SOURCE: “Arrian as Legate of Cappadocia,” in The English Historical Review, Vol. XLIV, October, 1896, pp. 625-40.

[In the essay below, Pelham surveys Arrian's works, noting ways in which Arrian's military experience informed his writings.]

That Arrian, the historian of Alexander the Great and the disciple of Epictetus, was also for a time governor of the important frontier province of Cappadocia is a fact which, though long known as well established, has received much less attention than it deserves. Yet it is remarkable enough that a Greek philosopher and man of letters should have been entrusted by a Roman emperor with a first-rate military command. It was, indeed, no uncommon thing, in the second century a.d., for Greeks to find admission into the Roman senate, and to be decorated with a consulship. More rarely a distinguished Greek was given some administrative post in a peaceful province, such as Asia.¹ But I know of no other instance, before the third century, in which the command of Roman legions and the defence of a Roman frontier were placed in Greek hands. And the significance of Arrian's appointment becomes greater when it is remembered that it was the doing of Hadrian, the emperor who, though by temperament and policy a lover of Greeks, was of all the Caesars the most solicitous for the efficiency both of the imperial army and of the frontier defences.

The interest attaching to Arrian's legateship does not, however, stop here. What we know of Roman frontier life and of the duties and difficulties of Roman frontier officers is mainly derived from the monuments, for it is a subject on which the literature is provokingly silent. Unless startled into attention by some serious reverse or brilliant success the world of letters knew little and cared less about what was passing in the distant camps and forts where the imperial troops kept constant watch and guard against the outside barbarians. To this prevailing indifference Arrian was naturally an exception, and he has given us from his own pen a unique glimpse of Roman frontier life, and of a Roman frontier force in the first half of the second century of the Christian era, at a period which marked an epoch in the history of the Roman frontier system. Nor is this all; Arrian was not only a trusted officer, but the intimate friend of Hadrian, and in the writings of Arrian the character and policy of Hadrian are reflected almost as clearly as the character and policy of Trajan in the letters of the younger Pliny.

We have no means of knowing how Arrian's appointment to Cappadocia (131 a.d.) was received in official circles, but some explanation of Hadrian's choice is supplied by what is known both of Arrian himself and of Hadrian's policy. The literary materials for a biography of Arrian are somewhat scanty, and have until quite recently been very badly used.² They consist chiefly of incidental notices in Arrian's own writings, in those of his friend and protégé Lucian, and in Dio Cassius, and of the meagre summaries of his career supplied by Photius and Suidas. A few additional facts and some important dates are furnished by inscriptions.

It should be noticed in the first place that Arrian, though by descent a Greek, and of a good family at Nicomedia, in Bithynia, where, as he tells us, he was born and bred, was also a Roman citizen, and that not, as is constantly stated, by the grace of Hadrian, but by birth. Both his nomen, Flavius, and his cognomen, Arrianus, are Roman, and the former proves that the Roman franchise came into his family as the gift not of Hadrian, but of one of the Flavian emperors. Now Domitian, the last of the Flavii, died in 98 a.d., when Arrian, who did not become legate of Cappadocia till 131 a.d., and who was still living in 171 a.d.,³ must have been a mere boy in his father's house at Nicomedia. It must therefore have been Arrian's father who received citizenship, possibly from Vespasian. It is moreover conceivable that Arrian not only inherited the Roman citizenship, but had Roman blood in his veins. His cognomen, Arrianus, certainly suggests the conjecture that his mother was a Roman lady belonging to the gens Arria,⁴ a family famous in the annals of Roman Stoicism. Such a connexion may well have influenced Arrian's philosophic views, and would certainly have been of service to him in his official career. In any case no one can read Arrian without being struck by the unusual combination in him of Roman and Greek. With the versatility, grace, and intellectual keenness of the latter he unites a genuinely Roman sobriety and capacity for affairs.

The boyhood and youth of the future legate were passed at Nicomedia. Like his favourite hero, Xenophon, he was already devoted to hunting, to the art of war, and to study,⁵ a combination of tastes which no doubt aided him in winning the favour of Hadrian. Of hunting, as practised by himself and his companions, and of a favourite hound, he tells us something in his treatise on hunting, which, though rewritten in later life, seems to have been partly at least composed in his younger days. It is possible too that his interest in the adventures and exploits of a famous Mysian brigand, whose history, according to Lucian, he afterwards wrote, may date from this period of his life.⁶ From Nicomedia Arrian passed, after the fashion of his day, to complete his education by a course of philosophic study. In going to Nicopolis and to the lecture room of Epictetus he was not improbably influenced by the traditions of his mother's family. His teacher, who had seen his sect persecuted, and had lived to see it patronised by the Caesars, was the most prominent representative of a Stoicism in which very little remained of the impracticable arrogance and contumacy of the days of Thrasea, and which aimed only, to quote Arrian's description of Epictetus's discourses, at raising men to better things.⁷ How long Arrian remained at Nicopolis it is impossible to say, but long enough certainly to become known as the most devoted and loyal of Epictetus's disciples. It may well have been during this period that he became a familiar figure at Athens, and it was conceivably by the wits of Athens that he was christened the younger Xenophon, a nickname of which he was evidently proud, and which was certainly justified by the strong similarity in tastes and accomplishments which existed between the disciple of Socrates and the disciple of Epictetus.

So far Arrian's career had differed but little from that of other young provincials of good family and fortune. As a Roman citizen, and probably of equestrian rank, he no doubt looked forward to a term of military service, and then to the customary round of municipal duties and honour in his native town. But at some date which cannot be precisely fixed an event occurred which altered his whole prospects in life. His introduction to Hadrian probably took place early in that emperor's reign—certainly several years before Arrian's consulship in 130 a.d. It was the beginning of a close friendship between the two men, and Hadrian's favour opened to Arrian a new career. As a Roman knight he might have risen to high place in the household of Caesar as a procurator or prefect. But the ancient magistracies of the state, and the great provincial commands to which they led, were reserved for men of senatorial rank, the homines laticlavii (‘the men of the broad stripe’). Senatorial dignity was not Arrian's by right of birth, and he must have received it from Hadrian. He may have been granted the ius lati clavi, and thus enabled to offer himself as a candidate for the quaestorship, an office through which the ordinary road lay to a seat in the senate and to the higher honores, or he may have been directly admitted to the senate with quaestorian rank (allectus inter quaestorios). In either case his promotion was assured and seems to have been rapid, though of his official career up to 130 a.d. no record remains. That he held the praetorship may be taken for granted, and he very probably gained useful experience as a legate of a legion and legate of a praetorian province. He obtained the consulship, as consul suffectus, in 130 a.d.,⁸ and in 131 a.d.,⁹ was made legate of the consular province of Cappadocia, a post which he held for at least seven years.

The new legate had unquestionably much in common with his master. Both were enthusiastic sportsmen, and zealous students of military tactics. That Hadrian, in pursuance of his policy of strengthening the defences of the empire, was bent on increasing the efficiency of the frontier troops is well known, and the combination in Arrian of scientific knowledge with practical ability may well have marked him out as a valuable ally in the work. Arrian moreover was a student of philosophy, a scholar, and a connoisseur. Hadrian at least aspired to be all three. But, apart from this congeniality in tastes and pursuits, there may have been weightier reasons for Arrian's appointment. Hadrian aimed above all things at the consolidation of the empire. He was consequently opposed not only to ambitious schemes for its expansion, but to the old-established view of the empire as a federation of allied communities under the leadership of Rome. The differences of race and political status, which the federal theory helped to keep alive, Hadrian did his best to sink in a sense of common citizenship. It was a policy which has often been called cosmopolitan, but which might more properly be described as imperialist. With such aims before him Hadrian would naturally welcome the chance of promoting to high office a man who was in many ways the ideal citizen of a united empire, a man who was Greek by descent, but born a Roman citizen, and probably with Roman blood in his veins.

In a legate of Cappadocia this mixture of the Greek and the Roman was especially appropriate, and indeed few provinces of the empire demanded a greater variety of qualifications in their governor For the Cappadocia of 131 a.d. was by no means the Cappadocia which on its annexation in 16 a.d. had been relegated to the care of an imperial procurator. Arrian's province included not only the ancient kingdom of Archelaus, but also the entire district lying between the northern boundary of that kingdom and the Black Sea and in addition the seaboard eastward from Trapezus to Dioscurias. The area was wide, and the population heterogeneous, comprising as it did the scattered pastoral inhabitants of the Cappadocian uplands, the dwellers in the Greek or half Greek towns of Pontus, and then unruly neighbours the tribes of the hills. Cappadocia, again, in 131 a.d. was a frontier province, with legions and legionary camps, and with a chain of frontier stations garrisoned, as elsewhere, by auxiliary troops. The duty of keeping both camps and garrisons in a high state of efficiency was not the least important of the legate's duties, especially under the rule of Hadrian. But on the Upper Euphrates the care of the frontier required more than the strict discipline and constant vigilance which was as a rule sufficient on such frontiers as those of the Rhine or the Danube, where no more serious danger was to be feared than a marauding raid by some restless, half savage tribe. For in this quarter of the empire the frontier question was political as well as military. It was necessary for the legate of Cappadocia to keep a watchful eye on Rome's great rival, Parthia, to check Parthian intrigue in Armenia, and to take care that none of the smaller potentates beyond the frontier, such as the Iberian king, did anything disrespectful to the majesty of the Roman people, or likely to disturb the Roman peace. Nor was this all. As the chief political officer in the near East he was bound to keep himself and his master informed of the movements of the restless peoples beyond the Caucasus, and even, as the Periplus shows, of the attitude and temper of the tribes and princes bordering the Black Sea. A threatened descent of Alans, or the death of a powerful ruler, such as the king of the Cimmerian Bosporus, was equally an event with which the legate of Cappadocia had to deal. In Arrian's own case both the political and military difficulties of the position had been increased by the unsettling effect produced by Trajan's momentary conquest beyond the Euphrates, and by Hadrian's prompt return to a defensive policy.

Three of Arrian's extant works belong to the period of his legateship—the Periplus of the Euxine Sea, the fragment styled The Expedition against the Alans, and the treatise on Tactics. The Periplus, as the reference to the death of King Cotys proves, was written in 131 a.d.—in the first year, that is, of his command. It is a unique specimen of a report made by a Roman frontier officer to his master, the emperor, and ranks with the letters of Pliny to Trajan as a document of the first importance for the history of provincial administration under the Caesars. The Expedition against the Alans describes the composition and marching order of the expeditionary force led by Arrian against the Alans, in 135 a.d., on the occasion of their invasion of Armenia, when, as Dio tells us, they retired ‘through fear of Flavius Arrianus, governor of Cappadocia.’¹⁰ It stands alone as a contemporary account by a Roman commanding officer of a Roman frontier force. The treatise on Tactics was written, as its author states, in the twentieth year of Hadrian's reign (137-8 a.d.) and the last year of Arrian's legateship. Its chief value consists in the fact that it is an exposition by one of Hadrian's most trusted officers of the cavalry tactics in use at the time on the frontiers, and of the reforms introduced by Hadrian himself. Outside these three important documents we possess only a few isolated references to Arrian's command. Dio mentions the Alan invasion; a rescript of Hadrian, addressed to Arrian, is quoted in the ‘Digest,’¹¹ and a single inscription records a dedication to Hadrian by the city of Sebastopolis ‘during the legateship of Flavius Arrianus.’ The date is 137 a.d.¹² We may, lastly, with some confidence assume that Arrian was the governor of Cappadocia who supplied Lucian with an escort when he went to expose the false prophet Alexander,¹³ a task which would command Arrian's sympathy both as a Roman official and as a man of letters. Lucian speaks of the governor in question as ‘my friend,’ and nothing is more likely than that the brilliant young provincial, whose native place, Samosata, was on the borders of Arrian's own province, should have sought and won the patronage of a Greek scholar, whom Lucian himself describes characteristically as ‘a foremost man among the Romans.’¹⁴

The process of creating a ‘scientific frontier’ along the line of the Upper Euphrates seems to have been gradual. As early as the reign of Nero the imperial government had realised that, for the protection of Eastern Asia Minor, no less than for effective action in Armenia, some nearer base of operations than Syria was needed;¹⁵ and the annexation of Pontus and Lesser Armenia, towards the close of Nero's reign, rendered possible the drawing of a continuous frontier line up to the Black Sea. Vespasian took the important step of permanently stationing a legion on the Upper Euphrates,¹⁶ and the legionary camp at Melitene probably dated from his time. The roads which, according to an inscription, were made under Domitian in Cappadocia, Lesser Armenia, and Pontus,¹⁷ presumably included the frontier road from Samosata northwards to Trapezus, along which Hadrian seems to have travelled in 124, and the line of the road was no doubt guarded by military stations. By the time of the accession of Hadrian a second legionary camp had been formed at Satala,¹⁸ and the entire frontier from Samosata to Trapezus, together with the Euxine coast as far as Dioscurias, had been placed under the command of a legate of consular rank. When, therefore, Hadrian visited the frontier, some seven years before Arrian's appointment, he must have found the frontier system fairly well developed. He travelled along the frontier northwards to Trapezus, and possibly visited also some of the Black Sea stations. Here, as on other frontiers, he reviewed the frontier force, the ‘army of Cappadocia;’¹⁹ and inspected the military stations; existing forts were remodelled and new forts constructed.²⁰ The Periplus, as will be seen, indicates that Hadrian's reforming activity left its mark on the Cappadocian no less than on the German, British, and African frontiers.

Arrian followed in his master's steps, and evidently inaugurated his command by a tour of inspection along the frontier. Of the earlier part of this tour we have no record, for the Periplus begins at the moment when, on nearing Trapezus from the south, he caught sight of the sea from the spot whence ‘both that other Xenophon and you’ viewed it. The results of the entire tour he embodied in an official report to Hadrian, written, as etiquette required, in Latin. He alludes to it as τα ‘Ρωμαικα γραμματα.²¹ The Periplus itself is part of a supplementary report, and deals primarily with the Black Sea stations from Trapezus to the limits of the Roman empire at Sebastopolis (Dioscurias). It contains, in addition, a summary account of the estuaries and harbours along the shores of the Black Sea, and especially along the northern coast, such as would give Hadrian the information necessary to enable him to act with vigour should any crisis in these quarters call for Roman intervention. The Periplus was clearly a less formal document than the Latin report. Arrian is writing in his native Greek, and he writes as a friend to a friend, as one man of letters to another, rather than as a legate to his imperator. Consequently together with reports on forts and harbours we get a good deal of lively chat about the weather, the antiquities, the traditions, and the scenery. Arrian is never long-winded or rhetorical; but his Greek discursiveness and keen curiosity temper pleasantly the military brevity of the Roman officer.

The first place noticed by Arrian is the ancient Greek seaport of Trapezus, the emporium and chief town of eastern Pontus. In Pliny's lists²² it appears as a free town, and that it was still free at the time of Arrian's visit may be inferred from the fact that he makes no reference to the presence of any imperial garrison²³ there, and that the hoplites from Trapezus, who marched with Arrian against the Alans, are classed with the native allies, not with the legions and auxiliaries of the regular army. The importance of Trapezus as a centre for the neighbouring tribes, and as a port where supplies and troops for the frontier could be landed, had evidently been appreciated by Hadrian, for when Arrian visited the place a new harbour was in process of construction by Hadrian's orders;²⁴ till then, it would seem, there had been only an open roadstead. Another memorial of Hadrian's visit is described by Arrian. On the high ground whence the older Xenophon first caught sight of the sea, a statue of Hadrian and altars had been set up. Both were probably connected with the official worship of Caesar. Trapezus still called itself a Greek city, but its Hellenism was clearly somewhat corrupt, for Arrian comments on the barbarous Greek of the inscriptions on the altars, as well as on the inferior workmanship of the statue.²⁵ Life in these remote Greek towns was not altogether peaceful even under Roman rule. In the highlands near Trapezus dwelt a Colchian tribe, the Sauni.

‘To this day,’ writes Arrian, ‘they are most warlike, and live at deadly feud with the people of Trapezus. They dwell in strongholds and have no king, and though long tributary to the Romans they are not punctual in paying their tribute, being engaged in robbery and pillage. Now, however, they shall pay regularly, or, with the help of God, we will root them out.’²⁶

It would seem, however, that here, as in other parts of the empire, these marauding highlanders maintained their ground, for the description given of them by Procopius²⁷ differs little from that just quoted.

On leaving Trapezus—not without sacrificing in the temple of Hermes, and offering a prayer for the welfare of his benefactor Hadrian—Arrian started on a cold and stormy voyage along the coast eastward, touching in turn at each of the military stations, which, as on the coasts of Britain, were situated at the mouths of the rivers, and performed the double duty of guarding the river mouth and protecting the trading settlements against attacks from the hill-sides of the interior. The first station visited was at the mouth of the Hyssus. The garrison consisted of infantry, ‘as you are aware,’ adds Arrian significantly, with twenty mounted spearmen, the latter being apparently intended to assist in foraging for supplies. The infantry were put through their exercises, and even the spearmen were obliged to show that they could use their weapons. It is worth noting, as characteristic of the permanence of the Roman frontier arrangements, that in the ‘Notitia Dignitatum’²⁸ ‘Ysiportus’ was still garrisoned by a regiment of infantry, the ‘Cohors Apuleia Civium Romanorum,’ which may possibly have been the one which Arrian reviewed.²⁹ Proceeding eastward, Arrian and his staff were detained for two days by stress of weather at a lonely little harbour, with a ruinous fort, which bore, however, the name of Athens. ‘We were not permitted,’ he remarks, ‘to sail past even Athens in Pontus as if it were some deserted and nameless harbour.’

The next station of importance was Apsarus, on a river of the same name. Pliny³⁰ mentions a fort there, called ‘in faucibus,’ evidently from its position in a narrow valley. In 131 a.d. Apsarus was a considerable military station, with a garrison of unusual strength. ‘The five cohorts,’ as Arrian calls them, implying that Hadrian knew all about them, as about the infantry at Hyssi portus, constituted a much larger force than was usual in frontier stations. Possibly the reason is to be found in the close proximity of the important valley of the Acampsis, leading, as it did, for a long distance into the interior. Arrian paid the troops and inspected the fort, examining not only the walls and fosse, but the hospital and granary. The results of his inspection he had, unfortunately for us, already given in the Latin report.

From Apsarus Arrian sailed along the coast to the Phasis. As he entered the mouth of the river he saw on his left hand the colossal statue of the Phasian goddess. She was represented seated on a throne supported by lions, and holding a cymbal in her hands. Clearly here, as at Arrian's own city of Nicomedia, and at a hundred other places in Asia Minor, the ‘Great Mother’ was the supreme protecting deity. But it was for relics of Jason and his Argonauts that Arrian most eagerly inquired. None, however, were to be found, with the exception of some fragments of a stone anchor, which might possibly have belonged to the good ship Argo. An iron anchor was also shown to him, but he rightly assigned it to a much later date than the time of Jason. From antiquarian gossip Arrian passes to matters more directly connected with the main object of his journey, and his account of the Roman station on the Phasis is of especial interest. The garrison consisted of 400 ‘picked soldiers.’ These … milites singulares, were a somewhat recent invention, probably dating from Trajan. They were auxiliaries, but differed from the ordinary auxiliary cohort or squadron in being composed of picked men of different nationalities, and possibly selected from different auxiliary corps. A regiment of equites singulares was included in the force which Arrian led against the Alans, but whether the garrison on the Phasis consisted of cavalry or infantry we are not told. The fort is described as occupying a position of great natural strength, and well adapted for the protection of vessels entering the river. It had originally been inclosed in the old way, by an earthen vallum, with wooden towers and probably a wooden palisade. But when Arrian visited the place these old-fashioned defences had been replaced by walls and towers of brick, with solid foundations, and carrying a sufficient equipment of military engines. Such reconstruction was probably common enough at the time, and may safely be connected with Hadrian's visit to the Cappadocian frontier in 124 a.d.

No less interesting, as illustrating a characteristic feature of Roman frontier life, is the account which Arrian gives of the civil settlement which was growing up near this, as near most other frontier stations. The evidence furnished by inscriptions and by the actual remains discovered in Germany, Africa, Britain, and elsewhere has thrown much light on the origin and character of these canabac, some of which developed into important towns and obtained the status of colonies or municipia. Ancient literature, however, says as little about them as about any other institution of Roman frontier life. Arrian's brief notice is, therefore, all the more valuable, and is in complete agreement with our other evidence. The fort on the Phasis was protected by a double fosse. Arrian decided to dig a third, for the better protection of the settlement, which had planted itself between the fort and the river. This settlement, which he describes as consisting partly of retired soldiers and partly of traders,³¹ was probably of comparatively recent date. A more important settlement of the same class was the one near the camp of the 12th legion at Melitene, which it would seem had already attained to the dignity of a municipality.³²

From the mouth of the Phasis Arrian passed on rapidly to Sebastopolis, with only a brief halt at the mouth of the Chobus. The reason for this halt, and ‘what we did there, you will learn,’ he says, ‘from the Latin report.’ Sebastopolis, anciently known as Dioscurias, marked the extreme limit of the Roman empire and of Arrian's province in this direction. Commercially it was important as the centre to which the polyglot tribes of the Caucasus came down for purposes of trade. In the time of the elder Pliny the old Greek town had been deserted, but he implies that a considerable business was done with the natives by Roman merchants.³³ It was probably to the later settlement that the name Sebastopolis was given, and the fort which Pliny mentions and which Arrian inspected may have been built for its protection. Arrian, however, mentions only the fort, which at the time was clearly a regular military station, with walls and fosse, hospital and granary.³⁴ The garrison consisted of cavalry, or possibly mounted infantry. In the ‘Notitia’ the regiment stationed at Sebastopolis was the Cohors I. Claudia equitata. From near Sebastopolis Arrian got a view of the higher peaks of the Caucasus, and one peak in particular was pointed out as that to which Prometheus was bound.

In addition to his brief reports on the military stations along the coast Arrian supplies information, such as a modern Anglo-Indian ‘political’ would be expected to furnish, as to the native chiefs and tribes inhabiting the neighbouring hill country. The settlement of this Hinterland had been, it seems, the work of Trajan and Hadrian, and the majority of the chiefs are described as holding their ‘kingships’ from one or other of these emperors. But the political interests of Rome and her sphere of influence extended here, as elsewhere, beyond the actual frontiers of her empire round the northern shores of the Black Sea, and it was the duty of the legate of Cappadocia to keep his master duly informed of any important event in these remote regions. Such, for instance, was the death of Cotys, king of the Cimmerian Bosporus, of which Arrian heard, while at Sebastopolis.³⁵ Cotys, like his predecessors, held his crown from the emperor, and his successor would have to be, in his turn, formally recognised by Rome. It is improbable that Hadrian contemplated any change in the old-established relations between the Roman government and this prosperous vassal state; nevertheless Arrian thought it proper to supply Hadrian with such a brief sketch of the northern seaboard of the Euxine as might enable the latter to act with effect if action were considered necessary. The sketch is highly business-like, but its monotony is relieved, after Arrian's manner, with antiquarian digressions, and in one place³⁶ by a picturesque account of the Island of Achilles, near the mouths of the Danube, with its ancient temple, tended only by countless sea birds.

The fragment entitled … The Expedition against the Alans shows us the legate of Cappadocia engaged, not in the inspection of the frontier forts and garrisons, but in preparing to meet and repel a threatened invasion of his province. The invaders were the Alans from the plains beyond the Caucasus, who once already in the reign of Vespasian had invaded Armenia, and who now, according to Dio, came at the invitation of the Iberian king Pharasmanes. The same historian tells us that the invasion came to nothing. The Parthian king purchased immunity for his own territories by gifts, while their invasion of Armenia was checked by fear of Flavius Arrianus, the legate of Cappadocia. Arrian seems to have promptly advanced to meet the enemy with a considerable force, and it is with the composition of this force, with its order of march, and with the tactics to be employed when face to face with the enemy, that the Expedition against the Alans is concerned.

The troops which Arrian led against the Alans belonged to the standing army of Cappadocia, the exercitus Cappadocicus, to use the territorial designation which came into fashion under Hadrian. What proportion of the entire army was called out for the expedition is not clear. We learn, however, that of the two available legions one, the 12th, was represented only by a detachment, and the same was the case with some of the auxiliary regiments. The probability is that Arrian, who may be supposed to have been making for northern Armenia, composed his force, as far as possible, of the troops nearest at hand. Hence the presence in full strength of the 15th legion, whose camp at Satala would be a convenient base of operations, while Melitene, the headquarters of the 12th legion, was a long way to the southward. For the same reason, no doubt, the native levies present were those belonging to the northern half of the province, to Lesser Armenia and Pontus. But the strength of the expeditionary force was very considerable. In addition to the 15th legion, and a substantial portion of the 12th, it included the whole or part of eighteen regiments of auxiliaries. Unless therefore the auxiliaries in the Cappadocian army were more numerous than in other frontier armies at the time, very few auxiliary corps can have been entirely unrepresented. The army of Upper Germany, for instance, consisted in 116 a.d. of two legions and nineteen auxiliary regiments, that of Britain in 146 a.d. of three legions and fourteen auxiliary corps. To the legions and auxiliaries must be added the native levies, of whom there were clearly a considerable body. These local troops were not a part of the regular army, but were called out, if necessary, to repel a threatened invasion. They were, with the exception of the hoplites from Trapezus, light troops, and both on the march and when the line of battle was formed were brigaded with a regiment of regulars.

Returning to the auxiliaries as the most characteristic and interesting element in the expeditionary force, we notice at once the predominance of cavalry over infantry, there being twelve cavalry and only five infantry regiments, a predominance which recurs in the army of Africa, whereas in Germany and Britain the proportion is reversed. Both cavalry and infantry were largely composed of archers, as we should expect to be the case in an army on the eastern frontier, and indeed the majority of the auxiliaries belonged to the eastern rather than the western half of the empire. Spain and Gaul were represented by one regiment each, Raetia by two, Dacia by one. On the other hand the African provinces supplied four, and there were also infantry from the Thracian Bosporus and cavalry from Cappadocia, Ituraea, and Arabia Petraea. It is noteworthy that among the officers mentioned are several Greeks. Such were the commander-in-chief himself, Daphnes, of Corinth, commanding the 4th Raetian cohort, Demetrios, commanding a cavalry brigade composed of four regiments, one being the 1st Raetian cohort, and Lamprocles, who commanded the infantry regiment from the Bosporus.

The army of Cappadocia was not of such old standing as some of the other frontier armies, and its definite organisation was probably the work of Trajan and Hadrian. The 12th legion, indeed, had been stationed at Melitene by Vespasian some sixty years before Arrian came out as legate, but it was not before Trajan's time that the 15th was established at Satala. As to the auxiliaries, both the Dacian cavalry and the mounted archers from Arabia Petraca must be subsequent to Trajan's Dacian and eastern campaigns. We know from an inscription³⁷ that the Spanish regiment in Arrian's army, the Ala Auriana, bore the title ‘Ulpia,’ and was therefore either raised by Trajan or served under him with distinction. The 1st Raetian cohort was certainly on the Danube in 108 a.d.,³⁸ and probably followed Trajan to the east. On the other hand the evidence of the ‘Notitia’³⁹ proves that not only the legions but some at least of the auxiliaries of the army of Cappadocia remained there as immovably as did the legions and auxiliaries in Britain. The dux Armeniae had altogether 26 corps under his command, and among them were several which formed part of Arrian's army. The 12th and 15th legions were still at Melitene and Satala; in addition we find still on the Cappadocian frontier the 4th cohort of Raetians, the Spanish ala Auriana, the ala Colonorum, a cohort from the Bosporus, an ala Dacorum, an ala Gallorum, and a cohors Petracorum. Nor is it unlikely, when we remember the permanence of Roman military stations when once the frontier defences had been organised, that the stations assigned to these corps in the ‘Notitia’ were those which they occupied in the time of Arrian. These stations, with one exception, that assigned to the Ala II. Gallorum, lay along the frontier line between the two legionary camps at Satala and Melitene, and are given in the Antonine ‘Itinerary’ as stations on the frontier road which led from Samosata past Melitene and Satala to Trapezus, a road certainly as old as the reign of Hadrian. Two other stations mentioned by Arrian in the Periplus reappear in the ‘Notitia.’ At Hyssi portus the latter places a cohors Apuleia civium Romanorum. In Arrian's time the garrison consisted of infantry, and in the expeditionary force was a regiment of Pεzοι 'Aπλανοι, who were brigaded on the march with the native levies from Trapezus, Rizus, and Colchis, and may well, therefore, have come from Hyssi portus. They were commanded by a Roman officer, and when the line of battle was formed they occupied a position on the left wing corresponding to that of the ‘Italian cohort’ on the right. The other station is Sebastopolis, where the cavalry garrison mentioned by Arrian may possibly have been the cohors I. Claudia equitata of the ‘Notitia.’ …

Both the order of march and the line of battle to be adopted in face of the enemy were evidently prescribed with reference to the peculiar tactics of the Alan horsemen and the danger of a sudden attack in a difficult country. The flanks and rear of the column were protected by light cavalry, while in advance of it rode the … the ‘guides’ or ‘scouts,’ a class of auxiliary troops best known under their Latin name of exploratores. The column was formed of three main divisions, corresponding to the three elements of which the force was composed. In the first were the auxiliary cavalry and infantry, with the mounted archers from Arabia Petraea at their head. In the second, forming the centre of the column, were the legions, preceded by two picked cavalry corps, the equites singulares and the equites legionarii, and by the artillery of the period. In the third division were the native levies and the baggage train, protected, as has been said, by a cavalry regiment from Dacia. The fighting order was clearly intended, in the first place, to resist the charge of the dreaded light horsemen of the east, and secondly to defeat any attempt on the part of the enemy to outflank the Roman force. On the extreme right and left, but on rising ground, were posted the native levies, mostly archers; in front of them, but at a lower level, Arrian placed some of his heavy auxiliary infantry. The local militia were thus protected against the enemy's charge, while able to shoot freely over the heads of the protecting infantry. In the centre were the legions drawn up eight deep, and behind them the rest of the auxiliary infantry. In the rear of the whole line were the cavalry. When the enemy attacked they were to be received with a might shout, and a discharge of missiles from the whole line. If this fire failed to check the charge of the Alans, which it is assumed would be directed against the centre, then, says this general order ‘let the front ranks of the legions prepare to receive the charge kneeling, shoulder to shoulder, with locked shields and levelled pikes.’ If the charge is thus repelled the cavalry are to follow up the retreating enemy, the infantry, after opening their ranks to let the cavalry pass, advancing to support them. At this point Arrian notes and provides for the likelihood that the enemy would rapidly wheel round and endeavour to outflank one of the wings. To meet this manœuvre the wings should be extended as far as possible. Should it, however, succeed, then the cavalry are to wheel round and charge the enemy on the flank, engaging them at close quarters, and using their broadswords and axes with deadly effect upon riders and horses unprotected by defensive armour. Here the fragment ends.

The third of the three of Arrian's extant works belonging to this period of his career has only an indirect bearing on his Cappadocian command. The treatise on ‘Tactics’ was written, as its author tells us, in the 20th year of Hadrian's reign (137-8), and it is closely connected with one part of Hadrian's frontier work, his endeavour to render the frontier troops as efficient as possible in the field. In this work Arrian was keenly interested, as a life-long student of tactics. He had written, he tells us, a treatise on infantry tactics for the benefit and probably at the suggestion of the emperor. The existing treatise is a popular exposition of cavalry tactics, with especial reference to the reforms introduced by Hadrian. Throughout Arrian impresses upon his readers the readiness with which the tactics of all the various peoples included within the empire, and of many outside it, had been adopted by Rome, so that the imperial army anticipated that political fusion which Hadrian of all the Caesars was most anxious to bring about. The treatise is in fact, a good commentary not only on the brief statement of Hadrian's biographer⁴⁰ as to his military reforms, but on the fragments of the speech addressed by Hadrian himself to the army of Africa, when just established in their new headquarters at Lambresis.⁴¹ Of its general drift no better summary can be given than is furnished by Arrian himself in the closing chapter. ‘The king,’ as in the Greek fashion he styles Hadrian,

has obliged his soldiers to practise barbaric movements, both those of the mounted archers of Parthia and the rapid evolutions of Sarmatians and Celts. They have been obliged also to learn the native war cries proper to such movements—those of the Celts, and the Dacians, and the Raeti. They have been trained also to leap their horses across trenches and over ramparts. In a word, in addition to their ancient exercises they have learnt all that has been invented by the king tending to grace or speed, or calculated to strike terror into the enemy, so that the words once applied to ancient Lacedaemon seem to me to apply to this present monarchy which Hadrian now holds for the 20th year, ‘where the strength of the young flourishes, and the clear-voiced muse, and broad justice, the helper of noble deeds.’

The close of Arrian's legateship preceded his master's death by a year at the most.⁴² With his retirement or recall from Cappadocia Arrian's official career seems to have ended, and it is not my business here to follow him in his later literary labours. I hope, however, that I have been able to give some idea of the value of the glimpse which Arrian allows us to get of an eastern frontier province at a critical moment in the history of the imperial frontiers.

Notes

1. Herodes Atticus was for a time corrector civitatum liberarum in the province of Asia.

2. The biography of Arrian in the new edition of Pauly's Real-Encyclopädie contrasts most favourably with all previous ones. It is more complete, and it is free from the blunders which disfigure, for instance, the article in Smith's Dictionary of Greek and Roman Biography.

3. Corp. Inscr. Attic. iii. 1032.

4. In older times such a cognomen would have indicated that Arrian was by birth an Arrius, and by adoption a Flavius. But in the first two centuries a.d. the cognomen frequently indicated the family of the mother, e.g. in the case of the emperor Vespasianus.

5. Cyneget. ii.

6. Lucian, Alexander, i.

7. Dissert. praef. i.

8. The date is fixed by stamps on bricks (Borghesi, iv. 157). The Dict. of Biogr. wrongly places Arrian's consulship after the Cappadocian legateship.

9. Arrian was already legate at the time of the death of Cotys II, king of the Cimmerian Bosporus, which took place in 131 a.d. (see Periplus, 17, and Liebenam, Legaten, p. 124).

10. Dio, 69, 15.

11. Digest, 49, 14, 2.

12. Liebenam, Legaten, p. 124.

13. Lucian, Alexander, 54.

14. Ibid. i.

15. Tacitus, Annals, xiii. 7.

16. Suetonius, Vespas. 8; Joseph. Bell. Jud. 7, 3. 1.

17. C. I. L. 3, 312.

18. See below, p. 636 seq.

19. Cohen, Médailles, ii. p. 153.

20. Dio, 69, 9.

21. Periplus, 6.

22. Pliny, Nat. Hist. vi. 11.

23. At a later period Trapezus was the headquarters of ‘Legio I. Pontica’ (Notit. Dig. Or. 38).

24. Periplus, 16.

25. Ibid. 1.

26. Ibid. 11.

27. Procop. De Aedif. iii. 6.

28. Not. Dig. Or. 381. …

29. Periplus, 11.

30. Procopius, De Aedif. iii. 5, states that under Trajan Melitene attained to πολεωs αξιωμα. He refers not to the camp but to the civil settlement. …

31. C. I. L. 3, Suppl. 6743.

32. Ibid. 3.

33. Not. Dia. Or. 38

34. Vita Hadriani, 10.

35. C. I. L. viii. 2532.

36. Liebenam, Legaten, p. 124.