Diyarbakir is a bustling city in the southeastern part of Turkey. Located about 80 kilometers north of the Syrian border, it is the capital of Diyarbakir province and the unofficial capital of Turkish Kurdistan. As you might imagine that doesn't make it the safest place to be right now, although when I went the Kurdish PKK still had a truce with the Turkish government and Daesh had yet to execute any (proven) attacks on Turkish soil. I got in and out just in time too. Nine days after I left the city a Daesh suicide bomber killed 32 youth activists in Suruç (in Şanlıurfa province about 200 km to the west) and shortly after that the PKK murdered three policemen in Diyarbakir in revenge. Now the entire region seems poised to begin a new wave of violence as Turkey takes advantage of the chaos to their south to crack down on their Kurdish problem, all in the name of fighting extremism. Yay.
So what made Diyarbakir worth the risk? The old city of Diyarbakir contains a nearly complete set of circuit walls that have survived with few modifications from the late Roman period. More impressive considering this fact is the city's classical antecedents. Once upon a time it was the city of Amida (even today the city is often called Amed in Kurdish). The capital of the province of Mesopotamia, this city produced many important figures but is most famous for undergoing two well-recorded sieges during late antiquity. These sieges are so well recorded in fact that in conjunction with the surviving city walls it is possible to actually identify many of the features in them. For that reason what follows will be a description of the sieges themselves as illustrated by the surviving remains. Due to space constraints this section will only cover the first siege under Constantius and Shapur II while the second will cover the siege under Anastasius and Kavad.
AD 359
For this siege we are immensely fortunate to have the detailed record of a participant, and an observant one at that. Ammianus Marcellinus was a military officer serving under Ursicinus, the
magister equitum per Orientem (general in charge of the East). He served in several of the major campaigns of the period, including Julian's Persian campaign. In the current campaign he served as a scout to identify Persian movements as they crossed the Tigris, and then fought in the defense of Amida.
It is worth mentioning before we begin that the exact date that the existing city walls were constructed is uncertain. Ammianus mentions that Constantius had fortified the city prior to the siege and some scholars take that to mean that the current walls are the ones built at this time. Other scholars prefer to date the walls to Valens' reign, after the resolution of the Persian war allowed a refortification program to succeed. Still others consider the walls to have reached their finished state only under the emperor Justinian in the sixth century.
Whatever the exact provenance of the existing walls it is clear that the city was already protected by an impressive set of defenses at this time. I tend to the view that the walls we see today were largely in existence at that date, and if they were later expanded and enhanced after the siege then this was a development along the original path of the walls and not likely to be radically different. There are after all Greek and Latin inscriptions set into the wall that date from the fifth century or earlier.
In any event, in the spring of 359 Shapur crossed the Tigris with a huge army and drove the frightened inhabitants and soldiers of Mesopotamia in front of him. The Romans had known he was coming as no less than Ammianus himself had led a scouting party to spy on him crossing the river. Advancing along the right bank of the Tigris he encountered nothing but a confused and ineffectual Roman response. Nisibis was the capital of the province at this point, but Shapur ignored that city in favor of marching unto Syria to prey on the relatively untouched cities in the interior. He planned on marching past Amida too, but circumstances were to conspire against him.
And so Shapur and his army marched up the Tigris until they neared the city of Amida. His arrival caused much chaos among the Romans. Ammianus himself was taken by surprise at the arrival and was forced to abandon his wounded friend and flee for the relative safety of the city walls. He was not alone in this and he was forced to spend the night trapped outside the wall amongst a stream of refugees hiding from Persian soldiers sent to scout the city.
This gate he was attempting to reach seems to have been the eastern one. Ammianus describes a
steep hill up which he had to climb and mentions his ambush near the Tigris. The plain on the eastern side of the city is most suited to military maneuvers as it is flat and relatively open. The hill leading up from this displays the same winding narrowness that Ammianus described. Indeed, the description matches Ammianus' exactly and can hardly be applied to any of the other gates. The ones to the north and west are on level ground and the one to the south, while located on a hill, is only a very minor incline.
The gate currently in place is known as the New Gate (Yenikapi), which I would radically suggest means that it is newer than the others. As it stands it is smaller than the other gates and appears to be located in a relatively recent section of wall. Ammianus' gate would presumably have been more similar to the monumental gates found on the other sides of the city. The path up to the entrance (pictured above) would at least have been similar, even if it is lacking the buildings that Ammianus describes leading up to it.
After securing two other forts in the area the Persian army returned to Amida to try and see if they could be encouraged to surrender. Shapur rode out in front of his entire army and paraded himself in front of the gates in a show of nonchalance and power that was undoubtedly intended to show his lack of fear towards his adversaries. This may have been at the southern gate since that faces the Tigris plain where his army would have been drawn up, or possibly at the northern gate near where the citadel containing the leaders of the defense were located. Ammianus is unclear.
What Ammianus is clear about is what happened on the following day. Grumbates, king of the Chionites, was parading outside the eastern gate along with his retinue. This was the same gate that Ammianus fled towards and was located at the top of a winding but relatively concealed path. Ammianus describes Grumbates as "a man of moderate strength, and with shriveled limbs, but of a certain greatness of mind and distinguished by the glory of many victories." He led one of several foreign contingents fighting in the Persian army. As had the Achaemenid Persians who faced Alexander and the Greeks, the Sassanian Persian kings called themselves the king of kings (shahanshah), and this ideology of dominance was enhanced by having subject kings (often related by blood or marriage) serving underneath them. Unlike the Romans the Persians maintained the ethnic groupings of their soldiers and thus had divisions from several different ethnic groups serving under their own leaders..
In any event, the king Grumbates was riding outside the walls, apparently scouting for weaknesses, when he was spotted by the Roman defenders. Deciding to rid themselves of the second most important man in the Persian army they waited until he was in range and then fired their ballista (a form of catapult) at him. Instead of hitting him they killed his son, who had been riding alongside him, and forced the other warriors into a fighting retreat during which they were forced to abandon the boy's body. Realizing their error they returned with reinforcements and the rest of the day was consumed with bitter fighting that only ended at nightfall when the Chionites were able to recover the young prince's corpse. Enraged at the attack and mad with grief Grumbates demanded revenge on the Romans who had killed his son. Shapur was in no position to refuse. The Chionites had only recently been pacified, and in fact the Persian king had spent the winter in their territory to ensure that they remained
obedient. Failing to avenge the death of his sub-king's son would risk everything Shapur had achieved in the east.
So after an immense funeral lasting seven days, during which the body was paraded around the city alongside effigies of heroes or ancestors, the siege was begun. From this point on the city was besieged on all sides. Grumbates and the Chionites were given the task of besieging the eastern gate where his son fell, and other gates were guarded by different sub-kings and allied contingents.
The
north gate fell under the responsibility of the Albani. These were the inhabitants of modern Georgia in the Caucasus, and had been under Persian influence for quite some time. Ammianus describes them as equal in rank to the Chionites, which shows that this stretch of wall was considered important. And for good reason.
The citadel was located in the north-east corner of the city, right near the northern gate. Nowadays this section is an archaeological museum and a popular spot to relax. At the time this fortress within a fortress would have housed those in charge of the defense, and indeed Ammianus mentions the depredations of the Persian army being witnessed from this place. With the two most important contingents in the Persian army camped outside this section must have been hard pressed. While the citadel itself is located on cliffs, as is the eastern section, the majority of the northern wall is on level ground. It would not be easy to defend. It seems probable that the northern wall was where many of the siege engines and devices were placed.
This north gate is currently known as the Mountain Gate (Dağkapi) or occasionally as Harput Gate. As with many of the places in Diyarbakir it has been given multiple names over the years. In Roman times it was called the Armenian Gate. This section of wall is almost entirely missing. This isn't due to the siege however, but was only destroyed in 1932 to improve air flow inside the city. Believe me the city needs it, but this is still a tragic waste. The gate itself survives flanked by two towers, one of which prominently displays a Greek inscription and the other of which features some nice Arabic carvings.
The
west gate was given to the Segestani (or more properly the Sistani or Sakistani). This eastern Iranian people had been subject to Persian rule since the time of Bahram III (c. 293). Sakistan had long been under the domination of the Suren family, one of the seven noble houses in the Parthian and later Persian empire. The most famous member of this family (known only as Surena to the Roman sources) was responsible for annihilating Crassus' army at the battle of Carrhae. Ammianus clearly knows of the Suren since he mentions a commander from this family during Julian's Persian campaign. Does his failure to mention one here imply that the Sistani contingent was under the control of another family? Were the Suren not actually in control of the region at this point? As ever internal Persian history remains remarkably obscure. At any rate this was a contingent with a long history of service to the Persian state whose reputation was quite high. Ammianus describes them as "the bravest warriors of all." To the west were also located the elephants.
This gate is currently known as Urfa Gate (Urfakapi) since it faces towards Şanlıurfa (ancient Edessa). The gate today is interesting as it has a large entrance for traffic and two smaller entrances for pedestrians. The stretches along both sides of the gate are parkland, and the walls are filled with green grass and trees that truly make the place one of the nicest on the wall while at the same time making it very difficult to photograph. The western stretch is completely flat, or near enough, and must have been the hardest to defend. Indeed, in 503 it was on the west that the wall fell and it may well have been the same in 359.
The Persians assaulted the wall in several places and even built mounds and siege engines for the assault. This was presumably on the western side, and possibly in the north as well, since none of the other sides are suitable for such an operation.
The
southern gate today is known as Mardin Gate (Mardinkapi) and is located next to a section of wall that has been removed to make room for the main road. The road leading from the south gate goes directly to the north gate, as does the west gate to the east gate. The modern town is still built around this quintessentially Roman design of four roads intersecting in the center and a series of minor roads leading off from that. The southern wall is built on top of cliffs to the east but flat to the
west and has a gentle slope that leads down to the Tigris River. The
section of walls from the southern gate to just before the northern one
is the best preserved in the city.
During the siege the south gate was the responsibility of the Gelani.
These were another Caucasian people who had migrated down to live on the
southwestern coast of the Caspian sea. This section of wall is described
by Ammianus as quiet and the Gelani contingent were probably not
Shapur's best troops.
Given (or perhaps because of) this general quiet it is surprising that this section of wall almost gave the Persians an early victory. Underneath the tower on one of these cliffs there was a cave and a set of stairs leading up into the interior of the city. This passage was used to fetch water from the river and worst of all was unguarded due to its remoteness and steepness. Which is why the Roman forces were taken by surprise when a force of seventy Persian bowmen led by a deserter snuck into the tower and began to attack the Romans inside the city. An assault was mounted by the Persian army and the Romans were hard pressed to defend themselves or take back the tower. Eventually, by relocating five of their ballistas they were able to drive the Persians out of the tower and save the day.
The obvious location for this tower is the one shown above. Its appearance is exactly as Ammianus describes, an isolated tower to the south that stands tall on top of a cliff under in which there are several caves. As a matter of fact, it is probably too obvious. This tower is known as the Maiden's Tower (Birca Keçikê) and it is located right next to the southern gate. It is probably the most conspicuous tower on the entire wall and a force stationed to the south and paying particular attention to the gate would have to be stationed right next to it. It seems far more likely that the tower described was one of the ones further to the east, where the terrain is similar although now there are houses built up against the edge of the cliff which would cover any entrances. This does however give a good idea of what the actual tower would have looked like and its situation with regards to the siege.
With this attempt to storm the city a failure the Persians intensified their siegeworks. There were two mounds constructed (on either the west or northern sides) and the Romans began building their own mounds to counter them. There are no remains visible of these mounds, which would undoubtedly have been leveled after the conflict and in any event the entire area has been covered over by houses in the 1600 years since. As can be seen from the picture to the right the area to the west is now heavily populated.
Two of the seven legions inside the city were Gallic ones who had previously served the usurper Magnentius. Ammianus has mixed feelings about these soldiers, praising, on the one hand, their fighting spirit even as he derides their impatience and uselessness at siegecraft. These soldiers were growing increasingly impatient at having to stay inside the walls and take the hits instead of charging out of the city and confronting their foe. After weeks of this they demanded to be let out to join in an actual fight in the hopes of driving the enemy off and relieving tension. Earlier raids by these legions had resulted in the gates being firmly locked and the high command was nervous lest their refusal to waste these soldiers in a fruitless battle would result in an even more suicidal attack. For that reason they consented to allow them to lead a sally outwards in an attempt to break through the siege.
Waiting for a moonless night the Gallic legions charged through a postern gate and charged the enemy encampments. They hoped to reach the tent of Shapur himself, but long before they reached it they were drawn to a halt and forced to retreat back behind the walls with 400 casualties. Given the diminished size of the legions in this period that may have been as much as a quarter of their original strength. The results of this raid were minimal but they did get a three day truce out of it as both sides buried their dead.
Tired of the stalling the Persians decided to press the siege. Captured ballistae were placed on top of their siege towers and from this higher angle were able to rain down serious artillery upon the defenders on the wall. The Persians charging the wall fared little better as their wicker
shields were not much protection against Roman arrows. Even the feared
cataphracts were kept at bay through fear of the ballistae on the walls. After a day of this the Romans were able to counteract the Persian strategy by positioning scorpions on the walls nearby in order to take down the ballista-bearing siege towers. Amida had its own artillery armory so the Roman forces, small as they were, did not have a shortage of weaponry. By the end of the second day this threat had been met. Apparently, Shapur himself was forced to join in the fighting in order to restore his soldiers' morale.
By this time it was nearing the end of summer. Diyarbakir is enormously hot in summer. While I was there the temperature reached 42
°C (107°F). Just walking around the walls in such weather was exhausting. I was carrying nothing heavier than my camera and I suffered from heat exhaustion (though I did run up and down the entire set of circuit walls). The locals deal with this by lying in the shade of the walls all day. I can't even imagine how people could wear boiling suits of armor not to mention doing something so strenuous as combat in that temperature.
Tiring of the siege, which had already consumed two whole months, the combat switched to attacking up the mounds. The Roman mound was of necessity higher than the Persian one and this had led to a greater instability. As the Romans were fighting atop it the entire thing collapsed under them. The Roman forces were thrown into chaos and couldn't stop the Persians forcing their way into the city.
Comes Aelianus and the legionary commanders were hung and the others led off into slavery. This was the end of the siege as the Persians sacked the city. It had held for 73 days.
Sources
http://penelope.uchicago.edu/Thayer/E/Roman/Texts/Ammian/18*.html
http://penelope.uchicago.edu/Thayer/E/Roman/Texts/Ammian/19*.html
City Map
In any event, in the spring of 359 Shapur crossed the Tigris with a huge army and drove the frightened inhabitants and soldiers of Mesopotamia in front of him. The Romans had known he was coming as no less than Ammianus himself had led a scouting party to spy on him crossing the river. Advancing along the right bank of the Tigris he encountered nothing but a confused and ineffectual Roman response. Nisibis was the capital of the province at this point, but Shapur ignored that city in favor of marching unto Syria to prey on the relatively untouched cities in the interior. He planned on marching past Amida too, but circumstances were to conspire against him.
The gate currently in place is known as the New Gate (Yenikapi), which I would radically suggest means that it is newer than the others. As it stands it is smaller than the other gates and appears to be located in a relatively recent section of wall. Ammianus' gate would presumably have been more similar to the monumental gates found on the other sides of the city. The path up to the entrance (pictured above) would at least have been similar, even if it is lacking the buildings that Ammianus describes leading up to it.
During the siege the south gate was the responsibility of the Gelani.
These were another Caucasian people who had migrated down to live on the
southwestern coast of the Caspian sea. This section of wall is described
by Ammianus as quiet and the Gelani contingent were probably not
Shapur's best troops.
Given (or perhaps because of) this general quiet it is surprising that this section of wall almost gave the Persians an early victory. Underneath the tower on one of these cliffs there was a cave and a set of stairs leading up into the interior of the city. This passage was used to fetch water from the river and worst of all was unguarded due to its remoteness and steepness. Which is why the Roman forces were taken by surprise when a force of seventy Persian bowmen led by a deserter snuck into the tower and began to attack the Romans inside the city. An assault was mounted by the Persian army and the Romans were hard pressed to defend themselves or take back the tower. Eventually, by relocating five of their ballistas they were able to drive the Persians out of the tower and save the day.
With this attempt to storm the city a failure the Persians intensified their siegeworks. There were two mounds constructed (on either the west or northern sides) and the Romans began building their own mounds to counter them. There are no remains visible of these mounds, which would undoubtedly have been leveled after the conflict and in any event the entire area has been covered over by houses in the 1600 years since. As can be seen from the picture to the right the area to the west is now heavily populated.
Waiting for a moonless night the Gallic legions charged through a postern gate and charged the enemy encampments. They hoped to reach the tent of Shapur himself, but long before they reached it they were drawn to a halt and forced to retreat back behind the walls with 400 casualties. Given the diminished size of the legions in this period that may have been as much as a quarter of their original strength. The results of this raid were minimal but they did get a three day truce out of it as both sides buried their dead.
No comments:
Post a Comment