MASTER NEGA TIVE NO. 92-81113 MICROFILMED 1993 COLUMBIA UNIVERSITY LIBRARIES/NEW YORK as part of the "Foundations of Western Civilization Preservation Project" Funded by the NATIONAL ENDOWMENT FOR THE HUMANITIES Reproductions may not be made without permission from Columbia University Library COPYRIGHT STATEMENT The copyright law of the United States - Title 17, United States Code - concerns the making of photocopies or other reproductions of copyrighted material. Under certain conditions specified in the law, libraries and archives are authorized to furnish a photocopy or other reproduction. One of these specified conditions is that the photocopy or other reproduction is not to be "used for any purpose other than private study, scholarship, or research.** If a user makes a request for, or later uses, a photocopy or reproduction for purposes in excess of **fair use,** that user may be liable for copyright infringement. This institution reserves the right to refuse to accept a copy order if, in its judgement, fulfillment of the order would involve violation of the copyright law. AUTHOR: HARDY, ERNEST GEORGE TITLE: STUDIES IN ROMAN HISTORY PLACE: LONDON DA TE : 1909 COLUMBIA UNIVERSITY LIBRARIES PRESERVATION DEPARTMENT Master Negative # BIBLIOGRAPHIC MICROFORM TARHRT Original Material as Filmed - Existing Bibliographic Record Restrictions on Use: A ^874. 06" ' H222 Hardy, Ernest George, 1852- Studies in Roman history. 2d series, by E. G. Hardy ... London, S. Sonnenscliein & CO. lim., 1909. xii. 307 p. 2 fold. maps. 17i'-. Contents. — I. The armies and frontier relations of the German prov- inces. — II. The four emperors* year. — ill. A military game of chess. D07 4 ,Q& C o py in IP^arnftrd t lOOO i H2^ ^Rome — Hist. — Empire, b. c. 3G-A. d. 476. 2. Germany — Hist. — Early period to 843. 3. Rome — History, Militar>'. 4. Rome — Defenses. Library of Congress I J DG271.H3 rs23dli 10—28807 'J TECHNICAL MICROFORM DATA ^5u±^ ^^_ REDUCTION RATIO: //^ IMAGE PLACEMENT: lA y^ IB IIB ■ — - DATE FILMED:___2,-_2>v^^l INITIALS HLMEDBY: RESEARCH PUBLICATIONS. INC WOODBRlbdrCT V c Association for Information and Image Management 1100 Wayne Avenue. Suite 1100 Silver Spring, Maryland 20910 301/587-8202 Centimeter iiii 4 5 iiiiliiiiliiiiliiiiiiiiiiiiiiliiiiliiiiliiiiliiii TTT Inches Mil 6 iiiliiii •I' 2 7 8 9 10 11 12 13 14 15 mm iiiliiMliiiiliiMliiiiliiiiliiiiliiiilMiiliiiilMii [mjlniiL I I I I I I I I I I I I rrr 1.0 I.I 1.25 y^ 112.8 ■ 5.0 " 2.5 la ll^-= ■ 6.3 2.2 ■ 80 II 4.0 2.0 l£. 1.8 1.4 1.6 I MRNUFnCTURED TO fillM STflNDfiRDS BY RPPLIED IMRGE, INC. m^-y^iomm y:>:f:;>v; :»:>-^»S:.!t?v*)i.:.E; .SS'.^SN W?^'§:MM:^M'&f''iM ^^^'^m:^^^: -.:.*-M"v ;!:!:■;«!:'' .-... :.':,. .••.'•.-.^,'.v'.'.S-.,-.H',-Asw.' -i * "^j: ^ ■^- Vi/'-'^j! -Vi.:" /■St ■» .»v-, ^t' .-:-v >t. ■M * 'it . \ ^?ftH 'M i.*» t Columbia «Bnit»ers(ttj> intl)fCitpofJlftti]?ork THE LIBRARIES it: IV f! ^ ! ]\\ ■H 1 ^ Michael Shinnick . STUDIES IN ROMAN HISTORY 11 STUDIES IN BY THE SAME AUTHOR STUDIES IN ROMAN HISTORY. (Series I) Crown Svo, cloth, gilt, 6s. Some Opinions of the Press. .. Dr. Hardy presents his case with the utm^^^^^ of language, and there is no PO/"tpth^^l^^^^^^ certainly l^ read to accept his conclusions. Altogether the l>x>k is on^^ Histmical Revtev>. with interest and deserves to be studied wittirespeci^s , ^^^ ^^ ^.^^g. •' The Essay on ' Christianity and ^j^^ Roman ^^vernmcu ^^.^^^ nized as a so/d contribution to the study of a difficult^ubjea..^^i^^^ ;i£:t^^6^:,'^in^9^s^^''^^^oUon. with sufhcient knowledge to ^^D^.'HSytolr^t'';^t^qd ac.^^^^^^^^ ^^nS.°^rhTve^Srrugrto%C;^^^^^^^^^ is indispensable -:bi^ai/?j^?^-i^ and the Roman Government, wh>ch w^P^^^^ ^^^^^^^^^ which ap- Another fifty pages is occupied ^y ^?f ^^le on tne ^^\ y^^ ^.^re accepted peared in the £»$^«* ^»^,^''»^«^,?f?*^uS>ns to know^^^^ when their author was by the Board of the Faculty as contributions to knowieag ^^^^ gJanted the degree ?* D.Utt in igoi ^^^^^ .^^^Val^^^^^^^^ of the impor- always been recognised as ^^f^/^^^^^/^if^ gLeedingly convenient to have them tant subjects of which they t'^eat and it is excewung^y important bear purposes."— Oy/of\e f , ^ (3 n in CVJ z 3 N VN TO I 9^Jl2>^ THE PRINCIPAL AND FELLOWS OF JESUS COLLEGE, OXFORD, TO MARK A CONNEXION WITH THE COLLEGE OF ONE THIRD OF A CENTURY PREFACE The publication, in its present form, of the first of these essays, on the German armies and frontier, requires some explanation, if not apology. It was written fifteen years ago, and was intended to be the first of a series of essays dealing with the armies and frontiers of the early Empire, one main object being to utilize the plentiful, but at that time Httle explored epigraphic evidence as to the auxiliary troops. Other engage- ments, however, prevented the pursuance of this plan, and when, after five or six years, I might have returned to it, I was no longer able to do so. Accordingly the one completed section of my work remained buried away and almost forgotten, till I unearthed it three years ago, at the time when the first series of these " Studies " was being prepared. I then made some attempt to verify the inscriptions cited from Brambach by the aid of vol. xiii. of the C.J.L., which had recently been published. The difficulties, however, in the way of doing this, without an expert assistant, and without indices, the volume containing them not having then appeared, were over- whelming and I was able to do no more than add a few references, and one or two fresh notes. Perhaps I should have been wise if I had stood by the decision I then came to, of finally putting the thing aside. I suppose that it vu VUl PREFACE is the natural, if not excusable, reluctance finally to throw away what represents many months' laborious investigation of Brambach and the inscriptions then only to be found scattered over German periodicals, that has made me change my mind and include the essay, for what it is worth, in the present volume. Of course from the point of view of the technical and expert examination of the German Limes, conducted by the Limes-Com- mission mostly since my paper was written, it may be regarded as in a sense antiquated. The account given here of the Limes is of a general character only, and de- pends upon the earlier evidence, and the views of Momm- sen and other German scholars who used that evidence. But my principal object was not so much to deal with the technicalities of the frontier arrangements in them- selves, as to show the distribution of the various parts of the German army along the frontier, and, where pos- sible, to trace the movements of those parts, both legions and auxiliaries, from one position to another, and to determine the strength and composition of the two armies at successive periods. As far as I have been able to look into the new volume of the C.I.L., I gather that my results are not seriously affected by later discoveries. At the same time, my work on the German auxiliaries would have been immensely lightened, and no doubt greatly improved, if I had had at my disposal the two articles by Cichorius on ala and cohors in the new edition of Pauly, the value of which slowly progressing work is perhaps best reaUzed by those who, Uke myself, are quite unable to use its exhaustive information. With regard, too, to the general character, object and direction of the Limes, I gather from the late Prof. Pelham's lucid r6sum6 of recent results, which came under my notice PREFACE IX too late to be more than referred to in one or two notes, that earlier theories have been verified, amplified and illustrated rather than corrected by the work of the Limes- Commission, while the details as to construction, the dates of particular fortresses and the system of inter- connecting roads would in any case have stood outside my subject. I must crave the indulgence of the probably very few readers, who will go into the epigraphical evidence, for possible mistakes in some of the references, and especi- ally in the numerals. I think I may say with confidence that these were all correct in the original draft of the article, for I never referred to an inscription, either in Brambach or elsewhere, without verifying it. But the difficulties in preparing this part of the volume for the press have been extraordinarily great. The original copy was in pencil, never perhaps very legible, and made still less so by lapse of time. It had therefore to be typed out, some of it more than once, while portions of it I was compelled myself to rewrite in Braille, a process in which numerals are very liable to go wrong. The obvious answer to all this is of course that I might have avoided the difficulties involved in presenting this con- fessedly imperfect piece of work to the pubUc, by leaving it alone. If this is the verdict pronounced by those most competent to judge, I shall feel that no less and no more than justice is done. The subject of the auxiUary branch of the German army has never yet been system- atically dealt with. It seems to me that it can best be treated frontier by frontier. With regard to Britain, we shall no doubt within measurable distance get from Prof. Haver field aU that can be desired. But what I, for one, should like to see, would be a thorough examination j^ PREFACE Of the evidence suppHed by the Danube provinces in connexion with the frontier arrangements generaUy, and particularly with regard to the auxihary troops, their numbers, their disposition and their movements. There are many more Diplomata for these provinces than for Germany, and the material has aheady been largely collected by Cichorius. If my own semi-post- humous effort should suggest this undertaking to some better equipped scholar, such a result would be its best justification. I need say Uttle about the other two essays. That on " The Four Emperors' Year " is not intended to compete with Mr. Henderson's Civil War and Rebellion %n the Roman Empire. It was indeed already written m BraiUe, though not typed out, before I had the pleasure of reading his three briUiant and graphic chapters. Pos- sibly, if I had known beforehand of his coming book, I should have amused myself with some other chapter of Roman history. Still, my object was not so exclusively as his the miU- tary history of the year, while even from the point of view of miUtary history, it may be interesting to compare an account of the first Itahan campaign, professedly based upon Tacitus, with Mr. Henderson's reconstruction in which Tacitus is set at nought. The last essay is merely a farewell to some lectures, formerly given to my pupils, and now rendered super- fluous by the regrettable removal of Caesar's Bellum Civile from the hst of historical texts taken up. I have to acknowledge the courtesy of Messrs. Mac- millan for allowing me, (i) to reprint, practically as they stood, between twenty and thirty pages from my intro- duction to Plutarch's Lives of Galba and Otho, pubUshed PREFACE XI by them in 1889 ; and (2) to reproduce the map of Roman Germany from the English translation of Mommsen's Provinces of the Roman Empire. Finally I desire to record my gratitude to Mr. J. C. Haire for the invaluable assistance which he has rendered me during the preparation of this volume. At spare times, during practically the whole of this year, he has been engaged in deciphering my illegible MSS., in verifying references and inscriptions, correct- ing typists' errors, and in a word, in producing some kind of order out of the chaos of written and re- written, corrected and re-corrected sheets. How, as a layman to the subject, he could have spent hour after hour in the bewildering company of Cohorts IV VindeHcorum, Aquitanorum Biturigum, and the rest of them, and retained his reason, I hardly understand. At any rate, without his help in not only doing all this, but also in reading practically the whole of the Histories to me, I should have been unable to advance a step. At the end of all he has corrected the proofs of the entire book, not the least difficult of his tasks in view of the comphcated character of the notes. E. G. H. Oxford, Dec. 12, 1908. CONTENTS THE ARMIES AND FRONTIER RELATIONS OF THE GERMAN PROVINCES PAGB § I. The German Army up to Varus § 2. From 9 a.d. to the Recall of Germanicus § 3. The German Legions and Auxilia under Tiberius and Claudius .... § 4. Events in Germany under Nero § $. The Batavian Rising § 6. The Gallic Empire ...... § 7. The Campaign of Petilius Cerealis § 8. The German Auxiliaries before 70 a.d. . § 9. The German Legions as Placed in 70 a.d. § 10. Frontier Arrangements under Vespasian and domitian ....... § II. The Rising of Saturninus and Changes Caused BY IT ....•••• § 12. The Upper German Limes .... 1 13. The Legions under Trajan and Hadrian § 14. The Auxilia of Upper Germany from the Flavian Period Onwards .... § 15. The Auxilia of Lower Germany from the Flavian Period Onwards .... § 16. Probable Position of the Auxilia along the Limes II THE FOUR EMPERORS' YEAR Chapter I. The Principate of Galea Chapter II. The Reign of Otho .... Appendix to Chapter II . Chapter III. Vitellius Ill A MILITARY GAME OF CHESS A Military Game of Chess ..... I 7 18 29 36 49 55 63 70 71 82 90 97 99 114 118 130 158 197 203 269 Map of Roman Germany Map of North Italy. at end of book »» xu The Armies and Frontier Relations of the German Provinces § I. The German Army up to Varus For the purposes of the present inquiry the military history of the German provinces may be said to begin after the defeat of Varus, an event which not only necessitated an entire rearrangement of the German armies on the Rhine, but occasioned a no less complete change of frontier policy in this direction. To under- stand this change, however, as well as to bring together the few scattered notices that we possess as to the com- position of the German armies before that date, a brief r6sum6 must be given of the events which led up to that catastrophe. The conquest of Gaul was ah-eady, when Augustus began his principate, an accom- plished fact, though the work of organization was still to be performed. With this work we are not here concerned, except to say that the whole of Gaul on the division of the provinces in 27 B.C. fell to the share of the princeps, though Narbonensis was after a few years reUnquished to the senate. The original frontier of the Gallic provinces as the result of Caesar's conquest was no doubt the Rhine, and to maintain this frontier against the unconquered and restless German tribes on the right bank, as well as to provide against the possibility of risings in Gaul itself, it was necessary to establish a strong 1 B 2 STUDIES IN ROMAN HISTORY military force, which was placed under the command of an imperial legate, who was responsible both for the civil and military administration of what were ulti- mately called the Tres Galliae. From the outset, no doubt, the legions were placed along the frontier, and the poHcy adopted at first was in principle a defensive one. This, however, by no means excluded the possi- bility of occasional expeditions across the river for the purpose of protecting the interests of Roman traders or punishing the frequent marauding invasions of the German tribes. An incident of this kind took place in 25 B.C. when M. Vinicius crossed the river to avenge the murder of some Roman merchants. ^ Nine years later in 16 B.C. a similar expedition led to more important results. The legate was at this time M. LolUus. a man "pecuniae quam recte faciendi cupidior." Having crossed the Rhine in an expedition against the Sugambri, Usipu and Tencteri, he met with a serious check at the hands of these tribes, leaving in the field the eagle of the legion v.^ The princeps himself was hastily sum- moned to the province, and it was this occasion which probably led to the adoption of a bolder and more aggressive poHcy. similar to that which was being pursued on the Danube. So while Tiberius was engaged in the latter province, Drusus was sent in 13 B.C. as the commander-in-chief of the GalUc legions. Our know- ledge of the campaigns (13-9 bx.) is derived from Dio Cassius, as Velleius hastens on to the exploits of his own hero Tiberius. Without entering into any detailed account of the campaigns, it may briefly be said that his original object was to push back the tribes of the middle Rhine, the Sugambri, Tencteri, Chatti, Cherusci, and > P.C. liii. 26. » VeU. ii. 97. THE ARMIES AND FRONTIER RELATIONS 3 Suevi, and that his plan of operations consisted in attack- ing these tribes from the North Sea by way of the great rivers. It was one result of this plan that the Batavi, in- habiting the island between the two arms of the Rhine at its mouth, were incorporated in the empire, a tribe which continued from this time to contribute numerous efficient auxiliaries to the Roman armies, while the Frisii further along the coast were also united to the empire by a similar but somewhat looser tie. With the help of these tribes the Fossa Drusiana was carried out to facili- tate communication between the Rhine and the Ems, while in the wars which followed, first the interior of the country between the Rhine and Weser and by 9 B.C. Germany up to the Elbe, was practically conquered. The tribes along the Rhine were pushed back from the river, the Sugambri further north, the Chatti further south-east, while the Suevi bodily migrated from the upper Main to the district now called Bohemia, where they afterwards formed the kingdom of Maroboduus and were destined still later as the " regnum Van- nianum " to form a buffer-state between the Danube provinces and the German tribes. The work of Drusus, interrupted by his death in 9 B.C., was carried on and completed by Tiberius, and by 7 B.C. the whole of Ger- many between the Rhine and the Elbe was practically a Roman province. ^ Civil administration was intro- duced,* and the ara Ubiorum was to have been for the new province what the ara Romae et Augusti at Lug- ' Veil. ii. 97 : " Sic perdomuit earn ut in formam stipendiariae redigeret provinciae." Tac. Ann. i. 59; Mon. Anc. v. 10: " Gallias et Hispanias provincias et Germaniam qua includit oceanus a Gadibus usque ad ostium albis fluminis pacavi." * Ann. i. 59 : " Germanos nunquam satis excusaturos quod inter Albim et Rhenum virgas et secures et togam viderint." 4 STUDIES IN ROMAN HISTORY dunum was for the Tres Galliae, the centre of the political cult of Rome and Augustus. As to the strength or constitution of the armies by means of which these results were produced, unfortunately little can be said. We shall see below that Varus was probably in command of six legions, and we may certainly say that Drusus could hardly have had fewer. The bearings of this upon the question as to the original number of legions retained by Augustus, and the difficulties in the way of Mommsen's theory, owing to the necessity of having to subtract twelve' legions, six for Germany, and as many for Illyricum from a total of eighteen, has been discussed elsewhere. 5 Considering that something like a field- army was required during the annexation of Germany and the Danube provinces, it seems more likely that some at least of the eight legions xiii-xx were raised for the purpose of these wars, than that all were hastily levied during the Pannonian rising. But though the limits of the empire were extended to the Elbe, the Rhine stiU remained the mihtary frontier, just as in Pannonia the Drave continued to be the mih- tary frontier long after the Danube was the pohtical one. Accordingly the winter quarters of the legions were on the left bank of the Rhine at Vetera near the mouth of the Lippe,« and at Mogontiacum opposite the mouth of the Main. It is possible that Vindonissa at the junction of the Aare and the Rhine was also at this early date the headquarters of a legion, but, if so, its object was rather to support in case of need the praefectus of Raetia than to defend the Gallic frontier. But naturally » See " Studies in Roman History," chap. xi. • This camp is proved by Hist. iv. 23 to have existed m the time of Augustus. THE ARMIES AND FRONTIER RELATIONS 5 the conquest of Germany involved the planting of mili- tary stations on the right bank of the river. Thus between Vetera and Mogontiacum there was apparently a line of fifty fortresses.' On the upper waters of the Lippe (ad caput Luppiae) Aliso was an important military station where indeed during the second command of Tiberius a legion seems to have been provisionally stationed.8 Further to the south another important fortress was planted in the territory of the Mattiaci, possibly at Saalburg, possibly at Hofheim, castellum in monte Tauno.^ Roads began to connect these forts with one another, and these, together with the canal of Drusus and the pontes longi of Ahenobarbus from the Ems to the lower Rhine did something towards opening out communication in the conquered country. From 6 B.C. to 4 A.D. Tiberius was in retirement, and, imder the legates who succeeded him, no further advance was made in the conquest or organization of Germany. Indeed, when in 4 a.d. he resumed the command, much of his work had to be done over again, and the campaigns of 4-6 A.D. were to a great extent a repetition of his earlier achievements. In the words of Velleius " per lustrata armis tota Germania est. . . denique quod nunquam antea spe conceptum, nedum opere temptatum erat, ad quadringentesimum miharium a Rheno usque ad flumen Albim . . . Romanus cum signis perductus exercitus.^io In 6 A.D. the time seemed to have arrived for making the Elbe the military as well as the political frontier, and at the same time for connecting and compacting the German frontier with the Danube by taking in the ex- ' Fhr. iv. 12. • Veil. ii. 105 ; D.C. Hv. 33 ; Ann. ii. 7. Ann, i. 56 and D C. Uv. 33. w VelL ii. 106. 6 STUDIES IN ROMAN HISTORY tensive kingdom of Maroboduus who, as yet untouched, had at his disposal an army of 70,000 infantry and 4,000 cavahy.^^ The invasion was to come from both sides, from Germany and Illyricum, and all the available forces were to be employed. The German legions— probably six in number— were left under the command of Sentius Saturninus who was to advance through the Hercynian forest, while Tiberius himself, collecting the Danube legions, also six in number, was on the point of marching from Carnuntum.i2 Thus Maroboduus, as he afterwards boasted, was threatened by twelve legions.^^ But the invasion was stopped by a formidable rising of the Pannonians in the rear of Tiberius. The Pannonian and Dalmatian auxiharies deserted the Roman standards, and almost at once an army of 200,000 infantry and 9,000 cavalry was collected, threatening to invade Italy, over- running Macedonia, and undoing at a blow twenty years' work of conquest in Pannonia and Dalmatia.^* The emergency was extreme ; in ten days the enemy might appear before Rome. Tiberius with six legions was quite unable to cope with the rebel forces. Three out of the six German legions probably at once joined his army, more it v/ould have been unsafe to take away. At Rome and in Italy fresh levies were held with frantic haste, probably four new legions were raised xvii-xx. Large numbers of freedmen were in the emergency drafted into cohortes as voluntarii cives Romani, while instructions were sent to Syria and Egypt to send across legions from those provinces. Of the four newly raised legions, three, xvii-xix were despatched to Germany " Veil. iL 109. " Veil. ii. 109. »« Ann. ii.. 46 : " se duodecim legionibus petitum duce Tiberio." ** Veil. ii. no. THE ARMIES AND FRONTIER RELATIONS 7 to fill the places of those sent to Tiberius, while the fourth, XX, without waiting for its full numbers to be made up, joined the army of Tiberius, ^^ who was now accordingly in command of ten legions, while the voluntarii from Italy, and auxiliaries probably sent from Germany and other parts, made up the number of seventy cohortes and ten alae.^® Later on, legions were led up ex trans- marinis provinciis, so that Tiberius was in command of the unprecedented number of fifteen legions.^' § 2. From 9 a.d. to the Recall of Germanicus During the three years in which Tiberius was occupied in putting down this formidable rising, Germany was under the command of QuinctiHus Varus, a man more suited to the otium castrorum which he had experienced as legate of Syria, than to the rough warfare of German campaigns. Looking upon the country as finally conquered he irritated the uncivilized Germans by insisting on the formahties of the Roman law-courts, and played a part more suited to the praetor urbanus in Rome than to the head of an army in the heart of Germany. ^^ The consequence was a conspiracy, against which Varus was warned in vain, organized and headed by Arminius, a prince of the Cherusci. Varus had been passing the " VeU. ii. 112. *• " Junctis exercitibus quique sub Caesare fuerant quique ad eum venerant, contractisque in una castra decern legionibus, septuaginta amplius cohortibus, decern alis, etc.," Veil. ii. 113. " Suet. Tib. 16 : " per quindecim legiones paremque auxili- orum copiam triennio gessit." ** " Ut se praetorem urbanum in foro jus dicere, non in mediis Germaniae finibus exercitui praeesse crederet," Veil. ii. 118. B STUDIES IN ROMAN HISTORY summer of 9 a.d. with no very definite purpose, in the heart of Germany at the head of three legions and some auxiHaries, and was returning towards AUso where a fourth legion was probably posted ; his legate Asprenas was at Mogontiacum, with two other legions. Misled by reports of a local rising, Varus was induced to leave the miUtary road leading to AUso, and soon found his army entangled in impenetrable forest and marsh ; and after several days' desperate resistance to hosts of Germans who knew the country and were under efficient leaders, the whole army, consisting of three legions, three alae and six cohortes, was annihilated. No greater disaster had befallen the Roman armies since the defeat and death of Crassus. Fortunately the remaining troops behaved well. The legion — or part of a legion — at Aliso, under the command of the praefectus cas- trorum, after closing their gates against overwhelming numbers of Germans, succeeded in cutting their way back to the Rhine, 1* while Asprenas with great prompti- tude led up his two legions from Mogontiacum in time to prevent an invasion of the GalUc provinces. 20 The disaster, however, was followed by important consequences. Augustus was old ; he had just passed through the crisis of the Pannonian rebellion ; he had within a year or two materially increased the number of the legions ; he was probably finding the aerarium militare established in 6 a.d. unequal to the strain put upon it. *• Veil. ii. 120. *o Veil, ad loc. cit. : " Reddatur verum L. Asprenati testi- monium, qui legatus sub avunculo suo Varo militans gnava virilique opera duarum legionum, quibus praeerat, exercitum immunem tanta calamitate servavit mature que ad inferiora hiberna descendendo vacillantium etiam cis Rhenum sitarum gentium animos confirmavit." THE ARMIES AND FRONTIER RELATIONS 9 He therefore completely changed his German policy. The ambitious scheme of extending the empire to the Elbe was given up : the kingdom of Maroboduus, instead of being incorporated in the empire, was to serve as a buffer-state between the Danube provinces and the German tribes, while the Rhine was once more con- stituted the poUtical as well as the mihtary frontier of the Empire. But defensive as the Roman policy was henceforth to be, the miUtary prestige of Rome had to be maintained, and above all, the strength of the German armies had to be increased. That the three legions lost with Varus were xvii, xviii, and xix is practically certain. Legion xix is definitely mentioned by Tacitus : ^i we have a funeral inscription to a centurion of xviii, who fell hello Variano,^^ while neither these two legions nor XVII are found again in the Roman miUtary history. As to the other three legions, we are in almost complete uncertainty. Legion v probably remained in Ger- many between the defeat of LoUius and the mutiny of the German legions when we next hear of it, though it may have been one of the legions replaced by the three sent to Varus. Legion i, called in some inscrip- tions Germanica, and which has certainly left no traces in any other province, is mentioned as belonging to the German army after the death of Augustus and may quite possibly have been one of the original German legions. 23 It is perhaps no improbable conjecture that it was the legion at Aliso, which, through losing its eagle in cutting its way through the German forces, was, in consideration of its bravery, reformed and presented 21 Ann. i. 60. ^^ Brambach, 209. *' Ann. i. 42. 10 STUDIES IN ROMAN HISTORY with a new standard by Tiberius.^* At any rate, it seems probable that the three legions at Aliso and Mogontiacum were among the eight which from this time formed the frontier army along the Rhine. This was the number which it was considered henceforward necessary to make up the normal strength of the German army. Augustus, on hearing of the defeat of Varus and the loss of the three legions, immediately held a levy in the city and raised a new legion out of the vernacula multitudo or non-citizen populace of Rome, probably at the same time and from the same source, filling up the gaps in legion i. The new legion was numbered XXI and was afterwards known by the cognomen Rapax.^^ At the same time a number of Galatian troops which had originally belonged to Deiotarus, then to Amyntas, and since the death of the latter had formed part of the non -legionary Roman army in the East, were now by a grant of the Roman civitas made into a legion, and sent to Egypt to supply the place of the legions sent by that province to Tiberius. This legion was numbered xxii, and probably at a later time received in remembrance of its origin, the cognomen, Deiotariana.^^ Fortunately the Pannonian rebellion was over before the rising in Germany took place, 2' so that Augustus, in addition to the two newly-raised legions, had the fifteen under ** See Domaszewski in Westdeutsche Zeitschrift Korrespon- denzhlatt, 1893, P- H5- 2* Tac. Ann. i. 31 : " Vernacula multitudo nuper acto in urbe dilectu." 2* Domaszewski has shown [W.D.Z. Kor. 1891, p. 393) that it was probably not till the enrolment of the new xxii by Claudius that the cognomen Deiotariana was conferred for purposes of distinction. In one inscription the Egyptian xxii is called, like the other Egyptian legion, Cyrenaica. " VelLii. 117. THE ARMIES AND FRONTIER RELATIONS IT Tiberius to distribute over the various frontiers. Of these, seven were required for the Danube provinces ; ^s four others and the new legion xxii had to replace in Syria and Egypt the five which had been brought over from those provinces, while four others, together with the new legion xxi, were sent to reinforce the three already in Germany. From this time the province of Germany, strictly speaking, ceased to exist. But partly to cover this retreat, partly to separate the command over the frontier armies from the administration of the Gallic provinces, the legions were placed under the command of two consular legati, and though stationed on the left or GalUc side of the Rhine, were called the armies of Upper and Lower Germany. The territories occupied by the legions were not provinces in the strict sense of the term until the end of the first century, though they were sometimes incorrectly spoken of as such ; but a strip of territory was ne^^ertheless detached from Gallia Belgica and placed under the civil administration, so far as there was room for anything of the kind, of the legates of the German armies. From this time, too, the command over the German legions was divided. The armies were two and independent of one another, the upper army and the lower, each consisting of four legions, and each under its own legatus.^^ The boun- dary between the two Germanies was the river Vinxt- bach. This is proved by two altars, one found on the north bank dedicated "Finibus et genio loci et Jovi O.M." l)y soldiers of leg. xxx belonging to the lower army, the other on the south bank dedicated by a soldier of leg. viii Aug. which was stationed in the upper pro- ** Ann. iv. 5. " Ann. i. 31 : " duo apud ripam Rheni, exercitus/* etc. THE ARMIES AND FRONTIER RELATIONS 12 STUDIES IN ROMAN HISTORY 13 vince.^ With this agrees the evidence of several mile- stones, e.g. one found between Oberwinter and Remagen marked " a col. Agripp. m.p. xxviii/' and one just beyond Remagen " a col. Agripp. m.p. xxx/'^^ since the distances measured from Koln prove them to have been in the lower province. On the other hand, another milestone found a little north of Brohl is marked a M[ogontiaco], which proves Brohl to have been in the upper province. This alteration came into force at once, though Tiberius from 10-12 a.d. and Germanicus from 13-16 A.D. by virtue of their proconsulate imperium were for the time in command of all the legions. ^2 But though Tiberius was sent at once to the Rhine, and though, in the three campaigns which followed, the Roman armies again penetrated into the heart of Ger- many, the object of the war wais merely to vindicate the prestige of the Roman arms and to crush out of the German tribes any inclination towards an invasion of the empire. In the year 13 a.d. his place was taken by Germanicus, *3 and it is in the next year, after the death of Augustus was made known, that we first get specific information as to the strength and composition of the German armies. The legions had seen hard and continuous service for years ; the summers had been occupied by continual expeditions, the winters by build- ing roads and forts. The length of service had been originally fixed by Augustus at sixteen years, 3* but finding that this imposed too great a burden on the finances, he afterwards changed it to twenty years. ^^ At the end of this time the soldiers were supposed to *® Bramb. 649-50. ^^ Br. 1 930-1. '* Ann. i. 31 : " regimen summae rei penes Germanicum." " Veil. ii. 123. " D.C. liv. 25. " D.C. Iv. 23. receive their honesta missio and the payment of 12,000 sesterces, while they were often planted in colonies, and it was to enable him to carry out this system of honesta missio after a given term that in 6 a.d. he established aerarium militare.^^ But the need of new legions and the critical wars in lUyricum and Germany made these regulations practically a dead letter. The sol- diers were retained, some of them till they had exceeded thirty campaigns : 3' even when they were discharged from the legions and freed from the military oath (exauctoratt) they were retained as veterans sub vexillo ^ I and their pensions were not paid ; and when the honesta missio came at last, the arrangements for colonial I settlements were incomplete and unsatisfactory. The accession of a new princeps seemed a fitting occasion 'to demand the redress of their grievances, and accord- lingly first in Pannonia, and then in Germania inferior, I serious mutinies took place. The four legions of Lower Germany were at this time i and xx stationed apparently I at the oppidum Ubiorum,^^ and v and xxi stationed at Vetera. *« For the time, however, they were all to- j gather in the territory of the Ubii under the command of the legate A. Caecina.*! The sedition commenced among the newly enrolled soldiers of legion xxi ; *2 the |other three legions made common cause and hoped for ^« D.C. Iv. 25 ; Mon. Anc. ii. 34. " Ann. i. 17 and 35. ^^ Possibly each legion had attached to it at this time a body )f 500 veterans. Conf. Ann. iii. 21 and an inscription C.I.L. iii. 2817; trib.(?) veteranorum leg. iv Mac : Tiberius had more than |io,ooo veterans in the Pannonian war. Veil. ii. 1 13. " Ann. i. 37 and 39 ; Br. 377. *® Ann. i. 45 ; Br. 218, 223. " C.I.L. iii. 1698. "» C.I.L. iii. 3272. The inscription relates to vexillationes of the Moesian legions in Thrace probablyat the time of the Thracian rising in 46 a.d. v Maced. and viii Aug. are certainly mentioned, but a hiatus before the former and the number of vexillationes, viz. 1 5, seems to point to three legions. 30 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 31 on account of the Thracian trouble. The normal gar- rison, however, of Moesia was two legions, and after the rising was put down iv Scythica may have been sent to Germany. This view is perhaps strengthened by the fact that somewhere about this time xiii Gem. must have been sent from Upper Germany to Pannonia. Up to 43 A.D. there were three legions here, viii Aug., IX Hisp., and xv Apoll. . In that year ix Hisp. was sent to Britain, while in 46 a.d., if Mommsen's supposi- tion is correct, viii Aug. was in Moesia. As one legion was insufficient for the Pannonian frontier, xiii was probably either at once, or very shortly, sent to take the place of viii Aug. at Poetovio.^^o There is therefore some possibility that, soon after 46 a.d., iv Scythica took the place of xiii Gem. in Upper Germany. ^^i In 58 A.D. upper Germany, according to this view, was reduced to three legions by the sending of iv Scythica to Armenia, but in 66 a.d., as we learn from Josephus,i22 there were again eight legions in Germany. Now in the year 66 123 and 68 i^* Spain was garrisoned by one 1*0 That Poetovio was the camp of viii Aug. is proved by C.I.L. iii. 3845. 4060. 4858. 5220; that XIII gem. was there m 69 A.D. is proved by Tac. Hist. iii. i ; that it must then have been in Pannonia for an appreciable time is shown by the fact that it sided with the other Illyrian legions, first with Otho and then with Vespasian, against the German legions. On the date of the transfer of xiii to Pannonia. see especially Ritterhng de leg. Rom. X. gem. p. 86 foil, and Schultze. de leg. Rom. XIII. gem. p. 34 foil. . ^ J »ai This is Mommsen's view. Res. GesL div. Aug. p. 69. and fom. Gesch. v. p. 120. For other instances in which Moesian troops were sent to Germany or vice versa, see Dipl. Ixvm. in C.I.L. iu. where an ala and two cohorts belonging to the German army are serving in Moesia. 1" B.J. ii. 16, 4. "* Josephus, B.J. loc. cit. »2* Hist. i. 16 ; V. 16 ; Suet. Galb. x. legion only, viz. vi Vict., x Gem. being employed elsewhere* That it was only temporarily absent, however, is shown by the fact that at the end of the year 68 it was again in Spain. 125 j^ is therefore no improbable conjecture that it was sent either in 58, or at some time between 4hat and 66, to Upper Germany, ^^e jj^ ^^iq year 58 we find further signs of restlessness on the part of the Frisii. One division of the tribe, perhaps not that which had been settled by Corbulo in 47, occupied some of the vacant territory on the right bank of the Rhine. ^27 Dubius Avitus, the legate of the lower army, induced their chieftains to lay the matter before Nero in Rome. The emperor's decision was against them, and the Frisii — not without the active intervention of an auxiliary ala— were driven out. But the possession of this empty tract was not even then secured, for another and a stronger tribe— the Ampsivarii— expelled from their own homes by the Chauci, put in a claim to the same territory ; i^s and it was only by the threat of joint action on the part of both upper and lower armies, that the intruders were finally got rid of. 129 We now come to the memorable year 68 a.d. and the events which led to the entire dislocation of the Roman army in Germany as well as in other places. Early in this year Julius Vindex, legate of Lugdunensis and a Gaul by birth, made an attempt to secure the independence >" Hist. ii. 58. "' This is Domaszewski's view, who shows from inscriptions, C.I.L. X. 5829 ; vi. 3538 ; Bramb. 896, that on other occasions the Upper German army was reinforced by the Spanish, Rh. Mus. 47, p. 215. "' Agfos vacuos et militum usui sepositos. "« It had formerly belonged to the Chamavi, then to the Tubantes and the Usipii. *"' Tac, Ann. xiii. 54-56. 32 STUDIES IN ROMAN HISTORY of the Gallic provinces. Collecting an army from the youth of his province, which Plutarch reckons at loo.ooo men. but which might certainly be described hke the Gallic cohorts raised by ViteUius as inter mama belh he tried with some success to secure the support of Galba, legate of Tarraconensis, by assuming the part of Hberator of the Roman world from Nero's tyranny, and then by offering to help that general to the impenal position. He probably also tried to win over Vergmms Rufus, legate of Upper Germany, in the same way. Shortly before this Nero had withdrawn some vextlla- tiones probably from both German armies ^^i for the two expeditions which he was contemplatmg, one against the Alani on the Caspian,^32 and another against the Ethiopians.133 Some of these vexilla had actuaUy got as far as Alexandria, when the news of the GalUc rising was brought. They were hastily recalled to put down the revolt,i3^ and a new legion which Nero had raised from Itahans for his Caspian expedition, i^* was also sent on towards Gaul. These precautions, however, were not necessary. Verginius Rufus had collected the whole of the upper army.^^^ and in view of the formid- able number of the insurgents summoned vextllahones also from the lower army.^^^ With this powerful army 130 Conf. adsertor Hbertatis and the account of D.C. Ixiii. 24 and ^"""1 Certainly from the lower ; see Hist. i. 42. where a soldier of leg. XV is mentioned as in Rome. rr • , : /: i3« Hist i 6. "' Hist. i. 31. " "''^' '• ^: .. 18* Suet. Nero 19. It was afterwards called leg. i Itahca, but at first unofficially, phalanx Alexandri Magni. i« Hist i 5V " bello adversus Vindicem universus adfuerat. 13. This is proved by Hist. i. 51 : " exercitus finibus p-ovin- ciarum discernebantur. Turn adversus Vindicem contractae legiones seque et Gallias expertae " ; i. 53 : " (exercitus superior) THE ARMIES AND FRONTIER RELATIONS 33 he had moved to Vesontio, and in the battle which ensued there Vindex and probably the whole army were destroyed. ^3' The victorious army wished to make Verginius emperor, following the example of the Spanish legion which had similarly saluted Galba. He, however, refused the honour. Meanwhile, the senate had con- firmed the election of Galba, and that emperor marched towards Rome. Distrustful of the German armies, he replaced Verginius by Hordeonius Flaccus, and Fonteius Capito, legate of Lower Germany, whose death he caused, by A. Vitellius.^^s It was probably at this time and by Galba's order that legio x was sent back from Germany to Spain, its place in the upper army being taken by the new legion I Italica, which was, however, for the present, left at Lugdunum ^^^ to maintain order in Gaul after the rest of the army had returned to their camps. The German armies, therefore, were discontented and disaffected. Events had shown " posse principem alibi quam Romae fieri." Hordeonius Flaccus was a nullity, and Vitellius supported by the intrigues of Valens in the lower army (he was legate of legio i) and Caecina in the upper (he was legate of leg. iv) took steps to make himself popular among the soldiers. i*° On January i, 69 A.D., legions iv and xxii at Mogontiacum refused the oath to Galba and went through the meaningless form of swearing allegiance to the senate and Roman in eo ipso sacramento vexillis inferioris Germaniae praeventus erat " ; and Plut. Galba, 10, where Valens, legate of leg. i is clearly with Verginius. *" Hist. i. 51 : " Caeso cum omnibus copiis Julio Vindice : other conflicting accounts are given by Plutarch and Dio Cassius. "» Hist. i. 8, »'» Hist. i. $9. "» Hist. i. 52. ■ 1W 34 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 35 people, 1*1 its auxiliaries sharing the attitude of the legions. 1*2 The lower army which had, though with a bad grace, taken the oath to Galba,i" now at Valens' instigation proclaimed Vitellius emperor, a step which was foUowed a few days later by the upper army ; while legio I ItaUca and the ala Tauriana at Lugdunum also deserted Galba. Steps were at once taken to march into Italy. Valens started with the whole of legio V, a few only being left behind at Vetera, i** and vexilla- Hones of i, xv and xvi probably made up by the Ger- manorum auxilia to almost their full strength. i*^ Cae- cina took with him the whole of xxi Rapax, and similarly strengthened vexilla of iv and xxii. With each legion probably went the whole of its auxiliary alae and co- hortes, so that if these were about equal to the legion- aries, Valens would have 20,000 of the latter and 20,000 of the former, Caecina 15.000 of each, which in fact are the numbers given by Tacitus i, 61, " Valenti ... ad quadraginta milia . . . triginta milia Caecina." The army of Valens, however, which was to march through Gaul and over the Cottian Alps was strengthened on its way by the eight Batavian cohorts temporarily stationed among the Lingones i*« and by the leg. i Itahca and the ala Tauriana at Lugdunum. 1*' These 1" i. 55. 1" i. 54. *" i- 55. '** "• 57. »*» Tacitus says ; " addita utrique Germanorum auxilia c quibus suas quoque copias supplevit." I take this to refer not to auxiliary cohorts, but to the fact that in this emergency the Germans who would naturally have served in the auxUia were drafted into the legions, so that the vexillationes of Valens and Caecina were practically whole legions, though being without the aquilae, technically vexilla. and consisting largely of recruits and peregrini (conf. ii. 21, peregrinum et externum) while the legions left behind in Germany with their aquilae had the places of the vexilla partly supplied from the same source. !*• i. 64. "' i. 64. would make the strength of Valens* army about 54,000, almost double that of Caecina. i*^ Valens, however, had the longer march, and was delayed in Gaul by having to send two cohorts of Tungri and an ala of Treveri to co-operate with the garrison of the Alpes Maritimae against the attack of Otho's fleet. Caecina accordingly reached Italy first, and after being repulsed from Placentia by Spurinna 1*® was defeated — though not decisively— by Paulinus and Celsus. Their army consisting of a vexillatio of leg. xiii from Pannonia, a new legion i Adjutrix raised by Nero from the classiarii of the fleet at Misenum and formerly enrolled as a legion by Galba, six auxiliary cohorts, two alae and three praetorian cohorts, "o could hardly have amoimted to more than half of the German army. Soon after this Valens and Caecina joined their forces and the decisive battle was fought at Bedriacum, ending in the defeat of Otho's army, who had decided to fight without waiting for the main bodies of the four Dalmatian and Pannonian legions. In the battle leg. i Italica especi- ally distinguished itself, I'^i while conflicts took place between xxi Rapax and i Adjutrix, and between v Alauda and xiii Gem.i«2 Meanwhile, Vitellius, who was to follow with a third army, had filled up the gaps in the legions left behind by levies in Gaul and had re- ceived veMla from the three British legions, i" He must also have taken with him the aquila of either leg. xxii or IV Mac. "4 Altogether he started with 60,000 ^** Prope duplicatus, Hist. ii. 30. »*• ii. 21-22. "" ii. 29. "1 ii. 41. 1" ii. 43. "' ii. 57 and 100. Of the 8.000 troops from Britain 6,000 were probably from the legions, and 2,000 from the auxiliaries. "* See ii. 100. Each army on starting had one legion with its aquila, though a second was with Valens owing to the addition 36 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 37 men ; most of these, however, he no doubt— after news of the victory of Bedriacum— sent back, especially the hastily-raised Gallic auxiUaries ingens numerus et prima statim defectione inter inania belli adsumptus}^^ since the Othonian legions were at once removed from Italy, VII Galbiana and xiii Gem. to Pannonia, xi to Dalmatia, I Adjutrix to Spain.i^* and xiv back to Britain. The actual army which eventually marched to Rome with Vitellius consisted of the 8 legions, I2 alae and 34 co- hortes.i" In Italy and Rome these troops naturally became demoralized. The cUmate weakened them, city Hfe corrupted them, and the flower of the legions and alae were drafted into the praetorian cohorts,^^^ so that when the troops had to march against the Flavian army,i5» consisting of the Moesian and Pannonian legions, its efficiency was in no degree commensurate with its size. Two of the legions, i ItaUca and xxi Rapa-x, allowed themselves to be routed at Cremona by some Moesian alae,i«o and the whole army was hope- lessly defeated at the second battle of Bedriacum. i" Tacitus gives a full account of the disposition of the legions engaged in the battle in Hist. iii. 22. The con- quered legions after the battle " ne manenti adhuc civiU bello ambiguae agerent " were dispersed along the lUyrican frontier. i«2 (For a full account of these campaigns, see next essay, chapters ii. and iii.) § 5. The Batavian Rising Meanwhile events were happening in Germany, which of I Italica at Lugdunum ; ii. 89 : " quattuor legionum aquilae per frontoem ttidemque circa e legionibus aliis vexilla." "• ii. 67. "' ii. 89. "• ii. 94- !•• iii. 18. *•* iii. 22-25. »•* iii. 35. "» ii 69. t»» ii, 100. led to an entire reconstruction and redistribution of the two Rhine armies. These armies, as we have seen, had been stripped of their best men by the necessities of the civil war in Italy. Not only had the greater part of the legionary troops been led across the Alps by Valens from Lower, and by Caecina from Upper Germany, but Vitellius had himself followed with all the best of the auxiliary cohorts and alae.-®^ Portions, however, of at least six legions were left behind in the Rhine camps, and these were no doubt strengthened by veterans, serving suh vexillo, and by new recruits from the Gallic states. At Vetera, there were detachments, amounting however to only 5,000 men, of the v Alauda and xv Prim. . At Novaesium was part of XVI, at Bonna part of i, while at Mogon- tiacum in the upper province were detachments of IV Mac. and xxii Prim. . In addition to these, the account of Tacitus shows that there were a number of cohorts and alae, mostly Gallic, some stationed in separate camps below and above Vetera, some quartered in the legionary camps. At Mogontiacum, there was a far more powerful but also more dangerous force, consisting of the eight Batavian cohorts, which had served with legion xiv, first in Upper Germany, and then in Britain. When legion xiv was recalled from Britain, as we have seen, by Nero, these Batavian cohorts, always at feud with the legion, separated from it. They had joined the army of Valens, fought at the first battle of Bedriacum, quarrelled again with the legion at Turin, and finally had been sent back to Germany by VitelUus.^®* *•' Thirty-four cohorts and twelve alae entered Rome with him. Hist. ii. 89. *•* Hist. i. 59 ; ii. 66, 69 and 97. p 38 StUDlES IN ROMAN ttlStORV It was the presence of these efficient but turbulent troops which was the chief cause of the crisis in Ger- many in 69 and 70 a.d. For it was in the home of these Batavians, the so-called insula Batavorum, be- tween the two arms into which the Rhine divides at its mouth, that the first troubles began. The Batavi, originally a branch of the Chatti, together with their neighbours the Cannenefates had first been brought within the empire by Drusus. From the first, they had been in usum proeliorum sepositi, not subjected to tribute, but compelled to contribute their youth to the Roman auxiliary service. Besides the eight or nine cohorts, ^*^ there was at least one ala Batavorum, and Batavians also served in the Rhine fleets. These troops were all commanded by men of their own race,^'* and had done good service in the German and British wars. In the last years of Nero, two brothers, be- longing to the royal race, Julius Paulus and Julius Civilis, were in command of two of the Batavian co- horts. Both incurred the suspicion of Rome, and while Paulus was executed by Fonteius Capito, legate of the lower army, Civilis was sent in chains to Rome. Released by Galba and sent back to his cohort in or near the " island," he was still angry and disaffected, and ready for any chance of revenge. The opportunity came in July 69, when Antonius Primus, whose ac- quaintance he had probably made in Italy, sent letters, urging him to support the cause of Vespasian, and with that end to prevent the Batavian cohorts, which Vitellius had again sent for, from leaving Germany,^*' *•■ The cohort commanded by Civihs himself was apparently not one of the eight serving with legio xiv. »•• iv. 12. "' Hist. iv. 19, THE ARMIES AND FRONTIER RELATIONS 39 and also to occupy the legions by a rising among his own people.^*® A levy was at this very time being carried out by order of VitelUus among the Batavi, and it was not difficult for Civilis to rouse his tribesmen to refuse it. The chiefs were assembled at a sacred ban- quet, and Civihs dexterously exaggerated the hard- ships of their subject position, at the same time pointing to the Roman camps, scantily garrisoned by old men and recruits. Even if the attempt to shake off the Roman yoke should fail, it might at least be repre- sented as a service to the cause of Vespasian.^** Thus a movement began, the nature of which was at first ambiguous. By the Batavi themselves, it was undoubtedly regarded as an attempt to secure freedom from Rome ; by Hordeonius Flaccus, who after the departure of ViteUius was in command of the whole Rhine army, it was looked upon as an in- tervention in the civil war in favour of Vespasian, and as he was himself secretly in sympathy with the same cause, he was at first inclined to ignore it.^'® Whether Civilis from the first imagined himself a Hannibal or a Sertorius, is uncertain. More probably, it was only after his first signal successes that this idea took shape in his mind. The first steps taken were to send en- voys to the Cannenefates, to tamper with the British auxiha, and above all, to secure the adhesion of the eight Batavian cohorts now at Mogontiacimi."^ The Cannenefates under their leader Brenno were the first to move. Sending messengers to rouse the Frisii, they attacked and destroyed the camps of two cohorts, stationed near the ** island," and threatened other camps in the neighbourhood. The praefects of these »•• iv. 13. »•• iv. 14. "• iv. 18. "1 iv. 15. 40 STUDIES IN ROMAN HISTORY. withdrew their cohorts and fired their own camps."* Civilis himself had not yet thrown off the mask, and even offered the cohort which he commanded, to put down the Cannenefates. But this was merely to gain time, while the Batavi and Frisii were arming. The few Roman troops that could be got together, con- sisting of a cohort of Tungri and a number of worthless Nervian auxilia, hastily raised by ViteUius, were con- centrated by a centurion in the upper part of the " island," near the river, with a small fleet to support them. Against this force Civilis advanced with an army of Batavians, Cannenefates and Frisii. In the conflict which followed, the Tungri deserted, the Batavian sailors betrayed the twenty-four ships to the enemy, and the small Roman force was annihi- lated. ^'^ The success, small in itself, opened out wider prospects to Civihs, while it aroused the alarm of Hor- deonius Flaccus. The Batavi began to be regarded as champions of freedom by the other German tribes, and offers of assistance came pouring in. The Gallic states were slower to move, but Civilis did his best to win them over, now by sending back to their homes the praefects of the captured Gallic cohorts, now by inflammatory messages, calling upon them to seize the chance of freedom. Let them recognize their own strength ; it was by means of Batavian and Belgian auxiliaries as much as by their own legions that the Romans had crushed Vindex ; now was the time for the Gallic auxilia to close in upon the few legionaries, and join the Germans in throwing off the yoke of Rome. ^'* When news reached Flaccus at Mogontiacum of stormed camps and captured cohorts, he at once sent orders nt iv. 15. "» iv. 16. 174 IV. 17. THE ARMIES AND FRONTIER RELATIONS 41 to Munius Lupercus, legate of legion xv, to lead out the two legions which he commanded at Vetera against the enemy. Lupercus accordingly strengthened his legionaries with some cohorts of Ubii and an ala Tre- verorum stationed near, and also taking with him some Batavian cavalry, crossed the Rhine and sought the rebel force. Civilis advanced, displaying the standards of the captured cohorts. Again treachery caused the Romans to be defeated. The Batavians turned upon their own side, the Treveri and Ubii fled in confusion, and the two legions were compelled to seek refuge within the camp at Vetera."^ Meanwhile, the messengers of Civilis had reached Mo- gontiacum and invited the veteran cohorts of Batavi and Cannenefates, who had been ordered back to Italy by VitelHus, to assist their fellow-tribesmen at home. As a cloak for mutiny, they demanded of Flaccus a donative, double pay and other privileges, and when these were refused, prepared to march towards Lower Germany. Flaccus in irresolution let them start, then sent orders to Herennius Gallus, legate of legion i at Bonna, bidding him intercept the mutinous troops, and promising himself to hang on their rear. With prompt co-operation, the Batavian cohorts might probably have been crushed, and the subsequent disasters averted, but Flaccus failed to do his part. At Bonna the Batavians were opposed, against the wishes of Gallus, by the legion and a number of Belgian auxilia, but the veteran cohorts, drawn up in deep columns, though inferior in number, easily broke through the Roman Unes, and passing the Colonia Agrippinensis, "8 iv. 18. 42 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 43 without doing further mischief, joined the army of CiviUs."« There could be no doubt now about the urgency of the danger. But Civilis had not entirely thrown off the mask of supporting Vespasian. He caused his own army to take the oath of allegiance to that emperor, and in order to make his attack upon the Roman camp appear merely a part of the civil war, he summoned the legions in Vetera to do the same! When they scornfully refused, he advanced against the fortress, now openly caUing the whole Batavian nation to arms. He was joined too by the Bructeri, a tribe between the Lippe and the upper Ems, and by the Tencteri, to the east of Colonia Agrippinensis, while envoys were sent and not without result to other German tribes further to the south, the Chatti, Usipii and Mattiaci."7 jhe short space during which Civilis was collecting and bringing up his army, was used by the commanders at Vetera in strengthening the walls and fortifications, and in destroying the settlements of merchants and camp followers which had grown up, in modum municipii round the camp. The residents had to be brought within the camp, an unfortunate circumstance, because the place was very imperfectly provisioned. "8 The camp at Vetera dated from the time of Augustus, and had not been built to resist the assault of an invading army so much as to guard the passage of the river, and to serve as a base of operations in case of a frontier war."» Its very size was now a certain disadvantage, since it was built for two full legions, and was now defended by not more than 5,000 men. However the position was far from desperate, 17« IV. 19, 20. 17T IV. 21, 17S IV. 22. 17» IV. 23. as there were four legions and many auxiliaries in the camps higher up the Rhine. StiU the garrison was cut off. CiviUs had brought up some ships, his German allies lined both banks of the river, and his cavalry were masters of the surrounding plain. i^o His army as it advanced to the attack was a motley one. The standards of the veteran cohorts were carried side by side with the effigies of wild beasts, the emblems of the uncivilized German tribes. A desperate assault was made upon the walls, the Batavians even attempt- ing under the instruction of deserters to use siege engines against the place. But the legionaries success- fully beat off the assailants, and Civilis, knowing that there were only provisions in the camp for a few days, and trusting to treachery within the walls, turned the attack into a blockade. ^^^ When news was brought to Mogontiacum that Vetera was besieged, the first act of Flaccus was to send urgent appeals for auxiliaries to the GaUic states. He then made a show of sending forward DiUius Vocula, legate of legion xxii, with a relief party of picked legionaries. But the soldiers were out of control. They suspected their general's loyalty to VitelUus, and these suspicions seemed to be confirmed by letters arriving from Ves- pasian which Flaccus was unable to keep back. Vocula alone could control the men, and it was probably on this account that the whole army marched in one body, Flaccus, an old man and an invalid, accompanying it on board the fleet. 1^2 Reaching Bonna they were joined by legion i, which, having good ground for complaint against Flaccus for not supporting it against »« IV. 22. *" iv. 23. "* iv. 24. 44 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 45 the Batavian cohorts, started a second mutiny. So amid disorder and discontent, the relieving army reached Colon ia Agrippinensis. Here the Gallic auxiliaries began to assemble, but so bitter was the feeling of the soldiers against Flaccus, that he placed the chief direction of affairs in the hands of Vocula. Another stage brought them to Novaesium, where the army was strengthened by legion xvi.^^^ Vocula was now in command of a considerable relief force ; ^^* but the soldiers were not in good temper or good spirits. Pay was scanty and provisions short. The Rhine, too, was unusually low, and this involved the necessity of bringing up all supplies by land, while it made the passage of the river easier for the enemy ; and therefore necessitated extra outpost duty along the banks. 1^^ The outlook indeed became more dangerous and threatening every day. The enemy were no longer only in front. All the north German tribes as far as the Taunus were under arms ; they were on both sides of the river devastating the lands of the Ubii, and even the Treveri. Some cohorts of the former were cut to pieces at Marcodurum, some distance west of the river. Mogontiacum itself was threatened by the Chatti, Usipii and Mattiaci. The Cugemi, a German tribe in Gaul, on the left of the Roman army, rose in revolt,^®' while a force was sent by Civilis across the Mosa, to attack the Menapii "* iv. 26. "* But Tacitus does not tell us what proportion of the two legions from Mogontiacum, or from the other two, had been detached from their headquarters. *** iv. 26. "• The Cugeri were probably the remnants of the Sugambri, brought over the Rhine by Drusus. and Morini in north-east Belgica.i^' All this meant threatened communications, and the relief of Vetera, still the most pressing, ceased to be the only anxiety of Vocula. Under the circumstances, he wisely or unwisely decided against an immediate attack upon the besieging army of CiviUs, and entrenched his army in the camp at Gelduba, half way between Novaesium and Vetera. Here he occupied his army with drill, fortification and entrenchment, and from here he thought it advisable to lead an expedition against the Cugemi on his left.^^^ Flaccus had remained behind at Novaesium, and Herennius Gallus was left in com- mand at Gelduba. When Vocula returned with his expeditionary force, he found the camp once more in a state of frenzied mutiny. This was caused by a disastrous skirmish with the Germans on the river, resulting in the loss of a corn-ship, and the soldiers, always suspicious of their leaders, had put Gallus in chains. It needed all the firmness and severity of Vocula to restore obedience. ^^^ All this time, the garrison at Vetera was gallantly holding out, not only without assistance, but without the news of approaching help. The blockade was varied from time to time by desperate assaults upon the camp, the Batavian veterans having charge of the engines of war, while the German tribesmen were en- couraged to throw themselves recklessly upon the entrenchments. Tacitus gives a graphic description of a desperate night attack upon the camp, and of the methods and counter methods on both sides. 1®® It was at this point that news arrived of the Flavian victory at Cremona. The news did not strengthen "' iv. 28. "• iv. 27. "» iv. 27. "• iv. 29, 3©- 46 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 47 the hands of the Roman generals on the Rhine, though it compelled Civilis to throw off the mask once for all, and openly to wage a war of independence. In the Roman camps, the Gallic auxiliaries without hesitation accepted Vespasian, but it was only sullenly and with reluctance and suspicion that the legions were induced by Flaccus at Novaesium, and Vocula at Gelduba to take the new oath.^*^ Events were precipitated by the determination of Civilis to take the offensive against Vocula. Still maintaining the blockade of Vetera, he sent the Bata- vian cohorts and some picked Germans to surprise the camp at Gelduba. Advancing rapidly, they rushed the camp of an ala at Asciburgium, and fell without warning upon Vocula's army. Vocula, taken by sur- prise, did what he could. The legionaries were formed up in the camp, the cavalry ordered to charge, and the auxilia sent out on the flanks. But the cavalry were driven back, the Nervian cohorts misbehaved, and the enemy were pouring in upon the legions, when providentially some cohorts of Vascones, raised by Galba, arrived upon the scene, and, attacking the Batavians in the rear, converted a disaster into a victory. The enemy, supposing that reinforcements had arrived from Novaesium or Mogontiacum, fled in confusion, suffering heavy loss, but not without some Roman prisoners with them."^ It was not till some days after this, that Vocula at last threw his army upon the besiegers of Vetera. Whether, as Tacitus implies, he might have at once followed up his success by doing this, it is impossible to decide. At any rate, Vocula had left the relief till the very last moment. Civilis had attempted, by displajdng to the besieged the captured standards and soldiers, to make them believe that the relieving force had been defeated, and it was only the devotion of one of the prisoners which prevented the ruse from being successful. When Vocula did reach Vetera, nothing could restrain the impetuosity of his men. In little or no order they rushed against the Germans, the besieged sallied out from all the gates, Civilis was thrown from his horse and supposed to be killed, and the whole German army was driven off.^*^ Vetera was relieved, but Vocula's course was by no means clear. Whether he would have gained much in the end by a more persistent pursuit of the enemy, is very uncertain. Probably he made a mistake in the course which he chose, but his motive must surely have been, not a selfish desire to prolong the war, as Tacitus suggests (non falso suspectus malle helium), but the saving of Roman pres- tige. He determined not to evacuate Vetera, but to relieve it of the non-combatants, to strengthen its fortifications, and to re-provision it. The presence of Vocula and his army was urgently called for higher up the river, but by these means Vetera might be ex- pected to hold out, till he could return for its permanent relief. Complications arose in the defection of the Gallic * states, and the ultimate treason of the legions, which Vocula could not have foreseen at the time, and which frustrated this programme, but it was not in itself an impracticable one. Of two errors of judg- ment, however, Vocula cannot be acquitted. In the first place, he trusted too much to the present dispersal of the Grerman army, and accordingly took insufficient "* iv. 31, 32. lit IV. 33. Its IV. 34. 48 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS precautions to get the convoys of provisions through from Novaesiuifi.i®* One was brought in safely, but a second, owing to the carelessness of the cohorts guarding it, was prevented from reaching Vetera. In the second place, he decided to reinforce his own army by taking i,ooo men from the garrison of Vetera. This half measure probably did more harm in discouraging the garrison than it added strength to his own force, i*^ Having made these arrangements, Vocula led back his army to Gelduba, and then to Novaesium, which was for the time his base. His back was hardly turned when CiviUs was again before Vetera. Soon after, Gelduba fell into his hands, while his cavalry even appeared outside the walls of Novaesium. i»« In Novaesium itself, there was taking place what was perhaps the worst of aU the disgraceful mutinies which characterized the civil war. The legions, know- ing that money had been conveyed into the camp, demanded a donative from Flaccus, who gave it in the name of Vespasian. The soldiers took the money, but in the drunken revels which followed, they re- sented the sUght put upon Vitellius, dragged Flaccus from his tent and murdered him. Vocula escaped the same fate by hiding. The soldiers then replaced the statues of Vitellius, though he was akeady dead, and being without a general, ran considerable risk of allowing the camp to faU into the hands of CiviUs, whose advance guard was outside. Soon, however, the two upper legions, iv and xxii, joined by the legion from Bonna, once more put themselves under w* These had to be conveyed by land, as the enemy were in command of the river. !•» iv. 35. "• iv. 36. ^ - 49 Vocula's command, and he then carried out what had no doubt been his original intention when he left Vetera, and marched hastily up the river to the relief of Mogontiacum, which was being hard pressed by the Chatti, Usipii and Mattiaci. These tribes he succeeded in driving back with severe loss.^®' Still, the situation was sufficiently critical. The Roman bank of the Rhine up to the Moselle was in German hands, the legionary camps of the lower pro- vince, though still untaken, were besieged and isolated, the territory of the Ubii was overrun and the colony threatened, the Treveri, though they had built a stock- ade and rampart to protect themselves, were haid pressed by their German neighbours, while some of the Gallic states in north-east Belgica were also in the power of Civilis. And with all this, there was as yet no hope or sign of reinforcements from Italy, for it was only the end of December, 69, and Mucianus had not yet arrived in Rome. § 6. The Gallic Empire Up to the close of 69, the Gallic states had been passively, and some of them actively loyal. The Treveri in particular had offered strenuous resistance to the German tribes, and the rest, though inchned to regard the recruiting system and the tribute as grievances, had still responded to the appeals of Flaccus for fresh auxiliaries, and had certainly shown no alacrity in following the suggestions of Civilis. But, as earlier events had shown, there was always the possibility of disaffection in Gaul, and the repeated disasters happen- ing to the Roman legions at last began to excite their 1«7 IV. 37- £ n 50 STUDIES IN ROMAN' HISTORY hopes. Nor were these disasters confined to the German armies. Exaggerated reports were spread and be- Ueved that the legions in Pannonia and Moesia wer hemmed in by the Sarmatae and Dacians, and that the British legions were in trouble. The events, too. in Rome, and especially the destruction of the Capitol created a profound impression. Some remnants of the Druids, always the upholders of Celtic nationahty, augured from this that the fates portended empire to the Transalpine nations."' But this unquiet feehng in Gaul might have had no results, if it had not been for the ambitious schemes of a few leading men. These were Julius Classicus, praefect of an ala Treverorum, and descended from the royal family of his tnbe Juhus Tutor, of the same state, and appointed by VitelUus praefect of some of the castella on the lower Rhine and Juhus Sabinus. of the Lingones. a man of httle capacity, but claiming descent from Caesar. These men concerted their schemes in the Coloma Agnppm- ensis On the Treveri and Lingones. both powerful states they could count,"' and messengers were sent Lr and wide among the other states. The passes of the Alps were to be occupied to stop any Roman reinforcements from Italy, the Gallic cohorts and alae were 'to be used against the legions, and the legions themselves, it was believed, if their commanders were removed, might be won over. In this way an independent imperium Galliarum was to be estab- lished "«« But the plan involved, at any rate at the outset, an aUiance with CiviUs, and accordingly im- "' The Lingones could put 70.000 menjnto the field. Front. Sir at eg. iii. 4. H- THE ARMIES AND FRONTIER RELATIONS 51 mediately after the death of Flaccus, messages were interchanged between Classicus and the German leader. At first all this was done in secret ; though Vocula seems to have been informed of it. He, however, was not in a position to act with effect. His legions at Mogontiacum were not to be trusted; they had even shown signs of preferring Civilis to Vespasian. But deeming his presence advisable in the lower pro- vince, where these plots were being concocted, he marched hastily down the river to the Colonia Agrip- pinensis, perhaps taking with him legio i, but almost certainly leaving iv and xxii at Mogontiacum. 201 At Kohl, Vocula, amid the difficulties thickening round him, did what he could. He commissioned Claudius Labeo, formerly praefect of an ala Batavorum and a rival of Civilis, to create a diversion by stirring up the Nervii and Baetasii, and attacking the Can- nenefates, and then, advancing from the camp at Novaesium with legion xvi, probably a few soldiers from legion 1,202 and a force of Gallic auxiliaries under Classicus and Tutor, 203 he made one last attempt to re- lieve Vetera. The catastrophe was now not far off. Near Vetera, Classicus and Tutor, on pretence of scouting, withdrew their force to a separate camp, and made common cause with Civilis. "» iv. 57. Tacitus is very vague as to the forces Vocula had with him. But no mention is made of iv and xxii in con- nexion with the events at Novaesium, and it had been shown that Mogontiacum was not safe from attack. It is clear from iv. 62 that legio i was at Bonna after the death of Vocula, and therefore probably Vocula left the legion, or most of it, at Bonna as he passed down. "' Tacitus vaguely speaks of " legiones." *•' They seem to have allayed his suspicions ; Tacitus says ae was Gallorum fraude inlectus (iv. 58). 52 STUDIES IN ROMAN HISTORY Vocula had no choice but to faU back on Nova^iurn. The place was strong and weU provisioned, and the army might have held out. but it was disaffected and demoralized, the GalUc leaders tampered with it by their emissaries, and in the end, Vocula, after a vain attempt to appeal to the honour of Roman so die^ was murdered, and Novaesium was surrendered to the Gauls Then followed the disgraceful spectacle of a Roman legion taking the oath of allegiance to the Gallic empire, and of Classicus administenng it, decked in the insignia of a Roman general.^* By this disgraceful surrender the fate of Vetera was decided. The gaUant garrison had so far endured aU the cxtiemities of famiivc. but when the last hope of relief >««8 gone, tbc two legions, though with far more ewttse followed the example of their comrades at Novac*ium. took the oath to the Gallic empire, and wrendcred Vctwa to Civilis and CUKjicus In spite of the stipulation that tl«ir Kvea should be *p*red. the army u-as treacherously attacked, after it had evacuatee fight- ing against Cbudius Labeo, that tbc north-eastern sutes of Bdlgka, the Ncrvii. Baetasii and Tungri gave in their adbesioi), not so much to the Gallic empire, as to the German cause Nor wa* it In tlie interest of the Gallk empire that Civilis organised a^ regular ooboits the Gallic tribe of the Suiiuci, to the west of Kobi.** Indeed, the only prominent states which committed thcnttdvcs to an anti-Roman jwlicy were the Trevcri and Lingoncs, 'powerful states no doubt, but neither likely to cede to the other the claim to be head of tl>e new empiie, and botli somewhat isolated from the rest of Gaul, owing to their late action in opposing the attempts of Vindex at a national rising. To the Trcveri both Qassicus and Tutor belonged, and their uifluienoc was great enoij(gh to carr>* the wliole tribe with them. But Julius Sabinus made it evident enough that he intended to claim the •« iv.ba. •^ iv. 66. . ...-^.--^.^^-^-. ~.-,^ 54 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 55 seat of empire for his own state. Causing himself to be saluted as Caesar, he made an attack upon the loyal state of the Sequani. It was his want of success in this expedition which, according to Tacitus, first caused the tide to turn in favour of Rome, and increased the waver- ing among the other GaUic states.a^e The position taken up by two other states was in the one case negatively, in the other positively favourable to Rome. The Ubii were loyal at heart to Rome, though they had been com- pelled to recognize the Gallic empire. But the Tencteri and other German tribes, hating the Ubii as untrue to their nationaUty, pressed for the destruction of the colony, and it was only by the adroit appeal of the Ubii to the arbitration of CiviUs and Vellaeda, the prophetess of the Bructeri, that terms were arranged between them and the Tencteri, which left the colony intact. Its destruction would have damaged Roman prestige even more than the capture of the camps.^w still more im- portant was the action of the Remi. This powerful state and ancient ally of Rome, summoned the con- cilium of the tres galliae to hold a special meeting in their territory to consider the question of peace and war. When it met, the Roman armies were already advancing, and this strengthened the hands of the Remi. In spite of blatant speeches from Valentinus, a young prince of the Treveri, letters were sent nomine Galliarum, to the Treveri and Lingones, bidding them desist from war with Rome.*" By this time, Mucianus had arrived in Rome, and one of his first acts was to provide for the security of the Rhine frontier. Annius Gallus and Petilius CereaUs were chosen as the legates, the former for Upper, the «>• iv. 67. *" iv. 64, 65. •" iv. 67. 68. latter for Lower Germany.212 As many as eight legions were ordered to march into Germany. Of these, only one, XXI Rapax, belonged to the old German army. Possibly a small detachment of the legion may have been left behind at Vindonissa, but if so, it had taken no part in connexion with the rising of Civilis. Three belonged to the victorious army of the Flavians, vii Claud, from Moesia, xi Claud, from Dalmatia, and viii Aug. from Moesia.213 Qne was 11 Adjutix, a legion raised by Antonius Primus out of the classiarii of Ravenna, and then presented with their eagle by Mucianus.2i* Qf the other three, one was famous xiv, which was to cross over from Britain, and the last two were the vi Victr., and i Adj. from Spain.^w § 7. The Campaign of Petilius Cerealis The danger of concerted action on the part of Gaul "* iv. 68. . • X 4- "» The MS. reading is vimxj viij, which most editors interpret as II and 8, but clearly the numbers of three legions underlie the symbols, and the third can only be 7 or 13. since the other Flavian legions are otherwise accounted for. r 21* Hist. iii. 50 : e classicis Ravennatihus, legionariam milt' Ham poscentibus, optimus quisque adsciti. It seems to me that the legion was formed from these Ravenna sailors, and was not the legio e classicis mentioned among the forces of VitelUus in Rome (iu. 55), as Mucianus would naturally give the preference to marines who had served on his side. "» The second Spanish legion is pma in the MS. This, by most editors, is given as decima. no doubt owing to a mistaken inference from v. 19. where the decima ex Hispama appears with Cerealis, but pma clearly represents prima. The legio i Adj has been supposed to have remained in Spain till the rising of Saturninus under Domitian. but Ritterling has lately shown {W.D.Z. 1893, pp. 107, 108) that some of its monuments m Germany date back to 73 (Br. 1141). 74 (Br. 1288). 76 (Br. 1142). and this, together with the MS. reading, appears to be con- clusive. I| 56 STUDIES IN ROMAN HISTORY was removed by the council held at Reims, but neither the Treveri nor the Lingones laid down their arms. There was however little common action among the leaders of the enemy. Civihs was still engaged chasing Labeo through the wilds of Belgica ; Sabinus was in hiding ; Classicus was reposing after his success, as if empire was already won, while Tutor had as yet taken no steps to bar the Alpine passes. The first of the re- inforcements to arrive was legion xxi Rapax, which reached the province by way of its old camp at Vindon- issa. Then came Sextilius FeUx from Noricum with a strong force of auxiUary cohorts by way of Raetia, and an ala singularium, under the command of Julius Briganticus, a nephew of Civilis.2i6 To oppose these troops Tutor strengthened his army of Treveri with some cohorts raised from the Vangiones and Tribocci, and also induced a few of the Roman legionaries at Trier to join him. These latter, however, as well as the German co- horts, in spite of a slight success against a detached cohort of Sextilius, deserted, as the Roman army ad- vanced, and Tutor, avoiding Mogontiacum, took up a position at Bingium, destroying the bridge over the Nava. But SextiUus, crossing the river by a ford, routed and dispersed his army. On this, the remnants of the two legions from Novaesium and Bonna took the oath to Vespasian, and when Tutor and Valentinus again drove the reluctant Treveri to take up arms, the legions withdrew to the friendly tribe of Mediomatrici.^i' At this point, Petilius CereaUs reached Mogontiacum. His own province was Lower Germany, and he was anxious to settle matters with Civilis as soon as possible. But it was necessary to dispose of Tutor and Valentinus tie iv. 70. '1' The chief town was Metz. THE ARMIES AND FRONTIER RELATIONS 57 first, and accordingly in three days' march he reached the Moselle, not far from Trier. His army was not a large one, for by an act of policy, which did much to conciliate the Gauls, he dismissed all the GaUic levies to their homes. He had therefore only legion xxi, the cohorts from Noricum, and some small remnants from the Mainz legions. To these were added similar remnants from the other two Vitellian legions, which joined him at Trier. Valentinus had occupied Rigodulum, a strong position some miles lower down the Moselle than Trier, with a large but inefficient force of Treveri. The place was protected on the flanks by the river and the hills, and was entrenched in front, but Cerealis rushed the position, dispersed the enemy, and next day advanced to the Colonia Treverorum.^i® Restraining the desire of his soldiers to destroy the wealthy colony, he addressed the Treveri in a speech, which whether due to him or to Tacitus, is an admirable exposition of the Roman policy in Gaul and Germany since the time of Caesar.^® But meanwhile, both Civilis and Classicus had become alive to their danger, and were concentrating their forces from every direction in the territory of the Treveri. Cerealis, a general of much dash and capacity, but little caution, hastily entrenched his camp in the plain outside the city, separated from it by the bridge. He was only just in time. The enemy were already occupying the hills in the neighbourhood. Civilis himself would have waited for his Transrhenane allies to join him, but he was overruled by Tutor and Classicus, who pointed out the necessity of striking a blow while Cerealis was still weak, and before the advancing legions should join him. Ac- cordingly, with their army in three divisions, the Bata- SIS IV. 71. ait IV. 73* 74. 58 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 59 m vian cohorts on the right, the Ubii and Lingones in the centre, and the Bnicteri and Tencteri on the left, a desperate night attack was made upon the Roman camp. Some poured down from the hills, others advanced be- tween the road and the river, and while one party seized the bridge the rest forced their way through the entrench- ments into the camp itself. There was all but a fatal disaster, but Cerealis, whose slackness was responsible for the danger, vindicated his reputation. Collecting a handful of men, he first recovered the bridge, then rushing to the camp, now full of the enemy, by example and upbraidings he raUied the legionaries. It was xxi Rapax which saved the day, but probably nothing would have availed had not the enemy been too intent on plundering the camp to complete the work of capturing it. A desperate struggle ended with the complete dis- persal of the German army.^^ The Treveri being now finally subdued, Cerealis pro- ceeded by forced marches to his own province. There was every need for haste, if the Colonia Agrippinensis was to be saved. Civilis was marching towards the city, the ambiguous position of which has aheady been men- tioned, intent on vengeance. Not only had the Agrip- pinenses recently massacred all the Germans resident in their city, but had also treacherously destroyed one of his most warUke and useful cohorts. This was composed of Chauci and Frisii. and had been stationed at Tolbia- cum on the borders of the Ubii. It was fortunate for the colony that Cerealis succeeded in reaching it first. By this time, legio xiv had been conveyed across by the classis britannica. It might have been used, as Civilis feared it would be, to devastate in co-operation "® iv. 77. 78. with the fleet the home of the Batavi themselves. In- stead of this, however, it was employed in Belgica in the somewhat easy task of reducing the Nervii and Tungri. The fleet, left without the support of the legion, was attacked and partially destroyed by the Cannenefates.^^i The war was now entering upon its last stage, but Civilis and Classicus, whose army was again mustering in the north, were still far from crushed. The country was cleared as far down the river as Koln, but no farther. A cavalry outpost, which had ventured as far as Novae- sium, was cut off by Classicus. 222 But at Koln Cerealis was joined by the rest of his legions, xiv from Belgica, II Adjutrix from Italy and vi Victr. from Spain.223 Strengthened by these and some new auxilia, Cerealis advanced against the German army, which was en- camped near Vetera, the scene of its former triumphs. The armies were separated by a wide plain which the arts of Civilis had turned into a vast morass. By throw- ing a dyke obUquely across the Rhine, he had partially flooded the left bank. When the battle began, this placed the Romans at a disadvantage. They were heavily armed, they did not know the ground, and they were disconcerted and even panic-stricken at seeing horses and men losing their feet in the swamp. On the other hand, the Germans were tall in body, they were used to the water, and this semi-naval battle suited them well. The first day's fighting was all in their favour, though they did not venture to attack the Roman camp. To prevent his legions from being discouraged, Cerealis appealed to each, reminding xiv of its conquest of Britain, urging vi not to forget that it had created an emperor, exhorting 11 to deserve its first laurels. "1 iv. 89. "» iv. 89. *" V. 14. J 6o STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 6l Next day, the battle was resumed. Cerealis placed the auxiUary cohorts and alae in the first line, the legions in the second, and some picked troops in the rear for emergencies. Civilis drew up his men not in Une but in column, the Batavians and Cugemi on his right, the Transrhenane tribes on the left, and the Gauls in the centre. The Roman soldiers were more cautious and more on the defensive than on the preceding day, and would not be tempted into the swampy ground. How- ever the Bructeri, by swimming over from the dyke, drove back the auxilia, which had to be supported by the legions. The battle was still uncertain, when a Batavian deserter offered to lead some cavahy round on the left, by a firm path on the edge of the morass, where the Cugerni were off their guard, and so to take the enemy in the rear. The outflanking movement was successfully carried out by two alae, with the result that the whole German army was thrown into confusion and routed. If only the Roman fleet had arrived in time, the campaign might have been ended on that day. Cereahs had ordered the fleet to be brought down, and had counted upon its presence, but he was apt, as Tacitus remarks, not to allow sufficient time for his orders to be executed. As it was, the German army, as well as its leaders, was able to escape across the Rhine.224 Next day, legio xiv was sent to Annius Gallus in the upper province, and its place in the army of Cereahs was taken by x Gemina from Spain. 225 "« V. 17, 18. *** V. 19. Whether xiv was selected for service in the upper army, to prevent it from taking part in the last desperate struggle with its old auxilia. is not certain. It was no doubt at this time that vii Claud, was sent from Upper Germany to Moesia, where Fonteius Agrippa was killed and where rein- m J Civilis was now driven into the " island," the native home of the Batavi. He could hardly cherish any illusions as to the final issue of the war, but for the mo- ment he was safe, as there was no bridge across the Rhine, and Cereahs had not as yet enough ships to make one. To make the passage of the river still more difficult, he cut the dam constructed by Drusus, and this had the effect of widening the left or western arm of the river, which separated the Romans from the Island, and of narrowing the right or eastern channel, which gave access to the Transrhenane tribes. 22« The end could hardly be far off. Cereahs began the construction of a bridge at Batavodurum,^^? where the two arms of the Rhine separate to form the Island, and when the Roman fleet came up, the home of the Batavi would be invaded. For the time, Cereahs had posted legion 11 at Batavodurum, to build the bridge, legion X at Arenacum— perhaps Cleve, while his auxiha were stationed at Grinnes, Vada and other places, "s But Civihs had not given up the struggle. He had indeed evacuated the chief town of the tribe, carrying out of it all that was valuable, but encouraged by fresh contin- gents from the Chauci, he gave Cerealis, whose reputation for slackness and over-confidence was well known, no rest. On one and the same day, four simultaneous attacks were made on the Roman divisions, at Bata- vodurum, Arenacum, Grinnes and Vada. The attacks on the last two forts were all but successful, being led by Civihs and Classicus in person. Cerealis however with picked bodies of cavalry came to the rescue of his forcements were urgently needed. See Joseph. Bell. Jud. vii. 23, 3, and Ritterling, W.D.Z. 1892. p. 114. . v -l ««• V. 19. ■■' Batavodutum was either identical with or at least close to Noviomagus. *** v. 20. 62 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 63 i| auxilia, and compelled the Germans to recross the river, though in the continued absence of the fleet he was unable to press his advantage. Even when the long expected fleet ^id arrive, it had to be fetched by Cerealis himself, and he lost part of it by the way. He had gone up the river, to superintend the rebuilding of the camps at Novaesium and Bonna, and was returning with the fleet, when he allowed him- self to be surprised by a night attack on the part of the enemy. While one division of the Germans threw the camp into confusion by cutting the tent ropes, another succeeded by means of grappling irons in towing off some of the ships, and among them the praetorian vessel which was conveyed up the Lippe as a present to Vellaeda.229 Cerealis however was now able, in spite of an empty naval demonstration made by Civilis at the mouth of the Mosa, to transport his army into the island, and though the autumn rains partially submerged it, and even threatened to wash away the Roman camp, the whole island was ravaged. All classes of the Batavians were now tired of the war, and ready to turn against Civilis, while even Vellaeda was induced by Cerealis to withdraw her influence and support. Of the final scenes of the war we have no knowledge, as the narrative of Tacitus breaks off in the middle of a conference between Cerealis and Civilis. ^^o Nor have we any information as to the campaign which Annius Gallus must have found necessary in Upper Germany. We know that his four legions were i Adj., viii Aug., XI Claud, and after the removal of vii Claud, to Moesia, xiv Gem. . That the Lingones submitted, and ■■• V. 22. «8o V. 22-25. apparently without a struggle, we know from a passage of Frontinus,23i but almost certainly a punitive ex- pedition must have been carried out against the Usipii and Mattiaci, and above all against the still formidable and menacing tribe of the Chatti. But in both provinces as well as in Gaul the power and prestige of the Roman Empire was re-established. § 8. The German Auxilia before 70 a.d. At the beginning of the Flavian era, therefore, Ger- many was garrisoned by an entirely new set of legions. Of the old ones, four whose eagles had remained in Ger- many, and had therefore been disgraced by taking the oath of allegiance to the Gallic empire, were cashiered. ^32 Of the other four legions whose eagles had accompanied them to Italy, i Itahca and probably v Alauda were sent to Moesia, while xxi, and, as we shall see ultimately, XXII were returned to Germany, though at first to the lower instead of the upper army. Before passing on to the Flavian period, it is necessary to cast a glance at the auxiliary troops of the German army in the period before Vespasian. It is especially important to do this, because the war with Civihs was, to a very great extent, a struggle between auxiliaries on the one hand and the legions on the other, and we have good reasons for supposing that the lessons taught by this war occasioned several important changes in "» Strateg. iv. 3, 14. »3« Leg. I never occurs again ; leg. iv Maced. was replaced by a new legion iv Flavia Firma, leg. xvi Gallica by xvi Flavia Felix ; as to the fate of leg. xv Primigenia there is some uncer- tainty. There is no evidence of its existence after this date, but if four legions were cashiered and only three new ones raised 11 Adj., iv F.F., xvi F.F., there would be a diminution in the number of legions which we should hardly have expected. ... . I . A. 64 STUDIES IN ROMAN HISTORY 1 ! #1' I relation to the auxiliary troops not only in Germany but in other provinces as well. We have already seen rea- sons for supposing that the number of auxiliaries was almost equal to that of the legionaries, ^33 and we find that Vitellius entered Rome with twelve alae and thirty- four cohorts. 234 These, however, were the auxiliaries of seven legions, since i Italica had only one ala attached to it, 285 and if we make allowance for this, and for the fact that Vitellius no doubt lost or left some bodies of these troops on his march, 23« we may perhaps assume that there were on an average about two alae and six cohorts for every legion ; 23? or eight alae and twenty- four cohorts for each army. Unfortunately, we have no diplomata for this period, but we are able to gain some information from Tacitus, and possibly some from in- scriptions. To the upper army certainly belonged the ala Petriana ^38 which accompanied Caecina's march. In an inscription at Mainz dated 56 a.d., it is called ala Gallorum Petriana. ^39 It is not mentioned in any of the German Diplomata which begin with 74 a.d., and in 124 A.D. it was in Britain, so that it may probably have been removed from Germany by Vespasian. Another "3 See above, where Valens with four legions had 40,000 men, Caecina with three, 30,000. "* Hist. ii. 89. "* i. 59. «*• Two cohorts were destroyed at Ticinum {Hist. ii. 68). «>» This would give, supposing that one cohort in eight, and one ala in five were miliariae, a total of about 37.000 men. as against the 40,000 legionaries. It, of course, by no means follows from this that the same proportion was always main- tained in Germany, or that it was the same in other provinces or that all the auxilia were definitely assigned to legions. On this question and on the statement of Tacitus with regard to the Batavian cohorts see above, in § 4. "» Hist. i. 70. "• WM.Z. 1883. p. 430- THE ARMIES AND FRONTIER RELATIONS 65 GalHc ala Sebosiana 240 or ala 11 Gallorum Sebosiana «" was in the same army ; an inscription relating to it is found at Worms, 2*2 and it was in Britain by 103 a.d. ; it was, therefore, in all probability, removed about this time. The ala Tauriana 243 ni^gt also have belonged to the upper army : it was afterwards called ala i Flavia Gallorum Tauriana, and has left no further trace in Germany. In addition to these, an ala Agrippiana has left traces at Worms 244 and may with some probability be assigned to the prae-Flavian period. 2*5 An ala Nori- corum may be inferred from two inscriptions of Mainz246 to have served in the province, and as there are no traces of it in the Diplomata it was probably removed in or before 70 a.d. An ala i Hispanorum is mentioned in an inscription of Worms, 2*7 and as it occurs in none of the Diplomata and was in Pannonia early in the second century, its service was probably before Vespasian. With regard to cohorts, we find cohortes of Galli, Lusitani and Britanni belonging to Caecina's army. 2*8 We know that there were Belgae among the auxilia of Verginius.2« A cohors Thracum is also with Caecina, which may have been the cohors iv Thracum which has left traces at Mainz and Wiesbaden 250 and which is not found later in the West, while a cohors vi Thracum was once in the province 251 but was in Pannonia by 84 a.d. Also a co- hors II Biturigum and i Cyrenaica are found in an in- scription of Mainz in close connexion with leg. iv Maced. which was stationed there, and, as we have seen, was »*« Hist. iii. 6. "1 Dipl. 21 in C.I.L. iii. "« Bramb. 894. *" Hist. i. 59. 244 Bramb. 893. '^^ At a later time it was certainly in the East, C.I.L. iii. 600. "• Bramb. 11 18, 1229. «« Bramb. 89. "• Hist. i. 70. "• Hist. iv. 17. "0 Bramb. 1290 and 1523. •" Bramb. 990 and 1009 at Mainz. r m y^i ■■■'I ; 66. STUDIES IN ROMAN HISTORY ml disbanded in 70 A.D., similarly a cohors iii Hi^Pf ^^^"^ of which traces are found at Vindonissa."^ In ad- dition to these, cohors vii Breucorum was at one time in Upper Germany "3 and probably at this period, smce from 85 A.D. we find it in Pannonia."* There was also a cohors AsturumetCallaecorum^^ which may have been the cohort II Asturum et Callaecorum in Pannoma after 80 A.D. A single inscription is found at Mainz of cohors I Lucensium Hispanorum, a cohort which was probably in the lower province under the Flavians, as it was under Trajan ^^^ Similarly cohors iv Dahnatarum, which by 103 A.D. was in Britain, was very Ukely in Germany before that date."' It has aheady been noticed that there were probably Ituraean bowmen in the German armies under Germanicus, and we find traces of a cohors I Ituraeorum at Mainz ^^s which was in Pannoma by 98 and of a cohors i Sagittariorum at Bingerbriick and Kreuznach,26o while an ala Aug. Ituraeorum is mentioned in one inscription of uncertain source, and as it was in Pannonia after 98, may have belonged to this period. Further, cohors v Raetorum and iv Vindelicorum are both found in the earUest Diploma of 74 A.D., and as there were Raeti and Vindehci in the German army under Germanicus ^62 we may prob- ably add them, as well as a cohors Raetorum et Vmdeh- corum, to the auxilia of this period.^*^ A cohors vi «" Insc. Helv. 344. '." Bramb. 897 at Worms. i»* Diploma xii. in C.I.L. iii. «»» Bramb. 1232 at Mainz. "" Bramb. 6b. "7 See Bramb. 741 and 742 at Bingerbruck. and 869 at Bmgen. «" Bramb. 1233-4, 1289, 1099. "• Bramb. 738-9. *" Bonn. Jahrb. Ixxvii. 42. »" Bramb. 2003. "* Ann.u. 17. »•» Bramb. 1235 Mainz. 895 Worms, THE ARMIES AND FRONTIER RELATIONS 67 Raetorum, too, was once at Vindonissa 2«4 and probably belonged to this period, while a cohors i Pannoniorum has left traces at Bingerbriick 265 and Wiesbaden, 28« and as this was in Egypt in 83 a.d., it was probably in Ger- many before Vespasian. A cohors Raetorum too has left traces at Worms, 26? Mainz, 268 and Andernach. 269 Finally we should notice that there were apparently cohorts of Vangiones and Nemetes in the upper army in 50 a.d. 270 With regard to the lower army the ala Picentina was with the lower army during the war with Civilis ; 2'! an ala Batavorum was under the command ' of Lupercus at Vetera. 2^2 An ala Treverorum formed part of Valens' army, 273 and as in 21 a.d. 274 we hear of an ala e Treveris conscripta and apparently under the command of a Treveran named JuHus Indus, we may with some probabihty identify this ala Treverorum with the ala Indiana Gallorum 275 ^hich probably remained in Germania inferior tiU Hadrian's time. 276 An ala Cannenefatium was mentioned in Lower Germany under Tiberius 277 and was probably transferred to the upper army by Vespasian. 278 Jhe ala Frontoniana Tungrorum has left traces in Lower Germany, 279 and as it was in Pannonia by 80 a.d. and afterwards in Britain, its ser- vices in Germany can only have belonged to this period, while the ala Longiniana which has left a number of "* I.H. 344. a«6 Bramb. 740. 741. "• Bramb. 1519 ; Bonn. Jahrb. Ixxvii. 42. "' Bramb. 892. a" Bramb. 1128. "» Bonn. Jahrb. Ixxiii. 15 and Ixxvii. 25. |^'« Ann. xii. 27. "1 Hist. iv. 62. "' ii. 14. "4 Ann. ii. 42 ; conf. 46. *" So described in the Dipl. of 134 a.d iii. 28. «'• Bramb. 307. «" Ann. iv. y^, Bramb. 271. at Novaesium ; conf. also 1994. *" iv. 18. conf. LimesblcUt, "• Dipl. of 74 A.D. 68 STUDIES IN ROMAN HISTORY traces in lower Germany, near Bonn.^^o was possibly here both before and after Vespasian. An ala Pompom is mentioned on a single inscription from Bonn.^i but we can say nothing as to its date. The ala Britanmca mentioned as part of Vitellius' army ^82 may have be- longed either to Upper or Lower Germany, and was prob- ably the ala i Flavia Augusta Britannica which was m Pannonia in the second century. ^^^ Of cohorts, Valens had at least two of Tungri ^84 while a third deserted from the lower army to Civilis.^ss There was a cohort of Batavi commanded by CiviUs and evidently stationed in the territory of the Batavi themselves. 28« There were some Nerviorum cohortes employed against CiviUs «" as weU as some Ubiorum auxiUa ^ss and some tumul- tuariae Belgarum cohortes. ^s^ In addition to these troops, all depending on the evidence of Tacitus, cohors v As- turum may probably have been in the province at this time • 290 a cohors viii Breucorum ^^^ is not afterwards heard of : cohors iii Lusitanorum ^'^ has left traces at Kohl but was in Pannonia early in the second century ; ^'^ tso Bramb. 498 ; W.D.Z. 1891. P- 296 ; 1892. p. 65 ; Bonn Jahrb. xciii. i6. .., rr „ •• ..r Ml Bonn. Jahrb. Ixxxviii. 128. «" Htst. 111. 41- "» Dipl. xxvi. in C.I.L. iu. "* Hist. ii. 14. "» iv. 16 "• That this was not one of the Batavian cohorts is proved bv the fact that while these were still at Mogontiacum, Civihs says : " sese cum cohorte. cui praeerat. Cannenefatem tumultum compressurum " {Hist. iv. 16). M7 Hist. iv. 33. *" iv. 18, 28. •»• Hist iv. 20. These were no doubt not old-established auxilia, but these hasty levies would certainly be drawn from the same materials as the ordinary auxilia. , MO See an inscription at Bonn (Bramb. 478) ; an inscription referring to it has been found on the top of Great St. Bernard. and possibly dates from its march back from Italy m 70 a. d. with the legions mentioned in Hist. iv. 68. {I.H. 35). Ml Bramb. 189 at Cleve ; 232 at Koln. M« Bramb. 312. "• Dipl. xxvi in C./X. iii THE ARMIES AND FRONTIER RELATIONS 69 cohors I Thracum is proved to have been in Lower Ger- many by several inscriptions 294 and was certainly in the upper army after 70 a.d. That this list of the auxilia of the two armies is very incomplete, is, of course, self-evident. We are able, however, in some cases with certainty, in others with more or less probability, to name six alae and twelve or fourteen cohorts for the upper army and 6 alae and 10 or 12 cohorts for the lower. In the following period, the Diplomata will come to our aid for this division of the army. One point, however, is already established for the prae-Flavian period, that the vast majority of the German auxilia consisted of Gauls and Germans, trained and disciplined in the Roman fashion, 296 and employed with the legions near their own homes, and commanded by officers of their own nationality, e.g. Civilis,296 Claudius Labeo,297 juHus Indus, 298 Julius Classicus!299 Thus we have Tungri, Batavi, Cannenefates, Ubii, Nervii, Treveri, Galli and Belgae, while there can be little doubt that Frisii, Sugambri, Baetasii, Lingones and Vangiones also formed part of the auxilia of this period, though we have no definite evidence in these cases. The dangerous rising of the Batavian people, and the Batavian auxili- aries proved clearly enough the dangers of this policy. In an emergency of such a nature the auxilia employed near their own homes were not to be trusted : Tungri, Nervii, Lingones, Frisii, Cannenefates, Batavi and Van- giones all in part or in whole turned against the legions, as the Pannonian auxilia had apparently done in the rising of 6 a.d. The lesson was not overlooked ; from *»* Bramb. 310 and 414 at Koln. and W.D.Z. 1892, p. 28 Remagen ; see also Bonn. Jahrb. Ixxxviii 1 10. *•» Ann. iii. 42. «»• Hist. iv. 16. "' iv. 66. "" Ann. iii. 42. "» Hist. ii. 14. % \i U a 70 STUDIES IN ROMAN HISTORY this time the auxiha were no longer commanded by their own countrymen but by Roman equites, and above all the Diplomata from 74 a.d. to 134 a.d. show us that with the fewest exceptions ^oo the German auxilia came from such provinces as Illyricum or Spain or Raetia. and if from Gaul, only from Aquitania its south-western portion. § 9. The German Legions as placed in 70 a.d. We have seen that of the eight German legions four— I IV Maced., xv Prim, and xvi Gallic were disbanded, and three others-v Alauda, probably xxii Prim., and i Italica certainly ^oi sent to the Danubian provinces ; that An- nius Gallus in Upper Germany had i Adj., xiv Gem. Mart. Vict VIII Aug. and xi Claudia, while PetiUus Cereahs in the lower province had 11 Adj., vi Vict., x G^m. and xxi Rapax. In 71 a.d., however, Cereahs was appointed legate of Britain, and he was probably accompanied to his new province by leg. 11 Adjutrix, which was to take the place of xiv Gem. and which we know to have been stationed at Chester during the Flavian penod. Its place was probably taken by xxii Primigema, which we shall see below was almost certainly in the province by 80 A.D., and which has left numerous traces probably belonging to the Flavian period. ^02 With regard to the 300 coh. Ill Gallorum in 74. which was. however, moved to Moesia by 82 ; ala i Cannenefatium. which was moved away from the neighbourhood of its home to the upper army ; a cohors I Germ. c. R., which does not. however, appear till 116 A.D. »oi Hist. iii. 35, and iii. 46. ... ." See Ritterling, De Legione X Gem. p. 68. This is to a certain extent conjecture, but not so entirely as Pfitzner s view that the legfon re Jained In Pannonia till the rising of Saturninus for its service in the latter province is not attested by epigraphic evidence. THE ARMIES AND FRONTIER RELATIONS 71 headquarters of these legions we may arrive at probable conclusions. No change seems to have been made either by Vespasian or Domitian in the frontier of the lower pro- vince, nor are any wars recorded during this period. During the war with Civilis, Tacitus says that leg. x was stationed at Arenacum, ii Adj. at Batavodurum. The former of these places described as Arenatium in the tabula Peutingeriana and as Harenatium in the Itiner- arium Antonini was probably Cleve ; the latter was either identical with or near to Noviomagus.^o^ xhis arrangement was no doubt only temporary. Leg. x was almost at once permanently posted at Noviomagus, where there are numerous inscriptions relating to it and also tegulae,304 while bricks are found marked l.x.g. in several other parts of Holland. ^os Of ii Adj. no distinct traces are found in the province ; and with regard to XXII Prim, which took its place, there is some doubt whether its headquarters were at Vetera ^os or Novio- magus. ^o? It has also left traces at Calcar,308 Vech- ten,309 and Dormagen.^io In all probability, Vetera was not restored till at any rate some time after the war, ' and two of the legions were moved lower down the river to support the various auxiliary corps stationed at such places as Cleve, Vada, Grinnes, Asciburgium, etc., from any fresh attack from the tribes along the North Sea. The other two legionary camps were certainly Novae- sium and Bonna.^" ^"3 Ritterling, De Leg. X Gem. p. 43. ^®* Bramb. 99, 116, 117, 118-22, 1285, 81. '»» Bramb. 236, 64, 137, 139^, e.g. at Gelduba (Bramb. 2456). *»• Bramb. 4B. iv. 223A ; W.D.Z. Kor. 1883, 5. '"' Bramb. i28d ; conf. also 23^, near Voorburg. ^^^ Br. 185a. 309 Br 50^ 310 Br. 288. '" Hist. V. 22 : " profectus Novaesium Bonnamque ad visenda castra quae hiematuris legionibus erigebantur." I 72 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 73 That XXI Rapax which, during its earlier services in the lower army had been at Vetera, was now posted at Bonna is ahnost certain ; while its inscriptions found at Vetera omit the cognomen Rapax, ^12 those at Bonna usually contain it.^^^ Leg. vi was probably stationed at Novaesium ^^^ though traces of its presence are found in considerable numbers almost all over the province, ^i^ In the upper province we have seen that the legions at the beginning of the Flavian period were i Adj.,3i« VIII Aug., XI Claud, and xiv Gem. Of these i Adj. and XIV were stationed together at Mogontiacum, which camp indeed they probably rebuilt, ^i' xi Claudia was at Vindonissa,3i8 in the neighbourhood of which its tegulae are found in great numbers, e.g. at Brunegg, Dalhken, Granischen, Kloten, Schleitheim, etc.^i^ On the right bank of the Rhine also traces of the legion are found at Heideggerhof,32o Hufingen,32i Beringen,322 Schaffhausen.323 it should be noticed ■1' The omission of the cognomen points to an early date. »" Bramb. 492. 496, 1983, 511c, 1-6. »" Bramb. 264a, 263, 279a. »i» E.g. Cleve, 1446 ; Dormagen. 2886 ; JuUch, 201 ; Bonn. Jahrb. Ixxxviii. 152; Calcar ; Br. 186; Vetera. 223c; Bonn, Jahrb. Ixxxix. 218 ; Koln, 376, 436, etc. »!• C.I.L. ii. 2477 proves that Spain had only one legion in 79 A.D. '" For leg. XIV ; the Mainz inscriptions are very numerous, the greater number, no doubt, belonging to this its second residence in the place. For leg. i Adj. see Bramb. 1091, 1 103-5, 1286, 1288, 1 141-48, and Cagnat. I'ann. ipigr. 1890, 25. »i« See Br. 105 1, 1377^. 1078. 947. 989. 1173-^. 1 181-85, 1196, 1044, 1061, 1177-80, 1 1 19, etc. «" I.H. 344; Bonn. Jahrb. Ixxvii. 194. For inscriptions at Vindonissa itself see I.H. 251, 253-6, 257-8. »" Bonn. Jahrb, i. 21. "^ B.J. Ixxi. 26. »" W.D.Z. Kor. 1887, 2. "» W.D.Z. Kor. 1885, 129. too that tegulae of leg. xi are found at Friedberg,^** and at Bergen near Hanau.^^s Legio viii Aug. was probably at Argentoratum, where it certainly was in the time of Ptolemy, ^^e § 10. Frontier Arrangements under Vespasian and domitian. Under Vespasian and Titus the lower German frontier seems to have been mostly quiet. An obscure reference in a poem of Statins addressed to Rutilius Gallicus,327 has led to the supposition that there may have been an expedition in about 79 a.d. against the Bructeri in con- sequence of v/hich Vellaeda, the prophetess of that tribe, was surrendered to the Romans. ^28 in the upper army, however, there is more distinct evidence of fighting. Under CorneHus Clemens there must have been an ex- pedition probably against the Chatti in 74 a.d. An inscription belonging to that year mentions him as hav- "* Bramb. 1417a; W.D.Z. Kor. 1883, 5. 3" Bramb. 1437a. 32« ii. 9, 9 ; see Bramb. 1884, 18946, conf. also 1904 at Ell near Strasburg ; W.D.Z. Kor. 1885, 141, near the mouth of the Marg. In addition to its traces on the limes, detachments of the legion seem to have been from time to time at Mainz (Bramb. 1377c), Wiesbaden (Br. 151 5, i537«). Nied (1501a), Friedberg (1417a, W.D.Z. Kor. 1883. 5). Baden (Bramb. 1673a), and Rottenberg (Bramb. Append. 42). "7 Silv. i. 4. 90. 3" Conf. Friedlander, Sitt. Gesch. iii. 15, p. 45 1- I gather from Mr. Henderson's book, Civil War and Rebellion in the Roman Empire, note H, that some fresh evidence has come to light about this expedition against the Bructeri. I have not been able to look up the references he gives, but perhaps he will allow me, for the benefit of any of my readers who may care to do so to state them. Conf. E. Ritterling, ap. Korrespondenz- blatt der Westd. Zeitschrift, xxv., 1906, pp. 20-8, and von Domaszewski in Lindenschmit's Heidn. Altert. Bd. v. Heft. vi. (1905). [July, IQ08I. 74 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 75 ing received '* triumphalia ornamenta ob res in Germania prospere gestas."329 w^^t is more certain is that the armies in both provinces were occupied with the rebuilding and strengthening of their hiberna ; ^30 ^jt^ the erection of castella along the frontier and the development of roads. In the lower province the policy of Claudius was still maintained, and all garrisons were on the left bank of the river, while the legionary camps were, as we have seen, Noviomagus, possibly Vetera, Novaesium and Bonna. There were auxiliary camps at Arenacum, Vada, Grinnes, Asciburgium, Gel- duba, Dormagen, and other places, the space between them being guarded by a series of smaller castella, which recent excavations would seem to show were at intervals roughly of about two miles. Two of the castella have recently been brought to light, one at Werthausen, about the distance named from Asciburgium, ^^^ and another at Reckberg near Grimlinghausen, about the same dis- tance from Novaesium. 332 As coins of Vespasian were found in these castella, they were probably built about this time. In Upper Germany the legionary camps were at Mogontiacum, Argentoratum and Vindonissa. But here, while no doubt such places as Speier, Worms, Andernach and several places in the neighbourhood of Vindonissa were occupied by detachments from the legions and by auxiliary troops, the army was not so entirely confined, even from the beginning of this "» Wilm. 1 142. »'» Notice especially the inscriptions at Mainz already alluded to in connexion with leg. i Adj. and xiv Gem., also the evidence to be alluded to shortly of extensive quarries at Brohl worked by vexillarii from the lower army. "1 See Bonn. Jahrb, 93. p. 271. »" W.D.Z. Kor, 1893, P. US- period, to the left bank. 33' The castellum in monte Tauno 334 possibly Saalburg or possibly Hochst on the Main had certainly never been given up ; there was certainly a garrison at Kastel, opposite to Mainz at the mouth of the Main— castellum Mogontiacense— connected with Mainz by a bridge either now or not much later. 335 The Ager Mattiacus, the district be tween the Taunus and the lower Main, was certainly occupied, if not garrisoned, by the Romans under Claudius, for Tacitus 336 teUs us that Curtius Rufus occupied his legions in working the silver mines in agro Mattiaco. He also expressly states that the Mattiaci were in the same relation to the empire as the Batavi,337 and this is confirmed by the existence of cohortes Mattiacorum. That the hot springs at Baden were known and used before the Flavian period seems to follow from Plin. H.N. xxxi. 2. There can be no doubt that the building of castella of which we have seen traces in the lower province was also carried on in the upper, and probably on both sides of the Rhine, as the Chatti were here still dangerous enemies in the immediate neighbourhood of the Taunus region. It was probably against the Chatti that Vespasian was engaged in a war in 74 a.d., probably of some import- ance, but of which we have only epigraphical evidence. An inscription dating from that year, 338 as we have al- 833 So the tegulae of leg. xiv which may probably be referred to the period between 70 and 90 a.d. are found at Gernstein {W DZ. Kor. 1886. p. 55). Hofheim (Bramb. 1503a). Wiesbaden (i537d), Hochst (1502a). Rodelheim (i423«). Nied (15016), Hedderkheim (1491^). and Friedberg (1417c. and W.D.Z Kor 1883, p. 5). Those of leg. i Adj. at Wiesbaden (15 370. and Hed- dernheim (1491a). "* ^^^- "• 4^' „ ^. v 3S6 Tegulae of leg. i Adj. are found there (Bramb. I377«)- 3" Ann. xi. 20. "' Germ. 19. "* ^ilm. 1142. ( ! ■11^ !i 76 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 77 ready seen, speaks of Cornelius Clemens the legate of the upper army as having received triumphalia ornamenta from Vespasian and Titus " ob res in Germania prospere gestas/' To this we may perhaps add that two brothers, Domitius Tullus and Domitius Lucanus, are described under Vespasian as having in successive years held the post of praefectus auxiliorum omnium adversus Germanos. There is some doubt, however, about the years in which they served in Germany, and as Domitius Tullus seems to have been made legate of the army in Africa soon after 70, I am inclined to think that they were either engaged in the war against Civilis, or in some recrudescence of it a year or two later ; at the same time their service may have had some connexion with the expedition under Cornelius Clemens in 74.339 To the south of the Taunus and Main regions we have already seen the ambiguous position of the Neckar valley and the whole of the extensive region known as the agri Decumates.34o Here an approach was certainly made und^r Vespasian towards a Roman occupation. Whether the foundation of Arae Flaviae —Rottweil— is to be ascribed to him or to Domitian is uncertain. It no doubt immediately followed upon the recognition of this district as pars ptovinciae, as the establishment of the imperial cult at once took "• SeeWilm. 1148. 1149. Professor Pclham brings this war into connexion with the occupation of the agri Decumates, to be noticed below. It seems more probable, however, that the two movements were distinct ; that the Chatti threatening the Taunus region were the object of the war. and that the Neckar region, not occupied by Germans and bounded by the friendly Hermanduri, was occupied, whether by Vespasian or Domitian. without fighting. '*" See on page 19. place on the creation of a new province. 3*^ It is cer- tain, however, that a road from Argentoratum to Offen- burg was constructed under Vespasian and was prob- ably continued to Rottweil. This seems to follow from an inscription of 74 on a milestone found at Offen- burg,3*2 which Zangemeister 3*3 j^^s with much prob- ability restored as follows : — Caesare Aug. Domitiano Cos . . . Cn. Cornelio Clemente leg. aug. pro pr. iter ab Argentorato.34* By 98 A.D. it is certain that the whole of the Neckar region was definitely incorporated in the upper pro- vince. This is proved J)y a passage in the Germania, chap. 29.3*5 " Non numeraverim inter Germaniae populos, quamquam trans Rhenum Danuviumque con- sederint, eos qui decumates agros exercent ; levissimus quisque Gallorum et inopia audax dubiae possessionis solum occupavere ; mox limite acto promotisque praesidiis, sinus imperii et pars provinciae habentur." By this time undoubtedly the imperial cult was established at arae Flaviae, and this place connected by roads both with Argentoratum and Vindonissa. •** Conf. the ara Ubiorum for Germany between the Rhine and the Elbe and the templum Claudio for Britain, and note the phrase arx aeternae dominationis in Ann. xiv. 31. »" Bramb. 1955. »" W.D.Z. 1884, p. 246. *** Cornelius Clemens has been referred to above under the year 74 and is also mentioned in the Diploma of that year. Generally, therefore, we may attribute the beginnings of this forward policy on the upper Rhine to Vespasian, but we cannot go so far as to say that the definite incorporation of the Neckar Valley or agri Decumates was his work. Whether the road from Vindonissa to Rottweil was begun by Vespasian is uncertain ; that from Mogontiacum through Heidelberg and Baden to Offenburg was certainly not completed till 100 a.d. under Trajan {W.D.Z. 1884, p. 237). *♦» Published in that year. 78 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 79 The limes referred to in this passage with the praesidia, probably ran up the left bank of the Neckar, and, if so, the fortresses would be Wimpfen, Bockingen, Ben- ningen, Cannstadt, Rottenburg and Rottweil. Whether these owe their origin to Vespasian or Domitian, or whether some of them were later still, it is possible that the investigations of the Limes-Commission will enable us to decide. At any rate, the words of Tacitus are conclusive as to the whole district being now in- cluded in the empire and protected by troops. But if it is uncertain whether the enlargement of the empire south of the Neckar was due to Vespasian or Domitian, it was certainly the latter to whom the important forward movement was due which led to the incor- poration of the whole Taunus district and the Main Valley down to the Neckar. This was the result which followed from the important war carried on by Domitian against the Chatti in 83 a.d. The date of this war is fixed (i) by the title of Germanicus which first occurs in 84 ; 3*« (2) by the fact that in this year the Usipii, not before subject to the empire, and who had been removed from the Lippe district to the east of the Mattiaci under Vespasian ^^^ were enrolled as auxili- aries 3*^ ; (3) by the fact that the triumph was before the recall of Agricola.^*® It was, no doubt, infinitely more important than the ex 'parte statements of Tacitus 350 would lead us to suppose. The war was an aggressive one,^^^ probably a sequel to that »«• Eck. vi. 378, 379. **' Mom. rom. Gesch. v. p. 136. '" Agric. 28. •*• Agric. 39. Conf. also coins with Germania capta dating from 84 to 85 (Eck. vi. 380) ; Cohen, Domitian, 139. 35 1. 483, 488, 503. **® Agric. 39 ; Gerrrt. 37 ; see also D.C. Ixvii. 4. »»i Suet. Dom. 6, under Vespasian, and necessitated an increase in the Upper German army. Legio xxi was almost certainly sent from the lower army. Its presence at Friedberg about this time is rendered very probable by Bergk's explanation of Bramb., 1416, who identifies the Sosius Senecio of the inscription with PHny's friend, consul in 99, and whose mihtary tribuneship would therefore naturally fall about fifteen years earlier. ^^^ A vexillatio, too, of IX Hispana took part in expeditione Germanica,^^^ and the fact that this legion is specially mentioned by Tacitus as being weakened at this time ^^^ makes it probable that this was the expedition alluded to. That the war was followed by some extension of the empire and with new frontier arrangements appears not only from expressions in contemporary poets, ^^^ but from statements, unfortunately not altogether unambiguous, of Frontinus, who probably himself took part in the war. 356 The first of these statements merely says generally that Domitian : — " contusa immanium nati- onum ferocia provinciis consuluit." The second passage is more important : — " limitibus per centum viginti milia passuum actis non mutavit tantum statum belli, sed et subjecit dicioni suae hostes, quorum refugia nudaverat.'* The third statement is " Imperator Caesar Augustus Germanicus eo bello, quo victis hostibus cog- nomen Germanici meruit, cum in finibus Cubiorum castella poneret, pro fructibus locorum, quae vallo **' The tegulae of the legion found at Hochst (Bramb. 15026) Hofheim (15036), Wiesbaden (1537^), Nied (1510c), Friedberg (1416, and W.D.Z. Kor. 1883, p. 5) no doubt belong to this period, and quite lately a tegula of the legion has been found at Saalburg on Taunus (Limesblatt. 9). «" Orell. 3569. »" Agric. 26. "* E.g. Mart. ix. 2, 3 ; ix. 7, i ; Stat. Silv. i i. 51 ; v. 2, 133. «•• Strateg. i. i, 8 ; i. 3, 10 ; ii. 11, 7, 8o STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 8i comprehendebat. pretium solvi jussit/' There can be no doubt that these passages all refer to the war against the Chatti, and therefore that the limites and the cas- tella had direct reference to that war. It is a more difficult question to decide where these limites, extending for 120 Roman miles, ran, and what castella are alluded to, as the name Cubiorum is no doubt corrupt. Ham- meran's view ^s? is that even before this time there was an offensive line facilitating operations against the Chatti from Kastel to Hochst and from there to the Wetterau and Taunus plateau ; the defence of the middle Rhine being based on the bridge existing at Kastel and the fortresses of Hochst, Heddemheim and Friedberg. We can, however, hardly suppose, in the absence of definite evidence, that Domitian's policy was so far anticipated as this would imply, though it may be true that forts of some kind already existed at Hochst, Heddemheim, Friedberg and per- haps Saalburg. The suggestion of Asbach that the 120 Roman miles of Frontinus stretched from Gross Krotzenburg to Lorch may be dismissed at once. Such a line would not have been the protection of the Taunus region against the Chatti which Domitian clearly desired, while, as we shall see below, the line from Miltenburg to Lorch was certainly later in date than the so-called Neckar-Mumling line from Gross Krotzen- burg to Wimpfen. On the whole, it seems probable that Domitian's line started from the Rhine either at Rheinbrohl or at some point to the south of it, and then generally followed the direction of Zugmantel, Saalburg, Capersburg, Friedburg, Oberflorstadt, Mar- kobel, Riickingen, Gross Krotzenburg, and then west by »•' W.D.Z. 1889, p. 287 foil. way of Kesselstadt and Hochst to the Rhine again at Kastel. This would seem approximately to correspond with the distance stated by Frontinus, and would include the Taunus district, the Wetterau and the district on the right bank of the lower Main. It would include the territories of the Mattiaci and the Usipii which are now protected against, and cut off from the dangerous Chatti. Inside this line would be the fortresses of Heddemheim, Wiesbaden and Okarben. It is not suggested necessarily that all the fortresses, afterwards certainly existing at the places named above, owed their origin to Domitian — whether they did or did not, the Limes-Commission will perhaps be able to decide — but only that this was the general direction of the limites and that there were castella along them.^^s in all probability Domitian must have completed his work of extending Upper Germany beyond the Rhine by connecting his Taunus and Main limes with the agri Decumates by fortresses running up the Main past Seligenstadt, Stockstadt, Niedernberg, Obemberg, to Worth, and from there, along what is now called the Neckar-Mumling line to Wimpfen on the Neckar, through Wurzburg, Schlossau, Oberscheidenthal and »" The result of the explorations undertaken by the Limes- Commission is apparently to show that Domitian's line left the Rhine near Ems and then passing round the north-western and northern portions of the Taunus plateau, taking in the Wetterau, joined the Main at Kesselstadt, and then ran down to the Rhine . The fortresses along the limes, it seems, were small earth forts only about 85 metres square, with burgi or wooden watchtowers between them. The real defensive force was posted behind this limes in a series of large stone base forts at Heddemheim, Okarben, Friedberg, connected by roads with the small earth forts on the frontier. See Professor Pelham, Transactions of the Royal Historical Society, 1906, p. 29 [July, 1908]. I 82 STUDIES IN ROMAN HISTORY Neckarburken. The fortresses at these places may not all have owed their origin to Domitian, but if they are later, must in most cases have replaced earUer ones.^^® § II. The Rising of Saturninus and Changes Caused by it From 83 A.D., as we have seen, the lower army was composed of three legions— vi, x, and xxii, while the upper army had five— i, viii, xi, xiv, and xxi ; and it is no doubt to this period that the tegulae are to be referred found at Mirebeau near Dijon, and containing the names of these five legions. 3«o Another tegula found in the same place contained the names only of VIII, XI, XIV and xxi ; from which Ritterling has in- ferred with some plausibility that at some time between 83 and 89, when the rising of Saturninus caused further changes, i Adj. was removed from Germany. 3«i It is quite probable that this was the case. In 87 A.D. Cornelius Fuscus was defeated and killed by the Da- cians,3«2 and as we know from D.C. Ixviii. 9, the eagle of a legion was lost.3«3 it was ahnost certainly at this 3»» One of the consequences of this rectification of the German frontier was probably the estabhshment of the two Germanies as provinces in the strict and technical sense of the term. The first official description of them as such comes under Hadrian (Mommsen in Sack Berichten, 1852. p. 233). »•« See Hermes, 19. p. 437. where Mommsen refers them to the year 70 a.d. when Cerealis was assembling his army ; but see Ritterling. W.D.Z. 1893. P- i^^. . . ^v, * s«i His argument depends mainly on the supposition tnat when detachments were sent as vexillationes from more than one legion in a province each legion contributed a share ; see he. cit, p. 117, note 38 and 39. '" Suet. Dom. 6. »" The legion destroyed on the occasion was probably v Alauda. THE ARMIES AND FRONTIER RELATIONS 83 time that Domitian created a new legion i Minervia.^s* which, however, he sent, not to the Danube where veteran legions were required, but to make up the four legions in Lower Germany ; 365 ^^dlQ i Adj. may have been sent to reinforce the army of the Danube. The movements of leg. i Adj. at this period are somewhat obscure. It may have been sent to the Danube at this time as Ritterling suggests, but if so, it was cer- tainly back again in Upper Germany by the time of Trajan, for we find this legion and xi Claudia mentioned on an inscription at Baden-Baden as dedicating a build- ing to Trajan. 366 Mommsen supposes that it had been sent back to Spain by Vespasian, and came to Upper Germany from that province at the time of the rising of Saturninus. It was certainly finally removed with XI Claudia to the Danube early in the second century. 36? At any rate, towards the end of m a.d. an event hap- pened which caused important changes in the German armies. The legate of the upper army was L. Antonius Saturninus, who, taking advantage of the fact that his legions were discontented and disgusted with the build- ing operations in the limes, caused himself to be pro- claimed imperator by the two legions at Mogontia- cum.368 Few details of the rising are known from '•^ D.C. Hv. 23. "« It was certainly there in 89 ; see below. '" Bramb. 1666. "' Ritterling, De Leg. X Gem. p. 72, argues from an inscription given in Bonn. Jahrb. 77, p. 70 that leg. i Minervia was created not later than 83 a.d. Schilling, however, proves the insuf- ficiency of the argument, and shows that new legions were as a rule created to supply the loss of old ones. Suet. Dom. 7: "L. Antonius apud duarum legionum hiberna res novas moliens." The two legions were xiv Gem. and XXI Rapax. For inscriptions relating to xxi Rapax at Mainz, see Bramb. 1057, 1206-7. 84 STUDIES TN ROMAN HISTORY historians. Great alarm was felt in Rome. 3«9 Domitian himself started for the seat of the war, probably with some praetorian cohorts; but almost immediately received intelligence that the rebellion was put down. 3'° Saturninus had entered into communication with some Germans across the Rhine, probably the Chatti, and only the sudden melting of the ice prevented them from crossing the river and joining him.^'i Meanwhile Appius Norbanus, the legate of some neighbounng province, had arrived on the scene. Saturninus, m the battle which followed, was defeated and killed. ^'^ Very different views have been held both as to the legions which joined Saturninus and as to the province from which Appius Norbanus marched against him. That the two legions at Mainz joined him seems clear from the words of Suetonius,^'^ but this alone would not account for the panic at Rome or for the hasty departure of Domitian. There were two other legions in the province at Argentoratum and Vindonissa, and if these had remained faithful, to say nothing of the Lower German army, and the Pannonian legions, which could easily have stopped his march into Italy, there could have been no occasion for panic. Besides, on occasions like this, the legions of a province usually made common cause ; all would share in the prestige of making an emperor, and all, too, had the same cause »•• Plut. vita Aemil. Paul. 25. a'o The date fixed by the Acta Frat. Arv. Henzen, p. 12 1-2, which shows that Domitian started from Rome on January 17. 89 A.D.. and that the victory was celebrated on the 24th or 2 5tn. "1 Suet. Dom. 6. »" That Appius Norbanus put down the rismg is proved Dy D.C. 67, II ; Aur. Vict. cap. n ; Mart. ix. 85. and by C./.L. vi. 1347 where Appius Maximus is called " confector belU Oer THE ARMIES AND FRONTIER RELATIONS 85 mamci. "» Dom. 6. for disaffection. Whether legions viii and xi had actually time to join his standard may perhaps be doubtful, but that they were considered both by Saturni- nus and Domitian as part of the rebel army, must cer- tainly be assumed. From where did Appius Norbanus march ? There can only be two alternatives — from Pannonia or from Lower Germany. Bergk, depend- ing on Mart. ix. 85, thinks he was procurator of Raetia, an office not belonging to the senatorial cursus honorum at all. Asbach, that he was legate of Lugdunensis, Schiller of Aquitania, both inermes provinciae ; Momm- sen 374 followed by Liebenam ^75 considers that the refer- ence in Martial ^^^ is conclusive in favour of Pannonia. ^'^ It must be admitted that, in the absence of positive and negative arguments against it, the passage, though still not easy to explain, is somewhat in favour of this view. But it must, nevertheless, be given up. The argument from Martial depends on the supposition that Norbanus marched through Noricum and Raetia, and that the battle took place somewhere near Vindonissa. But in this case he would have had to march more than twice as far as Saturninus, while some days would have elapsed before the news of the rising could reach him, after which he would have to collect his legions. It is hardly conceivable that if Saturninus was marching into Italy at all, he would not have crossed the Alps "* rotn. Gesch. v. p. 137. *'* Die Legaten, p. 213. ''• ix. 85 : " Cimi tua sacrilegos contra, Norbane, furores Staret pro domino Caesare sancta fides, Haec ego Pieria ludebam tutus in umbra, lUe tuae cultor notus amicitiae. Me tibi Vin- delicis raptum narrabat in oris, Nescia nee nostri nominis Arctos erat. •" See also Domaszewski in W.D.Z. Kor. 1892, 73. 86 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 87 before the arrival of Norbanus at Vindonissa.^'^ Again, there could have been no more than three legions in Pannonia at this time, and considering the unsettled state of the Danube frontier, and that the Dacian war was hardly over, we can scarcely beheve that all of these could be employed, and even if we add the Raetian auxiharies, the army could hardly be a match for the united forces of Upper Germany; while to suppose that two of the German legions joined him is an assump- tion without evidence, and in itself unlikely. In the next place, the German alUes of Saturninus were clearly only separated from him by the river at the time when the battle was fought ; ^'^ but it is hardly conceivable that the Chatti who till the thaw came might have joined Saturninus anywhere between Mogontiacum and Vindonissa, would have marched all the way from their own home on the right bank of the river. Nor is it at all probable that they would have joined him in a march southward at all. Finally, it is impossible to leave out of account the army of Lower Germany. If that had joined Saturninus, Norbanus with the Pannonian army would certainly have been unable to put down the rising as promptly as it was put down ; if, on the other hand, it was against him, Saturninus by every rule of prudence and strategy must have been prevented from marching towards Italy until he had either won over or conquered the lower army. Fortunately, we are 3" The natural plan for Norbanus would be to march to North Italy to intercept the rebel army, as the Pannonian legions perhaps intended to do in 69 ; but then the passage in Martial has no bearing on the subject. 3'» Suet. Dom, 6 : " cum ipsa dimicationis hora resolutus repente Rhenus transituras ad Antonium copias barbarorum inhibuisset." not without evidence on this point. The Lower German legions at this time were i Minervia, vi Val. Victr., x Gem. and xxii Prim., and with regard to all of these Ritterling ^^ has pointed out that in certain inscrip- tions they are described with the letters p.f.D. after them. Thus we have leg. i Min. p.f.D. ; ^^ leg. vi, Vic. p.D. ; 382 leg. x G. p.f.D. ; 383 jeg. xxii P. p.f.D. ^s* Now in no inscription earlier than Domitian's reign are any of these legions styled p.f., while in inscriptions and tegulae of Trajan's reign the letters are assigned to all of them. 385 But there are only three other legions which are styled pia fidelis before Trajan's reign, leg. vii Claudia p.f., leg. xi, Claudia p.f. and leg. 11 Adj. p.f. ; and of these the two former received the title owing to their fidelity towards Claudius in the rising of Camillus Scribonianus in Dalmatia, while no other legions except these four are styled p.f.D. That D stands for Domiti- ana is both probable in itself, and receives some support from the analogy of legions vii and xi, which often have the letters C.p.f. after them, and from that of VIII Aug. which is styled 38« pia fidelis constans Com- moda. It seems therefore in the highest degree prob- able that these four legions received the honorary title pia fidelis Domitiana for some important service ren- dered in Domitian's reign; and the analogy of the grant to legions vii and xi, as well as the fact that there was no war in Lower Germany at the time, seem to point unmistakably to the rising of Saturninus. ^^^ 380 De Leg. X Gem. pp. 120-1, and W.D.Z. 1893, pp. 203 foil. 381 Bonn. Jahrb. 57, p. 70. '" Bramb. 1892. 383 Br. 651. 3" Br. 673, 1626. 385 Ritterling, De Leg X Gem. p. 120-1. "* Wilm. 1459 38' The fact that D is not found in connexion with the legions in later times is, of course, due to the damnatio memoriae passed M 88 STUDIES IN ROMAN HISTORY But not only were these four legions styled p.f.D. In Bramb. 684 we have classis Germanica — the fleet, it is well known, belonged to Lower Germany — p.f.D.,^^^ in Br. 678 we have coh. 11 Asturum p.f.D., ^^' and in Br. 676 we have coh. 11 c.R. p.f.D. ; ^^ while finally two alae, the ala Indiana and the ala i Singularium, and at least two other cohorts — coh. i c.R. and coh. v Lucensium — certainly belonging to the lower army in Flavian times are styled p.f.^*^ When it is remem- bered that out of all the other auxiliary forces of the empire only one ala and five cohorts were styled p.f., the fact that at least two alae and four cohorts in Lower Germany are so styled, together with the four legions and the fleet, seems to make it almost certain that it was the lower army which defeated Saturninus and that Appius Norbanus was its legate. In all probability a second war with the Chatti was the necessary comple- tion of these events, ^^s Norbanus probably succeeded Saturninus as legate of the upper army.^^a It is prob- on Domitian by the senate. If Ritterling's view is correct all the inscriptions which have this letter date between 89 and 96 A.D. It appears from an inscription, as Mommsen points out {W.D.Z. Kor. 1886, p. 171) that on March 23, 86 a.d. leg. XXII was not called p.f. *** See also Bonn. Jahrb. Ixxi. pp. 107-9 and Ixxviii. p. 137. '*» That this belonged to Lower Germany is proved by Br. 666, C.I.L. ii. 1702, and W.D.Z. 1885, p. 222. »•» For this as part of the lower army see C.T.L. ix. 2958 ; Bonn. Jahrb. Ixxvii. p. 19. "1 Ritterling adds ala 11 Flavia mil. ; coh. iii Delmatarum and coh. 11 Hispanorum, but their attribution to the lower army is only conjectural. '»* Thus Suet. Dom. 6 says : "de Catthis Dacisque post varia proelia duplicem triumphum egit " ; and the interval between the war of 83 and the Dacian war was too great for the triumphs to have been celebrated together. ••' This is how Mommsen explains the tegulae found at THE ARMIES AND FRONTIER RELATIONS 89 able that the two legions at Mainz which commenced the revolt, xiv and xxi, were removed from the pro- vince and sent to Pannonia.^^* But leg. xxii Prim, seems to have been sent to Mogontiacum from the lower army ; ^^^ while the upper army was also strength- ened for the time by vii Gem. which Trajan hastily brought up from Spain. ^^^ In the last six years of Domitian the Danube was the scene of the greatest danger, and the German provinces were probably left with three legions each — i Min., vi Victr., and x Gem. in the lower ; xxii Prim., viii Aug., and xi Claud, in the upper province.^®' Mirebeau in the territory of the Lingones marked leg. viii Aug. L. Appio leg. {Hermes, 19, p. 438). ^•* Leg. XIV was certainly in Pannonia before the end of the first century {W.D.Z. Kor. 1891, p. 88). It is perhaps in favour of the supposition that leg. xiv was removed as early as this from the province that it has left no traces in the limes, Fried- berg being the furthest point eastwards where its tegulae are found, and leg. xxi was probably the legion destroyed by the Sarmatae in 92 a.d. (Suet. Dom. 6). 3»* It was in Upper Germany before Domitian's death, as is proved by Bramb. 1626 from Alpirsbach where it is styled p.f.D. ; see also a tegula, Bramb. iS77S, 31 ; it is proved to have been there in 97 by a comparison of Hadrian's cursus honorum {C.I.L. iii. 550) with Spartian {Hadr. 2 § 5) ; it is to this period of its service with the upper army that its tegulae and inscribed stones belong, found at Wiesbaden (Bramb. 1537), Florsheim. (1506), Hofheim (1503c), Nied (1501(f), Heddern- heim (1491c), and Friedberg {W.D.Z. Kor. 1883, p. 5). Some of its principal inscriptions at Mainz are Br. i377g, 1308--9, 1075, 1210, 1215, 1217, 1221, 1341-2, etc. '•• Plin. Paneg. 14, speaks of " legiones " ; but if the view taken above as of leg. i Adj. is correct, there could have been only one legion in Spain after 70 a.d. Traces of the presence of vii Gem. in Germany are found in Bramb. ^96, 15 12, and Henz. 6701, though it is not certain that they belong to this period. ^•' From 89 A.D. no hiberna were allowed to contain more than a single legion ; Suet. Dom. 7 ; " Geminari legionum castra prohibuit." 90 STUDIES IN ROMiWrHISTORY § 12. The Upper German Limes That in the years which followed, the work of forming the limes on the right bank of the Rhine was continued by the upper army, there is no reason to doubt, though we have no positive evidence on the subject, and it was probably this work which occupied Trajan there in 97 A.D., when he was made legate of the upper army by Nerva. It is not the object of this paper to enter into any minute account of the details of the Upper German Hmes. The whole question is in fact at the present time sub judice, as the Imperial Limes-Com- mission has been for the past year investigating the whole subject, and is as yet far from having concluded its task. Its investigations will, in all probability, throw much Ught on the date and origin of particular fortresses, and may possibly even enable the future historian to distinguish between the work of Domitian, Trajan, Hadrian and Antoninus Pius. The subject dealt with in my paper is rather the German army and its distribution than the strategic objects of the limes on their technical side, and enough is aheady known to show the general system of defence and precaution which the limes must have been intended to secure. 3®^ We have seen that Domitian was the first to under- take any regular rectification of the Upper German frontier, and that his work was probably a double one. In the first place he estabhshed some sort of limes for the Neckar district or agri Decumates, probably estab- lishing his castella mostly, if not entirely, along the 3»« The results of the Limes-Commission are being pubhshed m a Limesblatt. of the first six parts of which, up to December, 1893, I have made what use was possible. THE ARMIES AND FRONTIER RELATIONS 91 Neckar itself. In the second place, he took measures to protect the newly incorporated Taunus and Main region, probably forming a limes and establishing fortresses along an arc of about 120 miles from Kastel to the neighbourhood of Rheinbrohl. There can be no doubt that his castella were connected with one another by limites in the sense of boundary roads, but it would appear from Frontinus that a vallum or entrenched earthwork was a part of his arrangement. ^^^ That the greater part of the vallum traceable outside the castella of the Umes is far later than Domitian there is no doubt, but as Frontinus was a contem- porary writer we seem bound to assume that along some part of Domitian's line there was a vallum, and I should suggest that it was at those points where the nature of the country made inroads on the part of the Chatti most likely.*^ Whether, as suggested above, the connexion between these two limites, i.e., the series of fortresses from Gross Krotzenburg to Worth and the Neckar-Mumling line from Worth to Wimpfen, was also the work of Domitian, is quite uncertain ; but considering that there were thirteen years between the war with the Chatti and his death, and that such a connexion between the two was almost imperative, the suggestion seems at least probable. How far Trajan developed or improved upon the work of Domitian we cannot say, but it is tolerably safe to attribute some 3»» Conf. the expression locorum quae vallo comprehendehat. 400 It seems to me beyond doubt that the phrase " limite acto " in Germ. 29, refers not, as Furneaux seems to take it, to the " German limes," but only to the limes established for the agri Decumates. On the other hand, the " limitibus actis " of Frontinus refers as distinctly to the Umes established for the Taunus district. THE ARMIES AND FRONTIER RELATIONS 93 92 STUDIES IN ROMAN HISTORY important developments to Hadrian. He probably not only rebuilt, enlarged, and strengthened many of Domitian's fortresses, but to some extent also enlarged the circuit, especially on its northern side. If we are to keep to the 120 miles of Frontinus, it is impossible to suppose that his line ran as far north as Butzbach and Arnsburg, but both those places belong to the limes as we know it, and as Hadrian is known to have been peculiarly active on the frontiers, the extension is probably to be ascribed to him. From his time, therefore, it would seem that the limes— still a frontier hne and not a continuous earthwork— started from Rheinbrohl, *oi the northern extremity of the upper province, and ran south-east past Niederbieber *02 to Augst ; then crossed the Lahn near Ems— here there were two distinct fortresses —and still in the same direction, passed Holzhausen and Kemel. At the head waters of the Aar it ran along the northern slope of the Taunus plateau, passing Zugmantel, and then ascended the mountain peak to Alteburg near Heftrich, and then to Feldburg and Saalburg.*o3 It then passed through Capersburg, where it probably deviated from Domitian's Hne, and then due north to Langenhain— a large camp 163 x 201m.— near which, but some httle distance within the limes, was the impor- tant fortress of Butzbach. Still further north it reached Griiningen, a small camp standing at the extreme north of the limes. From here it turned to the south-east "1 For inscriptions of Rheinbrohl see C.I.L. xiii. ch. m especially 7691-7730. ^ ' *" The fortress was probably rebuilt and enlarged at a later time ; see C.I.L. xiii. ch. 136. «» The camp here seems to date from or to have been rebuilt by Hadrian. 1 I and crossed the Wetter at Arnsburg, the Horloff at Inheiden, passing Echzell, the Nidda at Oberflorstadt and the Nidder at Altenstadt, from where it ran south, past Markobel, across the Kinzig at Ruckingen, to Gross Krotzenburg on the Main. These last tWo were backed by the fortress of Hanau on the Main. From Gross Krotzenburg, as we have seen, the fortresses are on the left bank of the Main at Seligenstadt, Stock- stadt, Niedernberg, Obernberg at the mouth of the Mumling— while Aschaffenburg on the right bank was possibly also a camp— Worth, Trennfurt, and lastly at Miltenburg. From Worth there begins a double line of fortresses, of which the inner one, the so-called Neckar-Mumling line, may, as we have suggested, have been due either to Domitian or to Trajan. It ran in a southerly direc- tion past Vielbrunn, Eulbach, Wurtzburg, Schlossau, Oberscheidenthal, Neckarburken, Gundelsheim, then to Wimpfen on the Neckar. At this point the chain of fortresses was'connected with those dating probably from Domitian which were placed along the Neckar and which formed the Umes of the agri Decumates— Bockingen, Wahlheim, Benningen, Cannstadt, Kongen, Rottenburg and Rottweil. Whether this was the extent of the frontier line as left by Hadrian, it is perhaps impossible to decide with certainty, but in all probability the so-called limes Raeticus necessitated by the extension of Raetia beyond the Danube— an extension probably to be assigned to the time of Trajan— called for some further modifica- tion. This limes Raeticus ran from Kelheim on the Danube in a north-westerly direction, and then passing Gunzenhausen, turned slightly south-west, till it reached 94 STUDIES IN ROMAN HISTORY Lorch on the Rems. With the limes Raeticus itself we are not directly concerned, but the advisability of connecting the German limes with the Raetian led either Hadrian or Antoninus Pius to give a new direction to the former by running a second line of fortresses from Worth to Walldum and from there in a bee hne almost parallel to the Neckar-Mumling hne to Lorch, where it joined the Raetian hmes. The fortresses along this line were at Walldurn, Osterburken, Jags- thausen, Oehringen, Mainhardt, Murrhardt and Welzheim. Possibly the Limes-Commission will decide the ques- tion whether this extension of the German Hmes was due to Hadrian or to Antoninus Pius, the difficulty at present is that there are so few dated inscriptions. The earhest from the Neckar-MumHng hne was, till lately, 148 a.d., at Bockingen,-**** but quite recently a mihtary diploma has been found at Neckarburken,*®^ dated 134 a.d. The earhest from the Miltenburg-Lorch Hne is 169 a.d. at Oehringen. *o« But whether this outer hne was the work of Hadrian or not, it seems extremely probable from a passage of Spartianus ^"^ that Hadrian con- structed some sort of wooden palisade along the whole length of the limes as he left it, excepting those parts where the fortresses lay along either the Main or the Neckar : "In plurimis locis in quibus barbari non fluminibus sed hmitibus dividuntur, stipitibus magnis in modum muralis saepis funditus jactis at que connexis barbaros separavit." This passage of Spartian would seem to be a piece of negative evidence against the view that either the great earthwork or vallum which "* Bramb. 1583, 1590. *»» L,B. 28. *o« Bramb. 1558. "' Vit. Hadriani THE ARMIES AND FRONTIER RELATIONS 95 eventually protected the German hmes, or the stone wall which eventually protected the Raetian hmes, were the work of Hadrian. The whole length of the hmes from Rheinbrohl to Welzheim was about 236 miles. It would seem that the construction of the exterior hne from Worth to Lorch, by no means imphed the rehnquishment or even the disarmament of the inner or Neckar-Mumhng hne. Some idea of the size of the fortresses, which varied a good deal, may be gathered from the foUowing measurements. That at Feldburg was 93x78 m.*08; that at Langenhain 198x163 m.*09 ; that at Markobel 190 x 163 m. ; *io the two at Neckarburken respectively 90 X 78 and 160 X 133 m. ; *ii that at the Hunneburg 223 x 145 m. ; ^^^ those at Murrhardt and Lorch respectively 135 X 164 and 164 X 155 m. ; ^^^ that at Saalburg 221 X 146 m. ; Zugmantel 163 X 113 m. ; Niederbieber 264 X 198 in 414 Between these castra there were a number of castella situated at weak points in the hmes, usually 20-30 X 30-40 m. in size, e.g. one at Heidenstock 25 x 19 m. ; *i5 one at the Kaisergrun near Feldberg 26 x 24 m. ; *^« one at Honehausen near WaUdurn 40 X 46 m ; *i7 one at Haselburg near Miltenburg, 42 x 52 m. *i8 etc. Between these agam were a number of watch-towers, burgi ; in the nine miles north of Wall- dum there are fifteen of them.*i» a good general *o« L.B. 7. "• L.B. 3. *io L.B. ID. "" L.B. 27. "» L.B. 30. *" L.B. 40-1. *»* The castle at Richborough is 184 x 138 m. «" L.B. I. *^« L.B. 6. "' L.B. 11. *" L.B. 49. *i» See W.D.Z. 1884, where the cohors i Sequanorum et Rauracorum offers a thankoffering ob explicitum hurgum {C.I.L. xiii. 6509) ; see also Veget. iv. 10 : " castellum parvulum quod burgum vocant." 96 STUDIES IN ROMAN HISTORY account of the limes is given by T. Hodgkin in Arch- cBologia Aeliana, 1882. *2o We have seen that Trajan's share in the task of developing the frontier of the enlarged upper province is uncertain, but his activity was undoubtedly great in developing the interior of the country now taken in by the empire. A road was made by him, as is proved by a milestone found at Buhl dated 100 a.d. and marked a Mog. m.p. cxx,'*2i from Mogontiacum, crossing the river either at Mainz itself or at Worms, and passing through Ladenburg, Heidelberg and Baden to Buhl, and no doubt on to Rottenburg or Rottweil. This road was probably connected with Trajan's scheme for facilitating communications between the Danube provinces and Gaul,422 ^nd may have been carried out during the Dacian wars by leg. i. Adj. and xi Claudia. *23 But not only military fortresses and roads *24 v^e^e constructed, but municipal institutions were developed. These, as we should expect from the previous history of the country and the character of its inhabitants, were of a somewhat heterogeneous nature, and the civitates which grew up can hardly be reduced to a single type hke that which characterized the civitates in Gaul. Nor is it possible, except in one *" Professor Bury's account, not published when this article was written, seems to me extremely confused and unsatisfactory. He regards the line from Miltenburg to Lorch as the eastern boundary of the agri Decumates, and apparently attributes it to Vespasian. **^ See Zangemeister in W.D.Z. 1884, p. 237 foil. *** Aur. Vict. Caes. 13 : " et inter ea iter conditum per feras gentes quo facile ab usque Pontico mari in Galliam permeatur." *" Bramb. 1666. *** For the later development of roads in the agri Decumates, see an article in the Bonn. Jahrb. 1871. THE ARMIES AND FRONTIER RELATIONS 97 or two instances, to trace back with certainty their origin to Trajan. Eutropius, however, says of him *** " urbes trans Rhenum in Germania reparavit." The one civitas which with certainty can be ascribed to Trajan was the civitas Ulpia Sueborum Nicretum, the centre of which was probably Ladenburg — Lopodunum.*^ That the Mattiaci, like the Suebi, were formed into a civitas, is almost certain, but what its centre may have been there is no clear evidence to show.^^? Heddem- heim became the centre of the civitas Taunensium.*^^ Rottenburg became the civitas Sumelocennensis. **• Baden became the civitas Aquensium.^^o § 13. The Legions under Trajan and Hadrian The new and shortened frontier with all its defensive arrangements naturally led to a reduction in the German armies, as well as to some change in their stations. The construction of the limes, the taking in of the Neckar valley and the pushing northward of the Raetian frontier took all importance away from Vindonissa, and probably from the beginning of Trajan's reign, as long as three legions remained in the upper province, the third camp was Rottweil.*" The other two camps were still Mogontiacum and Argentoratum. In the lower province Bonna remained a legionary camp ; a new camp was built by Trajan a little north of Vetera and soon developed into colonia Ulpia Trajana. We have *" viii. 2. "• C.I.L. xiii. ch. 86. "' Bramb. 1313. *" C.I.L. xiii. ch. 112. "• C.I.L. xiii. ch. 81. *" C.I.L. xiii. ch. 77. The subject of this municipal develop- ment does not fall within the present paper, but see Professor Pelham's article already referred to, p. 32 foil. **i The size of the camp, 820 X400 m. seems to show that it was intended for a legion, and tegulae of leg. xi have been found there (Bramb. 1645a). 98 STUDIES IN ROMAN HISTORY seen that the legions at the beginning of Trajan's reign were i Min., x Gem. and vi. Victr. for the lower ; xxii Prim., VIII Aug. and xi Claud, for the upper province. But I Min. was certainly engaged in the second Dacian war *32 and it has recently been made very probable that it was also engaged in the first as well, in which case it must have been removed from Lower Germany soon after loo a.d.*^^ It is also known that xxii Primigenia was again in the lower province in about 103 or 104 a.d. This is proved by the inscriptions in the tuff-stone quarries at Rheinbrohl. These quarries were within the territory of the upper army, but during the Flavian times they seem to have been worked by vexillationes of the lower army for the rebuilding of their camps. None of the troops mentioned in these inscriptions can be shown to have belonged at the time to the upper army, and may all be safely assumed to belong to the lower. In Br. 660 vexillationes are mentioned of vi, x, and XXII, under Acutius Nerva. This person was consul designatus in 100 a.d.,*^* and may therefore probably have been legate of Lower Germany about 104-107. *3s "a C.I.L. vi. 3584 ; ii. 2424. *33 On the other hand Brambach, 680, shows it to have been one of the legions under Acutius Nerva, i.e. probably between 103 and 107, see next note. "* Plin. Ep. ii. 12. *35 The fact that these quarries though within Upper Germany were worked by troops belonging to the lower army is prob- ably correctly explained by Ritterling {De Leg. X Gem. p. 45). Though there were quarries in Lower Germany from which stone could be got, there were none so conveniently situated for the conveyance of stone all the way to their destination by water and down the stream. As far as we can tell, these quarries were only used between Nero's reign and Trajan's. The troops mentioned are le^. xv Prim. (Bramb. 685) and xvi (Br. 657), THE ARMIES AND FRONTIER RELATIONS 99 However, the removal of xxii to the lower army, if it took place, was only temporary, for in the period which follows, its headquarters were certainly Mogon- tiacum. By Hadrian's time, or by whatever date we suppose the work of the limes to have been completed, Upper Germany — and probably Lower Germany as v^ell — was garrisoned by only two legions. From the lower province leg. vi was transferred by Hadrian to Britain, while leg. x Gem. was after the Dacian war in Upper Pannonia. Leg. i Min., on the other hand, was sent back to the province, and together with Trajan's new legion xxx Ulpia Victrix formed the legionary army there. From the upper province leg. xi Claud, after being perhaps temporarily posted at Rottweil ^^^ was moved either by Trajan or Hadrian to Lower Moesia *37 and legions xxii and viii from that time form the legionary garrison, stationed at Mogontiacum and Argentoratum, while the numerous castella along the limes were occupied by the auxiliary troops. § 14. The Auxilia of Upper Germany from the Flavian Period Onwards With regard to the auxiliary forces of the two armies, which must both be earlier than 70 a.d. ; leg. xxi Rap. (Br. 656, 675), between 70 and 83 a.d. ; leg. vi Victr. p.f.D. {Bonn, Jahrb. 89, 48) ; leg. x p.f.D. (Br. 65), and leg. xxii Prim. p.f.D. (Br. 673, also 671-2, 674), all dating between 89 and 96 A.D. Then we get vexillationes of several legions together, e.g. VI, X and xxii (Br. 660), vi and x (Br. 662), i, vi and xi (Br. 680), all under the command of Acutius Nerva, and so dating probably between 103 and 107. In addition to the legions we find a vexillatio of coh. 11 Asturum p.f.D. (Br. 666, 678), and of coh. II c.R. p.f.D. (Br. 676), and coh. i c.R. (Br. 676). **• C.I.L. xiii. p. 212. **' Its tegulae are not found like those of xxii and viii along the limes. 100 STUDIES IN ROMAN HISTORY we are dependent for the Flavian and later period solely on inscriptions. It has been already seen that after the war with Civilis not only were almost all the original auxilia of the provinces removed elsewhere or cashiered, but that generally henceforth these troops are posted in provinces away from their own homes. The change is, to a large extent, illustrated in Germany. While with some few exceptions ^^^ the alae and cohorts of Gauls and Germans are found in such provinces as Britain and Pannonia, those found in the German armies come mainly from Spain, Thrace, Raetia, Dalmatia and Pannonia. Our principal sources of information for the upper army are five MiUtaria Diplomata, dated 74, 82, 90, 116, and 134, and it will be convenient to arrange them in tabular form : — It will be remembered that of the auxiUa traceable as belonging to the former period, the majority were raised in Eastern Gaul and Germany. These have now for the most part disappeared. The earlier auxilia, like the earlier legions, were either cashiered, i.e. probably the greater number of them were, or transferred to provinces further removed from their homes. Of troops belonging to the former period there are really only two — the co- horts of Thracians, Raeti, Vindelici and Delmatae may or may not have been in the province before — the ala Picentina and the ala Cannenefatium : but of these, the former had distinguished itself by its exceptional fidelity,*^® while the latter was transferred from the lower to the upper army. The great bulk of the cohorts now came from Aquitania, the Danube provinces and Spain, while of the alae, as their names seem to imply, I and II Flaviae Geminae were newly created by Ves- ♦•• See below, «» Hist. iv. 62. THE ARMIES AND FRONTIER RELATIONS lOl o 00 ged .So . * js • ^ I N g d S bib ^§ it I Q < 8. I I I I I I nS % d i s'l a u O 1—1 -ml ""^ . be a a > > ^ gos Jiiii > o C/) o.. S» fl 3 '■> - "* ^ 1^1 o ^ (A Q o ^-» rt rt a) 'mi •^ 1^ s. »* »H W I* M H n u a . 3 rt ii I 3 rt.w) a rt a o d2-H b rt cr °J3 a 3 o 1 1 1 II 1 1. III! I 3,2 cr 5^ (z^U l|d 3 o a 3 O ifiii 48^ cSiS g 8 |li|llll||ll.|l.|lll| o.t! •«-> 3 •3 cr < 3 cr dR ^". -•^ la • Q < •SsJ ^ •CN «s §0 ^2^ a >•§ rt 3 S(/) §§ > lal I 1^1 I I ^•^ 3-3 cr cr •d i— »H ►< 8 v\ a 3 u H bo 3 < ►< ►* ^ ^ ^ m^ C MM M M ^ > ^ 102 STUDIES IN ROMAN HISTORY pasian out of several of the disbanded troops.**® It will be noticed at once that the list of troops enumerated in the diplomata is not necessarily a complete list of all those in the army. Thus under the year 90 the ala Scubulorum and the cohors i Germanorum are omitted, but as they are in the diplomata of 82 and of 116, we may safely assume that they were in the province during the intervening time. Instances from other provinces show us that a troop may be absent from two or even three successive diplomata and then again reappear, and in such cases where there is no evidence to the con- trary, we must assume their continued presence in the province. From Vespasian to Domitian there seems to have been four or five alae in the province, under Trajan and Hadrian only one or two. That this diminu- tion in the cavalry had some connexion with the new frontier arrangements seems almost certain. Inscrip- tions seem to show that the fortresses along the limes were garrisoned exclusively by cohorts and detachments from the legion, while the alae quarters seem to have been either the legionary hibema or camps not far re- moved from them or from the Rhine, as e.g. at Wies- baden, Worms, Zahlbach, Heddernheim, Rottenberg, and possibly Rottweil.**^ It can hardly be accident that no traces of any of the alae are found in the stations of the limes. When the limes was constructed, the number of legions was, as we have seen, reduced, and the importance of these inner fortresses diminished. The two Flavian **" That this was the meaning of gemina as applied to legions or auxiliary corps, follows from Caes. B.C. iii. 4, i : " Effeceret . . . unam ex Cilicia veteranam quam factam ex duabus gemel- 1am appellabat." (Conf. D.C. Iv. 23.) *** If Bramb. 1643 mentions ala i Flavia. I THE ARMIES AND FRONTIER RELATIONS 103 alae, which have left traces at Mainz, Wiesbaden and Heddernheim, **2 were posted in Upper Germany at the beginning of Vespasian's reign. Ala i appears in all the diplomata up to ii6, and as it is missing in 134 and was certainly in Raetia in 141 a.d.,**^ it was probably re- moved by Hadrian : ala ii was moved earlier ; it was in Raetia by io8,*** and is not traceable in Germany after 82 a.d. The ala Cannenefatium has only left one inscription in Mainz. ^^^ it probably left Germany at the end of the first or the beginning of the second cen- tury, and was in Pannonia sup. in 138 a.d. 4*6 The ala Scubulorum of uncertain nationahty, and apparently stationed either at Mainz, 4*' Wiesbaden, ^^s or Worms, **« was certainly in the province till the end of Trajan's reign, when all evidence for his history ceases. The ala Picentina, stationed perhaps at Mainz, *5o was removed probably at the end of the first century to Britain where we find it in 124 a.d.^^i The ala Singularium included in the diploma of 90 a.d., probably only joined the upper army at the time of the rising of Saturninus. It was certainly with the lower army before ^52 and must soon have been moved to Raetia where we find it in 108 a.d.*^^ The ala Claudia Nova, which probably accompanied leg. XI Claudia from Dahnatia to Germany in 70 a.d., was in a few years sent to Moesia,*^* where it may pos- sibly be the i Claudia Gallorum mentioned in the diploma of 105 A.D. The diploma of 134 mentions only a single *" C.I.L. iii. 5906. **6 Bramb. 968. *" Bramb. 981, 1525. U68. *** Dipl. xxiv. in C.I.L. iii. **• Dipl. xxxvi. in C.I.L. iii. **' Br. II4S- "" Br. 1524. *" Br. 896. ^ "0 Br. 1 344, conf. 91 5 at Dienheim. *" Dipl. xxx. m C.I.L, UL *•* See below. **' Dipl. xxiv. in C.I.L. iiL *»* See the Dj/>/oma of 82. 104 STUDIES IN ROMAN HISTORY ala, the Indiana Gallonim. ^^^ It was probably only moved into the province by Hadrian from the lower army.*5« Its further history is unknown ; and its only traces in the upper province are at Mainz ; ^^^ Klein- Win terheim ; *S8 and Worms. ^59 Of the other six or seven alae proved by inscriptions to have been with the upper army, e.g., ala Agrippiana, *«o ala Hispanorum,*«i ala Noricorum,*«2 ala Rusonis,*«3 ala Sebosiana Gal- lorum,*«* ala Vallensium, *«5 there is no sufficient evi- dence which justifies us, in view of the silence of the diplomata, in placing any within this period. Of the cohorts, eleven were in the province during the whole of this time. Cohors i Asturum seems never to have been moved far from the Rhine— unless the fragmentary inscription from Mainhardt,4«« refers to this cohort ; its tegulae are found at Gernsheim on the right bank not far from Worms, "' while several inscriptions found at Andernach seem to show that it was at one time posted there. *«s It may, however, have been advanced as far as Hofheim and Heddemheim, where its tegulae are found. *«» The two cohorts i Aquitanorum veterana and I Aquitanorum Biturigum are not always to be dis- tinguished from one another on inscriptions, but the tegulae of one of the two have been found in some num- **» The complete title of this also is found only in the diploma, though Henzen {Bonn. Jahrb. 13, p. 77) had ateady conjectured that it was a Gallic corps. It was no doubt named after Julius Indus {Ann. iii. 44) and originally raised from the ci vitas Treverorum, his native state. "• See below. "' Bramb. 1087. *" Br. 924. "• Br. 891. *•« Br. 893. "^ Br. 889-90. *" Br. 1 1 18. 1229. *«3 Br. 1230 . *•* Br. 894. *" Br. 1631. *•« Br. 1621. *•' Bonn. Jahrb. Ixxxviii. 32. *" Bonn. Jahrb. Ixxii. 82 ; Ixxvii. 19 ; xc. 134. *•• C.I.L. xiii. chaps, iii and 112. THE ARMIES AND FRONTIER RELATIONS 105 bers at Amsburg, where the limes crosses the Wetter, *'» while a coh. i Biturigum seems to have been posted at Langenhain,*^! and it was probably posted on the northern Umes, and possibly at one time at Friedberg.*'* The cohors i Germanorum first appears in 82 A.D., and was still in the province in 248 a.d.*'^ Its headquarters were certainly Jagsthausen, where the limes crosses the Jagst.*'* Coh. II Aug. Cyrenaica was first, perhaps, at Neuenheim opposite Heidelberg, ^'^ but an inscription of this cohort has recently been found at Butzbach north- east of the Taunus plateau. *'« Cohors 11 Raetorum was certainly in the province from 82 to 134, and was therefore probably different from the similarly styled cohort which was in Raetia in 108,*" and again between 145 and 166. *78 It has left traces at Wiesbaden, *'» but later on was probably posted at Saalburg where its tegulae have been found, *8o and at Butzbach. The cohors III Aquitanorum is mentioned on all the diplo- mata. It has left traces on the Neckar-Mumling Une at Neckarburken,*^! at Stockstadt on the Main,*^* *'o Br. 1422&. *'* ■^'^' 9. ... *'« Br. 1417^ ; W.D.Z. Kor. 1883. p. 50 ; C.I.L. xm. 7399«. «3 W D.Z. Kor. 1886. p. 167 ; it must, therefore, be different from the cohors i Nervana Germanorum of British inscription {C.I.L. vU. 953. 1003)- , ^ , -D ./:r< *'* Br. 1608 ; W.D.Z. Kor. 1887. p. 136; conf. also Br. 1616 from Olnhausen in the immediate neighbourhood. *'» Br. 1708 ; W.D.Z. 1892, p. 304. .^, . «• C.I.L. xiii. 7432 ; L.B. No. 4. P- m. while possibly the coh. II Augusta at Heddernheim (Br. 1456 ; C.I.L. xm. 7432) may be the same troop. _ , ^ ,r , *" Dipi. xxiv. in C.I.L. iii. *" Pauly, Real-Encyc. Vol. iv. p. 327• *'» Br. 1520-22. "0 C.I.L. xiu. 7465-8 ; Br. 1431''. ^427 1 W.D.Z. 1885. "1 Br. 1728 ; L.B. 27 ; C.I.L. xui. 6493-4. «> Br. 17616 ; W.D.Z. 1886, p. 341. io6 STUDIES IN ROMAN HISTORY while its tegulae are also found at Ruckingen, north of the Kinzig,*83 and at Osterburken in the third century a.d.^s* The cohors iv Aquitanorum is also mentioned in all the diplomata ; its headquarters were probably Obernberg, an important fortress at the mouth of the MumHng ; 485 traces are also found of it at Milten- burg on the Main.^se Cohors iv Vindehcorum was per- haps first stationed at Kastel,487 or Wiesbaden, ^ss and seems several times to have changed its headquarters. Its numerous tegulae found at Amsburg on the Wetter seem to prove that it was once posted to guard the extreme north-eastern portion of the Umes. To this period belong the tegulae at Niederbieber. ^^^ Again its tegulae near Idstein,«o at Saalburg,49i Langenhain,^^ and Alteburg, point to its presence on the Taunus plateau, while those at Gross Krotzensburg,493 Echzell, Ruckingen^* and Stockstadt,495 show that at one time it guarded the Kinzig valley. Cohors v Delmatarum was with the German army till Hadrian's time, and has left traces on the inner Umes, at Bockingen and also at Mainz and Wiesbaden. «« Lastly cohors vii Raetorum was at one time posted in the neighbourhood of Vindonissa,*»7 but later seems to have been the garrison of Niederbieber and Ems, and apparently at one time of Niedernberg.*^^ *" Br. 14366. "4 c.i.L. xiii. 6577. *" Br. 1747-8 ; W.D.Z. 1885. p. 161. "• W.D.Z. Kor. 1883. p. 5 ; C.I.L. xiii. 6620, 6621. *" See its tegulae. Br. 13771. *" Br. 1537 ; conf. 1439 from Frankfort. "» Br. 7036. 4»o Br. 15506. 4»i Br 141 3^ "" ^^az°;8t P. :;r ^'- '''" '■ ''■''■'■ ''''■"■ '''■ «! 9°°*- *'*° ^'^■^- ^o'- 1883, p. 46, from MUtenburg. M ???•'• '.°^; '518. «' I.H. 344, 10. " C.I.L. xiu. ch. 134 ; xiU. 7733. 7765. ■Ml THE ARMIES AND FRONTIER RELATIONS 10? While the cohorts akeady mentioned remained, so far as we know, permanently in the province, cohors I Thrac. c R was apparently moved by Hadrian to Pannoma sup'., where we find it in 138 a.d.*»» The question of the various cohorts styled i Thracum is very compbcated ; see C I.L. iii, 109 note, but as this one was in Germany m 116 it can hardly be the same as the coh. Thrac. c.R. in Dacia in no or the coh. i Thrac. sag. in the same pro- vince in 157 and 158 a.d. It is more probably to be identified with the i Thrac. c.R. which was m Pann. sup. in 133, 138, 148, 149 and 154- Its station in Germany was perhaps Bendorf near Neuwied -«• whether the in- scriptions at Remagen «»i prove it to have been in the lower province may be left here uncertain. Two cohorts. Ill Gallorum and v Hispanorum, which were m the upper army in 74 a.d. were sent by Domitian at the beginning of disturbances on the Danube to Moesia where we find them in 82 a.d.,^'^ and where the former is traceable till about 114 A.D. when it went to Dacia.^"' Two others, ill Dehnatanim and I Flavia Damascenorum appear for the first time in the province in 90 A.D., and may possibly have been sent there with the ala Singularium, the former almost certainly from Lower Germany, to put down the rising of Saturninus. A tegula of cohors i Dama- scenorum is found at Saalburg. It seems also to have been present at Amsburg. The cohors iii De^at™ has left tegulae at Wiesbaden,"* Ruckingen,"-^ and Oberscheidenthal on the Neckar-Mumling line"" and Gross Krotzenburg. The cohors Damascenorum seems «. Dipl. xxxvii. in C.I.L. lit. "' W.D.Z. Kor. 1890, p. 18. ••> W.D.Z. 1892. p. 281 : Bonn. Jahrb. Ixxxvui. 210. ". Dipt. xiv. (as cited by Cichorius). '•« D^pl. xx., xxii. and xxxui. in C.I.L. iU. "• Br. 1 537*- «» Br. 1436 ; C.I.L. xiii. 742"- '" Bonn. Jahrb. Ixxvi. 242. io8 STUDIES IN ROMAN HISTORY to have been at one time in the neighbourhood of Fried- berg, m? and Strassheim, and afterwards to have been moved to Saalburg,6o« and Amsburg.«w Under Trajan, two other cohorts probably joined the army : the co^ hors I c.R. caUed p.f. on an inscription at Gross Krot- 2enburg."o certainly from the lower province,"^ and the cohors i Ligurum et Hispanorum probably from the Alpes Maritimae."2 of these the former seems to have passed through the same stations as cohors iv Vindeli- corum, being posted first at Arnsburg."^ and Ems on the Lahn,"4 then at Saalburg,"^ and again at Gross Krotzenburg,"6 and Sehgenstadt on the Main : ^i? the latter has left no traces in the province. In addition to these, a cohors ii Hispanorum seems to be included in the diploma of ii6 a.d.,"^ and its tegulae have been found at Stockstadt and Wimpfen on the Neckar-Mum- Hng Une,"« and Oedheim.^^o There is also an inscription from Remagen of the date 158 a.d.,«i which does not necessarily prove it to have been in the lower province at the time. The cohors 11 Aquitanorum mentioned m the^ diplomata of 82 and 90 has left traces at Ams- burg,«2 and as we have seen, a cohors 11 Biturigum was probably with the upper army before Vespasian. As ••' See tegulae there, Br. 1417^. ••• C.I.L. xiii. ch. cxviii. •" C.7.L. xiii. 741 1. •" See CJ.L. v. 7890, 7900. •" Bonn, Jahrb. Ixxxviii. iii. •" W.D.Z. 1885. p. 407. "• Br. 1412. •" See below. •»» Br. 1542c. ■»» Br. 14316. »" Br. 1407ft. ArZ ^' '^'L^^'^^y ^^ *^^ ^'"^ ^ *h^ Pannonian cohort (see diplomata of 60. 80. 84. 85). which was in Dacia by 108 [CJL. "• Br. 1759; C./.L. xiii. p. 238. »«» Bonn. Jahfb. xciii. 42. "a qJX. xiii. ch. 116. THE ARMIES AND FRONTIER RELATIONS 109 far as these five diplomata inform us, it would appear that while in this period the nimiber of alae gradually diminished, that of the cohortes was somewhat increased. While 12 cohorts are mentioned in 74 A.D., we have 17 in 116 A.D. The question arises whether this, as Momm- sen assumes, *23 ^^^s the whole auxiliary force of the province. In the first place, it is certain both from these diplomata and from those of other provinces, that it was by no means the case that each diploma necessarily enumerated all the auxilia in the province. That of 82, e.g., omits i Biturigum, 11 Aug. Cyrenaica, iv Vindeli- corum and v Delmatarum ; that of 90 omits i Germano- rum,52* though a comparison of the other diplomata makes it practically certain that these cohorts were in the province at these dates. But in the Pannonian diplomata we find coh. i Hispanorum mentioned in 60, and then after being omitted in 80, 84, 85, 98, and 133 reappearing in 138, while a cohors vii Breucorum, men- tioned in 85 and omitted in 98, 114, 145, and 160, reap- pears in 167. But the coh. 11 Treverorum, though omitted from all the diplomata, was ahnost certainly posted at Holzhausen*^^ while traces of a coh. Treverorum have been found at Zugmantel. ^^e Similarly coh!* i Sequanorum et Rauracorum, though again not men- tioned in the diplomata, is proved to have been at Mil- tenburg,527 ^nd near Schlossau on the inner limes,*** while a coh. i Helvetiorum has left traces at Oehringen and Bockingen. ^^^ To these we may add coh. i Pannoni- •■* rom. Gesch. v. p. 108. »«* This cohors was posted in the third century at Jagsthausen. As to where it was in earlier times there is no evidence. »" C.I.L. xiii. 761 5-8. "• C.I.L. xiii. 7612. «*' CJ.L. xiii 6604. 6609. •" C.I.L. xiii. 6503. 6509. •*• C./X. xiu. 6472, 6475. no STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS III orum and coh. iv Thracum, both of which have left traces at Wiesbaden. "0 it is therefore not altogether impos- sible that there were alae or cohorts in the province during all this period, not mentioned in the diplomata ; Still less impossible that others may have been introduced there, e.g., between 90 and 116, which, nevertheless, do not appear in the last two diplomata. In the case of the alae, as has been seen, we have no evidence to justify us in assuming anything of the kind, nor have we anything certain in the case of any of the cohorts, though the following point may perhaps be noted. The cohors IV Voluntariorum seems from a single inscription to have been at Obernburg ; ^^i in 148 and 149, it was in Pan- nonia sup. ; ^^^ where was it previous to that time ? Hardly in Pannonia, for it is absent from the seven diplo- mata in which it might have appeared previous to 148 ; hardly in Germania sup. all the time, for it is absent from our five diplomata. On the otherhand, of the cohortes Voluntariorum of which we have traces, all, with one or two exceptions, are found in the Danube provinces or in Germany, while in a passage already quoted from Macrobius ^^^ «' Caesar Augustus in Germania et Illyrico cohortes Ubertinorum complures legit quas voluntarias appellavit." It may then, Uke the cohors i c.R., have been in Germania inferior till Trajan, and then have been for a short time in the Upper German limes before being transferred to Pannonia. Of the cohors iii c.R., we cannot even say so much ; it has, however, left tegulae at Seligenstadt on the Main, cohors xxiv Voluntar- iorum must have been in the province for a considerable •w C.I.L. xiii. c. 124. »" Br. 1754. i»« Diph Ix. and Ixi. as cited by Cichorius. "' Sat. i. 14, 33. time either in this or the following period. It seems to have been stationed at Benningen ^3* or Murrhardt,^^^ the two corresponding fortresses on the Neckar-Mumling, and Miltenberg-Lorch Umites, while it has also left traces at Wiirzburg ^^^ and Oberscheidenthal ^^7 higher up on the Neckar-Mumling line. The cohors xxvi Voluntariorum c.R. was almost certainly in the province before the camp at Vindonissa was given up, since we find traces of it both at Vindonissa itself ^^ and at Benningen, 539 Heideggerhof,^*^ Oberscheidenthal and Wiirtzburg on the inner Umes.^^^ It must also — from the number of inscriptions and tegulae found there — have at some time been stationed at Baden ^*2 and later at Murrhardt on the Umes. In this, and in several other cases, we find the same cohort at corresponding fortresses built on the outer and the inner limes : e.g. coh. XXIV at Benningen and Murrhardt ; coh. i Helv. at Bockingen and Oehringen ; coh. iii Aquit. at Neckarbur- ken and Osterburken. Similarly cohors xxxii Volun. c. R. must have been at Heddernheim and Oberflorstadt ^*^ probably in the Flavian period,^** a cohors i Iturae- orum, which may probably have been in Germany before Vespasian ^*s has left traces at Rottweil, and was therefore probably in the army till at any rate Domitian's "* Br. 1596 ; W.D.Z. 1887, p. 46. »« Br. 1568-70 ; W.D.Z. Kor. 1886, p. i. "• Br. 1393. 6=»' Bonn. Jahrb. Ixxvi. 242. ^^^ I.H. 344. "» W.D.Z. Kor. 1887. 2. "0 Bonn. Jahrb. Ixxi. 20. »" C.I.L. xiii. p. 238. **' Br. 1673c, 2062, 1659, 1662, 1667. «" C./.L. xiii. c. 116. "* Br. 1467, 1480, 1496 ; W.D.Z. Kor. 1882, 4. *** Its four inscriptions in Mainz seem to testify to a consider- able service in the province, Br, 1099, ^233, 1289. .-s^s^iMKKIk . 112 STUDIES IN ROMAN HISTORY time **• though by 98 a.d. it was certainly in Pannonia. Lastly, a cohors i Helvetiorum was certainly at Bockingen on the Neckar-Mumling line in 148 a.d."' and has also left tegulae at Oehringen, the correspond- ing fortress on the main line.s*^ How long before this it may have been in the province we cannot tell. A cohors I Sequanorum et Rauracorum too has left numerous traces at Miltenberg, one inscription being dated 191 a.d.6« On the whole, therefore, we have no evidence supplementary to the diplomata which justifies us in estimating the auxiliary forces of the upper province at more than about twenty to twenty- four alae and cohorts during this period, or from 10,000 to 12,000 men ; while it appears from the diplomata that this force, both as regards numbers and constitu- tion, was almost, if not quite, independent of the legions in the province. Thus in 74 and 82 there were four legions in the province, in 116 and 134 only two, and of these two, only one, viii Aug., belonged to those four : and yet we find that out of the thirteen cohorts of the year 82, no fewer than twelve are still in the province in 116 and eleven in 134. We find nothing like a reconstruction of the auxiliary forces after the revolt of Satuminus, no change of consequence after the Dacian wars of Trajan, and practically no alteration made by Hadrian. That after the limes was completed, the auxiliary forces, consisting more and more of cohorts only, were stationed along its fortresses, seems clear »" W.D.Z. Kor. 1889. 17. »*» Br. 1583 ; see also 1566 and W.D.Z. 1887, p. 46. •*• Br. 1563c. »*• Br. 1740 ; see also 1744, W.D.Z. Kor. 1883, 46, and Br. 1738, at Steinbach. THE ARMIES AND FRONTIER RELATIONS 113 from epigraphical evidence. As to the strength of the garrisons we have no certain evidence. No doubt detachments from the two legions at Mogontiacum and Argentoratum helped in the task of defending the limes. Thus on the Neckar-Mumling line, we find traces of leg. VIII, at Oberscheidenthal,"o Wimpfen,"i Bockin- gen,552 and of leg. xxii at Stockstadt,^^^ Schlossau,e54 and Breuberg ; ^^5 on the Main we have traces of leg. VIII at Aschaffenberg,556 Miltenberg, ^s? of leg. xxii at Aschaffenburg,568 Obernburg,659 Seligenstadt ^^^ and Gross Krotzenburg ; ^si while in the Taunus district both legions have left traces at Saalburg,^®^ ^nd Butz- bach,5«3 and leg. xxii at Langenheim,^^* and leg. viii at Idstein.5«6 Similarly on the Miltenberg-Lorch limes, leg. viii appears at Osterburken,^^^ and Ohlhausen,^*' both legions at Oehringen, ^^^ and leg. xxii at Welz- heim.5«9 More than this we are hardly able to say. Von Cohausen thinks that there must have been not less than fifty of the large fortresses between Rheinbrohl and Lorch, and of course many more of the intermediate castella. Whether a cohort or a legionary detachment undertook two of these fortresses and the castella between, we cannot say. The fact that cohors iv Vindelicorum is found at the neighbouring fortresses of Saalburg and Langenhain, Ruckingen and Gross Krotzenburg ; iv Aquitanorum at Obemburg and Miltenburg ; coh. i c.R. at Ems, and Arzbach, SeUgen- "" W.D.Z. Kor. 1883. p. 140. "1 Br. 1391. »" Br. 1583-4. "3 Br. 1761. »" Br. 1736. "« Br. 1400. ««• Br. 1752. «" W.D.Z.Kor. 1883. 46. »" Br. 1757 "» W.D.Z. 1885. 161. "« Br. 1407. "1 Br. 1455^1. *" Br. I43ifland6. "^ Br. 1421a. "* L.B. g. ««5 Br. 1548. "« Br. 1729. "' Br. 1618. «•» Br. 1554, 15636, 1563a. "• Br. 1564. I 114 STUDIES IN ROMAN HISTORY stadt and Gross Krotzenburg ; coh. xxiv Vol"- <=R^ at Murrhardt and Benningen ; i Germanorum at Jagsthausen and Ohlhausen ; i Helvetiorum at Boclan- gen and Oehringen, and perhaps other cases which I have not noticed may possibly point to f "l^/'^!^!^ ~;;; iecture But while Mommsen is no doubt right n upposing that at ordinary times the effective strength of the garrison was not sufficient for really defensive purposes "» probably some of the camps were garnsoned more strongly than others ; e.g., there were two camps at Neckarburken, and it was always possible to collect a considerable number of soldiers at any S^ven pomt while reinforcements could always be humed forward from the headquarters of the legions at Mainz and Strassburg. § 15. THE AUXILIA OF LOWER GERMANY FROM THE Flavian Period onwards Our information as to the auxilia of Lower Germany for the period after 70 a.d., is much more ^complete As in the upper province the old auxiUaries were probably almost all either disbanded or transferred elsewhere and it is unUkely that either Batavi. Frisii Cannenefate. Treveri, Tungri or Belgae were employed so near their homes. Unfortunately, no diplomata have as yet come to our aid.'" The solitary fragment of one apparently .7. He points out that the garrison °/.the f^tellum of Kut- lowitza in Moesia inferior was garnsoned •" ■» ^-^^^^ f/'^*! six men of leg. xi Claudia {rom. Gesch. v. p. I43. and see tp" ^^.^.- i^Lijtt seen in Mr. Henderson's new t^k. «» r;„tS tt=s of X X rd one%ohort. ^The alae are THE ARMIES AND FRONTIER RELATIONS II5 dating from Domitian's reign suppKes us only with the name of a single ala — the ala Afrorum — and the fact that at least ten cohorts were in the province at the time. Ritterling's convincing conjecture that the entire army of Lower Germany, legions, alae and cohorts, received the cognomina p.f.D., as a recognition of their services against Satuminus in 89 a.d., gives a little further assistance, while any alae or cohorts mentioned on inscriptions from the Brohl quarries may be safely assigned to the lower army of this period. The ala Afrorum already mentioned was perhaps stationed at Koln.^'^ Q^e of its veterans has left a funeral inscription in Holland. ^'^ j^e ala Indiana Gallorum was, as we have seen, probably moved into Upper Germany by Hadrian : it is called p.f. in an inscription of Trajan's time,^'* and its presence in Lower Germany is attested by an inscription from Worringen north of Koln. The ala was in Britain at some time ^'^ but this may have been between 43 a.d. and 69 A.D., when, if identical with the ala Treverorum of Htst ii. 14, it was in Germany. The ala i Singularium is also called p.f. in a Raetian diploma of 107 a.d. ; it was certainly in the lower army in 69 a.d. ; s'« it is men- ala Noricorum, ala Singularium, ala Moesica, ala Siliana, ala Sulpicia and ala Indiana. The cohort is i Flavia Hispanorum. It does not appear whether this is a complete list of the troops mentioned in the diploma or whether the other cohorts are missing. In any case there must have been other cohorts in the province at this date, and as all the alae with the exception of the ala Siliana are mentioned from other evidence below, the diploma does not really affect the conclusions I have drawn. »'« Br. 317; Bonn. Jahrb. bd. 70; bcxxi. 92 ; W.D.Z, Kor. 1886, 10. •'» Br. 66. w« Wilm. 694. »" C./.L. vii 66. *'• Hist, iv. 70; V. 21. Il6 STUDIES IN ROMAN HISTORY tioned in close connexion with leg. vi Victr. and coh. ii Varcianorum, both belonging to the lower army, under Vespasian ; it probably only joined the upper army m 89 A.D., while by 108 it was in Raetia."' The ala Longima or Longiniana from the number of inscriptions relatmg to it "» must have been stationed at Bonn for a con- siderable time ; that it was not disbanded by Vespasian is proved by Wilm. 2198, and we have no traces of it in any other province. An ala Moesica has left an inscription at Deutz,"» and was certainly in the province in Trajan's time ; ^»'' while an ala Noricorum has left traces at uncertain dates, but some of them belonging probably to this period, at Calcar,^" Dormagen,'" and Koln <»» In addition to these alae an ala Sulpicia was probably at Koln in 187 a.d.'«« and has also left traces at Bonn,^'' though as to its earUer service m the province we can say nothing. There are single inscnp- tions relating to an ala Vocontiorum at Cleve,^'' to an ala I Thracum at Vechten,"' and to an ala Aug. Ituraeorum at some place unknown, ^^s As far, therefore, as we can form any judgment from the imperfect epigraphical evidence before us, there may have been six or seven alae in the province during this period. Of cohorts, 11 Asturum was certainly in Lower Germany under the »" Dipt. xxiv. in C.I.LAix. , „ ^, bo«« »" Br 498 : W.D.Z. Kor. 1891, p. 296 ; 1892. P- t>5 : Bom. Jahrb. xciii. i6. »'• Br. 438. . .» . i„ 5.0 See C.I.L. xi. 709, where a praefectus alae Moesuae is at the same time censor Germaniae inferions; conf. C.iX. vi. 3538, where we have a tribune of leg. vii Gem. in Germama becoming praefectus alae Moesicae. , , t , ;: „ Ml Br 168 187. •" Br. 285. '" Bonn. Jahrb. Ixxil. 21. •" Br. 314'; conf. 344- "' Bonn. Jahrb. Ixxxi. 89. M« Br. 161. "' Br. 56. "' Br. 2003. h THE ARMIES AND FRONTIER RELATIONS 117 Flavian emperors. It has left traces, together with other troops of the lower army, in the Brohl quarries, ^^^ while its cognomina p.f.D.^^^ show that it had a share in the defeat of Saturninus. A tegula at Vetera also ^®^ is probably to be attributed to it and not to coh. i.^^^ By Trajan it was probably moved to Britain, where we find it in 105 a.d.^^^ Cohortes i and 11 c.R. are styled respectively p.f. and p.f.D. Both are mentioned in inscriptions from the Brohl quarries,^®* and they may, therefore, with some certainty be attributed to the lower army of the period. ^^^ Cohors i certainly, cohors II possibly, were sent to the upper province by Trajan.^®* A cohors i Lucensium is styled p.f. on an inscription in Holland, dated between 104 and III A.D.^*' and was therefore in the province both under Domitian and Trajan, having been moved prob- ably from the upper army by Vespasian. A cohors xv Voluntariorum is almost certainly mentioned in the Brohl quarries, ^®^ and was also in Holland in 1916 A.D.^®® A cohors II Varcianorum has also left traces in the quarries,*^ as well as at Bonn ^^^ and Koln, while a cohors III Lusitanorum is mentioned on an inscription at Koln,®°2 and as it was in Pannonia after 114 a.d. "» Br. 666. 678. «»o Br. 678. "i W.D.Z. 1885. p. 222. *'^ Conf. also an inscription of Domitian 's time, of cohors secund. Astur. in Germ. C.I.L. ii. 4251. "3 Dipi. xxiii. in C.I.L. iii. "* Br. 670, 676. *•' See also C.I.L. vi. 3520, where cohors i is expressly de- scribed as " in Germania inferiore," and ix, 2958, which should no doubt be restored cohors 11 civi[um Rom]anorum Germ[an inf]erioris. "•• See diplomata of 116 and 134 ; and L.B. $6, for a tegula of cohors II found at Hunneburg on the Taunus. «" Bi-. 6b. "8 Bonn. Jahrb. bcxxi. 115. "» Br. 6a, conf. i4oh. •<»> Br. 664. •»i Bonn, Jahrb. Ixxxi. 112. "a Br. 312. Ii8 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS its German service probably belongs to the end of the first century. If we add to this a cohors Latabi (ensium) of which there are two inscriptions at Kohi ^^^ and a cohors VIII Breucorum which has left traces at Cleve «"* and Koln/<^5 vve have really exhausted the list of troops which can with any probabiUty be assigned to the lower army after Vespasian. ««« Probably the number of cohorts did not much, if at all, exceed that, viz. ten given on the fragmentary diploma of Domitian, and when we consider that of those mentioned above, four or five left the province at the end of the first or the beginning of the second century, we may perhaps suspect that in this case the reduction in the number of legions was accompanied by a corresponding reduction in the auxiliary forces. § i6. Probable Position of the Auxilia along the Limes With regard to Lower Germany, there is little more that can or need be said. Its auxiUary forces, as we have seen, are only imperfectly known. Its legions from the time of Trajan were only two, the new legion xxx Ulpia, stationed in a new camp built for it by Trajan, a few miles south of Vetera, and afterwards called the Colonia Tra- jana, and legion i Minervia, further to the north. The praesidia along the Lippe, and generaUy on the right bank of the Rhine, had been given up since the time of •03 Bonn. Jahrb. Ixxxii. 23 ; and W.D.Z. 1892, p. 232. •0* Br. 159. •0* Br. 232, possibly also at Vetera (Br. 22 ip). •«• Ritterling supposes that cohors m Delmatanim and il Hispanorum. which are both p.f., and both seem to have been moved into the upper province by Domitian or Trajan, came from the lower army. It is not improbable, but there is no evidence for it except their cognomina. 119 i Claudius ; though the Batavi, Frisii and Cannenefates still belonged to the empire. The frontier was of course guarded not only by the two legionary camps, but by a series of smaller fortresses garrisoned by the auxiliary troops, all along the river. The most important of these were no doubt, as Mommsen suggests, at Remagen, Bonn, Wessehng, Deutz,^^^ Dormagen, Biirgel, Neuss, Gelduba and Cleve. The object of these stations was not so much to pre- vent invasion as to regulate the crossing of the river, which was allowed only at certain places, at certain times, and to unarmed and not too numerous parties. We know from Hist iv. 65 that it was the duty of the Ubii to enforce such conditions at Koln. To a large extent, therefore, the duties of the auxiliary troops were only semi-military in character. At ordinary times, they were the duties of patrols, customs officers and coast- guard-men. In the discharge of these duties there can be no doubt that the classis Germanica played a useful and important part. Of important relations with the free German tribes we hear nothing till the Barbarian invasions of the third century. The piratical Chauci might now and again give trouble, but these incursions affected Belgica rather than Roman Germany. The affairs of the Bructeri engaged the attention of Trajan in 97, but his restoration of an exiled king was of more moment in suggesting some courtly phrases to Pliny in his " Panegyric " than from any important bearing upon the frontier relations in the lower province. «<»' This was an outpost of Koln on the right bank, rendered necessary by the more constant traffic across the river at this important point. 120 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 121 The development in the upper province Wcis more important, but, as far as literary sources are concerned equally obscure. Here, too, the legions were reduced to two, legion viii Aug. at Strasburg, and legion xxii Prim, at Mainz. But the auxiliary troops, or at least the in- fantry cohorts, were somewhat increased in number, and, posted almost entirely in the Trans-Rhenane district, must have exercised continual activity in patrolling, guarding and controUing the 230 miles of frontier line between Rheinbrohl on the Rhine, and Lorch on the Rems. We hear however of no wars, even against the Chatti, under Trajan, Hadrian or Antoninus Pius. The interior of the country was no doubt deve- loping, and civitates of various kinds sprang up. It was certainly during these reigns that the frontier, known as the Upper German Limes, was being carried to- wards its final stages. It has been suggested above that Domitian, partly in continuation of his father's work in the Neckar district, partly as the result of his own war against the Chatti, had estabhshed a " Umes *' of over 100 miles in the Taunus region, and another, mainly following the course of the Neckar, for the " agri Decu- mates." We have further assumed that either he or Trajan must have completed the protection of this Trans-Rhenane Germany by connecting these two *' limites " by a Une of forts from Gross Krotzenburg, first up the Main, and then from the mouth of the Mum- ling to Wimpfen on the Neckar. Hadrian, in all prob- ability, somewhat extended Domitian 's hne, especially on the north and north-east of the Taunus plateau, and rebuilt or enlarged many of Domitian's fortresses, made more systematic the arrangement of intermediate castella and burgi, and, as would appear from Spartianus, con- structed some kind of wooden palisade all along the frontier, except where the Main or the Neckar formed the boundary. As points of communication between this frontier line and the legionary headquarters, still on the left bank of the Rhine, we have seen that there were several important fortresses, at Friedburg, Hed- demheim, Baden-Baden, Wiesbaden, Hochst, and Kastel. But from Trajan onward, the old idea of Augustus and Tiberius of connecting and compacting the Danube and German frontiers was undoubtedly revived. The ex- tension of Raetia north of the Danube, the antecedents of the " limes Raeticus," and the ultimate change which made Raetia a legionary instead of a procuratorial pro- vince, belong to the history of the Danube frontier, which may possibly follow the present article. At any rate, when the Raetian Une was dravm from the Danube at Kelheim to Lorch on the Rems, the time came to modify the German hmes so as to join the Raetian. This modification took place under either Hadrian or more pro- bably under Antoninus Pius. The limes from Worth — the Neckar -Mumling line — to Wimpfen, ceased to be the actual frontier ; and an external limes was drawn from Miltenburg to Wahldum, and from that place in a bee-line to Lorch. This, as we shall see, did not neces- sarily imply the complete withdrawal of troops from the inner Une, though there was no doubt a tendency in that direction. So far, we have assumed the Limes to consist of a number of frontier roads, Uned by about fifty castra, varying in size, but as a rule large enough to contain 800 men.*^ At intervals between these larger camps, were ••* Von Cohausen supposes that there were about fifty of these larger fortresses. The general direction of the limes is already f 122 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 123 smaller castella, perhaps from fifty to eighty yards square, while between these again, were burgi, watch-towers, five or six yards square. From the time of Hadrian, this Limes was probably supplemented by a wooden palisade, perhaps running its entire length from the Danube to the Rhine. The only reference we have to anything like a continuous line of entrenchment is in one of the passages quoted from Frontinus,*^^ from which it would seem possible that part of the Taunus district was protected by him by some kind of vallum. But what has hitherto been primarily suggested by the terms, Limes Germanicus, and Limes Raeticus, is un- doubtedly, in the latter case, the stone wall, eight or ten feet high, the so-called " Teufels-Mauer," extending from Kelheim to Lorch ; and in the former, the enormous earthwork with a ditch in front of it, the " Pfahl- Graben," which ran outside the line of fortresses already described, from Lorch to the Rhine at Rheinbrohl. That this gigantic rampart wais originated, if not actually carried out, by some single emperor, seems in the highest degree probable.®^^ That these constructions formed the last step in the " Hmes " system, seems probable from two considerations. In the first place, the system known, but the investigations of the Limes-Commission may bring to Ught additional camps, and will almost certainly give more precise data for tracing the time order of its various portions. •<*• See p. 91. *^^ The different character of the two portions of the work is probably to be ascribed to the fact that the stone required was more easily procurable along the southern line than to any important difference in object or conditions. At the same time Mommsen is no doubt right in explaining the comparative absence of fortresses along the Raetian " limes " by the fact that the Her- munduri, who were on the other side of it, were very friendly to the Romans, and were allowed exceptional privileges in regard to crossing the frontier. of fortresses, castella, and watch towers might have an intelligible object and be capable of achieving it, without the addition of a wall or rampart ; while on the other hand such a rampart, by itself, could have had no appreciable result. In the second place, though it is impossible that the *' limes " arrangements, either with or without the rampart, could have been an effectual barrier against Barbarian invasion, yet the addition of a wall or continuous earthwork would undoubtedly put a fresh obstacle in the way of an invader, and therefore points to a time not so peaceful and secure as the first fifty years of the second century must have been. There is some evidence that the formidable wars under M. Aurelius against the Marcomanni on the Danube were not without influence upon the Upper German pro- vince, and encouraged the Chatti, perhaps even the Hermunduri, to threaten if not to attack the pro- vince. There is therefore much to be said in favour of the view that the " Teufels-Mauer " and the " Pfahl- Graben " date from the time of M. Aurelius and his son Commodus. With regard to the objects of the Limes— both before and after the construction of the exterior vallum and ditch— it is hardly possible that the investigations of the Limes-Commission will invalidate Mommsen's view, that it was not intended primarily to be a protection against the invasion of the province by the German tribes to the east and north-east.«ii Even the original Taunus " hmes " of Domitian, it would appear from the words of Frontinus, was mainly intended to cut off communications between the subject tribes within that •" See a short but instructive paper on " Der oberrheinische Limes " in the Gesammelte Schriften, vol. v. pp. 444 i- 124 STUDIES IN ROMAN HISTORY THE ARMIES AND FRONTIER RELATIONS 125 district and the Chatti outside it.^^^ its primary object must have been merely to control and regulate the pas- sage of the frontier, to insist on certain formalities on the part of individual travellers and small parties, to limit the passage to certain points, to exact the custom duties on all goods imported, and no doubt to prevent entirely the passage of large bodies either of immigrants or enemies. It must be remembered that the outer German limes alone, with at least fifty fortresses, ex- tended along about 230 miles ; that in addition there was the inner hne and the fortresses along the Neckar, probably not wholly denuded of troops, even after the ex- terior line was made ; while a certain number of troops were still required for several fortresses in the interior of the country, serving as points of communication between the Rhine headquarters and the garrisons along the Limes. To have garrisoned on anything like a war footing the larger camps, as well as the intermedi- ate castella and burgi, with adequate reserves at the supporting camps, would have been impossible without a force of at least 40,000 auxiliaries, in addition to the legions, which, we must remember, were now only two in number. That nothing Hke this force can be traced or assumed, we have akeady seen. The two legions, xxii Prim, and VIII Aug. were stationed at Mogontiacmn and Ar- gentoratum. Detachments from them were no doubt constantly employed for one purpose or another along the Limes, but still they formed no part of its regular garrison. At the same time, in the event of anything like an invasion, they could be concentrated at any point, and were still the backbone of the provincial •^* Quorum refugia nudaverat. army. Of the auxiliary cohortes, whose special duty it was to guard the frontier fortresses, there can hardly have been more than about 20,000 or at most 22,000 men. From the time of Trajan onwards, there are indications that in Germany, as elsewhere, the regular auxilia, who no longer necessarily or perhaps usually belonged to the tribe whose name they bore, were supplemented by bodies of Barbarians, called " numeri," raised from the tribes within or just outside the province ; and probably in some cases occupying lands on condition of such military service. In the Trans-Rhenane part of Upper Germany, we find several numeri Brittonum, distin- guished by appellations such as Nemaningentes,®!^ and Triputienses,®!* which appear to be territorial. As to the number of these local and barbarian soldiers, we can get no reliable estimate. Most of them were perhaps in the nature of a militia, but some of them were cer- tainly employed to garrison some of the fortresses, as at Niederbieber. The individual camps, and indeed the Limes as a whole, were quite capable of being efficiently defended with proper garrisons, and the connecting roads and rivers together with the base fortresses would always enable a certain concentration to be effected at threatened points. But at ordinary times, the fort- resses must have been weakly garrisoned, and the troops dispersed and scattered. Certain camps, owing to the importance of their position, were perhaps more strongly and continuously guarded than others, such as Nieder- bieber, Saalburg, Amsburg, and Gross Krotzenburg. Friedburg, Heddernheim, Baden Baden, and Wiesbaden, probably retained their garrisons almost to the end, but the evidence of inscriptions is somewhat in favour of the •i» Bramb. 1752. •^^ Br. 1732. I il 126 STUDIES IN ROMAN HISTORY view that there was a tendency for the Neckar camps and some of those on the Neckar-Mumling Une to be abandoned and for the great proportion of the cohorts to be placed along the line between Gross Krotzenburg and Niederbieber. The following attempts to locate the cohorts for the period after Hadrian, depending on inscriptions already referred to,^^^ can of course be no more than approximately correct, as very few of the inscriptions are dated. The northern section of the Limes, from Niederbieber in the Lahn valley to Arnsburg at the north-east extremity of the Taunus, seems to have been specially looked after by the following ten cohorts : — vii Raetorum ; iv Vindelicorum ; i civium Romanorum ; i Thracum ; i Aquitanorum Biturigum ; II Aquitanorum ; ii Flavia Cyrenaica ; i Flavia Damas- cenorum ; ii Treverorum and ii Hispanorum. Of these, the vii Raetorum seems to have found its way to the north from the neighbourhood of Vindonissa, and to have been stationed for a time at Niedernburg, on the Odenwald line, between Gross Krotzenburg and Worth. It has left traces at both Niederbieber and Ems, and was perhaps responsible for those two important camps, though the former was also guarded by two nimieri of Brittones. Coh. iv Vindelicorum has perhaps left more traces on the Limes than any other troop. It probably started from Kastel, Castellum Mogontiacense, was then perhaps moved to Friedburg, was at one time appar- ently at Gross Krotzenburg, and employed in the Main fortresses between that and Miltenburg. But its later and most permanent station seems to have been Arns- burg, and we find traces of it at Saalburg and Hedders- dorf and as far west as Niederbieber. Coh. i civium •" See § 14. THE ARMIES AND FRONTIER RELATIONS 127 Romanorum would seem to have served on the same two sections of the Limes as the last mentioned corps, having left traces at Arnsburg and Saalburg on the north Hne, and also at Gross Krotzenburg and Seligen- stadt on the eastern side. It may be conjectured that at some time it changed places with iv Vindelicorum, and was moved down to the Main, when the latter went to Arnsburg. Coh. 11 Augusta Cyrenaica, once stationed on the Neckar at Neuenheim, was afterwards the gar- rison of Butzbach. Coh. i Flavia Damascenorum, per- haps raised by Titus in the Jewish war, was posted at Saalburg. Coh. i Thracum, at Offenburg in the Flavian period, was later on at Bendorf. Coh. 11 Hispanorum, once at Wimpfenon the Neckar, was afterwards posted at Heddersdorf , but it would seem to have been stationed for a time at Stockstadt on the Main. A cohort, 11 or III Treverorum, probably had the task of protecting the neighbouring camps of Zugmantel and Holzhausen. Coh. I Biturigum Aquitanorum, which was once at Rott- weil, was afterwards the garrison of Langenhain, while we find traces of i and 11 Aquitanorum on the north line, especially at Arnsburg. To the section of the Limes between Arnsburg and Gross Krotzenburg we can only assign with certainty three cohorts : iii Delmatarum ; xxxii Voluntari- orum, and either i c.R. or iv Vindelicorum, which, as we have seen, were alternately stationed on this and the north line. Of these iii Delmatarum was first stationed at Wiesbaden, then apparently on the Neckar-Mumling line at Oberscheidenthal, and ultimately at Ruckingen. Coh. xxxii Voluntariorum was at Oberflorstadt ; while either i c.R. or iv Vindeli- corum was at Gross Krotzenburg. 128 STUDIES IN ROMAN HISTORY To the section between Gross Krotzenburg and Miltenburg, we can assign four or five cohorts : iii civium Romanorum ; iii and iv Aquitanorum ; vii Raetorum, and i Sequanorum et Rauracorum. Of , these. III c.R. is found at Seligenstadt. With regard to III Aquitanorum, it seems to have served both on this and on the next lower section of the Limes. At one time it was certainly on the Neckar-Mumling line at Neckarburken. Then, when the outer line was constructed, it was either moved forward to Osterburken, the corresponding camp, on the outer line, or, as is more probable, had to look after both stations. It may after- wards have been moved north, to Stockstadt, when II Hispanorum was moved from there to Heddersdorf. This may have been all the more necessary, as vii Raetorum, once certainly posted at Niedemburg, was also, as we have seen, removed from this part. Coh. iv Aquitanorum was certainly the garrison of Obemburg, and no doubt had to look after the neighbouring camps. Coh. I Sequanorum et Rauracorum was stationed at Miltenburg, but we have an example of the way in which these troops might have to operate in the neighbourhood in an inscription from near Schlossau, in which a vexillatio of this cohort offers a thank-offering— oJ burgum explicitum. On the Neckar-Mumling line, we have already seen that at one time, iii Delmata- rum was at Oberscheidenthal ; iii Aquitanorum at Neckarburken ; and ii Hispanorum at Wimpfen, but all these seem to have been removed, and probably the later arrangement was for the inner Une to be looked after, as far as was necessary, by the corps stationed along the outer Limes. To this outer line, between Miltenburg and Lorch, we THE ARMIES AND FRONTIER RELATIONS I29 can assign four cohorts : i Helvetiorum ; i German- orum ; v Delmatarum, and xxiv Voluntariorum. Of these I Germanorum was stationed at Jagsthausen ; i Helvetiorum at Oehringen ; v Dehnatarum at Bock- ingen; and xxiv Voluntariorum at^Murrhardt or Benningen. Of the Neckar camps, higher up than Wimpfen, the original limes of the agri Decumates, Rottweil alone perhaps retained a permanent garrison in iii Hispan- orum. Of the interior fortresses, Heddernheim was still garrisoned by i Asturum, and Baden Baden by XXVI Voluntariorum. Friedburg, which had been successively held by iv Vindelicorum and iv Aqui- tanorum was perhaps left without a garrison, when the north limes was strengthened. Wiesbaden still seems to have retained 11 Raetorum. With regard to the others we can say nothing. This was presumably the strength and generally speak- ing the distribution of the Upper German army, when the period of imminent danger from the Barbarians, began under Caracalla, was continued under Alexander Severus, and under Gallienus, led to the loss of all Trans- rhenane Germany, THE PRINCIPATE OF GALEA 131 THE FOUR EMPERORS' YEAR CHAPTER I The Principate of Galba An important crisis in the history of the Roman Empire is reached, when the Julio-Claudian line of emperors came to an end with the death of Nero. Hitherto the position of the princeps had been, though not strictly speaking hereditary, still practically confined to the members of a single family. ^ And the election of each new emperor had taken place at Rome with a certain amount of constitutional form, not always quite independently of the wishes of the soldiers, as the election of Claudius proved, ^ but at any rate without violence or bloodshed. More important still, the provincial armies on the frontiers had hitherto acquiesced without demur in each election sanctioned by the Senate. In all these respects the death of Nero marks a new epoch. Family dynasties after this time occur only at intervals and exceptionally ; the prae- torian cohorts frequently use the power afforded by their position in Rome and, above all, the secret was now dis- closed, " posse principem alibi quam Romae fieri." The natural consequence of this was the frequent election * Hist. i. 16 : " sub Tiberio et Gaio et Claudio unius familiae quasi hereditas fuimus." • Suet. Claud. 10. 130 of an emperor by one of the great provincial armies, a practice which was always liable to occasion, and often • did actually bring about, a civil war. In the administration of the empire, too, a change of spirit perhaps rather than of system dates from the same point. Under the first two emperors, the admin- istration had been vigorous and statesmanlike, but the evils to which hereditary government is hable came out strongly in the insane Caligula, the feeble and indolent Claudius, the ill-trained and ungovernable Nero. It is true that provincial government was still for the most part efficient, but nevertheless, a reformation in all directions was necessary, and began with Vespasian. It might indeed have begun with Galba, had not old age and intrigue overpowered him. But the better period, commenced under Vespasian, continued without a break, in spite of some cruelties at Rome during the last years of Domitian, for more than a century. The short period covered by the reigns of Galba, Otho and Vitellius does not, it is true, display to us any of the better features of the new era ; it is a period of disorder and transition ; it shows us the provincial* legions conscious of their own strength, sensitively jea- lous of one another, and rushing into civil war in support of their several candidates for the imperial position. At Rome, the government was absorbed in prepara tions for its own security, the praetorians extorted con- stant indulgencies, and between them and the Senate there was a mutual distrust which paralysed all political action. Still, the period is interesting from the diversity of the incidents which present themselves, quite apart from the more settled state of things which was ulti- mately to result from them. Not only do we get crowded II 132 STUDIES IN ROMAN HISTORY into little more than a year three short-lived emperors, but we have an abortive national rising in Gaul, two civil wars fought out on Italian soil in quick succession, and a critical and even desperate struggle on the Rhine against rebellious German tribes and unfaithful Gallic states. In these events Africa, Spain and Britain, the Gallic provinces and Germany, the Danube armies and even the distant East were all more or less involved, so that almost every portion of the empire was disturbed by the stirring events of these stormy months. The closing years of Nero's reign were increasingly marked by unworthy extravagances on the one hand and by acts of savage or suspicious cruelty on the other. As Dio Cassius says,^ virtue, wealth and ability were in his eyes so many crimes, and their possessors were never for a moment safe. If these cruelties had been confined to Rome, the tyranny might have lasted on indefinitely until some assassination scheme proved successful, but Nero seems to have been possessed by an increasing suspicion of his generals and provincial governors.* In the East Corbulo had been the most conspicuous figure, but on his return homewards after the Armenian War, he received and obeyed the Emperor's command to put an end to his own Ufe.^ In the West, Scribonius Rufus and Scribonius Proculus had for some time been entrusted with the two German armies, but in 67 a.d. they were sent for by the tyrant, and shared the fate of Corbulo.' When the commanders of the three most powerful armies fell a prey to the Emperor's jealousy, there would naturally be a feeling of insecurity and dis- trust among the provincial governors generally. Galba, ■ Ixiii. 17. * Plut. Galb. 3. » Dio Cass. Ixiii. 17. • Id. ib. THE PRINCIPATE OF GALBA 133 as we know from Suetonius,' was afraid to take any active step in his province, lest he should give offence, and no doubt he was only in the position of mahy others. This tension in the provinces was an acute danger to Nero's position ; and though in the excitement of his musical triumphs in Greece he disregarded it, his freed- man Helius, whom he had left in charge at Rome, was more observant, for he again and again urged Nero's return, and finally himself went to Greece to bring the Emperor back.® Meanwhile, however, the spark of disaffection was lighted in Gaul, where Julius Vindex, legate of Gallia Lugdunensis, raised an army from some of the Gallic civitates, and placed himself in open revolt against Nero. Gallia Lugdunensis was one of the unarmed provinces,^ but nevertheless Vindex soon fou Ann. iii. 40 ; Dio Cass. Ixiii. 22. " HisL iv. 17, 53. 57 " Plut. Galb. 4. i 138 STUDIES IN ROMAN HISTORY conqueror of Greece. His father, a consular, had been an industrious pleader in the courts. Galba was adopted by his wealthy step-mother, Livia Ocelhna, and was known during the greater part of his hfe as L. Livius . Sulpicius Galba. He had been a favourite of the Em- press Livia, who left him a legacy of fifty milUon ses- terces, which he never received. 26 Through her in- fluence, he entered upon the senatorial cursus honorum before the usual age. After being praetor and then legate of Aquitania, he was consul in 33 a.d. and under CaUgula was appointed to succeed Gn. Cornelius Len- tulus Gaetulicus as legate of Upper Germany, 26 where he was distinguished for his strict discipline. By Claud- ius he was chosen as proconsul of Africa, receiving the tritimphalia ornamenfa, and admission into three of the priestly colleges. 2? Under Nero, he remained in retire- ment till 60 A.D. when he was sent to Spain. Though Galba's age— he was now seventy-three— seemed to exclude him from Nero's jealousy, yet his relations with the Emperor had of late become strained. He had made no secret of his sympathy with the pro- vincial population against the exactions of the imperial procurators. 28 He had been somewhat ostentatiously inactive for fear of offending Nero, 29 and it was even said that he had intercepted secret orders to the pro- curators for his own murder. »<> The first communication of Vindex, sent before any overt act had been committed, he had ignored; but a second invitation definitely offering him the empire and bidding him place himself at the head of the Galhc troops met with a more serious « Suet. Galb. $, *' Suet. Galb. 8. » Suet. Galb. 9. " Suet. Galb. 6; Ann. vi. 30. " Plut. Galb. 4. »» Plut. Galb. 4. THE PRINCIPATE OF GALBA 139 consideration. While some of his friends advised delay, Titus Vinius, one of his legates.^i urged immediate acceptance of the offer, and a request from the legate of Aquitania for assistance against Vindex ^2 made some decision unavoidable. A court for the purpose of grant- ing manumissions had been fixed at Nova Carthago for April 2, 68 A.D. 33 On that occasion, events were pre- cipitated by the action of his soldiers and friends, who saluted him as Imperator. His troops however con- sisted of a single legion, vi Victrix.^^ The value of Vindex's support was quite uncertain, and as never before had an emperor been proclaimed outside Rome, it was no wonder that Galba for the moment shrank from the imperial title and the name of Caesar. He marked however clearly enough his defection from Nero and his alliance with Vindex by calling himself the legate of the Senate and Roman people. ^^ Vindex was now able, or hoped to be able, to cloak his real designs under the pretence of substituting Galba, whose wealth, birth and reputation was second to none of his contemporaries, for Nero, very much as Civilis afterwards used the name of Vespasian, as long as the state of affairs rendered necessary any concealment of his ulterior aims.^^ AH really depended upon the at- »* Titus Vinius Rufinus had been military tribune in Pannonia under Calvisius Sabinus. He passed through the usual cursus honorum under Claudius ; was legate of a legion after the praetorship, and then proconsul of Narbonensis {Hist. i. 48). " Suet. Galb. 9. »' The date is calculated by means of a statement by Dio Cassius, that Galba reigned nine months and thirteen days. ** It is uncertain where the second Spanish legion, x Gem., was at this time. " Plut. Galb. 5 ; Suet. Galb. 10. *• Hist. iv. 21. 140 STUDIES IN ROMAN HISTORY THE PRINCIPATE OF GALEA I^I titude which Verginius Rufus, legate of Upper Germany, might take up. 37 Verginius was in no way deceived as to the inten- tions of Vindex, and at once collected his troops, ^s being reinforced by detachments from Lower Germany, and also by the militia of the Lingones and Treveri.' who did not on this occasion share the designs of the other Galhc tribes. The first object was to secure V^ontio, which from its position was an important point, being situated at the junction of the roads from the upper Rhine to the province of Belgica. This place, the capital of the Sequani, had declared for Vindex, 39 and towards it both armies marched. Whether there was a personal interview between the two generals, as Dio Cassius states, is uncertain, ^o At any rate, both armies were eager for battle, and the leaders would probably not have had the power, even if they had had the wish, to postpone it.^i The result was what might have been expected. The Galhc levies were no match for the discipUned legions, and Vindex with 20,000 of his troops was left dead on the field. This event placed both Verginius and Galba in an awkward posi- »' Verginius Rufus was born in 15 a.d. near Comum of an equestrian family. He, however, obtained senatorial rank and passed through the usual cursus honorum, until he gained the consulship in 63 a.d. [Ann. xv. 23) ; he had only been appointed to his present position in the last year of Nero's reign " Hist. i. 5 1. 53. »» Dio Cass. Ixiii. 24. *» Ixiii. 24 ; his account is that Vindex and Verginius came to some agreement, against Nero, and that the battle was the result of an unfortunate misunderstanding between the two armies, Vindex committing suicide in despair, much to the grief of Verginius. It has already been noticed that this version is quite inconsistent with the epitaph of Verginius. as well as with the statements of both Tacitus and Plutarch *"- Plut. Galb. 6. tion. The latter had lost the ally on the strength of whose assurances he had taken up his independent attitude. The news, too, which quickly arrived, that Verginius had been urged by his legions to accept the empire for himself, threw him into the greatest alarm. He accordingly withdrew to Clunia, apparently with the intention of awaiting events, though he sent letters to Verginius, suggesting some common action. To Verginius, the offer of his soldiers caused hardly less embarrassment. Hitherto, he had apparently acted in Nero's name, but his soldiers, elated at their victory and conscious of their strength, were determined to appoint their own emperor, and all that he could do was to declare that the matter must stand over for the decision of the Senate. *2 Meanwhile, events in Rome freed the situation from what seemed abnost to threaten a dead-lock. The first news of the Galhc rising had reached Nero at Naples on March 19.^3 ^t first he made Ught of it, and even affected to regard it as a lucky means of re- plenishing his exchequer.** He, however, subsequently on his return to Rome put a price on the head of Vindex, and recalled certain troops to Italy, which he had detached from their headquarters, for a pro- posed expedition against the Alani, a tribe beyond the Caspian. *5 Other detachments, too, were recalled from Egypt, where an Ethiopian war had been meditated. The tidings of Galba's proclamation by his army was a ** Plut. Galh. 6 ; for Mommsen's view, already referred to as to the aims of Vindex andVerginius. see Hermes, xiii. p. 90 foil., and for a further discussion of the subject see Schiller, Geschichte des fom. Kaiserreichs unterNero,p. 261 foil; Hermes, xv.p, 627 foil. ; xvi. p. 147 foil. «» Suet. Nero, 40. " Plut. Galb. 5. " Hist, i. 6, 9, 70. ;i ' 142 STUDIES IN ROMAN HISTORY much more serious blow. Nero seems completely to have lost his head. Wild schemes occurred to him for having all the provincial governors murdered ; for a massacre of all the Gauls in Rome, of poisoning all the senators, setting fire to the city, and taking refuge in Egypt.*® Somewhat more serious acts were his assumption of the Consulship, the raising of a legion of marines from the fleet at Misenum,*' and the appoint- ment of Turpihanus and Rubrius Gallus as his generals.*^ By this time, however, in spite of Galba's supineness, other provincial governors had joined him,*^ and it became known in Rome that the Upper German army had declared against Nero, though not for Galba. This last news, while it drove Nero to despair,^® prob- ably convinced the two praetorian praefects, Tigelhnus and Nymphidius Sabinus, that Nero's star had fallen. Possibly not without their advice, Nero fled from the palace, and hid himself in the suburban residence of one of his freedmen ; ^^ while Nymphidius, persuading the praetorians that Nero had fled to Egypt, ^^ ^nd at the same time promising an unprecedented donative in Galba's name, induced many to proclaim that general emperor. The Senate, taken by surprise, and antici- pated in any designs it may have had for assuming the government itself, acquiesced and proclaimed Nero a public enemy. ^^ The news of this event and of Nero's ignominious end reached Galba in Spain within eight days, 5* and probably Verginius Rufus about the same time. Galba THE PRINCIPATE OF GALBA 143 *• Suet. Nero, 43. *^ Dio Cass. Ixiii. 27. *® Suet. Nero, 27. *' Suet. Nero, 49. «' Hist. i. 6. *• Plut. Galh. 16. •1 Suet. Nero, 48. " Plut. Galb. 7. " Plut. Galh. 2, hesitated no longer. He had already assembled a kind of provincial Senate, established a body-guard of evocati for his own protection, and enrolled from the provincials a new legion, vii Galbiana^after- wards Gemina — as well as some auxiliary troops, ^s He now formally assumed the titles of Imperator and Caesar, 5« and made immediate preparations for march- ing to Rome. His chief advisers were T. Vinius, his legate, M. Salvius Otho, legate of Lusitania, who' had been the first to join his standard, ^^ Cornelius Laco, whom he named his praetorian praefect, and Icelus, the freedman, who had brought the news of Nero's death. Nowhere had he any reason to expect serious resistance to his authority. Verginius Rufus, who had hitherto caused him the greatest anxiety, remained true to his own declarations, and though not without difficulty, induced his legions to acquiesce in the Senate's decision, and to take the oath to Galba. ^s Only in Africa were there symptoms of disorder. P. Clodius Macer, the legate of the army, had been urged by Nero to create a diversion against Galba, and with that end, to keep back the com supplies from Rome.ss Macer had obeyed these orders, but not entirely to Nero's advantage, for under the pretence of establishing again the Repubhc,«o he was really setting himself up as emperor. But the course of plunder and rapine in which he indulged soon made it clear that he was hardly a serious rival, «i and Galba found no difficulty in getting " Suet.Ga/fe.io. 5« Suet. Galb. 11. «' Plut. Galb 20 " Plut. Galb. 10. " Hist. i. 3. . •0 So he gave himself the old republican title of " propraetore Africae," and formed a new legion which he called legio Macer- iana liberatrix (Mionnet. i. p. 1 50). " Plut. Galb. 6, i 144 STUDIES IN ROMAN HISTORY rid of him and his claims through the agency of Gani- tianus, a procurator. '^ But while there was no opposition, there was little enthusiasm in Galba's favour, and it was perhaps in order to bind some portion at least of the empire to himself, that, in rewarding those to whom he owed his promotion, he associated with the Spanish provinces those states of Gaul which had identified themselves with Vindex.®^ That he still could have remained blind to the real meaning of that rising is almost im- possible, but he probably chose to close his eyes to all but Vindex's invitation to himself, and so he rewarded the Sequani and Aedui with the remission of a fourth part of their tribute, and with grants of the Roman citizenship.®* Whatever may have been the motive for these acts, and whether they were really the acts of Galba, or, as some believe, a job on the part of Vinius,** it was hardly politic to take these steps. If they bound the states to him, they still further alienated the German legions, whose services in defeating Vindex were thus worse than ignored. But Galba went even further. The hesitation of Verginius and the reluc- tance of his legions were well known to Galba, through the communications of Fabius Valens,** and he lost no time in recalling Verginius, and putting in his place Hordeonius Flaccus,'' who, as an old man and an invalid, was unlikely to excite any enthusiasm on the part of the troops. Besides this, he punished the Lingones and Treveri,®^ who had acted throughout with Verginius, with loss of territory, and Lugdimum, which had all along remained faithful to Nero, with " Hist. i. 7. " Plut. Galb. 15. •* Hist. i. 8. and 50. •» Plut. Galb, 18. •• Hist. i. 52. •' Plut. Galb. 10. •• Hist. i. 53. THE PRINCIPATE OF GALBA 145 sequestration of its revenues. '^ When we add to these acts the execution of Fonteius Capito, legate of Lower Germany, which, if not ordered by Galba was at any rate approved by him,'<> it will be seen that, even before he left his own province, he had done much to earn the reputation of strictness and cruelty which preceded him to the capital.'^ The events which were meanwhile happening there were not calculated to make him more lenient. Nym- phidius Sabinus, after Nero's death, found himself by means of the praetorian cohorts the most powerful personage in Rome. Tigellinus, his colleague in the praefecture, he got rid of ; ^2 and though the Senate was nominally carrying on the government, Nymphidius contrived to exercise, partly by flattery, partly by intrigue, partly through the force which backed him, a paramount influence over their proceedings. '^ But his aims went further than this. He had originally no doubt hoped to play Sejanus to Galba's Tiberius, but he had not calculated upon the influence of Galba's friends in Spain, and the rejection of his demands, first for the sole praefecture, and then for the dismissal of Vinius and Laco,*'* drove him to more decisive mea- sures. He now meditated the assumption of the im- perial position himself.'^ But he had miscalculated on the devotion of the praetorians. The exhortation of one of the tribunes, Antonius Onomastus, not to desert Galba for an upstart like Nymphidius, had an immediate effect ; and when Nymphidius presented himself to deliver an harangue, specially written for •• Hist. i. 65. " Plut. Galb. 8. '» Hist. i. 7. " Id. ib. '• Id. ib. 14. '* Suet. Galb. 14. '* Plut. Galb. 13. 146 STUDIES IN ROMAN HISTORY him by Cingonius Varro, one of his senatorial supporters, he was received with acclamations for Galba, and presently killed. By this time Galba was approaching the city, escorted by the new Spanish legion, which he had himself en- rolled. ^^ At Narbo he was met by a deputation of the Senate, which he received with courtesy and re- gard." But it was already becoming evident that the new emperor was ahnost entirely under the control of Vinius, Laco and Icelus, his paedagogi, as Suetonius calls them,'8 while M. Otho, who seems from the first to have cherished ambitious hopes, was craftily in- gratiating himself both with Galba and with Vinius. That severity should be shown to those who had aided Nymphidius in his treasonable attempt was only to be expected, but that men of senatorial rank like Cingonius Varro and Petronius Turpilianus should be executed without a trial was an unfortunate com- mencement of the reign. 79 Nor was Galba's entry into Rome a propitious one. As he approached the Mulvian bridge, he was met by the legion of marines whom Nero had enrolled, but without formally giving them their eagle, and thus establishing their legionary rank. This they now came in somewhat tumultuous fashion to demand of Galba, and when he refused to give them an immediate answer, some of them in their excitement drew their swords. Galba ordered his cavalry to charge, and a number of them were killed on the spot, while the legion was subjected to the severe punish- ment of decimation.80 Although Galba at the time refused to confirm the legionary character of these THE PRINCIPATE OF GALBA 147 '• Hist. i. 7. »• Plut. Ga/6. 15. " P\ut.Galb, n. •« Suet. Galb. 12. '• Galb. 14. dasstarit, when the request was made in this disorderly manner, we know from two military diplomata that he did so afterwards. From these, both dated December 22, 68 A.D., we find that he gave the civitas " veteranis qui militaverunt in legione prima adjutrice, qui vicena stipendia aut plura meruerunt, et sunt dimissi honesta missiones.'* What took place was probably this :— On December 22, Galba formally recognized the legion, gave it its eagle and its title— i Adjutrix— and the civitas to all the classiarii in it, while he also gave the civitas with the honesta missio to those who were retiring from service.^* The state of affairs in the city Uv^eded a very wary tread. The Senate, freed at last from Nero's tyranny, looked for a reformed government ; the city population, on the other hand, would be sure to resent any diminution of the amusements provided for them by Nero. The Praetorian cohorts, which had been induced to desert Nero by false pretences,82 were not likely to be satisfied without the payment of the huge donative promised them in Galba's name by Nymphidius, while lastly, in addition to the troops brought by Galba himself, there were a number of legionary detachments— ?;^:*:t7/a .—in the city, from the British, German and lUyrian armies.83 Neither these nor the praetorians were much predisposed in favour of an emperor who had haughtily declared that it was his habit to levy soldiers, not to buy them.^* In accordance with this maxim the donative was not paid, and indeed Galba found himself almost at once involved in difficulties from the ■^ For discrepancies between the accounts of this episode see Hist. i. 6 ; Suet. Galb. 12 ; Plut. Galb. 15. " Hist. i. 5. «3 Hist. i. 6. " Plut. Ga/6. 18. I 148 STUDIES IN ROMAN HISTORY emptiness of the treasury. Thoughtlessly, and prob- ably immediately after his election, he had remitted one of the regular taxes,^^ and it was now, in order to carry on the government, necessary to have recourse to extraordinary means. These were not well chosen, and served to confirm the reputation for meanness and avarice which Galba had already gained. Immense sums had been squandered by Nero on his favourites and courtiers, and especially on musicians and other artists. These Galba determined to recover, leaving to each recipient one tenth part of the spoils. But as the extra- vagant lives of these men had left little or nothing of their property, he went further still, and ordered all the real property to be reclaimed from those who had received it from these reckless spendthrifts. The result of this was wide-spread bankruptcy and distress, with by no means proportionate gain to the treasury. Hardly more dignified, though less widely unpopular, was the institution of an inquiry into gifts made to the various temple treasuries, and the appro- priation of whatever rested upon an insecure title.^* But while Galba was thus injudiciously niggardly with regard to the donative, and mean in his other financial measures, it was an open secret that his three favourites were quickly laying by huge fortunes by a shameless traffic in offices, privileges and immunities of all kinds.^' It was, no doubt, an exaggeration, but sufficiently true to be telling, when Otho declared that •• See coins with the legend " Quadragesima remissa." Cohen, i ; Suet. Galb. 138, 139, 198-200; Plut. Galb. 16; Hist. i. 20. •• Tac. Agric. 6. " Plut. Galh. 16 ; Hist. i. 33 ; Suet. Galb. 15; Dio Cass, Ixiv. 6, THE PRINCIPATE OF GALBA 149 in eight months Icelus had amassed as much as Poly- clitus or Vatinius had coveted.^^ Even in his most popular acts Galba seemed fated to give offence. None of the creatures of the Neronian r^ime were so hated as the rapacious ministers to his vices and luxury. Nymphidius had won a momentary popularity by offering up some of them to the popular fury,^^ and Galba performed an act no less just than politic, when he executed Helius, Polyclitus and Patrobius.»« But this was far from satisfying the populace, whose ven- geance could never be complete as long as the hated Tigellius lived. That worthy personage, however, had taken the best means to secure the protection of Vinius, and so completely was Galba under his favourite's influence, that he was induced to rebuke the revengeful spirit of the people, who wished, forsooth, to make his reign tyrannical by the execution of a sick old man.»i It was thus amid growing unpopularity among all classes, civil and military, that the year 69 a.d. began ; a year which was to see the deaths of three emperors and two civil wars fought out in Italy. The first act of this tragedy opens among the legions of the Rhine armies. In Upper Germany,®^ the soldiers were still smarting under the recall of Verginius ; their services to the empire against Vindex were overlooked, while the Gauls whom they had conquered could point to rewards and honours.^^ This discontent was fomented by the Lingones and Treveri, whose punishment by " Hist. i. 37. «» Plut. Galb. 8. »<» Id. ib. 17. »» Id. ib. •» There were here at the time three legions, xxii Primi- genia and iv Macedonica at Mogontiacum, and xxi Rapax at Vindonissa, while i Italica, a new legion, was at Lug- dunum. •» Hist. i. 53. f 150 STUDIES IN ROMAN HISTORY Galba was an insult to the legions with whom they had served. To check the mutinous inclinations, which Caecina, the legate of one of the legions, with all the spite of a detected swindler, did his best to encourage,^* Hordeonius Flaccus was little qualified. Galba had intended him to be a nonentity and a nonentity he proved, remaining throughout the disturbances which followed, the help- less spectator of events which he could not control. Besides the actual grievances of the troops, various unfounded rumours got abroad. It was whispered that Galba intended to decimate the legions,®^ and to cashier the bravest of the centurions ; while some legates of the Lingones, who had done much to foment the disorder, and had been sent away by Hordeonius in the night, were believed by many to have been made away with. The result was a secret understanding among the legions, and when, on January i, they were required to take the usual oath of fidehty to the emperor, they threw down his effigy from the standards, and insisted on taking the oath to the Senate and Roman people.^® In the army of Lower Germany, the legions went further still.®^ Fonteius Capito had been succeeded by Aulus Vitellius, a man of little force of character, but with a careless bonhomie which made him popular with the soldiers, and deriving prestige from the three consulships and censorship of his father. Born in •* Caecina, when quaestor in Baetica, had joined Galba's cause, and was rewarded by the command of a legion. Soon after, Galba, on discovering certain peculations on Caecina 's part, ordered his dismissal [Hist. i. 53). •» Htst.i. 51. »• Hist. i. 55. •' The four legions here were v Alauda, and xv Primigenia at Vetera, xvi at Novaesium, and i at Bonna. THE PRINCIPATE OF GALBA 151 15 A.D., and brought up at the court of Tiberius at Capreae, he had ingratiated himself with Caligula by his taste for charioteering, and with Claudius by his love of gambling. Nero's favour he won by flattering his skill as a citharoedus.^s Backed by these advan- tages, he passed through the usual honours, and reached the consulship in 48 a.d.^^ jje was also one of the fr aires Arvales,^^ and a member of the XV viri.^^^^ After his consulship, he governed Africa for one year as proconsul, and the next year succeeded his brother in the same province as legate of the African army.^^^^ As curator of public works in Rome he had been found guilty of dishonesty, and had been living in poverty and retirement when he was chosen by Galba, probably as a safe stop-gap, to be legate of the lower Rhine army. Reaching his province at the end of November, he at once ingratiated himself with his troops by various indulgences. From the first he overstepped the limits of a legate's authority, and the artifices of Fabius Valens, legate of legion i, paved the way for more decisive steps. This officer, who had been the first in the German army to take the oath to Galba, now by various re- presentations instigated the sluggish Vitellius to covet, if not to hope for, the imperial power. 1^3 On January I the oath was taken to Galba with marked reluctance and not without insult to the Emperor's effigy. On the night of the following day, news arrived from the upper army that the oath to Galba had been refused. •8 Suet. Vitell. 4. •» Ann. xi. 23. 100 Henzen, Act. fr. Arv. p. 87. 101 Hist. iii. 86 ; Eck. vi. p. 316. "« Suet. Vitell. 5 ; Hist. ii. 77- "' ^»^^- i- 53- 152 STUDIES IN ROMAN HISTORY But still no successor had been openly suggested, though, according to Plutarch, the attention of the upper army had already been drawn to Vitellius. ^°* The decisive step was now taken by Valens, who formally proclaimed him at Colonia Agrippinensis. The example was followed by all the legions of the pro- vince, then by those of Upper Germany as well as by the Lingones and Treveri-^^^ The example set by the Spanish legion was thus followed by the German troops who, disappointed in their hopes of Verginius, were sure to find and accept some more willing candidate for the position which they claimed to confer. The first rumour of disturbances in the German armies, though no definite news had as yet been re- ceived about the proclamation of Vitellius, coming, as it did, in addition to his domestic troubles, decided Galba to complete at once what he had for some time meditated, the adoption of an heir to the empire.^®* Augustus had on more than one occasion provided for the future by this means, though he had never looked beyond the members of his own family. Galba had now a wider field to select from, but it was essential that the heir and colleague of his choice should unite youth and vigour with an unsullied reputation, popular manners, and the qualities which appeal to and com- mand respect in soldiers. No doubt the right man was difficult to find, though candidates were not want- ing. This was the opportunity to which Otho from the first had been looking forward. He was popular with the praetorians, to whom he seemed the repre- sentative of Nero's indulgent regime. He had in- w* Galb. 6. "» Hist. i. 57. "• Hist. i. 12. THE PRINCIPATE OF GALBA 153 gratiated himself with suitors of all kinds ; he was the candidate favoured by T. Vinius,^^' and he had with promptitude and hberality supported Galba's cause at a time when others still held back.^^ But Galba, weak as he was, was conscientious in his desire to do what was best for the empire. ^^^ Otho's antecedents were not good ; he had shared Nero's orgies, though not perhaps his cruelties, and he was loaded with debts. Besides, he was vain, unprincipled, and without self- restraint ; no one in his senses would have made him heir even to a private estate, which needed economy and care, and therefore in spite of his influential sup- porters, Galba looked elsewhere. ^^° Cornelius Dolabella, a distant relative of the Emperor, and a man of noble family, was supported by some,^^^ but he was unpopular with the praetorians, and per- haps not altogether agreeable to Galba himself. The actual choice, though probably Galba's own mind was made up before, was the result of a sort of cabinet council at which Vinius, Laco, Marius Celsus and Ducenius Geminus, praefect of the city, were present. ^f^ "' Hist. i. 13. "« Plut. Galb. 20. i«> Id. ib. 21. *i® The family of Otho came from Terentinum. His grand- father, an eques, owed his admission to the Senate to the in- fluence of Livia Augusta. His father, said to have strongly resembled in features the Emperor Tiberius, passed through the senatorial cursus honorum, and became proconsul of Africa. M. Salvius Otho was born in 32 a.d. He was introduced to the circle of Nero's friends, and it was at his house that Nero and Agrippina dined on the day destined for the murder of the latter. He was the husband of Poppaea Sabina, and it was merely to get him out of the way that Nero sent him in 36 a.d. as legate of Lusitania, where he had remained in semi-exile. For the relations between Otho, Nero and Poppaea, see Hist. i. 13 ; Suet. Oih. 3 ; Plut. Galb. 19, and the clearer and more consistent account in Ann, xiii. 45. *" Plut. Galb, 23. "' Hist. i. 14. , 154 STUDIES IN ROMAN HISTORY THE PRINCIPATE OF GALEA 155 It fell upon L. Calpumius Piso Frugi Licinianus, a young man of noble ancestry and exalted character, many members of whose family had fallen victims either to Claudius or Nero, and who had himself only recently returned from exile. ^^^ xhere can be no doubt that Galba's intentions in choosing Piso were excellent, but nevertheless, the choice was neither judicious nor fortunate. Piso was more a stoic philo- sopher than a practical statesman. With him virtue was austerity ; he had no military experience and was utterly unknown to the legions, while the praetorians more than suspected that he shared Galba's views about a donative. Some concession even now on this last point might have postponed the crisis, but none was made. 11^ The adoption was publicly announced in the praetorian camp, but the coldness and silence with which it was received were signs even more ominous than the thunder and hghtning amid which it was performed. This was on January 10, and the reign was destined to last but six days longer. Otho had in effect staked everything on his hopes of being ap- pointed Galba's successor. He had given up his pro- vince and returned to Rome, where innumerable creditors were only forbearing while they waited to see what course his fortunes would take. If he acquiesced in his defeat everything was lost."^ Galba clearly did not trust him, Piso would almost certainly regard him as a rival. On the other hand, Galba had taken a false step ; Piso was agreeable to no order and no "3 His mother was a granddaughter of Sex. Pompeius. His father, Calpumius Piso Frugi, had been adopted by M. Licinius Crassus. "* Plut. Galb. 23. "« Plut. 0th. 5, party. Astrologers encouraged Otho still to hope for the empire. Crowds of discontented and disappointed men were ready to join him, and, above all, he had reason to believe that the praetorians would stand by him. But if anything was to be done, it must be done at once, before Piso's authority had begun to take root. Now the arts by which Otho had laboured to win popularity among the soldiers bore opportune fruit. ^^® Several of the praetorian cohorts had already received money presents from Otho under pretence of a compli- ment to Galba, ^^' and his agent and freedman, Onomas- tus, made no doubt that he could organize a movement among the corps. As his instrument for working upon the rest, he chose Barbius Proculus, a " tesserarius," and Veturius, an " optio," both belonging to the " speculatores,'* or mounted division of the praetorians. Humble as these men were, however, " two private soldiers undertook and accomplished the transference of the empire." The necessary preparations were made without exciting the suspicions of Laco, while Otho himself remained in the background. On the morning of January 15 all was ready, and Otho was hastily summoned, while attending Galba at a sacrifice, to meet his supporters. Excusing himself on the ground that he had to view a house which he intended to buy, he hastened to the forum, where twenty-three specu- latores saluted him emperor. Otho's courage almost gave way at the fewness of his supporters, but he was hurried off in a litter to the praetorian camp, and there acknowledged by the soldiers. Meanwhile, vague reports of a disturbance reached !!• Hist. i. 23. "' Id. ib. 11 156 STUDIES IN ROMAN HISTORY Galba, who was still engaged in sacrificing to the gods of an empire no longer his own, and it was deemed advisable to send Piso to the palace, in order to secure, if possible, the cohorts on guard there. Marius Celsus was sent on a similar errand to the Illyrian detachments stationed in the porticus Vipsania. Both, however, were unsuccessful. Of the other troops in the city, the legion of marines, to whom Galba had ultimately given their eagle, at once declared for Otho, while the German soldiers, who had been kindly cared for by Galba, remained undecided. ^^^ Round Galba himself was collected a Uttle band of friends and freedmen, who drew their swords and prepared to defend their master, while the crowd which now filled the Palatine demanded with shouts the death of Otho, whom in another hour they would proclaim emperor. Con- trary to the advice of Vinius, Galba determined to proceed to the forum. ^^® On the way he was met by a report of Otho's death, and even the soldier who had killed him was introduced with streaming sword. " Who gave you your order, fellow-soldier ? " was all the recognition which his supposed deed gained from Galba, who thus worthily maintained his ideas of discipline to the last. Arrived in the forum, Galba soon learned the truth ; Piso returned from a fruitless attempt to enter the camp, Marius brought back word of his failure with the Illyrian soldiers, and abready the horsemen of Otho were seen coming through the Basilica Pauli.^^o Galba's litter was hustled hither and thither by the crowd, as it hurried to secure good positions from which to view the expected tragedy. It was by the Lacus "• Hist. i. 31. »" Hist, i. 34. "» Plut. Galb. 26. THE PRINCIPATE OF GALBA 157 Curtii that Otho's soldiers came up with him. Accord- ing to some, Galba pleaded for a few days in which to pay the donative, according to others, he exclaimed, " What is this you do, fellow-soldiers ? I am your emperor, you are my soldiers." But according to the version which obtained most general belief, he merely bade them despatch and strike, if the good of the Republic required it. In this wise died Galba, a man, as Tacitus says, " Omnium consensu capax imperii, nisi imperasset." The fate of Galba was shared by Piso, who was killed at the Temple of Vesta by Vinius, who died asserting his own complicity with Otho, and by Laco and Icelus. The streets were filled with the disorderly soldiers, some carrying the heads of their victims on their spears, others seeking some means of distinguishing themselves and claiming reward, ^^i *** No fewer than one hundred and twenty claims of this kind were afterwards discovered by Vitellius, who by a judicious exercise of severity had all his claimants executed {Hist. i. 44). CHAPTER II The Reign of Otho Galea once dead, neither Senate nor populace hesitated to accept the accompHshed fact. Otho was without delay acknowledged by the servile Senate. The tri- bunician power was decreed him, as well as the titles of Caesar and Augustus, and the othtBr insignia of empire.^ But he found himself from the first in a most difficult position. He owed his election to the praetorians, whose prospects and his own were indeed inextricably boimd together. But it was also necessary, especially with the immediate prospect of a struggle with Vitellius, to win over the senatorial order as well. Unfortunately the praetorians were animated with the bitterest jealousy and suspicion of the Senate. This showed itself in a comparatively trifling matter at once. Marius Celsus, a leading senator and consul designatus, who had been faithful to Galba to the last, was demanded for punishment by the soldiers, and it was only by a ruse that Otho was able to save his life. 2 But the soldiers were quieted for the pre- sent by unusual indulgences. The system was abolished by which they had to buy from their centurions leave of absence and all other military privileges, while to compensate the centurions, an annual sum was to be » Hist. i. 47 ; Plut. Galb. 28. 168 * Hist. i. 45. THE REIGN OF OTHO 159 paid out of the fiscus.^ Further than this, the prae- torians were allowed to choose their own praefects, Plotius Firmus, who had been praefectus vigilum, and Licinius Proculus ; and according to Tacitus, even the praefectus urbi. For this last ofiice, they showed their regard for Nero's memory by selecting Flavins Sabinus, the elder brother of Vespasian, who had held the post for seven years under that emperor. On the other hand, to the Senate Otho showed respect and consideration. No punishment was inflicted on Galba's supporters ; Celsus was openly taken into favour.* The Consulships conferred in advance by Nero or Galba were uninterfered with ; the special honour of succeeding the emperor himself was accorded to Ver- ginius Rufus ; ^ the honours of a priesthood or an augurship were conferred on a number of senators who had passed through the ordinary career, while similar honours, as well as the restoration of their former rank, were granted to those who had been exiled under Nero.* These indulgences and favours were no doubt looked upon with some suspicion, but still for the moment at least they disposed the Senate to see in Otho quahties which they had not suspected.' The populace he was able to gratify by the long- deserved and hoped for punishment of Tigellinus, who met his death as unworthily as he had lived at Sinuessa. The title of Nero Otho, which he had half accepted to please the populace, he judiciously abandoned out of deference to the Senate.^ Nor were his favours con- ' Hist. i. 46 : Vitellius adopted the same system in the German army. {Hist. i. 58.) * Hist. i. 77. « Plut. 0th. 2. • Hist. i. 87. » Plut. 0/A. I. 'Plut. 0/A. 3. i6o STUDIES IN ROMAN HISTORY fined to Rome. Some attempt was made to conciliate the Spanish provinces. Hispalis in Baetica and Emerita in Lusitania were strengthened by the addition of new families, Baetica received the revenues of a number of Mauretanian towns, while in Gaul, Otho seems to have made some attempt to remove the bad impression produced by Galba's severity by a grant of the Roman ci vitas to the Lingones.^ At the same time, Africa and Cappadocia were conciliated by certain privileges, which were perhaps more valuable in appearance than in reality. Whether, as Tacitus suggests, Otho was merely postponing his vices and his revenge till he could indulge them in safety, or not, he at least showed some signs of rising to his position. But notwithstanding all his precautions, a chance incident showed how little at his disposal were the forces which had raised him to power. Intending probably to increase the number of praetorian cohorts, he had sent for one of the urban cohorts, hitherto stationed at Ostia, to the city. One of the praetorian tribunes was charged with the duty of arming the corps with its new equipments, but to prevent disorder this was to be done at night. The soldiers, however, seeing the armoury opened, suspected some treachery against Otho, and listening to no explanation, they killed the tribune and some centurions, and then rushed • Hist. i. 78. This grant of course came to nothing, as at this very time Valens and his army were in the territory of the Lingones. It has been suggested that, as this grant is mentioned, between privileges given to Spanish communities, the name of some Spanish tribe should be substituted for " Lingonibus." This is possible, but the Lingones with their 70,000 men would have been valuable allies to Otho, especially if his plan for invading Narbonensis had come to anything, and I am inclined to retain the traditional reading. THE REIGN OF OTHO 161 to the palace, demanding vengeance on the Senate. Otho at the time was entertaining a number of senators and their wives, and it was with difficulty that he hurried them off, before the angry soldiers forced their way into his presence. The end of the incident was hardly likely to reassert Otho's authority, or to reaissure the Senate. While two insignificant soldiers were selected for punishment, the rest were rewarded for their zeal by a present of 5,000 sesterces. ^° Appar- ently the number both of praetorian and urban co- horts had been increased by Claudius, the numbers attached to the latter following consecutively on those of the former. This was cohort xvii, cohort xviii being stationed at Lugdunum.^^ The state of uncertainty and suspense in the city passed all description. Rumours of all sorts about the movements and prospects of Vitellius drove the Senate ahnost to distraction. Placed as they were in Otho's power, the senators had to flatter him, and to flatter one who had himself been the adroitest flatterer of his time was itself no easy task.^^ jhe readiest plan was to abuse Vitellius, but this was not without its dangers, as no one knew how soon Vitellius might not be in Rome as emperor. To add to the desperation of their position, the streets were infested with disorderly soldiers, with whom noble birth, or wealth, or reputation were so many motives for out- rage. For the superstitious, omens were not wanting, while an extraordinary overflow of the Tiber caused widespread distress, by destroying quantities of com in the river docks, and by flooding the roads leading out of Rome towards the north. Indeed Otho's posi- " H%$t, i. 80 ; Plut. 0th. 3. " Hist. i. 64. *■ Hist. i. 85. M l62 STUDIES IN ROMAN HISTORY tion was as difficult as can be imagined. He might affect to discharge the duties of empire as if in pro- found peace, ^^ but, as a matter of fact, every action of his short and troubled reign had some reference to the impending struggle with the German legions. The misfortune was that while he tried to reconcile irreconcilable elements at home, he postponed till it was too late all active measures against his rival. On the other hand, ViteUius, though sluggish him- self, had in Valens and Caecina two enthusiastic and capable generals, whose whole interest for the present lay in furthering his cause. The ardour of the soldiers, however, needed little stimulus, nor were the Agrip- pinenses, as the Ubii now preferred to call themselves, the Treveri or the Lingones behind them in enthusiasm. Volunteers came in from every side, and those who could not serve themselves provided money, or horses, or arms.i* All felt that the prestige of the German army was at stake, and that their claim to appoint an imperator must not a second time be disappointed. The first step was to make sure of the neighbouring provinces, in order that no diversion might be made from the rear. Of the Gallic provinces, Belgica, under Valerius Asiaticus, and Lugdunensis, under Junius Blaesus, at once joined the cause. The accession of the latter involved that of a legion, i Itahca, and the ala Tauriana, both for the time stationed at Lug- dunum.^^ Aquitania and the Spanish provinces had at first declared for Otho, and Cluvius Rufus, Galba's successor in Tcirraconensis, was complimented on his loyalty by an edict of Otho, but as soon as military preparations began, these provinces found it safer to *8 i i.87. 14 i. 57. 15 1. 59. THE REIGN OF OTHO 163 support ViteUius. The same course and for the same reason was taken by Gallia Narbonensis. ^® Raetia, too, was ready to add its cohorts and alae to the Vitel- lian army, while, more important than all these, the British legions acquiesced in the choice of their German comrades.^' With these adhesions, ViteUius had no cause to fear any hostile movements in the West, and soon after the middle of January, 69 a.d., his army had begun the march for Italy. He decided to send forward the two divisions of his army by different routes into Italy. The upper army under Caecina was to march by the nearest and most direct route over the Pennine Alps down into the plain of north Italy. The lower army under Valens was to take a longer and more circuitous course, by way of south-east Gaul, and then over the Cottian Alps to the valley of the Po through Turin. He himself was to remain behind for the pre- sent, raising fresh recruits from Gaul and Germany, waiting for promised detachments from the British legions, and keeping a watchful eye on the Rhine frontier, which it was dangerous wholly to denude of troops. Nothing was more striking than the contrast between the soldiers and their new Emperor. While they were all eagerness for immediate action, he thought of Uttle but eating and drinking. If the success of the enter- prise had depended on the new Germanicus, as he aUowed himself to be caUed, Otho would have had Uttle cause for anxiety. To Valens was assigned the greater part of the lower army. While more or less weak detachments of the *• i. 77. " i. 59' 164 STUDIES IN ROMAN HISTORY four legions, i, v, xv, and xvi, were left to defend the camps at Vetera, Novaesium and Bonna, the rest of the legionary force, legion v Aland, being accompanied by its eagle, joined the expedition. To these were added most of the auxiliary cohorts and alae, this division amounting to 40,000 men.^^ The march of Valens through Gaul had the double object of im- pressing the Gallic states, and picking up reinforce- ments on the way. The march, long and circuitous in itself, was made far slower through the insubor- dination of the soldiers and the greed and rapacity of the general. After passing through the land of the Treveri, Valens entered that of Mediomatrici. At Divodurum, the chief town of the tribe, the soldiers from pure devilry attacked the citizens, and had killed 4,000 before Valens was willing or able to restrain them.i^ To avoid similar outrages, the other Gallic states did their best to appease the advancing army, and all the more, when news came of the death of Galba, and the election of Otho. This news reached Valens in the territory of the Leuci, and whatever hesi- tation there may have been among many states which had received benefits from Galba, they had nothing to hope for from Otho, and much to fear from Vitellius. In the territory of the Lingones, Valens was joined by eight Batavian cohorts, which had been separated from legion xiv, and had been ordered back to Britain. Efficient as these troops were, their insubordination increased the disorder of Valens' march, while their constant quarrels with the legionaries made their accession a doubtful advantage. Even the Aedui, though they had sided with Vindex and had been re- !• i. 61. !• i. 61. THE REIGN OP OTHO 165 warded by Galba, were forward with their offers of assistance. 20 At Lugdunum, where Valens was re- ceived with enthusiasm, he added to his army the legion I Ital., and the ala Tauriana, stationed there. This city had under Galba paid for its fidelity to Nero by the loss of its revenues, and it now saw the chance of avenging itself upon its old rival Vienna, which had been the centre of the national rising under Vindex. To excite the anger and covetousness of the soldiers was no difficult task, and it was only by a huge bribe to Valens that the AUobroges were able to save their city. 21 Nor was this the only instance in which safety was bought by presents to the general, who had actually given the order to fire the town of the Vocontii, when the payment demanded was made. It was thus amid cruelty, disorder and extortion that this division of the invading army approached the Alps. By this time, the other division of the army under Caecina had already reached Italy. Caecina had with him legion xxi Rapax from Vindonissa, and detachments from the two legions stationed at Mogon- tiacum. iv and xxii, together with the auxilia attached to these three legions. His total force amounted to 30,000 men. With this army he had been ordered to march direct for Italy, crossing over the Pennine Alps. 22 His march was quicker but even more cruel than that of Valens. A quarrel with the Helvetii, who had not yet heard of Galba's death, caused by the unscrupulous avarice of«the soldiers, gave rise to a short but disastrous campaign between the army and the local mihtia of the tribe. The latter, especiaUy «i L 66. " i. 61. «o ; i. 64. 166 STUDIES IN ROMAN HISTORY as they were attacked in the rear by the Raetian auxiliaries, were of course routed. Many of the tribes- men were butchered by their conquerors, a watering- place and health resort, which had grown up in the district, was destroyed, while the fate of Aventicum, the capital, was reserved for the decision of Vitellius himself, who was with difficulty induced to spare it.^^ Caecina had lost much valuable time in the territory of the Helvetii. It was now necessary that he should pursue with all speed his march to Italy. News which he received from the other side of the Alps made this course imperative. The ala Siliana, belonging to Otho's army, was north of the Po, possibly sent on as a small advance guard. This ala had formerly served under Vitellius in Africa, and after being sent to Egypt, was recalled by Nero to Italy. Bound by no ties of loyalty to Otho, mindful of its former general and finding itself unsupported by the main army, it was persuaded by its officers to desert to the ViteUian side, and it at once proceeded to secure for its new master four of the most important fortresses of the Transpa- dane district, Mediolanum, Eporedia, Vercellae and Novaria.2* This important news reached Caecina while still among the Helvetii. It was necessary to at once support this isolated squadron, and accordingly Caecina without delay sent forward the Gallic, Lusi- tanian and British auxiUaries, and the ala Petriana to- gether with some German veterans over the Pennine passes. For the moment he hesitated to follow this advance guard with his maiir army. He seems indeed to have had some thought of departing from the orders of Vitellius to reach Italy by the nearest route, and to 23 i.68. ■* i. 70. THE REIGN OF OTHO 167 have considered the advisabihty of crossing the for- midable Raetian Alps, in order to secure the province of Noricum. Noricum was under the command of the procurator, Petronius Urbicus, and a not incon- siderable force of auxiha. Petronius, hke the legates of Pannonia and Dalmatia, was on the side of Otho, and had akeady, as a means of preventing the invasion of his province, destroyed the bridges along the river Inn. This step on the part of Caecina might have considerably altered the course of the campaign ; it would certainly have delayed his arrival in Italy, and might have brought him into collision with the army of Pannonia, the natural support of the Norican auxiUaries, and it would have postponed and made more problematical his junction with Valens. On the other hand, the possession of Noricum was not essential to the ViteUians. That Uttle province was not necessary for his communications, and would certainly come over to ViteUius in the event of a successful campaign in Italy. Caecina therefore wisely determined to proceed by the original route. He accordingly followed his advance guard over the Pennine Alps, and descended into the plain of Transpadane Italy. His worst qualities he seems to have left behind on the northern side of the Alps. Marching down the valley of the Dorius, he kept his troops well in hand, preventing outrages on the towns and colonies, and made his way towards the Po. Whether it was Caecina with his main army which occupied Cremona, or whether that place had akeady been captured by his advance guard, is not certain. It is clear, however, that Cremona about this time feU into the hands of the ViteUians, and that a i68 STUDIES IN ROMAN HISTORY cohort of Pannonians, which held the town for Otho, was captured. 2^ It can hardly have been much later than the end of February, 2« that Caecina crossed the Alps, and whether by way of Cremona, or by some other route, he advanced with the intention of crossing the Po, and seizing the important stronghold of Placentia. Meanwhile Otho at last, though a month and a half too late, was making serious preparations to meet the invading armies. At first indeed he had attempted to appeal to ViteUius' well-known love of ease and self-indulgence by offering him large sums of money and some place of luxurious retreat on condition of his resigning his claims to the empire. Whether he went on to suggest that they should rule as joint emperors is uncertain. 27 At any rate, the correspondence, which had begun in civility, ended in undignified abuse, and mutual insults and accusations were bandied to and fro, of which all that could be said was that they were equally true on both sides, ^s Attempts, moreover, were made by both to tamper with the fidelity of the soldiers by the secret despatch of envoys, but this met with no better success, and only the recourse to arms remained. Formidable as were the forces on which ViteUius could rely, the prospects of Otho were by no means desperate. He had not, it is true, at his immediate disposal an army equal to the 80,000 or 90,000 men whom Valens and Caecina were leading into Italy. *» ii. 18. There is nothing to show whether this " cohors Pannoniorum " had been sent forward by Otho or by the Danube army. ■• " Hibemis adhuc Alpibus," i. 70. " Plut. 0th. 8. " i. 74. THE REIGN OF OTHO 169 In Rome he had 12,000 praetorians, four or five ur^>an cohorts, the legio i Adj., formed from the mannes, and some detachments from the German and Pan- nonian legions. To these he added a body of 2,000 gladiators from the training schools. These might perhaps, amount to some 25,000 men. But news had already arrived that the legions of Dahnatia, Pannonia and Moesia had taken the oath of allegiance to him, while the four legions of the two former provinces XI Claud, and the famous xiv Gem. in Dahnatia, and VII Galb. and xiii Gem. in Pannonia, had actually sent forward detachments of 2,000 men each ; the legions themselves, together with the auxilia, intending to follow 30 If only the main body of the Danube legions should arrive in time, Otho might weU hope to meet successfully the German armies. From a greater distance tidings arrived that the legions of Syria and Judaea had also accepted Otho as emperor, while Egypt and all the other Onental provinces took the same side. Even Africa, in spite of its recollection of the good government of ViteUius, glad to be freed from the rapacity of Clodius Macer, accepted first Galba and then Otho without hesita- tion In fact, all the distant and transmanne pro- vinces, actuated chiefly no doubt by the fact that Otho had Italy, the city of Rome, and the Senate on his side, maintained their loyalty to him, and looked upon VitelUus as an usurper.^^ It is probable that even Spain, in spite of the part taken by Otho m Galba's death, would have adopted the same course, had it not been more within striking distance of the German army. " i. 76. " ii. II. "i.76. 170 STUDIES IN ROMAN HISTORY But although Otho thus had on his side the greater half of the empire, and seventeen out of the thirty legions, as well as the proud but unserviceable city troops, 32 it was doubtful whether he would have time to put into the field a force equal to that of the united armies of Valens and Caecina. The eastern provinces would certainly take no part in the struggle. They would wait events. Even among the nearer Danube legions, there was unaccount- able slowness. Tacitus says, "e fiducia tarditas." But probably the real reason was the want of a single leader. It was rumoured afterwards that Antonius Primus, who so brilliantly led the Flavian invasion, offered himself as general of the Danube army and that Otho ignored the offer. If it had been accepted the issue of the campaign might well have been dif- ferent. As for the Moesian legions, it was improbable that they would arrive in sufficient force or with sufficient speed to enable them to take part in the campaign. The Moesian frontier was at this time in a critical condition. The Moesian legions had not only to guard against invasion on the part of the various Sarmatian and Dacian tribes, but also to protect the Greek coast cities on the Euxine against the Scythian nomads. ^3 At this very time there was trouble. Two auxiUary cohorts had been recently annihilated by the Roxolani, Otho's seventeen legions were : one in Rome, one in Africa seven on the Danube, three in Syria, three in Judaea, and two in Egypt. On the side of Vitellius were : eight in Germany three in Britain and two in Spain. " For Moesian frontier relations see an important inscription ( Wilm. 1 145), dating from the reign of Nero. THE REIGN OF OTHO 171 and during February 9,000 cavalry belonging to the same tribe, counting on the absorption of the legions in the civil war, crossed the Danube on a plundenng expedition. The occurrence, while it showed the danger of denuding the frontier of troops, by its for- tunate ending enabled Otho to pose for a moment as a conqueror and successful defender of the empire against barbarians. The Roxolani, intent on plunder and not expecting resistance, were attacked and ahnost annihilated by legio iii Gall, and some auxiUaries.^* This was a useful service on the part of Aponius Satuminus, legate of Moesia, but at ordinary times would perhaps hardly have been rewarded by the tnum- phal statue which Otho accorded him. The original plan of Otho ^ had been to block the passes of the Alps, and so to prevent the invading armies from entering Italy. If preparations for doing this had been begun in January, and the Danube army properly handled, this might have been done. But Otho took no step whatever till after Caecina had crossed the Alps.3« Then, after consultation with his miUtary advisers, Suetonius PauUnus, Marius Celsus and Annius Gallus, he decided on his plan of campaign. It was a double one. A part, though a very small one of his forces was to be conveyed by the fleet at Misenum along the coast of Italy, and to effect a land- ing in GaUia Narbonensis, thus creating a diversion in the rear of Valens, and with the possibiUty even of stirring up the rest of Gaul against ViteUius. The mam »« Legio III was a Syrian legion and had only recently been sent to Moesia, i. 79- . „ . • ,, as ^ 11^ 3« This is clear from 1. 87 and u. 1 1. 172 STUDIES IN ROMAN HISTORY army was to defend the line of the Po. and at the same time form a junction with the Danube army. If the former expedition had been sent out earUer, while Narbonensis was still wavering ; if adequate and efficient troops had been put on board, and, above all, if a skilful and responsible commander had been placed in charge, it might have had an important bearing upon the campaign. As it was, it did not start till Caecina had crossed the Pennine, and Valens was on the point of crossing the Cottian Alps.^' It was com- posed, in addition to the marines, of some urban co- horts and a certain number of praetorians, and it was commanded by two ex-centurions, an ex-tribune and a freedman. But these men were either incapable and without control over their soldiers, or rashly eager for fighting. As the fleet sailed northwards, the coast of Italy was treated as hostile ground, incursions were made, villages burnt and outrages committed. The province of the Maritime Alps, held for Vitellius by its procurator, Marius Maturus, was then attacked. The native militia were no match in the field for the trained soldiers, and fled to their town, Albintimilium, which was sacked by the Othonians, whose great aim was gain and plunder. Meanwhile, news reached Valens of the meditated attack upon Narbonensis, and that general, who had ahnost certainly by this time crossed the Alps, detached against the Othonians two Tungrian cohorts, several turmae of cavalry and the ala Treverorum, commanded by Julius Classicus. A part of this force remained at Forum Julii, to watch the fleet, and prevent a landing in Gaul, while the rest, reinforced by a cohors Ligurum, *' This is clear from ii. 87. THE REIGN OF OTHO 173 the garrison of the Maritime Alps, proceeded against the enemy. The two armies met near the coast in the province of the Maritime Alps, and as the fleet was able to co-operate with the land force, the CM:honians were victorious both in the first battle and also in an attempted surprise on the part of the Vitellians.^ The result, however, was not very decisive, and though Valens found it advisable to send further reinforcements, in the shape of some of the Batavian cohorts, ^^ these had apparently no fighting to do, and quite possibly, as Tacitus suggests, the motive of Valens in sending them was to separate some of these turbulent troops from his army. At any rate, both the Othonians and the Vitellians retreated after these engagements, the latter to Antipolis in Narbonensis, the former to Albin- gaunum in Liguria, and the Othonian expedition utterly failed either to embarrass Valens or to disturb Gaul. It had, however, the incidental effect of involving for the moment Corsica and Sardinia in the turmoil of contending parties. These provinces, exposed to the ravages of the fleet, and well out of reach of the Ger- man legions, had espoused Otho's cause, but a rash procurator, Decimus Pacarius, made an attempt by raising the local militia of Corsica to bring about a revolution in favour of Vitellius. The attempt failed, though beyond the death of Pacarius no serious consequences followed, the more important events in Italy causing this petty disturbance to be over- looked. *» " ii. 12-16. *• ii. 20. *o The proconsul of Sardinia at the time was Helvius Agrippa, a decree of whom is discussed by Mommsen, Gesamm. Schr, vol. V. p. 325. 174 STUDIES IN ROMAN HISTORY It was by Otho's land army in the vaUey of the Po that the decisive campaign was to be fought out. But before leaving Rome, Otho, in deference to the wish of the jealous praetorians, ordered ComeUus Dolabella, once a candidate for Galba's adoption, into a sort of semi-custody at Aquinum. At the same time, to make sure of the nobles and Senate, he compelled a number of magistrates and consulars to accompany him on his march, in appearance as advisers, in reaUty as hostages for the security of the city. . . . This was, however, by no means guaranteed by the removal of the leading men, for the results of a war so near at hand were akeady beginning to be felt by the people in the shape of rising prices and scarcity of money," and a mob so situated can never be depended upon. However, Otho had no alternative but to leave the city to itself. Delay in sunilar circumstances had akeady been fatal to Nero, and therefore with a final and somewhat cheap attempt to gain a parting popu- larity, by granting the remains of the recovered gifts of Nero to those who had just returned from exile, he entrusted the city to the Senate under the direction of his brother, Salvius Titianus, and started on his march northwards. Though Otho so far had had no experience of mihtary affairs, and was more at home in the use of the looking- glass than in the employment of arms and the discipline of a camp, he roused himself in this last scene of his Hfe, and shared with his soldiers aU the hardships of the campaign. He had, moreover, distinguished generals at his disposal in Suetonius PauHnus, the 41 i.89. THE REIGN OF OTHO 175 conqueror of Boudicca, Marius Celsus,*^ Annius Gallus and Vestricius Spurinna.*^ If only Otho throughout the campaign had listened to the advice of his senatorial commanders, rather than to the inexperienced Proculus,** he might, with the considerable forces at his disposal in Rome, and the support of the formidable Danube army, have brought the war to a successful issue. It was now, however, as we have seen, too late to take the course originally intended, and to defend the passes of the Alps against the two advancing divisions of the Vitellian army. News had akeady arrived that Raetia had declared for ViteUius, that the ala Siliana had deserted and carried over with it four important fortresses of North Italy, and that Caecina himself had crossed the Alps. There would appear to have been no Othonian forces north of the Po to oppose his southward advance untU he reached *' Paulinus had his first important military command in 42 A.D. as legate of the African army, and had taken part in the campaigns which resulted in the establishment of the two Mauretanian provinces (conf. Plin. Hist. Nat. v. i). After his consulship he became legate of Britain in 59 a.d. {Ann. xiv. 29). There he confirmed his reputation by putting down the desperate rising which at one time threatened the loss of the province. Owing to some want of judgment on his part in dealing with the conquered country, he was recalled in 61 a.d. From that time however, there was, as Tacitus says, " nemo ille tempestate mili- taris rei caUidior," Hist. ii. 31. Marius Celsus had won his mili- tary experience under Corbulo in the East. He was there legate of legio XV ApoU., which in 63 a.d. was sent from Pannonia to reinforce the eastern army. {Ann. xv. 25.) *' Of the earlier mihtary career of Annius Gallus nothing is known. He had, however reached the consulship under Nero, and we find him chosen as legate of Upper Germany by Mucianus. Vestricius Spurinna was bom in 24 a.d., as appears from Phn. Ep, iii. i, 10. But of his earlier career we have no know- ledge. «« i. %j^ 176 STUDIES IN ROMAN HISTORY THE REIGN OF OTHO 177 Cremona/where we find that a cohort of Pannonians was captured, no doubt at the time when Cremona was occupied.** From this point, therefore. Cremona, an important fortress on the left bank of the Po, became the base of the German army, and Tacitus states that the plain of Italy with aU its towns between the Alps and the Po were in the possession of the Vitellians." It is clear, however, that this statement can only be true of the western half of North Italy, as it appears from what followed that communications were still open on the north-east with Aquileiaand the Danube regions The detachments of the Pannonian and Dalmatian legions were advancing from that direction, perhaps with the original object of securing Cremona, to which a cohors Panuoniorum had been sent. The roads from MuUeut through Vicetia to Verona, and through Batavium and Ateste to Hostilia were still open. To maintam this communication with the Danube provmces, and at the same time to secure the line of the Po from Placentia towards the East were therefore clearly the first and essential steps to be taken by the Othonian eenerals." Whether some preliminary steps had been taken before this, to oppose the Vitellians north of the Po is uncertain. We only know that the ala Siliana was in North Italy, when it deserted to Caecma ; that a cohort of Pannonians was stationed at Cremona and that 100 cavalry and 1,000 soldiers from the fleet were cut off, apparently at the very beginnmg of the campaign, between Placentia and Ticmum>« But beyond these points, there is no indication that any troops were despatched to the north before the be- ginning of March. By that time, the Othonian plan of defending the line of the Po was seriously threatened by the capture of Cremona. This, however, might be retaken ; and at any rate, further advance might be stopped. Accordingly, Vestricius Spurinna and Annius Gallus were now sent forward with a strong advance guard. Spurinna with three praetorian cohorts, i,ooo legionaries and a few cavalry was just in time to throw himself into Placentia, a strong fortress on the right bank of the Po, and the obvious spot to bar the Vitel- lian passage of the river. *^ Gallus with legio i Adj., two praetorian cohorts and some cavalry marched more to the north-east, probably by way of Hostilia, in order to form a junction with the advancing con- tingents of the Illyrican army, and to secure the com- munications with the Danube district. Otho himself started a few days later, on March 14, with his main army. It consisted of the remaining praetorian co- horts, and a large force of Marines, while 2,000 gladiators were despatched either by Spurinna or Gallus to take up a position on the south of the Po, opposite Cremona. Otho's own force was under the command of Suetonius Paulinus and Marius Celsus, though he already showed signs of putting more trust in Proculus, one of the praetorian praefects. He was also accompanied by a considerable number of senators, including L. Vitellius, the brother of his rival, but these, during the actual campaign, were left behind at Mutina. Spurinna successfully occupied Placentia, but only just in time, for Caecina had already sent across the Po from Cremona bodies of his Batavian auxiliaries, in the direction of Placentia, and was preparing to follow *» ii. II and 17. ii. 46. « U. II. u. 17. *» ii. II and 18. N 178 STUDIES IN ROMAN HISTORY With his German legions.'" Spurinna's great difficulty was to curb the reckless impatience of his troop^- who, distrusting their general's loyalty to Otho, insisted on marching against the enemy. When ordered, however by Spurinna to fortify a camp in the open plain not far from the Po. they began to reaUze the hardship of a campaign, and this, aided by considerab e tact on the pa'rt of Spurinna, induced them to withdraw again into the city. This was only one instance of the common distrust shown by the Othonian soldiers towards their senatorial generals during this campaign. Spurmna then proceeded to strengthen the waUsand battlements^ Caecina meanwhile with a far superior force had crossed the Po, and a series of obstinate assaults u^n the town followed. These, however, were not marked by much judgment or skill on the part of Caecma. and in the end Spurinna with his handful of troops suc- ceeded in repelling the attack, though m the course of the fighting the famous amphitheatre outside the walls was destroyed." Placentia was thus saved though its importance to Otho was considerably lessened bv the Vitellian occupation of Cremona. Caecina. foiled and disappointed, led his army back across the Po into Cremona. Meanwhile Annius Gallus had crossed the Po in all probability at Hostilia. and was no do"bt marching towards Verona, with a view of -"^etrng the 8.000 legionaries on their way from Pannonia and Dalmatia. » U ',°' Tacitus does'not stete where Gallus crowed the p„ nor l^en that he had crossed it, nor does he mention the Po, nor even ^fat ne detachments, but his account of THE REIGN OF OTHO 179 News, however, reached him of Caecma's attack on Placentia, and fearing for the safety of that important position, garrisoned as it was by so small a force, he was proceeding to march to its relief with legio i Adj., 53 when a messenger from Spurinna reported that Caecina had raised the siege and was about to enter Cremona. On this, Gallus, who was apparently marching along the Via Postumia, halted at Bedriacum, where, in spite of the impatience of legio i to advance at once upon Cremona, he collected all his forces and encamped.^ Bedriacum was a place of no importance in itself, but its position made it of great strategical value to Gallus. It was situated on the Via Postumia, the road running from Genoa to Cremona, and from there through Bedriacum to Hostilia. According to the "Tabula Peutingeriana," it was twenty-two miles east of Cremona, and it stood at a point on the Via Postumia where another road diverged to the north- east, leading to Verona, thirty-five miles distant, and from there through Vicetia to Aquileia.^^ It is clear, therefore, that Gallus selected Bedriacum with a double object of threatening Cremona to the west, and of waiting for the main body of the Pannonian and Dalmatian legions, which would join him by way of Verona. the first fighting. The other vexilla must almost certainly have arrived too. 63 n. 23. 54 ii. 23. ** iii. 15. The SchoHast on Juv. ii. 99, basing his information on the historian Pompeius Planta, makes the distance from Cremona twenty miles. Mommsen believes that the site of Bedriacum is the modern Calvetone, between Piadene and Bozzolo. i8o STUDIES IN ROMAN HISTORY By this time, Otho with his main army had reached the Po and the greater part of it under the command of Suetonius PauUnus and Marius Celsus crossed the river and joined GaUus at Bedriacum. Otho hunself remained behind at Brixellum on the south bank of the rivei, and so deprived his troops of the inspiration of his personal presence." The two armies were now facing one another, at a distance of twenty-two miles, the Othonians at Bedri- acum, Caecina's force, not yet joined by Valens, m an entrenched camp outside the eastern walls of Cremona. The 2,000 gladiators of Otho under the command of Martins Macer were posted on the right bank of the Po, opposite Cremona, and these, with a view of harass- ing the ViteUians, crossed the river in boats and attacked the enemy's outposts. They inflicted con- siderable loss on some of the German auxiliaries, whom they drove back into Cremona. They were assisted in this movement by some forces from the main army ,^8 and when the generals thought it prudent not to push forward the attack at this point, and withdrew the gladiators again to the right bank, a mutiny all but broke out, the soldiers accusing Paulinus, Celsus and GaUus of treachery and bad faith. Several skirmishes took place between the cavalry outposts of the two armies, and the Othonians under the skilful du-ection of Pauhnus won some brilliant successes. But the ill-assorted army of praetorians and legionaries was under very imperfect control, and Otho, bewildered »• Plutarch. 0th. 5. explicitly states this, and is confirmed by Suet. 0th. 9. though the account of Tacitus is rather in favour of the view that he only left his army before the final battle. »' This appears from iii. 26. »« This is implied by u. 23. THE REIGN OF OTHO 181 by the constant complaints of his soldiers, and not himself wholly trustful of his generals, decided to send for his brother, Titianus, from Rome, and to entrust to him the supreme command. ^^ But before the arrival of Titianus, a more import- ant collision took place between the two armies. Caecina, vain and impetuous by nature, was chafing under repeated failures. His attempt on Placentia had miscarried, his auxiliaries had been routed by the force of gladiators, and so far fortune had been against him in all the skirmishes with the enemy. He was anxious, too, to obtain some decisive success before the arrival of his colleague, Valens. He therefore advanced from his camp with a picked force of auxiliaries and cavalry, and proceeding twelve miles from Cremona prepared an ambuscade, near a ruined temple of Castor and Pollux, concealing his auxiliaries in the woods, which at that spot overhung both sides of the Via Postumia. The cavalry were to advance still further along the road, and then by a feigned retreat to entice the enemy within reach of the concealed auxiliaries. Unfortunately for the success of this plan, the device was betrayed to the Othonian generals, who nevertheless advanced with a portion of their army, the cavalry being under the command of Celsus, the infantry under that of Paulinus. They arranged their forces in the following way. On the raised cause- way of the road itself, were three cohorts of praetorians, drawn up in deep columns. On the left was the de- tachment of legio XIII, with four auxiliary cohorts, and beyond them 500 cavalry ; on the right was legio I Adj., together with two auxiliary cohorts, and on w Plut. 0th. 4. l82 STUDIES IN ROMAN HISTORY their flank 500 cavalry. In the rear were posted 1,000, praetorian cavalry, ready either to complete a victory or to retrieve a repulse. When the Othonians came within sight of the Vitel- lian cavahy, Celsus with the 500 cavalry from either wing advanced against them. The ViteUians, according to arrangements, turned in flight. Celsus, however, aware of the stratagem, held back his soldiers, on which the Vitellian ambuscade, rising prematurely, charged in his direction. Celsus with his cavaky retreated, drawing the enemy after him. Meanwhile, Pauhnus had drawn back the line of infantry " en echelon,'* the three praetorian cohorts retiring along the road, while the legionary force on either wing similarly re- treated, the auxiliary cohorts remaining stationary in their original position. The result was that as the Vitellians advanced, they found themselves faced by the legionaries and praetorians in front, while they were hemmed in by the auxiliary cohorts on the two flanks. Celsus and his cavalry had meanwhile retreated between the lines of the legionaries, which had then closed up to protect them. The cavaky, however, had still an important part to play, for Celsus, as soon as he had retreated behind the infantry, again wheeled round with his two divisions, to the right and left, joined apparently by the Praetorian horsemen from the rear, and proceeded to attack the Vitellians on the two flanks. If Paulinus had only at once given the signal for his infantry to charge, the result would have been the annihilation of the Vitellian force. He, however, was cautious by nature, and deemed it ad- visable, before advancing, to have the ground cleared, and some dykes at the side of the road filled up. Owing THE REIGN OF OTHO 183 to this delay, the Vitellians were able partially to escape the charge of the cavalry by retking into the vineyards which lined the road. When, however, Paulinus did order an advance, the rout of the Vitellians was complete, as Caecina made the mistake of sending forward his supporting cohorts one by one, and so those akeady routed communicated thek disorder to those still undefeated. The difficulty of Caecina, too, was increased by the fact that a mutiny broke out in his camp, the soldiers there loudly complaining that they had not been allowed to share in the battle. So great indeed was the confusion, that if Paulinus had continued his victorious advance, there seemed the possibility even of storming the Vitellian camp. Paul- inus, however, and probably with wisdom, decided against this course. Nearly twelve miles would have had to be traversed ; a great part of the Vkellian army had not been engaged in the fighting, and to dash his weary troops against these would have clearly been to court disaster. Paulinus accordingly led back his forces into Bedriacum.*® Still the victory, though not decisive, had been a serious check to Caecina, and in spite of the discon- tented and mutinous spirit of his army Otho only showed his own incapacity in persisting in his intention to replace Paulinus and Celsus as commanders-in-chief by his brother Titianus, especially as the change would almost certainly mean the virtual command of the inexperienced Pro'culus. ' ' Meanwhile the army of Valens was at last approach- ing. Its march had been delayed by the rapacity of soldiers and general, and latterly by the insubordinate •« ii. 24-26. i84 STUDIES IN ROMAN HISTORY conduct of the Batavian auxiHaries. We have already seen that Valens had had to despatch a few auxihary troops against the Othonian fleet, and when news came of their defeat, determined to send some further remforcements. The Batavian cohorts, which had joined him in the territory of the Lingones, had been a constant source of disorder through their quarrels with the legionaries, and Valens took this opportunity of getting rid of some of these efficient but turbulent soldiers. On it becoming known, however, in his army that the Batavians were to be separated from the main body one of the worst mutinies of the war broke out. Whether the Batavians were actually despatched to the coast is uncertain. At any rate, Valens himself ahnost lost his life, and it was only through the tact of the camp-praefect that discipline was restored. When it was, Valens continued his march by way of Turin to Ticinum, and it was there that news arrived of Caecina's defeat.«i The result was all but another mutiny, the men declaring that it was through the rapacity and delay of Valens that they had missed the battle. However, there was no further delay. Spurinna with his 4,000 or 5,000 men at Placentia was quite unable to put any obstacle in the way, and Valens, with an army twice as large as that of Caecina,«2 and therefore not far short of 60,000 men, hastUy effected a junction with his coUeague at Cremona. Though there was already some jealousy between the two generals, this was for the moment put aside, and even the soldiers, sobered by the recent defeat, were now amenable to discipline. The combined army can •1 u. 30. 6S ii. 30. THE REIGN OF OTHO 185 hardly have numbered less than 85,000 men, while Otho's forces could not at the very outside have amounted to more than 35,000 men.«3 On the Vitellian side, the 40,000 of Valens had been increased by the legio i Ital., the ala Tauriana, and the eight Batavian cohorts, while Caecina had been strengthened by the ala Siliana. The Vitellians were eager for battle, but determined not to take the initiative.** Among the Othonians opinions were divided, and Otho himself for the first time came to the camp at Bedriacum, to attend a council of war. Paulinus and his fellow-generals were strongly in favour of delay. The Vitellian army, they represented, was not likely to grow larger than it was, since Vitellius would not venture to leave the Rhine frontier exposed by leading into Italy his remaining troops. It might even be- come smaller, since in an enemy's country its suppHes might easUy fail, and the warm climate of Italy would almost certainly affect the health of the German soldiers. On the other hand, they themselves were in a friendly country, well supplied with provisions, and, what was even more important, were expecting large reinforcements from the Danube army. In particular, the famous legio xiv was only a few days' march away. Meanwhile, they had all the prestige «' I assume that Otho had 25,000 in Rome to start with. Of these, about 4,000 were sent with the fleet ; 4»ooo were with Spurinna at Placentia. but these had now rejoined Otho. To these must probably be added 8,000 from the four legions of Pannonia and Dalmatia,and most likely the remainder of legion XIII. since its legate is present at Bedriacum ; whUe, lastly, the six auxiliary cohorts, mentioned in the battle against Caecina, had come from the Danube army. «4 u. 31. i86 STUDIES IN ROMAN HISTORY conferred by the possession of Rome, and the support of the Senate. Let Otho therefore protract the war.^s In opposition to their views, Otho was feverishly eager to escape from his present suspense and un- certainty by an immediate battle, and Titianus, who had now arrived,^** and Proculus upheld this view in the council of war, maintaining that at the present moment all were eager and confident,®' while delay would blunt their spirits and damp their zeal. There seems no reason to suppose that Paulinus and Celsus were actuated by any other motives than mili- tary considerations. It is no doubt true that the senatorial order generally was supporting Otho rather through necessity than from any preference for him over Vitellius, and indeed the character of both claimants was so mean and contemptible, that enthusiasm for either was out of the question. It was perhaps owing to this circumstance and what seemed the natural probabilities of the case, that the report afterwards arose, alluded to both by Tacitus and Plutarch,^ that the idea occurred to both armies of coming to some agreement, and either themselves choosing in common some third candidate, or entrusting the choice to the Senate. It was for this reason that Otho's generals wished for delay, Paulinus hoping that the choice might fall on himself, while the precipitancy of Otho was the necessary means of anticipating treason. With regard to the Vitellians, any such design on their pari •• ii. 32. •• L. Salvius Otho Titianus, the elder brother of the Emperoi , had been consul in 52 a.d. {Ann. xii. 52) ; as proconsul of Asia in 63 A.D. he was " in omnem aviditatem pronus " {Agric. 6) ; he was magister of the " fratres arvales " (Wilm. 2807). •' Plut. 0th. 8. " ii. 37> and Plut. 0th. 9. THE REIGN OF OTHO 187 is inconceivable. The interests of Valens and Caecina were bound up with those of ViteUius, while the German army was Uttle likely to give up its claims to appoint an emperor either to the Senate or to any one else. Nor could the Othonian generals possibly hope for such a way out of their uncomfortable position. They knew too well how suspicious their soldiers were of their every action. They had seen the experiment tried and fail of selecting a successor to the empire simply for his character and virtue. But however this may be, the opinion of Titianus and Proculus prevailed, and it was resolved to fight,*® while Otho, with an almost inconceivable want of judgment or want of spirit, allowed himself once more to be persuaded to retire to Brixellum, not only withdrawmg from his troops the inspiration of his presence, but, what was equally important, taking with him a considerable force of praetorians to guard his person.'® Valens and Caecina, in spite of their superior num- bers and the fighting spirit of their troops, determined on this occasion not to be the attacking party. The conditions of civil war were favourable for the trans- mission of news from one army to the other,'i and hearing through scouts or deserters the decision of the council of war, they remained quiet but alert, and ready to take advantage of the first false move on the part of the enemy. Meanwhile, partly to irritate the enemy, partly to strike a blow at the troublesome gladiators on the south bank, partly to give their men employment, but not with any serious intention of transferring the war to the right bank of the Po, they •• ii. 33. '• ii. 33. " ii- 34- i88 STUDIES IN ROMAN HISTORY began to construct a bridge of boats, opposite the camp of Macer and his gladiators. On the furthest boat they raised a tower from which to harass the Othonians as the work proceeded.'^ The latter first attempted to meet this annoyance by building a tower of their own, and, when this means failed, they sent fireships down stream against the bridge, which caused the Vitellians somewhat ignominiously to decamp. Shortly after, another skirmish near the same point once more exhibited the insubordination of the Othonian troops. A number of Batavian auxiUaries, belonging to the army of Valens, swam across to an island in the river and occupied it. Macer led his gladiators across in boats, but amid much confusion and fighting at a dis- advantage against the agile Germans, they were re- pulsed. The customary recriminations against their leader followed, and once more Otho was induced to change his general. Macer was deposed and the com- mand over the gladiators given to Flavins Sabinus. At the same time, Spurinna received orders to leave a small garrison at Placentia, and with the rest of his force to join the main army at Bedriacum.'^ Meanwhile, in obedience to the decision of the council of war, Titianus and Proculus, who had now assumed the command of Otho's army, proceeded as a first step to march out of Bedriacum, and to encamp for the night six miles along the Via Postumia.'* It was " Plut. 0th. lo. " ii. 35. 3^- '* Plut. 0th. 1 1 ; sostades. Tacitus, ii. 39, says " ad quartum a Bcdriaco." Plutarch, however, had himself been over the scene of the battle with Mestrius Florus, a member of Otho's suite, and therefore where his distances differ from those of Tacitus, which they only do on this one point, they should probably be accepted (Plut. 0th. 14). THE REIGN OF OTHO 189 significant of the new p'eneralship, which had super- seded that of Paulinus, that, although there were plenty of small streams in the neighbourhood, the spot selected was absolutely without water, and the troops were considerably inconvenienced even during the one night of their encampment there. Next day, there was another discussion as to the battle, and loud de- mands were made by the soldiers that the troops on the other side of the Po should join the main army, and that Otho should himself be present. And indeed the army which was thus recklessly advancing against an enemy strongly entrenched aud nimibering over 80,000 men, could not have amounted to more than 28,000.'^ However, a special messenger arrived from Otho, ordering an immediate advance, and Titianus and Proculus once more set out. Their plan was apparently to secure the co-operation of the gladiatorial force in the final attack by marching to the Po, at a point opposite to its confluence with the Adra — ^the modern Arda — a river which joins the Po from the south, about seven miles below Cremona, and therefore only a few miles from where the gladiators were en- camped. Here they might encamp for the night, and make their assault upon the ViteUian position next day, supported by the gladiators, who woiild at any rate detain some part of the enemy's force. The dis- tance from their present camp to the mouth of the Adra was fourteen miles, as the crow flies, but as they would follow the Via Postumia for a considerable part of the way, and then deflect to the left, the distance " Not only had Otho taken with him " valida manus " to Brixellum, but a strong force had been left behind at Bedriacum, which Tacitus calls ' ' magnam exercitus psirtem. " (ii. 3 3 and 44. ) igo STUDIES IN ROMAN HISTORY to be actually traversed would be about sixteen miles.'* It was at best a mad and reckless plan, only under- taken in obedience to the repeated orders of Otho. Its dangers were clearly pointed out by Paulinus and Celsus. These generals urged the practical certainty that they would not be allowed to take this course immolested by the enemy. By a short march of only about four miles, the Vitellians could attack them on the march, " incompositos in agmine," weary and encumbered with baggage, or, if they chose, could come upon them at the end of their march, while engaged in entrenching their camp.'' These objections were, however, overruled. Probably neither the impatient soldiers nor the unskilful generals — '* imperitia properantes " — were averse to fighting on the march, and the disastrous enterprise began. '^ '• That there was some idea of a second encampment is clear from the words of Tacitus ii. 40, " non ad pugnam sed ad bel- landum profecti," and " vallum molientes." " It seems clear that " incompositos in agmine " and " vallum molientes " represent two different points, the former probably the point where they would deflect from the Via Postumia, which Paulinus estimated as about four miles from Cremona, the latter at the confluence. '* The account given in the text depends upon the alteration of a single letter in Tacitus, " Adrae " for " Aduae," and the substitution of the " 50 stades " of Plutarch for the " ad quartum a Bedriaco " of Tacitus. The great crux of the campaign has of course always been the statement of Tacitus in ii. 40, " non ad pugnam, sed ad bellandum profecti confluentes padi et aduae fluminum, sedecim inde milium spatio distantes petebant." The difficulties in this statement are two. (i) The distance from the Othonian camp to the mouth of the Adua was not sixteen miles but twenty-five miles as the crow flies, since the confluence in question is seven miles to the west of Cremona. (2) It is perfectly inconceivable that, if any sort of flanking movement was in- tended, Tacitus would have been absolutely silent on the subject. He must have, of course, known that the Adua is west of Cre- THE REIGN OF OTHO 191 After the march commenced, things turned out almost exactly as Paulinus had predicted. The Othonians began their march towards Cremona with all their bag- gage mixed up among their lines. If they expectecf to find the enemy off his guard they were woefully mis- mona, and yet he entirely agrees with Plutarch in describing the affair as a reckless frontal attack. My suggestion avoids both these difficulties and brings Tacitus and Plutarch into complete harmony. Plutarch, 0th. 11, does not mention the objective, the confluence of the Po and the Adra at all, because, as things turned out, it was never reached. He states that the Othonians marched twelve miles — 100 stades — and then fell in with the enemy, which agrees exactly with the four miles which according to Paulinus in Tacitus the Vitellians would have to march. The sixteen miles of Tacitus were the estimated dis- tance of the march to the confluence. Mommsen in his well- known monograph, " Die zwei Schlachten bei Bedriacum," re- published in his Gesammelte Schriften,\o\. iv. pp. 354 foil., suggests that Tacitus in his sixteen miles has confused the ultimate objective of the Othonians with the end of their first day's march. He supposes that a wide outflanking movement was intended, with a view of compelling the Vitellians to fight by threatening their communications. The Othonian generals aimed at getting astride the Cremona-Brixia road, which he supposes might involve a march of about sixteen miles. Having secured this position by the first day's march, they would then, if not attacked by the Vitellians, by a second day's march reach the mouth of the Adua. But several difficulties are involved in this explana- tion. In the first place, no strategical reason is suggested for such an objective as the mouth of the Adua. It could not be to co-operate with the garrison of Placentia, for that garrison had already been recalled. It could not have been to protect Placentia, for there is no evidence that it had again been threatened. On the other hand, by reaching it, the Othonians would hopelessly cut themselves off from the re-inforcements they were expecting from the Danube army. Nor would a position on the Cremona-Brixia road have seriously threatened the Vitellian communications, or necessarily induce them to fight. Nor, as far as distances are concerned, does it seem pos- sible to reconcile the sixteen miles required to get astride this road with the four miles which, according to Tacitus, the Vitel- I 192 STUDIES IN ROMAN HISTORY taken, and before it was time to leave the Via Postumia, the troops found themselves within touch of the enemy, weary with their long march, confused and hindered by their baggage waggons, and distrustful of one another and of their generals.'* Caecina was on that day absent from the camp, engaged upon the construction of the bridge across the Po. A message reached him, however, that the Othonians were not far off, and he hastened back to the camp. Here he foimd that Vcdens had not been caught unprepared, and had already given the signal for the army to march out against the enemy. The battle began with a charg(3 of the Vitellian cavalry, which was, however, repulsed by some auxiliaries of the Danube army, and the Vitellians were only pre- vented from taking refuge within the entrenchments by the resolute conduct of legio i Italica. On the causeway of the road itself there was some stubborn hand to hand fighting, but no room for tactics. The ground on either side of the road was overgrown with vineyards, and intersected by trenches. Among the Othonians the officers lost their heads, and the men fought or fled as each was brave or cowardly. A re- port, due probably to treachery, spread that the enemy was ready to desert Vitellius. Accordingly the front ranks instead of attacking saluted ; and this not only lians would have to march. Besides, the account both of Tacitus and Plutarch shows that the battle took place near the Via Postumia, and only a very few miles from Cremona. Momm- sen's explanation, therefore, cannot be accepted. On the modi- fication and development of Mommsen's view in Mr. B. W. Henderson's Civil War and Rebellion in the Roman Empire, published after this note was written, see appendix at the end of this chapter. THE REIGN OF OTHO 193 7» n. 41-42. put them at a disadvantage against the Vitellians, who at once attacked them, but it also gave the im- pression to the ranks behind that treachery was medi- tated A number of the praetorians fled without striking a blow,8o but the rest of the army, though placed at every disadvantage, made an obstinate re- sistance. Especially conspicuous was the legio i Adj., which, though never before engaged in battle, main- tained in an open plain between the road and the Po a not unequal combat with the veteran legion xxi Rapax, which formed the flower of Caecina's army. In another part of the battle legio xiii from Pannonia was repulsed by the Vitellian legion v Alauda, while a detachment of the famous legio xiv, the conquerors of Britain, was overpowered by superior numbers. The chances of the Othonians, however, were from the first des- perate. They were outnumbered by nearly three to one, weary with a long march, and finally deserted by their generals. The gladiators from the other side of the Po had attempted to cause a diversion by crossing the river and attacking the Vitellians on the flank, but Alfenus Varus with the Batavian cohorts had met and repulsed them, cutting most of them to pieces as they fled back to the river. When after this success the Batavians charged the already wavering Othonians, their rout was complete, and the whole force was soon in full flight, pursued by the conquerors.^i The way back to Bedriacum was a long one, and no quarter was shown. In consequence the carnage was unusually heavy. At one particular spot, near the temple of Castor and Pollux, Plutarch, who went over the ground some years afterwards, was informed «*> Flut, Oth. 12. " ii. 41-43. 194 STUDIES IN ROMAN HISTORY that there was a pile of corpses as high as the pediments.s^ Probably a stand had been made here by some of the Othonians, who were cut to pieces to a man. Of the generals, Paulinus and Proculus, fearing the rage of the soldiers, by different routes avoided Bedri- acum,83 and subsequently repaired to Vitellius, whose punishment they escaped by a false confession of their own treachery, Paulinus basely asserting that the long and fatiguing march before the battle, which he had really opposed, was made in consequence of his advice.84 Annius Gallus, who being disabled by a fall from his horse had remained in camp, and Titianus and Celsus did their best to coUect the routed troops in Bedriacum. Celsus, however, made no secret of his conviction that the cause was lost. Otho, he said, would never wish his troops to continue the war after such a defeat. 85 The soldiers, however, were for the moment furious at their defeat, and, as usual, ready to throw the blame upon their officers. Some of these, and in particular the legate of legio xiii «« feU victims to the mutinous spirit of the men. The praetorians, in spite of their doubtful conduct in the battle, now that they were again behind their entrenchments were all in favour of resuming the struggle. The defeat was due to treachery, but Otho's position was still not desperate. A strong force had remained in Bedriacum which was still unvanquished. There were strong divisions on the south of the Po, under Flavins Sabinus, and especially under Otho himself. THE REIGN OF OTHO 83 ii. 44. •* ii. 6o. " Plut. 0th. 14. 85 Plut. 0th. 13. . u 1 8« His presence seems to prove that the whole legion had arrived. 195 The legions from Moesia were approaching. It was all that Annius Gallus could do to curb these reckless troops for the night by pointing out that whether the war was resumed or not, their hopes depended on union and obedience. Meanwhile the victorious Vitellians had bivouacked five miles from Bedriacum, not deeming it advisable to attack the camp that night, and counting upon a capitulation on the following day. Nor were they disappointed in this expectation. A night's reflection sobered the Othonian troops. Celsus and Gallus were sent to the ViteUian camp, and though there was some delay, during which Titianus once more attempted to man the walls, all serious thought of resistance was abandoned, and the camp at Bedriacum together with the whole army there was surrendered. ^^ The news of the battle reached Otho, who was anxiously expecting it, at Brixellum. He could cherish no iUusions as to what it meant. No doubt the prae- torians, who were still with him, were devoted and loyal. Like their comrades at Bedriacum, they were eager to continue the war. Their interests, indeed, were clearly bound up with Otho. Not only was he the emperor of their creation ; not only had he lavished privileges and indulgences upon them, not only had he endeared himself to them by his brave and soldier- like conduct during the march from Rome, but they knew fuU weU that from Vitellius, even if their lives were spared, there was nothing to be expected but dismissal from their privileged service. But Otho knew well that he had played and lost his last stake. It might be true that some of the Moesian " ii. 44, 45 ; Plut. Oth. 13. i 4 196 STUDIES IN ROMAN HISTORY army had reached Aquileia,®^ but in any case it was hopelessly cut off from him by the victorious Vitellian army at Bedriacum. All communication with the north-east of Italy was now broken, and the soldiers with Otho himself, however devoted, were too few to be seriously reckoned. The only alternatives for Otho were to await the executioners of Vitellius, or to face a voluntary death. He chose the worthier course. That he showed more dignity, more intrepidity and more consideration for his friends and followers in his last hours than we should have expected from so selfish a volup- tuary, is certain. But probably the rhetorical speech attributed to him by Tacitus is apocryphal, and to compare his death with Cato's argues more for Martial's imagination than his knowledge of history. ^^ After Otho's funeral, which took place amid an ex- traordinary display of devotion on the part of the praetorians, the soldiers clamoured round Verginius Rufus, who was with Otho at Brixellum, and urged upon him the acceptance of the empire. Verginius, however, refused again, as he had refused before, the dangerous honour, as well as the mission of treating with the Vitellian generals. This was finally under- taken by Rubrius Gallus, and the Othonian forces at Brixellum as well as those higher up the river under Flavins Sabinus followed the example of their comrades at Bedriacum, and submitted to Valens and Caecina.^ " ii. 46. *• Mart. V. 32. Tacitus believes, but surely without reason, that the war might have been resumed with some chance of success, ii. 46. *° ii. 51 ; Plut. 0th. 18. APPENDIX TO CHAPTER II Since this chapter was written, I have read with great interest and admiration Mr. Henderson's new book. Cm/ War and Rehelhon in the Roman Empire. In his account of the campaign between Otho and the Vitellian generals Mr. Henderson finds great fault with the narrative of th^'foHowi^ I^'^^ ^'''''*' ^""^^ ''P'''' ^""^ criticism are I. Tacitus misunderstands the reason of Caecina's hesitation before crossing the Alps. There was no idea Ir^^^'^V^^^^''"'^'''^' ^"^ ^^ P^^^^"^ t^^^"gh Raetia and over the Brenner Pass into Italy, so as to " penetrate " or thrust a wedge between the two divisions of the Othonian army at Placentia and Aquileia. 2. Tacitus fails to assign its proper place in the scheme of strategical defence to the expedition of the Othonian fleet 3. In describing Spurinna at Placentia as compelled by his soldiers to march out into the plain. Tacitus is repeating a silly story, which is merely the reflection of camp gossip. 4. Tacitus describes Caecina's bridge opposite Cremona trv T'^ ^'''^' "^^^"^^^ '^ ^^^ ^^^"y a^ example of tactical penetration." with the object of cutting off the Othonian army from Placentia. 5- The Othonian movement leading to the battle of w^f rT' /"" ^^' ^''''^ ^'^"^ ^ '^^^^^'^ f^^^tal attack. 1 n.l u '*^P ""^ ^ ^""^^^* '^^^' th^ author of which was Otho himself. x,.L^r"x*..^''''^P^ ^'■- Henderson's view, or admit the validity of his criticism on any of these points, and I have a ready dealt fully with them all in an article in the Journal of Philology, vol. xxxi. No. 61. With regard to the battle of Bedriacum, however I 197 I ig8 STUDIES IN ROMAN HISTORY must here recapitulate my objections to Mr. Hendersons theory, because, though the most ingenious, it seems to me the least tenable of all the explanations of this famous crux. Mr. Henderson accepts the traditional reading, " Con- fluentes Padi et Aduae," and bases his theory upon it. According to him, Otho himself, in opposition to the too cautious policy of Suetonius Paulinus, formed the brilliant conception of a strategical envelopment of the Vitellian army at Cremona. The force at Bedriacum was, after advancing some distance along the Postumian road, to make a flanking movement to the right, and, not with a view of forcing a battle but on the contrary counting on the immobility of the enemy, to pass close round the north side of Cremona, and without further halt, to fortify a position at the mouth of the Adua. Here it would be within touch of Placentia and would cut the Vitellian communications with Gaul. At the same time, by a simultaneous move- ment, the Danube army was to occupy Bedriacum, and then advance against Cremona. In this way Cremona would become the Metz of the campaign ; it would be enveloped by the two divisions of the Othonian army, trapped and forced to capitulate. To carry out this design, the Othonian generals on the first day advanced, not four miles as Tacitus states, nor six miles as Plutarch affirms, but as Mr. Henderson decides, fourteen miles. They are then eight miles from Cremona, and fifteen, as the crow flies, from the confluence, but Mr. Henderson saves the sixteen miles of Tacitus by allowing one mile extra for the flanking march. Then followed another discussion among the generals, Paulinus urging that they had come too far along the road, and might at any moment be attacked by the enemy, who had only to cover four miles in order to reach them — a characteristic under-estimate, according to Mr. Hen- derson, since the real distance was eight miles. Then arrived Otho's messenger, and either because the generals misunderstood him, or because Otho thinking the flank march now impossible, gave orders for a frontal attack, the army blundered on along the road, until it fell in with APPENDIX TO CHAPTER II 199 the Vitellians. Of the simultaneous movement of the Danube army, we hear nothing. Such are the essential points in Mr. Henderson's ex- planation. He maintains, perhaps not unfairly, that his view of Otho's strategical intention does not stand or fall with his manipulation of the distances, as stated by Tacitus. At any rate, I shall consider the two questions separately. In dealing with the conception of a strategical envelop- ment, it is essential to remember that it involved two parts, each necessary to the other, the flank march to the confluence and the simultaneous movement of the Danube army. I. Was the flank march, I will not say possible, but one which under the circumstances any sane general could have conceived to be possible ? It will help us to answer this question if we remember — (i) that the Othonians numbered less than 30,000, the Vitellians more than 80,000 men ; (2), that it was not to be a wide turning movement, like that suggested by Mommsen, but a march close round the walls of Cre- mona ; (3) that the design was, not to court attack, but to reach the objective unmolested. What chance was there of success ? Mr. Henderson himself in so many words admits that there was none, except on the supposition of the enemy's immobility. He persists, however, contrary to all prob- ability and to the express statements of Tacitus, in be- lieving that the Vite'lian generals had resolved to wait for their emperor before fighting, that they were quiescent in their entrenchments, unsuspecting, and lulled by a sense of false security. In opposition to all this, Tacitus says distinctly, that the Vitellians were resolved to fight without delay, that they had full information as to the designs of the enemy, and that they were keenly watching for the first false move. The immobility and false se- curity exist only in Mr. Henderson's imagination, and without them, the march to the confluence would be the height of madness. 200 STUDIES IN ROMAN HISTORY 2. But even if the confluence could have been reached, the move would still have been the height of madness, unless the second part of the programme was also carried out, the simultaneous movement of the Danube army. Mr. Henderson is himself very explicit on this point, and admits that the whole success of the strategic envelop- ment depended on the careful timing of this simultaneous movement. What evidence is there then that any con- siderable force was marching from Aquileia to Bedriacum ? That no part of the Moesian army could be counted upon for the movement is quite certain. There is some obscurity as to the whereabouts of the Moesian legions, but one thing is absolutely clear both from Tacitus and Suetonius, viz. that neither legions nor detachments had left Aquileia when news reached that city of the death cJf Otho. This is conclusive, because no troops could possibly take part in the supposed simultaneous movement, unless they left Aquileia at least four or five days before the battle. There remain, therefore, the armies of Pannonia and Dalmatia. The four legions and auxiliaries might number about 35,000 men. But of these, the whole of one legion, xiii, and 2,000 men from each of the other three, besides several thousand auxiliaries had already joined Otho, and would therefore take part in the flank march. There might still remain about 20,000 men. Where these troops were at this time, there is really no evidence. Some of them, and in particular the main body of legion xiv. may have left Aquileia, but two points seem clear— (i) Otho can hardly have counted much on this simultaneous movement, or he would not have thought it necessary to leave " magnam exercitus partem " behind at Bedriacum ; (2) no reinforcements from the Danube were near enough to Bedriacum after the battle to prevent the surrender of Otho's army. There is thus not a shred of evidence of the simultaneous movement of the Danube army, and so, upon examination, both portions of the scheme of envelopment utterly collapse. If Mr. Henderson's theory of a strategical envelopment APPENDIX TO CHAPTER II 201 will not bear investigation, his method of dealing with the narrative of Tacitus calls in my opinion for strong protest. It is hardly too much to say that in order to maintain the one word " Aduae," almost every other statement of the historian is ignored. I will confine my- self to his manipulation of the distances as given by Tacitus. Tacitus states — (i) that the first day's camp was *'ad quartum a Bedriaco " ; (2) that from this camp to the confluence was sixteen miles ; (3) that at the end of or in the course of the second day's march the Vitellians could attack after a march of four miles. Of course if we are to retain the Adua-Po confluence, these figures require some manipulation. Mommsen, recognizing this, suggested that Tacitus con- fused the length of the proposed second day's march — sixteen miles — with the distance from the final objective. This does not sound very probable. Mr. Henderson takes a more heroic course and boldly substitutes " ad quartum decimum " for " ad quartum." Even if this explained the other figures, which it does not, such a method would be wholly unjustifiable. There is no sus- picion that the text is corrupt, and Plutarch's figure proves that for some reason — Mr. Henderson could no doubt suggest several if he were put to it — the first day's march was a short one. However, by means of the impossible estimate that the flank march would involve only one extra mile, the alteration gives us the sixteen miles to the confluence. There remain the four miles of Paulinus to be accounted for. It would of course have been open to Mr. Hender- son to suggest that this precise distance was put into the mouth of Paulinus by Tacitus simply because the battle actually took place about that distance from Cremona. He does not, however, do this, but makes an ineffectual attempt to meet the difiiculty by giving a version of what Paulinus urged, inconsistent alike with both Tacitus and Plutarch. According to both of these authorities, Paulinus urged that in the course of the second day's march, not 202 STUDIES IN ROMAN HISTORY yet begun, the Othonians, weary after a long tramp, would find themselves within four miles of the enemy.' According to Mr. Henderson, he urged that they had already come too far, and that the enemy might attack them where they were by a march of four miles, the four miles being " a characteristic under-estimate " of the actual distance, eight miles, which separated the two armies. Comment on such a method of interpretation is surely superfluous. If these distances of Tacitus are worthless, why not wholly disregard them, as other statements even more import- ant are disregarded ? As it is, Mr. Henderson neither wholly accepts nor wholly rejects, but by an arbitrary method of his own, plays fast and loose with his authority. May I beg the reader who is interested in the question at issue to refer to my article in the Journal of Philology and to compare it with Mr. Henderson's statements to which I have there referred by chapter and verse. CHAPTER III Vitellius Our knowledge of this brief and inglorious campaign, lasting barely more than a month, and ending in the death of Otho, and the surrender of his troops at Bedria- cum and Brixellum, is mainly derived from Tacitus, though some details are added by Plutarch. The latter writer had himself gone over the ground, and some of the perplexities involved in the operations between Cremona and Bedrianum may perhaps be partly cleared up by his account. The general course of the cam- paign and the main causes of its issue are sufficiently obvious from Tacitus. Tacitus has, of course, been labelled by Mommsen as the most unmiUtary of his- torians, but in this case, there was little opportunity for him to go seriously wrong. Vague geographical knowledge, which too often vitiated his account of operations in Armenia or Germany or Britain, was, of course, out of the question when the movements of both armies were along the famihar roads of Trans- padane Italy, while Tacitus was an intimate friend of both Verginius Rufus, and Vestricius Spurinna, both competent generals, and both present, in an active or passive capacity, in the operations. No doubt there are several not unimportant points which Tacitus leaves uncertain. It is not clear at exactly what point MS f 204 STUDIES IN ROMAN HISTORY VITELLIUS 205 of the operations the attack of Otho's fleet upon the Maritime Alps and Narbonensis took place, nor where Valens was when he despatched the Batavian cohorts to the coast ; it is not stated when or by whom Cre- mona was captured by the Vitellians ; the movements of Annius Gallus are not directly described ; his passage of the Po is not mentioned, it is not stated where the vextlla of legions xiii and xiv joined him, and it is even a matter of doubt whether those of vii and xi had actually joined the Othonian army before the final battle. However, the answer to all these questions is at least suggested by Tacitus, and the only point on which it seems impossible to follow him, is his sugges- tion that Otho even after the battle of Bedriacum might have resumed the war with some prospect of success. For such a statement there seems no justification. His main army and with it all his ablest generals had surrendered at Bedriacum, and the ViteUians occupy- ing that position had blocked the road from Aquileia by way of Verona. The road through Pavia and Ateste was still for the moment open, but before Otho could collect his forces and get to it by way of Hostiha It would certainly be closed. Nor is it at all clear that any really considerable force was on its way from Aquileia. The four legions of Pannonia and Dalmatia had already sent forward vexilla of 2,000 men each, but these had almost certainly surrendered with the other Othonian troops at Bedriacum. 1 If the main body of those four legions was on its way, it could not amount to more than 10,000 or 12,000 men, while the force at Otho's disposal, even if we include the force of Spurinna as well as the gladiators, must have been well under 10,000. As for the three Moesian legions, in spite of vague reports that advance guards were reaching Aquileia, there is no evidence whatever that they had as yet made any serious move in that direction. What chance had Otho then of resuming the war, with a force of less than 10,000 men, and cut off from advancing rein- forcements of about the same number by a victorious enemy, whose army must have numbered 80,000 men ? In January there had been many chances in Otho's favour. He had possession of Rome and Italy ; he was supported ostensibly by the Senate, he was acknow- ledged by the armies of Africa, Egypt, Judea, and Syria ; the legions of Moesia, Pannonia and Dalmatia were prepared to give him active assistance. He had the praetorian cohorts, the new legion i Adjutrix, and a number of miscellaneous troops in Italy at his immedi- ate disposal. What was perhaps of hardly less import- ance, he had a number of able military advisers, includ- ing Suetonius Paulinus, whose reputation was not behind that of Corbulo or Vespasian. It was the delay, the want of firmness and judgment and the general incompetence of Otho himself, that had ruined all these chances. For nearly two months, as Tacitus says, he was occupied in Rome, almost as if no war was impending. If he gave orders for the advance into Italy of the lUyrican legions, he at least took no steps to see that their concentration was speedily effected. He took no steps to protect the passes of the Alps, or even to secure the fortresses of Transpadane Italy. Vercellae, Eporedia, Novaria and Mediolanum were t'1 K i 206 STUDIES IN ROMAN HISTORY VITELLTUS 207 dependent on the protection of the solitary ala Sihana ; Cremona was garrisoned by a single cohort of Panno- nians, while Placentia was apparently not defended at all. The expedition against Narbonensir^ conveyed by the fleet at Misenum, which if it had been sent out at once, with adequate instructions, and under a competent leader, might have led to important results, was de- layed till it became merely a desperate means of caus- ing a diversion, while, instead of being entrusted to a responsible commander, it was left to the reckless direc- tion of three incapable and unprincipled subordinates. When at last, but not before the first weeks of March, Otho's forces began to move northward, the scheme of defence adopted by Otho was the only possible one for securing the Hne of the Po from Placentia to Hostilia, and of forming a junction with the advance guard of the Illyrican army. Spurinna and Gallus did their part of the work well, but they were only just in time, and Otho's delay had involved the loss of Cremona, which if not the key of the position, at least seriously weakened the plan of maintaining a connected Po line. Indeed, if Caecina, instead of wasting several days in an ill-contrived assault upon Placentia, had at once occupied Bedriacum, he would apparently have antici- pated Gallus, and might have endangered his junction with the Illyrican reinforcements. No position under the circumstances could have been more favourable for the Othonians than Bedriacum. They were bound by the necessity of meeting their reinforcements to oper- ate to the east of Cremona, and therefore to abandon any attempt to prevent the junction between Caecina and Valens, which if Otho's arrangements had been made earlier, might possibly have been effected by a concentration of forces at Placentia. As it was, and after the arrival of Valens, Paulinus, Celsus and Gallus saw that there was no safe alternative but to wait for the rest of the Dalmatian and Pannonian army, if not for the Moesian legions. What foiled their plans was in the first place the mutinous, distrustful and insub- ordinate spirit of their troops. The disastrous effects of civil war upon the discipline and morale of the soldiers on both sides is very marked during the whole cam- paign, but there were special reasons for it in Otho's army.' There was not only the bitter rivaky between the legions of the Rhine and the Danube, which made them eager at all risks to fly at one another's throats— this was perhaps as strong in the Vitellian army— but the greater part of Otho's army consisted of the praetorian cohorts. These men, devoted as they were to Otho personally, were especially Uable to get out of hand. They were the most distinguished troops in the Roman army, but they were far from being the most efficient. Their pay was higher and their privileges greater, but they had had no experience of actual warfare, and they were spoilt by the indulgences of the city, and the donatives and largesses of Nero and Otho. More than this, they were intensely distrustful of their senatorial commanders, and only too ready to detect treachery to the Emperor of their choice in every order ; this was the experience of Spurinna at Placentia, of Martins Macer on the Po, and of Paulinus at Bedriacum. Un- fortunately for himself, this distrust was shared by Otho. Utterly incompetent as a general himself, he took the suicidal step of superseding his able and experienced generals by his elderly brother Titianus and the up- start praetorian praefect, Licinius Proculus. Then X ; M I : r I » l\ 208 STUDIES IN ROMAN HISTORY. followed the council of war. There can be no doubt that the advice of Paulinus and his colleagues to re- main on the defensive, until substantial reinforcements should arrive from Aquileia was dictated by prudence and a correct view of the situation. It might be a question which would arrive first, the reserve force with Vitellius, or the Moesian army, though all the chances seemed in favour of the latter. But at any rate, it would be a desperate risk, with an army of at most 30,000 or 35,000 men, to attempt either a direct attack, or any such turning movement as that suggested by Mommsen, in face of a veteran army of not less than 80,000 men. But the troops, and especially the praetorians, encouraged by their recent successes were mad for battle ; Titianus and Proculus took the same view, and Otho had apparently com- pletely lost his head, and decided on the fatal movement to advance upon Cremona. His order to Spurinna, to denude Placentia of its garrison, was perhaps of minor importance, as the fate of Placentia would clearly be decided by the result of the battle, but it showed the reckless determination of Otho to risk all upon a single stake. The battle lost, therefore, the war, as far as Otho was concerned, was over. Placentia was at the mercy of the Vitellians ; the path of the lUyrican army was blocked, and in spite of the mag- nanimous attitude ascribed to Otho by Tacitus, there was practically no course open to him but the one he took. Meanwhile at Rome the news of Otho's defeat and death created little excitement. The soldiers left behind in the city at once took the oath to the new emperor, and the fickle populace carried round decked VITELLIUS 209 with flowers and laurel the images of Galba, of whose death not three months ago they had been indifferent spectators. ^ Nor was the Senate less forward. The greater part of this august body had been compelled to accompany Otho on his march, and had been left behind at Mutina. guarded or watched by a small military force. After the battle at Cremona, their position was an anxious and even a critical one. To declare at once for Vitellius would have led to their massacre at the hands of the soldiers, who like his other troops were devoted to Otho, and Otho's fate was not yet beyond doubt. Pro- ceeding to Bononia on the way to the capital, they received news of the death of Otho, and were beginning to court the rising sun in the person of L. Vitellius, when they were struck with consternation by the report that the famous legion xiv had reached Brixellum, and had defeated the Vitellian army.^ However, letters from Valens soon removed all doubts, and the Senate hurrying back to Rome decreed to Vitellius all the honours and insignia of former principes. Letters of congratulation were at once addressed to him, and votes of thanks decreed to the German army.* Mean- while to the Italian colonies and municipia in the neighbourhood of the victorious army the triumph of Vitellius proved a serious and disastrous event. « The German legionaries treated them exactly as they had treated the towns in Gaul, and neither Valens nor * Hist. ii. 55. ' The story was iavented according to Tacitus in order to revive the validity of Otho's;diplomata or passports, which on the supposition of his death became useless {Hist, ii. 54). * His/, ii. 55. 6 Hist. ii. 56. 'It i 210 STUDIES IN ROMAN HISTORY Caecina had the power even if they had the will to check these outrages. ./-,.., Vitellius himself all this time was still m Gaul, busied in levying fresh troops, to supply the place of the veterans who had marched into Italy. From the Brit- ish army he received an important accession of 8,000 legionaries. « Leaving Hordeonius Flaccus with the remnants of the German legions and some auxiliaries to guard the Rhine, he himself began his march with the British vexilla and a number of hastily raised GaUic cohorts. Soon good news began to arrive. The two Mauretanian provinces, Caesariensis and Tingi- tana, had so far been on Otho's side. They were both of them at this time under the command of Lucceius Albinus, the procurator, with a force of nineteen co- horts and five alae, as well as a large number of Moorish tribesmen. With these, he was even meditating an attack on Spain in Otho's interest, and Cluvius Rufus, legate of Tarraconensis, deemed it advisable to send down the legion x Gemina, to oppose him.' This pre- caution, however, proved unnecessary, as Albinus, who showed signs of assuming the insignia of royalty and the name of Juba, was murdered, and Mauretania declared for VitelUus.s Proceeding down the Arar towards Lugdunum, he received news of the victory at Cremona and the death of Otho. At Lugdunum, ViteUius virtually entered upon the empire. Here he found Valens and Caecina who had travelled back to congratulate him, and also the leaders of the con- quered party, who were there to secure his pardon. « Hist. ii. 67. „ ■ n iKo'c ' This legion was apparently not in Spain during OaiDa s governorship, but it had now returned. * Hist. u. 68, 69. VITELLIUS 211 Paulinus and Proculus were contemptuously dismissed on their own confession of treachery. Titianus was excused on the score of his relationship to Otho, and his old age. Marius Celsus, who had negotiated the surrender of the troops at Bedriacum, was received with honour and not even deprived of the promised consul- ship.® In fact, towards the partisans of his rival, Vitel- lius, with a few exceptions, showed no signs of cruelty. He was, however, induced by false representations to order the execution of Dolabella, whom Otho had sent away to the colony of Aquinum.^^ The new emperor was already beginning to indulge in the gluttony and excesses which marked his short and ignoble reign, but he made a show of moderation by sending an edict to the Senate, in which he postponed the acceptance of the title of Augustus and refused that of Caesar. ^^ He at the same time made a faint attempt to mark a new regime by banishing astrologers from Italy and forbidding Roman knights to appear on the public arena. ^^ Before Vitellius left Lugdunum, two provincial mat- ters required his attention. The legate of Britain, Trebellius Maximus, had been driven from his province by the army, and a new governor had to be sent. Vitel- lius chose for the post Vettius Bolanus, who was at the time serving on his staff. ^^ From Spain, Cluvius Rufus, who had succeeded Galba as legate of Tarra- conensis, arrived at Lugdunum, in appearance to congratulate the new emperor, in reality to clear him- self of the suspicion of aiming at the chief power. He » Hist. ii. 70. 10 Hist. ii. 63. ^^ He had, however, already conferred the name of Germanicus I; I on his infant son {Hist. ii. 59). la 11. 73. 13 ii. 65. 212 STUDIES IN ROMAN HISTORY VITELLIUS 213 had, it seems, issued diplomata within his province, to which no emperor's name was attached. What- ever his own beUef may have been, Vitellius dismissed these charges, but ordered Cluvius, while still nominally retaining his province, to accompany him to Rome.^* On crossing the Cottian Alps to Italy, a more difficult problem presented itself in the treatment and disposal of the conquered army. One ill-advised step had al- ready been taken in the execution of some of the bravest centurions of the Othonian legions. This had fatally alienated the armies of Pannonia and Dalmatia. At present the surrendered troops were quartered in the various towns of North Italy, side by side with the victorious German legionaries. But there was the constant fear of an outbreak of ill-feeling and jealousy. The legion xiv was especially disaf!ected. It enjoyed a reputation greater perhaps than that of any other legion, and it loudly proclaimed that it was still unde- feated, as only a detachment of it had been present at the battle of Cremona. As a check upon its mutin- ous spirit its old auxiliaries, the Batavian cohorts, were quartered with it at Turin. Here a deadly con- flict was only prevented by the intervention of two praetorian cohorts, which sided with the legion. The legion was now ordered back to Britain, over the Graian Alps, special precautions being taken to prevent it from passing Vienna, which was suspected of being at best half-hearted in the Vitellian cause. ^^ The Batavian cohorts were for the time attached to Vitellius himself, but owing to a fresh quarrel with the legion v Alauda at Ticinum, were soon after sent back to the German 14 iL65. IS ii. 65. frontier, where they gave still more serious trouble by joining the movement of Civilis.^' Of the other Othonian troops, legion i Adjutrix was sent to the peaceful province of Spain, legions vii Galbiana and xi Claudia were sent back respectively to Pannonia and Dalmatia, while legion xiii Gemina was temporarily employed in building amphitheatres at Cremona and Bononia, where Caecina and Valens were preparing gladiatorial shows in honour of Vitellius. 1^ Soon after, this legion too was sent back to Pannonia. The restoration of these three disaffected legions to their old quarters on the north-east of Italy, especially as the three Moesian legions had also declared for Otho, was an ill-judged step, and was the immediate cause of disaster, but Vitellius was for the time elated by the news, true for the moment, that the Eastern legions had taken the oath of alle- giance to him.^^ It still remained to deal with the praetorian cohorts. These, under pretence of honourable dismissal, were ordered to give up their arms to their tribunes, and were disbanded. No greater punishment could have been inflicted upon them, and it is not surprising that most of them took service with the invading Flavian army. Within his own army, Vitellius found formidable diffi- culties, nor were his measures more judicious than to- wards the vanquished. His reckless despatch of the Batavian auxilia to Germany we have already noticed. But there was a mutinous spirit in the troops all through this civil war, and though the German legions were de- voted to the emperor of their choice, any trifling cause might produce an outbreak Uke that which had nearly »• u. 68. IT u. 74. " Hist. ii. 73. ill 214 STUDIES IN ROMAN HISTORY VITELLIUS 215 cost Valens his life earlier in the year. At Ticinum, a conflict arose between the legionaries and the auxiliaries which caused the destruction of two Gallic cohorts. On the same occasion the soldiers were with difficulty prevented from murdering Verginius Rufus, owing to some utterly unfounded suspicion. i» Partly perhaps to meet these difficulties, partly on the assumption that the civil war was over, but mainly under stress of financial considerations, Vitellius first sent back to their homes the numerous but inefficient GalUc auxiUaries, and then proceeded to dismiss large numbers of his legionaries, while all recruiting was for the time stopped. Under the circumstances, such short-sighted economy was bound to be at once dangerous to the empire and unpopular with the soldiers. *° Towards the end of May Vitellius reached Cremona, and after viewing the shows provided by Caecina was conducted over the scene of the battle by his generals. The traces of carnage were still unremoved, but the emperor showed no sign of emotion, and proceeded to Bononia, where the shows of Valens were witnessed. Unfortunately, the nearer he came to Rome, the greater licence prevailed in his army, and the less consideration or respect he showed for his late opponents. 21 The army which accompanied him numbered still 60,000 men, but they were no longer like the hardy veterans with whom Valens and Caecina had set out. The luxuries of Italy had enervated them, and disci- pline was corrupted by the vile crowd of camp followers who were allowed to accompany them. Vitellius him- self was surrounded by a motley retinue. Senators and knights, each bent on securing pardon or lucrative posts, i» Hist. ii. 68. «• ii. 69. " "• 87. actors, buffoons and charioteers, all flocked to the moving court. During his short reign of eight months, ViteUius is said to have spent on his table alone the sum of 20,000,000 sesterces. 22 But perhaps the most shameful feature of the whole reign was the entry into Rome. Seven miles from the city gates, a quarrel arose between the soldiers and the city mob which had poured out to witness the emperor's advance, and the soldiers had no hesitation in drawing their swords upon the defenceless multitude. 23 ViteUius himself was with difficulty induced to enter clad in the toga praetexta rather than the miUtary paludimentum, but his march none the less had all the appearance of an entry into a conquered and captured city. It was a gorgeous mili- tary spectacle, as described by Tacitus.2* pour legions with their eagles accompanied him, 25 and the greater part of four others.26 In addition to these, there were strong detachments from the British legions, and no fewer than twelve alae of auxiliary cavalry, and thirty- four auxiUary cohorts. It was an imposing army, worthy of a better general. The strange, rough-looking soldiers, who most of them now saw Rome for the first time in their fives, wandered through the streets, some to the forum, where Galba had fallen, with a morbid desire to see the exact spot, and all needed but sUght provocation to come to blows with the citizens. Never since the days of Sulla had the capital been so openly and helplessly at the mercy of a provincial army. ViteUius was now formaUy estabUshed as princeps. His salutation as Imperator by the army had already " Dio Cass. Ixv. 3. " ii- 88. " ii- 89. *5 I Italica, V Alauda, xxi Rapax and xxii Primigenia. *• I, IV Macedonica, xv Primigenia and xvi. ii 2l6 STUDIES IN ROMAN HISTORY VITELLIUS 217 been confirmed by the Senate : not only was he himself Augustus, but he honoured his mother with the title of Augusta. 2' The title of Caesar he still refused, though towards the end of his life, with a superstitious trust in the associations of the name, he accepted it.^^ The tribunicia potestas was duly conferred, and the office of Pontifex Maximus assumed. In the last capacity he chose the ill-omened day, July 18, the anniversary of the battle of the AUia, on which to issue an edict concerning the national religion. ^^ It is clear that Vitellius made some flimsy attempts to observe constitutional forms. Suetonius tells us, and an inscription confirms the state- ment, that he assumed the position of consul perpetuus. This was almost certainly intended as a concession to the Senate, and a return to the system of Augustus, prior to the year 23 b.c.^o In other ways, too, Vitellius made some attempts at constitutional government. While confirming many of the consulships promised in advance by his immediate predecessors, he contrived by shortening the period of office in some cases to find room for Valens and Caecina. He, however, made a point of canvassing the tribes with his candidates, as Augustus had done at first. ^^ He was constantly present in the Senate, however trivial the business, and when on one occasion he was opposed by Helvidius Priscus, prae- tor elect, he checked the indignation of his friends by remarking that it was no unusual thing for two senators a? 11. 90. ** iii. 55. One inscription gives him the title : Imp. Vitellius Caesar. Wilm. 916. " Suet. Vit. II. " Suet. Vii. 1 1 and C.I.L. vi. 929 ; A. ViteUius L. fil. imperator COS. perp. The perpetual consulship was offered both to Caesar and to Augustus. »» ii. 91. to disagree ; he had himself under Nero had controversies with Thrasea. What might, if permanent, have proved an important administrative change and an anticipation of Hadrian, was his appointment of Roman knights to the posts in the imperial household hitherto held by freedmen.32 A somewhat unreal concession to the Neronian exiles restored by Galba was the restoration of their rights over their freedmen.^^ It was perhaps a greater satisfaction to the Senate, when, as a means of raising money, a tax was levied on the freedmen of former emperors in proportion to the number of their slaves.®* But economy, administrative reform and constitu- tional government were alike illusory. The freed- man, Asiaticus, who had already received equestrian rank, was allowed to follow in the steps of Pallas and Patrobius and Icelus : ^^ Vitellius himself is said to have squandered 900,000,000 of sesterces during his short reign, and all real power was in the hands of Valens and Caecina, who vied with one another in extravagance, ostentation and flattery.'* Meanwhile the most complete want of discipline prevailed among the soldiers. To replace the disbanded praetorians, sixteen praetorian, and four urban cohorts were now enrolled. They were taken in both cases from the soldiers of the German legions, but no care was taken to make the more privileged service a reward for special merit. On the contrary, the soldiers were allowed to choose their own service and their own duties. As praefects of these new praetorians, were chosen Publilius Sabinus, praefect of an ala, and Julius »« Hist. i. 58. " ii. 95. as u. 92. »• ii. 92, S4 ii. 94- 2l8 STUDIES IN ROMAN HISTORY Priscus, a centurion. ^^ As for the legionaries, they were quartered about the city, in temples and porticoes, many of them in the unhealthy district of the Vatican, where disease made lamentable gaps in their ranks. All alike enervated and demoralized, were fast becoming formidable only to their friends.'^ So far the disturbances following on the death of Nero had been confined to the provinces of the West. Of these, none had been unaffected. Africa under Clodius Macer, Mauretania under Lucceius Albinus, had both been disturbed, but were both now held for Vitellius. The army in Spain had led the way in ad- vancing its own general to the imperial position. Gaul had been intimately connected with the course of events both through the rising of Vindex and the support it gave to the cause of Vitellius. The German legions had followed the example set in Spain, and their general was now abusing his short spell of power in Rome. Even remote Britain had sent detachments to support Vitellius and these too were now in the capital. The Danube army had not indeed proclaimed any emperor of its own, the legates of Dalmatia, Pan- nonia and Moesia being merely dignified nonentities, but they had resented the action of the German legions, and had supported Otho, though only a small part of the army had taken an actual share in the first Italian campaign. The smaller provinces of Noricum, Raetia and the Maritime Alps had all espoused one side or the other, and even Sardinia and Corsica had not been wholly unaffected by the general convulsion. The Eastern provinces alone had up to this point taken no part in the civil wars. But nearly one-third tT 11. 92, 94. >8 11. 93. VITELLIUS 219 of the Roman army was stationed in the Last : m wealth and resources the Eastern provinces were superior even to those of the West : and above all. there were two generals whose claims might well com- pare with those of Galba or Otho or Vitellms. The most important of the Eastern provinces was Svria This from the time of Augustus was garrisoned by four legions, Iii Gallica, vi Ferrata, x Fretensis, and XII Fulminata. As a rule, these Oriental legions tended to compare unfavourably with those of the West in discipline, efficiency and morale, but under Corbulo's command, in the course of the Armenian and Parthian wars, not only had this defect been remedied, but the Eastern army had been strengthened by no fewer than three legions from the West-iv Scythica, v Macedonica and xv ApoUinaris.^^ The outbreak of the Jewish War in 67 a.d. had so far prevented any of these from being sent back, though quite recently iii Gallica had been detached to reinforce the diminished Moesian army. Since 67 ad. byna had been under the command of C. Licinius Mucianus, a man of marked personality, much ability but some- what chequered career. He was as much a litterateur as a general, and though he was one of the few men of the period who in the end attained a third consulship, he had under Claudius been for some time in quasi- banishment as legate of Lycia and Pamphylia.^ How- ever, his abiUty. tact and prestige would no doubt 3. ^ee my Studies in Roman HU^^^^^^ ^^^ ^ ^^^^ 40 Hist. i. 10; Plin. Nat, Hist. xu. i. ^> ^. , lector and compiler of historical memoirs and letters (Tac. DiaL ToJ:^z7)Z. the author of a book on the curiosities of natural history frequently cited by the elder Phny. 220 STUDIES IN ROMAN HISTORV have secured him the support of the three Syrian legions, IV Scythica, vi Ferrata and xii Fuhn., if he had chosen to bid for empire. He seems, however, to have seen his way to much of the power without the responsibility of the supreme position. When the Jewish War broke out in 67 A.D. Nero had made Judaea into an independent military province, and had sent out as legate and com- mander in the war T. Flavius Vespasianus, assigning to him the remaining three legions, v Mac, x Fret, and XV Apoll. . Vespasian was of a very different type. Born in 9 a.d. of a comparatively humble family, near Reate in the Sabine country, he was with difficulty persuaded by his mother to take the latus clavus and enter upon the senatorial career. His earliest mihtary service was as tribunus mihtum in Thrace, but after his praetorship, he was, owing to the favour of Narcissus, sent to Germany as legate of a legion, whence he was transferred to Britain as legate of 11 Augusta. « Here he gained many victories in the south of the island, and was in the words of Tacitus, " monstratus fatis." For these successes he received the inumphaha ornamenta. He was consul for the last two months of 51. After remaining in retirement for some years owing to fear of Agrippina, he was after her death made proconsul of Africa, which Suetonius, differing from Tacitus, « says that he governed with honesty and dignity. On his return to Rome, he was reduced to considerable poverty and obscurity, and in 65 fell into sore disgrace for going to sleep during one of Nero's performances. ^^ However, when in 67 Cestius Gallus, legate of Syria, was defeated by the Jewish insurgents, Nero could find no more capable general to conduct the war. It was " Hist, ii. 49 ; Agric, 3. " ii. 17. 4S ^„„, ^^^ ^^^ VITELLIUS 221 not yet over, for though the Jews had been driven from town to town and from fortress to fortress, they were now at bay behind the fortifications of Jerusalem. Besides the six legions in Syria and Judaea, there were two more in Egypt, under the command of the praefect, Tiberius Alexander, a Jew by birth. In all three provinces, there were considerable forces of auxili- ary troops, both cohorts and alae. In Cappadocia also, a strong auxihary force was stationed. If to these regu- lar troops are added the native armies of such client kings as those of Armenia Minor, Sophene and Commagene, it will be seen that an army at least as large as that of Pompey in 49 B.C. could without much difficulty be got together. At first, however, both generals and armies had ac- quiesced in the election of Galba. Titus indeed, the elder son of Vespasian, who had already composed some slight friction between his father and Mucianus, had been sent to do homage on behalf of the Eastern army to the new emperor. Whether there was any truth in the report that Titus had been summoned by Galba with a view to being adopted is uncertain, but at any rate, on reaching Corinth, he received news of the death of Galba and the revolt of Vitellius. Deeming it inadvis- able to commit his father's cause by doing homage to Otho instead of Galba he returned by way of Rhodes and Cyprus to Syria.** Here he found the situation changing. The armies had indeed taken the oath of allegiance to Otho, but the action of the German army could not but rouse the jealousy of the powerful Eastern legions, whose services to the empire in recent years had certainly been second to those of no other troop. 44 u. 1 2. 222 STUDIES IN ROMAN HISTORY VITELLIUS 223 Messages began to pass between the armies. Mucianus, while withdrawing all claims for himself, was at first in favour of Titus, but the eyes of the soldiers were bent on Vespasian, and Mucianus cast aside his doubts and co-operated in the same cause. Titus was the life and soul of the movement, arranging, reconciling and encouraging. Vespasian himself was the most back- ward. He was well aware of the strength of the German army, he realized the difficulties of concentrating and bringing to bear against Italy his powerful but scattered forces. He had too the Jewish War still upon his hands. Much was hoped from the Danube army, where the legion III Gall., virtually a Syrian legion, was expected to prove a rallying point for his cause : but the legions of Dal- matia and Pannonia were more hostile to Vitellius than devoted to himself, and the Danube army was at present without a leader. Thus no definite step had been taken when news arrived of the victory of Bedriacum and the death of Otho. Sullenly and silently the armies even took the oath to Vitellius, but after a personal meeting between Mucianus and Vespasian, the mind of the latter was made up. Mucianus reviewed the forces at his disposal, pointed to the injudicious treatment of his army in Italy by Vitellius, and warned Vespasian by the fate of Corbulo that his only safety was to act with decision. *^ When the two parted, Mucianus for Antioch, Vespasian for Caesarea, the die was cast. The first open step was taken on July 2, when Tiberius Alexander caused Vespasian to be proclaimed Imperator by the two Egyp- tian legions at Alexandria. This day was reckoned as the dies imperii, or official commencement of the new reign.** Two days later, the example was followed *• ii. 76-78. *• ii. 79. by the legions in Judaea, and before July 15, Mucianus bad administered the same oath to the Syrian army. If any additional incentive to the soldiers was want- ing, Mucianus suppHed it by representing Vitellius as intending to transfer his own German legions to the luxurious Eastern provinces, and to send the Eastern legions to the toils and hardships of the German frontier. *^ Soon after this, an important council of war was held at Berytus. All the Eastern provinces, from Achaea to Armenia, from the Euxine to Egypt, had now declared for Vespasian. The meeting at Berytus was a splendid and a repre- sentative one. Vespasian and Mucianus were both pres- ent. Officers, high and low, legates, tribunes and cen- turions, troops of cavalry and infantry from the Syrian and Judaean armies, all flock to support the new em- peror. Client kings vied with one another. Sohaemus of Sophene was present with no contemptible force : Antiochus of Commagene, whose cavalry had rendered good service to Corbulo in Armenia, was also there. ^ Sohaemus had been made king of Sophene by Nero. It was a district between Cappadocia and Armenia, on the left bank of the Euphrates, important, as command- ing one of the passages of the river. It says much for the friendly relations with Armenia that it was thought safe to denude this little kingdom of its troops.*^ The first business of the council of war was to pro- *' ii. 80. ** Commagene had] been joined to Syria in 17 a. d., but Antiochus, son of the late king, had been restored by Caligula, then deposed and restored again by Claudius. His kingdom was finally incorporated Jwith Syria by Vespasian. See Ann. ii. 56 ; xii. 55 ; Dio Cass. hx. 8 ; Ix. 8. *» Ann, xiii. 7. 1 i: it . i « 1 r '' h I I i •B^l: 224 STUDIES IN ROMAN HISTORY VITELLIUS 225 vide the sinews of war. Levies were ordered, veterans recalled to the ranks : despatches sent to the legates in lUyricum, announcing the election of Vespasian, invit- ing co-operation, and ordering the disbanded praetorians to be won over by promises of renewed service. Orders were given for gold and silver money to be coined at Antioch, and arms to be manufactured in all the cities. Contributions of money, voluntary and involuntary, were collected, while Vespasian sought to gain over individuals by the grant of miUtary and civil posts, and even by admission to the Senate. ^° With regard to the miUtary situation, it was decided not to denude the Eastern provinces of their troops. This might to a certain extent be necessary later on, but at present, a force of not much more than 20,000 men, consisting of vi Ferrata, 13,000 veterans and some auxiUaries, were to cross into Europe, under the command of Mucianus. The Danube army was to muster at Aquileia, but not to advance before the arrival of Mucianus. To guard against any complications on the Eastern frontier, the other two Syrian legions re- mained behind. Such complications, however, were not at the moment likely. The Parthian king was well content with seeing his brother, Tiridates, recognized by Rome as the King of Armenia, and there was no special cause for quarrel between the two great empires. Indeed, in response to envoys sent by Vespasian, the great king offered a large force of Parthian cavalry, an offer, however, which was declined. The Jewish War was not completed and could not be discontinued. Accordingly Titus was to retain the army of Judaea, and press forward with the siege of »• ii. 81, 82. Jerusalem. Vespasian himself, it was decided, should not accompany the army advancing into Europe, but rather proceed to Egypt, in order to secure that important granary of the empire, on which Rome was largely de- pendent for her corn supply. There was some talk about the various fleets at the disposal of Vespasian, but little or no attempt seems to have been made to collect or use them. 51 When all these matters had been arranged, the assemblage at the colony of Berytus broke up. Mucianus himself lost no time setting out. He had a long march across Asia Minor before him to Byzantium. Only a few light-armed troops accompanied him, but his main army, the legio vi Ferrata and 13,000 veterans brought up the rear. The plan of operations, arranged with Vespasian, had been for the Eastern army to march through the Danube provinces, and form a junction with the lUyrican force at Aquileia. But Mucianus was more a colleague than a subordinate, and it was open to him to conduct the campaign as he chose. At Byzan- tium he had ordered the Pontic fleet to assemble, and he seems to have had some idea of adopting a plan of campaign in which the fleet would have played an im- portant part. Instead of making Aquileia his object- ive, he had some thought of marching his land force along the Via Egnatia to Dyrrhacium, and sending the fleet round to the Hadriatic. In this way, he would have been able to threaten an invasion of Italy at any point between Tarentum and Brundisium, and Vitellius, not knowing where to expect an attack, might have had dangerously to divide his forces. ^ 2 Why Mucianus did not take this course Tacitus does not inform us. As things turned out, owing to the " u. 82. 83. " ii. 85. h 226 STUDIES IN ROMAN HISTORY defection from Vitellius of the Ravenna fleet, Mucianus might probably have reached Rome in this way sooner than he actually did ; but this he could not have fore- seen, and even if he had succeeded in crossing to Italy, his force of 20,000 men was perhaps too small to venture over the Apennines. At any rate, he decided to keep to the original plan, and to march through Moesia and Pannonia to Aquileia. But it was not the Eastern army, or any part of it, which was destined to take part in the second Italian campaign. We have seen that the four legions of Pan- nonia and Dalmatia had sided with Otho, and had all sent forward detachments to join him. Possibly the whole of legion xiii had come up, as its legate was present at Bedriacum, and the legion was employed in building an amphitheatre at Cremona. The other legions had delayed to join, and whether they had ad- vanced to or beyond Aquileia, is uncertain. At any rate, they were all treated by the Vitellian leaders as conquered legions, and not very judiciously sent back to their former provinces, vii Galbiana, and ultimately xiii Gem. to Pannonia, and xi Claudia to Dalmatia. The other Dalmatian legion, xiv, was, as we have seen, re- moved to Britain. On the other hand, the statements of Tacitus involve the movements of the three Moesian legions in some obscurity. They were iii Gall., recently moved from Syria, vii Claudia and viii Augus- ta. These too had declared for Otho, but how far they had given effect to their adhesion, is not certain. A few soldiers of the Moesian army were with Otho at Brixellum and they reported that the legions themselves were at Aquileia. ^3 Tacitus seems to accept this, report. VITELLIUS 227 68 XI. 77. for in II. 85 he speaks of the Moesian legions as having entered Aquileia, as seizing the military chest, and even after news of the battle pulling down the standards of Vitellius. Probably, however, the Moesian legions, in the absence of an energetic legate, had done no more than send forward a small advance guard to Aquileia. 6* It was the Moesian army which took the lead in assert- ing the cause of Vespasian. First, legion in, really a Syrian legion, proclaimed him, and soon after, the other two, the legate of the province, Aponius Satuminus, apparently only joining the movement on the revolt of the last two legions. They at once announced their action to the Pannonian legions, which, however, needed no stimulus to take the same course. ^^ -pj^^y y^^^^ j^^ fact, proclaimed Vespasian on the first receipt of letters from the East. This was not due to the initiative of the consular legate, T. Tampius Flavianus, a relative of Vitellius, who indeed deserted his post for the time and retired to Italy, but to the zeal and energy of Antonius Primus, legate of legion vii. Antonius had ** The legio vii Claudia was clearly in Moesia when Aponius Satuminus made his attempt on the life of its legate, who escaped " per avia Moesiae ultra montem haemum " (ii. 85). Legio iii must have been in Moesia when it first proclaimed Vespasian and Aponius reported its revolt to VitelUus, while all three legions were certainly in Moesia when the Pannonian legions sent letters urging their advance (ii. 86). If the Moesian legions had really advanced to Aquileia, and had then been sent back to their province by the Vitellian leaders, they would certainly have been mentioned along with the other legions sent back in ii. 66, 67. We must, therefore, assume either that the Moesian troops spoken of by Tacitus as being at Aquileia, were only more or less small detachments of the Moesian legions, and this for every reason is the more probable view (Suet. Vesp. 6), or that the legions at Aquileia were those of Pannonia and not those of Moesia at all. 68 ii. 85. 4 t 228 Studies in roman history VITELLIUS 229 been condemned for forgery under Nero, but was re- stored to his senatorial rank by Galba, and appointed to the command of his new Spanish legion, vii Galbi- ana. He had offered his services as general to Otho in the war against Vitellius, and it was probably owing to the rejection of his offer that his legion took no part in the campaign. Now, at any rate, it was Antonius who took the lead and soon proved that he possessed brilHant, if somewhat reckless gifts of generalship. ^^ He was backed by Cornelius Fuscus, the procurator of Pannonia, a man who had deliberately given up the senatorial career for the equestrian, as a more rapid road to fortune. He had won over his native colony for Galba, and had been rewarded by the emperor with his present post." The legion in Dabnatia, xi Claudia, under the legate of the province, Pompeius Silvanus, hesitated for the present. ,and did not join the cause till after the passage of the Po,^^ but letters were sent by Antonius to legion i Adj. in Spain and to legio xiv in Britain, while messages were sent broadcast through every part of Gaul. The Pannonian army was mustered at Poetovio on the Drave. Among the officers opinions were divided ; some were in favour of waiting for the'arrival of Mucianus, of utilizing the fleets and other resources of Vespasian in the Far East. They pointed to the numerical strength of the German army concentrated in Italy, increased as it was by fresh forces from Britain. Delay, while »« ii. 85. . »' His subsequent fortune fully justified his promotion by Galba, and also his own choice of a career. He died praetorian praefect under Domitian. Suet. Dom. 6 ; Juv. Sat. iv. 112. " ui. 55- it would add to their own numbers, would carry still further the demoralization among the German troops, which had already begun. The best plan was to guard against attack by blocking the passes of the Pannonian Alps, and so in safety to wait for reinforcements. On the other hand, Antonius Primus in a vehement and effective speech urged immediate action. The Pannonian legions were eager for battle, smarting under their late defeat ; the Moesian army was equally im- patient, and as yet unbeaten. In morale, if not in num- bers they were superior to the Vitellian army. Even in the former campaign, the Moesian and Pannonian cavalry had ridden down the enemy ; now they had sixteen alae in addition to other tried auxilia. As for the fleet, Vespasian's ships were remote ; Vitellius had the two fleets at Ravenna and Miseniun, he commanded the Hadriatic, and if he chose to take the offensive by sea, what protection would the Pan- nonian Alps afford them ? Delay would not necessarily demoralize the German legions ; it might on the con- trary give them time to recover themselves, and it would certainly give time for the resources of Italy to be collected, and for reinforcements to come in from Britain, Germany, Gaul and Spain. Finally, turning to the consular legate, he bade him keep back his legions ; he was prepared to advance with the auxiliary infantry and cavalry alone. ^® Antonius was strongly supported in this policy by Cornelius Fuscus, and it carried the day. Urgent de- spatches were sent to the legate of Moesia to bring up his army. But there was even greater danger along the Danube frontier than in the East involved in the t I 6» lu. 2-4. 230 STUDIES IN ROMAN HISTORY VITELLIUS 231 removal of the defending armies. However, certain precautions were taken. In eastern Moesia some auxi- liary forces were left behind. *° Above western Moesia, the Sarmatae had recently been roughly handled by legion III : their princes were now compelled to follow with the Moesian army : an offer made by them to furnish some of their native cavalry being prudently declined. North of Pannonia there was less danger, as the Suevi had for years formed a vassal kingdom, the regnum Vannianum. But the two vassal kings, Sido and Italicus, with a picked band of their countrymen, were ordered to join the army. Further still to the west, it was deemed advisable to strengthen the garrison of Noricum, in view of the fact that a dangerous diversion might be made on the right flank of the advancing army by the Vitellian procurator of Raetia, especially if he was joined by reinforcements from Germany. Accordingly, Sextilius Felix was sent with eight cohorts, the ala Auriana, and the Norican Militia, to occupy the banks of the Inn which divided the two provinces.*^ These preliminary arrangements having been made, Antonius Primus with the veteran auxiliary cohorts and a strong body of cavalry, proceeded to cross the Pan- nonian Alps and occupied Aquileia. He was accom- panied by Arrius Varus, a soldier of skill and experi- ence, who had served under Corbulo in Armenia. ®2 There was no opposition, and Antonius began a rapid march along the road leading to Hostilia. Opitergium and Altinum opened their gates. At the latter, a garrison was left, since a force landed here by the Ra- •• We shall see below that their camps were seized by the invading Dacians. •* ii. 5. •- Ann. xiii. 9. venna fleet, the defection of which was not yet known, might cut off the advance guard from the main army. Then Patavium and Ateste were successively occupied. '^ Meanwhile at Rome, Vitellius had Hved at first in a fool's paradise. The first disturbing news came from Moesia, in the form of a letter from Aponius Saturninus, the legate, reporting the revolt of legio in.®* The im- portance of this news was, if understood, as far as possible concealed from the soldiers and the public. Vitellius declared that exaggerated reports were cir- culated by the disbanded praetorians, and that there was no danger of civil war. As a matter of fact, correct tidings of the Flavian preparations were kept back, partly by the blocking of the Pannonian Alps, partly by the Etesian winds, which retarded news from the East. Still, enough was known to arouse some misgivings. Orders for fresh auxiUaries were sent to Germany, Britain and Spain. Nothing, however, came of this. Hor- deonius Flaccus in Germany was already foreseeing serious trouble with the Batavians. Vettius Bolanus in Britain found plenty to occupy him in that unsettled province ; while in Spain, in the absence of a consular legate, the commanders of the three legions, i Adj., vi Victr., and x Gem., were waiting events. In Africa alone, among the provinces, was there any zeal shown in the cause of Vitellius. Here, the legion and auxiUaries, enrolled by Clodius Macer, and then disbanded by Galba, were again collected, and active recruiting went on among the provincials. But even this led to no result, as Valerius Festus, the legate of the African army, prob- ably aware of Vespasian's plan for invading the pro- vince from Egypt, held his hand.*^ " iii. 6. •* ii. 96. ^^ ii. 98. !l' llf' 232 STUDIES IN ROMAN HISTORY VITELLIUS 233 At last the gravity of the situation became clear, and Vitellius gave orders for Valens and Caecina to take the field. The former was, however, disabled by ill- ness and not able to start with his colleague. The army that left Rome under the command of Caecina was in numbers a powerful one. Large detachments or vexilla of four legions, i, iv, XV and xvi led the van. *• Then followed four legions with their eagles, v, xxi, XXII and i Ital. . Besides these, there were picked auxiUaries and detachments from three of the British legions. But the five months spent in Italy and Rome had woefully changed the spirit and appearance of the fine troops which had passed the Alps in the spring. Not only were their ranks thinned by injudicious dis- missals, but the men were enervated by riotous Uving under the hot Italian sun, while the smallest cause might lead to disobedience and mutiny. Still it was a formidable army and was at least obstinately loyal and devoted to Vitellius. So much could not be said for its general. Caecina, always vain and arrogant, had played a second part to Valens, both in the campaign against Otho and in the ignoble struggle for rewards and favour in Rome. He had probably already been tampered with by Flavins Sabinus, and he left Rome, if not resolved on treason, at least prepared to consider its advantages. Once more, the defending army was too late to pre- vent the enemy's entrance into Italy by blocking the Alpine passes, and accordingly once more it was the line of the Po that had to be defended. But this time Hos- •• A vexillum might vary in strength from a few hundred to the greater part of a legion. The term, however, always implies that the eagle of the legion is absent. tilia rather than Placentia was the point to be secured. It was necessary, however, to occupy Cremona, destined a second time to be the central point of the campaign, and if possible to secure the co-operation of the Ravenna fleet. Accordingly, Caecina at once sent forward some cavalry and auxiharies, to prevent Cremona from falling into the enemy's hands, as it had done in the former campaign, and later on detached from the main army to the same important position, the two legions i Italica and XXI Rapax.*^ If Valens had had his way, the army would have been still further divided, as he sent orders to the legions which had formed his army (i.e., i, v, xv, xvi, and I Ital.) to wait for him upon the road. Caecina, however, withholding this order, marched his army along the Via Flaminia in the direction of Hostilia. He himself, however, as the army neared its goal, turned aside to the fleet at Ravenna. There was good military reason for doing this. The Ravenna fleet might have been far more unportant to Vitellius in this campaign than the Misenum fleet, even if better directed, could possibly have been to Otho in the first. It was much nearer to the seat of the war : the enemy's line of march ran much closer to the coast, touching it, in fact, at Altinum : while the possession of the Hadriatic might, if things went well on land, be of vital import- ance in the later stages of the war. It is to be feared, however, that Caecina went to Ravenna not to utilize the fleet, but to betray it to the enemy. If Caecina had his grievances against Vitellius and Valens, so also had Lucilius Bassus, praefect of the Ravenna fleet. Raised by Vitellius from being praefect of a cavalry ala, he had had his eyes upon the •' ii. 100. 234 STUDIES IN ROMAN HISTORY higher position of praetorian praefect, and in his dis- appointment was ready to betray his trust. Plans were no doubt arranged between Caecina and Bassus, but nothing was done immediately, and Caecina fol- lowed his army to HostiHa. Whether Caecina had any thought of a forward movement against the small ad- vance guard of Antonius— this was all of the Danube army yet in Italy— is uncertain. He had at any rate pushed forward a detachment of three cohorts and the ala Sebosiana, a Galhc squadron, to occupy Forum Alieni, a town on the right bank of the river Adige, over which they had thrown a bridge. Antonius, as we have seen, had reached Patavium and Ateste, and the first actual collision between the two armies took place when Antonius surprised and captured this outpost. «» The incident was of no great importance in itself, as part of the garrison in escaping over the river had the presence of mind to destroy the bridge behind them, so that the Flavians were unable to press on further.' This, however, they would hardly have ventured in any case, since the main army of Caecina was not more than fifteen miles away, and in point of fact they returned to Patavium. The success, however, small as it was, bore important fruits, since it quickened the movements of the Danube legions. These had apparently still been retained at Poetovio by the cautious Flavianus, but on news of the victory at Forum Alieni, the two legions, vii Galbiana, and xiii Gem., crossed the mountains and joined Antonius at Patavium. The interval was used by Antonius in associating the Flavian cause with the memory of Galba, and with that end in re-erecting the statues of that •" iii. 6. VITELLIUS 235 emperor in the towns of the neighbourhood.** Even after the arrival of the two Pannonian legions, the Flavian army was not a large one. Caecina had by this time moved his army a few miles to the north of Hostilia, and was posted in a strongly entrenched camp, between that town and the river Tartarus. Be- fore him was that river, with a bridge across it, in his rear was the Po, while his two flanks were protected by the marshes formed by the Tartarus. ''^ The position was a strong one both for defence and attack, and if Caecina had only been loyal, there is no doubt that he might, as Tacitus points out, have easily crushed Antonius, who was still without the Moesian legions, especially as a force, at least equal to that of the Flavians, was now occupying Cremona. Caecina, however, made no sign, and Antonius determined to make the strong and wealthy colony of Verona his base of operations. There were several reasons for this choice. Verona lay on one of the direct roads from Aquileia, so that the Moesian legions could easily reach it : its situation in a wide plain would give him the opportunity of using to the best advantage his numerous cavalry, while lying as it did between Raetia and the Julian Alps, its occupation would enable him to bar the passage of any German reinforcements sent by way of Raetia. This danger was, however, really very small, as Antonius was in communication with Civilis, and was somewhat recklessly using him as a means of detaining all possible reinforcements in Germany.'^ In all these steps Antonius might be justified by suc- cess, but he was acting in direct disobedience to the orders cf both Mucianus and Vespasian. The latter " iii. 7. »» iii. 9. '^ iii. 8 ; iv. 14. a M i! 236 STUDIES IN ROMAN HISTORY VITELLIUS was now in Egypt, and was not without hopes that Vitellius, finding the com supply from Egypt and perhaps from Africa cut of!, might be inclined to sur- render without a struggle. At any rate, the Danube army was to remain concentrated at Aquileia, until the arrival of Mucianus with his Syrian army.'^ No doubt fortune and the treacherous inactivity of Caecina had so far saved Antonius from what might have been the disastrous effects of his rash disobedience. But now the chief danger was over. Verona was occupied, and the Moesian legions began to arrive, first vii Claudia with the legate of the province, Aponius Satuminus, and soon after, legions iii and viii. To make a demonstration of strength, it was resolved to surround Verona with Unes of entrenchments. This did not improve the temper of the soldiers, who had come to fight not to dig. There were also too many generals in his camp, and the two legates of Pannonia and Moesia were not only in the way, but suspected of being at best half-hearted in the Flavian cause. Two savage mutinies broke out among the Pannonian and Moesian legions, against their respective legates. It was all that Antonius with all his tact could do to save their Uves and secure their escape. '^ Meanwhile Caecina, before declaring himself, was wait- ing for the concerted revolt of the Ravenna fleet, ex- changing in the interval a curious correspondence with the Flavian leaders. These quasi-diplomatic inter- ludes seem to have been a feature, and not a dignified one, of the civil war. At last the news came. Bassus had httle difficulty in inducing the fleet to change sides. Many of the sailors had recollections of service " iii. 8 ; iv. 8. 7S m. 10-12. 237 under Otho, while many were themselves natives of Dahnatia and Pannonia, and were therefore in sym- pathy with the legions of those provinces. They wisely, however, deposed Bassus from his command, and appointed ComeUus Fuscus in his place.'* Caecina was less successful in his attempt to corrupt the loyalty of the German legions. The officers, indeed, he per- suaded at a secret meeting to proclaim Vespasian, and to place his name on the standards. He enlarged on the merits and fortunes of Vespasian, and the weak- ness of Vitellius. Gaul and Spain, he told tnem, were lost. Now the fleet at Ravenna in their rear had gone over to the enemy. But the common soldiers were loyal : they had created Vitellius emperor, and they would support him. They had conquered the Danube legions once already, and they would conquer them again. As for the fleet, what was that to be set against eight legions ? Legion v Alauda took the lead : the standards of Vitellius were replaced ; they put Caecina himself in irons, and placed themselves under the command of Fabius Fabullus, legate of legio v, and Cassius Longus, praefect of the camp.'* So far, the campaign, since the first rapid advance of Antonius, had somewhat hung fire. On the Flavian side, there had been, first the dilatory advance of the legions, and then the retarding influence of the two consular legates. These dignified persons, perhaps not without the connivance of Antonius, were now got rid of, and he was in supreme and undisputed com- mand. The ViteUians had been kept back by the treacherous inactivity of Caecina. From this point, however, events moved rapidly, 74 lU. 12. 71 ui. 14. 238 STUDIES IN ROMAN HISTORY VITELLIUS 239 too rapidly indeed, as far as the interests of Vitellius were concerned. For this Valens was in a great measure responsible. He had followed his colleague to the north, not indeed with an army, but with a troop of courtiers and courtezans, and his slow progress was marked by orgies and licentiousness. When he ap- proached Ravenna, he heard of the defection of the fleet. Two courses then suggested themselves. He might avoid Ravenna, and push on, either to Caecina at HostiHa, or to the Vitellian force at Cremona. Accord- ing to Tacitus, he would have had time to reach Caecina before his treason was accomplished, or to throw himself into Cremona before the battle. In this way, the whole fortune of the war might have been changed. Or he might have sent for the praetorians from Rome, and with them either have forced his way to the front, or neutralized the action of the fleet. He took neither course at once, and when he did send for reinforce- ments, either his request was not sufficiently urgent, or Vitelhus distrusted him, for only three cohorts and some British cavalry were sent. It was now too late to be present at the decisive action, and with such a force, httle could be done. What Valens did was prob- ably the best thing under the circumstances. Ari- minum, vital as it was through its position on the Via Flaminia, to the communications of the Vitellian army, was bound to be attacked by the Ravenna fleet. Valens therefore threw his small force into the place, while he himself crossed the Apennines and made his way to Pisa with fresh and extensive plans in his mind.'® The VitelUan army on the Tartarus had remained faithful to Vitellius, but in every sense it had lost its head. For his own purposes, Caecina had exaggerated the danger from the revolted fleet : he had declared their supplies to be threatened. The soldiers no longer felt their position secure. They were eager above all things to join their comrades at Cremona where they possibly expected, as Antonius did, that Valens would take the command. They accordingly broke up their camp, destroyed the bridge in front of them over the Tartarus, and fell back on Hostiha, with the view of an immediate march for Cremona. "'^ In order to understand what followed, it is necessary to clear up some obscurity in the account of Tacitus with respect to this march. Mommsen was the first to point out that they could not have taken the direct and obvious road from Hostilia to Cremona, the Via Postumia, involving a march of about sixty miles. Tacitus says that they reached Cremona after a day's march of thirty miles on the evening of the same day on which the Flavians had, as we shall see, fought their way to Cremona. If, therefore, the thirty miles' march on that day had been anywhere on the north side of the Po, the two armies must inevitably have come into colhsion in the course of that day's fighting. ''« It is therefore certain that this distracted and leaderless army actually crossed the Po at Hostiha, and found their way to Cremona by way of Mutina and Parma. There could be, as Mommsen says, no strategical justification for such a course. The Vitellians were as near Bedri- acum as Antonius was : they were nearer to Cremona. They had, too, a day's start, before Antonius could know of their intention. If it was desirable to reach Cremona without fighting, they could certainly have done so by f ! i..^' t < If.',' S4 8S iii. 32-34. m. 37. 86 iii. 42, 43. 248 STUDIES IN ROMAN HISTORY VITELLIUS 249 battle of Cremona, practically the whole of Gaul took the oath to Vespasian. The army on the Rhine, as far as the struggle between VitelUus and Vespasian was concerned, was a negligible quantity. The common soldiers were still faithful to Vitellius ; the officers secretly in favour of his rival, but both were helpless, in face of Civihs and his German army, who had not yet wholly thrown off the mask of asserting the Flavian cause. The legions were, before long, to undergo the disgrace of swearing allegiance to the GalUc empire. In Spain, there was no longer any hesitation. The Othonian legion i Adj. though it had not at first responded to the appeal of the Pannonian army, now openly declared for Vespasian, and brought over with it the other two legions, vi Victrix and X Gemina. In Britain, too, in spite of the hesitation of certain oJSicers, who owed their promotion to Vitellius, the same course was taken. Vespasian was well known in Britain, where his first military successes had been gained under Claudius, as legate of legio 11. Besides this, legion xiv was again in the province, and was from first to last hostile to Vitellius. Thus the whole of the West, hostile as it had seemed a few months ago, was now Flavian in its sympathies. The Maritime Alps came over after the capture of Valens, and, though Tacitus does not record the fact, Raetia also.^' By this time Mucianus was in Moesia. It was per- haps as well that he chose this route, though, as events turned out, the Ravenna fleet could have transported •' This is clear from the action of Sextilius Felix, procurator of Noricum, in marching through Raetia to Germany, iv. 70 ; for other provinces, see iv. 44. his army from Dyrrhacium to Italy. But the Dacians, whose restlessness and growing hostility were soon to lead to the Dacian wars of Domitian and Trajan, using the opportunity presented by the absence of the Moesian legions, crossed the Danube, captured several camps belonging to cohorts and alae, and were preparing to attack Singidunum, the chief legionary camp, when Mucianus averted the danger by sending against them legio VI Ferr. . He then transferred Fonteius Agrippa from the proconsulship of Asia to the command in Moesia, and placed under him some of the conquered Vitellian legions, thus attaining the double object of giving these dangerous troops emplo5mient, and pro- tecting Italy from the possibihty of a combined invasion from Dacia and Germany.^® Meanwhile, after the departure of Caecina and Valens, Vitellius had been lying torpid, slothful and self-indul- gent among the groves at Aricia. News arrived first of the defection of the fleet, then of the treason of Cae- cina, then of his imprisonment by his soldiers. The last point made the deepest impression. However, he returned to Rome, made a eulogistic speech to the people on the fidelity of the legions, and replaced Publi- lius Sabinus, a prot6g^ of Caecina as praefect of the praetorians, by Alfenus Varus. He even went so far as to appoint a successor to Caecina for the last day of his consulship (October 31). This, however, was not quite in accordance with his usual adherence to consti- tutional usage, for never before had a consulship been regarded as vacant, unless the previous holder had abdi- cated, or been deprived of his office by law.®* The Senate was in an awkward position. It was not safe, " iii. 46. «• iii. 38. Mil-* f 1 I 250 STUDIES IN ROMAN HISTORY VITELLIUS 251 as the fate of Junius Blaesus showed, to excite the suspicions of Vitellius, and yet no one ventured to com- mit himself against Vespasian, who might be emperor within a few months. These difficulties were vastly increased, when news came of the disaster at Cremona, and the advance of the victorious army across the Po. Vitellius made the fatal mistake of trying to suppress the news. This not only lost valuable time in which he might have been mustering his still considerable forces to meet the new emergency, but naturally led to exaggerated rumours as to the extent of the crisis. He was himself well informed by his spies, who were indeed carefully sent back with full intelligence by Antonius, but, to prevent the news from spreading, they were put to death. According to Tacitus, Vitellius was roused from this slumber of false security or wilful blindness by the act of a centurion, who, having been sent out by the Emperor and having brought full reports of the enemy's position and strength, left his master's presence and killed himself. At any rate, Vitellius did rouse himself at last. He had in Rome the sixteen praetorian and the four urban cohorts, which, as we have seen, he had created in May out of his own German legionaries. He had besides, a considerable force of cavalry, and also a legion, not, however, formally en- rolled, raised from the marines of the fleet at Misenum.^^ Of these forces, ViteUius now roused himself to send fourteen praetorian cohorts, the cavalry and the marines under Alfenus Varus and Julius Priscus to occupy •• Tacitus does not say which fleet, but Misenum was much nearer to Rome, and the Ravenna fleet had revolted some time before and had sent a body of marines for legionary service to the Flavian army. iii. 54, 55. the passes of the Apennines.®^ But instead of him- self accompanying the army, as even Otho had done under similar circumstances, Vitellius still remained at Rome, assigning consulships in advance, scattering broadcast over the provinces treaties, latin rights and other immunities, hoping by a reckless disregard of the future somehow to strengthen his present posi- tion. At last, urged by the soldiers at the front, he did leave Rome, and leaving it in charge, not of Flavius Sabinus, the praefectus urbi, but of his brother L. Vitellius with the remaining cohorts, he joined his army at Maevania. This was a position on the Via Flaminia, situated on the south-west slopes of the Apennines, and well adapted, if it had only been retained, for defen- sive purposes, though, as Tacitus points out, under other generals, the force of over 20,000 men was quite strong enough to have crossed the Apennines, and undertaken offensive operations. And, in spite of the time wasted, it was not yet too late to take this course. After Cremona, the Flavian army had not advanced with a speed corresponding to its first rapid dash. Antonius himself, indeed, was eager to push for- ward, but there were difficulties in the way. His men were out of control ; they stopped for pillage ; they treated Italy like a conquered country. Antonius, who now behaved as if the legions were his own, was obliged to humour them, even allowing the soldiers to elect their own centurions. Winter too was coming on, and the low-lying country near the Po was flooded. It was found necessary to leave the heavy baggage behind at Verona and even the legions for the present •1 iii. 55. i { Ul 252 STUDIES IN ROMAN HISTORY remained either there or at Hostilia. These were now joined by legio xi Claud, from Dahnatia, which had so far hesitated, and with it 6,000 Dahnatian auxilia.®^ From the enemy, there was no immediate fear of opposition. Ahnost as soon as Valens* back was turned, Cornelius Fuscus with his fleet had seized Ariminum, and secured the greater part of Umbria and Picenum.'^ For the moment, it was the Apennines that separ- ated the Flavian and Vitellian portions of Italy. An- tonius himself with the auxihary forces and some picked legionaries advanced along the Via Flaminia as far as Fanum Fortunae, being joined by the elite of the Ra- venna fleet, which demanded legionary service.®* Here a halt was ordered. Despatches arrived from Vespasian and Mucianus, both urging delay. The former had received the news of Cremona in Egypt. He at once advanced to Alexandria, and was preparing to invade Africa by sea and land. He was naturally anxious, if possible, to prevent further fighting, and to put pressure upon Vitellius by withholding the com supplies from the two great granaries of Rome.®^ The motives of Mucianus were more personal. If Antonius followed up his success and occupied Rome, his own share in the result would be small. A deadly jealousy already existed between him and Antonius. These retarding influences were not without effect in the army at Fanum Fortunae. By condoning breaches of disci- pline, Antonius might count upon the common soldiers, but letters were exchanged between the ofiicers and Mucianus, and however Antonius might pose in his despatches to Vespasian as the conqueror of Italy •t m. 49, so. •s m. 43. •4 lu. SO. •» iii. 48. VITELLIUS 253 and the donor of empire, he did not feel his position wholly secure. Mihtary considerations also demanded some pause. The baggage had to be brought up, and it was danger- ous for the advance guard to get too much out of touch with the legions. The commissariat too required attention, the convoys had to be brought up both by way of the sea and the Po.^® From the enemy, only vague rumours had as yet arrived. The praetorians were reported to have marched from Rome, and it was not doubted that the passes of the Apennines were already in the hands of the Vitellians. To make sure on this important point, and also, if possible, to find an easier route over the mountains than that presented by the Via Flaminia, Antonius and Varus sent forward a force of auxihary infantry and cavalry to explore the region of Umbria.*' To their surprise, it was discovered that the passes were still unoccupied,®^ and it was therefore decided to continue the advance. The advance guard accord- ingly crossed the Apennines, not indeed without diffi- culty, for the snow lay deep upon the mountains, but without any interference on the part of the enemy. They fully realized in the wintry passage how fortune had once more favoured their cause, for if the natural difficulties had been increased by a hostile force, the Flavian army would never have reached the other side. Vitellius had thrown away his last chance of suc- •• iii. 52. •' iii. $2. •• It is not clear whether the ViteUians were yet at Maevania, on the other side of the mountains, but at any rate the passes were clear. 254 STUDIES IN ROMAN HISTORY cessful resistance. He had reached the camp at Mae- vania, attended by a large number of reluctant senators, but his arrival led to no forward movement. There was still time to cross the Apennines, with a good chance of crushing Antonius and Varus before the legions joined them from the rear. His best officers advised this course : his troops were superior in number to the advance guard of the enemy, and still devoted and loyal. Instead of following this advice, he pre- sented to his soldiers a pitiable spectacle of military incapacity, and of reckless self-indulgence, and before long, on receiving news that the fleet at Misenum had revolted, he left his army under the praetorian praefect, and returned to the capital.^® The fleet was induced to revolt by means of forged letters from Vespasian, and some of the neighbouring towns, like Mintumae and Puteoli, joined in the movement. ^°<^ Capua, how- ever, remained faithful to Vitellius, and in any case, the best way of meeting the new danger was not to divide his forces, but to gain some speedy success over the enemy in the front. Vitellius, however, when Claudius Julianus, a former praefect of the fleet, whom he sent against the rebels with an urban cohort, went over to the enemy, and when the strong fortress of Terracina was also seized, completely lost his head. In the first place, he ordered his army to fall back from Maevania to Namia, a retreat of some thirty miles along the Via Flaminia, which left the enemy free to cross the Apennines. ^^^^ In the second place, he took •» iii. 56. iM iii. 57. 101 Tacitus does not state this in so many words, but he speaks of the army being at Narnia in iii. 58, and alludes to the effect of the retreat in iii. 59. VITELLIUS 255 the fatal step of dividing his forces, by sending his brother, L. Vitellius, with six praetorian cohorts and 500 cavab-y to meet the new danger in Campania. 102 Nothing could have been more suicidal. The immedi- ate consequence was a more serious defection than that in Campania, for the Marsi, Paeligni and Samiiites all now declared for Vespasian. io3 The Flavian army, too, finding no force to oppose it, after crossing the Apennines, advanced unmolested as far as Carsulae, only ten miles north of Namia. 10* This was a strong position, though, for the matter of that, the danger of attack was now small, and well supplied with provisions, owing to the wealthy and favourably disposed towns in its rear. It was perfectly safe to wait here for the legionary rear guard. Meanwhile, as the distance between the two camps was so small, there was a chance of winning over the ViteUian soldiers by negotia- tion without further fighting. The soldiers of Antonius were eager to advance on Rome, without delay. He himself, however, now that the campaign was drawing to a close, had uneasy visions of Mucianus and Vespasian behind him. He accordingly represented to his men how important it was to preserve intact the capital and the Senate, and not to increase the odium caused by the sack of Cremona. On the arrival of the legions, it seemed indeed that there would be no further opposition. The common soldiers at Narnia were still faithful to Vitellius, but there was a growing feeling of hesitation and uneasiness among the officers. Owing to treachery on the part of some of their officers, Interamna, where a body of ViteUian cavalry was stationed, was betrayed to the enemy, and "« iii. 59. "8 iii. 59. 104 iii. 60^ 256 STUDIES IN ROMAN HISTORY n since their right wing was now practically turned, the Vitellian generals followed their emperor's example, and abandoning their army retired to Rome. At this, the endurance of the soldiers gave way. There had up to this time been vague reports that Valens had reached Gaul, and was raising a fresh army in Germany. These were now dissipated. Valens had been put to death by the Flavian leaders, and his head was now displayed to his former soldiers. This removed all hesitation. The Vitellians marched down into the plain below Nar- nia, and surrendered to the Flavian army, drawn up to receive them. The prisoners of war were quartered, some at Namia, some at Interamna, in charge of detach- ments of the conquerors. ^^'^ Meanwhile in Rome, Vitellius was in a state of pitiable hesitation. At first, to meet the double danger from Campania and the north, he attempted to raise new legions from the city rabble and even from slaves. The old forms of the dilectus were revived. The tribes were convoked, the consuls charged with enrolling the recruits. Senators, knights and even freedmen were ordered to furnish slaves and money. These recruits were promised not only their discharge but even the re- wards of veterans as soon as victory should be secured. i*^ But VitelUus was only half in earnest with these warUke preparations. He was at the same time inter- changing letters with Antonius and Varus, and more than half inclined to accept their offer of a safe and luxurious retreat for himself and his family on condition of abdicating the empire.^^' The situation in the capital was indeed ambiguous. Flavins Sabinus, elder brother of Vespasian, was still "» iii. 6i, 62. »»• iii. 58 ; Suet. Vit, 15. "' iu, 63. \ VITELLIUS 257 praefect of the city. He was a man with a distin- guished career behind him. He had held his present post for twelve years under Nero; before that, he had been for seven years legate of Moesia. With such antecedents, he might have been expected to play an important part in this critical year. But he had served under Otho, then acknowledged Vitellius, and was ostensibly loyal to him, even while his brother's army was within a few days' march from Rome. Now on the one hand, he was invited by Antonius to escape with his nephew Domitian and join the Flavian army, and on the other he was urged by many of the leading men in Rome to anticipate Antonius and Varus, and to secure the capital for his brother. He might count upon the urban cohorts and perhaps the cohorts of vigiles; Vitellius had only three praetorian cohorts to depend upon, while the city mob, enthusiastic as it was for the moment in the cause of Vitellius, would always take the stronger side.^o® Sabinus, however, was an elderly man, and preferred more diplomatic methods. Several conferences were held between him and Vitellius, and at an interview in the temple of Apollo, witnessed by Cluvius Rufus and Silius Italicus, an arrangement was drawn up for the abdication of Vitellius. ^^^ If only either of these two men had acted with firmness, there might have been no further bloodshed. But the soldiers of Vitellius and other adherents, who had no hopes from the other side, urged the disgrace of such a surrender, the interests of his son Germanicus, the improbability of his own hfe being spared, and the devotion of his soldiers. On December 18, however, 108 iii. 64. lOf iii. 65. 258 STUDIES IN ROMAN HISTORY news arrived of the surrender at Namia, and Vitellius proceeded to carry out his compact with Sabinus. He left the palace in moummg robes, and proceeded to the forum. Here in sight of the assembled people, he offered his dagger to the consul, in token of laying down the power of life and death, and was already seeking the Temple of concord, in order to complete the abdication, when the surging crowd barred his passage and he was forced to return to the palace, his purpose unaccom- llshed.ii« Sabinus meanwhile, supposing that the programme was carried out, had detained his cohorts in their bar- racks, while the leading senators and knights had thronged to his house, on the assumption that Vespasian was the recognized emperor. Then came news of what had happened, and of the threatening attitude of the German soldiers and the populace. Sabinus had hope- lessly committed himself. Even if he had had his co- horts ready, he might have been unable to resist the enthusiasm for the moment, but he had only a small retinue, and as this was attacked by the Vitellians near the lacus Fundanus, he took what seemed the only course open to him, and threw himself into the Capitol with a miscellaneous body of soldiers and a crowd of senators and knights. During the night, a storm raged and the besiegers kept slack watch, so that Sabinus was able to bring into the Capitol his own family and Domi- tian, and also to send an urgent message to Antonius, informing him of the position, and begging for instant relief."^ The Flavian army had by this time left Namia, and was spending the Saturnalia at Ocriculum, only a few miles north of Rome.^^* »»• ui. 67, 68, "Mii. 70, »" iii. 7^- VITELLIUS 259 This delay on the part of the Flavian army which might easily have reached Rome by this time, was put down by some to the restraining orders of Mucianus by others to treachery on the part of Antonius It wa^ more probably due to a desire to avert anything Hke a catastrophe in the city. The negotiations between Sabmus and Vitellius were probably known in the camp and It was hoped that Vitellius would give way An- tonius too had not been entirely inactive. Petilius Cerealis, a relative of Vespasian, and legate of legion ix m Bntam, eight years before, had escaped from Rome and joined the army. Admitted to take rank with the other generals, he had already been sent with 1 000 cavalry, to traverse the Sabine country by cross roads and so to reach Rome by the Via Salaria and the CoUine gate. Had CereaUs used more despatch, the catastrophe m Rome might have been averted, but he was too late When he did arrive, the fighting in the city had already begun, and being attacked as he was threading his way among lanes and gardens, he was at first repulsed and dnven back to Fidenae."' Antonius, on receiving the message from Sabinus realized that no time was to be lost, and started with his whole army along the Via Flaminia direct for the city. He reached Saxa Rubra, outside the Mulvian bndge at night-fall. But though not twenty-four hours had passed since Sabinus had sent out his messenger Antonius was some hours too late. ' At dawn, Sabinus had sent an officer to Vitellius, com- plaining of his breach of faith in violating the agreement amved at. He had been loyal to ViteUius up to the moment of that agreement : after it, he was justified in "* iii. 79. '1 I.i 260 STUDIES IN ROMAN HISTORY regarding Vespasian as emperor. If Vitellius now re- nounced the agreement, it was against the advancing Flavian army that he should turn his arms, not against those whom he had tricked into rebelhon. Vitellius sent back some words of excuse, but he was now help- less in the hands of his infuriated soldiers. He was no longer emperor but only the cause of war. An attack was immediately made upon the Capitol from the forum. A resolute and capable commander ought to have been able to hold the place till relief came. Vitellius had only three cohorts, and there were at least three urban cohorts in the city,"* favourable to Sabinus, as well as the cohorts of vigiles. Sabinus could perfectly well have sent for them during the night, or even left the Capitol himself and brought them up. The attack was without method on either side. The assailants had no leader, and the leaders within the Capitol seem to have lost their heads. The besiegers swarmed over the colon- nades on the right of the forum, undeterred by the showers of tiles and stones hurled down by the defenders. Firebrands, too, were used on both sides, and the assail- ants were only prevented from rushing through the half -burning gates by a barricade which Sabinus ordered to be made of the statues adorning the temple, and commemorating the heroes of worthier days. Then the Vitellians attacked on the left, by way of the Grove of the Asylum and the Tarpeian rock with its hundred steps. Here the defenders were taken by surprise, and in the confusion which followed, the famous Capitoline temple caught fire and was burnt to the ground. The Capitol was taken ; the defenders, who failed to con- >»* The IV had been sent with Claudius Julianus against the rebel fleet. VITELLIUS 261 ceal themselves, were captured or put to the sword. Domitian escaped in the dress of an acolyte. Sabinus was seized, brought before Vitellius, murdered and thrown down the scalae Gemoniae into the Tiber. "^ The crisis was now at hand. The city was in a state of utter confusion ; the Flavians lay concealed to escape massacre ; the German soldiers were resolute and des- perate, and, as we have seen, drove back the cavalry of Cerealis ; even slaves and the city mob were being armed. It was clear that the Flavian army would not enter Rome without a struggle. The struggle would have been more desperate and the bloodshed far greater, if L. Vitellius had led his six praetorian cohorts back to the city. He might have done so, for he had just re- captured Terracina through the treachery of a slave, and put its garrison of gladiators and marines to the sword."* But he contented himself with sending messages of his success to Rome, and that danger was averted. At Saxa Rubra, Antonius was met with the fatal news that the Capitol was taken, Sabinus killed, and that the slaves and city mob were being armed. He now too heard for the first time of the repulse of Cerealis. There was no rest that night on either side. Vitellius, utterly distracted, at one moment urged the half-armed mob to save the city, at another called the Senate and caused envoys as weU as the Vestal Virgins to be sent to the Flavian leaders, begging for a day's respite, in order to effect a settlement. But the Flavian army was once more out of control. The envoys sent to Cerealis, who was again advancing, were roughly handled, and Arulenus Rusticus, the praetor, wounded. Those sent to ii» m. 70-72. "• iii. 76, jj» 26o STUDIES IN ROMAN HISTORY VITELLIUS a6i il it ill regarding Vespasian as emperor. If VitelUus now re- nounced the agreement, it was against the advancing Flavian army that he should turn his arms, not against those whom he had tricked into rebelhon. VitelUus sent back some words of excuse, but he was now help- less in the hands of his infuriated soldiers. He was no longer emperor but only the cause of war. An attack was immediately made upon the Capitol from the forum. A resolute and capable commander ought to have been able to hold the place till reUef came. Vitellius had only three cohorts, and there were at least three urban cohorts in the city,"* favourable to Sabinus, as well as the cohorts of vigiles. Sabinus could perfectly weU have sent for them during the night, or even left the Capitol himself and brought them up. The attack was without method on either side. The assailants had no leader, and the leaders within the Capitol seem to have lost their heads. The besiegers swarmed over the colon- nades on the right of the forum, undeterred by the showers of tiles and stones hurled down by the defenders. Firebrands, too, were used on both sides, and the assail- ants were only prevented from rushing through the half -burning gates by a barricade which Sabinus ordered to be made of the statues adorning the temple, and commemorating the heroes of worthier days. Then the Vitellians attacked on the left, by way of the Grove of the Asylum and the Tarpeian rock with its hundred steps. Here the defenders were taken by surprise, and in the confusion which followed, the famous CapitoUne temple caught fire and was burnt to the ground. The Capitol was taken ; the defenders, who failed to con- »»* The IV had been sent with Claudius Juhanus against the rebel fleet. ceal themselves, were captured or put to the sword. Domitian escaped in the dress of an acolyte. Sabinus was seized, brought before Vitellius, murdered and thrown down the scalae Gemoniae into the Tiber. ^^^ The crisis was now at hand. The city was in a state of utter confusion ; the Flavians lay concealed to escape massacre ; the German soldiers were resolute and des- perate, and, as we have seen, drove back the cavalry of Cerealis ; even slaves and the city mob were being armed. It was clear that the Flavian army would not enter Rome without a struggle. The struggle would have been more desperate and the bloodshed far greater, if L. Vitellius had led his six praetorian cohorts back to the city. He might have done so, for he had just re- captured Terracina through the treachery of a slave, and put its garrison of gladiators and marines to the sword. ^^' But he contented himself with sending messages of his success to Rome, and that danger was averted. At Saxa Rubra, Antonius was met with the fatal news that the Capitol was taken, Sabinus killed, and that the slaves and city mob were being armed. He now too heard for the first time of the repulse of Cerealis. There was no rest that night on either side. Vitellius, utterly distracted, at one moment urged the half-armed mob to save the city, at another called the Senate and caused envoys as well as the Vestal Virgins to be sent to the Flavian leaders, begging for a day's respite; in order to effect a settlement. But the Flavian army was once more out of control. The envoys sent to Cerealis, who was again advancing, were roughly handled, and .\rulenus Rusticus, the praetor, wounded. Those sent to ii» m. 70-72. "• iii. 76, 77, 262 STUDIES IN ROMAN HISTORY VITELLIUS 263 Antonius, together with the Vestal Virgins, escaped out- rage but were sent back with the grim reply, that after the death of Sabinus, no terms could be entertained, i" Antonius, fearful of outrage on the part of his men would gladly have waited till dayhght, but the soldiers would brook no delay. The army advanced in three lines, one along the Via Flaminia, towards the north gate of the city, another along the bank of the Tiber, by way of the Campus Martins, the third by the Via Salaria to the Colline gate. The half-armed populace was soon dispersed by a charge of the cavalry, but the Vitellians offered a desperate resistance at every point. But few in number and without leaders, they were gradually driven back within the city walls. Only at one point on the left, where the Flavians became entangled in narrow and sHppery roads, near the gardens of Sallust, did the Vitellians hold their ground till late in the day, when they were taken in the rear by the cavahy of Cerealis which had forced its way in by the ColUne gate. Street by street the ViteUians were forced back, fighting for every inch of ground. They received no help from the cowardly mob, which delighted in the hideous spectacle, as if it had been a gladiatorial show, intended for their amusement. While the soldiers fought, they piUaged. Loafers and prostitutes thronged the taverns, applauded the combatants and made off with the booty. The final episode was the storm of the praetorian camp, to which the surviving VitelHan soldiers retreated. Here the last stand was made, and here the German soldiers were cut down to a man with all their wounds in front. ^^^ Their end was at least worthy of brave men. That of Vitellius was as contemptible as most of his life had 1" iii. 80, 81. ii« iii. 8-i-84. I been. When all was lost, he left the palace by a back way, and was carried in a htter towards the Aventine, intending to take refuge in his wife's house, till he could escape to his brother at Terracina. But his courage failed him ; he dared not face the streets and returned to the deserted palace. He was soon dragged from an unsavoury hiding-place, and hurried through the forum, amid jeers and blows and execrations. His head forced up with sword points, he saw his statues thrown down, he noted the spot where not a year ago Galba had more worthily fallen, and then the Gemonian Stairs, where the body of Flavins Sabinus had been flung into the Tiber. Here he was despatched, and the most unworthy emperor of Rome disappeared from the scene. The Flavian period had begun, but there was as yet no ruler and no government. Domitian, on emerging from his conceahnent, was saluted Caesar by the soldiers, and conducted to his father's house. But he was im- mersed in his pleasures, and no direction of affairs was to be expected from him. The Senate it was at first impossible to convene, as the magistrates and senators were still in hiding. But no government could have saved Rome from the immediate consequences of the victorious entry of the Flavian army. For days it was the scene of plunder, outrage and murder. The entry of Vitellius had been bad enough, and the Ucence allowed to his soldiers not to be soon forgotten. But the ex- cesses of the Danube army put all this into the shade. Even Antonius who for the moment held the chief power, appropriated money and slaves from the im- perial household, as though they had been the spoils of Cremona. *^* »»» iv. 2. 264 STUDIES IN ROMAN HISTORY VITELLIUS 265 However, as the fury of the soldiers worked itself out, some order was restored. Military necessities were not forgotten. L. Vitellius with his six cohorts had still to be dealt with. The cavalry was sent for- ward to Aricia, a legionary force to Bovillae. No campaign, however, was necessary. Vitellius at once surrendered his army, which was conducted to Rome, and showed by its dogged and sullen behaviour what an increase of bloodshed would have resulted from its earUer arrival in Rome. L. Vitellius himself was executed, a man of far more capacity than his brother, but also of greater cruelty. The movement in Cam- pania was soon stamped out by Lucilius Bassus, at the head of some auxiUary cavalry. Capua alone suffered for its fidehty to Vitellius, legio iii being quartered there during the remainder of the winter."© When the Senate met, its first acts were the decree of honours to the victorious leaders, the consularia orna- menta to Antonius Primus, praetoria ornamenta to Arrius Varus and Cornelius Fuscus, the former of whom, either by letters from Vespasian or the authority of Antonius, had already been made praetorian praefect. On Vespasian and Titus was conferred the consulship for next year, on Domitian the praetorship with con- sular insignia. To Mucianus, in spite of the offence created by a somewhat arrogant letter, was granted the honour of a triumph, ostensibly for his successful repulse of the Dacian invasion of Moesia. But of course the most important act of the Senate was the formal recognition and appointment of Vespasian, the bestowal of the nomen imperatorium, and of the titles of Caesar and Augustus. Fragments of the 130 IV. 2, 3. senatorial decree, afterwards to be ratified by a law passed in the Comitia, conferring on Vespasian all the special privileges granted to former emperors, together with certain new ones, are still extant under the name of the Lex de imperio Vespasiani. i" On questions of administration, in spite of the some- what independent attitude taken up by Helvidius Priscus, and though the financial situation was declared by the two praetors of the aerarium to be critical,"* the Senate decided to wait for the arrival of Vespasian himself. The transition period between the Julio-Claudian and the Flavian lines came to an end with the entrance of Mucianus into Rome in January, 70 a.d. Under his capable control, order was restored to the capital, and peace to Italy. Antonius Primus lapsed into insignificance ; the legions were sent back to the pro- vinces, and above all, adequate steps were taken to re-establish Roman authority and Roman prestige in Gaul and Germany. "^ For fifteen months the whole Roman Empire had been convulsed. Nero's death had caused a breach of continuity in the imperial system, a result, bound to ensue sooner or later, of the vague and tentative way in which Augustus had treated the question of i«i Bruns. font, fur. Rom. ant. p. 192 foil. "• When these praetors of the aerarium were substituted for the praetorian praefecti, appointed by Nero in 56 a.d., is not clear. It was probably a concession made to the Senate by Galba, Otho or VitelUus {Ann. xiii. 29 ; Hist. iv. 9). »" See especially iv. 68. The events in Gaul and Germany, the rising of Civilis and the attempt of certain states to set up an imperium GaUiarum have already been to some extent dealt with in the essay on the German army, and I have therefore omitted here all reference to the subject. 266 STUDIES IN ROMAN HISTORY VITELLIUS 267 '■i the succession. The praetorians, ever since their con- centration under Tiberius, had made their influence felt in the selection of a new princeps. When a single provincial army, and that not one of the most important, disclosed the secret, " posse princi- pem alibi quam Romae fieri," nothing could have pre- vented civil war, except the absolutely convincing character of its choice. The suitability of Galba was not convincing. If he had been twenty years younger it might have been. As it was, the powerful German army, elated by their crushing defeat of Vindex, at once followed the example set in Spain, and offered the empire to Verginius Rufus. Vitellius was their second choice, and owed his elevation, not to his own merits or popularity, but partly to the machinations of Valens, partly to the prestige of his father, but chiefly to the determination of the German army not to accept the nominee of the Spanish legion. If the British army had had a capable legate, and especially if the proud legio XIV had been in the province, quite possibly another claimant to empire might have been put for- ward. As it was, the British legions supported the German. In Rome, the praetorians, proud of their privileged service, were not likely tamely to give up their right of selecting emperors, and had already set up Otho. It only remained for the Danube army to intervene. It had no candidate of its own, as the three provincial legates, Aponius Saturninus, Tampius Fla- vianus and Pompeius Silvanus were mere titled nonen- tities, but jealousy of the German army, to which it was hardly inferior in strength or reputation, led it to support the choice of the praetorians. That the Oriental legions, whose recent exploits had put those of the Western armies into the shade, would permanently allow the latter to dispose of the imperial position, was not probable. They had, too, not one but two worthy candidates. The Syrian legions did not actually take part in the civil war, simply because the Danube army had no hestitation in substituting the name of Vespasian for that of Otho on their standards, but they put forward the same claims, and would have maintained them by arms, if it had been necessary. Thus it was a war of rival armies. The military basis of the empire, so studiously kept out of sight by Augus- tus, had come to Ught with a vengeance. Even the war on the Rhine was very largely, though not so entirely as Mommsen represents it, a war between legionaries and auxiliaries typified by the rancorous feud between legion XIV and the Batavian cohorts. The whole history of the empire was during this short period diverted into a military channel. Military rivaky and esprit de corps ran riot. The armies were con- tending not to secure a good emperor, but to assert their own nominees. The claimants themselves were Uttle more than names to the soldiers who fought for them. It is true that much devotion and much obstin- ate fidelity were shown, especially by the German sol- diers to their unworthy emperor, but all the peculiar characteristics of the war are to be explained by this mad rivabry of army corps with army corps. No great generals emerged in the course of the war ; no permanent reputations were made, and at least one great reputation was sullied and lost. The mad impetuosity, the mutinous spirit, and the constant suspicions of the soldiers made sound strategy and skUful tactics aUke impossible. Valens and Caecina owed their sue- 268 STUDIES IN ROMAN HISTORY cess to the mistakes of their opponents ; Antonius Primus, dashing and capable as he was, owed more to the treachery of Caecina and the lethargy of ViteUius than to his own generalship, while Vocula on the Rhine was from first to last the leader of a forlorn hope. The empire, however, survived this desperate crisis, and with 70 a.d. the Flavian period commenced. i'ii t A MILITARY GAME OF CHESS Among all the wars recorded by Greek and Roman historians it is hardly too much to say that not one is attended with so extraordinary an interest as the civil war between Caesar and Pompey in 49 and 48 B.C. The political importance of the struggle it would be difficult to exaggerate. On its issue depended the future of Rome and the Roman Empire. What the victory of Caesar was destined to bring about there is no need to conjecture, for they are to be found in his own Dictatorship and in the Principate of Augustus. What would have been the consequences to the future history of Rome and of Europe, if Pompey, utterly devoid himself of political genius and political con- viction, and surrounded by all that was reactionary in the senatorial oligarchy, had realized his own con- fident expectations, it is impossible to say, though perhaps not difficult to imagine. But it is with the military and not with the political interest of the war that this paper is concerned. The war in fact is a perfect piece of military chess. It was fought out between the two foremost generals of the day. It was won by perhaps the most brilliant military leader the world has ever seen, and the account of it which has come down to us is the account of Caesar himself. Till ten years before the outbreak of this war, Pompey had stood almost without a rival as the one great general of the day. Il 270 STUDIES IN ROMAN HISTORY A MILITARY GAME OF CHESS 271 Caesar's military genius had lain entirely dormant till 61, when he had gained a certain reputation from his brief command in Further Spain, but it was only after his appointment to the Gallic provinces in 59 that he had r(3ally important military work to do. During the nine following years, in which Gaul was gradually conquered between the Bay of Biscay and the Rhine, Caesar's reputation as a general of consummate ability, daring and resource, was fully established. What was perhaps of still greater importance for his own career, was the fact that during these years, a trained and efficient army of eleven veteran legions was being formed, devoted to their leader, and ready to follow him in any enterprise. The civil war was, of course, due to far deeper causes than the personal jealousy between Caesar and Pompey. The position of Caesar, as the leader of 80,000 or 90,000 men, wholly independent of the Senate, and continued for ten years, was inconsistent with the theory of the Republican Constitution, and the Home Government was anxious to bring it to an end at the earliest possible moment. The only possible way of effecting this was, as things were, by means of an alliance with another military leader. Since 59, Pompey and Caesar had been ostensibly working together as members of the first Triumvirate, but Pompey was feeling himself more and more overshadowed by the growing miUtary prestige of Caesar. This was where the personal motive came in, which eventually led to an alliance between Pompey and the Senate. After this necessary introduction, we will proceed to the war as a game of Military Chess. Compared with modem wars, it was, of course, on a small scale. Even at the final battle of Pharsalia, less than 70,000 men were engaged. But on the other hand, it was by no means a miniature war. It was marked, within a period of twenty months, by critical campaigns in Italy, Spain, Africa and Greece. It was a war, if one may use the term, of extraordinary symmetry. There were very definite moves in the game, occasions on which now one general, now the other, had before him a choice of alternative courses. That the various episodes in the war were short, was due not to the absence of complicated conditions, or of those causes which often prolong wars, but to the brilliant rapidity and initiative of Caesar himself. Let us begin by casting a glance at the position, strength and prospects of the two rivals. Caesar had been for the past ten years in command of the Gallic provinces, both Transalpine and Cisalpine. The great importance of the former to him was that during its conquest he had got together and trained a powerful army of eleven legions, which were now at his absolute disposal. There had been times, and quite recently, during the rising of Vercingetorix, when Gaul would have been no help but a hindrance to Caesar in a struggle with Pompey. But at the present time the country was quiet, and the hopes of his opponents that Gaul might rise behind him were not realized. Beyond this negative advantage, there were supplies to be drawn from Gaul, as long as the war was in the West, and some useful Gallic cavalry. The possession of Cisalpine Gaul, the northern part of Italy, was of far more immediate moment. The Transpadane dis- trict supplied him with an excellent recruiting ground for his legions, and the country was the key of Italy from the north ft'- 272 STUDIES IN ROMAN HISTORY Pompey's resources, though not so concentrated, were to all appearance greater. For the past six years he had been governor of the two Spanish provinces, and though he was not personally in command of the army there, his own influence in the country was great, and his seven legions were under the command of three legates of his own, two of whom, Afranius and Petreius, were able soldiers. In Italy, too, his position was apparently a strong one. For years he had been allowed to retain in Italy several legions, raised for service in Spain. The soldiers, it is true, were not mobilized but on furlough, but they might be available on short notice. From the beginning of 49, too, Pompey was the recognized general of the Senate, and a general levy throughout Italy had been ordered. Pompey himself had declared that he had only to stamp with his foot, and legions would rise from the ground. Two legions he akeady had mobihzed at Capua, but these had dishonestly been obtained on the plea of a Parthian war from Caesar's army, and therefore were not wholly reliable, though their presence meant the reduction of Caesar's army to nine legions. In addition to Spain and Italy, and all the prestige attaching to the support of the Senate and the possession of Rome, Pompey could depend with certainty on all the resources of the East, both military and financial, owing to the influence he had gained there during his brilliant campaigns first against the Cilician pirates, and then against Mithridates and Tigranes. His father-in-law, Metellus Scipio, was now governor of Syria, and several of the client kings in the East, such as those of Galatia, Cappadocia and Conmiagene, were bound to him by ties of gratitude. The same A MILITARY GAME OF CHESS 273 influences would make him absolute master of the sea, whenever he chose to raise a fleet from his Eastern allies. Of the other provinces, the Senate, at the first outbreak of war, had sent out governors to secure Sardinia, Sicily and Africa, the chief corn-producing provinces. The war may be said to have commenced on January 7, when Caesar's ultimatum was rejected and he himself declared a public enemy. Caesar himself was in Cisal- pine Gaul, at Ravenna, but of his nine legions only one was with him. Of the rest, two were under orders to join him with all speed. But it was not only with Pompey himself that Caesar had to deal. There was always the possibility of disturbance in Gaul, and there was the very real danger of a diversion being made by Pompey's army in Spain, which might try to cross the Pyrenees, and so to march into Italy. Trebonius, therefore, one of his legates, was ordered to remain in the neighbourhood of the Rhine with three legions, while Fabius, another legate, with the remaining three, was to advance to Narbo, in order to guard against any such attempt on the part of the Spanish army. With these precautions taken in the rear, Caesar, on hearing that his ultimatum was rejected, had no hesi- tation in crossing the Rubicon, the boundary between his province and Italy, with a single legion and occupy- ing Ariminum. It was no rash step. He knew exactly when to expect two other legions ; fresh levies were being raised for him in the Transpadane district ; he was aware that Pompey had no forces reliable enough to lead against him, and that in any case some time must elapse before any concentration likely to stop him could take place. T f ii ;l t1 274 STUDIES IN ROMAN HISTORY A MILITARY GAME OF CHESS 275 Everything depended upon rapidity of movement, and accordingly within a few days, Caesar had ad- vanced along the coast as far as Ancona, placing a small garrison there, and at Firmum, Pisaurum, and Auximum. At the same time, to prevent his com- munications with Gaul from being cut of!, he sent Antony to occupy Arretium on the Via Cassia, and Curio to occupy Iguvium on the Via Flaminia. The effect of these two last movements was that the Pom- peian officers raising troops in Etruria and Umbria were forced to beat a hasty retreat. If Pompey had ever had the idea of advancing northward against Caesar, this was now given up, and in fact Rome itself was hastily evacuated by the Senate, which made a precipitate flight to Capua. Pompey with the two Caesarian legions, and whatever other forces he had raised, retreated to ApuUa, where he made Luceria his headquarters. Pompey in fact realized that his original plan of raising sufficient troops in Italy to oppose Caesar was hopeless, and he accordingly deter- mined to have all his forces within reach of Brundisium, with a view of transferring the war to Greece or the East. Under the circumstances, this was probably good policy, and Cicero's ignorant criticisms of him as an incapable general as well as an incapable politician have little point. There were, however, still two chances to be waited for, before deciding on the retreat to Brundisium. In the first place, Picenum, where Pompey's family and personal influence was great, might under the leader- ship of Lentulus Spinther offer a more strenuous re- sistance to Caesar. In the second place. Domitius Ahenobarbus had collected over three legions and with them had thrown himself into Corfinium. If only Picenum could bar Caesar's way a little longer, and Domitius be induced to join Pompey in Apulia, there might still be time to collect an army strong enough to try conclusions with Caesar in Italy. Both chances failed. Caesar's swoop southwards was far too rapid to allow of any concentration or resistance on the part of the Pompeian officers in Picenum, and the news came that Cingulum, Fanum, Asculum and the whole of Picenum was in Caesar's hands. With regard to Domitius, instead of obeying orders, he had a plan of his own and urged Pompey to join him at Corfinium and with their combined forces to bar Caesar's advance. Such a plan would inevitably have failed. Pompey's troops were not reliable, nor, as the event proved, were those of Domitius, while Caesar was now reinforced by another legion from Gaul. Letters were ex- changed between Pompey and his obstinate officer until it was too late for either course to be adopted. Caesar, realizing the importance of Corfinium, the citadel of central Italy, occupied a position to the south of it, and so cut Domitius of! from his chief. After a delay of only seven days, the soldiers of Domitius insisted on a capitulation, and took the oath of allegiance to Caesar. Hardly six weeks had passed since Caesar had crossed the Rubicon, and the way lay open to Rome. Rome, however, was not Caesar's object. If only he could prevent Pompey from withdrawing his army from Italy, the war might be ended at once. But Pompey had no further hesitation ; he was nearer to Brundisium, and Caesar had lost seven valuable days before Corfinium. He therefore succeeded in collecting all his troops, it iitl I u I 276 STUDIES IN ROMAN HISTORY which, including the levies in Campania and Apulia, amounted to about five legions, within that city, before the arrival of Caesar. Half of his army he at once sent across to Dyrrachium, but strangely, considering his command of the sea, he had neglected to provide ships enough for all, and he had to remain behind with the rest till the ships returned. Caesar threw lines of circumvallation round the city from the land side, and began to dam up the entrance to the harbour. But the ships came back before this was completed, and Pom- pey with considerable military skill succeeded in baffling Caesar and getting the rest of his army over to Greece. This ended the first division of the war. Caesar's strategy had been faultless. He had never for a moment hesitated, and his marvellous rapidity of movement had simply bewildered his antagonist. It was not yet the end of March, and Italy was cleared of the enemy. Of Pompey's action it is more hard to judge. His great fault lay in having been unprepared at the com- mencement. For years he had been at the head of an ItaUan army in posse. For months he had been deliberately preparing for the breach with Caesar, and it argues an extraordinary want of foresight, that when the time came, he was practically without an army of his own. He pubUcly announced, when Caesar crossed the Rubicon, that he had ten legions ready for action. If he had had half that number the issue of the whole war might have been different. Caesar's dispositions would have had to be essentially modified, and it is possible that he would not have been able to spare a sufficiently strong force to be an effectual check upon Pompey's Spanish army. As it was, Pompey was dependent upon the promptitude, capacity A MILITARY GAME OF CHESS 277 and obedience of his officers, and Caesar found no difficulty in disposing of them separately, and preventing anything like a formidable concentration of force. The move was with Pompey when he left Brundisium. There were two alternatives open to him. He might transfer his army to Epirus and Macedonia, and then make use of his influence in the eastern half of the empire to raise an overwhelming force, with which to return, as Sulla had done, for the reconquest of Italy. Or he might make a concentration of his present forces in Spain, where he already had seven legions. There was something to be said for this latter course, and there would have been more, if Pompey could only have foreseen that three of Caesar's legions would be detained by the siege of Massilia. Pompey would have been at the head of a really powerful army in Spain, and owing to the influence he had gained in the country during the Sertorian war, he could easily have raised a useful force of native Spaniards. As far as position and resources were concerned, he would have been on at least equal terms with Caesar, and the fleet which was already being collected could have secured Africa and perhaps even Sicily. However, Pompey decided against this course, and probably with wisdom. It would be staking everything upon a conflict on not unequal terms with his great antagonist, and it would have left almost unutilised that on which he had all along been counting, the resources and strength of the East. The real importance to Pompey of his Spanish army was that at the worst it might be depended upon to detain Caesar for at any rate the remainder of the year, till the army in the East was prepared, while, at the best, it might co-operate 278 STUDIES IN ROMAN HISTORY with that army in a jomt and overwhelming attack from east and west. Pompey's move, therefore, was to cross to Epirus, and there at his leisure to col- lect an army. At the same time, a fleet was to be got together which would blockade Italy, secure the corn-producing provinces, and finally transport his army back to Italy. Caesar had probably no hesitation in deciding on his own next move. To follow Pompey to Epirus was out of the question for several reasons. In the first place, he had no fleet, and though no doubt he might have marched his army by land through Illyricum, this would have involved loss of time, and time was all important. But apart from this, it was clearly impossible to transfer his army, or even a considerable part of it, to the East, while Spanish legions were still menacing Gaul and Italy. Caesar decided, therefore, to leave Pompey alone for the present, and to con- centrate his forces against the Pompeian army in Spain. All that he could do for the moment by way of preparation against Pompey or to frustrate his mtentions, was to give orders for ships to be collected from all quarters at Brundisium, and to secure Sicily and Africa. Curio was sent to the former with the legions which had surrendered at Corfinium, with orders, as soon as it was possible, to cross over into Africa. From Brundisium Caesar proceeded direct to Rome, where a few political arrangements had to be made. We shall not, however, mterrupt the thread of mili- tary events by touching on the political situation. The delay on Caesar's own part was brief, and it involved no delay in his preparations for the next move in the game, for he had already sent forward his three legions A MILITARY GAME OF CHESS 279 from Brundisium, to join Fabius, who was ordered at once to advance to the Pyrenees. After the briefest stay in Rome, Caesar himself hastened north to Gaul, leaving Antony and some of the fresh Italian levies to maintain order in Italy. On reaching Gaul, Caesar found that the powerful and important city of Massilia had been induced by Domitius Ahenobarbus, who had escaped there after the surrender of Corfinium, to declare for Pompey, and to close its gates against him. It was essential that the city, situated as it was be- tween Italy and Spain, should be reduced, and Caesar, though he had no time to direct the siege himself, was obliged to leave three legions with Trebonius to invest the place. At the same time he ordered some ships to be built at Arelate, and leaving them under the command of Decimus Brutus, to hem in the city from the sea, he himself hurried on to join his main army in Spain. Spain, as we have seen, had for the last six years been governed for Pompey by three legates. Terentius Varro was in the southern province with two legions, Petreius in the west with the same number, and Afranius in the north and east with three. Varro took no part in the campaign now to be described, but Petreius and Afranius joined their forces and having made some attempt, but too late, to prevent Fabius from crossing the Pyrenees, had taken up a strong position at Ilerda. Their combined force consisted of five legions, about 30,000 men, 5,000 Spanish cavalry, and eighty cohorts, between 30,000 and 40,000 of light-armed native troops, drawn from the tribes of Lusitania, Celtiberia and Cantabria. Ilerda, the modem Lerida, was situated on a considerable plateau ;V1 i 28o STUDIES IN ROMAN HISTORY A MILITARY GAME OF CHESS 281 just above the right or western bank of the Sicoris, a tributary of the Ebro, and about thirty miles above Octogesa, where the two rivers join. Another tribu- tary of the Ebro, the Cinga, flows down from the north- west, and between it and the Sicoris is a wide open plain, narrowing towards the Ebro, but about thirty miles across at the point where the two armies were engaged. Afranius had collected plentiful supplies within Ilerda, but his actual camp was on rising ground, a little to the south-west of the town. Between them was a smaller hill, and just below Ilerda, there was a stone bridge over the river. The position was strong and well chosen. It commanded the approach to the Ebro and the Celtiberian country beyond it. Ilerda was practically impregnable. It was in close com- munication with the camp, also strongly fortified, while the permanent bridge gave Afranius the command of both banks of the river. The native Spaniards were used to guerilla warfare, and many of them were able to cross the river independently of the bridge. Fabius had crossed the Pyrenees before the end of May, and when reinforced by the three legions from Italy, he had six legions, or about 36,000 men, under his command. In addition to these, he had 10,000 Gallic auxiliaries and 6,000 cavahry, half of whom had served in all Caesar's campaigns. With this army, slightly but not much inferior to that of Afranius, Fabius advanced into the head of the plain between the Cinga and the Sicoris, and pitched his camp some fifteen or twenty miles north of Ilerda, to await the arrival of Caesar. His communications with the left bank, from which all his supplies and reinforcements must reach him, were secured by two temporary bridges over the Sicoris, four miles distant from each other. The possible dangers of his position were soon shown, when one of the bridges was swept away, and part of his army, which was on the left bank, was all but cut off by the Pompeians. But for the existence of the second bridge a serious disaster must have followed. In the last days of June Caesar himself arrived with 500 fresh cavalry. From this point events marched rapidly. Caesar at once broke up the camp, leaving a small force to look after the bridges, and marched direct to Ilerda. As Afranius refused his ofier of battle, Caesar, while his front rank was drawn up to face the enemy, with the rest of his men speedily forti- fied a camp immediately to the west of the Pompeian camp and Ilerda. Afranius had everythmg to gain by delay. Caesar everything to lose. His communications with Gaul were at the best not very secure, and the Pompeians, having command of the left bank, might at any time attack his bridges from that side. He accordingly attempted by a sudden attack to seize the hill between Ilerda and the Pompeian camp, which Afranius had strangely omitted to protect. If this dash had been successful, Caesar would have been able to cut off the Pompeians both from their supplies in Ilerda, and from the stone bridge. Fortunately for Afranius, he was able to anticipate the movement, and to reach the hill first. There was hard fighting for several hours, and it was with great difficulty and not without severe loss that Caesar was able to extricate the picked body of men he had selected for this hazardous duty. The result was a decided check, if not a serious danger to Caesar's position. But worse was to follow. Is,. 282 STUDIES IN ROMAN HISTORY Heavy rains ensued, the Sicoris overflowed its banks, and both Caesar's bridges were swept away. The posi- tion was now critical, and might become desperate. Caesar was imprisoned in the partially flooded plain, between the two rivers ; the supplies were running short, and his soldiers had just sufliered defeat. The Pompeians on the other hand, had plenty of supplies at Ilerda, were in an impregnable position, and by means of their bridge could freely traverse the country on the left bank, and interfere with Caesar's convoys. It happened that at this very time, a large and im- portant convoy from Gaul was on its way to Caesar's camp, but being unable to cross the river, had only escaped capture at the hands of the Pompeian cavalry by a retreat to the mountains. Corn rose to famine prices in Caesar's camp, and it looked for the moment as if Caesar would have to attempt a retreat towards Gaul, which might easily have ended in a disaster. The news of his difficulties reached Rome in an ex- aggerated form, while the Pompeians were encouraged by the improbable rumour that Pompey himself was advancing to Spain by way of Mauretania. But Caesar was never so dangerous as when hard pressed, and within a very few days the situation was en- tirely changed. The immediate need was to establish communications with the missing convoy, and this Caesar effected by a brilliant and original device. He set his soldiers to work at constructing a number of coracles, such as he had seen in Britain, out of skins stretched on wicker frames. These, when completed, he had conveyed by night in carts twenty-two miles up the river, so as to be out of reach of the Pompeian cavaky. With the help of these coracles, some troops A MILITARY GAME OF CHESS 283 were conveyed across the river, the convoy safely got in, a fort built and entrenched on the left bank, and finally a permanent bridge thrown across the river. This was the turning point m the campaign. Caesar had now plenty of provisions, and his cavaky were able to range at will along both banks of the river. The advantages of the Pompeian position were largely neutralized, and sooner or later their supplies would begin to fail. Already several of the Spanish states had sent in thek submission to Caesar, and every such accession turned the scale in Caesar's favour. His one disadvantage lay in the long distance between his camp and his bridge. If this could be remedied, the Pompeian position would be completely turned. At a point a mile or two above Ilerda, the Sicoris was divided into several channels and Caesar chose this spot as best adapted to his purpose of divert- ing the water from the river by means of ditches, so as to make it fordable. The work was hurried on, and the water soon became shallow enough for the cavaky to pass over. Afranius, however, did not wait for this. He reaUzed that his position was turned, and determined to retreat to the Ebro. Ccltiberia, the country south of that river, was well disposed to Pompey, and there was every chance that a successful stand might be made there. Orders were accordingly sent to Octogesa for ships to be got ready to carry the army across the Ebro, and preparations were made for an immediate start. The distance to be traversed was about thirty miles. There were three possible routes, one down the right bank of the river, one following the left bank, and one somewhat further away from the left bank, through 284 STUDIES IN ROMAN HISTORY A MILITARY GAME OF CHESS 285 hilly * country, and for the last five miles over the mountains. As his object was to get a good start of Caesar, Afranius had clearly to choose between the two last. There seemed every chance of success, for though Caesar's cavaky could cross by the ford, and so would only lose three or four miles, his infantry, as it seemed, would have to march twenty-two miles up the river to the bridge, and then twenty-two miles down again, before they reached the starting pomt of the Pompeians. To obtain some protection against Caesar's cavah*y, Afranius chose the third route, which, though somewhat longer, was through ground less favourable for the operations of mounted troops. Therefore, abandoning the greater part of their stores at Ilerda, but leaving a sufficiently strong force to prevent Caesar from using the stone bridge, the Pom- peians started early in the day and marched with all speed. Caesar's cavalry were soon upon them, harass- ing their rear. Then was illustrated the effect of Caesar's personality upon his soldiers. The infantry from their camp on the right bank could see what was taking place on the other side, and they became so eager to take a share in the action, that they insisted on being allowed to attempt the ford. Caesar consented, and with only slight loss the whole army crossed to the left bank, and in spite of every effort on the part of the Pompeians, Caesar caught them up late in the afternoon. Both armies encamped for the night, and the next day was spent in reconnoitring. On the third day Caesar started with his army apparently in the direction of Ilerda, and the Pompeians, thinking that he had given up the pursuit, leisurely continued their march. Caesar, as soon as his movements were hidden by the high ground, had rapidly wheeled round to the right and was marching through rougher and more moun- tainous country with the object of getting between the Pompeians and the final defiles leading down to Octo- gesa. It was a desperate race, but Caesar won it. Afranius, on coming within sight of the mountain passes which meant safety for his demoralized army, found Caesar's army confronting him. Again the two armies encamped. Afranius made one last attempt to get forward by detaching some divisions to occupy a hill nearer to the passes. When these were cut to pieces by Caesar's cavalry, there was no course left except to try to regain Ilerda. It was already a hope- less position ; the cavaky had lost their nerve and were worse than useless, and Caesar might easily have stormed the camp if he had chosen. But he resisted the eagerness of his men to make the final attack, and allowed Afranius to return to his last camp. The Pompeians now made for the river, but they were harassed at every step by the cavalry, and they soon found the light-armed troops all round them. In the end the Pompeian generals found themselves com- pelled to surrender at discretion with their whole army. The surrender took place on August 6, only forty days after Caesar had joined his army in person. It was probably the most brilliant campaign in all Caesar's brilliant career. Afranius had had every advantage, a larger army, choice of position, plenty of time for preparation, abundant supplies, a friendly country round, and, what he could not have calculated upon, the rising of the river. Caesar had triumphed as usual 1^* i !<: ui 286 STUDIES IN ROMAN HISTORY i» through his promptitude, his resource, and, above all, by the marvellous way in which he upon occasion could turn his whole army into a flying squadron. The terms of surrender were moderate in the ex- treme. All soldiers belonging to Spain were dismissed at once to their homes, while the rest were marched under the charge of four of Caesar's legions as far as the borders of Italy, where they were disbanded. The generals were somewhat contemptuously allowed to join Pompey in the east. After Caesar's signal success against Afranius and Petreius, the settlement of further Spain gave him little trouble. Varro was a scholar rather than a soldier, and though he had made some preparations, and collected a quantity of stores at Gades, the whole province welcomed Caesar on his arrival with two legions, and Varro himself with his army at once made his submission. Caesar himself had still much work before him, and therefore leaving two legions under Q. Cassius, with orders to join him later at Brundisium, he embarked at Tarraco for Massilia. Here he found that Trebonius had just accomplished the difficult task of reducing Massiha. Caesar himself gives a detailed account of the siege operations, which is of great importance for miUtary history. The siege was not so complicated nor so prolonged as that of Syracuse, so vividly described by Thucydides, nor were the inci- dents so sensational as those in the siege of Plataea, but most of our knowledge of ancient siege warfare comes from these three descriptions. The difficulty of the operations on the land side, lay in the fact that between the lines of circumvallation which Trebonius in due course threw round the city, and the scarped A MILITARY GAME OF CHESS 287 cliff on which the city walls were built, there was a considerable dip in the ground which made it excep- tionally difficult to construct the " agger " from which the battering ram was to be applied to the walls. In this case, the " agger " was a huge embankment, sixty feet wide at the base, thrown right across the valley, between the besiegers' lines and the walls. Caesar's description of the construction of this '* agger," and of the various contrivances against the missiles and catapults of the Massilians, who were famous for their mechanical means of ofience, is very minute. He also describes the way in which a square brick re- doubt, built originally by the besiegers to prevent sorties from the principal gate, was elevated in the face of the enemies' fire and transformed into a tower with six stories, under cover of which the battering ram was to be run up to the wall in a covered gallery. It is clear that Trebonius, as we should expect in a tried officer of Caesar, knew his business well, and when Caesar arrived at Massilia in September, he found that the last sign of resistance in the west was overcome. It is doubtful whether it was at this time or some- what later, that Caesar received news of a disaster to his cause in Africa which was to give him much trouble in the future, though it in no way affected his immediate movements. Curio had found no difficulty in obtaining possession of Sicily, and in expelling Cato, who was loud in his reproaches against Pompey for giving him no support. Curio was in command of four legions, composed largely of the soldiers who had surrendered at Corfinium. In the course of July, acting on orders received from Caesar, he crossed over to Africa with two of his legions. The Pompeian officer in the pro- > ^h ttttl army by the end of ^^^^^^^^^^^^^^^ means of rapid marching to sei2:e some of the coast town nev could reach it. If only uaesar luu r Snded everything in person the f-^^^^ ^^^ succeeded. But when he «-Jf J^^^t"' ih culties presented t—e^ Jhe twelve^^g^^ ^^^^ ^^ r:me lonT«irtr^ Herda. th.ee from M.- siU^! and'two by sea from further Spain. Some of them. MILITARY GAME OF CHESS 293 and especially the two last, were late in arriving, and when they came, the long journey, and then the un- healthy climate of ApuHa had reduced their strength far below their proper numbers. More vexatious still was the fact that in spite of his orders given months before, he found that only enough ships had been collected to carry 15,000 infantry and 500 cavaky. The result was that Caesar not only started several weeks later than he had intended, but was only able to take across seven legions and a few hundred cavalry, a force which, owing to the thinned ranks of the legions, barely exceeded 20,000 men, and even these were without much of their baggage. By great good fortune Caesar escaped the fleet of Bibulus, and landed safely a little south of Ori- cum. Here however his good fortune ended, for the ships on their way back to fetch the rest of the army left with Antony were overtaken by Bibulus, who de- stroyed twenty transports. Caesar himself lost no time in carrying out his programme. First Oricum and then Apollonia were successfully occupied, and Pompey, who was still some way from the coast, when he heard of Caesar's landing, had only just time to throw himself between the advancing army and Dyrrhacium. Pompey encamped on the north bank of the river Apsus, half way between Apollonia and Dyrrhacium, while Caesar fixed his camp on the southern side of the same river. His scheme had so far failed, and his position was not without its dangers. He was virtually in the enemy's country ; he had no ships and was quite uncertain when the rest of his army would join him ; he was dependent for his supphes upon a few friendly states in southern Epirus, and Apollonia, his own base, had neither the advantageous situation of Dyrrhacium, nor the abund- 294 STUDIES IN ROMAN HISTORY ance of provisions and war material with which Pompey's depot was supplied. Pompey, on the other hand, was in a country selected by himself ; he was in communication with a powerful fleet, which might for an indefinite time prevent Caesar from receiving reinforcements; he had Dyrrhacium with all its stores behind him, and he commanded the great road to the East. With these advantages and with an army at least twice as numerous, it seemed that the next move was with Pompey. He might surely bring matters to a conclusion before Caesar's reinforce- ments could arrive. But Pompey knew, if his self-con- stituted staff of senatorial officers did not, the difference between his own untrained and miscellaneous army and the veteran legions, trained by ten years' constant fight- ing. He therefore declined Caesar's offer of battle, and decided to wear his opponent out by a policy of masterly inactivity. The time would come when Caesar must move, and then the blow might be struck. Caesar had no course open but to spend the remainder of the winter under canvas, and to wait for Antony. For about two months he waited, his position in the face of Pompey's 7,000 cavalry becoming more and more difficult. The delay was not the fault of Antony, who laboured under the old difficulty, the want of ships, and when they were procured, he found the sea too closely patrolled by the fleet of Bibulus for a passage to be practicable. He was even for a time himself block- aded at Brundisium. At last Caesar could wait no longer, and sent impera- tive orders for the crossing to be made at all risks. An- tony obeyed, and in accordance with Caesar's instruc- tions, made for the coast well to the north of Dyrrhacium, A MILITARY GAME OF CHESS 295 where there was less chance of the landing being opposed. With judgment and good fortune, Antony got safely across. As he sailed northward, his fleet was visible both to Pompey and Caesar, and there must have been much excitement in both camps. Some Rhodian galleys put out in pursuit from Dyrrhacium, but a storm of wind coming on, whHe it speeded the heavy ships of Antony on their course, wrecked the Ughter vessels of Pompey, and Antony effected a landing at Lissus on the borders of lUyricum. The time for action had come at last both for Pompey and Caesar. Several days' march still separated the latter from the reinforcements he had so long expected, and he was on the wrong side of the river Apsus. Pom- pey had not succeeded in preventing their landing, but it was the next best thing to find himself with a superior army and 7,000 cavaky between the two divisions of Caesar's army. Both camps were broken up, and both Pompey and Caesar started in a northerly direction, Pompey, to cut off and annihilate the weaker division of Antony, Caesar, to anticipate this by a junction with his Ueutenant. In spite of Pompey's superior means of knowing the country, and his nearer position, Caesar's rapidity of movement baffled his antagonist. Marching up the river to find a ford, and traversing the hiUy dis- trict of Scampa, he succeeded in reaching Antony before Pompey with his slower movement could prevent the junction. Caesar's position was now much stronger, his army amounting to over 30,000 men, but the campaign was stiU somewhat hanging fire, and his original design for ending the war by the seizure of Dyrrhacium was ap- parently as far off as ever. Pompey stiU refused to fight, 296 STUDIES IN ROMAN HISTORY and determined not to lose touch with his base of opera- tions, established his camp at Asparagium, south of the river Genusus, and not many miles to the south of Dyr- rhacium, to secure communications with which he threw a bridge over the river. Caesar, still hoping to provoke a battle, encamped not far ofl. Pompey was still determined to play a waiting game, especially as Scipio might now arrive at any time with his Syrian legions. But to remain inactive was the one thing which Caesar could not afford to do. The arrival of Scipio on the scene of action might make his position on the coast untenable, and he accordingly despatched Domitius Calvinus with two legions to Macedonia with the object of intercepting him. At the same time in view of the possible necessity of transferring the war into the interior, he sent two other officers with smaller forces, one to Thessaly, the other towards the Corinthian Gulf, partly to secure supplies, partly to win over fresh allies in those districts. As for Pompey's policy of inactivity, there might be a means of deranging it. The direct road to Dyrrhacium from the south was effectually barred by Pompey's army, but there was a roundabout and difficult route by which it might be possible to reach the city, and to secure the approach to it by way of the Via Egnatia Pompey must have either thought this an impracticable route, or considered that with his bridge across the Genusus he could in any case reach Dyrrhacium first. He had however once more not calculated on the mar- vellous mobility of Caesar's legions. Caesar determined upon a sudden swoop from the mountains down upon Dyrrhacium. If the movement succeeded, Pompey A MILITARY GAME OF CHESS 297 would be cut off from his dep6t, and therefore compelled to fight ; if it failed, he would probably have marched up the Via Egnatia, joined Calvinus, and attacked with overwhehning force the advancing army of Scipio. But Caesar knew the capabilities of his soldiers, and the plan did not fail. Starting as if for a foraging expedition, he marched eastward, and was already well on his way through the mountains before Pompey realized his object. It was again a sheer race between the two armies, and again Caesar won the race, and dropped down from the mountains between Pompey and Dyr- rhacium. Pompey halted, and instead of attempting to force his way through, or risk a battle, he fortified a camp on high ground, at a place called Petra, close to the sea and just above a convenient beach, where ships could easily land. Caesar's object therefore was only half achieved. He was playing a persistently attacking game, but the de- fence if not brilliant was sound, and the defensive posi- tion was one very hard to turn. Pompey, it is true, was cut off from his base by land, and yet by means of his ships he was in daily communication with it. At the same time, for Caesar to attack Dyrrhacium was out of the question with Pompey's army close behind hini. It was again Caesar's move. Had he sufficient pieces to effect the checkmate ? The move he decided upon was perhaps the most astounding in the whole war. It was nothing less than to blockade Pompey's camp with lines of circumvallation from sea to sea. His object in doing this was fourfold. He would cut off the supply of fodder for the horses and beasts of burden : he would deprive Pompey of the advantage of his formidable cavaky, which might otherwise seriously interfere with 298 STUDIES IN ROMAN HISTORY his own commissariat ; he would discredit Pompey's reputation in the eyes of his Oriental allies, and finally when once Pompey was shut in, Caesar might use part of his army in making an attempt upon Dyrrhacium itself. The nature of the ground to the north-east of Pompey's camp made the undertaking less impossible than it would otherwise have been. The country was rugged and mountainous, with a number of peaks or tors which needed little fortification, and which could easily be connected with one another by means of entrenched lines. The important thing, for every reason, was to draw the Hues as tightly round Pompey's camp as possible. In this way time would be saved, labour spared, and the inconvenience of the enemy increased. Pompey's reply to this move was to commence a counter Une of entrenchments of his own, and the longer and more ex- tensive he could make them, the longer Caesar's must obviously become. In this work of entrenchment, Pom- pey had several advantages. He had more men to work, and a shorter Une to construct, and he could always spare detachments to interfere with Caesar's workers. Caesar's original intention had been to run his lines straight to the sea from a strong position a little below Pompey's camp which he had occupied with one of his legions. From this position however after some hard fighting he was dislodged, and the result was that he had to include within his Unes several more heights to the south, before he could run his lines down to the sea. The work of entrenchment therefore turned out to be a far more laborious one than Caesar had at first con- templated, and in the end, his lines constituted an arc of about seventeen miles from sea to sea. Caesar him- A MILITARY GAME OF CHESS 299 self points out the paradoxical nature of the situation. The besieging army was hardly half the size of the be- sieged ; the besieged had abundance of provisions in their camp, while the besiegers were suffering from scarcity of supplies and were even reduced to making bread of a root which grew wild in the mountains. How- ever, as the summer advanced Caesar's position im- proved ; the corn began to ripen, and supplies came in more freely. On the other hand, while Pompey had plenty of provisions for his men, there was scarcity of water, and soon no fodder for the horses. Numbers of them died, and their bodies putrefying in the sun began to cause disease in the camp. There was from time to time some very hard fighting between the lines. On one occasion, when Caesar was himself absent with the object of making an attempt upon Dyrrhacium, Pompey made a determined attempt to break the enemy's Unes at a point due east of his own camp. The fighting lasted four days, but Pompey, though he nearly got through, was in the end driven back with some loss. It was just when the situation had become intolerable that information reached Pompey which decided his next move, and enabled him to disentangle himself from his awkward predicament. Two GalUc brothers, who had deserted from Caesar's camp owing to personal motives, revealed to Pompey the one weak spot in Caesar's Unes. At the extreme southern point, where his Unes reached the sea, there were two Unes of entrench- ment, 200 yards apart. The extremities of these at the sea were to have been connected by means of a trans- verse rampart, to prevent the possibiUty of the Unes being entered from the sea. By some oversight, of which Caesar was possibly not aware, this connexion had not 300 STUDIES IN ROMAN HISTORY A MILITARY GAME OF CHESS 301 been completed, and therefore this was for every reason the point to be 'attacked. Pompey at once acted upon the information. While a strong force marched down the coast, to attack Caesar's inner Une, another party was conveyed down in ships, and landed between the two Hnes, so that the defenders of the inner line were attacked both in front and from behind. They were overpowered and fled, and the arri- val of a few reinforcements only added to the confusion. Antony reached the spot with a legion in time to pre- vent more serious consequences, and Caesar himself was communicated with by means of smoke signals. On his arrival, he saw at a glance that his Une was irre- trievably broken and that the work of weeks was wasted. There was nothing for it but to call in his garrison all along the line, and to collect his army at the southern extremity. But before commencing the retreat which was inevitable, Caesar aimed one blow at his triumphant enemy, which almost fatally miscarried. A disused camp, standing a Uttle apart from the two armies, had been occupied by a party of Pompeians. Caesar sent out two detachments to cut them ofi. Through some mistake, one of these parties lost its way, and got to the wrong side of an earthen wall, which it mistook for the rampart of the camp. By this time, Pompey had sent up a large reinforcement, and it was with the great- est difficulty, and chiefly owing to Pompey's over-caution, that Caesar's force escaped annihilation. As it was, his losses were heavy, and he himself admits that the disaster was nearly fatal. Even apart from this the situation was sufficiently critical. Caesar's immediate object was to withdraw his army, his stores and his wounded to ApoUonia. But the retreat in face of a victorious enemy was no simple matter If Caesar had been Afranius and Pompey Caesar the retreat from Ilerda would doubtless have repeated itself. Caesar's men, after a stirring and tactful address from him, were eager to fight again at once, but Caesar decided that an interval of rest was necessary, and was also anxious about his suppUes. The retreat he effected with consummate skill. He at first sent on his baggage and the wounded, under the charge of one lecion then after an interval, his main army, keepmg however two legions behind in the camp, to lull the sus- picions of Pompey. These too he eventually led out, and as they were hampered with no baggage, they marched with great rapidity. There was of course much rear-guard fighting, and the passage of the river Genusus was not effected without a sharp conflict with Pompey s cavalry On the other side, Caesar occupied his old camp and kept his men within it. The Pompeians similarly occupied their camp at Asparagium, and assum- ing that Caesar was encamped for the night, scattered to get fodder and water, while some returned to fetch some of the baggage which in their haste they had left behind Caesar however at once started off again, and had gained eight miles before Pompey was ready to follow So the march continued for four days, when Pompey gave up the pursuit, and returned to Dyrrha- cium. ^ Caesar had extricated himself with bnUiant success, but he had to decide without delay on his next move. There was really Uttle choice. He was absolutely with- out ships ; even the few he had had at Oricum and those brought over with Antony had been recently destroyed by the younger Pompey, and therefore to return to Italy 303 STUDIES IN ROMAN HISTORY was impossible. On the other hand, he was anxious for the safety of Domitius Calvinus, who with two legions had so far successfully impeded the advance of Scipio, but who now might find himself cut oft between Pompey and Scipio. Caesar had also the best part of two legions in Acarnania and Aetolia under Longinus and Calvisius, while another of his officers, Fufius Calenus, was en- gaged in winning over southern Greece. He therefore decided to pick up Domitius and march his army into Thessaly. He had still two legions in Italy under Corni- ficius, and these were ordered to march towards Mace- donia by way of Ill5n:icum. Then leaving garrisons at Apollonia and Oricum in view of future eventuahties, he marched direct for Thessaly with seven legions. It was again Pompey 's move. He had to a certain extent a choice before him, but Caesar had made his decision with perfect confidence as to what Pompey would do. With his fleet ready at hand, he might convey his army back to Italy and make himself master of the west. This was the course urged upon Pompey by Afranius and others. He however rejected it for several reasons. It would be a mere postponement of the final struggle, for Caesar would certainly follow him, marching over- land by way of Illyricum, and Italy was less favourable to Pompey and more favourable to Caesar than Greece. Besides, it was enough to have avoided Caesar once by seeking a new scene of operations ; to repeat the same manoeuvre was obviously capable of unfavourable in- terpretation. Finally, such a course would involve the abandonment of Scipio and his Syrian army. Another course was to secure the whole seaboard of Epirus by the capture of Apollonia and Oricum. This however A MILITARY GAME OF CHESS 303 would have equally abandoned Scipio, and would have had little effect, as Caesar was now practically independ- ent of any special base. It only remained therefore to follow Caesar into Thessaly. This was the course loudly urged by the senators present in his camp, who consi- dered that it was merely a question of overtaking a fugitive army and striking the final blow. Pompey can hardly have shared this view. He must have been aware that by separating himself from his base at Dyr- rhacium, he was relinquishing what had hitherto been his strongest point, and that he would now be carrying on the war under conditions ahnost exactly smular to those experienced by Caesar. Still, his army was quite twice as large as Caesar's ; there was the chance of cutting off Domitius, and there was the certamty of meeting the Syrian reinforcements. Accordingly, both armies started at the same time, Caesar from ApoUoma, Pompey from Dyrrhacium. Caesar marched south-east, to enter Thessaly over the Pindus range : Pompey marched due east through Candavia, to effect his junction with Scipio. Domitius had a narrow escape. Caesar's messengers had failed to reach him, and he was marching along the Via Egnatia towards Heraclea, at the very same time when Pompey was nearmg the same place from Dyrrhacium. However, he was warned just in time, and striking south, he joined Caesar at Aegimium in Thessaly. Caesar was now practically choosing his own battle ^' It was now July, the fertile plain of Thessaly provided rest and recuperation for his legions. He finally chose a position, south of a little river, the Enipeus, and north of the town of Pharsalus. This was the meetmg point 304 STUDIES IN ROMAN HISTORY of several important roads, and secured him communica- tion with Fufius Calenus in southern Greece. His left was protected by the steep banks of the river, on his right was the open plain stretching to Pharsalus and beyond. Pompey, after joining Scipio at Larissa, es- tablished his camp about four miles further to the east, on rising ground, his right like Caesar's left being pro- tected by the river. Immediately behind his camp, was a lofty hill, situated within a bend of the river. Pom- pey's own design was still not to fight, but to wait and wear Caesar out, using his powerful cavalry to interfere with the enemy's supplies. This was probably sound policy, though it was not clear what was to come of it. But he found it increasingly difficult to resist his sena- torial officers, who regarded a decisive victory as a fore- gone conclusion. They were already securing good houses in the Forum, ready for their return to Rome, and disputing about the consulships, priesthoods and governorships which were to be the rewards of victory. They acted, as Plutarch puts it, as though they had to deal with some petty Armenian prince, rather than with Caesar, the hero of a hundred hard-fought battles. As Pompey's position was too strong to attack, Caesar had to content himself with a series of raids, strengthen- ing his 1,000 cavalry by mixing with them light-armed runners, a plan which he had learnt from the Germans. One day, however, when the signal for starting had already been given, Caesar observed that Pompey's army was at last drawn up in the plain below his camp. The last move in the game was now to be played, and Caesar made his dispositions for the battle. Pompey's tactics were judicious but obvious. With the best part of a legion left to guard his camp, he had about 55.ooo men, A MILITARY GAME OF CHESS 305 exclusive of his 7,000 Oriental cavalry. His right wing, protected by the river, consisted of the two Caesarian legions under Domitius Ahenobarbus. In the centre were the Syrian legions under Scipio. On the left the Cilician legion, and some veterans from the Spanish army. The other legions filled up the gaps between these. Out- side the left wing were the slingers and bowmen and above all the numerous cavalry. Pompey instructed his legionaries to remain stationary and await the attack of Caesar's lines, which having to charge some distance would enter the battle out of breath and in disorder. At the same time the cavalry would swoop round, out- flanking Caesar's right wing, and finally driving home their assault from flank and rear. This cavalry attack was in fact the main danger which Caesar had to meet. His fighting force consisted of eight legions, so reduced in numbers, that they only amounted to 22,000 men, with 1,000 cavalry posted on his right. On his left wing, where the river prevented any outflanking movement, Antony was in command with two weak legions. The rest were divided between Domitius Calvinus in the centre, and L. Sulla on the right. The army was drawn up in three lines, each legion having four out of its ten cohorts in the front rank. Caesar foresaw that his cavalry would be swept away by Pompey's far stronger force, and therefore to guard against the inevitable flanking movement, he detached six cohorts from the third line, and stationed them behind on the right, ready to meet the onset of the cavalry with their spears. It was this precaution which won the day. The battle began with the charge of Caesar's two front lines, the third being ordered to remain stationary. Caesar's veterans knew better than to fulfil the expecta- 3o6 STUDIES IN ROMAN HISTORY tions of Ponipey, and halted a short distance from the enemy, to take breath and dress their ranks. Then the hand-to-hand battle commenced. Meanwhile, Pompey's cavaky had scattered Caesar's weaker troops, but on proceeding to attack the enemy's right wing, were confronted by the serried ranks and formidable spears of Caesar's six cohorts. On these they could make no unpression, and being thrown into confusion, eventually turned and fled. At the same time the third line was ordered to charge, and their onset decided the struggle, hitherto doubtful, between the two main armies. Pompey's whole army was routed, and fled back to the camp, but without attemptmg to defend it, established themselves on the hill behind it. Many of their officers, Uke Domitius Ahenobarbus, were killed. Their general, Pompey, had made his way to Larissa, thence to seek the sea coast, and so next day, the army, without supplies, and cut off from water, accepted terms of capitulation. It was checkmate at last as far as Pompey was con- cerned. Possibly he despaired too soon. If instead of seeking refuge at the court of one of his Oriental allies, where even Cicero could foresee his fate, he had rejoined his fleet in the Hadriatic, he might have continued the struggle in a difterent way, or if he had repaired to Africa, as many of his officers did, Caesar might have had to face the next campaign at once, instead of at his leisure two years later. Throughout the Eastern campaign Pompey played a sound defensive game, his moves were cautious and judicious, and he never lost his head. But he lacked initiative, his methods were slow, and he never knew when to strike. Caesar, with his disregard of times and seasons, his A MILITARY GAME OF CHESS 307 sudden dashes, his record-breaking marches and his ap- parently reckless indifference to any one base of opera- tions, was not playing the game of war as Pompey under- stood it. But it was just these points, baffling to the stereotyped strategy of his opponent, which constituted Caesar's greatness and genius, and which make this cam- paign so unique in its interest. The one point in which Caesar failed, was the attempt to shut in Pompey's far superior force at Petra. Whether this was an error of judgment military critics must decide. It may be argued that Caesar might just as well have taken the course before the blockade which he took after its failure. He had failed to seize Dyrrhacium, but he was in command of the main road into Mace- donia, and if he had marched his army along it towards Scipio and his Syrian legions, Pompey would have been bound to follow, and the final battle might have come earlier than it did. On the other hand, the enterprise, impossible as it seemed, all but succeeded, at any rate it added infinitely to the interest of the campaign as a game of military chess. fjBtnklJIuHJ T T ^.VQ r. JT V >- -v \ > I-. MonvniLsen Roin.. &each.. V. The Rcnrum Hvqbwcujs are marked thus frorn'Maminsen^s Uvstory- of QoTTve" Vol /. bu permission oj^ Messrs Macmillan & Cq Lim ^.X. 188 4. H. rbn RitJiard HcntlcvH Son l«8' r L Oltroann hth . ( 874.06 H222 c. 2 i o tn rn >t h- rvj «>o >0 o o • CNJ <^ CNi r^ (M 00 X I COLUMBIA UNIVERSITY 1999 0026056135 i» •",' '5 ',,"1* . ,.. ;-'' . •Is^i^iPI hi' I