Hatra

Before Ardashir undertook any military activity directed at Roman Mesopotamia, he attempted to secure the western portions of his empire by dealing with Armenia and Hatra. Potter points out that neither Mesene in the Persian Gulf nor Adiabene in northern Mesopotamia appear on the SKZ, which may indicate that they were also a problem for Ardashir at this time.10 In 229, only a few years after his investiture as Shahanshah, Ardashir attacked the desert kingdom of Hatra, which Dio claimed he planned to use as a forward base from which to attack Roman possessions in the East.11 This was the first of two attacks Ardashir directed at the city, the second being approximately a decade later in 239/240. Hatra was abandoned by the Persians after its capture in 240 and not used as a base from which to attack the Romans, which casts some doubt on Dio’s claim.12 The motive for the first Persian attack on Hatra is more likely to have been part of Ardashir’s consolidation of victory over the Parthians as he mounted an expedition against Armenia and Media before the attack on Hatra.13 This was probably linked to Rome’s provision of support to both Armenia and Hatra. The second siege of Hatra was part of a broader campaign that saw the capture of Mesopotamia in 237/238 and an attack on the Euphrates in 239. By this time, there appears to have been a regular Roman military presence at Hatra as epigraphic evidence shows that an auxiliary cohort was there in the mid-230s and perhaps earlier.

A siege of Hatra was a major undertaking as the city’s desert location afforded it great protection. The necessary provisions to conduct a siege over any significant period of time were also very difficult to procure. Furthermore, the city was well defended and contained an ample internal water supply.14 Ardashir probably invested considerable resources in the first unsuccessful attack on Hatra, but the details of the siege reported by Dio are limited. The size of the Sasanian force and the state of the defences at Hatra at the time of the first Sasanian attack are not known. A number of scenarios can be conjectured on the basis of archaeological evidence and the more detailed surviving account of Septimius Severus’ attempts to take the city approximately 30 years earlier. Severus made two unsuccessful attempts to capture Hatra at the end of the second century and the emperor Trajan also unsuccessfully attacked the city in 117.15

Ardashir was soon to experience the difficulties associated with the prosecution of a siege of Hatra. Archaeological investigation of sections of the walls at Hatra drew Gawlikowski to conclude that ‘the whole system constituted a formidable triple barrier extending in depth for about 30 metres’.16 The defences of the city Ardashir encountered were probably much the same as those Septimius Severus encountered 30 years earlier as it is thought they were originally constructed in the middle of the second century AD.17 There also appears to be considerable evidence at Hatra for the successful Sasanian siege of 239/240.18

We are not informed of the length of the first siege of Hatra by Severus but Dio claimed that it accomplished nothing, whereas the second siege lasted 20 days and succeeded in breaching one of the walls.19 The second siege had been better planned and supplied than the first, but Severus still suffered major losses.20 Dio stated that Severus was within sight of his goal of capturing the city when he called off the siege in the hope that the Hatreni would come to terms - but this did not happen.21 Severus’ discontinuation of the siege was, according to Dio, the result of a mutiny of European troops who were part of the besieging force.22 Dio and Herodian both claimed that the surrounding desert and unhealthy climate only allowed a short siege, which was probably an important factor in the cause of the mutiny.23 Both authors, particularly Dio, attributed the failure to poor military discipline, but there were clearly considerable advantages for the defenders of the city.

The reasons for Ardashir’s attack on Hatra were twofold. First, the Romans had shown interest in capturing the city over the previous century. Second, it was probably part of a consolidation of power in the wake of defeating the Parthians because Hatra seems to have resisted aligning itself with the Sasanians.24 The Romans took advantage of this by supplying an auxiliary cohort, Cohors IX Maurorum Gordiana, to the city c.235.25 The duration of this cohort’s presence at the city is uncertain. Sartre suggests that Hatra became part of the Roman Empire from 217 or even as early as 198, but there is no direct evidence for this.26 While the evidence does not allow conclusive statements on the size and duration of the Roman presence at Hatra, it is an indication of Roman power and influence at the city, this undoubtedly being a concern to Ardashir in the early years of his reign.

The Arabic historian Tabari asserted that at the time of Shapur I’s attack on Hatra, c. 239, a noble named Daizan was based there and was known as the king of Mesopotamia.27 Daizan controlled the tribes in the area, according to Tabari, and his rule extended as far as Syria. It is probable that Daizan was a relative of Sanatruk II, or even Sanatruk himself who is referred to in another Arabic source of the early eleventh century. Sanatruk II came to power at Hatra by October 231 and was king when the city fell to the Sasanians in 240/241.28 Hatra’s territory, therefore, extended towards the Euphrates in the west. The potential for Hatra to assist invading Roman forces, now that it was well-disposed to the Romans, was also a problem for Ardashir. This is reflected in Tabari’s report that Shapur’s motive for attacking and destroying Hatra was its treachery while he was away in Chorasan dealing with problems on his own eastern frontier.29

Ardashir took a risk in mounting an attack on Hatra at this time. The strength of his forces for the siege is not known, but they were unable to achieve their task. The Sasanian Persian domination of Iran was still in its infancy and at the time Ardashir attacked Hatra, he still had much to do in order to subdue elements of opposition in Armenia. The Persian attack on Hatra may also have been the catalyst that drove Hatra to request military assistance from the Romans. Dio reported that as a result of his losses at Hatra, Ardashir retreated to Media, but this may have been associated with putting down elements of resistance there.30

In light of the above discussion, Dio’s claim that Ardashir wished to use Hatra as a base from which to attack the Romans must be questioned. Hatra would have been a poor choice for this purpose anyway. Located approximately 60 km west of the Tigris, it was in striking distance of Persian territory for the Romans. The city was located in difficult territory over 200 km east of the Euphrates and approximately 100 km south-east of the nearest Roman legionary base at Singara, making it a very difficult base for the Persians to attack from. Singara was itself a remote outpost for the Romans, who found it increasingly difficult to supply over the following century.31 Ardashir’s motives for the first attack on Hatra were probably more associated with consolidating the overthrow of the Parthi-ans, together with concerns over Roman interest in the city.