We have already seen that the list of cities provided in the SKZ inscription of Shapur I at Naqsh-i Rustam, near Persepolis, allows some consideration of the military organization of the middle Euphrates and that it complements the evidence of the Dura and Euphrates papyri. The SKZ also allows us to go geographically beyond the area containing most of the fortifications mentioned in the papyri to consider how the middle Euphrates and lower Khabur region was located in the broader military and administrative organization of Mesopotamia and Coele Syria in particular.
The inscription refers firstly to the defeat of a Roman army numbering 60,000 men at Barbalissos on the Euphrates before listing the individual cities that were captured as a part of the campaign in Syria and Cappadocia:
(10) The nation of Syria and whatever nations and plains that were above it, we set on fire and devastated and laid waste. And in that campaign <we took> (the following) fortresses and cities from the nation of the Romans: (11) the city of Anatha with its surrounding territory,112 Birtha Asporakos, (byrt ’arpn (Parthian only), Sura, Barbalissos, Hierapolis, (12) Beroea, Chalcis, Apamea, Rephanea, Zeugma, Ourima, (13) Gindaros, Larmenaza, Seleucia, Antioch, Cyrrhus, (14) another Seleucia, Alexandria (Alexan-dretta), Nikopolis, Sinzara (Larissa), Chamath (Epiphania),
(15) Aristia (Arethusa), Dichor, Doliche, Doura, Circesium, Germanicia, (16) Batna, Chanar and in Cappadocia, Satala, Domana, (17) Artangil, Souisa, Suid, Phreata, a total of thirty-seven cities with their surrounding territories.113
The direction and strategy of the Persian forces as they made their way up the Euphrates to Antioch has been described and debated ever since the discovery of the SKZ in the 1930s. Emphasis is often placed on the order in which the cities appear in the inscription and how this might indicate the route taken by the Persian army as it marched through Syria, with considerable work done to identify the sites captured.114 The first city listed in the SKZ is Anatha, which was discussed earlier and is known to have been an island fortress belonging to the Persians at the time of Julian’s invasion in 363. Little is known of its significance as a Roman fortification in the mid-third century; however, it may have been controlled by Palmyra from the early first century AD before becoming part of the complex system of fortifications on the Euphrates under Septimius Severus. As noted earlier, a number of smaller fortifications to the north and south of Anatha appear to have formed part of its military jurisdiction.
Following the capture of Anatha, Birtha Arupan and Birtha Asporakos, Sura is listed as the next city captured by Shapur I.115 The obvious omissions between Anatha and Birtha Arupan are Dura and Circesium, which are listed well after the capture of Antioch and cities within its vicinity. This has been the subject of considerable speculation and is dealt with below. Sura later became important as a Roman fortification at the head of the Strata Diocletiana, which ran from Damascus to the Euphrates through Palmyra in the early fourth century AD.116 Little is known of its fortification prior to this time. In the first century AD, Pliny referred to Sura as the place where the Euphrates turned east and flowed into the Palmyrene desert.117 It was noted in Chapter 2 that there is effectively no evidence for Sura as a Roman fortification as early as 75.118 It is possible that Sura was garrisoned by Legio XVI Flavia Firma from as early as the Severan period, but this too is speculative.119 Interestingly, Sura is not mentioned in any of the papyri from Dura, while epigraphic evidence from Dura attests to the presence of vexillations of Legio XVI Flavia Firma from c. 211-250, claimed by some to be garrisoned at Sura at this time.120 If Legio XVI Flavia Firma was the garrison of Sura during this period it is surprising that Sura is not mentioned in any of the papyri.
Sura does appear to have been an important bridge crossing in the middle of the third century AD, this being indicated by a reference in Oracula Sibyllina XIII. Sura was considered important enough in the middle of the third century for the author of the oracle to specifically describe it as the location through which Mariades passed as he fled to Persia, and it has been argued that Sura was the natural place for someone to pass through when leaving western Syria and heading to Persia.121 After leaving Antioch and arriving at the Euphrates, the crossing at Sura placed traders and troops on the left bank of the Euphrates along which they made their way to the Khabur confluence and then along the Euphrates towards Persia. From the Khabur confluence it would also be possible to travel north along its banks to reach Nisibis in Mesopotamia. The other major crossing of the Euphrates at Zeugma allowed access to Armenia, Osrhoene and Mesopotamia. Once the Euphrates was crossed at Sura, Circesium was the last Roman fortification on the left bank of the Euphrates in the fourth century, but in the middle of the third century it was only weakly defended.122 If there was a permanent crossing at Sura it is likely that a fortification was located there; because of this it may well have held Legio XVI Flavia Firma in the third century.
Following Sura, the next city listed in the SKZ is Barbalissos.123 Bar-balissos is mentioned only briefly in the surviving sources before the fourth century. Ptolemy simply described it as a town in Chalybonitis on the Euphrates, while in the late third/early fourth century AD it became well known as the site at which the Christian martyr Bacchus was killed.124 It was here that the SKZ declared a major victory over the Roman army, which saw a force of 60,000 men destroyed in 252/253.125 The defeat of the army at Barbalissos was one of the highlights of Shapur’s campaigns in Syria as it is described separately and before the long list of cities and fortifications captured, while the capture of Barbalissos itself forms part of that list. While there is no mention of this defeat in the Roman sources, and it is probable that the size of the defeated Roman army is exaggerated in the SKZ, a significant defeat should not be discounted as the Persian army appears to have marched without resistance to the north and south of Bar-balissos after the battle.
It is argued by Potter that the presence of a Roman force at Barbalissos indicates that Trebonianus Gallus was preparing for an offensive against
Figure 3.12 A cornfield on the right bank of the Euphrates close to the site of ancient Sura. The Euphrates today, as in antiquity, supports extensive agricultural production.
Persia in 252/253.126 While the SKZ’s claim that the Roman army numbered 60,000 is probably an exaggeration, Potter argues that a significant Roman force on the middle Euphrates would have taken too long to organize if it was a defensive operation, given the rapidity with which Shapur’s forces attacked.127 According to Potter, the defeated Roman army had originally been organized to deal with difficulties in Armenia, but when the Persians invaded this force marched to meet them on the Euphrates. Barbalissos was located on the right bank of the Euphrates and at the precise location where the river bends closest to Antioch. It lay approximately 150 km due east of Antioch and 100 km south of Zeugma.128 The location of Barbalissos made it an appropriate location at which to organize an army to protect Antioch and other key Syrian cities such as Apamea, Beroea and Cyrrhus. Almost a century earlier, this section of the river was the scene of intense fighting between Roman and Parthian forces prior to Lucius Verus’ invasion of Parthia. It is likely that a large component of the army that Shapur defeated at Barbalissos was comprised of troops withdrawn from fortifications further east and south, Dura Europos probably being one of them.
One of the most debated issues regarding the list of cities in the SKZ is the order in which Dura and Circesium appear. These fortifications lay
Figure 3.13 The Assad dam at Meskene near the site of ancient Barbalissos (modern Balis).
Between Anatha and the two Birthas, but they were not listed immediately after Anatha, appearing instead towards the end of the list of Syrian cities captured in Shapur’s first Syrian campaign. As the archaeological evidence for the Sasanian capture of Dura points to a date of 256/257 for its capture and the literary evidence indicates a date of 252/253 for the beginning of the campaign at Anatha, Rostovtzeff proposed that Dura fell twice, briefly in 253 and permanently in 256/257.129 Sprengling attempted to identify Birtha Arupan as Dura, the next city on the SKZ in the Parthian version, so as to fit with Rostovtzeff’s earlier idea; this is now thought to be incorrect.130 The belief that the city was taken twice was later quite convincingly refuted, but new readings of middle Persian dipinti from the frescoes of the Dura synagogue, together with analysis of other material, indicate that the city probably was occupied briefly before its final capture in 256/257, and as part of Shapur’s first campaign.131 This conclusion resurrects the old problem of the SKZ only listing the city once, in conjunction with Circesium, at what appears to be a considerable time lag from the first stage of the campaign. It is presently an insoluble question as to why Dura was not listed between Anatha and Birtha Arupan if it was captured at this time, and there is no archaeological evidence from Circesium to throw light on the subject. If Dura was captured in 252/253 after the capture of Anatha are we also to assume that Circesium was captured but not listed until later? It is possible that the SKZ, which was ultimately a work of Sasanian propaganda, only listed cities once because to list them twice would hint at some type of failure.
World History
