THE AUGUSTAN EMPIRE (44 B.C.—A.D. 70)

CHAPTER IX

THE EASTERN FRONTIER UNDER AUGUSTUS

I. THE DEFENCE OF EGYPT AND THE MILITARY OCCUPATION OF THE NILE VALLEY

WITH the fall of Alexandria on August 1, 30 b.c., the whole East lay at the feet of Octavian, and one of the immediate tasks confronting him was to provide for its security. Nowhere was the task so easy as in Egypt, which nature had surrounded with ramparts so formidable as to make it the most defensible of all Mediterranean lands. The physical geography of the country dictated the principles of the military policy to be adopted for its defence against external foes and for the maintenance of internal peace and order.

Egypt is a plane of limestone formation which slopes gradually down from the jagged crest of the mountain range that borders the Red Sea to the line marked by the chain of oases which stretches from the latitude of Assuan (Syene) north-westwards to Siwa, the site of the oracle of Ammon. Linked at its north-east corner by the barren peninsula of Sinai to the deserts of Arabia and Syria, this plane is a waste expanse, almost devoid of vegetation save where the Nile, collecting its waters from the highlands of Abyssinia and equatorial Africa, has cut through the middle of it a deep trench, which it has overlaid with fertile alluvium, refertilized by annual inundation. From the first cataract above Assuan, the southern boundary of Egypt proper, the alluvial belt is narrow while the stream passes through a long tract of sandstone and limestone rocks which hem it in; but within about a dozen miles of Thebes (Luxor) it opens out into a broad valley, varying in width from twelve to thirty miles, which runs all the way to Cairo. Below Cairo, at the edge of what was once the coastline, the parting of the waters has expanded the alluvial ribbon into the fan-shaped Deltaic tract that spreads out between Alexandria and Pelusium. The whole sea base of the Delta is occupied by a string of marshy lakes, whose waters filter through sand dunes to the sea, except on the west, where Lake Mareotis is penned by a spur of sandy limestone which provides Alexandria with a dry site and a harbour on an otherwise practically havenless coast. Access to the old harbour, which can now be used only by fishing boats, was difficult in ancient times owing to the narrowness of the entrance and the existence of rocks under the surface of the water; while communication between the city and the interior was pro­vided only by a strip of dry land between the lagoons, a strip much narrower in antiquity than it is today and easy to defend. It was not without justice that the ancients spoke of Alexandria as lying ‘close to’ Egypt rather than in it. Difficult of approach as it was from the Mediterranean, the valley of the Nile was protected on its western flank by the infinite waste of the Libyan Desert and on its eastern by the Arabian Desert—a sterile, calcareous, wadi-furrowed plateau, almost destitute of surface water and clothed with but little vegetation or none at all, which stretches away to the coastal range, whose eastern slopes fall steeply down to the desolate rocky shore of the Red Sea.

Egypt was truly an isolated land, ‘walled about on every side’, in the phrase of Josephus. Only on the south was there no better defined natural boundary than the first cataract of the Nile, and beyond it lay the kingdom of Ethiopia, a loose organization of desert tribes, which, though not powerful, was a potential source of trouble. Excavation has led to the belief that at the time of the Roman conquest Ethiopia was divided, as it had been in the third century b.c., into two kingdoms, a northern round Napata and a southern round Meroe; but it is more likely that there was then only a single kingdom with its capital at Meroe. The establishment of settled relations with this neighbouring State was one of the duties which fell to the first governor, C. Cornelius Gallus, the friend of Virgil.

Before dealing with this matter, Gallus was called upon to crush two local insurrections in Egypt itself. For the natives the Roman conquest merely meant one more change of masters, and they had offered no opposition to it; but the new yoke was to prove heavier than the old, and revolts broke out at Heroonpolis near the eastern edge of the Delta and in the district of Thebes of the Hundred Gates, where the rich and powerful priesthood of Ammon had proved a thorn in the side of earlier rulers. Probably in both cases, certainly in the latter, the outbreak was due to the pressure of taxation, the result either of stricter collection or of the imposition of a fresh burden, possibly the general poll-tax which was levied in the Roman period. The revolts were easily suppressed by Gallus, who vaunted his exploits in a vainglorious trilingual inscription which he erected on the island of Philae, close to Syene, on April 15, 29 b.c. In this document, which shows that his head was turned by vice-regal power, he boasts that he crushed the revolt of the Thebaid within fifteen days, after winning two pitched battles and capturing by storm or siege five towns, three of which were, in fact, parts of the one city of Thebes.

Gallus then proceeded southwards to regulate the frontier. His instructions, no doubt, were to impose ‘the friendship of the Roman people’ on the Ethiopians, or in plain words to establish a protectorate, a favourite but precarious Roman method of covering a frontier. After a military demonstration beyond the first cataract—‘a region which,’ he avers, ‘neither the Roman nor the Egyptian arms had ever reached’—he received at Philae envoys of the Ethiopian king and made a convention whereby the king was admitted to Roman protection and a subordinate tyrannies was appointed to rule the Triakontaschoinos, a district once under Ptolemaic rule, which perhaps extended from the first to the second cataract at Wadi Haifa. The purpose apparently was to create a buffer state under a chief whose dependence on Rome would be a guarantee of his loyalty. This settlement made, Gallus proceeded to indulge his vanity by setting up statues of himself throughout Egypt and inscribing a record of his achievements on the pyramids; he was recalled and disgraced by Augustus, and a deluge of accusations, followed by a decree of the Senate that he should be convicted in the courts and condemned to banishment and loss of property, drove him to suicide (26 b.c.).

The inadequacy of this first frontier policy was proved some four years after its adoption, when a large number of the Roman troops left in Egypt had been withdrawn for an expedition to Arabia under the command of Aelius Gallus. It was a favourable opportunity for a raid, which the Ethiopians were not slow to seize. In 25 b.c. they poured across the frontier, overpowered the garrison of three auxiliary cohorts planted there to defend it, and ravaged Philae, Syene and Elephantine, carrying off statues of Augustus as trophies and enslaving the inhabitants. The governor of Egypt, C. Petronius, soon appeared with a force of not quite 10,000 infantry and 800 cavalry, drove out the invaders, and pursued them into their own land. The Ethiopians retreated up the river to Pselcis (Dakke), where Petronius forced a battle ana easily routed the ill-organized horde, which had no better equipment than shields of ox-hide, axes and pikes; only a few had swords. After taking the town, Petronius advanced southwards and stormed Primis, a stronghold on the steep hill on the east bank of the Nile which is now crowned by the fort of Kasr Ibrim. Thence he pushed on to Napata, the northern capital of the kingdom, where was the queen’s son; the queen herself, a one-eyed lady of masculine character, had retired to a neighbouring fort, whence she sent envoys to negotiate; but Petronius stormed and destroyed the town, from which the prince effected his escape. Deterred by the difficulty of the country ahead, he advanced no farther. Contenting himself with the recovery of the prisoners and the statues of Augustus, he retraced his steps to Primis, which he refortified and garrisoned with a force of 400 men, whom he supplied with provisions to last two years. Then he returned to Alexandria and dispatched a thousand of his prisoners to Augustus, who had recently returned from Spain to Rome, which he reached in the early months of 24. Two years later, in 22, towards the end of the period for which Primis had been provisioned, the Ethiopian queen returned in force to attack the fort, but Petronius succeeded in relieving it and strengthened its fortifications, with the result that the queen gave up the struggle and opened negotiations. Ethiopian envoys were sent to Augustus, whom they found at Samos during the winter of 21—20, and they obtained terms of peace, of which nothing is recorded beyond the fact that they secured what they wanted, including a remission of the tribute that had been imposed on them.

The sequel of the Ethiopian war was the military occupation of the district called the Dodekaschoinos, which extended southwards from Syene as far as Hiera Sykaminos (opposite Maharraka). This district had been granted by the Ptolemies to the temple of Isis at Philae as ‘sacred land’, and it remained the property of the goddess, forming a zone intermediate between Egypt and Ethiopia, not included in the former, but attached for administrative purposes to its most southerly nome and guarded by a string of military posts, which prevented any violation of the southern frontier until the middle of the third century.

The establishment of a southern frontier zone made Egypt immune from invasion on every side. The nomadic inhabitants of the surrounding deserts might, indeed, raid a Nile village or a mining district or a caravan, but they were powerless to do more than produce a local disturbance of the peace. Egypt had thus no frontier region which required to be defended by a limes. The primary tasks of the Egyptian army were to guard the Nile valley against Arab raids and to maintain internal order. Once the country had settled down under the new regime, only a garrison of moderate size would be needed for such duties. Until then Augustus deemed it advisable, in view of the paramount economic importance of the country, to take no risks, and he placed under the command of the Prefect a force more than half the size of that which he probably left to face the Parthians, three legions with their complement of auxiliary troops, amounting to nine cohorts and three cavalry regiments, about 23,000 men in all. In addition to its garrison Egypt was provided with a naval squadron, classis Augusta Alexandrina, which shared with the Syrian fleet the duty of policing the southern coast of the Mediterranean and convoyed the Alexandrian merchant ships which transported Egyptian grain to Italy. Augustus also took over the Ptolemaic institution of a river-patrol, which policed the Nile from Syene to the sea, and placed it perhaps under the orders of the prefect of the fleet, who certainly was in command of it at a later period.

The distribution of the troops at the outset is partially indicated by two brief notices of Strabo. One legion, he tells us, was stationed at the capital Alexandria, and another near the apex of the Delta at Babylon (Old Cairo) on the east bank of the Nile, opposite the great pyramids; of the nine auxiliary cohorts three were stationed at Alexandria, three at Syene, and the other three ‘in the rest of the country’, by which he perhaps meant Middle Egypt; as regards the three alae no precise inference can be drawn from his superlatively vague statement that they were ‘similarly posted at suitable points’. The Alexandrian legion was XXII Deiotariana, a Galatian regiment which had been armed and trained in the Roman style by King Deiotarus and was taken over when Galatia was annexed in 25 b.c.; its fortress lay three miles from the east gate of the capital in the new suburb of Nicopolis, which Augustus founded in commemoration of his victory at the point where Egyptian resistance had been crushed in 30 b.c. The legion which was stationed at Babylon cannot be identified. The quarters of the third legion are not mentioned by Strabo, but there is small room for doubt that they lay in Upper Egypt, at Thebes or Coptos, and that the legion in question was III Cyrenaica. The oldest monuments of this legion connect it with the Thebaid, and although the evidence they furnish is not sufficient to prove the presence of the whole legion there, it is clear that a garrison was needed in this region. Coptos was the nodal point of the most important military and commercial roads in Egypt, and Thebes, with its powerful priesthood and celebrated cult, was a focus of native nationalism and had been a centre of disaffection in the early days of the Roman occupation.

Meagre as this evidence is, it enables us to discern the broad features of the military arrangements made by Augustus. The occupation of Lower Egypt was based on the triangle formed by Alexandria, Babylon, and Pelusium, communication between these three points being secured by river ways along the arms of the Nile, by canals, and by a network of roads not now traceable. Alexandria was the military, as it was the naval, base, and as time went on its importance steadily increased. The strategic value of Babylon lay in the fact that it commanded communications between the Delta and the upper valley of the Nile. Pelusium, ‘the key-position by land’, as Alexandria was by sea, always had a garrison, and along the coastal road leading from it through the sandy desert to Syria forts were always maintained as a defence against Arab raids. Doubtless military posts, held by small detachments of legionary or auxiliary troops, were soon established along the eastern edge of the Delta on the route connecting Pelusium with Babylon and Memphis by way of the Nile canal and Heroonpolis, and on the road leading from Pelusium by Serapeum to Clysma (Colzum) near the head of the Gulf of Suez; and probably the roads from Alexandria to Memphis and towards Cyrenaica were at all times held by detached posts. But on such matters of detail no information has survived for the period of the early Empire. After a time, and in all probability about a.d.7—although the fact is not definitely attested till a.d. 23—conditions in Egypt proved to be tranquil enough to permit a reduction of the garrison, and the legion at Babylon was withdrawn, its place being taken by an auxiliary regiment. The withdrawal of the legion was not accompanied by any appreciable reduction of the auxiliary establishment, so that the Egyptian garrison now numbered about 17,000 men all told.

About the military dispositions adopted by Augustus in the rest of Egypt there is little evidence. The conditions admitted of so little variation of policy that we may presume that, as in later times, garrisons were placed at suitable points in the Nile valley, such as Hermoupolis Magna (Ashmunen), where there was a customs station for goods brought down the river from Upper Egypt. A specially important point was Coptos, the collecting and forwarding centre not only for goods landed at the Red Sea ports of Myos Hormos (probably at Abu Shar) and Berenice (Bender el-Kebir) but also for the products of the numerous mines and quarries in the mountains of the Arabian Desert between the Nile and the Red Sea. This whole region was placed by Augustus under the care of a military officer who bore the title of praefectus Berenices (or Berenicidis or montis Berenicidis). The first recorded praefectus is described in a.d. io—ii as director of all the mines of Egypt. He administered the district, supervising the working of the mines and quarries, with the assistance of a procurator, and commanding the military posts established to provide for the security of the mining centres and of the desert roads leading to them and to the Red Sea ports. To the improvement of the facilities of travel on the trade routes Augustus paid special attention. In the early years of his reign wells were dug and cisterns built; their existence was known to Strabo, whose information about Egypt was gained during his visit to his friend, the prefect Aelius Gallus, and further evidence is furnished by an inscription of Coptos, belonging probably to the later years of Augustus, which records the construction by legionary and auxiliary detachments of cisterns at Myos Hormos and Berenice and at two points on the road connecting the latter with Coptos. These reservoirs were no doubt constructed, as in later times, within a fortified enceinte, guarded by soldiers. It was always necessary to protect these roads against the robber Bedouin, and it may well be that the system of convoys and convoy-dues, which is attested in the reign of Domitian, owed its institution to Augustus.

To the south of Thebes, where the Nile valley becomes steadily narrower, communications with the frontier were secured in later time by the establishment at various points of forts placed opposite each other on either bank of the river, but whether this system was as old as Augustus, cannot be said. About the military occupation of Lower Nubia, which made it unnecessary to increase the strength of the garrison at Syene, no details are known. The establishment of a fort at the extreme southern point, Hiera Sykaminos, may be taken for granted; and it may be presumed that the northern and southern ends of the granite gorge through which the river flows between Tafis (Tafa) and Talmis (Kalab-sheh)—‘the Gate,’ el-Bab, as the Arabs call it—were held by military posts, as well as Pselcis, the scene of Petronius’ victory, where the presence of a legionary cohort is attested in a.d. 27—8.

In Egypt frontier defence was a simple task. The duties of the troops were primarily of a police character. A large number of the soldiers were at all times dispersed over the country, manning the military posts which studded the land from the Mediterranean to Lower Nubia. Those that were stationed at Nicopolis had many police duties to discharge in Alexandria, mounting guard over the corn magazines or the papyrus factories or the mint. From time to time, as in other provinces, detachments of the various units were set to execute works of public utility, such as clearing and deepening the irrigation canals and maintaining and improving the roads. Once the southern frontier had been regulated, there was no actual fighting in Egypt itself for two and a half centuries, and field service was seen only by the detachments which were drafted off now and again to strengthen Roman armies campaigning in other parts of the East. The conditions of service and of recruitment were not calculated to maintain a high standard of military efficiency in the army of Egypt.

II. THE ROMANS AND ARABIA

The Ethiopian invasion of Egypt was prompted, as has been seen, by the withdrawal of about a third of the Egyptian garrison for an expedition which Augustus decided to launch against Arabia Felix under the command of Aelius Gallus, who had been the successor of Cornelius Gallus in the prefecture of Egypt. The decision of Augustus to embark on this aggressive campaign against a country that lay beyond the proper limits of the Roman Empire was a striking exception to his general policy of abstaining from the annexation of territory in the East which was not essential for the security of the frontiers.

Arabia in Roman times was very like what it is today, except that some of the more favoured regions have long since fallen far below the standard of prosperity and civilization which they then enjoyed. The character of the land has not changed. From the Red Sea a strip of coral beach, the Tihama, fringed with reefs and volcanic islands, slopes up to the mountain range of varying altitude which enfolds the inland plateaux—the two sandy, waterless deserts of Nefud and Roba el-Khali, separated by the tableland of Nejd, which is nowhere utterly waste or wholly waterless and has numerous oases. The coastal regions towards the north get some rain in winter, but under the burning summer sun surface water disappears, except in the higher valleys, where there are some oases. The inhabitants have always been nomadic and predatory, save where there is water enough to allow of agriculture and settled life. Farther south, in Yemen and the adjoining region of Hadramut, conditions are more favourable. The coastal range, rising to 9000 or 10,000 feet, is high enough to catch the moisture of the two monsoons and to send copious streams down its slopes, to be finally lost in the torrid strand of the Tihlma on the seaward side and to be stored for the irrigation of the fields on the plateau or to lose themselves in the steppe.

Here in the south-west corner of the peninsula was Arabia Felix par excellence, the El Dorado of the ancients, a happy land by contrast with the surrounding wastes and by reason of its precious products. From the third century b.c. the kingdom of Saba, the Sheba of the Bible, which occupied the southern half of Yemen and reached the sea on the west and south, had been famous in the Mediterranean world as the richest and most powerful state in Arabia. Its political organization was of a feudal type, and its prosperity was based on the cultivation of the soil and, above all, on the production and export of those aromatic substances— frankincense, myrrh, cinnamon and cassia—so highly prized by the ancient world for religious ceremonial and for the preparation of fragrant unguents, perfumes, spices and medicinal ointments. These articles and the Ethiopian (Somaliland) varieties of them, as well as gold, precious stones, pearls and other wares, were transported by the Sabaeans by land and sea to Egypt, Palestine and Syria. The overland route northwards traversed the desert to Leuke Kome, an entrepot in the territory of the Nabataean Arabs, situated most probably at the mouth of the Wadi el-Hamd (south of El-Widj), whence it led through Aelana, at the head of the gulf of Akaba, to Petra; there it forked north-westwards to Gaza and Rhinocolura on the Egyptian border and northwards to Syria, As this route was exposed throughout its entire length to the attacks of Arab tribes, the trade could only be carried on by the native chiefs, who organized the caravans and provided them with military escort. But the land route was not the only artery of communication with the West. The Sabaeans had also a large carrying trade by sea. Not only were their exports partly conveyed by ship to Leuke Kome and to the Egyptian Red Sea ports, but they were the commercial intermediaries between India and the Mediterranean. At the time of the expedition of Gallus the port of Aden, then called Eudaemon Arabia, was a great mart both for the export of Arabian goods to Egypt, Syria, and India and for the import of Indian wares, which were brought by Arabian and Indian merchants and re-shipped partly to Leuke Kome for transport to Syria, partly to Berenice and Myos Hormos, whence they were conveyed by camel to Coptos and thence down the Nile to Alexandria. Egyptian ships had some share in the Red Sea traffic, but it was mostly in Arab hands.

Greek writers have left glowing descriptions of the wealth and luxury of the Sabaeans (that is to say, the Sabaean aristocracy), which, if exaggerated, are not fictitious; they are supported by the existing remains of city walls and towers, temples, palaces, colonnades, sculptures and engineering works, of which the most striking is the great barrage drawn across the valley of the Wadi Dena, the river of Marib, to form a vast reservoir, from which the water was drawn off through sluices into strongly built channels to irrigate the land. Such pride in building and such constructive skill imply a developed civilization and a well-organized form of government with large financial resources; and this is confirmed by the coinage of the Sabaean and Sabaean-Himyarite kings, which from the third century b.c. was modelled on Athenian money, brought across the desert from Gaza, but adopted the Babylonian standard of weight.

Such was the state against which Augustus resolved to launch an expedition. His motives are indicated by Strabo, who owed his information to his intimate friend, Gallus. Encouraged by the promise of the co-operation of the friendly Nabataean Arabs, he proposed to make Arabia Felix a protectorate or a subject State and thereby bring its trade under Roman control, and incidentally to get possession of the proverbial wealth of its inhabitants, who plied a one-sided trade selling their wares at high prices and buying nothing in return. If the country became an annexe of Egypt, both the imperial exchequer and Roman subjects would profit materially by the substitution of maritime transport for the laborious and expensive transport by caravan, by the cheapening of Arabian products, and by the development of direct trade with India, which would be promoted by the control of the Arabian coast and Arabian waters. The potential financial gain may be gauged from the data supplied by Pliny for his own time.

The campaign opened in 25. The force at the disposal of Gallus amounted in all to about 10,000 men, composed of legionary and auxiliary troops, a contingent of 500 men sent by Herod, and one of 1000 furnished by the Nabataean king Obodas under the command of his vizier Syllaeus, who undertook to act as guide and to be responsible for the commissariat. Instead of concentrating his troops at the most southerly harbour of Egypt, Berenice, and transporting them under escort of a small fleet to the south Arabian coast, using his warships to secure his communications with Egypt, Gallus decided to assemble his men at Cleopatris (Arsinoe), close to Suez, and to transport them to Leuke Kome, the southernmost port within Nabataean territory, where he would have 900 miles of desert between him and his objective, the Sabaean capital Mariaba. He began by building 80 fighting ships before he realized that the enemy had no war fleet; then he proceeded to construct 130 transports. On these he embarked his troops, but the shoals and the coral reefs which fringe the coast of the Gulf of Suez and the islands and shores of the Red Sea proved too much for his pilots, and before he reached Leuke Kome after a fourteen days’ voyage, he lost many ships with their crews. No sooner had he arrived than his troops began to be attacked by scurvy and palsy of the legs, diseases endemic in the country owing to the scarcity of good water and vegetable food, and he was forced to spend the rest of the hot summer and the following winter at the port.

In the spring of 24 he started on his march. After many days’ journey through wastes, where water had to be carried on camel back, he reached the land of Aretas, a kinsman of Obodas, who gave him a friendly reception but whose barren country could provide no supplies except coarse grain, dates and butter. A whole month was needed to traverse it, and ahead lay the roadless desert of Ararene, where long detours had to be made to find food and water. A weary march of 50 days across it brought the army at last to the town of Negrana, which lay in the valley of Wadi Nejran, within the borders of Arabia Felix. The king fled, and his city was taken and destroyed. Six days’ further march led to a river, probably the Wadi Kharid, where the Arabs were waiting to offer battle. Their untrained host, equipped with double-edged axes or with bows, spears, swords and slings, was easily dispersed; Strabo gravely reports their losses as 10,000 men against two Roman casualties. This victory was followed by the capture of two other towns, Nasca, the Nashk of Arabian geographers and inscriptions, now el-Baida, and Athrula or Athlula, probably the Iathul or Iathlul of inscriptions, represented by the ruins of Barakish. In the latter town Gallus placed a garrison, and advanced to a city which Augustus and Pliny call Mariba, while Strabo calls it ‘Marsyaba(e), belonging to the tribe of the Rhammanitae, who were subject to Ilasaros’. After six days’ fruitless assault and siege he was forced by lack of water to retire. This city has usually been identified with the Sabaean capital Mariaba, now Marib, some days’ march to the east of Sana, the modern capital of Yemen, but the identification can hardly be correct. Lack of water neither was nor is a feature of Marib, famous in antiquity for its great reservoir and still supplied with excellent water by the Wadi Dena; nor is it credible that, if Gallus had really reached the Sabaean metropolis, the reports of his expedition would all have failed to mention the fact, or that Strabo, who in two previous passages speaks of it under its correct name, would have described it here as a town of an obscure tribe elsewhere unmentioned in Greek literature.

On his retreat from Mariba Gallus took a more direct route northwards through Negrana and across the desert to Egra, an unidentified village on the sea within Nabataean territory, covering the whole distance in sixty days, while his outward march had taken six months. From Egra he shipped the remnant of his army to Myos Hormos, whence he marched to Coptos and descended the Nile to Alexandria.

So ended this abortive expedition, planned in total ignorance of the physical and climatic conditions of the country and of the difficulties it presented to an invader. Its failure was attributed to the treachery of Syllaeus, but the proofs of treachery given by Strabo, who naively reproduces the official version, are mostly proofs of the credulity of himself and his public. While it is impossible to believe that the Nabataeans can have welcomed Roman interference in Arabia or can have genuinely wished for the success of the campaign, which would have seriously affected their profits from the caravan trade, it is plain that without the services of Syllaeus the Roman army could never have traversed the peninsula from north to south. Twenty years later he was tried and executed at Rome, but a belated charge of treachery in Arabia was not, as Strabo implies, one of the counts against him.

Doubtless the campaign of Gallus had a moral effect on people who had never known invasion, and it appears to have led to the establishment of friendly relations between the Sabaean-Himyarite kingdom and Rome. It was not, however, a cause of the rapid development of the maritime traffic of Egypt with India. That development had begun even before Gallus entered Arabia, as Strabo’s evidence shows. While on tour in Egypt during the prefecture of his friend, he learned that, whereas in the later Ptolemaic age not even twenty ships a year ventured to pass the straits of Bab el-Mandeb, now as many as 120 Alexandrian merchant vessels sailed from Myos Hormos to the Somali coast and India, bringing back valuable cargoes to be re-exported from Alexandria and to enrich the exchequer by double duties, for import and for export. In describing the enforced stay of Gallus at Leuke Kome, he observes that, although merchandise from Arabia and India was still conveyed by way of that port to Petra and Rhinocolura, most of it was then landed at Myos Hormos; and speaking of Berenice, he says that ‘all Indian and Arabian goods as well as those from Ethiopia that are brought down by the Arabian Gulf are now conveyed to Coptos’. It is plain that before the Arabian campaign transport by sea was gaining ground at the expense of the caravan trade. This was a natural result of the establishment in Egypt of a strong government which brought the country into close contact with Rome, the greatest market for Oriental products. The Egyptian share in this maritime traffic steadily increased, especially after the discovery of the periodicity of the monsoons and the use to which they could be put enabled mariners to sail direct from the Red Sea straits to northern India. Egyptian competition naturally crippled the trade of the southern Arabs, but it did not paralyse it. The Periplus of the Erythraean Sea, written in all probability towards the middle of the first century, shows that the Arabs still maintained an active traffic not only in the Red Sea but also in Persian and Indian waters, while they still exported Arabian products by way of Leuke Kome to the Mediterranean. But direct intercourse between Egypt and India was bound to cause the steady decline of Aden as an entrepot and the diversion of trade to Egypt was promoted by the measures which Augustus took to improve communications between the Red Sea ports and Coptos. Whether he took any further steps to foster maritime commerce and to protect it against the piracy which infested southern waters, we have no knowledge. There is no certain trace of a Roman fleet in the Red Sea before the reign of Trajan.

The experiences of 25—24 b.c. might have been expected to put an end to the idea of conquering Arabia, the more so as, within a few years of the retreat of Gallus, Augustus publicly adopted, and consistently adhered to, the policy of refraining from fresh conquests in the East. Yet in 1 b.c., when Gaius Caesar was sent on a mission to the East, there figured in his programme an Arabian expedition, which Pliny clearly understood to imply a revival of the old scheme of conquest. He narrates that an Oriental Greek, Dionysius of Charax (the geographer Isidorus was probably meant) was sent to the East by Augustus to put together all necessary information for the use of the crown prince, and that Juba, the learned king of Mauretania, dedicated to him volumes of literary material about Arabia, the fame of which had set his heart ablaze’. It is so difficult to believe that Augustus contemplated such a departure from the policy he had pursued for twenty years that there is something to be said for the view that the real goal of the expedition was to be, not Arabia Felix, but Arabia Petraea. The Nabataeans took their obligations as clients of Rome very lightly. On the death of King Obodas about 9 b.c., Aretas had assumed the kingship without asking the permission of Augustus and nearly lost his throne in consequence. Moreover, strained relations between the Nabataean vizier Syllaeus and King Herod had led the former to abet a revolt against the latter in Trachonitis and to shelter the rebels, who made plundering raids into Judaea and Syria. Herod appealed to the governor and the procurator of Syria, who gave judgment in his favour, but instead of complying with their verdict, Syllaeus set out for Rome. Herod then made reprisals by invading Arabia with the consent of the Roman officials. Syllaeus made capital out of this attack, and for a time he won the ear of Augustus, but in the end he was condemned to death for his misdeeds (among others the murder of an imperial finance officer Fabatus) and was executed shortly before Herod’s death (4 b.c.). The death of Herod did not improve the stability of the situation in this part of the East, and Augustus may have thought it advisable to send Gaius thither to secure the subservience of the Arabs and the maintenance of peace and order in the borderlands of Syria.

III. THE PARTHIAN AND ARMENIAN PROBLEMS: THE PERMANENT FACTORS IN THE SITUATION

In 30 b.c., after settling Egyptian affairs, Augustus proceeded to Syria. The Roman world expected that the conclusion of civil war would be followed by the chastisement, and even the conquest, of Parthia and by the restoration of Roman supremacy in Armenia. Its expectations were disappointed. Augustus left Syria for Asia, where he spent the winter setting in order the affairs of the peninsula, and he then returned to Rome, shelving for the moment the settlement of Eastern questions.

Had Caesar been the victor of Actium, he would certainly not have quitted the East without settling accounts with Parthia and re-establishing Roman authority in Armenia. Augustus was of a different mould. He was well aware that matters could not be left as they were; not merely Roman sentiment but the interests of the Empire demanded the restoration of Roman prestige and the establishment of such relations with Parthia as would ensure the security of the frontier lands. The circumstances of the moment were favourable for vigorous action. Phraates had recovered his throne, but he was not firmly seated on it. Augustus had a large force on the spot. True, many of his soldiers were looking forward to receiving their discharge and the rewards of service, and probably they would not have been in a mood to face the hard­ships of campaigning beyond the Euphrates. But these were hardly the considerations that weighed with Augustus. He had made up his mind to refrain from war in the East, if it could be avoided, and he believed that every vital interest could be secured without it. He judged that there was no immediate danger to be feared from Parthia or from Parthian influence in Armenia, now that the Roman empire was united under a single control. Should Phraates attempt any hostile action he had an instrument in his hands wherewith to threaten his security, in the person of Tiridates, the defeated pretender to the Arsacid throne, whom he permitted to live in Syria. He also held as hostages the brothers of the Armenian king Artaxes, and he created a menace to his western flank by establishing on the throne of Armenia Minor his implacable foe Artavasdes, the fugitive king of Media Atropatene. If danger from Parthia or Armenia might be regarded as negli­gible, the re-organization of the Roman empire was an urgent task which could not wait. That accomplished, he would take steps to secure the satisfaction of Roman honour and the establishment of stable relations with the Eastern monarchy, and he was sagacious enough to foresee that the instability of political conditions in the Arsacid kingdom was likely to offer an opportunity of attaining these objects by diplomatic methods, backed by a display of force. His line of policy was already foreshadowed when during the winter of 30 to 29 b.c. he gave a friendly reception to the envoys of Phraates and indicated that he would lend no support to his defeated rival Tiridates. Ten years were to elapse before a settlement was reached. Meanwhile he left in Syria a garrison of perhaps as many as four or five legions, supplemented no doubt by auxiliary troops. This was the only military force assigned to the Eastern front. Where the legions were quartered, is wholly unknown; but, judging from later arrangements, we may safely presume that none of them was stationed on the Euphrates, which was the natural line of a Parthian invasion.

Public opinion at Rome urged the subjugation of the East as far as India or even, as the poets would have it, as far as China, just as it had hailed with joy Caesar’s projected campaign to avenge the disaster of Carrhae, which it hoped would result in the subjection of Parthia. Some have thought that Caesar’s aim was to recover for the Mediterranean world and for Western civilization the eastern half of Alexander’s empire, where Hellenic culture was being submerged by its Oriental environment. If such a grandiose scheme, wholly divorced from practical possibilities, was really cherished by Caesar, it was certainly never present to the mind of his successor. Augustus took a sane view of Rome’s Eastern question, and if he could have freed himself from the shackles of political tradition, he might have brought about an understanding that would have saved the empire endless trouble and bloodshed without sacrificing any real Roman interest. He clearly realized that what the empire needed was, not expansion beyond the Euphrates, but consolidation and peace. He felt—instinctively perhaps, but none the less truly—that it was an empire based on the Mediterranean, and that that basis was the source of its strength and the condition of its cohesion.

The lands that surround the Mediterranean and the islands that stud it are uniform in climate and character, and they are knit together by the sea, which they dominate from Gibraltar to the Bosporus, as Napoleon saw. The Roman Empire was thus a natural formation, but its proper boundaries were the limits of the Mediterranean belt. Once the Romans had established themselves in the province of Asia, they were bound sooner or later to advance eastwards to the Euphrates; physical geography left no alternative, and the intervening lands were all within the Mediterranean area, permeable by Greek civilization, and consequently capable of absorption into the empire. Farther south, too, the river marked a natural dividing line up to the point where it enters the great desert which stretches from central Mesopotamia through Arabia to Egypt. To overpass that limit was to leave the Mediterranean world behind and to enter the alien domain of Oriental nationality and civilization. It is true that the Euphrates did not mark a sharply defined cultural boundary. From the dawn of history the plains of Mesopotamia had been in close contact with Syria, and far beyond the river Hellenic civilization, disseminated by the conquests of Alexander, still enjoyed living strength and influence; while the border peoples of Asia Minor had been strongly affected by Iranian blood and influences and were hardly touched by Hellenism. Nevertheless the Euphrates was the natural limit of Roman expansion, and conquest beyond it was a false policy, which, if seriously pursued, would have resulted in an overgrown and unstable empire, impossible to weld together and, even if maintainable for a time by exhausting efforts, doomed to fall to pieces under its own unbalanced weight.

The feud between Parthia and Rome owed its origin to the acts and arrangements of Pompey in the East, and it was aggravated by the aggression of Crassus. But there was only one abiding cause of hostile relations, the establishment of Roman suzerainty over Armenia. Save for this apple of discord, there was no reason why the two empires should not have lived side by side in peace and amity. The recognition of the Euphrates as the frontier between them, to which Lucullus had agreed and Pompey also had at first assented, would have ensured peaceful relations. Parthia was anything but an aggressive power. Her organization (which has already been described) was of such a character as to confine her almost entirely to a defensive role. The Arsacid kings never succeeded in welding their empire into a strong and united state. The vassal kingdoms which fringed their provinces (satrapies) were never brought under their effective sovereignty; the ‘King of Kings’ was no more than overlord of his feudatory princes; and overlordship, whatever it precisely implied, did not mean real control. All the Parthian institutions—political, social, and military—were of a feudal type. A landed aristocracy ruled over a population of half-free serfs and of slaves, and the most powerful of these magnates, owners of vast estates, governed the king’s provinces and led his armies. While constant in their allegiance to the principle that only an Arsacid should rule, the Parthian nobles were fickle in their loyalty to the individual king and as ready to welcome a change of sovereigns as to regret the change. Revolution and civil war were frequent, while the king, seated precariously on his throne, sought to safeguard his position by putting potential rivals and enemies out of the way, and his most dreaded foes were those of his own house.

Such internal disunion weakened the power of Parthia and rendered her incapable of a sustained offensive. Another cause of weakness was her military system. Parthia maintained no standing army; a field force consisted almost wholly of levies raised by nobles and landowners from their retainers, who were carefully trained to horsemanship and archery. Such a mobile force could win decisive victories, but it could not be kept together for any great length of time. One reason was the neglect of any system of commissariat, a defect emphasized by Dio as a cause of the in­ability of the Parthians to wage a continuous offensive war. Another was their invincible repugnance to prolonged campaigning even within their own dominions; nothing except an unsuccessful war was more apt to cause a revolution.

The weakness of Parthia’s offensive power and the absence of an aggressive tendency are facts of cardinal importance in judging the wisdom of the traditional Roman policy in regard to Armenia. Having conquered that country, Pompey had added it to the ring of vassal states on which Rome relied for the defence of her provinces in Asia Minor, hoping thereby to make it a counterpoise to Parthia and a barrier between her and Roman spheres of interest. This system of frontier defence was a convenient one for Rome, reluctant as she was to increase her responsibilities by expansion eastwards, and in itself it was sound enough as a provisional measure. In recording the assignment to king Archelaus of Cilicia Tracheia, a region which readily lent itself to brigandage and piracy, Strabo explains that the Romans deemed it better to place it under client-kings than under Roman governors, who would not always be on the spot nor have armed force at their disposal. The principle here stated had a wider application. Not only where districts were unruly but where the native peoples were too backward in civilization to be conveniently incorporated in the empire, the Romans realized that the task of governing and civilizing them was better entrusted to princes born and bred in the country. Their function was to promote the political and economic development of their realms by stimulating the growth of city life and the improvement of agriculture. When that development was sufficiently well advanced, their fiefs could be incorporated as provinces or parts of provinces, and they could be annexed at any moment that seemed good to the imperial govern­ment. Vassal status was by its nature transitional, and, as a rule, it was destined to lead ultimately to absorption.

Pompey’s action in applying this system to Armenia was not unnatural. For geographically Armenia is a continuation of the lofty plateau of Asia Minor. Towards the east of the tableland the mountain ranges which traverse it—apart from the Black Sea chain and the long range of Taurus on the south—converge, as they rise in height, towards the orographical roof of Armenia, the broad mass of Bingeul Dagh (Abus mons), which fills the space between the two arms of the northern Euphrates and, together with its prolongations towards Erzerum on the north and Great Ararat on the east, forms the watershed between the Euphrates and its affluents and the rivers that flow towards the Caspian Sea. Up to this central point the mountain ranges follow an easterly direction, enclosing between them the valleys which give access to the centre of Armenia. Then they trend south-eastwards in a direction parallel to the Caucasus mountains, affording easy communication with lesser Media (Atropatene, now the Persian province of Azerbaijan). While, therefore, Armenia is closely attached to Asia Minor, it is no less closely connected through Media with the Iranian plateau; and it is easily reached from the lowlands of northern Mesopotamia, where the Taurus barrier becomes narrow and sinks so low as to offer no serious obstacle to an army marching towards the Armenian plains.

Thus interposed between two great empires, Armenia was bound to be under the effective influence of one or the other, if it did not actually form an integral part of either. But beyond a doubt its natural connection was with Parthia, not with Rome. Its civilization was unaffected by Hellenism; its people had become completely Iranized, although the basis of their language is held to be Thracian. In manners, customs, and mode of life, in political and military organization, in religion, in dress, and through marriage ties, their affinities were with the Parthians. Here was an insuperable obstacle to the success of the policy of a Roman protectorate. There were other difficulties in the way. The claim to suzerainty was naturally regarded by the Arsacids as an encroachment on their domain and as a standing threat to the security of their realm. It was wholly incompatible with lasting peace between the two empires. Moreover, the lack of cohesion and of the spirit of aggressive militarism in the Parthian empire made the assertion of the claim superfluous; and it was in any case a claim which Rome could not enforce when her actual frontiers were far distant and she had no troops anywhere near to oferawe the vassal king, who might easily be induced by the pressure of circumstances to transfer his allegiance. Even after Actium, Rome had under her direct rule only the outer shell of the peninsula of Asia Minor, and not quite all of that.

IV. THE POLICY OF AUGUSTUS AND THE SETTLEMENT OF 20 b.c.

Such were the difficulties involved in the traditional policy in regard to Armenia. How was Augustus to deal with the problem? If he were dissatisfied with the traditional policy, there were two courses theoretically open to him. He might have boldly cut the knot and abandoned Armenia to Parthia, taking any measures that seemed to him necessary for the defence of the border lands. No vital interest would have been sacrificed. Had there been a strong and militant power established in Iran, the cession of Armenia would indeed have furnished it with an avenue of attack on Asia Minor; but such a power Parthia was not, and could never become without a radical transformation. In itself Armenia was a poor country, which had little attraction for Romans. Though it contained numerous valleys and plains of great fertility, one of the most favoured of which lies round the ancient capital Artaxata, it is predominantly an Alpine land, with a rigorous winter climate and a short summer season, which is hot and dry; it offered no rich field for Roman capital to exploit. While, however, the abandonment of Armenia would have been the best solution of the problem, it was hardly a practicable policy even for a ruler with the prestige of Augustus. A national claim which has become historic is not easily renounced. Roman sentiment and pride would have been offended by the repudiation of a legacy associated with the memories of Republican victories and triumphs, and the de facto Armenian king had to be punished for his massacre of every Roman subject left behind in his realm.

The alternative policy would have been to annex the country, and at a later date Augustus proclaimed to the world that he might have done so. This policy would have involved as a preliminary step the incorporation of all the dependent kingdoms of Asia Minor, most of which were not ripe for annexation. But the declaration of Augustus was merely intended to impress on the Romans the mastery of the situation which he had achieved by bloodless means. That he did not contemplate annexation is shown by the mere fact that he reduced his army at once to a size which was inadequate for the defence of the empire as it stood. The objections to conquests beyond the Euphrates have been already set forth. Geographical conditions would have compelled an advance beyond Armenia to the Caucasus and the Caspian, as Pompey found; and strategic considerations would have required the annexation of northern Mesopotamia as far as the river Chaboras (Khabur) and the range of Jebel Sinjar, which bounds on the north the desert of central Mesopotamia. Such a frontier, even with a well-affected population behind it, could only have been held at the point of the sword; the military and financial burdens of defence could not have been borne; the only result would have been perpetual conflict with Parthia; and the unity of the empire would have been broken by the eastward shifting of the centre of gravity. The policy of annexation could not commend itself to a sober statesman, and having rejected it, Augustus saw no alternative but to follow precedent.

His purpose, as has been said, was to settle the whole Eastern question, if he could, without actual war. Immediately after Actium he had held out a friendly hand to Phraates by declining to abet the refugee pretender Tiridates. From his domicile in Syria the latter made a second attempt to overthrow the Parthian king, and, failing, fled a second time to Augustus in 26—25 B.C. bringing with him the king’s youngest son, Phraates, whom he is said to have kidnapped. Augustus was then in Spain, from which he returned to Rome in the early months of 24. In the previous year he had advanced the Roman frontier in Asia Minor a considerable distance eastwards by annexing the kingdom of Galatia on the death of Amyntas, who lost his life in the course of a vigorously conducted campaign against the robber tribes of the northern front of the Taurus range, which made life and property insecure in the plains of southern Phrygia towards Pisidia ana Isauria. The new province at first included the whole kingdom of Amyntas, but its size was reduced in 20, when eastern Lycaonia together with Cilicia Tracheia, to which it was the key, was transferred to the rule of Archelaus, king of Cappadocia; thereafter it comprised Galatia proper, Pisidia, Isauria and western Lycaonia. By the annexation of this large tract of country the whole interior of the peninsula as far as the borders of Pontus and Cappadocia was brought under direct Roman administration.

No steps, however, were taken to bring the Taurus tribes under effective control, and (so far as is known) no legionary troops were stationed in the new province. As the Parthian question also remained in abeyance, Eastern affairs needed careful watching, and in 23 Augustus sent out Agrippa with secondary proconsular authority and a staff of legates to act as vice-regent in the East, a mission believed by the Roman public to be merely a pretext—not unwelcomed, according to one report, by Agrippa himself—for his temporary removal from Rome. Some months later, if we may take the order of events in Dio’s narrative as chronological, envoys from the Parthian king arrived in Rome with a demand for the surrender of Tiridates and the restoration of the young Phraates. Augustus declined to hand over the former, who might be useful to him as an instrument, but he gave an assurance that he would not lend him support against Parthia, and he sent back the king’s son on condition that the Roman standards and pri­soners of war should be restored.

A year passed, but Phraates gave no sign of complying with the condition, and Roman opinion, reflected by the poets, kept clamouring for vengeance on Parthia. Augustus decided that the time had come to apply pressure, and late in the year 22 he set out on a tour of inspection through Sicily, Greece, and the eastern provinces. From Samos, where he passed the winter of 21—20, he sent instructions to Tiberius, now twenty-one years of age, to bring a large legionary force, drawn no doubt from the armies of Macedonia and Illyricum, overland through Macedonia to Armenia. In that country chronic disunion reigned among the nobles, a weaker section leaning on Rome and a stronger favouring Parthia. The former had sent an embassy to Augustus to prefer charges against king Artaxes and to request his deposition in favour of his younger brother Tigranes, who had lived for ten years in Rome; and the mission of Tiberius was to place him on the throne. In the spring Augustus crossed to the mainland and, after instituting reforms in Asia and Bithynia, proceeded to Syria. His arrival and the news of the approach of Tiberius’ army, travelling with the rapidity with which news has always travelled in the East, had the desired effect on the Parthian king. Faced by the threat of attack from two sides, and knowing that a Roman invasion was likely to cost him his throne, he resolved to yield and, risking the displeasure of his own people, he restored the standards and such of the prisoners as still survived and did not seek to elude discovery. On May 12, in all probability, they were handed over by his representatives, and Augustus was acclaimed Imperator for the ninth time.

It was a notable diplomatic success, which Augustus ranked higher than a victory in the field. ‘I compelled the Parthians,’ he proudly wrote, ‘to restore the spoils and standards of three Roman armies and as suppliants to implore the friendship of the Roman People’. The submission of Parthia was sung by the poets with tedious iteration and immortalized by artists and by the Roman mints. On the cuirass of the fine statue of Augustus found in the villa of Livia at Prima Porta a relief depicting the surrender occupies the central place. In 18 b.c. coins were issued with the legend Caesar Augustus signis receptis inscribed round the figure of a Parthian on bended knee, proffering a standard with his right hand and holding out his left in an attitude of supplication. In honour of Augustus the Senate decreed a triumph, which he declined, and a triumphal arch, which was erected in the Forum next to the temple of the deified Juliusa, nd is figured on coins of 18—17 b.c. bearing the legend Civibus et signis militaribus) a Parthis recuperatis. The standards were taken to Rome to be dedicated to Mars Ultor. They were placed temporarily in a small round temple which Augustus ordered to be built on the Capitol on the model of that of Juppiter Feretrius, and they were subsequently transferred to the splendid new temple of the god in the Forum of Augustus which was dedicated on August 1, 2 b.c. The memory of the event was kept alive till the fourth century by the annual celebration of a thanksgiving and of Circensian games on May 12.

The settlement with Parthia made the task of Tiberius easy. The threat of invasion gave the pro-Roman party the upper hand, and before he reached Armenia, Artaxes had been murdered by his own kinsmen. Tiberius entered the country without opposition and in the presence of his legions solemnly placed the diadem on the head of Tigranes. Augustus announced to the world that he had conquered Armenia but that he refrained from annexing it, preferring to follow established precedent and hand it over to a client-king. The claim to conquest was stressed by contemporary writers and by the Roman mints, which issued coins bearing the legend Armenia capta (or recepta) and displaying the figure of an Armenian on bended knee, extending his hands in an attitude of surrender, or other symbolic representations of conquest. The restoration of Roman authority in Armenia was followed by a request from the people of Media Atropatene, now freed from the rule of Artaxes, that Augustus should choose them a king, and he appointed Ariobarzanes, son of their former king Artavasdes, who had died in Armenia Minor.

Such was Augustus’ solution of the Eastern question. Peace was to be maintained with Parthia on condition of her recognition of Roman suzerainty over Armenia. Despite the applause which greeted his successes, the renunciation of all idea of conquest in the East was not what Romans had expected, and Augustus thought it well to justify his policy. This he did in a communication to the Senate, in which he declared that he regarded any further extension of the empire as undesirable.

The submission of Phraates was no doubt resented by his subjects and was not calculated to improve the security of his position. It has been supposed, on slender evidence, that he was once more driven from his throne. Anyhow, about 10 or 9 b.c., he took the extraordinary step of handing over to M. Titius, the governor of Syria, his four legitimate sons—Vonones, Seraspadanes, Rhodaspes and Phraates—with their families to be permanently domiciled in Rome, where they were provided for in royal style at the public expense. His motives were variously interpreted. The Roman official explanation was that they were sent as pledges of his friendship. A sounder explanation is given by Strabo: Phraates was afraid of revolution, and as he knew that no rebel could prevail unless he were allied with an Arsacid, he removed his sons to deprive evil-doers of that hope. A contribu­tory motive is stated by Josephus. Phraates had an illegitimate son Phraataces, by an Italian slave girl sent him by Augustus, whom he afterwards made his legitimate queen under the name of Thea Urania Musa, and she persuaded him to remove his legiti­mate offspring in order to secure the succession for her son, who did in fact succeed. The sons were evidently content to go to Rome, as they made no attempt to resist or to escape, and their willingness indicates that they were as afraid of their father as he was of them. Fortune could have placed no more valuable gift in the hands of Augustus than this supply of Arsacid princes, who might be used to undermine the security of a hostile Parthian king, and were so to be used by his successors. Augustus was gratified by this crowning success of his Parthian policy, and did not fail to take an opportunity of impressing it on the people of Rome by showing off his hostages at a gladiatorial show.

V. THE BOSPORAN KINGDOM

Three years after his return from the East in 19 Augustus decided to send Agrippa thither again with plenipotentiary powers to superintend the government of the provinces and the vassal states and to deal with all matters that required settlement. One of the tasks that fell to him was the regulation of affairs in the Bosporan kingdom. The effective control of this outpost of civilization was a matter of importance for Rome. Alike for political and economic reasons it was imperative that its government should be in capable and loyal hands. Flanked on the west by the Scythians of the Crimea, and on the north and east by Sarmatian tribes strung out over the South Russian steppes and extending westwards as far as the Dniester, it was the only barrier against the establishment of an Iranian domination of the northern shores of the Euxine. It was even possible that such a development might be promoted from within the Bosporan king­dom by the emergence of a ruler inspired by Mithridatic ambitions. The government of the State was purely monarchical, the Greek population had no voice in it; its rulers were of Iranian or semi-Iranian lineage, and its Greek subjects were no longer of pure Greek blood, while the nearer Sarmatian tribes had become partially hellenized through intensive commercial intercourse with their neighbours. Such conditions might throw up an ambitious prince bent on casting off the yoke of Rome and uniting the adjoining tribes into a formidable Graeco-Iranian state on the northern shores of the Black Sea.

There were other reasons for keeping a vigilant eye on the Bosporan kingdom. Its economic importance was great. It was still the main source of food supplies for the cities of northern Asia Minor and of the Aegean; and when Roman troops were required in the northern section of the eastern frontier, as they had been required in 20 b.c., their provisionment depended chiefly on the produce of South Russia. Moreover, the Black Sea had to be policed not only in the interests of indispensable trade but also for the protection of the coastlands within and beyond the limits of actual Roman territory; and for this work the co-operation of an efficient Bosporan government was essential. The sea was infested by pirates. The Tauri of the south coast of the Crimea were notorious wreckers and freebooters, but they were not so formid­able as the wild tribes inhabiting the steep wooded slopes of the Caucasus, which made their living by piracy and slave raids. The Caucasian pirates, says Strabo, were masters of the sea; and he adds that, while native rulers took punitive measures against them when their subjects suffered, Roman territory was attacked with greater impunity through the negligence of the governors. Until the imperial government made up its mind to take over the task of policing the waters of the Euxine, it had no alternative but to impose the dutyon the client-kings of Bosporus and Pontus. There was plainly no lack of reasons why Rome should see to it that the government of the Bosporan kingdom was vested in strong and trustworthy hands.

Since Caesar’s death the throne had been held by Asander, probably one of the half-Greek citizens of Panticapaeum (Kertch), who had overthrown Pharnaces and secured the support of the people through his marriage with the late king’s daughter Dyna-mis (which may, however, have taken place before her father’s death). After crushing Caesar’s nominee, Mithridates of Pergamum, and ruling three years as archon, he had succeeded, perhaps by a bribe, in obtaining the recognition of Antony and the title of King, which appears on his coins from 41 b.c. Either then or at a later date he was also recognized by Octavian and enrolled among the amici populi Romani. He evidently proved a capable ruler, maintaining his territory against Scythian and Sarmatian invasion and keeping piracy in hand. At the time of Augustus’ sojourn in the East he was still on the throne, but shortly after a revolt was raised against him by an adventurer named Scribonius, who claimed to be a grandson of the great Mithridates; and it is said that, when he saw his troops beginning to go over to the usurper, he starved himself to death. He died in or just before 17 b.c. at the great age of 93. Scribonius alleged that he had been chosen by Augustus as Asander’s successor, and won the hand of the unscrupulous queen, who may be suspected of complicity in the plot; but she kept the sovereign power in her own hands, for she struck gold coins (of which only one, dated 17—16, survives) with her own head and name alone. Who Scribonius was, is unknown; he may well have been a provincial of good social standing—possibly even, like Mithridates of Pergamum, of royal descent—who took a Roman name; a low-class adventurer would not have been accepted by Dynamis as a husband.

When the news of these events reached Agrippa in Syria, he naturally declined to recognize the impostor and commissioned Polemo, king of Pontus, to attack him, promising him the throne and arranging, with the sanction of Augustus, that he should marry Dynamis; thereby his position would be legitimized, the acquiescence of the Bosporans would (it was hoped) be secured, and the reunion of the two parts of the old Mithridatic empire under the rule of an energetic and reliable vassal would obviate possible dangers and ensure stable conditions on the north-eastern fringe of the Empire. Before Polemo reached the Bosporus, Scribonius had been put to death by the Bosporans themselves, in the hope that his removal would save them from the rule of Polemo. Disappointed in their expectation, they offered resistance to the invader and, in spite of defeats, persevered in the struggle until in 14 b.c. Agrippa sailed with a fleet to Sinope and threatened to take the field against them. The threat sufficed. The Bosporans submitted, and Agrippa proceeded to order the affairs of the kingdom, imposing on it the obligation to supply contingents to the regular auxiliary forces of the imperial army, and arranging that the isolated city of Chersonesus, near Sebastopol, while retaining the autonomy which Augustus had apparently restored to it in 25, should place itself under the aegis of the Bosporan kingdom by the conclusion of an alliance which should secure it against the attacks of its Scythian neighbours. The kingdom was then handed over to Polemo, who duly wedded Dynamis. The importance which Augustus attached to the settlement of Bosporan affairs is shown by the honours decreed to Agrippa, among them a triumph, which in accordance with his fixed practice he declined.

The marriage of Polemo and Dynamis proved unhappy. Both were masterful personalities, and each was determined to rule; they separated after little more than a year, and Polemo then took to wife Pythodoris, daughter of a rich citizen of Tralles in the Maeander valley, by whom he had three children. Dynamis, as would seem most probable, fled across the straits and took refuge with a Sarmatian (or possibly Maeotian) tribe ruled by one Aspurgus, son of King Asandrochus, whose active support she secured by giving him her hand and with it the prospect of the Bosporan throne. With his help she organized a revolt against Polemo, which dragged on for several years. Despite various successes, among which is perhaps to be included the capture and destruction of the Greek colony of Tanais at the mouth of the Don, Polemo failed to crush the revolt and in 8 b.c. he had recourse to the stratagem of attacking the rebels under the guise of friendship, but the ruse was discovered, and he was captured and put to death. The tribesmen who formed the following of Aspurgus and styled themselves ‘Aspurgians’ were rewarded by being settled on the rich lands of the Taman peninsula between Phanagoreia and Gorgippia; the possession of these lands, which henceforth bore their name, ensured their support for the new régime.

Dynamis had won, but her position was critical; she had overthrown the nominee of Augustus, and the last word was with him. At this time he had many preoccupations: his hands were full in the West; in southern Asia Minor the subjugation of the predatory mountaineers of the Pisidian Taurus was perhaps in progress; and trouble may have already broken out in Armenia. No doubt he also realized the attachment of the Bosporans to the Mithridatic dynasty and their aversion from the Polemonian house, and he accepted the situation, recognizing the rule of the queen on condition that she should accept the full obligations of a Roman vassal. The condition was loyally observed. Dynamis received the title of ‘friend of the Roman People’, which regularly appears on her monuments. Her head, name, and title now disappeared from her coins and were replaced by the heads of Augustus and Agrippa, with nothing more than a humble monogram to indicate that she was still ruler; and for the rest of the Julio-Claudian period the Bosporan kings, with one significant exception, placed on their gold coins only monograms and dates, never their full name and title. She died in a.d. 7/8 about the age of seventy.

During her lifetime her husband Aspurgus was not officially associated with her in the government of the kingdom, nor did he immediately succeed her. For two years the throne was held by a ruler whose name is unknown. Then Aspurgus acceded to power (a.d. 19/11), buthe did not bear the royal title till a.d. 14/15, when it was conferred on him by Tiberius together with that of amicus Caesaris-populique Romani. At the same time he received Roman citizenship and adopted the names of his benefactor, Tiberius Iulius, which were borne by his son Cotys and all his successors for centuries. Aspurgus appears to have been a strong ruler as well as a loyal vassal; he maintained Bosporan territory intact as far as Tanais and is extolled by one of his officials as having ‘subdued’ the Scythians and the Tauri. He died soon after Tiberius in a.d. 37/8, leaving two sons, Mithridates and Cotys, the former borne to him perhaps by Dynamis, the latter by his second wife, a Thracian princess Gepaepyris.

VI.THE HOMANADENSIAN WAR

Not many years after Agrippa had completed his mission in the East military operations on a considerable scale were required to establish peaceful conditions in the southern borderlands of the province of Galatia. Between the province and the Levant stretched a broad belt of very wild country forming part of the great Taurus range, which walls off the central tableland from the sea. Lofty mountain ridges running athwart the general trend of the range from west to east and enclosing rough, rocky, rolling plateaux, seamed with precipitous ravines and deep cafions which have been cut by the rivers that drain to the Pamphylian and Cilician seas, make the whole belt exceedingly difficult to traverse and furnished a terrain ideally suited to be a nursery of brigands and pirates. The tribes that occupied the northern front of this mountainous tract—Pisidians on the west and, adjacent to them, two tribes racially akin to their southern neighbours in Cilicia Tracheia, the Homanades in and round the valley of lake Trogitis and the Isaurians in the Lycaonian hills—were the scourge of the fertile countryside which Strabo calls ‘Phrygia adjoining Pisidia’. Amyntas, the last king of Galatia, had made a gallant effort to reduce those lawless mountaineers and had achieved considerable success. After capturing Pisidian strongholds, among them Cremna, he proceeded to attack the Homanades and succeeded in storming most of their fastnesses, reputed to be almost impregnable, and in slaying their chief; but he fell into an ambuscade laid by the dead man’s wife and lost his life.

Although he left sons behind, Augustus decided to annex the kingdom (25 b.c.), whether because he thought that the interests of the empire demanded its incorporation or because for one reason or another he deemed none of the sons suitable for the throne. For many years, however, he took no step to carry through the work of pacification which Amyntas had begun; he confined himself to the defensive measure of founding at Antioch, perhaps in 19, under the name of Colonia Caesarea, a garrison colony of soldiers discharged from two legions, V Gallica and VII (Macedonica), which was designed to aid in checking the raids of the mountain tribes. To ensure the effective protection of territory which it has annexed is the first duty of any government, and the delay in taking decisive action is hardly to be explained by the consideration that much more urgent tasks were awaiting accomplishment or that the coercion of the Homanades, which Amyntas had nearly achieved, was so serious an undertaking that it had to be postponed until the reorganization of the empire was complete. It is more likely that the establishment of Roman administration had produced a lull in the marauding activities of the tribe; otherwise the inaction of Agrippa between 16 and 13 would not be intelligible.

If quiescent for a time, the Homanades must have caused serious trouble at some date between 12 b.c. and a.d. 1, and Augustus resolved to deal with them. He entrusted the task to P. Sulpicius Quirinius, a man of humble birth who had forced his way up by his military talents. Among his services was the successful conduct of a campaign which he had been commissioned, no doubt as proconsul of Crete and Cyrene, to wage against two tribes of the Libyan desert, the Marmaridae and the Garamantes, which had menaced the security of the Cyrenaic cities. As a reward he was given one of the two eponymous consulships for 12, an honour which about that time was reserved for members of the highest nobility, and some time thereafter he was chosen to carry out the subjugation of the Taurus region. The main attack on the offending tribe could be made only from the north, and in all probability Quirinius was appointed governor of Galatia and Pamphylia, which appear to have formed a single governorship in the time of Augustus. The inclusion of Pamphylia in his command gave him control of both sides of the mountain belt, the only way (as the Romans were well aware) of ensuring the subjugation of mountain tribes. About the forces employed there is no evidence, but it is probable that they included two or three legions, drawn from Syria and perhaps Egypt. Nor is anything recorded about the course of the operations beyond the fact that Quirinius resorted to the method of blockading the mountain strongholds of the tribe, forty-four in number (which implies that he had a large force under his command), and starved them into submission, capturing 4000 men. He doubtless began by occupying the fertile but marshy valley of lake Trogitis, in which lay the chief tribal centre, Homana, and driving the tribesmen back on their hill forts, which he stormed when famine had exhausted their power of resistance. His prisoners, who represented the total number of the surviving adult tribesmen, were removed from the mountains to the plain and settled in the neighbouring cities. In recognition of his successful accomplishment of a difficult task Quirinius was awarded the triumphal insignia, and the colony of Antioch expressed its gratitude to him by electing him duumvir, an office in which he was represented by a deputy who was a distinguished citizen of the town.

A war of subjugation was usually followed by a reorganization of the country concerned, and this may have been the date of the establishment of five colonies of veteran soldiers which Augustus planted in the region south of Antioch to assist in the maintenance of law and order, three of them in Pisidia (Olbasa, Comama and Cremna) and two in Lycaonia (Parlais and Lystra), all called Iulia Augusta (or Iulia) with added epithets. They were connected with the older colonia and administrative centre Antioch by a system of roads, officially designated by the hybrid name of viae Sebastae, which were in process of construction in 6 b.c. during the governorship of Cornutus Aquila. The subjugation of the Homanades was complete and final, but the severity of their treatment did not suffice to cow all the tribes of the mountain area. In a.d. 6 a fresh outbreak of brigandage on the part of the neighbouring Isaurians developed into a regular war, of which no details have survived. Thereafter peace seems to have reigned in the Galatian borderlands, and its hill tribes settled down to assimilate Graeco-Roman culture; but their neighbours in Cilicia Tracheia, removed from civilizing influences, remained untamed barbarians, ready to descend from their mountains on plundering raids and destined in later centuries, under the name of Isaurians, to emerge as a militant force and even to give an emperor to Constantinople.

VII.THE REVOLT OF ARMENIA AND THE MISSION OF GAIUS CAESAR

With the results of his policy in Armenia Augustus had small reason to be satisfied. His nominee Tigranes II died, after a brief term of rule, at an uncertain date not later than 6 b.c., and then, in the words of Augustus, Armenia ‘revolted and rebelled’. The anti-Roman faction placed on the throne his son Tigranes III and his daughter (perhaps by a different wife) Erato, who were joined in wedlock in eastern fashion. Augustus commissioned Tiberius to proceed to Armenia, and when he declined the commission and retired to Rhodes, he ordered the installation of Artavasdes, probably a younger brother of the late king, with whom he had been taken to Rome. He was supported by Roman troops, but his reign was short. Some time before 1 b.c. he was driven out by Parthian aid, together with the Roman troops sent to support him, and Tigranes regained the throne. It was a serious blow to Roman prestige, and the vexation of Augustus is reflected in his silence about this unfortunate nominee. Being deprived of the co-operation of Tiberius and being himself too old to under­take another journey to the East, he was forced to send the youthful Gaius Caesar to deal with the situation, causing to be conferred on him the secondary proconsular power, with authority overriding that of provincial governors, which had formerly been wielded by his father Agrippa, and providing him with an advisory staff, headed by M. Lollius, who had acquired some knowledge of the East as organizer of the province of Galatia. This frontier province had in recent years crept a little farther eastwards; in 6 b.c. all inland Paphlagonia had been added to it on the death of king Deiotarus Philadelphus, and in 3—2 b.c. the adjoining principalities of Amasia and Sebastopolis in inland Pontus, with the city of Megalopolis-Sebasteia (Sivas)—an important nodal point in the system of roads, which hitherto had belonged to the Polemonian kingdom of Pontus—were incorporated in the pro­vince under the distinguishing title of Pontus Galaticus. These districts carried the main road from the Bosporus towards Armenia, which was afterwards to become a great trunk-road connecting the Armenian frontier with Moesia. Roman territory was thus advanced a considerable way towards the Euphrates, though between the two there still intervened three vassal states, Polemonian Pontus now under the rule of Polemo’s widow, the able queen Pythodoris, Armenia Minor, and the realm of Archelaus, king of Cappadocia.

All preparations completed, Gaius travelled by way of Athens across the Aegean. At Samos (or Chios) he met his stepfather Tiberius, who had come from Rhodes to pay his respects but was accorded a cold reception, thanks to the hostile influence of Lollius. From Samos he went out of his way to Egypt, perhaps merely to gain personal knowledge of the most important economic dependency of Rome, just as he had previously been sent round European provinces. There is no clear trace of any activity on his part in the country, and little probability in the view that the object of his visit was to set on foot preparations in the Red Sea ports for the Arabian expedition which was believed to be the intended climax of his triumphal progress. From Egypt he sailed direct to Syria, where he entered on his consulship for a.d. 1.

Two years earlier (2 b.c.) there had been a dynastic revolution in Parthia, where the old king Phraates had been murdered and succeeded by his son Phraataces. The servile descent of the new king made him unacceptable to the Parthian nobility, and his policy of actively abetting the revolution in Armenia was probably adopted in the hope of strengthening his position. On hearing of the mission of Gaius, he had sought to negotiate with Augustus, proffering an explanation of what had occurred and requesting, as a condition of peace, the return of his four half-brothers domiciled in Rome, who were a potential menace to his security. Augustus naturally declined to part with those convenient weapons of offence, and brusquely branding him as a usurper, ordered him to withdraw from Armenia. Phraataces sent a haughty reply, and for the time Armenia remained under Parthian control; but its de facto king Tigranes, after the death of his expelled rival Artavasdes, dispatched envoys to Rome with gifts and a humble petition for the Armenian throne. Satisfied, apparently, with this acknowledgment of Roman suzerainty, Augustus accepted the gifts and bade Tigranes go with good hope to Gaius in Syria.

Some time after the arrival of the crown prince in the province, Phraataces, believing that resolute action was portended and apprehensive of the disaffection of his own subjects, changed his mind and determined to come to terms with Rome. He receded from his demand for the return of his brothers and declared his readiness to refrain from interference in Armenia. On these terms a concordat was concluded, and in the following spring (as would appear) a personal interview between the Roman prince and the Great King took place on the neutral soil of an island in the river Euphrates, which was thereby definitely recognized as the boundary between the two states. Velleius Paterculus, then a young legionary tribune, describes the scene which he witnessed. The Roman and Parthian armies were drawn up on either bank and the representatives of East and West landed on the island with an escort of equal size. An exchange of banquets followed, the Parthian yielding precedence to the Roman and dining first on the western bank, while Gaius crossed to the Parthian side to return the compliment. The meeting, which can hardly have taken place without the approval of Augustus, was significant as a public acknowledgement by the imperial government that it recognized the Parthian empire as an independent state subsisting side by side with the Roman, not indeed on a footing of equality, but with equal rights of sovereignty, however grudgingly they might be accorded. Before taking his leave, Phraataces revealed to Gaius that Lollius had been abusing his high trust by accepting huge bribes from Eastern potentates; the charge proved to be true, and Gaius renounced his friendship. Within a few days thereafter Lollius died, probably by his own hand, and was replaced by P. Sulpicius Quirinius, whose military ability and experience well fitted him to act as adviser to the young prince.

Meanwhile Tigranes had fallen in ‘a war with barbarians’, stirred up perhaps by the philo-Parthian nobles in resentment at his submission to Rome, and his death was followed by the abdication of his sister and wife Erato. The Armenian royal family was now extinct, and Gaius in the name of Augustus gave the crown to Ariobarzanes, king of Media since 20 b.c., who, says Tacitus, was readily accepted on account of his singularly handsome person and his noble character. It was not merely for those qualities that he was chosen, but because the Median royal stock was connected with the Armenian by marriage ties, and both countries had been recently ruled by Artaxes. The Parthian faction, however, refused to accept him and raised a revolt (a.d. 2), which Gaius proceeded to suppress. While attacking the fortress of Artagira (in the province of Ararat, perhaps near Kagyzman in the Araxes valley, some 80 miles west of Artaxata), he allowed himself to be lured on the 9th of September to a conversation with the commandant of the fort, a man named Addon, who pretended to have secret information to give him concerning the treasures of the Parthian king and treacherously wounded him. Thereupon the fort was besieged and captured after a long resistance, and the revolt was crushed, but Gaius did not recover from his wound. His health, never robust, and his spirits were broken. He begged Augustus to let him retire into the seclusion of a Syrian city and was with difficulty persuaded at least to return to Italy. Thither he set sail on an ordinary merchant ship, and died at the Lycian port of Limyra on February 21, a.d. 4.

VIII.THE ECLIPSE OF ROMAN INFLUENCE IN ARMENIA

Such was the tragic end of an expedition which had started with great pomp and high hopes. The net result was the establishment of a modus vivendi with Parthia and the restoration of Roman supremacy in Armenia. The understanding with Parthia lasted, despite dynastic changes, beyond the lifetime of Augustus, but ill fortune continued to attend his dispositions in Armenia. Soon after peace had been restored to that distracted country king Ariobarzanes died, to be succeeded by his son Artavasdes, who, like his father, was at the same time ruler of Media. Within a short time he was murdered, and Augustus then set on the throne Tigranes IV, whom he describes as a scion of the royal stock of Armenia. He was, beyond reasonable doubt, a grandson of Herod the Great, son of Alexander and Glaphyra, daughter of the Cappadocian king Archelaus, whose first wife was in all probability a princess of the Armenian royal house. With his appointment the record of Augustus ends, and as there appears to be no allusion in the Res Gestae to provincial and foreign affairs after a.d. 6, that year was perhaps the approximate date of his accession. He was evidently soon deposed by the Parthian faction. His reign was so brief that Tacitus omits it altogether, naming as the successor of Artavasdes a queen called Erato, who was quickly expelled; she was clearly the wife of Tigranes III, brought back for a moment to power. Her fall was followed by an interregnum, which lasted till the death of Augustus.

Meantime dynastic revolutions had taken place in Parthia. Soon after the death of Gaius, Phraataces was driven from the throne and replaced by Orodes (Hyrodes), a prince of Arsacid descent, whose excessive cruelty and irascible temper quickly led to his assassination. A Parthian embassy was then sent to Rome to ask for the return of Vonones, the eldest of the sons of Phraates IV, who had lived with his brothers in the capital since 9 b.c., to fill the vacant throne. The embassy was probably that which was sent on to Tiberius while he was engaged in completing the conquest of Germany (a.d. 4—6), and the evidence of Parthian coins indicates that the reign of Vonones began between the autumn of a.d. 6 and a.d. 8—9. His popularity was short-lived: his foreign habits and his indifference to riding and hunting and the other interests of a Parthian gentleman excited the contempt of the nobility; it was intolerable to them that the Arsacid throne should be held by a slave who had brooked bondage for so many years and that it should be bestowed by the Roman emperor, as though Parthia were a Roman province. So they called to the throne Artabanus, an Arsacid on his mother’s side, brought up among the Dahae of the steppe country north of Hyrcania, the original homeland of the Arsacids. According to Josephus, he had become king of Media (Atropatene), presumably after the death of the last representative of the native royal house. After an initial defeat he succeeded in establishing himself in a.d. 11—12, and Vonones fled to Armenia, then without a ruler, in the hope of securing the crown.

Such was the position at the end of Augustus’ reign. Armenia was kingless and had passed from Roman control. The whole involved story of failure after failure to keep the country under Roman influence, when all allowance has been made for ill fortune, shows the hopelessness of the task to which Rome had committed herself. The policy of setting a Romanized prince on an eastern throne, whether Armenian or Parthian, was doomed to failure, for a prince sophisticated by long residence in Rome could never win the favour of an Oriental aristocracy; and when, with the exhaustion of the supply of native princes, Augustus varied his policy by choosing nominees from an Eastern royal house friendly to Rome, such nominees, even if suitable in themselves, had to contend against the natural predominance of Parthian sympathies among the majority of the nobles. It was only by the rarest combination of personal qualities and tastes that a ruler appointed by Rome might succeed in maintaining his position; in general, there could be no guarantee of the stability of Roman arrangements unless a military force were stationed in the country or within striking distance of it.

IX. THE DEFENCE OF THE EUPHRATES FRONTIER

Augustus made the province of Syria the pivot of frontier defence in the East. In 27 b.c. he placed it under a legatus pro praetore of consular rank, combining with it Cilicia campestris, as geographical and political conditions suggested. The Cilician plain is cut off from Asia Minor by the broad, high wall of Taurus, over which winds for nearly seventy miles the pass of the Cilician Gates, ascending to about 4300 feet at the summit. On the other hand, it is closely attached to Syria, and communication between the two is easy. Cilicia included the eastern slopes of the northern part of the Amanus range and it was joined to Cyrrhestice by the pass over the broad col (3140 feet) of Arslanli Bel which led to Nicopolis (Islahiyeh), while on the south it was easily reached from Syria by the ‘Syrian Gates’, the Beilan pass, which barely reaches an altitude of 2000 feet. Politically it was separated from Roman territory on the west and north-west by a wedge of hill country which was not under direct Roman rule. The attachment of Cilicia to Syria was therefore natural.

The earliest evidence about the garrison of Syria, which relates to the time when Quinctilius Varus was governor (6—4 b.c.), puts its total strength at three legions. They are to be identified as III Gallica, VI Ferrata, and X Fretensis. In the later years of Augustus (in all probability) a fourth legion, XII Fulminata, was added. From these facts it has usually been inferred that, after the battle of Actium, Augustus assigned to Syria only three legions, and that the garrison remained at that strength during the greater part of his reign. If the inference were valid, it would prove conclusively that Augustus never anticipated any serious danger of aggression on the part of Parthia. It is, however, in itself improbable that, at a time when the course of events in the East could not be foreseen and when no solution of outstanding questions was in sight, the military establishment could have been fixed on such a low scale; and certain facts, notably the transference of two legions from the oversea provinces to Illyricum in a.d. 7 (one at least of which must have belonged to the garrison of Syria), indicate that the normal strength of the garrison was not less, and possibly more, than four legions, and that it had been temporarily reduced when Varus was governor by the absence of one or two legions on service in Galatia. During the Julio-Claudian period the military establishment remained fixed at four legions with their complement of auxiliary troops. The relative smallness of the garrison is striking when it is remem­bered that the governor had to keep his eye not only on his own province and frontier but also on the various vassal states to north and south, and that there was no other force anywhere else in the East to undertake military operations in Asia Minor or Armenia. The responsibility attaching to the Syrian command is reflected in the fact that it was the highest in rank of all imperial governorships and was held in the later stages, or at the end, of a senator’s career.

The legions were quartered not in isolated fortresses, as in the West, but in or close to towns. This was a result of the physical configuration of the country, which precluded the sharp division into civil and military areas that was so marked a feature of provinces like Gaul or Britain; for the fully and partially civilized parts of Syria, the coastal fringe and the inland valleys, were flanked by mountain ranges, which were occupied by robber tribes, and protection was needed against their depredations. The conditions of service made the soldier’s life more agreeable than in the West, but they were not equally favourable to the maintenance of a high standard of discipline and efficiency. Immediate contact with luxurious city life, easy and profitable service in a pleasant climate with no active foe on the frontier, the frequent breaking up of the units into detachments for police duties and for the execution of works of military and public utility, and the recruitment of the legions from eastern provinces—all these factors combined to relax military discipline and soon reduced the efficiency of the Syrian formations to a level far below that of the Western legions.

Concerning the distribution of the Syrian forces under Augustus our sources are silent, but the broad features of the system he adopted are indicated by the incidental mention of the standing quarters of two legions four years after his death. The military occupation was limited to the northern part of the province, which alone was under direct Roman administration. The defence of the rest against the raids of the Bedouins of the desert, as well as the maintenance of internal order, was entrusted to the vassal states of Emesa and Ituraea, which continued to rule all the southern portion of the province from Arethusa in the upper valley of the Orontes to the borders of Herod’s kingdom. The Ituraean prince, the ‘tetrarch’ and high-priest Zenodorus, retained for a time almost the whole of his hereditary principality, the valley of Chalcis and Heliopolis with its two flanking ranges of Lebanon and Antilebanon, which were the home of robber tribes, and the adjacent region beyond the Jordan as far as the Hauron, where brigandage and the robbery of caravans were the chief occupation of the natives. But his misgovernment and positive encouragement of robbery for his own profit led in 24 b.c. to military intervention and to the transference of the Transjordan districts of Trachonitis, Batanaea and Auranitis to Herod; and when he died in 20, the intervening district between Galilee and Trachonitis was added to Herod’s territory. Six years later the principality was further broken up by the assignation of a large tract in the centre of it to the colonies of veteran soldiers which Agrippa established at Berytus (Beirut) and Heliopolis (Baalbek) to act as  garrisons and assist in holding the Lebanon tribes in check. Nevertheless these hardy mountaineers continued to give trouble; about a.d. 6/7, during the governorship of Quirinius, a punitive expedition had to be sent against them; but, if difficult to reduce to order, they furnished excellent material for auxiliary regiments. What remained of the Ituraean principality was subsequently divided into the three ‘tetrarchies’ of Chalcis, Abila in Antilebanon, and Arca at the northern end of Lebanon, which survived under separate or combined rule till the end of the first century.

The administrative and military headquarters of the province were at Antioch, which can hardly have been left without a legionary garrison, any more than its rival Alexandria, in view of the character of its inhabitants, light-hearted, frivolous, intransigent people, inclined to turbulence. The other legions were stationed on or near the main roads leading north and south from the capital, in positions from which they could doubtless be easily concentrated at need for operations in any direction, but which were plainly chosen with an eye to the maintenance of peace within the province rather than to its defence against external foes. On the north the city of Cyrrhus (Khoros) was selected as the military centre of the quadrilateral forming the northern corner of the province, which extended from Mt. Amanus and the Cilician border to the Euphrates and abutted on the southern boundary of the client-kingdom of Commagene, left by Augustus under the rule of its old dynasty. Cyrrhus, a Seleucid foundation, occupied a strong position commanding the roads to Commagene and to Zeugma (Seleuceia ad Euphratem), situated at Balkis Kaleh, five miles above Birejik, which derived its name from a famous pontoon bridge constructed by Seleucus Nicator across the Euphrates, but no longer existing in the Roman period. It was a suitable station for a legion which had to serve the double purpose of covering the capital and of maintaining law and order in a region where brigandage had long been endemic. The legion placed there was no doubt X Fretensis, which occupied the position in a.d. 18. On the south of Antioch one legion, VI Ferrata, was quartered at or close to Laodiceia (Latakia) on the coast; and we may presume that the remaining legionary site was, as in later times, at Raphaneae (Rafmyeh) near the borders of the southern principalities, 15 miles north-west of Emesa, on the high ground which slopes down from the coastal range to the Orontes valley.

No legion lay on the Euphrates either in the reign of Augustus or for a long time after his death. It cannot indeed be supposed that there were no troops at all on the river. A position of the strategic importance of Zeugma, which was also a customs station for goods entering Syria from the East, cannot have been left undefended; we may suppose that it was guarded by an auxiliary force, but the distribution of the auxiliary troops is a blank in our knowledge. Scanty as the evidence is, it is sufficient to show that the disposition of the legionary forces was inspired primarily by considerations of internal security. Once a settlement with Parthia was reached and relations between the two empires were adjusted, the main function of the Syrian army was to maintain peace and order in the province and in the adjoining vassal states, not to stand on guard against a menacing foe ready to seize any opportunity of violating the frontier. The effect on the morale of the troops was soon to be revealed: when a serious campaign was contemplated, they always proved unfit to take the field and had to be reinforced by legions brought from the West.

CHAPTER X

EGYPT UNDER THE EARLY PRINCIPATE