xvi. 12) and Armageddon, whatever may be meant by it, as the rendezvous

of the Orientals for the collective attack on the West. Certainly the author, writing in the Roman empire, hints these far from patriotic hopes more than he expresses them.

[2] This holds true even in some measure for the chronology. The official historiography of the Sassanids reduces the space between the last Darius and the first Sassanid from 558 to 266 years (Nöldeke, _Tabari_, p. 1).

[3] The viceroys of Persis are called in their title constantly “Zag Alohin” (at least the Aramaean signs correspond to these words, which were presumably in pronunciation expressed in the Persian way), son of God (Mordtmann, _Zeitschrift für Numismatik_, iv. 155 f.), and to this corresponds the title θεοπάτωρ on the Greek coins of the great-kings. The designation “God” is also found, as with the Seleucids and the Sassanids.--Why a double diadem is attributed to the Arsacids (Herodian, vi. 2, 1) is not cleared up.

[4] Τῶν Παρθυαίων συνέδριόν φησιν (Ποσειδώνιος) εἶναι, says Strabo, xi. 9, 3, p. 515, διττόν, τὸ μὲν συγγενῶν, τὸ δὲ σοφῶν καὶ μάγων, ἐξ ὧν ἀμφοῖν τοὺς βασιλεῖς καθίστασθαι (καθίστησιν in MSS.); Justinus, xvii. 3, 1, _Mithridates rex Parthorum ... propter crudelitatem a senatu Parthico regno pellitur_.

[5] In Egypt, whose court ceremonial, as doubtless that of all the states of the Diadochi, is based on that ordained by Alexander, and in so far upon that of the Persian empire, the like title seems to have been conferred also personally (Franz, _C. I. Gr._ iii. 270). That the same occurred with the Arsacids, is possible. Among the Greek-speaking subjects of the Arsacid state the appellation μεγιστᾶνες seems in the original stricter use to denote the members of the seven houses; it is worthy of notice that _megistanes_ and _satrapae_ are associated (Seneca, _Ep._ 21; Josephus, _Arch._ xi. 3, 2; xx. 2, 3). The circumstance that in court mourning the Persian king does not invite the _megistanes_ to table (Suetonius, _Gai._ 5) suggests the conjecture that they had the privilege of taking meals with him. The title τῶν πρώτων φίλων is also found among the Arsacids just as at the Egyptian and Pontic courts (_Bull. de corr. Hell._ vii. p. 349).

[6] A royal cup-bearer, who is at the same time general, is mentioned in Josephus, _Arch._ xiv. 13, 7 = _Bell. Jud._ i. 13, 1. Similar court offices are of frequent occurrence in the states of the Diadochi.

[7] Tacitus, _Ann._ xv. 2, 31. If, according to the preface of Agathangelos (p. 109, Langlois), at the time of the Arsacids the oldest and ablest prince bore rule over the country, and the three standing next to him were kings of the Armenians, of the Indians, and of the Massagetae, there is here perhaps at bottom the same arrangement. That the Partho-Indian empire, if it was combined with the main land, was likewise regarded as an appanage for the second son, is very probable.

[8] These are doubtless meant by Justinus (xli. 2, 2), _proximus maiestati regum praepositorum ordo est; ex hoc duces in bello, ex hoc in pace rectores habent_. The native name is preserved by the gloss in Hesychius, βίσταξ ὁ βασιλεὺς παρὰ Πέρσαις. If in Ammianus, xxiii. 6, 14, the presidents of the Persian _regiones_ are called _vitaxae_ (read _vistaxae_), _id est magistri equitum et reges et satrapae_, he has awkwardly referred what is Persian to all Inner Asia (comp. _Hermes_, xvi. 613); we may add that the designation “leaders of horsemen” for these viceroys may relate to the fact that they, like the Roman governors, united in themselves the highest civil and the supreme military power, and the army of the Parthians consisted preponderantly of cavalry.

[9] This we learn from the title σατράπης τῶν σατραπῶν, attributed to one Gotarzes in the inscription of Kermanschahân in Kurdistan (_C. I. Gr._ 4674). It cannot be assigned to the Arsacid king of the same name as such; but perhaps there may be designated by it, as Olshausen (_Monatsbericht der Berliner Akademie_, 1878, p. 179) conjectures, that position which belonged to him after his renouncing of the great-kingdom (Tacitus, _Ann._ xi. 9).

[10] Still later a troop of horse in the Parthian army is called that “of the free:” Josephus, _Arch._ xiv. 13, 5 = _Bell. Jud._ i. 13, 3.

[11] The oldest known coin with Pahlavi writing was struck in Claudius’s time under Vologasus I.; it is bilingual, and gives to the king in Greek his full title, but only the name Arsaces, in Iranian merely the native individual name shortened (_Vol._).

[12] Usually this is restricted to the large silver money, and the small silver and most of the copper are regarded as of royal coinage. But by this view a singular secondary part in coinage is assigned to the great-king. More correctly perhaps the former coinage is conceived of as predominantly destined for dealings abroad, the latter as predominantly for internal intercourse; the diversities subsisting between the two kinds are also explained in this way.

[13] The first ruler that bears it is Phraapates about 188 B.C. (Percy Gardner, _Parthian Coinage_, p. 27).

[14] Thus there stands on the coins of Gotarzes (under Claudius) Γωτέρζης βασιλεὺς βασιλέων ὑὸς κεκαλουμένος Ἀρταβάνου. On the later ones the Greek legend is often quite unintelligible.

[15] While the kingdom of Darius, according to his inscriptions, includes in it the Gādara (the Gandhâra of the Indians, Γανδαρῖτις of the Greeks on the Cabul river) and the Hîdu (the dwellers by the Indus), the former are in one of the inscriptions of Asoka adduced among his subjects, and a copy of his great edict has been found in Kapurdi Giri, or rather in Shahbaz Garhi (Yusufzai-district), nearly 27 miles north-west of the point where the Cabul river falls into the Indus at Attock. The seat of the government of these north-west provinces of Asoka’s kingdom was (according to the inscription _C. I. Indicar._ i. p. 91) Takkhasilâ, Τάξιλα of the Greeks, some 40 miles E.S.E. of Attock, the seat of government for the south-western provinces was Ujjênî (Ὀζήνη). The eastern part of the Cabul valley thus belonged at any rate to Asoka’s empire. It is not quite impossible that the Khyber pass formed the boundary; but probably the whole Cabul valley belonged to India, and the boundary to the south of Cabul was formed by the sharp line of the Suleiman range, and farther to the south-west by the Bolan pass. Of the later Indo-Scythian king Huvishka (Ooerke of the coins), who seems to have resided on the Yamunâ in Mathurâ, an inscription has been found at Wardak not far northward from Cabul (according to information from Oldenberg).

[16] The Egyptian merchant named in note 3 makes mention, c. 47, of “the warlike people of the Bactrians, who have their own king.” At that time, therefore, Bactria was separated from the Indus-empire that was under Parthian princes. Strabo, too (xi. 11, 1, p. 516) treats the Bactro-Indian empire as belonging to the past.

[17] Probably he is the Kaspar--in older tradition Gathaspar--who appears among the holy three kings from the East (Gutschmid, _Rhein. Mus._ xix. 162).

[18] The most definite testimony to the Parthian rule in these regions is found in the description of the coasts of the Red Sea drawn up by an Egyptian merchant under Vespasian, c. 38: “Behind the mouth of the Indus in the interior lies the capital of Scythia Minnagara; but this is ruled by the Parthians, who constantly chase away one another” (ὑπὸ Πάρθων συνεχῶς ἀλλήλους ἐνδιωκόντων). The same is repeated in a somewhat confused way, c. 41; it might here appear as if Minnagara lay in India itself above Barygaza, and Ptolemy has already been led astray by this; but certainly the writer, who speaks as to the interior only from hearsay, has only wished to say that a large town Minnagara lay inland not far from Barygaza, and much cotton was brought thence to Barygaza. The numerous traces also of Alexander, which occur according to the same authority in Minnagara, can be found only on the Indus, not in Gujerat. The position of Minnagara on the lower Indus not far from Hyderabad, and the existence of a Parthian rule there under Vespasian, appear hereby assured.--With this we may be allowed to combine the coins of king Gondopharus or Hyndopherres, who in a very old Christian legend is converted to Christianity by St. Thomas, the apostle of the Parthians and Indians, and in fact appears to belong to the first period of the Roman empire (Sallet, _Num. Zeitschr._ vi. 355; Gutschmid, _Rhein. Mus._ xix. 162); of his brother’s son Abdagases (Sallet, _ib._ p. 365), who may be identical with the Parthian prince of this name in Tacitus, _Ann._ vi. 36, at any rate bears a Parthian name; and lastly of king Sanabarus, who must have reigned shortly after Hyndopherres, perhaps was his successor. Here belongs also a number of other coins marked with Parthian names, Arsaces, Pacorus, Vonones. This coinage attaches itself decidedly to that of the Arsacids (Sallet, _ib._ p. 277); the silver pieces of Gondopharus and of Sanabarus--of the others the coins are almost solely copper--correspond exactly to the Arsacid drachmae. To all appearance these belong to the Parthian princes of Minnagara; the appearance here of Indian legend alongside of the Greek, as of Pahlavi writing among the late Arsacids, suits this view. These, however, are not coins of satraps, but, as the Egyptian indicates, of great-kings rivalling those of Ctesiphon; Hyndopherres names himself in very corrupt Greek βασιλεὺς βασιλέων μέγας ἀυτόκρατωρ, and in good Indian “Maharajah Rajadi Rajah.” If, as is not improbable, under the Mambaros or Akabaros, whom the Periplus, c. 41, 52, designates as ruler of the coast of Barygaza, there lurks the Sanabarus of the coins, the latter belongs to the time of Nero or Vespasian, and ruled not merely at the mouths of the Indus, but also over Gujerat. Moreover, if an inscription found not far from Peshawur is rightly referred to king Gondopharus, his rule must have extended up thither, probably as far as Cabul.--The fact that Corbulo in the year 60 sent the embassy of the Hyrcanians who had revolted from the Parthians--in order that they might not be intercepted by the latter--to the coast of the Red Sea, whence they might reach their home without setting foot on Parthian territory (Tacitus, _Ann._ xv. 25), tells in favour of the view that the Indus valley at that time was not subject to the ruler of Ctesiphon.

[19] That the great kingdom of the Arsacids of Minnagara did not subsist much beyond the time of Nero, is probable from the coins. It is questionable what rulers followed them. The Bactro-Indian rulers of Greek names belong predominantly, perhaps all of them, to the pre-Augustan epoch; and various indigenous names, _e.g._ Maues and Azes, fall in point of language and writing (_e.g._ the form of the ω Ω) before this time. On the other hand the coins of the kings Kozulokadphises and Oemokadphises, and those of the Sacian kings, Kanerku and his successors, while all are clearly characterised as belonging to one coinage by the gold stater of the weight of the Roman aureus, which does not previously occur in the Indian coinage, are to all appearance later than Gondopharus and Sanabarus. They show how the state of the Indus valley assumed a national Indian type in ever increasing measure in contrast to the Hellenes as well as to the Iranians. The reign of these Kadphises will thus fall between the Indo-Parthian rulers and the dynasty of the Sacae, which latter begins with A.D. 78 (Oldenberg, in Sallet’s _Zeitschr. für Num._ viii. 292). Coins of these Sacian kings, found in the treasure of Peshawur, name in a remarkable way Greek gods in a mutilated form, Ηρακιλο, Σαραπο, alongside of the national Βουδο. The latest of their coins show the influence of the oldest Sassanid coinage, and might belong to the second half of the third century (Sallet, _Zeitschr. für Num._ vi. 224).

[20] The Indo-Greek and the Indo-Parthian rulers, just as the Kadphises, make use on their coins to a large extent of the indigenous Indian language and writing alongside of the Greek: the Sacian kings on the other hand never used the Indian language and Indian alphabet, but employ exclusively the Greek letters, and the non-Greek legends of their coins are beyond doubt Scythian. Thus on Kanerku’s gold pieces there sometimes stands βασιλεὺς βασιλέων Κανήρκου, sometimes ραο νανοραο κανηρκι κορανο, where the first two words must be a Scythian form of the Indian Rajâdi Rajah, and the two following contain the personal and the family name (Gushana) of the king (Oldenberg, _l. c._ p. 294). Thus these Sacae were foreign rulers in India in another sense than the Bactrian Hellenes and the Parthians. Yet the inscriptions set up under them in India are not Scythian but Indian.

[21] Arrian, who, as governor of Cappadocia, had himself wielded command over the Armenians (_contra Al._ 29), always in the _Tactica_ names the Armenians and Parthians together (4, 3, 44, 1, as respects the heavy cavalry, the mailed κοντοφόροι and the light cavalry, the ἀκροβολισταί or ἱπποτοξόται; 34, 7 as respects the wide hose); and, where he speaks of Hadrian’s introduction of barbaric cavalry into the Roman army, he traces the mounted archers back to the model of “the Parthians or Armenians” (44, 1).

[22] Caesar’s illegitimate son Πτολεμαῖος ὁ καὶ Καῖσαρ θεὸς φιλοπάτωρ φιλομήτωρ, as his royal designation runs (_C. I. Gr. 4717_), entered on the joint rule of Egypt in the Egyptian year 29 Aug. 711/2, as the era shows (Wescher, _Bullet. dell’ Inst._ 1866, p. 199; Krall, _Wiener Studien_, v. 313). As he came in place of Ptolemaeus the younger, the husband and brother of his mother, the setting aside of the latter by Cleopatra, of which the particulars are not known, must have taken place just then, and have furnished the occasion to proclaim him as king of Egypt. Dio also, xlvii. 31, places his nomination in the summer of 712 {42 B.C.} before the battle of Philippi. It was thus not the work of Antonius, but sanctioned by the two rulers in concert at a time when it could not but be their object to meet the wishes of the queen of Egypt, who certainly had from the outset ranged herself on their side.

[23] This is what Augustus means when he says that he had brought again to the empire the provinces of the East in great part distributed among kings (_Mon. Ancyr._ 5, 41: _provincias omnis, quae trans Hadrianum mare vergunt ad orientem, Cyrenasque, iam ex parte magna regibus eas possidentibus ... reciperavi_).

[24] The decorum, which was as characteristic of Augustus as its opposite was of his colleague, did not fail him here. Not merely in the case of Caesarion was the paternity, which the dictator himself had virtually acknowledged, afterwards officially denied; the children also of Antonius by Cleopatra, where indeed nothing was to be denied, were regarded doubtless as members of the imperial house, but were never formally acknowledged as children of Antonius. On the contrary the son of the daughter of Antonius by Cleopatra, the subsequent king of Mauretania Ptolemaeus, is called in the Athenian inscription, _C. I. A._ iii. 555, grandson of Ptolemaeus; for Πτολεμαίου ἔκγονος cannot well in this connection be taken otherwise. This maternal grandfather was invented in Rome, that they might be able officially to conceal the real one. Any one who prefers--as O. Hirschfeld proposes--to take ἔκγονος as great-grandson, and to refer it to the maternal great-grandfather, comes to the same result; for then the grandfather is passed over, because the mother was in the legal sense fatherless.--Whether the fiction, which is in my view more probable, went so far as to indicate a definite Ptolemaeus, possibly to prolong the life of the last Lagid who died in 712 {42 B.C.}, or whether they were content with inventing a father without entering into particulars, cannot be decided. But the fiction was adhered to in this respect, that the son of Antonius’s daughter obtained the name of the fictitious grandfather. The circumstance that in this case preference was given to the descent from the Lagids over that from Massinissa may probably have been occasioned more by regard to the imperial house, which treated the illegitimate child as belonging to it, than by the Hellenic inclination of the father.

[25] It is in itself credible that Antonius concealed the impending invasion from Phraates as long as possible, and therefore, when sending back Monaeses, declared himself ready to conclude peace on the basis of the restitution of the lost standards (Plutarch, 37; Dio, xlix. 24; Florus, ii. 20 [iv. 10]). But he knew presumably that this offer would not be accepted, and in no case can he have been in earnest with those proposals; beyond doubt he wished for the war and the overthrow of Phraates.

[26] The account of the matter given by Strabo, xi. 13, 4, p. 524, evidently after the description of this war compiled by Antonius’s comrade in arms Dellius, and, it may be conjectured, at his bidding (comp. _ib._ xi. 13, 3; Dio, xlix. 39), is a very sorry attempt to justify the beaten general. If Antonius did not take the nearest route to Ctesiphon, king Artavazdes cannot be brought in for the blame of it as a false guide; it was a military, and doubtless still more a political, miscalculation of the general in chief.

[27] The fact of the deposition and execution, and the time, are attested by Dio, xlix. 32, and Valerius Maximus, ix. 15, ext. 2; the cause or the pretext must have been connected with the Armenian war.

[28] The account of the seizure of Armenia is wanting, but the fact is clearly apparent from Tacitus, _Ann._ xi. 9. To this connection probably belongs what Josephus, _Arch._ xx. 3, 3, tells of the design of the successor of Artabanus to wage war against the Romans, from which Izates the satrap of Adiabene vainly dissuades him. Josephus names this successor, probably in error, Bardanes. The immediate successor of Artabanus III. was, according to Tacitus, _Ann._ xi. 8, his son of the same name, whom along with his son thereupon Gotarzes put out of the way; and this Artabanus IV. must be here meant.

[29] The statement of Petrus Patricius (_fr._ 3 Müll.) that king Mithradates of Iberia had planned revolt from Rome, but in order to preserve the semblance of fidelity, had sent his brother Cotys to Claudius, and then, when the latter had given information to the emperor of those intrigues, had been deposed and replaced by his brother, is not compatible with the assured fact that in Iberia, at least from the year 35 (Tacitus, _Ann._ vi. 32) till the year 60 (Tacitus, _Ann._ xiv. 26), Pharasmanes, and in the year 75 his son Mithradates (_C. I. L._ iii. 6052) bore rule. Beyond doubt Petrus has confused Mithradates of Iberia and the king of the Bosporus of the same name (I. 316, note 1), and here at the bottom lies the narrative, which Tacitus, _Ann._ xii. 18, presupposes.

[30] If the coins, which, it is true, for the most part admit of being distinguished only by resemblance of effigy, are correctly attributed, those of Gotarzes reach to Sel. 362 Daesius = A.D. 51 June, and those of Vologasus (we know none of Vonones II.) begin with Sel. 362 Gorpiaeus = A.D. 51 Sept. (Percy Gardner, _Parthian Coinage_, pp. 50, 51), which agrees with Tacitus, _Ann._ xii. 14, 44.

[31] Gorneae, called by the Armenians _Garhni_, as the ruins (nearly east of Erivân) are still at present named. (Kiepert.)

[32] Even after the attack Tiridates complained _cur datis nuper obsidibus redintegrataque amicitia ... vetere Armeniae possessione depelleretur_, and Corbulo presented to him, in case of his turning as a suppliant to the emperor, the prospect of a _regnum stabile_ (Tacitus, _Ann._ xii. 37). Elsewhere too the refusal of the oath of fealty is indicated as the proper ground of war (Tacitus, _Ann._ xii. 34).

[33] The report in Tacitus, _Ann._ xiii. 34-41, embraces beyond doubt the campaigns of 58 and 59, since Tacitus under the year 59 is silent as to the Armenian campaign, while under the year 60, _Ann._ xiv. 23 joins on immediately to xiii. 41, and evidently describes merely a single campaign; generally, where he condenses in this way, he as a rule anticipates. That the war cannot have begun only in 59, is further confirmed by the fact that Corbulo observed the solar eclipse of 30th April 59 on Armenian soil (Plin. _H. N._ ii. 70, 180); had he not entered the country till 59, he could hardly have crossed the enemy’s frontier so early in the year. The narrative of Tacitus, _Ann._ xiii, 34-41, does not in itself show an intercalation of a year, but with his mode of narrating it admits the possibility that the first year was spent in the crossing of the Euphrates and the settling in Armenia, and so the winter mentioned in _c._ 35 is that of the year 58-9, especially as in view of the character of the army such a beginning to the war would be quite in place, and in view of the short Armenian summer it was militarily convenient thus to separate the marching into the country and the conduct proper of the war.

[34] From the representation of Tacitus, _Ann._ xv. 6, the partiality and the perplexity are clearly seen. He does not venture to express the surrender of Armenia to Tiridates, and only leaves the reader to infer it.

[35] This is said by Tacitus himself, _Ann._ xv. 10: _nec a Corbulone properatum, quo gliscentibus periculis etiam subsidii laus augeretur_, in naive unconcern at the severe censure which this praise involves. How partial is the tone of the whole account resting on Corbulo’s despatches, is shown among other things by the circumstance that Paetus is reproached in one breath with the inadequate provisioning of the camp (xv. 8) and with the surrender of it in spite of copious supplies (xv. 16), and the latter fact is inferred from this, that the retiring Romans preferred to destroy the stores which, according to the capitulation, were to be delivered to the Parthians. As the exasperation against Tiberius found its expression in the painting of Germanicus in fine colours, so did the exasperation against Nero in the picture of Corbulo.

[36] The statement of Corbulo that Paetus bound himself on oath in presence of his soldiers and of the Parthian deputies to send no troops to Armenia till the arrival of Nero’s answer, is declared by Tacitus, _Ann._ xv. 16, unworthy of credit; it is in keeping with the state of the case, and nothing was done to the contrary.

[37] As, according to Tacitus, _Ann._ xv. 25 (comp. Dio, lxii. 22), Nero dismissed graciously the envoys of Vologasus, and allowed them to see the possibility of an understanding if Tiridates appeared in person, Corbulo may in this case have acted according to his instructions; but this was rather perhaps one of the turns added in the interest of Corbulo. That these events were brought under discussion in the trial to which he was subjected some years after, is probable from the statement that one of the officers of the Armenian campaign became his accuser. The identity of the cohort-prefect, Arrius Varus, in Tacitus, _Ann._ xiii. 9, and of the primipilus, _Hist._ iii. 6, has been without reason disputed; comp. on _C. I. L._ v. 867.

[38] In Ziata (Charput) there have been found two inscriptions of a fort, which one of the legions led by Corbulo over the Euphrates, the 3d Gallica, constructed there by Corbulo’s orders in the year 64 (_Eph. epigr._ v. p. 25).

[39] Nero intended _inter reliqua bella_, an Ethiopian one (Plin. vi. 29, comp. 184). To this the sending of troops to Alexandria (Tacitus, _Hist._ i. 31, 70) had reference.

[40] As the aim of the expedition both Tacitus, _Hist._ i. 6, and Suetonius, _Ner._ 19, indicate the Caspian gates, _i.e._ the pass of the Caucasus between Tiflis and Vladi-Kavkas at Darial, which, according to the legend, Alexander closed with iron gates (Plin. _H. N._ vi. 11, 30; Josephus, _Bell. Jud._ vii. 7, 4; Procopius, _Pers._ i. 10). Both from this locality and from the whole scheme of the expedition it cannot possibly have been directed against the Albani on the western shore of the Caspian Sea; here, as well as at another passage (_Ann._ ii. 68, _ad Armenios, inde Albanos Heniochosque_), only the Alani can be meant, who in Josephus, _l. c._ and elsewhere appear just at this spot and are frequently confounded with the Caucasian Albani. No doubt the account of Josephus is also confused. If here the Albani, with consent of the king of the Hyrcanians, invade Media and then Armenia through the Caspian gates, the writer has been thinking of the other Caspian gate eastward from Rhagae; but this must be his mistake, since the latter pass, situated in the heart of the Parthian kingdom, cannot possibly have been the aim of the Neronian expedition, and the Alani had their seats not on the eastern shore of the Caspian but to the north of the Caucasus. On account of this expedition the best of the Roman legions, the 14th, was recalled from Britain, although it went only as far as Pannonia (Tacitus, _Hist._ ii. 11, comp. 27, 66), and a new legion, the 1st Italic, was formed by Nero (Suetonius, _Ner._ 19). One sees from this what was the scale on which the project was conceived.

[41] In what connection he refused to Vespasian the title of emperor (Dio, lxvi. 11) is not clear; possibly immediately after his insurrection, before he had perceived that the Flavians were the stronger. His intercession for the princes of Commagene (Josephus, _Bell. Jud._ vii. 7, 3) was attended by success, and so was purely personal, by no means a protest against the conversion of the kingdom into a province.

[42] The four Syrian legions were the 3d _Gallica_, the 6th _ferrata_ (both hitherto in Syria), the 4th _Scythica_ (hitherto in Moesia, but having already taken part in the Parthian as in the Jewish war), and the 16th _Flavia_ (new). The one legion of Palestine was the 10th _fretensis_ (hitherto in Syria). The two of Cappadocia were the 12th _fulminata_ (hitherto in Syria, moved by Titus to Melitene, Josephus, _Bell. Jud._ vii. 1, 3), and the 15th _Apollinaris_ (hitherto in Pannonia, but having taken part, like the 4th _Scythica_, in the Parthian as in the Jewish war). The garrisons were thus changed as little as possible, only two of the legions already called earlier to Syria received fixed stations there, and one newly instituted was moved thither.--After the Jewish war under Hadrian the 6th _ferrata_ was despatched from Syria to Palestine.

[43] At this time (comp. _C. I. L._ v. 6988), probably falls also the Cappadocian governorship of C. Rutilius Gallicus, of which it is said (Statius, i. 4, 78): _hunc ... timuit ... Armenia et patiens Latii iam pontis Araxes_, with reference presumably to a bridge-structure executed by this Roman garrison. That Gallicus served under Corbulo, is from the silence of Tacitus not probable.

[44] That war threatened to break out under Vespasian in the year 75 on the Euphrates, while M. Ulpius Trajanus, the father of the emperor, was governor of Syria, is stated by Pliny in his panegyric on the son, c. 14, probably with strong exaggeration; the cause is unknown.

[45] There are coins dated, and provided with the individual names of the kings, of (V)ologasus from the years 389 and 390 = 77-78; of Pacorus from the years 389-394 = 77-82 (and again 404-407 = 92-95); of Artabanus from the year 392 = 80-1. The corresponding historical dates are lost, with the exception of the notice connecting Titus and Artabanus in Zonaras, xi. 18 (comp. Suetonius, _Ner._ 57; Tacitus, _Hist._ i. 2), but the coins point to an epoch of rapid changes on the throne, and, apparently, of simultaneous coinage by rival pretenders.

[46] This is proved by the detached notice from Arrian in Suidas (_s. v._ ἐπίκλημα): ὁ δὲ Πάκορος ὁ Παρθυαίων βασιλεὺς καὶ ἄλλα τινὰ ἐπικλήματα ἐπέφερε Τραιανῷ τῷ βασιλεῖ, and by the attention which is devoted in Pliny’s report to the emperor, written about the year 112 (_ad Trai._ 74), to the relations between Pacorus and the Dacian king Decebalus. The time of the reign of this Parthian king cannot be sufficiently fixed. There are no Parthian coins with the king’s name from the whole period of Trajan; the coining of silver seems to have been in abeyance during that period.

[47] That Axidares (or Exedares) was a son of Pacorus and king of Armenia before Parthomasiris, but had been deposed by Chosroes, is shown by the remnants of Dio’s account, lxviii. 17; and to this point also the two fragments of Arrian (16 Müller), the first, probably from an address of a supporter of the interests of Axidares to Trajan: Ἀξιδάρην δὲ ὅτι ἄρχειν χρὴ Ἀρμενίας, οὔ μοι δοκεῖ εἶναί σε ἀμφίλογον, whereupon doubtless the complaints brought against Parthomasiris followed; and the answer, evidently of the emperor, that it is not the business of Axidares, but his, to judge as to Parthomasiris, because he--apparently Axidares--had first broken the treaty and suffered for it. What fault the emperor imputes to Axidares is not clear; but in Dio also Chosroes says that he has not satisfied either the Romans or the Parthians.

[48] The remnants of Dio’s account in Xiphilinus and Zonaras show clearly that the Parthian expedition falls into two campaigns, the first (Dio, lvi. 17, 1, 18, 2, 23-25), which is fixed at A.D. 115 by the consulate of Pedo (the date also of Malalas, p. 275, for the earthquake of Antioch, 13 Dec. 164 of the Antiochene era = A.D. 115 agrees therewith), and the second (Dio, c. 26-32, 3), which is fixed at A.D. 116 by the conferring of the title _Parthicus_ (c. 28, 2), took place between April and August of that year (see my notice in Droysen, _Hellenismus_, iii. 2, 361). That at c. 23 the titles _Optimus_ (conferred in the course of A.D. 114) and _Parthicus_ are mentioned out of the order of time, is shown as well by their juxtaposition as by the later recurrence of the second honour. Of the fragments most belong to the first campaign; c. 22, 3 and probably also 22, 1, 2 to the second.--The acclamations of _imperator_ do not stand in the way. Trajan was demonstrably in the year 113 imp. VI. (_C. I. L._ vi. 960); in the year 114 imp. VII. (_C. I. L._ ix. 1558 _et al._); in the year 115 imp. IX. (_C. I. L._ ix. 5894 _et al._), and imp. XI. (Fabretti, 398, 289 _et al._); in the year 116 imp. XII. (_C. I. L._ viii. 621, x. 1634), and XIII. (_C. I. L._ iii. D. xxvii.). Dio attests an acclamation from the year 115 (lxviii. 19), and one from the year 116 (lxviii. 28); there is ample room for both, and there is no reason to refer imp. VII. precisely, as has been attempted, to the subjugation of Armenia.

[49] The pungent description of the Syrian army of Trajan in Fronto (p. 206 f. Naber) agrees almost literally with that of the army of Corbulo in Tacitus, _Ann._ xiii. 35. “The Roman troops generally had sadly degenerated (_ad ignaviam redactus_) through being long disused to military service; but the most wretched of the soldiers were the Syrian, insubordinate, refractory, unpunctual at the call to arms, not to be found at their post, drunk from midday onward; unaccustomed even to carry arms and incapable of fatigue, ridding themselves of one piece of armour after another, half naked like the light troops and the archers. Besides they were so demoralised by the defeats they had suffered that they turned their backs at the first sight of the Parthians, and the trumpets were regarded by them, as it were, as giving the signal to run away.” In the contrasting description of Trajan it is said among other things: “He did not pass through the tents without closely concerning himself as to the soldiers, but showed his contempt for the Syrian luxury, and looked closely into the rough doings of the Pannonians (_sed contemnere_--so we must read--_Syrorum munditias, introspicere Pannoniorum inscitias_); so he judged of the serviceableness (_ingenium_) of the man according to his bearing (_cultus_).” In the Oriental army of Severus also the “European” and the Syrian soldiers are distinguished (Dio, lxxv. 12).

[50] This is shown by the _mala proelia_ in the passage of Fronto quoted, and by Dio’s statement, lxviii. 19, that Trajan took Samosata without a struggle; thus the 16th legion stationed there had lost it.

[51] It may be that at the same time Armenia also revolted. But when Gutschmid (quoted by Dierauer in Büdinger’s _Untersuchungen_, i. 179), makes Meherdotes and Sanatrukios, whom Malalas adduces as kings of Persia in the Trajanic war, into kings of Armenia again in revolt, this result is attained by a series of daring conjectures, which shift the names of persons and peoples as much as they transform the causal nexus of events. There are certainly found in the confused coil of legends of Malalas some historical facts, _e.g._ the installation of Parthamaspates (who is here son of king Chosroes of Armenia) as king of Parthia by Trajan; and so, too, the dates of Trajan’s departure from Rome in October (114), of his landing in Seleucia in December, and of his entrance into Antioch on the 7th Jan. (115) may be correct. But, as this report stands, the historian can only decline to accept it; he cannot rectify it.

[52] Fronto, _Princ. hist._ p. 209 Naber: _cum praesens Traianus Euphratis et Tigridis portoria equorum et camelorum trib_[_utaque ordinaret, Ma_]_cer_ (?) _caesus est_. This applies to the moment when Babylonia and Mesopotamia revolted, while Trajan was tarrying at the mouth of the Tigris.

[53] Nearly with equal warrant, Julian (_Caes._ p. 328) makes the emperor say that he had not taken up arms against the Parthians before they had violated right, and Dio (lxviii. 17) reproaches him with having waged the war from ambition.

[54] Hadrian cannot possibly have released Armenia from the position of a Roman dependency. The notice of his biographer, c. 21: _Armeniis regem habere permisit, cum sub Traiano legatum habuissent_ points rather to the contrary, and we find at the end of Hadrian’s reign a contingent of Armenians in the army of the governor of Cappadocia (Arrian, _c. Alan._ 29). Pius did not merely induce the Parthians by his representations to desist from the intended invasion of Armenia (_vita_, 9), but also in fact invested them with Armenia (coins from the years 140-144, Eckhel, vii. p. 15). The fact also that Iberia certainly stood in the relation of dependence under Pius, because otherwise the Parthians could not have brought complaints as to its king in Rome (Dio, lxix. 15), presupposes a like dependent relation for Armenia. The names of the Armenian kings of this period are not known. If the _proximae gentes_, with the rule of which Hadrian compensated the Parthian prince nominated as Parthian king by Trajan (_vita_, c. 5), were in fact Armenians, which is not improbable, there lies in it a confirmation as well of the lasting dependence of Armenia on Rome as of the continuous rule of the Arsacids there. Even the Ἀυρήλιος Πάκορος βασιλεὺς μεγάλης Ἀρμενίας, who erected a monument in Rome to his brother Aurelius Merithates who died there (_C. I. Gr._ 6559), belongs from his name to the house of the Arsacids. But he was hardly the king of Armenia installed by Vologasus IV. and deposed by the Romans (p. 74); if the latter had come to Rome as a captive, we should know it, and even he would hardly have been allowed to call himself king of Great Armenia in a Roman inscription.

[55] As vassals holding from Trajan or Hadrian, Arrian (_Peripl._ c. 15) adduces the Heniochi and Machelones (comp. Dio, lxviii. 18; lxxi. 14); the Lazi (comp. Suidas, _s. v._ Δομετιανός), over whom also Pius put a king (_vita_, 9); the Apsilae; the Abasgi; the Sanigae, these all within the imperial frontier reaching as far as Dioscurias = Sebastopolis; beyond it, in the region of the Bosporan vassal-state, the Zichi or Zinchi (_ib._ c. 27).

[56] This is confirmed not only by Arrian, _Peripl._ c. 7, but by the officer of Hadrian’s time _praepositus numerorum tendentium in Ponto Absaro_ (_C. I. L._ x. 1202).

[57] Comp. p. 75 note 2. The detachment probably of 1000 men (because under a tribune) doing garrison duty in the year 185 in Valarshapat (Etshmiazin) not far from Artaxata, belonged to one of the Cappadocian legions (_C. I. L._ iii. 6052).

[58] Hadrian’s efforts after the friendship of the Oriental vassal-princes are often brought into prominence, not without a hint that he was more than fairly indulgent to them (_vita_, c. 13, 17, 21). Pharasmanes of Iberia did not come to Rome on his invitation, but complied with that of Pius (_vita Hadr._ 13, 21; _vita Pii_, 9; Dio, lxix. 15, 2, which excerpt belongs to Pius).

[59] We still possess the remarkable report of the governor of Cappadocia under Hadrian, Flavius Arrianus, upon the mobilising of the Cappadocian army against the “Scythians” among his minor writings; he was himself at the Caucasus and visited the passes there (Lydus, _de Mag._ iii. 53).

[60] This we learn from the fragments of Dio’s account in Xiphilinus, Zonaras, and in the Excerpts; Zonaras has preserved the correct reading Ἀλανοί instead of Ἀλβανοί; that the Alani pillaged also the territory of the Albani, is shown by the setting of the exc. Ursin. lxxii.

[61] So he is named in Lucian, _Hist. conscr._ 21; if the same calls him (_Alex._ 27) Othryades, he is drawing here from a historian of the stamp of those whom he ridicules in that treatise, and of whom another Hellenised the same man as Oxyroes (_Hist. conscr._ c. 18).

[62] Syria was administered when the war broke out by L. Attidius Cornelianus (_C. I. Gr._ 4661 of the year 160; _vita Marci_, 8; _C. I. L._ iii. 129 of the year 162), after him by Julius Verus (_C. I. L._ iii. 199, probably of the year 163) and then by Avidius Cassius presumably from the year 164. The statement that the other provinces of the East were assigned to Cassius’s command (Philostratus, _vit. Soph._ i. 13; Dio, lxxi. 3), similarly to what was done to Corbulo as legate of Cappadocia, can only relate to the time after the departure of the emperor Verus; so long as the latter held the nominal chief command there was no room for it.

[63] A fragment probably of Dio (in Suidas _s. v._ Μάρτιος), tells that Priscus in Armenia laid out the Καινὴ πόλις and furnished it with a Roman garrison, his successor Martius Verus silenced the national movement that had arisen there, and declared this city the first of Armenia. This was Valarshapat (Οὐαλαρσαπάτ or Οὐαλεροκτίστη in Agathangelos), thenceforth the capital of Armenia. Καινὴ πόλις was, as Kiepert informs me, already recognised by Stilting as translation of the Armenian Nôr-Khalakh, which second name Valarshapat constantly bears in Armenian authors of the fifth century alongside of the usual one. Moses of Chorene, following Bardesanes, makes the town originate from a Jewish colony brought thither under king Tigranes VI., who according to him reigned 150-188; he refers the enclosing of it with walls and the naming of it to his son Valarsch II. 188-208. That the town had a strong Roman garrison in 185 is shown by the inscription _C. I. L._ iii. 6052.

[64] That Sohaemus was Achaemenid and Arsacid (or professed to be) and king’s son and king, as well as Roman senator and consul, before he became king of Great Armenia, is stated by his contemporary Jamblichus (c. 10 of the extract in Photius). Probably he belonged to the dynastic family of Hemesa (Josephus, _Arch._ xx. 8, 4, _et al._) If Jamblichus the Babylonian wrote “under him,” this can doubtless only be understood to the effect that he composed his romance in Artaxata. That Sohaemus ruled over Armenia before Pacorus is nowhere stated, and is not probable, since neither Fronto’s words (p. 127 Naber), _quod Sohaemo potius quam Vologaeso regnum Armeniae dedisset aut quod Pacorum regno privasset_, or those of the fragment from Dio (?) lxxi. 1: Μάρτιος Οὐῆρος τὸν Θουκυδίδην ἐκπέμπει καταγαγεῖν Σόαιμον ἐς Ἀρμενίαν point to reinstatement, and the coins with _rex Armeniis datus_ (Eckhel, vii. 91, comp. _vita Veri_, 7, 8) in fact exclude it. We do not know the predecessor of Pacorus, and are not even aware whether the throne which he took possession of was vacant or occupied.

[65] This is shown by the Mesopotamian royal and urban coins. There are no accounts in our tradition as to the conditions of peace.

[66] The beginning of the Ursinian excerpt of Dio, lxxv. 1, 2, is confused. Οἱ Ὀρροηνοὶ, it is said, καὶ οἱ Ἀδιαβηνοὶ ἀποστάντες καὶ Νίσιβιν πολιορκοῦντες καὶ ἡττηθέντες ὑπὸ Σεουήρου ἐπρεσβεύσαντο πρὸς αὐτὸν μετὰ τὸν τοῦ Νίγρου θάνατον. Osrhoene was then Roman, Adiabene Parthian; from whom did the two districts revolt? and whose side did the Nisibenes take? That their opponents were defeated by Severus before the sending of the embassy is inconsistent with the course of the narrative; for the latter makes war upon them because their envoys make unsatisfactory offers to him. Probably the supporting of Niger by subjects of the Parthians and their concert with Niger’s Roman partisan are now strictly apprehended as a revolt from Severus; the circumstance that the people afterwards maintain that they had intended rather to support Severus, is clearly indicated as a makeshift. The Nisibenes may have refused to co-operate, and therefore have been attacked by the adherents of Niger. Thus is explained what is clear from the extract given by Xiphilinus from Dio, lxxv. 2, that the left bank of the Euphrates was for Severus an enemy’s land, but not Nisibis; therefore the town need not have been Roman at that time; on the contrary, according to all indications, it was only made Roman by Severus.

[67] As the wars against the Arabians and the Adiabenians were in fact directed against the Parthians, it was natural that the titles _Parthicus_, _Arabicus_, and _Parthicus Adiabenicus_, should on that account be conferred on the emperor; they are also so found, but usually Parthicus is omitted, evidently because, as the biographer of Severus says (c. 9), _excusavit Parthicum nomen, ne Parthos lacesseret_. With this agrees the notice certainly belonging to the year 195 in Dio, lxxv. 9, 6, as to the peaceful agreement with the Parthians and the cession of a portion of Armenia to them.

[68] That Armenia also fell into their power is indicated by Herodian, v. 9, 2; no doubt his representation is warped and defective.

[69] When at the peace in 218 the old relation between Rome and Armenia was renewed, the king of Armenia gave himself the prospect of a renewal of the Roman annual moneys (Dio, lxxviii. 27: τοῦ Τιριδάτου τὸ ἀργύριον ὃ κατ’ ἔτος παρὰ τῶν Ῥωμαίων εὑρίσκετο ἐλπίσαντος λήψεσθαι). Payment of tribute proper by the Romans to the Armenians is excluded for the period of Severus and the time before Severus, and by no means agrees with the words of Dio; the connection must be what we have indicated. In the fourth and fifth centuries the fortress of Biriparach in the Caucasus, which barred the Dariel pass, was maintained by the Persians, who played the part of masters here after the peace of 364, with a Roman contribution, and this was likewise conceived as payment of tribute (Lydus, _de Mag._ iii. 52, 53; Priscus, _fr._ 31, Mull.).

[70] Artaxares names his father Papacus in the inscription, quoted at p. 83, note 1, king; how it is to be reconciled with this, that not merely does the native legend (in Agathias ii. 27) make Pabek a shoemaker, but also the contemporary Dio (if in reality Zonaras, xii. 15, has borrowed these words from him) names Artaxares ἐξ ἀφανῶν καὶ ἀδόξων, we do not know. Naturally the Roman authors take the side of the weak legitimate Arsacid against the dangerous usurper.

[71] Strabo (under Tiberius) xv. 3, 24: νῦν δ’ ἤδη καθ’ αὐτοὺς συνεστῶτες οἱ Πἐρσαι βασιλέας ἔχουσιν ὑπηκόους ἑτέροις βασιλεῦσι, πρότερον μὲν Μακεδόσι, νῦν δὲ Παρθυαίοις.

[72] When Nöldeke says (_Tabari_, p. 449), “The subjection of the chief lands of the monarchy directly to the crown formed the chief distinction of the Sassanid kingdom from the Arsacid, which had real kings in its various provinces,” the power of the great-kingdom beyond doubt is thoroughly dependent on the personality of the possessor, and under the first Sassanids must have been much stronger than under the last decayed Arsacids. But a contrast in principle is not discoverable. From Mithradates I., the proper founder of the dynasty, onward the Arsacid ruler names himself “king of kings,” just as did subsequently the Sassanid, while Alexander the Great and the Seleucids never bore this title. Even under them individual vassal-kings ruled, _e.g._ in Persis (p. 81, note 2); but the vassal-kingdom was not then the regular form of imperial administration, and the Greek rulers did not name themselves according to it, any more than the Caesars assumed the title of great-king on account of Cappadocia or Numidia. The satraps of the Arsacid state were essentially the Marzbans of the Sassanids. Perhaps rather the great imperial offices, which in the Sassanid polity correspond to the supreme administrative posts of the Diocletiano-Constantinian constitution, and probably were the model for the latter, were wanting to the Arsacid state; then certainly the two would be related to each other much as the imperial organisation of Augustus to that of Constantine. But we know too little of the Arsacid organisation to affirm this with certainty.

[73] According to the Persian records of the last Sassanid period preserved in the Arabic chronicle of Tabari Ardashir, after he has cut off with his own hand the head of Ardawan and has assumed the title Shahan-shah, king of kings, conquers first Hamadhan (Ecbatana) in Great Media, then Aderbijan (Atropatene), Armenia, Mosul (Adiabene); and further Suristan or Sawad (Babylonia). Thence he returns to Istachr unto his Persian home, and then starting afresh conquers Sagistan, Gurgan (Hyrcania), Abrashahr (Nisapur in the Parthian land), Merv (Margiane), Balkh (Bactra), and Charizm (Khiva) up to the extreme limits of Chorasan. “After he had killed many people and had sent their heads to the fire-temple of Anahedh (in Istachr), he returned from Merv to Pars and settled in Gor” (Feruzabad). How much of this is legend, we do not know (comp. Nöldeke, _Tabari_, p. 17, 116).

[74] The title runs in Greek (_C. I. Gr._ 4675), Μάσδασνος (Mazda-servant, treated as a proper name) θεὸς Ἀρταξάρης βασιλεὺς βασιλέων Ἀριανῶν ἐκ γένους θεῶν; with which closely agrees the title of his son Sapor I. (_ib._ 4676), only that after Ἀριανῶν there is inserted καὶ Ἀναριανῶν, and so the extension of the rule to foreign lands is brought into prominence. In the title of the Arsacids, so far as it is clear from the Greek and Persian legends of coins, θεός, βασιλεὺς βασιλέων, θεοπάτωρ (= ἐκ γένους θεῶν) recur, whereas there is no prominence given to the Arians and, significantly, to the “Mazda-servant”; by their side appear numerous other titles borrowed from the Syrian kings, such as ἐπιφάνης, δίκαιος, νικάτωρ, also the Roman αὐτοκράτωρ.

[75] Frawardin, Ardhbehesht, etc. (Ideler, _Chronologie_, ii. 515). It is remarkable that essentially the same names of the months have maintained themselves in the provincial calendar of the Roman province Cappadocia (Ideler, i. 443); they must proceed from the time when it was a Persian satrapy.

[76] Such is the account of the trustworthy Dio, lxxviii. 1; the version of Herodian, iv. 11, that Artabanus promised his daughter, and at the celebration of the betrothal allowed Antoninus to cut down the Parthians present, is unauthenticated.

[77] If there is any truth in the mention of the Cadusians in the biography, c. 6, the Romans induced this wild tribe, not subject to the government in the south-west of the Caspian Sea, to fall at the same time upon the Parthians.

[78] The subsequently received chronology puts the beginning of the Sassanid dynasty in the Seleucid year 538 = 1st Oct. A.D. 226-7, or the fourth (full) year of Severus Alexander, reigning since spring 222 (Agathias, iv. 24). According to other data king Ardashir numbered the year from the autumn A.D. 223-4 as his first, and so doubtless assumed in this the title of great-king (Nöldeke, _Tabari_, p. 410). The last dated coin as yet known of the older system is of the year 539. When Dio wrote between 230 and 234, Artabanus was dead and his adherents were overpowered, and the advance of Artaxares into Armenia and Mesopotamia was expected.

[79] The emperor remained probably in Palmyra; at least a Palmyrene inscription, _C. I. Gr._ 4483, mentions the ἐπιδημία θεοῦ Ἀλεξάνδρου.

[80] The incomparably wretched accounts of this war (relatively the best is that drawn from a common source in Herodian, Zonaras, and Syncellus, p. 674) do not even decide the question who remained victor in these conflicts. While Herodian speaks of an unexampled defeat of the Romans, the Latin authorities, the Biography as well as Victor, Eutropius, and Rufius Festus, celebrate Alexander as the conqueror of Artaxerxes or Xerxes, and according to these latter the further course of things was favourable. Herodian vi. 6, 5, offers the means of adjustment. According to the Armenian accounts (Gutschmid, _Zeitschr. der deutschen morgenländ. Gesellschaft_, xxxi. 47) the Arsacids with the support of the tribes of the Caucasus held their ground in Armenia down to the year 237 against Ardashir; this diversion may be correct and may have tended to the advantage of the Romans.

[81] The best account is furnished by Syncellus, p. 683 and Zonaras, xii. 18, drawing from the same source. With this accord the individual statements of Ammianus, xxiii. 5, 7, 17, and nearly so the forged letter of Gordian to the Senate in the Biography, c. 27, from which the narrative, c. 26, is ignorantly prepared; Antioch was in danger, but not in the hands of the Persians.

[82] So Zonaras, xii. 19, represents the course of affairs; with this Zosimus, iii. 33, agrees, and the later course of things shows that Armenia was not quite in Persian possession. If, according to Euagrius,