CLASSICAL PHILOLOGY

Volume XXXVII JULY 1942 Number 3

THE MID-THIRD CENTURY OF THE CHRISTIAN ERA

A. T. OLMSTEAD

IBBON long since impressed on the popular mind the signifi- G cance of the third century as one of the great turning-points in world-history. Succeeding historians have only deepened the impression. The second century basked in the Antonine peace. It considered itself the true successor to the famous classical culture in life, literature, and religion. The forms of republican government were still observed. The slow but steady economic and social decline went unnoticed by the literate.

Oriental monarchy with all its pomp and ceremony ruled the fourth century. An oriental religion, Christianity, had conquered the Roman Empire. Wars against barbarian invaders were constant, and the frontier steadily receded. Economic and social security had been recovered only through a hardening of the arteries, with the conse- quent refusal of energetic movement. From the Western viewpoint, the world was in decay.

We of the West find it difficult to realize that the exact reverse was the situation in the true Orient east of the Euphrates. At the begin- ning of the third century, this Orient was ruled by Parthia, whose best years were long past. The empire was torn by revolts of pre- tenders to the throne; the feudal chiefs were all-powerful. Even the weak Roman armies could penetrate at will to the heart of the Par- thian empire, and the devastation of their soldiers added to the pre- vailing chaos. Wealth rapidly declined, culture almost disappeared. [Cuasstcan Putotoey, XXXVII, Jury, 1942] 241

242 A. T. OLMSTEAD

At the close, all Parthian territory and much more besides owed fealty to a powerful Sasanid family, little troubled by the feudal no- bility. A new religion which claimed to be revived Zoroastrianism gave fresh impulse to an already abounding life. Accumulated wealth produced an art whose high quality we are only now beginning to ap- preciate. Roman attacks on Parthia were terribly avenged when Sasanid armies repeatedly advanced to the very limits of the Roman Empire in the east.

For the beginning and close of this century we possess sources which, if neither ample nor too trustworthy, do at least permit the construction of a fairly adequate history. Until 229 we may utilize fragments and an epitome of the excellent Cassius Dio; until 238 the valuable statements which another contemporary, Herodian, conceals beneath his vile rhetoric.

The period which follows, when the Roman Empire was in far more danger of complete collapse, in east as in west, than at the conven- tional date of 476, is the most obscure portion of all imperial Roman history. We should not have expected much historical writing during the actual time of breakdown; but the almost miraculous, if partial, recovery which ensued should have been celebrated by more able writers, who, incidentally, described the collapse. Of one such his- torian, Dexippus of Athens, we have some information, though the “fragments” are disappointing; other contemporary sources are little more than names. We must therefore turn to the later chroniclers.

The last great Latin historian, Ammianus Marcellinus, dealt with this period in detail; but this portion has been lost, and we have only a few backward references in preserved books. We try to use the so- called Augustan Histories, perhaps the worst source material the his- torian has attempted to utilize. Scores of dissertations have led almost entirely to negative results: the Augustan Histories do not come from the authors to whom they are assigned, they were not written in the days of Diocletian and Constantine, most of their documents and many of their sources are fakes. To the problem of their sources I can add one small but significant item: confusion of r and d in a proper name shows that one original source was in Aramaic characters.! The unfortunate historian must pick out facts which appear to be true,

1 Poll. Tyr. Trig. 2.

Tue Mip-Tuirp CENTURY 243

but how can he check them? Worst of all, for some of the most crucial years—from 244 to 253—the Augustan Histories themselves are lost.

If we turn to still later chroniclers, we soon discover that a few out- standing events are repeated over and over again in the epitomes. There is unique material in Johannes Malalas, but so chuckleheaded is he, even for his own day—the reign of Justinian—that we hesitate to believe him without further check.? The fullest and apparently best narrative is that of Zonaras, who lived in the twelfth century! Where, as in the case of Josephus, his source is preserved, Zonaras follows it closely. Our editions of Cassius Dio fill out from Zonaras the gaps in fragments and epitome. When, therefore, after the extracts from Dio we find the story going on without a break, we may be sure that Zonaras is following another good source; his immediate source was Peter the Patrician, from whom many direct quotations are pre- served.’ Peter’s source must have been contemporary and excellent; we can only call him Dio’s Continuator.

As for oriental sources, the vast Talmudical literature is yet to be properly digested; Syriac gives little not extracted from Greek, with the notable exception of the Chronicle of Arbela by Mshiha Zkha, who adds a few precious bits.‘ Armenian historians are translated by Langlois, who has used the modern system of transliteration and com- pels us regularly to retransliterate back to the pronunciation of the time when the great sound shift had not yet been completed.®

In recent years the value of these historians has been unduly de- preciated. For the collapse of the Arsacid kingdom we must use Aga- thangelus and Moses. Agathangelus, secretary to the first Christian king, Tiridates, wrote in Greek; comparison with the later translation into the native language exhibits only too plainly the curse of Ar- menian historiography—interpolation on a large scale. Even our pres- ent Greek edition is not free from later addition, but to the careful

2 Cf. Chronicle of John Malalas, translated from the Church Slavonic by Matthew Spinka, in collaboration with Glanville Downey (1940).

3U. P. Boissevain, C. de Boor, Th. Biittner-Wobst, Excerpta historica iussu Imp. Constantini Porphyrogeniti confecta (1903-10).

4A. Mingana, Trois sources syriaques (1908); E. Sachau, Die Chronik von Arbela (“Abhandl. berl. Akad.,"” No. 6 [1915]). ;

5 Collection des historiens anciens et modernes de l’ Arménie, 1 (1880); II (1869). A new edition is urgently needed.

244 A. T. OLMSTEAD

student the original and accurate narrative only stands out the more clearly.

Throughout the centuries, Moses of Chorene’s History has been “to the majority of his readers the most authentic and trustworthy book next to the Bible.” Recent criticism has been actually savage in its attacks.® A test which has attempted to set every alleged fact in its proper place in my own narrative manuscript of the later ancient oriental history gives the following results: The legendary period is worthless, though no better and no worse than similar Greek attempts, by which it has been strongly influenced. The narrative before the third century A.D. is a distorted picture, based ultimately on Greek sources. From that time onward it is good history, free from serious interpolation, in utmost contrast to the usual run of Armenian his- tories. There is no historical mistake nor is there any sure sign of bias which demands a date later than 440—the point at which the narra- tive closes. The data on the pagan religion, when tested by known facts about Sasanid Zoroastrianism, are of so trustworthy a character that a date of composition not too long after the conversion of Ar- menia to Christianity must be postulated.

Although of much later date, the Arabic historians have decided value. Since our study is confined to the relations of Persia and Rome, it will be unnecessary here to enter into the complicated question of sources. For our present purpose it will be enough to point out the general agreement of scholars that what we use to supplement the Greek writers has come to us, along with much legendary material, from late Sasanid chronicles written in Pahlavi.”

As the stream of papyri and inscriptions grows thinner, we are driven to the coins, whose value for aesthetics and for purchase is in direct reverse to their value for history. In this connection, Sasanid coins should not be neglected. The numerous Sasanid reliefs are of

6 Summarized by A. O. Sarkissian, Jour. Amer. Orient. Soc., LX (1940), 73 f.

7 Our foundation is Th. Néldeke, Geschichte der Perser und Araber (1879), an anno- tated translation of this portion of Tabari; other Arab historians give essentially the same story, rarely with authentic additions; cf., e.g., Barbier de Maynard, Les Prairies d’or (1861-77), Arabic text and French translation of Mas’udi.

8 F. D. Paruck, Sasanian Coins (1924).

Tue Mip-Tuirp CENTURY 245

\

extraordinary importance.’ Thus far, of contemporary inscriptions we have had only a few examples.'°

But Persia has at last relented and has given us a new inscription which equals the immortal Behistun Inscription of Darius the Great in its historical importance. Early in 1936 Dr. Erich Schmidt, director of the Iranian Expedition of the Oriental Institute, University of Chicago, discovered a badly damaged Parsik inscription on the east wall of the so-called Kaaba of Zoroaster, which faces the Nagqsh-i- Rustam cliff, in which are the tombs of this same Darius and of his next successors. In December of that year Mr. Donald McCown, tem- porarily in charge at Persepolis, uncovered for me a part of the in- scription and discussed with me his own copy. From the remarkably clear photographs made by the expedition photographer, Mr. Boris Dubensky, Professor Martin Sprengling, of the University of Chicago, published a transliteration and running interpretation in the American Journal of Semitic Languages, supplemented by an article in the Zeitschrift der deutschen morgenldindischen Gesellschaft. His articles were used by A. Christensen in the last volume of the Cambridge Ancient History. Sprengling assigned the inscription to Narseh I and thought the lists of place names marked the borders of the empire; Henning, in the Bulletin of the School of Oriental Studies, proved that the in- scription came from Shahpuhr I and that the list of place names indi- cated the route of his campaigns."

The position of the Parsik on a side wall of the structure had al- ready led us to expect, on the analogy of the shorter inscription of Shahpuhr at the near-by Naqsh-i-Rejab,” Pahlavik and Greek trans- lations on the west and back walls, and these were later uncovered by

°F. Sarre and E. Herzfeld, Iranische Felsreliefs (1910), for the region about Per- sepolis, to be supplemented by the excellent photographs taken by Mr. Boris Dubensky, of the Iranian Expedition of the Oriental Institute under the direction of Dr. Erich Schmidt, and by N. C. Debevoise, ‘‘The Rock Reliefs of Ancient Iran,’’ Jour. Near East. Stud., I (1942), 76 ff.

10 Collected by E. Herzfeld, Paikuli (1924).

1M. Sprengling, ‘‘A New Pahlavi Inscription,’”” Amer. Jour. Semit. Lang., LIII (1936), 126 ff.; ‘‘Zur Parsik-Inschrift an der ‘Kaaba des Zoroasters,’”’ Zischr. d. deutsch. morgenlind. Gesellsch., XCI (1937), 652 ff.; A. Christensen, Camb. Anc. Hist., XII (1939), 109 f.; W. B. Henning, ‘The Great Inscription of Sapur I,’”’ Bull. School Orient. Stud., IX (1939), 823 ff.

Herzfeld, Paikuli, p. 86.

246 A. T. OLMSTEAD

Dr. Schmidt. The photographs arrived at Chicago in the summer of 1939, and through the courtesy of Professor John A. Wilson, director of the Oriental Institute, I was permitted to study them. Publication of my own historical results has naturally been withheld until the for- mal publication by Professor Martin Sprengling, to whom publication of the inscription has been intrusted.'’

For the badly damaged Parsik and the almost complete Pahlavik original, as well as for the sections of value only to Iranian students, I must now refer to Professor Sprengling’s articles. Readers of Classi- cal Philology will, however, be more interested in the Greek version. This version is both more and less easy to decipher than appears at first glance. We are gratified to discover that one huge lacuna, where a repair block has fallen out, is only apparent; the break was there al- ready when the inscription was engraved; this is proved by the way in which the words straddle the break, for example, BAC! on one side and A€ WC on the other. In numerous other cases of apparent lacunae the pitting of the stone had already been effected by the elements, and these too are straddled.

Although such defects greatly detract from the looks of the record, they demonstrate an important historical fact. The native name, Qa‘aba Zardusht, simply implies a sacred structure connected with the Zoroastrian religion. Herzfeld argued that we should read Qabr Zardusht, ‘Tomb of Zoroaster,” and therefore identified the similar but more ruined structure at Parsagadae as the ‘“Tomb of Cambyses.”’ At any rate, since the latter structure was in the city founded by Cyrus and abandoned as capital after Darius, the early Achaemenid date of both should have been assured.

Sarre drew attention to a coin of the Fratadara dynasty which ruled Persepolis during the Hellenistic age, on which he believed that this very structure was represented. Above it was the Auramazda symbol, on one side the king poured a libation, on the other was a standard. Therefore, he argued, the monument was a fire temple.'4

Our new inscription devotes a third of its text to the establishment of fire temples by Shahpuhr; this fact would seem to confirm Sarre’s

13M. Sprengling, ‘‘From Kartir to Shahpuhr I,’’ Amer. Jour. Semit. Lang., LVII (1940), 330 ff.; ‘“‘Shahpuhr I, the Great: On the Kaabah of Zoroaster (KZ),’’ ibid., pp. 341 ff.; ‘‘Pahlavi Notes,’ zbid., LVIII (1941), 169 ff.

14 E. Herzfeld in Sarre and Herzfeld, op. cit., pp. 3 ff.; F. Sarre, Die Kunst des alten Persiens (1925), 14 f., Pl. 51.

Tue Mip-Tuirp CENTURY 247

explanation of the coin. The walls of our structure had suffered dam- age on three contiguous blocks; a repair block of irregular size had been inserted, and this, too, had fallen out before our inscription was incised. Again the architectural history points back to the Achaemenid period.

Our difficulties in decipherment are, however, greatly increased by injuries to the surface. In a large proportion of the inscription we can detect certain traces of the letters from the photographs only if we already know what should be there. Fortunately, the Pahlavik origi- nal is almost complete and very clear. We find special difficulty in the last third of the Greek version, where transliterations of Persian per- sonal names and of Persian official titles predominate, and here the Greek can be successfully read only by constant reference to the other versions.

Greek letters are those characteristic of the third century, especially the rounded sigma and epsilon and the W-shaped omega. In the be- ginning the letters were rather well carved, but the engraver soon be- came careless. There are a few cases of ligatures, especially at the end of the lines. Sentences are marked off by the triangular wedge, a reminiscence of the word-divider in the Achaemenid cuneiform.

The language is a fair sample of what was written in the third cen- tury. This is well shown in the spelling, where there is complete lack of consistency in the use of epsilon, iota, and epsilon iota. In one case the epsilon iota is used where the classical form has only epsilon. There is at least one case of crasis, kas for xai eis. At the end of line 4 and at the beginning of line 5 xai is repeated by error. Latin words might be expected in Greek of this date, but it is surprising to find so many this far east. These include xagréAdous, line 12, as far as I know its earliest use in Greek; Kaioap as title in lines 8f., 19, 24; voupeias, lines 45, 52; dvvwvns, line 65. Among the words not in Liddell and Scott are the Latin voupeias and xaoréAdouvs, eicévypador, line 44; Tpérrou, line 62; etyovpou, line 66. mupeta, line 38 and often, is new in the sense of “fire altar.’’ ¢7a8agdpov is used in line 64 for crabndopov. ywlodvXakos, line 66, is more exact than the common yafodiAaé. TaOvns, line 58, is a vulgar form for darvn. While duaéa as a Persian loan word for “wagon” is well known, auaéao7ov appears to be a trans- literation of a Persian word meaning “wagon horse.” éevtias, line 8, is cited in the eighth edition only from First Maccabees but is missing

248 A. T. OLMSTEAD

in the last revision. The formula xal of ‘Pwyato. Pikirrov Kaicapa avnyopevoay, lines 8f., is already found in Plutarch Galba 2, though naturally our inscription substitutes the more familiar Caesar for im- perator.

There is no obvious violation of Greek grammar, but the style is utterly non-Greek. The double re... . re is the lone attempt to give a “classical” tone. Comparison with the Pahlavik original proves it an extremely literal translation, quite as bad as Aquila’s notorious version of the Hebrew Bible. Thus we add an important example to the scanty list of secular works in translation Greek; that it is found so late is not without significance in a discussion of translation Greek in our Gospels.

For the extraordinary percentage of words which are in no sense Greek but are direct transliterations of the Pahlavik original, we have no parallel. Even Theodotion’s translation of the Hebrew Bible does not rate so high. As a result, large portions are completely unintel- ligible to the epigraphist familiar only with Greek. Naturally, this transliteration is of the utmost value to the student of Pahlavi in dis- covering the actual pronunciation of the mid-third century. This is the task of the philologist ; here we can note only that the Greek trans- literations of the personal and local names and of the various official titles were derived from oral dictation and not from the written origi- nal. Thus, for instance, the name of the god Mithra was still written mtr in both Parsik and Pahlavik, but the Greek has Meer, proving that the Middle Persian Meher was already in use. Naturally, this led to inconsistencies in the transliteration; Shahpuhr could be “Sapores,”’ “Sapore,” and “Sabur”; Artakhshatr is ‘“Artaxaros,” ‘“Artaxir,” and “Artaxeir.”’

It is indeed most fortunate that this article has been so long pre- vented from reaching the press, and thus the evidence of this superb inscription could be included. But the true heroine who made possible this article—the Ariadne who gave the clue to the labyrinth—has not yet been mentioned. All these years she has been perfectly well known to a little group of enthusiasts who have devoted their lives to an out- of-the-way corner of the historical field, and yet the professional his- torians have virtually ignored her. The reason becomes clear when we discover that this invaluable source, our one absolutely contem- poraneous narrative, is herself a perfect—fake!

Tue Mip-Tuirp CENTURY 249

Perhaps the strangest phenomenon in the modern writing of ancient history has been the almost complete ignoring of the pseudepigrapha, Jewish and Christian, by the secular historians. It is the stranger be- cause their dates and the thinly veiled names of their historical char- acters have long been known to the students of the pseudepigrapha. But they in turn have been little interested in secular history, save as it determined the meaning of their own documents. All that is neces- sary for the historian is to bring the two sets of facts together. This has been done briefly for the pre-Christian pseudepigrapha;' another article must deal with the post-Christian.

Such popularity as the pseudepigrapha have hitherto enjoyed has been mostly to the advantage of the Jewish apocalyptic, written gen- erally in Hebrew, rarely in Aramaic, but now preserved only in Greek or other translations. Their chief use has been in providing a background for the New Testament, though they are quite as valuable for secular history. Of far more importance for the history of the Greece and Rome familiar to us are the so-called Oracula Sibyllina. They are a terrible hodgepodge—pagan, Jewish, and Christian seemingly inextricably intermingled—yet their elements are now all worked out for us to use. They repel us by the Sibyl’s claim that Homer lied when he imi- tated her, by her limping, unpoetical hexameters, by her pretense of knowing the future when she knows only the past, but what she tells us of that past should never be ignored: she is always speaking as a contemporary.

From her we learn of the Galatians invading Greece and Anatolia, of the advance of Rome into the eastern Mediterranean. She rejoices over the downfall of Macedonia, of Corinth and Rhodes, and of Delos. She exults when Italy is tearing herself to pieces in the Social War. One Sibyl urges Mithradates on to his massacre of Italians in Asia; another, more canny, warns Greek cities to beware. Here is a new angle on Antony and Cleopatra.

The well-advertised Pax Romana means to our Sibyl only oppres- sion, which one day will be terribly revenged. Reference after refer- ence in the notes to Suetonius and other late authors should not blind us to the fact that her evidence is contemporary. If she tells us that Domitian killed Titus, that Nerva was assassinated, her testimony

1 A. T. Olmstead, ‘‘Intertestamental Studies,’’ Jour. Amer. Orient. Soc., LVI (1936), 242 ff.

250 A. T. OLMSTEAD

should be considered. She gives us a new picture of Hadrian. From oracles about Septimius Severus and his successors numerous details may be added or straightened out. The series goes down into the Muslim period.'®

Immediately after the oracles dealing with the Severan period comes the Thirteenth Sibylline. It is significant that the recently published twelfth volume of the Cambridge Ancient History cites it in the Bib- liography only under the wars between Persia and Rome but shows no trace of use in the text. Yet in 1869 Alexandre had already given so exact an interpretation that Geffcken and Rzach have found little to change.!”

By our use of the hitherto unemployed oriental sources, including the great inscription of Shahpuhr, and still more through the Thir- teenth Sibylline, it is now possible to re-write the history of the mid- third century. An unacknowledged continuator of the Twelfth Oracle, our pretended Sibyl begins her “great word,” which “God orders to sing” by observing “how God to kings gives power and takes”; God alone ‘‘determines bounds for them, both life and baneful death.” She is enjoined to “bring tidings to kings of royal rule” and is shocked to behold how “all perish, child and ancient, who gives laws to assem- blies.”” The earth is filled with ‘‘many wars and battles, homicides, famine and pestilences, earthquakes, mighty thunderbolts, and many ways of the Assyrians throughout the whole world, pillage and robbing of temples.’’!®

She is sketching briefly the breakdown of imperial rule which fol- lowed the assassination of Alexander Severus, the last representative of oriental rule over the Roman Empire, though the bowmen from Osrhoene, the vassal kingdom of Edessa, vainly attempted to con- tinue the tradition by putting forward as imperial claimant their trib- une, Titus Quartinus.'? Included are also the savage reign of the

16 Bibliography: M. Rzach, ‘“‘Sibyllinische Orakel,’”’ in Pauly-Wissowa, R.-E., II (2d ser., 1923), 2165-69, to which add M.S. Terry, The Sibylline Oracles (1899), a useful verse translation. Best edition: Joh. Geffcken, Die Oracula Sibyllina (1902). Best dis- cussions: Geffcken, Komposition und Entstehungszeit der Oracula Sibyllina (‘‘Texte und Untersuchungen zur Gesch. der altchristlichen Literatur,” Vol. VIII, No. 1 [new ser., 1902]); Rzach, op. cit., cols. 2117-65.

17 Geffcken, op. cit., pp. 59 ff.; Rzach, op. cit., cols. 2158 ff.

18 Orac. Sibyl. xiii. 1-12.

19 Herodian vii. 1. 9 f.; 2. 1; Capitol. Maximin. 11; Poll. Tyr. Trig. 32; cf. CIL, XIII, 677a, name of Osrhoenians chiseled out.

Tue Mip-Tuirp CENTURY 251

Illyrian Maximinus (235-38), his harsh treatment of the senate, the revolt of the two Gordians in Africa and that of Maximus and Bal- binus in Italy, the forced association of the young Gordian III (238- 44), and the subsequent murder of the older emperors.”°

The aged founder of the Sasanid empire, Artakhshatr, took advan- tage of the complete disorder to invade the Roman Empire. Shortly after the accession of Maximinus, Nisibis and Carrhae were cap- tured.24 To our Sibyl, Mesopotamia is the “whole world,” through which passed the “‘many ways of the Assyrians,” bringing “‘pillage and robbing of temples.”

Artakhshatr celebrated his victories with new coins which intro- duced the crown surmounted by the globe—Sasanid equivalent of the mural crown—and thus set the model for his successors.”” Maximus determined to win back the conquests but was murdered just as he was about to proceed against the Persians.”* Shortly thereafter, Ar- takhshatr associated with himself his son Shahpuhr (241-72) as king of Airan. On coins the youth, facing his father, wears the diadem, earflap, and crested helmet which distinguish the crown prince.*4 Somewhere about this time, Hatra, the powerful Arab city in the des- ert west of Ashur, which had successfully resisted both Trajan and Severus, was taken; from that day the well-preserved ruins have never been inhabited. But the tale of the last king’s traitor daughter and her terrible punishment have remained a stock theme in Arabic lit- erature.”®

Shahpuhr’s coronation ceremony may yet be seen on the rock wall of the Auhrmazd sanctuary just south of Stakhr, where his father carved the first Sasanid relief. On his own horse, in Pahlavi and Greek,

20 E. Hohl, “Julius 526, 527,” in Pauly-Wissowa, R.-E., X (1917), 852 ff.; A. Stein,

“Clodius 50,” R.-E., IV (1901), 88 ff.; ‘‘Caelius 20,” R.-H., III (1899), 1258 ff.; P. W. Townsend, ‘‘The Administration of Gordian III,” Yale Class. Stud., IV (1934), 59 ff.

1 The date is fixed by the disappearance of Roman coins from these two cities during the reign of Maximinus, though they reappear under Gordian III (G. F. Hill, Greek Coins of Arabia, Mesopotamia, and Persia [1922], pp. xcii, cix). Syncell. 681, using Euse- bius, Zonar. xii. 18, and Capitol. Gordian. 23. 5, wrongly date the outbreak of the Persian War to 241. Capitol. op. cit., 26. 6, to Nisibis and Carrhae adds Edessa, Artaxanses, and Antioch, the last also in the unauthentic letter to the senate (27. 5), but he is wrong. The conquest of Antioch in 241 is still accepted by Townsend, op. cit., p. 127.

* Paruck, op. cit., p. 77. 23 Capitol. Maz. et Balb. 13. 5.

*4 Paruck, op. cit., pp. 315 f.; ef. Tabari, pp. 25 ff.; Masudi, II, 159 f.; Elias Nisib., p. 61.

* Tabari, pp. 33 ff.

252 A. T. OLMSTEAD

we may read: “The image is this of the Masdasnian God Saporos, king of the kings of the Ariani and Anariani, from the stock of the gods, son of the Masdasnian god Artaxaros, king of the kings of the Ariani, from the stock of the gods, grandson of the god Papakos, king.’”’> We trace the development of the titulary with the expanding empire, until Shahpuhr may claim rule of the non-Aryans; the use of Geos proves that there was actual survival, in however attenuated form, of the ancient oriental concept of king-godship. Virtually the same titulary is found in the new inscription, whose breaks are thus safely filled.?’

In founding the Sasanid empire, Artakhshatr had destroyed the Parthian, whose royal family was Arsacid. The ruling family in Ar- menia was also Arsacid, and this fact automatically made its king Chosrhoes the bitterest of enemies to the Sasanids and friend to the Romans. Renewed Sasanid invasion of Mesopotamia, which a weak- ened Rome could not protect, brought the danger close home; and it was presumably at this time that Chosrhoes added to his regular troops Alans and Iberians and opened the “Gates of the Alans,” Dariel in the Caucasus, to permit barbarian attacks on the Persians.”*

To meet these barbarian attacks Shahpuhr must turn to his north- ern borders. We hear from Mshiha Zkha of a war in his first year against the Khuarazmians; this much-later people must be substi- tuted for some earlier people, perhaps the Zuari, whose tower is men- tioned by Agathangelus. The Medes of the mountains were defeated in a great battle. From there he attacked the Gelians, Dailumians, and Gurganians, who inhabited the mountains near the last sea, the Caspian. In Shahpuhr’s own list we find Guarathranos, king of the Gelani, probably the native king installed after the conquest; the Gur- ganians—Parsik Gurgan and Pahlavik Gurkan—are found among the satrapies. The more easterly conquests, suggested by this list, do not at present concern us.”°

26 Sarre and Herzfeld, Felsreliefs, pp. 92 ff.; Herzfeld, Paikuli, p. 86.

27 Cf. Sprengling, ‘‘From Kartir,’’ p. 333.

28 For this period Armenian chronology is dubious, and our only guide is probability. Agathangelus (10) places this in the year after the end of the Parthian empire. Moses Chor. (ii. 71) appears to do the same, then states that Chosrhoes sent news to Philip; yet in ii. 72 the war is against Artashir or Artakhshatr.

29 Mshiha Zkha, pp. 33, 110 (Mingana); p. 64 (Sachau). More easterly conquests: Sprengling, ‘‘From Kartir,”’ p. 337; “‘Shahpuhr,”’ pp. 344 ff. It should be noted that no

THe Mip-Tuirp CENTURY 253

Shahpuhr’s absence gave opportunity for Chosrhoes to ravage As- syria and also Babylonia up to the gates of Ctesiphon, the Sasanid capital. A return attack by the shah was beaten off. Chosrhoes re- turned to Valarshabat and pleased his subjects by sacrificing at the Seven Altars the national offering—white mules, horses, and sheep— with the gift of one-fifth the booty, silk robes and gold and silver ves- sels enriched with precious stones.*°

“Then an uprising of the industrious Persians there shall be, In- dians, Armenians, and Arabs together’: Shahpuhr had returned from his eastern frontier. ‘‘And to them again a Roman king, insatiate for war, leading spearmen against Assyrians, shall approach, a young Ares; unto Euphrates, deep flowing, silvery, shall warlike Ares stretch forth his deadly spear.’’*!

Gordian came east in 242, accompanied by the praetorian prefect, his father-in-law Timesitheus, and was met by the philosopher Ploti- nus.** Contributions of supplies would be expected; the senate and people of Palmyra, in October of this year, suddenly bethought them- selves of Julius Aurelius Zenobius, who is also Zabdila, how he had been general of the colony on the visit of the god Alexander Caesar and had at great expense provisioned the vezillationes. Now at this late date they decreed a statue in his honor, doubtless hoping that again he would show his munificence.**

Coins of Edessa with the busts of the emperor and of his consort indicate that for some time the colony was military headquarters. More numerous coins prove that during the visit the Osrhoenian king- dom was restored in the person of Abgar Phraates. On the reverse we see the youthful emperor seated and receiving the bearded, oriental- garbed king, bearing the statue of Victory in honor of the local con- quests for which he has been given the title. Then Abgar is seen alone,

conclusion can be drawn from the alleged list of independent states in Pollio’s life of Valerian. Velenus, king of the Cadusii (2. 1) is absurd, for they are good Persians; Artevasdas, king of the Armenians (3. 1) is the satrap who, after the collapse of the kingdom, brought with him the infant Tiridates for refuge with the Romans; against the Bactrians, Hiberians, Albanians, and Tauro-Scythians, who refused to accept Shah- puhr’s letters (4. 1), we may place the mention in Shahpuhr’s list of Iberia, Albania, Herat (Bactria), and Sakistan-Sacastene, the Scyths.

30 Agathangel. 10 f.; Mos. Chor. ii. 71. 31 Orac. Sibyl. xiii. 13-18. 32 Porphyr. Vit. Plotin. 3.

33 CTS, III, 3932; IGRR, III, 1033; J. B. Chabot, Choir d’inscriptions de Palmyre (1922), p. 62.

254 A. T. OLMsTEAD

on horseback; on the latest coins he is depicted only by the royal bust. At the same time, near-by Carrhae was recovered; for the last time the metropolis minted Roman coins.*4

The young emperor won a series of skirmishes, then went on to Resaina, ‘Head of the Spring,” the source of the Chaboras River, once the location of the Mitannian capital, Washshukkanni. Taken under Lucius Verus, its colonial coinage started with Caracalla, who pre- sumably garrisoned there the third Legio Parthica and built the city wall recently discovered by Dr. Calvin McEwan of the Oriental Insti- tute, though the dystyle temple pictured on the coins is yet to be found. Shahpuhr was already attempting its siege; in a great battle he was defeated. The recovery of Nisibis (Septimia Colonia Nesibis) was the natural consequence; and again the colony struck coins, as did, for the only time in its history, the colony Aurelia Septimia Singara. Gor- dian followed down the course of the Chaboras to the Euphrates and then along that river, aiming at Ctesiphon (243).*° So great was the danger that Shahpuhr hastily collected troops to defend his capital; among them was a young pagan named Ahadabuhi, “come to Ctesi- phon on account of the war against the Romans,” who later was to become Christian bishop of Arbela.*®

Timesitheus was poisoned by Philip the Arabian, who took his place as praetorian prefect; his brother Priscus had already held this office; Philip was promoted from the vice-prefecture.*” Difficulties in secur- ing food were exaggerated by Philip, who thus compelled Gordian to retreat to Zaitha, twenty miles down the river from Circesium and up stream from Dura,** where he was compelled to share the throne with Philip, toward the end of 248; soon after, in February of 244, Philip slew Gordian at Zaitha; “betrayed by his colleague,” declares the Sibyl, “‘he shall fall down in the ranks, smitten by the gleaming iron.’’*

34 Hill, op. cit., pp. 111 ff.; 89 f.

35 Hill, op. cit., pp. 121, 125 ff., 134 ff.; Amm. Mare. xxiii. 5. 17; Capitol. Gord. 26. 5 f.; Zosim. i. 18. 3; Zonar. xii. 18; Syncell. 681; Eutrop. ix. 2. 2; Johan. Ant. 147.

36 Mshiha Zkha, pp. 36, 113 (Mingana); 66 (Sachau).

37 See the doctoral dissertation of Laurence L. Howe, The Pretorian Prefect from Commodus to Diocletian (1941).

38 Eutrop. ix. 2. 3; Amm. Mare. xxiii. 5. 8.

39 Orac. Sibyl. xiii. 17 f.; Amm. Mare. xxiii. 5. 7 f., 17; Capitol. Gord. 28 ff.; Victor 27; Eutrop. ix. 2 f.; Sext. Ruf. 22; Zosim. i. 18 f.; iii. 14. 2; 32. 4; Oros. vii. 19; Syncell. 681; Johan. Ant. 147; Malal. (ed. A. Schenk, Graf von Stauffenberg) 62; Zonar. xii. 18; ef. K. F. Lehmann, Kaiser Gordian III (1911); Townsend, loc. cit.; P. von Rohden, ‘Antonius 60,’’ Pauly-Wissowa, R.-E., I (1894), 2619 ff.

Tue Mip-Tuirp CENTURY

Shahpuhr tells a very different story:

And when first over the kingdom of the peoples I was established, Gordianos Kaisar from all the dominion of the Romans a force of Guththi [Goths] and German peoples collected. And into Assyria, against the people of the Aryans and us he made an inroad. And into the mountains of Assyria, in the Mesichise from opposite a great war arose. And Gordianos Kaisar was killed, and we annihilated the army of the Romans. And the Romans proclaimed Philippos as Kaisar. And Philippos the Kaisar came for supplica- tion. And offering in return for the life of his friends five hundred thousand denarii, he gave them to me, and for tribute to me it shall be. And on account of this we named Misichen Peros-Sabur.*°

If we accept Shahpuhr’s word, Sprengling is indeed justified in

writing: “It was fought on what Shahpuhr considered his ancestral territory, ....and it was to Shahpuhr’s mind a successful defense against unprovoked, aggressive Roman attack and attempt at con- quests beyond recognized Roman borders.’”’ Actually, these had been “ancestral” only since the “unprovoked, aggressive’ Persian attack of 238. It ended in a complete defeat of the Roman army with the death of its Caesar on the field of battle, Philipp’s surrender at discretion, the payment of a stiff ransom not miscalled reparations, and the Roman empire on the eve of its thousandth anniversary tributary on its eastern boundary to its new rival world-empire Iranshahr There is no indication of treason on the part of Philipp; nor of his having been at fault in the defeat; nor of any particular weakness shown by him in concluding peace on the best terms possible, when no other recourse was open to him; nor of his having ceded Nisibis; least of all of his having had a hand in the death of Gordian, who is here clearly repre- sented as having fallen in open battle.”

Undoubtedly, this is what Shahpuhr would have us believe, but is it true? It is no detraction from the value of this splendid new dis- covery when we point out that, while the shah was a contemporary and in a unique position to know the truth, he was also determined to make out for himself the best case possible, whatever the cost to the truth. That Philip killed Gordian is not only the unanimous agree- ment of a long line of Greek and Roman chroniclers, backed by the statement that the tomb monument was in existence a century later; it is proved by the reference of the contemporary Christian Sibyl, who had far less reason to slander the protector of her fellow-religionists than Shahpuhr.

40 Kaaba ins., Gr. 6-10. 41 Sprengling, ‘‘Shahpuhr,”’ pp. 363 f.

256 A. T. OLMSTEAD

We cannot set this “victory” into the account of the campaign given by the classical writers. The reason is given by the name of the city at which the battle was fought. Its name was Mishikaman, Greek ‘‘Mesichise”’ or ‘‘Misiche,”’ which the king changed into Peroz Shahpuhr, “Victory of Shahpuhr.”’ As Sprengling has seen, this is the “modern Faysabur, about a hundred and fifty miles north of Mosul.” There is no difficulty in drawing the conclusion that this is an other- wise unknown battle in connection with the Armenian campaign and has nothing whatever to do with the events hitherto mentioned.

Later on, Shahpubr himself gives away his case. In the list of cities captured on his next campaign he begins with Anatha, modern Ana on its palm-covered island in the Euphrates. This, then, marks the southeastern limit of Gordian’s advance toward Ctesiphon, and it re- mained in Roman hands for the next ten years. Actually, the frontier fell back to the advantage of the Romans!

From literary sources we had known that Philip made a peace with the Persians.*? Shahpuhr claims a huge payment which he says meant tribute. No doubt the payment is true; it is by far the earliest men- tion of those “contributions,” so familiar in later centuries. In theory Rome paid a subvention to Persia for the support of the garrisons at the northern passes which protected both empires from incursions of the northern barbarians; in practice the Sasanids treated these yearly payments as tribute from vassals and employed the funds as they saw fit.

Now that we have learned how Philip purchased a peace, we are amazed to find him daring to assume in 244 the title of Persicus Max- imus and in 245 that of Parthicus Maximus.** Perhaps a reason may be found. The generally trustworthy, though twelfth-century, Zona- ras states that Mesopotamia and Armenia were abandoned by the peace but that when the Romans learned of the surrender they were so upset that Philip broke the peace and a little after recovered the lands.** This statement has been generally discredited, but it must be true. Philip did recover Mesopotamia. The use of two different titles in his first two years must refer to two different expeditions in the two years

42 Zosim. i. 19. 1; Syncell. 683; Zonar. xii. 19.

43 CTL, III, 4634; 10619; 14354, 6; VI, 1097; Dessau, 506 f.

44 Zonar. xii. 19.

Tue Mip-Tuirp CENTURY 257

244 and 245. Further proof that Philip did hold all Mesopotamia is found in the coins of Nisibis, called Julia in his honor, and in inscrip- tions which give his brother Priscus the title of prefect (eparch) of Mesopotamia.** But once more the best witness is our Sibyl:

“Straightway then shall rule a purple loving warrior, from Syria ap- pearing, terror of Ares, and his son Caesar, and oppress all the earth. One name is to them both, on one and twenty, five hundred is placed.” Thus, not too obscurely, our Sibyl indicates the two Philips by the numerical value of their name. “But when these in war shall rule and become lawgivers, there shall indeed be rest from war, a little, but not for long.”*® Zonaras might well have had our Sibyl in mind.

What Shahpuhr thought of this broken peace he has made clear: “And the Caesar again lied and unto Armenia he did injustice.’’*” That this refusal to live up to treaty obligations took place almost im- mediately after the peace is indicated by the Sibyl, who goes on at once: “But when the wolf makes faithful oath against the white toothed dogs, but then deceives, tearing the woolly sheep but oaths casts off, then to strife unlawful of overweening kings in wars, Syrians shall perish terribly, Indians, Armenians, Arabs, Persians, Babylo- nians, one another through hard fighting shall destroy.”’**

Earlier Latin poets had often referred to the Armenian sheep; the term ‘“‘woolly sheep” was therefore appropriate for our oracle-monger to apply to the Armenians themselves. The later native historians place the fall of the Arsacid dynasty in Armenia immediately after that of the Arsacid dynasty in Parthia, and older scholars have fol- lowed their testimony.*® Our earliest historian, Agathangelus, makes it clear, however, that the capture took place at least a dozen years after that date, that is, not earlier than 238; the confused account of Moses of Chorene mentions Philip; Zonaras placed it after Volusianus, about 253; and this has generally been accepted in recent years.®° The evi-

4 Zosim. i. 20. 2; Le Bas-Waddington, III, 2077 f.; Dessau, 1331, 8847; JGRR, III,

1202; J. Cantineau, Inventaire des inscriptions de Palmyre (1930), ITI, 22; Hill, op. cit., pp. 122 ff.

46 Orac. Sibyl. xiii. 21-27. 47 Kaaba, Gr. 10. 48 Orac. Sibyl. xiii. 28-34. 49 So, e.g., G. Rawlinson, The Seventh Great Oriental Monarchy (n.d.), I, 50 ff.

50In 252; M. Besnier, L’Empire romain de l’événement des Sévéres au Consile de Nicée (1937), p. 177, who makes Shahpuhr appoint Artavasdas as successor of Chos- rhoes.

258 A. T. OLMSTEAD

dence of the Sibyl, supported indirectly by that of Shahpuhr, brings it close to the year 245.

Abandoned by his Roman ally Philip, Chosrhoes fought bravely on and with success until he was assassinated by a pretended fugitive, Anak, of his own Arsacid house, whose immediate death robbed him of his promised reward. Shahpuhr invaded Armenia and overcame the desperate resistance of the satraps, who were ultimately forced to flee into Roman territory; among them was Artavasdas, who brought with him Tirdat (Tiridates), the infant son of Chosrhoes.*!

In our inscription, immediately after the titularies, are given the countries which belonged to the regularly organized empire. We have, of course, Persis, Parthia, Chusistene, Assyria, Adiabene, Arabia, and Adurbadene, classical Media Atropatene, and