I-.-T- %.>;-■;-■... . V)?f7' f$| 10RNELL UNIVERSITY .LIBRARY. ©j is**. This book is not to be taken § m from tiie Reading Room. ■ 3 k itf*t>c n K »-.. : •' .- i , r e r u ft w AT OSFC E to ■■'■■ r ' : C W - r '.' W&W^0M t The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31 924071 1 92441 CORNELL UNIVERSITY LIBRARY 3 1924 071 192 441 THE HISTORY OF ROME THE HISTORY OF ROME BY THEODOR MOMMSEN TRANSLATED WITH THE AUTHOR'S SANCTION AND ADDITIONS BY WILLIAM P. DICKSON, D.D., LL.D. PROFESSOR OF DIVINITY IN THE UNIVERSITY OF GLASGOW THE PROVINCES, FROM CAESAR TO DIOCLETIAN" PART I WITH EIGHT MAPS BY PROFESSOR KIEPERT LONDON RICHARD BENTLEY & SON, NEW BURLINGTON STREET $ttWfefjws in ©rtrtnsrjj to fger ffflajestg tfje ©tweti 1886 THE PROVINCES OF THE ROMAN EMPIRE FROM CAESAR TO DIOCLETIAN BY THEODOR MOMMSEN TRANSLATED WITH THE AUTHOR'S SANCTION AND ADDITIONS BY WILLIAM P. DICKSON, D.D., LL.D. PROFESSOR OF DIVINITY IN THE UNIVERSITY OF GLASGOW VOL. I WITH EIGHT MAPS BY PROFESSOR KIEPERT LONDON RICHARD BENTLEY & SON, NEW BURLINGTON STREET $ttftlts[jms in ©r&mat2 to |§er fKsjestg tfre ©urat 1886 /COR: (T~ i UNIVE^ TO LEOPOLD KRONECKER AND RICHARD SCHONE IN GRATEFUL REMEMBRANCE PREFACE A WISH has often been expressed to me that the History of Rome might be continued, and I have a desire to meet it, although it is difficult for me, after an interval of thirty years, to take up again the thread at the point where I had to let it drop. That the present portion does not attach itself immediately to the preceding, is a matter of little moment ; the fifth volume would be just as much a fragment without the sixth as the sixth now is without the fifth. Besides, I am of opinion that, for the purposes of the cultured public, in whose minds this History is intended to promote an intelligent conception of Roman antiquity, other works may take the place of the Two Books, which are still awanting between this (the Eighth) and the earlier ones, more readily than a substitute can be found for that now issued. The struggle of the Repub- licans in opposition to the monarchy erected by Caesar, and the definitive establishment of the latter, are so well presented in the accounts handed down to us from antiquity that every delineation amounts essentially to a reproduction of their narrative. The distinctive character of the monarchical rule and the fluctuations of the monarchy, as well as the general relations of government influenced by the personality of the individual rulers, x PREFACE. which the Seventh Book is destined to exhibit, have been at least subjected to frequent handling. Of what is here furnished — the history of the several provinces from the time of Caesar to that of Diocletian, — there is, if I am not mistaken, no comprehensive survey anywhere accessible to the public to which this work addresses itself; and it is owing, as it seems to me, to the want of such a survey that the judgment of that public as to the Roman imperial period is frequently incorrect and unfair. No doubt such a separation of these special histories from the general history of the empire, as is in my opinion a preliminary requisite to the right understanding of the history of the imperial period, cannot be carried out completely as regards various sections, especially for the period from Gallienus to Diocletian ; and in these cases the general picture, which still remains to be given, will have to sup- ply what is wanting. If an historical work in most cases acquires a more vivid clearness by an accompanying map, this holds in an especial degree true of our survey of the Empire of three Continents according to its provinces, and but few of its readers can have in their hands maps adequate for the pur- pose. These will accordingly be grateful, along with me, to my friend Dr. Kiepert, for having, in the manner and with the limits suggested by the contents of these volumes, annexed to them, first of all, a sheet presenting a general outline of the Orbis Romanus, which serves moreover in various respects to supply gaps in those that follow, and, in succession, nine special maps of the several portions of the empire drawn — with the exception of sheets S, 7, 8, 9 — on the same scale. The ancient geographical names PREFACE. xi occurring in the volumes, and the more important modern ones, are entered upon the maps ; names not mentioned in the volumes are appended only, in exceptional cases, as landmarks for the reader's benefit. The mode of writing Greek names followed in the book itself has been displaced by the Latinising spelling — for the sake of uniformity — in several maps in which Latin names preponderate. The sequence of the maps corresponds on the whole to that of the book ; only it seemed, out of regard for space, desirable to present on the same sheet several provinces such as, e.g. Spain and Africa. PREFATORY NOTE BY THE TRANSLATOR. When I learned from Dr. Mommsen that he was about to issue, after so considerable an interval, a continuation of his History, I had some thoughts of leaving the task of its translation to other and less occupied hands. But, when he expressed a desire that I should introduce this new portion also to the English public, I felt it but due to him as well as to the publishers and to those who had favourably received my rendering of the earlier volumes, that I should attempt to meet his wish and what I might presume to be theirs. I have endeavoured to prepare the translation accord- ing to the principles and method adopted in the earlier volumes (and explained in my preface to Volume I.), so far as there can be uniformity in applying them after an interval of five-and-twenty years ; and, if in my desire to reproduce the form as well as the matter of the book, I have at times followed the mould of the German too closely, I trust that the reader may not at least often be at a loss for the meaning. The task has been in so far longer and more difficult, that there is a much larger pro- portion of matter in the form of notes. The present volumes differ indeed considerably in character from those preceding them ; but, while the reader will miss, as Dr. Mommsen has himself remarked, the graphic description and portrait-painting of the earlier portions, he will find compensation in the presence of other and fresh elements PREFA TOR Y NOTE. xiii of interest, more especially in the copious and masterly- use of materials gleaned from Jthe epigraphic stores 1 which Dr. Mommsen has done so much to collect and to make accessible. To prevent misconception, it may be well to add that in translating the work I am not to be held as accepting all its principles and verdicts. Whether, and when, the missing link of the Fifth volume will be supplied, Dr. Mommsen leaves as open questions. I have thought that the convenience of different readers would be best met, under the cir- cumstances, by issuing the volumes, after a fashion common in Germany, with a general and a special title, so that they may either take their place in continuation or be procured by themselves. The special title chosen is somewhat elliptical in point of grammar ; but the fuller form would have been cumbrous. 1 The chief epigraphic works referred to are usually quoted under the initials, or other very abbreviated form, of the title, and, as they may not be known to all readers, I subjoin a brief explanation : C. I. L. represents the great Corpus Dittenberger, of which five volumes inscriptionum Latinarum projected or parts have appeared since 1877. and authorised by the Royal Aca- C. I. Rh., or sometimes "Brambach," demy of Sciences of Berlin, in the denotes the Corpus inscriptionum preparation and superintendence of Rhenanarum, edited by Wilhelm ) which Dr. Mommsen has had a chief Brambach, and issued in 1867. part. Begun upwards of twenty " Orelli" or " Henzen" or " Orelli- y ears ago, it now extends to sixteen Henzen" indicates the Inscrip- volumes (or half volumes) folio. tionum Latinarum selectarum am- Eph. Ep. or Epigr., is the Ephemeris plissima collectio, edited by Orelli epigraphica, issued under the sane- and (in the supplementary third tion of the Archaeological Institute volume) Henzen, and published at at Rome, and edited by Mommsen, Zurich, 1828-1856. Henzen, and others, as a Supplement Bull, de corr. Hell, denotes the Bul- la the Corpus above named. It letin de correspondence Hette'nique, was begun in 1872, and has reached published by the Ecole Francaise a sixth volume. at Athens from 1877 onward, of C. I. Gr. — Corpus inscriptionum which nine or ten volumes have Graecarum prepared under the aus- appeared. pices of the Berlin Academy, by " Le Bas" or " Waddington" or Boeckh and Franz, and subsequently " Le Bas -Waddington," refers to by E. Curtius and A. Kirchhoff. 4 the Voyage archiologique en Grhe vols. fol. 1828-1877. et en Asie mineure, by Philippe Le C. I. A. or Att. — Corpus inscrip- Bas and William Henry Wadding- tionum Atticarum, edited under ton, of which portions have ap- the like authority, by Koehler and peared at intervals since 1847. xiv PREFATORY NOTE. I have taken the opportunity, in verifying various references, to correct several such errors as are apt to occur in the frequent use of figures ; have broken up some of the longer paragraphs, and added considerably to the marginal headings ; and have drawn up on my own part an Index, which seemed to me more necessary even for the present than for the earlier volumes. In prepar- ing it I have attempted a mean between a mere notice of the more salient matters and a full list of names, and have sought to meet the wish of a correspondent by index- ing the names under the familiar surname rather than (as in the Index to the earlier volumes) under the gentile name. As regards the maps which have been specially pre- pared, as the preface states, for the book by the well- known German cartographer, Professor Kiepert, I have deemed it best simply to append them, as they stand delineated by him, to the English book. Beyond the apparent incongruity of the German titles the reader will have little difficulty in using them, and will be glad to be in possession of Dr. Kiepert's own work. I have, however, sought to obviate any such difficulty by prefixing an explanatory leaflet. WILLIAM P. DICKSON. Glasgow College, November 1886. CONTENTS BOOK EIGHTH THE PROVINCES AND PEOPLE, FROM CAESAR TO DIOCLETIAN PAGE Introduction ... .3 CHAPTER I. The Northern Frontier of Italy ... 7 CHAPTER II. Spain ....... 63 CHAPTER III. The Gallic Provinces . . . .78 CHAPTER IV. Roman Germany and the Free Germans . . 117 xvi CONTENTS. PAGE CHAPTER V. Britain .... .170 CHAPTER VI. The Danubian Lands and the Wars on the Danube 195 CHAPTER VII. Greek Europe . . . . .252 CHAPTER VIII. Asia Minor ...... 320 BOOK EIGHTH. THE PROVINCES AND PEOPLE FROM CAESAR TO DIOCLETIAN. Go through the world and converse with every one. Firdusi. »v VOL. I. INTRODUCTION. The history of Rome under the Empire presents problems similar to those encountered in the history of the earlier Republic. Such information as may be directly obtained from literary tradition is not merely without form and colour, but in fact for the most part without substance. The list of the Roman monarchs is just about as trustworthy and just about as instructive as that of the consuls of the republic. The great crises that convulsed the state may be dis- cerned in outline ; but we are not much better informed as to the Germanic wars under the emperors Augustus and Marcus, than as to the wars with the Samnites. The republican store of anecdote is very much more decorous than its counterpart under the empire ; but the tales told of Fabricius and of the emperor Gaius are almost equally insipid and equally mendacious. The internal development of the commonwealth is perhaps exhibited in the traditional accounts more fully for the earlier republic than for the imperial period ; in the former case there is preserved a picture — however bedimmed and falsified — of the changes of political order that were brought at least to their ultimate issue in the open Forum of Rome ; in the latter case the arrangements are settled in the imperial cabinet, and come before the public, as a rule, merely in unimportant matters of form. We must take into account, moreover, the vast extension of the sphere of rule, and the shifting of the vital development from the centre to the circumference. The history of the city of Rome widens out into that of the country of Italy, and the latter into that of the Mediterranean world ; and of what we are most concerned to know, we learn the least. The Roman state of this epoch resembles a mighty 4 INTRODUCTION. tree, the main stem of which, in the course of its decay, is surrounded by vigorous offshoots pushing their way upwards. The Roman senate and the Roman rulers soon came to be drawn from any other region of the empire just as much as from Italy ; the Quirites of this epoch, who have become the nominal heirs of the world -subduing legion- aries, have nearly the same relation to the memories of the olden time as our Knights of St. John have to Rhodes and Malta ; and they look upon their heritage as a right capable of being turned to profitable account — as an endowment provided for the benefit of the poor that shrink from work. Any one who has recourse to the so-called authorities for the history of this period — even the better among them — finds difficulty in controlling his indignation at the telling of what deserved to be suppressed, and at the suppression of what there was need to tell. For this epoch was also one productive of great conceptions and far-reaching action. Seldom has the government of the world been conducted for so long a term in an orderly sequence ; and the firm rules of administration, which Caesar and Augustus traced out for their successors, maintained their ground, on the whole, with remarkable steadfastness not- withstanding all those changes of dynasties and of dynasts, which assume more than due prominence in a tradition that looks merely to such things, and dwindles erelong into mere biographies of the emperors. The sharply-defined sections, which — under the current conception, misled by the superficial character of such a basis — are constituted by the change of rulers, pertain far more to the doings of the court than to the history of the empire. The carrying out of the Latin-Greek civilising process in the form of perfecting the constitution of the urban community, and the gradual bringing of the barbarian or at any rate alien elements into this circle, were tasks, which, from their very nature, required centuries of steady activity and calm self- development ; and it constitutes the very grandeur of these centuries that the work once planned and initiated found this long period of time, and this prevalence of peace by land and sea, to facilitate its progress. Old age INTRODUCTION. 5 has not the power to develop new thoughts and display creative activity, nor has the government of the Roman empire done so ; but in its sphere, which those who belonged to it were not far wrong in regarding as the world, it fostered the peace and prosperity of the many nations united under its sway longer and more completely than any other leading power has ever succeeded in doing. It is in the agricultural towns of Africa, in the homes of the vine-dressers on the Moselle, in the flourishing townships of the Lycian mountains, and on the margin of the Syrian desert that the work of the imperial period is to be sought and to be found. Even now there are various regions of the East, as of the West, as regards which the imperial period marks a climax of good government, very modest in itself, but never withal attained before or since ; and, if an angel of the Lord were to strike the balance whether the domain ruled by Severus Antoninus was governed with the greater intelligence and the greater humanity at that time or in the present day, whether civilisation and national prosperity generally have since that time advanced or retro- graded, it is very doubtful whether the decision would prove in favour of the present. But, if we find that this was the case, we ask of our surviving books for the most part in vain how it came to be so. They no more give an answer to this question than the traditional accounts of the earlier republic explain the mighty phenomenon of the Rome, which, in the footsteps of Alexander, subdued and civilised the world. The one void as little admits of being filled up as the other. But it seemed worth our making the attempt for once to turn away our eyes from the pictures of the rulers with their bright or faded, and but too often falsified, colours, as well as from the task of linking into a sem- blance of chronological order fragments that do not fit each other ; and, instead of this, to collect and arrange such materials as tradition and the monuments furnish for a de- scription of the Roman provincial government. It seemed worth while to collate the accounts accidentally preserved by the one or by the other, to note traces of the process of growth embedded in its results, and to view the general 6 INTRODUCTION. institutions in their relation to the individual provinces, along with the conditions given for each by the nature of the soil and of the inhabitants, so as to work out by the im- agination — which is the author of all history as of all poetry — if not a complete picture, at any rate a substitute for it. In this attempt I have not sought to go beyond the epoch of Diocletian. A summary glance, at the utmost, into the new government which was then created may fitly form the keystone of this narrative; to estimate it fully would require a separate narration and another frame for its setting — an independent historical work, carried out in the large spirit and with the comprehensive glance of Gibbon, but with a more accurate understanding of details. Italy and its islands have been excluded ; for the account of these cannot be dissociated from that of the general government of the empire. The external history, as it is called, of the imperial period is dealt with as an integral part of the provincial administration ; what we should call imperial wars were not carried on under the empire against those outside of its pale, although the conflicts called forth by the rounding off, or the defence, of the frontier sometimes assumed such proportions as to make them seem wars between two powers similar in kind, and the collapse of the Roman rule in the middle of the third century, which for some decades seemed as though it were to become its definitive end, grew out of the unhappy conduct of frontier-defence at several places simultaneously. Our narrative opens with the great work of pushing forward, and of regulating the frontier towards the north, which was partly carried out and partly failed under Augustus. At other points we bring together the events that occurred on each of the three chief arenas for frontier-defence— the Rhine, the Danube, the Euphrates. The remainder of the narrative is arranged according to provinces. Charms of detail, pictures of feeling, sketches of character, it has none to offer ; it is allowable for the artist, but not for the historian, to reproduce the features of Arminius. With self-denial this book has been written • and with self-denial let it be read. CHAPTER I. THE NORTHERN FRONTIER OF ITALY. The Roman Republic extended its territory chiefly by Northern means of the sea towards the west, south, and east : little ^°™ e daiy was done towards extending it in the direction, in which empire. Italy and the two peninsulas dependent upon it to the west and east are connected with the great mainland of Europe. The region which lay behind Macedonia was not subject to the Romans, nor yet even the northern slope of the Alps ; only the inland region behind the south coast of Gaul ,had been annexed by Caesar to the empire. Looking to the position occupied by the empire in general, this state of things could not be allowed to continue ; the fact that the inert and unstable rule of the aristocracy had been superseded could not but tell with preeminent effect in this sphere of action. Caesar had not charged the heirs of his dictatorial power with the extension of Roman territory on the north slope of the Alps and on the right bank of the Rhine so directly as with the conquest of Britain ; but in reality such an enlargement of the bounds suggested itself far more naturally, and was more necessary, than the subduing of the transmarine Celts, and we can readily understand why Augustus took in hand the former and omitted the latter. The task was divided into three great sections — the operations on the northern frontier of the Graeco- Macedonian peninsula, in the region of the middle and lower Danube, in Illyricum ; those on the northern frontier of Italy itself, in the region of the upper Danube, in Raetia and Noricum ; lastly, those on the right war. 8 NORTHERN FRONTIER OF ITALY, book vm. bank of the Rhine, in Germany. Though conducted for the most part independently, the military political mea- sures in these regions had yet an inward connection ; and, as they all had their origin from the free initiative of the Roman government, they can only be understood in their success or in their partial failure, when they are looked at from a military and political point of view as a whole. We shall, therefore, in our account of them, follow the connection of place rather than the order of time ; the structure, of which they are but parts, is better viewed in its internal compactness than according to the succession of the several buildings composing it. Dalmatian The prelude to this great aggregate of action was formed by the measures which Caesar the Younger, so soon as he had his hands free in Italy and Spain, under- took on the upper coasts of the Adriatic and in the inland region adjacent to them. In the hundred and fifty years that had elapsed since the founding of Aquileia, the Roman merchant had doubtless from that centre possessed himself more and more of the traffic ; yet the state, directly as such, had made little progress. Considerable trading settlements had been formed at the chief ports of the Dalmatian coast, and also, on the road leading from Aquileia into the valley of the Save, at Nauportus (Upper Laybach) ; Dalmatia, Bosnia, Istria, and Carniola were deemed Roman territory, and the region along the coast at least was actually subject ; but the founding of towns in a legal sense still remained to be done, quite as much . as the subduing of the inhospitable interior. Here, however, another element had to be taken into account. In the war between Caesar and Pompeius the native Dalmatians had as decidedly taken part for the latter as the Roman settlers there had taken the side of Caesar ; even after the defeat of Pompeius at Pharsalus, and after the Pompeian fleet had been driven from the iv. 434. Illyrian waters (iv. 456), the natives continued their resistance with energy and success. The brave and able Publius Vatinius, who had formerly taken a very effect- ive part in these conflicts, was sent with a strong army to chap. i. NORTHERN FRONTIER OF ITALY. 9 Illyricum, apparently -in the year before Caesar's death, and that merely as the vanguard of the main army, with which the Dictator himself intended to follow in order to over- throw the Dacians, who just then were putting forth their rising power (iv. 305), and to regulate the state of affairs iv. 291. in the whole domain of the Danube. The execution of this plan was precluded by the daggers of the assassins. It was fortunate that the Dacians did not on their part penetrate into Macedonia ; Vatinius himself fought against the Dalmatians unsuccessfully, and sustained severe losses. Thereafter, when the republicans took up arms in the East, the Illyrian army joined that of Brutus, and for a considerable time the Dalmatians remained free from attack. After the overthrow of the republicans, Antonius, to whom, in the partition of the empire, Mace- donia had fallen, caused the insubordinate Dardani in the north-west and the Parthini on the coast (eastward from Durazzo) to be put to rout in the year 715, when the 39. celebrated orator Gaius Asinius Pollio gained triumphal honours. In Illyricum, which was under Caesar, nothing could be done so long as the latter had to direct his whole power to the Sicilian war against Sextus Pompeius ; but after its successful termination Caesar personally threw himself with vigour into this task. The small tribes from Doclea (Cernagora), as far as the Iapydes (near Fiume), were in the first campaign (719) either brought back to 35. subjection or now for the first time subdued. It was not a great war with pitched battles of note, but the mountain- conflicts with the brave and desperate tribes, and the cap- ture of the strongholds furnished in part with Roman appliances of war, formed no easy task ; in none of his wars did Caesar display to an equal extent his own energy and personal valour. After the toilsome subjugation of the territory of the Iapydes, he marched in the very same year along the valley of the Kulpa to the point where it joins the Save ; the strong place Siscia (Sziszek) situated at that point, the chief place of arms of the Pannonians, against which the Romans had never hitherto advanced with success, was now occupied and destined as a basis io NORTHERN FRONTIER OF ITALY, book vm. for the war against the Dacians, which Caesar purposed next 34. 33- to undertake. In the two following years (720, 721), the Dalmatians, who had for a number of years been in arms against the Romans, were forced to submit after the fall of their fortress Promona (Promina, near Dernis, above Sebenico). Still more important than these military suc- cesses was the work of peace, which was carried on about the same time, and which they were intended to secure. It was doubtless in these years that the posts along the Istrian and Dalmatian coast, so far as they lay within the field of Caesar's rule, Tergeste (Trieste), Pola, Iader (Zara), Salonae (near Spalato), Narona (at the mouth of the Narenta), as well as Emona (Laybach), beyond the Alps, on the route from Aquileia over the Julian Alps to the Save, obtained, through the second Julian law, some of them town-walls, all of them town-rights. The places themselves had probably all been already long in existence as Roman villages ; but it was at any rate of essential importance that they were now inserted on a footing of equal privilege among the Italian municipia. Prepara- The Dacian war was intended to follow ; but the civil Dacian "^ war stepped in before it a second time. It summoned the war. ruler not to Illyricum.but to theEast,and the heavings of the great decisive struggle between Caesar and Antonius reached even to the distant region of the Danube. The people of the Dacians, united and purified by king Burebista iv. 291. (Boerebistas, iv. 305), now under king Cotiso, found itself courted by the two antagonists — Caesar was even accused of having sought the king's daughter in marriage, and having offered to him in turn the hand of his five-year-old daughter Julia. It is easy to understand how the Dacian should, in view of the invasion planned by the father and ushered in by the son with the fortification of Siscia, have attached himself to the side of Antonius ; and had he done what people in Rome feared — had he, while Caesar was fighting in the East, penetrated from the north into de- fenceless Italy ; or had Antonius, in accordance with the proposal of the Dacians, sought the decision of the struggle not in Epirus but in Macedonia, and drawn thither the chap. i. NORTHERN FRONTIER OF ITALY. n Dacian bands to help him, the fortunes of the war might perhaps have ended otherwise. But neither the one nor the other took place ; moreover, at that very time the Dacian state, created by the vigorous hand of Burebista, again went to pieces ; internal troubles, perhaps also the attacks from the north by the Germanic Bastarnae and by the Sarmatian tribes that subsequently environed Dacia on all sides, prevented the Dacians from interfering in the Roman civil war, in the decision of which their future also was at stake. Immediately after that war was decided, Caesar set himself to regulate the state of things on the lower Danube. But, partly because the Dacians themselves were no longer so much to be dreaded as formerly, partly because Caesar now ruled no longer merely over Illyricum, but over the whole Graeco-Macedonian peninsula, the latter became the primary basis of the Roman operations. Let us picture to ourselves the peoples, and the relations of the ruling powers, which Augustus found there. Macedonia had been for centuries a Roman province. Macedoni- As such, it did not reach beyond Stobi to the north and the Rhodope mountains to the east ; but the range of Rome's power stretched far beyond the frontier proper of the country, although varying in compass and not fixed in point of form. Approximately the Romans seem to have been the leading power at that time as far as the Haemus (Balkan), while the region beyond the Balkan as far as the Danube had been possibly trodden by Roman troops, but was by no means dependent on Rome. 1 Be- yond the Rhodope mountains the Thracian dynasts, who were neighbours to Macedonia, especially those of the Odrysians (ii. 309), to whom the greatest portion of the "■ 29°- south coast and a part of the coast of the Black Sea were subject, had been brought by the expedition of Lucullus 1 Dio, li. 23, expressly says this as irpa-yjua, irpbs roiis 'Pw/iatovs 5jv m iirel Si 2 9- to the year 725 : rias p.iv o$v tclvt' t &„ Te At/j-ov inrepifiriaav k.o.1 rty 9pq.KT)v iiroiow (i.e. so long as the Bastarnae T ty AecSeXijTfii' ho-iroi/5ov airoU ofoav attacked only the Triballi — near Ka.Ttdpap.oi> k. t. X. The allies in Oescns in Lower Moesia, and the Dar- Moesia, of whom Dio, xxxviii. 10 dani in Upper Moesia), oiBtv crtplai speaks, are the coast towns. 12 NORTHERN FRONTIER OF ITALY, book vm. iv. 39 . (iv. 41) under the Roman protectorate ; while the inhabit- ants of the more inland territories, especially the Bessi on the upper Maritza, were perhaps called subjects, but were not so, and their incursions into the settled territory as well as retaliatory expeditions into theirs were of constant occur- 60. rence. Thus, about the year 694, Augustus' own father, 43- Gaius Octavius, and in the year 711, during the prepara- tions for the war against the triumvirs, Marcus Brutus had fought against them. Another Thracian tribe, the Denthe- letae (in the district of Sofia), had, even in Cicero's time, on an incursion into Macedonia, threatened to besiege its capital Thessalonica. With the Dardani, the western neighbours of the Thracians, a branch of the Illyrian family, who inhabited southern Servia and the district of Prisrend, Curio, the predecessor in office of Lucullus, had fought successfully ; and ten years later Cicero's colleague in the consulate, Gaius Antonius, unsuccessfully in the 62. year 692. Below the Dardanian territory, again, there were settled close to the Danube Thracian tribes, the once powerful but now reduced Triballi in the valley of the Oescus (in the region of Plewna), and farther on, along both banks of the Danube to its mouth, Dacians, or, as on the right bank of the river they were usually called by the old national name which was retained also by their Asiatic kinsmen, Mysians or Moesians, probably in Burebista's time a part of his kingdom, now once more split up into different principalities. But the most powerful people between the Balkan and the Danube at that time were the Bastarnae. We have already on several occasions met with this brave and numerous race, the eastmost branch of ii. 290. the great Germanic family (ii. 308). Settled, strictly speaking, behind the Transdanubian Dacians beyond the mountains which separate Transylvania from Moldavia, at the mouths of the Danube and in the wide region from these to the Dniester, they were themselves outside of the Roman sphere ; but from their ranks especially had both king Philip of Macedonia and king Mithridates of Pontus formed their armies, and in this way the Romans had often already fought with them. Now they had crossed the chap. i. NORTHERN FRONTIER OF ITALY. 13 Danube in great masses, and established themselves north of the Haemus ; in so far as the Dacian war, as planned by Caesar the father and then by the son, had doubtless for its object to gain the right bank of the lower Danube, it was not less directed against them than against the Dacian Moesians on the right bank. The Greek coast towns in the barbarian land, Odessus (near Varna), Tomis, Istropolis, hard pressed by these movements of the nations surging around them, were here as everywhere from the outset clients of the Romans. At the time of Caesar's dictatorship, when Burebista was at the height of his power, the Dacians had executed that fearful devastating raid along the coast as far down as Apollonia, the traces of which were not yet obliterated after a century and a half. It may probably have been this invasion that at first induced Caesar the elder to under- take the Dacian war ; and after that the son now ruled also over Macedonia, he could not but feel himself under obligation to interfere here at once and with energy. The defeat which Cicero's colleague, Antonius, had sustained near Istropolis at the hands of the Bastarnae may be taken as a proof that these Greeks needed once more the aid of the Romans. In fact soon after the battle of Actium (725) Marcus 29- Licinius Crassus, the grandson of him who had fallen ti o n J of a " at Carrhae, was sent by Caesar to Macedonia as Moesiaby governor, and charged now to carry out the cam- paign that had twice been hindered. The Bastarnae, who just then had invaded Thrace, submitted without resistance, when Crassus had them summoned to leave the Roman territory ; but their retreat was not suffi- cient for the Roman. He, on his part, crossed the Haemus, 1 at the confluence of the Cibrus (Tzibritza) with the Danube, defeated the enemy, whose king, Deldo, was left on the field of battle ; and, with the help of a Dacian prince adhering to the Romans, took prisoners all that had 1 When Dio says (li. 23) : tV Serdica, the modern Sofia, on the 2e-y%™ri)i' Ka\oviiivqv Trpoaeiroli)aa.To upper Oescus, the key to the Moesian koX is ttjv MvaiSa ^W/3aAe, the town country, spoken of, doubtless, can only be i 4 NORTHERN FRONTIER OF ITAL Y. book viii. escaped from the battle and sought shelter in a neighbour- ing' stronghold. Without offering further resistance the whole Moesian territory submitted to the conqueror of the Bastarnae. These returned next year to avenge the defeat which they had suffered ; but they once more suc- cumbed, and, with them, such of the Moesian tribes as had again taken up arms. Thus these enemies were once for all expelled from the right bank of the Danube, and the latter was entirely subjected to the Roman rule. At the same time the Thracians not hitherto subject were chastised, the national shrine of Dionysos was taken from the Bessi, and the administration of it was entrusted to the princes of the Odrysians, who generally from that time, under the protection of the Roman supreme power, exer- cised, or were assumed to exercise, supremacy over the Thracian tribes south of the Haemus. The Greek towns, moreover, on the coast of the Black Sea were placed under its protection, and the rest of the conquered territory was assigned to various vassal-princes, on whom devolved accordingly, in the first instance, the protection of the fron- tier of the empire ; 1 Rome had no legions of her own 1 After the campaign of Crassus Had Moesia been originally a part of the conquered land was probably the province of Illyricum, it would have organised in such a. way that the retained this name ; for on the division coast went to the Thracian empire, as of a province the name was usually Zippel has shown (Rom. Illyricum, retained, and only a defining epithet p. 243), and the western portion was, added. But the appellation Illyricum, just like Thrace, assigned in fief to which Dio doubtless reproduces I.e., the native princes, in place of one of was always in this connection restricted whom must have come the praefectus to the upper (Dalmatia) and the lower civitatium Moesiae el Triballiae (C. /. (Pannonia). Moreover, if Moesia was L. v. 1838), who was still acting a part of Illyricum, there was no under Tiberius. The usual assump- room left for that Prefect of Moesia tion that Moesia was at first combined and Triballia, or in other words for with Illyricum, rests only on the cir- his kingly predecessor. Lastly, it cumstance that in the enumeration of is far from probable that in 7 2 7 a 27> the provinces apportioned in the year command of such extent and import- . 727 between emperor and senate in ance should have been entrusted to a Dio, liii. 12 it is not named, and so single senatorial governor. On the was contained in " Dalmatia. " But other hand, everything admits of easy this enumeration does not extend at explanation, if small client - states all to the vassal -states and the pro- arose in Moesia after the war of curatorial provinces, and so far all is in Crassus ; these were as such from the due keeping with that assumption. On outset under the emperor, and, as the the other hand, weighty arguments senate did not take part in their suc- tell against the usual conception, cessive annexation and conversion into chap. i. NORTHERN FRONTIER OF ITALY. 15 left for these distant regions. Macedonia thereby became an inland province, which had no further need of military- administration. The goal, which had been contemplated in those plans of Dacian warfare, was attained. Certainly this goal was merely a provisional one. But before Augustus took in hand the definitive regula- tion of the northern frontier he applied himself to re- organise the provinces already belonging to the empire ; more than ten years elapsed over the arrangement of things in Spain, Gaul, Asia, and Syria. How, when what was needful in these quarters was done, he set to work on his comprehensive task, we have now to tell. Italy, which bore sway over three continents, was still, Subjuga- as we have said, by no means absolutely master in her Aip S ° fthe own house. The Alps, which sheltered her on the north, were in all their extent, from one end to the other, filled with small and but little civilised tribes of Illyrian, Raetian, or Celtic nationality, whose territories in part bordered closely on those of the great towns of the Transpadana — that of the Trumpilini (Val Trompia) on the town of Brixia ; that of the Camunni (Val Camonica above the Lago d'Iseo) on the town of Bergomum ; that of the Salassi (Val d Aosta) on Eporedia (Ivrea) — and whose neighbourhood was by no means wont to be peaceful. Often enough conquered and j proclaimed at the Capitol as vanquished, these tribes, in spite of the laurels of the men of note that triumphed over \ them, were constantly plundering the farmers and the merchants of Upper Italy. The mischief was not to ! be checked in earnest until the government resolved to cross the Alpine chain and bring its northern slope also under their power ; for beyond doubt numbers of these depredators were constantly streaming over the mountains to pillage the rich adjoining country. In the direction a governorship, this might easily be Pamphylia, can only have had as his unnoticed in the Annals. It was province Pannonia or Moesia, and, as . completed in or before the year 743, at that time Tiberius was acting as seeing that the governor, L. Calpur- legate in Pannonia, there is left for nius Piso then waging war against the him only Moesia. In 6 A.D. there Thracians, to whom Dio (liv. 34) certainly appears an imperial governor erroneously assigns the province of of Moesia. 1 6 NORTHERN FRONTIER OF ITALY, book vm. of Gaul also similar work had to be done; the tribes in the upper valley of the Rhone (Valais and Vaud) had indeed been subdued by Caesar, but are also named among those that gave trouble to the generals of his son. On the other side, the peaceful border-districts of Gaul complained of the constant incursions of the Raeti. The numerous expeditions arranged by Augustus on account of these evils do not admit, or require, historical recital ; they are not recorded in the triumphal Fasti and do not fall under that head, but they gave to Italy for the first time settled life in the north. We may mention 16. the subjugation of the already named Camunni in 738 — by the governor of Illyria, and that of certain Ligurian 14. tribes in the region of Nice in 740, because they show- how, even about the middle of the Augustan age, these insubordinate tribes pressed directly upon Italy. If the emperor subsequently, in the collective report on his im- perial administration, declared that violence had not been wrongfully employed by him against any of these small tribes, this must be understood to the effect that cessions of territory and change of abode were demanded of them, and they resisted the demand ; only the petty cantonal union formed under king Cottius of Segusio (Susa) sub- mitted without a struggle to the new arrangement. Subjuga- The southern slopes and the valleys of the Alps Raeti* ^ formed the arena of these conflicts. The establishment of the Romans on the north slope of the 1, mountains and in i S . the adjoining country to the northward followed in 739. The two step-sons of Augustus reckoned as belonging to the imperial house, Tiberius the subsequent emperor, and his brother Drusus, were thereby introduced into the career of generalship for which they were destined ; very secure and very grateful were the laurels put before them in prospect. Drusus penetrated from Italy up the valley of the Adige into the Raetian mountains, and achieved here a first victory; for the farther advance his brother, then governor of Gaul, lent him a helping hand from Helvetia ; on the lake of Constance itself the Roman triremes defeated the boats of the Vindelici ; on the emperor's day, the 1st chap. i. NORTHERN FRONTIER OF ITALY. 17 August 739, in the vicinity of the sources of the Danube, 1 5- was fought the last battle, whereby Raetia and the land ' of the Vindelici — that is, the Tyrol, East Switzerland, and Bavaria — became thenceforth constituent parts of the Roman empire. The emperor Augustus had gone in person to Gaul to superintend the war and the organisa- tion of the new province. At the point where the Alps abut on the Gulf of Genoa, on the height above Monaco, a monument commanding a wide prospect of the Tyrrhene Sea, and not even yet wholly effaced, was erected some years later by grateful Italy to the emperor Augustus, because under his government all the Alpine tribes from the Upper to the Lower sea — the inscription enumerates forty-six of them — had been brought under the power of the Roman people. It was no more than the simple \ I truth ; and this war was what war ought to be — the \, guardian and the guarantee of peace. A task more difficult doubtless than that of the war Organisa- proper was the organisation of the new territory ; the more p 0n t ? f especially as considerations of internal policy exerted to some extent a very disturbing influence on it. Since, as things stood, the preponderance of military power might not be located in Italy, the government had to take care that the great military commands were removed as far as possible from its immediate vicinity ; indeed one of the motives that conduced to the occupation of Raetia itself was the desire to remove the command, which probably up to this time could not have been dispensed with in Upper Italy itself, definitively away from that region, as was thereupon actually done. It might most naturally have been ex- pected that there would be created on the north slope of the Alps a great centre for the military posts indispensable in> the newly acquired territory ; but a course the very opposite \\ \ of this was followed. Between Italy on the one hand, and \ ~ \ the great commands on the Rhine and Danube on the \ other, there was drawn a girdle of small governorships, which were not merely all filled up by the emperor, but were also filled up throughout with men not belonging to the senate. Italy and the province of southern Gaul were VOL. 1. 2 1 8 NORTHERN FRONTIER OF ITALY, book vm. separated by the three small military districts of the Maritime Alps (department of the Maritime Alps and the province of Cuneo), the Cottian Alps with Segusio (Susa) as its chief town, and probably the Graian Alps (East Savoy). Among these the second, administered by the already named cantonal prince, Cottius, and his descend- ants for a time under the form of clientship, 1 was of most importance, but they all possessed a certainjmilitary power, and were primarily destined to maintain public safety in the territory concerned* .and ^above~air"orrtherim- portantriiflperiaL highways traversing it. The upper valley of the Rhone again— that is, the Valais, and the newly conquered Raetia — were placed under a commander of higher standing not in rank, but doubtless in power; a corps, relatively speaking, considerable was here for the time being indispensably requisite. In order, how- ever, to provide for its being diminished as far -as possible, Raetia was in^great measure depopulated by the. removal of its inhabitants. The' circuit was closed by the similarly organised province of Noricum, embracing the largest part of what is now German Austria. This wide and fertile region had submitted without substantial resistance to the Roman rule, probably in the form of a dependent principality emerging in the first instance, but of its prince erelong giving place to the imperial procurator, from whom, for that matter, he did not essentially differ. Some, at all events, of the Rhenish and Danubian legions had their fixed quarters in the immediate neighbourhood, on the one hand of the Raetian frontier at Vindonissa, on the other of the Norican frontier at Poetovio, obviously to keep in check the adjoining province; but in that intermediate region as little were there armies of the first rank with legions under senatorial generals, as there were senatorial 1 The official title of Cottius was position was beyond doubt held for not king, like that of his father life, and, under reservation of the Donnus, but "president of the can- superior's right to confirm it, also tonal union" (praefectus civitatium),&s hereditary ; so far therefore the union he is named on the still standing arch was certainly a principality, as it is of Susa erected by him in honour of usually so termed. 98. Augustus in the year 745-6. But the chap. i. NORTHERN FRONTIER OF ITALY. 19 governors. The distrust towards the corporation governing the state alongside of the emperor finds very forcible expression in this arrangement. Nex t to the p rotection of the peace of Italy the chief Roa^isand aim of this organisation was to secure its communications theAipV" with the north, which were of not less urgent importance for traffic than in a military point of view. With special energy Augustus took up this task ; and he doubtless deserved that his name should still live at the present day in those of Aosta and Augsburg, perhaps also in that of the Julian Alps. The old coast-road, which Augustus partly renewed, partly constructed, from the Ligurian coast through Gaul and Spain to the Atlantic Ocean, can only have served purposes of traffic. The road also over the Cottian Alps, already opened up by Pompeius (iv. 28), w. 27. was finished under Augustus by the already mentioned prince of Susa, and named after him ; in like manner a trading route, it connects Italy, by way of Turin and Susa, with the commercial capital of south Gaul, Arelate. But the military line proper — the direct connection between Italy and the camps on the Rhine — led through the valley of the Dora Baltea from Italy partly to Lyons the capital of Gaul, partly to the Rhine. While the republic had con- fined itself to bringing into its power the entrance of that valley by founding Eporedia (Ivrea), Augustus possessed himself of it entirely by not merely subjugating its inhabi- I tants — the still restless Salassi, with whom he had already fought during the Dalmatian war — but extirpating them \ outright; 36,000 of them, including 8000 fighting men, j were sold under the hammer into slavery in the market- place of Eporedia, and the purchasers were bound not to grant freedom to any of them within twenty years. The camp itself, from which his general Varro Murena had achieved their final defeat in 729, became the fortress, 25. which, occupied by 3000 settlers taken from the imperial guard, was to secure the communications — the town Augusta Praetoria, the modern Aosta, whose walls and gates then erected are still standing. It commanded subsequently two Alpine routes, as well that which led 20 NORTHERN FRONTIER OF ITALY, book vm. over the Graian Alps or Little St. Bernard, along the upper Isere and the Rhone to Lyons, as that which ran over the Poenine Alps, the Great St. Bernard, to the valley of the Rhone and to the Lake of Geneva, and thence into the valleys of the Aar and the Rhine. But it was for the first of these roads that the town was designed, as it originally had only gates leading east and west; nor could this be otherwise, for the fortress was built ten years before the occupation of Raetia ; in those years, moreover, the later organisation of the camps on the Rhine was not yet in existence, and the direct connection between the capitals of Italy and Gaul was altogether of the foremost importance. In the direction of the Danube we have already mentioned the laying out of Emona on the upper Save, on the old trade-road from Aquileia over the Julian Alps into the Pannonian territory. This road was at the same time the chief artery for the military communi- cation of Italy with the region of the Danube. Lastly, with the conquest of Raetia was connected the opening of the route which led from the last Italian town Tridentum (Trent), up the Adige valley, to the newly established Augusta in the land of the Vindelici, the modern Augs- burg, and onward to the upper Danube. Subsequently, when the son of the general who had first opened up this region came to reign, this road received the name of the Claudian highway. 1 It furnished the means of connection, indispensable from a military point of view, between Raetia and Italy ; but in consequence of the comparatively small importance of the Raetian army, 1 We know this road only in the it led to the Danube, is attested ; the shape which the emperor Claudius, connection of the making of this road the son of the constructor, gave to it ; with the founding of Augusta Vindeli- originally, of course, it cannot have cum, though this was at first only a been called via Claudia, but only market-village (forum), is more than via Augusta, and we can hardly re- probable (C. I. L. iii. p. 711); in gard as its terminus in Italy Altinum, what way Augsburg and the Danube in the neighbourhood of the modern were reached from Meran we do not Venice, since, under Augustus, all the know. Subsequently the road was imperial roads still led to Rome, rectified, so as to leave the Adige at That the road ran through the upper Bautzen, and to lead up the Eisach Adige valley is shown by the milestone valley over the Brenner to Augsbure found at Meran ( C. I. L. v. 8003 ) ; that chap. i. NORTHERN FRONTIER OF ITALY. 21 and doubtless also in consequence of the more difficult communication, it never had the same importance as the route of Aosta. The Alpine passes and the north slope of the Alps \ were thus in secure possession of the Romans. Beyond j the Alps there stretched to the east of the Rhine the land i of the Germans ; to the south of the Danube that of the j Pannonians and the Moesians. Here, too, soon after the occupation of Raetia, the offensive was taken, and nearly , contemporaneously in both directions. Let us look first \ at what occurred on the Danube. The Danubian region, to all appearance up to 727 27. administered along with Upper Italy, became then, on the ^ ec - t ^ n of reorganisation of the empire, an independent administra- tive district, Illyricum, under a governor of its own. It consisted of Dalmatia, with the country behind it, as far as the Drin — while the coast farther to the south had for long belonged to the province of Macedonia — and of the Roman possessions in the land of the Pannonians on the Save. The region between the Haemus and the Danube as far as the Black Sea, which Crassus had shortly before brought into dependence on the empire, as well as Noricum and Raetia, stood in a relation of clientship to Rome, and so did not belong as such to this province, but withal were primarily dependent on the governor of Illyricum. Thrace, north of the Haemus, still by no means pacified, fell, from a military point of view, to the same district. It was a continued effect of the original organisation, and one which subsisted down to a late period, that the whole region of the Danube from Raetia to Moesia was compre- hended as a customs-district under the name Illyricum in the wider sense. Legions were stationed only in Illyricum proper, in the other districts there were probably no im- perial troops at all, or at the utmost small detachments ; the chief command was held by the proconsul of the new province coming from the senate ; while the soldiers and officers were, as a matter of course, imperial. It attests the serious character of the offensive beginning after the conquest of Raetia, that in the first instance the co-ruler 22 NORTHERN FRONTIER OF ITALY, book vm. Agrippa took over the command in the region of the Danube, to whom the proconsul of Illyricum had to become de iure subordinate ; and then, when Agrippa's 12. sudden death in the spring of 742 broke down this com- bination, Illyricum in the following year passed into im- perial administration, and the imperial generals obtained the chief commands in it. Soon three military centres were here formed, which thereupon brought about the administrative division of the Danubian region into three parts. The small principalities in the territory conquered by Crassus gave place to the province of Moesia, the governor of which henceforth, in what is now Servia and Bulgaria, guarded the frontier against the Dacians and Bastarnae. In what had hitherto been the province of Illyricum, a part of the legionaries was posted on the Kerka and the Cettina, to keep in check the still troublesome Dalmatians. The chief force was stationed in Pannonia, on what was then the boundary of the empire, the Save. This distribution of the legions and organisation of the provinces cannot be fixed with chronological precision ; probably the serious wars which were waged simultane- ously against the Pannonians and the Thracians, of which we have immediately to speak, led in the first instance to the institution of the governorship of Moesia, and it was not till some time later that the Dalmatian legions and those on the Save obtained commanders-in- chief of their own. First Pan- As the expeditions against the Pannonians and the Germans were, as it were, a repetition of the Raetian campaign on a more extended scale, so the leaders, who were put at their head with the title of imperial legates, were the same — once more the two princes of the imperial house, Tiberius, who, in the place of Agrippa, took up the command in Illyricum, and Drusus, who went to the Rhine, both now no longer inexperienced youths, but men in the prime of their years, and well fitted to take in hand severe work. Immediate pretexts for the waging of war in the region of the Danube were not wanting. Marauders from noman war of Tiberius. chap. i. NORTHERN FRONTIER OF ITALY. 23 Pannonia, and even from the peaceful Noricum, carried pillage in the year 738 as far as Istria. Two years there- 16. after the Illyrian provincials took up arms against their masters, and, although they returned to obedience without offering opposition when Agrippa took over the com- mand in the autumn of 741, yet immediately after his 13. death the disturbances are alleged to have begun afresh. We cannot say how far these Roman accounts correspond to the truth ; certainly the pushing forward of the Roman frontier, required by the general political situation, formed the real motive and aim of the war. As to the three campaigns of Tiberius in Pannonia from 742 to 744 we 12, are very imperfectly informed. Their result was stated by the government as the establishment of the Danube as the boundary for the province of Illyricum. That this river was thenceforth looked upon in its whole course as the boundary of Roman territory, is doubtless correct ; but a subjugation in the proper sense, or even an occupa- tion, of the whole of this wide domain by no means took place at that time. The chief resistance to Tiberius was offered by the tribes already at an earlier date declared Roman, especially by the Dalmatians ; among those first effectively subdued at that time, the most noted was that of the Pannonian Breuci on the lower Save. The Roman armies, during these campaigns, hardly ever crossed the Drave, and did not in any case transfer their standing camp to the Danube. The region between the Save and Drave was at all events occupied, and the headquarters of the Illyrian northern army were transferred from Siscia • on the Save to Poetovio (Pettau) on the middle Danube, while in the Norican region recently occupied the Roman garrisons reached as far as the Danube at Carnuntum (Petronell, near Vienna), at that time the last Norican town towards the east. The wide and vast region between the Drave and the Danube, which now forms western Hungary, was to all appearance at that time not even militarily occupied. This was in keeping with the whole plan of the offensive operations that were begun ; the object sought was to be in touch with the Gallic army, 24 NORTHERN FRONTIER OF ITALY, book vm. and for the new imperial frontier in the north-east the natural base was not Buda, but Vienna. Thracian Complementary in some measure to this Pannonian warofPiso. exped i t j on G f Tiberius was that which was simultaneously undertaken against the Thracians by Lucius Piso, per- haps the first governor that Moesia had of its own. The two great neighbouring nations, the Illyrians and the Thracians, of whom we shall treat more fully in a sub- sequent chapter, stood alike at that time in need of subjugation. The tribes of inland Thrace showed themselves still more obstinate than the Illyrians, and far from sub- 16. ordinate to the kings set over them by Rome ; in 738 a Roman army had to advance thither and come to the help of the princes against the Bessi. If we had more exact accounts of the conflicts waged in the one quarter 13. "• as in the other in the years 741 to 743, the contemporary action of the Thracians and Illyrians would perhaps appear as concerted. Certain it is that the mass of the Thracian tribes south of the Haemus and presumably also those settled in Moesia took part in this national war, and that the resistance of the Thracians was not less obstinate than that of the Illyrians. It was for them at the same time a religious war ; the shrine of Dionysos, 1 taken from the Bessi and assigned to the Odrysian princes well disposed to Rome, was not forgotten ; a priest of this Dionysos stood at the head of the insurrection, and it was directed in the first instance against those Odrysian princes. One of them was taken and put to death, the other was driven • away ; the insurgents, in part armed and disciplined after the Roman model, were victors in the first engagement over Piso, and penetrated as far as Macedonia and into the Thracian Chersonese ; fears were entertained for Asia. 1 The locality "in which the his son (Suetonius, Aug. 94), and Bessi honour the god Dionysos," which Herodotus already mentions and which Crassus took from them (ii. in; compare Euripides, Hec. and gave to the Odrysians (Dio, li. 1267) as an oracular shrine placed 25), is certainly the same Liberi under the protection of the Bessi. fatris Incus, in which Alexander Certainly it is to be sought north- sacrificed, and the father of Augustus, wards of Rhodope ; it has not yet cum per secreta Thraciae exercitum been discovered. duceret, asked the oracle respecting chap. i. NORTHERN FRONTIER OF ITALY. 25 Ultimately, however, Roman discipline gained the superi- ority over these brave opponents ; in several campaigns Piso mastered the resistance, and the command of Moesia, instituted either already on this occasion or soon after- wards on " the Thracian shore," broke up the connection of the Daco-Thracian peoples, by separating the tribes on the left bank of the Danube and their kinsmen south of the Haemus from each other, and permanently secured the Roman rule in the region of the lower Danube. The Germans still more than the Pannonians and Attack of the Thracians gave the Romans occasion to feel that the the Ger " • • r mans. existing state of things could not permanently continue. The boundary of the empire since Caesar's time had been the Rhine from the lake of Constance to its mouth (iv. 2 5 8). It was not a demarcation of peoples, for already iv. 247. of old in the north-east of Gaul the Celts had on various occasions mingled with Germans, the Treveri and Nervii would at least gladly have been Germans (iv. 244), and on iv. 233. the middle Rhine Caesar himself had provided settlements for the remnant of the hosts of Ariovistus — Triboci (in Alsace) Nemetes (about Spires), Vangiones (about Worms). Those Germans on the left of the Rhine indeed adhered more firmly to the Roman rule than the Celtic cantons, and it was not they that opened the gates of Gaul to their countrymen on the right bank. But these, long accustomed to predatory raids over the river and by no means forgetting the half successful attempts on several occasions to settle there, came unbidden. The only Germanic tribe beyond the Rhine, which already in Caesar's time had separated from their countrymen and placed themselves under Roman protection, the Ubii, had to give way before the hatred of their exasperated kinsmen and to seek protection and new abodes on the Roman bank (716) ; Agrippa, although personally present 38- in Gaul, had not been able, amidst the pressure of the Sicilian war then impending, to help them otherwise, and had crossed the Rhine merely to effect their transference. From this settlement of theirs our Cologne subsequently grew up. Not merely were the Romans trading on the 26 NORTHERN FRONTIER OF ITALY, book vm. 25- 1 6. Defeat of Lollius. IS- German war of Drusus. 33- 13- right bank of the Rhine subjected to various injuries by the Germans, so that even in 729 an advance over the Rhine was executed, and Agrippa in 734 had to expel from Gaul Germanic hordes that had come thither from the Rhine ; but in 7 3 8 the further bank was affected by a more general movement, which terminated in an invasion on a great scale. The Sugambri on the Ruhr took the lead, and with them their neighbours the Usipes on the north in the valley of the Lippe, and the Tencteri on the south ; they attacked the Roman traders sojourning among them and nailed them to the cross, then crossed the Rhine, pillaged the Gallic cantons far and wide, and, when the governor of Germany sent the legate Marcus Lollius with the fifth legion against them, they first cut off its cavalry and then put the legion itself to disgraceful flight, on which occasion even its eagle fell into their hands. After all this they returned unassailed to their homes. This miscarriage of the Roman army, though not of importance in itself, was not to be despised in presence of the Germanic movement and even of the troublesome feeling in Gaul ; Augustus himself went to the province attacked, and this occurrence may possibly have been the immediate occasion for the adoption of that great movement of offence, which, beginning with the Raetian war in 739, led on to the campaigns of Tiberius in Illyricum and of Drusus in Germany. Nero Claudius Drusus, born in 716 by Livia in the house of her new husband, afterwards Augustus, and loved and treated by the latter like a son — evil tongues said, as his son — the very image of manly beauty and of winning grace in converse, a brave soldier and an able general, a pronounced panegyrist, moreover, of the old republican system, and in every respect the most popular prince of the imperial house, took up, on the return of Augustus to Italy (741), the administration of Gaul and the chief command against the Germans, whose subjugation was now contemplated in earnest. We have no adequate means of knowing either the strength of the * army then stationed on the Rhine, or how matters stood with the chap. i. NORTHERN FRONTIER OF ITALY. 27 Germans ; this much only is clear that the latter were not in a position suitably to meet the compact attack. The region of the Neckar formerly possessed by the Helvetii (iii. 182), then for long a debateable border-land between Si. 173. them and the Germans, lay desolate and dominated on the one side by the recently subdued district of the Vindelici, on the other side by the Germans friendly to Rome about Strassburg, Spires, and Worms. Farther northward, in the region of the upper Main, were settled the Marcomani, perhaps the most powerful of the Suebian tribes, but from of old at enmity with the Germans of the middle Rhine. Northward of the Main followed first in the Taunus the Chatti, farther down the Rhine the already named Tencteri, Sugambri, and Usipes ; behind them the powerful Cherusci on the Weser, besides a number of tribes of secondary rank. As it was these tribes on the middle Rhine, with the Sugambri at their head, that had carried out that attack on Roman Gaul, the retaliatory expedition of Drusus was directed mainly against them, and they too combined for joint resistance to Drusus and for the institution of a national army to be formed from the contingents of all these cantons. The Frisian tribes, however, on the coast of the North Sea did not join the movement, but persevered in their peculiar isolation. It was the Germans who assumed the offensive. The Sugambri and their allies again seized all the Romans whom they could lay hold of on their bank, and nailed to the cross the centurions among them, twenty in number. The allied tribes resolved once more to invade Gaul, and even divided the spoil beforehand — the Sugambri were to obtain the people, the Cherusci the horses, the Suebian tribes the gold and silver. So they attempted in the beginning of 742 again to cross the Rhine, and hoped'for 12. - the support of the Germans on the left bank of the river, and even for an insurrection of the Gallic cantons just at that time excited by the unwonted matter of the census. But the young general took his measures well ; he nipped the movement in the Roman territory before it was well set agoing, drove back the invaders even as they were 28 NORTHERN FRONTIER OF ITALY, book vm. crossing the river, and then crossed the stream on his own part, in order to lay waste the territory of the Usipes and Sugambri. This was a repulse for the time ; the plan of the war proper, designed on a grander scale, started from the acquisition of the North Sea coast and of the mouths of the Ems and the Elbe. The numerous and valiant tribe of the Batavi in the delta of the Rhine had been incor- porated — to all appearance, at that time and by amicable concert — in the Roman empire ; with its help a communi- cation by water was established from the Rhine to the Zuyder See, and from the latter to the North Sea, which opened up for the Rhine-fleet a safer and shorter way to the mouths of the Ems and Elbe. The Frisians on the north coast followed the example of the Batavi and likewise submitted to the foreign rule. It was doubtless still more the moderate policy than the military preponderance of the Romans, which paved the way for them here ; these tribes remained almost wholly exempt from tribute, and were drawn upon for war-service in a way which did not alarm, but allured them. From this basis the expedition proceeded along the coast of the North Sea ; in the open sea the island of Burchanis (perhaps Borchum off East Friesland) was taken by assault ; on the Ems the fleet of boats of the Bructeri was vanquished by the Roman fleet ; Drusus reached as far as the Chauci at the mouth of the Weser. The fleet indeed on its return homewards encountered dangerous and unknown shallows, and, but for the Frisians affording a safe escort to the shipwrecked army, it would have been in a very critical position. Nevertheless, by this first campaign the coast from the mouth of the Rhine to that of the Weser had been gained for Rome. After the coast was thus acquired, the subjugation of the" interior began in the next year (743). It was materially facilitated by the dissensions among the Germans of the middle Rhine. For the attack on Gaul attempted in the previous year the Chatti had not furnished the promised contingent ; in natural, but still far from politic, anger the Sugambri had suddenly assailed the land of the Chatti with all their force, and so their own territory as chap. i. NORTHERN FRONTIER OF ITALY. 29 well as that of their next neighbours on the Rhine was occupied without difficulty by the Romans. The Chatti thereupon submitted to the enemies of their enemies without resistance ; nevertheless, they were directed to evacuate the bank of the Rhine and to occupy instead of it that district which the Sugambri had hitherto possessed. Not less did the powerful Cherusci farther inland on the middle Weser succumb. The Chauci settled on the lower stream were now assailed by land as they had been before by sea ; and thus the whole territory between the Rhine • and Weser was taken possession of, at least at the places of decisive military importance. The return was certainly, just as in the previous year, on the point of being almost fatal ; at Arbalo (site unknown) the Romans found them- selves surrounded on all sides in a narrow defile by the Germans and deprived of their communications ; but the firm discipline of the legions, and the arrogant confidence of success withal on the part of the Germans, changed the threatened defeat into a brilliant victory. 1 In the next year (744) the Chatti revolted, indignant at the loss 10. of their old beautiful home ; but now they for their part remained alone, and were, after an obstinate resistance, and not without considerable loss, subdued by the Romans (745). The Marcomani on the upper Main, who after 9- the occupation of the territory of the Chatti were next exposed to the attack, gave way before it, and retired into the land of the Boii, the modern Bohemia, without inter- fering from this point, where they were removed beyond the immediate sphere of the Roman power, in the conflicts on the Rhine. In the whole region between the Rhine and Weser the war was at an end. Drusus was able in 745 to set foot on the right bank of the Weser in 9- the canton of the Cherusci, and to advance thence to the Elbe, which he did not cross, and presumably was in- structed not to do so. Several severe combats took place ; successful resistance was nowhere offered. But on Death of the return-march, which led apparently up the Saale and 1 That the battle at Arbalo (Plin. shown by Obsequens, 72, and so the H.N.sX. 17, 55) belongs to this year, is narrative in Dio, liv. 33, applies to it. 3 o NORTHERN FRONTIER OF ITALY, book vm. Continu- ance of the war by- Tiberius. 8, 7- thence to the Weser, a severe blow befell the Romans, not through the enemy but through an incalculable misfor- tune. The general fell with his horse and broke his thigh-bone ; after thirty days of suffering he expired in the distant land between the Saale and Weser, 1 which had never before been trodden by a Roman army, in the arms of his brother who had hastened thither from Rome, in the thirtieth year of his age and in the full consciousness of his vigour and of his successes, long and deeply lamented by his adherents and the whole people — perhaps to be pronounced fortunate, because the gods granted to him to depart from life young, and to escape the disillusions and embitterments which tell most painfully on those highest in station, while his brilliant and heroic figure continues still to live in the remembrance of the world. In the course of things, as a whole, the death of the able general made — as might be expected — no change. His brother Tiberius arrived early enough not merely to close his eyes, but also with his firm hand to bring the army back and to carry on the conquest of Germany. He commanded there during the two following years (746, 747), in the course of which there were no con- flicts on a larger scale, but the Roman troops showed themselves far and wide between the Rhine and Elbe, and where he died) ; we may be allowed to seek it in the region of the Weser. The dead body was then conveyed to the winter-camp (Dio, lv. 2) and there burnt; this spot was regarded, according to Roman usage, also as the place of burial, although the depositing of the ashes took place in Rome, and to this is to be referred the honorarius tumulus with the annual obsequies (Sueton. /. c). Probably we have to seek for this place at Vetera. When a later author (Eutropius, vii. 13) speaks of the monumentum of Drusus at Mentz, this is doubtless not the tomb, but the elsewhere mentioned Tropaeum (Floras, ii. 30 : Marcoman- orum spoliis et insignibus quendam editum tumulum in tropaei modum excoluif). 1 That the fall of Drusus took place in the region of the Saale we may be allowed to infer from Strabo, vii. I, 3, p. 291,' although he only says that he perished on the march between Salas and Rhine, and the identification of the Salas with the Saale rests solely on the resemblance of name. From the scene of the mishap he was then transported as far as the summer camp (Seneca, Cons, ad Marciam 3 : ipsis ilium hostibns aegrum cum vene- ratione el pace mutua prosequentibus nee optare quod expediebat audentibus), and in that camp he died (Sueton., Claud. 1). This camp lay in the heart of the barbarian land (Valerius Max. v. 5. 3) and not veiy far from the battlefield of Varus (Tacitus, Ann. ii. 7> where the vetus ara Druso sita is certainly to be referred to the place chap. i. NORTHERN FRONTIER OF ITALY. 31 when Tiberius made the demand that all the countries should formally acknowledge the Roman rule, and at the same time declared that he could only accept that acknowledgment from all the cantons simultaneously, they complied without exception ; last of all the Sugambri, for whom indeed -there was no real peace.' What pro- gress in a military point of view had been made, is shown by the expedition, falling a little later, of Lucius Domitius Ahenobarbus. The latter was able, as governor of Illyricum, probably from Vindelicia as a basis, to assign to a restless horde of Hermunduri settlements in the land of the Marcomani itself; and on this expedition he reached as far as, and beyond, the upper Elbe, without meeting with resistance. 1 The Marcomani in Bohemia were completely isolated, and the rest of Germany between the Rhine and Elbe was a Roman province — though still by no means reduced to tranquillity. Of the military-political organisation of Germany, as camp on at that time planned, we have but a very imperfect ?^ eft knowledge, because, on the one hand, there is an utter Rhine. want of accurate information as to the arrangements made in earlier times to protect the Gallic eastern frontier, and, on the other hand, those made by the two brothers were in great part destroyed by the subsequent develop- ment of affairs. There was no attempt to move the Roman frontier- guard away from the Rhine ; to this matters might perhaps come, but they had not yet done so. Just as was the case in Illyricum at that time with the Danube, the Elbe was doubtless the political boundary of the empire, but the Rhine was the line of frontier- defence, and from the camps on the Rhine the connections in rear ran to the great towns of Gaul and to its ports. 2 1 What we learn from Dio, lv. 10, Bohemia itself, which would involve partly confirmed by Tacitus, Ann. iv. still greater difficulties, is not required 44, cannot be apprehended otherwise, by the narrative. Noricum and Raetia must have been 2 To a connection in rear of the put under this governor as an excep- camp on the Rhine with the port of tional measure, or the course of Boulogne we might perhaps take the operations induced him to pass beyond much disputed notice of Florus, ii. 30, the limit of his governorship. The to refer : Bonnam (or Bormam) et Ges- assumption that he marched through soriacum pontibus iunxit classibusque 32 NORTHERN FRONTIER OF ITALY, book vm. The great headquarters during these campaigns was what was afterwards named the " Old Camp," Castra vetera, (Birten near Xanten), the first considerable height below Bonn on the left bank of the Rhine, from a military point of view corresponding nearly to the modern Wesel on the right. This place, occupied perhaps since the beginning of the Roman rule on the Rhine, had been instituted by Augustus as a stronghold for curbing Germany ; and, if the fortress was at all times the basis for the Roman defensive on the left bank of the Rhine, it was not less well chosen for the invasion of the right, situated, as it was, opposite to the mouth of the Lippe which was navigable far up, and connected with the right bank by a strong bridge. The counterpart to this " Old Camp," at the mouth of the Lippe was probably formed by that at the mouth of the Main, Mogontiacum, the modern Mentz, to all appearance a creation of Drusus ; at least the already mentioned cessions of territory imposed on the Chatti, as well as the constructions in the Taunus to be mentioned further on, show that Drusus clearly perceived the military importance of the line of the Main, and thus also that of its key on the left bank of the Rhine. If the legionary camp on the Aar was, as it would seem, instituted to keep the Raeti and Vindelici to their obedience (p. 1 8), it may be presumed to have been laid out about this time ; but then it had merely an out- ward connection with the Gallico-German military arrange- ments. The legionary camp at Strassburg hardly reaches back to so early a time. The line from Mentz to Wesel formed the basis of the Roman military dispositions. That Drusus and Tiberius had — apart from the Narbonese province which was then no longer imperial — the gover- norship of all Gaul as well as the command of all the firmavit, with which is to be com- safest land - route between the two pared the mention by the same author stations for the fleet available for of forts on the Maas. Bonn may transport, though the writer, prob- reasonably have been at that time ably bent on striking effect, awakens the station of the Rhine-fleet ; Bou- by his pointed mode of expression logne was in later times still a fleet- conceptions which cannot be in that station. Drusus might well have form correct, occasion to make the shortest and chap. i. NORTHERN FRONTIER OF ITALY. 33 Rhenish legions, is an ascertained point ; apart from these princes, the civil administration of Gaul may at that time perhaps have been separated from the com- mand of the troops on the Rhine, but scarcely was the latter thus early divided into two co-ordinate commands. 1 Correlative to these military arrangements on the left Positions bank of the Rhine were those adopted on the right. In tankVfthe the first place the Romans took possession of the right Rhine. bank itself. This step affected above all the Sugambri, in whose case certainly retaliation for the captured eagle and the crucified centurions contributed to it. The envoys sent to declare their submission, the most eminent men of the nation, were, at variance with the law of nations, treated as prisoners of war, and perished miserably in the Italian fortresses. Of the mass of the people, 40,000 were removed from their homes and settled on the shores of Gaul, where they subsequently, perhaps, meet us under the name of the Cugerni. Only a small and harmless remnant of the powerful tribe was allowed to remain in their old abodes. Suebian bands were also transferred to Gaul, other tribes were pushed farther into the interior, such as the Marsi and doubtless also the Chatti ; on the middle Rhine the native population of the right bank was everywhere dislodged or at any rate weakened. Along this bank of the Rhine, moreover, fortified posts, fifty in number, were instituted. In front of Mogontiacum the territory taken from the Chatti, thenceforth the canton of the Mattiaci in what is now Wiesbaden, was brought within the Roman lines, and the height of Taunus strongly 1 As to the administrative parti- under that assumption ; here, doubt- tion of Gaul there is, apart from the less, the hiberna inferiora appear, separation of the Narbonensis, an viz. that of Vetera (Velleius, ii. 120), utter absence of accounts, because it and the counterpart to it, the superi- rested only on imperial ordinances, ora, can only have been formed by and nothing in reference to it came that of Mentz ; but this was not into the records of the senate. But under a colleague of Varus, but under the first information of the existence his nephew, who was thus subordi- of separate Upper and Lower German nate to him in command. Probably commands is furnished by the cam- the partition only took place, in paigns of Germanicus, and the battle consequence of the defeat, in the last of Varus can hardly be understood years of Augustus. VOL. I. 3 NORTHERN FRONTIER OF ITALY, book vm. fortified. 1 But above all the line of the Lippe was taken possession of from Vetera ; of the two military roads furnished at intervals of a day's march with forts, on the two banks of the river, the one on the right bank at least is as certainly the work of Drusus as the fortress of Aliso in the district of the sources of the Lippe, probably the present village of Elsen, not far from Paderborn, 2 is attested to have been so. Moreover, there was the already mentioned canal from the mouth of the Rhine to the Zuider See, and a dyke drawn by Lucius Domitius Ahenobarbus through the marshy flat country between the Ems and the lower Rhine— the so-called " long bridges." Besides, there were detached Roman posts scattered through the whole region ; such are subsequently mentioned among the Frisians and the Chauci, and in this sense it may be correct that the Roman garrisons reached as far as the Weser and the Elbe. Lastly, the army encamped in winter, no doubt, on the Rhine ; but in summer, even though no 1 The praesidium constructed by Drusus in monte Tauno (Tacitus, Ann. i. 56), and the potipiov h X&ttois 7rap' avrQ tq 'PtJ^ associated with Aliso (Dio, liv. 33), are probably identical, and the special position of the canton of the Mattiaci is evidently connected with the construction of Mogontiacum. 2 That the " fort at the confluence of the Lupias and the Helison," in Dio, liv. 33, is identical with the oftener mentioned Aliso, and this must be sought on the upper Lippe, is subject to no doubt ; and that the Roman winter -camp at the sources of the Lippe {ad caput Lupiae, Velleius, ii. 205), the only one of the kind, so far as we know, on German ground, is to be sought just there, is at least very probable. That the- two Roman roads running along the Lippe, and their fortified places of bivouac, led at least as far as the region of Lippstadt, the researches of Hblzermann in par- ticular have shown. The upper Lippe has only one confluent of note, the Alme, and as the village of Elsen lies not far from where the Alme falls into the Lippe, some weight may be here assigned to the similarity of name. To the view, supported among others by Schmidt, which places Aliso at the confluence of the Glenne (and Liese) with the Lippe, the chief ob- jection is that the camp ad caput Lupiae must then have been different from Aliso, and in general this point lies too far from the line of the Weser, while from Elsen the route leads directly through the Doren defile into the Werra valley. Schmidt, who does not adhere to the identification of Aliso and Elsen, remarks generally ( Westfdlische Zeitschrift fur Gesch. und Alterthiwiskunde, xx. p. 259), that the heights of Weser (not far from Elsen), and generally the left margin of the valley of the Alme, are the centre of a semicircle formed by the mountains in front, and this high- lying, dry region, allowing an exact look-out as far as the mountains, which covers the whole country of the Lippe and is itself covered in front by the Alme, is well adapted for the starting-point of a march towards the Weser. chap. i. NORTHERN FRONTIER OF ITALY. 35 expeditions properly so called were undertaken, uniformly in the conquered country, as a rule near Aliso. The Romans, however, did not make mere military Organisa- arrangements in the newly acquired domain. The Ger- tion . ofthe ,,.,,., province mans were urged, like other provincials, to have law Germany. administered to them by the Roman governor, and the summer expeditions of the general gradually developed into the usual judicial circuits of the governor. The accusation and defence of the accused took place in the Latin language ; the Roman advocates and legal assessors began, on the right as on the left side of the Rhine, their operations, sorely felt everywhere, but here deeply exasper- ating to the barbarians, who were unaccustomed to such things. Much was lacking to the full carrying out of the provincial organisation ; a formal assessment of taxa- tion, a regulated levy for the Roman army, were not yet thought of. But as the new cantonal union had just been instituted in Gaul in connection with the divine adoration of the monarch there introduced, a similar arrangement was made also in the new Germany. When Drusus consecrated for Gaul the altar of Augustus at Lyons, the Germans last settled on the left bank of the Rhine, the Ubii, were not received into this union ; but in their chief place, which, as regards position, was for Germany nearly what Lyons was for the three Gauls, a similar altar for the Germanic cantons was erected, the priesthood of which was, in the year 9, administered by the young Cheruscan prince Segimundus, son of Segestes. Political differences, however, in the imperial family Retirement broke down or interrupted the full military success. The from'the" 15 discord between Tiberius and his stepfather led to the chief com- former resigning the command in the beginning of 748. 6 _ The dynastic interest did not allow comprehensive military operations to be entrusted to other generals than princes of the imperial house ; and after the death of Agrippa and Drusus, and the retirement of Tiberius, there were no able generals in that house. Certainly in the ten years, when governors with the ordinary powers bore sway in Illyricum and in Germany, the military operations 36 NORTHERN FRONTIER OF ITAL Y. book viii. there may not have undergone so complete an interrup- tion as they appear to us to have done, seeing that tradition, with its courtly colouring, does not in its report deal out equal measure to campaigns conducted by, and to those conducted without, princes ; but the arrest laid on them was unmistakable, and this itself was a retrogres- sion. Ahenobarbus, who, in consequence of his alliance by marriage with the imperial house — his wife was the daughter of a sister of Augustus — had greater freedom of action than other officers, and who in his Illyrian governorship had crossed the Elbe without encountering resistance, afterwards as governor of Germany reaped no laurels there. Not merely the exasperation, but the courage also, of the Germans was again rising, and in the year 2 the country appears again in revolt, the Cherusci and the Chauci under arms. Meanwhile at the imperial court death had interposed, and the removal of the young sons of Augustus had reconciled the latter and Tiberius. Tiberius Scarcely was this reconciliation sealed by his adoption commander as a son anc ' proclaimed (4), when Tiberius resumed in chief. the work where it had been broken off, and once more in this and in the two following summers (5-6) led the armies over the Rhine.. It was a repetition of, and an advance upon, the earlier campaigns. The Cherusci were brought back to allegiance in the first campaign, the Chauci in the second ; the Cannenefates, adjoining the Batavi, and not inferior in bravery, the Bructeri, settled in the region of the sources of the Lippe and on the Ems, and various other cantons, submitted, as did also the powerful Langobardi, here first mentioned, dwelling at that time between the Weser and Elbe. The first campaign led over the Weser into the interior ; in the second at the Elbe itself the Roman legions confronted the Germanic general levy on the other bank. From the year 4 to 5 the Roman army took up, apparently for the first time, its winter quarters on German soil at Aliso. All this was attained without any considerable conflicts ; the circumspect conduct of the war did not break resist- ance, but made it impossible. This general aimed, not chap. i. NORTHERN FRONTIER OF ITALY. 37 at unfruitful laurels, but at lasting success. The naval expedition, too, was repeated ; like the first campaign of Drusus, the last of Tiberius was distinguished by the navigating of the North Sea. But the Roman fleet this time advanced farther ; the whole coast of the North Sea, as far as the promontory of the Cimbri, that is, the extremity of Jutland, was explored by it, and it then, sailing up the Elbe, joined the land-army stationed on the latter. The emperor had expressly forbidden the crossing of the river ; but the tribes beyond the Elbe — the Cimbri just named, in what is now Jutland, the Charudes to the south of them, the powerful Semnones between the Elbe and the Oder — were brought at least into relation to the new neighbours. It might have been thought that the goal was reached. Campaign But one thing was still wanting to the establishment of ^faro-' the iron ring which was to surround the Great Germany ; boduus. it was the establishment of a connection between the middle Danube and the upper Elbe — the occupation of the old home of the Boii, which with its mountain -cincture planted itself like a gigantic fortress between Noricum and Germany. The king Maroboduus, of noble Marco- mafrian lineage, but in his youth by prolonged residence in Rome introduced to its firmer military and political organisation, had after his return home — perhaps during the first campaign of Drusus and the transmigration, thereby brought about, of the Marcomani from the Main to the upper Elbe — not merely raised himself to be prince of his people, but had also moulded his rule not after the loose fashion of the Germanic kings, but, one might say, after the model of the Augustan. Besides his own people, he ruled over the powerful tribe of the Lugii (in what is now Silesia), and the body of his clients must have extended over the whole region of the Elbe, as the Langobardi and the Semnones are described as subject to him. Hitherto he had observed entire neutrality in presence of the other Germans as of the Romans. He gave perhaps to the fugitive enemies of the Romans an asylum in his country, but he did not actively mingle in 38 NORTHERN FRONTIER OF ITALY, book vm. the strife, not even when the Hermanduri had settlements assigned to them by the Roman governor on Marcomanian territory (p. 31), and when the left bank of the Elbe became subject to the Romans. He did not submit to them, but he bore all these occurrences without inter- rupting, on that account, his friendly relations with the Romans. By this certainly not magnanimous and scarce even so much as prudent policy, he had gained this much, that he was the last to be attacked ; after the completely successful Germanic campaigns in the years 4 and 5 his turn came. From two sides — from Germany and Noricum — the Roman armies advanced against the Bohemian mountain - circle ; Gaius Sentius Saturninus, advanced up the Main, clearing the dense forests from Spessart to the Fichtelgebirge with axe and fire ; while Tiberius in person, starting from Carnuntum, where the Illyrian legions had encamped during the winter of the years 5-6, advanced against the Marcomani. The two armies, amounting together to twelve legions, were even in number so superior as almost to double that of their opponents, whose fighting force was estimated at 70,000 infantry and 4000 horsemen. The cautious strategy of the generals seemed on this occasion also to have quite ensured success, when a sudden incident interrupted the farther advance of the Romans. Pannonian The Dalmatian tribes and the Pannonians, at least insun-ec- of the region of the Save, for a short time obeyed the Roman governors ; but they bore the new rule with an ever increasing grudge, above all on account of the taxes, to which they were unaccustomed, and which were relent- lessly exacted. When Tiberius subsequently asked one of the leaders as to the grounds of the revolt, he answered that it had taken place because the Romans set not dogs and shepherds, but wolves, to guard their flocks. Now the legions from Dalmatia were brought to the Danube, and the men capable of arms were called out, in order to be sent thither to reinforce the armies. These troops made a beginning, and took up arms not for, but against, Rome. Their leader was one of the Daesitiatae (around tion chap. i. NORTHERN FRONTIER OF ITALY. 39 Serajevo), Bato. The example was followed by the Pannonians, under the leadership of two Breuci, another Bato and Pinnes. All Illyricum rose with unheard of rapidity and unanimity. The number of the insurgent forces was estimated at 200,000 infantry and 9000 horsemen. The levy for the auxiliary troops, which had taken place more especially among the Pannonians to a considerable extent, had diffused more widely a knowledge of Roman warfare, along with the Roman language and even Roman culture. Those who had served as Roman soldiers formed now the nucleus of the insurrection. 1 The Roman citizens settled or sojourning in large number in the insurgent regions, the merchants, and above all, the soldiers, were everywhere seized and slain. The independent tribes, as well as those of the provinces, entered into the movement. The princes of the Thracians, entirely devoted to the Romans, certainly brought their considerable and brave bands to the aid of the Roman generals ; but from the other bank of the Danube the Dacians, and with them the Sarmatae, broke into Moesia. The whole wide region of the Danube seemed to have conspired to put an abrupt end to the foreign rule. The insurgents were not disposed to await attack, but planned an invasion of Macedonia, and even of Italy. The danger was serious ; the insurgents might, by cross- ing the Julian Alps, stand in a few days once more before Aquileia and Tergeste — they had not yet forgotten the way thither — -and in ten days before Rome, as the emperor himself expressed it in the senate, to make sure at all events of its assent to the comprehensive and urgent military preparations. In the utmost haste new forces were raised, and the towns more immediately 1 This and not more is what same phenomena as are met with in Velleius says (ii. 1 10) : in omnibus the case of the Cherascan princes, Pannoniis non disciflinae (=military only in increased measure ; and they training) tantummodo, sed linguae quo- are quite intelligible when we bear in que notitia Ro'manae, plerisque etiam mind the Pannonian and Breucian litterarum usus et familiaris animo- alae and cohortes raised by Augustus. rum erat exercitatio. These are the 40 NORTHERN FRONTIER OF ITALY, book vm. threatened were provided with garrisons ; in like manner whatever troops could be dispensed with were despatched to the threatened points. The first to arrive at the spot was the governor of Moesia, Aulus Caecina Severus, and with him the Thracian king Rhoemetalces ; soon other troops followed from the transmarine provinces. But above all Tiberius was obliged, instead of penetrating into Bohemia, to return to Illyricum. Had the insurgents waited till the Romans were engaged in the struggle with Maroboduus, or had the latter made common cause with them, the position might have been a very critical one for the Romans. But the former broke ground too early, and the latter, faithful to his system of neutrality, condescended just at this time to conclude peace with the Romans on the basis of the status quo. Thus Tiberius had, no doubt, to send back the Rhine -legions, because Germany could not possibly be denuded of troops, but he could unite his Illyrian army with the troops arriving from Moesia, Italy, and Syria, and employ it against the insurgents. In fact the alarm was greater than the danger. The Dalmatians, indeed, broke repeatedly into Macedonia and pillaged the coast as far as Apollonia ; but there was no invasion of Italy, and the fire was soon confined to its original hearth. Nevertheless, the work of the war was not easy ; here, as everywhere, the renewed overthrow of the subjects was more laborious than the subjugation itself. Never in the Augustan period had such a body of troops been united under the same command ; already in the first year of the war the army of Tiberius consisted of ten legions along with the corresponding auxiliary forces, and in addition numerous veterans who had again joined of their own accord and other volunteers, together about I 20,000 men ; later he had fifteen legions united under his banners. 1 In the first campaign (6 A.D.) the contest was 1 .If we assume that of the twelve form the army there, the Illyrian legions who were on the march army of Tiberius numbered seven, against Maroboduus (Tacitus, Ann. ii. and the number of ten (Velleius ii. 46), as many as we find soon after 113) may fairly be referred to 'the in Germany, that is, five, went to contingents from Moesia and Italy, chap. i. NORTHERN FRONTIER OF ITALY. 41 waged with very varying fortune ; the large places, like Siscia and Sirmium, were successfully protected against the insurgents, but the Dalmatian Boto fought as ob- stinately and in part successfully against the governor of Pannonia, Marcus Valerius Messalla, the orator's son, as his Pannonian namesake against Aulus Caecina governor of Moesia. The petty warfare above all gave much trouble to the Roman troops. Nor did the following year (7), in which along with Tiberius his nephew the young Germanicus appeared on the scene of war, put an end to the ceaseless conflicts. It was not till the third campaign (8) that the Romans succeeded in subduing in the first instance the Pannonians, chiefly, as it would seem, through the circumstance that their leader, gained over by the Romans, induced his. troops all and sundry to lay down their arms at the river Bathinus, and surrendered his colleague in the supreme command, Pinnes, to the Romans, for which he was recognised by them as prince of the Breuci. Punishment indeed soon befell the traitor ; his Dalmatian namesake caught him and had him executed, and once more the revolt blazed up among the Breuci ; but it was speedily extinguished again, and the Dalma- tian was confined to the defence of his own home. There Germanicus and other leaders of division had in this, as in the following year (9), to sustain vehement conflicts in the several cantons ; in the latter year the Pirustae (on the borders of Epirus) and the canton to which the leader himself belonged, the Daesitiatae, were subdued, one bravely defended stronghold being reduced after another. Once more in the course of the summer Tiberius himself took the field, and set in motion all his fighting force against the remains of the insurrection. Even Bato, shut up by the Roman army in the strong Andetrium (Much, above that of fifteen to the contingents from legions brought up by A. Caecina and Egypt or Syria, and to the further Plautius Silvanus ex transmarinis pro- levies in Italy, whence the newly vinctis ; firstly, the transmarine troops raised legions went no doubt to Ger- could not be at once on the spot, and many, but those thereby relieved went secondly, the legions of Caecina were to the army of Tiberius. Velleius of course the Moesian. Comp. my (ii. 112) speaks inaccurately, at the commentaiy on the Man. Ancyr. 2d very beginning of the war, of five ed. p. 7 r ■ tt\iov KolvktI\i.oi> S^frou vlbv Qi&p enne Afrique, ii. p. 44, comp. p. 52). \ov\ ■Ka.at\% ApcT7J[s Ivejca]. The base which once supported the chap. i. NORTHERN FRONTIER OF ITALY. 45 The position of the army was what was then the normal one. There were at least five legions in the pro- vince, two of which had their winter-quarters at Mogon- tiacum, three in Vetera or else in Aliso. The latter had taken up their summer encampment in the year 9 on the Weser. The natural route of communication from the upper Lippe to the Weser leads over the low chain of heights of the Osnirig and of the Lippe Forest, which separates the valley of the Ems from that of the Weser, through the Doren defile into the valley of the Werra, which falls into the Weser at Rehme, not far from Minden. Here therefore, approximately, the legions of Varus at that time were encamped. As a matter of course this summer camp was connected with Aliso, the base of the Roman position on the right bank of the Rhine, by a road supplied with depots. The good season of the year came to its close, and they were making ready for the return march, when the news came that a neighbouring canton was in revolt ; and Varus resolved, instead of lead- ing back the army by that depot-route, to take a circuit and by the way to bring back the rebels to allegiance. 1 So they set out ; the army consisted, after numerous detach- ments, of three legions and nine divisions of troops of the second class, together about 20,000 men. 2 When the 1 The report of Dio, the only one the storming of the camp are both which hands down to us a somewhat known to the better tradition, and connected view of this catastrophe, that in their causal connection. The explains the course of it sufficiently, ridiculous representation of the Ger- if we only take further into account mans breaking in at all the gates into — what Dio certainly does not bring the camp, while Varus is sitting on. into prominence — the general re- the judgment-seat and the herald is lation of the summer and winter summoning the parties before him, is camps, and thereby answer the ques- not tradition, but a picture manufac- tion justly put by Ranke ( Wetige- tured from it. That this is in utter schichle, iii. 2, 275), how the whole antagonism to the description by army could have marched against a Tacitus of the three bivouacs, as well local insurrection. The narrative of as to sound reason, is obvious. Florus by no means rests on sources 2 The normal strength of the three originally different, as that scholar alae and the six cohorles is not to be assumes, but simply on the dramatic calculated exactly, inasmuch as among accumulation of motives for action, them there may have been double divi- such as is characteristic of all his- sions {miliaria/) ; but the army cannot torians of this type. The peaceful have numbered much over 20,000 dispensing of justice by Varus and men. On the other hand, there ap- 46 NORTHERN FRONTIER OF ITALY, book vm. army had removed to a sufficient distance from its line of communication, and penetrated far enough into the path- less country, the confederates in the neighbouring cantons rose, cut down the small divisions of troops stationed among them, and broke forth on all sides from the defiles and woods against the army of Varus on its march. Arminius and the most notable leaders of the patriots had remained to the last moment at the Roman headquarters to make Varus secure. On the very evening before the day on which the insurrection burst forth they had supped in the general's tent with Varus ; and Segestes, when announcing the impending outbreak of the revolt, had adjured the general to order the immediate arrest of himself as well as of the accused, and to await the justi- fication of his charge by the facts. The confidence of Varus was not to be shaken. Arminius rode away from table to the insurgents, and was next day before the ramparts of the Roman camp. The military situation was neither better nor worse than that of the army of Drusus before the battle at Arbalo, and than had, under similar circumstances, often been the plight of Roman armies. The communications were for the moment lost ; the army, encumbered with heavy baggage in a pathless country and at a bad rainy season in autumn, was separated by several days'' march from Aliso ; the assailants were beyond doubt far superior in number to the Romans. In such cases it is the solid quality of the troops that is de- cisive ; and, if the decision here for once was unfavourable to the Romans, the result was doubtless mostly due to the inexperience of the young soldiers, and especially to the want of head and of courage in the general. After the attack took place the Roman army continued its march, now beyond doubt in the direction of Aliso, amidst constantly increasing pressure and increasing demoralisa- tion. Even the higher officers failed in part to do their pears no reason for assuming a account for the comparatively small material difference of the effective number of the anxilia, which were strength from the normal. The always by preference employed for numerous detachments which are this duty, mentioned (Dio, lvi. 19) serve to chap. i. NORTHERN FRONTIER OF ITALY. 47 duty ; one of them rode away from the field of battle with all the cavalry, and left the infantry to sustain the conflict alone. The first to despair utterly was the general himself ; wounded in the struggle, he put himself to death before the matter was finally decided, so early indeed, that his followers still made an attempt to burn the dead body and to withdraw it from being dishonoured by the enemy. A number of the superior officers followed his example. Then, when all was lost, the leader that was left sur- rendered, and thereby put out of his own power what re- mained open to these last — an honourable soldier's death. Thus perished the Germanic army in one of the valleys of the mountain-range that bounds the region of Miinster, in the autumn of the year 9 A.D. 1 The eagles fell — all three 1 As Germanicus, coming from the Ems, lays waste the territory between the Ems and Lippe, that is, the region of Miinster, and not far from it lies the Teutoburgiensis saltus, where Varus's army perished (Tacitus, Ann. i. 61), it is most natural to understand this description, which does not suit the flat Miinster region, of the range bounding the Miinster region on the north-east, the Osning ; but it may also be deemed applicable to the Wiehen mountains somewhat farther to the north, parallel with the Osning, and stretching from Minden to the source of the Hunte. We do not know at what point on the Weser the summer camp stood ; but in ac- cordance with the position of Aliso near Paderborn, and with the con- nections subsisting between this and the Weser, it was probably some- where near Minden. The direction of the march on the return may have been any other excepting only the nearest way to Aliso ; and the catas- trophe consequently occurred not on the military line of communication between Minden and Paderborn itself, but at a greater or less distance from it. Varus may have marched from Minden somewhat in the direction of Osnabriick, then after the attack have attempted from thence to reach Paderborn, and have met with his end on this march in one of those two ranges of hills. For centuries there have been found in the dis- trict of Venne at the source of the Hunte a surprisingly large number of Roman gold, silver, and copper coins, such as circulated in the time of Augustus, while later coins hardly occur there at all (comp. the proofs in Paul Hofer, der Feldztig des Germani- cus imjahre 16, Gotha, 1884, p. 82, f.) The coins thus found cannot belong to one store of coins on account of their scattered occurrence and of the difference of metals, nor yet to a seat of traffic on account of their proximity as regards time ; they look quite like the leavings of a great extirpated army, and the accounts before us as to the battle of Varus may be reconciled with this locality. As to the year of the catastrophe there should never have been any dispute ; the shifting of it to the year 10 is « mere mis- take. The season of the year is in some measure determined by the fact that between the arrangement to cele- brate the Illyrian victory and the arrival of the unfortunate news in Rome there lay only five days, and that arrangement probably had in view the victory of 3d Aug., though it did not immediately follow on the latter. Accordingly the defeat must have taken place somewhere in Sep- 48 NORTHERN FRONTIER OF ITAL Y. . book viii. of them — into the enemy's hand. Not a division cut its way through, not even those horsemen who had left their comrades in the lurch ; only a few who were isolated and dispersed were able to effect their escape. The captives, especially the officers and the advocates, were fastened to the cross, or buried alive, or bled under the sacrificial knife of the German priests. The heads cut off were nailed as a token of victory to the trees of the sacred grove. Far and wide the land rose against the foreign rule ; it was hoped that Maroboduus would join the movement ; the Roman posts and roads on the whole right bank of the Rhine fell without further trouble into the power of the victors. Only in Aliso, the brave commandant Lucius Caedicius, not an officer, but a veteran soldier, offered a resolute re- sistance, and his archers were enabled to make the en- campment before the walls so annoying to the Germans, who possessed no weapons for distant fighting, that they converted the siege into a blockade. When the last stores of the besieged were exhausted, and still no relief came, Caedicius broke up one dark night ; and this remnant of the army, though burdened with numerous women and children, and suffering severe losses through the assaults of the Germans, in reality ultimately reached the camp at Vetera. Thither also the two legions stationed in Mentz under Lucius Nonius Asprenas had gone on the news of the disaster. The resolute defence of Aliso, and the rapid intervention of Asprenas, hindered the Germans from following up the victory on the left bank of the Rhine, and perhaps the Gauls from rising against Rome. Tiberius The defeat was soon compensated, in so far as the Rune "* 6 Rnine armv was immediately not simply made up to its strength, but considerably reinforced. Tiberius once more took up the supreme command, and though for the year following on the battle of Varus (10) the history of the war had no combats to record, it is probable that arrange- ments were then made for the occupation of the Rhine- tember or October, which also ac- the march back from the summer to cords with the circumstance that the the winter camp, last march of Varus was evidently chap. i. NORTHERN FRONTIER OF ITALY. 49 frontier by eight legions, and simultaneously for the division of this command into that of the upper army, with Mentz as its headquarters, and that of the lower with the head- quarters at Vetera, an arrangement, as a whole, which thereupon remained normal for centuries. It could not but be expected that this increase of the army of the Rhine would be followed by the energetic resumption of operations on the right bank. The Romano -German conflict was not a conflict between two powers equal in the political balance, in which the defeat of the one might justify the conclusion of an unfavourable peace ; it was the conflict of a great civilised and organised state against a brave but, in a political and military aspect, barbarous nation, in which the ultimate result was settled from the first, and an isolated failure in the plan as sketched might as little produce any change as the ship gives up its voyage because a gust of wind drives it out of its course. But it was otherwise. Tiberius, doubtless, went across the Rhine in the following year (11), but this expedition did not resemble the former one. He remained during the summer on that side, and celebrated the emperor's birth- day there, but the army kept to the immediate neighbour- hood of the Rhine, and of expeditions on the Weser and on the Elbe there was nothing said. Evidently the object was only to show to the Germans that the Romans still knew how to find the way into their country, and per- haps also to make such arrangements on the right bank of the Rhine as the change of policy required. The great command embracing both armies was re- Germani tained, and retained accordingly in the imperial house. Germanicus had already exercised it in the year 1 1 along with Tiberius ; in the following year ( 1 2), when the administration of the consulate detained him in Rome, Tiberius commanded alone on the Rhine ; with the begin- ning of the year 13 Germanicus took up the sole command. The state of things was regarded as one of war with the Germans ; but these were years of inaction. 1 1 Tacitus, Ann. I. 9, and Dio, state of war ; but nothing at all is Ivi. 26, attest the continuance of the reported from the nominal campaigns VOL. I. 4 cus on the Rhine. 50 NORTHERN FRONTIER OF ITALY, book vm. The fiery and ambitious hereditary prince bore with reluctance the constraint imposed on him, and we can understand how, as an officer, he should not forget the three eagles in the hands of the enemy, and how, as the son of Drusus, he should wish to re-erect his structure that had been destroyed. Soon the opportunity presented itself, and he took it. On the 19th August of the year 14, the emperor Augustus died. The first change in the throne of the new monarchy did not pass over without a crisis, and Germanicus had opportunity of proving by deeds to his father that he was disposed to maintain allegiance to him. But at the same time he found in it warrant for resuming, even unbidden, the long-wished-for invasion of Germany ; he declared that he had by this fresh campaign to repress the not inconsiderable ferment that had been called forth among the legions upon the change of sovereign. Whether this was a real reason or a pretext we know not, and perhaps he did not himself know. The commandant of the Rhine army could not be debarred from crossing the frontier anywhere, and it always to a certain degree depended on himself how far he should proceed against the Germans. Perhaps too, he believed that he was acting in the spirit of the new ruler, who had at least as much claim as his brother to the name of conqueror of Germany, and whose announced appearance in the camp on the Rhine might; doubtless, be conceived of, as though he were coming to resume the con- quest of Germany broken off at the bidding of Augustus. Renewed However this may be, the offensive beyond the offensive. , . Rhine began anew. i.ven in the autumn of the year 14, of the summers of 12, 13, and 14, was in Rome for the administration and the expedition of the autumn of of the consulate, which he retained 14 appears as the first undertaken by throughout the year, and which was Germanicus. It is true that Germani- still at that time treated in earnest ; cus had been proclaimed as Imper- this explains why Tiberius, as has ator probably even in the lifetime of now been proved (Hermann Schulz, Augustus (Mon. Ancyr. p. 17); but Quaest. Ovidianae, Greifswald, 1883, there is nothing to hinder our referring p. 15), still went to Germany in the this to the campaign of the year 11, in year 12, and resigned his Rhenish which Germanicus commanded with command only at the beginning of proconsular power alongside of Tiber- the year 13, on the celebration of the ius (Dio, lvi. 25). In the year 12 he Pannonian victory. chap. i. NORTHERN FRONTIER OF ITALY. 51 Germanicus in person led detachments of all the legions at Vetera over the Rhine, and penetrated up the Lippe pretty far into the interior, laying waste the country far and wide, putting to death the natives, and destroying the temples, such as that of Tanfana held in high honour. Those assailed — chiefly Bructeri, Tubantes, and Usipes, — sought to prepare the fate of Varus for the crown-prince on his way home ; but the attack recoiled before the energetic bearing of the legions. As this advance met with no censure, but on the contrary, thanks and marks of honour were decreed to the general for it, he went farther. In the opening of the year 15 he assembled his main force, in the first instance on the middle Rhine, and advanced in person from Mentz against the Chatti as far as the upper confluents of the Weser, while the lower army, farther to the north, attacked the Cherusci and the Marsi. There was a certain justification for this proceeding in the fact that the Cherusci favourably disposed towards Rome, who had, under the immediate impression of the disaster of Varus been obliged to join the patriots, were now again at open variance with the much stronger national party, and invoked the intervention of Germanicus. In reality he was successful in liberating Segestes, the friend of the Romans, when hard pressed by his countrymen, and at the same time in getting posses- sion of his daughter, the wife of Arminius. Segestes' brother Segimerus, once the leader of the patriots by the side of Arminius, submitted. The internal dissensions of the Germans once more paved the way for the foreign rule. In the very same year Germanicus undertook his main expedition to the region of the Ems ; Caecina marched from Vetera to the upper Ems, while he in person went thither with the fleet from the mouth of the Rhine ; the cavalry moved along the coast through the territory of the faithful Frisians. When reunited the Romans laid waste the country of the Bructeri and the whole territory between the Ems and Lippe, and thence made an expedition to the disastrous spot where, six years before, the army of Varus had perished, to erect a Caecina. 52 NORTHERN FRONTIER OF ITALY, book vm. monument to their fallen comrades. On their farther advance the Roman cavalry were allured by Arminius and the exasperated hosts of the patriots into an ambush, and would have been destroyed had not the infantry come up and prevented greater mischief. More serious dangers attended the return homeward from the Ems, which followed at first the same routes as the march thither. Retreat of The cavalry arrived at the winter camp uninjured. See- ing that the fleet was not sufficient for conveying the infantry of four legions, owing to the difficulty of navigation — it was about the time of the autumnal equinox — Germanicus disembarked two of them and made them return along the shore ; but inadequately acquainted with the ebbing and flowing of the tide at this season of the year, they lost their baggage and ran the risk of being drowned en masse. The retreat of the four legions of Caecina from the Ems to the Rhine resembled exactly that of Varus ; indeed, the difficult, marshy country offered perhaps still greater difficulties than the defiles of the wooded hills. The whole mass of natives, with the two princes of the Cherusci, Arminius and his highly esteemed uncle Inguiomerus, at their head, threw themselves on the retreating troops in the sure hope of preparing for them the same fate, and filled the morasses and woods all around. But the old general, experienced in forty years' of war service, remained cool even in the utmost peril, and kept his despairing and famishing men firmly in hand. Yet even he might not perhaps have been able to avert the mischief but for the circumstance that, after a successful attack during the march, in which the Romans lost a great part of their cavalry and almost the whole baggage, the Germans, sure of victory and eager for spoil, in opposition to Arminius' advice, followed the other leader, and instead of further surrounding the enemy, attempted directly to storm the camp. Caecina allowed the Germans to come up to the ramparts, but then burst forth from all the doors and gates with such vehemence upon the assailants that they suffered a severe defeat, and in consequence of it the further retreat took place without chap. i. NORTHERN FRONTIER OF ITALY. 53 material hindrance. Those at the Rhine had already- given up the army as lost, and were on the point of casting off the bridge at Vetera, to prevent the Germans at least from penetrating into Gaul ; it was only the resolute remonstrance of a woman, the wife of Germanicus and daughter of Agrippa, which frustrated the desperate and disgraceful resolve. The resumption of the subjugation of Germany thus began not quite successfully. The territory between the Rhine and Weser had indeed been again trodden and traversed, but the Romans had no decisive results to show, and the enormous loss in material, particularly in horses, was sorely felt, so that, as in the times of Scipio, the towns of Italy and of the western provinces took part in patriotic contributions to make up for what was lost. For the next campaign (16) Germanicus changed his Campaign plan of warfare. He attempted the subjugation of Germany ° 5 e yeai on the basis of the North Sea and the fleet, partly because the tribes on the coast, the Batavi, Frisians, and Chauci, adhered more or less to the Romans, partly in order to shorten the marches — in which much time was spent and much loss incurred — from the Rhine to the Weser and Elbe and back again. After he had employed this spring, like the previous one, for rapid advances on the Main and on the Lippe, he, in the beginning of summer, embarked his whole army at the mouth of the Rhine in the powerful transport-fleet of 1000 sail which had mean- while been made ready, and arrived in reality without loss, at the mouth of the Ems, where the fleet remained. Thence he advanced, as may be conjectured, up the Ems as far as the mouth of the Haase, and then along the latter as far up as the Werra-valley, and through this to the Weser. By this means the carrying of the army, 80,000 strong, through the Teutoburg Forest, which was attended with great difficulties, particularly as to provisions, was avoided. A secure reserve for supplies was furnished in the camp beside the fleet, and the Cherusci on the right bank of the Weser were assailed in flank instead of in front. Here the Romans encountered the levy en 54 NORTHERN FRONTIER OF ITALY, book vm. masse of the Germans, again led by the two chiefs of the patriot party, Arminius and Inguiomerus. What warlike resources were at their disposal is shown by the fact that on two occasions, one shortly after the other, in the Cheruscan country — first on the Weser itself and then somewhat farther inland 1 — they fought in the open field against the whole Roman army, and in both hardly contested the victory. The latter certainly fell to the Romans, and of the German patriots a considerable number were left on the fields of battle. No prisoners were taken, and both sides fought with extreme exaspera- tion. The second tropaeum of Germanicus spoke of the overthrow of all the Germanic tribes between the Rhine and Elbe ; the son placed this campaign of his alongside of the brilliant campaigns of his father, and reported to Rome that in the next campaign he should have the subjugation of Germany complete. But Arminius escaped, although wounded, and continued still at the head of the patriots ; and an unforeseen mischief marred the success won by arms. On the return home, which the greater part of the legions made by sea, the transport- fleet encountered the autumn storms of the North Sea. The vessels were dashed on all sides upon the islands of the North Sea, and as far as the British coasts. A great portion were destroyed, and those that escaped had for the most part to throw horses and baggage overboard, and to be glad of saving their bare life. The loss of vessels was, as in the times of the Punic war, equivalent to a defeat. Germanicus himself, cast adrift alone with the admiral's ship on the desolate shore of the Chauci, was in despair at this misfortune, and on the point of seeking his death in that ocean the assistance of which he had at the beginning of this campaign invoked so 1 The hypothesis of Schmidt ( West- near the village of Bergkirchen, which fdl. Zeitschrift, xx. p. 301)— that the lies to the south of this— will not be first battle was fought on the Idis- far removed from the truth, and may tavisian field somewhere near Biicke- at least help us to realise the matter, burg, and the second, on account of In this, as in most of the accounts of the morasses mentioned on the occa- battles by Tacitus, we must despair sion, perhaps on the Steinhudersee, of reaching an assured result. situation. chap. i. NORTHERN FRONTIER OF ITALY. 55 earnestly and so vainly. Doubtless afterwards the loss of men proved not to be quite so great as it had at first appeared, and some effective blows which the general, on his return to the Rhine, inflicted on the nearest barbarians, raised the sunken courage of the troops. But, taken as a whole, the campaign of the year 16, as compared with that of the preceding year, ended in more brilliant victories doubtless, but also in much more serious loss. The recall of Germanicus was at the same time the The altered abolition of the command-in-chief of the Rhenish army. The mere division of the command put an end to the conduct of the war as heretofore pursued ; the circum- stance that Germanicus was not merely recalled, but obtained no successor, was tantamount to ordaining the defensive on the Rhine. Thus the campaign of the year 1 6 was the last which the Romans waged in order to subdue Germany and to transfer the boundary of the empire from the Rhine to the Elbe. That this was the aim of the campaigns of Germanicus is shown by their very course, and by the trophy that celebrated the frontier of the Elbe. The re-establishment, too, of the military works on the right bank of the Rhine, of the forts of the Taunus, as well as of the stronghold of Aliso and the line con- necting it with Vetera, belonged only in part to such an occupation of the right bank as was in keeping with the restricted plan of operations after the battle of Varus ; in fact it had a far wider scope. But the designs of the general were not, or not quite, those of the emperor. It is more than probable that Tiberius from the outset allowed rather than sanctioned the enterprises of Germanicus on the Rhine, and it is certain that he wished to put an end to them by recalling him in the winter of 16-1 7. Beyond doubt, at the same time, a good part of what had been at- tained was given up, and in particular the garrison was with- drawn from Aliso. As Germanicus, even in the following year, found not a stone left of the memorial of victory erected in the Teutoburg Forest, so the results of his victories dis- appeared like a flash of lightning into the water, and none of his successors continued the building on this basis. 56 NORTHERN FRONTIER OF ITALY, book vm. Motives for If Augustus gave up the conquered Germany as lost the change after the battle f Varus, and if Tiberius now, when the of policy. ' 1 j ■ conquest had once more been taken in hand, ordered it to be broken off, we are well entitled to ask, What motives guided the two notable rulers in this course, and what was the significance of these important events for the general policy of the empire ? The battle of Varus is an enigma, not in a military but in a political point of view — not in its course, but in its consequences. Augustus was not wrong when he demanded back his lost legions, not from the enemy nor from fate, but from the general ; it was a disaster such as unskilled leaders of division from time to time bring about for every state. We have difficulty in conceiving that the destruction of an army of 20,000 men without further direct military consequences should have given a decisive turn to the policy at large of a judiciously governed uni- versal empire. And yet the two rulers bore that defeat with a patience as unexampled as it was critical and hazardous for the position of the government in relation to the army and to its neighbours ; they allowed the conclusion of peace with Maroboduus, which, beyond doubt, was meant to be in strictness a mere armistice, to become withal definitive, and made no further attempt to get the upper valley of the Elbe into their hands. It must have been no easy thing for Tiberius to see the collapse of the great structure begun in concert with his brother, and after the latter's death almost completed by himself ; the energetic zeal with which, as soon as he had again entered on the government, he took up the Ger- manic war which he had begun ten years ago, enables us to measure what this self-denial must have cost him. If, nevertheless, the self-denial was persevered in not merely by Augustus, but also after his death by Tiberius him- self, there is no other reason to be found for it than that they recognised the plans pursued by them for twenty years for the changing of the boundary to the north as incapable of execution, and the subjugation and mastery of the region between the Rhine and the chap. i. NORTHERN FRONTIER OF ITALY. 57 Elbe appeared to them to transcend the resources of the empire. If the previous boundary of the empire ran from the The Elbe middle Danube up to its source and to the upper Rhine, frontier - and thence down that river, it was, at all events, materi- ally shortened and improved by being shifted to the Elbe, which in its head -waters approaches the middle Danube, and to its course throughout ; in which case, probably, besides the evident military gain, there came into view also the political consideration that the keeping of the great commands as far as possible remote from Rome and Italy was one of the leading maxims of the Augustan policy, and an army of the Elbe would hardly have played such a part in the further development of Rome as the armies of the Rhine but too soon undertook. The preliminary conditions to this end, the overthrow of the Germanic patriot-party and of the Suebian king in Bohemia, were no easy tasks ; nevertheless they had already once stood on the verge of succeeding, and with a right conduct of the war these results could not fail to be reached. But it was another question whether, after the institution of the Elbe frontier, the troops could be with- drawn from the intervening region ; this question had been raised in a very serious way for the Roman govern- ment by the Dalmato-Pannonian war. If the mere im- pending movement of the Roman Danube -army into Bohemia had called forth a popular rising in Illyricum, that was only put down by the exertion of all their mili- tary resources after a four years' conflict, this wide region might not be left to itself either at the time or for many years to come. Similar, doubtless, was the state of the case on the Rhine. The Roman public was wont, indeed, to boast that the state held all Gaul in subjection by means of the garrison at Lyons 1200 strong ; but the government could not forget that the two great armies on the Rhine not merely warded off the Germans, but also had a very material bearing on the Gallic cantons that were not at all distinguished by submissiveness. Stationed on the Weser or even on the Elbe, they would not have ren- abandon ment. 58 NORTHERN FRONTIER OF ITALY, book vm. dered this service in equal measure ; and to keep both the Rhine and the Elbe occupied was beyond their power. And its Thus Augustus might well come to the conclusion that with the strength of the army as it then stood — con- siderably increased indeed of late, but still far below the measure of what was really requisite — that great regulation of the frontier was not practicable ; the question was thus converted from a military one into one of internal policy, and especially into one of finance. Neither Augustus nor Tiberius ventured to increase still further the expense of the army. We may blame them for not doing so. The paralysing double blow of the Illyrian and the Germanic insurrections with their grave disasters, the great age and the enfeebled vigour of the ruler, the increasing disinclina- tion of Tiberius for initiating any fresh and great under- taking, and above all any deviation from the policy of Augustus, doubtless co-operated to induce this result, and did so, perhaps, to the injury of the state. By the de- meanour of Germanicus, not to be approved but easily to be explained, we perceive how keenly the soldiers and the youth felt the abandonment of the new province of Ger- many. In the poor attempt to retain, at least nominally, the lost Germany with the help of the two German can- tons on the left of the Rhine, and in the ambiguous and uncertain words with which Augustus himself in his account of the case lays or forgoes claim to Germany as Roman, we discern how perplexed was the attitude of the government towards public opinion in this matter. The grasping at the frontier of the Elbe was a mighty, perhaps a too bold stroke, undertaken possibly by Augustus — who did not generally soar so high — only after years of hesita- tion, and doubtless not without the determining influence of the younger stepson who was in closest intercourse with him. But to retrace too bold a step is, as a rule, not a mending of the mistake, but a second mistake. The monarchy had need of warlike honour unstained and of unconditional warlike success, in quite another way than the former burgomaster-government ; the absence of the numbers 17, 18, and 19 — never filled up since the battle chap. i. NORTHERN FRONTIER OF ITALY. 59 of Varus — in the roll of regiments, was little in keep- ing with military prestige, and the peace with Maroboduus, on the basis of the status quo, could not be construed by the most loyal rhetoric into a success. The assumption that Germanicus began those far-reaching enterprises in opposition to the strict orders of his government is forbid- den by his whole political position ; but the reproach that he made use of his double position, as supreme commander of the first army of the Rhine and as future successor to the throne, in order to carry out at his own hand his politico-military plans, is one from which he can as little be exempted as the emperor from the no less grave reproach of having started back perhaps from the forming, or per- haps only from the clear expression and the sharp execu- tion, of his own resolves. If Tiberius at least allowed the resumption of the offensive, he must have felt how much was to be said for a more vigorous policy ; he may per- haps, as over-considerate people do, have left the decision, so to speak, to destiny, till at length the repeated and severe misfortunes of the crown-prince once more justified the policy of despair. It was not easy for the government to bid an army halt which had brought back two of the three lost eagles ; but it was done. Whatever may have been the real and the personal motives, we stand here at a turning-point in national destinies. History, too, has its flow and its ebb ; here, after the tide of Roman sway over the world has attained its height, the ebb sets in. North- ward of Italy the Roman rule had for a few years reached as far as the Elbe ; since the battle of Varus its bounds were the Rhine and the Danube. A legend — but an old one — relates that the first conqueror of Germany, Drusus, on his last campaign at the Elbe, saw a vision of a gigantic female figure of Germanic mould, that called to him in his own language the word " Back ! " The word was not spoken, but it was fulfilled. Nevertheless the defeat of the Augustan policy, as Germans the peace with Maroboduus and the sufferance of the Q^mans Teutoburg disaster may well be termed, was hardly a vic- tory of the Germans. After the battle with Varus the 6o NORTHERN FRONTIER OF ITALY, book vm. hope must doubtless have passed through the minds of the best, that a certain union of the nation would accrue from the glorious victory of the Cherusci and their allies, and from the retiring of the enemy in the west as in the south. Perhaps in these very crises the feeling of unity may have dawned on the Saxons and Suebians formerly confronting each other as strangers. The fact that the Saxons sent from the battle-field the head of Varus to the king of the Suebians, may be nothing but the savage expression of the thought that the hour had come for all Germans to throw themselves in joint onset upon the Roman empire, and thus to secure the frontier and the freedom of their land, as they could alone be secured, by striking down the hereditary foe in his own home. But the cultured man and the politic king accepted the gift of the insurgents only in order to forward the head to the emperor Augustus for burial ; he did nothing for, but also nothing against, the Romans, and persevered unshaken in his neutrality. Immediately after the death of Augustus there were fears at Rome of the Marcomani invading Raetia, but apparently without cause ; and when Ger- manicus thereupon resumed the offensive against the Germans from the Rhine, the mighty king of the Marco- mani looked on inactive. This policy of finesse or of cowardice dug its own grave amidst a Germanic world fiercely excited, and drunk with patriotic successes and hopes. The more remote Suebian tribes but loosely con- nected with the empire, the Semnones, Langobardi, and Gothones, declared off from the king, and made common cause with the Saxon patriots ; it is not improbable that the considerable forces, which were evidently at the disposal of Arminius and Inguiomerus in the conflicts with Ger- manicus, flowed to them in great part from these quarters. Fan of Soon afterwards, when the Roman attack was sud- boduus. denl y broken off, the patriots turned (17) to assail Maro- boduus, perhaps to assail the kingly office in general, at least as the latter administered it on the Roman model. 1 1 The statement of Tacitus (Ann. of the republicans against the mon- ii. 45), that this was properly a war archists, is probably not free from a chap. i. NORTHERN FRONTIER OF ITALY. 61 But even among themselves divisions had set in ; the two nearly related Cheruscan princes, who in the last struggles had led the patriots, if not victoriously, at any rate bravely and honourably, and had hitherto constantly fought shoulder to shoulder, no longer stood together in this war. The uncle Inguiomerus no longer tolerated his being second to his nephew, and at the outbreak of the war passed to the side of Maroboduus. Thus matters came to a decisive battle between Germans and Germans, nay, between the same tribes; for Suebi as well as Cherusci fought in both armies. Long the conflict wavered ; both armies had learned from the Roman tactics, and on both sides the passion and the exasperation were alike. Arminius did not achieve a victory properly so called, but his antagonist left to him the field of battle ; and, as Maro- boduus seemed to have fared the worst, those who had hitherto adhered to him left him, and he found him- self confined to his own kingdom. When he asked for Roman aid against his overpowerful countrymen, Tiberius reminded him of his attitude after the battle of Varus, and replied that now the Romans in turn would remain neutral. His fate was rapidly decided. In the very following year (18) he was surprised in his royal abode itself by a prince of the Gothones, Catualda, to whom he had formerly given personal offence, and who had there- upon revolted from him with the other non- Bohemian Suebi ; and, abandoned by his own people, he with diffi- culty made his escape to the Romans, who granted to him the asylum which he sought — he died many years after- wards, as a Roman pensioner, at Ravenna. Thus the opponents as well as the 1 rivals of Arminius End of had become refugees, and the Germanic nation looked to rmmius - none else than to him. But this greatness was his danger and his destruction. His own countrymen, especially his own clan, accused him of going the way of Maroboduus and of desiring to be not merely the first, but also the wish to transfer Hellenico - Roman called forth not by the nomen regis, views to the very different Germanic as Tacitus says, but by the certum world. So far as the war had an imperium visque regia of Velleius (ii. ethico-political tendency, it would be 108). 62 NORTHERN FRONTIER OF ITALY, book vm. lord and the king of the Germans — whether with reason or not, and whether, if he wished this, he did not perhaps wish what was right, who can say ? The result was a civil war between him and these representatives of popular freedom ; two years after the banishment of Maroboduus he too, like Caesar, fell by the dagger of nobles of republican sentiments near to his person. His wife Thusnelda and his son born in captivity, Thumelicus, on whom he had never set eyes, marched at the triumph of Germanicus (26th May, 17) among the other Germans of rank, in chains to the Capitol ; the old Segestes was for his fidelity to the Romans provided with a place of honour, whence he might look on at the public entry of his daughter and his grandson. They all died within the Roman empire ; with Maroboduus the wife and son of his antagonist met in the exile of Ravenna. When Tiberius remarked at the recall of Germanicus that there was no need to wage war against the Germans, and that they would of them- selves take care to do what was requisite for Rome, he knew his adversaries ; in this, at all events, history has pronounced him right. But to the high-spirited man who, at the age of six-and-twenty, had released his Saxon home from the Italian foreign rule, who thereafter had been general as well as soldier in a seven years' struggle for that freedom regained, who had staked not merely person and life, but also wife and child for his nation, to fall at the age of thirty-seven by an assassin's hand — to this man his people gave, what it was in their power to give, an eternal monument in heroic song. of the con- quest. CHAPTER II. SPAIN. The accidents of external policy caused the Romans to Conclusion establish themselves on the Pyrenaean peninsula earlier than in any other part of the transmarine mainland, and to institute there two standing commands. There, too, the republic had not, as in Gaul and Illyricum, confined itself to subduing the coasts of the Italian sea, but had rather from the outset, after the precedent of the Barcides, contemplated the conquest of the whole peninsula. With the Lusitanians (in Portugal and Estremadura) the Romans had fought from the time that they called themselves masters of Spain ; the " more remote province " had been instituted, strictly speaking, against these tribes and simultaneously with the " nearer " one ; the Callaeci (Gallicia) became subject to the Romans a century before the battle of Actium ; shortly before that battle the sub- sequent dictator Caesar had, in his first campaign, carried the Roman arms as far as Brigantium (Corunna), and con- solidated afresh the annexation of this region to the more remote province. Then, in the years between the death of Caesar and the sole rule of Augustus, there was unceas- ing warfare in the north of Spain ; no fewer than six governors in this short time won triumphs there, and per- haps the subjugation of the northern slope of the Pyrenees was effected chiefly in this epoch. 1 The wars with the 1 There triumphed over Spain — 720 C. Norbanus Flaccus (consul in 34. apart from the doubtless political 716), between 720 and 725 L. Mar- 38, 34, 29. 36 triumph of Lepidus — in 718 Cn. cius Philippus (consul in 716) and 38. 40. Domitius Calvinus (consul in 714), in Appius Claudius Pulcher (consul in 64 SPAIN. book vm. cognate Aquitanians on the north side of the mountains, which fall within the same epoch, and the last of which 27. was victoriously ended in the year 727, must stand in connection with these events. On the reorganising of the 27. administrative arrangements in 727 the peninsula went to Augustus, because there was a prospect of extensive military operations there, and it needed a permanent gar- rison. Although the southern third of the more remote province, thenceforth named from the river Baetis (Guad- alquivir) was soon given ' back to the government of the senate, 1 by far the greater portion of the peninsula remained constantly under imperial administration, includ- ing the greater part of the more remote province, Lusi- tania and Callaecia, 2 and the whole of the large nearer one. Immediately after the institution of the new supreme con- trol Augustus resorted in person to Spain, with a view, in 26, 25. his two years' stay (728, 729), to organise the new ad- ministration, and to direct the occupation of the portions of the country not yet subject. This he did from Tarraco as his headquarters, and it was at that time that the seat of government of the nearer province was transferred from New Carthage to Tarraco, after which town this province is thenceforth usually named. While it appeared 38, 28. 716), in 726 C. Calvisius Sabinus tus have belonged to Lusitania, just 39, 26. (consul in 715), and in 728 Sex. Ap- as Asturias also must have been at 29. puleius (consul in 725). The his- first attached to this province. Other- torians mention only the victory wise the narrative in Dio, liv. 5, is achieved over the Cerretani (near not intelligible ; T. Carisius, the Puycerda in the eastern Pyrenees) by builder of Emerita, is evidently the Calvinus (Dio, xlviii. 42 ; comp. governor of Lusitania, C. Furnius the Velleius, ii. 78, and the coin of Sabi- governor of the Tarraconensis. With nus with Osca, Eckhel, v. 203). this agrees the parallel representation 1 As Augusta Emerita in Lusi- in Florus, ii. 33, for the Drigaecini of 25' tania only became a colony in 729 the MSS. are certainly the PpLyauavoi, (Dio, liii. 26), and this cannot well whom Ptolemy, ii. 6, 29, adduces have been left out of account in the among the Asturians. Therefore list of the provinces in which Augus- Agrippa, in his measurements, com- tus founded colonies (Mori. Ancyr. p. prehends Lusitania with Asturia and 1 19, comp. p. 222), the separation of Callaecia (Plin. H. N. iv. 22, 1 18), and Lusitania and Hispania Ulterior must Strabo (iii. 4, 20, p. 166) designates not have taken place till after the the Callaeci as formerly termed Lusi- Cantabrian war. tani. Variations in the demarcation 2 Callaecia was not merely occu- of the Spanish provinces are men- pied from the Ulterior province, but tioned by Strabo, iii. 4, 19, p. must still in the earlier time of Augus- 166. CHAP. II. SPAIN. 65 necessary on the one hand not to remove the seat of ad- ministration from the coast, the new capital on the other hand commanded the region of the Ebro and the commu- nications with the north-west and the Pyrenees. Against the Astures (in the provinces of Asturias and Leon), and above all, the Cantabri (in the Basque country and the province of Santander), who obstinately held out in these mountains and overran the neighbouring cantons, a war- fare attended by difficulties and heavy losses was pro- longed — with interruptions, which the Romans called victories — for eight years, till at length Agrippa succeeded in breaking down the open resistance by destroying the mountain towns and transplanting their inhabitants to the valleys. If, as the emperor Augustus says, from his time the Military or- coast of the ocean from Cadiz to the mouth of the Elbe f n a ? h s e ation obeyed the Romans, the obedience in this corner of it was North- far from voluntary and little to be trusted. Matters were west ' still apparently far from having reached a proper pacifica- tion in north-western Spain. There is still mention in Nero's time of war-expeditions against the Asturians. A still clearer tale is told by the occupation of the country, as Augustus arranged it. Callaecia was separated from Lusitania and united with the Tarraconensian province, to concentrate in one hand the chief command in northern Spain. Not merely was this province then the only one which, without bordering on an enemy's country, obtained a legionary military command, but no fewer than three legions 1 were directed thither by Augustus — two to Asturia, 1 These were the Fourth Mace- 44). All three were on occasion of donian, the Sixth Victrix, and the the Batavian war sent to the Rhine, Tenth Gemina. The first of these and only one returned from it. For went, in consequence of the shifting in the year 88 there were still several of quarters of the troops occasioned legions stationed in Spain (Plin. by the Britannic expedition of Clau- Paneg. 14; comp. Hermes, iii. 118), dius, to the Rhine. The two others, of which one was certainly the Seventh although in the meanwhile employed Gemina already, before the year 79, elsewhere on several occasions, were doing garrison-duty in Spain (C. I. L. still, at the beginning of the reign of ii. 2477) ; the second must have been Vespasian, stationed in their old gar- one of those three, and was probably rison-quarters, and with them, instead the First Adiutrix, as this soon after of the Fourth, the First Adiutrix newly the year 88 takes part in the Danu- instituted by Galba (Tacitus, Hist. i. bian wars of Domitian, and is under VOL. I. 5 66 SPAIN. book viii. one to Cantabria ; and, in spite of the military pressure in Germany and in Illyricum, this occupying force was not diminished. The headquarters were established between the old metropolis of Asturia, Lancia, and the new Asturica Augusta (Astorga) in Leon that still at present bears his name. With this strong occupation of the north-west is probably connected the construction of roads undertaken there to a considerable extent in the earlier imperial period, although we are not able to demonstrate the connection in detail, seeing that the allocation of these troops in the Augustan age is unknown to us. Thus there was estab- lished by Augustus and Tiberius for the capital of Col- laecia, Bracara (Braga), a connection with Asturica, that is, with the great headquarters, and not less with the neigh- bouring towns to the north, north - east, and south. Tiberius made similar constructions in the territory of the Vascones and in Cantabria. 1 Gradually the occupying force might be diminished, and under Claudius one legion, under Nero a second, might be employed elsewhere. But these were regarded only as drafted off, and still at the beginning of the reign of Vespasian the Spanish garrison had resumed its earlier strength ; it was reduced, in the strict sense, only by the Flavian emperors, by Vespasian to two, by Domitian to one legion. From thence down to the time of Diocletian a single legion, the Seventh Gemina, and a certain number of auxiliary contingents garrisoned Leon. No province under the monarchy was less affected by outward or by inward wars than this land of the far west. While at this epoch the commanderships of the troops assumed, as it were, the place of the competing parties, the Trajan stationed in upper Germany, which alone is named on inscriptions which suggests the conjecture that it of Tiberius and of Nero, and that as was one of the several legions brought starting point of an imperial road in 88 from Spain to upper Germany, (C. /. L. ii. 4883, 4884), just as the and on this occasion came away from Asturian camp was at Leon. Augusto- Spain. In Lusitania no legions were briga also (to the west of Saragossa) stationed. and Complutum (Alcali de Henares 1 The camp of the Cantabnan to the north of Madrid) must have legion may have been at the place been centres of imperial roads not on Pisoraca (Herrera on the Pisuerga, account of their urban importance but between Palencia and Santander), as places of encampment for troops CHAP. II. SPAIN. 67 Spanish army played throughout a secondary part in that respect ; it was only as helper of his colleague that Galba entered into the civil war, and mere accident car- ried him to the first place. The force holding the north- west of the Peninsula, which even after its reduction still strikes us as comparatively strong, leads us to infer that this region had not been completely obedient even in the second and third centuries ; but we are unable to state anything definite as to the employment of the Spanish legion within the province which it held in occupation. The struggle against the Cantabrian^had been waged with the help of vessels of war ; subsequently the Romans had no occasion to institute a permanent naval station there. It is not till the period after Diocletian that we find the Pyrenaean peninsula, like the Italian and the Graeco- Macedonian, without a standing garrison. That the province of Baetica was, at least after the incursions beginning of the second century, visited on various ^ th ^ occasions from the opposite coast by the Moors — the pirates of Rif — we shall have to set forth in detail when we survey the affairs of Africa. We may presume that this serves to explain why, although in the senatorial provinces elsewhere imperial troops were not wont to be stationed, by way of exception Italica (near Seville) was provided with a division of the legion of Leon. 1 But it chiefly devolved on the command stationed in the province of Tingi (Tangier) to protect the rich south of Spain from these incursions. Still it happened that towns like Italica and Singili (not far from Antequera) were besieged by the pirates. If preparation was anywhere made by the republic introduc- for the great all-significant work of the imperial period — 'AaHan the Romanising of the West — it was in Spain. Peaceful municipal intercourse carried forward what the sword had begun ; Roman silver money was paramount in Spain long before it circulated elsewhere outside of Italy ; and the mines, the culture of the vine and olive, and the relations of 1 With this we may connect the only temporarily and with a detach- fact that the same legion was, though ment, on active service in Numidia. 68 SPAIN. book viii. traffic produced a constant influx of Italian elements to the coast, particularly in the south-west. New Carthage, the creation of the Barcides, and from its origin down to the Augustan age the capital of the Hither province and the first trading port of Spain, embraced already in the seventh century a numerous Roman population ; Carteia, opposite to the present Gibraltar, founded a generation before the age of the Gracchi, was the first transmarine civic community with a population of Roman origin Hi. 4. (iii. 4) ; the old and renowned sister-town of Carthage, Gades, the modern Cadiz, was the first foreign town out of Italy, that adopted Roman law and Roman language iv. S43- (iv. 573). While thus along the greatest part of the coast of the Mediterranean the old indigenous as well as the Phoenician civilisation had already, under the republic, conformed to the ways and habits of the ruling people, in no province under the imperial period was Romanising so energetically promoted on the part of the ruling power as in Spain. First of all the southern half of Baetica, between the Baetis and the Mediterranean, obtained, partly already under the republic or through Caesar, is. 14- partly in the years 739 and 740 through Augustus, a stately series of communities with full Roman citizenship, which here occupy not the coast especially, but above all the interior, headed by Hispalis (Seville) and Corduba (Cordova) with colonial rights, Italica (near Seville) and Gades (Cadiz) with municipal rights. In southern Lusitania, too, we meet with a series of equally privileged towns, particularly Olisipo (Lisbon), Pax Julia (Beja), and the colony of veterans founded by Augustus during his abode in Spain and made the capital of this pro- vince, Emerita (Merida). In the Tarraconensis the burgess-towns are found predominantly on the coast — Carthago Nova, Ilici (Elche), Valentia, Dertosa (Tortosa), Tarraco, Barcino (Barcelona) ; in the interior only the colony in the Ebro valley, Caesaraugusta (Saragossa), is conspicuous. In all Spain under Augustus there were numbered fifty communities with full citizenship ; nearly fifty others had up to this time received Latin rights, CHAP. II. SPAIN. 69 and stood as to inward organisation on a par with the burgess -communities. Among the rest the emperor Vespasian likewise introduced the Latin municipal organisation on occasion of the general imperial census instituted by him in the year 74. The bestowal of burgess-rights was neither then, nor generally in the better imperial period, extended much further than it had been carried in the time of Augustus ; * as to which probably the chief regulative consideration was the restricted right of levy in regard to those who were citizens of the empire. The indigenous population of Spain, which thus Romanis- became partly mixed up with Italian settlers, partly led Julians 6 towards Italian habits and language, nowhere emerges so as to be clearly recognised in the history of the imperial period. -Probably that stock, whose remains and whose language maintain their ground up to the present day in the mountains of Biscay, Guipuscoa, and Navarre, once filled the whole peninsula, as the Berbers filled the region of north Africa. Their language, different from the Indo-Germanic, and destitute of flexion like that of the Finns and Mongols, proves their original independence ; and their most important memorials, the coins, in the first century of the Roman rule in Spain embrace the peninsula, with the exception of the south coast from Cadiz to Granada, where the Phoenician language then prevailed, and of the region northward of the mouth of the Tagus and westward of the sources of the Ebro, which was then probably to a large extent practically independent, and certainly was utterly uncivilised. In this Iberian territory the south-Spanish writing is clearly distinguished from that of the north province ; but not less clearly both are branches of one stock. The Phoenician immigration here confined itself to still narrower bounds than in Africa, and the Celtic mixture does not modify the general uniformity of the national 1 The expression used by Josephus referred to the bestowal of Latin rights (contra Ap. ii. 4), that "the Iberians by Vespasian, and is an incorrect state- were named Romans," can only be ment of one who was a stranger. 70 SPAIN. book viii. development in a way that we can recognise. But the conflicts of the Romans with the Iberians belong mainly to the republican epoch, and have been formerly described ii. 209 f. (ii. 221 f). After the already mentioned last passages of arms under the first dynasty, the Iberians vanish wholly out of sight. To the question, how far they became Romanised in the imperial period, the information that has come to us gives no satisfactory answer. That in the intercourse with their former masters they would have always occasion to make use of the Roman language, needs no proof; but under the influence of Rome the national language and the national writing disappear even from public use within their own communities. Already in the last century of the republic the native coinage, which at first was to a large extent allowed, had become in the main set aside ; from the imperial period there is no Spanish civic coin with other than a Latin legend. 1 Language. Like the Roman dress, the Roman language was largely diffused even among those Spaniards who had not Italian burgess-rights, and the government favoured the de facto Romanising of the land. 2 When Augustus died the Roman language and habits prevailed in Andalusia, Granada, Murcia, Valencia, Catalonia, Aragon ; and a good part of this is to be accounted for not by colonising but by Romanising. By the ordinance of Vespasian 1 Probably the most recent monu- held good. On the contrary, at this ment of the native language, that time probably the converse was of admits of certainty as to its date, is a frequent occurrence. For example, coin of Osicerda — which is modelled the right of coining was allowed on after the denarii with the elephant the footing that the legend had to be that were struck by Caesar during Latin. In like manner public build- the Gallic war — with a Latin and ings erected by non-burgesses were Iberian legend (Zobel, Estudio his- described in Latin ; thus an inscrip- torico de la moneda antigua espaHola, tion of Ilipa in Andalusia (C. I. L. ii. \\). Among the wholly or ii. 1087) runs : Urchail Atitta f(ilius) partially local inscriptions of Spain Chilasurgun portas fornixes) aedifi- several more recent may be found ; cand(a) cttravit de s(ua) p(ecuma), public sanction is not even probable That the wearing of the toga was in the case of any of them. allowed even to non- Romans, and 2 There was a time when the com- was a sign of a loyal disposition is munities of peregrini had to solicit shown as well by Strabo's expression from the senate the right to make as to the Tarraconensis togata, as Latin the language of business ; but by Agricola's behaviour in Britain for the imperial period this no longer (Tacitus, Agric. 21). chap. ii. SPAIN. 7 1 previously mentioned the native language was restricted de jure to private intercourse. That it held its ground in this, is proved by its existence at the present day ; what is now confined to the mountains, which neither the Goths nor the Arabs ever occupied, must in the Roman period certainly have extended over a great part of Spain, especially the north-west. Nevertheless Romanising certainly set in very much earlier and more strongly in Spain than in Africa ; monuments with native writing from the imperial period can be pointed to in Africa in fair number, hardly at all in Spain ; and the Berber language at present still prevails over half of north Africa, the Iberian only in the narrow valleys of the Basques. It could not be otherwise, partly because in Spain Roman civilisation emerged much earlier and much more vigorously than in Africa, partly because the natives had not in the former as in the latter the free tribes to fall back upon. The native communal constitution of the Iberians The Span- was not perceptibly to our view different from the Gallic, ^unity!" From the first Spain, like the Celtic country on either side of the Alps, was broken up into cantonal districts ; the Vaccaei and the Cantabri were hardly in any essential respect distinguished from the Cenomani of the Trans- padana and the Remi of Belgica. The fact that on the Spanish coins struck in the earlier epoch of the Roman rule it is predominantly not the towns that are named, but the cantons, — not Tarraco but the Cassetani, not Saguntum but the Arsenses — shows, still more clearly than the his- tory of the wars of the time, that in Spain too there once subsisted larger cantonal unions. But the conquering Romans did not treat these unions everywhere in like fashion. The Transalpine cantons remained even under Roman rule political commonwealths ; the Spanish were, like the Cisalpine, simply geographical conceptions. As the district of the Cenomani is nothing but a collective expression for the territories of Brixia, Bergomum, and so forth, so the Asturians consist of twenty-two politically independent communities, which to all appearance do not 72 SPAIN. book vin. legally concern each other more than the towns of Brixia and Bergomum. 1 Of these communities the Tarracon- ensian province numbered in the Augustan age 293, in the middle of the second century 275. Here, therefore, the old canton-unions were broken up. This course was hardly determined by the consideration that the compact- ness of the Vettones and the Cantabri seemed more hazardous for the unity of the empire than that of the Sequani and the Treveri ; the distinction doubtless was chiefly based on the diversity of the time and of the form of conquest. The region on the Guadalquivir became Roman a century and a half earlier than the banks of the Loire and the Seine ; the time when the foundation of the Spanish organisation was laid was not so very far from the epoch at which the Samnite confederacy was dissolved. There the spirit of the old republic prevailed ; in Gaul the freer and gentler view of Caesar. The smaller and powerless districts, which after the dissolu- tion of the unions became the pillars of political unity — the small cantons or clans — became changed in course of time, here as everywhere into towns. The beginnings of urban development, even outside of the communities that attained Italian rights, go far back into the republican, 1 These remarkable arrangements to one of the Cantabrian populi {EpA. are clear, especially from the lists of Ep. ii. p. 243). But even for the Spanish places in Pliny, and have larger canton, which indeed was once been well exhibited by Detlefsen the political unit, there are no other (Philologus, xxxii., 606 f.). The designations than these, strictly speak- terminology no doubt varies. As the iiig, retrospective and incorrect ; gens designations civitas, popithis, gens, in particular is employed for it even belong to the independent community, in the technical style {e.g., C. I. L. they pertain dejure to these portions; ii. 4233 Intercat[iensis~] ex gente Vac- thus, e.g. there is mention of the X caeorum). That the commonwealth in «Wto , «-oftheAutrigones,ofthe^077 Spain was based on those small dis- populi of the Asturians, of the gens tricts, not on the cantons, is clear as Zoel'arum (C. I. L. ii. 2633), which well from the terminology itself as is just one of these twenty-two tribes, from the fact that Pliny in iii. 3, 18 The remarkable document which we places overagainst those 293 places possess concerning these Zoelae (C. /. the civitates contribute aliis; more- L. u. 2633) informs us that this gens over it is shown by the official at was again divided into gentilitates, census aecipiendos civitatium XXIII which latter are also themselves called Vasconum et Vardulorum (C. I. L gentes, as this same document and vi. 1463) compared with the censor other testimonies (EpA. Ep. ii. p. 243) civitatis Remorum foederatae IC I. prove. Civis is also found in reference L. xi. 1855, comp. 2607) chap. ii. SPAIN. 73 perhaps into the pre-Roman, time; subsequently the general bestowal of Latin rights by Vespasian must have made this conversion general or very nearly so. 1 In reality there were among the 293 Augustan communities of the province of Tarraco 114, and among the 275 of the second century only twenty-seven, that were not urban communities. Of the position of Spain in the imperial administration Levy. little is to be said. In the levy the Spanish provinces played a prominent part. The legions doing garrison- duty there were probably from the beginning of the prin- cipate raised chiefly in the country itself ; when afterwards on the one hand the occupying force was diminished, and on the other hand the levy was more and more restricted to the garrison-district proper, Baetica, sharing in this respect the lot of Italy, enjoyed the dubious blessing of being totally excluded from military service. The auxiliary levy, to which especially the districts that lagged behind as regards urban development were subjected, was carried out on a great scale in Lusitania, Callaecia, Asturica, and not less in the whole of northern and inland Spain ; Augustus, whose father had formed even his body- guard of Spaniards, recruited in none of the territories subject to him (setting aside Belgica) so largely as in Spain. For the finances of the state this rich country was beyond doubt one of the most secure and most produc- tive sources ; but we have no detailed information trans- mitted to us. The importance of the traffic of these provinces admits Trade and commerce. 1 As the Latin communal constitu- 2633, and Eph. Ep. ii. 322; and if tion is unsuited for a community not isolated ones from this period should organised as a town, those Spanish be found with non- Roman names, it communities, which still after Vespa- must always be a question whether sian's time lacked urban organisation, this is not simply due to actual neg- must either have been excluded from ligence. Presumptive proofs of non- the bestowal of Latin rights or have Roman communal organisation, com- had special modifications to meet paratively frequent in the scanty their case. The latter may be re- inscriptions that certainly date before garded as having more probability. Vespasian (C. /. L. ii. 172, 1953, Inscriptions, even of the gentes, sub- 2633, 5048), have not been met with sequent to Vespasian's time, show a by me in inscriptions that are cer- Latin form of name, as C. I. L. ii. tainly subsequent to Vespasian. 74 SPAIN. book vin. of being inferred in some measure from the careful provi- sion of the government for the Spanish roads. Between the Pyrenees and Tarraco there have been found Roman milestones even from the last times of the republic, such as no other province of the West exhibits. We have already remarked that Augustus and Tiberius promoted road-making in Spain mainly for military reasons ; but the road formed by Augustus at Carthago Nova can only have been constructed on account of traffic, and it was traffic mainly that was served by the imperial highway named after him, and partly regulated partly constructed anew by him. This road, continuing the Italo-Gallic coast-road and crossing the Pyrenees at the Pass of Puycerda, went thence to Tarraco, then pretty closely followed the coast by way of Valentia as far as the mouth of the Jucar, but thence made right across the interior for the valley of the Baetis, 1 then ran from the arch of Augustus — which marked the boundary of the two provinces, and with which a new numbering of the miles began — through the province Baetica to the mouth of the river, and thus connected Rome with the ocean. This was certainly the only imperial highway in Spain. Afterwards the government did not do much for the roads of Spain ; the communes, to which these were soon in the main entrusted, appear, so far as we see, to have pro- vided everywhere — apart from the tableland of the interior — communications to such an extent as was required by the state of culture in the province. For, mountainous as Spain is and not without steppes and waste land, it is yet one of the most productive countries of the earth, both through the abundance of the fruits of the soil and through its riches of wine and oil and metals. To this were early added manufactures, especially in iron wares and in woollen and linen fabrics. In the valuations under Augustus no Roman burgess-community, Patavium excepted, had such a number of rich people to show as the Spanish Gades 1 The direction of the via Augusta ii. 4920-4928) as those found between is specified by Strabo (iii. 4, 9, p. Tarragona and Valencia {ibid. 4949- 160) ; to it belong all the milestones 4954). and lastly, the numerous ones ab which have that name, as well those lano Augusto, qui est ad Baetem, ox ab from the region of Lerida (C. /. L. arm, unde inci fit Baetica, ad oceanum. chap. ii. SPAIN. 75 with its great merchants spread throughout the world ; and in keeping with this was the refined luxury of manners, the castanet -players who were here at home, and the Gaditanian songs, which circulated, like those of Alex- andria, among the elegant Romans. The nearness of Italy, and the easy and cheap intercourse by sea, gave at this epoch, especially to the Spanish south and east coasts, the opportunity of bringing their rich produce to the first market of the world, and probably with no country in the world did Rome pursue so extensive and constant a traffic on a great scale as with Spain. That Roman civilisation pervaded Spain earlier and more powerfully than any other province, is confirmed by evidence on various sides, especially in respect to religion and to literature. It is true that in the territory that was still at a later Religious period Iberian, and remained tolerably free from immigra- n es ' tion — in Lusitania, Callaecia, Asturia — the native gods, with their singular names, ending mostly in -icus and -ecus, such as Endovellicus, Eaecus, Vagodonnaegus, and the like, maintained their ground still even under the principate at the old seats. But not a single votive stone has been found in all Baetica, which might not quite as well have been set up in Italy. And the same holds true of Tarraconensis proper, only that isolated traces are met with on the upper Douro of the worship of Celtic gods. 1 No other province shows an equally energetic Romanising in matters of ritual. Cicero mentions the Latin poets at Corduba only to The Spani- censure them ; and the Augustan age of literature was l^ still in the main the work of Italians, though individual literature, provincials helped in it, and among others the learned librarian of the emperor, the philologue Hyginus, was born as a bondsman in Spain. But thenceforward the Spaniards undertook in it almost the part, if not of leader, at any rate of schoolmaster. The natives of Corduba, Mar- 1 At Clunia there was found a so long continuing among the western dedication to the Mothers (C. I. L. Celts — at Uxama, one set up to the ii. 2776) — the only Spanish example Lugoves (ib. 2818), a deity that re- of this worship so widely diffused and curs among the Celts of Aventicum. 7 6 SPAIN. book viii. cus Porcius Latro, the teacher and the model of Ovid, and his countryman and friend in youth, Annaeus Seneca, — both only about a decade younger than Horace, but for a considerable time employed in their native town as 'teachers of eloquence, before they transferred their activity in that character to Rome — were the true and proper representa- tives of the school-rhetoric that took the place of the republican freedom and sauciness of speech. Once, when the former could not avoid appearing in a real process, he came to a stand-still in his address, and only recovered his fluency when, to please the famous man, the court was transferred from the tribunal to the school-hall. Seneca's son, the minister of Nero and the fashionable philosopher of the epoch, and his grandson, the poet of the sentimental opposition to the principate, Lucanus, have an importance, as doubtful in literature as it is indisputable in history, which may in a certain sense be put to the account of Spain. In the early times of the empire, likewise, two other provincials from Baetica, Mela under Claudius, Columella under Nero, gained a place among the recog- nised didactic authors who cultivated style — the former by his short description of the earth, the latter by a thorough, in part poetical, picture of agriculture. If, in the time of Domitian, the poet Canius Rufus from Gades, the philo- sopher Decianus from Emerita, and the orator Valerius Licinianus from Bilbilis (Calatayud not far from Sara- gossa) are celebrated as literary notabilities by the side of Virgil and Catullus and by the side of the three stars of Corduba, this is certainly done on the part of one likewise a native of Bilbilis, Valerius Martialis, 1 who himself yields to none among the poets of this epoch in elegance and plastic power, or yet in venality and emptiness, and we are justified in taking into account withal the fact of 1 Thecholiambics(i. 6i) run thus: — Nasone Peligni sonant, „ . Duosque Senecas unicumque Lucanum Verona docti syllabas amat vatis, Facunda loquitur Corduba, Maronefehx Mantua est, Gaudent iocosm Canio sm G Censetur Apona Lruto suo tellus Emerita Deciano meo : Mellaque nee Flacco minus, t. t;,;„;„„„ „/„„•.. j \. ** r, rj j.i j-i ■ , -j- * r ; -l e, L.icimane, gloriabitur nostra, Apollodoro plaudit imbnfer Nilus, Nec me taceH( Bimu _ chap. ii. SPAIN. 77 their being fellow-countrymen ; yet the mere possibility of weaving such a garland of poets shows the importance of the Spanish element in the literature of the time. But the pearl of Spanish-Latin authorship is Marcus Fabius Quintilianus (35-95) from Calagurris on the Ebro. His father had already acted as a teacher of eloquence in Rome ; he himself was brought to Rome by Galba, and occupied, especially under Domitian, a distinguished position as tutor of the emperor's nephews. His text- book of rhetoric and, in some degree, of the history of Roman literature, is one of the most excellent which we possess from Roman antiquity, pervaded by fine taste and sure judgment, simple in feeling as in presentation, in- structive without weariness, pleasing without effort, con- trasting sharply and designedly with the fashionable litera- ture that was so rich in phrases and so empty of ideas. It was in no small degree due to him that the tendency be- came changed at any rate, if not improved. Subsequently, amidst the general emptiness the influence of the Spaniards comes no further into prominence. What is, historically, of special moment in their Latin authorship is the com- plete clinging of these provincials to the literary develop- ment of the mother -country. Cicero, indeed, scoffs at the clumsiness and the provincialisms of the Spanish votaries of poetry ; and even Latro's Latin did not meet . the approval of the equally genteel and correct Roman by birth, Messalla Corvinus. But after the Augustan age nothing similar is again heard of. The Gallic rhetors, the great African ecclesiastical authors have, as Latin writers, retained in some measure a foreign complexion ; no one would recognise the Senecas and Martial by their manner and style as belonging to one or to another land ; in hearty love to his own literature and in subtle understanding of it never has any Italian surpassed the teacher of languages from Calagurris. CHAPTER III. THE GALLIC PROVINCES. Like Spain, southern Gaul had already in the time of the republic become a part of the Roman empire, yet neither so early nor so completely as the former country. The two Spanish provinces were instituted in the age of Hannibal, the province Narbo in that of the Gracchi ; and, while in the former case Rome took to itself the whole Peninsula, in the latter it was not merely content, down to the last age of the republic, with the possession of the coast, but even of this it directly took only the smaller and the more remote half. The republic was not wrong in designating what it so possessed as the town-domain of Narbo (Narbonne) ; the greater part of the coast, nearly from Montpellier to Nice, belonged to the city of Massilia. This Greek community was more a state than a city, and through its powerful position the equal alliance subsisting from of old with Rome obtained a real significance, such as had no parallel in any second allied city. It is true, nevertheless, that the Romans were for these neighbouring Greeks, still more than for the more remote Greeks of the East, shield as well as sword. The Massaliots had probably the lower Rhone as far up as Avignon in their pos- session ; but the Ligurian and the Celtic cantons of the interior were by no means subject to them, and the Roman standing camp at Aquae Sextiae (Aix) a day's march to the north of Massilia, was, quite in the true and proper sense, instituted for the permanent protection of the wealthy Greek mercantile city. It was one of the chap. in. THE GALLIC PROVINCES. 79 most momentous consequences of the Roman civil war, that along with the legitimate republic its most faithful ally, the city of Massilia, was politically annihilated, was converted from a state sharing rule into a community which continued free of the empire and Greek, but pre- served its independence and its Hellenism in the modest proportions of a provincial middle-sized town. In a poli- tical aspect there is nothing more to be said of Massilia after its capture in the civil war ; the town was thence- forth for Gaul only what Neapolis was for Italy — the centre of Greek culture and Greek learning. Inasmuch as the greater part of the later province of Narbo only at that time came under direct Roman administration, it is to this epoch in particular that the erection of it in a certain measure belongs. How the rest of Gaul came into the power of Rome Last con- has been already narrated (iv. 240 ff.) Before Caesar's |£ree Gauls' Gallic war the rule of the Romans extended approximately iv. 230 f. as far as Toulouse, Vienne, and Geneva ; after it, as far as the Rhine throughout its course, and the coasts of the Atlantic Ocean on the north as on the west. This sub- jugation, it is true, was probably not complete, in the north-west perhaps not much less superficial than that of Britain (iv. 296). Yet we are informed of supplemental iv. 283. wars, in the main, merely as regards the districts of Iberian nationality. To the Iberians belonged not merely the southern but also the northern slope of the Pyrenees, with the country lying in front, Beam, Gascony, and western Languedoc 1 ; and it has already been mentioned (p. 63) that when north-western Spain was sustaining the last conflicts with the Romans, there was also on the north 1 The domain of Iberian coins ing of Narbo (636 u. c.) — under the 118. reaches decidedly beyond the Pyre- governor of Hither Spain. There are nees, though the interpretation of no Aquitanian coins with Iberian individual coin - legends, which are legends any more than from north- among others referred to Perpignan western Spain, probably because the and Narbonne, is not certain. As all Roman supremacy, under whose pro- these coinings took place under Ro- tection this coinage grew up, did not, man authorisation, this suggests the so long as the latter lasted, i.e. per- question whether this portion of the haps up to the Numantine war, subsequent Narbonensis was not at an embrace those regions, earlier date — namely before the found- 8o THE GALLIC PROVINCES. BOOK VIII. Insurrec- tions. Under Tiberius, side of the Pyrenees, and beyond doubt in connection therewith, serious fighting, at first on the part of Agrippa 3 8. in the year 716, then on the part of Marcus Valerius Messalla, the well-known patron of the Roman poets, who 28, 27. in the year 726 or 727, and thus nearly at the same time with the Cantabrian war, vanquished the Aquitanians in a pitched battle in the old Roman territory not far from Narbonne. In respect of the Celts nothing further is mentioned than that, shortly before the battle of Actium, the Morini in Picardy were overthrown ; and, although during the twenty years of almost uninterrupted civil war our reporters may have lost sight of the comparatively insignificant affairs of Gaul, the silence of the list of triumphs — -here complete — shows at any rate that no further military undertakings of importance took place in the land of the Celts during this period. Subsequently, during the long reign of Augustus, and amidst all the crises — some of them very hazardous — of the Germanic wars, the Gallic provinces remained obedient. No doubt the Roman government, as well as the Germanic patriot party, as we have seen, constantly had it in view that a decisive success of the Germans and their advance into Gaul would be followed by a rising of the Gauls against Rome ; the foreign rule cannot therefore at that time have stood by any means secure. Matters came to a real insurrection in the year 21 under Tiberius. There was formed among the Celtic nobility a widely -ramified conspiracy to overthrow the Roman government. It broke out prematurely in the far from important cantons of the Turones and the Andecavi on the lower Loire, and not merely the small garrison of Lyons, but also a part of the army of the Rhine at once took the field against the insurgents. Nevertheless the most noted districts joined ; the Treveri, under the guidance of Julius Florus, threw themselves in masses into the Ardennes ; in the immediate neighbourhood of Lyons the Haedui and Sequani rose under the leadership of Julius Sacrovir. The compact legions, it is true, gained the mastery over the rebels without much trouble ; chap. m. THE GALLIC PROVINCES. 81 but the rising, in which the Germans in no way took part, shows at any rate the hatred towards the foreign rulers, which still at that time prevailed in the land and particularly among the nobility — a hatred which was certainly strengthened, but was not at first produced, by the pressure of taxes and the financial distress that are designated as causes of the insurrection. It was a greater feat of Roman policy than that Gradual which enabled it to become master of Gaul, that it knew £* ^jf' 011 how to retain the mastery, and that Vercingetorix found no successor, although, as we see, there were not entirely wanting men who would gladly have walked in the same path. This result was attained by a shrewd com- bination of terrifying and of winning — we may add, of sharing. The strength and the proximity of the Rhine army was beyond question the first and the most effective means of preserving the Gauls in the fear of their master. If this army was maintained throughout the century at the same level, as will be set forth in the following section, it was so probably quite as much on account of their own subjects, as on account of neighbours who afterwards were by no means specially formidable. That even the temporary withdrawal of these troops imperilled the continuance of the Roman rule, not because the Germans might then cross the Rhine, but because the Gauls might renounce allegiance to the Romans, is shown by the rising after Nero's death, in spite of its vacillation ; after the troops had marched off to Italy to make their general emperor, an independent Gallic empire was proclaimed in Treves, and those soldiers who were left were taken bound to allegiance towards it. But although this foreign rule, like every such rule, rested primarily and mainly on superior power — on the ascendancy of compact and trained troops over the multitude — it by no means rested on this exclusively. The art of partition was here successfully applied. Gaul did not belong to the Celts alone ; not merely were the Iberians strongly re- presented in the south, but Germanic tribes were settled in considerable numbers on the Rhine, and were of VOL. I. 6 82 THE GALLIC PROVINCES. book vm. importance still more by their conspicuous aptitude for war, than by their number. Skilfully the government knew how to foster and to turn to useful account the antagonism between the Celts and the Germans on the left of the Rhine. But the policy of amalgamation and of reconciliation operated still more powerfully. Policy of What measures were taken with this view we shall ex- Uon. Sama " plain in the sequel. Seeing that the cantonal constitution was spared, and even a sort of national representation was conceded, and the measures directed against the national priesthood were taken gradually, while the Latin language was from the beginning obligatory, and with that national representation there was associated the new worship of the emperor ; seeing that, on the whole, the Romanising was not undertaken in an abrupt way, but was cautiously and patiently pursued, the Roman foreign rule in the Celtic land ceased to be such, because the Celts themselves became, and desired to be, Romans. The extent to which the work had already advanced after the expiry of the first century of the Roman rule in Gaul is shown by the just mentioned occurrences after Nero's death, which, in their course as a whole, belong partly to the history of the Roman commonwealth, partly to its relations with the Germans, but must also be mentioned, at least by way of slight glance, in this connection. The overthrow of the Julio-Claudian dynasty emanated from a Celtic noble and began with a Celtic insurrection ; but this was not a revolt against the foreign rule like that of Vercingetorix or even of Sacrovir ; its aim was not the setting aside, but the transforming, of the Roman govern- ment. The fact that its leader reckoned descent from a bastard of Caesar one of the patents of nobility of his house, clearly expresses the half- national, half- Roman character of this movement. Some months later certainly, after the revolted Roman troops of Germanic descent and the free Germans had for the moment overpowered the Roman army, some Celtic tribes proclaimed the independence of their nation; but this attempt proved a sad failure, not through the eventual interference of chap. m. THE GALLIC PROVINCES. 83 the government, but from the very opposition of the great majority of the Celtic cantons themselves, which could not, and did not, desire to fall away from Rome. The Roman names of the leading nobles, the Latin Roman legend on the coins of the insurrection, the travesty ^neerfeit throughout of Roman arrangements, show most clearly as foreign. that the deliverance of the Celtic nation from the yoke of the foreigners in the year 70 was no longer possible, just because there was such a nation no longer ; and the Roman rule might be felt, according to circumstances, as a yoke, but no longer as a foreign rule. Had such an opportunity been offered to the Celts at the time of the battle of Philippi, or even under Tiberius, the insurrection would have run its course, not perhaps to another issue, but in streams of blood ; now it ran off into the sand. When, some decades after these severe crises, the Rhine army was considerably reduced, they had just given the proof that the great majority of the Gauls were no longer thinking of separation from the Italians, and the four generations that had followed since the conquest had done their work. Subsequent occurrences here were crises within the Roman world. When that world threatened to fall asunder, the West as well as the East separated itself for some time from the centre of the empire ; but the separate state of Postumus was the work of necessity, not of choice, and the separation was merely de facto; the emperors who bore sway over Gaul, Britain, and Spain, laid claim to the dominion of the whole empire quite as much as their Italian anti- emperors. Certainly traces enough remained of the old Celtic habits and also of the old Celtic unruliness. As bishop Hilary of Poitiers, himself a Gaul, complains of the overbearing character of his countrymen, so the Gauls are, even in the biographies of the later Caesars, designated as stubborn and ungovernable and inclined to insubordination, so that in dealing with them tenacity and sternness of government appear specially requisite. But a separation from the Roman empire, or even a renouncing of the Roman nationality, so far as there was 84 THE GALLIC PROVINCES. book vm. any such in the case, was in these later centuries nowhere less thought of than in Gaul ; on the contrary, the develop- ment of the Romano-Gallic culture, of which Caesar and Augustus had laid the foundation, fills the later Roman period just as it fills the Middle Ages and more recent times. Organisa- The regulation of Gaul was the work of Augustus. tiu-eeGa'ufs * n t ' ie adjustment of imperial affairs after the close of the civil wars the whole of Gaul, as it had been entrusted to Caesar or had been further acquired by him, came — with the exception merely of the region on the Roman side of the Alps, which had meanwhile been joined to Italy — under imperial administration. Immediately after- 27. wards Augustus resorted to Gaul, and in the year 727 completed in the capital Lugudunum the census of the Gallic province, whereby the portions of the country brought to the empire by Caesar first obtained an organised land-register, and the payment of tribute was regulated for them. He did not stay long at that time, for Spanish affairs demanded his presence. But the carrying out of the new arrangement encountered great difficulties and, in various cases, resistance. It was not mere military affairs that gave occasion to Agrippa's 19- stay in Gaul in the year 735, and that of the emperor 16-13. himself during the years 738-741 ; and the governors or commanders on the Rhine belonging to the imperial 16. house, Tiberius, stepson of Augustus, in 738, his brother I 2 D 9 ' 3 9 s!' Drusus ' 742-745, Tiberius again, 745 -747, 757-759, 9-11, 12- 7 6 3-765, his son Germanicus, 766-769, had all of them J S- the task of carrying on the organisation of Gaul. The work of peace was certainly no less difficult and no less important than the passages of arms on the Rhine ; we perceive this in the fact that the emperor took in hand personally the laying of the foundation, and entrusted the carrying it out to the men in the empire who were most closely related to him and highest in station. It was only in those years that the arrangements, established by Caesar amidst the pressure of the civil wars, received the shape which they thereafter in the main retained. They extended over the old as over the new province ; chap. in. THE GALLIC PROVINCES. 85 but Augustus gave up the old Roman territory, along with that of Massilia, from the Mediterranean as far as the Cevennes, as early as the year 732, to the senatorial 22. government, and retained only New Gaul in his own administration. This territory, still in itself very exten- sive, was then broken up into three administrative districts, over each of which was placed an independent imperial governor. This division attached itself to the threefold partition of the Celtic country — already found in existence by the dictator Caesar, and based on national distinctions — into Aquitania inhabited by Iberians, the purely Celtic Gaul, and the Celto- Germanic territory of the Belgae ; doubtless too it was intended in this administrative partition to lay some measure of stress on these distinctions, which tended to favour the progress of the Roman rule. This, however, was only approximately carried out, and could not be practically realised otherwise. The purely Celtic region between the Garonne and Loire was attached to the too small Iberian Aquitania ; the whole left bank of the Rhine, from the Lake of Geneva to the Moselle, was joined with Belgica, although most of these cantons were Celtic ; in general the Celtic stock so preponderated that the united provinces could be called " the three Gauls." Of the formation of the two so-called " Germanies," — nominally the compensation for the loss or abeyance of a really Germanic province, in reality the military frontier of Gaul — we shall speak in the following section. Matters of law and justice were arranged in an alto- Law and gether different way for the old province of Gaul and for J ustlce - the three new ones ; the former was Latinised at once and completely, in the latter the subsisting national state of things was in the first instance merely regulated. This contrast of administration, which reaches far deeper than the formal diversity of the senatorial and imperial admini- stration, was doubtless the primary and main occasion of the diversity, still continuing at the present day in its effects, between the regions of the Langue d'oc and Provence and those of the Langue d'oui. 86 THE GALLIC PROVINCES. book vm. Romanis- The Romanising of the south of Gaul had not in the re- southern 6 publican period advanced so far as that of the south of Spain, province. The eighty years lying between the two conquests were not to be rapidly overtaken ; the military camps in Spain were far stronger and more continuous than the Gallic ; the towns of Latin type were more numerous in the former than in the latter. Here doubtless in the time of the Gracchi and under their influence Narbo had been founded, the first burgess-colony proper beyond the sea ; but it remained isolated, and, though a rival of Massilia in commercial intercourse, to all appearance by no means equal to it in importance. But when Caesar began to guide the destinies of Rome, here above all — in this land of his choice and of his star — neglect was retrieved. The colony of Narbo was strengthened, and was under Tiberius the most populous city in all Gaul. Thereupon four new burgess-communities were laid out, chiefly in the domain ■v. 542- ceded by Massilia (iv. 572), the most important among them being, from a military point of view, Forum Julii (Frejus), the chief station of the new imperial fleet, and for trade Arelate (Aries), at the mouth of the Rhone, which soon — when Lyons rose and trade was tending more and more towards the Rhone — outstripping Narbo, became the true heir of Massilia and the great emporium of Gallo- Italic commerce. What further he himself did, and what his son did in the same sense, cannot be definitely dis- tinguished, and historically little depends on the distinc- tion ; here, if anywhere, Augustus was nothing but the executor of Caesar's testament. Everywhere the Celtic cantonal constitution gave way before the Italian com- munity. The canton of the Volcae in the coast region, formerly subject to the Massaliots, received through Caesar a Latin municipal constitution on such a footing, that the " praetors " of the Volcae presided over the whole district embracing twenty-four townships, 1 until not long there- after the old arrangement disappeared even in name, and 1 This is shown by the remarkable fr[aetor] Volcar\um\ dat— the oldest inscription of Avignon (Herzog. Gall, evidence for the Roman organisation Narb. n. 403) : T. Carisius T. f. of the commonwealth in these regions chap. in. THE GALLIC PROVINCES. 87 instead of the canton of the Volcae came the Latin town of Nemausus (Nimes). In a similar way the most con- siderable of all the cantons of this province, that of the Allobroges, who had possession of the country northward of the IseYe and eastward of the middle Rhone, from Valence and Lyons to the mountains of Savoy and to the lake of Geneva, obtained, probably already through Caesar, a like urban organisation and Italian rights, till at length the emperor Gaius granted the Roman franchise to the town of Vienna. So in the province as a whole the larger centres were organised by Caesar, or in the first age of the empire, on the basis of Latin rights, such as Ruscino (Roussillon), Avennio (Avignon), Aquae Sextiae (Aix), Apta (Apt). Already at the close of the Augus- tan age the country along both banks of the lower Rhone was completely Romanised in language and manners ; the cantonal constitution throughout the province was prob- ably set aside with the exception of slight remnants. The burgesses of the communities on whom the imperial fran- chise was conferred, and no less the burgesses in those of Latin rights, who had acquired for themselves and for their descendants the imperial franchise by entering the imperial army or by the holding of offices in their native towns, stood in law on a footing of complete equality with the Italians, and, like them, attained to offices and honours in the imperial service. In the three Gauls, on the other hand, there were no Lugu- towns of Roman and Latin rights, or rather there was only dunum - one such town 1 there, which on that account belonged to none of the three provinces or belonged to all — the town of Lugudunum (Lyons). On the extreme southern verge of imperial Gaul, immediately on the border of the muni- cipally-organised province, at the confluence of the Rhone and the Saone, on a site equally well chosen from a mili- tary and from a commercial point of view, this settlement 1 Noviodunum (Nyon on the lake civitas Equestrium (Inscrip. Helvel. of Geneva) alone perhaps in the three 115), it seems to have been inserted Gauls may be compared, as regards among the cantons, which was not the iv. 242. plan, with Lugudunum (iv. 254) ; but, case with Lugudunum. as this community emerges later as 88 THE GALLIC PROVINCES. book vm. 43- had arisen in the year 711 during the civil wars, primarily in consequence of the expulsion of a number of Italians settled in Vienna. 1 Not having originated out of a Celtic canton, 2 and hence always with a territory of narrow limits, but from the outset composed of Italians and in possession of the full Roman franchise, it stood forth unique in its kind among the communities of the three Gauls— as respects its legal relations, in some measure resembling Washington in the North American Federa- tion. This unique town of the three Gauls was at the same time the Gallic capital. The three provinces had not any common chief authority, and, of high imperial officials, only the governor of the middle or Lugudunen- sian province had his seat there ; but when emperors or princes stayed in Gaul they as a rule resided in Lyons. Lyons was, alongside of Carthage, the only city of the Latin half of the empire which obtained a standing garrison after the model of that of the capital. 3 The only mint for imperial money, which we can point to with certainty in the West for the earlier period of the empire, is that of Lyons. Here was the headquarters of the transit-dues which embraced all Gaul ; and to this as a centre the Gallic network of roads converged. But not merely had all government institutions, which were common to Gaul, their native seat in Lyons ; this Roman town became also, as we shall see further on, the seat of the 1 The persons earlier driven forth venues accruing to the city (Tacitus, from Vienna by the Allobroges (o£ £k Hist. i. 65) may have been conferred Ohivv-qs rijs 'NapflavTjo-tas iirt> t&v upon it possibly at the expense of 'KWofiplytiiv irori tKirttxbvTes), in Dio, Vienna. xlvi. 50, cannot well have been other 2 The ground belonged formerly than Roman citizens, for the founda- to the Segusiavi (.Plin. H. JV. iv. tion of a burgess-colony for their bene- 15, 107 ; Strabo, p. 186, 192), one fit is intelligible only on this sup- of the small client - cantons of the position. The "earlier" expulsion Haedui (Caesar, B. G. vii. 75) ; but probably stood connected with the in the cantonal division it counts not rising of the Allobroges under Catug- as one of these, but stands for itself 61. natus in 693 (iv. 223). The explana- as /j.TjTp6woKi.s (Ptolem. ii. 8, II, iv. 213. tion why the dispossessed were not 12). brought back, but were settled else- 3 This was the 1200 soldiers with where, is not forthcoming; but various whom, as Agrippa the king of the reasons prompting such a course may Jews says in Josephus {Bell. Jud. ii. be conceived, and the fact itself is not 16, 4), the Romans held in subjection thereby called in question. The re- the whole of Gaul. chap. in. THE GALLIC PRO VINCES. 89 Celtic diet of the three provinces, and of all the poli- tical and religious institutions associated with it — of its temples and its yearly festivals. Thus Lugudunum rapidly rose into prosperity, helped onward by the rich endowment combined with its metropolitan position and by a site uncommonly favourable for commerce. An author of the time of Tiberius describes it as the second in Gaul after Narbo ; subsequently it takes a place there by the side of, or before, its sister on the Rhone, Arelate. On occasion of the fire, which in the year 64 laid a great part of Rome in ashes, the Lugudunenses sent to those burnt out a subsidy of 4,000,000 sesterces (£43,500), and when the same fate befel their own town next year in a still harder way, the whole empire paid its contribution to them, and the emperor sent a like sum from his privy purse. The town rose out of its ruins with more splen- dour than before ; and it has for almost two thousand years remained amidst all vicissitudes a great city up to the present day. In the later period of the empire, no doubt, it fell behind Treves. The town of the Treveri, named Augusta probably from the first emperor, soon gained the first place in the Belgic province ; if still in the time of Tiberius Durocortorum of the Remi (Rheims) is named the most populous place of the province and the seat of the governors, an author from the time of Claudius already assigns the primacy there to the chief place of the Treveri. But Treves became the capital of Gaul * — we may even say of the West — only through the remodel- ling of the imperial administration under Diocletian. After Gaul, Britain, and Spain were placed under one supreme administration, the latter had its seat in Treves ; and thenceforth Treves was also, when the emperors stayed in Gaul, their regular residence, and, as a Greek of the fifth century says, the greatest city beyond the 1 Nothing is so significant of the grammar of both languages in all the position of Treves at this time as the capitals of the then subsisting seven- ordinance of the emperor Gratianus teen Gallic provinces, over and above of the year 376 {Cod. Theod. xiii. their municipal salary, a like addition 3, II), that there should be given to from the state chest: but for Treves the professors of rhetoric and of the this was to be on a higher scale. 9 o THE GALLIC PROVINCES. book vm. Alps. But the epoch when this Rome of the north received its walls and its hot baths, which might well be named by the side of the city walls of the Roman kings and of the baths of the imperial capital, lies beyond the limits of our narrative. Through the first three centuries of the empire Lyons remained the Roman centre of the Celtic land, and that not merely because it occupied the first place in population and wealth, but because it was, like no other in the Gallic north and but few in the south, a town founded from Italy, and Roman not merely as regards rights, but as regards its origin and its character. The can- As the Italic town was the basis for the organisation tonal or- f th th prov i nce S o the canton was for the northern, gamsation r > . of the three a nd predominantly indeed the canton of the Celtic tor- Gauls ' merly political, now communal, organisation. The importance of the distinction between town and canton is not primarily dependent on its intrinsic nature ; even if it had been one of mere legal form, it would have separated the nationalities, and would have awakened and whetted, on the one hand, the feeling of their belonging to Rome, on the other hand, that of their being foreign to it. The practical diversity of the two organisations may not be estimated as of much account for this period, since the elements of the communal organisation — the officials, the council, the burgess-assembly — were the same in the one case as in the other, and distinctions going deeper, such as perhaps formally subsisted, would hardly be tolerated long by Roman supremacy. Hence the transition from the cantonal organisation to the urban was frequently effected of itself and without hindrance — we may even say, with a certain necessity, in the course of development. In consequence of this the qualitative dis- tinctions of the two legal forms come into little pro- minence in our traditional accounts. Nevertheless, the contrast was certainly not a mere nominal one, but as regards the competence of the different authorities, judi- cature, taxation, levy, there subsisted diversities which were of importance, or at any rate seemed important, for adminis- tration, partly of themselves, partly in consequence of custom. chap. in. THE GALLIC PROVINCES. 91 The quantitative distinction is definitely recognisable, character The cantons, at least as they present themselves among the Celts and the Germans, are throughout tribes more than townships ; this very essential element was peculiar to all Celtic territories, and was often covered over rather than obliterated even by the subsequent Romanising. Mediol- anum and Brixia were indebted for their wide bounds and their lasting power essentially to the fact that they were, properly speaking, nothing but the cantons of the Insubres and the Cenomani. The facts, that the territory of the town of Vienna embraced Dauphin^ and Western Savoy, and that the equally old and almost equally considerable townships of Cularo (Grenoble) and Genava (Geneva) were down to late imperial times in point of law villages of the colony of Vienna, are likewise to be explained from the circumstance that this was the later name of the tribe of the Allobroges. In most of the Celtic cantons one town- ship so thoroughly preponderates that it is one and the same thing whether we name the Remi or Durocortorum, the Bituriges or Burdigala ; but the converse also occurs, as e.g. among the Vocontii Vasio (Vaison) and Lucus, among the Carnutes Autricum (Chartres) and Cenabum (Orleans) balance each other ; and it is more than ques- tionable whether the privileges which, according to Italic and Greek organisation, attached as a matter of course to the ring-wall in contrast to the open field, stood de jure, or even merely de facto, on a similar footing among the Celts. The counterpart to this canton in the Graeco-Italic system was much less the town than the tribe ; we have to liken the Carnutes to the Boeotians, Autricum and Cenabum to Tanagra and Thespiae. The specialty of the position of the Celts under the Roman rule as com- pared with other nations — the Iberians, for example, and the Hellenes — turns on this, that these larger unions con- tinued to subsist as communities in the former case, while in the latter those constitutional elements, of which they were composed, formed the communities. Older diversities of national development belonging to the pre - Roman epoch may have co-operated in the matter ; it may pos- of the cantons. 92 THE GALLIC PROVINCES. book vm. Influence of the cantonal constitu- tion. sibly have been more easily practicable to take from the Boeotians the joint diet of their towns than to break up the Helvetii into three or four districts ; political unions maintain their ground even after subjugation under a central power, in cases where their dissolution would bring about disorganisation. Yet what was done in Gaul by Augustus or, if it be preferred, by Caesar, was brought about not by the force of circumstances, but chiefly by the free resolution of the government, as it alone was in keeping with the forbearance otherwise exercised towards the Celts. For there was, in fact, in the pre-Roman time and even at the time of Caesar's conquest a far greater num- ber of cantons than we find later ; in particular, it is remarkable that the numerous smaller cantons attached by clientship to a larger one did not in the imperial period become independent, but disappeared. 1 If subse- quently the Celtic land appears divided into a moderate number of considerable, and some of them even very large, canton-districts, within which dependent cantons nowhere make their appearance, this arrangement had the way no doubt paved for it by the pre-Roman system of client- ship, but was completely carried out only under the Roman reorganisation. This continued subsistence and this enlargement of the cantonal constitution must have been above all influ- ential in determining the further political development of Gaul. While the Tarraconensian province was split up into two hundred and ninety-three independent communi- 1 In Caesar there appear doubtless, taken on the whole, the same cantons as are thereafter represented in the Augustan arrangement, but at the same time manifold traces of smaller iv. 226. client-unions (comp. iv. 237); thus as "clients" of the Haedui are named the Segusiavi, the Ambivareti, the Aulerci Brannovices, and the Bran- novii (B. G. vii. 75), as clients of the Treveri the Condrusi (B. G. iv. 6) as clients of the Helvetii the Tulingi and Latobriges. With the exception of the Segusiavi, all these are absent from the Lyons diet. Such minor cantons not wholly merged into the leading places may have sub- sisted in great number in Gaul at the time of the conquest. If, according to Josephus (Bell. Jud. ii. 16, 4), three hundred and five Gallic cantons and twelve hundred towns obeyed the Romans ; these may be the figures that were reckoned up for Caesar's successes in arras ; if the small Iberian tribes in Aquitania and the client-cantons in the Celtic land were included in the reckoning, such numbers might well be the result. chap. in. THE GALLIC PROVINCES. 93 ties (p. 72), the three Gauls numbered together, as we shall see, not more than sixty -four of them. Their unity and their recollections remained unbroken ; the zealous adoration, which throughout the imperial period was paid among the Volcae to the fountain-god Nemau- sus, shows how even here, in the south of the land and in a canton transformed into a town, there was still a vivid sense of the traditional tie that bound them together. Communities with wide bounds, firmly knit in this way by inward ties, were a power. Such as Caesar found the Gallic communities, with the mass of the people held in entire political as well as economic dependence, and an overpowerml nobility, they substantially remained under Roman rule ; exactly as in pre-Roman times the great nobles, with their train of dependents and bondsmen to be counted by thousands, played the part of masters each in his own home, so Tacitus describes the state of things in Tiberius's time among the Treveri. The Roman government gave to the community comprehensive rights, even a certain military power, so that they under certain circumstances were entitled to erect fortresses and keep them garrisoned, as was the case among the Helvetii ; the magistrates could call out the militia, and had in that case the rights and the rank of officers. This prerogative was not the same in the hands of the president of a small town of Andalusia, and of the president of a district on the Loire or the Moselle of the size of a small province. The large-hearted policy of Caesar the elder, to whom the outlines of this system must necessarily be traced back, here presents itself in all its grand extent. But the government did not confine itself to leaving Diet of the with the Celts their cantonal organisation ; it left, or three a s ' rather gave, to them also a national constitution, so far as such a constitution was compatible with" Roman supremacy. As on the Hellenic nation, so Augustus conferred on the Gallic an organised collective representa- tion, such as they in the epoch of freedom and of disorganisation had striven after, but had never attained. Unde'r the hill crowned by the capital of Gaul, where 94 THE GALLIC PROVINCES. book vm. the Saone mingles its waters with those of the Rhone, on the ist August of the year 742, the imperial prince Drusus, as representative of the government in Gaul, consecrated to Roma and to the Genius of the ruler the altar, at which thenceforth every year on this day the festival of these gods was to be celebrated by the joint action of the Gauls. The representatives of all the cantons chose from their midst year by year the " priest of the three Gauls," who on the emperor's day presented sacrifice to the emperor and conducted the festal games in connection with it. This representation of the land had not only a power of administering its own property by means of officials, who belonged to the chief circles of the provincial nobility, but also a certain share in the general affairs of the country. Of its immediate interference in politics there is certainly no other trace than that, in the serious crisis of the year 70, the diet of the three Gauls dissuaded the Treveri from rising against Rome ; but it had and used the right of bringing complaints as to the imperial and domestic officials acting in Gaul ; and it co- operated, moreover, if not in the imposition, at any rate in the apportionment of the taxes, 1 especially seeing that these were laid on not according to the several provinces but for Gaul in general. The imperial government certainly called into existence similar institutions in all the provinces, and not merely introduced in each of them the centralisation of sacred rites, but also — what the 1 This is indicated not only by the I. L. ii. 4248) ; thus doubtless the inscription in Boissieu, p. 609, where diets of all provinces were invested the words tot\f[us cens\us Galliaruni] with the apportionment of the taxes, are brought into connection with the The imperial finance -administration name of one of the altar -priests, of the three Gauls was at least, as a but also by the honorary inscription rule, so divided that the two western erected by the three Gauls to an provinces (Aquitania and Lugudunen- imperial official a censibus accipiendis sis) were placed under one procurator, (Henzen, 6944). He appears to have Belgica and the two Germanies under conducted the revision of the land- another ; yet there were probably not register for the whole country, just legally fixed powers for this purpose, as formerly Drusus did, while the A regular taking part in the levy may valuation itself took place by commis- not be inferred from the discussion saries for the individual districts. A held by Hadrian — evidently as an sacerdos Romae et Augusti of the Tar- extraordinary step — with representa- raconensis is praised ob curam tabulari tives of all the Spanish districts censualis fideliter administratam (C. (vita, 12). chap. in. THE GALLIC PROVINCES. 95 republic had not done — conferred on each one an organ for bringing requests and complaints before the govern- ment. Yet Gaul had in this respect, as compared with all other parts of the empire, at least a privilege de facto, as indeed this institution is here alone found fully- developed. 1 For one thing, the united diet of the three provinces necessarily had a more independent position in presence of the legates and procurators of each of them, than, for example, the diet of Thessalonica in presence of the governor of Macedonia. But then, in the case of institutions of this nature, far less depends on the measure of the rights conferred than on the weight of the bodies therein represented ; and the strength of the individual Gallic communities was transferred to the diet of Lyons, just as the weakness of the individual Hellenic com- munities to that of Argos. In the development of Gaul under the emperors the diet of Lyons to all appearance promoted essentially that general Gallic homogeneity, which went there hand in hand with the Latinising. The composition of the diet, which is known to us Composi- with tolerable accuracy, 2 shows in what way the question ^° 1 For the area Galliarum, the Celtic portion north of the Garonne freedman of the three Gauls (Henzen, fourteen (iv. I, 1, p. 177). Tacitus 6393), the adlector arcae Galliarum, (Ann. iii. 44) names as the total inquisitor Galliarum, iudex arcae number of the Gallic cantons sixty- Galliarum, no other province, so far four, and so does, although in an as I know, furnishes analogies ; and incorrect connection, the scholiast on of these institutions, had they been the Aeneid, i. 286. A like total general, the inscriptions elsewhere number is pointed to by the list given would certainly have preserved traces, in Ptolemy from the second century, These arrangements appear to point which adduces for Aquitania seven- to a self-administering and self-taxing teen, for the Lugudunensis twenty- body (the adlector, the meaning of five, for the Belgica twenty - two which term is not clear, occurs as an cantons. Of his Aquitanian cantons official in collegia, C. I. L. vi. 355 ; thirteen fall to the region between Orelli, 2406) ; probably this chest the Loire and Garonne, four to that defrayed the doubtless not inconsider- between the Garonne and the Pyrenees, able expenditure for the temple- In the later one from the fifth century, buildings and for the annual festival, which is well known under the name The area Galliarum was not a state- of Notitia Galliarum, twenty-six fall chest. to Aquitania, twenty - four to the 2 As the total number of the com- Lugudunensis (exclusive of Lyons), munities recorded on the altar at twenty-seven to Belgica. All these Lyons, Strabo (iv. 3, 2, p. 192) numbers are presumably correct, each specifies sixty, and as the number of for its time. Between the erection the Aquitanian communities in the of the altar in 742 and the time of 12. 9 6 THE GALLIC PROVINCES. BOOK VIII. of nationalities was treated by the government. Of the sixty, afterwards sixty-four, cantons represented at the diet, only four fall to the Iberian inhabitants of Aquitania — although this region between the Garonne and the Pyrenees was divided among a very much larger number of, as a rule, small tribes — whether it was that the others were excluded altogether from representation, or that those four represented cantons were the meeting -places of canton -unions. 2 Afterwards, probably in the time of Trajan, the Iberian district was separated from the Lyons diet, and had an independent representation On the other hand, the Celtic cantons in given to it. Tacitus (for to this his statement is doubtless to be referred), four cantons may have been added, just as the shifting of the numbers from the second to the fifth century may be referred to individual changes still in good part demonstrable, Considering the importance of these arrangements, it will not be super- fluous to exhibit them in detail, at least for the two western provinces. In the purely Celtic middle province the three lists given by Pliny (first century), Ptol- emy (second century), and the Notitia (fifth century), agree in twenty- one names : Abrincates — Andecavi — Aulerci Cenomani — Aulerci Diab- lintes — Aulerci Eburovici — Baiocasses {Bodiocasses Plin. , Vadicasii Ptol.) — Carnutes — Coriosolites (beyond doubt the Samnitae of Ptolemy) — Haedui — Lexovii — Mddae — Namnetes — Osismii — Parisii — Redones — Senones — Tricassini — Turones — Veliocasses (Rotomagenses) — Veiieti — Unelli ( Con- stantia) ; in three more : Caletae — Segusiavi — Viducasses, Pliny and Ptolemy agree, while they are want- ing in the Notitia, because in the meanwhile the Caletae were put to- gether with the Veliocasses or the Rotomagenses, the Viducasses with the Baiocasses, and the Segusiavi were merged in Lyons. On the other hand, instead of the three that have disappeared, there appear two new ones that have arisen by division : Aureliani (Orleans), a branch from the Carnutes (Chartres), and Autessio- durum (Auxerre), a branch from the Senones (Sens). There are left in Pliny two names, Boi — Atesui ; in Ptolemy one, Arvii; in the Notitia one, Saii. For Celtic Aquitania the three lists agree in eleven names : Arverni — Bituriges Cubi — Bituriges Vivisci (Burdigalenses) — Cadurci — Gabales — Lemovici — Nitiobriges (Agimunses) — Petrucorii — Pictones — - Ruteni — Santones ; the second and third agree in the 1 2th of Vellauni, which must have dropped out in Pliny ; Pliny alone has (apart from the pro- blematic Aquitani) two names more, Ambilatri and Anagnules ; Ptolemy one otherwise unknown, Datii ; per- haps Strabo's number of fourteen is to be made up by two of these. The Notitia has, besides these eleven, other two, based on splitting up the Albigenses (Albi on the Tarn), and the Ecolismenses (Angoule'me). The lists of the eastern cantons stand related in a similar way. Although sub- ordinate differences emerge, which cannot be here discussed, the char- acter and the continuity of the Gallic cantonal division are clearly apparent. 1 The four represented tribes were the Tarbelli, Vasates, Auscii, and Convenae. Besides these Pliny en- umerates in southern Aquitania no less than twenty-five tribes— most of them otherwise unknown — as standing on a legal equality with those four. '' Pliny and, presumably here too CHAP. III. THE GALLIC PROVINCES. 97 that organisation, with which we have formerly become acquainted, were substantially all represented at the diet, and likewise the half or wholly Germanic, 1 so far as at the time of the institution of the altar they belonged to the empire. That there was no place in this cantonal representation for the capital of Gaul was a matter of course. Moreover, the Ubii do not appear at the diet of Lyons, but sacrifice at their own altar of Augustus : this was, as we saw (p. 35), a remnant, which was allowed to subsist, of the intended province of Germany. While the Celtic nation in imperial Gaul was thus con- Restricted solidated in itself, it was also guaranteed in some measure ^anchfae of against Roman influences by the course pursued as regards the Gauls the conferring of the imperial franchise for this domain, citizenship The capital of Gaul no doubt was, and continued to be, a Roman burgess -colony, and this was essentially bound following older sources of informa- tion, Ptolemy know nothing of this division ; but we still possess the un- couth verses of the Gascon farmer (Borghesi, Opp. viii. 544), who effected this change in Rome, beyond doubt in company with a number of his countrymen, although he has pre- ferred not to add that it was so : — Flamen, item dumvir, quaestor pagi- q\ue\ magister Vents ad Augustum legato (sic) munere functus pro novem optinuit populis seiungere Galhs: urbe redux Genio pagi hanc dedicat aram. The oldest trace of the administrative separation of Iberian Aquitania from the Gallic is the naming of the "district of Lactora" (Lectoure) alongside of Aquitania in an inscrip- tion from Trajan's time (C. /. L. v. 875 : procurator provinciarum Lugu- duniensis et Aquitanicae, item Lac- torae). This inscription certainly of itself proves the diversity of the two territories rather than the formal severance of the one from the other ; but it may be otherwise shown that soon after Trajan the latter was VOL. I. carried out. For the fact that the separated district was originally divided into nine cantons, as these verses say, is confirmed by the name that thenceforth continued in use, Novempopulana ; but under Pius the district numbers already eleven com- munities (for the dilectator per Aqui- tanicae XI populos, Boissieu, Lyon, p. 246, certainly belongs to this con- nection), in the fifth century twelve, for the Notitia enumerates so many under the Novempopulana. This increase is to be explained similarly to that discussed at p. 95, note 2. The division does not relate to the governorship ; on the contrary, both the Celtic and the Iberian Aquitania remained under the same legate. But the Novempopulana obtained under Trajan its own diet, while the Celtic districts of Aquitania, after as before, sent deputies to the diet of Lyons. 1 There are wanting some smaller Germanic tribes, such as the Baetasii and the Sunuci, perhaps for similar reasons with those of the minor Iberian ; and further, the Cannenefates and the Frisians, probably because it was not till later that these became subjects of the empire. The Batavi were represented. 98 THE GALLIC PROVINCES. book vm. up with the peculiar position which it occupied and was intended to occupy in contradistinction to the rest of Gaul. But while the south province was covered with colonies and organised throughout according to Italian municipal law, Augustus did not institute in the three Gauls a single burgess-colony ; and probably even that municipal ius, which under the name of " Latin " formed an intervening stage between burgesses and non-burgesses, and afforded to its more notable holders burgess-rights in law for their persons and their descendants, was for a considerable time withheld from Gaul. The personal bestowal of the fran- chise, partly, according to general enactments, on the soldiers sometimes at their entering on, sometimes at their leaving, service, partly out of special favour on individuals, might certainly fall to the lot also of the Gaul ; Augustus did not go so far as the republic went in prohibiting the Helvetian, for example, once for all from acquiring the Roman franchise, nor could he do so, after Caesar had in many cases given the franchise in this way to native Gauls. But he took at least from burgesses proceeding from the three Gauls — with the exception always of the Lugudunenses — the right of candidature for magistracies, and therewith at the same time excluded them from the imperial senate. Whether this enactment was made primarily in the interest of Rome or primarily in that of the Gauls, we cannot tell ; certainly Augustus wished to secure both points — to check on the one hand the intrusion of the alien element into the Roman system, and thereby to purify and elevate the latter, and on the other hand to guarantee the continued subsistence of the Gallic idiosyn- crasy after a fashion, which precisely by its judicious reserve promoted the ultimate blending with the Roman character more surely than an abrupt obtrusion of foreign institutions would have done. Admission The emperor Claudius, himself born in Lyons and, as ofmdivi- t-ijQgg ^q sco ffed at him said, a true Gaul, set aside in dual com- ' 7 munities to great part these restrictions. The first town in Gaul right". which certainly received Italian rights was that of the Ubii, where the altar of Roman Germany was constructed ; chap. in. THE GALLIC PROVINCES. 99 there Agrippina, the subsequent wife of Claudius, was born in the camp of her father Germanicus, and she procured in the year 50 colonial rights, probably Latin, for her native place, the modern Cologne. Perhaps at the same time, perhaps even earlier, the same privilege was pro- cured for the town of the Treveri Augusta, the modern Treves. Some other Gallic cantons, moreover, were in this way brought nearer to the Roman type, such as that of the Helvetii by Vespasian, and also that of the Sequani (Besangon) ; but Latin rights do not seem to have met with great extension in these regions. Still less in the time of the earlier emperors was the full right of citizen- ship conferred in imperial Gaul on whole communities. But Claudius probably made a beginning by cancelling Setting the legal restriction which excluded the Gauls that had ^ °^ he attained to personal citizenship of the empire from the franchise. career of imperial officials ; this barrier was set aside in the first instance for the oldest allies of Rome, the Haedui, and soon perhaps generally. By this step equality of position was essentially obtained. For, according to the circumstances of this epoch, the imperial citizenship had hardly any special practical value for the circles that were by their position in life excluded from an official career, and was of easy attainment for wealthy peregrini of good descent, who wished to enter on this career and on that account had need of it ; but it was doubtless a slight keenly felt, when the official career remained in law closed against the Roman burgess from Gaul and his descendants. While in the organising of administration the national Celtic and character of the Celts was respected so far as was at all lan ^ age . compatible with the unity of the empire, this was not the case as regards language. Even if it had been practicable to allow the communities to conduct their administration in a language, of which the controlling imperial officials could only in exceptional cases be masters, it undoubtedly was not the design of the Roman government to erect this barrier between the rulers and the ruled. Accord- ingly, among the coins struck in Gaul under Roman rule, ioo THE GALLIC PROVINCES. book vm. and monuments erected on behalf of the community, there has been found no demonstrably Celtic inscription. The use of the language of the country otherwise was not hindered ; we find as well in the southern province as in the northern monuments with Celtic inscription, written in the former case always with the Greek, 1 in the latter always with the Latin 2 alphabet; and probably at least several of the former, certainly all of the latter, belong to the epoch of Roman rule. The fact that in Gaul, outside of the towns having Italian rights and the Roman camps, inscribed monuments occur at all in but small number, is in all probability to be accounted for mainly by supposing that the language of the country, treated as dialect, appeared just as unsuited for such employment as the unfamiliar imperial language, and hence the erection of memorial-stones did not become generally adopted here as in the Latinised regions ; the Latin probably may at that time in the greater part of Gaul have had nearly the same position, as it had subsequently in the earlier Middle ages over against the popular language of the time. The vigorous survival of the national language is most dis- tinctly shown by the reproduction of the Gallic proper names in Latin, not seldom with the retention of non- Latin forms of sound. The facts that spellings like Lousonna and Boudicca with the non-Latin diphthong on found their way even into Latin literature, that for the aspirated dental, the English th, there was even employed in Roman writing a special sign (B), that Epadatextori- gus is written alongside of Epasnactus, and Dirona along- 1 Thus there was found in Nemau- best (Mowat ; Bull. {fig. de la Gauls, sus a votive inscription written in p. 2$i.) the main inscription is' Latin, the Celtic language, erected Marpefio but on the reliefs of the lateral sur- NanavtriKapo (C. I. L. xi. p. 383), faces, which appear to represent * i.e., to the Mothers of the place. procession of nine armed priests, there 2 For example, we ^ read on an stand explanatory words appended : altar-stone found in Neris-les-Bains, Senani Useiloni ... and Eurises, which (Allier ; Desjardins, Giographie de la are not Latin. Such a mixture is also Gaule romaine, ii. 476) ; Bratronos met with elsewhere, e.g., in an in- Nantonicn Epadatextorici Leucullo scription of Arrenes (Creuse, Bull. Suio rebelocitoi. On another, which {pig. de la Gaule, i. 38) ; Sacer Peroco the Paris mariners' guild under Tibe- ieuru (probably=^aV) Duorico v(ot- rius erected to Jupiter the highest and tern) s{olvit) l(ibens) m(erito). chap. in. THE GALLIC PROVINCES. 101 side of Sirona — make it almost a certainty that the Celtic language, whether in the Roman territory or beyond it, had in or before this epoch undergone a certain regulation in the matter of writing, and could already at that time be written as it is written in the present day. Nor are evidences wanting of its continued use in Gaul. Evidences When the names of towns Augustodunum (Autun), Augus- °. f con ' tonemetum (Clermont), Augustobona (Troyes), and various of Celtic, similar ones arose, Celtic was necessarily still spoken even in middle Gaul. Arrian, under Hadrian, gives in his dis- quisition on cavalry, the Celtic expression for particular manoeuvres borrowed from the Celts. Irenaeus, a Greek by birth, who towards the end of the second century acted as a clergyman in Lyons, excuses the defects of his style by saying that he lives in the country of the Celts, and is compelled constantly to speak in a barbarian language. In a juristic treatise from the beginning of the third century, in contrast to the rule of law that testa- mentary directions in general are to be drawn up in Latin or Greek, any other language, e.g., Punic or Gallic, is allowed for fidei commissa. The emperor Alexander had his end announced to him by a Gallic fortune-teller in the Gallic language. Further, the church father Jerome, who had been himself in Ancyra as well as in Treves, assures us that the Galatians of Asia Minor and the Treveri of his time spoke nearly the same language, and compares the corrupt Gallic of the Asiatic with the corrupt Punic of the African. The Celtic language has main- tained itself in Brittany, just as in Wales, to the present day ; but while the province no doubt obtained its present name from the insular Britons who, in the fifth century fled thither before the Saxons, the language was hardly imported for the first time with these, but was to all appearance handed down from one generation to another there for thousands of years. In the rest of Gaul naturally during the course of the imperial period Roman habits step by step gained ground ; but the Celtic idiom was put an end to here, not so much by the Germanic immi- gration as by the Christianising of Gaul, which did not, 102 THE GALLIC PROVINCES. book vm. as in Syria and Egypt, adopt and make a vehicle of the language of the country that was set aside by the govern- ment, but preached the Gospel in Latin. Romanis- In the progress of Romanising, which in Gaul, apart in&eE^r from the southern province, continued to be left in sub- stance to inward development, there is apparent a remark- able diversity between the eastern Gaul and the west and north — a difference, which turns doubtless in part, but not solely, on the contrast between the Germans and the Gauls. In the occurrences at and after Nero's fall this diversity comes into prominence even as exercising a political influence. The close contact of the eastern cantons with the camps on the Rhine and the recruiting of the Rhenish legions, which took place especially here, procured earlier and more complete entrance for Roman habits there than in the region of the Loire and the Seine. On occasion of those quarrels the Rhenish cantons — the Celtic Lingones and Treveri, as well as the Germanic Ubii or rather the Agrippinenses — went with the Roman town of Lugudunum and held firmly to the legitimate Roman government, while the insurrection, at least, as was observed, in a certain sense national, originated from the Sequani, Haedui, and Arverni. In a later phase of the same struggle we find under altered party-relations the same disunion — those eastern cantons in league with the Germans, while the diet of Rheims refuses to join them. Native While the Gallic land was thus in respect of language urement. treated in the main just like the other provinces, we again meet with forbearance towards its old institutions in the regulations as to weights and measures. It is true that, alongside of the general imperial ordinance, which was issued in this respect by Augustus, the local observances continued in many places to subsist agreeably to the tolerant, or rather indifferent, attitude of the government in such things ; but it was only in Gaul that the local arrangement afterwards supplanted that of the empire. The roads in the whole Roman empire were measured and marked according to the unit of the Roman mile (1.48 kilom.), and up to the end of the second century chap. in. THE GALLIC PROVINCES. 103 this applied also to those provinces. But from Severus onward its place was taken in the three Gauls and the two Germanies by a mile correlated no doubt to the Roman, but yet different and with a Gallic name, the leuga (2.22 kilometres), equal to one and a half Roman miles. Severus cannot possibly have wished in this matter to make a national concession to the Celts ; this is not in keeping either with the epoch or with that em- peror in particular, who stood in an attitude of expressed hostility to these very provinces ; it must have been con- siderations of expediency that influenced him. These could only be based on the fact that the national road- measure, the leuga or else the double leuga, the German rasta, which latter corresponds to the French lieue, con- tinued to subsist in these provinces after the introduction of the unit of road -measure to a much greater extent than was the case in other countries of the empire. Augustus must have extended the Roman mile formally to Gaul and placed the itineraries and the imperial high- ways on that footing, but must have in reality left to the country the old road-measurement ; and so it may have happened that the later administration found it less incon- venient to acquiesce in the double unit for postal traffic 1 than to continue .to make use of a road-measure practi- cally unknown in the country. Of far greater significance is the attitude of the Roman Religion of government to the religion of the country ; in this beyond ' ecountI ^- doubt the Gallic nationality found its most solid support. Even in the south province the worship of non-Roman deities must have held its ground long, much longer than, for example, in Andalusia. The great commercial town of Arelate, indeed, has no other dedications to show than to gods worshipped also in Italy ; but in Frejus, Aix, Nlmes, and the whole coast region generally, the old Celtic divinities were in the imperial epoch not much less worshipped than in the interior of Gaul. In the Iberian part of Aquitania also we meet numerous traces of the 1 The posting-books and itineraries Toulouse that here the leugae be- do riot fail to remark at Lyons and gin. 104 THE GALLIC PROVINCES. book vm. indigenous worship altogether different from the Celtic. All the images of gods, however, that have come to light in the south of Gaul bear a stamp deviating less from the usual type than the monuments of the north ; and, above all, it was easier to manage matters with the national gods than with the national priesthood, which meets us only in iv. 225. imperial Gaul and in the British Islands, — the Druids (iv. 236). It would be vain labour to seek to give any con- ception of the internal character of the Druidic doctrine, strangely composed of speculation and imagination ; only some examples may be allowed to illustrate its singular and fearful nature. The power of speech was symbolically represented in a bald-headed, wrinkled, sunburnt old man, who carries club and bow, and from whose perforated tongue fine golden chains run to the ears of the man that follows him — betokening the flying arrows and the crush- ing blows of the old man mighty in speech, to whom the hearts of the multitude willingly listen. This was the Ogmius of the Celts ; to the Greeks he appeared as a Charon dressed up as Herakles. An altar found in Paris shows us three images of the gods with annexed inscription ; in the middle Jovis, on his left Vulcan, on his right Esus " the horrid with his cruel altars," as a Roman poet terms him, and yet a god of commerce and of peaceful dealing j 1 he is girded for labour like Vulcan, and, as the latter carries hammer and tongs, so he hews a willow tree with the axe. A frequently recurring deity, probably named Cernunnos, is represented cowering with crossed legs ; on its head it bears a stag's antlers, on which hangs a neck chain, and holds in its lap a money- bag ; before it stand cattle and goats — apparently, as if it were meant to express the ground as the source of riches. The enormous difference of this Celtic Olympus — void of all chasteness and beauty and delighting in quaint and fantastic mingling of things very earthly — from the simply human forms of the Greek, and the simply human 1 The second Berne gloss on Lucan, ible, says of him : Hesum Mercurium i- 44S> which rightly makes Teutates colunt, si quidem u. mercatoribus Mars, and seems also otherwise cred- colitur. chap. in. THE GALLIC PROVINCES. 105 conceptions of the Roman, religion enables us to guess the barrier which stood between these conquered and their conquerors. With this were connected, moreover, very serious practical consequences ; a comprehensive traffic in secret remedies and charms, in which the priests played at the same time the part of physicians, and in which, alongside of the conjuring and the blessing, human sacri- fices occurred, and healing of the sick by the flesh of those thus slain. That direct opposition to the foreign rule pre- vailed in the Druidism of this period cannot at least be proved ; but, even if this were not the case, it is easy to conceive that the Roman government, which elsewhere let alone all local peculiarities of worship with indifferent toleration, contemplated this Druidical system, not merely in its extravagances but as a whole, with apprehension. The institution of the Gallic annual festival in the purely Roman capital of the country, and with the exclusion of any link attaching it to the national cultus, was evidently a counter-move of the government against the old religion of the country, with its yearly council of priests at Chartres, the centre of the Gallic land. Augustus, how- ever, took no further direct step against Druidism than that of prohibiting any Roman citizen from taking part in the Gallic national cultus. Tiberius in his more energetic way acted with decision, and prohibited altogether this priesthood with its retinue of teachers and healing practi- tioners ; but it does not quite speak for the practical success of this enactment that the same prohibition was issued afresh under Claudius : it is narrated of the latter that he caused a Gaul of rank to be beheaded, simply because he was convicted of having brought into applica- tion the charms customary in his own country for a good result in proceedings before the emperor. That the occupation of Britain, which had been from of old the chief seat of these priestly actings, was in good part resolved on in order thereby to get at the root of the evil, will be fully set forth in the sequel (p. 185). In spite of all this the priesthood still played an important part in the revolt which the Gauls attempted after the downfall of 106 THE GALLIC PROVINCES. book vm. the Claudian dynasty ; the burning of the Capitol — so the Druids preached — announced the revolution in affairs, and the beginning of the dominion of the north over the south. But, although this oracle came subsequently to be fulfilled, it was not so through this nation and in favour of its priests. The peculiarities of the Gallic worship doubtless still exerted their effect even later ; when in the third century a distinctive Gallo-Roman empire came into exist- ence for some time, Hercules played the first part on its coins partly in his Graeco-Roman form, partly as Gallic Deusoniensis or Magusanus. But of the Druids there is no further mention, except only so far as the sage women in Gaul down to the time of Diocletian passed under the name of Druidesses and uttered oracles, and the ancient noble houses still for long boasted of Druidic progenitors on their ancestral roll. The religion of the country fell into the background still more rapidly perhaps than the native language, and Christianity, as it pushed its way, hardly encountered in the former any serious resistance. Economic Southern Gaul, withdrawn more than any other pro- condition. v i nce by it s position from hostile assault, and, like Italy and Andalusia, a land of the olive and the fig, rose under the imperial government to great prosperity and rich urban development. The amphitheatre and the sar- cophagus-field of Aries, the "mother of all Gaul," the theatre of Orange, the temples and bridges still standing erect to this day in and near Nimes, are vivid witnesses of this down to the present time. Even in the northern provinces the old prosperity of the country was enhanced by the lasting peace, which, certainly with lasting pressure of taxation, accrued to the land by means of the foreign rule. " In Gaul," says a writer jof the time of Vespasian, " the sources of wealth are at home, and flood the earth with their abundance." 1 Perhaps nowhere do equally 1 Josephus, Bell. Jud. ii. 16, 4. testimonies accord. Nero hears of the There king Agrippa asks his Jews revolt not unwillingly occasione nata whether they imagined themselves to spoliandarum iure belli opulentissi- be richer than the Gauls, braver than marum provhwiarum (Suetonius, the Germans, more sagacious than Nero, 40 ; Plut. Galb. 5) ; the booty the Hellenes. With this all other taken from the insurgent army of chap. in. THE GALLIC PROVINCES. 107 numerous and equally magnificent country-houses make their appearance, — especially in the east of Gaul, on the Rhine and its affluents ; we discern clearly the rich Gallic nobility. Famous is the testament of a man of rank among the Lingones, who directs that there should be erected for him a memorial tomb and a statue of Italian marble or best bronze, and that, among other things, his whole implements for hunting and fowling be burned along with him. This reminds us of the elsewhere men- tioned hunting- parks enclosed for miles in the Celtic country, and of the prominent part which the Celtic hounds for the chase and Celtic huntsmanship play in the Xenophon of Hadrian's time, who does not fail to add that the hunting system of the Celts could not have been known to Xenophon the son of Gryllos. To this connection belongs likewise the remarkable fact that in the Roman army of the imperial period the cavalry was, properly speaking, Celtic, not merely inasmuch as it was pre- eminently recruited from Gaul, but also because the manoeuvres, and even the technical expressions, were in good part derived from the Celts ; we see here how, after the disappearance of the old burgess-cavalry under the republic, the cavalry became reorganised by Caesar and Augustus with Gallic men and in Gallic fashion. The basis of this notable prosperity was agriculture, towards the elevation of which Augustus himself worked with energy, and which yielded rich produce in all Gaul, apart perhaps from the steppe-region on the Aquitanian coast. The rearing of cattle was also lucrative, especially in the north, particularly the rearing of swine and sheep, which soon acquired importance for manufactures and for export ; the Menapian hams (from Flanders) and the Atrebatian and Nervian cloth -mantles (near Arras Vindex is immense (Tac. Hist.i. 51). lacessiti iure victoriae id solum vobis Tacitus (Hist. iii. 46) calls the Haedui addidimus quo pacem tueremur, nam pecunia dites el voluptatilms opulentos. neque quies gentium sine armis neque The general of Vespasian is not wrong arma sine stipendiis neque stipendia in saying to the revolted Gauls in sine tributis haberi queunt. The Tac. Hist. iv. 74 : liegna bellaque per taxes doubtless pressed heavily, but Gallias semper fuere, donee in nostrum not so heavily as the old state of ins concederetis; nos quamquam totiens feud and club-law. 108 THE GALLIC PROVINCES. book vm. and Tournay) went forth in later times to the whole empire. Culture of Of special interest was the development of the culture the vine. f t jj e v j ne Neither the climate nor the government was favourable to it. The " Gallic winter " remained long proverbial among the inhabitants of the southern lands ; as, indeed, it was on this side that the Roman empire extended farthest towards the north. But narrower limits were drawn for the Gallic cultivation of the vine by Italian commercial competition. Certainly the god Dionysos accomplished his conquest of the world on the whole slowly, and only step by step did the drink pre- pared from grain give way to the juice of the vine ; but it was a result of the prohibitive system that in Gaul beer maintained itself at least in the north as the usual spirit- uous drink throughout the whole period of the empire ; and even the emperor Julian, on his abode in Gaul, came into conflict with this pseudo-Bacchus. 1 The imperial govern- ment did not indeed go so far as the republic, which placed under police prohibition the culture of the vine Hi. 177; ii. and olive on the south coast of Gaul (iii. 175; ii. 398) ; 37S ' but the Italians of their time were withal the true sons of their fathers. The flourishing condition of the two great emporia on the Rhine, Aries and Lyons, depended in no small degree on the market for Italian wine in Gaul ; by which fact we may measure what importance the culture of the vine must at that time have had for Italy. If one of the most careful administrators who held the imperial office, Domitian, issued orders that in all the provinces at least the half of the vines should be destroyed 2 — which, 1 This epigram on " barley- wine " On an earthen ring found in Paris is preserved (Anthol. Pal. ix. 368) : (Mowat, Bull. Spig. de la Gaule, ii. Tis v66ev els, Aibvme ; /xi. yhp rbv no; iii. 133), which is hollow and &ki)8U B&k X oi>, adapted for the filling of cups, the ovetmytypwKu-TtoALhotSaptoov. drin ker says to the host : coio, con- KHV0 Vt7Z®" '" TPy0V ' ? «**(*) ^ noditu is a misspelling] Host, thou hast more in the cellar ; the flask is ™,;„ fi„r„,'»,» ™1*„„ t„< abes ' est reple{n)da—' TTJ Tevirj fSorpiuv rev^aii a7r : a.(jTa.xiu)v. T ere X P^ KaKieiv Aw/jrpiw, 01) empty ;" and to the barmaid : a#tffl, Ai6vv.) from Nemausus, belonged to this province. Generally, as was natural, Roman literature extended its circulation also over this region ; the poets of Domitian's time sent their free copies to friends in Tolosa and Vienna. Pliny, under Trajan, is glad that his minor writings find even in Lugudunum not merely favourable readers, but booksellers who push their sale. But we cannot produce evidence for the south of any such special influence, as Baetica exercised in the earlier, and northern Gaul in the later, imperial period, on the intellectual and literary ii2 THE GALLIC PROVINCES. book vm. development of Rome. The fair land yielded richly wine and fruits ; but the empire drew from it neither soldiers nor thinkers. Literature Gaul proper was in the domain of science the promised in imperial i anc j Q f teaching and of learning ; this presumably was due to the peculiar development and to the powerful influence of the national priesthood. Druidism was by no means a naive popular faith, but a highly developed and pretentious theology, which in the good church-fashion strove to enlighten, or at any rate control, all spheres of human thought and action, physics and metaphysics, law and medicine ; which demanded of its scholars unwearied study, it was said, for twenty years, and sought and found these its scholars pre-eminently in the ranks of the nobility. The suppression of the Druids by Tiberius and his successors must have affected in the first instance these schools of the priests, and have led to their being at least publicly abolished ; but this could only be done effectively when the national training of youth was brought face to face with the Romano-Greek culture, just as the Carnutic council of Druids was confronted with the temple of Roma in Lyons. How early this took place in Gaul, without question under the guiding influence of the government, is shown by the remarkable fact that in the formerly mentioned revolt under Tiberius the insurgents attempted above all to possess themselves of the town of Augustodunum (Autun), in order to get into their power the youths of rank studying there, and thereby to gain or to terrify the great families. In the first instance these Gallic Lycea may well have been, in spite of their by no means national course of training, a leaven of distinctively Gallic nationality ; it was hardly an accident that the most important of them at that time had its seat, not in the Roman Lyons, but in the capital of the Haedui, the chief among the Gallic cantons. But the Romano-Hel- lenic culture, though perhaps forced on the nation and received at first with opposition, penetrated, as gradually the antagonism wore off, so deeply into the Celtic charac- ter, that in time the scholars applied themselves to it chap. in. THE GALLIC PROVINCES. 113 more zealously than the teachers. The training of a gentleman, somewhat after the manner in which it at present exists in England, based on the study of Latin and in the second place of Greek, and vividly reminding us in the development of the school-speech, with its finely cut points and brilliant phrases, of more recent literary phenomena springing from the same soil, became gradually in the West a sort of chartered right of the Gallo- Romans. The teachers there were probably at all times better paid than in Italy, and above all were better treated. Quintilian already mentions with respect among the pro- minent forensic orators several Gauls ; and not without design Tacitus, in his fine dialogue on oratory, makes the Gallic advocate, Marcus Aper, the defender of modern eloquence against the worshippers of Cicero and Caesar. The first place among the universities of Gaul was subse- quently taken by Burdigala, and indeed generally Aquit- ania was, as respects culture, far in advance of middle and northern Gaul ; in a dialogue written there at the begin- ning of the fifth century one of the speakers, a clergyman from Chilon-sur-Saone, hardly ventures to open his mouth before the cultivated Aquitanian circle. This was the sphere of working of the formerly-mentioned professor Ausonius, who was called by the emperor Valentinian to be teacher of his son Gratian (born in 359), and who has in his miscellaneous poems raised a monument to a large number of his colleagues ; and, when his contemporary Symmachus, the most famous orator of this epoch, sought a private tutor for his son, he had one brought from Gaul in recollection of his old teacher who had his home on the Garonne. By its side Augustodunum remained always one of the great centres of Gallic studies ; we have still the speeches which were made before the emperor Con- stantino, asking, and giving thanks for, the re-establish- ment of this school of instruction. The representation in literature of this zealous scholastic activity is of a subordinate kind, and of slight value — declamations, which were stimulated especially by the later conversion of Treves into an imperial residence VOL. 1. 8 ii4 THE GALLIC PROVINCES. book vm. and the frequent sojourn of the court in the Gallic land, and occasional poems of a multifarious character. The making of verses was, like the supply of speeches, a necessary function of the teaching office, and the public teacher of literature was at the same time a poet not exactly born, but bespoken. At least the depreciation of poetry, which is characteristic of the otherwise similar Hellenic literature of the same epoch, did not prevail among these Occidentals. In their verses the reminiscence of the school and the artifice of the pedant predominate, 1 and pictures of vivid and real feeling, as in the Moselle- trip of Ausonius, but rarely occur. The speeches, which we are indeed in a position to judge of only by some late addresses delivered at the imperial palace, are models in the art of saying little in many words, and of expressing absolute loyalty with an equally absolute lack of thought. When a wealthy mother sent her son, after he had acquired the copiousness and ornateness of Gallic speech, onward to Italy to acquire also the Roman dignity, 2 this was certainly more difficult of acquisition for these Gallic rhetoricians than the pomp of words. For the early Middle age such performances as these exercised decisive influence ; through them in the first Christian period Gaul became the seat proper of pious verses and withal the last refuge of scholastic literature, while the great mental movement within Christianity did not find its chief representatives there. Construe- In the sphere of the constructive and plastic arts the ptotcart c hmate itself called forth various phenomena unknown, or known only in their germs, to the south proper. Thus the heating of the air, which in Italy was usual only for 1 One of the professorial poems of Obstitit nostrae quia, credo, mentis Ausonius is dedicated to four Greek Tardior sensus, neque disciplinis grammarians : — Appulit Graecis puerilis aevi Sedulum cunctis studium docendi ; Noxius error. Fructus exilis tenuisque sermo ; Sed, quia nostro docuere in aevo, r S " ch thoughts have frequently Commemorandi. f ° und utterance, but seldom in Sap- _, . . ., phic measure. This mention is the more men- . p „ .. „. . . . ., . , , , , , Romana gramtas, Hieronymus, tonous, seeing that he had learned 7-., r „„ „ „° ■,,• „ J .1 ■ ■,. 1 , /• n. ■£/■ i2 Sj P- 9 2 9> Vail, nothing suitable from them : — * J r * = chap. in. THE GALLIC PROVINCES. 115 baths, and the use of glass windows, which was likewise far from common there, were comprehensively brought into application in Gallic architecture. But we may perhaps speak of a development of art peculiar to this region, in so far as figures and, in progress of time, representations of scenes of daily life emerge in the Celtic territory with relatively greater frequency than in Italy, and replace the used-up mythological representations by others more pleasing. It is certainly almost in the sepulchral monuments alone that we are able to recognise this tendency to the real and the genre, but it doubtless prevailed in the practice of art generally. The arch of Arausio (Orange), from the early imperial period, with its Gallic weapons and standards ; the bronze statue of the Berlin museum found at Vetera, representing apparently the god of the place with ears of barley in his hair ; the Hildesheim silver-plate, probably proceeding in part from Gallic workshops, show a certain freedom in the adoption and transformation of Italian suggestions. The tomb of the Julii at St. Remy, near Avignon, a work of the Augustan age, is a remarkable evidence of the lively and spirited reception of Hellenic art in southern Gaul, as well in its bold architectural structure of two square storeys crowned by a peristyle with conic dome, as also in its reliefs which, in style most nearly akin to the Pergamene, present battle and hunting scenes with numer- ous figures, taken apparently from the life of the persons honoured, in picturesque animated execution. It is remarkable that the acme of this development is reached — by the side of the southern province — in the district of the Moselle and the Maas. This region, not placed so completely under Roman influence as Lyons and the headquarter- towns on the Rhine, and more wealthy and civilised than the districts on the Loire and the Seine, seems to have in some measure produced of itself this exercise of art. The tomb of a man of rank in Treves, well known under the name of the Igel Column, gives a clear idea of the tower -like monuments, crowned with pointed roof and covered on all sides with representations n6 THE GALLIC PROVINCES. book vm. of the life of the deceased, that are here at home. Frequently we see on them the landlord, to whom his peasants present sheep, fish, fowls, eggs. A tomb- stone from Arlon, near Luxemburg, shows, besides the portraits of the two spouses, on the one side a cart and a woman with a fruit -basket, on the other a sale of apples above two men squatting on the ground. Another tombstone from Neumagen, near Treves, has the form of a ship ; in this sit six mariners plying the oars ; the cargo consists of large casks, alongside of which the merry-looking steersman seems — one might imagine — to be rejoicing over the wine which they contain. We may perhaps bring them into connection with the serene picture which the poet of Bordeaux has preserved to us of the Moselle valley, with its magnificent castles, its many vineyards, and its stirring doings of fishermen and of sailors, and find in it the proof that in this fair land, more than fifteen hundred years ago, there was already the pulsation of peaceful activity, serene enjoyment, and warm life. CHAPTER IV. ROMAN GERMANY AND THE FREE GERMANS. THE two Roman provinces of Upper and Lower Germany Limitation were the result of that defeat of the Roman arms and of Q e f°^ Roman policy under the reign of Augustus which has been already (p. 55 f.) described. The original province of Germany, which embraced the country from the Rhine to the Elbe, subsisted only twenty years, from the first campaign of Drusus, 742 u. G, down to the battle of 12. Varus and the fall of Aliso, 762 u. C. ; but as, on the a.d. 9. one hand, it included the military camps on the left bank of the Rhine — Vindonissa, Mogontiacum, Vetera — and, on the other hand, even after that disaster, more or less considerable portions of the right bank remained Roman, the governorship and the command were not, in a strict sense, done away by that catastrophe, although they were, so to speak, placed in suspense. The internal organisation of the Three Gauls has been already set forth ; they embraced the whole country as far as the Rhine without distinction of descent — except that the Ubii, who had only been brought over to settle in Gaul during the last crises, did not belong to the sixty-four cantons, while the Helvetii, the Triboci, and generally the districts elsewhere held in occupation by the Rhen- ish troops, doubtless did so belong. The intention had been to gather together the German cantons between the Rhine and Elbe into a similar association under Roman supremacy, as had been constituted in the case of the Gallic cantons, and to bestow upon it, in the altar n8 ROMAN GERMANY book vm. to Augustus of the Ubian town — the germ of the modern Cologne — an executive centre similar to that which the altar of Augustus at Lyons formed for Gaul ; for the more remote future the transference of the chief camp to the right bank of the Rhine, and the restoration of the left, at least in the main, to the governor of the Belgica, were doubtless in contemplation. But these projects came to an end with the legions of Varus ; the Germanic altar of Augustus on the Rhine became or remained the altar of the Ubii ; the legions permanently retained their standing quarters in the territory, which properly belonged to the Belgica, but — seeing that a separation of the military and civil administration was, according to the Roman arrangement, excluded — was placed, so long as the troops were stationed there, for administrative purposes also under the commandants of the two armies. 1 For, as was formerly stated, Varus was probably the last commandant of the united army of the Rhine ; on the increase of the army to eight legions, which was con- sequent upon that disaster, the division of it to all appearance also ensued. What we have to describe in this section therefore is not, strictly speaking, the circum- stances of a Roman province, but the fortunes of a Roman army, and, as most closely connected therewith, the fortunes of the neighbouring peoples and adversaries, so far as these are interwoven with the history of Rome. Upper and The two headquarters of the army of the Rhine were always Vetera near Wesel and Mogontiacum, the modern Mentz, both doubtless older than the division of the command, and one of the reasons for introducing that division. The two armies numbered in the first century 1 This division of a province among not at all easy to see why the two three governors is without parallel Germanic ones !had districts within elsewhere in Roman administration, the Belgica assigned to them instead The relation of Africa and Numidia of districts of their own. Nothing offers doubtless an external analogy, but the taking back of the frontier, but was politically conditioned by the while the hitherto subsisting name was position of the senatorial governor to retained — just as the Transdanubian the imperial military commandant, Dacia continued subsequently to sub- while the three governors of Belgica sist by name as Cis - Danubian — were uniformly imperial ; and it is explains this singular peculiarity. Lower Ger many. chap. iv. AND THE FREE GERMANS. 119 four legions each, thus about 30,000 men 1 ; at or be- tween those two points lay the main bulk of the Roman troops, besides one legion at Noviomagus (Nimeguen), another at Argentoratum (Strassburg), and a third at Vin- donissa (Windisch not far from Zurich) not far from the Raetian frontier. To the lower army belonged the not inconsiderable fleet on the Rhine. The boundary between the upper and the lower army lay between Andernach and Remagen near Brohl, 2 so that Coblenz and Bingen fell to the upper, Bonn and Cologne to the lower military district. On the left bank there belonged to the upper German administrative circuit the districts of the Helvetii (Switzer- land), the Sequani (Besancon), the Lingones (Langres), the Rauraci (Basle), the Triboci (Alsace), the Nemetes (Spires), and the Vangiones (Worms) ; to the more re- stricted lower German circuit belonged the district of the Ubii, or rather the colony Agrippina (Cologne), those of the Tungri (Tongern), the Menapii (Brabant), and the Batavi, while the cantons situated farther to the west, including Metz and Treves, were placed under the dif- ferent governors of the three Gauls. While this separation has merely administrative significance, on the other hand the varying extent of the two jurisdictions on the right 1 The strength of the auxilia of the division. At the minimum, there- upper army may be fixed for the epoch fore, the figure indicated above re- of Domitian and Trajan with tolerable suits as the normal state of the auxilia certainty at about 10,000 men. A of this army, and it cannot have been document of the year go enumerates materially exceeded. But the auxilia four alae and fourteen cohortes of this of lower Germany, whose garrisons army ; to these is to be added at least were less extended, may well have one cohort (/ Germanorum), which, it been smaller in number, can be shown, did garrison-duty there 2 At the frontier bridge over the as well in the year 82 as in the year rivulet Abrinca, now Vinxt, the old 116; whether two alae which were boundary of the archdioceses of there in the year 82, and at least Cologne and Treves, stood two al- three cohorts which were there in tars, that on the side of Remagen 1 16, and which are absent from the dedicated to the Boundaries, the list of the year 90, were doing gar- Spirit of the place, and Jupiter (Fini- rison work there in 90 or not, is bus et Genio loci et lovi Optimo doubtful, but most of them probably maximo) by soldiers of the 30th were away from the province before lower German legion ; the other on 90 or only came into it after 90. Of the side of Andernach, dedicated to those nineteen auxilia one was certainly Jupiter, the Genius of the place, and {coh. I Damascenorum), another per- Juno, by a soldier of the 8th Upper haps (ala I Flavia gemind), a double Germanic (Brambach, 649, 650). izo ROMAN GERMANY book vm. bank coincides with the varying relations to their neigh- bours and the advancing or receding of the bounds of the Roman rule conditioned by those relations. With these neighbours confronting them, matters on the lower and on the upper Rhine were regulated in ways so diverse, and the course of events was so thoroughly different that here the provincial separation became historically of the most decisive importance. Let us look first at the develop- ment of things on the lower Rhine. Lower Ger- We have formerly described how far the Romans had many- subjugated the Germans on both banks of the Rhine. The Germanic Batavi had been peacefully united with the empire not by Caesar, but not long afterwards, perhaps by Drusus (p. 2 8). They were settled in the Rhine delta, that is on the left bank of the Rhine and on the islands formed by its arms, upwards as far at least as the Old Rhine, and so nearly from Antwerp to Utrecht and Leyden in Zealand and southern Holland, on territory originally Celtic — at least the local names are predomi- nantly Celtic ; their name is still borne by the Betuwe, the lowland between the Waal and the Leek with the capital Noviomagus, now Nimeguen. They were, especially com- pared with the restless and refractory Celts, obedient and useful subjects, and hence occupied a distinctive position in the aggregate, and particularly in the military system, of the Roman empire. They remained quite free from taxation, but were on the other hand drawn upon more largely than any other canton in the recruiting ; this one canton furnished to the army 1000 horsemen and 9000 foot soldiers ; besides, the men of the imperial body-guard were taken especially from them. The command of these Batavian divisions was conferred exclusively on native Batavi. The Batavi were accounted indisputably not merely as the best riders and swimmers of the army, but also as the model of true soldiers, and in this case certainly the good pay of the Batavian body-guard, as well as the privilege of the nobles to serve as officers, considerably confirmed their loyalty. These Germans accordingly had taken no part either preparatory to, or consequent upon, chap. iv. AND THE FREE GERMANS. 121 the disaster of Varus ; and if Augustus, under the first impression of the terrible news, discharged his Batavian guard, he soon became convinced of the groundlessness of his suspicion, and the troop was a short time afterwards reinstated. On the other bank of the Rhine next to the Batavi, in Cannene- the modern Kennemer district (North Holland beyond fates ' Amsterdam), dwelt the Cannenefates, closely related to them but less numerous ; they are not merely named among the tribes subjugated by Tiberius, but were also treated like the Batavi in the furnishing of soldiers. The Frisians, adjoining these further on, in the coast district Frisians. that is still named after them, as far as the lower Ems, submitted to Drusus and obtained a position similar to that of the Batavi. There was imposed on them instead of tribute simply the delivery of a number of bullocks' hides for the wants of the army ; on the, other hand they had to furnish comparatively large numbers of men for the Roman service. They were the most faithful allies of Drusus as afterwards of Germanicus, useful to him in con- structing canals as well as especially after the unfortunate North Sea expeditions (p. 53). They were followed on the east by the Chauci, a widely extended tribe of sailors Chauci. and fishermen along the coast of the North Sea on both sides of the Weser, perhaps from the Ems to the Elbe ; they were brought into subjection to the Romans by Drusus at the same time with the Frisians, but not, like these, without resistance. All these Germanic coast tribes sub- mitted either by agreement or at any rate without any severe struggle to the new rule, and as they had taken no part in the rising of the Cherusci, they still continued after the battle of Varus in their earlier relations to the Roman empire ; even from the more remote cantons of the Frisians and the Chauci the garrisons were not at that time withdrawn, and the latter still furnished a contingent to the campaigns of Germanicus. On the renewed evacua- tion of Germany in the year 1 7 the poor and distant land of the Chauci, difficult of protection, seems certainly to have been given up ; at least there are no later evidences ROMAN GERMANY BOOK VIII. Limes and desert-fron- tier on the lower Rhine. of the continuance of the Roman dominion there, and some decades later we find them independent. But all the land westward of the lower Ems remained with the empire, whose boundary thus included the modern Netherlands. The defence of this part of the imperial frontier against the Germans not belonging to the empire was left in the main to the subject maritime cantons themselves. Farther up the stream a different course was taken ; a frontier-road was here marked off, and the land lying between it and the Rhine was depopulated. With the frontier-road drawn at a greater or less distance from the Rhine, the Limes} was associated the control of frontier- 1 Limes (from limus, across) is a technical expression foreign to the state of things under our [German] law, and hence not to be reproduced in our language, derived from the fact that the Roman division of land, which excludes all natural boundaries, separates the squares, into which the ground coming under the head of private property is divided, by inter- mediate paths of a definite breadth ; these intermediate paths are the limites, and so far the word always denotes at once the boundary drawn by man's hand, and the road con- structed by man's hand. The word retains this double signification even in application to the state (Rudorff, Grom. Inst. p. 289, puts the matter incorrectly) ; limes is not every im- perial frontier, but only that which is marked out by human hands, and arranged at the same time for being patrolled and having posts stationed for frontier-defence (Vita Hadriani, 12; loeis in quibus larbari non fluminibus, sed limitibus dividuniur), such as we find in Germany and in Africa. Therefore there are applied to the laying- out of this limes the terms that serve to designate the construc- tion of roads, aperire (Velleius, ii. 121, which is not to be understood, as Mullenhoff, Zeitschr. f. d. Alterth., new series, ii. p. 32, would have it, like our opening of a turnpike), munire, agere (Frontinus, Strat. i. 3, 10 : limitibus per cxx m. p. actis). Therefore the limes is not merely a longitudinal line, but also of a certain breadth (Tacitus, Ann. i. 50; castra in limite local). Hence the construc- tion of the limes is often combined with that of the agger — that is, of the road-embankment (Tacitus, Ann. ii. 7 : cuiuta novis limitibus aggeri- busque permunita), and the shifting of it with the transference of frontier- posts (Tacitus, Germ. 29 : limite acto promotisque praesidiis). The Limes is thus the imperial frontier-road, destined for the regulation of frontier- intercourse, inasmuch as the crossing of it was allowed only at certain points corresponding to the bridges of the river boundary, and elsewhere forbidden. This was doubtless effect- ed in the first instance by patrolling the line, and, so long as this was done, the limes remained a boundary road. It remained so too, when it was fortified on both sides, as was done in Britain and at the mouth of the Danube ; the Britannic wall is also termed limes (p. 187, note 2). Posts might also be stationed at the allowed points of crossing, and the intervening spaces of the frontier-roads might be in some way rendered impassable. In this sense the biographer of Hadrian says in the above-quoted passage that at the limites he stipitibus magnis in chap. w. AND THE FREE GERMANS. 123 intercourse, as the crossing of this road was forbidden altogether by night, and, as regards armed men, by day, and was permitted in the case of others, as a rule, only under special precautions for security and on payment of the prescribed transit-dues. Such a road was drawn opposite to the headquarters on the lower Rhine, in what is now Miinster, by Tiberius after the disaster of Varus, at some distance from the Rhine, seeing that between it and the river stretched the " Caesian forest," the more precise position of which is not known. Similar arrange- ments must have been made at the same time in the valleys of the Ruhr and the Sieg as far as that of the Wied, where the province of the lower Rhine ended. This road did not necessarily require to be militarily occupied and arranged for defence, although of course the defence of the frontier and the fortification of it always aimed at making the frontier-road as far as possible secure. A chief means for protecting the frontier was the depopula- tion of the tract of land between the river and the road. " The tribes on the right bank of the Rhine," says a well- informed author of the time of Tiberius, "have been in part transferred by the Romans to the left bank, in part withdrawn of their own accord into the interior." This applied, in what is now the Miinster country, to the Ger- manic stocks earlier settled there of the Usipes, Tencteri, Tubantes. In the campaigns of Germanicus these appear dislodged from the Rhine, but still in the region of the Lippe, afterwards, probably in consequence of those very expeditions, farther southward opposite to Mentz. Their old home lay thenceforth desolate, and formed the ex- tensive pasture -country reserved for the herds of the modum muralis saepis funditus iactis of the republic ; and beyond doubt atque conexis barbaros separavit. By this conception of the limes only this means the frontier-road was con- originated with the institution of the verted into a. frontier-barricade pro- chain of posts enclosing the state, vided with certain passages through where natural boundaries were want- it, and such was the limes of ing — a protection of the imperial upper Germany in the developed shape frontier, which was foreign to the to be set forth in the sequel. We republic, but was the foundation of the may add that the word is not used Augustan military system, and above with this special import in the time all, of the Augustan system of tolls. 124 ROMAN GERMANY BOOK VIII. Conflicts with the Frisii and Chauci under Claudius. lower Germanic army, on which in the year 58 first the Frisii and then the Amsivarii, wandering homeless, thought of settling, without being able to procure leave from the Roman authorities to do so. Farther to the south at least a portion of the Sugambri, who likewise were sub- jected in great part to the same treatment, remained settled on the right bank, 1 while other smaller tribes were wholly dislodged. The scanty population tolerated within the Limes were, as a matter of course, subjects of the empire, as is confirmed by the Roman levy taking place among the Sugambri. In this way matters were arranged on the lower Rhine after the abandonment of the more comprehensive pro- jects, and thus a not inconsiderable territory on the right bank was still held by the Romans. But various incon- venient complications arose in connection with it. Towards the end of the reign of Tiberius (28) the Frisians, in consequence of intolerable oppression in the levying of tribute in itself small, revolted from the empire, slew the people employed in levying it, and besieged the Roman commandant acting there, with the rest of the Roman soldiers and civilians sojourning in the territory, in the fortress of Flevum, where, previous to the extension of the Zuyder See that took place in the Middle Ages, lay the eastmost mouth of the Rhine, near the modern island Vlieland beside the Texel. The rising assumed such proportions that both armies of the Rhine marched in concert against the Frisians ; but still the governor Lucius Apronius accomplished nothing. The Frisians gave up the siege of the fortress, when the Roman fleet brought up the legions ; but it was difficult to get near the Frisians 1 The Sugambri transplanted to the left bank are not subsequently mentioned under this name, and are probably the Cugerni dwelling below Cologne on the Rhine. But that the Sugambri on the right bank, whom Strabo mentions, were at least still in existence in the time of Claudius, is shown by the cohort named after this emperor, and thus certainly formed under him, doubtless of Sugambri (C. /. Z. iii. p. 877); and they, as well as the four other probably Augustan cohorts of this name, con- firm what Strabo also in a strict sense says, that these Sugambri belonged to the Roman empire. They disap- peared doubtless, like the Mattiaci, only amidst the tempests of the migra- tion of nations. chap. iv. AND THE FREE GERMANS. 125 themselves in a country so much intersected ; several Roman corps were destroyed in detail, and the Roman advanced guard was so thoroughly defeated that even the dead bodies of the fallen were left in the power of the enemy. The matter was not brought to a decisive action, nor yet to a true subjugation ; Tiberius, the older he grew, became ever less inclined to larger enterprises, which gave to the general in command a position of power. With this state of things was connected the fact that in the immediately succeeding years the neighbours of the Frisians, the Chauci, became very troublesome to the Romans ; in the year 41 the governor Publius Gabinius Secundus had to undertake an expedition against them, and six years later (47) they even pillaged far and wide the coast of Gaul with their light piratical vessels under the leader- ship of the Roman deserter Gannascus, by birth one of the Cannenefates. Gnaeus Domitius Corbulo, nominated governor of Lower Germany by Claudius, put a stop to the doings of these forerunners of the Saxons and Normans, and thereupon vigorously brought back the Frisians to obedience, by organising anew their common- wealth and stationing a Roman garrison among them. Corbulo had the intention of chastising the Chauci also ; The occu- at his instigation Gannascus was put out of the way — Iherigh? against a deserter he held himself entitled to take this bank aban- course — and he was on the point of crossing the Ems and advancing into the country of the Chauci, when not only did he receive counter-orders from Rome, but the Roman government in general completely altered its attitude on the lower Rhine. The emperor Claudius • directed the ' governor to remove all Roman garrisons from the right bank. We may well conceive that the imperial general with bitter words commended the good fortune of the free commanders of Rome in former days ; in this step certainly there was a conclusive admission of defeat, which had been but partially owned after the battle of Varus. Probably this restriction of the Roman occupation of Germany, which was not occasioned by any pressure of immediate necessity, was called forth by the quent posi- 126 ROMAN GERMANY book viii- resolve just then adopted to occupy Britain, and finds its justification in the fact that the troops were not sufficient for accomplishing both objects at once. That the order was executed, and matters remained afterwards in that position, is proved by the absence of Roman military in- scriptions on the whole right bank of the lower Rhine. 1 Only isolated points for crossing and sally-ports, such as, in particular, Deutz opposite Cologne, formed exceptions from this general rule. The military road keeps here to the left bank and strictly to the course of the Rhine, while the traffic-route running behind it, cutting off the windings, pursues the straight line of communication. Here on the right bank of the Rhine there is no evidence of Roman military roads, either through the discovery of milestones or otherwise, its subse- The withdrawal of the garrisons did not imply giving up possession, strictly speaking, of the right bank in this province. It was looked upon by the Romans thenceforth somewhat as the commandant of a fortress looks upon the ground that lies under his cannon. The Cannenefates and at least a part of the Frisians 2 were afterwards subject, as before, to the empire. We have already remarked that subse- quently in the Miinster country the herds of the legions still pastured, and the Germans were not allowed to settle there. But the government thenceforth relied — for the defence of such border-territory on the right bank as still existed 1 The fortress of Niederbiber, not to the time of Germanicus, as gens far from the point at which the Wied tumfida. Probably this is connected falls into the Rhine, as well as that of with the distinction between the Arzbach, nearMontabaur, in the region Frisii and Frisiavones in Pliny, H. N. of the Lahn, belong to upper Ger- iv. 15, ioi, and between the Frisii many. The special significance of maiores and minores in Tacitus, Germ. the former stronghold, the largest 34. The Frisians that remained fortress in upper Germany, turned Roman would be the western ; the on the fact that it, in a military point free, the eastern ; if the Frisians of view, closed the Roman lines on generally reach as far as the Ems the right bank of the Rhine. (Ptolem. iii. 11, 7), those subsequently 2 The levies {Eph.Epigr.y. p. 274) Roman may have settled perhaps to require us to assume this, while the the westward of the Yssel. We may Frisians, as they come forward in the not put them elsewhere than on the year 58 (Tacitus, Ann. xiii. 54) coast that still bears their name; the rather appear independent ; the elder designation in Pliny, iv. 17, 106, Pliny also (H. N. xxv. 3, 22) under stands isolated, and is beyond doubt Vespasian names them, looking back incorrect. chap. iv. AND THE FREE GERMANS. 127 in this province — in the north on the Cannenefates and the Frisians, and farther up the stream substantially on the space left desolate ; and, if it did not directly forbid, at any rate did not give scope to Roman settlement there. The altar stone of a private person found at Altenberg (circuit of Miilheim), on the river Dhiin, is almost the only evidence of Roman inhabitants in these regions. This is the more remarkable, as the prosperity of Cologne would, if special hindrances had not here stood in the way, have of itself carried Roman civilisation far and wide on the other bank. Often enough Roman troops may have traversed these extensive regions, perhaps even have kept the roads — which were here laid out in large number during the Augustan period — in some measure passable, and possibly laid out new ones ; sparse settlers, partly remains of the old Germanic population, partly colonists from the empire, may have settled here, similar to those that we shall soon find in the earlier imperial period on the right bank of the upper Rhine ; but the highways, like the possessions, lacked the stamp of durability. There was no wish to undertake here a labour of similar extent and difficulty to that which we shall become acquainted with further on in the upper province, or to provide here, as was done there, military defence and fortification for the frontier of the empire. Therefore the lower Rhine was crossed doubtless by Roman rule, but not, like the upper Rhine, also by Roman culture. For the double task of keeping the neighbouring Gaul The situa- in obedience and of keeping the Germans of the right "nd'cer" 1 ' bank aloof from Gaul, the army of the lower Rhine would, many after even after abandoning the occupation of the region on the N g ro a right of the river, have quite sufficed, and the peace with- out and within would not presumably have been inter- rupted, had not the downfall of the Julio-Claudian dynasty, and the civil or rather military war thereby called forth, exercised a momentous influence on these relations. The insurrection of the Celtic land under the leadership of Vindex was no doubt defeated by the two Germanic armies ; but Nero's fall nevertheless ensued, and when 128 ROMAN GERMANY book vm. the Spanish army as well as the imperial guard in Rome appointed a successor to him, the armies of the Rhine did the same ; and in the beginning of the year 6g the greater portion of these troops crossed the Alps to settle the point on the battle-fields of Italy, whether its ruler was to be called Marcus or Aulus. In May of the same year the new emperor Vitellius followed, after arms had decided in his favour, accompanied by the remainder of the good soldiers inured to war. The blanks in the garrisons ot the Rhine were no doubt filled up for the exigency by recruits hastily levied in Gaul ; but the whole land knew that they were not the old legions, and it soon became apparent that these were not coming back. If the new ruler had had in his power the army that placed him on the throne, at least a portion of them must have returned to the Rhine immediately after the defeat of Otho in April ; but the insubordination of the soldiers still more than the new complication which soon set in with the pro- clamation of Vespasian as emperor in the East, retained the German legions in Italy. Prepara- Gaul was in the most fearful excitement. The rising th^insur- of Vindex was, as we formerly remarked (p. 82), in itself rection. directed not against the rule of Rome but against the rulers for the time being ; but it was none the less on that account a warfare between the armies of the Rhine and the levy en masse of the great majority of the Celtic can- tons ; and these were none the less subjected to pillage and maltreatment resembling that of the conquered. The tone of feeling which subsisted between the provincials and the soldiers was shown, for instance, by the treatment which the canton of the Helvetii experienced as the troops destined for Italy marched through it. Because a courier despatched by the adherents of Vitellius to Pannonia had here been seized, the columns on the march from the one side, and the Romans stationed as a garrison in Raetia on the other, entered the canton, pillaged the villages far and wide, particularly what is now Baden near Zurich, chased those who had fled to the mountains out of their lurking-places, and put them to death by thousands or chap. iv. AND THE FREE GERMANS. 129 sold the captives under martial law. Although the capital Aventicum (Avenches, near Murten) submitted without resistance, the agitators of the army demanded that it should be razed, and all that the general granted was that the question should be referred not somehow to the em- peror, but to the soldiers of the great headquarters ; these sat in judgment on the fate of the town, and it was merely the turn of their caprice that saved the place from destruction. Outrages of this nature brought the provincials to extremities ; even before Vitellius left Gaul, a certain Mariccus, from the canton of the Boii, dependent on the Haedui, came forward a god on earth, as he said, and destined to restore the freedom of the Celts ; and people flocked in troops to his banner. But the exas- peration in the Celtic country was not of so very great moment. The very rising of Vindex had most clearly shown how utterly incapable the Gauls were of releasing themselves from the Roman embrace. But the tone of feeling of the Germanic districts Rising of reckoned as belonging to Gaul — in the modern Nether- \^ &3X& ' lands — of the Batavi, the Cannenefates, the Frisians, whose auxiliaries. distinctive position has already been dwelt on, had a somewhat greater importance ; and it happened that, on the one hand, these very tribes had been exasperated to the utmost, and on the other, that their contingents were accidentally to be found in Gaul. The bulk of the Bata- vian troops, 8000 men, assigned to the 14th legion, had for a considerable time a place along with the latter in the army of the upper Rhine, and had then under Claudius, on occasion of the occupying of Britain, gone to that island, where this corps shortly before had, by its incomparable valour, gained the decisive battle under Paullinus for the Romans ; from this day onward it occupied indisputably the first place among all the divi- sions of the Roman army. When it was recalled on account of this very distinction by Nero, in order to go off with him to the war in the East, the revolution break- ing out in Gaul had brought about a quarrel between the legion and its auxiliary troops ; the former, faithfully VOL. I. 9 the move- 130 ROMAN GERMANY book vm. devoted to Nero, hastened to Italy ; the Batavi, on the other hand, refused to follow. Perhaps this was con- nected with the fact that two of their most noted officers, Civiiis. the brothers Paulus and Civilis, had, without any reason and without respect to many years of faithful service and honourable wounds, been shortly before put on trial as suspected of high treason, and the former executed, the latter placed in captivity. After the downfall of Nero, to which the revolt of the Batavian cohorts had materially contributed, Galba released Civilis and sent the Batavians back to their old headquarters in Britain. While they, on the march thither, were encamped among the Lingones (Langres), the legions of the Rhine revolted from Galba and proclaimed Vitellius emperor. The Batavi, after consid- erable hesitation, ultimately joined the movement ; Vitellius did not forgive them for this hesitation, but did not venture directly to call to account the leader of the powerful corps. Progress of Thus the Batavians had marched with the legions of lower Germany to Italy and had fought with their usual valour in the battle of Betriacum for Vitellius, while their old legionary comrades confronted them in the army of Otho. But the arrogance of the Germans exasperated their Roman comrades in victory, however much these acknowledged their valour in battle ; the very generals in command did not trust them, and even made an attempt to divide by detaching them — a course, which, in this war, where the soldiers commanded and the generals obeyed, was not capable of being carried out, and had almost cost the general his life. After the victory they were commis- sioned to accompany their hostile comrades of the 14th legion to Britain ; but when matters came to a skirmish between the two at Turin, the latter alone went to Britain, and the Batavians to Germany. Meanwhile Vespasian had been proclaimed emperor in the East, and, while in consequence of this Vitellius gave to the Batavian cohorts marching orders for Italy as well as ordered new compre- hensive levies among the Batavi, commissioners of Ves- pasian opened communications with the Batavian officers to hinder this departure, and to provoke in Germany chap. iv. AND THE FREE GERMANS. 131 itself a rising which should detain the troops there. Civilis entered into the suggestion. He resorted to his home, and gained easily the assent of his own people as well as the neighbouring Cannenefates and Frisians. The insurrection broke out among the former ; the camps of the two cohorts in the neighbourhood were surprised and the Roman posts seized ; the Roman recruits fought ill ; soon Civilis with his cohort — which he had caused to follow, ostensibly to employ it against the insurgents — threw him- self openly into the movement, along with the three Germanic cantons renounced allegiance to Vitellius, and summoned the other Batavians and Cannenefates, who just then were breaking up from Mentz for the march to Italy, to join him. All this was more a soldiers' rising than an insurrection its char- of the province, or even a Germanic war. If at that time acter ' the Rhine legions were fighting with those of the Danube, and further with these and the army of the Euphrates, it was but in keeping that the soldiers of the second class, and above all their most distinguished troop, the Batavian, should enter independently into this divi- sional warfare. Any one who compares this movement among the cohorts of the Batavians and the Germans on the left of the Rhine with the insurrection of those on the right bank of the Rhine under Augustus, may not overlook the fact, that in the later rising the alae and cohorts took up the part of the general levy of the Cherusci ; and, if the perfidious officer of Varus released his nation from the Roman rule, the Batavian leader acted in the commission of Vespasian ; in fact, perhaps, on the secret directions of the governor of his province privately inclined towards Vespasian, and the rising in the first instance was directed simply against Vitellius. It is true that the position of things was such that this soldiers' revolt might change itself at any moment into a German war of the most dangerous kind. The same Roman troops who covered the Rhine against the Germans of the right bank were, in consequence of the corps -warfare, placed in an attitude of hostility to armies on the Rhine. 132 ROMAN GERMANY book vm. the Germans on the left bank ; the parts were of such a nature, that it seemed almost easier to exchange them than to carry them out. Civilis himself may possibly have left it to depend on the sequel, whether the movement would end in a change of emperor or in the expulsion of the Romans from Gaul by the Germans. state of the The command of the two armies on the Rhine was held at this time, after the governor of lower Germany had been made emperor, by his former colleague in upper Germany, Hordeonius Flaccus, a gouty man advanced in years, without energy and without authority, either, moreover, in fact secretly holding to Vespasian, or at any rate very much suspected of such faithlessness by the legions, who zealously adhered to the emperor of their own making. It is characteristic of him and of his position that, to clear himself of the suspicion of treason, he gave orders that the government despatches on arrival should be sent unopened to the eagle -bearers of the legions, and these should read them in the first instance to the soldiers, before they forwarded them to their address. Of the four legions of the lower army which had primarily to do with the insurgents two, the 5 th and the 15 th, were stationed under the legate Munius Lupercus in the headquarters at Vetera ; the 1 6th, under Numisius Rufus, in Novaesium (Neuss) ; the 1st, under Herennius Gallus, in Bonna (Bonn). Of the upper army, which then numbered only three legions, 1 one, the 21st, remained in its stated quarters Vindonissa, aloof from these events, if it had not rather been drawn off wholly to Italy ; the two others, the 4th Macedonian and the 2 2d, were stationed at the headquarters Mentz, where Flaccus also was present ; and in point of fact, his able legate Dillius Vocula exercised the chief command. The legions had throughout only half of their full complement, and most of the soldiers were half-invalids or recruits. First con- Civilis, at the head of a small number of regular troops, but of the collective levy of the Batavi, Cannene- 1 The fourth upper German legion on account of the Armeno-Parthian was sent in the year 58 to Asia Minor war (Tacitus, Ann. xiii. 35). fiicts. chap. iv. AND THE FREE GERMANS. 133 fates, and Frisians, advanced from his home to the attack. In the first instance,, on the Rhine he met with remnants of the Roman garrisons driven from the northern cantons and a division of the Roman Rhenish fleet ; when he attacked them, not merely did the ships' crews, consisting in great part of Batavians, go over to him, but also a cohort of the Tungri — it was the first revolt of a Gallic division ; such Italian soldiers as were present were slain or taken prisoners. This success brought at length the Germans on the right of the Rhine into the move- ment. What they had long vainly hoped for — the rising Participa- of the Roman subjects on the other bank — now came to l ' on of the J Germans be fulfilled, and as well the Chauci and the Frisians on on the right the coast, as above all, the Bructeri on both sides of the ^ n | upper Ems as far down as the Lippe, the Tencteri on the middle Rhine opposite to Cologne, and in lesser measure the tribes adjoining these on the south- — Usipes, Mattiaci, Chatti — threw themselves into the struggle. When, on the orders of Flaccus, the two weak legions marched out from Vetera against the insurgents, these could already confront them with a numerous contingent drawn from beyond the Rhine ; and the battle ended, like the combat on the Rhine, with a defeat of the Romans through the defection of the Batavian cavalry, which belonged to the garrison of Vetera, and through the bad behaviour of the cavalry of the Ubii and of the Treveri. The insurgents and the Germans who flocked to them siege of proceeded to invest and besiege the headquarters of the Vetera - lower army. During this siege news of the events on the lower Rhine reached the other Batavian cohorts in the neighbourhood of Mentz ; they at once wheeled round towards the north. Instead of ordering them to be cut down, the weak-minded commander-in-chief allowed them to go, and when the commandant of the legion in Bonn sought to intercept them, Flaccus did not support him as he might have done and had even at first promised. So the brave Germans dispersed the Bonn legion and succeeded in joining Civilis — henceforth the compact core of his army, in which now the banners of the Roman cohorts 134 ROMAN GERMANY book vm. stood by the side of the animal-standards from the sacred groves of the Germans. But still the Batavian held, at least ostensibly, by Vespasian ; he swore in the Roman troops in Vespasian's name, and summoned the garrison of Vetera to join him in declaring for the latter. These troops, however, saw in this, probably with warrant, a mere attempt to overreach them, and repelled it as resolutely as they repelled the assailing hosts of the enemy, who soon found themselves compelled by the superiority of Roman tactics to change the siege into a blockade. But, as the leaders of the Roman army had been taken by surprise in these events, provisions were scarce and speedy relief was urgently called for. In order to bring it, Flaccus and Vocula set out with their whole force from Mentz, drew to themselves on the way the two legions from Bonna and Novaesium as well as the auxiliary troops of the Gallic cantons \ appearing at the word of command in large numbers, and approached Vetera. Vocula. But instead of throwing at once the whole force from within and without on the besiegers, however great their superiority in numbers, Vocula pitched his camp at Gel- duba (Gellep on the Rhine, not far from Krefeld) a long day's march distant from Vetera, while Flaccus lay farther back. The worthlessness of the so-called general and the ever increasing demoralisation of the troops, above all, the distrust towards the officers, which frequently went so far as to maltreat and attempt to kill them, can alone at least explain this halting. Thus the mischief gradually thickened on all sides. All Germany seemed desirous to take part in the war ; while the besieging army constantly obtained new contingents from that quarter, other bands passed over the Rhine, which in this dry summer was unusually low, partly in the rear of the Romans into the cantons of the Ubii and the Treveri to lay waste the valley of the Moselle, partly below Vetera into the region of the Maas and the Scheldt ; further bands appeared before Mentz and made pretext of besieging it. Then came the accounts of the catastrophe in Italy. On the news of the chap. iv. AND THE FREE GERMANS. 135 second battle at Betriacum in the autumn of the year 69 the Germanic legions gave up the cause of Vitellius as lost and took the oath, though reluctantly, to Vespasian, perhaps in the hope that Civilis, who had in fact inscribed the name of Vespasian on his banners, would then make his peace. But the German swarms, who had meanwhile poured themselves over all northern Gaul, had not come to install the Flavian dynasty ; even if Civilis had ever wished this, he now had no longer the power. He threw off the mask, and openly expressed — what indeed was long settled — that the Germans of north Gaul intended, with the help of their free countrymen, to shake off the Roman rule. But the fortune of war changed. Civilis attempted to Relief of surprise the camp of Gelduba ; the attack began success- Vetera - fully, and the defection of the cohorts of the Nervii brought Vocula's little band into a critical position. Then suddenly two Spanish cohorts fell on the rear of the Ger- mans ; what threatened to be a defeat was converted into a brilliant victory ; the flower of the assailing army remained on the field of battle. Vocula indeed did not advance at once against Vetera, as he possibly might have done, but he penetrated into the besieged town some days later after a renewed vehement conflict with the enemy. It is true that he brought no provisions ; and, as the river was in the power of the enemy, these had to be procured by the land-route from Novaesium, where Flaccus was encamped. The first convoy passed through ; but the enemy, having meanwhile assembled again, attacked the second column with provisions on its way, and com- pelled it to throw itself into Gelduba. Vocula went off thither to its support with his troops and a part of the old garrison of Vetera. When they had arrived at Gelduba, the men refused to return to Vetera and to take upon themselves the further sufferings of the siege in prospect ; instead of this they marched to Novaesium, and Vocula, who knew that the remnant of the old garrison of Vetera was in some measure provisioned, had for good or evil to follow. 136 ROMAN GERMANY EOOK VIII. Mutiny of Roman troops. Insurrec- tion in Gaul. In Novaesium meanwhile mutiny had broken out. The soldiers had come to learn that a largess destined for them by Vitellius had reached the general, and com- pelled its distribution in the name of Vespasian. They had scarcely received it, when, in the wild carousing which ensued upon the largess, the old grudge of the soldiers broke out afresh ; they pillaged the house of the general who had betrayed the army of the Rhine to the general of the Syrian legions, slew him, and would have prepared the same fate for Vocula, if the latter had not escaped in disguise. Thereupon they once more proclaimed Vitel- lius emperor, not knowing that he was already dead. When this news came to the camp, the better part of the soldiers, and in particular the two upper German legions, began in some measure to reflect ; they again exchanged the effigy of Vitellius on their standards for that of Ves- pasian, and placed themselves under the orders of Vocula ; he led them to Mentz, where he remained during the rest of the winter 69-70. Civilis occupied Gelduba, and thereby cut off Vetera, which was most closely blockaded ; the camps of Novaesium and Bonna were still held. Hitherto the Gallic land, apart from the few insurgent Germanic cantons in the north, had kept firmly by Rome. Certainly partisanship ran through the several cantons ; among the Tungri, for example, the Batavi had a strong body of adherents, and the bad behaviour of the Gallic auxiliary troops during the whole campaign may probably have been in part called forth by such a temper of hos- tility to the Romans. But even among the insurgents there was a considerable party favourably disposed to Rome ; a Batavian of note, Claudius Labeo, waged a par- tisan warfare not without success against his countrymen in his home and its neighbourhood, and the nephew of Civilis, Julius Briganticus, fell in one of these combats at the head of a band of Roman horse. All the Gallic cantons had without more ado complied with the injunction to send contingents ; the Ubii, although of Germanic descent, were in this war mindful simply of their Romanism, and they as well as the Treveri had offered brave and success- chap. iv. AND THE FREE GERMANS. 137 ful resistance to the Germans invading their territory. It is easy to understand how this was so. The position of things in Gaul was still much as it was in the days of Caesar and Ariovistus ; a liberation of their Gallic home from the Roman dominion by means of those hordes, which, in order to lend to Civilis the help of his country- men, were just then pillaging the valleys of the Moselle, Maas, and Scheldt, was tantamount to a surrender of the land to its Germanic neighbours ; in this war, which had grown out of a feud between two corps of Roman troops into a conflict between Rome and Germany, the Gauls were, properly speaking, nothing but the stake and the booty. That the tone of feeling among the Gauls, in spite of all their well-founded general and special com- plaints as to the Roman government, was predominantly anti-Germanic, and that the materials for kindling such a national rising suddenly bursting into flame and reckless of consequences, as had -spread through the people in an earlier time, were wanting in this Gaul now half- Romanised, events up to this time had most clearly shown. But amidst the constant misfortunes of the Roman army the courage of the Gauls hostile to the Romans gradually grew stronger, and their defection completed the catas- trophe. Two Treveri of note, Julius Classicus, the com- mander of the Treverian cavalry, and Julius Tutor, com- mandant of the garrisons on the banks of the middle Rhine, Julius Sabinus one of the Lingones, descended, as he at least boasted, from a bastard of Caesar, and some other men of like mind from different cantons, professed in thoughtless Celtic fashion to discern that the destruc- tion of Rome was written in the stars and announced to the world by the burning of the Capitol (Dec. 69). So they resolved to set aside the Roman rule and to The Gallic set up a Gallic empire. For this purpose they took the em P ire course of Arminius. Vocula allowed himself to be really induced by falsified reports of these Roman officers to set out, with the contingents placed under their command and a part of the Mentz garrison, in the spring of 70 for the lower Rhine, in order with these troops and the legions of RO.VAX GERMAXY BOOK VIII. Capitula- tion of the Romans. End of the Gallic em- pire. Bonna and Novaesium to relieve the hard-pressed Vetera. On the march from Novaesium to Vetera, Classicus and the officers in concert with him left the Roman arm}- and proclaimed the new Gallic empire. Vocula led the legions back to Novaesium ; Classicus pitched his camp imme- diately in front of it. Vetera could not now hold out long ; the Romans could not but expect after its fall to find themselves confronted by the whole power of the enemy. The Roman troops refused to face this prospect and entered into a capitulation with the revolted officers. In vain Vocula attempted once more to urge the ties of dis- cipline and of honour ; the legions of Rome allowed a Roman deserter from the ist legion to stab the brave general on the order of Classicus, and themselves delivered up the other chief officers in chains to the representative of the empire of Gaul, who thereupon made the soldiers swear allegiance to that empire. The same oath was taken at the hands of the perfidious officers by the gar- rison of Vetera, which, compelled by famine, at once sur- rendered, and likewise by the garrison of Mentz, where but a few individuals avoided disgrace by flight or death. The whole proud army of the Rhine, the first army of the empire, had surrendered to its own auxiliaries ; Rome had surrendered to Gaul. It was a tragedy, and at the same time a farce. The Gallic empire lapsed, as it could not fail to do. Civilis and his Germans were doubtless, in the first instance, well content that the quarrel in the Roman camp delivered the one as well as the other half of their foes into their hands ; but he had no thought of recognising that empire, and still less had his allies from the right bank of the Rhine. As little would the Gauls themselves have anything to do with it — a result, to which certainly the split between the eastern districts and the rest of the country, which had already become apparent at the rising of Vindex, materially contributed. The Trcveri and the Lingones, whose leading men had instigated that camp- conspiracy, stood by their leaders, but they remained chap. iv. AND THE FREE GERMANS. 139 virtually alone ; only the Vangiones and Triboci joined them. The Sequani, into whose territory the Lingones marched to induce their accession, drove them summarily homeward. The esteemed Remi, the leading canton in Belgica, convoked the diet of the three Gauls, and, although there was no lack there of orators on behalf of political freedom, it resolved simply to dissuade the Treveri from the revolt. How the constitution of the new empire would have turned out, had it been established, it is difficult to say ; wc learn only that Sabinus, the great-grandson of Caesar's concubine, named himself also Caesar, and in this capacity allowed himself to be beaten by the Sequani ; whereas Classicus, who had not such ascendency at his command, assumed the insignia of Roman magistracy, and thus played perhaps the part of republican proconsul. In keeping with this there exists a coin, which must have been struck by Classicus or his adherents, exhibiting the head of Gallia, as the coins of the Roman republic show that of Roma, and by its side the symbol of the legion, with the genuinely audacious legend of " fidelity " (fides). At first, doubtless, on the Rhine the imperialists, in concert with the insurgent Germans, had full freedom. The remnants of the two legions that had capitulated in Vetera were put to death, in opposition to the expostulation and to the will of Civilis ; the two from Novaesium and Bonna were sent to Treves ; all the Roman camps on the Rhine, large and small, with the exception of Mogon- tiacum, were burnt. The Agrippincnses found them- selves in the worst plight. The imperialists had certainly confined themselves to requiring from them the oath of allegiance ; but the Germans in this case did not forget that they were, properly speaking, the Ubii. A message of the Tencteri from the right bank of the Rhine — this was one of the tribes whose old home the Romans had laid desolate and used as pasture -ground, and which had in consequence of this been obliged to seek other abodes — demanded the razing of this chief seat of the Germanic apostates, and the execution of all their citizens of Roman the Ro- mans. 140 ROMAN GERMANY book vm. descent. This would probably have been resolved on had not Civilis, who was personally under obligation to them, as well as the German prophetess Veleda in the canton of the Bructeri, who had predicted this victory, and whose authority the whole insurgent army recognised, interceded on their behalf. Advent of The victors were not left long to contend over the booty. The imperialists certainly gave the assurance that the civil war in Italy had broken out, that all the provinces were overrun by the enemy, and Vespasian was probably dead ; but the heavy arm of Rome was soon enough felt. The newly confirmed government could despatch its best generals and numerous legions to the Rhine ; and certainly an imposing display of power was there needed. Annius Gallus took up the command in the upper, Petillius Cerialis in the lower province ; the latter, an impetuous and often incautious, but brave and capable officer, took action in the proper sense. Besides the 2 1 st legion from Vindonissa, five came from Italy, three from Spain, one along with the fleet from Britain, and, in addition, a further corps from the Raetian garrison. This and the 21st legion were the first to arrive. The imperialists had possibly talked of blocking the passes of the Alps ; but nothing was done, and the whole country of the upper Rhine lay open as far as Mentz. The two Mentz legions had no doubt sworn allegiance to the Gallic empire, and at first offered resist- ance ; but, so soon as they perceived that a larger Roman army confronted them, they returned to obedience, and the Vangiones and Triboci immediately followed their example. Even the Lingones submitted — merely upon a promise of mild treatment — without striking a blow on the part of their 70,000 men capable of bearing arms. 1 The Treveri themselves had almost done the same ; but they were prevented from doing so by the nobility. The 1 Frontinus, Strat. iv. 3, 14. In sur-Beze, about fourteen miles north- their territory the advancing troops east of Dijon, men of at least five must have constructed a reserve of the advancing legions had ex- station and a depot ; according to ecuted buildings here (Hermes, xix. tiles recently found near Mirabeau- 437). chap. iv. AND THE FREE GERMANS. 141 two surviving legions of the lower Rhenish army that were stationed here had, on the first news of the approach of the Romans, torn the Gallic insignia from their standards, and withdrew to the Mediomatrici that had remained faithful (Metz), where they submitted to the mercy of the new general. When Cerialis arrived at the army, he found a good part of the work already done. The insurgent leaders exerted themselves, it is true, to the utmost — at that time by their orders the legionary legates delivered up at Novaesium were put to death — but in a military sense they were impotent, and their last political move — that of offering the Roman general himself the sovereignty of the Gallic empire — was worthy of the beginning. After a short combat Cerialis occupied the capital of the Treveri, the leaders and the whole council having taken refuge with the Germans. This was the end of the Gallic empire. More serious was the struggle with the Germans. Last Civilis, with his whole fighting strength, the Batavi, the o/afml contingent of the Germans, and the refugee bands of the Gallic insurgents, suddenly assailed the much weaker Roman army in Treves itself. The Roman camp was already in his power, and the bridge of the Moselle occupied by him, when his men, instead of following up the victory which they had won, began prematurely to pillage, and Cerialis, compensating for his imprudence by brilliant valour, restored the combat and ultimately drove the Germans out from the camp and the town. There was no further success of importance. The Agrippinenses again joined the Romans, and killed the Germans, who were staying among them, in their houses ; a whole Germanic cohort encamped there was shut up and burnt in its quarters. Whatsoever in Belgica still held to the Germans was brought back to obedience by the legion arriving from Britain ; a victory of the Cannenefates over the Roman ships which had landed the legion, and other isolated successes of the brave Germanic bands, above all, of the more numerous and better managed Germanic ships, did not change the general position of the war. H2 ROMAN GERMANY book viii. On the ruins of Vetera Civilis confronted the foe ; but he had to give way to the Roman army, which had mean- while been doubled, and at length, after an obstinate resistance, had to leave his own home to the enemy. As ever happens, discord ensued in the train of mis- fortune. Civilis was no longer sure of his own men, and sought and found protection from them among his opponents. Late in the autumn of the year 70 the unequal struggle was decided ; the auxiliaries now on their part surrendered to the burgess -legions, and the priestess Veleda went as a captive to Rome. Nature of When we look back on this war, one of the most the Roman singular a nd most dreadful in all ages, we cannot but task and 11, 1 its issue. own that hardly ever has an army had a task set before it equally severe with that of the two Roman armies on the Rhine in the years 6g and 70. In the course of a few months soldiers successively of Nero, of the senate, of Galba, of Vitellius, and of Vespasian ; the only support to the dominion of Italy over the two mighty nations of the Gauls and the Germans, while the soldiers of the auxiliaries were taken almost entirely, and those of the legions in great part, from those very nations ; deprived of their best men, mostly without pay, often starving, and beyond all measure wretchedly led, they were certainly expected to perform feats inwardly and outwardly super- human. They ill sustained the severe trial. This was less a war between two divisions of the army, like the other civil wars of this terrible time, than a war of soldiers, and above all of officers, of the second class against those of the first, combined with a dangerous insurrection and invasion of the Germans, and an incidental and insignificant revolt of some Celtic districts. In Roman military history Cannae and Carrhae and the Teutoburg Forest are glorious pages compared with the double disgrace of Novaesium ; only a few individual men, not a single troop, preserved a pure escutcheon amidst the general dishonour. The frightful disorganisa- tion of the political and, above all, of the military system, which meets us on the fall of the Julio-Claudian dynasty, chap. iv. AND THE FREE GERMANS. 143 appears — more clearly even than in the leaderless battle of Betriacum — in those events on the Rhine, to which the history of Rome never before and never after exhibits a parallel. The very extent and general diffusion of these mis- Conse- deeds rendered a corresponding chastisement impossible. theBata° f It deserves to be acknowledged that the new ruler, who ™n war. happily had remained in person aloof from all these occur- rences, in a genuine statesmanly fashion allowed the past to be past, and exerted himself only to prevent the repetition of similar scenes. That the prominent culprits, whether from the ranks of the troops or from the insur- gents, were brought to account for their crimes, was a matter of course ; we may measure the punishment by the fact that when five years afterwards one of the Gallic insurgent leaders was discovered in a lurking-place, in which his wife had up to that time kept him concealed, Vespasian gave him as well as her over to the executioner. But the renegade legions were allowed to share in the fighting against the Germans, and to atone for their guilt to some extent in the hot conflicts at Treves and at Vetera. It is true, nevertheless, that the four legions of the lower Rhenish army were all dismissed, as was one of the two upper Rhenish legions that took part — one would gladly believe that the 2 2d was spared in honour- able remembrance of its brave legate. Probably a consider- able number of the Batavian cohorts met with the same fate, and not less, apparently, the cavalry regiment of the Treveri, and perhaps several other specially prominent troops. Still less than against the rebellious soldiers could proceedings be taken with the full severity of the law against the insurgent Celtic and German cantons ; that the Roman legions demanded the razing of the Treverian colony of Augustus — this time for the sake not of booty but of vengeance — is at least as intelligible as the destruction, desired by the Germans, of the town of the Ubii ; but as Civilis protected the one so Vespasian protected the other. Even the Germans on the left of the Rhine had, on the whole, their previous position left 144 ROMAN GERMANY book viii. to them. But probably — we are here without certain tradition — there was introduced in the levy and the em- ployment of the auxilia an essential change, which diminished the danger involved in the auxiliary system. The Batavi retained freedom from taxation and a still privileged position as regards service ; a part of them, not altogether inconsiderable, had withal championed in arms the cause of the Romans. But the Batavian troops were considerably diminished, and, while hitherto — as it would appear of right — officers had been placed over them from their own nobility, and the same had been at least fre- quently done as respects the other Germanic and Celtic troops, the officers of the alae and cohortes were afterwards taken predominantly from the class from which Vespasian himself was descended — from the good urban middle class of Italy and of the provincial towns organised after the Italian fashion. Officers of the position of the Cheruscan Arminius, of the Batavian Civilis, of the Treverian Clas- sicus do not henceforth recur. As little is the previous close association of troops levied from the same canton met with subsequently; on the contrary, the men serve, without distinction as to their descent, in the most various divisions ; this was probably a lesson which the Roman military administration gathered from this war. It was another change, probably suggested by this war, that while hitherto the majority of the auxiliaries employed in Germany were taken from the Germanic and neighbouring cantons, thenceforth the Germanic auxiliary troops found prepon- derantly employment outside of their native country, just like the Dalmatian and Pannonian troops in consequence of the war with Bato. Vespasian was a soldier of sagacity and experience ; it is probably in good part a merit of his if we meet with no later example of revolt of the auxilia against their legions. Later That the insurrection, which we have just narrated, of TheRoman the Germans on the left of the Rhine— although it, in Germans consequence of the accidental completeness of the accounts lower 6 preserved respecting it, alone gives us a clear insight into Rhine. the political and military relations on the lower Rhine chap. iv. AND THE FREE GERMANS. 145 and in Gaul generally, and therefore deserved to be nar- rated in more detail — was yet called forth more by outward and accidental causes than by the inner necessity of things, is proved by the apparently complete quiet which now ensued there, and by the — so far as we can see — unin- terrupted status quo in this very region. The Roman Ger- mans were merged in the empire no less completely than the Roman Gauls ; of attempts at insurrection on the part of the former there is no further mention. At the close of the third century, the Franks invading Gaul by way of the lower Rhine included in their seizure the Batavian territory ; yet the Batavians maintained them- selves in their old though diminished settlements, as did like- wise the Frisians, even during the confusions of the great migration of peoples, and, so far as we know, preserved allegiance even to the decaying empire as a whole. When we turn from the Romans to the free Germans The free to the east of the Rhine, we find offensive action on their ? e T ans ' on tne part not less brought to an end with their participation lower in that Batavian insurrection, than the attempts of the me ' Romans to bring about an alteration of the frontier on a grand scale in those regions came to a close with the expeditions of Germanicus. Of the free Germans, those dwelling next to the Bmcteri. Roman territory were the Bructeri on both banks of the middle Ems, and in the region of the sources of the Ems and Lippe ; for which reason they took part before all the other Germans in the Batavian insurrection. To their canton belonged the maiden Veleda, who sent forth her countrymen to the war against Rome and promised them the victory, whose utterance decided the fate of the town of the Ubii, and to whose high tower the captive senators and the captured admiral's ship of the Rhenish fleet were sent. The overthrow of the Batavi affected them also ; and perhaps, in addition, a special counterblow of the Romans when that virgin was subsequently led as a cap- tive to Rome. This disaster, as well as feuds with the neighbouring tribes, broke their power ; under Nero a king whom they did not wish was obtruded on them by VOL. I. 10 146 ROMAN GERMANY BOOK VIII. Cherusci. Lango- bardi. Semnones. force of arms on the part of their neighbours with the passive assistance of the Roman legate. The Cherusci, in the region of the upper Weser, in the time of Augustus and Tiberius the leading canton in central Germany, is seldom mentioned after the death of Arminius, but always as sustaining good relations to the Romans. When the civil war, which must have continued to rage among them even after the fall of Arminius, had swept away the whole family of their princes, they re- quested from the Roman government the last of that house, Italicus, a brother's son of Arminius living in Italy, to be their ruler ; it is true that the return home of one who was brave but answered more to his name than to his lineage, kindled the feud afresh, and, when he was driven off by his own people, the Langobardi placed him once more on the tottering throne. One of his successors, king Chariomerus, so earnestly took the side of the Romans in Domitian's war with the Chatti, that he after its close, when driven away by the Chatti, fled to the Romans and invoked — although vainly — their interven- tion. Through those perpetual inward and outward feuds the Cheruscan people was so weakened that it henceforth disappears from active politics. The name of the Marsi is no longer met with at all after the expeditions of Ger- manicus. That the tribes dwelling farther to the east on the Elbe as well as all the more remote Germans took as little part in the struggles of the Batavians and their allies in the years 69 and 70, as these took in the German wars under Augustus and Tiberius may, considering the de- tailed character of the narrative, be described as certain. Where they meet us subsequently they never appear in a hostile attitude to the Romans. That the Langobardi reinstated the Roman king of the Cherusci, has already been mentioned. Masuus, the king of the Semnones, and — what is remarkable — along with him the prophetess Ganna, who was held in high repute among this tribe famous for its special credulity, visited the emperor Domitian in Rome, and met with a friendly reception at his court. In the regions from the Weser to the Elbe chap. iv. AND THE FREE GERMANS. 147 during these centuries various feuds may have raged, the balance of power may in various cases have shifted, various cantons may have changed their name or joined another combination ; as regards their relations to the Romans a permanent frontier-peace set in, after it came to be gener- ally felt that these had positively abandoned the subjuga- tion of this region. Even invasions from the far East cannot have materially disturbed it at this epoch ; for they could not but have reacted on the Roman guarding of the frontier, and we should not have lacked information had more serious crises occurred in this domain. All this is confirmed by the reduction of the army of the lower Rhine to half of its former amount, which occurred we know not exactly when, but within this epoch. The army of the lower Rhine, with which Vespasian had to fight, numbered four legions ; that of the time of Trajan presumably the same number, at least three j 1 probably already under Hadrian, certainly under Marcus, there were not more than two — the 1st Minervian and the 30th of Trajan — stationed there. Germanic affairs in the upper province developed Upper themselves after another fashion. Of the Germans on the left of the Rhine who belonged to this province, the Triboci, Nemetes, Vangiones, there is nothing historically worth mentioning, except that they, for long settled among the Celts, shared the destinies of Gaul. Here too the Rhine always remained the chief line of defence for the Romans. All the standing camps of the legions were at all times on the left bank of the Rhine ; not even that of 1 Under the legate Q. Acutius 22d Primigenia drafted off elsewhere. .Nerva, who was probably the consul But it is far more probable — seeing of the year 100, and so administered that all the legions were not always lower Germany after that year, there taking part in the detachments to the were stationed, according to inscrip- stone quarries at Brohl — that these tions of Brohl (Brambach, 660, 662, four legions were doing garrison-duty 679, 680), in this province four le- at the same time in lower Germany, gions, the 1st Minervia, 6th Victrix, These four legions are probably just 10th Gemina, 22d Primigenia. As those that came to lower Germany each of these inscriptions names only on the- reorganisation of the Germanic two or three, the garrison may then armies by Vespasian (p. 159 note), have consisted only of three legions, only that the 1st Minervia was put by if during the governorship of Acutius Domitian in the place of the 21st, the 1st Minervia came in place of the probably broken up by him. 148 ROMAN GERMANY book viii. Argentoratum was transferred to the right bank, when the whole region of the Neckar was Roman. But while in the lower province the Roman rule on the right bank of the Rhine was restricted in course of time, here on the other hand it was extended. The project of Augustus /|to connect the camps on the Rhine with those on the Danube by advancing the imperial frontier in an east- ward direction — which, if it had been carried out, would have enlarged upper more than lower Germany — was perhaps never completely abandoned in this command, and was resumed subsequently, though on a more modest scale. Historical tradition does not give us the means of presenting a connected view of the operations continued with this object for centuries, the construction of roads and walls pertaining thereto, and the wars waged on this account ; and even the great military structure still exist- ing, whose rise and progress — likewise embracing centuries — must include in itself a good part of that history, has hitherto not been investigated throughout, as it well might be, by the eyes of military experts. The hope that unified Germany would combine for the investigation of this its oldest historical monument, has not been fulfilled. We shall here attempt to put together what has hitherto been brought to light on the subject from the fragments of the Roman annals or of the Roman strongholds. Mogonti- On the right bank, not far from the northern end of the province, there stretches in front of the level or hilly country of the lower Rhine, in a direction from west to east, the range of the Taunus, which abuts on the Rhine opposite to Bingen. Parallel to this mountain-range, shut off on the other side by the spurs of the Odenwald, stretches the plain of the Jawer Main-valley, the true ^access to„-the..iaterior of Germany, dominated by the key of the position at the point where the Main falls Jnto the Rhine, Mogontiacum or Mentz, from the time of Drusus down 1'cHfie— end of Rome the stronghold out of which the Romans sallied to attack Germany from Gaul, 1 as it 1 According to the ingenious de- deutsche Zeitschrift, iii. 307 ff), it is cipherings of Zangemeister {West- established that a military road was acum. chap. iv. AND THE FREE GERMANS. 149 is at the present day the true barrier of Germany against France. Here the Romans, even after they had abandoned their rule in the region of the upper Rhine generally, retained not merely the tete-de-pont on the other bank, the castellum Mogontiacense (Castel), but also that plain of the Main itself in their possession ; and in this region a Roman civilisation might establish itself. This land Mattiaci. originally belonged to the Chatti, and a Chattan tribe, the Mattiaci, remained settled here even under Roman rule ; but, after the Chatti were compelled to cede this district to Drusus, it remained a part of the empire. The hot springs in the immediate neighbourhood of Mentz (aquae Mattiacae, Wiesbaden) were used by the Romans de- monstrably in Vespasian's time, and already even long before : silver was worked here under Claudius ; the Mattiaci already furnished troops to the army at an early date like other subject districts. They took part in the general rising of the Germans under Civilis ; but, after they were vanquished, the earlier relations were re- established. From the end of the second century we find the community of the Taunensian Mattiaci under authorities ,gj ganised jiftgxjhj;_Ro man model. 1 The Chatti, although thus driven away from the, Chatti. Rhine, appear in the~sequel as the most powerful among \ the_tribes_qfMnland Germany who came into, contact j with Jthe Romans ; the lead which, under Augustus and Tiberius, had been possessed by the Cherusci on the middle Weser, passed, amidst the constant feuds with these their southern cognate neighbours, over to the latter. All the wars between Romans and Germans, of which we have any knowledge from the time after the death of Arminius down to the time when the migrations of the already laid out under Claudius on ensium, with Duoviri, Aediles, De- the left bank of the Rhine from Mentz curiones, Sacerdotales, Seviri ; peculiar as far as the frontier of the upper and characteristic of a frontier town German province. are the hastiferi civitatis Mattiacorum, 1 The full name c(ivitas) M(attia- probably to be taken as a municipal corum) Ta(unensium) appears on the militia (Brambach, 1336). The oldest inscription of Castel in Brambach, C. dated document of this community is /. Rh. 1330; it occurs frequently as of the year 198 (Brambach, 956). civitas Mattiacorum or civitas Taun- iSo ROMAN GERMANY book vm. ]) peoples began at the end of the third century, were waged j against the Chatti ; as in the year 4 1 under Claudius by Galba, who became afterwards emperor ; and in the year 50 under the same emperor by Publius Pomponius Secundus, celebrated as a poet. These were the usual border incursions, and the Chatti had taken a part, but only a secondary one, in the great Batavian war (p. 133). But in the campaign which the Emperor Domitian under- took in the year 8 3 the Romans were the aggressors ; and this war led, not indeed to brilliant victories, but doubtless to a considerable and momentous pushing for- ward of the Roman frontier. 1 At that time the frontier- line was arranged, as we find it thenceforth drawn ; and within that line, which in its most northern portion was not far removed from the Rhine, must have been included a great part of the Taunus and the region of the Main as far as above Friedberg. The Usipes, who, after their already-mentioned expulsion from the region of the Lippe, appear about the time of Vespasian in the neighbourhood of Mentz, and may have found new settlements to the east of the Mattiaci on the Kinzig or in the Fuldan dis- trict, were then annexed to the empire, and, at the same time with them, a number of smaller tribes thrown off by the Chatti. Thereupon, when in the year 88, under the governor Lucius Antonius Saturninus, the upper German 1 The accounts of this war have probably corrupt) castella poneret, and been lost; its time and place admit i. 3, 10: limitibus per cxx. m.p.actis, of being determined. As the coins which is here brought into immediate give to Domitian the title Germanicus connection with the military opera- after the beginning of the year 84 tions, and hence may not be separated (Eckhel, vi. 378, 397), the campaign from the Chattan war itself and re- falls in the year 83. Accordant with ferred to the agri decumates, which this is the levy of the Usipes, which had for long been in the Roman falls on this same year, and their power. The measure of 108 miles desperate attempt at flight (Tacitus, is very conceivable for the military Agr. 28 ; comp. Martialis, vi. 60). line which Domitian planned at the It was an aggressive war (Suetonius, Taunus (according to Cohausen's esti- Dom. 6 : expeditio sponte suscepta ; mates, Rom Grenzwall, p. 8, the later Zonaras, ' xi. 19 ; "KerfKarficras tiv& Limes from the Rhine round the tux Tripoli 'Pfyov tuv haw6vioiv). Taunus as far as the Main is set down The shifting of the line of posts is at 137 miles), but is much too small attested by Frontinus, who took part to admit of its being referred to the in the war, Strat. ii. 11, 7 : cum in line of connection from thence to finibus Cubiorum (name unknown and Ratisbon. CHAP. IV. AND THE FREE GERMANS. 151 army rose against Domitian, the war was on the point of renewal ; the revolted troops made common cause with the Chatti ; a and it was only the interruption of the com- munications, when the ice broke up on the Rhine, that made it possible for the regiments which had remained faithful to settle matters with the revolters before the dangerous contingent arrived. It is stated that the Roman rule extended from Mentz towards the interior 80 leugae, and thus even beyond Fulda ; 2 and this 1 The Germans (Suetonius, Dom. 6) could only be the Chatti, and their earlier allies, perhaps in the first in- stance just the Usipes and those sharing their fate. The insurrection broke out in Mentz, which alone was a double camp of two legions. Satuminus was assailed from Raetia by the troops of L. Appius Maximus Nor- banus. For the epigram of Martial, ix. 84, cannot be understood other- wise, the more especially as his con- queror, of senatorial rank as he was, could not administer a regular com- mand in Raetia and Vindelicia, and could only be led into this region by a case of war emerging, as indeed the sacrilegi furores clearly point to the insurrection. The tiles of this same Appius, which have been found in the provinces of upper Germany and Aquitania, do not warrant the mak- ing him legate of the Lugdunensis, as Asbach (Westdeutsche Zeitschrift, iii. 9), suggests, but must be referred to the epoch after the defeat of An- tonius (Hermes, xix. 438). Where the battle was fought remains doubt- ful ; the region of Vindonissa most naturally suggests itself, to which point Satuminus may have gone to meet Norbanus. Had Norbanus encountered the insurgents only at Mentz, which in itself seems con- ceivable, these would have had the crossing of the Rhine in their power, and the contingent of the Germans could not have been hindered by the breaking-up of the Rhine from rein- forcing them. 2 The detached notice is found subjoined to the Veronese provincial list (Notitia dignitatum, ed. Seeck, p. 253) : nomina civitatum trans Renum flwvium quae sunt; Usiphorum (read Usiporum) — Tuvanium (read Tubantum) — Nictrensium — Novarii — Casuariorum : istae omnes civitates trans Renum in formulam Belgicae primae redactae trans castellum Mon- tiacese: namlxxx.leugas trans Renum Romani possederunt. Istae civitates sub Gallieno imperatore a barbaris oc- cupatae sunt. That the Usipes after- wards dwelt in this region, is confirmed by Tacitus, Hist. iv. 37, Germ. 32 ; that they belonged to the empire in the year 83, but had perhaps been made subject only shortly before, is plain from the narrative, Agr. 28. The Tubantes and Chasuarii are placed by Ptolemy, ii. II, II, in the vicinity of the Chatti ; that they shared the fate of the Usipes is accordingly probable. No certain identification of the other two cor- rupt names has hitherto been found ; perhaps the Tencteri had a place here, or some of the small tribes named with these only in Ptolemy, ii. 11,6. The notice in its original form named Belgica simply, as the province was only divided by Diocletian, and named it rightly in so far as the two Germanies belonged geographically to Belgica. The specified measurement carries us, if we follow the Kinzig valley to the north-east, beyond Fulda nearly to Hersfeld. Inscriptions have been found here far eastward beyond the Rhine, as far as the Wetterau ; Friedberg and Butzbach were mili- tary positions strongly garrisoned ; at Altenstadt between Friedberg and 152 ROMAN GERMANY book viii s account appears worthy of credit, if we take into con- sideration that the military frontier-line, which certainly seems not to have gone far above Friedberg, doubtless kept here also within the territorial boundary. The region But not merely was the valley of the lower Main in Neckar. fr° nt °f Mentz brought within the military frontier-line ; in south-western Germany also the boundary was pushed forward in a still greater degree. The region of the Neckar, once possessed by the Celtic Helvetii, then for long a debateable borderland between these and the ad- vancing Germans, and therefore named the Helvetian desert, subsequently perhaps occupied partially by the Mar- comani, before these retreated to Bohemia (p. 29), came on the regulation of the Germanic boundaries after the battle of Varus into the same position as the greater portion of the right bank of the lower Rhine. Here, too, there must have been a frontier-line already at that time marked off, within which Germanic settlements were not tolerated. Thereupon individual, mostly Gallic, immigrants, who had not much to lose, settled down, as on an unenclosed moor, in these fertile but little protected regions, which went at that time by the name of agri decumates} This private occupation, which was, it may be conjectured, merely tolerated by the government, was followed by the formal taking possession of it probably under Vespasian. As already, about the year 74, a highway was carried from Strassburg on the right bank of the Rhine as far as Offenburg, 2 there must have been instituted about this Biidingen there has been found an by the first who took possession upon inscription of the year 242 (Bram- payment of the tenth ; but neither is bach, C. I. Rh. 1410) pointing to it linguistically proved that decumas protection of the frontier [collegium can mean "liable for a tenth," nor iuventutis). are we acquainted with such arrange- 1 What the designation agri decu- ments in the imperial period. More- mates (for the latter word is at any- over it should not be overlooked that rate to be connected with agri) oc- the description of Tacitus refers to curring only in Tacitus, Germ. 29, the time before the institution of the means, is uncertain. It is possible line of the Neckar ; it does not suit that the territory regarded in the the latter period any more than does earlier imperial period certainly as the designation, which doubtless is not property of the state or rather of the clear, but is at any rate certainly con- emperor, like the old ager occupa- nected with the earlier legal relation. tonus of the republic, might be used 2 This has been proved by Zange- chap. iv. AND THE FREE GERMANS. 153 time in this region a more earnest protection of the fron- tier than the mere prohibition of Germanic settlement furnished. What the father had begun the sons carried out. Perhaps even through the construction — whether by Vespasian, by Titus, or Domitian — of the "Flavian altars" 1 at the source of the Neckar, near the modern Rottweil — a settlement of which indeed we know nothing but the name — there was procured for the new upper Germany on the right of the Rhine a centre similar to what the Ubian altar was formerly intended to become for Great Germany, and soon afterwards the altar of Sarmizegetusa became for the newly-conquered Dacia. The first institu- tion of the frontier-defence, to be described further on, by which the Neckar valley was brought within the Roman line, is thus the work of the Flavii, chiefly, doubtless, of Domitian, 2 who thereby carried further the construction at the Taunus. The military road on the right of the Rhine from Mogontiacum by way of Heidelberg and Baden in the direction of Offenburg — the necessary con- sequence of this annexation of the Neckar region — was, as we now know, 3 constructed by Trajan in the year 100, and was a part of the more direct communication established by that emperor between Gaul and the line of the Danube. There was employment for the soldiers at these works, but hardly for their arms ; there were no Germanic tribes dwelling in the region of the Neckar, and still less can the narrow strip on the left bank of the Danube, which was thereby brought within the frontier line, have cost serious struggles. The nearest Germanic people of note there, the Hermunduri, had more friendly meister (Westdeutsche Zeitschrift, iii. this or that deceased Flavian emperor p. 246). as to his own Genius. 1 The fact that here several altars 2 That the transfer took place were dedicated, while elsewhere at shortly before Tacitus wrote the Ger- these central sanctuaries only one is mania in the year 98, he himself states, mentioned, may be explained perhaps and that Domitian was its author, fol- by the cultus of Roma falling into lows from the fact 'that he does not the background by the side of that of name the author, the emperors. If at the very outset 3 This, too, has been document- several altars were erected, which is arily established by Zangemeister probable, perhaps one of the sons (Westdeutsche Zeitschrift, iii. 237 f.). caused altars to be set up as well to 154 ROMAN GERMANY book vm. dispositions towards the Romans than any other tribe had, and carried on lively commercial intercourse with them in the town of the Vindelici, Augusta ; of the fact that this advance met with no resistance from them, we shall find traces further on. Under the following reigns of Hadrian, Pius, and Marcus, further progress was made with these military arrangements. The upper We cannot historically follow out the mode in which L?™ s anic the frontier-fence between the Rhine and the Danube — still in great part subsisting as regards its foundations at the present day — came into existence, but we are able to recognise not merely the course which it took but also the purpose which it served. The work was as to its nature and purpose different in upper Germany from what it was in Raetia. The upper German frontier- fence, with a length in all of about 250 Roman miles (228 English miles 1 ) begins immediately at the northern boundary of the province, embraces, as has been already said, the Taunus and the plain of the Main as far as the district of Friedberg, and turns thence southward to the Main, which it meets at Grosskrotzenburg above Hanau. Following the Main thence as far as Worth, it here takes the direction of the Neckar, which it reaches somewhat below Wimpfen and does not again leave. Afterwards in front of the southern half of this frontier-line a second was laid out, which follows the Main by way of Worth as far as Miltenberg, and thence is led for the most part in a straight direction to Lorch between Stuttgart and Aalen. Here to the upper German frontier-fence is joined on the Raetian, only 120 miles (108 English) long ; it leaves the Danube at Kelheim above Ratisbon and runs thence, twice crossing the Altmuhl, in a curve westward likewise as far as Lorch. 1 This measurement holds for the case of the older line of the Neckar line of forts from Rheinbrohl to Lorch the rampart is considerably shorter, {C6ha.usen,derJ?dm.GrenzwaH,-p.j{.). since, instead of that from Miltenberg For the earthen rampart there falls to to Lorch, here comes in the much be deducted the stretch of the Main shorter one of the Odenwald from from Miltenberg to Grosskrotzenburg, Worth to Wimpfen. of about thirty Roman miles. In the chap. iv. AND THE FREE GERMANS. 155 The upper Germanic Limes consists of a series of forts which are distant from each other, at the most, half a day's march (about nine English miles). Where the lines of connection between the forts are not closed by the Main or the Neckar, as stated above, there was introduced an artificial barrier, at first perhaps merely by a palisade, 1 after- wards by a continuous wall of moderate height with fosse in front on the outside and watch-towers built in at short intervals on the inner side. 2 The forts are not introduced into the wall, but constructed immediately behind it at a distance seldom exceeding one-third of an English mile. The Raetian frontier-fence was a mere barrier, pro- The Raet- duced by piling up quarry-stones ; there were no fosses ian Limes ' or watch-towers, and the forts, constructed behind the Limes without regular succession and at unequal intervals (none nearer than two and a half to three miles), stand in no immediate connection with the barrier-line. As to the order in time of the constructions there is no definite testimony ; it is proved that the upper Germanic line of the Neckar was in existence under Piiis, 3 that placed in front 1 If, as is probable, the statement wall stood before their eyes in its in- that Hadrian blocked the imperial tegrity and significance. Whether frontier-roads by palisades against the the "region" Palas which Ammianus barbarians (p. 122) relates in part and mentions (xviii. 2, 15) is connected perhaps primarily to the upper Ger- with this is doubtful, manic, the wall, of which remains are 2 In such an one recently discovered extant, was not his work ; whether between the forts of Schlossau and this may have carried palisades or not, Hesselbach, 1850 yards from the no report would mention these and former, about three miles from the pass over the wall itself. Dio, Ixix. latter, there has been found a votive in- 9, says that Hadrian revised the de- scnpMan(KorrespcmdenzUattder West- knee of the frontier throughout the deutschen Zeitschrift, 1 Jul. 1884), empire. The designation of the pale which the troop that built it — a de- [Pfakl] or pale-ditch \Pfahlgraberi\ tachment of the 1st cohort of the cannot be Roman ; in Latin the Sequani and Raurici under command stakes, which, driven into the wall of of a centurion of the 22d legion, the camp, form a palisade-chain for it, erected as thanksgiving ob burgum are called not pali, but valli or sudes, explic(itum). These towers thus were just as the wall itself is never other burgi. than ■vallum. If the designation in 3 The oldest dated evidence for use from of old for this purpose ap- these is two inscriptions of the garri- parently along the whole line among son of Bockingen, opposite Heilbronn, the Germans was really borrowed on the left bank of the Neckar of the from the palisades, it must have been year 148 (Brambach, C. I. Rli. 15831 of Germanic origin, and can only have 1 590). proceeded from the time when this 1 56 ROMAN GERMANY book viii. of it from Miltenberg to Lorch under Marcus. 1 The idea of a frontier-bar was common to the two structures, otherwise so different ; the preference in the one case for the piling up of earth — whence the fosse for the most part resulted of itself — in the other case, for layers of stone, probably depended only on the diversity of the soil and of the materials for building. It was common to them, further, that neither the one nor the other was constructed for the defence, as a whole, of the frontier. Not merely was the hindrance, which the piling up of earth or stone presented to the assailant, slight in itself ; but along the line we meet everywhere with commanding positions, morasses lying in the rear, a want of outlook towards the country in front, and similar clear indications of the fact, that in the tracing of it warlike purposes generally were not contemplated. The forts are of course arranged for defence, each by itself, but they are not connected by paved cross-roads ; and so the individual garrison relied for support not on those of the neighbouring forts, but on the rear-base, to which the road led, whereby each was kept garrisoned. More- over, these garrisons were not dovetailed into a military system of frontier defence ; they were rather fortified posi- tions for a case of need than strategically chosen for the occupation of the territory, as indeed the very extent of the line itself, compared with the number of troops at disposal, excludes the possibility of its defence as a whole. 2 1 The oldest dated evidence for the 2 As to the distribution of the existence of this line is the inscription upper German troops there is a want of vkus Aurelii (Oehringen) of the of sufficient information, but not en- year 169 (Brambach, C. I. Rh. 1558), tirely of data on which to rest. Of doubtless only private, but certainly the two headquarters in upper Ger- not set up before the construction of many, that of Strassburg can be shown this fort belonging to the Miltenberg- to have been after the construction of Lorch line ; little later is that of Jags- the line of the Neckar occupied but thausen, likewise belonging to that weakly, and was probably more an line, of the year 179 (C. I. Rh. 1618). administrative than a military centre Accordingly vkus Aurelii might take ( Westdeutsches Correspondenzblatt, its name from Marcus, not from Cara- 1884, p. 132). On the other hand, calla, though it is attested of the latter the garrison of Mentz always demanded that he constructed various forts in a considerable portion of the aggre- these regions and named them after gate strength, all the more because .it himself (Dio, lxxvi. 13). was probably the only compact body CHAP. IV. AND THE FREE GERMANS. 157 Thus these extensive military structures had not, like the Object of Britannic wall, the object of checking the invasion of the these struc " tllT6S enemy. The intention rather was, that, like the bridges over the river-frontier, so the roads on the land-frontier should be commanded by the forts, but in other respects, like the river as the water -boundary, so the wall on the landward should hinder the uncontrolled crossing of the frontier. Other uses might be combined_with this ; the preference, often apparent, for the rectilineal direction points to its application for signals, and occasionally the structure may have been used directly for purposes of war. But the proper and immediate object of the struc- ture was to prevent the crossing of the frontier. The fact, withal, that watch -posts and forts were erected, not on the Raetian but on the upper Germanic frontier, is of troops on a large scale in all upper Germany. The other troops were distributed partly to the Limes, whose forts, according to Cohausen's esti- mate {Rom. Grenzwall, p. 335), were on an average five miles apart from one another, and so in all about fifty; partly to the interior forts, especially on the line of the Odenwaldfrom Giindels- heim to Worth; that the latter, at least in part, remained occupied even after the laying out of the outer Limes, is at least probable. Owing to the inequality in size of the forts still mea- surable, it is difficult to say what num- ber of troops was required to make them capable of defence. Cohausen (/. c. p. 340) reckons to a middle- sized fort, including the reserve, 720 men. As the usual cohort of the legion as of the auxiliaries numbered 500 men, and the fort-buildings must necessarily have had regard to this fact, the garrison of the fort in the event of siege must be estimated on an average at least at this number. After the reduction the upper German army could not possibly have held the forts, even of the Limes alone, simultaneously in this strength. Much less could it, even before the reduction, have kept the lines between the forts even barely occupied with its 30,000 men (p. 119); and, if this was not possible, the simultaneous occupation of all the forts had in fact no object. To all appearance each fort was planned in such a way that, when duly garrisoned, it could be held ; but, as a rule — and on this frontier the state of peace was the rule — the individual fort was not on a war-footing, but only furnished with troops, in so far that posts might be stationed in the watch - towers, and the roads as well as the byways might be kept under inspection. The stand- ing garrisons of the forts were, it may be conjectured, very much weaker than is usually assumed. We pos- sess from antiquity but a single record of such a garrison ; it is of the year 155, and relates to the fort of Kut- lowitza, to the north of Sofia (Eph. Epigr. iv. p. 524), for which the army of lower Moesia, and in fact the nth legion,- furnished the garrison. This troop numbered at that time, besides the centurion in command, only 76 men. The Raetian army was, at least before Marcus, still less in a position to occupy extensive lines ; it numbered then at the most 10,000 men, and had, besides the Raetian Limes, to supply also the line of the Danube from Ratisbon to Passau. 158 ROMAN GERMANY book vm. explained by their different relations to the neighbours, in the former case to the Hermunduri, in the latter to the Chatti. The Romans in upper Germany did not con- front their neighbours as they confronted the Highlanders of Britain, in whose presence the province was always in a state of siege ; but the repulse of predatory invaders as well as the levying of the frontier-dues demanded at any rate ready and near military help. The upper German army, and in keeping with it the garrisons on the Limes, might be gradually reduced, but the Roman pilum could never be dispensed with in the land of the Neckar. It might, however, be dispensed with in presence of the Hermunduri, who, in Trajan's time, alone of all the Germans, were at liberty to cross the frontier of the empire without special control and to trade freely in the Roman territory, especially in Augsburg, and with whom, so far as we know, border- collisions never took place. There was thus at this period no occasion for a similar structure on the Raetian frontier ; the forts north of the Danube, which can be shown to have subsisted already in Trajan's time, 1 sufficed here for the protection of the frontier and the control of frontier -intercourse. This accords with the observation that the Raetian Limes, as it stands before our eyes, corresponds only with the more recent upper Ger- manic barrier- line perhaps laid out for the first time under Marcus. Then occasion for it was not wanting. The wars of the Chatti, as we shall see (p. 161 ), seized at this time also on Raetia ; the strengthening too of the garrison of the province might reasonably stand in con- nection with the erection of this Limes, which, however little it was arranged for military ends, was at any rate doubtless constructed with a view to its being a frontier- bar, though of less strong character. 2 1 This is proved by the document man one for military occupation. A of Trajan of the year 107, found at weaker frontier-bar of that sort may Weissenburg. reasonably, even before the Marco- 2 The investigations hitherto as to manian war, have been chosen to face theRaetianLimeshavebutlittlecleared the Hermunduri ; nor does what up the destination of this work ; this Tacitus says of their intercourse in only is made out that it was less Augusta Vindelicum by any means ex- adapted than the analogous upper Ger- elude the existence at that time of a chap. iv. AND THE FREE GERMANS. 159 In a military as well as a political sense the shifting Their of the frontier, or rather the strengthening of the frontier- effect ' fence, was effective and useful. While formerly the Roman chain of forts in upper Germany and Raetia probably went up the Rhine by way of Strassburg to Basel and along by Vindonissa on the lake of Constance, then from thence to the upper Danube, now the upper Ger- man headquarters were in Mentz and the Raetian in Ratisbon, and generally the two chief armies of the empire were brought considerably nearer to each other. The legionary camp of Vindonissa (Windisch near Zurich) became thereby superfluous. The army of the upper Rhine might, like the neighbouring one, be reduced after some time to the half of its former strength. The original number of four legions, which was only acci- dentally diminished to three during the Batavian war, subsisted, at all events, probably still under Trajan j 1 but under Marcus the province was only occupied by two legions, the 8th and the 2 2d, of which the former was stationed at Strassburg, the second at the headquarters Mentz, while most of the troops, broken up into smaller posts, were stationed along the frontier-wall. Within the new line urban life flourished almost as on the left bank of the Rhine ; Sumelocenna (Rottenburg on the Neckar), Aquae (civitas Aurelia Aquensis, Baden), Lopodunum Raetian Limes. Only in that case we note), the 2d Adiutrix probably to should expect that it would not end at Britain (p. 174, note 4), the 1 3th Lorch, but would join the line of the Gemina (if this came to Germany at Neckar ; and in some measure it does all) to Pannonia ; the other seven re- this, inasmuch as at Lorch instead of mained, namely, in the lower pro- the Limes comes the Rems, which falls vince the 6th, 10th, 2ist, and 22d into the Neckar at Canstatt. (p. 147, note), in the upper the 8th, 1 Of the seven legions which at nth, and 14th. To the latter was Nero's deathwere stationed in the two probably added in the year 88 the 1st Germanies (p. 132), Vespasian broke Adiutrix, once more sent from Spain up five; there remained the 21st and to upper Germany (p. 65 note). That the 22d, to which, thereupon, were under Trajan the 1st Adiutrix and the added the seven or eight legions in- nth were stationed in upper Ger- troduced for the suppression of the many is shown by the inscription of revolt.thelst Adiutrix,2d Adiutrix,6th Baden-Baden (Brambach, C. I. Rli. Victrix, 8th and 10th Gemina, nth, 1666). The 8th and the 14th, it can 13th (?), and 14th. Of these, after be shown, both came with Cerialis to the close of the war, the 1st Adiutrix Germany, and both did garrison duty was sent probably to Spain (p. 65, there for a considerable period. i6o ROMAN GERMANY EOOK VIII. Germany under Marcus. (Ladenburg), had, if we except Cologne and Treves, to fear no comparison as respects Roman urban development with any town of Belgica. The rise of these settlements was chiefly the work of Trajan, who began his govern- ment with this act of peace j 1 " the Rhine Roman on both its banks " is what a Roman poet entreats the yet unseen ruler speedily to send to them. The great and fertile region, which was placed in this way under the protection of the legions, needed that protection and was worthy of it. Doubtless the battle of Varus marks the beginning of the ebb of Roman power, but only in so far as its advance was thereby ended, and the Romans thenceforth contented themselves in general with shielding more vigorously and continuously what was retained. Down to the beginning of the third century the Roman power on the Rhine showed no indications of tottering. During the war with the Marcomani under Marcus all remained quiet in the lower province. If a legate of Belgica had at that time to call out the general levy against the Chauci, this was presumably a piratical expedi- tion, such as often visited the north coast at this time, just as earlier and later. The surge of the great movement of peoples reached to the sources of the Danube and even as far as the region of the Rhine ; but it did not shake the 1 Traian was sent by Nerva in the year 96 or 97 as legate to Germany, probably to the upper, as at that time Vestricius Spurinna seems to have pre- sided over the lower. Nominated here as co-regentinOctoberofthe year97,he received the accounts of Nerva's death and of his nomination as the Augustus in February 98 at Cologne. He may have remained there during the winter and the following summer ; in the winter 98-99 he was on the Danube. The words of Eutropius, viii. 2 : urbes trans Rhenum in Germania reparavit (whence the often misused notice in Orosius, vii. 12, 2, has been copied), which can only be referred to the upper province, but naturally apply not to the legate, but to the Caesar or the Augustus, obtain a confirmation through the civitas Ulpia s(altus?) Nlyicerini?) Lopochmum of the inscrip- tions. The "restoration" may stand in contrast not to the institutions of Domitian, but to the irregular germs of urban arrangements in the Decu- mates-land before the shifting of the military frontier. There is no indica- tion pointing to warlike events under Trajan ; that he planned and gave his name (Ammianus, xvii. 1, 11) to a castellum in Alamannorum solo — ac- cording to the connection, on the Main not far from Mentz — is as little proof of such events as the circum- stance that a. later poet (Sidonius, Carm. vii. 115), mixing up old and new, makes Agrippina under him the terror of the Sugambri — that is, in his sense, of the Franks. the Alamanni. chap. iv. AND THE FREE GERMANS. 161 foundations there. The Chatti, the only considerable Germanic tribe on the upper German and Raetian border- fence, pushed forward in both directions, and were pro- bably at that time even among the Germans invading Italy, as will be shown further on when we describe this war. At any rate the reinforcement of the Raetian army at that time ordained by Marcus, and its conversion into a command of the first class with legion and legates, can only have taken place in order to check the attacks of the Chatti, and proves that they did not treat them lightly as regards the future. The already-mentioned strengthening of the border-defence would likewise stand connected with this movement. These measures must have sufficed for the next generation. Under Antoninus the son of Severus a new and more war with severe war once more (2 1 3) broke out in Raetia. This also was waged against the Chatti ; but by their side a second people is named, which we here meet for the first time — the Alamanni. Whence they came, we know not. According to a Roman writing a little later they were a conflux of mixed elements ; the appellation also seems to point to a league of communities, as well as the fact that afterwards the different tribes comprehended under this name stand forth — more than is the case among the other great Germanic peoples — in their separate character, and the Juthungi, the Lentienses, and other Alamannic peoples not seldom act independently. But that it is not the Germans of this region who here emerge allied under the new name and strengthened by the alliance, is shown as well by the naming of the Alamanni alongside of the Chatti, as by the mention of the unwonted skiful- ness of the Alamanni in equestrian combat. On the con- trary it was certainly, in the main, hordes coming on from the East that lent new strength to the almost extinguished German resistance on the Rhine ; it is not improbable that the powerful Semnones, in earlier times dwelling on the middle Elbe, of whom there is no further mention after the end of the second century, furnished a strong con- tingent to the Alamanni. The constantly increasing mis- VOL. I. -II l62 ROMAN GERMANY BOOK VIII. Severus Antoninus. government in the Roman empire naturally contributed its share, although only in a secondary degree, to the shifting of power. The emperor took the field in person against the new foe ; in August of the year 2 1 3 he crossed the Roman frontier, and a victory over them on the Main was achieved or at least celebrated ; further forts were con- structed ; the tribes of the Elbe and of the North Sea sent deputies to the Roman ruler, and wondered when in receiving them he wore their own dress, with silver-mounted jacket, and hair and beard coloured and arranged after the German fashion. But thenceforth the wars on the Rhine are incessant, and the aggressors are the Germans ; the neigh- bours formerly so pliant had as it were exchanged charac- ters. Twenty years later the inroads of the barbarians on the Danube as on the Rhine were so constant and so Alexander, serious, that the emperor Alexander had on their account to break off the less immediately dangerous Persian war and to resort in person to the camp of Mentz, not so much to defend the territory as to purchase peace from the Ger- mans by large sums of money. The exasperation of the soldiers at this led to his murder (a.D. 235), and thereby to the fall of the Severian dynasty, the last that existed at all until the regeneration of the state. His successor Maximinus, a rough but brave Thracian who had risen from the position of a common soldier, compensated for the cowardly conduct of his predecessor by an energetic expedition into the heart of Germany. The barbarians did not yet venture to face a strong and well-led Roman army ; they retreated to their forests and morasses, and the brave emperor, following them even thither, fought in front of all hand to hand. From these conflicts, which were doubtless directed from Mentz primarily against the Alamanni, he could with right call himself Germanicus ; and even for the future the expedi- tion of the year 236, for long the last great victory which the Romans gained on the Rhine, bore some fruit. Although the constant and bloody changes on the throne and the grave disasters in the East and on the Danube allowed the Romans no time to breathe, during the next Maxi- minus. chap. iv. AND THE FREE GERMANS. 163 twenty years, if peace was not strictly preserved on the Rhine a greater disaster did not occur. It appears even that one of the upper German legions was at that time sent to Africa without its place being supplied, and so upper Germany was held as tolerably secure. But when in the year 253 the different generals of Rome were once more fighting each other for the imperial dignity, and the Rhine-legions marched to Italy to fight out the cause of their emperor Valerianus against the Aemilianus of the Danube-army, this seems to have been the signal 1 for the Germans pushing forward especially towards the lower Rhine. 2 These Germans were the Franks, who appear The here for the first time, perhaps new opponents only in ra s ' name ; for, although the identification of them, already to be met with in later antiquity, with tribes formerly named on the lower Rhine — partly, the Chamavi settled beside the Bructeri, partly the Sugambri formerly mentioned subject to the Romans — is uncertain and at least inade- quate, there is here greater probability than in the case of the Alamanni that the Germans hitherto dependent on Rome on the right bank of the Rhine, and the Germanic tribes previously dislodged from the Rhine, took at that time — under the collective name of the "Free" — the offensive in concert against the Romans. So long as Gallienus himself remained on the Rhine, Gaiiienus. he, notwithstanding the small forces that were at his dis- posal, kept his opponents to some extent in check, pre- vented them from crossing the river, or drove out again the intruders, although he doubtless ceded to one of the Germanic leaders a portion of the desired territory on the 1 Not merely the causal connec- perhaps even if the coin in Cohen, n. tion, but even the chronological sue- 54, is to be trusted, the title Ger- cession of these important events is manicus maximus ter. obscure. The account, relatively the 2 That the Germans, against whom best, in Zosimus, i. 29, describes the Gallienus had to fight, are to be Germanic war as the cause why Va- sought at least chiefly on the lower lerian immediately on ascending the Rhine, is shown by the residence of throne in 253 made his son joint-ruler his son in Agrippina, where he can with equal rights ; and Valerian bears only have remained behind as nominal the title Germanicus maximus as representative of his father. His early as 256 (C /. L, viii. 2380 ; biographer also, <.. 8, names the likewise in 259 C. I. L. xi. 826), Franks. 164 ROMAN GERMANY BOOK VIII. river-bank, under the condition of his acknowledging the Roman rule and defending his possession against his countrymen — which indeed almost amounted to a capitu- lation. But when the emperor, recalled by the still more dangerous position of affairs on the Danube, resorted thither and left behind as representative in Gaul his elder son still in boyhood, one of the officers, to whom he had intrusted the defence of the frontier and the guardianship Postumus of his son, Marcus Cassianius Latinius Postumus, 1 got 1 It is difficult to form a concep- tion of the degree of historical falsifi- cation which prevails in a portion of the Imperial Biographies ; it will not be amiss to present here a specimen of it in the account of Postumus. He is here called (no doubt in an in- serted document) Iulius Postumus (Tyr. 6), on the coins and inscrip- tions M. Cassianius Latinius Postu- mus, in the epitomised Victor, 32, Cassius Labienus Postumus. — He reigns seven years ( Gall. 4) ; Tyr. 3, 5 ; the coins name his tr. p. X., and Eutropius, ix, 10, gives him ten years. — His opponent is called Lollianus, ac- cording to the coins Ulpius Cornelius Laelianus, Laelianus in Eutropius ix. 9 (according to the one class of manu- scripts, while the other follows the interpolation of the biographers) and in Victor (c. 33), Aelianus in the epi- tome of Victor. — Postumus and Vic- torinus rule jointly according to the biographer ; but there are no coins common to both, and consequently these confirm the report in Victor and Eutropius that Victorinus was the successor of Postumus. — It is a pecu- liarity of this class of falsifications that they reach their culmination in the documents inserted. The Cologne epitaph of the two Victorini (Tyr. 7), hie duo Victorini tyranni ( !) siti sunt criticises itself. The alleged com- mission of Valerian, whereby the latter communicates to the Gauls the nomination of Postumus, not only praises prophetically the gifts of Pos- tumus as a ruler, but names also various impossible offices ; a Trans- rhenani limitis dux et Galliae prae- ses at no time existed, and Pos- tumus &pxty & KeXroiS CTpaTiojroiv ifiireinaTev/itvos (Zosimus, i. 38) can only have been praeses of one of the two Germanies, or, if his command was an extraordinary one, dux per Germanias. Equally impossible is, in the same quasi-document, the tribu- natus Vocontiorum of the son, an evident imitation of the tribunates, as they emerge in the Notitia Dign. of the time of Honorius. — Against Postumus and Victorinus, under whom the Gauls and the Franks fight, Gallienus marches with Aureolus, afterwards his opponent, and the later emperor Claudius ; he himself is wounded by a shot from an arrow, but is victori- ous, without any change being pro- duced by the victory. Of this war the other accounts know nothing. Postumus falls in the military insur- rection instigated by the so-called Lollianus, while according to the re- port in Victor and Eutropius, Postu- mus becomes master of this Mentz insurrection, but then the soldiers kill him because he will not deliver up Mentz to them for plunder. As to the elevation of Postumus, by the side of the narrative which agrees in the main with the ordinary one, that Postumus had perfidiously set aside the son of Gallienus entrusted to his guardianship, stands another evidently invented to clear him, according to which the people in Gaul did this, and then offered the crown to Postu- mus. The tendency to eulogise one who had spared Gaul the fate of the Danubian lands and of Asia and had saved it from the Germans, comes chap. iv. AND THE FREE GERMANS. 165 himself proclaimed by his men as emperor and besieged in Cologne Silvanus the guardian of the emperor's son. He was successful in capturing the town and in getting into his power his former colleague as well as the imperial boy, whereupon he had them both executed. But during this confusion the Franks burst over the Rhine, and not merely overflowed all Gaul, but penetrated also into Spain and indeed pillaged even the coast of Africa. Soon after- wards, when the capture of Valerian by the Persians had filled up the measure of misfortune, all the Roman land on the left bank of the Rhine in the upper province was lost, passing doubtless to the Alamanni, whose eruption into Italy in the last years of Gallienus necessarily pre- supposes this loss. He is the last emperor whose name is found on monuments on the right of the Rhine. His coins celebrate him on account of five great victories over the Germans, and not less are those of his successor in the Gallic rule, Postumus, full of the praise of the German victories of the deliverer of Gaul. Gallienus in his earlier years had taken up the struggle on the Rhine not without energy, and Postumus was even an excellent officer and would gladly have been a good regent ; but amidst the utter unruliness which then prevailed in the Roman state or rather in the Roman army, the talent and ability of the individual profited neither himself nor the commonwealth. A series of flourishing Roman towns was at that time laid desolate by the invading barbarians, and the right bank of the Rhine was for ever lost to the Romans. The re-establishment of peace and order in Gaul was Aureiianus. primarily dependent on the cohesion of the empire generally ; so long as the Italian emperors stationed their troops in the Narbonensis to set aside the Gallic here and everywhere (most obviously an honourable secondary part. This at Tyr. 5) to light ; with which is con- narrative, not confused but thoroughly nected the fact that this report knows falsified, must be completely set aside ; nothing of the loss of the right bank the reports on the one hand in Zosi- of the Rhine and of the expeditions mus, on the other in the Latins draw- of the Franks to Gaul, Spain, and ing from a common source — Victor Africa. It is further significant that and Eutropius, short and confused as the alleged progenitor of the Con- they are, can alone be taken into stantinian house is here provided with account. 1 66 ROMAN GERMANY book vm. rival, and the latter in turn made as though he would cross the Alps, effective operations against the Germans were of themselves excluded. It was only after that, about the year 272, 1 the then ruler of Gaul, Tetricus, weary of his ungrateful part, had himself brought about the submission of his troops to Aurelianus, the emperor recognised by the Roman senate, that the thought of warding off the Germans could be again entertained. The raids of the Alamanni, who had for almost ten years ravaged upper Italy as far down as Ravenna, had a stop put to them for long by the same able ruler who had brought Gaul back to the empire, and he emphatically defeated one of their tribes, the Juthungi, on the upper Danube. If his government had lasted he would doubt- less have renewed the protection of the frontier also in Gaul; after his speedy and sudden end (275) the Germans once more crossed the Rhine and devastated the country far and wide. Probus. His successor Probus (from 276), also an able soldier, not merely drove them out afresh — he is said to have taken from them seventy towns — but also advanced again on the aggressive, crossed the Rhine, and drove the Germans back over the Neckar. He did not, how- ever, renew the lines of the earlier time, 2 but contented 1 The rule of Postumus lasted ten 41, 2), this appears to be superficially years (p. 164, note 1). That the elder reckoned according to the Chronicle son of Gallienus was already dead in of Jerome. The usual exact numbers 259, we learn from the inscription of are unattested and deceptive. Modena, C. I. L. xi. 826 ; the 2 According to the biographer, revolt of Postumus thus falls certainly c. 14, 15, Probus brought the Ger- in or before this year. As the mans of the right bank of the Rhine captivity of Tetricus cannot well be into dependence, so that they were placed later than 272, immediately tributary to the Romans and defended after the second expedition against the frontier for them (omnes jam bar- Zenobia, and the three Gallic rulers bari vobis arant, votes jam serviunt et reigned, Postumus for ten years, contra interiores geiites militant) ; the Victorinus for two (Eutropius, ix. 9), right of bearing arms is left to them Tetricus for two (Victor, 35), this for the time, but the idea is, on brings the revolt of Postumus to further successes, to push forward the somewhere about 259 ; yet such frontier and erect a province of Ger- numbers are frequently somewhat mania. Even as free fancies of a deranged. When the duration of Roman of the fourth century — more the expeditions of the Germans into they are not — these utterances have a Spain under Gallienus is definitely certain interest, stated at twelve years (Orosius, vii. chap. iv. AND THE FREE GERMANS. 167 himself with erecting and occupying at the more important positions of the Rhine tetes de pont on the other bank, — that is, he reverted nearly to such arrangements as had subsisted here before Vespasian. At the same time the Franks were defeated by his generals in the northern province. Great masses of the vanquished Germans were sent as forced settlers to Gaul, and above all to Britain. In this way the frontier of the Rhine was won back and handed over to the later empire. No doubt, like the rule on the right bank of the Rhine, peace on the left had passed away beyond recall. The Alamanni stood in a threatening attitude opposite to Basel and Strassburg, the Franks opposite to Cologne. By their side other tribes presented themselves. The fact that the Burgundiones, once settled beyond the Elbe, advancing > westward as far as the upper Main, threatened Gaul, is first mentioned under the emperor Probus ; a few years later the Saxons, in concert with the Franks, began their attacks by sea on the north coast of Gaul as on the Roman Britain. But under the — for the most part — vigorous and capable emperors of the Diocletiano- Constantinian house, and even under their immediate successors, the Romans kept the threatening inundation of peoples within measured bounds. To depict the Germans in their national development Romanis- is not the task of the historian of the Romans : for him ™s of the ' (jermans. they appear only as hindering or as destroying. An interpenetration of the two nationalities, and a mixed culture thence resulting, such as the Romanised land of the Celts presented, Roman Germany has none to show ; or — so far as concerns our conception of it — it coincides with the Romano-Gallic all the more, that the Germanic territories on the left bank of the Rhine, which remained for a considerable time in the Roman possession, were pervaded throughout with Celtic elements, and even those on the right, deprived for the most part of their original population, obtained the majority of the new settlers from Gaul. Communal centres, such as the Celtic system [ possessed in large number, were wanting to the German ) Roman Germanis- ing. 1 68 ROMAN GERMANY book viii. element. Partly on that account, partly in consequence of outward circumstances, the Roman element was able, as has been already brought "out (p. 102), to develop itself sooner and more fully in the Germanic east than in the Celtic regions. The encampments of the army of the Rhine, all of which fell within Roman Germany, were of essential influence in this respect. The larger of them obtained, partly through the traders who attached themselves to the army, partly, \ and above all, through the veterans who remained in their wonted quarters even after their discharge, an urban appendage — a town of huts (canabae), separate from the military quarters proper ; everywhere, and particularly in Germany, towns proper grew in time out of these at the legionary camps and especially the headquarters. At their head stood the Roman town of the Ubii, originally the second largest camp of the army of the lower Rhine, then from the year 50 onward a Roman colony (p. 99), exercising the most important effect in elevating Roman civilisation in the region of the Rhine. Here the camp-town gave place to that of the Roman plantation ; subsequently urban rights were obtained, without shifting the quarters of the troops, by the settlements belonging to the two great camps of the lower Rhine — Ulpia Noviomagus, in the land of the Batavi, and Ulpia Traiana, near Vetera — from Trajan, and in the third century by the military capital of upper Germany, Mogontiacum. No doubt these civil towns always retained a subordinate position by the side of the military centres of administration independent of them. If we look beyond the limit where this narrative closes, we certainly find, instead of the Romanising of the Germans, in some measure a Germanising of the Romans. The last phase of the Roman state was marked i by its becoming barbarian, and especially becoming Ger- i manised ; and the beginnings of the process reach far (back. It commences with the peasantry in the colonate, passes on to the troop as modelled by the emperor Severus, seizes then on the officers and magistrates, and ends with chap. iv. AND THE FREE GERMANS. 169 the Romano-Germanic mixed states of the Visigoths in Spain and Gaul, the Vandals in Africa, above all, with the Italy of Theoderic. For the understanding of this last phase there is certainly needed an insight into the political development of the one as of the other nation. No doubt in this respect Germanic research stands so far at a disadvantage, as the political arrangements into which these Germans "entered as servants" or joint rulers are well known, far better than the systematic history of the same epoch, while over the contemporary condition of the Germans floats that gray morning -haze in which sharp outlines are lost. German heathenism, apart from the far north, perished before the time of which we have knowledge ; and the religious elements, which are never wanting in a national war, we know doubtless for the 1 Sassanidae, but not for the Marcomani. The beginnings of the political development of the Germans are delineated for us in part by the picture of Tacitus — party-coloured, hampered by modelling itself on the ideas of a fading past, and but too often keeping silence as to elements of really decisive moment — while in part we must take them from the hybrid states which arose on formerly Roman soil and had Roman elements everywhere inwoven. How the people and the league of peoples, how king and nobles, how freedom and non-freedom, were moulded in those ■ circles from which Arminius and Theoderic came forth, we have no such definite and precise knowledge as we have of the contemporary conditions of the antiquated civilisation with which that youthful vigour strove, and in concert with which it — conquering and conquered — called the newer world of culture into life. However research may attempt to carry its torch into the early stages of Germanic growth, we shall never be able to picture to ourselves the two antagonists with an equally vivid clearness. CHAPTER V. BRITAIN. Caesar and NlNETY-SEVEN years elapsed from the time when Roman Emperors" troops had entered, subdued, and again abandoned the great island in the north-western ocean, before the Roman government resolved to repeat the voyage and permanently to occupy Britain. Certainly Caesar's Britannic expedition had not been, like his campaigns against the Germans, a mere forward movement of defence. So far as his arm reached, he had made the individual tribes subject to the empire, and had regulated their annual tribute to it in this case as in Gaul. The leading tribe, too, which was to be firmly attached to Rome by its privileged position and thereby to become the fulcrum of Roman rule, was found ; the Trinovantes (Essex) were to take up on the Celtic island the same part — more advantageous than honourable — as the Haedui and the Remi on the Gallic continent. The bloody feud between the prince Cassivellaunus and the princely house of Camalodunum (Colchester) had been the immediate cause of the Roman invasion ; to reinstate this house Caesar had landed, and the object was for the moment attained. Beyond doubt Caesar never deceived himself as to the fact that that tribute, as well as this protectorate, were in the first instance mere words ; but these words were a programme which could not but bring about, and was intended to bring about, the permanent occupation of the island by Roman troops. Caesar himself did not get so far as permanently to chap.v. - BRITAIN. 171 organise the affairs of the subject island ; and for his successors Britain was a perplexity. The Britons who had become subject to the empire certainly did not long pay — perhaps never paid at all — the tribute which was due. The protectorate over the dynasty of Camalodunum must have been still less respected, and had simply as its effect, that princes and scions of that house again and again appeared in Rome and invoked the intervention of the Roman government against neighbours and rivals. Thus king Dubnovellaunus, probably the successor of the prince of the Trinovantes confirmed by Caesar, came as a refugee to Rome to the emperor Augustus, and so, later, one of the princes of the same house came to the emperor Gaius. 1 In fact the expedition to Britain was a necessary part of the heritage left by Caesar. Already during the dual rule Caesar the younger had projected such an expedition, and had only desisted from it on account of the more urgent necessity of procuring quiet in Illyricum, or on account of the strained relation with Antonius, which proved useful to the Parthians in the first instance as well as to the , Britons. The courtly poets of the earlier years of Augustus celebrated variously in anticipation the Britannic conquest ; the programme of Caesar was thus accepted and adopted by his successor. When the monarchy was consolidated, all Rome thereupon expected that the close of the civil war would be followed by the 1 To all appearance the political p. 200 ; comp. Tacitus, Ann. ii. 24). relations between Rome and Britain Cunobelinus, according to the coins in the time before the conquest are the son of king Tasciovanus, of whom to be regarded essentially as arising history is silent, dying as it would out of the restoration and guarantee seem in advanced years between 40 (B. G. v. 22) of the principality of and 43, and so running parallel in his the Trinovantes by Caesar. That government with the latter reign of king Dubnovellaunus, who along with Augustus and those of Tiberius and anotherquiteunknownBritannicprince Gaius, resided in Camalodunum (Dio, sought protection with Augustus, ruled Ix. 2 1 ) ; around him and his sons the chiefly in Essex, is shown by his coins preliminary history of the invasion (my Man. Ancyr. 2d ed., p. 138 f.). turns. To what quarter Bericus, who We have to seek also mainly there came to Claudius (Dio, lx. 19), be- the Britannic princes who sent to longed we do not know, and other Augustus and recognised his supremacy British dynasts may have followed the (for such apparently we must take to example of those of Colchester ; but be the meaning of Strabo, iv. 5, 3, these stand at the head. 172 BRITAIN. book vm. Britannic expedition ; the complaints of the poets as to the dreadful strife, without which the Britons would long since have been led in triumphal procession to the Capitol, became transformed into the proud hope of adding to the empire the new province of Britain. The 27. expedition was, moreover, repeatedly announced (727, 26. 728), yet Augustus, without formally abandoning the undertaking, soon desisted from carrying it out ; and Tiberius, faithful to his maxim, adhered in this question also to the system of his father. 1 The worthless thoughts of the last Julian emperor roamed doubtless also over the ocean ; but serious things he was incapable of even planning. It was the government of Claudius that first took up the plan of the dictator afresh and carried it out. Thereasons What were the determining motives, on the one side agains" d the as on tne °ther, may be at least partially discerned, occupation Augustus himself laid it down that the occupation of the island was not necessary from a military point of view — seeing that its inhabitants were not in a position to annoy the Romans on the continent — and was not advantageous for the finances ; that what was to be drawn from Britain flowed into the exchequer of the empire in the form of import and export duties at the Gallic harbours ; that at least a legion and some cavalry would be requisite as garrison, and after deduction of its cost from the tribute of the island not much would be left. 2 All this was indisputably correct, and, in fact, by no means enough ; experience showed later that a legion was far from sufficient to hold the island. We must further take into account, what the government certainly had no occasion to say, that, considering the state of weakness to which the Roman army had been brought 'by the internal policy of Augustus, it could not but appear very 1 Tacitus, Agr. 13, consilium id the free traffic and therewith the divus Augustus vocabat, Tiberius prae- produce of the customs would decline, ceptum. must doubtless be taken as conced- 2 The exposition in Strabo, ii. 5, 8, ing the proposition that the Roman p. 115; iv. 5, 3, p. 200, gives rule and the Roman tribute affected evidently the governmental version, injuriously the prosperity of the That, after annexation of the island, subjects. chap. v. BRITAIN. 173 hazardous to banish a considerable fragment of it, once for all, to a distant island of the North Sea. There was presumably only the choice of keeping aloof from Britain or increasing the army on its account ; and with Augustus considerations of internal policy always outweighed those of an external character. But yet the conviction of the necessity for subduing Conviction Britain must have predominated with Roman statesmen. °ity S pre- eS Caesar's conduct would be inconceivable if we do not pre- dominant. suppose that conviction in his case. Augustus at first formally recognised, and never formally disowned, the aim proposed by Caesar, notwithstanding its inconvenience. It was precisely the governments that were the most far-seeing and most tenacious of purpose — those of Claudius, Nero, and Domitian — that laid the foundation for the conquest of Britain, or extended the work ; and, after it had taken place, it was never regarded in any such light as, possibly, the conquest by Trajan of Dacia and Mesopotamia. If the maxim of government, elsewhere adhered to almost inviolably, that the Roman empire had simply to fill, but not to extend, its bounds, was permanently set aside only in respect of Britain, the cause lies in the fact that the Celts could not be subdued in such a way as Rome's interest demanded, on the continent alone. This nation was to all appearance more connected than separated by the narrow arm of the sea which parts England and France ; the same names of peoples meet us on the one side and on the other ; the bounds of the individual states often reach over the Channel ; the chief seat of the priestly system, which here more than anywhere else pervaded the whole nationality, was from of old the islands of the North Sea. These islanders indeed were not able to wrest the continent of Gaul from the Roman legions ; but, if the conqueror of Gaul himself, and further the Roman government in Gaul, pursued other aims than in Syria and Egypt — if the Celts were to be annexed as members to the Italian nation — this task was doubtless impracticable, so long as the subjugated and the free Celtic territories touched each other over the sea, and the enemy of the 174 BRITAIN. BOOK VIII. Occasion for the war. Cunobe- linus. Military arrange- ments for occupying the island. Romans as well as the Roman deserter found an asylum in Britain. 1 In the first instance the subjugation of the southern coast sufficed for this purpose, although the effect was naturally the greater, the farther the free Celtic territory was pushed back. The special regard of Clau- dius for his Gallic home and his knowledge of Gallic relations may also have played a part in the matter. 2 What furnished occasion for the war was the fact that that very principality which sustained a certain dependence on Rome under the leadership of its king Cunobelinus — this was Shakespeare's Cymbeline — ex- tended widely its rule, 3 and emancipated itself from the Roman protectorate. One of his sons — Adminius, who had revolted against his father, came to the emperor Gaius de- siring protection, and upon his successor refusing to deliver up to the British ruler these his subjects, the war arose in the first instance against the father and the brothers of this Adminius. The proper motive for it, indeed, was the in- dispensable need for completing the conquest of a nation hitherto but half vanquished and keeping closely together. That the occupation of Britain could not ensue without a contemporary increase of the standing army was also the view of those statesmen who gave occasion to it ; three of the Rhine-legions and one from the Danube were destined thither, 4 but at the same time two newly instituted legions were assigned to the Germanic armies. An able 4 The three legions of the Rhine were the 2d Augusta, the 14th, and the 20th ; from Pannonia came the 9th Spanish. The same four legions were still stationed there at the begin- ning of the government of Vespasian ; the latter called away the 14th for the war against Civilis, and it did not return to Britain, but, in its stead, probably the 2d Adiutrix. This was presumably transferred under Domi- tian to Pannonia ; under Hadrian the 9th was broken up and replaced by the 6th Victrix. The two other legions, the 2d Augusta and the 20th, were stationed in England from the beginning to the end of the Roman rule. 1 Suetonius, Claud. 17, specifies as cause of the war : Britanniam tunc tumultuantem ob non redditos trans- fugas ; which O. Hirschfeld justly brings into connection with Gai. 44 : Adminio Cunobellini Britannorum re- gis filio, qui pulsus a patre cum exigua manu transfugerat, in deditionem re- cepto. By the tumultuari are doubt- less meant at least projected expedi- tions for pillage to the Gallic coast. The war was certainly not waged on account of Bericus (Dio, Ix. 19). 2 Mona was in like manner after- wards receptaculum perfugarum (Taci- tus, Ann. xiv. 29). s Tacitus, Ann. xii. 37 : pluribtts gentibus imperitantem. chap. v. BRITAIN. 175 soldier, Aulus Plautius, was selected as leader of this expedition, and at the same time as first governor of the province ; it departed for the island in the year 43. The soldiers showed themselves reluctant, more doubtless because of the banishment to the distant island than from fear of the foe. One of the leading men, perhaps the soul of the undertaking, Narcissus, the emperor's cabinet- secretary, wished to instil into them courage ; they did not allow the slave to utter a word for their shouts of scoffing, but did withal as he wished and embarked. The occupation of the island was not attended by any Course of special difficulty. The natives stood, in a political as in !^ e n occupa " a military point of view, at the same low stage of develop- ment which Caesar had previously found in the island. Kings or queens reigned in the several cantons, which had no outward bond of conjunction and were at perpetual feud with one another. The men were doubtless possessed of bodily strength, endurance, and bravery — despising death ; and were in particular expert horsemen. But the Homeric war-chariot, which was still a reality here, and on which the princes of the land themselves wielded the reins, as little held its ground against the compact squad- rons of Roman cavalry as the foot soldier without coat of mail and helmet, defended only by the small shield, was with his short javelin and his broad sword a match in close combat for the short Roman knife, or even for the heavy pilum of the legionary, and the plummet and arrow of the light Roman troops. To the army of about 40,000 well-trained soldiers the natives could oppose no corre- sponding defensive force. The disembarkation did not even encounter resistance ; the Britons had accounts as to the reluctant temper of the troops and no longer expected the landing. King Cunobelinus had died shortly before ; the opposition was led by his two sons Caratacus and Togodumnus. The invading army had its march at once directed to Camalodunum, 1 and in a rapid course of victory 1 The identification, based only on not be correct ; these first conflicts dubious emendations, of the Boduni must have taken place between the and Catuellani in Dio. Ix. 20, with coast and the Thames, tribes of similar name in Ptolemy, can- 176 BRITAIN. book vm. it reached as far as the Thames ; here a halt was made, chiefly perhaps to give the emperor the opportunity of plucking the easy laurels in person. So soon as he arrived, the river was crossed ; the British levy was beaten, on which occasion Togodumnus met his death ; Camalo- dunum itself was taken. His brother Caratacus, no doubt, obstinately continued the resistance, and gained for himself, in victory or defeat, a proud name with friend and foe ; nevertheless, the progress of the Romans was not to be checked. One prince after another was beaten and deposed — the triumphal arch of Claudius names eleven British kings as conquered by him ; and what did not succumb to the Roman arms yielded to the Roman largesses. Numerous men of rank accepted the possessions which the emperor conferred on them at the expense of their countrymen ; various kings also submitted to the modest position of vassals, as indeed Cogidumnus the king of the Regni (Chichester) and Prasutagus the king of the Iceni (Norfolk) bore rule for a series of years as depen- dent princes. But in most districts of the island, which had hitherto been monarchically governed throughout, the conquerors introduced their communal constitution, and gave what was still left to be administered into the hands of the local men of rank — a course which brought in its train wretched factions and internal quarrels. Even under the first governor the whole level country as far as the Humber seems to have come into Roman power ; the Iceni, for example, had already submitted to him. But it was not merely with the sword that the Romans made way for themselves ; veterans were brought to Camalo- dunum immediately after its capture, and the first town of Roman organisation and Roman burgess -rights, the " Claudian colony of victory," was founded in Britain, destined to be the capital of the country. Immediately afterwards began also the profitable working of the British mines,, particularly of the productive lead-mines ; there are British leaden bars from the sixth year after the invasion. Evidently with like rapidity the stream of Roman mer- chants and artisans poured itself over the field newly chap. v. BRITAIN. 177 opened up ; if Camalodunum received Roman colonists, Roman townships, which soon obtained formally urban organisation, were formed elsewhere in the south of the island as a mere result of freedom of traffic and of immi- gration, particularly at the hot springs of Sulis (Bath), in Verulamium (St. Albans to the north-west of London), and above all in the natural emporium of trading on a great scale — Londinium at the mouth of the Thames. The advance of the foreign rule asserted itself every- where, not merely in new taxes and levies, but perhaps still more in commerce and trade. When Plautius after four years of administration was recalled, he entered Rome in triumph, the last private who attained such honour, and honours and orders were lavished on the officers and soldiers of the victorious legions ; triumphal arches were erected to the emperor in Rome, and thereafter in other towns, on account of victory achieved " without any losses whatever;" the crown -prince born shortly before the invasion received, instead of his grandfather's name, that of Britannicus. We may discern in these matters the unmilitary age disused to victories with loss, and the extra- vagance in keeping with political dotage ; but, if the invasion of Britain has not much significance from a mili- tary standpoint, testimony must withal be borne to the leading men that they set about the work in an energetic and persistent fashion, and that the painful and dangerous time of transition from independent to foreign rule in Britain was an unusually short one. After the first rapid success, it is true, there were developed difficulties and even dangers, which the occupa- tion of the island brought not merely to the conquered but also to the conquerors. They were masters of the level country, but not of the Resistance mountains or of the sea. The west above all gave trouble B rit ^ 1 to the Romans. No doubt in the extreme south-west, in what is now Cornwall, the old nationality maintained itself, probably more because the conquerors concerned them- selves but little about this remote corner than because it directly rebelled against them. But the Silures in the VOL. 1. 12 1 78 BRITAIN. book vm. south of the modern Wales, and their northern neighbours the Ordovici, perseveringly defied the Roman arms ; the Mona. island Mona (Anglesey), adjacent to the latter, was the true focus of national and religious resistance. It was not the character of the ground alone that hindered the ad- vance of the Romans ; what Britain had been for Gaul, that the large island Ivernia was now for Britain, and especially for this west coast ; the freedom on the one side of the channel did not allow the foreign rule to take firm root in the other. We clearly recognise in the laying out of the legionary camps that the invasion was here arrested. Under the successor of Plautius the camp for the 14th legion was laid out at the confluence of the Tern with the Severn near Viroconium (Wroxeter, not far from Shrewsbury) j 1 presumably about the same time, to the south of it, that of Isca (Caerleon = Castra legionis) for the 2d ; to the north that of Deva (Chester = Castra) for the 20th ; these three camps shut off the region of Wales towards the south, north, and west, and pro- tected thus the pacified land against the mountains that remained free. Into this region the last prince of Camalodunum, Caratacus, threw himself, after his home had become Roman. He was defeated by the successor of Plautius, Publius Ostorius Scapula, in the territory oi the Ordovici, and soon afterwards delivered up by the terrified Brigantes, with whom he had taken refuge, to the Romans (5 1), and conducted with all his adherents to Italy. In surprise he asked, when he saw the proud city, how the masters of such palaces could covet the poor huts of his native country. But with this the west was by no means subdued ; the Silures above all persevered in obstinate 1 Tacitus, Ann. xii. 31 [P. Osto- the so-called "English Pompeii." rius) cuncta castris ad . . ntonam The epitaph of a soldier of the 20th (MSS. read castris antoam) et Sab- has also been found there. The camp rinam fiuvios cohibere farat. So the described by Tacitus was perhaps passage is to be restored, only that common at first to the two legions, the name of the river Tern not else- and the 20th did not go till after- where given in tradition cannot be wards to Deva. That the camp at supplied. The only inscriptions found Isca was laid out immediately after in England of soldiers of the 14th the invasion is plain from Tacitus, legion, which left England under Ann. xii. 32, 38. Nero, have come to light at Wroxeter, chap. v. BRITAIN. 179 resistance, and the fact that the Roman general announced his purpose of extirpating them to the last man did not contribute to make them more submissive. The enter- prising governor, Gaius Suetonius Paullinus, attempted Paullinus. some years later (61) to bring into Roman power the chief seat of resistance, the island of Mona, and in spite of the furious opposition with which he was met, and in which the priests and the women took the lead, the sacred trees, beneath which many a Roman captive had bled, fell under the axes of the legionaries. But out of the occupation of this last asylum of the Celtic priesthood there was de- veloped a dangerous crisis in the subject territory itself; and the governor was not destined to complete the con- quest of Mona. In Britain, too, the alien rule had to stand the test of Boudicca. national insurrection. What was undertaken by Mithra- dates in Asia Minor, by Vercingetorix among the Celts of the continent, by Civilis among the subject Germans, was attempted among the insular Celts by a woman, the wife of one of those vassal -princes confirmed by Rome, the Queen of the Iceni, Boudicca. Her deceased husband had, to secure the future of his wife and his daughters, bequeathed his sovereignty to the emperor Nero, and divided his property between the latter and his own rela- tives. The emperor took the legacy and, in addition, what was not to fall to him ; the princely cousins were put in chains, the widow was scourged, the daughters mal- treated in more shameful fashion. Then came other wrongs at the hands of the later Neronian government. The veterans settled in Camalodunum chased the earlier possessors from house and homestead as it pleased them, without the authorities interfering to check them. The presents conferred by the emperor Claudius were confis- cated as revocable gifts. Roman ministers, who at the same time trafficked in money, drove in this way the Britannic communities, one after the other, to bankruptcy. The moment was favourable. The governor Paullinus,. more brave than cautious, found himself, as we have said, with the flower of the Roman army in the remote island of 180 BRITAIN. book vm. Mona, and this attack on the most sacred seat of the national religion exasperated men's minds as much as it paved the way for insurrection. The old vehement Celtic faith, which had given the Romans so much trouble, burst forth once more, for the last time, in a mighty flame. The weakened and far separated camps of the legions in the west and in the north afforded no protection to the whole south-east of the island with its flourishing Roman towns. Attack on Above all, the capital, Camalodunum, was utterly dunum" defenceless ; there was no garrison. The walls were not completed, although the temple of their imperial founder, the new god Claudius, was so. The west of the island, probably kept down by the legions stationed there, seems not to have taken part in the rising, and as little the non- subject north ; but, as frequently occurred in Celtic revolts, in the year 61 on a concerted signal all the rest of the subject territory rose in a moment against the foreigners, the Trinovantes, driven out of their capital, taking the lead. The second commander, who at the time repre- sented the governor, the procurator Decianus Catus, had at the last moment sent what soldiers he had to its pro- tection ; they were 200 men. They defended themselves with the veterans and the other Romans capable of arms for two days in the temple ; then they were overpowered, and all that was Roman in the town perished. The like fate befell the chief emporium of Roman trade, Londinium, and a third flourishing Roman city, Verulamium (St. Albans, north-west of London), as well as the foreigners scattered over the island ; it was a national Vesper like 'that of Mithradates, and the number of victims — alleged to be 70,000 — was not less. The procurator gave up the cause of Rome as lost, and fled to the continent. The Roman army, too, became involved in the disaster. A number of scattered detachments and garrisons succumbed to the assaults of the insurgents. Quintus Petillius Cerialis, who held the command in the camp of Lindum, marched on Camalodunum with the 9th legion ; he came too late to save it, and, assailed by an enormous superiority of force, lost in the battle all his infantry ; the camp was chap. v. BRITAIN. 181 stormed by the Brigantes. The same fate well-nigh over- took the general-in-chief. Hastily returning from the island of Mona, he called to him the 2d legion stationed at Isca ; but it did not obey the command, and with only about 10,000 men Paullinus had to take up the unequal struggle against the numberless and victorious army of the insurgents. If ever soldiers made good the errors of their leader it was on the day when this small band — chiefly the thenceforth celebrated 14th legion — achieved, doubt- less to its own surprise, a full victory, and once more established the Roman rule in Britain. Little was want- ing to bring the name of Paullinus into association with that of Varus. But success decides, and here it remained with the Romans. 1 The guilty commandant of the legion that remained aloof anticipated the court-martial, and threw himself upon his sword. The queen Boudicca drank the cup of poison. The otherwise brave general was not indeed brought to trial, as seemed to be at first the inten- tion of the government, but was soon under a suitable pretext recalled. The subjugation of the western portions of the island Subjuga- was not continued at once by the successors of Paullinus. ''° n °1 . J West Bn- The able general Sextus Julius Frontinus first under tain. Vespasian forced the Silures to recognise the Roman 1 A worse narrative than that of peared there with a personal escort, Tacitus concerning this war, A nn. xiv. without the corps which he had with 31-39, is hardly to be found even in him in Mona — which indeed has no this most unmilitary of all authors, meaning. The bulk of the Roman We are not told where the troops troops, as well those brought back were stationed, and where the battles from Mona as those still in existence were fought ; but we get, instead, signs elsewhere, can, after the extirpation and wonders enough and empty of the 9th legion, only have been words only too many. The import- stationed on the line Deva — Viro- ant facts, which are mentioned in the conium — Isca ; Paullinus fought the life of Agricola, 31, are wanting in the battle with the two legions stationed main narrative, especially the storm- in the first two of these camps, the ing of the camp. That Paullinus 14th and the (incomplete) 20th. That coming from Mona should think not Paullinus fought because he was ob- of saving the Romans in the south- liged to fight, is stated by Dio, lxii. east, but of uniting his troops, is intel- 1-12, and although his narrative can- ligible ; but not why, if he wished to not be otherwise used to correct that sacrifice Londinium, he should march of Tacitus, this much seems required thither on that account. If he really by the very state of the case, went thither, he can only have ap- 1 82 BRITAIN. book viii. rule ; his successor Gnaeus Julius Agricola, after obstinate conflicts with the Ordovici, effected what Paullinus had not achieved, and occupied in the year 78 the island of Mona. Afterwards there is no mention of active resistance in these regions ; the camp of Viroconium could probably about this time be dispensed with, and the legion thereby set free could be employed in northern Britain. But the other two legionary camps still remained on the spot down to the time of Diocletian, and only disappeared in the later state of the occupying force. If political con- siderations may have contributed to this (p. 190), yet the resistance of the west was probably continued even later, perhaps supported by communications with Ivernia. Moreover, the complete absence of Roman traces in the interior of Wales, and the Celtic nationality maintaining itself there up to the present day, tell in favour of this view. Subjuga- In the north the camp of the 9th Spanish legion in Northern Lindum (Lincoln) formed the centre of the Roman position Britain. to the east of Viroconium. In closest contact with this camp in north England was the most powerful principality of the island, that of the Brigantes (Yorkshire) ; it had not properly submitted, but the queen, Cartimandus, sought to keep peace with the conquerors and showed herself compliant to them. The party hostile to the Romans had attempted to break loose here in the year 50, but the attempt had been quickly suppressed. Caratacus, beaten in the west, had hoped to be able to continue his resistance in the north, but the queen delivered him, as already stated, to the Romans. These internal dissensions and domestic quarrels must have partly interfered with the rising against Paullinus, in which we find the Brigantes in a leading position, and which fell with all its weight upon this very legion of the north. Meanwhile the Roman party of the Brigantes, however, was influential enough to obtain the restoration of the government of Cartimandus after the insurrection was defeated. But some years afterwards the patriotic party there, supported by the watchword of revolt from Rome, chap. v. BRITAIN. 183 which during the civil war after the downfall of Nero filled all the west, brought about a new rising of the Brigantes against the foreign rule, at the head of which stood Cartimandus's former husband, set aside and offended by her — the veteran warrior Venutius. It was only after prolonged conflicts that the mighty people was subdued by Petillius Cerialis, the same who had fought unsuccess- fully under Paullinus against these same Britons, now one of the most noted generals of Vespasian, and the first governor of the island nominated by him. The gradually slackening resistance of the west made it possible to combine one of the three legions hitherto stationed there with that stationed in Lindum, and to advance the camp itself from Lindum to the chief place of the Brigantes, Eburacum (York). But, so long as the west offered serious resistance, nothing further was done in the north for the extension of the Roman bounds ; at the Caledonian forest, says an author of the time of Vespasian, the Roman arms were arrested for thirty years. It was Agricola who first, after his work was over in Agricola. the west, energetically set himself to the subjugation also of the north. First of all, he created for himself a fleet, without which the provisioning of the troops in these mountains, which afforded few supplies, would have been impossible. Supported by this fleet he reached, under Titus (80), as far as the estuary of the Tava (Frith of Tay), into the region of Perth and Dundee, and employed the three following campaigns in gaining an exact knowledge of the wide districts between this frith and the previous Roman boundary on the two seas, in breaking everywhere the local resistance, and in constructing intrenchments at the fitting places ; with reference to which, in particular, the natural line of defence which is formed by the two friths running deeply into the land, of Clota (Clyde) near Glasgow, and Bodotria (Forth) near Edinburgh, was selected for the reserve. This advance called the whole Highlands under arms ; but the mighty battle which the united Caledonian tribes offered to the legions between the two friths of Forth and Tay at the Graupian 184 BRITAIN. BOOK VIII. Caledonia abandoned. Probable grounds for this policy. mountains ended with the victory of Agricola. Accord- ing to his view the subjugation of the island, once begun, had to be also completed, nay, even extended to Ivernia ; and in favour of that course there might be urged, with respect to Roman Britain, what the occupation of the island had brought about with respect to Gaul. More- over, with an energetic carrying out of the occupation of the islands as a whole, the expenditure of men and money for the future would probably be reduced. The Roman government did not follow these counsels. How far personal and spiteful motives may have co-operated in the recall of the victorious general in the year 85, who for that matter had remained longer in office than was usually the case elsewhere, must be left undetermined. The coincidence of the last victories of the general in Scotland and the first defeats of the emperor in the region of the Danube was certainly in a high degree annoying. But for the putting a stop to the operations in Britain, 1 and for the calling away, which apparently then ensued, of one of the four legions with which Agricola had executed his campaigns to Pannonia, a quite sufficient explanation is furnished by the military position of the state at that time — the extension of the Roman rule to the right bank of the Rhine in upper Germany and the outbreak of the dangerous wars in Pannonia. This, indeed, does not explain why, withal, an end should be put to the pressing forward towards the north, and northern Scotland as well as Ireland should be left to themselves. That thenceforth the government desisted not on account of accidents of the situation for the moment, but once for all, from pushing forward the frontier of the empire, and amidst all change of persons adhered to this course, we are taught by the whole later history of the island, and taught especially by the laborious and costly wall-structures to be mentioned immediately. Whether the completion of the conquest was renounced by them in the true interest of the state, is another question. That the imperial 1 Tacitus, Hist. i. 2, sums up the result in the words perdomita Britannia et statim missa. chap. v. BRITAIN. 185 finances would only suffer loss by this extension of the bounds was even now urged, quite as much as it formerly was against the occupation of the island itself; but could not be decisive of the matter. 1 In a military point of view the occupation was capable of being carried out, as Agricola had conceived it, beyond doubt without material difficulty. But the consideration might turn the scale, that the Romanising of the regions still free would have to en- counter great difficulty on account of the diversity of race. The Celts in England proper belonged throughout to those of the continent ; national name, faith, language, were com- mon to both. If the Celtic nationality of the continent had found a support in the island, on the other hand the Romanising of Gaul necessarily carried its influence over to England, and to this especially Rome owed the fact that Britain became Romanised with so surprising rapidity. But the natives of Ireland and Scotland belonged to another stock and spoke another language ; the Briton understood their Gaelic probably as little as the German understood the language of the Scandinavians. The Caledonians — with the Iverni the Romans hardly came into contact — are described throughout as barbarians of the wildest type. On the other hand, the priest of the oak (Derwydd, Druidd) exercised his office on the Rhone as in Anglesey, but not in the island of the west nor in the mountains of the north. If the Romans had waged the war chiefly to bring the domain of the Druids entirely into their power, this aim was in some measure attained. Beyond doubt at another time all these considerations would not have induced the Romans to renounce the sea-frontier on the north when brought so near to them, and at least Caledonia would have been occupied. But the Rome of that time was no longer able to leaven farther regions with Roman habits ; the productive power and the progressive spirit of the people had disappeared 1 The imperial finance-official under yip e$ BpirracWa undoubtedly took place soon after the HpiyivTW rty iroXX^v, &n iTreafialvtiv year 108 (C / L. vii. 241), and /col otroi aiv 87r\ois ijp^av is rty Yev- the substitution for it of the 6th ovvlav poipav (unknown ; perhaps, as Victrix. The two notices which point O. Hirschfeld suggests, the town of to this incident (Fronto, p. 217 the Brigantes, Vinonia) irriKSovs Naber : Hadriano imperium obtinente 'Pw/mJaw, it follows from this, not quantum militum a Britannis cat- that there were Brigantes also in sum ? Vita, 5, Britanni teneri sub Caledonia, but that the Brigantes in Romano, dicione non potcrant), as well the north of England at that time as the allusion in Juvenal, xiv. 196 : ravaged the settled land of the caste/la Brigantum, point to a revolt, Britons, and therefore a part of their not to an inroad. territory was confiscated. ciiAi'. v. />'A77'.7/,V. 189 of the Roman power was henceforth just as apparent here as on the Danube ami on the Kuphrates. In (he turbulent early years of Seyerus's reign the Caledonians hail broken their promise not to interfere with the Roman subjects, and, resting on their support, their southern neighbours, the Macules, had compelled the Roman governor Lupus to ransom captive Romans with large sums. For this the heavy arm of Sevcrus lighted on them not long before his death; he penetrated into their own territory and com- pelled them to cede considerable tracts, 1 from which indeed, after the old emperor had died in .mi at the camp of Kburacum, his sons at once of their own accord withdrew the garrisons, to be relieved o( their burdensome defence. From the third century hardly anything is told us of Cn\» the fate o( the island. Since none of the emperors down ^"i's'oois to Diocletian and his colleagues derived the name of con- queror from the island, there were probably no more serious conflicts in that quarter ; and, although in the region lying between the walls o( Pius and of Hadrian the Roman system doubtless never gained a linn footing, yet at least the wall of 1 ladrian seems to have rendered even then the service for which it was intended, and the foreign civilisation seems to have developed in security behind it. In the time oi~ Diocletian we find the district between the two walls evacuated, but the lladrianic wall occupied still as before, and the rest o( the Roman army in cantonments between it and the headquarters Kburacum, to ward otf the predator)- expeditions, thenceforth often mentioned, of the Caledonians, or — as they are now usually called— -the "tattooed" (jV.'.'A and the Scots streaming in from lvernia. The Romans possessed a standing licet in Britain ; kwi. but, as the marine always remained the weak side of Roman warlike organisation, the Hritish licet was tem- porarily of importance only under Agricola. If, as is probable, the government had reckoned on 1 That he hail the >losi£n of t>imj>- (.'..-.) or with the Wuiklin-; of the taji the whole notth uvulei (ho Roman wall, ami is ilonbtless as fabulous as |\i\\ev (I Mo. Kwi. 1.0 is not vow the Roman loss of 50,000 men without eom|\*tiWc cither with the cession the matter even coming to a battle. i go BRITAIN. BOOK VIII. Garrison and admi- nistration in the 2d and 3d centuries. Taxation and levy. being able to take back the greater part of the troops sent to the island, after it had been occupied, this hope was not fulfilled ; only one of the four legions sent thither was, as we have seen, recalled under Domitian ; the three others must have been indispensable, for no attempt was ever made to shift them. To these fall to be added the auxili- aries, who were called out apparently in larger proportion than the burgess-troops for the far from inviting service in the remote island of the North Sea. In the battle at the Graupian Mount in 84 there fought, besides the four legions, 8000 infantry and 3000 horsemen of the auxiliary soldiers. For the time of Trajan and Hadrian, when of these there were stationed in Britain six alae and twenty-one cohorts, together about 15,000 men, we shall have to estimate the whole British army at about 30,000 men. Britain was from the outset a field of command of the first rank, inferior to the two Rhenish commands and to the Syrian perhaps in rank, but not in importance, towards the end of the second century probably the most highly esteemed of all the governorships. It was owing only to the great distance that the British legions appear in the second rank amidst the corps- partisanships of the earlier imperial period ; in the corps-warfare after the extinction of the Antonine house they fought in the first rank. But it was one of the consequences of the victory of Severus that the governorship was divided. Thenceforth the two legions of Isca and Deva were placed under the legate of the upper province, the legion of Eburacum and the troops at the walls — consequently the main body of the auxiliaries — under the legate of the lower -province. 1 Probably the transference of the whole garrison to the north, which, as was above remarked, would doubtless have been appropriate on mere military grounds, was not carried out — partly because it would have put three legions into the hands of one governor. That financially the province cost more than it brought in (p. 172), can accordingly excite no surprise. For the military strength of the empire, on the other hand, Britain 1 The division results from Dio, Iv. 23. chap. v. BRITAIN. 191 was of considerable account ; the balance of proportion between taxation and levy must have had its application also to the island, and the British troops were reckoned alongside of the Illyrian as the flower of the army. At the very beginning seven cohorts were raised from the natives there, and these were constantly increased onward to the time of Hadrian ; after the latter had brought in the system of recruiting the troops as far as possible from their garrison-districts, Britain appears to have furnished the supply, at least in great part, for its strong garrison. There was an earnest and brave spirit in the people ; they bore willingly the taxes and the levy, but not the arro- gance and brutality of the officials. As a basis for the internal organisation of Britain, the Communal cantonal constitution existing there at the time of the °io^ amsa " conquest offered itself, which differed, as we have already remarked, from that of the Celts of the continent essen- tially only in the fact that the several tribes of the island, apparently all of them, were under princes (iv. 233). But iv. 222. this organisation seems not to have been retained, and the canton (civitas) to have become in Britain as in Spain a geographical conception ; at least we can hardly otherwise explain the facts that the Britannic tribes, taken in the strict sense, disappear as soon as they fall under Roman rule, and of the individual cantons after their subjugation there is virtually no mention at all. Probably the several principalities, as they were subdued and annexed, were broken up into smaller communities ; this was facilitated by the fact that there did not exist on the island, as there did on the continent, a cantonal constitution organised without a monarchic head. With this is doubtless con- nected the circumstance that, while the Gallic cantons possessed a common capital and in it a political and religious collective representation, nothing similar is stated as to Britain. The province was not without a concilium and a common cultus of the emperor ; but, if the altar of Claudius in Camalodunum 1 had been even approximately 1 To it doubtless the epigram of rens) : oceanusque luas ultra se respiat Seneca applies (vol. iv. p. 69, Bah- aras. The temple too, which accord- 192 BRITAIN. BOOK VIII. what that of Augustus was in Lugudunum, something would doubtless have been heard of it. The free and great political remodelling, which was given to the Gallic country by Caesar and confirmed by his son, no longer fits into the framework of the later imperial policy. We have already mentioned the founding, nearly con- temporary with the invasion of Britain, of the colony Camalodunum (p. 1 76), as it has also been already noticed that the Italian urban constitution was early introduced into a series of British townships. Herein, too, Britain was treated more after the model of Spain than after that of the Celtic continent. Prosperity. The internal condition of Britain must, in spite of the general faults of the imperial government, have been, at least in comparison with other regions, not unfavourable. If the people in the north knew only hunting and pastur- ing, and the inhabitants there as well as those adjoining them were always ready for feud and rapine, the south developed itself in an undisturbed state of peace, especially by means of agriculture, and along with it by cattle-rear- ing and the working of mines, to a moderate prosperity. The Gallic orators of Diocletian's time praise the wealth of the fertile island, and often enough the Rhine-legions received their corn from Britain. Roads. The network of roads in the island, which was uncom- monly developed, and for which in particular Hadrian did much in connection with the building of his wall, was of course primarily subservient to military ends ; but along- side of, and in fact taking precedence over the legionary camps Londinium occupies in that respect a place which brings clearly into view its leading position in traffic. Only in Wales were these imperial roads solely in the immediate neighbourhood of the Roman camps, from Isca to Nidum ing to the satire of the same Seneca (viii. 3), was erected to Claudius during his lifetime in Britain, and the temple certainly identical therewith of the god Claudius in Camalodunum (Tacitus, Ann. xiv. 31), is probably to be taken not as a sanctuary for the town itself, but after the analogy of the shrines of Augustus at Lugudunum and Tarraco. The delecti sacerdotes, who specie religionis omnes fortunas effun- debant, are the well-known provincial priests and purveyors of spectacles. chap. v. BRITAIN. 193 (Neath) and from Deva to the point of crossing to Mona. Roman Britain sustained a relation to Romanising Roman similar to that of northern and central Gaul. The national andcuiture. deities, the Mars Belatucadrus or Cocidius, the goddess Sulis treated as equivalent to Minerva, after whom the modern city of Bath was named, still received much worship on the island even in the Latin language. The language and manners that penetrated thither from Italy were yet more an exotic growth on the island than on the continent ; still towards the close of the first century the families of note there shunned as well the Latin language as the Latin dress. The great urban centres, the seats proper of the new culture, were more weakly developed in Britain ; we do not precisely know what English town served as seat for the concilium of the province and for the common worship of the emperor, or in which of the three legion-camps the governor of the province resided ; if, as it seems, the civil capital of Britain was Camalodunum, and the military capital Ebu- racum, 1 the latter can as little measure itself with Mentz as the former with Lyons. The rained sites even of places of note, of the Claudian veteran-town Camalodunum, and the populous mercantile town Londinium, and not less the camps of the legions for several hundred years, at Deva, Isca, Eburacum, present inscribed stones only in trifling number ; towns of name with Roman rights like the colony Glevum (Gloucester), and the municipmm Veru- lamium, have hitherto yielded not a single one ; the custom of setting up memorial-stones, on the results of which we are for such questions largely dependent, never rightly prevailed in Britain. In the interior of Wales and in other less accessible districts no Roman monuments at all have come to light. But there exist withal 1 The command stationed here (Vita Severi, 22). The praetorium, was, at least in later times, without situated probably on the coast below question the most important among the Eburacum (Itin. Ant. p. 466), may Britannic ; and there is also mention have been the summer seat of the here (for it is beyond doubt Ebura- governor, cum that is in view) of a Palatium VOL. I. I 3 194 BRITAIN. book vm. clear traces of the stirring commerce and traffic brought into prominence by Tacitus, such as the numerous drink- ing-cups which have come out of the ruins of London, and the London network of roads. If Agricola exerted him- self to transplant municipal emulation in the embellish- ment of one's native city by buildings and monuments to Britain, as it had been transferred from Italy to Africa and Spain, and to induce the islanders of note to adorn the markets of their home and to erect temples and palaces, as this was usual elsewhere, he was but in a slight degree successful as regards the public buildings. But it was otherwise as regards private economics ; the stately country-houses constructed and embellished in Roman fashion, of which now nothing is left but the mosaic pave- ments, are found in southern Britain — so far north as the region of York 1 — as frequently as in the land of the Rhine. The higher scholastic training of youth penetrated gradually from Gaul into Britain. It is specified among Agricola's administrative successes that the Roman tutor began to find his way into the leading houses of the island. In Hadrian's time Britain is described as a region conquered by the Gallic schoolmasters, and " even Thule speaks of hiring a professor for itself." These schoolmasters were in the first instance Latin, but Greeks also came ; Plutarch tells of a conversation which he held at Delphi with a Greek teacher of languages from Tarsus returning home from Britain. If in modern England, apart from Wales and Cumberland, the old native language has disappeared, it has given way not to the Angles or to the Saxons, but to the Roman idiom ; and, as usually happens in border-lands, in the later imperial period no one stood more faithfully by Rome than the man of Britain. It was not Britain that gave up Rome, but Rome that gave up Britain — the last that we learn of the island is the urgent entreaty of the population addressed to the emperor Honorius for protec- tion against the Saxons, and his answer, that they might help themselves as best they could. 1 None have been found to the (both somewhat north of York). See north of Aldborough and Easingwold Bruce, The Roman Wall, p. 6 1. CHAPTER VI. THE DANUBIAN LANDS AND THE WARS ON THE DANUBE. As the frontier on the Rhine was the work of Caesar, so Arrange- the frontier on the Danube was the work of Augustus. Augustus. When he came to the helm, the Romans were in the Italian peninsula hardly masters of the Alps, and in the Greek peninsula hardly masters of the Haemus (Balkan) and of the coast districts along the Adriatic and the Black Sea ; nowhere did their territory reach the mighty stream which separates southern from northern Europe. As well northern Italy as the Illyrian and Pontic commercial towns, and still more the civilised provinces of Macedonia and Thrace, were constantly exposed to the predatory ex- peditions of the rude and restless neighbouring tribes. When Augustus died there were substituted for the one province of Illyricum, which had barely attained to inde- pendent administration, five great Roman administrative districts, Raetia, Noricum, Lower Illyria or Pannonia, Upper Illyria or Dalmatia, and Moesia ; and the Danube became in its whole course, if not everywhere the military, at any rate the political, frontier of the empire. The com- paratively easy subjugation of these wide territories, as well as the grave insurrection of the years 6-g, and the abandon- ment, thereby occasioned, of the formerly cherished purpose of shifting the boundary-line from the upper Danube to Bohemia and to the Elbe, have been formerly described It remains that we should set forth the development of these provinces in the time after Augustus and the relations of the Romans to the tribes dwelling beyond the Danube. ation in Raetia. 196 THE DANUBIAN LANDS book vi 11. Latedviiis- The destinies of Raetia were so closely interwoven with those of the upper German province that we might refer for them to the earlier narrative. Roman civilisation here, taken as a whole, underwent but little development. The highlands of the Alps with the valleys of the upper Inn and the upper Rhine embraced a weak and peculiar population, probably the same as had once possessed the eastern half of the north -Italian plain, perhaps akin to the Etruscans. Driven back thence by the Celts, and perhaps also by the Illyrici, it held its ground in the northern mountains. While the valleys opening to the south, like that of the Adige, were attached to Italy, these offered to the southerns little room and still less incite- ment for settlement and founding of towns. Farther northward on the plateau between the lake of Constance and the Inn, which was occupied by the Celtic tribes of the Vindelici, there would doubtless have been room and place for Roman culture ; but apparently in this region, which could not become, like the Norican, an immediate continuation of Italy, and which, like the adjacent so- called Decumates-land, was probably in the first instance of value for the Romans merely as separating them from the Germans, the policy of the earlier imperial period had rather repressed culture. We have already indicated (p. 18) that immediately after the conquest there were thoughts of depopulating the district. Alongside of this lies the fact, that in the earlier imperial period no community with Roman organisation originated here. It is true that the founding of Augusta Vindelicorum, the modern Augsburg, was a necessary part of the laying out of the great road which was carried, simul- taneously with the conquest itself, by the elder Drusus through the high Alps to the Danube (pp. 19, 20) ; but this rapidly flourishing place was, and remained for above a century, a market-village, till at length Hadrian in this respect left the path prescribed by Augustus and made the land of the Vindelici share in the Romanising of the north. The bestowal of Roman urban rights on the chief place of the Vindelici by Hadrian may be connected chap. vi. AND THE WARS ON THE DANUBE. 197 with the fact that, nearly about the same time, the mili- tary frontier was pushed forward on the upper Rhine, and Roman towns arose in the former Decumates - land ; nevertheless in Raetia ever afterwards Augusta remained the only larger centre of Roman civilisation. The mili- tary arrangements exercised an influence in keeping it back. The province was from the first under imperial administration, and could not be left without a garrison ; but special considerations, as we have formerly shown, compelled the government to send to Raetia simply troops of the second class, and, though these were not inconsider- able in number, the smaller headquarters of alae and cohortes could not have exercised a civilising and town- forming effect like the camp of the legion. Under Marcus certainly, in consequence of the Marcomanian war, the Raetian headquarters, Castra Regina, the modern Ratisbon, was occupied by a legion ; but even this place appears to have remained in the Roman time a mere military settle- ment, and hardly to have stood on a line in urban develop- ment with the camps of second rank on the Rhine, such as e.g. Bonna. That the frontier of Raetia was already in Trajan's The Rae- time pushed forward from Ratisbon westward some dis- tian Limes - tance beyond the Danube, has already been observed (p. 158); and it has been there also shown that this territory was probably annexed to the empire without applying force of arms, similarly with the Decumates-land. It was likewise already mentioned that the fortifying of this territory was perhaps connected with the incursions of the Chatti extending thus far under Marcus, as also that these and subsequently the Alamanni in the third century visited as well this country in front as Raetia itself, and ultimately under Gallienus wrested it from the Romans. The neighbouring province of Noricum was doubtless The itaiis- in the provincial arrangement treated similarly to Raetia, Noricum. but in other respects had a different development. In no direction was Italy so open for land-traffic as towards the north-east ; the commercial relations of Aquileia, as well through Friuli with the upper Danube and with the iron- 198 THE DANUBIAN LANDS book viii. works of Noreia, as over the Julian Alps with the valley of the Save, here paved the way for the Augustan exten- sion of the frontier as nowhere else in the region of the Danube. Nauportus (Upper Laybach) beyond the pass was a Roman trading village already in the time of the republic ; Emona (Laybach), a Roman burgess - colony, afterwards formally incorporated with Italy, but substan- tially belonging to Italy from the time of its foundation by Augustus. Hence, as has already been noticed (p. 1 8), the mere proclamation was probably enough for the conversion of this " kingdom " into a Roman province. The population, originally doubtless Illyrian, afterwards in good part Celtic, shows no trace of that adherence to the national ways and language which we perceive among the Celts of the west. Roman language and Roman manners must have found early entrance here ; and by the emperor Claudius the whole territory, even the northern portion separated by the Tauern chain from the valley of the Drave, was organised in accordance with the Italian municipal constitution. While in the neighbouring lands of Raetia and Pannonia the monuments of Roman lan- guage are either wanting or appear withal only at the larger centres, the valleys of the Drave, the Mur, and the Salzach and their affluents are filled far up into the moun- tains with evidences of the Romanising which here took deep hold. Noricum adjoined, and was as it were a part of, Italy ; in the levy for the legions and for the guard, so long as the Italians were here at all preferred, this preference was extended to no other province so fully as to this. As respects military occupation what applies to Raetia applies also to Noricum. For the reasons already developed there was in Noricum, during the first two centuries of the empire, only a camp of alae and coJiortes. Carnuntum (Petronell, near Vienna), which in the Augustan age belonged to Noricum, was, when the Illyrian legions were sent thither, annexed for that very reason to Pan- nonia. The smaller Norican encampments on the Danube, and even the camp of Lauriacum (near Enns), instituted chap. vi. AND THE WARS ON THE DANUBE. 199 by Marcus for the legion sent by him to this province, were of no importance for the urban development. The large townships of Noricum, such as Celeia (Cilli), in the valley of the Sann, Aguontum (Lienz), Teurnia (not far from Spital), Virunum (Zollfeld, near Klagenfurt), in the north Juvavum (Salzburg), originated purely out of civil elements. Illyricum, that is the Roman territory between Italy The iiiy- and Macedonia, was in the republican time united, as to nan stock - its lesser portion, with the Graeco-Macedonian governor- ship, as to its greater, administered as a land adjacent to Italy, and, after the institution of the governorship of Cisalpine Gaul, as a portion of the latter. The territory coincides to a certain degree with the widely diffused stock from which the Romans named it ; it is the same whose scanty remnant still at the present day, at the southern end of its formerly far-extended possessions, has preserved its own nationality and its old language under the name of Skipetars, which they assign to themselves, or, as their neighbours call them, the Arnauts or Albanians. It is a member of the Indo-Germanic family, and within it doubtless most closely akin to the Greek branch, as is in keeping with its local relations ; but it stands by the side of the Greek at least as independent as the Latin and the Celtic. This nation in its original extent filled the coast of the Adriatic Sea from the mouth of the Po through Istria, Dalmatia, and Epirus, as far as Acarnania and Aetolia, and also in the interior upper Macedonia, as well as the modern Servia and Bosnia and the Hungarian territory on the right bank of the Danube ; it bordered thus on the east with the Thracian tribes, on the west with the Celtic, from which latter Tacitus expressly dis- tinguishes them. It is a vigorous type of a southern kind, with black hair and dark eyes, very different from the Celts, and still more from the Germans ; sober, tem- perate, intrepid, proud people, excellent soldiers, but little accessible to civic organisation, shepherds more than agriculturists. They did not attain any great political development. On the Italian coast they were confronted 200 THE DANUBIAN LANDS book viii. probably, in the first instance, by the Celts ; the probably Illyrian tribes there, especially the Veneti, became, through rivalry with the Celts, at an early date pliant subjects of the Romans, its reia- At the end of the sixth century of the city the Rome!" th founding of Aquileia and the subjugation of the peninsula ii. 196. of Istria (ii. 207 f.) farther narrowed their limits. Along the east coast of the Adriatic Sea the more important islands and the southern harbours of the mainland had long been occupied by the bold Hellenic mariners. When thereupon in Scodra (Scutari), to a certain extent in olden time as now the central point of the Illyrian land, the rulers began to develop a power of their own, and especially to make war upon the Greeks at sea, Rome, even before the Hannibalic war, struck them down with a strong hand, and took the whole coast under its protectorate ii. 74 f. (ii. 77 f.), which soon, after the ruler of Scodra had shared war and defeat with king Perseus of Macedonia, brought "■ 3°3- about the complete dissolution of this principality (ii. 321). At the end of the sixth century of the city, and in the first half of the seventh, after long years of conflict, the coast between Istria and Scodra was also occupied by iii. 172. the Romans (iii. 180 f.). In the interior the Illyrians were little touched by the Romans during the republican period ; but instead the Celts, advancing from the west, must have brought under their power a good portion of originally Illyrian territory, such as Noricum, afterwards preponderantly Celtic. The Latobici also in the modern Carniola were Celts ; and in the whole territory between the Save and Drave, just as in the Raab valley, the two great stocks were settled promiscuously, when Caesar Augustus subjected the southern districts of Pannonia to the Roman rule. Probably this strong admixture of Celtic elements contributed its part, along with the level character of the ground, to the early decline of the Illyrian nation in the Pannonian districts. Into the southern half, on the other hand, of the regions in- habited by Illyrians there penetrated of the Celts only the Scordisci, whose establishment on the lower Save chap. vi. AND THE WARS ON THE DANUBE. 201 as far as Morava, and raids as far as the vicinity of Thessalonica, have been formerly mentioned (iii. 1 84 f.). Hi 176 f. But the Greeks here gave place to them in some measure ; the sinking of the Macedonian power, and the desolation of Epirus and Aetolia, must have favoured the extension of the Illyrian neighbours. Bosnia, Servia, above all Albania, were in the imperial period Illyrian, and Albania is so still. It has already been mentioned that Illyricum was, The pro- according to the design of the dictator Caesar, to be ™. ce ° f n " constituted as a special governorship, and this design came into execution on the partition of the provinces between Augustus and the senate ; that this governor- ship, at first committed to the senate, passed to the emperor on account of the need for waging war there ; that Augustus divided this governorship and rendered effective the rule, which hitherto on the whole had been but nominal, over the interior both in Dalmatia and in the region of the Save ; and, lastly, that he subdued, after a severe struggle of four years, the mighty national insurrection which broke out among the Dalmatian as among the Pannonian Illyrians in the year 6. It remains that we relate the further fortunes, in the first instance, of the southern province. After the experience attained in the insurrection it Dalmatia seemed requisite not merely to employ the forces raised in 1^^ civi Illyricum abroad rather than as hitherto in their native lfcation. country, but also to keep in subordination the Dalmatians as well as the Pannonians by a command of the first rank. This rapidly fulfilled its object. The resistance, which the Illyrici under Augustus opposed to the unwonted foreign rule, expended its rage in the one violent storm ; after- wards our reports record no similar movement, even of but a partial kind. For the southern or, according to the Roman expression, the Upper Illyricum — the province Dalmatia, as it was usually called from the time of the Flavii — a new epoch began with the government of the emperors. The Greek merchants had indeed founded on the coast lying nearest to them the two great emporia of 202 THE DANUBIAN LANDS book viii. Apollonia (near Valona) and Dyrrachium (Durazzo) ; for that very reason this portion had already under the republic been consigned to Greek administration. But farther northward the Hellenes had settled only on the adjacent islands Issa (Lissa), Pharos (Lesina), Black-Cor- cyra (Curzola), and thence maintained intercourse with the natives particularly along the coast of Narona and in the townships adjacent to Salonae. Under the Roman republic the Italian traders, who here entered upon the heritage of the Greek, had settled in the chief ports Epi- taurum (Ragusa Vecchia), Narona, Salonae, Iader (Zara), in such numbers that they could play a not unimportant part in the war of Caesar and Pompeius. But it was only through Augustus that these townships received strength- ening by the settlement of veterans there, and — what was the main thing — urban rights ; and at the same time partly the energetic suppression of the piratic retreats still exist- ing in the islands, partly the subjugation of the interior and the pushing forward of the Roman frontier towards the Danube, tended to benefit especially these Italians settled on the east coast of the Adriatic Sea. Above all the capital of the country, the seat of the governor and of Salonae. the whole administration, Salonae rapidly flourished and far outstripped the older Greek settlements Apollonia and Dyrrachium, although to the latter town there were sent likewise under Augustus Italian colonists, not indeed veterans but dispossessed Italians, and the town was erected as a Roman burgess-community. It may be con- jectured that in the prosperity of Dalmatia and the arrested development of the Illyro-Macedonian coast the distinction between the imperial and the senatorial government played an essential part — as regards better administration, as well as a privileged position with the real holder of power. With this, moreover, may be connected the fact, that the Illyrian nationality held its ground better in the sphere of the Macedonian governorship than in that of the Dalmatian ; in the former it still lives at the present day ; and in the imperial period — apart from the Greek Apol- lonia and the Italian colony of Dyrrachium — while the two chap. vi. AND THE WARS ON THE DANUBE. 203 languages of the empire were made use of in the interior, that of the people must have continued to be the Illyrian. In Dalmatia, on the other hand, the coasts and the islands, so far as they were at all adapted thereto — the inhospi- table stretch to the north of the Iader necessarily was left behind in the development — were communalised after the Italian organisation, and soon the whole coast spoke Latin, somewhat as it speaks at the present day Venetian. The advance of civilisation into the interior had to Civilisation encounter local difficulties. The considerable streams of l ^ lioT e Dalmatia form waterfalls more than watercourses ; and even the establishment of land-routes meets unusual diffi- culties from the nature of its mountain -network. The Roman government made earnest exertions to open up the country. Under the protection of the legionary camp of Burnum in the valley of the Kerka and in that of Cettina under the protection of the camp of Delminium — which camps must have been here too the channels of civilisation and of Latinising — the cultivation of the soil developed itself after the Italian fashion, as also the planting of the vine and the olive, and in general Italian organisation and habits. On the other hand, beyond the watershed be- tween the Adriatic Sea and the Danube the valleys less favourable for agriculture from the Kulpa to the Drin remained during the Roman period in a primitive state, similar to that exhibited by Bosnia at the present day. The emperor Tiberius certainly had various roads made by the soldiers of the Dalmatian camps from Salonae into the valleys of Bosnia ; but the later governments ap- parently allowed the difficult task to drop. On the coast and in the districts adjoining the coast Dalmatia soon needed no further military protection ; Vespasian could already withdraw the legions from the valleys of the Kerka and the Cettina and employ them elsewhere. Amidst the decay of the empire in the third century Prosperity Dalmatia suffered comparatively little ; indeed, Salonae "letian. probably only reached at that time its greatest prosperity. This, it is true, was occasioned partly by the fact that the regenerator of the Roman state, the emperor Diocletian, 204 THE DANUBIAN LANDS book viii. was by birth a Dalmatian, and allowed his efforts aimed at the decapitalising of Rome to redound chiefly to the benefit of the capital of his native land ; he built alongside of it the huge palace, from which the modern capital of the province takes the name Spalato, within which it has for the most part found a place, and the temples of which now serve it as cathedral and as baptistery. 1 Diocletian, however, did not make Salonae a great city for the first time, but, because it was such, chose it for his private residence ; commerce, navigation, and trade must at that time in these waters have been concentrated chiefly at Aquileia and at Salonae, and the city must have been one of the most populous and opulent towns of the west. The rich iron mines of Bosnia were largely worked at least in the later imperial period ; the forests of the pro- vince likewise yielded massive and excellent timber ; even of the flourishing textile industry of the land a reminis- cence is still preserved in the priestly " Dalmatica." Alto- gether the civilising and Romanising of Dalmatia form one of the most peculiar and most significant phenomena of the imperial period. The boundary between Dalmatia and Macedonia was at the same time the political and linguistic demarcation of the West and East. As the spheres of rule of Caesar and Marcus Antonius came into contact at Scodra, so did those of Rome and Byzantium after the partition of the empire in the fourth century. Here the Latin province of Dalmatia bordered with the Greek province of Macedonia ; and the younger sister stands here alongside of the elder vigorous in aspiration and excelling in energy of effort. Pannonia While the southern Illyrian province and its peaceful Trajan" government soon ceased to be prominent in a historical aspect, northern Illyricum, or as it is usually called, Pan- nonia, forms in the imperial period one of the great mili- tary and thereby also political centres. In the army of the Danube the Pannonian camps have the leading posi- tion like the Rhenish in the west, and the Dalmatian and the Moesian attach themselves to them, and subordinate 1 The baptistery is perhaps the tomb of the emperor. chap. vi. AND THE WARS ON THE DANUBE. 205 themselves under them, in like manner as the legions of Spain and Britain were subordinate to those of the Rhine. Roman civilisation stands and continues here under the • influence of the camps, which did not remain in Pannonia as in Dalmatia only for some generations, but were per- manent After the subduing of the insurrection of Bato, the regular garrison of the province amounted at first to three, afterwards apparently only to two, legions ; and the further development was conditioned by their standing quarters and the shifting of these forward. When Augustus after the first war against the Dalmatians had selected Siscia, at the point where the Kulpa falls into the Save, as his chief stronghold, after Tiberius had subdued Pan- nonia at least as far as the Drave, the camps were pushed forward to the latter, and at least one of the Pannonian headquarters was thenceforth found at Poetovio (Pettau), on the borders of Noricum. The reason why the Pan- nonian army remained wholly or in part in the valley of the Drave can only have been the same as led to the con- struction of the Dalmatian legionary camps ; they needed troops here to keep in obedience their subjects as well in the neighbouring Noricum as above all in the region of the Drave itself. On the Danube watch was kept by the Roman fleet, which is already mentioned in the year 50, and presumably originated on the erection of the province. There was not yet perhaps a legionary camp on the river itself under the JuliorClaudian dynasty, 1 in connection 1 That there were no legions sta- small number also of inscriptions of tioned on the Danube itself in the Italici found in the camps of the year 50, follows from Tacitus, Ann. Danube (Eph. Ep. v. p. 225) points xii. 29 ; otherwise it would not have to their later origin. Certainly there been necessary to send a legion thither have been found in Camuntum some to receive the accession of the Suebi. epitaphs of soldiers of the 1 5th legion The laying out also of the Claudian which, from their outward form and Savaria suits better, if the town was from the absence of cognomen, ap- then Norican, than if it already be- pear to be older (Hirschfeld, Arch. longed to Pannonia; and, as the Epigraph. Mittheilungen, v. 217). assignment of this town to Pannonia Such determinations of date cannot coincides certainly as to time with the claim full certainty, where a decade is like severance of Carnuntum and with concerned ; nevertheless it must be the transference of the legion thither, conceded that the former arguments all this may probably have taken place also furnish no full proofs, and the only in the period after Claudius. The translocation may have begun earlier, 2o6 THE DANUBIAN LANDS book viii. with which we may note that the state of the Suebi im- mediately adjoining the province in front was at that time immediately dependent on Rome, and sufficed in some measure to protect the frontier. Then, as with the camps of Dalmatia, Vespasian apparently did away also with the camps on the Drave and transferred them to the Danube itself ; thenceforth the great headquarters of the Pannonian army were the formerly Norican (p. 198) Carnuntum (Petronell, to the east of Vienna), and along with it Vin- dobona (Vienna). Urban de- Civil development, such as we meet in Noricum and veiopment. on fa s coas t f Dalmatia, shows itself likewise in Pannonia only at some districts situated on the Norican frontier, and in part belonging originally to Noricum ; Emona and the upper valley of the Save stand on an equality with Nori- cum, and if Savaria (Stein, on the Anger) received the Italian municipal constitution at the same time with the Norican towns, that place must doubtless, so long as Car- nuntum was a Norican town, have belonged also to Noricum. It was only after the troops were stationed on the Danube that the government set to work to give urban organisation to the country behind. In the western territory originally Norican, Scarbantia (Oedenburg, on the Neusiedler See) obtained urban rights under the Flavii, while Vindobona and Carnuntum became of themselves camp -towns. Between the Save and Drave Siscia and Sirmium received urban rights under the Flavii, as on the Drave Poetovio (Pettau) under Trajan, Mursa (Eszeg) under Hadrian colonial rights — to mention here only the chief places. That the population, pre- dominantly Illyrian but in good part also Celtic, opposed no energetic resistance to the Romanising, has already been mentioned ; the old language and the old habits disappeared where the Romans came, and kept their ground only in the more remote districts. The districts — wide, but far from inviting for settlement — to the east of possibly under Nero. For the con- the inscription, attesting such a struc- struction or extension of this camp by ture, of Carnuntum, dating from the Vespasian we have the evidence of year 73 (Hirschfeld, /. c). cian stock. chap. vi. AND THE WARS ON THE DANUBE. 207 the river Raab and to the north of the Drave as far as the Danube were probably reckoned even from the time of Augustus as belonging to the empire, but perhaps in a way not much differing from Germany before the battle of Varus ; urban development neither then nor later found a true soil here, and in a military point of view this region was for a long time occupied but little or not at all. This state of matters changed in some measure only in conse- quence of the incorporation of Dacia under Trajan ; the pushing forward of the Pannonian camps towards the east frontier of the province, to which that step gave occasion, and the further internal development of Pannonia, will be better described in connection with the wars of Trajan. The last portion of the right bank of the Danube — the The Thra- mountain-land on the two sides of the Margus (Morava), and the flat country stretching along between the Haemus and the Danube — was inhabited by Thracian tribes ; and it appears necessary in the first instance to cast a glance at this great stock as such. It runs parallel in a certain sense to the Illyrian. As the Illyrians once filled the regions from the Adriatic Sea to the middle Danube, so the Thracians were formerly settled to the east of them, from the Aegean Sea as far as the mouths of the Danube, and not less on the one hand upon the left bank of the Danube, particularly in the modern Transylvania, on the other hand beyond the Bosporus, at least in Bithynia and as far as Phrygia. Herodotus is not wrong in calling the Thracians the greatest of the peoples known to him after the Indians. Like the Illyrian, the Thracian stock attained to no full development, and appears more as hard- pressed and dispossessed than as having any historically memorable course of its own. But, while the language and habits of the Illyrians have been preserved — though in a form worn down in the course of centuries — to the present day, and we with some right transfer the image of the Palikars from more recent history to that of the Roman imperial period, the same does not hold good of the Thracian stock. There is manifold and sure attestation that the tribes of the territory, which in conse- 208 THE DANUBIAN LANDS book viii. quence of the Roman provincial division has ultimately retained the name Thracian, as well as the Moesians be- tween the Balkan and the Danube, and not less the Getae or Daci on the other bank of the Danube, all spoke one and the same language. This language had in the Roman empire a position similar to that of the Celts and of the Syrians. The historian and geographer of the Augustan age, Strabo, mentions the likeness of language among the peoples named ; in botanical writings of the imperial period the Dacian appellations of a number of plants are specified. 1 When his contemporary, the poet Ovid, had opportunity given to him in the far-off Dobrudscha to reflect on his too dissolute course of life, he used his leisure to learn Getic, and became almost a poet of the Getae : — Ahpudet! et Getico seripsi sermone libellum .... Et filacui (gratare mihi) coepique poetae Inter inhwnanos nomen habere Getas. But while the Irish bards, the Syrian missionaries, and the mountain valleys of Albania secured a certain con- tinued duration for other idioms of the imperial period, the Thracian disappeared amidst the fluctuations of peoples in the region of the Danube and the overpowerful influence of Constantinople, and we cannot even determine the place which belongs to it in the pedigree of nations. The descriptions of manners and customs of particular tribes belonging to it, as to which various notices have been pre- served, yield no individual traits valid for the race as a whole, and for the most part bring into relief merely singu- larities such as appear among all peoples at a low stage of culture. But they were and remained a soldier-people, not less useful as horsemen than for light infantry, from the times of the Peloponnesian war and of Alexander down to that of the Roman Caesars, whether they might 1 We know whole sets of Thracian, first two also frequently occur isolated Getic, Dacian names of places and in their other half (Bithus, Zipa). A persons. Remarkable in a linguistic similar group is formed by the com- point of view is a group of personal pounds with — ■ poris, such as Mucaporis names compounded with — centhus : (as Thracian, Bull. I. c, as Dacian in Bithicenthus, Zipacenthus, Disacen- numerous cases), Cetriporis, Rhasky- thus, Tracicenthus, Linicenthus {Bull, ports, Bithoporis, Dirdiporis. de Corr. Hell. vi. 179), of which the chap. vi. AND THE WARS ON THE DANUBE. 209 range themselves against them or subsequently fight for them. Their wild but grand mode of worshipping the gods may perhaps be conceived as a trait peculiar to this stock — the mighty outburst of the joy of spring and youth, the nocturnal mountain-festivals of torch-swinging maidens, the intoxicating sense -confusing music, the flowing of wine and the flowing of blood, the giddy festal whirl frantic with the simultaneous excitement of all sensuous passions. Dionysos, the glorious and the ter- rible, was a Thracian god ; and whatever of the kind was specially prominent in the Hellenic and the Roman cultus, was connected with Thracian or Phrygian customs. While the Illyrian tribes in Dalmatia and Pannonia, The Thra- after the overthrow of the great insurrection in the last p a te. p ™ ' years of Augustus, did not again invoke the decision of arms against the Romans, the same did not hold true of the Thracian stock ; the often-shown spirit of independ- ence and the wild bravery of this nation did not fail it even in its decline. In Thrace, south of the Haemus, the old principate remained under Roman supremacy. The native ruling house of the Odrysae, with their residence Bizye (Wiza), between Adrianople and the coast of the Black Sea, was already in the earlier period the most pro- minent among the princely families of Thrace ; after the triumviral period there is no further mention of other Thracian kings than of those of this house, so that the other princes appear to have been made vassals or superseded under Augustus, and only members of this family were thenceforth invested with the Thracian kingly office. This was done, probably, because during the first century, as will be shown further on, there were no Roman legions stationed on the lower Danube ; Augustus expected the frontier at the mouth of the Danube to be protected by the Thracian vassals. Rhoemetalces, who in the second half of the reign of Augustus ruled all Thrace as a Roman vassal-king, 1 and his children and grandchildren therefore 1 Tacitus, Ann. ii. 64, says this Thracian mountains, and especially expressly. Of free Thracians, viewed the Rhodope of the Bessi, maintained from the Roman stand - point, there even in the state of peace an attitude were at that time none ; but the as regards the princes installed by VOL. I. 14 210 THE DANUBIAN LANDS book vm. played in this country nearly the same part as Herod and his descendants in Palestine ; unconditional devotedness towards the lord -paramount, a decided inclination to Roman habits, hostility to their own countrymen who clung to the national independence, mark the attitude of the Thracian ruling house. The great Thracian insur- rection of the years 741-743, of which we have formerly spoken (p. 24), was directed in the first instance against this Rhoemetalces and his brother and co-regent Cotys who perished in it, and, as he at that time was indebted to the Romans for reinstatement into his dominion, so he some years afterwards rendered to them his thanks when, on occasion of the rising of the Dalmatians and the Pannonians, to which his Dacian kinsmen adhered, he kept faithfully to the Romans, and bore an essential part in its overthrow. His son Cotys was more Roman, or rather Greek, than Thracian ; he traced back his pedigree to Eumolpus and Erichthonius, and gained the hand of a kinswoman of the imperial house, the great granddaughter of the triumvir Antonius ; and not merely did the Greek and Latin poets of his time address him in song, but he himself was all but accounted a Getic poet. 1 The last of the Thracian kings, Rhoemetalces, son of the early de- ceased Cotys, was reared in Rome, and, like the Herodian Agrippa, a youthful playmate of the emperor Gaius. Province of But the Thracian nation by no means shared the Roman leanings of the ruling house, and the government gradually became convinced in Thrace as in Palestine that the tottering vassal-throne, only maintained by constant inter- ference of the protecting power, was of use neither for them nor for the country, and that the introduction of direct administration was in every respect to be preferred. The emperor Tiberius made use of the quarrels that arose in the Thracian royal house to send to Thrace in the Rome, that could hardly be designated Thessalonica {AntJiol. Planud. iv. 75), as subjection ; they acknowledged the the same poet who celebrated also the king doubtless, but obeyed him, as conqueror of the Thracians, Piso (p. Tacitus says {I.e. and iv. 46, 51), 24), and a Latin epistle inverse ad- only when it suited them. dressed to Cotys by Ovid (ex Ponto, 1 We have still a Greek epigram, ii. 9), dedicated to Cotys by Antipater of Thrace. chap. vi. AND THE WARS ON THE DANUBE. 211 year 19a Roman governor, Titus Trebellenus Rufus, under cover of exercising guardianship over the princes that were minors. Yet this occupation was not accomplished with- out resistance, ineffectual doubtless, but serious on the part of the people, who, particularly in the mountain- valleys, troubled themselves little about the rulers ap- pointed by Rome, and whose forces, led by their family- chiefs, hardly felt themselves to be soldiers of the king, and still less soldiers of Rome. The sending of Trebellenus called forth in the year 21a rising, in which not merely did the most noted Thracian tribes take part, but which threatened to assume greater proportions ; messengers of the insurgents went over the Haemus to enkindle the national war in Moesia, and perhaps still further. Mean- while the Moesian legions appeared in right time to relieve Philippopolis, which the insurgents besieged, and to suppress the movement. But, when some years later (25) the Roman government ordered levies in Thrace, the men refused to serve beyond the bounds of their own country. When no regard was paid to this refusal, the whole mountains rose and a struggle of despair ensued, in which the insurgents, constrained at length by hunger and thirst, threw themselves in great part on the swords of the enemy or on their own, and preferred to renounce life rather than their time-honoured freedom. The direct government continued in the form of exercising wardship in Thrace up to the death of Tiberius ; and, if the emperor Gaius at the commencement of his reign gave back the rule to the Thracian friend of his youth just as to the Jewish, a few years after, in the year 46, the government of Claudius definitely put an end to it. This final annexation of the kingdom, and conversion of it into a Roman province, also encountered an equally hopeless and equally obstinate resistance. But with the introduction of direct adminis- tration the resistance was broken. The governor, at first of equestrian, and from Trajan's time of senatorial, rank, never had a legion ; the garrison sent into the country, though it was not stronger than 2000 men, along with a small squadron stationed at Perinthus, was sufficient, in 212 THE DANUBIAN LANDS BOOK VIII. Moesia. Hellenism and Ro- manism in Thrace. connection with the precautionary measures otherwise taken by the government, to keep down the Thracians. The laying out of military roads was begun immediately after the annexation ; we find that the buildings requisite in the state of the country for the accommodation of travellers at the posting stations were already, in the year 6 1, erected by the government and opened to traffic. Thrace was thenceforth an obedient and important pro- vince of the empire ; hardly any other furnished so numerous men for all parts of the war-forces, especially for the cavalry and the fleet, as this old home of gladiators and of mercenary soldiers. The serious conflicts which the Romans had to sustain with the same nation on the so-called " Thracian shore " [Ripa Thraciae], in the region between the Balkan and the Danube, and which led to the institution of the Moesian command, form an essential constituent part of the regula- tion of the northern frontier in the Augustan age, and have been already described in their connection (p. 1 3 f.) Of resistance similar to that offered by the Thracians to the Romans nothing is reported from Moesia ; the tone of feeling there may not have been different, but in the level country and under the pressure of the legions encamped at Viminacium the resistance did not emerge openly. Civilisation came to the Thracian tribes, as to the Illyrian, from two sides ; that of the Hellenes from the coast and from the Macedonian frontier, the Latin from the Dalmatian and Pannonian frontier. Of the former it will be more appropriate to treat when we attempt to describe the position of the European Greeks under the imperial rule ; here it suffices generally to bring out the fact that not merely did that rule protect the Greek element, where it found it, and the whole coast, even that subject to the governor of Moesia, always remained Greek; but that the province of Thrace, whose civilisation was begun in earnest only by Trajan, and was throughout a work of the imperial period, was not guided into a Roman path, but became Hellenised. Even the northern slopes of the Haemus, although administratively belonging to chap. vi. AND THE WARS ON THE DANUBE. 213 Moesia, were comprehended in this Hellenising ; Nicopolis on the Jantra and Marcianopolis, not far from Varna, both foundations of Trajan, were organised after a Greek model. Of the Latin civilisation of Moesia the same holds And in true as of that of the adjoining Dalmatian and Pan- oesia ' nonian interior ; only, as was natural, it emerges so much the later, weaker, and more impure, the farther remote it is from its starting-point. It followed predo- minantly here the encampments of the legions, and with these advanced eastward, starting from the probably oldest camps of Moesia at Singidunum (Belgrade) and Viminacium (Kostolatz). 1 It is true that, in keeping with the character of its armed apostles, it kept at a very low stage in upper Moesia, and left room enough for the play of the primitive conditions. Viminacium obtained Italian urban rights from Hadrian. Lower Moesia, between the Balkan and the Danube, in the earlier imperial period, remained probably throughout in the condition which the Romans found subsisting there ; not till the legion-camps on the lower Danube were founded at Novae, Durostorum, and Troesmis, which, as will be set forth further on (p. 227), probably did not take place till the beginning of the second century, did this part of the right bank of the Danube become a seat of so much Italian civilisation as was compatible with camp -arrangements. Thenceforth civil settlements arose here too — particularly on the Danube 1 It is one of the most seriously be older than Hadrian's time ; the felt blanks of the Roman imperial remains of the upper - Moesian are history that the standing quarters of hitherto so scanty that they at least the two legions, which formed under do not hinder our carrying back their the Julio-Claudian emperors the gar- origin a century further. When the rison of Moesia, the 4th Scythica and king of Thrace in the year 18 takes the 5th Macedonica (at least these arms against the Bastarnae and were stationed there in the year 33 ; Scythians (Tacitus, Ann. ii. 65), this C. I. L. iii. 1698) cannot hitherto be could not have been put forward even pointed out with certainty. Probably as a pretext, had lower - Moesian le- they were Viminacium and Singi- gionary camps been alreadyat that time dunum in what was afterwards upper in existence. This very narrative shows Moesia. Among the legion-camps of that the warlike power of this vassal- lower Moesia, of which that of prince was not inconsiderable, and Troesmis in particular has numerous that the setting aside of an uncompliant monuments to show, none appear to king of Thrace demanded caution. 214 THE DANUBIAN LANDS BOOK VIII. Hermun- duri. Mar- comani. itself, between the great standing camps, the towns con- stituted after the Italian model, Ratiaria, not far from Widin, and Oescus at the confluence of the Iskra with the Danube — and gradually the region approached the level of the Roman culture then subsisting, though of itself on its decline. In the construction of highways in lower Moesia the rulers displayed manifold activity after the time of Hadrian, from whence the oldest milestones hitherto found there proceed. If we turn from the survey of the Roman rule, as it took shape from Augustus onward in the lands on the right bank of the Danube, to the relations, and the inhabit- ants of the left, what we should have to remark as to the most westerly region has already in the main been said in the description of upper Germany ; and in particular it has been noticed (p. 158) that the Germans next adjoining Raetia, the Hermunduri, were of all the neighbours of the Romans the most peaceful, and, so far as is known to us, never fell into conflict with them. We have already stated that the people of the Mar- comani, or, as the Romans usually term them in earlier times, the Suebi, after it had in the Augustan age found new settlements in the old land of the Boii, the modern Bohemia, and had acquired through king Maroboduus a more fixed political organisation, remained indeed an on- looker during the Romano -German wars, but was pre- served through the intervention of the Rhenish Germans from the threatened Roman invasion ; and, not less, that the reaction of the renewed abandonment of the Roman offensive on the Rhine overthrew this too neutral state. The position of paramount power, which the Marcomani under Maroboduus had gained over the more remote peoples in the region of the Elbe, was thereby lost ; and the king himself died as an exile on Roman soil (p. 61). The Marcomani and their eastern neighbours of kindred stock, the Quadi in Moravia, fell under Roman clientship, in so far as in their case, nearly as in that of Armenia, the pretenders contending for the mastery leaned in part for support on the Romans, and these claimed, chap. vi. AND THE WARS ON THE DANUBE. 215 and according to circumstances also exercised, the right of investiture. The prince of the Cotones, Catu- alda, who had in the first instance overthrown Maro- boduus, could not maintain himself long as his successor, especially as Vibilius king of the neighbouring Hermun- duri took part against him ; he too had to pass over into Roman territory, and like Maroboduus to invoke the imperial favour. Tiberius then induced a Quadian of rank vannius. Vannius to take his place ; for the numerous train of the two banished kings, which was not allowed to remain on the right bank of the Danube, Tiberius procured settle- ments on the left in the March valley, 1 and procured for Vannius recognition on the part of the Hermunduri friendly with Rome. After a thirty years' rule the latter was over- thrown in the year 50 by his two nephews Vangio and Sido, who revolted against him, and gained for themselves the neighbouring peoples, the Hermunduri in Franconia, the Lugii in Silesia. The Roman government, which Vannius solicited for support, remained true to the policy 1 That the regnum Vannianum bordered, the Jazyges on the south, (Plin. H. N. iv. 12, 81), the Suebian the Bastarnae on the north, with state (Tacitus, Ann. xii. 29 ; Hist, the Quadi of the March valley. Ac- iii. S,2i), must be referred, not merely, cordingly the Marus is the March, as might appear from Tacitus, Ann. and the demarcation is formed by the ii. 63, to the dwellings of the people small Carpathians that stretch be- that went over with Maroboduus and tween the March and the Waag. If Catualda, but to the whole territory thus those retainers were settled inter of the Marcomani and Quadi, is shown flumen Marum et Cusum, then the clearly by the second report, Ann. Cusus not elsewhere mentioned is, xii. 29, 30, since here, as opponents provided the statement is correct, not of Vannius alongside of his own insur- the Waag, or even, as MiillenhofF gent subjects, there appear the peoples supposed, the Eipel falling into the bordering on Bohemia to the west and Danube below Gran, but an affluent north, the Hermunduri and Lugii. of the Danube westward of the March, As boundary towards the east Pliny perhaps the Gusen near Linz. The I.e. designates the region of Camun- narrative in Tacitus xii. 29, 30, also turn (Germanorum ibi confinium) requires the territory of Vannius to more exactly the river Marus or have reached to the west even beyond Duria, which separates the Suebi the March. The subscription to the and the regnum Vannianum from first book of the Meditations of the their eastern neighbours, whether we emperor Marcus iv KovdSois irpis t may refer the dirimens eos with Miil- Tpavoiq,, proves doubtless that then lenhoff (Sitzungsberichie der Berliner the state of the Quadi stretched as Akademie 1 883, p. 871) to the Jazy- far as the river Gran ; but this state ges, or, as is more natural, to the is not coincident with the regnum Bastarnae. In reality both doubtless Vannianum. 2i6 THE DANUBIAN LANDS book vm. of Tiberius ; it granted to the overthrown king the right of asylum, but did not interfere, especially as the successors, who shared the territory between them, readily acknow- ledged the Roman supremacy. The new prince of the Suebi, Sido, and his co-ruler Italicus, perhaps the successor of Vangio, fought in the battle, which decided between Vitellius and Vespasian, with the Roman army of the Danube on the side of the Flavians. In the great crises of the Roman rule on the Danube under Domitian and Marcus we shall again meet their successors. The Suebi of the Danube did not belong to the Roman empire ; coins probably struck by them show doubtless Latin in- scriptions, but not the Roman standard, to say nothing of the image of the emperor ; taxes proper and levies for Rome did not here take place. But, in the first century particularly, the Suebian state in Bohemia and Moravia was included within the sphere of Roman power ; and, as was already observed, this was not without its influence on the stationing of the Roman frontier-guard. Jazyges. In the plain between the Danube and Theiss eastward from the Roman Pannonia, and between this and the Thracian Daci, there was inserted a section of the people — probably belonging to the Medo-Persian stock — the Sar- matae, who living nomadically as a nation of shepherds and horsemen filled in great part the wide east-European plain ; these were the Jazyges, named the " emigrants " (jieTavdarat) in distinction from the chief stock which remained behind on the Black Sea. The designation shows that they only advanced at a comparatively late period into these regions ; perhaps their immigration falls to be included among the assaults, under which about the time of the battle of Actium the Dacian kingdom of Bure- bista broke down (p. 1 1). They meet us here at first under the emperor Claudius ; the Jazyges supplied the Suebian king Vannius with the cavalry for his wars. The Roman government was on its guard against the alert and pre- datory bands of horsemen, but did not otherwise sustain hostile relations to them. When the legions of the Danube marched to Italy in the year 70 to place Vespasian on the chap. vi. AND THE WARS ON THE DANUBE. 217 throne, they declined the contingent of cavalry offered by the Jazyges, and appropriately carried with them only a number of the men of chief rank, in order that these should meanwhile be pledges for quiet on the denuded frontier. More serious and continuous watch was needed farther Dad. down on the lower Danube. There, beyond the mighty stream, which was now the boundary of the empire, were settled in the plains of Wallachia and the modern Transylvania the Daci ; in the eastern flat country, in Moldavia, Bessarabia, and onward, in the first instance, the Germanic Bastarnae, and then Sarmatian tribes, such as the Roxolani, a people of horsemen like the Jazyges, at first between the Dnieper and Don (iii. 295), then advanc- iii. 281. ing along the sea-shore. In the first years of Tiberius the vassal prince of Thrace strengthened his troops to ward off the Bastarnae and Scythians ; in the latter years of Tiberius it was urged among other proofs of his government more and more neglecting everything, that he suffered the in- roads of the Dacians and the Sarmatae to pass unpunished. How matters went on in the last years of Nero on either side of the mouths of the Danube is approximately shown by the accidentally preserved report of the governor of Moesia at that time, Tiberius Plautius Silvanus Aelianus. The latter "brought upwards of 100,000 men dwelling beyond the Danube, with their wives and children, and their princes or kings over the river, so that they became liable to pay tribute. He suppressed a movement of the Sar- matae before it came to an outbreak, although he had given away a great part of his troops for the carrying on of war in Armenia (to Corbulo). A number of kings hitherto un- known or at feud with the Romans he brought over to the Roman bank, and compelled them to prostrate themselves before the Roman standards. To the kings of the Bas- tarnae and Roxolani he sent back their sons, who had been made captive or recovered from the enemy, to those of the Dacians their captive brothers, 1 and took hostages from several of them. Thereby the state of peace for the province 1 Regibus Bastarnarum et Roxo- tos aut hostibus ereptos remisit (Orelli, lanorum filios, Dacorum fratrum cap- 750) is miswritten ; it must lunfratres, 218 THE DANUBIAN LANDS book vm. was confirmed as well as further extended. He induced also the king of the Scythians to desist from the siege of the town Chersonesus (Sebastopol) beyond the Borys- thenes. He was the first who, by great consignments of corn from this province, made bread cheaper in Rome.'' We perceive here clearly as well the agitated vortex ot peoples on the left bank of the Danube under the Julio- Claudian dynasty, as also the strong arm of the imperial power, which even beyond the stream sought to protect the Greek towns on the Dnieper and in the Crimea, and was able also in some measure to do so, as will be further set forth when we describe the state of Greek affairs. inade- The forces, however, which Rome had here at her dis- Roman° f P osa l) were more than inadequate. The insignificant gar- forces, rison of Asia Minor, and the fleet, likewise small on the Black Sea, were of account at most for the Greek inhabit- ants of its northern and western coasts. A very difficult task was assigned to the governor of Moesia, who with his two legions had to protect the bank of the Danube from Belgrade to the mouth ; and the aid of the far from obedient Thracians was under the circumstances an addi- tional danger. Especially towards the mouth of the Danube there was wanting a sufficient bulwark against the barbarians now pressing on with increasing weight. The withdrawal on two occasions of the Danubian legions to Italy in the troubles after Nero's death provoked still more at the mouth of the Danube, than on the lower Rhine, incursions of the neighbouring peoples, at first of the Roxolani, then of the Dacians, then of the Sarmatae, that is, probably the Jazyges. There were severe conflicts ; in one of these engagements, apparently with the Jazyges, the brave governor of Moesia, Gaius Fonteius Agrippa, fell. Nevertheless, Vespasian did not proceed to increase the army of the Danube ; J the necessity of strengthening the Asiatic garrisons must have appeared still more urgent, or at anyrate fratrum filios. In like about the year 70 two legions, the manner afterwards per quae is to be 1 3th Gemina and the 15th Apollinaris, read for per quern and rege instead of in room of which latter during its regem. participation in the Armenian war for 1 In Pannonia there were stationed some time the 7th Gemina came in chap. vi. AND THE WARS ON THE DANUBE. 219 and the economy specially enjoined at that time forbade any increase of the army as a whole. He contented him- self with pushing forward the great camps of the army of the Danube to the frontier of the empire, as the pacifica- tion of the interior allowed, and the relations subsisting at the frontier, as well as the breaking up of the Thracian troops brought about by the annexation of Thrace, impera- tively required. Thus the Pannonian camps were brought away from the Drave, opposite to the Suebian kingdom, to Carnuntum and Vindobona (p. 206), and the Dalmatian from the Kerka and the Cettina to the Moesian bank of the Danube, 1 so that the governor of Moesia thenceforth disposed of double the number of legions. A shifting of the proportions of power to the disad- Dacianwar vantage of Rome set in under Domitian, 2 or rather the ° ian- ° mi ~ consequences of the insufficient frontier-defence were then reaped. According to the little we know of the matter, ( C. I. L. hi. p. 482) . Of the two legions added later, 1st Adiutrix and 2d Adiu- trix, the first still at the beginning of the reign of Trajan lay in upper Ger- many (p. 159, note 1), and can only have come to Pannonia under Trajan ; the second stationed under Vespasian in Britain can only have come to Pan- nonia under Domitian (p. 1 74, note 4). The Moesian army numbered after the union with the Dalmatian under Vespasian probably but four legions, consequently as many as the two armies together previously — the later upper -Moesian, 4th Flavia and 7 th Claudia, and the later lower-Moesian, 1st Italica and 5th Macedonica. The positions shifted by the marching to and fro of the year of the four em- perors (Marquardt, Staatsverw. ii. 435), which temporarily brought these le- gions to Moesia, need not deceive us. The subsequent third lower-Moesian legion, the Eleventh, was still under Trajan stationed in upper Germany. 1 Josephus, Bell. lud. vii. 4, 3 : irkeloai. koX fielfrai (frvXa.Ka.is rbv rdirov difKaflev, us elcot rots flapp&pois rty Sidficunv Te\£ws ddiivarov. By this seems meant the transference of the two Dalmatian legions to Moesia. Whither they were transferred we do not know. According to the Roman custom elsewhere it is more probable that they were stationed in the en- virons of the previous headquarters Viminacium than in the remote region of the mouths of the Danube. The camp there probably originated only at the division of the Moesian com- mand and at the erection of the inde- pendent province of lower Moesia under Domitian. 2 The chronology of the Dacian war is involved in much uncertainty. That it had begun already before the war with the Chatti (83), we learn from the Carthaginian inscription (C. /. L. viii. 1082) of a soldier decorated three times by Domitian, in the Dacian, in the German, and again in the Dacian war. Eusebius puts the outbreak of the war, or rather the first great conflict, in the year Abr. 2101 or 2102 = A.D. 85 (more exactly I Oct. 84 — 30 Sept. 85) or 86, the triumph in the year 2106 = 90; these numbers indeed have no claim to complete trust- worthiness. With some probability the triumph is placed in the year 89 (Henzen, Acta Arval. p. 116). 220 THE DANUBIAN LANDS book vm. the change of affairs hinged, quite like the similar one in Caesar's time, upon a single Dacian man ; what king Burebista had planned, king Decebalus seemed destined Decebaius. to execute. How much the real moving-spring lay in his personality, is shown by the story that the Dacian king Duras, in order to bring the right man into the right place, retired from his office in favour of Decebalus. That Decebalus first of all organised in order to strike, is shown by the reports as to his introduction of Roman discipline into the Dacian army, and his enlisting people of capacity among the Romans themselves, and even by the condition proposed by him to the Romans after the victory, that they should send him the necessary work- men to instruct his people in the arts of peace as of war. On what a great scale he set to work is shown by the connections which he formed, westward and eastward, with the Suebi and the Jazyges, and even with the Parthians. The assailants were the Dacians. The governor of the province of Moesia, who first went to oppose them, Oppius Sabinus, lost his life on the field of battle. A number of smaller camps were conquered ; the larger were threatened, the possession of the province itself was at stake. Domi- tian in person resorted to the army, and his representa- tive — he himself was no general and remained in the background — the commandant of the guard, Cornelius Fuscus, led the army over the Danube ; but he paid for the incautious proceeding by a severe defeat, and he too, the second in supreme command, fell before the enemy. His successor, Julianus, a capable officer, defeated the Dacians in their own territory in a great battle near Tapae, and was on the way to achieve lasting results. But, while the struggle with the Dacians was in suspense, Domitian had threatened the Suebi and Jazyges with war, because they had omitted to send to him a contingent against the former ; the messengers, who came to excuse this, he caused to be executed. 1 Here too misfortune 1 The fragment, Dio, lxvii. 7, 1, order of events to a time before the Dind., stands in the sequence of the negotiation with the Lugii. Comp. Ursinian excerpts before lxvii. 5, Hermes, iii. 115. 1, 2, 3, and belongs also in the chap. vi. AND THE WARS ON THE DANUBE. 221 pursued the Roman arms. The Marcomani achieved a victory over the emperor himself; a whole legion was surrounded by the Jazyges and cut down. Shaken by this defeat, Domitian, in spite of the advantages gained by Julianus over the Dacians, hastily concluded with these a peace, which did not indeed prevent him from confer- ring the crown upon the representative of Decebalus in Rome, Diegis, just as if the latter were a vassal of the Romans, or from marching as victor to the Capitol, but which in reality was equivalent to a capitulation. What Decebalus, on the advance of the Roman army into Dacia, had scoffingly offered — to dismiss to his home uninjured every man for whom a yearly payment of two asses was promised to him — became almost true : in the peace the incursions into Moesia were bought off with a fixed sum to be paid yearly. Here a change had to be effected. Domitian, who Dacianwar was doubtless a good administrator of the empire, but ofTra i an - obtuse to the demands of military honour, was followed after the short reign of Nerva by the emperor Trajan, who, first and above all a soldier, not merely tore in pieces that agreement, but also took measures that similar things should not recur. The war against the Suebi and Sarmatae, which was still being continued at Domitian's death (96), was happily ended, as it would seem, under Nerva in the year 97. The new emperor went, even before he held his entrance into the capital of the empire, from the Rhine to the Danube, where he stayed in the winter 98-99, but not to attack the Dacians at once, but to prepare for the war : to this time belongs the con- struction — joining itself on to the roads formed in upper Germany — of the road completed on the right bank of the Danube in the region of Orsova in the year 100 (P- l 5 3)- For the war against the Dacians, in which, as in all his campaigns, he commanded in person, he did not set out till the spring of 101. He crossed the Danube below Viminacium, and advanced against the not far distant capital of the king, Sarmizegetusa. Decebalus with his allies — the Buri and other tribes dwelling to the 222 THE DANUBIAN LANDS book viii. northward took part in this struggle — offered resolute resistance, and it was only by vehement and bloody con- flicts that the Romans cleared their way ; the number of the wounded was so great that the emperor put his own wardrobe at the disposal of the physicians. But victory did not waver ; one stronghold after another fell ; the sisters of the king, the captives from the former war, the standards taken from the armies of Domitian, fell into the hands of the Romans ; for the king, intercepted by Trajan himself and by the brave Lusius Quietus, nothing was left but complete surrender (102). Trajan demanded nothing less than the renunciation of the sovereign power and the entrance of the Dacian kingdom into the client- ship of Rome. The deserters, the arms, the engines of war, the workmen once supplied for these by Rome, had to be delivered up, and the king personally to kneel before the victor ; he divested himself of the right to make war and peace, and promised military service ; the fortresses were either razed or delivered to the Romans, and in these, above all in the capital, there remained a Roman garrison. The strong bridge of stone, which Trajan caused to be thrown over the Danube at Drobetae (opposite Turnu Severinului), secured the communication even in the bad season of the year, and gave to the Dacian garrisons a reserve-support in the near legions of upper-Moesia. Second But the Dacian nation, and above all the king him- war . self, did not know the art of accommodating themselves to dependence, as the kings of Cappadocia and Mauretania had understood it ; or rather they had merely taken upon them the yoke in the hope of ridding themselves of it again on the first opportunity. The signs of this were soon apparent. A portion of the arms to be delivered up was kept back ; the fortresses were not given over as had been stipulated ; an asylum was still granted, moreover, to Roman deserters ; portions of territory were wrested from the Jazyges at enmity with the Daci- ans, or perhaps the occurrence of violations of the frontier on their part was not taken patiently ; a lively and chap. vi. AND THE WARS ON THE DANUBE. 223 suspicious intercourse was maintained with the more re- mote natives still free. Trajan could not but be con- vinced that his work was but half done ; and, rapid in resolution as he was, he, without entering upon further negotiations, declared war once more against the king three years after the conclusion of peace (105). Gladly would the latter have avoided it ; but the demand that he should give himself a captive spoke too clearly. No- thing was left but a struggle of despair, and all were not ready for this ; a great part of the Dacians submitted without resistance. The appeal to the neighbouring peoples to enter jointly into measures for warding off the danger that threatened even their freedom and their national existence sounded without effect ; Decebalus and the Dacians that remained faithful to him stood alone in this war. The attempts to make away with the imperial general by means of deserters, or to purchase tolerable terms by the release of a high officer taken prisoner, likewise broke down. The emperor marched once more as victor into the enemy's capital, and Dece- balus, who up to the last moment had struggled with fate, put himself to death when all was lost (107). This time Trajan made an end ; the war concerned no longer the freedom of the people, but its very existence. The native population were driven out from the best part of the land, and these districts were reoccupied with a non-national population brought in from the mountains of Dalmatia, or the mines, and otherwise preponderantly, as it would appear, from Asia Minor. In several regions, no doubt, the old population yet remained, and even the language of the country maintained its ground. 1 These Dacians, as well as the sections dwelling beyond the bounds, still gave trouble to the Romans — subsequently, for example, under Commodus and Maximianus ; but they stood isolated, and dwindled away. The danger with which the vigorous Thracian race had several times threatened 1 Arrian, Tact. 44, mentions among battle-cries : KeXnicoiis piv tojs KeX- the changes which Hadrian introduced rots lirireda-iv, Tctlkovs Si roU Virais, into the cavalry, that he allowed to 'Peun/coiis Si & ko.1 tup d/i0i KivSiSov 2 The Moesian army gave away m{G>v in and as he had modus. already for several years nominally shared the throne with his father, he entered with the latter's death at once into pos- session of unlimited power. Only for a brief time did the nineteen years' old successor allow the men who had en- joyed his father's confidence — his brother-in-law Pompei- anus, and others who had borne with Marcus the heavy burden of the war — to rule in his spirit. Commodus was in every respect the opposite of his father ; not a scholar, but a fencing-master ; as cowardly and weak in character, as his father was resolute and tenacious of purpose ; as in- dolent and forgetful of duty, as his father was active and conscientious. He not merely gave up the idea of incor- porating the territory won, but voluntarily granted even to the Marcomani conditions such as they had not ventured to hope for. The regulation of the frontier-traffic under Roman control, and the obligation not to injure their neighbours friendly to the Romans, were matters of course ; but the garrisons were withdrawn from their country, and there was retained only the prohibition of settlement on the border-strip. The payment of taxes and the furnish- ing of recruits were doubtless stipulated for, but the former were soon remitted, and the latter were certainly not fur- nished. A similar settlement was made with the Quadi ; and the other Transdanubians must have been similarly dealt with. Thereby the conquests made were given up, and the work of many years of warfare was in vain ; if no more was wished for, a similar arrangement of things might have been reached much earlier. Nevertheless the colonate. chap. vi. AND THE WARS ON THE DANUBE. 237 Marcomanian war secured in these regions the supremacy of Rome for the sequel, in spite of the fact that Rome let slip the prize of victory. It was not by the tribes that had taken part in it that the blow was dealt, to which the Roman world-power succumbed. Another permanent consequence of this war was con- The nected with the removals, to which it gave occasion, of the Transdanubians over into the Roman empire. Of them- selves such changes of settlement had occurred at all times ; the Sugambri, transplanted under Augustus to Gaul, the Dacians sent to Thrace, were nothing but new subjects or communities of subjects added to those for- merly existing, and probably not much different were the 3000 Naristae, whom Marcus allowed to exchange their settlements westward of Bohemia for such settlements within the empire, while the like request was refused to the otherwise unknown Astingi on the Dacian north frontier. But the Germans settled by him not merely in the land of the Danube, but in Italy itself at Ravenna, were neither free subjects nor strictly non-free persons ; these were the beginnings of the Roman villanage, the colonate, the in- fluence of which on the agricultural economy of the whole state is to be set forth in another connection. That Ravennate settlement, however, had no permanence ; the men rose in revolt and had to be conveyed away, so that the new colonate remained restricted primarily to the provinces, particularly to the lands of the Danube. The great war on the middle Danube was once more The ad- followed by a six years' time of peace, the blessings of ^^hmen which could not be completely neutralised by the internal misgovernment that was constantly increasing during its course. No doubt various isolated accounts show that the frontier, especially the Dacian, which was most exposed, remained not without trouble ; but above all, the stern military government of Severus did its duty here, and at least Marcomani and Quadi appear even under his im- mediate successors in unconditional dependence, so that the son of Severus could cite a prince of the Quadi before him and lay his head at his feet. The conflicts occurring 238 THE DANUBIAN LANDS book vim. at this epoch on the lower Danube were of subordinate importance. But probably at this period a comprehensive shifting of peoples from the north-east towards the Black Sea took place, and the Roman frontier -guard on the lower Danube had to confront new and more dangerous opponents. Up to this time the antagonists of the Romans there had been chiefly Sarmatian tribes, among whom the Roxolani came into closest contact with them ; of Germans there were settled here at that time only the Bastarnae, who had been long at home in this region. Now the Roxolani disappear, merged possibly among the Carpi apparently akin to them, who thenceforth were the nearest neighbours of the Romans on the lower Danube, perhaps in the valleys of the Seret and Pruth. Goths. By the side of the Carpi came, likewise as immediate neighbours of the Romans at the mouth of the Danube, the people of the Goths. This Germanic stock migrated, according to the tradition which has been preserved to us, from Scandinavia over the Baltic towards the region of the Vistula, and from this to the Black Sea ; in accordance with this the Roman geographers of the second century know them at the Vistula, and Roman history from the first quarter of the third at the north-west coast of the Black Sea. Thenceforth they appear here constantly on the increase ; the remains of the Bastarnae retired before them to the right bank of the Danube under the emperor Probus, the remains of the Carpi under the emperor Dio- cletian, while beyond doubt a great part of the former as of the latter mingled among the Goths and joined them. On the whole this catastrophe may be designated as that of the Gothic war only in the sense in which that which set in under Marcus is called the war of the Marcomani ; the whole mass of peoples set in movement by the stream of migration from the north-east to the Black Sea took part in it ; and took part all the more, seeing that these attacks took place just as much by land over the lower Danube as by water from the north coasts of the Black Sea, in an inextricable complication of landward and maritime piracy. Not unsuitably, therefore, the learned chap. vi. AND THE WARS ON THE DANUBE. 239 Athenian who fought in this war and has narrated it, prefers to term it the Scythian, as he includes under this name — which, like the Pelasgian, forms the despair of the historian — all Germanic and non-Germanic enemies of the empire. What is to be told of these expeditions will here be brought together, so far as the confusion of tradi- tion, which is only too much in keeping with the confusion of these fearful times, allows. The year 238 — a year also of civil war, when there Gothic were four emperors — is designated as that in which the wars- war against those here first named Goths began. 1 As the coins of Tyra and Olbia cease with Alexander (f 235), these Roman possessions situated beyond the boundary of the empire had doubtless become some years earlier a prey to the new enemy. In that year they first crossed the Danube, and the most northerly of the Moesian coast towns, Istros, was the first victim. Gordianus, who emerged out of the confusions of this time as ruler, is designated as conqueror of the Goths ; it is more certain that the Roman government at any rate under him, if not already earlier, agreed to buy off the Gothic incursions. 2 As was natural, the Carpi demanded the same as the emperor had granted to the inferior Goths ; when the demand was not granted, they invaded the Roman territory in the 1 The alleged first mention of the 2 Petrus Patricius Jr. 8. The ad- Goths in the biography of Caracalla, ministration of the legate of lower c. 10, rests on a misunderstanding. Moesia here mentioned, Tullius If really a senator allowed himself Menophilus, is fixed by coins cer- the malicious jest of assigning to the tainly to the time of Gordian, and murderer of Geta the name Geticus, with probability to 238-240 (Bor- because he on his march from the ghesi, Opp. ii. 227). As the begin- Danube to the east had conquered ning of the Gothic war and the some Getic hordes (tumultuariis proe- destruction of Istros are fixed by liis), he meant Dacians, not the Dexippus {vita Max. et Balb. 16) Goths, scarcely at that time dwelling at 238, it is natural to bring into there and hardly known to the connection with these events the un- Roman public, whose identification dertaking of tribute ; at any rate it with the Getae was certainly only a was then renewed. The vain sieges later invention. — We may add that of Marcianopolis and Philippopolis by the statement that the emperor the Goths (Dexippus, Jr. 18, 19) Maximinus (235-238) was the son of may have followed on the capture of a Goth settled in the neighbouring Istros. Jordanes, Get. 16, 92, puts Thrace, carries us still further back ; the former under Philippus, but is in yet not much weight is to be attached chronological questions not a valid to it. witness. 240 THE DANUBIAN LANDS book viii. year 245. The emperor Philippus — Gordianus was at that time already dead — repulsed them, and energetic action with the combined strength of the great empire would probably here have checked the barbarians. Dedus. But in these years the murderer of an emperor reached the throne as surely as he found in turn his own murderer and successor ; it was just in the imperilled regions of the Danube that the army proclaimed against the emperor Philippus first Marinus Pacatianus, and, after he was set aside, Traianus Decius, which latter in fact vanquished his antagonist in Italy, and was acknowledged as ruler. He was an able and brave man, not unworthy of the two names which he bore, and entered, so soon as he could, resolutely into the conflicts on the Danube ; but what the civil war waged in the meanwhile had destroyed, could no longer be retrieved. While the Romans were fighting with one another the Goths and the Carpi had united, and had under the Gothic prince Cniva invaded Moesia denuded of troops. The governor of the province, Trebonianus Gallus, threw himself with his force into Nicopolis on the Haemus, and was here besieged by the Goths ; these at the same time pillaged Thrace and besieged its capital, the great and strong Philippopolis ; indeed they reached as far as Mace- donia, and invested Thessalonica, where the governor Priscus found this just, a fitting moment to have himself proclaimed as emperor. When Decius arrived to combat at once his rival and the public foe, the former was doubt- less without difficulty set aside, and success also attended the relief of Nicopolis, where 30,000 Goths are said to have fallen. But the Goths, retreating to Thrace, con- quered in turn at Beroe (Alt-Zagora), threw the Romans back on Moesia, and reduced Nicopolis there as well as Anchialus in Thrace and even Philippopolis, where 100,000 men are said to have come into their power. Thereupon they marched northwards to bring into safety their enormous booty. Decius projected the plan of in- flicting a blow on the enemy at the crossing of the Danube. He stationed a division under Gallus on the bank, and hoped to be able to throw the Goths upon this, and to chap. vi. AND THE WARS ON THE DANUBE. 241 cut off their retreat. But at Abrittus, a place on the Moesian frontier, the fortune of war, or else the treachery of Gallus, decided against them. Decius perished with His death, his son, and Gallus, who was proclaimed as his successor, began his reign by once more assuring to the Goths the annual payments of money (25 i). 1 This utter defeat of Roman arms as of Roman policy, the fall of the emperor, the first who lost his life in conflict with the barbarians — a piece of news which deeply moved men's minds even in this age demoralised by its familiarity with misfortune — the disgraceful capitulation following thereon, placed in fact the integrity of the empire at stake. Serious crises on the middle Danube, threatening probably the loss of Dacia, must have been the immediate consequence. Once more this was averted ; the governor of Pannonia, Marcus Aemilius Aemilianus, a good soldier, achieved an im- portant success of arms, and drove the enemy over the frontier. But Nemesis bore sway. The consequence of this victory, achieved in the name of Gallus, was, that the army renounced allegiance to the betrayer of Decius and chose their general as his successor. Once more there- fore civil war took precedence of frontier-defence ; and, while Aemilianus no doubt vanquished Gallus in Italy but soon afterwards succumbed to his general Valeri- anus (254), Dacia was lost for the empire — how, and Loss of to whom, we know not. 2 The last coin struck by this Dacia - province, and the latest inscription found there, are of the year 255, the last coin of the neighbouring Vimi- nacium in upper Moesia of the following year ; in the first years of Valerianus and Gallienus therefore the bar- barians occupied the Roman territory on the left bank 1 The reports of these occurrences some measure combined. The same in Zosimus, i. 21-24, Zonaras, xii. source lies at the bottom of the im- 20, Ammianus, xxxi. 5, 16, 17 penal biographies and Jordanes ; but (which accounts, down to that con- both have disfigured and falsified cerning Philippopolis, are fixed as it to such a degree that use can be belonging to this time by the fact made of their statements only with that the latter recurs in Zosimus), great caution. Victor, Cats. 29, is although all fragmentary or in dis- independent. order, may have flowed from the 2 Perhaps the irruption of the report of Dexippus, of which fir. 16, Marcomani in Zosimus, i. 29, refers 1 9, are preserved, and may be in to this. VOL. I. 16 the Black Sea, 242 THE DANUBIAN LANDS book viii. of the Danube, and certainly also pressed across to the right. Before we pursue further the development of affairs on the lower Danube, it appears necessary to cast a glance at piracy, as it was then in vogue in the eastern half of the Mediterranean, and the maritime expeditions of the Goths and their allies originating from it. Piracy on That the Roman fleet could at no time be dispensed with on the Black Sea, and piracy there was probably never extirpated, was implied in the very nature of the Roman rule as it had taken shape on its coasts. The Romans were in firm possession only perhaps from the mouths of the Danube as far as Trapezus. It is true that on the one hand Tyra at the mouth of the Dniester and Olbia on the bay at the mouth of the Dnieper, on the other side the Caucasian harbours in the regions of the modern Suchum-Kaleh, Dioscurias and Pityus, were Roman. The intervening Bosporan kingdom in the Crimea also stood under Roman protection, and had a Roman garrison subject to the governor of Moesia. But on these shores, for the most part far from inviting, there were only those posts formerly held either as old Greek settlements or as Roman fortresses ; the coast itself was desolate or in the hands of the natives filling the interior, who, compre- hended under the general name of Scythians, mostly of Sarmatian descent, never were, or were to become, subject to the Romans ; it was enough if they did not directly lay hands on the Romans or their clients. Accordingly, it is not to be wondered at, that even in the time of Tiberius the pirates of the east coast not merely made the Black Sea insecure, but also landed and levied contributions on the villages and towns of the coast. If, under Pius or Marcus, a band of the Costoboci dwelling on the north- western shore fell upon the inland town Elateia situated in the heart of Phocis, and came to blows under its walls with the citizens, this event, which certainly only by accident stands forth for us as isolated, shows that the same phenomena which preceded the downfall of the government of the senate were now renewed, and even chap. vi. AND THE WARS ON THE DANUBE. 243 with the imperial power maintaining itself outwardly unshaken not merely individual piratical ships, but squad- rons of pirates cruised in the Black and even in the Mediterranean seas. The decline of the government, clearly discernible after the death of Severus, and above all after the end of the last dynasty, manifested itself then, as was natural, especially in the further decay of marine police. The accounts, in detail far from trust- worthy, mention already in the time before Decius the appearance of a great fleet of pirates in the Aegean Sea ; then under Decius the plundering of the Pamphylian coast and of the Graeco- Asiatic islands ; under Gallus maraudings of pirates in Asia Minor as far as Pessinus and Ephesus. 1 These were predatory expeditions. These comrades plundered the coasts far and wide, and made even, as we see, bold raids into the interior ; but nothing is mentioned of the destruction of towns, and the pirates shunned coming into collision with Roman troops ; the attack was chiefly directed against such regions as had no troops stationed in them. Under Valerian these expeditions assume a different Maritime character. The nature of the raids varies so much from e ?p, ed 3?°? s of the Goths the earlier, that the raid, in itself not specially important, and allies. of the Borani against Pityus under Valerian could be designated by intelligent reporters precisely as the beginning of this movement, 2 and that the pirates were 1 Ammianus, xxxi. 5, 15 ; duobus siege of Pityus, and are more a part navium milibus perrupto Bosporo et of the migration of peoples than pir- litoribus Propontidis Scythicamm gen- atical raids. The number of the ships tium catervae transgressae ediderunt might indeed be transferred hither by quidem acerbas terra marique strages : error of memory from the expedition sed amissa suorum parte maxima of the year 269. To the same con- reverterunt ; whereupon the cata- nection belongs the notice in Zosimus, strophe of the Decii is narrated, and i. 28, as to the Scythian expeditions into this is inwoven the further notice : into Asia and Cappadocia as far as obsessae Pamphyliae civitates (to which Ephesus and Pessinus. The account must belong the siege of Side in as to Ephesus in the biography of Dexippus himself, fr. 23), insula Gallienus, u. 6, is the same, but populate complures, as also the siege transposed as to time, of Cyzicus. If in this retrospect all 2 In the case of Zosimus himself is not confused — which cannot well we should not expect complete under - be assumed to be the case with Am- standing of the matter ; but his voucher mianus — this falls before those naval Dexippus, who was a contemporary expeditions which begin with the and took part in the matter, knew 244 THE DANUBIAN LANDS book viii. for a long time called in Asia by the name of this tribe not otherwise known to us. These expeditions proceed no longer from the old native dwellers beside the Black Sea, but from the hordes pressing behind them. What had hitherto been piracy begins to form a portion of that migratory movement of peoples to which the advance of the Goths on the lower Danube belongs. The peoples taking part in it are very varied and in part little known ; in the later expeditions the Germanic Heruli, then dwelling beside the Maeotis, appear to have played a leading part. The Goths also took part, but, so far as sea-voyages are concerned — and tolerably exact reports of these are before us — not in a prominent manner ; strictly speaking, these expeditions are more correctly termed Scythian than Gothic. The maritime centre of these aggressions was the mouth of the Dniester, the port of Tyra. 1 The Greek towns of the Bosporus, abandoned through the bankruptcy of the imperial power, without protection to the hordes pressing onward, and expecting to be besieged by them, consented, half under compulsion, half voluntarily, to convey in their vessels, and by their mariners, the in- convenient new neighbours over to the nearest Roman possessions on the north coast of Pontus — for which these neighbours themselves lacked the needful means and the needful skill. It was thus that the expedition against Pityus was brought about. The Borani were landed and, confident of success, sent back the ships. But the resolute commander of Pityus, Successianus, repelled the attack ; and the assailants, fearing the arrival of the other Roman garrisons, hastily withdrew, for which they had difficulty in procuring the necessary transports. But the well why he termed the Bithynian the year 264, must be that to Trnpezus, expedition the Sevripa lipoSos (Zos. just as the Bithynian therewith con- i. 35) ; and even in Zosimus we nectcd must be that which Zosimus discern clearly the contrast, intended terms the second ; here indeed evcry- by Dexippus, between the expe- thing is confused, dition of the Borani against Pityus 1 This is said by Zosimus, i. 42, and Trapezus and the traditional and follows also from the relation of piratic voyages. In the biography of the Bosporans to the first (i. 32), and Gallienus the Scythian expedition to that of the first to the second expedi- Cappadocia, narrated at c. II, under tion (i. 34). chap. vi. AND THE WARS ON THE DANUBE. 245 plan was not given up ; in the next year they came back, and, as the commandant had meanwhile been changed, the fortress surrendered. The Borani, who this time had To Tra- retained the Bosporan vessels and had them manned by pezus ' pressed mariners and Roman captives, possessed them- selves of the coast far and wide, and reached as far as Trapezus. Into this well fortified and strongly garrisoned town all had fled, and the barbarians were not in a position for a real siege. But the leadership of the Romans was bad, and the military discipline so on the decline that not even the walls were occupied ; so the barbarians scaled them by night, without encountering resistance, and in the great and rich city enormous booty, including a number of ships, fell into their hands. They returned successful from the far distant land to the Maeotis. Excited by this success, a second expedition of other To Bithy- but neighbouring Scythian bands was in the following nia ' winter directed against Bithynia. It is significant of the unsettled state of things that the instigator of this move- ment was Chrysogonus, a Greek of Nicomedia, and that he was highly honoured by the barbarians fqr its successful result. This expedition was undertaken — as the necessary number of ships was not to be procured — partly by land partly by water ; it was only in the neighbourhood of, Byzantium that the pirates succeeded in possessing them- selves of a considerable number of fishing-boats, and so they arrived along the Asiatic coast at Chalcedon, whose strong garrison on this news ran off. Not merely this town fell into their hands, but also along the coast Nicomedia, Chios, Apamea ; in the interior Nicaea and Prusa ; Nicomedia and Nicaea they burnt down, and reached as far as Rhyndacus. Thence they sailed home, laden with the treasures of the rich land and of its con- siderable cities. The expedition against Bithynia had already been To Greece. undertaken in part by land ; all the more were the attacks that were directed against European Greece composed of piratical expeditions by land and sea. If Moesia and 246 THE DANUBIAN LANDS book viii. Thrace were not permanently occupied by the Goths, they yet came and went there as if they were at home, and roved from thence far into Macedonia. Even Achaia expected under Valerian invasion from this side ; Thermopylae and the Isthmus were barricaded, and the Athenians set to work to restore their walls that had lain in ruins since the siege by Sulla. The barbarians did not come then and by this route. But under Gallienus a fleet of five hundred sail, this time chiefly Heruli, appeared before the port of Byzantium, which, however, had not yet lost its capacity of defence ; the ships of the Byzantines successfully repulsed the robbers. These sailed onward, showed themselves on the Asiatic coast before Cyzicus not formerly attacked, and arrived from thence by way of Lemnos and Imbros at Greece proper. Athens, Corinth, Argos, Sparta, were pillaged and destroyed. It was always something that, as in the times of the Persian wars, the citizens of the destroyed Athens, two thousand in number, laid an ambush for the retiring barbarians, and, under the leadership of their equally learned and brave captain, Publius Herennius Dexippus, of the old and noble family of the Kerykes, with support of the Roman fleet, inflicted a notable loss on the pirates. On the return home, which took place in part by the land route, the emperor Gallienus attacked them in Thrace at the river Nestus and put to death a consider- able number of their men. 1 Theim- In order completely to survey the measure of mis- penaigov- f or t une we mu st take into account that in this empire eminent of ' r the Gothic going to shreds, and above all in the provinces overrun period. ky the enemy, one officer after another grasped at the crown, which hardly any longer existed. It is not worth the trouble to record the names of these ephemeral 1 The report of Dexippus as to this the event is placed under Claudius, expedition is given in extract by through error or through falsification, Syncellus, p. 717 (where dxeXiWos which grudged this victory to Gallienus. must be read for (WXipres), Zosimus, The biography of Gallienus narrates i. 39, and the biographer of Gallienus, the incident apparently twice, fn;st c. 13. Fr. 22 is a portion of his shortly in c. 6 under the year 262 ; own narrative. In the continuator then better, under or after 265, in of Dio, on whom Zonaras depends, u. 13. chap. vi. AND THE WARS ON THE DANUBE. 247 wearers of the purple ; it marks the situation that, after the devastation of Bithynia by the pirates, the emperor Valerian omitted to send thither an extraordinary com- mandant, because every general was, not without reason, regarded by him as a rival. This co-operated to produce the almost thoroughly passive attitude of the government in presence of this sore emergency. Yet, on the other hand, undoubtedly a good part of this irresponsible passiveness is to be traced to the personality of the rulers : Valerian was weak and aged, Gallienus vehement and dissolute, and neither the one nor the other was equal to the guidance of the vessel of the state in a storm. Marcianus, to whom Gallienus after the invasion of Achaia had committed the command in these regions, operated not without success ; but the matter did not gain any real turn for the better so long as Gallienus occupied the throne. After the murder of Gallienus (268), perhaps on the Gothic vic- news of it, the barbarians, again led by the Heruli, but Qaudius this time with united forces, undertook an assault on the imperial frontier, such as there had not been hitherto, with a powerful fleet, and probably at the same time by land from the Danube. 1 The fleet had much to suffer from storms in the Propontis ; then it divided, and the Goths advanced partly against Thessaly and Greece, partly against Crete and Rhodes ; the chief mass resorted to Macedonia and thence penetrated into the interior, beyond doubt in combination with the bands that had marched into Thrace. But the emperor Claudius, who marched 1 In our traditional accounts this of the expedition, in the first instance expedition appears as a pure sea- against Tomis and Marcianopolis, it voyage, undertaken with (probably) is more than probable that in it the 2000 ships (so the biography of procedure described by Zos. i. 34 was Claudius ; the numbers 6000 and followed, and a portion marched by goo, between which the tradition in land ; and under this supposition even Zosimus, i. 42, wavers, are probably a contemporary might well estimate both corrupt) and 320,000 men. It the number of assailants at that figure, is, however, far from credible that The course of the campaign, parti- Dexippus, to whom these statements cularly the place of the decisive battle, must be traced back, can have put shows that they had by no means to the latter figure in this way. On the do merely with a fleet, other hand, considering the direction 248 THE DANUBIAN LANDS book vra. up in person with a strong force, brought relief at length to the Thessalonians oft besieged but now reduced to extremity ; he drove the Goths before him up the valley of the Axius (Vardar) and onward over the mountains to upper Moesia ; after various conflicts, with changing fortune of war, he achieved here in the Morava valley near Naissus a brilliant victory, in which 50,000 of the enemy are said to have fallen. The Goths retired broken up, first in the direction towards Macedonia, then through Thrace to the Haemus, in order to put the Danube between themselves and the enemy. A quarrel in the Roman camp, this time between infantry and cavalry, had almost given them once more a respite ; but, when it came to fighting, the cavalry could not bear to leave their comrades in the lurch, and so the united army was once more victorious. A severe pestilence, which raged in all the years of distress, but especially then in those regions, and above all in the armies, did great injury doubtless to the Romans — the emperor Claudius himself succumbed to it — but the great army of the Northmen was utterly extirpated, and the numerous captives were incorporated Renewed in the Roman armies or made serfs. The hydra of the Dan? ° f mm t ai y revolutions, too, was in some measure subdued ; ube-fron- Claudius, and after him Aurelian, were masters in the empire after another fashion than could be said of Gallienus. The renewal of the fleet, towards which a beginning had been made under Gallienus, would not be wanting. The Dacia of Trajan was, and remained, lost ; Aurelian withdrew the posts still holding out there, and gave to the possessors dislodged or inclined for emigration new dwellings on the Moesian bank. But Thrace and Moesia, which for a time had belonged more to the Goths than to the Romans, returned under Roman rule, and at least the frontier of the Danube was once more fortified. Character We may not assign to these Gothic and Scythian ° f *? expeditions by land and by sea, which fill up the twenty wars. years 250-269, such significance, as if the hordes moving forth had been minded to take permanent possession of the countries which they traversed. Such a plan cannot chap. vi. AND THE WARS ON THE DANUBE. 249 be shown to have existed even for Moesia and Thrace, to say nothing of the more remote coasts ; hardly, moreover, were the assailants numerous enough to undertake invasions proper. As the bad government of the last rulers, and above all the untrustworthiness of the troops, far more than the superior power of the barbarians, called forth the flooding of the territory by land and sea robbers, so the re-establish- ment of internal order and the energetic demeanour of the government of themselves brought its deliverance. The Roman state could not yet be broken if it did not break itself. But still it was a great work to rally the govern- ment again as Claudius had done it. We know somewhat less even of him than of most regents of this time, as the probably fictitious carrying back of the Constantinian pedigree to him has repainted his portrait after the tame pattern of perfection ; but this very association, as well as the numberless coins struck in his honour after his death, show that he was regarded by the next generation as the deliverer of the state, and in this it cannot have been mistaken. These Scythian expeditions were at all events a prelude of the later migration of peoples ; and the destruction of cities, which distinguishes them from the ordinary piratic voyages, took place at that time to such an extent that the prosperity as well as the culture of Greece and Asia Minor never recovered from it. On the re-established frontier of the Danube Aurelian The Danu- consolidated the victory achieved, inasmuch as he con- t^J^ ducted the defensive once more offensively, and, crossing the of the 3d Danube at its mouth, defeated beyond it not only the Carpi, who thenceforth stood in client-relation to the Romans, but also the Goths under king Canabaudes. His successor Probus took, as was already stated, the remains of the Bastarnae, hard pressed by the Goths, over to the Roman bank, just as Diocletian in the year 295 took the remnant of the Carpi. This points to the fact that beyond the river the empire of the Goths was consolidated ; but they came no further. The border-fortresses were reinforced ; counter-Aquincum (contra Aquincum, Pesth) was con- structed in the year 294. The piratic expeditions did not century. ment. 250 THE DANUBIAN LANDS book vm. entirely disappear. Under Tacitus hordes from the Maeotis appeared in Cilicia. The Franks, whom Probus had settled on the Black Sea, procured for themselves vessels, and sailed home to their North Sea, after plunder- ing by the way on the Sicilian and African coasts. By land, too, there was no cessation of arms, as indeed all the numerous Sarmatian victories of Diocletian, and a part of his Germanic, would fall to the regions of the Danube ; but it was only under Constantine that matters again came to a serious war with the Goths, which had a successful issue. The preponderance of Rome was re-established after the Gothic victory of Claudius as firmly as before. niyrising of The war-history which we have just unfolded did not forceandof *" a ^ to react wl ^ a general and lasting effect upon the inter- the govern- nal organisation of the Roman political and military system. It has already been pointed out that the corps of the Rhine, holding in the early imperial period the leading position in the army, yielded their primacy already under Trajan to the legions of the Danube. While under Augustus six legions were stationed in the region of the Danube and eight in that of the Rhine, after the Dacian wars of Domitian and Trajan in the second century the Rhine-camps numbered only four, the camps of the Danube ten, and after the Marcomanian war even twelve, legions. Inasmuch as since Hadrian's time the Italian element, apart from the officers, had disappeared from the army, and, taken on the whole, every regiment was recruited in the district in which it was quartered, the most of the soldiers of the Danubian army, and not less the centurions who rose from the ranks, were natives of Pannonia, Dacia, Moesia, Thrace. The new legions formed under Marcus proceeded from Illyricum, and the extraordinary supple- mental levies which the troops then needed were probably likewise taken chiefly from the districts in which the armies were stationed. Thus the primacy of the Danu- bian armies, which the war of the three emperors in the time of Severus established and increased, was at the same time a primacy of Illyrian soldiers ; and this reached a very emphatic expression in the reform of the guard under chap. vi. AND THE WARS ON THE DANUBE. 251 Severus. This primacy did not, properly speaking, affect the higher spheres of government, so long as the position of officer still coincided with that of imperial official, although the equestrian career was accessible to the com- mon soldier through the intervening link of the centurion- ate at all times, and thus the Illyrians early found their way into that career ; as indeed, already, in the year 235, a native Thracian, Gaius Julius Varus Maximinus, in the year 248 a native Pannonian, Trajanus Decius, had in this way attained even to the purple. But when Gallienus, in a distrust certainly but too well justified, excluded the class of senators from serving as officers, what had hitherto held good as to the soldiers became necessarily extended to the officers also. It was thus simply a matter of course that the soldiers belonging to the army of the Danube, and mostly springing from Illyrian districts, played thence- forth the first part also in government, and, so far as the army made the emperors, these were likewise as to the majority Illyrians. Thus Gallienus was followed by Claudius the Dardanian, Aurelianus from Moesia, Probus from Pannonia, Diocletianus from Dalmatia, Maximianus from Pannonia, Constantius from Dardania, Galerius from Serdica ; as to the last named, an author writing under the Constantinian dynasty brings into prominence their descent from Illyricum, and adds that they, with little culture but good preliminary training by labour in the field and service in war, had been excellent rulers. Such service as the Albanians for a long time rendered to the Turkish empire, their predecessors likewise rendered to the Roman imperial state, when this had arrived at simi- lar disorder and similar barbarism. Only, the Illyrian regeneration of the Roman imperial order may not be conceived of as a national reorganisation ; it was simply the propping up, by soldiers, of an empire utterly reduced through the misgovernment of rulers of gentler birth. Italy had wholly ceased to be military ; and history does not acknowledge the ruler's right without the warrior's power. CHAPTER VII. GREEK EUROPE. Hellenism With the general intellectual development of the Hellenes heUeiSm" ^ e political development of their republics had not kept equal pace, or rather the luxuriant growth of the former had — just as too full a bloom bursts the cup that con- tains it — not allowed any individual commonwealth to acquire the extent and stability which are preliminary conditions for the thorough formation of a state. The petty-state-system of individual cities or city-leagues could not but be stunted in itself or fall a prey to the barbarians. Panhellenism alone guaranteed alike the continued exist- ence of the nation and its further development in presence of the alien races dwelling around it. It was realised by the treaty which king Philip of Macedonia, the father of Alexander, concluded in Corinth with the states of Hellas. This was, in name, a federal agreement, in fact, the sub- jection of the republics to the monarchy, but a subjection, which took effect only as regards external relations, seeing that the absolute generalship in opposition to the national foe was transferred by almost all towns of the Greek main- land to the Macedonian general, while in other respects freedom and autonomy were left to them ; and this was, as circumstances stood, the only possible realisation of Pan- hellenism and the form regulating in substance the future of Greece. It subsisted in presence of Philip and Alex- ander, though the Hellenic idealists were reluctant, as they always were, to acknowledge the realised ideal as such. Then, when the kingdom of Alexander fell to pieces, all chap. vii. v 6W ri? 12, which cannot be defended either McLKeSovlq, Trpov&pwTo, in which case for Augustus's time or for that of Dio, the remaining part of Epirus appears but because Tacitus on the year 1 7 to be assigned to the province of (Ann. ii. 53) reckons Nicopolis to Illyricum (reckoned here by Strabo Achaia. But at least from the time — erroneously as regards his time of Trajan Epirus with Acarnania — among the senatorial). To take forms a procuratorial province of its ixi%pi inclusively is — apart from con- own (Ptolem. iii. 13; C. /. Z. iii. siderations of fact — unsuitable for this 536; Marquardt, Staatsalth. v. 1,331). very reason, because according to the Thessaly and all the country north- closing words the regions previously ward of Oeta constantly remained named "are assigned to Macedonia." with Macedonia. chap. vii. GREEK EUROPE. 257 the several cities were admonished to regulate their own public affairs. Moreover certain general rules were laid down by the leading power for the several communal constitutions, and according to this scheme these were reorganised in an anti- democratic sense. It was only within these limits that the individual community retained autonomy and a magistracy of its own. It retained also its own courts ; but the Greek stood at the same time de jure under the rods and axes of the praetor, and at least could be sentenced — on account of any offence which admitted of being regarded as rebellion against the lead- ing power — by the Roman officials to a money-fine or banishment, or even capital punishment. 1 The communi- ties taxed themselves ; but they had throughout to pay to Rome a definite sum, on the whole, apparently, not on a high scale. Garrisons were not assigned, as formerly in the Macedonian period, to the towns, for the troops stationed in Macedonia were in a position, should need arise, to move also into Greece. But a graver blame than that falling on the memory of Alexander through the destruction of Thebes rests on the Roman aristocracy for the razing of Corinth. The other measures, odious and exasperating as in part they were, particularly as imposed by foreign rule, might, taken as a whole, be unavoidable and have in various respects a salutary operation ; they were the inevitable palinode of the original Roman policy — in part truly impolitic — of forgiving and forgetting towards the Hellenes. But in the treatment of Corinth mercantile selfishness had after an ill-omened fashion shown itself more powerful than all Philhellenism. Amidst all this, the fundamental idea of Roman 1 Nothing gives a clearer idea of the and to the organisation given by the position of the Greeks in the last cen- Romans to the Achaeans [ij iiroSoBeiaa tury of the Roman republic than the tois 'A^atois v, t&v 5' tiXhuv p.€Tplws . . . tfyyotip^voL, robs 5£ (3ap(3&povs irpbs tt]v e/caorots airrtbv odcrav tptiaiv TrcuSeiWres. 1 But the Hellenic literati remained grateful to their colleague and patron. In the Apollonius- romance (v. 41) the great sage from Cappadocia re- fuses Vespasian the honour of his company, because he had made the Hellenes slaves, just as they were on the point of again speaking Ionic or Doric, and writes to him various billets of delectable coarseness. A man of Soloi, who broke his neck and then became alive again, and on this occasion saw all that Dante be- held, reported that he had met with Nero's soul, into which the agents of the world-judgment had driven flaming nails, and were employed in turning it into a viper ; but a heavenly voice had interposed, and ordered them to transform the man — on account of his Philhellenism when on earth — into a less repulsive animal (Plutarch, De sera nu?n. vind., at the end). 2 At least in the ordinance of Hadrian regarding the deliveries of oil to the community incumbent on the Athenian landowners (C. I. A. iii. 18), the decision was indeed given to the Boule and the Ekklesia, but appeal to the emperor or the pro- consul was allowed. chap. vii. GREEK EUROPE. 263 coinage, .without even putting the emperor's head on its coins, and even on Spartan coins of the first imperial period it is frequently wanting. In Athens even the old reckoning by drachmae and oboli continued ; only that, it is true, the local Attic drachma of this period was nothing but small money current on the spot, and as to value circulated as obolus of the Attic imperial drachma or of the Roman denarius. Even the formal exercise of the right of war and peace was in individual treaties granted to such states. 1 Numerous institutions quite at variance with the Italian municipal organisation remained in existence, such as the annual change of the members of council and the daily allowance -moneys of these and the jurymen, which, at least at Rhodes, were still paid in the imperial period. As a matter of course, the Roman government nevertheless exercised continuously a regulative influence over the constitution even of the freed communities. Thus, for example, the Athenian constitution was, whether at the end of the republic or by Caesar or Augustus, modi- fied in such a way that the right of bringing a proposal before the burgesses belonged no longer to every burgess, but, as according to the Roman arrangement, only to definite officials ; and among the great number of officials, who were mere figures, the conduct of business was placed in the hands of a single one — the Strategos. Certainly in this way various further reforms were carried out, the presence of which, in dependent as in independent Greece, we everywhere discern, without being able to determine the time and occasion of the reform. Thus the right, or rather the wrong, of asylums, which, as survivals of a lawless period, had now become pious retreats for bad debtors and criminals, was certainly, if not set aside, at least restricted in this province also. The institution of proxenia — originally an appropriate arrangement, that may be compared to our foreign consulates, but politically dangerous through the bestowal of full civil rights and 1 What Strabo reports (xiv. 3, 3, of alliance, except when the Romans p. 665) of the Lycian cities-league, in allowed it or it operated for their ad- his time autonomous — that it had not vantage — may probably be, without the right of war and peace and that ceremony, held to relate also to Athens. Greek cities. 264 GREEK EUROPE. book viii. often also of the privilege of exemption from taxes on the friendly foreigner, especially considering the extent to which it was granted — was set aside by the Roman, government, apparently only at the beginning of the im- perial period ; in room of which thereupon came, after the Italian fashion, the empty city-patronate, which did not come into contact with the system of taxation. Lastly, the Roman government, as wielding supreme sovereignty over these dependent republics just as over the client- princes, always regarded it as its right, and exercised the power, to cancel the free constitution in case of misuse, and to take the town into its own administration. But partly the sworn agreement, ' partly the powerlessness of these nominally allied states, gave to these treaties a greater stability than is discernible in the relation to the client-princes. Diets of the While the freed communities of Achaia retained their previous legal position under the empire, Augustus con- ferred on those communities of the province, in which free- dom was not granted or possessed, a new and better legal position. As he had given to the Greeks of Europe a common centre in the reorganised Delphic Amphictiony, he allowed also all the towns of the province of Achaia, so far as they were placed under Roman administration, to constitute themselves as a collective union, and to meet annually in Argos, the most considerable town of non-free Greece, as a national assembly. 1 Thereby not merely 1 At all events the hitherto known were united with the Koivbv proper of presidents of the Koivbv twv 'A.%atSiv, the Achaeans into that wider league, whose home is made out, are from whose existence and diets in Argos Argos, Messene, Corone in Messenia are vouched for by the inscriptions of (Foucart-Lebas, ii. 305), and there Acraephia mentioned in the next note, have been hitherto found among them We may add that alongside of this not merely no citizens of the freed Koivbv of the Achaeans there subsisted communities, such as Athens and a still narrower one of .the district of Sparta, but also none of those belong- Achaia in the proper sense, whose ■ing to the confederation of the Boeo- representatives met in Aegium (Paus- tians and allies (p. 259). Perhaps anias, vii. 24, 4), just as the Kowbv this Koivbv was legally restricted to twv 'ApK&dui> (Arch. Zeit. 1879, p. the territory, which the Romans called 139, n. 274), and numerous others, the republic of Achaia — that is, that If, according to Pausanias,- v. 12, 6, of the Achaean league at its over- 0! iravres "EWtjvcs set up statues in throw — and the Boeotians and allies Olympia to Trajan, and al it rb' Axa^bv chap. vii. GREEK EUROPE. 265 was the Achaean league, dissolved after the Achaean war, reconstituted, but also the enlarged Boeotian union for- merly mentioned (p. 259) was engrafted on it. Probably it was just by the laying together of these two domains that the demarcation of the province of Achaia was brought about. The new union of the Achaeans, Boeo- tians, Locrians, Phocians, Dorians, and Euboeans, 1 or, as it is usually designated like the province, the union of the Achaeans, presumably had rights neither more nor less than the other provincial diets of the empire. A certain control of the Roman officials must have been intended in the case, and for that reason the towns not placed under the proconsul, like Athens and Sparta, must have been excluded from it. This diet withal, like all similar ones, must have found the centre of its activity chiefly in the common cultus embracing the whole land. But, while in the other provinces this cultus of the land pre- ponderantly attached itself to Rome, the diet of Achaia was rather a focus of Hellenism, and was perhaps meant to be so. Already under the Julian emperors it regarded itself as the true representative of the Greek nation, and assigned to its president the name of Helladarch, to itself even that of " the Panhellenes." 2 The assembly thus rekowai. 7r6\eis to Hadrian, and no where the union especially set up its misunderstanding has here crept in, memorials, it names itself for the the latter dedication must have taken most part no doubt rh Koivbv twv place at the diet of Aegium. 'A^cuSe, but shows often enough the 1 So (only that the Dorians are same tendency ; e.g. when t6 Koivbv wanting ; comp. p. 259, note 2) the tGiv 'Axcttwy II. M\iov 'Apiaruva . . . union is termed on the inscription ffivira.vresoV'^tKkr]vesd,v^ff77i(7av{Arch. of Acraephia (Keil, Syll. Inscr. Boeot. Zeit. 1880, p. 86, n. 344). So too a. 31). But this very document, in Sparta, oi "EXXijxes set up a statue along with the contemporary one, to Caesar Marcus dmi toO koivov t&v C. I. Gr. 1625, furnishes a proof 'A^aiSe (C. /. Gr. 1318). that the union under the emperor 2 In Asia, Bithynia, lower Moesia, Gaius, instead of this doubtless strictly the president of the Greek towns be- official appellation, designated itself longing to the province is also called also on the one hand as union of the 'EXXaSctpx';', without more being Achaeans, on the other as rb Kowbv thereby expressed than the contrast rSc TLaveKKfywv, or ^ trtWSos r&v with the non-Greeks. But, as the 'EXXijx&n', also tA tup 'Axeufflc ko.1 name of Hellenes is employed in TlaveW-fivav cwtSpiov. This gran- Greece in a certain contrast to the diloquence is nowhere so glaringly strictly correct one of Achaeans, this prominent as in those Boeotian petty is certainly suggested by the same country-towns ; but even in Olympia, tendency which was most clearly 266 GREEK EUROPE. BOOK VIII. The Pan- hellenion deviated- from its provincial basis, and its modest ad- ministrative functions fell into the background. These Panhellenes therefore took to themselves this ^Hadrian name b y an abuse of language, and were simply tolerated in Athens, by the government. But as Hadrian created a new Athens, so he created also a new Hellas. Under him the representatives of all the autonomous or non-autono- mous towns of the province of Achaia were allowed to constitute themselves in Athens as united Greece, as the Panhellenes. 1 The national union, often dreamed of and never attained in better times, was thereby created, and what youth had wished for old age possessed in imperial fulness. It is true that the new Panhellenion did not obtain political prerogatives ; but there was no lack of what imperial favour and imperial gold could give. There arose in Athens the temple of the new Zeus Panhellenios, and brilliant popular festivals and games were connected with this foundation, the carrying out of which pertained to the collegium of the Panhellenes, and primarily to the priest of Hadrian as the living god who founded them. One of the acts, which these performed every year, was the offering of sacrifice to Zeus the Deliverer at Plataeae, in memory of the Hellenes that fell there in marked in the Panhellenes of Argos. Thus we find arpariyybs tov kolvov tlov 'Axlillov Kal irpoGT&Tqs dta fiLov tlov "EXKfywv [Arch. Zeit. 1877, p. 192, n. 98), or on another document of the same man irpotrT&TTjs 6ll\ §iov tov kolvov tlov 'Ax°-llov (Lebas-Foucart, n * 3°5) » an &p&s tols "EWtjctiv o~6v- waaiv (Arch. Zeit. p. 195, n. 106) arpaTijybs aavv /cptrois fip£as ttjs "E\- XdSos (ill- 1877, p. 40, n. 42) a-Tpa- TTiybs KaVEKKadiipxvs (»'*• J 876, n. 8, p. 226), all likewise on inscriptions of the kolvov tlov 'Axclllov. That in this kolvov, though it may perhaps be deemed to refer merely to the Pelo- ponnesus (p. 264, note), the Panhel- lenic tendency none the less asserted itself, may well be conceived. 1 The Hadrianic Panhellenes name themselves to KOLvbv crvviSpiov tlov 'EXXtJpow tlov els IIXaT^cts (tvvl6vtlov (Thebes : Keil, Sytt. Inscr. Boeot. n. 31, comp. Plutarch, Arist. 19, 21); KOLvbv ttjs "EWddos (C. I. Gr. 5852); Tb UaveWfyLov (ib.). Its president is termed 6 &pxo>v twv HavtWfyuv (C. I. A. iii. 681, 682; C. I. Gr. 3832, comp. C. I. A. iii. 10 : cl[vt]- ipXLOv tov Upur&Tov 6,[yavos tov U.]av[eX\\ijvlov), the individual deputy H.ave'Wriv [e.g. C. I. A. iii. 534; C. I. Gr. 1 124). Alongside of these in the period subsequent to Hadrian the Kotvbv tlov 'A-xllllov and its o-Tparrrybs or 'EXXaffdpxijs still occur, who are probably to be distinguished from those just mentioned, although the latter now sets up his honorary decrees not merely in Olympia, but also in Athens (C. I. A. 18 ; second example in Olympia, Arch. Zeit. 1879. P- 52). chap. vii. GREEK EUROPE. 267 battle against the Persians, on the anniversary of the battle, the 4th Boedromion : this marks its tendency. 1 Still more clearly was this shown in the fact that the Greek towns outside of Hellas, which appeared worthy of the national fellowship, had ideal certificates of Hellenism issued to them by the assembly in Athens. 2 While the imperial rule in its whole wide range The decay encountered the devastations of a twenty years' civil ofHellas - war, and in many places its consequences were never entirely healed, probably no domain was so severely affected by them as the Greek peninsula. Fate had so arranged, that the three great decisive battles of this epoch — Pharsalus, Philippi, Actium — were fought on its soil or on its coast ; and the military operations, which with both parties led up to these battles, had here above all demanded their sacrifices of human life and human happiness. Even Plutarch was told by his great-grand- father how the officers of Antonius had compelled the citizens of Chaeronea, when they no longer possessed slaves or beasts of burden, to drag their last grain on their own shoulders to the nearest port to be shipped for the army ; and how thereupon, just as the second convoy was about to depart, the accounts of the battle of Actium arrived as glad news of relief. The first thing that Caesar did after the victory was to distribute the enemy's stores of grain that had fallen into his power among the famishing population of Greece. This heaviest measure 1 That the remark of Dio of Prusa, Cibyra in Phrygia (C. I. Gr. 5882), Or. xxxviii. p. 148 R., as to the issued from the Kowhv rijs 'EXXdSos dispute of the Athenians and the by a Sbyfm tov Tlaj>eX\Tj!>lav ; and for Lacedaemonians inrtp Tys irpo- Magnesia on the Maeander ( C. I. Att. wopwdas, refers to the festival at iii. 16). In both the good Hellenic Plataeae, is evident from (Lucian) descent of the corporations concerned EpuTes 18, is irepl irpoironTrelas is brought out along with their other d7wxioi5/iei'cu HKaraiao-iv. The sophist services to the Hellenes. Charac- Irenaeus also wrote irepl ttjs 'ABvjvalav teristic are also the letters of recom- Trpoirop.Tretas (Suidas, s. v.), and Her- mendation, with which these Pan- mogenes, de ideis, ii. p. 373. Walz hellenes furnish a man who had gives as the topic spoken of 'kBfjvaioi merited well of their commonwealth KaX AaxedaifidvLOL irepl t^s irpo7rv elo-e\aT}s iri- fipaid-qs ipiflelov re 7rA«y, ivri t 'Avo.k- e'lvaro JHik6tto\u>, Oetrjv ir6\iv ' &vn Toplav, Se vIktis "Apyeos ' Ap,(pi\6xov re, iciu binrbaa tpoipos iLva£ ratfnje Sixvvrai 'Ak- paitraro wbtckip ndSos. Anthol. Gr. ix. 553. 296 GREEK EUROPE. book vm. the Macedonian towns Thessalonica and Cassandreia, Demetrius Poliorcetes the Thessalian town Demetrias, and Lysimachus the town of Lysimachia on the Thracian Chersonese out of a number of surrounding townships divested of their independence. In keeping with the Greek character of the foundation Nicopolis was, according to the intention of its founder, to become a Greek city on a great scale. 1 It obtained freedom and autonomy like Athens and Sparta, and was intended, as already stated, to wield the fifth part of the votes in the Amphictiony representing all Hellas, and to do so, like Athens, without alternating with other towns (p. 254). This new Actian shrine of Apollo was erected quite after the model of Olympia, with a quadriennial festival, which even bore the name of " Olympia " alongside of its own, had equal rank and equal privileges, and even its Actiads as the former had its Olympiads ; 2 the town of Nicopolis stood related to it like the town of Elis to the Olympian temple. 3 Everything properly Italian was carefully avoided in the erection of the town as well as in the religious arrangements, however natural it might be to mould after the Roman fashion the " city of victory " so intimately associated with the founding of the empire. Whoever considers the arrangements of Augustus in 1 When Tacitus, Ann. v. 10. names Nicopolis a colonia Romana. . the statement is one liable to be mis understood, but not exactly incorrect but that of Pliny (ff. AT. iv. i, 5) colonia Augusti Actium cum . . civitate libera Nicopolitana, is errone (apxala). As competitive games are frequently called lv 'Q\ip.inos ra "Aktux, ij ir&Kis 'HXeiuv Kal ^ 'OXvfiinicil /3ou\l} Strabo, vii. 7, 6, p. 325 ; 'Akth£s, (Arch. Zeit. 1876, p. 57 ; similarly Josephus, Bell. Jud. i. 20, 4; ibid. 1877, pp. 40, 41 and [elsewhere). ' AKTiovUrji oftener. As the four Moreover the Spartans, as the only great Greek national festivals are, as Hellenes that took part in the victory is well known, termed t) ireplodos, at Actium, obtained the conduct and the victor crowned in all four (im/ii£keia) of the Actian games TepioSovticris, so in C. I. Gr. 4472 (Strabo, vii. 7, 6 p. 325) : their t?js irepibSov is appended also to the relation to the |8ouX'J) 'Akticuctj of games of Nicopolis, and the former Nicopolis we do not know. irepi68os is designated as the ancient chap. vii. GREEK EUROPE. 297 Hellas in this connection, and especially this remarkable corner-stone, will not be able to resist the conviction that Augustus believed that a reorganisation of Hellas under the protection of the Roman principate was practicable, and wished to carry it out. The locality at least was well chosen for it, as at that time, before the foundation of Patrae, there was no larger city on the whole Greek west coast. But what Augustus may have hoped for at the commencement of his sole rule, he did not attain, and perhaps even subsequently abandoned, when he gave to Patrae the form of a Roman colony. Nicopolis remained, as the extensive ruins and the numerous coins show, comparatively populous and nourishing j 1 but its citizens do not appear to have taken a prominent part in commerce and manufactures or otherwise. Northern Epirus, which, like the adjoining Illyricum bordering on Macedonia, was in greater part inhabited by Albanian tribes and was not placed under Nicopolis, continued during the imperial period in its primitive condition, which still subsists in some measure at the present day. " Epirus and Illyricum," says Strabo, " are in great part a desert ; where men are found, they dwell in villages and in ruins of earlier towns ; even the oracle of Dodona," — laid waste in the Mithradatic war by the Thracians (iii. 3 1 2), — " is extinct like everything in. 296. else." 2 Thessaly, in itself a purely Hellenic district as well Thessaiy. as Aetolia and Acarnania, was in the imperial period separated administratively from the province of Achaia 1 The description of its decay in Hadrian, who is named Zeis Aw&wtuos the time of Constantius (Paneg. 11,9) (C. /. Gr. 1822), visited Dodona is an evidence to the Opposite effect (Diirr, Reisen Hadrians, p. 56) he for the earlier times of the empire. did so as an archaeologist. A con- 2 The excavations at Dodona sultation of the oracle during the have confirmed this ; all the articles imperial period is only reported — found belong to the pre- Roman period and that not after the most trust- except some coins. Certainly a restor- worthy manner — in the case of the ation of the building took place, the emperor Julian (Theodoretus, Hist. time of which cannot be determined ; Ecd. iii. 21). perhaps it was quite late. When 298 GREEK EUROPE. book viii. and placed under the governor of Macedonia. What holds true of northern Greece applies also to Thessaly. The freedom and autonomy which Caesar had allowed generally to the Thessalians, or rather had not with- drawn from them, seem to have been withdrawn, on account of misuse, from them by Augustus, so that subsequently Pharsalus alone retained this legal position j 1 Roman colonists were not settled in the district. It retained its separate diet in Larisa, and civic self-adminis- tration was left with the Thessalians, as with the dependent Greeks in Achaia. Thessaly was far the most fertile region of the whole peninsula, and still exported grain in the fourth century ; nevertheless Dio of Prusa says that even the Peneus flows through waste land ; and in the imperial period money was coined in this region only to a very small extent. Hadrian and Diocletian exerted themselves to restore the roads of the country, but they alone, so far as we see, of the Roman emperors did so. Macedonia. Macedonia, as a Roman administrative district under the empire, was materially curtailed as compared with the Macedonia of the republic. Certainly, like the latter, it reached from sea to' sea, inasmuch as the coast as well of the Aegean Sea from the region of Thessaly belonging to Macedonia as far as the mouth of the Nestus (Mesta), as of the Adriatic from the Aous 2 as far as the Drilon (Drin), was reckoned to this district ; the latter territory, not properly Macedonian but Illyrian 1 The ordinance of Caesar is at- and so the Thessalians were brought tested by Appian, B. C. ii. 88, and before the tribunal of the emperor Plutarch, Caes. 48, and it very well (Suetonius, Tib. 8). Presumably the accords with his own account, B. C. two incidents and likewise the loss iii. 80 ; whereas Pliny, H. N. iv. 8, of freedom stand connected. 29, names only Pharsalus as a free 2 In the time of the republic Scodra town. In Augustus' time a Thessalian seems to have belonged to Macedonia of note, Petraeos (probably the partisan (iii. 181); in the imperial period this '"• J 73- of Caesar, B. C. iii. 35), was burnt and Lissus are Dalmatian towns, and alive (Plutarch, Praec. ger. reip. 19), the mouth of the Drin forms the doubtless not by a private crime, but boundary on the west, according to resolution of the diet, chap. vii. GREEK EUROPE. 299 land, but already in the republican period assigned to the governor of Macedonia (iii. 44), remained with the m, 42. province also during the time of the empire. But we have already stated that Greece south of Oeta was separated from it. The northern frontier towards Moesia and the east frontier towards Thrace remained indeed in so far unaltered, as the province in the imperial period reached as far as the Macedonia proper of the republic had reached, viz. on the north almost as far as the vale of the Erigon, eastward as far as the river Nestus ; but while in the time of the republic the Dardani and the Thracians, and all the tribes of the north and north-east adjoining the Macedonian territory, had to do with this governor in their circumstances of peaceful or warlike contact, and in so far it could be said that the Macedonian boundary reached as far as the Roman lances, the Macedonian governor of the imperial period bore sway only over the district assigned to him, which no longer bordered- on neighbours half or wholly independent. As the defence of the frontier was transferred in the first instance to the kingdom of the Thracians which had come under allegiance to Rome, and soon to the governor of the new province Moesia, the governor of Macedonia was from the outset relieved of his command. There was hardly any fighting on Macedonian soil under the empire ; only the barbarian Dardani on the upper Axius (Vardar) still at times pillaged the peaceful neighbouring province. There is no report, moreover, from this province of any local revolts. From the more southerly Greek districts this — the Nation- most northerly — stood aloof as well in its national basis aities - as in the stage of its civilisation. While the Macedonians proper on the lower course of the Haliacmon (Vistritza) and the Axius (Vardar), as far as the Strymon, were an originally Greek stock, whose diversity from the more southern Hellenes had no further significance for the present epoch, and while the Hellenic colonisation embraced within its sphere both coasts — on the west with Apollonia and Dyrrachium, on the east in particular 300 GREEK EUROPE. book viii. with the townships of the Chalcidian peninsula — the interior of the province, on the other hand, was filled with a confused mass of non-Greek peoples, which must have differed from the present state of things in the same region more as to elements than as to results. After the Celts who had pushed forward into this region, the Scordisci, had been driven back by the generals of the Roman republic, the interior of Macedonia fell to the share especially of Illyrian stocks in the west and north, of Thracian in the east. Of both we have already spoken previously ; here they come into consideration only so far as the Greek organisation, at least the urban, was probably introduced — as in the earlier, 1 so also in the imperial period — among these stocks only in a very limited measure. On the whole, an energetic impulse of urban development never pervaded the interior oi Macedonia ; the more remote districts hardly reached — at least as to substance — beyond the village-system. Greek The Greek polity itself was not a spontaneous growth polity. j n thjg country, obeying a king, as it was in Hellas proper, but was introduced by the princes, who were more Hellenes than their subjects. What shape it had is little known ; yet the civic presidency of politarchs uniformly recurring in Thessalonica, Edessa, Lete, and not met with elsewhere, leads us to infer a perceptible, and indeed in itself prob- able, diversity of the Macedonian urban constitution from that elsewhere usual in Hellas. The Greek cities, which the Romans found existing, retained their organisation and their rights ; Thessalonica, the most considerable of them, also freedom and autonomy. There existed a league and a diet (koivov) of the Macedonian towns, similar to those in Achaia and Thessaly. It deserves mention, as an evi- dence of the continued working of the memories of the 1 The towns founded in these tinctively Macedonian politarchs have regions outside of Macedonia proper epigraphic attestation (inscription of bear quite the character of colonies the year 197 A.D., t&i> irepl 'AX#f- proper ; e.g. that of Philippi in the avSpov QCklirirov iv hepploirig iroKi.- Thracian land, and especially that of Tapx&v, Duchesne and Bayet, Mission Derriopus in Paeonia (Liv. xxxix. 53), au mont Aihos, p. 103). for which latter place also the dis- chap. vii. GREEK EUROPE. 301 old and great times, that still in the middle of the third century after Christ the diet of Macedonia and individual Macedonian towns issued coins on which, in place of the head and the name of the reigning emperor, came those of Alexander the Great. The pretty numerous colonies of Roman burgesses which Augustus established in Mace- donia, Byllis not far from Apollonia, Dyrrachium on the Adriatic, on the other coast Dium, Pella, Cassandreia, in the region of Thrace proper Philippi, were all of them older Greek towns, which obtained merely a number of new burgesses and a different legal position, and were called into life primarily by the need of providing quarters in a civilised and not greatly populous province for Italian soldiers who had served their time, and for whom there was no longer room in Italy itself. The granting of Italian rights certainly took place only to gild for the veterans their settlement abroad. That it was never intended to draw Macedonia into the development of Italian culture is evinced, apart from all else, by the fact that Thessalonica remained Greek and the capital of the country. By its side flourished Philippi, properly a mining town, constituted on account of the neighbouring gold mines, favoured by the emperors as the seat of the battle which definitively founded the monarchy, and on account of the numerous veterans who took part in it and subse- quently settled there. A Roman, not colonial, municipal constitution was obtained already in the first period of the empire by Stobi, the already mentioned most northerly frontier-town of Macedonia towards Moesia, at the conflu- ence of the Erigon with the Axius, in a commercial as in a military point of view an important position, and which, it may be conjectured, had already in the Macedonian time attained to Greek polity. In an economic point of view little was done on the Economy. part of the state for Macedonia under the emperors ; at ^ oads and least there is no appearance of any special care on their part for this province, which was not put under their own administration. The military road already constructed under the republic right across the country from 302 GREEK EUROPE. book viii. Dyrrachium to Thessalonica, one of the most important arteries of intercourse in the whole empire, called forth re- newed effort, so far as we know, only from emperors of the third century, and first from Severus Antoninus ; the towns adjacent to it, Lychnidus on the Ochrida-lake and Heraclea Lyncestis (Bitolia), were never of much account. Yet Macedonia was, economically, better situated than Greece. It far excelled it in fertility ; as still at present the province of Thessalonica is relatively well cultivated and well peopled, so in the description of the empire from the time of Constantius, at all events when Constan- tinople was already in existence, Macedonia is reckoned among the specially wealthy districts. If for Achaia and Thessaly our documents concerning the Roman levy are absolutely silent, Macedonia on the other hand was drawn upon, in particular for the imperial guard, to a considerable extent, more strongly than the most of the Greek dis- tricts — on which, no doubt, the familiarity of the Mace- donians with regular war -service and their excellent qualifications for it, and probably also the relatively small development of the urban system in this province, had an important bearing. Thessalonica, the metropolis of the province, and its most populous and most industrial town at this time, represented likewise under various forms in literature, has also secured to itself an honourable place in political history by the brave resistance which its citizens opposed to the barbarians in the terrible times of the Gothic invasions (p. 248). Thrace. If Macedonia was a half-Greek, Thrace was a non- Greek land. Of the great but for us vanished Thracian stock we have formerly (p. 207) spoken. Into its domain Hellenism came simply from without ; and it will not be superfluous in the first instance to glance back and to set forth how often Hellenism had previously knocked at the gates of the most southerly region which this stock pos- sessed, and which we still name after it, and how little it chap. vii. GREEK EUROPE. 303 had hitherto penetrated into the interior, in order to make clear what was left for Rome here to overtake and what it did overtake. Philip, the father of Alexander, first Philip and subjected Thrace, and founded not merely Calybe in the Alexander - neighbourhood of Byzantium, but also in the heart of the land the town which thenceforth bore his name. Alex- ander, here too the precursor of Roman policy, arrived at and crossed the Danube, and made this stream the northern boundary of his empire ; the Thracians in his army played by no means the least part in the subjugation of Asia. After his death the Hellespont seemed as though it would become one of the great centres of the new formation of states, and the wide domain from thence to the Danube 1 as though it would become the northern half of a Greek empire, and would promise for the residence of Lysimachus, the former governor of Thrace — the town of Lysimachia, Lysima- newly established in the Thracian Chersonese — a like chus ' future as for the residences of the marshals of Syria and Egypt. But this result was not attained ; the independ- ence of this kingdom did not survive the fall of its first ruler (281 B.C., 473 U.C.). In the century which elapsed from that time to the establishment of the ascendancy of Rome in the East, attempts were made, sometimes by the Seleucids, sometimes by the Ptolemies, sometimes by the Attalids, to bring the European possessions of Lysimachus under their power, but all of them without lasting result. The empire of Tylis in the Haemus, which the Celts not Empire of long after the death of Alexander, and nearly at the same y IS ' time with their permanent settlement in Asia Minor, had founded in the Moeso-Thracian territory, destroyed the seed of Greek civilisation within its sphere, and itself succumbed during the Hannibalic war to the assaults of the Thracians, who extirpated these intruders to the last man. Thenceforth there was not in Thrace any leading power at all ; the relations subsisting between the Greek coast-towns and the princes of the several tribes, which would probably correspond approximately to those before 1 That for Lysimachus the Danube was the boundary of the empire, is evident from Pausanias, i. 9, 6. 3°4 GREEK EUROPE. BOOK VIII. Later Mace- donian rulers. Alexander's time, are illustrated by the description which Polybius gives of the most important of these towns : "Where the Byzantines had sowed, there the Thracian barbarians reaped, and against these neither the sword nor money is of avail ; if the citizens kill one of the princes, three others thereupon invade their territory, and, if they buy off one, five more demand the like annual payment." The efforts on the part of the later Macedonian rulers to gain once more a firm footing in Thrace, and in par- ticular to bring under their power the Greek towns of the south coast, were opposed by the Romans, partly in order to keep down the development of Macedonia's power gener- ally, partly in order not to allow the important " royal road " leading to the East — that along which Xerxes marched to Greece and the Scipios marched against Antiochus — to fall in all its extent into Macedonian hands. Already, after the battle at Cynoscephalae, the frontier -line was drawn nearly such as it thenceforth remained. The two last Macedonian rulers made several attempts, either directly to establish themselves in Thrace or to attach to themselves its individual rulers by treaties ; the last Philip even gained over Philippopolis once more, and put into it a garrison, which, it is true, the Odrysae soon drove out afresh. Neither he nor his son succeeded in placing matters on a permanent footing ; and the in- dependence conceded by Rome to the Thracians after the breaking up of Macedonia destroyed whatever Hellenic germs might still be left there. Thrace itself became — in part already in the republican, and more decidedly in the imperial period — a Roman vassal -principality, and then in 46 a Roman province (p. 211); but the Hellenis- ing of the land had not passed beyond the fringe of Greek colonial towns, which in the earliest period had been established round this coast, and in course of time had sunk rather than risen. Powerful and permanent as was the hold of Macedonian civilisation on the East, as weak and perishable was its contact with Thrace ; Philip and Alexander themselves appear to have reluctantly under- chap. vii. GREEK EUROPE. 305 taken, and to have but lightly valued, their settlements in this land. 1 Till far into the imperial period the land remained with the natives ; the Greek towns that were still left along the coast, almost all on the decline, remained without any Greek land in their rear. This belt of Hellenic towns stretching from the Greek Macedonian frontier to the Tauric Chersonese was of very Thrace' and unequal texture. In the south it was close and compact °n the from Abdera onward to Byzantium on the Dardanelles ; yet ac ea " none of these towns held a prominent position in later times with the exception of Byzantium, which through the fertility of its territory, its productive tunny fisheries, its uncommonly favourable position for trade, its industrial diligence, and the energy of its citizens — heightened merely and hardened by its exposed situation — was enabled to defy even the worst times of Hellenic anarchy. Far more scantily had the settlements developed them- selves on the west coast of the Black Sea ; among those subsequently belonging to the Roman province of Thrace Mesembria alone was of some importance ; among those subsequently Moesian Odessus (Varna) and Tomis (Kiis- tendje). Beyond the mouths of the Danube and the boundary of the Roman empire, on the northern shore of the Pontus, there lay amidst the barbarian land Tyra 2 and Olbia ; further on, the old and great Greek mercantile cities in what is now the Crimea — Heraclea or Chersonesus and Panticapaeum — formed a stately copestone. All these settlements enjoyed Roman protection, after Under the Romans had become generally the leading power on pr °™c n0 n. the Graeco-Asiatic continent ; and the strong arm, which often came down heavily on the Hellenic land proper, prevented here at least disasters like the destruction of Lysimachia. The protection of these Greeks devolved in 1 Calybe near Byzantium arose ac- colonists corresponding with that de- cording to Strabo (vii. 6, 2, p. 320) scription. However little these re- ipCklinrov rod 'A/iivrov toi)s irov-qpo- ports deserve trust, they yet in their t&tovs ivTavda ISpiaavTos. Philip- coincidence express the Botany-Bay popolis is alleged even according to character of these foundations, the account of Theopompus (fr. 122 2 Yet the northern Bessarabian Miiller) to have been founded as line, which perhaps is Roman, reaches UovripdwoXis, and to - have received as far as Tyra (p. 226). VOL. I. 20 3°6 GREEK EUROPE. BOOK VIII. the republican period partly on the governor of Mace- donia, partly on the governor of Bithynia, after this became Roman ; Byzantium subsequently remained with Bithynia. 1 We may add that in the imperial period, after the erection of the governorship of Moesia and subsequently of that of Thrace, the supplying of protection devolved on these. Protection and favour were granted by Rome to these Greeks from the first ; but neither the republic nor the earlier imperial period made efforts for the extension of Hellenism. 2 After Thrace had become Roman, it was divided into land-districts ; 3 and almost down to the end of the first century there is no record of the laying out of a town there, with the exception of two colonies of Clau- dius and Vespasian — Apri in the interior not far from 1 That Byzantium was still in Trajan's time under the governor of Bithynia, follows from Plin. ad Trai. 43. From the congratulations of the Byzantines to the legates of Moesia we cannot infer their having belonged to this governorship, which from their situation was hardly pos- sible ; the relations to the governor of Moesia may be explained from the commercial connections of the city with the Moesian ports. That Byzan- tium was in the year 53 under the senate, and so did not belong to Thrace, is plain from Tacitus, Ann. xii. 62. Cicero (in Pis. 35, 86 ; de prov. cons. 4, 6) does not attest its having belonged to Macedonia under the republic, since the town was then free. This freedom seems, as in the case of Rhodes, to have been often given and often taken away. Cicero, I.e., ascribes freedom to it ; in the year 53 it is tributary ; Pliny (H. JV. iv. 1 1, 46) adduces it as a free city ; Vespasian withdraws its freedom (Sue- tonius, Vesp. 8). 2 This is proved by the absence of coins of the Thracian inland towns, which could be assigned in metal and style to the older period. That a number of Thracian, especially Odry- sian, princes .coined in part even at a very early period, proves only that they ruled over places on the coast with a Greek or half-Greek popula- tion. A similar judgment must be formed as to the tetradrachmae of the "Thracians," which stand quite isolated (Sallet, Num. Zeitschrift, iii. 241). — The inscriptions also found in the interior of Thrace are throughout of Roman times. The decree of a town not named found at Bessapara, now Tatar Bazarjik, to the west of Philippopolis, by Dumont (Inscr. de la Thrace, p. 7), is indeed assigned to a good Macedonian time, but only from the character of the writing, which is perhaps deceptive. 3 The fifty strategies of Thrace (Plin. H. N. iv. 11, 40; Ptolem. iii. 11, 6) are not military districts, but, as is apparent with special clearness in Ptolemy, land-districts, which corre- spond with the tribes (arpaniyla McuSirf, Beo-ffiKi; k. t. X.) and form a contrast to the towns. The designa- tion (TTparrp/bs has, just like praetor, lost subsequently its original military value. Here perhaps the analogy of Egypt, which likewise was divided into urban domains under urban ma- gistrates and into land-districts under strategoi, served primarily as a basis. A CTpaT-qyis 'Actti/ojs irepl TUpwOov from the Roman period occurs in Eph. epigr. ii. p. 252. chap. vii. GREEK EUROPE. 307 Perinthus, and Deultus on the most northern coast. 1 Domitian began by introducing the Greek urban constitu- PhiUppo- tion into the interior, at first for the capital of the country, 0°^. towns Philippopolis. Under Trajan a series of other Thracian with civic townships obtained like civic rights ; Topirus not far nght3, from Abdera, Nicopolis on the Nestus, Plotinopolis on the Hebrus, Pautalia near Kostendil, Serdica now Sofia, Augusta Traiana near Alt-Zagora, a second Nicopolis on the northern slope of Haemus, 2 besides, on the coast, Trai- anopolis at the mouth of the Hebrus ; further, under Hadrian Adrianopolis, the modern Adrianople. All these towns were not colonies of foreigners but polities of Greek organisation, composed after the model set up by Augustus in the Epirot Nicopolis ; it was a civilising and Hellenis- ing of the province from above downwards. A Thracian diet existed thenceforth in Philippopolis just as in the properly Greek provinces. This last offshoot of Hellenism was not the weakest. The country was rich and charm- ing — a coin of the town Pautalia praises the fourfold blessing of the ears of grain, of the grapes, of the silver, and of the gold ; and Philippopolis as well as the beautiful valley of the Tundja were the home of rose-culture and of rose-oil — and the vigour of the Thracian type was not broken. Here was developed a dense and prosperous population ; we have already mentioned the largeness of the levy in Thrace, and few territories stand on an equality with Thrace at this epoch in the activity of the urban mints. When Philippopolis succumbed in the year 251 1 In Deultus, the coloniaFlavia Paris since Severus also administratively; but Deultensium, veterans of the eighth not merely does Ptolemy adduce it in legion, were provided for (C. /. L. vi. Thrace, but the places where the Had- 3828). Flaviopolis on the Chersonese, rianic terminal stones (C. I. L. iii. 749, the old Coela, was certainly not a comp. p. 992) are found, appear to colony (Plin. iv. 1 1, 47), but belongs assign it likewise to Thrace. As this to the peculiar settlement of the im- Greek inland town fitted neither the perial menials on this domanial pos- Latin town - communities of lower session (JEph. epigr. v. p. 83). Moesia nor the kow6v of the Moesian 2 This town Nik67toXis ri irepl Atfwv Pontus, it was assigned at the first of Ptolem. iii. II, 7, HuciiroXis irpis organising of the relations to the koo>6v "Icrrpov of the coins, the modern Ni- of the Thracians. Subsequently it kup on the Jantra, belongs to lower must, no doubt, have been attached Moesia geographically, and, as the to one or the other of those Moesian names of governors on the coins show, groups. 3o8 GREEK EUROPE. BOOK VIII. to the Goths (p. 240), it is said to have numbered 100,000 inhabitants. The energetic part taken by the Byzantines in favour of the emperor of the Greek East, Pescennius Niger, and the several years' resistance which the town even after his defeat opposed to the victor, show the resources and the courage of these Thracian towns- men. If the Byzantines here, too, succumbed and lost even for a season their civic rights, the time, for which the rise of the Thracian land paved the way, was soon to set in, when Byzantium should become the new Hellenic Rome and the chief residence of the remodelled empire. In the neighbouring province of lower Moesia a similar development took place, although on a smaller scale. The Greek coast-towns, the metropolis of which, at least in the Roman period, was Tomis, were, probably on the consti- tuting of the Roman province of Moesia, grouped as the " Five-cities-league of the left shore of the Black Sea," or as it was also called, '' of the Greeks," that is, the Greeks of Tomis and this province. Later there was annexed to this league, as Pentapoiis. a sixth town, that of Marcianopolis, constructed by Trajan not far from the coast on the Thracian frontier, and organised, like the Thracian towns, after the Greek model. 1 Lower Moesia. 1 The KOivbv Tijs TLevrairbXebis is found on an inscription of Odessus, C. I. Gr. 2056 c, which may fairly belong to the earlier imperial period, the Pontic Hexapolis, on two inscrip- tions of Tomis probably of the second century A.D. (Marquardt, Staatsverw. i. 2 p. 305 ; Hirschfeld, Arch, epigr. Mitth. vi. 22). The Hexapolis in any case, and in accordance there- with probably also the Pentapoiis, must have been brought into harmony with the Roman provincial boundaries, that is, must have included in it the Greek towns of lower Moesia. These are also found, if we follow the surest guides, — the coins of the imperial period. There were six mints (apart from Nicopolis, p. 282, note) in lower Moesia : Istros, Tomis, Callatis, Dio- nysopolis, Odessus, and Marcian- opolis, and, as the last town was founded by Trajan, the Pentapoiis is thereby explained. Tyra and Olbia hardly belonged to it ; at least the numerous and loquacious monuments of the latter town nowhere show any link of connection with this city -league. It is called Koivbv twv 'EXXfyw on an inscription of Tomis, printed in the Athenian Pandora of 1st June 1868 [and in Anc. Gr. Inscr. in the British Museum, ii. n. 175] : 'AyaBrj rixv- Kara rd S6%avra t9) Kpa- 7TJTa, AXXct Kal fiovXevrty Kal tu>v irpojrevbvTtav Xafllas HVas TriXcus, Kal rip ap\i4peiav atip.fsiov airrov 'lovXlav ' 'ATroXa&mjc Trdcrris Teip-Tjs x&P ir - chap. vii. GREEK EUROPE. 309 We have already observed that the camp-towns on the bank of the Danube, and generally the townships called into life by Rome in the interior, were instituted after the Italian model ; lower Moesia was the only Roman province intersected by the linguistic boundary, inasmuch as the Tomitanian cities-league belonged to the Greek, the Danubian towns, like Durostorum and Oescus to the Latin, linguistic domain. In other respects essen- tially the same holds true of this Moesian cities-league, as was remarked regarding Thrace. We have a description of Tomis from the last years of Augustus, doubtless by one banished thither for punishment, but certainly true in substance. The population consists for the greater part of Getae and Sarmatae ; they wear, like the Dacians on Trajan's column, skins and trousers, long waving hair and unshorn beard, and appear in the street on horseback and armed with the bow, with the quiver on their shoulder, and the knife in their girdle. The few Greeks who are found among them have adopted the barbarian customs, including the trousers, and are able to express themselves as well or better in Getic than in Greek ; he is lost, who cannot make himself intelligible in Getic, and no man understands a word of Latin. Before the gates rove predatory bands of the most various peoples, and their arrows not seldom fly over the protecting city-walls ; he who ventures to till his field does it at the peril of his life, and ploughs in armour — at anyrate about the time of Caesar's dictatorship ; on occasion of the raid of Burebista, the town had fallen into the hands of the barbarians, and a few years before that exile came to Tomis, during the Dalmato-Pannonian insurrection, the fury of war had once more raged over this region. The coins and the inscrip- tions of that city accord well with these accounts, in so far as the metropolis of the " left- Pontic cities-league " in the pre-Roman period coined no silver, which several other of these towns did ; and, in general, coins and inscriptions from the time before Trajan occur only in an isolated way. But in the second and third centuries it was remodelled and may be termed a foundation of Trajan with very 3 1 ° GREEK E UROPE. book viii. much the same warrant as Marcianopolis, which likewise quickly attained to considerable development. The bar- rier formerly mentioned (p. 227) in the Dobrudscha served at the same time as a protecting wall for the town of Tomis. Behind this wall commerce and navigation were flourishing. There was in the town a society of Alexandrian merchants with its own chapel of Serapis j 1 in municipal liberality and municipal ambition the town was inferior to no Greek town of middle size ; it was still even now bilingual, but in such a way that, alongside of the Greek language always retained on the coins, here on the border, where the two languages of the empire came into contact, the Latin is also often employed even in public monuments. Tyra. Beyond the imperial frontier, between the mouths of the Danube and the Crimea, the Greek merchant had made few settlements on the coast ; there were here only two Greek towns of note, both founded in remote times by Miletus, Tyra at the mouth of the river of the same name, the modern Dniester, and Olbia on the bay into which the Borysthenes (Dnieper) and the Hypanis (Bug) fall. The forlorn position of these Hellenes amidst the barbarians pressing around them, in the time of the Diadochi as well as during the earlier rule of the Roman iii. 282. republic, has already been described (iii. 297). The em- perors brought help. In the year 56, and so in the exemplary beginning of Nero's government, Tyra was annexed to the province of Moesia. Of the more remote Olbia. Olbia we possess a description from the age of Trajan ; 2 1 This is shown by the remarkable 2 Olbia, constantly assailed in war inscription in Allard (LaBulgarie ori- and often destroyed, suffered, accord- entale, Paris, 1863, p. 263): 6effl ing to the statement of Dio (Borysth. /ieydKu Sopdir[iSi Kal] tols avvvdots p. 75, n.), about 1 50 years before his Beois [koI ™ ai]TOKpi.Topi T. Al\l\\Q &\vTuvdvu 2e/3acr™ ~Eiaefl- 100 A.D., and so probably in the ex- let] ko.1 M. 'AiprjXlu Oirtipia ~Kalaa.pi. pedition of Burebista (iv. 305), its KapTriav ' Ai>ov§loivos tS o&cu twv 'AXe|- last and most severe conquest (t7)p avBptuv top pai/ibv iic ray ISloiv aviBrjicev reXevralav ko.1 fieylarriv &\wnv). ~Et\ov frous Kf' [jaivos] app,ovdl a iwl Upioiv Si, Dio continues, koX Tairqv Tirai [Kjoppotfrou rod Kal iiapaTluvos [IloXii]- ko.1 to.s &\\as ras iv tois i.piaripoh tov pivov tov ko.1 Aov[yelvov]. The mari- IISvtov iriXeis P-typ 1 'AjroXXowios ner's guild of Tomis meets us several (Sozopolis or Sizebolu, the last Greek times in the inscriptions of the town. town of note on the Pontic west coast); chap. vii. GREEK EUROPE. 31 1 the town was still bleeding from its old wounds ; the wretched walls enclosed equally wretched houses, and the quarter then inhabited filled but a small portion of the old considerable city-circuit, of which individual towers that were left stood far off in the desolate plain ; in the temples there was no statue of the gods which did not bear traces of the hands of the barbarians ; the inhabitants had not forgotten their Hellenic character, but they dressed and fought after the manner of the Scythians, with whom they were daily in conflict. Just as often as by Greek names, they designated themselves by Scythian, i.e. by those of Sarmatian stocks akin to the Iranians j 1 in fact, in the royal house itself Sauromates was a com- mon name. These towns were indebted doubtless for their very continued existence less to their own power than to the good-will or rather the self-interest of the natives. The tribes settled on this coast were neither in a position to carry on foreign trade from emporia of their own, nor could they dispense with it ; in the Hellenic coast -towns they bought salt, articles of clothing and wine, and the more civilised princes protected in some measure the strangers against the attacks of the barbarians proper. The earlier rulers of Rome must have had scruples at undertaking the difficult protection of this remote settlement ; nevertheless Pius, when the Scythians once more besieged them, sent to them Roman auxiliary troops, and compelled the barbarians to offer peace and fur- nish hostages. The town must have been incorporated BBev 5ij koX atpbSpa Taireivh rot irp&y- both, which is suspected in him as a fmra fcarta-H] 7&v Ta&r-rj EMiJpw, token of servile attitude towards the ru>v fikv oiK^Ti trwoiKKrdeiff&v irb\eu>v t Romans. Thus a century later matters twv Sk especially for mercantile intercourse, to be poms? 15 ' left like Olbia and Tyra to the administration of chang- ing municipal officials and a far distant governor; therefore it was entrusted to hereditary princes — a course further recommended by the circumstance that it might not seem advisable to transfer directly to the empire the relations which this region sustained to the surrounding tribes. The rulers of the Bosporan house, in spite of their Achae- menid pedigree and their Achaemenid mode of reckoning time, felt themselves thoroughly as Greek princes, and traced back their origin, after the good Hellenic fashion, to Herakles and the Eumolpids. The dependence of these Greeks on Rome — the royal in Panticapaeum, as the republican in Chersonesus — was implied in the nature of things, and they never thought of rising against the pro- tecting arm of the empire ; if once, under the emperor Claudius, the Roman troops had to march against an in- subordinate prince of the Bosporus, 2 yet withal this 1 In the only vivid narrative from the Bosporan history which we pos- sess, that of Tacitus, Ann. xii. 15-31, concerning the two rival brothers, Mithradates and Cotys, the neigh- bouring tribes, the Dandaridae, Siracae, Aorsi, are under rulers of their ownnot legally dependent on the Roman prince of Panticapaeum. — As to titles,- the older Panticapaean princes are wont to call themselves archons of the Bos- porus, that is, of Panticapaeum, and of Theudosia, and kings of the Sindi and of all the Maitae and other non- Greek tribes. In like manner what is, so far as I know, the oldest among the royal inscriptions of the Roman epoch names Aspurgos, son of Asandrochos (Stephani, Comptes rendus de la comm. four 1866, p. 128), as fiaaCkeiovTO. Travrbs 'Booffirbpov, 0eo8ov /ityas. Be- yond doubt this was the effect of the Roman sovereignty, with which a vas- sal-prince placed over other princes was not very compatible. 2 This was the king Mithradates, installed by Claudius in the year 41, who some years afterwards was de- posed and replaced by his brother Cotys ; he lived afterwards in Rome, and perished in the confusions of the four-emperor-year (Plutarch, Galba, 13, 15). The state of the matter, however, is not clear either from the hints in Tacitus, Ann, xii. 1 5 (comp. Plin. H. N. vi. 5, 17), or from the report (confused by the interchange chap. vii. GREEK EUROPE. 317 region itself, amidst the fearful confusion in the middle of the third century, which especially affected it, never broke away from the empire even when it was falling to pieces. 1 The prosperous merchant-towns, permanently in need of military protection amidst a flux of barbaric peoples, held to Rome as the advanced posts to the main army. The garrison was doubtless chiefly raised in the land itself, and to create and manage it was beyond doubt the main task of the king of the Bosporus. The coins, which were struck on occasion of the investiture of such a king, exhibit doubtless the curule chair and the other honorary presents usual at such investiture, but also by their side shield, helmet, sword, battle-axe, and war-horse; it was no peaceful office which this prince undertook. The first of them, whom Augustus appointed, fell in conflict with the barbarians, and of his successors, e.g. king Sauro- mates, son of Rhoemetalces, fought in the first years of Severus with the Siracae and the Scythians — perhaps it was not quite without reason that he stamped his coins with the feats of Herakles. By sea, too, he had to be active, especially in keeping down the piracy which never ceased in the Black Sea (p. 242) ; that Sauromates like- wise is credited with having brought the Taurians to order and chastised piracy. Roman troops, however, were also of the two, Mithradates of Bosporus, extinction of the family of the Sauro- and Mithradates of Iberia) in Petrus matae, suffice. The civic privileges Patricius/r. 3. The Chersonese tales in arid the localities of the city, for the ex- the late ConstantinusPorphyrogenitus, planation of which these mirabilia are de adm. imp. t. 53, do not, of course, invented, certainly deserve attention, come into account. The bad Bosporan x There are no Bosporan gold or king . Sauromates, ~Kpii>6pov (not pseudo-gold coins without the head of PijcrKOTrApou) vl6$, who with the Sar- the Roman emperor, and this is always matians wages war against the em- that of the ruler recognised by the perors Diocletian and Constantius, Roman senate. That in the years as well as against the Chersonese faith- 263 and 265, when in the empire ful to the empire, has evidently arisen elsewhere after the captivity of Vale- from a confusion of names between rian Gallienus was officially regarded the Bosporan king and people ; and as sole ruler, two heads here appear just as historical as the variation on on the coins, is perhaps due only to the history of David and Goliath, is want of information ; yet the Bos- the despatch of the mighty king of the porans may at that time have made Bosporans, Sauromates, by the small another choice amid the many pre- Chersonesite Pharnaces. The kings' tenders. The names are at this time names alone, e.g. besides those named, not appended, and the effigies are not the Asander, who comes in after the to be certainly distinguished. this vassal- prince, 3 1 8 GREEK E UR OPE. book viii. stationed in the peninsula, perhaps a division of the Pontic fleet, certainly a detachment of the Moesian army ; their presence even in small numbers showed to the barbarians that the dreaded legionary stood behind these Greeks. In another way still the empire protected them ; at least in the later period there were regularly paid from the im- perial chest to the princes of the Bosporus sums of money, of which they stood in need, in so far as the buying off of the hostile incursions by stated annual payments pro- bably became a standing practice here — in what was not directly territory of the empire — still earlier than else- where. 1 Position of That the centralisation of the government had its application also in reference to this prince, and he stood to the Roman Caesar on a footing not much different from that of the burgomaster of Athens, is in various ways apparent ; it deserves mention that king Asander and the queen Dynamis struck gold coins with their name and their effigy, whereas king Polemon and his immediate suc- cessors, while retaining the right of coining gold, seeing that this territory as well as the adjoining barbarians were for long accustomed exclusively to gold currency, were induced to furnish their gold pieces with the name and the image of the reigning emperor. In like manner from Polemon's time the prince of this land was at the same time the chief priest for life of the emperor and of the imperial house. In other respects the administration and the court retained the forms introduced under Mithradates after the model of the Persian grand monarchy, although the chief secretary (ap%t,ypafjifjLaTev<;) and the chief chamberlain (ap^c- /eoiTcoveiTiis) of the court of Panticapaeum stood related to the leading court-officers of the great kings, as the enemy of the Romans Mithradates Eupator to his descendant Tiberius Julius Eupator, who, on account of his claim to the Bosporan throne, appeared as a suitor at Rome at the bar of the emperor Pius. 1 This we may be allowed to be- narrates not merely piSais fyioia, but lieve at the hands of the Scythian a very myth, of whose kings Leucanor Toxaris in the dialogue placed among and Eubiotes the coins, as may well those of Lucian (c. 44) ; for the rest he be conceived, have no knowledge. chap. vii. GREEK EUROPE. 319 This northern Greece remained valuable for the em- Trade and pire on account of its commercial relations. Though these ™ t ^ erce at this epoch were doubtless less important than in earlier Bosporus. times, 1 yet the mercantile intercourse continued very lively. In the Augustan period the tribes of the steppes brought slaves and skins, 2 the merchants of civilisation articles of clothing, wine, and other luxuries to Tanais ; in a still higher degree Phanagoria was the dep6t for the exports of the natives, Panticapaeum for the imports of the Greeks. Those troubles in the Bosporus in the Claudian age were a severe blow for the merchants of Byzantium. That the Goths began their piratic voyages in the third century by pressing the Bosporan vessels to lend them involuntary aid, has been already mentioned (p. 244). It was doubt- less in consequence of this traffic, indispensable for the barbarian neighbours themselves, that the citizens of Cher- sonesus maintained their ground even after the withdrawal of the Roman garrisons, and were able subsequently — when in Justinian's time the power of the empire once more asserted itself in this direction — to return as Greeks into the Greek empire. 1 As respects the export of grain, pressed by the Roman troops to de- the notice in the report of Plautius liver 10,000 slaves (Tacitus, Ann. xii. (p. 2 1 8), deserves attention. 17), it may be allowable to infer a 2 From the offer of a township of lively import of slaves from these the Siracae (on the Sea of Azoff) hard regions. CHAPTER VIII. ASIA MINOR. The great peninsula which is washed on three sides by the three seas, the Black, the Aegean, and the Mediterranean, and which is connected towards the east with the Asiatic continent proper, will, so far as it belongs to the frontier- territory of the empire, be dealt with in the next section, which treats of the region of the Euphrates and the relations between the Romans and Parthians. Here we have to set forth the peaceful relations, more especially of the western districts, under the imperial government. The natives The original, or at any rate pre-Greek, population of these wide regions held its ground in many places to a considerable extent down to the imperial period. The greatest part of Bithynia certainly belonged to the formerly discussed Thracian stock ; Phrygia, Lydia, Cilicia, Cappadocia, show very manifold and not easily unravelled survivals of older linguistic epochs, which in various forms reach down to the Roman period ; strange names of gods, men, and places meet us everywhere. But, so far as our view reaches — and it is but seldom allowed to penetrate here very deeply — these elements appear only losing ground and waning, essentially as a negation of civilisation or — what seems to us here at least to coincide with it — Hellenising. We shall return at the fitting place to the individual groups of this category ; so far as concerns the historical develop- ment of Asia Minor in the imperial period there were and the colonists, chap. vin. ASIA MINOR. 321 but two active nationalities, the two which were the last immigrants, the Hellenes in the beginnings of the historical period, and the Celts during the troublous times of the Diadochi. The history of the Hellenes of Asia Minor, so far Hellenic as it forms a part of Roman history, has already been fentstkfcui- set forth. In the remote age, when the coasts of the ture. Mediterranean were first navigated and settled, and the world began to be apportioned among the progressive nations at the expense of those left behind, the flood of Hellenic emigration had poured no doubt over all the shores of the Mediterranean Sea, but yet nowhere — not even towards Italy and Sicily, — in so broad a stream as over the Aegean Sea rich in islands, and the adjacent charming coast of anterior Asia rich in harbours. There- after the west -Asiatic Greeks themselves had taken an active part, above all the rest, in the further conquest of the world, and had helped to settle from Miletus the coasts of the Black, and from Phocaea and Cnidus those of the Western, Sea. In Asia Hellenic civilisation doubtless laid hold of the inhabitants of the interior, the Mysians, Lydians, Carians, Lycians ; and even the Persian great power remained not unaffected by it. But the Hellenes themselves possessed nothing but the fringe of coast, including at the utmost the lower course of the larger rivers and the islands. They were not able here to gain continental conquests and a power of their own by land overagainst the powerful native princes ; moreover the interior of Asia Minor, highlying and in great part but little capable of cultivation, was not so attractive for settlement as the coasts, and the communications of the latter with the interior were difficult. Essentially in consequence of this, the Asiatic Hellenes attained still less than the European to inward union and to great power of their own, and early learned submissiveness in presence of the lords of the continent. The national Hellenic idea first came to them from Athens ; they became its allies only after the victory, and did not remain so in the hour of danger. What Athens had VOL. I. 21 of new centres 322 ASIA MINOR. book viii. wished to provide, and had not been able to furnish for these clients of the nation, was accomplished by Alexander ; Hellas he was obliged to conquer, Asia Minor saw in the conqueror simply its deliverer. Formation Alexander's victory in fact not merely made Asiatic Hellenism secure, but opened up for it a wide, almost boundless, future ; in the process of continental settlement, which, in contrast to the merely littoral, marked this second stage of Hellenic world-conquest, Asia Minor took part to a considerable extent. Yet of the great centres for the newly formed states there was none that came to the old Greek towns of the coast. 1 The new period required new formations in general, and above all, new towns, to serve at once as Greek royal residences and as centres of populations hitherto non^Greek, that were to be brought to Greek habits. The great political development moves around the towns of royal foundation and of royal name, Thessalonica, Antioch, Alexandria. With their masters the Romans had to contend ; the possession of Asia Minor they gained almost throughout, as a man gets an estate from relations or friends, by bequest in a testament ; and, however heavy was the burden at times of Roman government on the regions thus acquired, there was not added here the sting of foreign rule. Doubtless the Achaemenid Mithradates confronted the Romans in Asia Minor with a national opposition, and the Roman misrule drove the Hellenes into his arms ; but the Hellenes themselves never undertook anything similar. Therefore there is little to be told of this great, rich, and important possession in a political respect ; and all the less, inasmuch as what has been remarked in the previous section concerning the national relations of the Hellenes generally to the Romans holds good in substance also for those of Asia Minor. The pro- The Roman administration of Asia Minor was never organised in a systematic way, but the several territories 1 Had the state of Lysimachus Ephesos-Arsinoe strengthened by the endured it would probably have been transference of the inhabitants of otherwise. His foundations, Alex- Colophon and Lebedos, tended in andria in the Troad and Lysimachia, the direction indicated. vinces of Asia Minor. chap. viii. ASIA MINOR. 323 were, just as they came to the empire, established without material change of their limits as Roman administrative districts. The states which king Attalus III. of Pergamus bequeathed to the Romans, formed the province of Asia ; those of king Nicomedes, which likewise fell to them by- inheritance, formed the province of Bithynia ; the territory taken from Mithradates Eupator formed the province of Pontus united with Bithynia. Crete was occupied by the Romans on occasion of the great war with the pirates ; Cyrene, which may also be mentioned here, was taken over by them according to the last will of its ruler. The same legal title gave to the republic the island of Cyprus ; to which was here added the need for the suppression of piracy. This had also laid a basis for the formation of the governorship of Cilicia ; the land was annexed to Rome completely by Pompeius at the same time with Syria, and the two were administered jointly during the first century. Possession of all these lands was already acquired by the republic. In the imperial period a number of territories were added, which had formerly belonged but indirectly to the empire : in 729 u. c. the 25- kingdom of Galatia, with which there had been united a part of Phrygia, Lycaonia, Pisidia, and Pamphylia ; in 747 U. c. the lordship of king Deiotarus, son of Castor, 7 . which embraced Gangra in Paphlagonia and probably also Amasia and other neighbouring places ; in 1 7 A. D. the kingdom of Cappadocia ; in 43 the territory of the confederation of the Lycian towns ; in 63 the north-east of Asia Minor from the valley of the Iris to the Armenian frontier ; Lesser Armenia and some smaller principalities in Cilicia probably by Vespasian. Thereby the direct imperial administration was carried out throughout Asia Minor. As dependent principalities, there remained only the Tauric Bosporus, of which we have already spoken, and Great Armenia, of which the next section will treat. When, on the introduction of the imperial government, senatorial the administrative partition was made between it and that a "^ ™ of the senate, the whole territory of Asia Minor, so far govem- as it was at that time directly under the empire, fell to men ' 324 ASIA MINOR. book viii. the latter body ; the island of Cyprus, which at first had come under imperial administration, was likewise transferred, a few years later, to the senate. Thus arose the four senatorial governorships of Asia, Bithynia and Pontus, Cyprus, Crete and Cyrene. Only Cilicia, as part of the Syrian province, was placed at first under imperial administration. But the territories that subsequently came to be directly administered as parts of the empire were here, as throughout the empire, placed under imperial governors ; thus even under Augustus there was formed from the inland districts of the Galatian kingdom the province of Galatia, and the coast district of Pamphylia was assigned to another governor, under which latter Lycia was also placed under Claudius. Moreover Cap- padocia became an imperial governorship under Tiberius. Cilicia also naturally remained, when it obtained governors of its own, under imperial administration. Apart from the fact that Hadrian exchanged the important province of Bithynia and Pontus for the unimportant Lyco-Pamphylian one, this arrangement remained in force, until towards the end of the third century the senatorial share in administra- tion generally was, with the exception of some slight remnants, superseded. The frontier was in the first period of the empire formed throughout by the dependent principalities ; after their annexation the imperial frontier did not, apart from Cyrene, touch any of these adminis- trative districts, excepting only the Cappadocian, so far as to this at that time was apportioned also the north- eastern border-district as far as Trapezus ; 1 and even this 1 Nowhere have the boundaries of xiii. 7, xiv. 26 ; Josephus,-^»i!. xx. the vassal states and even of the 8, 4), who still possessed it in the provinces changed more than in the year 60 ; in the year 75 the district north-east of Asia Minor. Direct was Roman (C. I. L. iii. 306), and imperial administration was intro- probably one of the legions garrisoning duced here for the districts of king Cappadocia from Vespasian's time Polemon, to which Zela, Neocaesarea, was stationed from the first in the Trapezus belonged, in the year 63 ; Lesser-Armenian Satala. Vespasian for Lesser Armenia, we do not know combined the regions mentioned, as exactly when, probably at the begin- well as Galatia and Cappadocia, into ning of the reign of Vespasian. The one large governorship. At the end last vassal king of Lesser Armenia, of the reign of Domitian we find of whom there is mention, was the Galatia and Cappadocia separated Herodian Aristobulus (Tacitus, Ann. and the north - eastern provinces chap. viii. ASIA MINOR. 325 governorship bordered not with the foreign land proper, but in the north with the dependent tribes on the Phasis, and farther on with the vassal-kingdom of Armenia, which belonged de jure and in more than one sense de facto to the empire. In order to gain a conception of the condition and the development of Asia Minor in the first three centuries of one era, so far as this is possible in the case of a country as to which we have no direct historical tradition, we must, looking to the conservative character of the Roman provincial government, begin with the older territorial divisions and the previous history of the several regions. The province of Asia was the old kingdom of the Asia. Attalids, the west of Asia Minor as far north as the Bithynian and as far south as the Lycian frontier ; the eastern districts at first separated from it, the Great Phry- gia, had already in the republican period been again attached to it (iii. 288), and the province thenceforth iii. 274- reached as far as the country of the Galatians and the Pisidian mountains. Rhodes too and the other smaller islands of the Aegean Sea belonged to this province. The The coast- original Hellenic settlement had, besides the islands and towns - the coast proper, occupied also the lower valleys of the larger rivers; Magnesia on the Sipylus, in the valley of the Hermus, the other Magnesia and Tralles in the valley of the Maeander, had already before Alexander been founded as Greek towns, or had at any rate become such ; the Carians, Lydians, Mysians, became early at least half Hellenes. The Greek rule, when it set in, found not much to do in the coast districts ; Smyrna, which centuries before had been destroyed by the barbarians of the attached to Galatia. Under Trajan governor. Consequently — apart from at first the whole district is once more a short interruption under Domitian in one hand, subsequently (Eph. Ep. — the legate of Galatia had nothing V. a. 1 345) it is divided in such a to do with the defence of the frontier, way that the north-east coast belongs and this, as was implied in the nature to Cappadocia. On that footing it of the case, was always combined remained, at least in so far that with the command of Cappadocia Trapezus and so also Lesser Armenia and of its legions, were thenceforth constantly under this tenor. 326 ASIA MINOR. book vm. interior, rose at that time from its ruins, in order speedily to become one of the first stars in the brilliant belt of the cities of Asia Minor ; and if the rebuilding of Ilion at the sepulchral mound of Hector was more a work of piety than of policy, the laying out of Alexandria on the coast of the Troas was of enduring importance. Pergamus in the valley of the Caicus flourished as the court-residence of the Attalids. The in- In the great work of Hellenising the interior of this province in keeping with the intentions of Alexander, all the Hellenic governments, Lysimachus, the Seleucids, the Attalids vied with each other. The details of the foun- dations have disappeared from our tradition still more than the warlike events of the same epoch ; we are left dependent mainly on the names and the surnames of the towns ; but even these suffice to make known to us the general outlines of this activity continuing for centuries, and yet homogeneous and throughout conscious of its aim. A series of inland townships, Stratonicea in Caria, Peltae, Blaundus, Docimeium, Cadi in Phrygia, the Mysomace- donians in the district of Ephesus, Thyatira, Hyrcania, Nacrasa in the region of the Hermus, the Ascylaces in the district of Adramytium, are designated in documents or other credible testimonies as cities of the Macedonians ; and these notices are of a nature so accidental, and the townships in part so unimportant, that the like designation certainly extends to a great number of other settlements in this region ; and we may infer an extensive settling of Greek soldiers in the districts indicated, probably con- nected with the protection of anterior Asia against the Galatians and Pisidians. If, moreover, the coins of the considerable Phrygian town Synnada combine with the name of their city that of the Ionians and the Dorians as well as that of the common Zeus (Zev? iravhrjiioi), one of the Alexandrids must have summoned the Greeks in common to settle there ; and the summons was certainly not confined to this single town. The numerous towns, chiefly of the interior, the names of which are traceable to the royal houses of the Seleucids or the Attalids, or chap. viii. ASIA MINOR. 327 which have otherwise Greek names, need not here be adduced ; there are found in particular among the towns certainly founded or reorganised by the Seleucids several that were in later times the most flourishing and most civilised in the interior, e.g. in southern Phrygia Laodicea, and above all Apamea, the old Celaenae on the great military road from the west coast of Asia Minor to the middle Euphrates, already in the Persian period the entre- p6t for this traffic, and under Augustus, next to Ephesus, the most considerable city of the province of Asia. Although every case of assigning a Greek name is not to be connected with a settlement by Greek colonists, we may be allowed at any rate to reckon a considerable portion of these townships among Greek colonies. But even the urban settlements of non-Greek origin, which the Alexandrids found in existence, turned of themselves into the paths of Hellenising, as indeed the residence of the Persian governor, Sardes, was organised even by Alexander himself as a Greek commonwealth. This urban development was completed when the its position Romans entered upon the rule of interior Asia ; they ^^^ e themselves did not make special exertions to promote it. That a great number of the urban communities in the eastern half of the province reckon their years from that of the city 670, is due to the fact that then, after the 84. close of the Mithradatic war, these districts were brought by Sulla under direct Roman administration (iii. 328) ; iu. 312. these townships did not receive city-rights only then for the first time. Augustus occupied the town of Parium on the Hellespont and the already-mentioned Alexandria in Troas with veterans of his army, and assigned to both the rights of Roman burgess - communities ; the latter was thenceforth in Greek Asia an Italian island like Corinth in Greece and Berytus in Syria. But this was nothing but a provision for soldiers ; of the foundation of towns proper in the Roman province of Asia under the emperors there is little mention. Among the not numerous towns named after emperors there it is only perhaps in the case of Sebaste and Tiberiopolis, both in Phrygia, and of 328 ASIA MINOR. book viii. Hadrianoi on the Bithynian frontier, that no older name of the city can be pointed out. Here, in the mountain- region between Ida and Olympus, dwelt Cleon in the time of the triumvirate, and a certain Tilliborus under Hadrian, both half robber-chiefs, half popular princes, of whom the former even played a part in politics ; in this asylum of criminals the foundation of an organised urban community by Hadrian was at all events a benefit. Otherwise in this province, with its five hundred urban communities, the province richest in cities of the whole state, not much more was left to be done in the way of foundation ; there was room at the most perhaps for division, that is, for detaching such hamlets as developed themselves de facto into urban communities, from the earlier communal union and making them independent, as we can point to a case of the kind in Phrygia under Constantine I. But from Hellenising proper the sequestered districts were still far remote when the Roman government began ; especially in Phrygia the language of the country, perhaps similar in character to the Armenian, held its ground. If from the absence of Greek coins and of Greek inscriptions we may not with certainty infer the absence of Hellenising, 1 yet the fact that the Phrygian coins belong almost throughout to the Roman imperial period, and the Phrygian inscrip- tions as regards the great majority to the later times of the empire, points to the conclusion that, so far as Hel- lenic habits found their way at all into the regions of the province of Asia that were remote and difficult of access to civilisation, they did so in the main only under the emperors. For direct interference on the part of the imperial administration this process, accomplishing itself in silence, gave little opportunity, and traces of such inter- ference we are not able to show. Asia, it is true, was a 1 Urban coining and setting up of that it was the promised land of muni- inscriptions are subject to so manifold cipal vanity, and our memorials, in- conditions that the want or the abun- eluding even the coins, have for by dance of the one or the other do not far the greatest part been called forth per se warrant inferences as to the by the fact that the government of absence or the intensity of a definite the Roman emperors allowed free phase of civilisation. For Asia Mi- scope to this vanity, nor in particular we must take note chap. viii. ASIA MINOR. 329 senatorial province, and we may here bear in mind that with the government of the senate all initiative fell into abeyance. Syria, and still more, Egypt, became merged in their Urban capitals ; the province of Asia and Asia Minor generally had no single town to show like Antioch and Alexandria, but their prosperity rested on the numerous middle-sized towns. The division of the towns into three classes, which are distinguished as to the right of voting at the diet, as to the apportionment of the contributions to be furnished by the whole province, even as to the number of town-physicians and town-teachers to be appointed, 1 is eminently peculiar to these regions. The urban rival- ries, which appear in Asia Minor so emphatic and in part so childish, occasionally even so odious — as, for example, the war between Severus and Niger in Bithynia was properly a war of the two rival capitals Nicomedia and Nicaea — belong to the character of Hellenic politics in general, but especially of those in Asia Minor. We shall mention further on the emulation as to temples of the emperors ; in a similar way the ranking of the urban deputations at the common festivals in Asia Minor was a vital question — Magnesia on the Maeander calls itself on the coins the " seventh city of Asia " — and above all the first place was one so much desired, that the government ultimately agreed to admit several first cities. It fared similarly with the designation of " metropolis." The proper metro- polis of the province was Pergamus, the residence of the Attalids and the seat of the diet. But Ephesus, the de facto capital of the province, where the governor was obliged to enter on his office, and which boasts of this " right of reception at landing " on its coins ; Smyrna, in constant rivalship with its Ephesian neighbour, and, in defiance of the legitimate right of the Ephesians to 1 "The ordinance," says the jurist tion how it is to be applied to pro- Modestinus, who reports it (Dig. vinces otherwise organised. What xxvii. I, 6, 3) "interests all pro- the biographer of Pius, t. 1 1, reports vinces, although it is directed to the as to the distinctions and salaries people of Asia." It is suitable, in granted by Pius to the rhetoricians, fact, only where there are classes of has nothing to do with this enact - towns, and the jurist adds an instruc- ment. 33° ASIA MINOR. book viii. primacy, naming itself on coins "the first in greatness and beauty ;" the very ancient Sardis, Cyzicus, and several others strove after the same honorary right. With these their wranglings, on account of which the senate and the emperor were regularly appealed to — the " Greek follies," as men were wont to say in Rome — the people of Asia Minor were the standing annoyance and the standing laughing-stock of the Romans of mark. 1 Bithynia. Bithynia did not stand on a like level with the Attalid kingdom. The older Greek colonising had here confined itself merely to the coast. In the Hellenistic epoch at first the Macedonian rulers, and later the native dynasty which walked entirely in their steps, had — along with a regulation of the places on the coast, which perhaps on the whole amounted to a changing of their names — also opened up in some measure the interior, in particular by the two successful foundations of Nicaea (Isnik) and Prusa on Olympia (Broussa) ; of the former it is stated that the first settlers were of good Macedonian and Hellenic descent. But in the intensity of the Hellen- ising the kingdom of Nicomedes was far behind that of the citizen prince of Pergamus'; in particular the eastern interior can have been but little settled before Augustus. This was otherwise in the time of the empire. In the Augustan age a successful robber-chief, who became a convert to order, reconstructed on the Galatian frontier the utterly decayed township Gordiou Kome, under the name of Juliopolis ; in the same region the towns Bithynion- Claudiopolis and Crateia-Flaviopolis probably attained Greek civic rights in the course of the first century. 1 Dio of Prusa, in his address to towns, as Nicaea and Nicomedia never the citizens of Nicomedia "and of act together. " The Romans deal with Tarsus, excellently lays it down that you as with children, to whom one no man of culture would have such presents trifling toys ; you put up empty distinctions for himself, and with bad treatment in order to ob- that the greedy quest of the towns for tain a name ; they name your town titles was altogether inconceivable ; the first in order to treat it as the how it is the sign of the true petty- last. By this you have become a townsman to cause a display of such laughing-stock to the Romans, and attestations of rank on his behalf; they call your doings ' Greek follies ' " how the bad governor always screens ("EXX^cuci a/iapr-tinaTa). himself under this quarrelling of chap, vm. ASIA MINOR. 331 Generally in Bithynia Hellenism took a mighty upward impulse under the imperial period, and the tough Thracian stamp of the natives gave a good foundation for it. The fact that, among the inscribed stones of this province known in great number, not more than four belong to the pre-Roman epoch, cannot well be explained solely from the circumstance that urban ambition was only fostered under the emperors. In the literature of the imperial period a number of the best authors and the least carried away by exuberant rhetoric, such as the philosopher Dio of Prusa, the historian Memnon of Heraclea, Arrianus of Nicomedia, Cassius Dio of Nicaea, belong to Bithynia. The eastern half of the south coast of the Black Sea, Pontus. the Roman province of Pontus, had as its basis that portion of the kingdom of Mithradates, of which Pompeius took direct possession immediately after the victory. The numerous smaller principalities, which Pompeius at the same time gave away in the interior of Paphlagonia and thence eastward to the Armenian frontier, were, after a shorter or longer subsistence, on their annexation partly attached to the same province, partly joined to Galatia or Cappadocia. The former kingdom of Mithradates had been far less affected than the western regions either by the older or by the younger Hellenism. When the Romans took possession directly or indirectly of this territory, there were, strictly speaking, no towns of Greek organisation there ; Amasia, the old residence of the Pontic Achaemenids, and still their burial-place, was not such ; the two old Greek coast-towns, Amisus and Sinope that once commanded the Black Sea, had become royal residences, and Greek polity would hardly be given to the few townships laid out by Mithradates, e.g. Eupatoria (iv. 1 S 2). But here, as was already shown in detail (iv. 151 f), w. 146. the Roman conquest was at the same time the Hellenising ; Pompeius organised the province in such a way as to make the eleven chief townships of it into towns, and to distribute the territory among them. Certainly these artificially created towns with their immense districts — that of Sinope had along the coast an extent of 70 332 ASIA MINOR. book viii. miles, and bordered on the Halys with that of Amisus — resembled more the Celtic cantons than the Hellenic and Italian urban communities proper. But at any rate Sinope and Amisus were then reinstated in their old positions, and other towns in the interior, such as Pom- peiopolis, Nicopolis, Megalopolis, the later Sebasteia, were called into life. Sinope obtained from the dictator Caesar the rights of a Roman colony, and beyond doubt iv. 544- also Italian settlers (iv. 574). More important for the Roman administration was Trapezus, an old colony of Sinope ; the town, which in the year 63 was joined to the province of Cappadocia (p. 324, note), was, as the station of the Roman Black Sea fleet and so in a certain measure the base of operations for the military corps of this province, unique in all Asia Minor. Cap- Inland Cappadocia^ was in the Roman power after the padocia. erection of the provinces of Pontus and Syria ; of its annexation in the beginning of the reign of Tiberius, which was primarily occasioned by the attempt of Armenia to release itself from the Roman suzerainty, we shall have to give an account in the following section. The court, and those immediately connected with it, had »i- 57- become Hellenised (iii. 59), somewhat as the German courts of the eighteenth century adapted themselves to French habits. The capital, Caesarea, the ancient Mazaca, like the Phrygian Apamea, an intermediate station for the great traffic between the ports of the west coast and the lands of the Euphrates, and in the Roman period, as still at the present day, one of the most flourishing commercial cities of Asia Minor, was, at the instigation of Pompeius, not merely rebuilt after the Mithradatic war, but probably also furnished at that time with civic rights after the Greek type. Cappadocia itself was at the beginning of the imperial period hardly more Greek than Brandenburg and Pomerania under Frederick the Great were French. When the country became Roman, it was divided, accord- ing to the statements of the contemporary Strabo, not into city-districts, but into ten prefectures, of which only two had towns, the already-mentioned capital and Tyana ; chap. viii. ASIA MINOR. 333 and this arrangement was here on the whole not more changed than in Egypt, though individual townships sub- sequently received Greek civic rights ; e.g. the emperor Marcus made the Cappadocian village, in which his wife had died, into the town Faustinopolis. It is true that the Cappadocians now spoke Greek ; but the students from Cappadocia had much to endure abroad on account of their uncouth accent, and of their defects in pronuncia- tion and modulation ; and, if they learned to speak after an Attic fashion, their countrymen found their language affected. 1 It was only in the Christian period that the comrades in study of the emperor Julian, Gregory of Nazianzus and Basil of Caesarea, gave a better sound to the Cappadocian name. The Lycian cities in their secluded mountain-land did Lycia. not open their coast for Greek settlement, but did not on that account debar themselves from Hellenic influence. Lycia was the only district of Asia Minor in which early civilising did not set aside the native language, and which, almost like the Romans, entered into Greek habits without becoming externally Hellenised. It is characteristic of their position, that the Lycian confederation as such joined the Attic naval league and paid its tribute to the Athenian leading power. The Lycians not merely prac- tised their art after Hellenic models, but probably also regulated their political organisation early in the same way. The conversion of the cities-league, once subject to Rhodes, but which had become independent after the third Macedonian war (ii. 325) into a Roman province, u. 307. which was ordained by the emperor Claudius on account of the. endless quarrels among the allies, must have furthered the progress of Hellenism ; in the course of the imperial period the Lycians thereupon became completely Greeks. The Pamphylian coast-towns, like Aspendus and Perga, Pamphyiia and Cilicia. 1 Pausanias of Caesarea in Philo- X^ av > awriWuv 5k to /j.r/KvvS/j.ei'a nal stratus ( Vitae soph. ii. 1 3) places before pqicivuv ra ppaxta. Vita Apoll. i. 7 ; Herodes Atticus his faults : ira^ef? V 7^wtt (on account of the travelling on 36. not till the year 7 1 8 that Amyntas circuit) /«jre /ied' 6VXuc (which at all obtained Galatia, Lycaonia, and Pam- events were wanting to the later legate phylia (Dio, xlix. 32). of Galatia). 336 ASIA MINOR. book viii. Minor. If in doing so he, as was already observed (p. 324), assigned the small Pamphylian coast-district to a governor of its own and separated it from Galatia, this was evidently done because the mountain-land lying between the coast and the Galato-Lycaonian steppe was so little under con- trol that the administration of the coast region could not well be conducted from Galatia. Roman troops were not stationed in Galatia ; yet the levy of the warlike Galatians must have meant more than in the case of most provincials. Moreover, as western Cilicia was then placed under Cap- padocia, the troops of this dependent prince had to take part in the work. The Syrian army carried out the chastisement in the first place of the Homonadenses ; the governor, Publius Sulpicius Quirinius, advanced some years later into their territory, cut off their supplies, and compelled them to submit en masse, whereupon they were distributed to the surrounding townships and their former territory was laid waste. The Clitae, another stock settled in western Cilicia nearer to the coast, met with similar chastisements in the years 3 6 and 5 2 ; as they refused obedience to the vassal - prince placed over them by Rome, and pillaged land and sea, and as the so-called rulers of the land could not dispose of them, the imperial troops were on both occasions brought in from Syria to subdue them. These accounts have been accidentally preserved ; numerous similar incidents have certainly been lost to remembrance. Ksidian But Augustus attempted the pacification of this region colonies. a i so by way of settlement. The Hellenistic governments had, so to speak, isolated it ; not merely retained or seized a footing everywhere on the coast, but also founded in the north-west a series of towns — on the Phrygian frontier Apollonia, alleged to have been founded by Alexander himself, Seleucia Siderus and Antiochia, both from the time of the Seleucids, further in Lycaonia, Laodicea Katakekaumene, and the capital of this district which doubtless originated at the same time, Iconium. But in the mountain-land proper no trace of Hellenistic settle- ment is found, and still less did the Roman senate apply chap. vni. ASIA MINOR. 337 itself to this difficult task. Augustus did so ; and only here in the whole Greek coast we meet a series of colonies of Roman veterans evidently intended to acquire this district for peaceful settlement. Of the older settlements just mentioned, Antiochia was supplied with veterans and reorganised in Roman fashion, while there were newly laid out in southern Lycaonia Parlais, in Pisidia itself the already - mentioned Cremna, as well as further to the south Olbasa and Comama. The later governments did not continue with equal energy the work so begun ; yet under Claudius the " iron Seleucia '' of Pisidia was made the " Claudian ; " while in the interior of western Cilicia Claudiopolis, and not far from it, perhaps at the same time, Germanicopolis were called into life, and Iconium, in the time of Augustus a small place, was brought to considerable development. The newly-founded towns remained indeed unimportant, but still notably restricted the field of the free inhabitants of the moun- tains, and general peace must at length have made its triumphal entrance also here. As well the plains and mountain-terraces of Pamphylia as the mountain-towns of Pisidia itself, e.g. Selga and Sagalassus, were during the imperial period well peopled and the territory carefully cultivated ; the remains of mighty aqueducts and singu- larly large theatres, all of them structures of the Roman imperial period, show, it is true, only mechanical skill, but bear traces of a peaceful prosperity richly developed. The government, it is true, never quite mastered isaurians. brigandage in these regions, and if in the earlier period of the empire its ravages were kept in moderate bounds, the bands once more emerge as a warlike power in the troubles of the third century. They now pass under the name of Isaurians, and have their chief seat in the moun- tains of Cilicia, from whence they plunder land and sea. They are mentioned first under Severus Alexander. That under Gallienus they proclaimed their robber-chief em- peror, is probably a fable ; but certainly under the emperor Probus such an one, by name Lydius, who for long had pillaged Lycia and Pamphylia, was subdued in VOL. I. 22 338 ASIA MINOR. book viii. the Roman colony Cremna, which he had occupied, after a long and obstinate siege by a Roman army. In later times we find a military cordon drawn round their terri- tory, and a special commanding general appointed for the Isaurians. Their savage valour even procured for those of them, who chose to take service at the Byzantine court, for a time a position there such as the Macedonians had possessed at the court of the Ptolemies ; in fact one from their ranks, Zeno, died as emperor of Byzantium. 1 Gaiatia. Lastly, the region of Galatia, at a remote period the chief seat of the Oriental rule over anterior Asia, and pre- serving in the famed rock-sculptures of the modern Bog- hazkoi, formerly the royal town of Pteria, reminiscences of an almost forgotten glory, had in the course of centuries become in language and manners a Celtic island amidst the waves of eastern peoples, and remained so in internal organisation even under the empire. The three Celtic tribes, which, on the great migration of the nation about the time of the war between Pyrrhus and the Romans, had arrived in the heart of Asia Minor, and there, like the Franks in the East during the middle ages, had con- solidated themselves into a firmly knit soldier-state, and after prolonged roving had taken up their definitive abode on either side of the Halys, had long since left behind the times when they issued forth thence to pillage Asia Minor, and were in conflict with the kings of Asia and Pergamus, provided that they did not serve them as mercenaries. They too were shattered before the superior power of the i. 273. Romans (ii. 290), and became not less subject to them in Asia than their countrymen in the valley of the Po and on the Rhone and Seine. But in spite of their sojourn of several hundred years in Asia Minor, a deep gulf still separated these Occidentals from the Asiatics. It was not merely that they retained their native language and 1 Amidst the great unnamed ruins century after Christ, on which muti- of Sarajik, in the upper valley of lated parts of men— heads, arms, legs the Limyrus, in eastern Lycia (comp. —are produced in relief, as emblems Ritter, Erdkunde, xix. p. 1 1 72), stands we might imagine, as the coat of arms , a considerable temple-shaped tomb, of a civilised robber-chief (communi- certainly not older than the third cation from Benndorf ). chap. viii. ASIA MINOR. 339 their nationality, that still each of the three cantons was governed by its four hereditary princes, and the federal assembly, to which deputies were sent by all in common, presided in the sacred oak -grove as supreme authority over the Galatian land (ii. 232); nor was it that continued a. 219. rudeness as well as warlike valour distinguished them to advantage as well as to disadvantage from their neighbours ; such contrasts between culture and bar- barism existed elsewhere in Asia Minor, and the superficial and external Hellenising — such as neigh- bourhood, commercial relations, the Phrygian cultus adopted by the immigrants, and mercenary service brought in their train — must have set in not much later in Galatia than e.g. in the neighbouring Cappadocia. The contrast was of a different kind ; the Celtic and the Hellenic invasion came into competition in Asia Minor, and to the distinction of nationality was added the spur of rival conquest. This was brought clearly to light in the Mithradatic crisis ; by the side of the command of Mithradates to murder the Italians went the massacre of the whole Galatian nobjlity (iii. 322), and, m. 306. in keeping therewith, the Romans in the wars against the Oriental liberator of the Hellenes had no more faithful ally than the Galatians of Asia Minor (iv. 5 6, 1 49). iv. 53, 143. For that reason the success of the Romans was theirs The Gaia- also, and the victory gave to them for a time a leading ^JJ, ms " position in the affairs of Asia Minor. The old tetrarchate was done away, apparently by Pompeius. One of the new cantonal princes, who had approved himself most in the Mithradatic wars, Deiotarus, attached to himself, besides his own territory, Lesser Armenia and other portions of the former Mithradatic empire, and became an incon- venient neighbour to the other Galatian princes, and the most powerful among the dynasts of Asia Minor (iv. 149). After the victory of Caesar, to whom he iv. 143. occupied an attitude of hostility, and whose favour he was unable to gain even by help rendered against Pharnaces, the possessions gained by him with or without consent of the Roman government were for the most 34° ASIA MINOR. book viii. part again withdrawn ; the Caesarian Mithradates of Pergamus, who on the mother's side was sprung from the Galatian royal house, obtained the most of what Deiotarus lost, and was even placed by his side in Galatia itself. But, . after the latter had shortly after- wards met his end in the Tauric Chersonese (p. 313), and Caesar himself had not long afterwards been murdered, Deiotarus reinstated himself unbidden in possession of what he had lost, and, as he knew how to submit to the Roman party predominant on each occasion in the East as well as how to change it at the 40. right time, he died at an advanced age in the year 714 as lord of all Galatia. His descendants were portioned off with a small lordship in Paphlagonia ; his kingdom, further enlarged towards the south by Lycaonia and all the country down to the coast of Pamphylia, was 36. transferred, as was already said, in the year 718 by Antonius to Amyntas, who seems to have conducted the government already in the last years of Deiotarus as his secretary and general, and, as such, had before the battle of Philippi effected the transition from the republican generals to the triumvirs. His further fortunes have been already told. Equal to his predecessor in sagacity and bravery, he served first Antonius, and then Augustus as chief instrument for the pacification of the territory not yet subject in Asia Minor, till he there met his 25. death in the year 729. With him ended the Galatian kingdom, and it was converted into the Roman province of Galatia. Theinhabi- Its inhabitants were called Gallograeci among the Romans even in the last age of the republic ; they were, adds Livy, a mixed people, as they were called, and degenerate. A good portion of them must have descended from the older Phrygian inhabitants of these regions. Of still more weight is the fact, that the zealous worship of the gods in Galatia and the priesthood there have nothing in common with the ritual institutions of the European Celts ; not merely was the Great Mother, whose sacred symbol the Romans of Hannibal's time tants. chap. vin. ASIA MINOR. 341 asked and received from the Tolistobogi, of a Phrygian type, but her priests belonged in part at least to the Galatian nobility. Nevertheless, even in the Roman province of Galatia the internal organisation was pre- dominantly Celtic. The fact that even under Pius the strict paternal power foreign to Hellenic law subsisted in Galatia, is a proof of this from the sphere of private law. In public relations there were in this country still only the three old communities of the Tectosages, the Tolistobogi, the Trocmi, who perhaps appended to their names those of the three chief places, Ancyra, Pessinus, and Tavium, but were essentially nothing but the well- known Gallic cantons, which also indeed were not with- out their chief place. If among the Celts of Asia the conception of the community as town gains the predomi- nance earlier than among the European, 1 and the name Ancyra more quickly dispossesses that of the Tectosages than in Europe the name Burdigala dispossesses that of the Bituriges, and there Ancyra even as foremost place of the whole country calls itself the " mother - city " (jirjT pair okis), this certainly shows — what could not in fact be otherwise — the influence of Greek neighbourhood and the incipient process of assimilation, the several phases of which the superficial information that survives to us does not allow us to follow out. The Celtic names keep their hold down to the time of Tiberius ; afterwards they appear only isolated in the houses of rank. That the Romans after the erection of the province — as Language in Gaul they allowed only the Latin language — allowed in Jeans' 6 Galatia alongside of this only the Greek in business-dealings, was a matter of course. What course was taken earlier we know not, as we do not meet with pre-Roman written monuments in this country at all. As the language of conversation the Celtic maintained its ground with 1 The famous list of services ren- subsequently disappears ; but in the dered to the community of Ancyra full title, e.g. of the inscription, C. I. of the time of Tiberius (C. I. Gr. Gr. 401 1, from the second century, 4039) designates the Galatian com- Ancyra always bears the name of the munities usually by I0vos, sometimes people : r/ injTpbiroKis rrjs TaKartas by 7rv\i<} 7r6Xeioj>, Benndorf, Lyk. ko.1 ttjs iravryyipews (ib. p. 254). Reise, i. 122 ; letters to the Hellenes Another document of the same league in Asia, C. I. Gr. 3832, 3833; 3 from the time of Antigonus is given foSpes "EXKr/ves in the address to the in Droysen, ffellenismus, ii. 2, 382 ff. diet of Pergamus, Aristides, p. 517. — So two other trowd are to be taken, An fi/jfas tov koipov tuv iv Bidvvtf which refer to a narrower circle than 'EXK^vuv, Perrot, Expl. de la Galatie, the province, such as the old one of p. 32 ; letter of the emperor Alexan- the thirteen Ionic cities, that of the der to the same, Dig. xlix. I, 25. — Lesbians (Marquardt, Staatsverw. i. Dio, li. 20: tois %hoi%, "EXhjvas ltT■ Tacitus, Ann. iv. i S> 55. The " (imperial) temple-keeper "(i/ew/rfpos)- town which possesses a temple dedi- and, if one of them has several to cated by the diet of the province (the show, the number is appended In Koivhv rris 'Arias k. r. \. ) bears on that this institution one may clearly dis- account the honorary predicate of the cern how the imperial worship ob- CHAP. VIII. ASIA MINOR. 3A7 only one president and one chief priest, as only one temple, now not merely had as many chief priests to be appointed as there were provincial temples, but also, seeing that the conduct of the temple-festival and the execution of the games pertained not to the chief priest but to the land-president, and the rival great towns were chiefly concerned about the festivals and games, there was given to all the chief priests at the same time the title and the right of presidency, so that at least in Asia the Asiarchy and the chief priesthood of the provincial temples coincided. 1 Therewith the diet and the civil tained its full elaboration in Asia Minor. In reality the neocorate is general, applicable to any deity and any town ; titularly, as an honorary surname of the town, it meets us with vanishing exceptions only in the im- perial cultus of Asia Minor — only some Greek towns of the neighbour- ing provinces, such as Tripolis in Syria, Thessalonica in Macedonia, participated in it. 1 However little the original diver- sity of the presidency of the diet and the provincial chief-priesthood for the cultus of the emperor can be called in question, yet not merely in the case of the former does the magisterial character of the president, still clearly recognisable in Hellas, whence the organisation of the k.ou/6. generally proceeds, fall completely into the shade in Asia Minor, but here in fact, where the koiv6i> has several ritual centres, the 'As i-jriTeXotivTuv), while it is the very essence of the Asiarchate. To all appearance the rivalries of the towns have here led to the result, that, after there were several temples of the emperor dedicated by the pro- vince in different towns, the Agono- thesia was taken from the real presi- dent of the diet, and, instead, the titular Asiarchate and the Agonothesia were committed to the chief priest of each temple. In that case the 'Aaitipxys teal apxtepeis ij ir6\eav is explained on the coins of the thirteen Ionic towns (Mionnet, iii. 61, 1), and on Ephesian inscriptions the same Ti. Julius Re- ginus may be named sometimes 'Atri- apxys /3' va&v t&v £i> E0&ry (Wood, Inscr. from the gi-eat theatre, p. 18), sometimes apxiepebs /3' vawv t&v tv 'E07riiXi;s), may be assumed for them. In the two retail dealers (KdirriKos), five gold- third century here, as everywhere, the smiths (aipdpios thrice, xP vff ^X oos construction even of the imperial twice), one of whom is also presbyter, highways was transferred to the com- four coppersmiths (xaX/ciTmros once, munes (Smyrna: C. I. L. iii. 471; xahKtis thrice), two instrument-makers Thyatira, Bull, de corr. Hell. i. 101 ; (dp/j.wopd