at Constantinople, the year in which Honorius — if we can give Honorius the praise or blame of any deed good or bad — used, in the phrase of the next generation, his left hand to cut oif his right, by the slaughter of Stilicho himself*. But at the beginning of the year Stilicho is still in favour and Honorius contracts the second of his strange marriages with the daughters of the great Vandal f. Disputes with Alaric, now known to be alive, follow; he is ready for warfare in the East, for which Stilicho, with Constantino in Gaul, no longer designs him. With the tyrant at Aries, his counsel now was to send no less a champion than Alaric himself, in the character of a Koman general, to win back the lost provinces for their lawful prince. He himself, Stilicho, will undertake the affairs of the East, while the West-Gothic king represents the true majesty of Eome beyond the Alps \. A day was to come before long when a West- Gothic king was to go on such an errand, but the * The saying of the barbarian to the last Valentinian after the murder of Aetius (Procop. Bell. Vand. i. 4) ; oti avrov rrfv Be^iav rfi erepa X"P' dnoTefiav fir], t Zosimos, V. 28. X Zosimos, V. 31. Stilicho determines yvafirjv ttjv dplarrjv flvai Koi Tjj iroXiTela \v(TiT€\ova-av, 'AXdpixov fiev eniarpaTeva-at ra rvpavvca, tS)V re (Tvv avra ^ap^apap ayovra fxepos Koi TeXij 'Pw/iaifKci Koi T]y(p.ouas, ot KoivfjcrovcTiv avTia tov Trokep,ov, tt]v eaav Be avTos KaroKri^eiTOai pacrtXecus KeXfvoiTos Koi ypafifiara nepl tov irpaKriov StSdcros. 60 Western Europe in the Fifth Century, [ii. work for which Alaric was destined was of another kind. Yet another Goth was sent this very year to do the work of Honorius against Constantine. It is hard, though we are chiefly following one authority, to put the facts together out of a most confused narrative. We hear of the growing influence of Olympius at the court of Honorius, an influence used to bring about the downfall of Stilicho. We hear of Honorius at Ticinum, while Stilicho is at Bologna. We get a picture of the Emperor haranguing the troops who are to march, under whose command we are not told, against the tyrant at Arelate. A mutiny breaks forth, a mutiny which, it is implied, is in some way connected with the intrigues of Olympius against Stilicho. And it is most significant, though we cannot fully understand the significance, that the outbreak of the soldiers led to the slaughter of the two officers, Limenius and Chariobaudes, who had fled before Constantine to Honorius, and who must have joined him quite lately "^^ They were already in the * See above, p. 50, note. Honorius is under the influence of Olym- pius. Then (v. 32), says Zosimos, fKTaKKrjBevTcov els to. ^aa-iXtia tS)V arpaTicoTQiv ecjiaiveTO re avrols 6 ^acriXevs Koi els top Kara KcovcTTavnvov rov Tvpavvov 7rape6ap(Tvv€ TtoKepov. Then come the mysterious words, Trept be '2re\L)^a)vos ovdevos KivrjBevros ecpaivero vevcov toIs arpariaiTats 'OXvfnnos Koi axiivep a.vapi.fxvT)crKa>v S)v ervxev avrols ev Trapa^varci 8i.aKe)(6eLS. Then comes the slaughter of Limenius and Chariobaudes, and of [Vin- centius * magister equitum' and Salvius, 6 8e dopea-TiKav rdyparos Trpoea-ras, and] some others. They would thus seem to have been in the interest of Stilicho. As Honorius had been only four days at Ticinum, they could have only just joined him there ; but they need not have come stiaight from Trier, Aries, or whatever part of Gaul they started from. II.] A Tyrant of the West. 6i interest of Stilicho, and on their fate presently follows the fate of Stilicho himself. Yet we read elsewhere that it was at Stilicho's bidding that Sarus, the valiant Goth whose name we so often meet in the history of these times, was sent with a force into Gaul to bring back the land into the obedience of Honorius *. The campaign of Sarus is undoubted ; but we have no means of fixing the relations between his campaign and the force that he held and the contemplated march of the troops that broke out into mutiny at Ticinum. Anyhow the newly-built-up throne of Constantine was threatened. Are we to suppose that, after embarking on so hazardous an enterprise, he shrank from personal danger, or that he was conscious of a lack of military skill 1 Some accounts represent him, at a later time at least, as more active at the table than in the camp f. Certain it is that it was not Constantine in person who met the army of Sarus in battle. While the barbarians were marching and harrying throughout the land without let or hindrance, two Roman armies met, both doubtless largely made up of barbarian soldiers. The cause of Constantine was defended by his lieutenant Justinian ; but the fortune of war was on the side of legitimacy. Sarus gained a victory which carried with it the * Sarus has been already mentioned by Zosimos, v. 30 ; but our account of this campaign comes wholly from vi. 2 ; Kara tovtovs Toiis \p6vovs ^apov tov arpaTTjyov (KnifiTVH pera crTparevparos Kara Kcov- (TTavrivov '2t e\i \ciiv. t " Constantinus guise et ventri deditus," says Eenatus Profuturus Frigeridus in Greg. Tur. ii. 9. 62 Western Europe in the Fifth Century, [ii. death of Justinian and of the greater part of his army, and the winning of great spoil by the army of Honorius *. Of the details of the fight, of the place, of the exact time, we hear nothing ; but it is clear that it was fought somewhere in the lower Khone- land, and it would seem that the routed army could have been only a small part of the forces of Con- stantino. Where he himself was at the moment we are not told ; we know only that, after the battle, he deemed it wise to secure himself in one of the strong cities of the land, but in one which lies a good way to the north of his newly chosen capital. Many of those cities are greater in old renown, many are richer in abiding remains of Imperial power, but none holds a stronger site ; none looks naore proudly from its height on the great river at its feet, than the city in which Constantino sought shelter against the attack of Sarus. The walls of the Gaulish Valentia do not still stand in witness of those days like the walls of Arelate and the true Vienna ; but in those days the city of the Sagellauni was one of the great fortresses of the land. Its name might suggest the thought of the great prince who had bestowed that name on the recovered regions of the island that Constantino had forsaken. But while the Valentia of Britain did indeed preserve the name of Valen- tinian, the Valentia of Gaul was of older date ; it bore the name of Kome herself, and the Valentia by the Ehone might pass as not only the colony but the Zosimos, VI. 2 J 6 6e [^SaposJ 'louortwai/w to (XTparrjya fiera T^f Bvvafieas tijs s (f>iKop top apbpa, bovs 8e Koi Xa/Scbf opKovs dpaipel irapaxprjpa, prjbeva tS>p opKcop itoirjcrdpevos \6yov. \ lb. KoiPfTTaPTiPov 8e arrpaTrjyov KaTaarrjaavros 'EScojStyp^oi' ^pdyKOV uPTa TO yeVof, TepopTiop 8e dno rrjs Bperrapias o/a/iw/ieyoi', detcras 6 'Sdpos 64 Western Europe in the Fifth Century, [ii. His object was now to get back into Italy ; the generals of Constantine overtook him with a great force and brought him to great straits. But the words of our story would seem to imply that this was rather by harassing his march than by an actual battle. He escaped into Italy with great difficulty, and that only by help which we should hardly have looked for. Alongside of the new scourges of Gaul, barbarian invasion and civil war, a far older scourge had either lived on or had shown itself again. The Bagaudse, the Jacquerie of more than a hundred years earlier, were still in force, at any rate on the Gaulish slopes of the Alps. They met Sarus, with what objects we are not told, but we are given to understand that his passage into Italy was made secure by a timely gift to the Bagaudse of the spoil which he had won in his victory over Justinian *. Nothing can be more unsatisfactory than this kind of T^i/ tS>v crTpaTTjymv tovtcou nepi to. TTokefiia Tretpav ofiov Koi avbp'iav ave)(a)pr](r€ riji 'BakevTias, enra TrciXiopKr](ras airfjv fjfiepas. The British Gerent is of qourse merely a form of Gerontios ; but it has become a specially British name, and it is worthy of notice that we see it so early. * Zosimos, vi. 2 ; KaradpaiiovToov 8e avTov rS>p KcovoTavrivov crrpaTifyoiv fiera fjLeylaTrjs BvvacrTeias, pins ■nap avTutv ruxfl T^y eVi rfjv 'iraXlav Trapobov. KarabpafiovTcov would seem to imply skirmishes rather than a regular battle ; Swaa-reia is a singular word for a military force, and there is something strange in this quite taking for granted of the Bagaudse. He does not mention them elsewhere, and his bringing them in in this way might almost suggest that there is something in the view of Dubos- (i. 205, of. Gibbon, v. 223, ed. Milman) that the Bagaudae were, now at least, something of a local militia rather than mere freebooters. See Ducauge in voc. [See my Hist. Essays, Series iv. p. 118.] II.] A Tyrant of the West. 65 story. We put up with the mere annalist, who records victory and defeat without attempting to explain their causes ; but here we are told just enough to awaken our curiosity without satisfying it. But in any case the enterprise of Sarus alto- gether broke down ; he had slain in war and by treason two generals of Constantine ; but their death seems only to have led to the advancement of more competent successors. Whatever might become of Britain and Italy, the tyrant from the island was now the only representative of Eoman dominion in Gaul. His power was at all events firmly established in his own south-eastern corner, which Vandals, Suevians, and Alans, on their march from the Belgian lands to the Pyrenees, would be likely to leave untouched *. And as if Gaul were a separate realm and Italy a hostile land, he strengthened himself against a second invasion from beyond the Alps, by placing garrisons in their three chief passes, Cottian, Pennine, and Maritime f. Olympiodoros, U. S. ; Kpare^ iravTcav tS)V fiepStp rrjs TaXaTias fi^XP'- ''^'' "AXneav rmv fisTu^v 'iraXtas re Koi TaXaTias. Zosimos (vi. 2), in a passage part of which has already been quoted, seems to copy the definition ; irdvra re olKetwardfievos ra a-Tpareviiara fiexpi t5>v "AXneav ovra Totv opi^ovaSiv TdXarLav koi 'IraXiav, dvXaKas dpKovcras eyKarao'Tjiijai Tais "AXireaiv. ^a-av 8e aSrai rpeis, al ras en\ ttjv 'iraXiav diro KeXrau KaKfWev eneKdva ras obovs aTroKXeiovarai, Korrtai Hoivivai Mapiripai. Sozomen too (ix. 11) brings in the Cottian Alps, though he makes no mention of the expedition of Sarus. His summary of events F 66 Western Europe in the Fifth Century, [it. Constantine was now undisputed master of Gaul, at least of the remnant of Gaul that clave to Eome. Britain he had left behind him. If he aspired to the dominion of Italy, he prudently put oif any attempts on that side, till he had made himself master of all the provinces beyond the Alps. He was bound, for his own ends, to extend his dominion from Gaul in the geographical sense, to Gaul in the widest official meaning of the word, and to complete his possession of the Gaulish prefecture by the acquisition of Spain. The great peninsula of the west was one of the most flourishing parts of the Homan dominion, and the one which had suffered least from barbarian invasion. Since the Teutonic harryings in the days of Gallienus Spain had been untouched by strangers, free from any oppression save what it may have suffered at their hands who represented the power of Eome witbiii its borders *. The legions that were regularly runs ; Ufpaia^eis Se Kffli'oraiTeti'os aTTo BpeTTavias eVt Bovmviav noKiv ttjs TaXarias nepl BaXaacrav Kfifievrju, irpooTjydyeTO tovs irapa TaKarais Koi AKOviravols, arpaTiooTas' Ka\ tovs Tjj8f vutjkoovs TrepieiroiTjcrev iavTci, p^XP^ t5)v pera^v 'iraXias koI TaXaTias opav, as KoTTias AXirfis 'Ptopaioi KaXoviTi. KoDvaraPTa 8e tov Trpea^vTepov rav avrov vteav, ov varepov fiaaiXems crx^pa evedvae, Kaiaapa t6t€ avayopevcras, nfnopt^fv tls S/rawai/. His whole story seems to come from the same source as that of Zosimos, though there are odd differences. The use of 'Papaloi, as in Procopius and long after in Constantine Porphyrogenitos, shows the difficulties sometimes felt by those who were Romans by political allegiance but not Latins iu speech. * This seems implied in the emphatic though somewhat involved words of Orosius ; " Irruptse sunt Hispaniae ; csedes vastationes- que passse sunt : nihil quidem novum. Hoc enim nunc per biennium, illud quo hostilis gladius ssevit sustinere a barbaris quod per cc quondam annos passse fiierant a Romanis, quod etiam sub i II.] A Tyrant of the IVest. 67 quartered in Spain, and which were doubtless largely made up of natives of Spain, claimed the defence of the land as their special work, and resented any intrusion of strangers as a breach of their local privileges *. But the land had commonly been passive in revolutions, and had readily accepted such rulers as bore sway on the other side of the Pyrenees f. But at this particular moment, an element had to be reckoned with in Spain which would hardly have passed for a political influence in any other province. Spain had given the world a dynasty. Theodosius, like Trajan before him, had come forth to rule the Empire from the most western of its provinces, and to rule it, like his great countryman, so as to leave a memorable name behind him. The sons of Theodosius, princes of Spanish descent, still ruled, or at least reigned, at Constantinople and at Eavenna. The kinsfolk of the Imperial house, though not marked out from other men by titles or offices known to the Empire at large, were men of wealth imperatore Gallieno per annos propemodum xii Germanis evex-ten- tibus exceperunt." * So witnesses Zosimos (vi. 5) a little later ; Tav eV 'l^vpia crpaToneSav ifxirurrevBrfvai, Kara to avvrjBes rrju (jivXaic^v ahrjaavTav, Kot fifj ^evois eTnTpanrjpai rfju ttjs xiapas dacfxiXfiav. t Such is the remark of Gibbon (v. 223, ed. Milman); "His throne was soon established by the conquest, or rather submission of Spain ; which yielded to the influence of regular and habitual subordination, and received the laws and magistrates." He adds ; " The only opposition which was made to the authority of Constan- tine proceeded not so much from the powers of government, or the spirit of the people, as from the private zeal and interest of the family of Theodosius." F 2 68 Western Europe in the Fifth Century, [ii. and influence in their own land, attached to the throne of their Imperial kinsmen and acknowledged by those kinsmen as men bound to them by the ties of blood. To the mass of the people of Spain it might seem most natural that Spain and Gaul should go together; to the members of the Theodosian house and to all who shared their feelings the first object of all was that the land of Theodosius should abide in the allegiance of the sons of Theodosius. Constantino had therefore to look, not so much for any general resistance in arms on the part of the province or its regular defenders, as for whatever amount of opposition in any shape could be stirred up by a few powerful men. But that opposition was likely to be of a very dangerous kind. Con- stantino is described as fearing a joint attack from two branches of the Theodosian family, from the Emperor in Italy by the way of the Alps and from his kinsmen in Spain by way of the Pyrenees. Lest his dominion should fall when thus assaulted on both sides, Constantino determined to forestall all attacks from the Spanish side, and at once to begin the occupa- tion of the peninsula *. The date is not hard to fix. We are still in the year 408, the year of the campaign of Sarus and of the death of Stilicho. That year saw also the death of Arcadius and the beginning of the Zosimos, VI. 4 ; twi/ airodi fev 'l/3?7p/a] iravTcap i6vS>v iyKparrjs yevea-dai ^ovXofxfvos, coare koI rfjv dp^fju av^rjaai Koi afta Tf)V tS>v 'Ovapiov (Tvyyfvav avTodi bvvaa-Tfiav eKKoyj/^ai. beos yap avTov elarjei fifj jrore bvvapiv avvayayovres rap avrodi err paricorav avroi peu aira Bia^dvres ttjv Ilvpr]vr)v ineXdoKv, dno 8e ttjs 'iraXlas 6 ^aaiKevs 'Ovapios eTrmip'^^tas avra TO arpaToneBa t^s Tvpavvibos, KVKXa TravraxoGev TrepiKa^cnv, napaXvaeifv. 69 II.] A Tyrant of the West. long reign of the younger Theodosius *. It saw also the operations of the forces of Constantine in Spain. Those operations, it has been truly remarked, imply some kind of treaty or understanding with the barbarians who, it must never be forgotten, were still ranging through Gaul at pleasure f. The relations between him and them, the way in which each side seems to act with no seeming hindrance on the part of the other, form one of the great puzzles of our story. Some of the vain agreements with the invaders of Gaul, so darkly hinted at by a con- temporary J, must surely have taken place at this stage. It would almost seem that for a while (408) the pen- insula submitted without any opposition to the ruler of Gaul and to the officers whom he sent to represent himf. But if so, this submission was only for a * Theodosius was now in tlie sixth year of his reign as his father's colleague. Born in 401, he became Augustus in 402; he took his first consulship in 403, and kept his g'Mmg'wewnaZm in 407. See Sozomen, viii. 4, and the Fasti. He was as much Emperor before as he was now, only the style now was "Honorius et Theodosius Augg." instead of " Arcadius, Honorius, et Theodosius." t Wietersheim, ii. 161 ; " Auch muss zu Beginn dieses Jahres eine Art von friedlichen Yertragnisse zwischen ihm [Constantin] und die Eingedrungenen Barbaren bestanden haben, so dass er ohne Gefahr eines jeden Angriffs durch dieselben an Ausdehnung seiner Herrschaf auf Spanien denken konnte." X See above, p. 52, and note. § Orosius, u. s. ; "Misit in HispaniasywcZices, quos cum provinciae obedienter accepissent, duo fratres . . . tueri sese patriamque moliti sunt." After this comes the mission of Constans. From the other writers one would think that Constans was sent as the first step on the part of his father towards action in Spain. It seems 70 Western Europe in the Fifth Century, [ii. moment. Among the kinsmen of Honorius, four brothers, bearing the names of Didymus, Yerenianus, Theodosius or Theodosiolus, and Lagodius — we may- mark a certain tendency to Greek names in the Theodosian house — held a high position for birth and wealth in different parts of Spain*. Two of them, Didymus and Yerenianus, now raised the standard of legitimacy, the standard of their own house. The other two seem to have taken no part in the enter- prise f . Didymus and his brother, we are pointedly more likely that they should have left out an earlier mission than that Orosius should have imagined two missions when there was only one. * Theodosius, Theodosiolus, Didymus, Lagodius, Arcadius. On the other hand there is Honorius. Galla Placidia is called after her mother, the daughter of Valentinian. Sozomen (ix. 1 2) speaks of Theodosiolus and Lagodius eV irepais inapxiais diarpl^ovres, while the other brothers carried on the war. t The accounts here are somewhat hard to bring into agreement in detail. The clearest account is that of Sozomen (ix. 11, 12), who does not put the events in exactly the same order as Zosimos (vi. 4). Orosius moralities more, but goes less into detail ; from Olympiodoros we unluckily hear nothing again tUl a later stage. After the passage quoted above, Sozomen goes on ; 6 Se [Kwi/o-tos] TO ZQvos KaroKa^av, ap^^ovrai Ihiovs KaT€(TTT]v 8e ev rfj avXfj rd^ecov ap^ovrds re ttoKitikovs dfia Kai arpaTKOTiKovs KaracrTqaas, ayei 8id tovtup en' eKeivovs oi yefei tw II.] A Tyrant of the West. 71 told, did not themselves assume the tyranny in oppo- sition to the tyrant ; so to do, it seems to be implied, /3ao-tXfi GeoSaxrtM TrpoarjKOVTfs to. rrji 'l^rjpias (TweTapdrTovro irpdyfiara, irpoTepov fieu vrpos avrov '^mvaravra 8ia tS>v iv ttj Avairavia arparonebaiv apdpevoi noXefiov, eirel 8e Tr\eoveKTei(Tdai crvvrja-BovTO, TrXrjdos olnermu Koi yfcopywv iinaTpaTexKravTes Koi tvnpa ^paxel KaTaa-rrjaavTes avrov tls peyicrrov Kivhwov. oKKa KavravQa rrfs iknlhos diajxapTovTfs KavaravTi (tvv tois p yvvai^lv rjaav iv (f>v\aKiJ. Orosius goes on from the place already quoted ; " Fratrea juvenes nobiles et locupletes Didymus et Verenianus . . . plurimo tempore servulos tantum suos ex propriis preesidiis colligentes ac vernacula alentes sumptibus, nee dissimulato proposito absque cujusque inquietudine ad Pyrensei claustra tendebant. Adversus hos Constantinus Constantem filium suum, proh dolor, ex monacho Csesarem factum, cum barbaris quibusdam qui quondam in fcedus recepti atque in militiam allecti Honoriani vocabantur, in His- panias misit. Hie apud Hispanias prima mali labes. Nam inter- fectis illis fratribus qui tutari private prsesidio Pyrensei alpes moliebantur," &c. Here is first the difference already pointed out that Orosius makes the brothers rise against officers already sent by Constantine, and makes Constans come against them, while in the other two versions the brothers seem at least not to rise till after the coming of Constans, though, in the way that both Sozomen and Zosimos tell the story, the chronological order is not strictly observed, and their words might be understood of an earlier rising. This so far confirms the version of Orosius, though the "judices" whom he makes Constantine send first of all before Constans must surely be the same as the ap^ovres set up by Constans in Sozomen. The apxovres in Zosimos are in a marked way connected with the ev avXjj rd^eis, who seem to be the same as the Honoriani in Orosius. Then, whenever the movement of Didymus and Verenianus took place, our authorities seem hopelessly at variance as to its geography. Orosius makes them collect an army of slaves and peasants and occupy the Pyrensean frontier before Constans comes. Both Zosimos and Sozomen make Lusitania the first seat of war ; 72 Western Europe in the Fifth Century, [ii. would have been the most natural course for men in their position ; they strove for their country and for only Zosimos makea the brothers begin their enterprise with such regular troops as were quartered in Lusitania ; when these are defeated, they gather an army of peasants and slaves, who for a while bring Constans to great straits, but who are afterwards defeated. In Sozomen the brothers, after making up some differ- ences between themselves which are unexplained and of which we hear nothing elsewhere, appear in Lusitania at the head of their irregular army, and seem for a while to be successful in a guerrilla warfare. Then new forces come to the help of Constans, and they are defeated. One may suspect that this last is another version of the coming of Constans himself, and the word a-vfifxaxia suggests the Honorians. In any case, we have, as in Orosius, two sendings of forces to Spain on the part of Constantine. The " rustic army of the Theodosian family," as Gibbon calls it, appears in all the versions. It is the one thing about which all the accounts agree, and we therefore accept it as the one thing about which we may be really certain. But our accounts do not agree as to its sphere of action ; or rather Orosius gives us a clear and probable version, while Sozomen and Zosimos are quite vague. Ill Orosius they occupy, or at any rate set out to occupy, the passes of the Pyrenees (" tendebant," " moliebantur," imply rather an attempt than an actual occupation). Yet we can hardly get rid of the mention of Lusitania, a land which is mentioned by both Z6simos and Sozomen, though they differ as to what happened there. I think, on the whole, that we may infer, First, That Constantine sent agents or troops into Spain twice, the second time under the command of his son Constans. This is distinctly asserted by Orosius and is partly confirmed by Sozomen. Secondly, That the movement of the brothers was a rising against the first occupation, and that Constans was sent to put down their rising. This again is distinctly asserted by Orosius, and several expressions in the other two writers (though they tell another stoi'y) help to confirm it. Thirdly, That the rustic army is the most authentic part of the story, as being asserted by all three writers, and that its main II.] A Tyrant of the West. 73 their lawful prince at once against the tyrant and against the barbarians who followed him *. But if action was directed, seemingly unsuccessfully, towards guarding the Pyrenees. This is distinctly asserted by Orosius, and it is consistent with the language of the other two. Fourthly, That something happened in Lusitania before the march of the rustic army towards the Pyrenees. This is the hardest part of the story. Orosius says nothing about Lusitania. In Sozomen the action there is successful action of the rustic army. In Zosimos it is unsuccessful action of regular troops before the rustic army is got together. He makes the rustics fight better than the regulars, which, though unexpected, is quite possible. In the other stories there are no regulars on the side of the brothers. Yet one cannot help thinking that the twofold action of the rustics in Sozomen, first in Lusitania, then somewhere else, is the same as the action in Zosimos, first of the regulars in Lusitania, then of the rustics somewhere else. Zosimos can hardly have imagined his regulars ; so that so far his account has the preference to Sozomen. Only Sozomen represents the first action, whether of rustics or regulars, as successful, Zosimos as unsuccessful. In this kind of warfare, there might be many alternations of success, but the gathering of a second army slightly favours the version of Z6simos. On the whole the probabilities of the case would seem to be met by such an account as I have given in the text. * So Orosius ; " Non assumpserunt adversus tyrannum quidem tyrannidem, sed imperatori justo adversus tyrannum et harharos tueri sese patriamque suam moliti su^it." He goes on ; " Quod ipso gestae rei ordine patuit. Nam tyrannidem nemo nisi celeriter maturatam secreto invadit, et publicse arma cujus summa est assumpto diademate ac purpura videri ante quam sciri." It would seem to be from the phrase " tyrannum et barbaros," where " barbaros " surely means the Honorians, that Isidore of Seville (who seems to be followed by Fauriel, i. 51) developed the picture which is given at the beginning of the "Historia Wandalorum " of Didymus and Verenianus defending the passes of the Pyrenees against the Alans, Suevians, and Vandals during 74 Western Europe in the Fifth Century. [ii. two of the four brothers were united as to ends, they were not at first of one mind as to means. It was the whole time of their sojourn in Gaul. His story is made out of Orosius in a curious way. He first copies the passage quoted above with some noteworthy changes. He leaves out the " multseque cum his alise gentes " of Orosius, doubtless because Vandals, Alans, and Suevians were the only nations whom he could see in the later history of Spain. Then he changes the order of tlie words of Orosius, " Francos proterunt, Rhenum transeunt, Gallias invadunt " into " transito Rheno, Gallias irruunt, Francos proterunt," because he was most used to Franks in Gaul, and hardly understood the process of getting into Gaul by fighting Franks on the other side of the Rhine. Then, whereas Orosius surely means simply that they reached the Pyrenees without hindrance either from man or nature, and then shrank for a while from attempting anything so strange as the mountain passes, Isidore has the passes ready guarded by the kinsmen of Honorius. It is possible that they may have commanded the native forces in Spain ; but there is no reason to think that they did, as this piece of Isidore is full of confusion. He goes on ; " cujus [Pyrenaei] obice per Didymum et Verenianum nobilissimos et potentissimos fratres ab Spania tribus annis repulsi per circumjacentes Gallise provincias vagabantur. Sed postquam iidem fratres, qui privato praesidio Pyrensei claustra tuebantur ob suspicionem tyrannidis insontes et nulla culpa noxii a Constantio Csesare interfecti sunt, memoratse gentes Spanianas provincias irrumpunt." But it was against Constans, not against Vandals and Alans, that the brothers gathered their " privatum prsesidium," and Isidore seems to have jumbled together Constantine, Constans, and Constantius. This version of Orosius has been followed by Fauriel, i. 51, where Didymus and " Valerian " appear at a very early stage of the story as guarding the Pyrensean passes at the head of the Pyrensean mountaineers, and with them driving back the bar- barians. All this seems to come out of Orosius' phrase of the "obex," as improved by Isidore. The "barbari" of Orosius' narrative whom Didymus and Verenian oppose are surely the Honorians of Constantine. The Vandals, Suevians, and Alans are II.] A Tyrant of the West. 75 only after some unexplained differences among them- selves that Didymus and Verenianus agreed on any combined action. The general course of events is clear ; but it is not easy to put together our various short notices into a connected story. It would seem that Lusitania was the part of Spain in which the brothers had most influence, and that in which they first took up arms. One account reads as if a regular legion quartered on that side of the country joined the cause which they supported. It was seemingly at this stage that the Caesar Constans was sent from Gaul by his father to put down the revolt and to bring its leaders before him in bonds *. He came at the head of the barbarian allies whom his father had found in Gaul. They bore the name of Hono- rians, but they were enlisted on behalf of Constantino against the prince whose name they bore. A motley gathering of troops of various nations, Scots, Moors, and Germans, they ranked among the household troops of the Empire, but they were likely to be indifferent as to which of two rival Augusti they drew their swords to support f. Constans took with not seen in Orosius' narrative between the words " His per Gallias bacchantibus " and the words " Perdita Pyrensei custodia claustris- que patefactis." * Sozomen, ix. 11 ; 6 be [Kwi/oras] to Wvos napoKa^wv^ apxovras IBiovs KaT(6r]Ke Koi Secrfiiovs avrm dxdrjvai irpocreTa^e AiSvixov Koi Bepeviavov, Tovs 'Ovapiov a-vyyevf'ii. Surely this sounds more like orders given by Constantine to his son on his setting out from Gaul. t The passages in the Notitia (§ 38) about these Honoriaci — Honoriani, as in the record, is doubtless the more correct form — are discussed by Gibbon, ch. xxx. note 99 (v. 224) ; Wietersheim, ii. 162; Hodgkin, i. 743, n. The name shows that they could not be 76 Western Europe in the Fifth Century, [ii. him the British general Gerontius, and he took with him also as a civil lieutenant a man chiefly memor- able as the forefather of one of his own descendants. Apollinaris, grandfather of the famous Sidonius of Auvergne, came of a senatorial house which ranked high among the nobility of his own province and of all Gaul. The highest office in the Western lands, the praetorian prsefecture of the Gauls, was almost hereditary in his house. But he was the first of his line, as his admiring grandson tells us, to embrace the new creed of the Empire and to have the cross signed upon his brow *. He did not scruple to accept regiments of very long standing : says Wietersheim. Gibbon must have been smiling when he suggested that the Scots were "influenced by any partial aflfection for a British prince." Wietersheim lays the seat of war on the west coast, and supposes that the march of Constans was made in connexion with a landing by sea, which is not mentioned in our authorities. [Mr. Hodgkin believes that the " Honorians " did not form " one division of the army," nor " ever necessarily acted together."] * Zosimos, VI. 4 ; a-Tparriyov fieu TepevTiov e^oiv, ' AnoWivdpiov 8e t^s avXiis vnapxov. For Terentius, who does not appear in any other part of the story, I venture to read Gerontius. Sidonius speaks of his grandfather, Ep. iii. 12 (iii. 1 Baret) and V. 9 (v. 20 Baret), in the former of which letters he gives his epitaph. Some of the lines run ; " Prsefectus jacet hie Apollinaris Post praetoria recta Galliarum. Moerentis patriae sinu receptus, Consultissimus utilissimusque Euris, militise, forique cultor, Exemploque aliis periculoso Liber sub dominantibus tyrannis. Hoc sed maxima dignitas probatur, Quod frontem cruce, membra fonte, purgans II.] A Tyrant of the West. 77 his office, seemingly as the successor of Limenius, at the hands of the actual ruler of Gaul, and to help that ruler's son in his attempt to add Spain to his father's dominion. The adhesion of such a man to the cause of Constantino is the best witness to the general acquiescence, to say the least, of the Gaulish lands in the transfer of Imperial power to his hands. The joint march of Constans, Gerontius, and Apol- linaris was met at some stage, seemingly on the Spanish side of the Pyrenees, by an irregular army of slaves and peasants, a force which Didymus and Yerenianus had seemingly kept for some while at their own cost *. Their object was to bar the passes of the Pyrenees against the invaders from Gaul, a work for which Spanish guerrilla troops would be excellently fitted in any age. For this it would seem they came too late. Their efforts were indeed not wholly without success ; they are vaguely said to have put Constans in great danger f. But in the end they were routed, and their leaders, Didymus and Yerenianus, were taken prisoners, with their Primus de numero patrum suorum Sacris sacrilegis renuntiavit," In the other letter he tells us how ApoUinaris and his friend Eusticus " in Constantino inconstantiam, in Jovino facilitatem, in Gerontio perfidiam, singula in singulis, omnia in Dardano crimina simul exsecrarentur." He goes on to mention the offices of his father and his friend of the next generation under Honorius and Valentinian the Third, when " unus Galliarum prsefuit parti, alter soliditati" — the " soliditas," one would think, only of so much as was left. * See the passage from Sozomen quoted in note, p. 66. t See the passage from Orosius quoted in note, p. 66. 78 Western Europe in the Fifth Century, [ii. wives. The other brothers, who were in some other part of Spain, took fright at the fate of their kins- men, and fled, Theodosiolus to Honorius in Italy and Lagodios to Constantinople *. He could hardly have got thither till the latter part of the year 408, when he found the young Theodosius already the only Emperor in the East. Constans now, as a Caesar ruling in Spain, estab- lished his court at Cassaraugusta, the modern Zaragoza, a choice not unconnected with the greater events which we shall presently mention. He had so utterly cast aside his monastic vows that he had taken to him a wife ; whether he had brought her with him to Spain or had found her there, we are not told. He was now summoned by his father into Gaul to discuss the afiairs of their common Empire. He obeyed ; he left his wife in his Csesarean palace at Zaragoza, and entrusted Gerontius with, the com- mand of the Honorian troops and with the defence of Spain. He then hastened to his father, taking with him the captive kinsmen of Honorius, Didy- mus and Verenianus. They were presently put to death by order of Constantino ; of the fate of the wives who shared their captivity we hear nothing f. Zosimos, vi. 4 ; onep dKrjKoores ol tovtodv abehf^oi Qeohoaios re Koi Aaya>8ios, 6 (lev fts rrjP 'iraKiav 8i€(f)vyev, 6 8e els r^v iaav Staa-ai&els dvtxo>pr]cr€. S6zomen, ix. 12 ; eV irepais Se fnapxiais biarpi^ovres Qfo- 8oaia>kos Koi AayatSios ol airSiv a.8e\(f)ol evyov(rt, rfjv irarpiba, (cat 8iatru>- (pvrai, QeodoaioKos pev els 'iraXlau npos 'Ova>piov tov ^aaiKea, AaywStoi 8e npos Qeoboatov fls dvaroXrfv. I cannot think, with Gibbon, that these two brothers had any share in the war. t Orosius at this point tells us nothing of the doings of Con- II,] A Tyrant of the West. 79 Constantine was thus, to all appearance, undis- puted ruler of Spain and of so much of Gaul as the Vandals, Suevians, and Alans were not at any par- ticular moment laying waste. In the lands on the Khone the retreat of Sarus had left him without a rival. But he was at this moment the only repre- sentative of Eoman power beyond the Alps. His position in the Western world was clearly better than that of the Augustus at Kavenna, threatened every moment by Alaric, and now left without the arm of Stilicho to guard him. That Honorius should outlive both Alaric and Constantine, that he should die an undisputed Emperor, master of so much of the stantine and Constans. He is carried away from the subject by a torrent, partly of declamation, partly of valuable historical matter, to which we shall have to look presently. From Sozomen we might almost have fancied that Didymus and Vereniauus were put to death in Spain. See the passage, ix. 11, quoted in note, p. 66. He now goes on ; 'O fitv Kavcrras TaiiTa bunrpa^afiepos, inavrjXde npos TOP irarepaj (f>povpav KaTacrTTj(ras dno ratv (TTpanaTaiv t^s e;ri Tijs Siravias napoBov. Z6simos (vi. 5) is clear ; Tavra Kara rfiv 'l^tjpiav 6 Kavarai 8ian pa^d/Jievos iTravrjXde irpos rbv irarepa iavTov KavarapTivov, fwayojMevos Beprjviavop Kal ^ibvpiov, KaraXnrav re avTodi tov arpaTtjyov Tepovrioi/, ay.a Tois diro roKarlas orpaTiaTais, ({)v\aKa t^s dno KeXrav eVt t^v ^I^ijpiav irapodov . . . Bepijviapos ptv ovv koi Ai8vfiios as KcovaravTlvov d)(d4vTiS dvTjpe'dr](rav napaxp^fia. These troops from Gaul are clearly the Honorians, of whom, as we shall presently see, Orosius has much to say at this point. We have also got the help of one of the fragments of the otherwise unknown writer — Henatus Profuturus Frigeridus, preserved to us by Gregory of Tours, ii. 9 ; " Accito Constantinus tyrannus de Hispaniis Constante filio iterumque tyranno, quo de summa rerum consultarent praesentes ; quo factum est ut Constans, iustrumento aulae et conjuge sua Caesaraugust^ dimissis, Gerontio inter Hispanias omnibus creditis ad patrem continuato itinere decurreret." 80 Western Europe in the Fifth Century. West as was still left to Borne, and that the power of Kome should be yet restored over no small part of the West from which it seemed to have passed away, is one of the strangest things in the strange times which we are studying. III. [CONSTANTINE EMPEROR AND MAXIMUS TYRANT.] We left Constantine undisputed master, undis- puted emperor, within so much of Gaul and Spain as obeyed any Emperor at all. Some parts of those lands were still harried at pleasure by detachments of the great host that had crossed the Ehine on the last day of the year 406. Some parts, it may be, were throwing off the dominion of Kome altogether. Britain, the land from which Constantine had set forth, was, not so much throwing off the dominion of Eome, as slipping away from it without effort on either side. The dominions of Constantine in the West were painfully smaller than the dominions of Valentinian and Theodosius. But within them he had no Eoman rival. The master of Italy, far less master in Italy than Constantine was in Gaul, had striven to shake his throne, and he had failed. Throughout the provinces beyond the Alps, the adventurer from Britain, like other adventurers from Britain before him, was " Dominus Noster ; " he was Augustus, he was "Pius," "Felix," and "Pater Patriae." As such his name was graven on inscriptions; his G 82 Western Europe in the Fifth Century, [iii. image and superscription was, in all the Western lands, the image and superscription of Caesar. What then was lacking to him ? Something which it is not easy to define. With all his success, he was still, in the eyes of men of his own time, as he abides in the pages of history, Constantine the Tyrant. In using that name in these ages, just as in using it in the days of the old Greek commonwealths, we must throw aside that modern abuse of it by which it is vaguely applied to any ruler whom it is meant to brand as an oppressor. This abuse is closely allied to the kindred abuse of other technical terms of Greek and Roman politics, which make it dan- gerous, even in writing Greek or Eoman history, to use the original words in their original meaning without some kind of qualification. At least from the days of Herodotus to the days of William of Malmesbury, the word " tyrant " had a definite mean- ing; and it is wonderful to see how little the meaning of the word in William of Malmesbury has changed from its meaning in Herodotus. The change in the use of the word is simply the change which is implied in a changed state of things. A tyrant is one who takes to himself power without any lawful claim to take it. The name has nothing to do with his use of power when he gets it. Undoubtedly he who gains power wrongfully is under many temptations to use it badly; but his using of it badly is not implied in the mere name of tyrant. The Greek tyrants, as a rule, were oppressors ; but even among them the rule was not universal ; there is no contradiction in terms in speaking of a just Ill,] Constantine and Maximus. 83 and merciful tyrant *. The Roman rulers to whom the name was transferred by a happy analogy, hold a higher place ; they are average Emperors, good or bad as may happen. The difference between the Greek and the Roman use comes from the different shapes which the tyranny, that is the un- lawful assumption of power, took among the Greek commonwealths and under the Roman Empire. The Greek tyrant had overthrown a commonwealth ; even if it was an oligarchy and not a free democracy that he had overthrown, even if a large part of the community welcomed him as the destroyer of oli- garchy, he had still overthrown a commonwealth ; he had put his own personal will in the place of a svstem of law and order of some kind ; and if he himself sometimes kept his popularity for life, all traces of good will commonly vanished under the rule of his son. That such a tyrant had no means of giving a formal legitimacy to his power is clear on the face of things. When tyrants of exactly the same kind, tyrants of cities, again showed them- selves in the commonwealths of mediaeval Italy, the means of thus wiping out the original stain was supplied by the power of the Emperor, supreme over all. Not a few of the hereditary dukes and marquesses of Italy were tyrants whom the Imperial authority had raised to the rank of lawful princes. But the old Greek commonwealths knew no overlord ; there was no external power that could change Polykrates or Peisistratos into an outwardly lawful * As for instance Strabo (xiii. p. 631) speaks of the tyrants of Kibyra ; ervpapvelro be del' awc^povas de o/nws. G 2 84 Western Europe in the Fifth Century, [iii. ruler of Samos or of Athens. It is perfectly plain that the tyrants of the sixth and fifth centuries before Christ were well pleased to be spoken of as fiacriXevs and that flatterers in prose and verse won their favour by so speaking of them. It is not clear that any tyrant before Agathokl^s received or as- sumed the title in any formal way*. In his day the rise of the various Macedonian princes had made kingship familiar to Greek thought. The Koman tyrant, on the other hand, though he came under the same definition as having taken power to himself without lawful authority, had reached power in a very different way from his Greek predecessor. He had in no way changed the constitution of the state. He had neither suppressed a democracy nor delivered men from an oligarchy. He had simply set up his own power instead of the power of some other prince, and there was no presumption that his rule would be any worse than that of the prince whom he supplanted. He was guilty of whatever amount of human suffering was caused by a revolution wrought by violence ; he was not guilty of any general dis- turbance of the order of things. And it was easy for him, as it was not for the Greek tyrant, to * I except the case of any cities where the old lawful kingship or some survival of it may have gone on, a point which I may have to discuss elsewhere. See also Plass, Die Tyrannis, i. 262. I doubt if the first Hieron, for instance, was called ^aaiXevs by any but flatterers, Pindar and others. The saying of Diodoros (xi. 26 [see my Hist. Sic. vol. II. Appendices I. and XIII.]) about Gelon hardly proves it. On the other hand the second Hieron was undoubtedly jSao-iAeus by a real popular vote. Kingship had then become familiar. III.] Constantine and Maxzmus. 85 obtain a formal and regular confirmation of his authority. In the middle of the third century the most common way of reaching Empire was through the mutinies of the army. The soldiers murdered the reigning Emperor; they chose another in his place ; and the Senate presently voted him all the offices, powers, and titles which together made up the practical sovereignty of the Eoman common- wealth. He who received his commission from the senate, that extraordinary commission always renewed out of which the Empire grew, became a lawful Emperor; he who could not obtain it remained a tyrant. In the times which we have now reached, the power of the Senate has dwindled away. The Fathers indeed appear by fits and starts, under the strange circumstances of the time, with something nearer to their old authority than had been seen for a long time ; but, in the absence of any definite law of succession, it is no longer the vote of the Senate which stands forth as the main source of legiti- mate power. The Empire is becoming more like an ordinary kingdom, able to pass, either by hereditary descent to the children of the last prince or by adoption to some successor or colleague of his choosing. The joint rule of several princes was now familiar, and this system suppHed an easy means of bestowing formal legitimacy on a successful tyrant. When the tyrant had won a certain part of the Empire, and saw no hope of winning the rest, when the lawful prince kept a certain part of the Empire and saw no hope of winning back the rest, a compromise was easy. The lawful prince could admit the tyrant as his 86 Western Europe in the Fifth Century, [iii. colleague in the Empire, and thus, while raising him to the same level as himself, he could keep at least the rank of primus inter pares. The agreement of course, like other agreements, needed not to be kept any longer than was convenient. If either of the new colleagues found a good opportunity of overthrowing his Imperial brother, of taking his dominions to himself or bestowing them on a colleague whom he liked better, that opportunity was seldom missed. The thing had happened over and over again. The lives of Carausius, of Maximus, of the great Constan- tine himself, supply many and instructive examples. Constantine then, master of E-oman Gaul and Spain, still felt that there was something lacking to his position, and he hastened to make it good. He had torn away the Western lands from the dominion of HoDorius ; the armies of Honorius had failed to recover the lands that he had torn away; he was seemingly safe in Gaul, while Honorius was anything but safe in Italy. Yet he now stoops, as it might seem to us, to ask his defeated enemy to raise him from his irregular position to a lawful place at his own side. It does indeed mark the force of tradi- tional feeling that Constantine, called to the throne by an army which had shown itself able to maintain him there, still felt himself the upstart, the usurper, the tyrant, and owned the higher position of the Emperor who had come to the diadem by peaceful means, by a line of those adoptions and associations of sons and colleagues which passed for lawful suc- cession. The tyrant therefore sought for the acknow- ledgement of his claims by the lawful prince ; he III.] Constantine and Maximus. 87 sought for his admission as a third Augustus to the imperial fellowship of Honorius and his young nephew in the East. He sent an embassy (409), an embassy of eunuchs — the soldiers from Britain had conformed to the depraved fashion of the time — to the court of Eavenna, asking the Emperor's forgive- ness for his taking on himself the imperial rank ; it was not, his commissioners were bidden to say, his own act ; the presumptuous step had been forced upon him by his soldiers. It is implied, though it is not said in so many words, that Constantine de- manded the confirmation of their choice, and his own recognition as an Imperial colleague. Honorius was in no position to resist or refuse ; with Alaric and his Goths at no great distance, it was not for him to plunge into another w^ar which might end as the enterprise of Sarus had ended '^'. A domestic reason * The embassy is recorded in a fragment of OlympiodSros, p. 450 ; "Ort Y^fuvaravTwos els Tvpavvlba apBeis Trpeo-jSeverat npos 'Ovapiov^ aKcav pev ku\ dnb rStv arpaTKOTcov ^laadus dnoXoyovpevos ap^ai, (rvyyvaprjv be alraiv, Koi ttjp ttjs ^aaiKeias d^iav Koivatviav. Koi ^aaCkevs 8ia to. ivearrjKOTa Bvcrx^pfj recos KaraSe'p^erai ttiv ttjs ^acriXeias Koivaviav. It IS here that he stops to explain how Constantine came to be tyrant, Kara ras BpeTTavias 8e 6 KcavaravTivos ervyxaveu dvrjpyopevpevos, (TTaaei tS)v eKelcre (TTpaTidT&v els Tavrrjv dvrjypevos rfiv dp)(r]v. He then goes on with the passage quoted above. Z6sirao3 records this first embassy, v. 43. The last words of the chapter before fix the date to the eighth consulship of Honorius and third of Theodosius, that is the year 409 ; eVi Tovra KmvdTavTivos 6 rvpawos evvovxovs Trpos 'Ovaypiov eoTfXXf, crvyyvaprjv, alrmv eueKa tov ttju ^acriKeiav dve(T)(eadai Xa^elv, pT}8e yap eK -npoaipeaems ekecrdai ravTqv, oKka avdyKrjs avT^ rrapa tS)V (TTpaTicoTUiV e7ra\6ei(n]s. Tavrrjs dKrjKows 6 ^atnXevs Trjs aiTTjaeas, dempoau re as ov pabiou avroi t&v crvv 'AXapix^ ^ap^dpcov ov Troppoi ovratv irepl TToKepav irepav diavoeladaij . . . evSt'Swcri rals alrTjcreatv. 88 Western Europe in the Fifth Century, [iii. also moved him — in this matter Honorius himself may have exercised some measure of personal will. His kinsfolk were in the hands of Constantine — Theodosiolus had brought that news with him ; neither he nor Honorius knew that they had been actually put to death before the embassy had been sent, and he deemed that afavourable answer to the demands of their gaoler might be to their advantage*. Honorius therefore acknowledged the claims of Constantine ; he sent him a robe of the imperial purple f. The Boman world, so much of it as was still ruled from Eavenna, Constantinople, and Aries, had again three masters. It would seem that some formality was lacking in this transaction. Or it may simply be that Honorius was stirred to some sign of enmity when the news of the death of Didymus and Yerenianus reached him, when he thus saw how he had been in some sort cajoled into an acknowledgement of the tyrant of Gaul. It is certain that later in the year (409) Constantine sent another message to Eavenna, a message carried this time by a more honourable messenger. Its bearer was Jovius, who is described as a man of high culture and of other merits, but whom we have no means to identify with, or' to distinguish from, other bearers of his own and like names. He came to Honorius when that prince was not in a position to refuse anything ; Alaric was on the point of laying ZoSimOS, V. 43 J Tpoffert -ye \6yov iroiovfievos arvyyevap olneiav irapa Tov Tvpdvvov Karaxpyiivaiv {ovtol be rjtrav Beprjviavos Koi AibvfjLios) , . . tS)V fiev ovv (Tvyyepmv eveKa fj-araiav ei)(e (jipovriBa, irpo TavTrjs t^s Trpea^eias a'KO(Tv 'Ovcapiov tov ^aaikeas atreip. t lb. ; oTTeXoyeiTO Xiyap as ov Kara irpoaipetrip di-^prjprai KcavaTavTiPov. X lb. J (rvvTerapayfiepov 8e top 'Opoipiov diaadfxevos, evXoyop €(pacrK€P eivcu Tols Trepl ttjp 'irdXlap evaaxoXovfiepa (f>povTiv aTparicDTatv ■naveiv ttcos e^o^e araa-cv. Olympiodoros says ; fifTepxerai Kara ttju Pa^evvav eTTt rov Trpanroo'iTov 'Evcre^iov ^ bwacrTela. os pera iKavof xpovov AXko^i)^ov iiTrjpeia Koi vnodfjKTj 8T]po(Tta aal iir o-^eai rov ^aaiKecos pa/SSots avaipe'irai. Avvaareia is a remarkable word to be used of the ascendency of a minister, even of an eunuch under Honorius. TIL] Constantine and Maximus. 91 At this moment Constantine steps in ; we read in two independent narratives that he entered Italy with an army; but we get exactly opposite state- ments as to the motive which took him thither. In one version he is marching to Eavenna, to confirm or to carry out his engagements with Honorius, that is doubtless to give help to his Italian colleague against the Goth *. In the other version the master of Gaul and Spain sets out to add Italy to his dominions f. We may therefore assume with safety that the one version represents the purpose that was openly avowed, and the other the purpose which was commonly suspected. There is no reason to suppose any open breach with Honorius so soon after the npatrroViTos should be noticed as one of the Latin official names which were creeping into Greek, though as yet sparingly. The exact force of vivo6r]Kr] hr\\iouia I do not profess to understand, any more than Labbe did. * So says Olympiodoros, p. 452 ; KtBvorami'oy 6 rvpawos . . . eTreiyonevos irpbs 'Pd^ewav &aT€ aireia-aadai 'Ovcoplco. This was doubtless what was given out publicly. t Sozomen, ix. 12 ; KavaravTlvos reas kcto. yvafirfv iipaTTeiv 8okS)V, KcovcrraVTa tov viov avTi Kaicrapos ^acriKea Karaarriaas, e^ovXfvero tt)V 'iroKiav KaToKa^eiv. That was doubtless what the court of Eavenna feared. It certainly seems strange that Gregory's authority, Eenatus Profuturus Frigeridus, altogether leaves out Constantine's Italian expedition. It is just when it should come that he tells us (Greg. Tur. ii. 9) that Constantine was "gulae et ventri deditus," having " nullum ex Italia metum." Does that mean after his return from Italy ] I do not see on what ground Wietersheim (ii. 166) places the Italian expedition in 410. Surely the whole story of Eusebius and Allobich fixes it to 409, while Alaric was still only threatening Eome. 92 Western Europe in the Fifth Century, [iii. second embassy. Constantine appears to have as- sumed the consulship in partnership with Honorius * ; and on the whole it is most likely that it was now, when he was at the height of his power, that he raised his son CoDstans, who might pass for the conqueror of Spain, from the rank of Csesar to that of Augustus f. There would thus be four acknow- * I miglit not have found out this consulship of Honorius and Constantine, which is not to be found in any fasti and which was most likely unheard of outside of Constantine' s dominions, but from the mention of it by Tillemont, v. 570, and the references there given. The insci'iption which is given in Gruter 1072 comes from the church of Saint PauUinus at Trier, and, being in Greek, throws, as usual, some light on the spelling and pronunciation of that tongue. It runs thus ; ENGA . KEITE • EYCEBIA • EN EIPHNHI . OYCA • lEPOKfiMITI • AHO . 1 . KfiMHC AAAANQN ZHCAC HMEP • O nPOC . ETQN T • EN YHATEIA . ONOPIOY TO H • KAI KQNCTANTINOY • TO • A • MHNI HANHMOY "ii . HMEPA • KI . B • EN EIPHNH. The year is 409. One must suppose that Constantine, on receiving the purple from Honorius, declared himself consul, without regard to the rights of Theodosius. How much lost history might have been kept if all makers of epitaphs had put the consuls. t So says Sozomen in the passage (ix. 12) quoted above, where he distinctly places it before the Italian expedition of Constantine. I am not sure that this is really contradicted by Zosimos, who does not mention that expedition, when he says (vi. 13), after recording the hostageship of Placidius, to /xev Kara ttjv 'iraXiav kv tovtois ^V KcovarapTipos 6e tw 7rai6i Kavara ro Bidbrifia irepideis Koi avrl tov Kaiaapos jSao'tXeu TrerroiriKas, ATroWivdpiov TrapaXvaas Trjs dp)^rjs erepov dvT avTOV vnapxop Inedei^ev. Sozomen's date seems more distinctly given as i III.] Constantine and Maximus. 93 ledged Imperial colleagues, Ilonorius, Theodosius, Constantine, and Constans ; the making of Emperors was still for a moment in Eoman hands ; it was very soon to pass to the Goth. Thus, in all outward seeming, help was coming from Aries to Eavenna. But it was deemed at the court of Eavenna that such help was likely to be dangerous; it was believed that there were high officials about the Italian Augustus who were ready to displace him in favour of his Gaulish brother. Allobich, slayer of Eusebius, had won power, but not confidence ; he was suspected of being in league with Constantine to transfer to him the whole dominion of the West*. It would seem that Honorius, as princes sometimes do, conspired against his minister and found instruments ready to rid him of the suspected traitor. An opportunity was found as Allobich was riding, according to custom, in a solemn procession before his sovereign. Allobich Avas cut down by the loyal assassins, and the Emperor, springing down from his horse, gave God thanks in the hearing of all men for having preserved him from a manifest traitor f. So sultan-like had the a date, while that of Zosimos comes in more casually. And Olympiodoros (p. 453), though he too only mentions the matter incidentally, clearly places it before the revolt of Gerontius. That revolt took place Kava-ravrivov rov Tvpdvvov Koi KavaravTOs tov TraiSos, OS Trporepou fiev Kaia-ap, eTreira 8e Koi /Sao-iXevs eKex^ipOTOvrjTO, TovTCiv r]TTr]6ivT(i>v Koi Tre(f)evy6TO)v, * Sozomen, ix. 12; ov \^AW6^ixov] 817 (rrparriyov 'Ovmplov ovra koL VTTOTTTOV 0)5 Kwi/oTajTiVo) TTpaypaTevojj.evQV naaav rf/v Trpbs Tr\v ovavovs eni^ovXov oTraXXayety. 'O Kparmv seems an odd word for the Emperor. On this, passage Gibbon (vol. v. p. 289) remarks that the " assassination of Allobich, in the midst of a public procession, is the only circumstance of his life in which Honorius discovered the faintest symptom of courage or resentment." * 01ympiod6ros, p. 452 ; KavaravTlvos . . . t6v 'AXKo^ixov davarov liaBav . . . (})o^r]6f\s VTTotTTpecjiei. Sozomen, ix. 12 ; irapapd^as ras KoTTias AXTTcts, riKev els Ai^epSiva ttoKiv rrjs Aiyovpias. ixeXXcov 8e Trepaiovadai tov 'Hpi8av6v, t^v avTrju oSof dve(rTpe(f)e, fiaBoav tov 'AXKofiixov Bavarov , . . Kcovo-tuvt'ivos 8e (pevycov Tfju 'AprjXaTov KaTfXaj3e. III.] Constantine and Maximus. 95 change of rulers in the first instance, and the land might, it would seem, have settled down quietly again after the movement of the kinsmen of Honorius, if the new administration had not wounded local feeling in a very tender point, Spain, as we have seen, had been used to be defended by the arms of her own children. . The legions that served in Spain had been Spanish legions, and the keeping of the Pyrensean passes had been by usage entrusted to what we may call a national militia. Spain had no frontiers through which the barbarians could make their way ; she was not therefore, like Italy and the East, accustomed to have her borders guarded by one body of barbarians hired to keep out another body of their fellows. But now Constantine and Constans were guilty of the fatal, yet not unnatural, mistake of removing the local force, and entrusting the moun- tain passes to the keeping of their own barbarian allies, the Honorians. These troops were further indulged, by their commander Gerontius, it would seem in excessive licence in the way of plunder ; they were above all allowed to harry the district of Valentia, which, doubtless as having supported the cause of Didymus and Verenianus, was dealt with as an enemy's country. The demand of the Spanish legions that the barbarians might be withdrawn, and the old state of things restored, was refused, and great dis- content arose *, To quiet or suppress that discontent * This very interesting notice of tlie local usages of tlie Spanish provinces comes in different shapes from Z6simos, Orosius, and S&zomen. The first of these (vi. 5), to the passage quoted above p. 75, mentioning Gerontius as left in command in Spain, adds,, 96 Western Europe in the Fifth Century, [iil. the new Augustus Constans was sent. He went, as far as we can see from our fragmentary authorities, about the time of his father's Italian expedition. It is plain that the Spanish troubles were laid to the charge of the officers whom Constans had left to represent his father in the peninsula. He now took with him a general named Justus, destined, it would seem, to supplant Gerontius, while ApoUinaris lost his office of Prsefect, which was bestowed on a certain Decimius Eusticus, who had hitherto been Master of the Offices *. The wrath of Gerontius was naturally KaiToi ye ratv iv 'l^rjpla aTpaTOTreSatv ifnncrrevdrjval Kara to (rvmjdes Tr]V pav CLKovovres. TraprjpeXrjKOTcov re tS>v eniTpaTrevTcov irapa KavcrravTos rfjv (ppovpav rrjs irapohov, iTaprjkdov els 'lanaviav. He then, at the beginning of his next chapter, records the elevation of Maximus as happening eV rovra. Sozomen here does not personally connect Gerontius with the barbarian entry, but he attributes it to the negligence or treachery of the Honorians under his command. Orosius is to the same effect, but somewhat more explicit. After the words cited above from vii. 40, he adds, "Prodita Pyrenei custodia, claustrisque patefactis cunctas gentes quae per Gallias vagabantur, Hispaniarum provinciis immittunt, iisdemque ipsi adjunguntur." Olympiodoros (p. 453) seems distinctly to connect Gerontius personally with the barbarian inroad ; TepovTios 6 arpaTrjyos rfjv irpos Tovs jSap^apovs dapevicras elprjvrjv, Ma^ipov . . . /SacrtXea duayopevei. So Zosimos, vi. 5. Constans enters Spain with Justus, e'0' m TepovTios a)(66pevos, Koi tovs avrodi TrepnTOLrjcrdpevos crrpaTitiTas, inaviarijcn KcovaravTivcp tovs ev KeXrois ^ap^dpovs. He goes On to speak of Constantine's dealings in return with other barbarians, but says nothing about Maximus. Renatus too (Greg. Tur. ii. 9) is clear on this head, though it is hard to work in some of the details of his story; "Ab Hispania nuntii commeant a Gerontio Maximum . . . imperio praeditum atque in se [Constantinum et Constantem] comitatu gentium barbararum accinctum parari." III.] Constantine and Maximus. 99 one whom Gerontius might think good to clothe with the purple. We are so seldom taken behind the scenes, so seldom allowed to study the motives of the actors in this most confused story, that we can merely guess why Gerontius, instead of laying claim to the Imperial dignity in his own person, set up a certain Maximus as Emperor or tyrant. The proclamation of some rival Emperor was his only chance ; but we can do no more than guess at the causes which made Gerontius forbear from placing the diadem on his own brow. We see easily why at this very moment Alaric w^as setting up a puppet Emperor in Italy for his own ends, why later in the century Eicimer set up and put down Emperors at pleasure. For the days had not yet come for an avowed barbarian to mount the throne of the Caesars in his own person. Stilicho, charged with plotting the elevation of his son Eucherius, is a nearer case to this of Gerontius. But StiHcho was said to come of the stock of the Vandals*. The lapse of another generation, the connexion by marriage between his house and that of the Emperor's, may have caused the son to be looked on as more Eoman than the father. But Gerontius would seem to have been a provincial of the province of Britain, as good a Eoman then, by the edict of Antoninus, as any man in Spain, Gaul, or Italy. It is therefore by no means easy to see, why, when he risked himself * If we may believe his enemy Orosius (vii. 38), " Comes Stilico Wandalorum imbellis, avarae, perfidse, et dolosse gentis genere editus." " Imbellis " at least is a strange epithet ; but Livy and Gregory of Tours have equally strange sayings about Latins and West-Goths. H2 ^-^fC- 100 Western Europe in the Fifth Century, [iii. and all that belonged to him in a struggle for power, in a struggle against Honorius and Constantine at once, he did not at least run the risk on his own behalf and in his own name. Whatever were his motives, the fact is clear. It was not himself but Maximus whom Gerontius chose for the dangerous honour. But who was Maximus 1 That one among our authorities who is on the whole the most trust- worthy, but whose evidence has come down to us in the most fragmentary state, seems to call him the son of Gerontius, in which case we should have the closest parallel of all to the alleged designs of Stilicho. He was, it is said, serving among the domestics, the household troops doubtless of Constantine and Con- stans. Other writers speak more vaguely of Maximus as a friend or dependant of Gerontius. In any case, just as with Constantine himself, the name of the renowned British tyrant of the last century may have gone some way towards securing his elevation, though we are also told that Gerontius deemed him a man personally fit for the post*. Maximus therefore * 01ympiod6ros (453) calls Maximus the son of Gerontius; Ma^t/Ltof Tov eavTov Traiba, els ttju bofieariKoiv rd^iv reXovvra, ^aaiXea avayopevei. Sozomen's account (ix. 1 3) is, TepovTios 6 tS>p Kmva-TavTivov (TTparrjywv apiaros, 8v(Tpevfjs avrSt yiyovev, eTTiTrjdfiou re tls Tvpavviba Md^ipop TOV avTov olKelov vofiLcras, ^acrCKiKr^v iveSvaev e(r6rJTa. Renatus, in the passage before quoted, calls him " Maximum, unum e clientibus suis." Wietersheim prefers the witness of Olympiodoros, which is doubtless the best in itself. But it is hard to see how a son could be mistaken for anything else^ while a stranger might be more easily mistaken for a son. Orosius (vii. 42) puts the whole story of Gerontius and Maximus out of date ; " Constantem filium Constantini Gerontius comes suus, vir nequam, magis quam probus, apud Viennam interfecit, atque in ejus locum Maximum quemdam III.] Constantine and Maximus. loi assumed the purple and held his court at Tarragona*. Master of at least the north-eastern corner of Spain, he found himself better able to maintain his authority against other representatives of the Koman power than he was against the common enemies of the Eoman name. We cannot have a better illustration of the way in which these tyrants rose and fell than in the story of Gerontius, a story full of striking adventure, on which we have now entered. As Constantine has done by Honorius, so Gerontius now does by Con- stantine. All alike are Emperors to those who accept their dominion, tyrants to all beyond its bounds. The truth is that, during the whole life of the Eoman power, down to the disputes of a Palaiologos and a Kantakouzenos, the only chance for a man at the head of an army who had fallen under the suspicion of the master whom he was supposed to serve was to assume the purple himself It v/as a frightful risk ; but he might succeed ; otherwise he had no hope. Thus the Empire was torn in pieces by the personal interests of particular men, at a moment when no one frontier was safe against foreign enemies. Yet the wonderful thing is how often the Empire came together again. What strikes us at every step in the tangled history of these times substituit." This account would be true, if only tbe order of the two facts was turned about. This neglect of chronology comes of the fact that Orosius, after recording the events at the Pyrenees, goes off into an edifying discourse on the doings of tlie barbarians in Spain, and now comes back to give a " catalogus tyrannorum." * S6zomen, U. S. ; eV TapuKovrj bidyeiv e'iaa-ev. 102 Western Europe in the Fifth Century, [iii. is the wonderful life which the Koman name and the Koman power still kept when it was thus attacked on every side from without and torn in pieces in every quarter from within. The personal good luck of Honorius has been noticed both in older and in later times*; like the Persian conqueror of old, he overcame most of his enemies without stirring from his hearth t, and those whom he could not over- come he at least outlived. But the good luck, if not of the local Rome, at least of the wider Romania, is still more to be noticed. Whatever blows fall, some- thing escapes, and that something commonly lives ; it grows again, and wins back part at least of what had been lost. At this moment the whole West is overrun by barbarian invasion. Britain falls away; Gaul is ravaged from the Rhine to the Pyrenees ; the greater part of Spain, as we shall presently see, is cut up into barbarian kingdoms. By a blow more striking and terrible than all in its historic and dramatic aspect, Rome itself has been entered and sacked by a barbarian enemy. Yet the Roman name and the Roman power live on. The dominion of the conqueror of Rome passes to a successor who is ready to act as the soldier of Rome and who aspires to be the son and brother of her princes. While Italy is thus saved by the exchange of Alaric for Atawulf, neither Gaul nor Spain is wholly lost. * Sozomen (ix. 16) is strong on this head. See also Procopius, Bell. Vand. i. 2, and Gibbon. t ^sch. Pers. 860-63 ; oercas S' fiXf ttoXcis iropov ov dia^as "AXvos TroTafiolo, ovb d(f) iarias (rvdds. III.] Constantine and Maximus. i03 A corner of Gaul escapes barbarian ravage ; a corner of Spain escapes barbarian partition. And if at this moment neither Gaul nor Spain is in the obedience of Kavenna, if each land has its own Emperor or tyrant, yet the tyrants at once turn their arms against one another, and all presently yield to the fortunate star of the lawful prince. And if that lawful prince wins his victories by deputy, one at least of his enemies suffers defeat by deputy also. Maximus then is tyrant at this moment in Spain, reigning at Tarragona, but without any such acknow- ledgement of his position as Constantine had won from the unwilling Honorius. His immediate enemy was Constantine, whose power in Spain he had overthrown ; more immediately again it was Constans by whom his father Constantine had been represented in Spain. But Constans, though the greater part of his father's forces were under his command*, could not stand against the movement which had raised Maximus to power. He and his prsefect, Decimius Eusticus, who, we may gather, was specially unpopular, fled into Gaul to Constantine f. From his capital at ToO TrXeioj'os t^s Bvvdfieas fiepovs ovtos iv 'l^rjpia says Z6sim0S (vi. 5), but his account is confused, and lie mentions nothing of the acts of Constans, or indeed of Constantine, after Constans went into Spain with Justus. t It is again excessively hard to put our accounts together. Olympiodoros (453) tells us how, Kavaravriuov koI KwvcTTavTos . . . fjTTTjdevrav Koi TrecjyevyoTotv, Gerontius sets up Maximus as Emperor, fiVa eTTibia^as KavaravTa, KareTrpd^aTO dvaipedrjuai, Koi Kara irodas furero, Smokcoi; Ka\ rbv Trarepa KcovaravTlvov. One would think from this that Constantine as well as Constans was in Spain ; yet every other picture places him at Gaul, and 01ympiod6ros himself speaks of 104 Western Europe in the Fifth Century, [in. Aries that prince — an acknowledged colleague of Honorius and Theodosius — had to keep, if he could, so much of Gaul as was still Roman from the attack which was threatening from Spain. Maxim us him- self did not stir, any more than Honorius ; but him at Aries directly after. Nor would any one find out the presence of Constans at Vienne. Sozomen (ix. 12, 13) is a little clearer, though he tells his tale in a somewhat strange order. After the return of Constantine from Italy comes the passage cited in a former note ; Kcai'o-Tawti'os 6e (fifvycov ttjv 'AprjXaTou KareXajSe, Kara TovTov 8e KoX Kavcrras 8' avTov nais (f)evyo>v tK ttjs 'icrnavias. Then, as an explanation, follows his second account of the barbarian invasion of Spain above quoted ; then Gerontius makes Maximus Emperor and leaves him at Tarragona, alrhs he KavaravTlva inea-TpaTeva-ev, iv ■7rap68a Ka>v(TTavra tov vlov airov iv Bievvrj ovra avaipedrjvai irapaa-Kevdaas, eVei Se efiaBe KcovaTavrlvos ra Kara Md^ifjiov, 'E86^i)^ov fiev tov avrov (TTpaTTjybv nepav rov 'Ftjvov TreTro/i^ev, ^pdyycov re Koi AXafiavociv avp.- yLa)(j.av TrpoTpeyJAajjievov, KavaTavTi 8e rat avrov TratSt Biewrjs koi Ta>v T^8e noXetov rr^v i\o- xprjl^ariav e^atKeike. This, bating the special metaphor, is almost translated in the words of the Peterborough Chronicle, 1087; " He wses on gitsunge befeallan, and grsedinesse he lufode mid ealle.' + This story is told by Olympiodoros (450), but it is not easy to fix the date, and it must have been after our time, after his marriage with Placidia. Olympics lost his power, as is described by Zosimos, v. 46, but rose to power again, and on his second fall, was thus dealt with by Constantius ; e^eVfo-e ttjs apx^s. eha ndKiv fiTf^rf ravTrjs, etra eKireaav, ponaiKois vaTepou vno Kavcrravriov os ijydyero II\aKi8iav Traiofifvos dvaipelrai, Tas aKoas npoTepop eKKOiifU. He adds the moral comment J koL rj 8Ur] top avocriovpyov els rekos ovk apr](Tav. Olympiodoros (454) is less clear, but he gives us the reason for the desertion of the troops of Gerontius ; Tepovnos, Trapayevofievcov OvK(j)iKa koi KcovcrTavTioVf (f>evy€i. Koi KaraKijcpBels, on iyKparas rjpxe tov oiKdov (TTparov, vn avTwv eKeivav em^ovXeveTai. We should hardly find out from this that he got back to Spain, but that he did so is plain from the words which next follow in Sozomen. I 2 116 Western Europe in the Fifth Century, [iii. Through the defection of so great a part of the soldiers of Gerontius, the two besieging armies must have been largely made up of the same men. Meanwhile it will be remembered that the Frankish Edobich, now, at all events, the best officer in the service of Constantino, had gone beyond the Rhine to seek for allies for his master among Franks and Alamans. His mission was not in vain. Aries did not yield in a moment. Warfare beneath its walls lasted longer than it had lasted beneath the walls of Valence or seemingly beneath those of Vienne. The siege was already in its fourth month (ill) when the news came that Edobich was drawing near with a vast and motley host of barbarians to the relief of Constantino*. Constantius and Wulfilas were troubled * I have here ventured to take a date from Kenatus and a fact from other writers. In the extracts made by Gregory (ii. 9) from Eenatus, Edobich goes to collect Frankish and Alamannian allies, and we hear no more of him. But " vix dum quartus obsidionis Constantini mensis agebatur," not Edobich, but Jovinus, who has already assumed Empire, comes with a vast barbarian host, and then Constantine is given up and sent into Italy. Sozomen, on the other hand, records the mission of Edobich and its issue. He comes back with the troops he has gathered, fights Constantius, and dies as in the text. Then Constantine abdicates. Sozomen then mentions the overthrow of Jovinus, but without mentioning the time of his usurpation. It seems to me that, as far as the whole campaign of Constantius is concerned, Sozomen gives a coherent and probable account. Henatus may have done the same, if we had his full text ; but we have only the account that we can put together out of fragments quoted from him by Gregory. "What becomes of Edobich? The march of Jovinus is not mentioned elsewhere. Why should Constan- tine or the defenders of Aries surrender — seemingly to Constantius III.] Constantine and Maximus. 117 at the tidings ; for a moment they even, like Sarus, made up their minds to leave Gaul and await the enemy in Italy. But the march of Edobich was too — because of the coming of Jovinus? Why did not Constantius stay to fight Jovinus % But if we accept Sozomen's version, the whole is clear. With the failure of Edobich, Constantine's hopes of relief are at an end, and he surrenders. The work of Constantius, in Gaul at least, is done ; the usurpation of Jovinus, we must suppose, comes later. Is it not most likely that there is some confusion in Gregory's extracts from Renatus, and that the host which came in the fourth month of the siege was really that of Edobich and not that of Jovinus % Gregory does not always copy things accurately, as we may see by his quotations from Sidonius, where we can test him. The withdrawal of Constantius, the quiet surrender of Gaul to Jovinus, which Renatus, as we have him, implied, have naturally puzzled both Gibbon and Wieter- sheim. No one but Eenatus seems to put the usurpation of Jovinus before the fall of Constantine. Orosius does not follow strict chronological order, for he mentions the death of Constans and adds " in ej us locum [Gerontius] Maximum quemdam substituit.' ' But when he has got rid of Maximus, he says emphatically, " Jovinus postea vir Galliarum nobilissimus in tyrannidem mox ut assurrexit, cecidit." Prosper Tiro (whatever he is worth) places the fall of Constantine in 41 1, and the usurpation of Jovinus (" tyrannidem post Constantium invadit") in 413, the same year as his overthrow. Marcellinus kills Constantine in 411, and in 412 has "Jovinus et Sebastianus in Galliis tyrannidem molientes occisi sunt." The higher authority of Prosper places the fall of Constantine in 411, and in 413 has "Jovinus et Sebastianus fratres in Galliis regno arrepto interempti." In all these there is no very distinct or trustworthy statement of the date of the usurpation of Jovinus. The casual mention in Prosper and Marcellinus, though suggesting a later date than that of Renatus, does not amount to a direct statement. Idatius alone is explicit, and I think decisive, on the whole matter; 118 Western Europe in the Fifth Century, [iii. speedy to allow this timid scheme to be carried out. The besiegers of Aries were on the left, the eastern, side of the Khone ; Edobich seems to have been marching southward along the western bank. When the news came that he was actually encamped in their near neighbourhood, on the peninsula that is covered by the Julian Colony, the furthest point to the north-east of the dreary region of the Camargue, all thoughts of retreat were cast aside by the generals of Honorius. They determined to face the enemy boldly. They crossed the river to give battle to the new comers. Both this fact, and the scheme of action that was planned between the Eoman and the Gothic commander, a scheme which showed no lack either of skill or of daring, seem to show that the host of Edobich could hardly have reached even the wall of the Colony, and that the battle must have been fought at some little distance from Aries itself*. For the followers of Edobich, unlike the "xvii [411] Constantinus post triennium invasse tyrannidis ab Honorii duce Constantio intra Gallias occiditur. xviii [412] Jovinus et Sebastianus fratres intra Galliam, et in Africa Heraclianus pari tyrannidis inflantur insania. xix [413] Jovinus et Sebastianus oppressi ab Honorii ducibus Narbona interfecti sunt." I cannot think that the authority of this very clear statement is weakened by the inaccuracy of placing the death of Constantine (whom he had not mentioned before), as well as his reign and overthrow, " intra Gallias." I hold therefore that Jovinus did not set himself up till after the death of Constantine, and that the army of Jovinus spoken of in Gregory is really the army of Edobich whose fate is described by Sozomen. * I think I see something like this change of purpose in the not III. J Constantine and Maximus. im followers of Gerontius, did meet the army of Con- stantius in open fight. According to the plan arranged between him and Wulfilas, Constantius himself*, at the head of the infantry, awaited the attack of the enemy. Wulfilas, with the horse, seemingly a small body, lurked in ambush at no great distance. The host of Edobich, eager for battle, marched by the hidden foes without suspect- ing their presence, and met the troops of Constantius face to face. At a given signal Wulfilas and his horsemen dashed out of their lurking-place and charged straight on the rear of the enemy. The battle was at once decided; the barbarian host was broken ; some fled ; some were slain ; the more part threw down their arms, craved for mercy, and received itf. Edobich fled ; he had, in old Teutonic guise, like Englishmen ages after, waged the actual battle on foot ; the horse was but a means to take the warrior to and from the field. When the day very emphatic language of Sozomen (ix. 14). Edobich is said to be coming, tovto 8e koI roi/s 'Ovaplov arparrfyovs ov fierpicas ipeiv eKeXevcrev, ovk dyadrfv ij'yij- crdfievos kokov ^evo86)(ov ttju (rvvovalav eaea-dai avra rj rfj arpaTia, TIL] Constantine and Maxtmus. 121 And so the man who slew his friend in the day of danger was sent away empty by the man who refused to reward crime even when he gained by it*. The overthrow and death of Edobich sealed the fate of Constantine. Seeing no longer any hope of Empire, or indeed of life if he still laid claim to Empire, he put aside his diadem and purple ; he betook himself to a church — already perhaps a church of Saint Trophimus — for sanctuary. He there found a bishop who perhaps deemed that in such a case he might dispense with the precept to lay hands suddenly on no man. Constans son of Con- stantine had of a monk become Caesar ; Constantine himself was now of an Augustus to become a Christian presbyter f. In that character he deemed that his life at least would be safe. But no great harshness was to be feared from Constantius. The defenders of the city, on receiving the general's oath for their safety and for that of their fallen prince, threw open their gates, and the people of Aries at least had no need to complain of any breach of faith on the part of the conqueror if. No blood was shed by * Sozomen seems to quote a proverb; Kara Kevrjs, tovto 8fj roO Xoyou, xavav anrjKBe. t Again tlie fullest and clearest account is that of S6zomen (ix. 15), who alone helps us to some geography; /xera r^f vUr^v dvTorepaicodeicrqs avdii npos ttjp ttoKiv t^s 'Ova>ptov arpaTias, paBav Kojj/oraj/Tti'os dvaipe'iadai 'Edo^ixov, airos i^ eavrov ttjv aXovpyida Kai ra (rvfi^oka TTJs ^acnXetas aTredero, Koi KardKa^av ttjv eKKkrjaiav, x^i'POTOveiTai Trpecr^vTfpos. So Olympiodoros, 453; Kcova-Tavrlvos KaTa(f)vya)v els evKTTjpiov, TrpeafivTepos Tore ^eipoToveiTai. X S6zomen, u. S. ; SpKovs re wporepov Xd^ovres ot ea-a retxSyv dvolyovai Tas TTvXas Kai ^etSoCs d^iovvrai ndvres. It is from Olympiodoros that we 122 Western Europe in the Fifth Century, [iii. Constantius. But Constantine and his younger son Julian the Nohilissimus were sent to E-avenna to ahide the judgement of Honorius. The Emperor remembered the slaughter of his kinsmen and did not hold himself bound by the oath of his general. Messengers of death were sent to meet the prisoners, and the priest Constantine and his son were beheaded at some point of their journey, either on the Mincio or at some point nearer to Ravenna *. learn that the promise of safety was specially made to Constantine personally ; SpKcov avra vnep (TcoTrjpias dodevTcoVj Koi rots iroKiopKovuiv ai TTiikat ttjs TroXeas avaneTovvvvTai. * Olympiodoros, 454; TrepTrerai. a-vv ra via Kcuva-Tavrlvos TTpos 'OvoapwV 6 8e fivrjcriKaKoov avTois virep tS>v dvey^iav avrov ovs irvyxave KavcrravTlvos dveXav Tvpo TpiaKovra rrjs 'Pa^evvrjs pi\la>v irapa tovs opKovs irpoa-TdTTei. avToiis dvaipe6r)vai. The geography of Renatus (Greg. Tur. ii. 9) seems different; " Reserata urbe Constantinus deditur. Confestimque ad Italiam directus, missis a principe obviam per- cussorihus, super Mintiam flumen capite truncatus est." Surely no point on the Mincio can be within thirty miles of Eavenna; yet the exactness, in different ways, of both accounts is remark- able. Sozomen does not mention the place ; KavaTavrivos a/ia 'lovKiava rm naiSi irapaTzepCpdels els 'iraXiav, irpXv (pddaai Kara rrjv 686v KTivvvrai. It is not wonderful that writers who were not telling the story in the same detail as Sozomen or even as Olympiodoros should have left out the sending into Italy, and have fancied that Con- stantine was put to death at Aries. It mattered a good deal for the characters of Constantius and Honorius ; but it mattered not at all for the general course of things. So Orosius tells the whole story in a few words ; " Constantius comes in Gralliam cum exercitu profectus Constantinum imperatorem apud Arelatum civitatem clausit cepit et occidit." So Idatius in the passage quoted already. Prosper Tiro has simply under 411; "Constantinus tyrannus occiditur." Marcellinus puts the whole story of Constan- tine under 411; "Constantinus apud Gallias invasit imperium, III.] Constantine and Maximus. 123 Just at this stage of our story we cannot complain of any lack of personal incident. We part for a moment from the meagre entries of annalists and from fragments pieced together from this source and that, to listen to such a story as the fate of Edobich and its punishment. But the stirring story of the fate of Edobich is tame compared with the thrilling tale of the fate of Gerontius. Flying, as we have seen, from Aries, he betook himself to Spain, deeming that there at least he might reign in the name of the tyrant of his own making. But his hold on the Spanish province was gone. The troops that had been left in Spain scorned the commander who had fled*. They plotted his death, and besieged him in his own house. He had with him his wife Nouuechia, a few slaves, and a faithful Alan. In one version he too is a slave ; in a more likely shape of the story he is an honourable companion in warfare t. The most detailed account of the death filiumque suum ex monacho Csesarem fecit. Ipse apud Arelatum civitates occiditur ; Constans filius apud Viennam capite pleetitur." Any one would think that father and son were put to death in the same interest. Prosper himself, who has recorded the revolt of Constantine in its place in 407, sums up his later story under 411; " Constantinus per Honorii duces Constantius et Ulphilam, apud Arelatum oppidum victus et captus est, cujus filium Constantem in Hispania regnare orsum Gerontius comes in Maximum quemdam tyrannidem transferens interemerat.'' * Sozomen, U. S. ', oi 8e iv 'laivavlq. crrpaTi&Tai evKaTa(pp6vT}Tov otto TrjS (jivy^s 86^avTa tou TepoPTiov i^ovXevcravTO dveXtlv. + In Olympiodoros he is fls avvayaviaTris 'AXavos to yevos, els BovXovs avTov apiGpovfievos. In Sozomen he rises to the rank of eh 'AXavos emrrjBetos. Surely this is no slave, but a thegn or Bepdirasvy a gesiiS or 8ios iraipos. 124 Western Europe in the Fifth Century, [iii. of Gerontius comes from an ecclesiastical historian who seems suddenly to take up a character oddly mingled between a pagan philosopher and a writer of romance. Gerontius and his few comrades, attacked by night, defend themselves from the upper stage of the house which we must conceive as a strong tower capable of offering some effective resistance. Not a few such miniature fortresses in Ireland and in the border shires of England will enable us to call up the scene. Through the embrasures of the battlements of his pele-tower, sheltered no doubt by the wooden roof coming down on the battlements, Gerontius, his Alan friend, and seemingly the slaves also, did no small execution among the assailants. Themselves almost beyond the reach of missiles, they shot at the besiegers till full three hundred of them were slain, when their stock of arrows failed them. What follows we should hardly beheve if it came from a lighter source than an ecclesiastical history. It was night, and for a while the attacks of the besiegers seem to have ceased. The slaves escaped from the house ; Gerontius, and therefore we may suppose, his wife and his faithful comrade, might have done the same. But Gerontius, restored to his wife, hke Odysseus, after a long absence, could not bring himself, even when the lives of both were at stake, to leave a besieged tower that sheltered her. His Alan thegn *, his true Oepdircov — * The details of the story all come from Sozomen. Olympio- doros says only, Tzvp Kara ttjs oIkius ovtov dvrjyl^av' 6 8e irpos Toi/s inavatnavTas Kaprepas ipa^^ero. He then mentions the presence of the Alan. But in Sozomen we read; ^pa^dnevoi vvKxap airov ] iir.] Constantine and Maximus. 125 we are hardly wrong if we use either the Teutonic word or its Greek equivalent — tarried with his lord and friend, a doomed groomsman at the renewed wedding. The day dawned, but it brought with it to Nounechia only a morning-gift of death. With the light the besiegers was again active ; their weapons had failed; they now brought fire to the attack, and the three felt that there was no longer hope. But they would not fall alive into the hands of their enemies. First of all Gerontius smote off the head of the faithful Alan, who offered himself to the stroke, a gesi^ who would not outlive his eld,er. Then the weeping Nounechia craved a last gift of the husband who was so strangely to die for love of her ; let her be slain by his hand rather than pass into the power of others. She thrust herself eagerly against the weapon; Gerontius yielded to her prayer, and the faithful wife died by a stroke of the same sword wielded by the same hand that had ended the days of the Alan. Gerontius now stood alone beside the dead ; the stroke of the sword failed him ; he then grasped the trusty dagger that hung by his thigh, and drove it to his heart *. It might seem that rr]V oiKiav Karebpafiov, 6 Se fieff ivos 'Akavov iTTinjdelov Koi oXiyav oiKerav, avadep To^evav, vnep tovs rpiaKociovs dpaipei (Trpariaras' iirCKei^dvTmv he tS)V jSeXSi/, (jievyovaiv oi olKerai, KaOivres avroiis tov olKTjfiaros Xddpq. TepovTios fie tov Ictov BiacadTJi/ai dwdfievos, ovx eiKero, KaraaxeBus epcori Nouw;^tas rrjs avTov yap,€Trjs. If lie could escape, surely she could also. * Olympiodoros records the three deaths in a few words ; reXos Toi; re 'AXavov koi ttjv yvvaiKa^ tovto 7rpo6vp.ovfifVovs, dvaipet, eTrtKaTacr^a^ft 8e Koi eavTov. Sozomen enlarges ; Trept Se Tfjv eco TTvp en^dKovrav rfj oiKia rav aTparKorav, ovk e^cop konrop craTtjpias eX;riSa, Ikovtos tov 126 Western Europe in the Fifth Century, [iir. all these details of deeds of which no witness was left could hardly have been inferred even from a more careful examination of the dead bodies than was likely to be made when wrathful enemies at last made their way into a house which was perhaps already burning, But we must tell the tale as we find it, and specially we must not leave out the comment. Nounechia, so our ecclesiastical guide tells us, a Christian woman, died with a courage worthy of her faith, and left a memory which ought never to be forgotten'"'. It is for some moral ductor diibitantium to rule whether we have here truly a case of " homicide by necessity." The ordinary historian may keep himself to the humbler work of wondering at the minute knowledge of the guide whom he has to follow. So, we are to believe, died Gerontius the Briton, who had helped to set up one tyrant in Gaul, and who had set up another in Spain of his own hand. His former (TVvovTOS avTO) A\avov dnoTenvei rfjv Kev evia ttjs 'Pdfiaicov dpxrjs drroaT^vai Koi Kad iavrov ^loreveiv^ ovKeri rots Tovrav inraKovovTa vopois. t See above, p. 44. IV.] The Barbarian Invaders. 147 what we are instead of being like our neighbours in Gaul and Spain. That there is in any part of the world an English folk speaking the English tongue is largely owing to the facts which lurk in the short statement that the Britons took up arms and set free their cities. The existence of a British people in Britain, a British people free, bearing arms and knowing well how to wield them, was an essential condition of the growth of an English people in Britain. When our turn soon came to take our greatest part in the general Wandering,we had another work to do from that which fell to the lot of Goths, Vandals, and Franks. They had hardly more to do than to receive the submission of Eomans; the con- quest was so easy that they themselves were con- quered ; in speech, in much besides speech, the Goth and the Frank became Eomans. We had not to receive the submission of Eomans but to overcome the long and stubborn resistance of in- dependent Britons. The Eoman of Gaul made in the end the moral conquest of the Frank, because he never overcame him, never faced him, on the field of battle. The Briton had no chance of making the moral conquest of the Angle or the Saxon, because year after year he withstood him, face to face and hand to hand, in defence of a land which was his own land and not the land of a foreign master. The difference is written on the whole history of the fifth and sixth centuries. The Angle and the Saxon won Britain in fight, in fight, not against Eomans, but against Britons. The Teutonic invaders of Britain did not turn their arms against one another till they were L 2 148 Western Europe in the Fifth Century, [iv. well settled in the land. The Frank won Gaul in fight ; but it was almost wholly in fight with fellow Teutons that he won it. Save in the new-born British peninsula, there were no avowed Celtic enemies to fight with ; with Bomans, that is with Celts who had become Eomans, the Frank had to fight only at that one stage when he won the Eoman remnant of Syagrius. And there again we are followed by the thought whether, at Constantinople at least, Syagrius was not held for a tyrant and Chlodowig for a loyal officer of Augustus. In truth Gaul is what it now is, Britain is what it now is, because there was no day on Gaulish soil like the day when Saxon Cerdic had to fall back for a moment before the might of British Arthur. Britain, forsaken by Eome, had fallen away from Borne. Terminus had withdrawn within the lands on his own side of the stream of Ocean. And Eome herself had presently to look the fact in the face ; she had to come as near to formally acknowledging the fact as the proud forms of Eoman diplomacy would allow. Another passage of Zosimos, thrust strangely into the narrative of a wholly different series of events, tells us again in a casual way that Honorius sent letters to the cities of Britain bidding them guard themselves* (4lo). If we can put any * In Zosimos, vi. 10, in the midst of the story of Alaric and Attains, we read suddenly ; 'Opcoplov 8e ypdfifiacri npos ras iv BpeTravlq Xpr](Tapevov noXfis v Kai arpaTiav Xdyov d^iav.fs 'la-iraviav re Koi TaWiav a>s 8ov\a>a6/iepus arpara 154 Western Europe in the Fifth Century, [iv. the ground is of course a guess to explain the frequent finding of Koraan coins ; but one would think that there must be some groundwork in fact for the space of nine years which the story makes between the time when Eoman Emperors ceased to rule in Britain and the time when the Komans themselves left Britain. But it is certainly hard to find in the year 418, the year of the twelfth con- sulship of Honorius and the eighth of the younger Theodosius, anything recorded in which we can recognize the minutely dated fact of our own Chronicler. But it is of the deepest importance that, throughout this story, not only in the English Chronicler so long after, but in the British Jeremiah of the next century, the Komans in Britain and out of Britain are looked /xeyaXoj eare^aXXev. Then we hear of Honorius designing to fly to Africa and generally of his relations to Attains, down to the death of Alaric. Then we get hack to Gaul and Britain. Constantino is, strangely enough, overcome by the "West-Goths under Atawulf, a clear confusion between two sets of events. Gaul seems to be thus looked on as recovered to the Empire ; but Britain was not won back, and remained under tyrants ; 6 tSv OvKnyoTdav a-rparos, rjyov- fievov (T(pivvfioi BpiTTcoves. Toaavrr) be ff TCdvhe rav idvaiu TroXvavBpoania (f)aiveTat oiiaa more dva tvclv eros Kara ttoKKovs ivBivhe jjieTavicrrdfievoi ^vv yvpai^l koi TfOKnv es ^pdyyovs XcopovtTiv. ol 8e avTovs evoiKi^ovaiv es yrjs ttjs (T(j)eTepas ttjv ipr}p.0Tepav 8oKov(rav eivai, koL ott avrov ttjv vrjdov •npocnroulcrBai (^aaiv. aycrre ap,eKei oil TToXXa irporepop 6 ^pdyyav fiacriXevs irn Trpecr^eia tuv ol emrrjdeicov Tivas irapa ^aaikea 'lovtrriviapov es Bv^dvriov crreiXas avdpas avTois eit ratv 'AyyiXatv ^vviirep^^e, (jiiKoTiiiovpevos as koI tj vrjaos f}8e irpbs avrov apxerai. There is something very strange in this account of the great numbers of the three nations in Britain and their overflow into Gaul. It must be some confused version of the Armorican migration, to which we are just coming; but, as Procopius tells the tale, the settlers may just as well have been Angles or Frisians as Britons. That two of the three nations were conquerors who drove out the third he gives no hint. It is possible that he may have mixed up the flight of the Britons with crossings of Saxon invaders from one side of the Channel to the other. The passage about the wall is stranger still. But nothing brings out better the main point, the distinction between the state of Britain and that of Gaul. M 162 Western Europe m the Fifth Century, [iv. the more memorable landings of Hengest, ^^EUe, and Cerdic. It is possible that Saint German, on his mission to Britain, may have come across warriors from some of those Teutonic colonies of unrecorded date which grew into the later kingdoms of East- Anglia, Deira, Bernicia, and Mercia*. The chief diiflculty is that the strong language of the annalist could hardly be used of a time when the lands which were to be Kent, Sussex, and Wessex were still British, lands whose fate would be much more likely to interest a continental writer than the lands further to the North. But these points do not immediately concern us. Our business now is rather to take the Eomans out of Britain than to bring the English into it. It is enough for us that, before the end of the reign of Honorius, before the end of the years with which we are specially concerned, the first land that bore the British name had ceased to be one of the lands to which decrees went forth from Caesar Augustus. The last land of the West to be won, it was the first to fall away. Between the conquest of Britain and its separation another part of the Empire had seen the conquest and the separation of Dacia. But Dacia had not fallen away in the same sense as Britain ; it had rather been found wise to give it up to an invading enemy. But now that the insular Britain had set the example, that example was followed by a land which soon came to be reckoned as a second Britain, if indeed it had not begun to put on that character already. At least from this time, most likely even from an earlier time, the * See Norman Conquest, i. 26. IV.] The Barbarian Invaders. 163 north-western peninsula of Gaul, balancing in its geographical position the south-western peninsula of Britain, was beginning to take to itself the name and the nature of a British land. We may believe that, even in the most flourishing days of Eoman dominion, this corner of Gaul, so well fitted, as the experience of later ages has shown, to be the last abiding-place of an ancient folk and an ancient speech, had kept traces of the tongue and the traditions of ancient days which were little dreamed of in Eomanized Lugdunum and Burdigala. Such relics of former times needed only to be strengthened, to be kept up by settlers from other lands where they had never died out, and there might again come into being, in this one corner of the West, a land as purely Celtic as though no part of Gaul beyond the Alps had ever been reckoned as a province of Eome. Such a strengthening was undoubtedly supplied by the im- migration of Britons from the insular Britain fleeing before the swords of Teutonic conquerors. Such, to quote no other writer, not to dwell on long-abiding tradition, is the distinct judgement of Einhard, the very clear assertion of a very clear-headed man*. That assertion it would need some strong contemporary evidence to set aside, and no such evidence is forthcoming. Indeed the saying of * See the distinct statement of Einhard, Ann. 786. Charles the Great " exercitum in Brittanniam cismarinam mittere constituit. Nam cum ah Anglis ac Saxonibus Brittannia insula fuisset invasa, magna pars incolarum ejus traiciens, in ultimis Gallise finibus Venetorum et Coriosolitarum regiones occupavit. Is populus a regibus Francorum subactus et tributarius factus," &c. M 2 164 Western Europe in the Fifth Century. [IT. Procopius about the crowds of Britons who yearly took refuge in the dominions of the Franks, is the saying of a writer with nothing hke the clearness of Einhard, but much nearer to the time ; and it looks the same way. That there was an Armorican migration, a migration from the greater Britain to the land which became the lesser, there can be no reasonable ground for doubting. The only question is as to its date. And we may be sure that it began early in the days of Teutonic conquest in the insular Britain. For in the sixth century the continental, the lesser Britain is distinctly marked as a land having a settled being of its own, with its own people, its own princes, quite apart from anything in the rest of Gaul. It is plain that, long before the end of the fifth century (468), there was a British people in this part of Gaul, Britons of the Loire, who played a considerable part in Gaulish affairs, who appear as the allies of the Koman and the Frank, as the enemies of the Goth and the Saxon, as spreading themselves inland as far as the land, perhaps as the city, of the Bituriges, and as driven out of that distant possession by the arms of the Gothic Euric. The Britons of Gaul, the Britons of the Loire, had their deeds recorded in annals which formed part of the materials both of the Goth Jordanis and of Gregory of Tours, and there is more than one reference to their presence in the writings of Sidonius of Auvergne*. At this date at least they are * The first mention of Britanny or Britons in Gregory of Tours is in the passage ii. 18, 19, which seems clearly to be copied from annals. The earliest fact about them is "Brittani de Bituricis IV.] The Barbarian Invaders. les a recognized people, one of the nations of Gaul, with a prince of their own, called of some a king, who a Gothis expulsi sunt, multis apud Dolensen vicum peremptis." This must not be taken for the Breton g-wasi-metropolis of Dol. The place is D^ols in Berry. This note comes among a series of entries from which we gather that Eomans, Franks, and Bretons — Wealas in short and those who were to become Wealas — were on one side, while Saxons and Goths are on the other. The date seems to be 468. Later notices in Gregory are common. The event recorded by Gregory is told at greater length by Jorda- nis, Getica, 45; "Anthemius imperator Brittonum solatia postulavit, quorum rex Kiotimus cum duodecim millibus veniens in Bituricas civitatem Oceano e navibus egressus susceptus est. Ad quod rex Vesegothorum Eurichus innumerum ductans advenit exercitum diu- que pugnans Eiotimum Brittonum regem, antequam Romani in ejus societatem conjungerentur, effagavit." Riotimus fled to Burgundy. Sidonius refers to all this in his letter to Vincentius about the affair of the prefect Arvandus (see Gibbon, eh. xxxvi. vol. vi. p. 198, ed. Milner). Arvandus was said to have dictated a treasonable letter to Euric ; " Hsec ad regem Gothorum carta videbatur emitti, pacem cum Grseco imperatore dissuadens, Britannos super Ligerim sitos impugnari oportere demonstrans, cum Burgundionibus jure gentium Gallias dividi debere." In iii. 9 we have a letter from Sidonius to E,iothamus, clearly the same as the Eiotimus of Jordanis, in which he speaks of one who " mancipia sua, Britannis clam soUicitantibus, deplorat." This is about 472. The phrase " Grgecus imperator" is odd. It is not a Eoman, though it is a Gothic, way of speaking of the Eastern colleague, and it has been thought not to mean Leo, but to be a sneering way of pointing at Anthemius, the Western Emperor sent by Leo. Yet Sidonius could write (Can. xxii. 30) in the Panegyric of this very Anthemius, " Salve, sceptorum columen, regina Orientis, Orbis Eoma tui, rerum mihi principe misso, Jam non Eoo solum veneranda Quiriti, Imperii sedes, sed plus pretiosa, quod exit Imperii genitrix." 166 Western Europe in the Fifth Century, [iv: played a part in the general politics of the land. This prince, Eiotimus by name, appears by that name in the story of Jordanis and he is numbered among the correspondents of the poet-bishop. And this people is found ranged alongside of the same allies and in face of the same enemies against whom we should look to find them ranged. The Wecdas of either world, Uiim-Wehli, Gal-Welsh, Br ef -Welsh, with their ally the Frank, still the faithful soldier of Rome, against the more-abiding Teutonism of the Goth and the still young barbaric life of the Saxon. The continental Britons could hardly have gained this position, if their first migration had happened after 449. We may rather believe that the migration of those who fled from the Saxon seax merely strengthened a British element which had already taken root on Gaulish soil. The beginnings of this earlier British settlement have been with much likelihood attributed to the days of the elder tyrant Maximus *. Their coming however made no im- mediate change in the provincial nomenclature of the Empire. The only Britain known to the Notitia Imperii is still the island ; the continental Britain, perhaps already so called in common speech, is not entered among the divisions of Gaul. The Lesser Britain was in no way distinguished from the Greater in either the older or the younger form of the Koman tongue, as in the tongue of the Saxon conqueror it has come to be by a slight difference in the form of the name. But in the great survey of the Empire the Lesser Britain is still hidden under * See Wietersheim, ii. 71, 166. I IV.] The Barbarian Invaders. i67 the general name of Armorica, a name then of far wider extent, taking in at least so much of Gaul as lay between the Seine and the Loire *. The Armorican name seems afterwards to have? shrunk Tip into a synonym for the Lesser Britain ; but we should be led astray if we put so narrow a sense upon the word even in the sixth century. At Constantinople, in the days of the Gothic war, the Armorican na.me took in those lands between Seine and Loire which became the kernel of Francia in the later sense, while the lesser Britain seems to have shared the fate of the greater, to have become the subject of the wildest fables, and to have been looked on, no longer as a peninsula of the mainland, but as another island like the land whose name it had taken f. * See the "Dux Tractus Armoricani," Notitia Imperii, iii, 106. His jurisdiction takes in Avranches, Coutances, and Rouen, and is further extended over five provinces, Aquitania Prima et Secunda, Senonia, Secunda Lugdunensis et Tertia. He too has "Littus Saxonicum." See Booking's Dissertation, v. 817 et seqq. t Whatever we make of Procopius' account of the 'Ap^opvxoi in Bell. Goth. i. 12, we cannot doubt, First, That the word is the same as Armorici or Aremorici, and, Secondly, That it is not meant to be confined to the peninsula of Britanny. His story is most likely a confused account of the conquest of the Roman land of Syagrius by Chlodowig. Its inhabitants did become one people with the Franks, which the Bretons have never done. So when Sidonius in the passage quoted above, p. 39, uses the old formal phrase of "Aremoricus tractus," he certainly does not mean Britanny only ; he is perhaps specially thinking of the lands that were to be Normandy and Anjou. Of the wonderful stories of Procopius I have spoken already. He clearly got his tales about Britain and Britanny from some quite different source from that where he found his notice of the 168 Western Europe in the Fifth Century, [iv. At the stage which we have now reached, when the insular Britain had fallen away from the dominion of Rome, the example of the islanders is said to have been followed by a considerable part of the Gaulish mainland. If we can accept the geography of our only informant, the spirit of independence spread far beyond the region which did in the end put on a character apart from the rest of Gaul. It was not merely the new continental Britain, but the whole Armorican land and other provinces besides, which asserted their independence of a power which could no longer defend them against barbarian inroads. They drove out the officers of the Eoman government and set up an independent state of their own *. We yearn to know the form of its constitution ; but such knowledge is denied us. We may gather from an incidental source that the revolution was not brought about without changes within as well as without, changes, it would seem, social as well as political. But from the same source it would also seem that the independence of Armorica, at least in the wider sense, was not lasting. A few years later (417-420), a poet of Southern Gaul could rejoice that Exuperantius, seem- ingly Prsefect of the Gauls, had brought back peace to the shores of Armorica and had restored the reign of law and freedom. The poet's standard of freedom may have been different from that of a large part of the inhabitants of Armorica. The effect of the * Zosimos, vi. 5 J 6 ^Apfiopixos anas Kai erepai TdKarcov eirapxiai, Bperavvoiis fiiiirjcTdnevoi, koto, top ktov (T(pas TJkevdepaaav rponov, e/c- jSaXXoucrai p.iv tovs 'P(op,aiovs ap)^ovTas, olKelov 8e Kar i^ovcriav noKlrevpa KadiiTTaaai. IV.] The Barbarian Invaders. i69 renewed rule of order was that men were no longer slaves to their own bondmen *. We need many more details before we can judge of the exact force of these words, whether they need imply such a revolution as had happened of old in the Etruscan Volsinii, when personal slaves actually set themselves in the seats of their masters. It may be only a poet's dark way of describing changes which put power into new hands, perhaps in the districts to which such a picture would apply, into the hands of the old natives of the land strengthened by the new settlers of kindred race. The whole subject is dark, and we can hardly get beyond probable guesses. We hear of further Armorican revolts, and, when the Franks made their way into central Gaul, we find the eastern part of Armorica in the wide sense, to a great extent a Eoman land, a land which clings to its Koman standing when Rome herself obeyed a barbarian king. But long before that time, as we have just seen, that part of Armorica which formed the continental Britain was a distinct land, with its own people and princes. The inference seems to be that the restora- tion of Roman power by Exuperantius was abiding, at * Kutilius Namatianus (i. 213) speaks of a kinsman of his " Cujus Aremoricas pater Exsuperantius oras Nunc postliminium pacis amare docet. Leges restituit libertatemque reducit, Et servos famulis non sinit esse suis." The date (see Tillemont, v. 659) of the poem is shown by an astronomical argument to be either 417 or 420. In Prosper Tiro (427?) we read, "II [Theodosii] in Galliis Exsuperantius praefectus a militibus interficitur." That may be the date of the undoing of his work. 170 Western Europe in the Fifth Century. least for some generations, in Armorica in the wider sense, but that in the peninsula which was becoming British, if the Eoman power was ever really again set up, it was cast off again in one of the later revolts. This restoration of the Eoman power in Armorica was, we can hardly doubt, connected with another change in the affairs of Gaul which brought two other Teutonic nations to the front in that land, and led to a lasting settlement of one of them which has affected geography ever since. The Franks, as the ruling, or indeed as a leading, people in Gaul, hardly come within the strict range of our present inquiry; the fascination of our own settlement in the second of our three great homes, a fascination the stronger because of the darkness in which our coming is enwrapped, has carried us on that head somewhat beyond our proper limits. But we have come in due order to the first settlements of the West-Goths and the Burgundians within the lands of the Empire, and to the events in the history of the Empire itself, the rise and fall of more than one tyrant, by which those settlements were accompanied. And before all it will bring before us one of the noblest forms in the whole history of our race, one of the men to whose lot it fell to shape the fates of ages, the kingly form of Atawulf the Goth. V. [WEST-GOTHS AND BUEGUNDIANS.] We have to deal now with the settlement on Gaulish ground of the West-Goths and of the Bur- gundians. The two names call up widely different thoughts. The Goths seem to belong wholly to the past ; the nation is gone ; the name is gone ; it is mere accident through which the people of Atawulf and the people of Gaiseric seem still to give kingly titles to the sovereigns of Northern Europe, But the Burgundian name is so familiar as the name of a land of modern Gaul, its intermediate history calls up associations so utterly alien to our present tale, that it is a little hard to picture to ourselves Burgundians, like Goths or Vandals or Saxons, as playing their part in the Wandering of the Nations. The Burgun- dian name seems in a manner out of place, almost as the English name does. Yet when we compare the history of the two nations, of the modern-sound- ing Burgundians and of the Goths who seem to belong to so much more distant an age, we shall find that, if the Goths were less abiding as a name — it may be doubted whether they were less abiding as a nation — they were much longer-lived as a political power. The Burgundians, as a people and kingdom, enjoyed little more than a century of independence, and that independence tempered by 172 Western Europe in the Fifth Century, [v. a degree of deference to the Empire unusual among the nations of Gaul. The Gothic dominion, on the other hand, was not swept away, even in Gaul, till the days of Saracen conquest in the West. Yet the name of Gothic has been for some ages swept away from Gauhsh soil, while the endless changes in the meaning of the word Burgundy, from the time of the first Burgundian settlement down to quite modern days, have been among the standing puzzles of geography. Both these nations now begin to play an important part in Gaulish history. The Goths show themselves for the first time on Gauhsh soil in the year that followed the fall of Con- stantine (412). Very short had been the time of peace, the time of union under the acknowledged princes of East and West. Perhaps within a twelvemonth of Con- stantine's overthrow, tyrants again show themselves in Gaul, tyrants who have, as before, to be put down by barbarian help ; but who show more distinctly than before how very largely their power rested on barbarian support. In the year that we have just spoken of we read in our annals that the West- Goths under Atawulf entered Gaul and that Jovinus assumed the purple at MaiDz, by the help of the Alan Goar and of the Burgundian Gunthachar. And in the following year (413) we read that the Burgun- dians obtained the part of Gaul next to the Ehine*. * The entries in Prosper are ; " 412. Gothi rege Athaulfo Gallias ingressi. 413. Burgundiones partem Gallise propinquantem Eheno obti- nuerunt. Jovinus et Sebastianus fratres in Gallia regno arrepto interempti." v.] West'Goths and Burgundians. 173 It must strike us at once that we have now come to regular political action in a region whose name we have as yet heard only as suffering passing ravage. One cannot doubt that the authority of Constantine had been acknowledged throughout Eastern Glaul. That would be pretty well shown by his being acknowledged at once at Trier and at Aries ; but Trier is the only point north of the Ehoneland where we see distinct traces of him. It is very hard to keep ourselves from already speaking of that land as Bur- gundy, though the events with which we are now concerned are enough to show how much such a name would be before the time. We have come, not to the first of all the Burgundies in the world, but to the first Burgundy within the bounds of Gaul. And that Burgundy finds itself, not on the lower Bhone, but on the middle Rhine. The centre of action is at Mainz, a city of which we heard as grievously suffering in the great invasion of five This makes rather too short work of two years. The other Prosper has ; " xviii. [Honorii]. Eursum alia prsedatio Galliarum, Gothis qui AJarico duce Romam ceperant, Alpes transgredientibus. xix. Jovinus tyrannidem post Constantinum invadit." Idatius (see above, p. iiS, note) only mentions the usurpation of Jovinus and Sebastian in 412. He does not mention the Goths till next year at Narbonne. Orosius in the Catalogue of tyrants (vii. 42) says; "Jovinus postea vir Galliarum nobilissimus in tyrannidem mox ut assurrexit cecidit. Sebastianus frater ejusdem hoc solum ut tyrannus more- retur elegit. Nam continuo ut est creatus occisus est." So Philostorgios, xii. 8 ; KaTo. Se tovs airovs xpoi/ovs 'Ico^iavos re iiraviarrf , els (pQopav CLTvia^r} Koi 2e/3acmai/os ahek^os avTOv rols Xa-ois inocpdaXixTja-as, 174 Western Europe in the Fifth Century, [v. years earlier ; but which may have risen from its ruins as easily as Trier. Of the actors in the move- ment, one we have heard of already. He is the Alan King Goar who had been won over to the Eoman service, but, like most of his fellows, was not specially scrupulous as to his strict allegiance to any one Roman prince over another*. His partner in setting up the new Augustus was the head of one of the two Teutonic nations who are now winning themselves homes in Gaul, Gunthachar the Burgundian. The name of his people has long been familiar in the history of the Empire, and a generation or more before Gunthachar they had played a great part in some of the wars on the Gaulish frontier. But, as there is no ground for the legend which claimed for them a Eoman origin, neither is there any ground for the behef of some scholars that they were, before the times with which we are dealing, already settled on Gaulish soil f. Burgundians also * The best account of the whole story comes from Olympiodoros (p- 454 et seqq.). The passage which at present concerns us is this ; 'lo^lvos ev Movi'SiaKa ttjs irepas Tepfiavias Kara a-rrovbrjv Toap rod AXavov Koi TovvTiapiov os (jivKap)(os exprniari^e tS)v BovpyovvTiovcav (j)v\upxos Tvpavvos dvT)yopevdr]. Goar is the Alan chief who joined the Eomans in 407. He is therefore not marked as (f)v\apxos, while Gunthachar, head of the Burgundian nation, is. These names in -char, as they are now written, are of course the same as our names in -here. There can be no doubt that the Movv^mkov of Olympiodoros is Moguntiacum, or Mainz. The form may possibly show that the name was already beginning to be shortened. Mediomatrici had fully sunk to Mettis in the course of the next century. t These points are discussed at great length by Albert Jahn, Geschichte der Burgundionen und Burgundiens, i. 237 et seqq. v.] West-Goths and Burgundians. 175 find tlieir place in some of the vaguer lists of the nations which took a part in the great movement of the year 406"'. But we have no distinct account of their share, if they had any, in the transactions of the last six years. There is nothing to show that they bore any part in the general harrying of Gaul ; they clearly had none in the partition of the lands. Whether they took any part in the wars of Con- stantino and his enemies depends on a single most confused passage*. On the whole we may safely say that, if the Burgundians took any part at all in the great events of those memorable years, it was not as chief actors, but in the way in which, in those days of wandering, stray detachments of almost any nations may get mixed up in the acts of any other. But if the Burgundians stood aloof from these greater movements, they might be thereby the better able to settle quietly, almost without notice, in some convenient region near to their older seats beyond the Ehine. Such a settlement they had clearly made by the year following the elevation of Jovinus (413), the year in which their occupation of part of Gaul The zeal of this author to refer to every writer from the earliest days to our own day who has said a word upon the subject sometimes makes it hard to dig out his own conclusions. But he seems to show with clearness that there is no reason to suppose any lasting settlement of the Burgundians in Gaul before that with which we are now concerned. * In the passage from Gregory quoted above, p. 1 16, Burgundians are mentioned in the army said there to have been led by Jovinus, that is really by Edobich. They may have been there along with the Franks and Alemans, or the mention of them may be owing to the same confusion as the mention of Jovinus. 176 Western Europe in the Fifth Century, [v. is recorded. And we cannot help connecting the two events which are brought so close together, the elevation of Jovinus and the Burgundian occupation. We may be sure that the Burgundian help which Jovinus received was paid for by the new Emperor with a formal grant of Gaulish territory to the Bur- gundian king and people. Jovinus and his power lasted but for a moment ; but the settlement of the Burgundians, or at least of their name, was for ever. Setting aside the north-eastern corner of Gaul that was held by the various tribes of Franks, the settle- ment of Gunthachar was the first Teutonic settlement in Gaul, as distinguished from mere harrying. It was the first establishment of a regular Teutonic kingdom, even if a kingdom dependent on the Empire, as distinguished from these mere planta- tions of prisoners or mercenaries as immediate sub- jects of Eome*. The march of Atawulf into Gaul, the elevation of Jovinus, the establishment of the Burgundians, were all made possible by the withdrawal of Constantius from Gaul after the fall of Constantino, whether he withdrew to rest in Italy or to fight in Spain. A new and in some points dark period now opens, a period in which it is not hard to follow the mere order of events, but in which the connexion of events and the working of causes baffle us at every step. Most hard of all is it to account for the course of Atawulf and his West-Goths. They now left Italy for Gaul. We know the fact ; we know the date ; at causes and motives we are left to guess. If * See Wietersheim, i. v.] West-Goths and Bur gimdians. \ii Atawulf designed any such territorial settlement in Italy as was before long carried out by his successor Wallia, his design at least remained a design that bore no fruit. But if the difficulties of the story are increased, a special interest is added to it by a certain vein of personal romance. The policy of princes and nations was just now largely influenced by the fact that the foremost men of two nations were rival and honourable suitors for the hand of the same bride. PJacidia, the daughter of Theo- dosius, the sister of Honorius, the captive of Alaric, was sought in marriage alike by the King of the West-Goths and by Constantius, already Count and conqueror and to be Consul and Emperor. It adds to the singularity of the case, while it does honour to every side of the character of the Gothic King, that the prize eagerly striven for by such mighty candidates was actually in the power of one of them. Placidia was still the captive of the Goths, but the King of the Goths was Atawulf. Her master was the man who spoke that memorable speech which traced out, which perhaps did much to rule, the coming history of the world. It was indeed a lucky chance for us which brought Orosius to hear the man of Narbonne, the stout soldier of the wars of Theodosius, tell to Jerome in his cell at Bethlehem the words which he had himself hearkened to in his own city in friendly talk with the Gothic King. That the words are truly the words of Atawulf we cannot doubt; the evidence is as good as evidence can be. The thoughts are far more likely to have sprung up freely in the mind of a Goth who wondered N 178 Western Europe in the Fifth Century. [V. at the new world around him than to have been devised by either a Koman monk or a Eoman senator of that day. We seize then our rare chance of listening to the inmost thoughts of one of the men who have indeed made history. We cannot dwell too often on those words so deep with meaning in which Atawulf declared the great change between his earlier and his later thoughts. He had once dreamed of overthrowing the Koman power, of changing Ro- mania into Gothia and placing Atawulf in the place of Caesar Augustus. The lesson of his life had taught him better. The rule of Rome was the rule of law ; by the law of Rome alone could the world be ruled ; he, the Gothic king, would wield the Gothic sword in the cause of Rome ; he would keep the nations under the shelter of the Roman peace and the obedience of the Roman law*. The man who could speak words like these is at once stamped as holding his place among the wisest and noblest of the world's heroes and sages. Atawulf, like Poly- bios, had his lot cast in one of the great turning- points of the world's history, and, like Polybios, he understood the memorable age in which he lived. Not all the lore, not all the experience of the friend of Philopoimen and of Scipio had taught him a clearer insight and a wider view than was revealed to the untutored warrior whom the Goths had heaved on the shield when Alaric was lost to them. For fourteen hundred years men have been consciously, or * The wouderful passage just at the end of the last book of Orosius' Histories has been quoted over and over again. It cannot be read too often. y.] West-Goths and Burgundians. 179 unconsciously, carrying out Atawulfs teaching, though not always in the lofty spirit of the man who taught the lesson. If we may take the Goth, the noblest form of the Teutonic family, as the representative of the whole household, we may say that all later history has been the carrying out of a process by which Romania has become Gothia without ceasing to be Romania, and Gothia has become Romania without ceasing to be Gothia. If not Atawulf, yet Charles became Rome's Caesar and Augustus without ceasing to be the Teutonic king that he was born to be. The Gothic sword wielded on behalf of the laws of Rome has been in truth the symbol of the whole history of the European world since the day when the foresight of Atawulf first made it so. The Goth then is the champion of Rome ; but we must remember that the champion of Rome is not necessarily the champion of Honorius. Atawulf no longer thought of placing himself in the seat of Caesar Augustus ; but he kept to himself the power of choosing between rival Caesars and Augusti. And he did not this time choose the one whom it would have been most easy for him to use as a puppet for his own purposes. The whole story is dark ; we are not told why Atawulf led his army into Gaul ; but we know that he carried with him a deposed Emperor and the sister of a reigning Emperor. An honourable lover, he would take Placidia to wife, but he would take her only with her own consent and that of her brother. A wise statesman, he was not insensible to the advantage which he might gain in negotiations with the brother from the fact that he had the sister in his N 2 180 Western Europe in the Fifth Century, [v. power. And if he had Placidia in his power, he had Attains also. The Emperor whom Alaric had set up and put down as was convenient at each particular moment, was still in the Gothic camp, " for the occa- sional purpose," as it has been inimitably put, " of acting the part of a musician or a monarch *." But Attalus could also play a third part, that of a coun- sellor to his Gothic patron ; only his career in this third character is less intelligible than in either of the other two. It is said to have been by his counsel that Atawulf, champion of Eome, having crossed into Gaul, acknowledged as the representative of Eome the prince who had been just set up by Alan and Burgundian helpf . Jovinus was indeed, so far as we can see, the acknowledged Emperor in so much of Gaul as admitted any Emperor at all \. All men had submitted to him, save only the praefect Dardanus, a puzzling character, the honoured correspondent of * Gibbon, ch. xxxi. + Our fullest narrative here comes from Olympiodoros, pp. 454 et seqq. He now says ; 7rp6s ov ['loiSIz/oj'J Trapayevtadai "AttoXos AoaovXcfiov irapaivei' koI Trapayeverat ap.a tov TrXrjdovs. X The Chronicle known as Prosper Tiro gives a clear summary of events, though more than one year seems to be rolled together ; "Jovinus tyrannidem post Constantinum invadit. Industria viri strenui qui solus tyranno non cessit, Dardani, Ataulfus, qui post Alaricum Gothis imperitabat, a societate Jovini avertitur. Salustius quoque et Sebastianus occisi. Valentia nobilissima Galliarum civitas a Gothis effringitur, ad quam se fugiens Jovinus contulerat." This division of the two first clauses, given in the note in Eoncalli (cf. Jahn, Burgundiens, i. 311), alone makes sense. As commonly stopped it would mean that Dardanus helped Jovinus. I I v.] West-Goths and Burgundians. isi contemporary saints, of Augustine and of Jerome, but whom a later saint, our own Sidonius, describes as uniting the characteristic sins of all the tyrants. The inconstancy of Constantine, the recklessness of Jovinus, the faithlessness of Gerontius, were blame- worthy each by itself; in Dardanus all were found together *. Yet the career of Dardanus at this time, if harsh and cruel, specially perhaps to the chosen land of Sidonius, is certainly not marked by reckless- ness or perfidy. He is at least faithful to his master, and serves him well alike in diplomacy and in warfare. That he should do all in his power to keep Atawulf on the side of Honorius was a matter of course ; why Attains should try to enlist him for Jovinus is less clear at first sight. Yet it may be that he had given up all hope of his own restoration to power, but still, as was likely enough, cherished a spite against Honorius and was inclined to support any enemy of his. And we can perhaps understand that Jovinus might at once be afraid of such an ally as Atawulf and might distrust the counsellor who had advised his march. But when we are told that Jovinus reproached Attains in riddles, we feel that we have got into the region of riddles ourselves f. Anyhow the advances of Atawulf to Jovinus were not received in a friendly * Of Dardanus bishop Sidonius ( Ep. v. 9) says ' cum in Constantino inconstantiam, in 3 oymo facilitatem, in Gerontio perfidiam, singula in singulis, omnia in Dardano crimina simul exsecrarentur." Au- gustine, ep. 87, calls him "illustrius mihi in caritate Christi quam in hujus sseculi dignitate," and ends " nee tua indignitas parvuli." This talk is only theological, + Olymp. ib. ; 'lo/Stvos aviaxm kin TJj 'ASaouXi' (TVfi- fidxov TTpoexfov. His enmity to Atawulf conies out at the very end of ZosimoSj vi. 13 J Avafievas e^tov irpos avrbv \^dpov\ 'ArdouX^os Zk. Tivos Tvpoka^ovaris dWoTpioTrjTos. So Olympiodoros (p. 449) after recording the captivity of Placidia and the elevation of Attalus (a. D. 410), adds, koL on 2dpov, koL avrov TotBov ovra, Koi nXrjdovs p-ev oXiyov endp)(ovTa (p-xpi- ydp 8i,aKocria>v fj TpioKocriiav 6 Xaos e^eruvero) aWws be Tjpco'iKov Tiva koX iv pd^ais dKarayavia-rov, tovtov on 'Pcopmoi fjTaipicravTO d' f'xdpos 'AXapiXdi ovra, aairovhov ix^pov 'A\dpixov enoirjcravTO, The enmity is here carried back from Atawulf to Alaric. Jordanis in his Getica makes Sarus a king. ;}i Olymp. p. 455 ; 2dpos rjv aTrooraf 'Ovapiov, on BeXXepiSoi', 6s ^v avTCi dopfariKos, dvaipedevros oi8e\s Xoyos ra jSaciXet Trjs dvaipefreati aide Tov (jiovov ylverai ttcnrpa^is. Jordanis (Komana, 321) makes Sarus a "EexGothorum." Sozomen v.] West-Goths and Burgundians. i83 giance to a prince who did no justice *, and betook himself to the obedience of Jovinus. It was only with a handful of men, eighteen or twenty in all, that Sarus made his way into Gaul. But his enemy was there with a force greater beyond measure. Atawulf met his enemy at the head of ten thousand Goths, where we are not told, but at some point doubtless between Narbonne and the Alps. Sarus, true to his old character, would neither flee nor sur- render. He fought against these overwhelming odds in a way worthy of the renown of his former exploits, till he was taken alive and put to death f. Atawulf was not likely to feel more kindly towards the man to whom Sarus had sought to join himself, nor was Jovinus likely to feel more kindly towards the man who had deprived him of such a helper as Sarus. (ix. 15, see Dahn, Kbnige des Germanen, v. 57) lias been thought to reckon him among Roman tyrants. His words are, dSoK^Tws dvaipovvrai 'lo/3tafdf re Koi Md^ifios ot Trpoeiprjfievoi rvpavvoi, Koi "Edpos kol SKKoi TrXeTcrrot iirX tovtocs ini^ovKevdavT^s rfj 'Ovapiov /SacrtXeia. Sarus, by trying to join Jovinus, certainly brought himself under this last head ; but he seems to be distinguished from the rvpawM. Sozomen in an earlier passage (ix. 9) calls him Sdpos rts ^dp^apoi t6 yivos, els aKpov to. ndkepia r](rKripevQS. * There is a certain likeness to Honorius in the picture of Stephen in the [Old English] Chronicle ; only Stephen could fight like Sarus himself. + Olymp. 455 ; 2dpos e/neXXe rrpos 'lojSii/oi/ napayevecrdaf dXX' 'ASciouX- ^05 TOVTO paOcav, TrpoviravTid^ei ^'^'"^"^ 8eKa a-vveirayopevos aTpaTiarrjv [(TTpaTKOToiv ?] €)(ovn av8pas irepl avrbv '2,dpv Tvpdvvav Ke(})aXas Koi elprjVTjv e^^"'' X On the former career of Heraclian, see, for Ms slaughter of Stilicho, Zosimos, v. 37 ; for his defence of Africa, Sozomen, ix. 8 ; Orosius, vii. 29. v.] West-Goths and Burgundtans. iss career, like the taking of Rome itself, lies apart from our main subject ; we are concerned with Heraclian simply as illustrating the abundance of the crop of tyrants, perhaps as showing the brood on a somewhat loftier scale than Constantino, Maximus, or Jovinus. But we have no need to dwell on his invasion of Italy, his fleet which men likened to that of Xerxes, his battle on Italian soil, of his own defeat, his flight to his own Africa, the slaughter at Carthage of his army; they are needful only to set before us the nature of the time in which Atawulf and Constantius played their part *. These dangerous rivals were now drawing nearer to each other's path. Atawulf may well have dreamed that the heads of Jovinus and Sebastian should be the price of the daughter of Theodosius, as the foreskins of the Philistines had been the price of the daughter of Saul. He may have as yet seen in Constantius at worst a hostile negotiator and not a hostile lover. A treaty was agreed to, oaths were exchanged, and the promise of tyrants' heads was before long fulfilled. The geography of the story is wholly dark ; we do not know how far south Jovinus and Sebastian had shown themselves in person. Most likely they were still on their way southwards, with * His revolt is recorded by Prosper, 413, and referred to by Olympiodoros, 457, where Constantius is said to have got rich out of the goods of Heraclian 6s rvpawlda neXerStv dprjpTjTai. But the grand flourish, comes from Orosius, vii. 42 ; " Nam habuisse tunc iii. M. naves dicitur, quern numerum nee apud Xerxem quidem prseclarum ilium Persarum regem nee Alexandrum magnum vel quemdam alium regis fuisse historise ferunt." 186 Western Europe in the Fifth Century, [v. Aries as their most likely goal, where their empire and their Hves were cut short. The head of Se- bastian was soon obtained, we" are not told where or how, and was duly sent to Honorius *. But before the head of Jovinus could follow it, the Gaulish Valentia, the city which had lately (l43) stood a siege on behalf of Constans against the forces of Gerontius, had now to stand another on behalf of the present tyrant against the power of Atawulf and the West- Goths. We have no details of the siege, but our single notice seems to point to a stout resistance followed by a storm. *' Valentia, the noble — hardly the noblest — city of the Gauls, where Jovinus had sought for shelter, was broken down by the Goths t." Dardanus, there seems reason to believe, stood with Atawulf before Valentia ; but there is no need to suppose that Constantius, whose eyes seem just now to have turned towards African affairs \, was at this time in Gaul. The next point of the Gothic march was Narbonne, which city the Gothic army entered in the time of vintage. It may be that the King and his Roman colleague were there before them. Any- * Olympiodoros, p. 455 ; av [Trpeo-jSea)!/] iTroa-TpeyjrdvTmu koi opxav fieaiTevaavToiv, '2e^aaTiavov fxev TrefnreTai, rw ^acriXei fj KecpaXrj. t See above, p. 180 [citing from Prosper Tiro]. X One or two things might suggest that Constantius was at this time, if not actually in Africa, yet engaged with African affairs. A law of 412 (Cod. Theod. vii. 18. 17), addressed to Constantius as " magister militum," has wholly to do with Africa. Orosius (vii. 42) rejoices how " his diebus, prsecipiente Honorio et adjuvante Constantio, pax et unitas per universam Afiicam ecclesise catholicse reddita est." Lastly, it appears from Olympiodoros, p. 457, that Constantius received the confiscated property of Heraclian. v.] West-Goths and Burgundians. isr how it seems to have been to Narbonne that Jovinus was brought as a captive. The old colony of Narbo, the colony of Mars, the city which gave its name to the whole Mediterranean land of Gaul, now becomes for a while the chief centre of our story. The first town of Gaul, it would seem, to be held by a Gothic king and a Gothic army, it remained the abiding seat of Gothic dominion north of the Pyrenees long after the Gothic name had passed away from the Loire and even from Garonne. A special creation of Rome, the first established seat of the Gaulish dominion of Rome, the commercial rival which went far for a while to supplant the ancient wealth and greatness of Messalia, Narbo Martins was still in the days of our kings and tyrants one of the foremost of Gaulish cities, but it does not now supply us with the same opportunities for tracing the memory of those times in still abiding monuments which we have so freely enjoyed at Arelate and Vienna, The balance between it and Messalia has been restored by physical changes. The haven of Messalia has been for ages growing greater and greater ; the haven of Narbo has passed away far more utterly than that of Arelate. The great mart of Roman trade in Gaul has now become wholly an inland town ; the stronghold of the Roman, the Goth, and the Saracen, has become an unwalled town; no works of Imperial days either crown its slight hill or watch over its narrow river ; memorials of those days are not lacking, but they are wholly of the kind which are treasured in museums, not of the kind which stand forth first of objects to catch the 188 Western Europe in the Fifth Century, [v. beholder's eye at Arelate and at Nemausus. The Narbo of the days of Atawulf and Placidia gathers round it so many interests that there is no city of which we should be better pleased to call up a living picture as it stood when the Gothic host entered its gates. But this is denied us. We cannot see the scene of the doom of Jovinus as we can see the scene of the doom of Constantine. For the captive of Valentia became the victim of Narbo ; Jovinus was slain by the hand of Dardanus. His head and the head of Sebastian went ia due form to Bavenna, perhaps to Carthage *. It might be well that Africa, * The words of Olympiodoros are ; 'lo/SIi/os Se iitto 'ASaovX^ou TTokiopKovnevos, iavTov endldacn, Koi TreynTTcrat eKeivos ra jSatriXei, ov avBein-Tjcras AdpSavos 6 enapxos dvaipei. This account needs to be explained and filled up from the other authorities. Thus it is from Prosper Tiro that we learn where it was that Jovinus was besieged ; " Valentia nobilissima Galliarum civitas a Gothis effringitur, ad quam se fugiens Jovinus contulerat." Then again the words that follow might make one think that Jovinus was sent alive to Eavenna, and that Dardanus killed him there. But Dardanus, the one loyal man in Gaul, was the inapxos of Honorius there and not at Eavenna. We may perhaps infer from the word TtepneTai that Atawulf designed to send Jovinus alive to Honorius, but that the act of Dardanus hindered him. We thus get the meaning of the entry of Idatius, ' Jovinus et Sebas- tianus oppressi ab Honorii ducibus Narbona interfecti sunt," followed by " Gothi Narbonam ingressi vindemise tempore." Whatever we say of Sebastian, Jovinus is not " oppressus " at Narbonne, but both are put to death at Narbonne. Nor is it needful to bring Constantius to Narbonne for that purpose; Dardanus is at any rate one " dux Honorii," and Atawulf might be called another. They may have been at Narbonne, though the whole Gothic army did not get there till a little later. About the sending of heads to Carthage see above, p. 127, n. v.] West-Goths and Burgundians. i89 restored to the allegiance of its lawful prince, should know that the arm of the lawful prince could strike in other provinces also. A third brother, Sallustius, shared the fate of Jovinus and Sebastian *. And we hear that the re-establishment of the authority of Honorius was accompanied by harsh doings in Auvergne, a land which, we may therefore infer, had been zealous for Jovinus. Many men of rank were put to death, among them Decimius Kusticus, prsefect of the Gauls under Constantino and again prsefect under Jovinus f. He had, it may be re- membered, supplanted Apollinaris, the grandfather of the saint and poet, who may therefore be conceived to have had no special love for him. Yet he was a chief man of Auvergne, he died among others of the chief men of Auvergne, by the act of the generals of Honorius, that is, we can hardly doubt, by the act of Dardanus. The man who slew Jovinus with his own hand was surely the man by whose bidding, perhaps also by whose hand, Decimius, Agrsetius, and the other Arvernian nobles met their end. In this slaughter wrought in his adopted country we at once see the ground for the excessive bitterness which Sidonius displays towards Dardanus. * Prosper Tiro (just before the entry of the fall of Valence) writes [as cited above], " Salustius quoque et Sebastianus occisi." Jahn, i. 313. + Gregory of Tours (ii. 9) quotes Eenatus Profuturus Frigi- redus as saying " Hisdem diebus praefectus tyrannorum Decimius Eusticus, Agrsetius ex primicerio notariorum Jovini, multique nobiles apud Arvernos capti a ducibus Honorianis crudeliter interempti sunt." 190 Western Europe in the Fifth Century. [V. The authority of Honorius was thus yet again acknowledged throughout the whole extent of Eoman Graul. And this time its acknowledgement was en- forced by the help of the Gothic sword. But the extent of Roman Gaul was lessened by the same process. The settlement of the Burgundians west of the Ehine was a fact which had to be dealt with. They had not as yet reached any of the lands to which they were to give their name in times to come. Dijon, Geneva, Vienne, Aries, were not as yet seats of Burgundian power. The first Burgundian land in Gaul was, as the chronicler says, in the regions near to the Ehine. It lay among those lands on the Gaulish side of the river which still specially kept the name of Germany. It was at Mainz that the Burgundian king set up his Emperor ; Worms was the traditional home of Burgundian kingship. It was then the land of Mainz, Worms, Speyer, stretch- ing southwards along the river into the land of Elsass, perhaps as far as Strassburg, perhaps not, which Jovinus had given over to his allies as the price of his diadem *. How was the land thus occupied affected by the overthrow of the power of Jovinus % It is plain that the Burgundians did not withdraw to their own homes. Gunthachar and his people appear again among the nations of Gaul twenty years later t. And though they then appear as enemies * Jahn (Geschichte der Burgundionen und Burgimdiens, i. 324) traces out the geography very clearly ; but I do not see why they may not (p. 329) have reached as far as Strassburg. t See the entries in Prosper and elsewhere under the year 435. They are fully discussed by Jahn, i. 341 et seqq. v.] West-Goths and Burgundians. ]9i of Eome, yet on the whole the Burgundians are found more closely connected with the Empire than any other of the Teutonic powers. A hundred j^ears and more after this time, when Emperors no longer reigned at Eome- or Eavenna, the Burgundian kings still acknowledged the supremacy of their successors at Constantinople, and ruled over their Eoman sub- jects under titles held by the grant of the Eoman Augustus *. Our authorities are utterly silent as to the whole matter, except as to the bare fact of the Burgundian settlement. But it seems impossible to avoid the conclusion that the counsellors of Honorius — was it the act of Atawulf or of Constantius '? — acknowledged a fact which it would be hard to undo, and that Gunthachar was commissioned as a lieutenant of the Empire in the lands which had been granted to him by Jovinust. His position would thus be that which was the formal position of so many bar- barian kings, the position of Atawulf and of Wallia, the position of Odowakar and the great Theodoric, perhaps of Chlodowig himself. Gunthachar was Bur- gundian king to his own Burgundians ; he was the partrician or proconsul of the Empire to the Eomans of the ceded land. We have no picture of the Bur- * See above all things the letters from Sigisraund to Anastasius, " Sigismundus rex domno imperatori/' among the epistles of Alcimus Avitus, 83, 84. One MS. adds emphatically; "ab Avito episcopo dictata est sub nomine domni Sigismundi regis ad imperatorem." But the king must have known what he was sending, and the letters are the letters of a vassel to his lord. Theodoric appears as " rector Italiae." t See Jahn, i. 315. 192 Western Europe m the Fifth Century, [v. gundians from the hand of Salvianus ; but it is quite in conformity with this position of their kings that their rule in Gaul seems to have been acknowledged as that which dealt out the least measure of hard- ship to the Koman inhabitants. The Burgundians dealt with the older people of the land, not as subjects, but as friends and brothers. There was not, at least not in the beginning, the unhappy difference of religion to sharpen the difference of nationality. The Burgundians were converted to Christianity, if not before their settlement in Gaul, at any rate while their settlement was still fresh. And they were con- verted to it in its Catholic form *. The Arianism of some of the later Burgundian kings is undoubted, and the belief of the kings was doubtless followed by at least part of the nation. Later in the century the strife between Arian and Catholic in the Burgundian kingdom becomes an important element in the politics of Gaul f. But Burgundian Arianism seems in no sort to have been, with Goths and Vandals, a national * The good character and catholic belief of the Burgundians comes from a passage in Orosius (vii. 32) earlier than his descrip- tion of the events with which we have been mainly concerned, a passage in which he mentions this mythical Eoman origin, and also the odd derivation of their name from hurgus. He adds, " Gallise hodieque testes sunt in quibus praesumpta possessione consistunt, quamvis providentia Dei omnes Christiani modo facti Catholica fide, nostrisque clericis quibus obedirent receptis blande mansuete innocenterque vivant, non quasi cum subjectis Gallis sed vere cum fratibus Christianis." t Tliis comes out largely in the writings both of Avitus and of Gregory of Tours. See above all the " Collatio Episcoporum " in Peiper's edition of Avitus, p. 161. ,v.] West-Goths and Burgundians. 193 faith adopted in the first moment of conversion, but a rare case of the falling away of Catholics to the heretical teaching. At any rate heresy never was universal ; the kingly house itself was never without Catholic members *. And, somewhat later than our present time, we hear of Burgundians beyond the Ehine still abiding in heathendom till, at the moment of a Hunnish inroad, they too entered the Catholic fold f. At that date Gunthachar was still reigning over the colony of his people in Gaul. The mention of Huns reminds us that he is one of the chosen heroes of Burgundian story, and that his name, like those of so many of the princes of these ages, found its way into the great Teutonic epic of the Nibelungs \. * On the religion of the Burgundians see Jahn, i. 122, 385. Chrotechildis, whom the French have made into Clotilde, is of course the great case of later Burgundian orthodoxy. + This comes from the passage of S6krates, vii. 30, discussed by Jahn, i. 337. The Burgundians come in; edvos ia-Ti ^ap^apov, irepav tov 'Ptjvov exoov rfjv oiKrjaiv' Bovpyov^laves KoKovvraL. It has been questioned whether this really belongs to Burgundians east of the Ehine so late as 430, and whether it is not a confusion with the earlier conversion of the Burgundians in Gaul. In strict geography the words irepav tov 'Ptjvov written at Constantinople ought to mean the left bank. But this would imply more accurate study of the map than Sokrates had a chance of. His knowledge would come from Western informants, who by "trans Ehenum" would mean the right bank. The Burgundians are converted Kara vovv Xafi^avovTes on 'Pcofiaiatv 6 Qeos ttrxvpais toIs (po^ovfievois avrbv ^or)6ei. Here we seem to have almost got back to the Hebrew notion of a national God; but one of the greatest facts in the history of the world lurks beneath the phrase. X See Jahn, i. 341 et al. . 194 Western Europe in the Fifth Century. [V. Jovinus had been raised to his short day of Empire by the joint help of Burgundians and Alans, of that branch of the Alans who, under Goar, had entered the Koman service when the mass of the nation went on to their harrying in Gaul and their settlement in Spain. But while we can in some sort trace the history of Guntachar and his Burgundians from this time onwards, we seem to lose sight of Goar and his Alans. We get a singular glimpse somewhat later of an Alan king and an Alan army in an alliance with the Goths of which they are weary* (415). We have to guess at the time and circumstances under which this union was formed ; but it would be nothing wonderful if, after Jovinus had yielded to Atawulf, the Alans were either constrained or found it prudent to' join the side of the conquerors. There seems to be no later mention of Goar or his people ; they must have been merged among some of the other settlers in Gaul, or else have joined their brethren in Spain, who were before long to be merged among the Suevians. No lasting settlement of mere Asiatic barbarians was to be made in the Cisleithan lands of Europe. But other Teutonic people besides the Burgundians were stirring at this time on the eastern frontiers. If the Burgundians had shown themselves at Mainz and Worms, the Franks were at work somewhat further to the north. By that name we must just now understand, not the Franks within the Ehine who were Eoman allies and had so lately done their duty in that character. The Franks, of whom we now get * This comes from the Eucharisticon of Paulinus of Pella, of whom more anon. v.] West-Goths and Burgundians. 195 a glimpse are the still untamed Franks who lived beyond the boundary stream and who had not 3-et obtained any settlement within the Empire. The head of Gaul, Augusta of the Treveri, which could hardly have recovered from its sack by Vandals, Alans, or Suevians, was now again taken, sacked, and burned by the Franks *. We have no distinct record of the several takings of Trier, four of which, it must be remembered, came within the memory of Salvianus. It may indeed have been after this sack [and not the earlier one] that the people of Trier drew on themselves the stern preacher's indignant rebuke for thinking, as soon as the enemy was gone, of the games of the circus before all things f. But the language of Salvian himself shows that even this last blow did not separate the capital of Valentinian and Maximus from the Empire. Whatever Trier suffered now, the damage must have been so far repaired that it lived on as a city and as a Eoman city. But the Franks, whether defending the Empire or * This again comes from Gregory's quotations of Kenatus Profuturus Frigeridus. After tlie passage already quoted in note, p. 104, comes " Treverorum civitas a Francis direpta incensaque est secunda irruptione." t Salvian, vi. 13, says "expugnata est quater urbs Gallorum opulentissima." But the most striking passage comes at vi. 15, and refers to the third taking of Trier; "ter continuatis ever- sionibus summa urbe Gallorum, cum omuis civitas combusta esset, malis et post excidia crescentibus." Then "pro summo deletes urbis remedio circenses ab imperatoribus postulabant." It is possible that Frigeridus and Salvian may reckon the sieges differ- ently and that they may refer to the same taking. If so, the Emperors must be Honorius and, for form's sake, Tbeodosius. O 2 196 Western Europe in the Fifth Century, [v. sacking its cities, do not as yet form the great centre of Gaulish history. At this time a higher interest gathers round the Burgundian, and a higher still round the Goth. At this moment, if Gunthachar was in form a Eoman oflS.cer, Atawulf was so yet more distinctly. He was so all the more because he was not, like Gunthachar, the ruler of any acknowledged territorial possessions. But his friendship with his formal overlord was not unbroken. The restoration of Placidia was wished for, most hkely by her brother, certainly by her Illyrian lover ; while his Gothic rival had assuredly no mind to give her up. He was the less likely to do so as long as her detention could be diplomatically justified, as long as the plighted price of her release, corn for the feeding of the landless Goths, and that in a year of hunger, still remained unpaid*. In the very year of the fall of Jovinus (413), Goths and Eomans are again in arms against one another. It can hardly be doubted that Atawulf now aimed at a great Gothic settlement in Southern Gaul, much like that which was afterwards carried out by his successor Wallia. We find him attacking several of the great cities of that region, and as entering into possession of some of them. We know not in what '^ The policy of Atawulf and Constantius is well marked out by Olympiodoros. We read now (p. 456), 'ASdovX^os iXKaKihiav d'lrrjTelTO Kara crnovSrjv fidXia-Ta KcDvaravTiov, bs varepov avrfj koI els ydfiovs iX^v^ev. aXka tuiv ivpos Abdov\(f)ov VTrocrx^cecnv nf) Tre paivofiepav, Kal fidkicrra r^j cnroTropmaf, oiire Taiirrjv ajreStSow, Koi els pdx^v ep-eXera ra T^s elpfjvrjs 8ia\vfa-6ai. So in a fragment of PMlostorgius, xii. 4, it can only be Constantius who appears as iXnidas rpecfxov, as avrbs KaTanoikfpr]v tls to bovvai, ovhev Se rfTTOV SixoXoyovvTcov, el Xn/Soiev TiXaKi8iav napao'^eiv, Koi 6 ^dp^apos ra opoia vTroKpivero. Then follows liow the barbarian TTpos MaaadKiav, noKiP ovrm Ka\ovfievT]v, irapayevofievos, SoXw TavTrjv Xa^tiv ^TTiCfV. evda jrkrjyels, Boi>t](f)aTiov tov yepvaioTarov jSaXdvros, fidikis top OavaTOP Bia(f)vyuip, els Tag olKelas vnfx^PW^ (TKrivas, ttjp ttoXiv iv evBvpia \i,nav Koi 61 enaipap Koi evos TOV ydp,op peXerap IlXaKiHias, KaporaPTiov ravTTjp anaiToiiPTos, ^apvTepas TrpovTfipep alTTjaeis, ipa 8ia t^p divoTv\iav evXoyop 86^rj ttjp KaTdaxeaip avTfjs irenoirjKepac. I It is when recording the consulship of Constantius that Olym- piodoros (p. 457) mentions the grant of Heraclian's property to him (see p. 185) and gives the personal description of him (see p. 112). § See Diet, of Christ. Biog. s. n., and Hodgkin, i. 831, n. v.] PVest-Goths and Burgundtans. 199 children *. She had to depart to make room for the august bride, and that was all. For now, at the be- ginning of the year (414), came that famous bride-ale of Narbonne, which it was fondly hoped would be far other than bale to many men, Gothic and Komanf. At the wedding of Atawulf the Gothic king took his place alongside of the daughter and sister of Em- perors, while a deposed Emperor led the choir in the wedding-song. The tale has been often told, and in modern Narbonne we shall seek in vain for any sign of the spot, for any trace of the house of Ingenuus which beheld the celebration of the marriage rites. Those rites were gone through in due order according to Eoman usage ; the bridegroom conformed to the national uses of the bride ; the stranger conformed to the national uses of the land in which he was sojourning %. Goth and Eoman rejoiced with equal * The iraidla, a eK rris Trporepas yavaiKos eTvy)(avev 'AdaovXos yap-iKets SpiKlais rfi UXaKibla avvelneTo, t^v yap oar paKivqv ^vaiv . . . t See the song of the bride-ale of Norwich, if it was Norwich, in the Chronicles, 1075. X Olympiodoros (457) and after him Mr. Hodgkin (i. 832) describe the wedding with much lively detail. I am most con- cerned with the first words, 'AdaovXcfxa aTrovBfj Kal vnodTjKT} Kavdidiavov 6 irpos HXaKLdlav avvTeXeiTat ydpos. The epithalamium is sung 'AttuXov irpSiTov (Ittovtos, and we read, avvTeXelrai 6 ydpos, nai^ovrav koi Xaipovrav 6p.ov tS>v re ^ap^dpav Koi rS>v iv avrols 'Papaiojv. All the ceremonies were Roman, and Atawulf wore a Roman dress. Idatius 200" Western Europe in the Fifth Century. [v. joy at the wedding which was in truth the symbolic wedding of Gothia and Eomania, the setting forth in a visible shape of the lofty schemes which were working in the mind of Atawulf. The Gothic King, soldier a,nd champion of Eome, was now the brother- in-law of Kome's elder Emperor. But in those days the soldier of Eome, without forsaking the service of Eome, might shift his obedience almost at pleasure from one Eoman prince to another. The prince who at the bride-ale had his turn as musician had again before the year was out his turn as monarch. So soon were the Imperial and royal allies, the Eoman and Gothic brothers-in-law, again at variance. Con- stantius had won back his influence with Honorius, and he was likely to be more wroth than ever with the rival who was in actual possession of the prize that had been so long sought for by both. So, wherever the power of the Goth reached, the Eome from whose cause he never fell away was to be repre- sees in the marriage tlie fulfilment of the prophecy of Daniel, " qui ait filiam regis Austri sociandam regi Aquilonis." Jordanis (Getica, 31) gives quite an unexpected place to the marriage, which he strangely fits in before the Gothic march into Gaul. In fact his story is utterly confused; but he, or Cassiodorus before him, quite understood the significance of the event. " Cujus [Honorii] germanam Placidiam Theodosii imperatoris ex altera uxore filiam ab urbe captivam abduxit [Atauulfus], quam tamen ob generis nobilitatem formseque pulcritudinem et integritatem casti- tatis attendens, inForo Juli^milise civitate suo matrimonio legitime copulavit, ut gentes hac societate comperta quasi adunatam Gothis rem publicam efficacius terrerentur, Honorioque Augusto quamvis opibus exhausto tamen jam quasi cognatum grato animo derelin- quens, Gallias tendit." v.] West-Goths and Burgundians. 201 sented by another chief. Attalus Augustus appears once more in Gaul under the patronage of Atawulf, as he had appeared for one moment in Italy under the patronage of Alaric * . Wherever the brother-in-law of Honorius had practical dominion in Gaul, there not Honorius but Attalus was Emperor. Of his acts in that character we know at least one. He bestowed a great oflSce on a man who was not eager for it, a Eoman of high position and descent, whose singular autobiography throws a good deal of light on these times and reveals to us some particular events of which we might otherwise never have heard. This is Paulinus, distinguished from many other bearers of his name with some of whom he has sometimes been confounded, as Paulinus of Pella. He notes with some pride that he was a native of the same city as Alexander, though in his day it had become needful to point out the royal seat of the Macedonian kings as being near their own crea- tion of Thessalonicaf. But he was not a man of Macedon, but of Gaul. His family was of Bourdeaux or perhaps of Bazas, and he was the grandson of Decimus Magnus Ausonius, poet and consul, some have thought through his son Hesperius, others through his daughter married to Thalassius ]. His * We here lose Olympiodoros for a season, but the new elevation of Attalus is recorded by Prosper, 414 ; " Attalus Gothorum consilio et prsesidio tyrannidem resumit in Galliis." t Paulinus, 24 ; "Editus ut Pellius, inter cunabula quondam Eegis Alexandri prope mcenia Thessalonices." { The point is fully argued in the Preface to the new edition of 202 Western Europe in the Fifth Century, [y. birth east of Hadria was owing to the official employ- ments of his father, who at the time of his birth, in the year in which the Goths crossed the Danube (376), held the vice-prsefectship of Macedonia*. An ap- pointment to the African proconsulship carried father and son to Carthage ; thence the child came by way of Eome to the city of his forefathers, to see his grandfather in the glories of a consulship enjoyed wholly on the banks of the Garonne f (379)- The record of his life, his studies, his pleasures, his affairs, the youthful errors which he confesses^, throws light, like every such record, on the life of the age in which he lived. The piece of detail most worthy of notice is that where, in his somewhat lumbering Latin hexameters, he tells us that he preferred the Greek authors to the Latin, seemingly — was it the result Symmachus, as before by Leipziger in his Dissertation (Breslau, 1858), p. 3. * Paulinus, 26 ; " Patre gerente vices illustris prsefecturse." + lb. 43-49 ; "Majorum in patriam tectisque advectus avitis, Burdigalam veni .... Tunc et avus primum illic fit mihi cognitus, anni Ejusdem consul, nostra trieteride prima." As the consulship of Ausonius was in 376, this fixes the dates for the whole life of PauKnus ; for he is very careful in always men- tioning his own age, though less so in giving the names of other people. X Paulinus, 156 et seqq. The distinction which he draws on this head, and the pointed contrast he makes between " culpa " and " crimen," are worth noting. He was (166) " Contentus domus illecebris famulantibus uti." v.] West-Goths and Burgundians. 203 of mere birth in Macedonian air % — that in his youth the Greek tongue came more familiar to him than the Latin, though in his later days he came to set more store by the tongue which was more native to a man of Bourdeaux, even if casually born at Pella *, From his thirty-first year (407), the year when enemies poured into the bowels of the Eoman realm f, his tale of his own life becomes for a while an important contemporary authority for the history of the time. It is from him that we learn the relations between the Goths and the Alans and the Gothic occupation * Paulinus, 72 ; "... Exacto primo post tempore lustri Dogmata Socratis et bellica plasmata Homeri, EiToresque legens cognoscere cogor Ulixis. Protimis ad libros etiam transire Maronis, Vix bene comperto jubeor sermone Latino, Colloquio Graiorum adsuefactis famulorum, Quos mibi jam longos ludorum vinxerat usus ; Unde labor puero, fateor, fuit hie mihi major, Eloquium librorum ignotce ajpjprehendere linguce.'' This is as curious as Orderic's seeming ignorance of French when he was taken from Shrewsbury to Saint Evroul. The child, born at Pella, is taken before he is three years old to Carthage, Eome, Bourdeaux ; yet Latin is " ignota lingua," + lb. 232. The date of the great invasion of Gaul is accurately marked ; " Sed transacta sevi post trina decennia nostri, Successit duplicis non felix cura laboris ; Publica quippe simul clade in commune dolenda, Hostibus infusis Eomani in viscera regni." His father dies, and he has a dispute with his brother about his mother's dowry. 204 Western Europe in the Fifth Century, [v. of Bourdeaux *. Of that citj Atawulf was still in possession when he again gave the diadem to Attains. Paulinus was one of its chief citizens; he won the favour of the Gothic king, and we shall presently see that he was on intimate terms with an Alan king, doubtless Goar. He obtained the special favour of having no Gothic guest quartered on his house f. But the presence of the strangers who had entered the city in peace was no great burthen ; the Goth knew the duties of a ruler, and the peace of King Atawulf may well have been better kept than the peace of Honorius Augustus. The virtual ruler now proclaimed the empire of the prince under whom, while Alaric lived, he had once held a high military command, and under whose renewed sove- reignty he doubtless rose higher still %. Our poet carefully points out that Attains had in truth no power, no revenue, no soldiers of his own ; he was a Eoman prince wholly by the grace of the Goth. Not out of love for the helpless tyrant, but out of mixed fear and regard for his Gothic master, Paulinus acknowledged the empire of Attains, and * See above, p. 197. t Paulinus, 282 ; " Otia nota domus specialia commoda plura, Omnibus heu nimium blandis magnisque referta Delitiis, cunctisque bonis in tempore duro, Hospite tunc et quae Gothico jam sola careret." X "When Alaric was " magister utriusque militise " {arpaTT]y6s eKarepas Bwdpfcas) under Attalus, Atawulf was " comes domesticorum equitum " {fjyefiuv rav inneav So/xeo-TtKwi/ KaXovfiepwv), Sozomen, ix. 8. Atawulf now doubtless held the higher place. v.] West-Goths and Burgundmns. 205 received from him the post of count of tlie private largesses. The post was not a pleasing one, as Attains had no revenues from which to be bountiful. Yet he submitted ; it was the will of the Goth ; the rule of the Goth was a fact in Aquitaine, and many Eomans had learned how to flourish under it *. The Augustus of Bourdeaux and Narbonne had thus a strong helper of another people ; but the Augustus of Eavenna had found a strong helper among his own people. Constantius was now the counsellor of Honorius ; and Constantius could act as well as counsel. The man who had lost Placidia * Paulinus, 292 ; " Addita majoris nova est quoque causa labor is, Ut me conquirens solatia vana tyrannus Attalus absentem casso onoraret honoris Nomine, privatse comitivam largitionis Dans mihi, quam sciret nullo consistere censu ; Jamque suo ipse etiam dedisset fidem regno, Solis quippe Gothis fretus male jam sibi notis Quos ad praesidium vitse prsesentis habere Non etiam imperii poterat per se nihil ipse, Aut opibus propriis aut ullo milite nixus. Unde ego non partes infirmi omnino tyranni, Sed Gothicam fateor pacem me esse secutnm, Quse tunc ipsorum consensu optata Gothorum, Paulo post aliis cessit mercede redempta, Nee penitenda manet cum jam in republica nostra Cernamus plures Gothico florere favore." It is to be hoped that Paulinus' Greek verses, if he made any, were better than his Latin. I do not profess to understand every word, and the last lines seem to refer to a later time when the Goths were in fall possession of Aquitaine. But the general sense must be much as I have given it in the text, and in any case we see a " Pax Gothica " supplanting the " Pax Eomana." 206 Western Europe in the Fifth Century. [V. was sent, with the new rank of Patrician, against the man who had won her. Constantius entered Gaul ; if Boniface, with only pubHc motives for action, had once proved too strong for Atawulf, much more might Constantius, with his own quarrel to stir him to yet further zea]. So it proved. Eoman mihtary science, when combined, as it was in the case of the new Patrician, with a stout heart and a strong arm and a private grudge to boot, proved too skilful for the simpler valour of the Goth. Boniface had saved Marseilles from Atawulf ; Constantius now succeeded in driving him out of all Gaul, and that, if we rightly understand the somewhat dark language of our authorities, without any actual fighting. The work seems to have been done by skilful combinations which cut off the Gothic host from the coast and all supplies. While Constantius kept his own head- quarters at Aries, he constrained Atawulf to depart from Narbonne and from the whole land *. Again * At this point Olympiodoros fails us. Prosper only speaks casually of " Grothi ad Hispanias migrantes." Idatius speaks of " Ataulfus a patricio Constantio pulsatus, ut relicta ISTarbona His- panias peteret." It is from Orosius that we get the nearest approach to an account of the campaign ; " Constantius comes apud Arelatum Gallise urbem subsistens magna rerum gerundarum industria Gothos a Narbona expulit, atque abire in Hispaniam cogit, interdicto praecipue et intercluso omni conatu navium et peregrinorum usu commerciorum Gothos." Jordanis (Getica, 31) has quite another story. Franks, Bur- gun dians, Vandals, Alans, all flee out of Gaul for fear of the Goths, and take refuge in Spain. Ata-wulf follows them, and at the end of three years is master of Gaul and Spain both. " Tali ergo casu Galliae Atauulfo patema venienti. Confirmato ergo Gothus regno in Galliis Spanorum casu coepit dolere, eosque deliberans a Vanda- v.] West-Goths and Burgundtans. 207 we get some details of great interest and singularity, from the poet of the Eucharisticon. The Goths had entered Bourdeaux in peace, and they had kept peace in it ; but when Atawulf was driven to leave Gaul, and his bidding came that his army was to leave the city, they did not leave it in peace. Attains was still acknowledged by Atawulf as Emperor, and, according to this theory, whatever Constantius might be doing at Aries, the people of Bourdeaux were harmless subjects of a prince in alliance with the Gothic King, and were therefore entitled to the full protection of the Gothic peace. But at such a moment rage and disappointment trampled on all such subtleties as this. The Goths were giving way before Koman enemies ; they had a Koman city in their power ; and, though they had entered it as friends and had dwelled in it as friends, yet, when they had to leave it against their wills, they dealt with it as if it had been taken by storm. Whether at the bidding of Atawulf or not, Bourdeaux was plundered and burned — burned that is doubtless in the way that cities were burned, and from which they so speedily recovered *. The count of the lorum cursibus eripere, suas opes Barcilona cum certis fidelibus derelictas plebeque imbelli, interiores Spanias introivit, ubi ssepe cum Vandalis decertans, tertio anno postquam Gallias Spaniasque domuisset, occubuit." Here must surely be a confusion between Atawulf and Wallia. * Paulinus, 308 ; " Tristia qugeque tamen perpessis antea multis Pars ego magna fui quorum privatus et ipse, Cunctis quippe bonis propriis patriseque superstes, Namque profecturi regis prsecepto Ataulfi, 208 Western Europe in the Fifth Century, [v. largesses fared no better than others, or rather worse. His privilege of having no Goth quartered in his house now turned to his loss. Not a few others found the horrors of the sack much lessened by the personal kindness of their Gothic guests ; Paulinus had no such friend to help him''\ His rank as the minister of an allied prince went for nothing ; his goods were plundered ; his house was burned ; he, his mother, and his household, escaped with the loss of all. They were fain to be thankful that they were spared in life and limb, and that the chaste Goths, true to their picture as drawn by Salvian, did no wrong to the honour of any of the female members of the company f. Nostra ex urbe Gothi fuerant qui in pace recepti, Non aliter nobis quam belli jure subactis Aspera quseque omni urbe irrogavere cremata." The " preefecturus rex Ataulfus" clearly points to the departure of Atawulf into Spain. In so confused a writer we may hope that the " prseceptus " referred only to the departure and not to the sack of Bourdeaux. * Paulinus, 286 (after the lines quoted, p. 204, note) ; " Quod post eventu cessit non sero sinistro : Nullo ut quippe domum speciali jure tuente Cederet in preedam populo permissa abeunti; Nam quosdam scimus summa humanitate Gothorum Hospitibus studuisse suis prodesse tuendis." t lb. 315 ; " In quam me inventum comitem tum principis ejus, Imperio cujus sociatos non sibi norant, Nudavere bonis simul omnibus, et genitricem Juxta meam mecum communi sorte subactos, Uno hoc se nobis credentes parcere captis. Quod nos immunes poena paterentur abire, v.] IVest-Goths and Burgundians. 209 All that Paulinus tells us about Bourdeaux is a distinct addition to our knowledge. But for him we should never have found out that the Gothic occupation under Atawulf stretched so far to the west. What follows is yet more remarkable, as it gives us our last glimpse on Gaulish soil of the vanishing race of the Alans, and of their relations towards the West-Goths. We are admitted to the personal acquaintance of an Alan king, who cannot fail to be that Goar of whom we have twice heard. He who had helped to set up Jovtnus was now the fellow- soldier of the patron and officer of Attains. Paulinus and his company, fleeing from Bourdeaux, made their way to Bazas, a city of Novempopulania, lying to the south-west of Bourdeaux, a little way ofl" the left bank of the Garonne. This too was for Paulinus an ancestral city; if his descent from Ausonius was through his daughter, it was most likely the home of the family of Thalassius *. How he and his party were able to enter is not clear; for he found a strange state of things within and without the town. Without it was besieged by a mixed host of Goths and Alans, Cunctorumque tamen comitum simul et famulorum, Eventum fuerant nostrum qusecumque secutse Illseso penitus, nullo adtemptante, pudore." He is glad that his married daughter had left the country already. * lb. 328 ; "Nee postrema tamen tolerati meta laboris Ista fuit nostri quam diximus ; illico namque Exacto laribus patriis tectisque crematis Obsidio hostilis vicina excepit in urbe Vasatis patria majorum et ipsa meorum." P 210 Western Europe in the Fifth Century. [V. minded clearly in their unwilling retreat to do all mischief that might be to the land which they were leaving. This is intelligible enough. Greater curio- sity is awakened by Paulinus' picture of the internal state of Bazas. The slaves were in revolt, and a few young men of free birth — Catilina has his likeness in many times and places — joined with them in a con- spiracy for the general slaughter of the nobles*. Are we, if the Bagaudse were the mere Jacquerie that they are commonly painted, to suppose civic as well as rural Bagaudsel But in any case it is not wonderful if the confusions that must have followed on successive barbaric invasions had stirred society to its lowest depths. Still servile conspiracies are seldom successful, and Bazas was not, any more than Armoricat, to be as Yolsinii or as Hayti. The revolt was put down w^ith the deaths of a few only of the guilty, and Paulinus is specially thankful to the Providence which allowed him the double satis- faction of forgiveness and of vengeance by causing the man who specially tried to murder him to be punished, but by the hand of another J. But the * Paulinus, 333 ; " Et gravior multo, circumfusa hostilitate, Factio servilis paucorum mixta furori Insano juvenum . . . licet ingenuorum, Armata in csedem specialem nobilitatis." + See above, pp. 64, 168, 169. X Paulinus, 337 ; " Quae tu, juste Deus, insonti a sanguine avertens, Illico paucorum sedasti morte reorum, Instantemque mihi specialem percussorem Me ignorante alio jussisti ultore perire.'' v.] West-Goths and Burgundians. 211 danger within the walls suggested to Paulinus a hazardous scheme of dealing with the besiegers who lay without them. He remembered his old friendship with the Alan king, and he knew that it was not of his own will that he and his people were serving with the Goths against the Eomans. He contrived to make his way without hindrance to the camp of Goar, and asked that by his help he and his family might be allowed to leave the town. To the amazement of Paulinus, the Alan answered that he could not help him, that he could not even allow him to go back into Bazas, unless he were himself admitted into the town. He knew that, if the pro- posed escape were allowed, the wrath of the Goths would be heavy against Paulinus, while he, Goar, was anxious for an opportunity of escaping from Gothic supremacy *. The discourse between Paulinus * lb. 343 ; " Sed mihi tarn subiti concusso sorte pericli Quo me intra urbem percelli posse videi'em, Subrepsit -fateor nimium trepido novus error ; Consilio, ut me prsesidio regis dudum mihi carl Cujus nos populus longa obsidione premebat, Urbe ab obsessa sperarem abscedere posse, Agmine carorum magno comitante meorum, Hac tamen hos nostros spe sollicitante paratus, Quod scirem, imperio gentis cogente Gothorum, Invitum regem populis incumbere nostris. Explorandi igitur studio digressus ab urbe, Ad regem intrepidus nullo obsistante petendi. Laetior ante tamen prima quam affarer amicum, Alloquio gratumque magis fore quam mihi rebar. Perscrutato autem ut potui interius viri voto P 2 212 Western Europe in the Fifth Century, [v. and the Alan king is a little less clearly explained than we could have wished, but we may gather that Goar expressed his wish to change from the Gothic to the Eoman side — or, if we like so to put it, from the side of Attains to that of Honorius — and that he went within the walls of Bazas in company with Paul in us, in order to make a treaty to that effect with the chief men of the city. These were doubt- less the members of the curia, the forefathers of those senatorial families who appear so often in the Gaulish history of the next century; in the days of Honorius political life seemed to be falling back into its original elements, and the senate of Bazas, like the senate of Home, might be called to act for itself in matters of peace and war*. The agreement, whatever its exact terms, was made and carried Prsesidium se posse mihi prsestare negavit Extra urbem posito, nee tutus jam sibi prodens Ut visum remeare aliter pateretur ad urbem, Ipse nisi mecum mox susciperetur in urbe, Gnarus quippe Gotbos rursum mihi dira minari, Seque ab ipsorum cupiens absolvere jure." * Paulinus, 364 ; " Obstupui fateor pavefactus conditione Proposita et nimio indicti terrore pericli, Sed miserante Deo afflictis qui semper ubique Imploratus adest, paulo post mente resumpta, Ipse licet trepidus et adbuc nutantis amici Consilium audacter studui pro me ipse fovere, Ardua dissuadens quse scirem omnino neganda, Praestanda et prius quam mox tentanda perurgens, Quse non sero probans vir prudens ipse secutus, Illico consultis per se primatibus urbis, ^ Rem cceptam accelerans una sub nocte peregit; "^'•3 West-Goths and Burgundians. 213 out. The wife and son of the Alan king were given to the Eomans as hostages for his good faith. He and his people from the enemies became the friends of the Romans of Bazas, and they undertook to guard the city which they had the day before been besieging. But it would seem that they guarded it only from without. An unarmed crowd of both sexes, Paulinus himself, now restored to his friends, being doubtless among them, thronged the walls of Bazas to see the unexpected deliverers by whom they were set free from fear of the Gothic enemy. Close under the walls was the Alan host which had streamed together from all quarters, the women thronging along with their armed husbands. Bazas was closely fenced in by barbarian arms and barbarian waggons, but they were there for the protection and not for the assault of the town. When the Goths saw their army lessened by so important a part of it as their late Alan allies, they deemed that all hope of taking Bazas had passed from them. They marched away, by what exact course it did not concern Paulinus to tell us ; but they must have made their way to join the army which had been driven to leave Narbonne. When the Gothic enemy was gone, the Alan deliverer did not long tarry; he too marched away, we know not whither ; it is the last that we hear of Goar, the last that we hear of his peof)le on Gaulish soil. Bazas was, for the moment at least, free from the presence alike of barbarian friends and Auxiliante Deo cujus jam munus habebat, Quo nobis populoque suo succurrere posset." Does the last line but one imply that Goar was a Christian ? 214 Western Europe m the Fifth Century. \y. of barbarian enemies *. We should be glad of other such like tales of Gaulish towns during these memor- able years. The Goth had now to withdraw, not only from Bourdeaux and Bazas, but from all Gaul ; * Paulinus, 377; " Concurrit pariter cunctis ab sedibus omnis Turba Alanaram armatis sociata maritis; Prima uxor regis Eomanis traditur obses, Adjuncto pariter regis caro quoque nato. Eeddor et ipse meis pactse inter fcedera pacis, Communi tanquam Gotbico salvatus ab boste. Yallanturque urbis pomoeria milite Alano, Acceptaque dataque fide certaret parato, Pro nobis nuper quos ipse obsederat bostis, Mira urbis faeies cujus magna undique muros Turba indiscreti sexus circumdat inermis. Subjecta exterius; muris hserentia nostris Agmina barbarica plaustris vallantur et armis, Qui se truncatam parte agminis baud mediocris, Circumjecta videns populorum turba Gothorum, Illico diffidens tuto se posse morari Hoste intestine subito in sua viscera verso, Nil tentare ausa ulterius properanter abire, Sponte sua legit cujus non sero secuti Exemplum et nostri quos diximus auxiliares Discessam fidem pacis servare parati Eomanis quoquo ipsos sors oblata tulisset." I can bardly tbink with Fauriel (i. 134) tbat the line (381) " Eeddor et ipse meis pactse inter fcedera pacis " means tbat Paulinus was given up as a bostage to the Alans. But Fauriel is the only writer who has used the witness of Paulinus at any length. Dahn has some references. I doubt whether Gibbon had actually seen the poem. The story of Paulinus goes on for forty years longer. It contains much interesting personal matter, and we shall have to turn to it once more for an illustration of general history. v.] West-Goths and Burgundians. 215 Atawulf had to give up all immediate hopes of dominion north of the Pyrenees. He and his Goths passed into Spain. He took with him his puppet Emperor and his Koman queen, the sister of the lawful Augustus of whom he was now again the enemy. This was a strange moment in the strangely chequered career of Placidia, As far as Komania was concerned, she had sunk into nothingness. She was a banished woman among a strange folk, a folk at war with her house, and if not formally at war with her country, yet kept from being so only because they had set up the enemy of her house as the nominal ruler of her country. As far as Gothia was concerned, she was the wife of a loving husband, the queen of a mighty king, the royal lady of what still seemed to be a loyal people. An exile from Kome and Kavenna, she had come to share a kingly throne, if only the throne of a barbarian, in the elder home of the Theodosian house. And presently it seemed as if the line of Theodosius and the line of the Baits were to be alike continued in a common representative of Gothia and Eomania. At Barcelona, the new seat of her husband's power, Placidia bore a son (415) who might look to be one day heaved on the shield as a Gothic king, and to wear the diadem of his childless uncle in the palace of Bavenna and on the capitol of Eome. The babe received the name of his Koman grandfather, and the birth of the youngest Theodosius seemed to open a way towards a reconciliation between the families and the nations of his parents. Both Atawulf and 216 Western Europe in the Fifth Century. [V. Placidia sought for peace and friendship with Honorins *. The claims of Attains were again for- gotten ; whether he was actually sent to Italy as a peace-offering is not quite clear ; anyhow he was cast aside by the Goths ; he was taken at sea by officers of Honorius, delivered up to Constantius, and kept for the judgement of the lawful Emperor. Between Honorius and the first Csesar the likeness is not great ; yet the fate of Attains has something in common with the fate of Vercingetorix. Each kept his captive to adorn his triumph ; for two years after this time (417) Honorius again entered Eome with the ancient ceremonies of a conqueror. Attains was, like Perseus, Jugurtha, or Tetricus, led before the triumphal chariot, but as he escaped the fate of Vercingetorix, he escaped also the harder fate of Jugurtha. Gains Julius could slay, but he did not mutilate, nor did he, like Gains Marius, condemn the victim to a lingering death. Honorius could * Olympiodoros, p. 258 ; *A6aovX0or, rexOevTos airS (K Trjs U\a- Ktoias TraiSoy, v ;^a)paff els yecopylav anoKKTjpcocrdiievoi, This can mean only the Aquitanian cession, for which Clinton's date is 418. Prospers order of events is on the other hand clear and careful, and it is hardly set aside by the Paschal Chronicle (i. 573, ed. Bonr.) which places ra emviKia to. Kara "AttoKov tov Tvpavvov in the consulship of Theodosius (VII) and Palladius, that is 416. Olympiodoros (452) mentions the fate of Attains casually, when recording his first elevation ; /xera xpovov nva fiacri\evei. eira Kadaipelrai^ Koi fjura ravra varepop enl 'Vd^evvav Trapayeyovas, Kal tovs TTJs Sextos ;)(eip6? 8aKTv\ovs dKpaTr]puicr6els i^op'ia TrapajrefineTai. The mutilation however was clearly done, not at Eavenna but at Kome among the ceremonies of the triumph. Philostorgius (xii.) gives some curious details ; vnep rod ^ruiaros dva^as \^Ovaipioi\, 6 TTju ■KpQ>Ti]U aiirm ^adfiiba top "AttoKop Bia^aiveip {meriOei .... Sf^ta? ;)(ei/>6f dneTefie roiis 8vo 8aKTv\ovs, wf 6 nep dvTixf'ip, o 8e Xtp^ai/os f-^ei rrjv KXrj(nv. Koi els AiTrdpap rfjv ptJctov tovtop (j)vyabevf(., fMTjbfvos aXXov KaKov Trpbs TTflpap KaraaTTjcras, dWa Koi ras els top jSi'ov ;(p6i'as 7rapa(Tx6fiepos. It must be remembered that Attalus had once threatened Honorius with banishment to an island, and that either he or Jovius — more likely Jovius — had added the threat of mutilation. Cf. Olymp. p. 452, with Zosimos, vi. 9. * Olymp. p. 458 ', KavaravTiov 8e koi tS)V nepl KcoparravTiop dvri- TrpoTTOPTap, e/iefei' anpuKTOs ^ tovtov koi UXaKibias 6pp.r}. v.] West-Goths and Burgundians. 219 Constantius might rejoice that the son and the husband of his beloved were taken away out of his path, and that without any crime on his part. The infant Theodosius, born to be the hope of two nations and the tie between two periods of the world's history, died to the deep grief of both his parents, and was buried in a casket of silver in a church outside the walls of Barcelona *. The death of the father (415) soon followed on the death of the child. Atawulf had foes of his own nation. Some told how a certain Eoforwulf could not endure the king's jeers at his small stature. Others told how in times past a king of some branch of the Gothic folk had been slain at Atawulf's bidding, how his faithful follower, Dubius by name, cherished vengeance for his slain lord, and one day gave Atawulf a deadly wound, as the King was going the round of his horses f. Atawulf * lb. 459; Tek€VTr}(TavTOS tov iraibos, irevdos fieya iroiovaiv eV avrm Ka\ BdnTovaiv iv XdpvaKi Karadevres apyvpa irpo t^s "RapKeKKcDvos ev Tivi evKTTjpia. This last phrase of bedehouse, is perhaps a sign of the probable paganism of 01ympiod6ros. Anyhow it is an early case of burial inside a church. The American use of " casket " seems exactly to translate Xdpva^. t This is the story in Jordanis, Getica, 31; " Occubuit gladio ilia perforata Euerwulfi, de cujus solitus erat ridere statura." Olympiodoros says ; dvaipeirai ^A8dov\(j)os, els eTriTi]pr]criv t5>v olKeiav tTTJTtov, ms f'i6c(TTo avTa, bcaTpi^ccv iv tw Ittttww. dvaipa. be avrbv eis tS)v olKeicov VotBcop, AovjSioy Tovvofia, e)(6pav iraXaiau Kaipo(f)vS.aKT](ras' TrdXai yap rfv 6 tovtov dfcrnoTr}!, p.oipas Fot^jk^s pfj^, vno 'A8aov\(jiov dvTjprjpevos' e^ ov Koi tov Aov^iov Xa/Swi/ 'ABdov\(j)os aKeiaxraTo' 6 8e T«5 TrpuTO) heairoTT] dpvvtov tov Sfirtpov fiie^prjaaTO. I agree With Mr. Hodgkin, i. 415, that the former lord cannot be Saras. But may not the Aov^ios of Olympiodoros and the Euerwulfus — clearly Eoforwulf, a grand wild beast name — of Jordanis be the same ? 220 Western Europe in the Fifth Century, [v. died, but not till he had given his dying charge to a brother. A childless widow among a strange folk, there was no longer a place for Placidia in the Gothic camp ; Atawulf could bring himself to bid that she should be given back to her own people, even at the risk of handing her over to the arms of his rival. And, to the last faithful to his mission, the Gothic King, the beginner of the world of modern Europe, died with a worthy bidding on his lips. The last words of Atawulf were the counsel that his Goths should ever dwell, if so it might be, in peace and friendship with the Eomans *. A king of a moment followed Atawulf, the successor whom Atawulf would have least wished to follow him. Deep in the next century men in other Teutonic kingdoms remarked on the little regard which the Goths showed to the claims of birth in disposing of their crown. They had an evil practice, so it seemed Aov/3ios does not sound like a Gothic name — unless one could fancy something like Duhha — and it might be a Latin nickname. Philostorgius (xii. 2) brings in the death of Atawulf with a singular phrase J ov tvoKv to fieaov koi ttoXXo. dpafiarovpyfja-as i^ opyrjs 'A8aov\(f)os vTTo Tivos Ta>v olneicov dnoa-cfidTTeTat. Orosius laments his being cut off while he was so earnestly striving for peace ; " Cumque eidem paci petendse atque offerendse studiosissime insisteret, apud Barchilonem Hispanise urbem dolo suorum, ut fertur, occisus est." Idatius, who places the death of Atawulf in 416, has a singular phrase; "per quemdam Gothum apud Barcelonam inter familiares fabulas jugulatur." Does that mean a friendly chat in the stable 1 * Olympiodoros, 459 ; TfXfvrau 8e 'A8dov\(f>os irpoairaTTe tw Wia d8e\(f)a uTTodovvai ttjv ItXaKidlav koi, eiri tvvaivTO, rfju 'Pwfjialtov (f)i\iav iavTois irepinoirjaaa-dai. This brother does not seem to be spoken of elsewhere. v.] West-Goths and Burgundtans. 221 in Frankish Gaul, of killing their kings, and setting up whom they would in their place *. Nobility of birth, the lofty stock of the Balti andthe Amali, was indeed respected among both Eastern and Western Goths; but in the succession to the Gothic crowns we see neither the Frankish rule which deemed that every son of a king had a right to be a king nor yet the English rule by which the nation chose for itself among the kingly house. Atawulf had left children and a brother ; but some strange passing influence gave for one week the cynehelm of the West-Goths to Sigeric the brother of the slain Sarus. A party favourable to his house and hostile to the house of Atawulf already had the upper hand for the moment t. We are told that Sigeric, no less than Atawulf, sought for peace with Eome ; but it is more certain that he treated the widowed sister of the Emperor with the deepest insult, Placidia, mourning for her child, mourning for her husband, was forced to walk undistinguished * Greg. Tur. iii. 30 ; " Sumpserant enim Gothi banc detestabilera consuetudiiiem, ut si quis eis de regibus non placuisset, gladio eum adpeterent, et qui libuisset animo, bunc sibi statuerent regem." t " Segericus rex a Gothis creatur," says Orosius ; the name is one of a familiar type. In Greek hands it changes a little ; as 01ympid6ros says ; biaboxos 6 tov '2dpov n8eX(f)os ^lyyepixos, arnovhfj fiaXkov Koi BvvaaTeia ij dxoXovdia koL vofia yiverai. This is again a curious use of the word bwaa-Teia. The fate of the children is emphatically marked by Olympiodoros ; to re TratS/a a sk ttjs erepas yvvaiKOS (Tvyxuvev 'ASaovX^o) yeyet>T]p.epa, avelXe, ^la rS>v tov iiruTKOTTov ^lyrjadpov koXttcov aTroamdcras. Then follows the treatment of Placidia — fts v^piv ' A8aov\(f)ov. See Hodgkin, i. 835. Jordania also (Getica, 31) records the election and death of Sigeric. In the chronicles his short dominion passes without notice. Wallia would seem to have immediately succeeded Atawulf. 222 Western Europe in the Fifth Century, [v. among a crowd of captives who were driven before the horse of Sigeric. Her step-children, the brood of the Sarmatian woman, were torn from the arms of the Bishop Sigesar, a Goth and an Arian, and slaughtered without mercy. But on the seventh day the murderer was himself slain, and a worthier choice was now made. Wallia, the wise and valiant, was heaved on the shield. We hear nothing of his descent or of his earlier deeds ; but what he did in a short reign showed that he had well learned the lesson of Atawulf. The great hindrance to peace, the personal rivalry between Atawulf and Constantius, was now at an end. We know nothing of the domestic relations of Wallia at the time, but he at least did not give Placidia a third suitor. She was given back to her brother, Constantius being the officer whose duty it was to receive her, and the Goths received the long-promised payment of corn in exchange *. The memory of her noble Goth * Olympiodoros says of Sigeric, enTO. ^^epas ap^as, avaipfirai, fjyefioav 8e rav rordoiv OvaXlas KaOiaTarai. In a later extract (p. 462) he records the restoration of Placidia with some details which are not found elsewhere, especially the payment of the wheat ; EvTrXowrtOf 6 fiayKTrpiavos wpos OvaXiav, os tS>v Tordav expTjudri^e (])vXap^os, aTTOOTeXAerat, e(f) a anovbds re deadai elprjviKas nal aTrdXa^e'iu TTjV nXoKiblav, 6 8e erot'/Ltcos 8e)^eTai, Koi airovTakevTOs avra (titov iv iwpidaiv i^rjKovTa, drroXveTai IlXaKtBia Trapatodelaa 'EvTrXovTlco irpbs Oviapiov TOP oiKeiov avTtjs dSeX^df. Orosius records the death of Sigeric and the election of Wallia, and adds, somewhat later ; " Pacem optimam cum Honorio im- peratore datis tutissimis obsidibus pepigit. Placidiam imperatoris sororem honorifice apud se honesteque habitam fratri reddidit." Philostorgius (xii. 4) records the payment of corn ; but he is v.] West-Goths and Burgundians. 223 lived in her heart, but she at last became the un- willing bride of the man who had so long waited for her. Constantius, count, patrician, consul, held for seven months the rank of Augustus, and even in that short space learned that the diadem did not bring happiness *. Placidia Augusta, mother of Honoria, saw her daughter, if not wedded, yet wooed, by a barbarian of another stamp, from her own Atawulf. Attila claimed her as his ; but at least the blood of Emperors did not actually mingle with the blood of the Hun. Mother of the last Valentinian, the last Eoman prince who could claim even female in rather a hurry about Aquitaine, as he is rather too late with Attalus; Ik tovtov [on the death of Atawulf] top ^dp^apov irpos 'Ovapiov anevdeTai ^(ovardvTLOs seemingly] Koi ttjv olKeiav dSeX^jji/, Koi top AttoKov T(5 /SatrtXei jrapaTidevTai avTo), cn/rrjirecrl re be^iovdevTts, Koi polpav Tiva rrjs tS>v TaXaToyu ^(Oipas els yeapyiav a.TTOKKrjpaxrap.evoi, Prosper's entry is a little mysterious; "Athaulfus a quodam suorum vulneratus interiit, regnumque ejus Wallia, jperemptis qui idem cwpere intelligehantur, invasit." The words in Italics are an odd way of pointing at Sigeric. Under the next year, 416, his entry is, "Placidiam Theodosii imperatoris filiam, quam Eomse Gothi ceperant, quamque Ataulfus conjugem habuerat, Wallia pacem Honorii expetens reddit, ejusque nuptias Constantius promeretur." In this last entry he also is in too great a hurry. * Idatius records the marriage ; but he places it in the same year with her restoration and with the death of Atawulf. Olympiodoros (p. 464) fixes it to the first day of the eleventh consulship of Honorius and second of Constantius, that is to January 1, 417. He brings out the unwillingness of Placidia very strongly ; TroXXa nev avrrj dvavevova-a TrapeaKevaae Koi t&v avrrjs opyt^eadat BepanovTOiv. reXos Se eV t^ t^s vnareias fjl^epo. dno j^eipos ravTijv 6 ^aaiKeiis koi d8f\v, eyx^ipiC^i Trapadovs KcivaTavTia), 224 Western Europe in the Fifth Century, [v. descent from the stock of Theodosius, she knew exile and she knew rule. Her memory still lives among the columns and mosaics of Eavenna ; it is but a few centuries since she was there in her bodily presence *. The reign of Wallia forms the last stage of our story (415-419). He was the direct founder of the Gothic power in Gaul ; he was the indirect founder of the more famous, but hardly in truth more memorable, Gothic power in Spain. At him and his works and the works of those who followed him we must at least look so far as to see the West-Gothic king- dom definitely change from a wandering people to an established territorial power. That power has, beyond all others, a threefold position. It was the Goth who was called, in the forefront of all the nations of Western Europe, to bear the assault of the Saracen, to bridge over the time when the strife was between the older and the newer life of Europe, between the elder power of Eome and the younger power of the Turk, and the time when both had to strive against wholly alien foes from Africa and Asia. Into those days it is not our present business to follow him ; but we must see this power established in the lands in which we have as yet seen him only as a wanderer. Of the three lands whose revolutions during some most eventful years we have under- taken to trace, Britain has passed away into a world * On the tomb of Placidia and her embalmed body, which sat there in Imperial state till late in the sixteenth century, see the various accounts of Eavenna and Hodgkin, i. 887, 888. v.] West-Goths and Burgundians. 225 of fable to come forth again into the world of history under a guise wholly unlike that of either of her fellows. In Spain and Gaul we have still to see some shadow of a return to settled order brought about by the sword of the West-Goth. VI. [WALLIA AND THE SETTLEMENT OF AQUITAINE.] Wallia, King of the West-Goths, is one of the men to whom we may be inclined to think that later ages have hardly done justice. The dispensing of his- toric fame is always liable to be somewhat accidental ; it was specially so in the times with which we are now dealing. Our actual narratives are so painfully meagre and piecemeal ; and it is so purely a matter of chance whether any other record of this or that prince or other leading man happens to be preserved. We can hardly fancy that the glory of the great Theodoric could ever have been wholly obscured or brought down to the level of an ordinary barbarian king. Yet from direct narrative we should know hardly anything of his Italian reign; we should know far more — that is, if any human effort could remember the story — of the endless intrigues in which he and his namesake figured while the East- Goth still abode on the eastern side of the Hadriatic. It is to the good luck that has preserved to us the whole mass of the state-papers of one of the most memorable of reigns that we owe that, though there are few kings whose reigns it would be harder to record in detail in the shape of annals, there are few whom we can more fully call up in every detail of Wallia and the Settlement of Aqmtame. 227 his internal government and his foreign policy. A lesser, but not contemptible bearer of his name, stands before ns as a living man and not a mere name in a chronicle, because our prelate and poet at Auvergne has by good luck drawn us the full- length portrait of a neighbour whom he dreaded but whom he could not help respecting*. Of Atawulf himself, of the clear sight with which he spanned the ages, of the keen grasp with which he learned the place in the world's history that was meant for him, we should have had but the faintest glimmerings, if a citizen of Narbonne had not told the tale to a saint at Bethlehem in the hearing of a pilgrim from Tarra- gona. On Wallia Orosius has bestowed only a few lines of narrative prose, while Sidonius has bestowed on him the chance gift of a casual mention, taking to be sure the shape of a few sounding hexameters, enough perhaps for a barbarian king, in the long panegyric with which he hails a short-lived Em- peror f. We can judge of him only by his acts, as they are recorded in the meagre materials out of which we have to patch his story. In them he stands forth as the worthy successor of Atawulf, as the man who carried on the work of Atawulf, as the Goth wielding his sword in the cause of Kome, as the prince who found a settled dwelling-place for his people, who established Gothia as a known part of the earth's surface, and that without wiping out Eomania to make room for it. Wallia waged many wars ; but he waged them all, according to the teaching of * See the picture of the second Theodoric in Sidonius, i. 2. t Sidonius, Carm. ii. 363, &c. Q2 228 Western Europe in the Fifth Century, [vi. Atawulf, as the soldier of the Empire. It is said by one who was writing while Wallia was acting that Wallia was chosen to the West- Gothic kingship in order that he might be the enemy of the Empire, but that he really showed himself its faithful friend *. We see here either a change of purpose in Wallia himself, like the change of purpose which we have seen in Atawulf, or else a difference of objects between Wallia and his people, in which the warlike instincts of the nation submitted in the end to the direction given to them by the King. It seems certain that the first enterprise which Wallia designed was a direct attack on the lands of the Empire, on a province which had been spared invasion for many years. Wallia proposed to forestall with his Goths the work which Gaiseric afterwards carried out with his Vandals, to pass the bounds of Europe and to found a Teutonic dominion in Africa which could have been founded only at the expense of Eome. It was the second time during these wars and settlements that the Goths, after so long a history as a nation ever moving by land, ven- tured, as they had once done so long before, to risk their fate on the waters of the Mediterranean. Alaric, flushed with the spoils of Eome, had designed to brave Skylla and Charybdis and to make Sicily, perhaps a Gothic dominion, perhaps only a field for * So at least says Orosius, vii. 43 ; " Segericus rex a Gothis creatus cum itidem judicio Dei ad pacem pronus esset, nihilominus a suis interfectus est. Deinde Vallia successit in regnum, ad hoc electus a Gothis ut pacem infringeret, ad hoc ordinatus a Deo ut pacem confirmaret." VI.] Wallia and the Settlement of Aquitaine. 229 Gothic plunder. The dangers of the strait had been too much for him, and the Gothic fleet was dashed in pieces *. So now Wallia, as the firstfruits of his reign, gathered a fleet to bear his warriors to their African conquest ; but his enterprise shared the fate of that of his predecessor ; another Gothic fleet was dashed in pieces by a mighty storm in the narrow sea between the pillars of H^rakl^s f. Then Wallia thought of the ill luck of Alaric ; he learned that destiny did not design him and his people for warfare with Kome or for warfare on the sea. He would keep himself to the element on which his people had done great things and would there act as the ally and soldier of Eome. He gladly listened to the advances of the Eoman envoy Euplutius, and the peace was concluded be- tween Wallia and the patrician Constantius \. He is * See the strange story of the image in 01ympiod6ros, 453. t Orosius, vii. 43; " Territus maxime judicio Dei, quia cum magna superiore abhinc anno Gothorum manus instructa armis navigiisque transireinAfricammolii'etur,in duodecim millibus passuumGaditani freti tempestate correpta, miserabili exitu perierat, memor etiam illius acceptge sub Alarico cladis, cum in Siciliam Gothi transire conati, in conspectu suorum miserabiliter arrepti et demersi sunt." X lb.; "Pacem optimam cum Honorio imperatore datis lectissimis obsidibus pepigit." So Idatius, who places it in the twenty-second year of Honorius, that is 416 ; " Cui [Ataulfo] succedens Wallia in regno cum patricio Constantio pace mox facta." So Prosper speaks of "Wallia pacem Honorii expetens" in the consulship of Theodosius VII. and Palladius ; that is also 416. This seems to be the right year for the peace and the restoration of Placidia. Only Prosper has put her second marriage too early, and Idatius has put the death of Atawulf too late. The name of the negotiator comes from Olympiodoros (462); 230 Western Europe in the Fifth Century, [vi. named as the actor ; and to him the peace was specially interesting, as it was the peace by which Placidia was at last restored to her countryman, and the way opened for her marriage with himself. In the wider view of things this peace — 'pax optima as it is called by the devout Orosius — was marked by the engagement made by the Groths to win back Spain to the obedience of the Empire from the dominion of the Vandals, Suevians, and Alans, by whom so large a part of it was still possessed. If the Yandals really had made a treaty with the Empire, it went for nothing when so promising an alliance offered itself, and one which so much better suited the personal objects of Constantius *. In observance of his new engagements, Wallia, during his short reign, waged many wars in Spain, but always to the at least nominal advantage of the Empire. Yet it is hard to believe, though our authority is the absolutely contemporary Orosius, who recorded the exploits of Wallia in his own land as the best news of the day, that either Wallia or the barbarian king generally sent messages to the Emperor, setting forth the state of things with "EvirXovTios 6 fiayiaTpiavos rrpos Ova\iav, os rav TotBohv e)(prjfidTi^e cbv- Xapxos, aTTOoreXXfTat, €0' o) (nrov8ds re decrdai elpriviKas Koi dnoXa^elv rfiv HXaKiBiav, k. t. X. Wallia is (}>v\apxos again ia 465 ; he was riyepLav in 659. * On this peace see Dahn, i. 145. Procopius (Bell. Vand. i. 3) makes an agreement between Honorius and Godegisl {rore ^vii^aiva. FoSiyto-KXa) 'Ovapios ecf) a 8fj ovk eVt Xu/i?; rrjs ■)(i)pas iBpvaovraiY But Godegisl had been killed long before; see above, p. 28. Orosius also seems to refer to something of the kind in words which will be quoted in the next note. VI.] Walha and the Settlement of Aquitaine. 231 a plainness of speech unusual among the princes of any age. Let Honorius, he is made to say, abide at peace, and take hostages from all ; in the war between him and the other barbarians, whichever side won, whichever side was overthrown, the loss was the loss of barbarians, the gain in any case would belong to the Emperor and the Eepublic*. With or without this clear understanding of what he was doing, Wallia set forth to bring back that part of Spain which was in the hands of the newly settled barbarian powers, that is to say, all the peninsula save the Eoman corner in the north-east and the few points which still held out elsewhere. Of these powers two were broken in pieces, that one most utterly which seemed most thoroughly out of place. Non-Aryan invaders were not to rule abidingly in Western Europe till they came in quite another shape from that of the half-Teutonized Turanian. * Idatius (Eoncalli, vol. i. p. 19) is emphatic on this head; ""Wallia rex Gothorum Eomani nominis causa intra Hispanias csedes magnas efficit barbarorum." Orosius, vii. 43, adds some strange details ; "EomansB securitati periculum suum obtulit [Wallia] ut adversus cseteras gentes quae per Hispanias consedissent sibi pugnaret, et Ro- mania vinceret; quamvis HalanorumcoeteriVanddlorum Suevorumque reges eodem nobiscum placito despecti [al. depecti] forent, mandantes imperatoriHonorio; Tu cum omnibus pacem habe omniumqueobsides accipe : nos nobiscum confligimus ; nobis perimus, tibi vincimus : immortali vero qusestu erit reipublicse tuse si utrique pereamus. Quis hsec crederet, nisi res doceret ? " How strictly contemporaiy Orosius was comes out strongly in the words that follow ; " Ita- que nunc quotidie apud Hispanias geri bella gentium et agi strages ex alterutro barbarorum crebris certisque nuntiis disci- mus, prsecipue Valliam Gothorum regem insistere patrandse paci ferunt." 232 Western Europe in the Fifth Century, [vi. At that moment the Alans were the greatest power of central Spain, cut off indeed from the straits and from the Pyrenees, but stretching from the Ocean to the inner sea, from the haven of Odysseus to the haven of Asdrubal. Their dominion has on the map almost the air of a kingdom of Castile with a kingdom of Portugal added. To the north-west the Suevians under Hermenrich and the AsdingianYandals under Guntheric between them held the Gallician horn of Europe ; south of Anas the Silingian Yandals held the land of Bsetica, the land to which some have thought that they gave their name. The moun- tainous frontier of Gaul, and the land on either side of Ebro, the land of Tarraco and Csesaraugusta, was still held, either by the Boman or by those whom neither Koman nor Saracen could fully overcome. To enlarge this Imperial remnant at the cost of all the settlers of the last few years, the sword of Wallia was now drawn. The Alans, under their king Atax, were so utterly overthrown that they ceased to be a people and a kingdom ; the remnant that escaped from the Goth commended themselves to the Yandal King Guntheric, and lost themselves in the greater mass of his people. Here the report of the contemporary annalist is borne out by later history. The Alans now vanish from Spanish history. It is more startling when the same author says that the Silingian Vandals in Bsetica were all cut off by King Wallia. For that is just the corner of Spain in which the Vandal power lived on till its voluntary departure beyond the straits, and where it showed not a little vigour VI.] WalUa and the Settlement of Aquitaine. 233 a few years after this time. A contemporary Spaniard must be supposed to know the geography of his own country; and, if we allow for some- what of exaggeration, if we grant the survival of a remnant which was capable of again becoming a great people by the immigration of a kindred folk, the statement becomes intelligible. We have an entry a little earlier by which it seems that a Vandal king, Frithbald by name, was taken and sent as a trophy to Honorius. But as he was taken by craft without dealing of handstrokes, we may be tempted to guess that those who took him were Eomans rather than Goths *. Anyhow, as long as * Our accounts of these wars are very meagre. The clearest account is that of Idatius. After the entry in note, p. 229, placed in 416, come the words " Alanis et Wandalis Silingis in Lusitania et Bsetica sedentibus adversatur [Wallia]." Then comes under the same year, only seemingly with some doubt as to the manuscript authority ; "Fredbalum regem gentis Wandalorum sine uUo certamine in- geniose captum ad imperatorem Honorium destinat." (The nominative seems to be Constantius, whose marriage comes just before. Only that was certainly in the next year, 417. Honorio XI. et Constantio II. Coss.) Then comes the entry quoted in note, p. 231, under the year 417 (Honorius XXIY.). Then in the same year; " Wandali Silingi in Baetica per Walliam regem omnes extincti. " Alani qui "Wandalis et Suevis potentabantur, adeo csesi sunt a Gothis, ut, exstincto Addace rege ipsorum, pauci qui superfuerant, abolito regni nomine, Gunderici regis Wandalorum qui in Gallsecia resederat, se patrocinio subjugarent." One finds less help than one looked for in Pallmann, Geschichte der Vblkerwanderung, i. 259, and Dahn, Kbnige der Germanen, V. 56. Wietersheim (Band iv. 178-180) brings out more points in a few words. 234 Western Europe tn the Fifth Century, [vi. Wallia remained in Spain, the Gothic sword, wielded, according to the bidding of Atawulf, in the cause of Eome, went on and conquered, and the other barbarian settlers in the land were cut short before the joint advance of Goth and Roman. It must be borne in mind throughout the story that all that Wallia did was done in the name of Bome ; all the conquests that he won were held to be restored to the dominion of her Emperor. The dominion, whether of Honorius or of Wallia, seems to have been fully established in western and central Spain, when, it is hard to say from what motive, the loyal conqueror was taken away from his career of victory in the peninsula to enjoy the reward of his labours in a magnificent grant on the other side of the Alps. The West- Goths, before long to be so famous a power in Spain, turned away from the land of which they had been allowed a glimpse and no more *. Their kings were presently to reign on the Garonne ; it was not for several generations that they were to reign on the Tagus. Spain was left to be torn in pieces by the warfare of the barbarians with one another, and by the struggles of the Roman officers against the Vandals, who became great again as soon as Wallia's back was turned. The next year (419), when Wallia was no more, we read of a fierce strife between * Idatius (418) seems pointedly to mark how the work of "Wallia in Spain was cut short ; " Gothi, intermisso certamine quod agehant, per Constantium ad Gallias revocati, sedes in Aquitanica a Tolosa usque ad Oceanum acceperunt." VI.] Wallia and the Settlement of Aquitaine. 235 Vandals and Suevians — Alans have passed away — in their Gallician corner. The Suevians, destined to keep their place in that region for many generations, had the upper hand; and the remnant, under the guidance of the Eoman Count Asterius, joined their brethren in Bsetica (420), leaving the Suevians successors to the great Alan dominion in central Spain, which they were to hold till successors of Wallia came back again *. This Yandal migration from Gallicia strengthened the feeble remnant of the nation which had been left in the south, and the Vandals again became a powerful people in Spain (422) under the dynasty which had ruled in their short-lived Gallician territory. The Vandals of Bsetica soon called for a Koman force to be sent against them under Castinus, the magister militum, and that Boman force did not go without Gothic help. And if our tale is told truly, here was a case of that Gothic faithlessness of which it startles us to hear in the declamation of Salvian. The besieged Vandals — we are not told the place of the siege — pressed by hunger, were on the point of surrender, when the Eoman commander unwisely risked a pitched battle, and forsaken by his allies — so the Eoman or Spanish annalist tells us — made his way back as a beaten man to Tarragona f. * Idatius, 420 ; " Wandali, Suevorum obsidione diiaissa, instante AsterioHispaniarum comite, et subvicario Maurocello,aliquantis Bra-r cera in exitu suo occisis, relicta Gallsecia ad Beeticam transierunt." t lb. 422 ; " Castinus magister militum cum magna manu et auxiliis Gothorum bellum in Bsetica Wandalis infert. quos jam ad inopiam vi obsidionis arctaret, adeo ut se tradere jam pararent. 236 Western Europe in the Fifth Century, [vi. Three years later the Vandals of Bsetica had again grown to such power that Guntheric could make Himself master of two of the great cities of Spain, New Carthage on the eastern sea, and His- palis, Seville, on her great river flowing westward to the Ocean *. Either these great cities had held out all along, or they had been won back for Rome by Wallia. Seville now passed aw^ay from the Eoman power for ever ; New Carthage was again to become a possession of the Republic when the conquests of Justinian again stretched its dominion to the Ocean. But the later Vandal history is no part of our story, which, at this its last stage, gathers mainly round the West-Goths. The Gothic allies who failed Castinus must have been fetched from the land which was by this time occupied by the Gothic feudatories — it is hard to keep ourselves from the use of that and of kindred words — of the Empire in Gaul. There now was the great seat of Gothic power, the first land within the western border of Eome held by any Gothic people as an established territorial possession. The West-Goths and their king received the second Aquitaine to dwell in and to till t- It was not a land that was new to them. inconsulte publico certamine confligens, auxiliorum fraude deceptus ad Terraconam victus effugit." * Idatius, 424 ; " Wandali Baliaricas insulas prsedantur, deinde, Carthagine Spartaria et Hispali eversa et Hispaniis deprsedatis, Mauritaniani invadunt." + Proaper (419) describes the grant with some accuracy; " Con- stantius patricius pacem firmat cum Wallia, data ei ad habitandam secunda Aquitania et quibusdam civitatibiis confinibus provin- VI.] Wallia and the Settlement of Aquitaine. 237 They had appeared, as friends and as enemies, before more than one of its cities in the days when Atawulf marched through Gaul as the soldier of Attains. The settlement which, we may be sure, had been then designed by Atawulf, but which had been hindered by the successes of Constantius, became a real and memorable fact under Wallia. The land now (418) became the possession of the West-Goths and their king. It was given them to dwell in, to dwell in nominally as subjects and sol- diers of the Empire, in truth to make the land that was thus granted to them the kernel of a great, and for those days abiding Gothic power. The second Aquitaine, the land that lies between the mouths of the two mighty Ocean rivers of Gaul, and which is watered by them and their great tributaries, was a noble prize indeed. Its renowned cities call up the memories of many a stirring day in the later history of our own people, and they had already begun to win their place in the annals of the world and of the Church. Poitiers, on her peninsula, with the monuments of unrecorded days looking down from the other side, steep and woody, of her encircling stream — not yet the city of courts and ciarum." Philostorgius (xii. 4) witnesses that his fellow-sectaries were to till the ground ; (xolpdv nva rrji twv TaKaTav xapa^ ks yecopyiau anoKKrfpaxTaiievoi, The tilling of the ground by the Goths is referred to also by Merobaudes, Frag. viii. 1 3 ; " Csesareoque diu manus obluctata labori Sustinet acceptas nostro sub consule leges, Et quamvis Geticis sulcum confundat aratris Barbara vicinae refugit consortia gentis." 238 Western Europe in the Fifth Century. [Vi. minstrels, not yet the city of the holy Eadegund, but already the city of the most famous of the Hilaries. The Arian Goth when he entered her gates, entered as master into the home of the champion of orthodoxy, yet not minded, we may believe, to disturb his successors in the baptistery, well nigh without fellow beyond the mountains, which has outlived the church of Hilary's own worship, nor yet in his basilica which had already doubtless in some earlier shape crowned the hill from which the beacon-fire was to flash up to heaven, when, within a hundred years from Wallia's entry, the Frankish convert to the faith of Hilary * marched to break down the Arian dominion in the Aquitanian land. The Goth entered too a second time within the gates of Burdigala, where Atawulf had entered as an ally, and whence his host had marched as destroyers. He now held the city by the estuary of Ocean f, its amphitheatre doubtless still standing whole, perhaps for Wallia, like Theodoric, to wonder at the sports that pleased his Eoman subjects. Besides these more famous cities, the second Aquitaine took in also Saintes and Angouleme and Agen ; it took in the Petracorian city \ by the Dordogne, not yet the borough of Saint Fronto on * Greg. Tur. ii. 37. See Sketches of French Travel, " The House of Hilary," and " Churches of Poitiers." t " Burdigalam veni cujus speciosa Garumna Mcenibus Oceani refluas maris invehit undas, Navigeram per portam quae portum spatiosum Haec etiam maris spatiosa includit in urbe." Paulini Euch. 44. His grandfather before him had been there. X See Hist. Essays, vol. iv. p. 131, for notices of Perigueux, &c. VL] Wallia and the Settlement of Aquitaine. 239 his hill, but still the Vesonna of the Eoman, looking up across the stream to the older home of the Gaul ; so much at least of Yesonna as, in the years of havoc that had just gone by, had been fenced in with the mighty stones of earlier buildings, to guard at least an inner remnant from the flood of barbarian ravage. The Goth entered on the walls, the gates, the amphitheatre, the temple outside the narrowed en- closure, its mighty round tower still perhaps clothed with its marbles and surrounded by its columns, or perhaps standing as a fresh-made ruin, raw and gaping, to tell of the passage of beleaguering Yandals, Alans, or Suevians. He held the land of hills and streams and dwellings deftly hollowed in the hill- sides, dwellings of races whose record had passed away before the coming of the Goth or the coming of the Gaul. But the fief of Wallia and his people was not shut in within the bounds of the second Aquitaine ; it stretched into the first. The head of Aquitaine, Avaricum, Bituriges, Bourges, one day to be the seat of Aquitanian kings and Aquitanian patriarchSj formed no part of the first Gaulish heritage of the Goth. The Arvernian land and city, the land and city where the fellowship of Sidonius and Gregory has made us more at home than on any other spot of Gaulish soil, was one of the latest of Gothic conquests, and never knew Wallia as its master. But within the bounds of the first Aquitaine he ruled over the Eutenian city, one day to be Ehodez with its famous tower, over the land and city of the Cadurci, Cahors of evil name, with her peninsula and her bridge, where 240 Western Europe m the Fifth Century, [vi. Eoman walls still guard the memory of men who fought well to save Gaul from the Eoman power- he ruled over the Lemorican and the Albigensian cities, each already seated by its river, each doubt- less already with its great church in its freshness displacing some holy place of pagan days, but whose chief renown was to come in later times. But if the new land of the Goth did not take in the whole of the first Aquitaine, it overleaped the bounds of Aquitaine in the widest sense. It stretched into the older Roman land of Narbo. The city which had seen the wedding of Atawulf and Placidia was not at once to pass into the hands of Atawulf 's successor ; but the Goth now won the city from which his kings were presently to reign on both sides of the Pyrenees. Tolosa, whence Ceepio carried off, as men deemed, the gold of Brennus, Tolosa, seated on no hill-top, but planted by the fierce stream of the broad Garonne, and looking back to the hills which the skill of later times has taught to guard her, Tolosa, whose capitol has proclaimed her to all ages as the true child of Eome, Tolosa, where the first basilica of the holy Saturninus must have already arisen beyond her walls, that renowned city now passed into the hands of the Goth to become his kingly seat. There, as at Narbo Martins, we shall seek in vain for traces of his presence. The traveller is told that the castle or palace of the West-Gothic kings stood where the paltry palace of justice of modern times now stands. That is all the help that he gains to call up the picture of Toulouse as the head of a Gothic kingdom. For VI.] Wallia and the Settlement of Aquitaine. 241 the abiding monuments of Gothic rule, though of Gothic rule later than the days of Wallia, he must go to a place which does not seem as yet to have been reckoned as a city, which was not as yet a possession of the Goth, to the wondrous hill crowned by the twofold walls and towers of Carcassonne. Before the great barbarian invasion Aquitaine and the land of Novempopulania to the south of it were held to be the fairest regions of Gaul. The sternest prophet of the age, in order to rebuke the ungrateful wickedness of its people, has drawn a living picture of the richness of the land itself. It is to be noticed that he does not dwell specially on the greatness and splendour of its cities. And indeed, with the single exception of Bourdeaux — for Toulouse lies beyond the bounds of Aquitaine — none of the Aquitanian cities of which we have just spoken, with all the surpassing charm of their sites, their history, and their monuments, can claim a place in the first rank of the cities of Gaul. In the whole of Wallia's possessions, no city, save the two Bourdeaux and Toulouse, could at all stand by the side of Narbonne or of the great cities east of Ehone. What Salvianus specially enlarges on is the richness of the land itself. It is the marrow of all. the Gauls, the breast of all fruitfulness, and more than fruitfulness, of pleasant- ness and beauty and all delight. The meadows, the vineyards, the orchards, the cornfields, the groves, the fountains that watered them, the streams that flowed among them, made the masters of that land seem as if it was not a share of the common earth which had become their portion, but that they had 242 Western Europe in the Fifth Century, [vi. become possessors of the image of paradise *. But the men thus highly favoured, the Christian Eomans of Aquitaine, had shown themselves indeed unworthy of the gifts of Heaven. They were given up to every kind of vice, to unchastity above all. The Eoman of Aquitaine seems to have been the foulest of sinners, save only the Koman of Africa. Such a people needed the chastisement of barbarian inva- sion to slay some and to reform the restf. We should be glad to know exactly in what case the land stood at the moment of Wallia's entry. From the general picture of the passage of the bar- barians which we looked at long ago, we- may fancy that the cities had greatly suffered ; Yesonna, with the narrowed enclosure of its walls, is a living witness of the shifts to which men were driven to defend them- selves. But even the cities, as in the case of Trier, seem to have sprung up again with wonderful ease to some measure of prosperity, and the fertile land, its cornfields, vineyards, and orchards, might be agaiiL smiling now that ten years had passed since the flood of mere havoc had passed over them. And now * Salvianus, vii. 2 ; " N"emini dubium est Aquitanos et Novem- populanos meduUam fere omnium Galliarum et uber totius fecundi- tatis habuisse, nee solum fecunditatis, sed quae prseponi interdum fecunditati solent, jucunditatis, pulcritudinis, voluptatis. Adeo illic omnis admodum regio aut intertexta vineis aut florulenta pratis aut distincta culturis aut consita pomis aut amoenata lucis aut inrigua fontibus aut interfusa fluminibus aut crinita messibus fuit, ut vere possessores ac domini terrae illius non tam soli istius portionem quam paradisi imaginem possidere videantur." t lb. 1 2 ; " Sed paulatim id ipsum tamen, ut dum pars clade cseditur, pars exemplo emendaretur." VI.] Wallia and the Settlement of Aquitaine. 243 milder visitors had come ; the chaste Goths were there to dwell in the land and rule it and cleanse it from its defilements, the Goths, such true models of virtue, that notwithstanding their heresy, heresy which the presbyter of Massalia hardly deems to have been their fault, they might dare to look with some hope for a place in the kingdom of heaven *. The barbarian heretic, in whose dominions none was unclean save the Catholic Eomant, thus sat down to dwell in the land of the Eoman, in his stately cities, amid his goodly fields and vineyards, by the side of his cooling founts and streams. He came in not as a conqueror of the Eoman, but as in some sort his fellow-subject, at least the faithful soldier of his Emperor, rewarded for his faithful service with lands within his Empire. But it is hard to see how the Goth could be settled on the lands of the Eoman except at the cost of the Eoman. If not a conqueror in form, he must have been strongly tempted to become a conqueror in practice. The almost received law of such settlements was that the faithful soldiers of the Empire received as their wages two-thirds of the lands of its peaceful citizens. It is not clear whether this system was regularly * Salvianus, v. 2 ; " Errant ergo, sed bono animo errant, non odio sed affectu Dei, honorare se Dominum atque amare credentes. Quamvis non rectam habeant fidem, illi tamen hoc perfectam Dei sestimant caritatem. Qualiter pro Hoc ipso falsse opinionis errore in die judicii puniendi sint, nullus potest scire nisi judex." + lb. vii. 6 ; " Esse inter Gotbos non licet scortatorem Gotbum ; soli inter eos praejudicio nationis et nominis permittuntur impuri esse Eomani." 23 ; " Apud Gotbos impudici non sunt nisi Komani, jam apud "Wandalos nee Eomani." E 2 244 Western Europe in the Fifth Century, [vi. carried out in the Gothic settlement of Aquitaine *, and it is remarkable that in one case where we happen to know something of the details, we see a much greater regard to earlier rights of property than we should have looked for. Chastity was not the only virtue of the Goth, Even in grasping the lands of others, he could sometimes be touched with the natural feeling of just dealing between man and man, even when man and man took the shape of barbarian and Eoman, of conqueror and conquered. Paulinus of Bourdeaux and Pella, Paulinus grandson of Ausonius, driven from his own city to dwell in exile and poverty at Marseilles, had his fortunes in some measure raised again by the justice or bounty of one of the new settlers. A Goth who had coveted the last remnant of Paulinus' great estates sent its owner a payment, not, the owner thought, equal to the full value of the land, but a payment which made to the banished man the difference between poverty and comfort, a payment which, if the Goth had had the mind to refuse, the Eoman had assuredly no means of enforcing f. And from the picture which * See Dahn, K. G. v. 70. t Paulini Eucharisticon, 570 ; " Ut cum jam penitus fructus de rebus avitis Sperare ulterius nullos me posse probasses, Cunctaque ipsa etiam quae jam tenuatus habere Massilise potui, amissa jam proprietate, Conscripta adstrictus sub conditione tenerem, Emptorem mihi ignotum de gente Gothorum Excires, nostri quondam qui juris agellum Mercari cupiens pretium transmitteret ultro, Haud equidem justum, verumtamen accipienti VI.] Pf/allta and the Settlement of Aquitaine. 245 Paulinus gives of the relations between Eoman and Goth during the earlier occupation of Bourdeaux we may infer that his case did not stand alone. We have seen that the Gothic guest, the delicate euphemism for the stranger who was quartered on the lands of the Eoman, showed himself not uncommonly the friend and protector of the host. So in the more lasting settlement, if the Eoman of Aquitaine had to surrender two-thirds of his land to the Goth — and, even without such formal division, the transfer of land cannot fail to have been large — we may be- lieve that the Eoman often enjoyed what was left to him with greater security under barbarian fellow- ship than if he had possessed the whole when subject to those exactions of Imperial rule under which Salvian paints every Eoman land as groaning. A third Teutonic kingdom had thus arisen in Gaul. The West-Gothic kingdom was now far greater than those of the Franks or the Burgundians ; it was the first of Gaulish powers ; it was presently, by extension beyond the Pyrenees, to become for a while the first of all powers beyond the Alps. Of the other two Teutonic nations which had settled on Gauhsh soil, one hardly knows how to speak of the Franks. The Safians, under their long-haired kings, are dwelling on lands of the Empire ; they are in form subjects and soldiers of the Empire, and in the last character we have more than once seen them do good service. But though they have come geographically Votivum, fateor, posset quo scilicet una Et veteres lapsi census fulcire ruinas Et vitare nova cari mild damna pudoris." 246 Western Europe in the Fifth Century, [vi. within the Eoman boundary, they have not in any but a purely military sense come within the Eoman pale. They have not come into the Roman world in the same way in which Goths and even Burgun- dians have come into it. The Franks still stand outside almost like the Saxons themselves. Sixty years later, they have not yet adopted the religion of the Empire ; they are not even Arian Christians. The Frank, soldier of Kome, perhaps all the more because he is the soldier of Eome, has not yet convinced himself, as the Burgundian has already done, that the God of the Eomans is stronger than the gods of his fathers*. When that conviction was at last brought home to his mind, the conse- quences were memorable indeed. For the military defence of the Empire he is better to be trusted than any other of its nominal vassals ; but he has rent away a certain portion of the earth from fellowship with the Eoman and Christian world in a way that even the revolted Briton, whether in his island or on the mainland, has not done. The Burgundian was a later settler on Imperial soil than the Frank ; but he became a member of the Eoman and Christian world far more speedily. Still he was a new-comer, and was only gradually making his way from his first Ehenish home, from the land of Mainz and Worms, to those cities of the Ehoneland which became the dwelling-place and the burying-place of his kings, but which we have had to look at mainly as the * Sokrates, vii. 30 ; Kara vovv Xa^^dvovres OTi 'Pafuiiav 6 Q(6s l