The First Golden Era of Sasanian Iran

[Semper Victor]

8.9 FROM THE BATTLE OF CTESIPHON TO THE BEGINING OF THE RETREAT.

After the fall of Maiozamalcha, the Roman army crossed a series of canals, and an attempt to stop them by the Sasanians was foiled (Ammianus XXIV, 4, 31). The relative ease with which the Roman army reached the outskirts of Ctesiphon is briefly mentioned in many other sources, like Libanius’ Epistle 1402, 2–3, Eutropius X, 16, 1; Festus, Breviarium, 28, p. 67. 18–19; Gregory of Nazianzus, Oration V,9; Socrates, III, 21, 3, Sozomenus VI, 1, 4, Malalas, XIII, pp. 329, 23–330 and Zonaras, XIII, 13, 1.

At this point, the Roman army reached the western bank of the Tigris, the location of the twin cities of Vēh-Ardaxšīr (called Coche or Kōḵē in Syriac Christian texts, or Māhōzē in the Babylonian Talmud, and still called anachronistically “Seleucia” by Greek and Latin authors like Libanius or Ammianus)on the western bank of the river and Ctesiphon (Middle Persian Ṭīsfūn) on its eastern bank. Ctesiphon has never been excavated, and Vēh-Ardaxšīr was only excavated briefly in the late 1920s by an American team and later in 1964 by an Italian one which managed to draw a map of the southwestern quarter of the city.

Reconstruction of the surroundings of Ctesiphon / Vēh-Ardaxšīr in Late Antiquity. The riverbed of the Tigris has changed significantly across the centuries. The westernmost course is the one that existed from the III century BCE to the I century CE, and as you can see, it reached the Hellenistic foundation of Seleucia; while Ctesiphon (founded by the Arsacids in the II century CE stood quite far removed on the eastern bank of the river. Vologesias / Walagašapat was founded by the Arsacid king Vologases I (Walagaš in Parthian, Walāxš in Middle Persian and Balāš in New Persian and Arabic) in he second half of the I century CE, probably after the Tigris changed course and left Seleucia without its river port. Vēh-Ardaxšīr was founded by the first Sasanian Šahān Šāh Ardaxšīr I in 230 CE just opposite the river from Ctesiphon, since then the Tigris has again changed course and now its riverbed crosses almost straight through the center of the ancient city. The irregular pink shape in the map is the excavated area of the old city, and the thick black line the part of the ancient wall that has been discovered (and which has led archaeologists to believe it was a “round city” as many other Arsacid and Sasanian foundations.

Based on the Italian excavations, Vēh-Ardaxšīr was a typical Sasanian city, built on a round plan and was heavily fortified, with circular walls studded with semicircular towers at regular intervals of 35 meters. The wall was very thick (10 meters wide at the foundations) and was protected by a wide moat, probably filled by water and built in mudbrick; all in all strikingly similar to the fortifications of better-known Central Asian cities that were also part of the Sasanian empire (permanently or temporarily) like Merv or Balḵ. From Christian and Jewish texts, we also know that the city had a citadel (known in Middle Persian as Grondagan or Garondagan, and as Aqra d’Kōḵē in Syriac) where the Sasanian governor, functionaries and garrison resided. The city lodged also one of the royal mints of the empire (Ctesiphon had another).

According to Tabarī, two bridges united Vēh-Ardaxšīr with Ctesiphon (one of them built by Šābuhr II early on his reign). Nothing is known for sure about Ctesiphon proper, as the German excavations before the WWI and the American ones in the 1920s only covered the Aspanbar area south of Ctesiphon proper, where in the VI century Khusrō I and his successors built the great palace of which still stands the throne hall, as well as pleasure gardens, walled hunting grounds, and where the members of the Sasanian court also built their private palaces near the royal residence. But all this did not exist yet in 363 CE, Ctesiphon proper is supposed to have been another walled city built on a round plan as was typical of Arsacid and Sasanian foundations, following the Central Asian custom. Inside the walls stood the “White Palace”, the palace of the Sasanian kings, which was still standing when the Arabs took the city in the VII century CE. We know also, according to Jewish and Christian texts, that most of the population of Vēh-Ardaxšīr was Christian or Jewish, and that the city was the see of a bishopric. In both cities, the majority of the population would have been formed by native Syriac speakers, with a small group of Iranian speakers forming the administrative, clerical and military elite.

Mesopotamia is flat, but that doesn’t mean that the environs of Ctesiphon were open terrain suited for army maneuvering. Apart from the large obstacle of the river Tigris, the terrain was crisscrossed by a dense network of irrigation canals, some of which (like the Naarmalcha canal) were large obstacles for troop movements (as François Paschoud showed in the paper I mentioned in the last chapter, due to hydraulic reasons and to the incline of the terrain, a major canal like the Naarmalcha must’ve had its water surface risen several meters above the surrounding terrain, and it was flanked by major earth dams). This would have been a problem for both armies; the Sasanian cavalry would’ve been denied the sort of open, flat terrain that its cavalry needed, but the Romans also found their advance encumbered by frequent ambushes and by floods caused by the Sasanian defenders who tried to delay their advance by opening locks and floodgates whenever possible. On top of it all, the whole area was densely settled and built up, and other than the “twin cities” there were multiple palaces, villages, country estates and smaller fortified towns and cities.

As I said in the previous chapter, there’s still a lack of understanding among scholars about the exact layout of the Naarmalcha, and this has been the cause for different interpretations of the arrival of the Roman army to Ctesiphon. Let’s see the different accounts by Ammianus, Zosimus, Libanius and we’ll also take a look at Cassius Dio’s account of Trajan’s campaign in 116-117 CE.

Ammianus Marcellinus (Res Gestae Libri XXXI, XXIV, 6, 1):

Then we came to an artificial river, by name Naarmalcha, meaning "the kings' river", which at that time was dried up. Here in days gone by Trajan, and after him Severus, had with immense effort caused the accumulated earth to be dug out, and had made a great canal, in order to let in the water from the Euphrates and give boats and ships access to the Tigris. It seemed to Julian in all respects safest to clean out the same canal, which formerly the Persians, when in fear of a similar invasion, had blocked with a huge dam of stones. As soon as the canal was cleared, the dams were swept away by the great flow of water, and the fleet in safety covered thirty stadia and was carried into the channel of the Tigris. Thereupon bridges were at once made, and the army crossed and pushed on towards Coche (i.e. Vēh-Ardaxšīr). Then, so that a timely rest might follow the wearisome toil, we encamped in a rich territory, abounding in orchards, vineyards, and green cypress groves. In its midst is a pleasant and shady dwelling, displaying in every part of the house, after the custom of that nation, paintings representing the king killing wild beasts in various kinds of hunting; for nothing in their country is painted or sculptured except slaughter in diverse forms and scenes of war.

Zosimus (New History, III ,24, 2):

They advanced to a very broad sluice or channel, said by the country people to have been cut by Trajan, when he made an expedition into Persia. In this channel runs the river Narmalaches and discharges itself into the Tigris. The emperor caused it to be cleansed, in order to enable his vessels to pass through it into the Tigris, and constructed bridges over it for the passage of his army.

Libanius (Oration XVIII, 244-247)

Doing things of this sort, they at length arrive at the cities, so long objects of their desire, the which, in place of Babylon, adorn the land of the Babylonians. Through the midst of these runs the river Tigris, and after passing by them some little distance, unites with the Euphrates. At this point, what was to be done could not be discovered; for if the soldiers should pass along in the flotilla it was impossible to approach the towns; whilst if they attacked the towns, their boats would be useless to them, and if they should sail up the Tigris, the labor would be excessive, and they would have to pass in the middle between the cities. Who then solved the difficulty? It was not a Calchas, nor a Tiresias, nor any one of the diviners; the emperor seized some prisoners out of those dwelling in the neighborhood, and made inquiry about a navigable canal (this too from his books) constructed by the ancient kings, and leading from the Tigris into the Euphrates, at some distance from the two cities (i.e. Ctesiphon and Vēh-Ardaxšīr). Of these prisoners, the youthfulness of the one was entirely unsuspicious of his design in putting the question, whilst the one of advanced age told the truth because there was no help for it (for he perceived that the emperor was as exactly informed about the locality as anyone of the natives, so much had he, though distant, got a view of the place in books). The elder prisoner therefore tells, both where the canal is, and in what way it is closed up, and that it had been filled up with earth, and sowed over with corn at the part next its opening. At the nod of the commander all the obstruction was taken out, and of the two streams the one is seen drained dry; the other bore along the flotilla which kept side by side with the army; whilst the Tigris coming down upon those in the cities greater than before, inasmuch as it had received the waters of the Euphrates, occasioned them great alarm, in the belief that it would not spare their walls.

Cassius Dio (Roman History, Book LXVIII, 28, 1-2)

Trajan had planned to conduct the Euphrates through a canal into the Tigris, in order that he might take his boats down by this route and use them to make a bridge. But learning that this river has a much higher elevation than the Tigris, he did not do so, fearing that the water might rush down in a flood and render the Euphrates unnavigable. So, he used hauling-engines to drag the boats across the very narrow space that separates the two rivers (the whole stream of the Euphrates empties into a marsh and from there somehow joins the Tigris); then he crossed the Tigris and entered Ctesiphon.

If we take the three parallel narratives about this part of Julian’s expedition, Paschoud remarks that the less reliable one in this case is Ammianus, even if he accompanied it and was a direct eyewitness (let’s remember that he wrote his work almost thirty years later after he’d retired from the army and settled in Rome). For starters, a comparison with the other two accounts shows clearly that in this passage he’s confusing the Naarmalcha with the so-called “Canal of Trajan”. And on top of it, he’s obviously forgotten that he’d already written about the Naarmalcha in a previous passage (quoted in the previous chapter). Zosimus’ account is more precise, but it only specifies in a very indirect way that the Canal of Trajan was dry when Julian reached it. Of the three accounts, Libanius’ is the most helpful with geographical details in this case. In his account, he specifies that the Naarmalcha only reached the Tigris downstream from Seleucia (a deserted city by now) and Ctesiphon, but that another canal built by an ancient emperor (evidently, the Canal of Trajan) branched off from the Naarmalcha and reached the Tigris upstream from Ctesiphon.

Paschoud carries on with his study by pointing out that Cassius Dio’s text lets us know that the topography of the region had not changed much in two centuries, and that what IV century authors called “the Canal of Trajan” had not been made by that emperor. Dio’s passage has only arrived to us in an abbreviated form (through John Xiphilinus’ epitome) but it seems that Trajan did not seek to build a new canal, but that he merely repaired an existing one that was dry by then, as it happened again in 363 CE. Paschoud hypothesizes that this canal must’ve been much older, perhaps related to the foundation of the Hellenistic settlement of Seleucia on the Tigris; after the foundation of this city it would have become necessary to build a waterway that allowed shipping to reach its city docks without having to sail up the Tigris stream (the Naarmalcha reached the Tigris south of Seleucia’s location, same as with Ctesiphon and Vēh-Ardaxšīr).

Notice also how Dio’s text also show how the Romans were well aware of the difference in topographical levels between the Euphrates and Tigris. And two of the three narratives from the IV century attest to these effects: Ammianus describes that after reopening this blocked canal, it filled up quickly and the current was swift; Libanius also describes a brusque rise of the level of the waters of the Tigris between “Seleucia” (probably Vēh-Ardaxšīr) and Ctesiphon after Julian had this old canal repaired.

The Naarmalcha main role would have been as the principal irrigation canal in this region, and as irrigation was done by gravity, it was necessary for its water level to be higher than that of the surrounding land. The higher the level compared to that of its surrounding fields, the further the secondary irrigation channels that branched out could carry their water, and more land could be brought into cultivation. Today’s landscape in central and southern Iraq is thus still crisscrossed by these ancient canals (now dry) with their dams that lie well above the surrounding countryside; this practice was facilitated by the abundant amounts of alluvial sand and other deposits carried by the Euphrates, that meant that in its initial stages it was the water itself that brought the construction materials for the canals (although this also implied that later they would have needed constant dredging to keep their beds in working order).

Paschoud’s reconstruction (which I find personally convincing) goes against the reconstructions by Dodgeon & Lieu, as well as by Syvänne, according to whom the Roman supply fleet reached the Tigris directly through the Naarmalcha downstream of Ctesiphon, while Paschoud (reading correctly the ancient sources as we’ve seen above, especially Libanius) stated that the Roman fleet reached the Tigris via the so-called “Canal of Trajan” upstream of Ctesiphon. This will be important for later, when we will consider Julian’s decision to burn the fleet.

Julian’s attack on Ctesiphon according to Dodgeon & Lieu (from “The Roman Eastern Frontier and the Persian Wars (AD226–363): A Documentary History”)

The Romans came upon a palace built in Roman style and left it untouched (probably on 15 May), and they also found Šābuhr II’s hunt reserve well stocked with game (Ammianus XXIV, 5, 1–2, Zosimus III, 23, 1–2 and Libanius, orations XVII, 20 and XVIII, 243). At the site of the ruins of the ancient Hellenistic city of Seleucia (west of Vēh-Ardaxšīr, which was used by the Sasanian kings as a site for public executions, as stated in Christian Syriac martyrs’ acts) Julian saw the impaled bodies of the relatives of the Persian commander who surrendered Pērōz-Šābuhr (Ammianus XXIV, 5, 4). Here Nabdates, the former Persian commander of Maiozamalcha, who had surrendered to the Romans and was their prisoner but had “grown insolent”, was burnt alive with 80 of his men (Ammianus XXIV, 5, 4). A strong fortress nearby was captured by the Romans (probably on 16 May: Ammianus XXIV, 5, 6–11). This could be the Meinas Sabath (identified variously as either the Arsacid settlement of Vologesias/Walagašapat or as a suburb of Vēh-Ardaxšīr) mentioned in Zosimus III, 23, 3. The fleet then sailed down the Naarmalcha and the army reached the walls of Vēh-Ardaxšīr. (Ammianus XXIV, 6, 1–2, Zosimus III, 24, 2, Libanius, Oration XVIII, 245–7 and Malalas, XII, 10–16 -identical to Magnus of Carrhae’s Fragment 7).

This is a key moment in the narrative. Notice that according to some scholars (like Paschoud) the fleet reached the Tigris upstream from Ctesiphon / Vēh-Ardaxšīr (opening the locks of the so-called “Canal of Trajan”; hence the flooding in Vēh-Ardaxšīr recorded by Libanius) while according to others like Syvänne and Dodgeon & Lieu, the fleet reached the Tigris downstream from the twin cities.

This stucco panel (dated tentatively to the VI century CE) found near the great palace of Khusrō I in the Aspanbar area is thought to have belonged to the private palace of one of the high-ranking members o the Sasanian court. It’s one of the very few excavated artifacts from Sasanian Ctesiphon. Interiors of Sasanian palaces and temples were usually intricately covered with stucco decoration, which was later painted, either applied in situ or “prefabricated” and later fixed to the inner walls of the building (like this case). Two identical panels have been preserved, one in the Met in New York and the other at the Museum für Islamische Kunst in Berlin. According to Aboulala Soudavar, the Pahlavi inscription reads as “afzun” which can be translated into English as “to increase”, while the pearl roundel and the open wings are symbols for the ancient Iranian concept of “farr” (“divine” or “kingly” glory).

In his Oration V, the Christian author (contemporary of the events) Gregory of Nazianzus stated that both cities were “strongly defended”. As we’ve seen, archaeological digs in Vēh-Ardaxšīr confirm this, and the same is assumed to be true by scholars with respect to Ctesiphon on the other bank of the Tigris. The Babylonian Talmud also informs us that the civilians of Māhōzē were organized in a militia that was mobilized in case of a military emergency; so the Romans had to contend with two very large and populous cities that were very well fortified and defended, and which couldn’t be cut off from the outside world, unless they were both besieged at the same time (as they were united by bridges) and the besiegers cut the Tigris both upstream and upstream from the cities. And on top of it all, the Romans would have needed to do this without having defeated before the main Sasanian army led by the Šahān Šāh, which as we will see was about to arrive in scene. For the sake of comparison, in the VII century CE it took the Muslim Arab conquerors a year to besiege and conquer both cities, and only because they’d destroyed previously the main Sasanian field army at al-Qadisiyyah and a defected Sasanian engineer built eight catapults for them. Julian did not have that amount of time available, and probably he’d planned the invasion of Ērānšahr in the same way he’d planned his commando raids against the Franks and Alamanni in the Rhine. But this enemy was different.

Julian partially unloaded the fleet and used the boats to ferry soldiers across the Tigris against a strong Sasanian defense; a bitter battle was fought before the gates of Ctesiphon and won by the Romans, but (according to Ammianus) they were unable to exploit the victory because of a lack of discipline. Let’s see the different accounts:

Ammianus Marcellinus: Res Gestae Libri XXXI, 6, 4-15:

Since thus far everything had resulted as he desired, the Augustus now with greater confidence strode on to meet all dangers, hoping for so much from a fortune which had never failed him that he often dared many enterprises bordering upon rashness. He unloaded the stronger ships of those which carried provisions and artillery, and manned them each with eight hundred armed soldiers; then keeping by him the stronger part of the fleet, which he had formed into three divisions, in the first quiet of night he sent one part under comes Victor with orders speedily to cross the river and take possession of the enemy's side of the stream. His generals in great alarm with unanimous entreaties tried to prevent him from taking this step but could not shake the emperor's determination. The flag was raised according to his orders, and five ships immediately vanished from sight. But no sooner had they reached the opposite bank than they were assailed so persistently with firebrands and every kind of inflammable material, that ships and soldiers would have been consumed, had not the emperor, carried away by the keen vigor of his spirit, cried out that our soldiers had, as directed, raised the signal that they were already in possession of the shore, and ordered the entire fleet to hasten to the spot with all the speed of their oars. The result was that the ships were saved uninjured, and the surviving soldiers, although assailed from above with stones and every kind of missiles, after a fierce struggle scaled the high, precipitous banks and held their positions unyieldingly. History acclaims Sertorius for swimming across the Rhine with arms and cuirass; but on this occasion some panic-stricken soldiers, fearing to remain behind after the signal had been given, lying on their shields, which are broad and curved, and clinging fast to them, though they showed little skill in guiding them, kept up with the swift ships across the eddying stream.
The Persians opposed to us serried bands of mail-called horsemen in such close order that the gleam of moving bodies covered with densely fitting plates of iron dazzled the eyes of those who looked upon them, while the whole throng of horse was protected by coverings of leather. The cavalry was backed up by companies of infantry, who, protected by oblong, curved shields covered with wickerwork and raw hides, advanced in very close order. Behind these were elephants, looking like walking hills, and, by the movements of their enormous bodies, they threatened destruction to all who came near them, dreaded as they were from past experience.
Hereupon the emperor, following Homeric tactics, filled the space between the lines with the weakest of the infantry, fearing that if they formed part of the van and shamefully gave way, they might carry off all the rest with them; or if they were posted in the rear behind all the centuries, they might run off at will with no one to check them. He himself with the light-armed auxiliaries hastened now to the front, and now to the rear.
So, when both sides were near enough to look each other in the face, the Romans, gleaming in their crested helmets and swinging their shields as if to the rhythm of the anapaestic foot, advanced slowly; and the light-armed skirmishers opened the battle by hurling their javelins, while the earth everywhere was turned to dust by both sides and swept away in a swift whirlwind. And when the battle-cry was raised in the usual manner by both sides and the trumpets' blare increased the ardor of the men, here and there they fought hand-to‑hand with spears and drawn swords; and the soldiers were freer from the danger of the arrows the more quickly they forced their way into the enemy's ranks. Meanwhile Julian was busily engaged in giving support to those who gave way and in spurring on the laggards, playing the part both of a valiant fellow-soldier and of a commander. Finally, the first battle-line of the Persians began to waver, and at first slowly, then at quick step, turned back and made for the neighboring city with their armor well heated up. Our soldiers pursued them, wearied though they also were after fighting on the scorching plains from sunrise to the end of the day, and following close at their heels and hacking at their legs and backs, drove the whole force with Pigranes, the Surena, and Narseus, their most distinguished generals, in headlong flight to the very walls of Ctesiphon. And they would have pressed in through the gates of the city, mingled with the throng of fugitives, had not the general called Victor, who had himself received a flesh-wound in the shoulder from an arrow, raising his hand and shouting, restrained them; for he feared that the excited soldiers, if they rashly entered the circuit of the walls and could find no way out, might be overcome by weight of numbers (…).
After their fear was past, trampling on the overthrown bodies of their foes, our soldiers, still dripping with blood righteously shed, gathered at their emperor's tent, rendering him praise and thanks because he had won so glorious a victory, everywhere without recognition whether he was leader or soldier, and considering the welfare of others rather than his own. For as many as 2,500 Persians had been slain, with the loss of only seventy of our men. Julian addressed many of them by name, whose heroic deeds performed with unshaken courage he himself had witnessed, and rewarded them with naval, civic, and camp crowns.

Festus, Breviarium 28:

Having pitched his camp opposite to Ctesiphon on the bank where the Tigris joins the Euphrates, he spent the day in athletic contests to relieve the enemy of their watchfulness. In the middle of the night he embarked his soldiers and suddenly carried them across to the far bank. They struggled over the escarpment, where the ascent would have been difficult even in daytime when no one was trying to stop them. They threw the Persians into confusion by the unexpected terror and the armies of their entire nation were turned to flight. The soldiers would have victoriously entered the open gates of Ctesiphon if the opportunity for booty had not been greater than their concern for victory.

Gregory of Nazianzus, Oration V, 9-10:

Now, the first steps in his (i.e. Julian’s) enterprise, excessively audacious and much celebrated by those of his own party, were as follows. The entire region of Assyria that the Euphrates flows through, and skirting Persia, there unites itself with the Tigris; all this he took and ravaged, and captured some of the fortified towns, in the total absence of anyone to hinder him, whether he had taken the Persians unawares by the rapidity of his advance, or whether he was outgeneraled by them and drawn on by degrees further and further into the snare (for both stories are told); at any rate, advancing in this way, with his army marching along the river’s bank and his flotilla upon the river supplying provisions and carrying the baggage, after a considerable interval he reached Ctesiphon, a place which, even to be near, was thought by him half the victory, by reason of his longing for it.
From this point, however, like sand slipping from beneath the feet, or a great storm bursting upon a ship, things began to go black for him; for Ctesiphon is a strongly fortified town, hard to take, and very well secured by a wall of burnt brick, a deep ditch, and the swamps coming from the river. It is rendered yet more secure by another strong place, the name of which is Coche, furnished with equal defenses as far as regards garrison and artificial protection, so closely united with it that they appeared to be one city, the river separating both between them. For it was neither possible to take the place by general assault, nor to reduce it by siege, nor even to force a way through by means of the fleet principally, for he would run the risk of destruction; being exposed to missiles from higher ground on both sides, he left the place in his rear, and did so in this manner. He diverted a not inconsiderable part of the river Euphrates, the greatest of rivers, and rendered it navigable for vessels, by means of a canal, of which ancient vestiges are said to be visible; and thus joining the Tigris a little in front of Ctesiphon, he saved his boats from one river by means of the other river, in all security; in this way he escaped the danger that menaced him from the two garrisons. But, as he advanced, a Persian army suddenly started up, and continually received fresh reinforcements, but did not think it advisable to stand in front and fight it out, without the greatest necessity (although it was in their power to conquer, from their superior numbers); but from the tops of the hills and narrow passes they shot arrows and threw darts, whenever opportunity served, and thus readily prevented his further progress. Hence he is reduced to great perplexity, and, not knowing to what side to turn, he finds out an unlucky solution for the difficulty.

Sozomenus, Ecclesiastical History, VI, 5-8:

As he was journeying up the Euphrates, he arrived at Ctesiphon, a very large city, whither the Persian monarchs had now transferred their residence from Babylon. The Tigris flows near this spot. As he was prevented from reaching the city with his ships, by a part of the land which separated it from the river, he judged that either he must pursue his journey by water, or quit his ships and go to Ctesiphon by land; and he interrogated the prisoners on the subject. Having ascertained from them that there was a canal which had been blocked up in the course of time, he caused it to be cleared out, and, having thus effected a communication between the Euphrates and the Tigris, he proceeded towards the city, his ships floating along by the side of his army. But the Persians appeared on the banks of the Tigris with a formidable display of horse and many armed troops, of elephants, and of horses; and Julian became conscious that his army was besieged between two great rivers, and was in danger of perishing, either by remaining in its present position, or by retreating through the cities and villages which he had so utterly devastated that no provisions were obtainable; therefore he summoned the soldiers to see horse-races, and proposed rewards to the fleetest racers. In the meantime he commanded the officers of the ships to throw over the provisions and baggage of the army, so that the soldiers, I suppose, seeing themselves in danger by the lack of necessary provisions, might turn about boldly and fight their enemies more desperately. After supper he sent for the generals and tribunes and commanded the embarkation of the troops. They sailed along the Tigris during the night and came at once to the opposite banks and disembarked; but their departure was perceived by some of the Persians, who defended themselves and encouraged one another, but those still asleep the Romans readily overcame. At daybreak, the two armies engaged in battle; and after much bloodshed on both sides, the Romans returned by the river, and encamped near Ctesiphon.

Zosimus, New History, III, 24-25:

Soon after his execution, the army marched to Arintheus, and searching all the marshes found in them many people whom they made prisoners. Here it was that the Persians first collected their forces and attacked the advanced party of the Roman army. They were however routed and preserved their lives by flying to a neighboring city. The Persians on the other side of the river attacked the slaves who had the care of the beasts of burden, and those who guarded them; they killed part of them and made the rest prisoners. This being the first loss which the Romans had sustained occasioned some consternation in the army.
They advanced to a very broad sluice or channel, said by the country people to have been cut by Trajan, when he made an expedition into Persia. In this channel runs the river Narmalaches and discharges itself into the Tigris. The emperor caused it to be cleansed, in order to enable his vessels to pass through it into the Tigris, and constructed bridges over it for the passage of his army.
While this was in agitation, a great force of Persians, both horse and foot, was collected on the opposite bank, to prevent their passage should it be attempted. The emperor, discerning these preparations of the enemy, was anxious to cross over to them, and hastily commanded his troops to go on board the vessels.
Perceiving, however, the opposite bank to be unusually lofty, and a kind of fence at the top of it, which formerly served as an enclosure to the king's garden, but at this time was a rampart, they exclaimed that they were afraid of the fire-balls and darts that were thrown down. The emperor, however, being very resolute, two barges crossed over full of foot soldiers, which the Persians immediately set on fire by throwing down on them a great number of flaming darts.
This so increased the terror of the army, that the emperor was obliged to conceal his error by a feint, saying, "They are landed and have rendered themselves masters of the bank. I know it by the fire in their ships, which I ordered them to make as a signal of victory."
He had no sooner said this, than without further preparations they embarked in the ships and crossed over, until they arrived where they could ford the river, and then leaping into the water, they engaged the Persians so fiercely, that they not only gained possession of the bank, but recovered the two ships which came over first, and were now half burnt, and saved all the men who were left in them.
The armies then attacked each other with such fury, that the battle continued from midnight to noon of the next day. The Persians at length gave way, and fled with all the speed they could use, their commanders being the first who began to fly. Those were Pigraxes, a person of the highest birth and rank next to the king, Anareus, and Surena.
The Romans and Goths pursued them, and killed a great number, from whom they took a vast quantity of gold and silver, besides ornaments of all kinds for men and horses, with silver beds and tables, and whatever was left by the officers on the ramparts.
It is computed, that in this battle there fell of the Persians two thousand five hundred, and of the Romans not more than seventy-five. The joy of the army for this victory was lessened by Victor having received a wound from an engine.

The accounts by Ammianus and Zosimus are the most complete ones, and they mostly agree on broad terms. What they describe is a bold crossing of the Tigris by the Roman army in successive assault waves against a Sasanian army massed on the opposite bank, and deployed in a classic Sasanian battle order: a first line formed by heavy cavalry (savārān) a second line of infantry massed in closed ranks and lastly a line of war elephants (presumably to act as regrouping points for the infantry if it was routed by the Romans, to cover a retreat if necessary and to offer a good shooting platform to Sasanian bowmen given the height advantage offered by these animals; and finally perhaps also to ensure the infantry kept up its fighting spirit, or else).

As it was a purely frontal battle without any possibility for maneuvering, this was the sort of combat in which the Roman infantry excelled, and it’s noteworthy that the Sasanian forces were able to put up such a long fight. Notice that this was still not Šābuhr II’s main army, but merely the cavalry screen of Sūrēn (one of the Sasanian commanders in the battle) alongside with local forces, probably drawn from the garrisons of the twin cities. Nothing is known about the other two Sasanian commanders, Pigranes (“Pigraxes” in Zosimus) and Narseus (“Anareus” in Zosimus; Middle Persian Narsē); who were perhaps the military governors (marzbānan) of the twin cities. This time, Julian’s audacity and boldness (and sheer luck) served him well, and his night amphibious assault was victorious, although his strategic situation had not improved in the slightest.

The British reenactor Nadeem Ahmad from the reenacting group “Eran ud Turan” in the full garb of a Sasanian spāhbed.

It’s now when Julian had to cope with the reality of his strategic position: he’d won several tactical victories but he still found himself in a position that could at the very best be described as a strategic stalemate, and which was bound to worsen with every passing day. The army encamped, but apparently Julian didn’t even consider seriously to start a formal siege of Ctesiphon, instead he took a decision that all the ancient sources criticize strongly, and which according to them was the reason for the final failure of the campaign. Some of the sources (mostly Christian ones) attribute it to Julian being fooled by a devious “Persian” trick (a common place when dealing with easterners in classical authors: they were deceitful, devious, etc.). According to Libanius (Oration XVIII, 257-259) and Socrates of Constantinople (Ecclesiastical History, III, 21, 4–8) Julian rejected peace overtures from Šābuhr II. Ammianus says nothing of the sort; he only wrote that a council of war was held (at a place called Abuzatha by Zosimus; scholars have been unable to find this location, judging from the context of the narrative, it must’ve been located east of Ctesiphon) and a decision to march inland was made (probably on 5 June). These are the sources:

Ammianus Marcellinus: Res Gestae Libri XXXI, XXIV, 7, 1–3:

Having held council with his most distinguished generals about the siege of Ctesiphon, the opinion of some was adopted, who felt sure that the undertaking was rash and untimely, since the city, impregnable by its situation alone, was well defended; and, besides, it was believed that the king would soon appear with a formidable force. So the better opinion prevailed, and the most careful of emperors, recognizing its advantage, sent Arintheus with a band of light-armed infantry, to lay waste the surrounding country, which was rich in herds and crops; Arintheus was also bidden, with equal energy to pursue the enemy, who had been lately scattered and concealed by impenetrable by-paths and their familiar hiding-places. But Julian, ever driven on by his eager ambitions, made light of words of warning, and upbraiding his generals for urging him through cowardice and love of ease to lose his hold on the Persian kingdom, which he had already all but won; with the river on his left and with ill-omened guides leading the way, resolved to march rapidly into the interior.

Zosimus: New History, III, 26, 1:

Upon the following day the emperor sent his army over the Tigris without difficulty, and the third day after the action he himself with his guards followed them. Arriving at a place by the Persians termed Abuzatha, he halted there five days.

Libanius, Oration XVIII, 261:

Nevertheless, this man (i.e. Julian) though invited to peace, went up to the walls (of Ctesiphon) and challenged the besieged to battle, saying that what they were doing was fit for women, but what they shunned for men. On their replying that he must seek out the king, and show himself to him, he was anxious to see and pass through Arbela, either without a battle, or after fighting a battle; so that in company with Alexander's victory at that place his own might become the theme of song. His intention was to traverse all the land which the Persian empire comprises; even more, the adjacent regions also; but he retreated because no reinforcements came to him, neither of his own side, nor from his ally: the latter through the false play of the prince of that nation; whilst the second army, according to report, because some of their men had been shot at the very beginning whilst bathing in the Tigris, had thought it better worth their while to wage war on the natives. Add to this, the quarrelling of the generals with one another had bred cowardice in those under their command; for whenever the one leader was gaining victories, the other, by recommending inaction, gave advice that pleased his men.
This state of things, however, did not discourage the emperor; he did not approve of their being absent, yet he proceeded as he had planned to do if they had joined him, and extended his views as far as Hyrcania and the rivers of India.

Eunapius of Sardes, fragments 22.3 and 27:

(He says) that the war against the Persians reached its peak under Julian and that either by invoking the gods or by calculation he comprehended from afar the disturbances of the Scythians, like waves on a smooth sea. Thereupon he says to someone in a letter, ‘The Scythians are now lying quiet, equally, they will not do so in the future’. His forethought for the future extended over such a period that he knew in advance that they would remain
quiet only for his own time.
(He says) that Julian, having previously revealed the plain before Ctesiphon as an orchestra for war, as Epaminondas said, now paraded it as a stage for Dionysus, providing relaxation and amusement for the troops.
(He says) that there was such an abundance of the provisions in the suburbs of Ctesiphon that the overriding danger faced by the troops was that of being destroyed by luxury.
(He says) that mankind, besides, seems generally inclined towards and readily given over to envy. And since the troops have no means of taking sides fairly about what is done, “From the tower”, they say, “they judged the Achaeans”, each one of them desirous of being versed in military matters and possessed of more than usual good sense. To some, then, any matter was the subject of foolishness, but he who followed the arguments right from the beginning went back to his own domain (…).

There is an oracle which was given to him while he was waiting at Ctesiphon: Zeus, the all-wise, once destroyed a race of Earth-born giants most hateful to the blessed ones who dwell in the Olympian halls. Julian the godlike, Emperor of the Romans, contending for the cities and long walls of the Persians, fighting hand-to-hand, destroyed them with fire and the valiant sword, subdued without pause their cities and many races. Seizing also, with heavy fighting, the German soil of those people of the west he laid waste their land.

John Zonaras, Extracts of History, XIII, 13, 10:

Some, therefore, say that Julian was deceived in this way. Others say he gave up the siege of Ctesiphon because of its strength and, since the army was running out of necessities, he considered returning (…)

Notice that for the first time Ammianus mentions the fear that Šābuhr II was approaching with the main Sasanian army. At this moment, Julian took the controversial decision of burning his transport fleet (Dodgeon and Lieu date it to the timeframe between 11–15 June). These are the sources reporting on Julian’s decision:

Ammianus Marcellinus: Res Gestae Libri XXXI, XXIV, 7, 3–5:

But Julian, ever driven on by his eager ambitions, made light of words of warning, and upbraiding his generals for urging him through cowardice and love of ease to lose his hold on the Persian kingdom, which he had already all but won; with the river on his left and with ill-omened guides leading the way, resolved to march rapidly into the interior. And it seemed as if Bellona herself lighted the fire with fatal torch, when he gave orders that all the ships should be burned, with the exception of twelve of the smaller ones, which he decided to transport on wagons as helpful for making bridges. And he thought that this plan had the advantage that the fleet, if abandoned, could not be used by the enemy, or at any rate, that nearly 20,000 soldiers would not be employed in transporting and guiding the ships, as had been the case since the beginning of the campaign.

Zosimus: New History, III; 26, 2–3:

Meanwhile he consulted about his journey forward and found that it was better to march further into the country than to lead his army by the side of the river. There being now no necessity to proceed by water. Having considered this, he imparted it to his army, whom he commanded to burn the ships, which accordingly were all consumed, except eighteen Roman and four Persian vessels, which were carried along in wagons, to be used upon occasion. Their route now lying a little above the river, when they arrived at a place called Noorda they halted, and there killed and took a great number of Persians.

Libanius: Oration XVIII, 262–3:

But when the army was already on the move in that direction, and part was actually marching off, the other part collecting the baggage, some god diverts him from his first scheme, and, as the poet hath it, "warned him to think of his return." The flotilla, according to his original design, had been given for prey to the flames; for better so than to the enemy. The same thing would probably have been done, even though the former plan (of advancing) had never been contemplated; but that of returning had carried the day; because the Tigris, swift and strong, running counter to the prows of the boats, forced them to require a vast number of hands (to tow them up the stream); and it was necessary for those engaged in towing to be more than half the army; this meant that the fighting men were to be beaten, and after them everything else was gained by the enemy without fighting for it. Besides all this, the burning of the fleet removed every encouragement to laziness, for whoever wished to do nothing, by feigning sickness obtained conveyance in a boat; but when there were no vessels, every man was under arms. Since therefore it was impossible, however much they wished it, to keep so many vessels, it was decided not to be expedient even to retain those that had been saved (they were fifteen in number, reserved for making bridges); for the stream being too violent for the skill of the boatmen, and the multitude of hands, used to carry the boat with those embarked therein into the hands of the enemy; so that if it behooves the side that is injured to complain of the conflagration, it will be the Persian that has to grumble; and full often, they say, he did complain.

Ephraim the Syrian: Hymns against Julian, III, 15:

The king saw that the sons of the East had come and deceived
him,
the unlearned (had deceived) the wise man, the simple the
soohsayer.
They whom he had called, wrapped up in his robe,
had, through unlearned men, mastered his wisdom.
He commanded and burned his victorious ships,
and his idols and diviners were bound through the one deceit.

Gregory of Nazianzus: Oration V, 11–12:

A Persian of considerable standing, following the example of that Zopyrus employed by Cyrus in the case of Babylon, then pretended that he had had some quarrel, or rather a very great one and for a very great cause, with his king, and was on that account very hostile to the Persian cause, and well-disposed towards the Romans. He gained the emperor’s confidence through his pretense as follows: “Your Highness, what means all this, why are there so many shortcomings in so important an enterprise? What need is there of this provision fleet, this superfluous burden; a mere incentive to cowardice; for nothing is so unfit for fighting, and fond of laziness, as a full belly, and the having the means of saving oneself in one’s own hands? But if you will listen to me, you will burn this flotilla: what a relief to this fine army will be the result! You yourself will take another route, better supplied and safer than this; along which I will be your guide (being acquainted with the country as well as any man living), and will cause you to enter into the heart of the enemy’s country, where you can obtain whatever you please, and so make your way home; and me you shall then recompense, when you have actually ascertained my good will and sound advice”.
And when he had said this and gained credence to his story (for rashness is credulous, especially when God goads it on), everything went wrong at once. The boats became the prey of the flames. They were low on victuals.
Everywhere there was ridicule, and the whole venture resembled a suicide attempt. Hope vanished when the guide disappeared along with his promises. They were surrounded by the enemy and battle waged on all sides. It was difficult to advance and provisions were not easy to procure. In despair, the army became disenchanted with their commander. There was no hope for safety left, but one wish alone, as was natural under the circumstances, the ridding themselves of bad government and bad generalship.

Sozomenus: Ecclesiastical History, VI, 1, 9:

The emperor, being no longer desirous of proceeding further but wishing only to return to the (i.e. Roman) empire, burnt his vessels, as he considered that they required too many soldiers to guard them; and he then commenced his retreat along the Tigris, which was to his left. The prisoners, who acted as guides to the Romans, led them to a fertile country where at first they found an abundance of provisions.

Theodoret of Cyrrhus: Ecclesiastical History, III, 25, 1,

Julian’s folly was yet more clearly manifested by his death. He crossed the river that separates the Roman Empire from the Persian, brought over his army, and then forthwith burnt his boats, so making his men fight not in willing, but in forced obedience.

Festus: Breviarium 28:

After winning such great glory, when he received a warning from his retinue concerning the return, he put greater trust in his own purpose and burnt his fleet. He was misled by a deserter who had surrendered for the purpose of leading him astray and was induced to follow a direct route to Madaeana (perhaps Media?).

John Zonaras: Extracts of History, XIII, 13, 2–9:

Then suddenly affairs turned to the worse for him and he and the majority of his army perished. For the Persians in despair decided to rush headlong into destruction in order to do something really terrible to the Romans. Therefore, two men in the guise of deserters hurried to the Emperor and promised him victory over the Persians if he followed them. They advised him to leave the river and burn the galleys he had brought along with the other cargo vessels, so that the enemy could not use them, while they would lead his army to safety through a different way. It would quickly and safely reach the inner parts of Persia and conquer it with ease. The wicked man in his derangement believed them although many told him, and even Hormisdas himself, that it was a trap. But he set fire to his ships and burnt them all except twelve. There were seven hundred galleys and four hundred cargo vessels. After they had been completely burnt, when many of the tribunes objected that what was said by the deserters was a trap and a trick, he reluctantly agreed to examine the false deserters. Questioned under torture, they disclosed their conspiracy.

Some ancient sources put forward the theory that Julian was led astray by one or more Sasanian double agents (Gregory of Nazianzus, Ephraim the Syrian, Festus, Socrates of Constantinople, Sozomenus, Philostorgius, John Malalas, the anonymous Passion of Artemius, and John Zonaras). But in my opinion it’s unnecessary to resort to such novelesque tales to explain Julian’s decision. If he was marching “inland” and possibly along the Diyala River, the fleet would’ve been useless. But even in case of a retreat to the north, the fleet would’ve been mostly an encumbrance due to a simple reason: upstream from Ctesiphon (near present-day Baghdad), the topographical incline of the Tigris riverbed becomes too steep to allow navigation if it’s not with engine-propelled boats or by pulling the boats from the banks (as Libanius correctly stated). Considering that by now a good part of the supplies carried by the fleet would’ve been consumed anyway, at least part of the boats would’ve been empty and so pulling them would’ve been completely useless. And in the case that (as Syvänne and Dodgeon & Lieu assume), the Roman fleet had reached the Tigris downstream from Ctesiphon / Vēh-Ardaxšīr, pulling it past the twin cities would’ve been simply impossible. Still, as almost all the sources (even Ammianus, who was present there) sharply criticize Julian’s decision, it’s to be assumed that the fleet still carried supplies, that now would have to be carried by the pack animals of the army. It’s now when the reason for the attacks of Suren’s cavalry against the pack animals of the army and against its foraging parties became sharply evident: the army simply lacked enough pack animals to carry all the supplies. And possibly it would have also lacked enough of them to pull the fleet upstream, unless Julian used part of his manpower to pull the boats (indeed, both Ammianus and Libanius name this as the reason for Julian’s decision).

A second council of war was held (at Noorda according to Zosimus, perhaps modern Djsir Nahrawan near the
river Tamarra) and the decision to head for Corduene (roughly modern Iraqi Kurdistan) rather than to return via Āsūrestān was agreed upon (or imposed by Julian). The army struck camp on 16 June and marched due north towards the river Douros (considered by most scholars to be the Diyala River). Let’s see Ammianus’ account:

Ammianus Marcellinus (Res Gestae Libri XXXI, XXIV, 8, 2-6):

After these words the prisoners were led away, and a council was held to discuss the situation. And after much interchange of opinion, the inexperienced mob crying that we must return by the way we had come, and the emperor steadfastly opposing them, while he and many others pointed out that it was out of the question to go back through a flat country of wide extent where all the fodder and crops had been destroyed, and where what remained of the burnt villages was hideous from the utmost destitution; moreover, since the frosts of winter were now melting the whole soil was soaked, and the streams had passed the bounds of their banks and becoming raging torrents. Still another difficulty faced the undertaking, in that in those lands heated by the sun's rays, every place is filled with such swarms of flies and gnats that their flight hides the light of day and the sight of the stars that twinkle at night. And since human wisdom availed nothing, after long wavering and hesitation we built altars and slew victims, in order to learn the purpose of the gods, whether they advised us to return through Assyria, or to march slowly along the foot of the mountains and unexpectedly lay waste Chiliocomum, situated near Corduena; but on inspection of the organs it was announced that neither course would suit the signs. Nevertheless, it was decided, since all hope of anything better was cut off, to seize upon Corduena. Accordingly, on the sixteenth day of June, camp was broken, and the emperor was on his way at break of day, when smoke or a great whirling cloud of dust was seen; so that one was led to think that it was herds of wild asses, of which there is a countless number in those regions, and that they were travelling together so that pressed body to body they might foil the fierce attacks of lions. Some believed that Arsaces (i.e. the Armenian king) and our generals (i.e. Sebastianus and Procopius) were coming at last, aroused by the reports that the emperor was besieging Ctesiphon with great forces; and some declared that the Persians had waylaid us.

Notice the chronology, as reconstructed by Dodgeon & Lieu: the Roman army started its invasion (crossing of the Khabur river) on April 4, and the battle of Ctesiphon happened shortly after May 16. In a month and a half, the Roman army had conducted a “lightning” invasion, took two major fortified Sasanian towns/cities (Pērōz-Šābuhr and Maiozamalcha) and reached the enemy capital. But then, this rapid pace of events stagnates; by June 5 Julian decides to “march inland” and between June 11-15 the Roman fleet is burnt and the army departs the environs of Ctesiphon. It seems, according to Ammianus’ account (and the same is suggested by some of the other accounts too) that Julian and his generals really didn’t know what to do. Harrel proposed that Julian’s decision to follow the Diyala River upstream (if that’s in effect the Douros of the classical sources) was an attempt to invade the Iranian Plateau itself: the valley of this river is the main pass across the Zagros, and it leads directly to Ecbatana / Hamadān. This would be a move beyond rash even for a commander given to bold moves like Julian; and Harrel suggests that it was perhaps a way to try to force a decisive field battle against Šābuhr II’s main army. Syvänne, although very critical of Julian’s abilities as a commander, thinks that this was a move merely dictated by geography and that Julian intended indeed to retreat northwards as swiftly as possible. This is also perfectly reasonable: if we look at a map, we can see that the Diyala River joins the Tigris a few km. upstream from Ctesiphon, and so if the Roman army was retreating north along the eastern bank of the Tigris, it would need to find a bridge or a ford to cross it; that justifies perfectly following the river upstream until some crossing point was found.

The decision to retreat along the eastern bank of the Tigris is also a rational one and resorting to elaborate conspiracy theories is unnecessary to explain Julian’s decision. As Libanius points out, the route the army had followed from the Roman border couldn’t be followed now, as the invaders had destroyed all the fields and harvests. Upstream from Ctesiphon, the western bank of the Tigris is mostly barren land (steppe or desert) as it can’t be irrigated by gravity, but the eastern bank is crossed by several rivers that descend from the Zagros Mountains (the Diyala, the Greater Zab and the Little Zab) that allow for the existence of irrigated agriculture, and most importantly: its fields and orchards had not been devastated still by the invading army.

Map of Sasanian Mesopotamia. Notice the course of the Diyala River and its confluence with the Tigris upstream from Ctesiphon.

Another factor that could have influenced Julian’s decision is the possibility to join forces with the Roman army of Sebastianus and Procopius (which should’ve invaded Media or at least Adiabene / Nodšēragān but which had done nothing at all) or the Armenian army, which apparently had managed to invade Media; that could’ve been a factor in the initial march upstream the Diyala River towards Media (Chiliocomum was the classical name for the Nisaean Plain around Ecbatana / Hamadān). But, for one reason or another, neither of these armies joined forces with Julian.

And anyway, between Julian’s army and them (and a safe passage back towards Roman territory) now stood Šābuhr II with his main army, blocking his way.

 

[Semper Victor]

8.10 THE ROMAN RETREAT, THE DEATH OF JULIAN AND THE SECOND PEACE OF NISIBIS.


(NOTE: I forgot to specify in previous chapters that all the dates provided by Ammianus follow the Julian calendar.)

We can’t know for sure what did Julian imagine when he ordered his retreat / march up the Diyala River, but he must’ve soon realized that the Šahān Šāh was not going to make things easy for him. The Roman army was now isolated deep into enemy territory, with supplies running dangerously low, and with three rivers (plus the Tigris) between it and the safety of the fortresses of Roman Mesopotamia. All the sources (even those favorable to Julian) agree that the retreat became a veritable Via Crucis for the Roman forces, hampered by heat (it was now late June in Mesopotamia), hunger (the Sasanians had begun to apply a scorched earth policy), thirst and constant attacks by the Sasanian cavalry. The Roman army managed to cross the Douros against strong Sasanian opposition. These are the sources:

Ammianus Marcellinus: Res Gestae Libri XXXI, XXV,1, 1–3:

Now this night, which was lighted by the gleam of no stars, we passed as is usual in difficult and doubtful circumstances, as fear prevented anyone from daring to sit down or to close his eyes in sleep. But no sooner had the first light of day appeared, than the glittering coats of mail, girt with bands of steel, and the gleaming cuirasses, seen from afar, showed that the king's forces were at hand. Our soldiers, inflamed by this sight, since only a small stream separated them from the enemy, were in haste to attack them, but the emperor restrained them; however, a fierce fight took place not far from our very rampart between our outposts and those of the Persians, in which Machameus, general of one of our battalions, fell. His brother Maurus, later a general in Phoenicia, tried to protect him, and after cutting down the man who had killed his brother, he terrified all who came in his way, and although he was himself partly disabled by an arrow through his shoulder, by main strength he succeeded in bringing off Machameus, already pale with approaching death, from the fray.
And when, because of the almost unendurable heat and the repeated attacks, both sides were growing weary, finally the enemy's troops were utterly routed and fled in all directions. As we withdrew from the spot, the Saracens followed us for some distance but were forced to retreat through fear of our infantry; a little later they joined with the main body of the Persians and attacked with greater safety, hoping to carry off the Romans' baggage; but on seeing the emperor they returned to the cavalry held in reserve.

Zosimus: New History, III, 26, 4–5:

Advancing thence to the river Durus (sic), they constructed a bridge over it for their passage. The Persians had burnt up all the forage of the country, so that the cattle of the Romans were ready to perish with hunger. They were collected into several parties awaiting the Romans, whom they imagined to be but a small number, and presently afterward uniting into one body they proceeded towards the river.
Here, while the advanced guard engaged with a party of Persians, an enterprising man, named Macanaeus, entered among them and killed four of them. For that bold action they all fell upon him and struck him down. His brother, Maurus, upon seeing this, attempted to rescue at least his dead body from the Persians, and killed the man who had given him the first wound; nor did he desist, though frequently shot at, until he had brought off his brother and delivered him to the army still alive.

Libanius: Oration XVIII, 264:

In this way they marched on, drinking of the waters of the Tigris, and keeping that river on the left hand, whilst they were passing through a country more fertile than the former, so that they added to the captives whom they already had, with all confidence. But when they were at the end of the planted land, and in the middle of that bare of trees, though no less fertile, proclamation is made that they must load themselves with provisions for twenty days; for thus long was the march to the noble city which, at the same time, is the boundary of our empire. Then for the first time is beheld the battle-array of the Persians: no disorderly multitude; abundance of gold upon their armor. But when one or two of our vanguard had fallen, and when all had joined battle, neither horseman nor foot-soldier stood the shields on our side, but turned and fled, being inferior in this one branch of warfare.

Notice that curiously, Libanius doesn’t mention the crossing of the Douros River, Zosimus mentions it and Ammianus describes it as “a small stream” (which has caused many scholars to believe this could be the Diyala River). But all three accounts agree that for the first time the Roman army clashed now with the main Sasanian army. The Roman infantry prevailed, but as it was to be expected, the Sasanian cavalry could retreat with impunity, as it usually happened in such clashes. Mention is also made by Ammianus of the “Saracen” allies of the Sasanians also joining in the harassment of the Roman army; as their forces too were formed by cavalry, they could also employ “hit and run” tactics with impunity.

Syvänne’s description of this encounter follows closely Ammianus’ account. following Dodgeon & Lieu’s chronological restitution, he dates the encounter to June 17. The two armies were deployed with the “small stream” between them, and although the Roman soldiers wanted to attack, Julian wisely restrained them; an attack across the river could have disorganized the Roman formation and have made it vulnerable to a charge of the Sasanian savārān. Or the Sasanian cavalry could have retreated, forcing the Romans into a pursuit under the heat of the day, and then counterattack suddenly once the Roman infantry was tired enough. The roman “outposts” (the excursatores) though approached the Sasanian army to harass it and then became locked in a fierce fight that ended in a Roman defeat; all the while the main Roman force remained static, and Syvänne thinks it was probably formed in a hollow square formation, the most advisable formation for a infantry-heavy army when faced by an enemy much superior in cavalry. This inaction by the Romans finally forced the Sasanian army to attack; which (according to Syvänne) would have consisted in the standard Iranian tactic of combining “waves” of heavy cavalry and mounted archers, that would only have pressed the attack home if the Roman formation showed signs of disorder. As Ammianus says nothing of the sort, Syvänne thinks that the Roman infantry resisted the rain of arrows steadfastly and that the Sasanian forces finally abandoned the fight due to the exhaustion of men and horses combined with the heat of the day; but that the attack was then resumed by the Sasanian Arab allies, in two episodes, the second of which needed the personal presence of Julian (and his elite guard units) to force the enemy into retreat.

British reenactor Nadeem Ahmad wearing the garb of a Sasanian Šahān Šāh, with a crown very similar to the one worn by Šābuhr II in his coinage.

If we look back to the start of the invasion, we’ll see that before leaving Roman territory, Rome’s Arab foederati joined forces with Julian’s army, but later they disappear from Ammianus’ account. Syvänne thinks that they probably deserted Julian because the emperor stopped paying them subsidies, and he proposes that a reason for this stop could be Julian’s profligacy at Pērōz-Šābuhr, where he not only bribed the Sasanian defenders, but he also had to appease his troops with a second donativum (he had already handed out a first one at the crossing of the Khabur). It’s an interesting (if completely unsupported by the sources) hypothesis; the absence of the Arab foederati became painfully obvious during the Roman retreat, and if Julian did indeed make such a fragrant mistake, it’s the sort of thing that his admirers (like Ammianus, Libanius and Zosimus) would try to discard as discreetly as possible. But against it, it can also be said that the sources hostile to Julian (and there are many of them) would not have let such a shiny opportunity to drag Julian’s reputation go to waste.

After crossing the Douros, the army kept marching eastwards (according to Dodgeon & Lieu) in the direction of Barsaphtas; the location of Hucumbra (or Symbra, identified by Ernst Herzfeld in 1948 with ‘Ukbara, about 50 km north of Baghdad; it was a city founded by Šābuhr II with the name of Wuzurg-Šābuhr) was reached on June 17. as usual, the sources are Ammianus and Zosimus, but Libanius’ account (his Oration XVIII, known as the Funeral Oration for Julian) becomes very vague after the battle on the Douros and mentions only frequent skirmishes, giving no specific details):


Ammianus Marcellinus, Res Gestae Libri XXXI, XXV, 1, 4:

Leaving this region we came to an estate called Hucumbra, where contrary to our expectation we refreshed ourselves for two days, procuring everything that was useful and an abundance of grain; then we moved on after immediately burning everything except such things as time allowed us to carry off.

Zosimus, New History, III, 27, 1-2:

Afterwards, arriving at the city of Barsaphtas, they found the forage as before burnt up by the barbarians. Perceiving a party of Persians and Saracens, who dared not even look at the Roman army, but immediately fled, the Romans were unable to judge their design, until the Persians, by collecting together into a considerable body, shewed that they had a design upon the beasts of burden.
Upon which the emperor, who immediately armed himself, proceeded with greater expedition against them than the rest of the army. The Persians, unable to sustain the force of his charge, fled to places with which they were well acquainted. The emperor then continued his march to Symbra, which lies between two towns named Nisbara and Nishanadalba (…)

The two accounts differ considerably; and for once Zosimus is more informative. According to Ammianus, when the Roman army reached Hucumbra, they found supplies in there (i.e. for some reason the Sasanians had not implemented scorched earth policies in this location) and the Roman army stopped in there for two days to rest and (presumably) to forage. Zosimus tells us that after reaching Barsaphtas they found supplies left unburned by the enemy, but that the beasts of burden of the Roman army were attacked by the enemy cavalry; which confirms, as Syvänne points out, that the repeated attacks similar to this one against the Roman pack animals since before the battle of Ctesiphon were part of a deliberate Sasanian strategy. Again, the Sasanian attackers dispersed once the Romans counterattacked.

On June 20, the army resumed its march; at this point, the Romans began to suffer from shortage of supplies as the effects of the scorched earth policy of the Sasanians were becoming apparent:

Ammianus Marcellinus, Res Gestae Libri XXXI, XXV, 1, 10:

We then advanced for seventy stadia, while every kind of supplies grew less, since the grass and grain had been burned and every man had to snatch from the very flames whatever produce and fodder he could carry.

Zosimus, New History, III, 26, 4:

Advancing thence to the river Durus (sic), they constructed a bridge over it for their passage. The Persians had burnt up all the forage of the country, so that the cattle of the Romans were ready to perish with hunger. They were collected into several parties awaiting the Romans, whom they imagined to be but a small number, and presently afterward uniting into one body they proceeded towards the river.

John Chrysostom, Homily upon Saint Babylas against Julian and the Heathens XXII/122:

For in fact Julian, at the head of an army whose numbers had never been surpassed under any previous emperor, was fully expecting to overrun the whole of Persia on the spur of the moment, as it were, and without any effort. However, he fared as wretchedly and pitiably as if he was accompanied by an army of women and young children rather than men. For one thing, he brought them to such a pitch of desperation through shortage of provisions that they were reduced to consuming some of their cavalry horses and slaughtering others as they wasted away of hunger and thirst. You would have thought Julian was in league with the Persians, anxious not to defeat them but to surrender his own forces, for he had led them into such a barren and inhospitable region that he in fact surrendered without being defeated.

Theodoret of Cyrrhus, Ecclesiastical History, III, 25, 4:

His soldiers had not enough to eat and drink; they were without guides; they were marching astray in a desert land. Thus, they saw the folly of their most wise emperor.

John Zonaras, Extracts of History, XIII, 13, 13–14:

However, the Romans were hard-pressed by lack of necessities, so Julian, uncertain as to what he should do and from where he ought to return, chose to make the journey through the mountains.

Libanius however in his Oration XVIII said nothing about the Roman army suffering from a scarcity of supplies. The army continued its march up the right bank of the Tigris, passing Danabe, Synce and Accete (all three unidentified locations). In between Danabe and Synce the Sasanians raided the Roman column and lost Adaces (or Daces, according to Zosimus), a distinguished “satrap”, in the combat. Sources:


Ammianus Marcellinus, Res Gestae Libri XXXI, XXV, 1, 5–6:

On the following day, as the army was advancing more quietly, the Persians unexpectedly attacked the last division, which on that day chanced to have the duty of bringing up the rear, and would have slain them with little trouble, had not our cavalry, who were nearby, quickly noticed this, and, spreading widely over the open valleys, prevented so great a disaster, inflicting wounds on those who came up with them. In this battle Adaces, a distinguished satrap, fell; he had once been sent as an envoy to the emperor Constantius and kindly received. The man who killed him brought his armor to Julianus and received the reward which he deserved.

Zosimus, New History, III, 27, 4:

From thence they proceeded to a place between the cities of Danabe and Synca, where the Persians attacked the rear of the army and killed a great number. Their own loss, however, greatly exceeding that of the Romans, and having the disadvantage from many causes, they fled. In this engagement, Daces, a great satrap, was killed. He had formerly been sent on an embassy to the emperor Constantius with proposals of peace.

Although both Ammianus and Zosimus give only a very brief account of this encounter, Syvänne thinks it was a serious attack, as both sources state clearly that the Roman rearguard was almost overwhelmed, and only the intervention of the Roman cavalry prevented a rout of this part of the Roman column. According to Ammianus, Julian proceeded to punish a cavalry unit, the Numerus Tertiaci, for having been the first to flee in front of the Sasanian attack, endangering the whole rearguard. He had their cavalry lances broken and their standards taken from them and ordered the men to follow on foot with the servants of the camp. The only one scathed was its tribune, whom Julian considered to have fought valiantly. On June 20, the Roman army reached Accete (location unknown) and it again found all the crops burnt:

Zosimus, New History, III, 28, 1:

The enemy, upon seeing that the Romans approached a town called Accete, burnt all the produce of the country; but the Romans hastened and, extinguishing the fire, took what they could save for their own use.

On May 22, as the Roman column passed through a district called Maranga (or Maronsa according to Zosimus; identified by Ernst Herzfeld in 1948 with Tell-Hir in Iraq, about 54 km from the modern city of Samarra.), the Sasanian army attacked again the Roman column, but once more the Romans managed to repeal the attack:

Ammianus Marcellinus, Res Gestae Libri XXXI, XXV, 1, 11–19:

Leaving this place as well, the whole army had come to a district called Maranga, when near daybreak a huge force of Persians appeared with Merena, general of their cavalry, two sons of the king, and many other magnates. Moreover, all the companies were clad in iron, and all parts of their bodies were covered with thick plates, so fitted that the stiff joints conformed with those of their limbs; and the forms of human faces were so skillfully fitted to their heads, that, since their entire bodies were plated with metal, arrows that fell upon them could lodge only where they could see a little through tiny openings fitted to the circle of the eye, or where through the tips of their noses they were able to get a little breath. Of these some, who were armed with pikes, stood so motionless that you would think them held fast by clamps of bronze. Hard by, the archers (for that nation has especially trusted in this art from the very cradle) were bending their flexible bows with such wide-stretched arms that the strings touched their right breasts, while the arrow-points were close to their left hands; and by a highly skillful stroke of the fingers the arrows flew hissing forth and brought with them deadly wounds. Behind them the gleaming elephants, with their awful figures and savage, gaping mouths could scarcely be endured by the faint-hearted; and their trumpeting, their odor, and their strange aspect alarmed the horses still more. Seated upon these, their drivers carried knives with handles bound to their right hands, remembering the disaster suffered at Nisibis; and if the strength of the driver proved no match for the excited brute, that he might not turn upon his own people (as happened then) and crush masses of them to the ground, he would with a mighty stroke cut through the vertebra which separates the head from the neck. For long ago Hasdrubal, brother of Hannibal, discovered that in that way brutes of this kind could quickly be killed. Although these sights caused no little fear, the emperor, guarded by troops of armed men and with his trustworthy generals, full of confidence, as the great and dangerous power of the enemy demanded, drew up his soldiers in the form of a crescent with curving wings to meet the enemy. And in order that the onset of the bowmen might not throw our ranks into confusion, he advanced at a swift pace, and so ruined the effectiveness of the arrows. Then the usual signal for battle was given, and the Roman infantry in close order with mighty effort drove the serried ranks of the enemy before them. And in the heat of the combat that followed, the clash of shields, the shouts of the men, and the doleful sound of the whirring arrows continued without intermission. The plains were covered with blood and dead bodies, but the Persian losses were greater; for they often lacked endurance in battle and could with difficulty maintain a close contest man to man, since they were accustomed to fight bravely at long range, but if they perceived that their forces were giving way, as they retreated they would shoot their arrows back like a shower of rain and keep the enemy from a bold pursuit. So by the weight of great strength the Parthians were driven back, and when the signal for retreat was given in the usual manner, our soldiers, long wearied by the fiery course of the sun, returned to their tents, encouraged to dare greater deeds of valor in the future.
In this battle (as was said) the loss of the Persians was clearly the greater, while that of our men was very slight. But noteworthy among the various calamities of the combats was the death of Vetranio, a valiant fighter, who commanded the legion of the Zianni.

Zosimus, New History, III, 28, 2:

In their march from this place they came to a town called Maronsa, where the Persians again attacked the rear-guard, and killed amongst others Vetranio, the captain of a troop, and a brave soldier. They also took several ships, which fell into their power by being considerably behind the army.

As we can see by Ammianus’ description, this was clearly the most serious attack the Roman army had endured to this point. It happened on June 23 (the day after the Romans reached Accete), and Syvänne attempts a reconstruction of the battle according to Ammianus’ account. First, Syvänne assumes that the Sasanian force was blocking the Roman retreat, and that this was what forced Julian to attack (which goes contrary to Zosimus’ account; according to whom the Sasanians attacked again the rearguard). The Sasanian commander according to Syvänne would’ve been the Ērān-spāhbed (the higher rank in the army after the king), with Merena being a corruption of the Middle Persian Mihrān, which was one of the great families of the Iranian wuzurgān. He was accompanied according to Ammianus by two of the sons of the Šahān Šāh, and his army was formed entirely by an elite force of armored cavalry (savārān) and elephants. According to Ammianus, these armored cavalrymen were partly armed with cavalry lances (“pikes”) and partly with bows, with the cavalry in front of the elephants, in a classical Sasanian battle array. I have no idea where he gets this idea, but Syvänne thinks that the Sasanian army would’ve been organized in five corps of 6,000 men each (it’s a formation that exists in later Arabic and New Persian treatises and which is described as “having been used by the ancient Persians” but I’ve found nothing in Ammianus or in Zosimus to suggest that it was being used in this battle). Despite its lack of support in the sources, Syvänne’s reconstruction makes sense if we consider Julian’s actions. A cavalry army of 30,000 men deployed in a line would’ve been long enough to outflank easily the Roman army if it had attacked in a phalanx or linear formation, so Julian, showing he was a good tactician, deployed his army in a convex formation to deny the wings of his army to the Sasanian cavalry (and probably posting the Roman cavalry at both ends of the convex semicircle for good measure and attacked the Sasanian battle array at a quick pace, to shorten the time his soldiers would’ve been exposed to the superior Sasanian archery, forcing the Sasanians into the sorts of hand-to-hand close combat the Romans preferred. Julian’s tactics worked, and according to Ammianus the Romans won the encounter with light losses.

Reconstruction of the battle of Maranga according to Ilkka Syvänne (Military History of Late Rome 361-395).

Despite this success, Zosimus furnishes an important detail: the Sasanians managed to capture some of the boats that the Romans still kept, as they were lagging behind the army. This was probably due to the fact that they had to be towed against the current of the river (quite strong in this part of its course), but it was a small disaster for the Romans, who were already desperately short of supplies and beasts of burden. A truce of three days was agreed upon after the battle, while the Roman army’s supply situation worsened:

Ammianus Marcellinus, Res Gestae Libri XXXI, XXV, 2, 1:

After this three days were devoted to a truce, while each man gave attention to his own wound or his neighbor’s, but since we were without supplies we were tormented by hunger that was already unendurable; and because grain and fodder had everywhere been burned, and both men and animals experienced extreme danger, a great part of the food which the pack-animals of the tribunes and generals carried was distributed even to the lowest soldiers, who were in dire want.

The Romans continued their march northwards along the eastern bank of the Tigris, and on June 25 reached Toummara (according to Ammianus; location unidentified, although François Paschoud located it “not far south of Samarra”) after an exhausting march, during which the Romans regretted the earlier decision to burn the fleet:

Zosimus, New History, III, 28, 3:

The Romans from thence passed hastily along by some villages and arrived at a place called Tummara (sic). Here they regretted the burning of their ships; for the cattle were so exhausted with the fatigue of travelling in an enemy's country, that they were not able to carry all the necessaries; and the Persians collected all the provender they could, and stored it in their strongest fortresses that it might not fall into the hands of the Romans. When they were thus situated they perceived the Persian army, with which they engaged, and having considerably the advantage, they killed a great number of Persians.

At Toummara, the Roman army was again attacked by the main Sasanian army, and although once more the Romans were able to repeal the attack, the Sasanian assault almost succeeded, and more importantly: Julian was mortally wounded in the fight and died shortly afterwards. The sources report it thus:

Ammianus Marcellinus, Res Gestae Libri XXXI, XXV, 3:

When we marched on from this place, the Persians, since their frequent losses made them dread regular battles with the infantry, laid ambuscades, and secretly attended us, from the high hills on both sides watching our companies as they marched, so that the soldiers, suspicious of this, all day long neither raised a palisade nor fortified themselves with stakes. And while the flanks were strongly protected and the army, as the nature of the ground made necessary, advanced in square formation, but with the battalions in open order, it was reported to the emperor, who even then unarmed had gone forward to reconnoiter, that the rear guard had suddenly been attacked from behind. Excited by the misfortune, he forgot his coat-of‑mail, and merely caught up a shield in the confusion; but as he was hastening to bring aid to those in the rear, he was recalled by another danger: the news that the van, which he had just left, was just as badly off. While he was hastening to restore order there without regard to his own peril, a Parthian band of mailed cavalry on another side attacked the center companies, and quickly overflowed the left wing, which gave way, since our men could hardly endure the smell and trumpeting of the elephants, they were trying to end the battle with pikes and volleys of arrows. But while the emperor rushed hither and thither amid the foremost ranks of the combatants, and as the Persians turned in flight, they hacked at their legs and backs, and those of the elephants. Julianus, careless of his own safety, shouting and raising his hands tried to make it clear to his men that the enemy had fled in disorder, and, to rouse them to a still more furious pursuit, rushed boldly into the fight. His guards, who had scattered in their alarm, were crying to him from all sides to get clear of the mass of fugitives, as dangerous as the fall of a badly built roof, when suddenly (no one knows whence) a cavalryman's spear grazed the skin of his arm, pierced his ribs, and lodged in the lower lobe of his liver. While he was trying to pluck this out with his right hand, he felt that the sinews of his fingers were cut through on both sides by the sharp steel. Then he fell from his horse, all present hastened to the spot, he was taken to camp and given medical treatment. And soon, as the pain diminished somewhat, he ceased to fear, and fighting with great spirit against death, he called for his arms and his horse in order by his return to the fight to restore the confidence of his men, and troubling nothing about himself, to show that he was filled with great anxiety for the safety of the others; with the same vigor, though under different conditions, with which the famous leader Epaminondas, when mortally wounded at Mantinea and carried from the field, took particular care to ask for his shield. And when he saw it near him, he died of his terrible wound, happy; for he who gave up his life without fear dreaded the loss of his shield. But since Julianus's strength was not equal to his will, and he was weakened by great loss of blood, he lay still, having lost all hope for his life because, on inquiry, he learned that the place where he had fallen was called Phrygia. For he had heard that it was fate's decree that he should die there. But when the emperor had been taken to his tent, the soldiers, burning with wrath and grief, with incredible vigor rushed to avenge him, clashing their spears against their shields, resolved even to die if it should be the will of fate. And although the high clouds of dust blinded the eyes, and the burning heat weakened the activity of their limbs, yet as though discharged by the loss of their leader, without sparing themselves, they rushed upon the swords of the enemy. On the other hand, the exulting Persians sent forth such a shower of arrows that they prevented their opponents from seeing the bowmen. Before them slowly marched the elephants, which with their huge size of body and horrifying crests, struck terror into horses and men. Further off, the trampling of the combatants, the groans of the falling, the panting of the horses, and the ring of arms were heard, until finally both parties were weary of inflicting wounds and the darkness of night ended the battle. On that day fifty Persian grandees and satraps fell, besides a great number of common soldiers, and among them the distinguished generals Merena and Nohodares were slain. The boastfulness of antiquity may view with amazement the twenty battles of Marcellus in various places; it may add Sicinius Dentatus, honored with a multitude of military crowns; it may besides admire Sergius, who (they say) was wounded twenty-three times in different battles, and whose last descendant Catiline tarnished the glorious renown of these victories with an indelible stain. Yet the joy in our success was marred by sorrow. For while the fight went on everywhere after the withdrawal of the leader, the right wing of the army was exhausted, and Anatolius, at that time chief marshal of the court, was killed. Salutius, the prefect, was in extreme danger, but was saved by the help of his adjutant, and by a fortunate chance escaped death, while Phosphorius, a counselor who chanced to be at his side, was lost. Some of the court officials and soldiers, amid many dangers, took refuge in a neighboring fortress, and were able to rejoin the army only after three days.
While all this was going on, Julianus, lying in his tent, addressed his disconsolate and sorrowful companions as follows: "Most opportunely, friends, has the time now come for me to leave this life, which I rejoice to return to Nature, at her demand, like an honorable debtor, not (as some might think) bowed down with sorrow, but having learned from the general conviction of philosophers how much happier the soul is than the body, and bearing in mind that whenever a better condition is severed from a worse, one should rejoice rather than grieve. Thinking also of this, that the gods of heaven themselves have given death to some men of the greatest virtue as their supreme reward. But this gift, I know well, was given to me, that I might not yield to great difficulties, nor ever bow down and humiliate myself; for experience teaches me that all sorrows overcome only weaklings but yield to the steadfast. I do not regret what I have done, nor does the recollection of any grave misdeed torment me; either when I was consigned to the shade and obscurity, or after I attained the Principate, I have preserved my soul, as taking its origin from relationship with the gods, stainless (in my opinion), conducting civil affairs with moderation, and making and repelling wars only after mature deliberation. And yet success and well-laid plans do not always go hand in hand, since higher powers claim for themselves the outcome of all enterprises. Considering, then, that the aim of a just rule is the welfare and security of its subjects, I was always, as you know, more inclined to peaceful measures, excluding from my conduct all license, the corrupter of deeds and of character. On the other hand, I depart rejoicing that, so often as the state, like an imperious parent, has exposed me deliberately to dangers, I have stood four-square, accustomed as I am to tread underfoot the storms of fate. And I shall not be ashamed to admit, that I learned long ago through the words of trustworthy prophecy, that I should perish by the sword. And therefore I thank the eternal power that I meet my end, not from secret plots, nor from the pain of a tedious illness, nor by the fate of a criminal, but that in the mid-career of glorious renown I have been found worthy of so noble a departure from this world. For he is justly regarded as equally weak and cowardly who desires to die when he ought not, or he who seeks to avoid death when his time has come. So much it will be enough to say, since my vital strength is failing. But as to the choice of an emperor, I am prudently silent, lest I pass over some worthy person through ignorance, or if I name some of whom I consider suitable, and perhaps another is preferred, I may expose him to extreme danger. But as an honorable foster-child of our country, I wish that a good ruler may be found to succeed me”.
After having spoken these words in a calm tone, wishing to distribute his private property to his closer friends, as if with the last stroke of his pen, he called for Anatolius, his chief court-marshal. And when the prefect Salutius replied "He has been happy," he understood that he had been slain, and he who recently with such courage had been indifferent to his own fate, grieved deeply over that of a friend. Meanwhile, all who were present wept, whereupon even then maintaining his authority, he chided them, saying that it was unworthy to mourn for a prince who was called to union with heaven and the stars. As this made them all silent, he himself engaged with the philosophers Maximus and Priscus in an intricate discussion about the nobility of the soul. Suddenly the wound in his pierced side opened wide, the pressure of the blood checked his breath, and after a draught of cold water for which he had asked, in the gloom of midnight he passed quietly away in the thirty-second year of his age. Born in Constantinople, he was left alone in childhood by the death both of his father Constantius (who, after the decease of his brother Constantinus, met his end with many others in the strife for the succession to the throne) and of his mother Basilina, who came from an old and noble family.

Zosimus, New History, III, 28, 4–29, 1:

Upon the following day, about noon, the Persians drew up in a large body, and once more attacked the rear of the Roman army. The Romans, being at that time out of their ranks, were surprised and alarmed at the suddenness of the attack yet made a stout and spirited defense. The emperor, according to his custom, went around the army, encouraging them to fight with ardor.
When by this means all were engaged, the emperor, who sometimes rode to the commanders and tribunes, and was at other times among the private soldiers, received a wound in the heat of the engagement, and was borne on a shield to his tent. He survived only till midnight. He then expired, after having nearly subverted the Persian Empire.

Libanius, Oration XVIII, 268–74:

Up to this point he was advancing in his career of victory, and it is a pleasure to me to speak, but thenceforth, O ye gods and genii and vicissitudes of fortune! to what a tale am I compelled! Do you wish that I should hold my tongue, and end the history with its more auspicious part? It would be to you much comfort, instead of a source of lamentation. What, then, is your decision: must we shut up, or proceed? You appear to me to be sorrow-stricken by the fact, but to crave the account of it. It is indeed necessary I should speak out and put an end to the false reports current concerning his end. For when the Persian was already reduced to despair, having been manifestly conquered, and in fear lest our troops should occupy the best places in his kingdom, and winter there; when he was choosing envoys, was counting out presents (amongst which was a crown), and intending, it is said, to send them on the following day, together with a supplication for peace, and to leave him (i.e. Julian) the arbiter of the terms: a part of the army is separated from the rest, from some troops having to resist an attack, and the others going on without perceiving it, and a brisk breeze at the same time stirring up the dust and producing a cloud, and giving cover to those who wished to commit the crime, the emperor hastened up with one attendant for the purpose of uniting the broken line, when a horseman's spear cast at him, being without armor (for he, on account, I suppose, of his being so much the stronger, did not even arm himself), passed through his arm and entered into his side. The hero fell, and seeing the blood pouring forth, but wishing to conceal the disaster, remounted his horse, when, as the blood betrayed the wound, he kept crying out to those he successively met "not to be alarmed at his hurt, for that it was not mortal". He said this, but at the same time was sinking under the danger; and is carried to the tent, to the black bed, the lion's hide, and mattress, for such was his couch. And when the surgeons pronounced there was no hope of life, the army receiving the news of his death all set up a wail, all beat the breast, by all was the around moistened with tears; their weapons escaped from their hands, and were thrown away; for they thought that not even one to carry the tidings would return from thence home. The Persian king dedicated to the gods his saviors, those gifts which he ought to have sent to him; and had customary table set up before him, having before that made the earth serve in its place; he decorated his head according to custom, which had been neglected all the time of his danger; and all things that he would have done if his adversaries had been swallowed up root and branch by the earth opening her mouth, the same way did he behave because a single young man was come to his end. Both sides, therefore, gave their vote that the existence of the Romans was locked up in him, the one as they mourned, the other as they exulted; the one as they deemed they were lost, the other as they believed they had already conquered. One may discern his excellence even from his dying words; for when all who stood round him had fallen a-weeping, and not even the masters of philosophy could master their feelings, he rebuked the others, and especially the latter persons, because "when his past life was bringing him to the islands of the Blessed, they wept for him as though he had spent his life so as to deserve Tartarus”. The scene was like the prison that contained Socrates; those present resembled those that were present with that philosopher; the wound the cup of poison; the words his words; whilst the circumstance that Socrates shed not a tear was paralleled by our hero's doing the same. But when his friends besought him to name a successor to the empire, inasmuch as he saw nobody like himself at hand, he referred the election to the army; and him he urges to do his best to save the troops, for that he in preserving them had endured every toil.

Eunapius of Sardes, fragments 20 and 26:

When the troop of mounted cataphracts over and above four hundred rushed down upon the rear-guard.

[Concerning the end of Julian, the Apostate and Soldier, the reply of the oracle was as follows]
But when that time comes when you shall tame the Persian blood with your scepter and bring them beneath your rule, driven back as far as Seleucia by your sword, then indeed a chariot bright as fire leads you to Olympus, a chariot swallowed up in the turmoil of whirlwinds and lightning, leaving behind the wretched distress of human limbs. And you shall come to the ancestral halls of the ethereal light, whence you came, led astray to take the form of a human body.

Inspired by such eloquent words as these, and even more by prophecies, they say he was most agreeably exalted above mortal destruction …

There is an oracle which was given to him while he was waiting at Ctesiphon: “Zeus, the all-wise, once destroyed a race of Earth-born giants most hateful to the blessed ones who dwell in the Olympian halls. Julian the godlike, Emperor of the Romans, contending for the cities and long walls of the Persians, fighting hand-to-hand, destroyed them with fire and the valiant sword, subdued without pause their cities and many races. Seizing also, with heavy fighting, the German soil of those people of the west he laid waste their land”.

Festus, Breviarium, 28:

Retracing his route upstream along the Tigris (with the river on his right), he exposed the flank of his troops. When he wandered along the column too incautiously, after the dust was stirred up and he lost sight of his own men, he was pierced through the abdomen as far as the groin by the lance of an enemy horseman. Amidst the excessive loss of blood, when in spite of having been wounded he had restored the ranks of his army, and after a long address to his companions, he breathed forth his lingering life.

Ephraim the Syrian, Hymns against Julian, III, 16:

When he saw that his gods were refuted and exposed,
and that he was unable to conquer and unable to escape,
he was prostrated and torn between fear and shame.
Death he chose so that he might escape in Sheol
and cunningly he took off his armor in order to be wounded
so that he might die without the Galileans seeing his shame.

Gregory of Nazianzus, Oration V, 13–14:

Up to this point, such is the universal account; but thence-forward, one and the same story is not told by all, but different accounts are reported and made up by different people, both those present at the battle, and those not present; for some say that he was hit by a dart from the Persians, when engaged in a disorderly skirmish, as he was running hither and thither in his consternation; and the same fate befell him as it did to Cyrus, son of Parysatis, who went up with the Ten Thousand against his brother Artaxerxes, and by fighting inconsiderately threw away the victory through his rashness. Others, however, tell some such story as this respecting his end: that he had gone up upon a lofty hill to take a view of his army and ascertain how much was left him for carrying on the war; and that, when he saw the number considerable and superior to his expectation, he exclaimed: “What a dreadful thing if we shall bring back all these fellows to the land of the Romans!” as though he begrudged them a safe return. Whereupon one of his officers, being indignant and not able to repress his rage, ran him through the bowels, without caring for his own life. Others tell that the deed was done by a barbarian jester, such as follow the camp, “for the purpose of driving away ill humor and for amusing the men when they are drinking”. Some give this honor to one of the Saracens. At any rate, he received a wound truly seasonable (or mortal) and salutary for the whole world, and by a single cut from his executioner he pays the penalty for the many entrails of victims to which he has trusted (to his own destruction); but what surprises me, is how the vain man, that fancied he learned the future from that means, knew nothing of the wound about to be inflicted on his own entrails! The concluding reflection is for once very appropriate: the liver of the victim was the approved means for reading the future, and it was precisely in that organ that the arch-diviner received the fatal thrust.

John Chrysostom, Homily upon Saint Babylas against Julian and the Heathens, XXII/123:

Even those who were eyewitnesses or forced to experience the many disasters which befell the campaign could hardly begin to describe the full picture, for it defies description. To come to the gist of the matter, Julian died disgracefully and without honor, for some say that he was mortally wounded by one of his own baggage-bearers, out of resentment at the appalling predicament of the army. Another version says that he died at the hands of an unknown assassin, recounting only that he was struck down and that he requested that he should be buried in Cilicia, where his body remains to this day.

Socrates of Constantinople, Ecclesiastical History, III, 21, 9–18:

Wherefore the latter, convinced of the uselessness of them, was constrained to prepare for conflict, and therefore on the next day after the rejection of his embassy, he drew out in order of battle all the forces he had. The Romans indeed censured their prince for not avoiding an engagement when he might have done so with advantage: nevertheless, they attacked those who opposed them, and again put the enemy to flight. The emperor was present on horseback and encouraged his soldiers in battle; but confiding simply in his hope of success, he wore no armor. In this defenseless state, a javelin was thrown at him unexpectedly and pierced his arm and entered his side. In consequence of this wound he died, and the identity of the killer was unknown. Some say that a certain Persian deserter hurled the javelin; others assert that one of his own men was the author of the deed, which indeed is the best corroborated report. But Callistus, one of his bodyguards, who celebrated this emperors’ deeds in heroic verse, says, in narrating the particulars of this war, that the wound of which he died was inflicted by a demon. This is possibly a mere poetical fiction, or perhaps it was really the fact; for vengeful furies have indeed destroyed many persons. Be the case however as it may, this is certain, that the ardor of his natural temperament rendered him incautious, his learning made him vain, and his affectation of clemency exposed him to contempt. Thus, Julian ended his life in Persia, as we have said, in his fourth consulate, which he bore with Sallust his colleague. This event occurred on the 26th of June, in the third year of his reign, and the seventh from his having been created Caesar by Constantius, being at that time in the thirty first year of his age.

Sozomenus, Ecclesiastical History, VI, 1, 13–16:

In the heat of the conflict which ensued, a violent wind arose; and the sky and the sun were totally concealed by the clouds, while the air was at the same time mixed with dust. During the darkness which was thus produced, a horseman, riding at full gallop, directed his lance against the emperor, and wounded him mortally. After throwing Julian from his horse, the unknown assailant secretly went away. Some conjectured that he was a Persian; others, that he was a Saracen. There are those who insist that he who struck the blow was a Roman soldier, who was indignant at the imprudence and temerity which the emperor had manifested in exposing his army to such peril …

Theodoret of Cyrrhus, Ecclesiastical History, III, 25, 5–7,

In the midst of their murmuring and grumbling, they suddenly found him who had struggled in mad rage against his Maker wounded to death. Ares who raises the war-din had never come to help him as he promised; Loxias had given lying divination; he who rejoices in the thunderbolts had hurled no bolt on the man who dealt the fatal blow; the boasting of his threats was dashed to the ground. The name of the man who dealt that righteous stroke no one knows to this day. Some say that he was wounded by an invisible being, others by one of the Nomads who were called Ishmaelites; others by a trooper who could not endure the pains of famine in the wilderness.

Philostorgius, Ecclesiastical History, VII, 15:

The apostate Julian undertook an expedition against the Persians, relying upon the prophecies of the heathen oracles in different quarters, that his might would prove irresistible. But a certain old man, one of those who had long since been discharged from the Persian service, approached the Apostate as he was making war in Persia. And when he had brought the Romans into the greatest straits by leading them into a pathless desert, in which a very great portion of the army perished, he gave the enemy, like the prey of a hunter, into the hands of his countrymen. For the Persians rushed upon the Romans, having joined to their forces as allies some Saracen horsemen who were armed with spears. One of them thrust a spear at Julian, which struck him forcefully on the thigh near the groin; and when the spear was drawn out, it was followed by a quantity of bile and blood also. Subsequently, one of the bodyguards of the emperor immediately attacked the Saracen who had wounded the king and cut off his head: while the Romans immediately placed the mortally wounded Julian on a shield and carried him off into a tent. Many even thought that the fatal blow was struck by Julian’s own friends, so sudden and unexpected was it, and so much at a loss were they to know where it came from. But the wretched Julian took up in his hands the blood which flowed from his wound, and cast it up towards the sun, exclaiming in a clear voice, “Take thy fill!” and he added curses upon the other gods as villains and destroyers. In his train was a most distinguished physician, one Oribasius, a native of the Lydian city, Sardis. But the wound was far beyond all medical art, and carried Julian off after three days of suffering, after he had enjoyed the dignity of Caesar for five years, and the imperial throne two years and a half from the death of Constantius. Philostorgius in this passage writes that Julian sprinkled his blood towards the sun and cursed his gods. But most historians write that he used this act as an expression of hatred against our Lord and only true God, Jesus Christ.

John Malalas, Chronographia, XIII, 21–333:

On the next day, the twenty-sixth of June, he brought out the Persians who had misled him and examined them. They confessed with the words “For the sake of our country and our king, that he might be saved, we gave ourselves to death and deceived you. Now, as your slaves, we die”. He believed them and did not kill them but gave them his promise if they would lead the army out of the desert area.
About the second hour of the same day, the Emperor Julian was walking among the army and urging them not to behave in an undisciplined manner when he was wounded by someone unknown. He went into his tent and died during the night as Magnus (i.e. Magnus of Carrhae), whom we referred to earlier, relates. However, Eutychianus, the historian from Cappadocia, who was a soldier and a vicarius of his unit of the Primoarmeniaci (Legio I Armeniaca?), and who was himself present in the war, wrote that the Emperor Julian entered Persian territory by way of the Euphrates for fifteen days’ marching. There he was victorious and conquered and took everything as far as the city called Ctesiphon which was the seat of the Persian king. The latter fled to the territory of the Persarmenians, while Julian decided with his Council and his army to set out for Babylon on the next day and to take it by night. While he slept he saw in a dream an adult man, wearing a cuirass, approaching him in his tent near the city of Ctesiphon, in a city called Asia and smiting him with a spear. Distraught, he awoke and cried out. The eunuchs of the bedchamber and the bodyguard and the unit who guarded the tent arose and came to him with royal torches. When the Emperor Julian realized that he had been fatally wounded in the armpit, he asked them, “What is the name of the village where my tent is?”, and they told him that it was called Asia. Then he immediately cried out, “O Sun, you have slain Julian”, and having lost blood, he died at the fifth hour of the night in the year 411 of the era of Antiochus the Great.

Passion of Artemis, 69:

But immediately, in this state of distress, they encountered against their will the army of the Persians. A battle occurred and while Julian himself was rushing here and there and organizing his men, he fell to the spear, so it is said, of a soldier; but, as others record, to the spear of a Saracen serving with the Persians. But in the true Christian version, which is ours, the spear belonged to the Lord Christ who was ranged against him. For suddenly a bow from the skies stretched taut and launched a missile at him as at a target and it pierced through his side and wounded him in the abdomen. And he wailed deeply and woefully and thought that our Lord Jesus Christ was standing before him and exulting over him. But he, filled with darkness and madness, received his own blood in his hands and sprinkled it into the air, and when he became breathless shouted out, saying: “You have won, Christ. Take your fill, Galilean!” Thus, he ended his life in a most fearful and hateful manner after many reproaches had been heaped upon his own gods.

Quite obviously, the most detailed and reliable account is Ammianus’ one. This time, the Sasanian attack was well-coordinated, with successive attacks against different parts of the Roman column to try to overwhelm the defenders. Syvänne guesses (he calls it an “educated guess”, for he has no documentary proof in this respect) that the marching order of the Roman army was the same one followed during the first part of the invasion, when it was advancing along the Euphrates. This time, the army would have been marching northwards with the Tigris on its left, but far from the river (as Ammianus describes that the army was attacked from both flanks). The infantry would be deployed in a hollow square, with the baggage train in its center. The cavalry would be distributed along the flanks of the army, with a vanguard and a rearguard also formed by cavalry. The cavalry rearguard would’ve been commanded by Secundinus and the cavalry vanguard by Julian himself, and it would’ve been formed by the elite scholae units.

Reconstruction of the battle of Toummara (or “Sumara”) according to Ilkka Syvänne (Military History of Late Rome 361-395).

The Sasanian attack was well planned; as it seems to have been a sudden attack and to have proceeded very quickly, Syvänne assumes that only the Sasanian cavalry and elephants were involved (and indeed Ammianus does not mention the Sasanian infantry at any point). First, a Sasanian force attacked the Roman rearguard, causing it to stop.

When Julian received the news, he immediately turned back with his elite cavalry vanguard to go in the rearguard’s help, most probably by crossing through the hollow interior of the square. The attack was so sudden that Julian, who was not wearing armor (remember it was summer in Mesopotamia) just had time to pick up a shield before rushing to the danger point. But when he was hurrying to the rearguard, he received news that the vanguard was under attack too, and so he turned back and managed to restore the situation, at great personal risk. Syvänne states (and I agree with him) that the Sasanian plan was to force the Roman column to stretch out and to stop, weakening its flanks as the Romans sent help to both the rearguard and the vanguard. Because effectively, it was at this moment when the Sasanians attacked both flanks of the Roman column. The most dangerous attack was the one against its left flank (for some reason, the Romans were marching inland, and this flank was not protected by the Tigris), commanded by Nevitta. Here, a unit of “Parthian” cavalry (Syvänne assumes that Ammianus meant here “Parthians from the region of Parthia” as opposed to an antiquated use of the term instead of the then common term of “Persians”; in a previous part of his work Ammianus had referred to the “Parthians” as the elite of the Sasanian heavy cavalry) supported by elephants attacked the Roman cavalry screen and dispersed it quickly (yet another show of the inferiority of Roman cavalry vs. its Sasanian counterparts) and then proceeded to subject the infantry of the Roman left flank to a deadly attack with its kontoi and arrows that threatened to cause the Roman lines to break.

Julian and his escort rushed to the spot and succeeded in causing the enemy to flee; Julian’s guardsmen then proceeded to pursue the retreating enemy, and Julian joined them; it was at this moment that he was struck by a cavalry hasta (equestris hasta) that grazed his arm and lodged in his liver. He fell from his horse and his guards immediately took him back inside the square. Ammianus states explicitly that Julian was not killed by Roman conspirators, although his use of the expression equestris hasta is surprising, because this was a Roman cavalry thrusting weapon; the Latin term for “javelin” in Ammianus’ time was lancea, and the heavy and very long Sasanian cavalry pike was called contus in Latin. If you read the above sources, there are almost as many theories about the identity of Julian’s killer as authors: a disaffected Roman, a Sasanian cavalryman, an Arab in Sasanian service, an Arab in Roman service, a demon, the pagan gods / Fate or even Jesus Christ Himself.

The so-called Column of Julian in Ankara (ancient Ancyra, Turkey). It’s been traditionally assumed to have been built in occasion of Julian’s visit to this city in 362 CE, but some scholars think it was built a century or more later.

Upon seeing this, the Sasanian retreating forces regrouped and attacked again the left flank, where the fight would continue until the late afternoon. As this happened in the left flank, the right flank under Hormisdas (and perhaps also Arintheus, if the command arrangements were the same as during the first part of the invasion) also came under attack and here the Sasanian attackers were very successful; the cavalry screen was also overwhelmed and the attackers were able to break through the infantry lines and enter the hollow center of the square, where they killed many of the court officials that had accompanied the expedition, like the Magister Officiorum Anatolius, and the Praetorian Prefect of the East Salutius escaped death narrowly when his horse was killed (his secretary Phosphorius was also killed near him).

As Syvänne states, it’s unclear how did the Romans manage to escape complete disaster and force the Sasanians to retreat, because all the sources concentrate on narrating Julian’s death, but they managed to do so, inflicting heavy casualties on their enemy, among them their commander the (supposed) Ērān-Spāhbed Mihrān, the Romans’ old enemy Nohodares, as well as many other high-ranking nobles and officers. Following Ammianus’ account, the following day (June 27) the commanders of the army called to a meeting all the commanders of legions and cavalry turmae (which according to Syvänne meant all the leaders of units stronger than 512 men on paper, plus the officers and supernumeraries) to elect a new emperor. As was to be expected, the officers were divided in two blocks: one led by Arintheus, Victor and other old members of Constantius II’s staff, and were mainly officers of the eastern army; the western or “Gallic” block was formed by officers who had served (and risen in the ranks) in Gaul under Julian and was led by Nevitta and Dagalaifus. In the end, they agreed to elect the Praetorian Prefect of the East Salutius as new augustus: he was the only member of the “high command” present at the site who was a Roman by birth, and he was a pagan who had refused to prosecute Christians, which made him a good consensus choice. But Salutius refused to accept on the grounds of his advanced age and poor health. This caused a real problem, as there were no Romans of high enough rank left for the post.

It was at this moment that some “hotheads” (in Ammianus’ words) chose Jovian, an officer of the Domestici (primicerius domesticorum) as the new augustus. His sole recommendation for the post was the reputation his recently retired father Varronianus, who had been commander of the Domestici (comes domesticorum). According to Ammianus, Jovian was elected by the camp-followers, but it’s difficult to accept his claim, he was probably acclaimed as augustus by his companions of the Protectores Domestici, who dressed him in purple robes in his tent and then presented him to the troops, and the troops acclaimed him as their new emperor without even knowing who he was. Syvänne suggests that the reason for Ammianus’ strange claim that the Domestici had nothing to do with this strange event is that he himself was an officer of this corps, and did want to hide or at least obscure his own role in what was essentially a (bloodless) coup.

The following day (June 27) the signifer (standard bearer) of the Legio Ioviani, who had been commanded by Varronianus and who was a mortal enemy of both Jovian and his father, deserted immediately to the Sasanians after hearing the news of Jovian’s rise to the purple. He informed Šābuhr II that Julian had been killed and that “a weak man” had been chosen as successor, and the Šahān Šāh immediately ordered a new attack against the Roman army, which took place the following day, as recorded in the sources:

Ammianus Marcellinus, Res Gestae Libri XXXI, XXV, 6, 2–3:

But when we accordingly were just beginning to leave, the Persians attacked us, with the elephants in front. By the unapproachable and frightful stench of these brutes, horses and men were at first thrown into confusion, but the Ioviani and Herculiani, after killing a few of the beasts, bravely resisted the mail-clad horsemen. Then the legions of the Iovii and the Victores came to the aid of their struggling companions and slew two elephants, along with a considerable number of the enemy. On our left wing some valiant warriors fell, Iulianus, Macrobius and Maximus, tribunes of the legions which then held first place in our army.

Zosimus, New History, III, 30, 2–4:

When Jovian had assumed the purple and the diadem, he directed his course homewards with all possible speed. Arriving at the castle of Suma, he was attacked by the Persian cavalry, accompanied by a great number of elephants, which committed great devastation in the right wing of the army, in which were placed the Ioviani and Herculiani. These were the appellations of two legions, so named from Diocletian and Maximian, the former of whom assumed the surname of Jupiter, and the latter that of Hercules.
Although at first they were unable to sustain the shock of the elephants, yet when the Persians with their horses and elephants in one body approached them, and happened to arrive at a rising ground, on which were the carriages of the Romans and those who had the care of them, they availed themselves of the advantage to throw darts from above upon the Persians, with which they wounded the elephants. Upon feeling the smart of their wounds, the elephants, in their usual manner, immediately fled, breaking the line of the cavalry. The soldiers were thus enabled to kill the elephants in their flight, and numbers of the enemy.
There fell also on the Roman side, three tribunes, Iulianus, Maximianus, and Macrobius. They then marched forward four days, continually harassed by the enemy, who followed them when they were proceeding, but fled when the Romans offered any resistance. At length, having gained some distance of the enemy, they resolved to cross the Tigris.

Reenactors from the group Legio Iovani Seniores in historical garb; the design on their shields is the one that appears in the Notitia Dignitatum.

The Romans were thus able to repulse yet another Sasanian attack, in which the Sasanians had put their elephants in front of their cavalry in the hope of being able to break the Roman formation. But despite their victory, the loss of three legionary tribunes (the title that the commander of a legion received during the IV century CE) seems to indicate that the Romans also suffered important losses. Zosimus’ account, which this time is more detailed than Ammianus’, also informs us that the Sasanian harassment continued for the next four days while the Romans kept retreating. A few days later the army found refuge at Charcha (identified by Ernst Herzfeld in 1948 with Karkh Fairuz in Iraq, about 11 km north of Samarra; according to Dodgeon and Lieu, this would’ve happened around June 29-30) and on July 1 it reached Dura (not Dura Europos; identified by Ernst Herzfeld with Dur Arabaya, about 5 km north of Charcha), where they were delayed for four days by Sasanian attacks:

Ammianus Marcellinus, Res Gestae Libri XXXI, XXV, 6, 9–11:

And from here, having completed a march of thirty stadia, on the first of July we reached a city called Dura. Our horses were tired, and their riders, who marched on foot and fell to the rear, were surrounded by a throng of Saracens, and would at once have perished, had not some squadrons of our light-armed cavalry brought help to them in their distress. We found these Saracens hostile for the reason that they had been prevented by Julian's order from receiving pay and numerous gifts, as in times past, and when they complained to him, had received the simple reply that a warlike and watchful emperor had steel and not gold. In this place the persistence of the Persians delayed us for four days. For when we began to march, they followed us, and by frequent onsets forced us to turn back; if we halted to do battle with them, they little by little retired and harassed us by continual delays. But now (since to those who are in fear of the worst even false reports are commonly welcome) the rumor was circulated that the frontiers of our possessions were not far distant; whereupon the army, with mutinous bluster, demanded that they be allowed to cross the Tigris.

Notice Ammianus’ remark about the “Saracens” that seems to confirm Syvänne’s hypothesis about the desertion of the Roman Arab allies because Julian had not been paying them. The Romans seized a bridgehead on the opposite bank of the Tigris by a daring raid, but the strong current prevented the Tigris from being successfully bridged:

Ammianus Marcellinus, Res Gestae Libri XXXI, XXV, 6, 11 – 7, 4:

In this place the persistence of the Persians delayed us for four days. For when we began to march, they followed us, and by frequent onsets forced us to turn back; if we halted to do battle with them, they little by little retired and harassed us by continual delays. But now (since to those who are in fear of the worst even false reports are commonly welcome) the rumor was circulated that the frontiers of our possessions were not far distant; whereupon the army, with mutinous bluster, demanded that they be allowed to cross the Tigris. The emperor, as well as the generals, opposed them, and pointing to the river, which was in flood, since the dog-star had already risen, begged them not to trust themselves to the dangerous currents, declaring that very many could not swim, and adding that the scattered bands of the enemy had beset the banks of the swollen stream in various places. But when these warnings, though several times repeated, had no effect, and the loud shouts of the excited soldiers threatened violence, Jovian reluctantly consented that the Gauls, mingled with the northern Germans, should enter the river first of all, to the end that if these were swept away by the force of the stream, the obstinacy of the rest might be broken down; or if they accomplished their purpose without harm, the rest might try to cross with greater confidence. For this attempt the most skillful men were chosen, who from early childhood were taught in their native lands to cross the greatest of all river, and as soon as the quiet of night gave an opportunity for concealment, as if starting all together in a race, they gained the opposite bank more quickly than could have been expected, and after trampling under foot and killing a great number of the Persians, who had been posted to guard the places, but from a feeling of security were buried in quiet sleep, they raised their hands and waved their mantles, to show that their bold attempt had succeeded. 15 When this was seen from afar, the soldiers, now eager to cross, were delayed only by the promise of the pontoon builders to make bridges of bladders from the hides of slain animals.
While these vain attempts were being made, King Sapor, both when far away and when he had come near, learned from the true accounts of scouts and deserters of the brave deeds of our men and the shameful defeats of his army, accompanied by a greater loss of elephants than he had ever known in his reign; also that the Roman army, inured to constant hardship after the loss of their glorious leader, were looking out (as they said), not for their safety, but for revenge, and would end the difficulties of their situation by either a decisive victory or a glorious death. This news filled his mind with fear for many reasons: for he knew by experience that the troops scattered in great numbers through the provinces could easily be assembled by one little ticket, and he was aware that his own subjects, after the loss of so many men were in a state of extreme panic, and, besides, that in Mesopotamia a Roman army had been left which was not much smaller. More than all, it dulled his anxious mind that five hundred men together in one swim had crossed unharmed the swollen river, had slain his guards, and had roused their comrades who had remained behind to similar boldness.
Meanwhile our men, since the raging waters prevented bridges from being made, and everything edible had been used up, passed two days in wretchedness, deprived of everything useful; excited by hunger and wrath, they were in a state of frenzy and eager to lose their lives by the sword rather than by starvation, the most shameful kind of death.

Zosimus, New History, III, 30, 4–5:

There fell also on the Roman side, three tribunes, Iulianus, Maximianus, and Macrobius. They then marched forward four days, continually harassed by the enemy, who followed them when they were proceeding, but fled when the Romans offered any resistance. At length, having gained some distance of the enemy, they resolved to cross the Tigris.
For this purpose, they fastened skins together, and floated over. When the greater part had gained the opposite bank, the commanders crossed over in safety with the remainder. The Persians, however, still accompanied them, and followed them with a large army so assiduously, that the Romans were in perpetual danger, both from the unfavorable circumstances in which they were placed, and from the want, of provisions.

Ammianus embellishes his tale as much as possible by praising the bravery of the Roman army and the cowardice of the “Persians”, but the truth is that the Roman army was trapped without supplies on the eastern bank of the Tigris and virtually surrounded. a small vanguard of “Gauls and Germans” who had experience in commando operations and river crossings (like Julian’s campaign in Gaul had showed) had managed to seize a bridgehead on the western bank, but the rest of the army could not cross due to the strong current. In this hopeless situation, Jovian and the top officers of the army accepted the peace terms offered by Šābuhr II, who then allowed the Roman army to retreat unhindered. The sources tell it thus:


Ammianus Marcellinus, Res Gestae Libri XXXI, XXV, 7, 5–14:

However, the eternal power of God in heaven was on our side, and the Persians, beyond our hopes, took the first step and sent as envoys for securing peace the Surena and another magistrate, being themselves also low in their minds, which the fact that the Roman side was superior in almost every battle shook more and more every day. Nevertheless, they offered conditions which were difficult and involved, for they pretended that from feelings of humanity the most merciful of kings would allow the remnants of the army to return, if the emperor and his most distinguished generals would comply with his demands. In reply to this, Arintheus was sent to him with the prefect Salutius, but, while a deliberate discussion was going on as to what ought to be determined, four days passed by, full of torments from hunger and worse than any death. If the emperor, before letting these envoys go, had used this space of time to withdraw gradually from the enemy's territories, he could surely have reached the protection of Corduena, a rich region belonging to us, and distant only a hundred miles from the spot where all this took place. Now, the king obstinately demanded the lands which (he said) were his and had been taken from him long ago by Maximianus; but, in fact, as the negotiations showed, he required as our ransom five provinces on the far side of the Tigris: Arzanena, Moxoëna, and Zabdicena, as well as Rehimena and Corduena with fifteen fortresses, besides Nisibis, Singara and Castra Maurorum, a very important stronghold. And whereas it would have been better to fight ten battles than give up any one of these, the band of flatterers pressed upon the timid emperor, harping upon the dreaded name of Procopius, and declaring that if he returned on learning of the death of Julianus, he would with the fresh troops under his command easily and without opposition make himself emperor. Jovian, inflamed by these dangerous hints too continually repeated, without delay surrendered all that was asked, except that with difficulty he succeeded in bringing it about that Nisibis and Singara should pass into control of the Persians without their inhabitants, and that the Romans in the fortresses that were to be taken from us should be allowed to return to our protection. To these conditions there was added another which was destructive and impious, namely, that after the completion of these agreements, Arsaces, our steadfast and faithful friend should never, if he asked it, be given help against the Persians. This was contrived with a double purpose, that a man who at the emperor's order had devastated Chiliocomum might be punished, and that the opportunity might be left of presently invading Armenia without opposition. The result was that later this same Arsaces was taken alive, and that the Parthians amid various dissensions and disturbances seized a great tract of Armenia bordering on Media, along with Artaxata.
When this treaty was concluded, lest anything contrary to the agreements should be done during the truce, distinguished men were given on both sides as hostages: from our side Nemota, Victor, and Bellovaedius, tribunes of famous corps, and from the opposite party Bineses, one of the distinguished magnates, and three satraps besides of no obscure name. And so, a peace of thirty years was made and consecrated by the sanctity of oaths; but we returned by other routes, and since the places near the river were avoided as rough and uneven, we suffered from lack of water and food.

Zosimus, New History, III, 31, 1–2:

Although the Roman army was in this condition, the Persians were willing to treat for peace, and for that purpose sent Surena with other officers to the Roman camp. Jovian, upon hearing this, sent to them Sallustius, prefect of the court, together with Aristaeus, who, after some discussion, agreed on a truce for thirty years. The conditions were, that the Romans should give up to the Persians the country of the Zabdiceni, and that of the Gordyeni, Rhemeni, and Zaleni, besides fifteen castles in those provinces, with the inhabitants, lands, cattle, and all their property; that Nisibis should be surrendered without its inhabitants, who were to be transplanted into whatever colony the Romans pleased.
The Persians also deprived the Romans of great part of Armenia, leaving them but a very small part of it. The truce having been concluded on these conditions, and ratified on both sides, the Romans had an opportunity of returning home unmolested, neither party offering or sustaining any injury, either by open force, or secret machination.

Chronicle of the Pseudo-Joshua the Stylite, 7:

In the year 609 (297–298 CE) the Greeks got possession of the city of Nisibis, and it remained under their sway for sixty-five years. After the death of Julian in Persia which took place in the year 674 (362–363 CE), Jovian, who reigned over the Greeks after him, preferred peace above everything, and for the sake of this he allowed the Persians to take possession of Nisibis for one hundred and twenty years, after which they were to restore it to its (former) masters.

John Malalas, Chronographia, XIII:

Jovian himself and his army left the desert for the fertile land of Persia and he pondered anxiously how he might leave Persia. When Sabbourarsakios (i.e. Šābuhr II), the Persian king, learnt of the death of the emperor Julian, he was distressed by a great anxiety. From the country of Persarmenia he dispatched as envoy one of his highest nobility called Suren to the Roman emperor, with a request and plea for peace. The holy Emperor Jovian gladly received him and consented to receive his embassy for peace, stating that he was also sending an ambassador to the Persian king. When Suren, the Persian ambassador, heard this, he asked the Emperor Jovian to sign a peace treaty immediately and forthwith. And Jovian gave the appointment to a senator of his, the patrician Arintheus, and entrusted the entire negotiations to him. He agreed to stick to what was sanctioned by him or was signed. The emperor disdained in person to make a peace treaty with a man of senatorial rank, albeit a Persian envoy. A relaxation of hostilities for three days was granted for the deliberations about peace. An agreement was struck between the Roman patrician Arintheus and Suren, the Persian senator and ambassador; the Romans would give to the Persians all the so-called province of Mygdonia (i.e. Roman Mesopotamia) and its chief city called Nisibis, empty and with only its walls, without its inhabitants. And when this was settled and a peace was put into writing, the Emperor Jovian took with him one of the satraps, a Persian in the company of the ambassador and called Junius. The purpose of this was to safeguard him and his expedition out of Persian territory and (for the Persians) to take over the province and its main city.

John Zonaras, Extracts of History, XIII, 14, 4, 6:

He (i.e. Jovian) accepted the title of emperor and made a treaty with the Persians which was hardly honorable, but necessary at the time. He conceded two famous cities to them, Nisibis and Singara, and transferred the inhabitants of them elsewhere, who, stressed by the violence of grief, spoke to him in terms far removed from the respect which they owed him. He abandoned to them (i.e. the Sasanians) some provinces and rights which had belonged to the Romans for a long time. When the hostages had been handed over from one side to the other, the Romans left to return to their country; but they suffered great discomfort throughout their whole journey, and were extremely hard pressed by hunger and thirst.

In this peace, Šābuhr II got practically all that he wanted:

• The satrapies “beyond the Tigris”.
• Most of Roman Mesopotamia, including its capital, Nisibis.
• A compromise by the Romans to not help the Armenian king against him.

Apart from Ammianus’ account, the other most interesting one is the one by John Malalas, because he basically copied the lost chronicle of Magnus of Carrhae, who like Ammianus had been an officer in the Roman army during the ill-fated expedition. For example, he informs us that the Sasanian ambassador who negotiated with Jovian and his staff was Surēn, one of the greatest among the wuzurgān and who had commanded the cavalry force that had shadowed Julian’s advance to Ctesiphon. Ammianus (and other authors) considered the peace to have been dishonorable and that Jovian had covered the name of Rome with shame, and later in his account he expressly blamed Jovian for not breaking the peace treaty as soon as he’s reached the nearest Roman fortress. He expressly accused Jovian of caring more about securing his throne than about keeping the Roman borders intact; he probably was right in this (after all, all the Roman augusti had acted in the same way) because Julian’s cousin and presumptive heir Procopius was in the area commanding another sizeable army; most probably Jovian’s first worry would have been to reach Constantinople before Procopius in order to secure his throne.

Map showing Julian’s invasion of Ērānšahr and the retreat begun by him and ended by Jovian. The borders in this map are the ones after the Second Peace of Nisibis.

Scholars are not sure why Šābuhr II agreed to leave the westernmost part of northern Mesopotamia in Roman hands, but if he did so willingly, he showed political vision, because this border would last until the last Roman-Sasanian war in the VII century CE. With Nisibis, the Sasanian king managed to gain a bulwark against Roman invasions, that would have it much more difficult now to reach Ctesiphon or to menace the Zagros passes, and at the same time he allowed the Romans to retain a screen of fortified cities east of the Euphrates (Carrhae, Edessa, and even Amida) that offered a minimum of protection for Syria, Cilicia and Cappadocia against a Sasanian invasion.

Due to the symbolism of Nisibis, this treaty is known as the Second Treaty of Nisibis (or the Second Peace of Nisibis), even if (as we have seen) it was signed nowhere near this city. After the loyalty that the population of this city had showed to the Roman cause, Šābuhr II had enforced in the treaty that the city would be delivered to him empty of its inhabitants, who were offered three days to leave the city with whatever material possessions they could carry; most of them would be resettled by the Roman authorities in Edessa (the famed theological school of Nisibis would thus become the school of Edessa for some time), and according to Tabarī, the Šahān Šāh settled in this city 12,000 Persians “of good families” from Staķr in Pārs (the cradle of the House of Sāsān).

A more detailed map showing the new trace of the Roman-Sasanian border in upper Mesopotamia (the Armenian border to the north of the Tur Abdin mountains is not correct and would not be in place until the 430s CE).

To Ammianus, the most shameful parts of this peace were the treason to the citizens of Nisibis and the abandonment of Rome’s loyal Armenian allies to the revenge of Šābuhr II. This king, at the age of 54, had reached the goal he’d been pursuing since the beginning of his personal reign and had restored the western borders of Ērānšahr to what they had been before Narsē’s shameful defeat against Galerius. This was to be the zenith of Šābuhr II’s reign, for until his death in 379 CE his rule would know more failures than successes. This peace would also signal the beginning of a long period of more or less peaceful coexistence between the Roman and Sasanian empire in the Middle East. There would still be some minor wars in the following decades until the 430s CE, when both empires reached a definitive settlement for the “Armenian question” and from then until 502 CE (the start of the “Anastasian War” both empires would be fully at peace.

The main reason for this sudden penchant for pacifism was not to be found in political or ideological changes in either of these empires, but the appearance of new and very dangerous enemies in each case. In 376 CE, the Goths and other “barbarian” peoples crossed the lower Danube seeking refuge within the Roman empire against the onslaught of the “European” Huns. And in Central Asia, the Sasanians would have to cope with the raise of new powerful dynasties of nomadic origin (probably Hunnic too in this case); the Kidarites and Hephthalites, who would pose a worse danger for Ērānšahr than any of the Roman emperors to date. The remainder of the IV century CE and all of the V century CE would be a time of survival struggle for the Sasanians against these new and dangerous enemies from the East.

This is perhaps the most famous portrait of Šābuhr II that has arrived to our days; sculpture made in silver and preserved at the Met in New York.

 

[Avernite]

So the Romans were hopelessly stranded but got a mostly decent loss rather than a total disaster. I wonder, might that imply the Sassanian army had suffered rather seriously, so it needed Rome out of the way, rather than risk any extra losses destroying them?

 

[LizarDAMN]

Well, if they destroy the field army, they would still need to siege the towns they wanted and that takes time and money.

 

[Semper Victor]

So the Romans were hopelessly stranded but got a mostly decent loss rather than a total disaster. I wonder, might that imply the Sassanian army had suffered rather seriously, so it needed Rome out of the way, rather than risk any extra losses destroying them?

That is exactly the reason offfered by Ammianus, together with the menace of the (supposedly nearby) intact Roman army led by Procopius and Sebastianus. Syvänne offers also another reason for Shapur II's willingness to negotiate: that the Armenian army had broken into Media and was looting this rich and populous region of his empire, and he could not ignore this indefinitely.

He also criticized Jovian and the top brass for the time wasted in negotiations. According to him (and Syvänne thinks that he was correct) the Roman army was only 100 miles south of Corduene, a friendly region (one of the Transtigritan satrapies), and the Roman army could have covered that distance in the fourt days that were spent in negotiations (25 miles per day). To Syvänne, it's quite obvious (I agree with him, and Ammianus seems to do the same, although he doesn't say so explicitly) that the Sasanian side lengthened the negotiations on purpose, to let the Roman army stew a bit more.

But Syvänne presents a series of reasons that are obviated by Ammianus ansd which would have weighted heavily in the mind of Jovian and his generals and officials:

• The Roman army was in a state close to mutiny due to the lack of food; by this time Jovian and his officers had barely prevented a disaster when the soldiers had pressed them to just cross the swollen waters of the Tigris by whatever means they had. Even crossing the Tigris would have solved nothing; the nearest Roman fortifications (Singara and Bezabde) were either destroyed or in Sasanian hands, and the northern Mesopotamian plain in the area between the crossing point and Nisibis is either desert of semi desertic steppe, which would have worsened the sufferings of the Roman army and offered flat, open and hard terrain perfect for the Sasanian cavalry.
• According to Ammianus and Zosimus, at least part of the cavalry had lost their horses; the beasts were so weakened by hunger that the cavalrymen had been forced to dismount and walk besides them. These horses were so tired and weakened that thay actually lagged behind the Roman infantry and were targeted by the harassing Arab and Iranian cavalry.
• Covering 100 miles in four days would have been perfectly possible for a Roman army marching at top speed and not being harassed and with plenty of supplies, and without weakened men and horses. Not only was the army not in such a pristine condition, but it was quite obvious that Shapur II was not going to offer them that luxury, as their previous advance northwards had been much slower than that under the constant harassment and attacks of the enemy.

In what I sort of agree with Ammianus is that as soon as the Roman army reached Nisibis, Jovian should have rejected the treaty (sacrificing the hostages given to Shapur II, though) and that in this case Jovian decided to honor the treaty out of self interest in order to secure his position against Procopius (Julian's cousin).

 

[Eusebio]

What do you think would have happened if Julian hadn't got himself killed? Would Shapur have negotiated a similar treaty if he was still leading the army? And if so, and Julian repudiates it as soon as he gets out of Mesopotamia, how does his political situation look? Would he be able to claim the campaign as a kind of draw in that scenario?

 

[Semper Victor]

What do you think would have happened if Julian hadn't got himself killed? Would Shapur have negotiated a similar treaty if he was still leading the army? And if so, and Julian repudiates it as soon as he gets out of Mesopotamia, how does his political situation look? Would he be able to claim the campaign as a kind of draw in that scenario?

I think that there was little hope for Julian to come out of the campaign alive. The campaign was his pet project; he had decided he was going to defeat the Sasanian empire the same way he'd defeated a Frankish or Alamannic tribe, and that in itself was a fatal mistake. In other words, he'd gambled all his prestige in a campaign that had little hopes of ending in a victory. As we've seen, there was discontent among the troops already at the start of the campaign, and Julian had to distribute money to them when crossing the Khabur river and later at Pirisabora. And we should add to this the strange string of "accidents" that befell him and the army during the march from Antioch to Circesium. If Julian had lived past the battle of Toummara / Sumere, I'm pretty sure he would've been eventually killed either by mutinied troops or by a conspiration of his generals. As it happened, his death was suspicious enough in real life (Syvänne thinks he was most probably murdered by a member of his guard).

 

[Acheron]

I think that there was little hope for Julian to come out of the campaign alive. The campaign was his pet project; he had decided he was going to defeat the Sasanian empire the same way he'd defeated a Frankish or Alamannic tribe, and that in itself was a fatal mistake.

If you say "the same he'd defeated a Frankish or Allamannic tribe", do you mean that Julian was ignorant of the size, strength and/or organization of the Sassanids or that he conducted the campaign in a way similar to his earlier ones, with no or too little regard to the new environment?

 

[Semper Victor]

If you say "the same he'd defeated a Frankish or Allamannic tribe", do you mean that Julian was ignorant of the size, strength and/or organization of the Sassanids or that he conducted the campaign in a way similar to his earlier ones, with no or too little regard to the new environment?

A bit of the former, and definitely 100% of the latter. Attacking directly Ctesiphon without defeating decisively the Sasanian field army first (like the Arabs did at al-Qadissiyah) would lead inevitably to this result. It's the same problem that Shapur II encountered time and again in his direct attacks against Nisibis, until he finally changed his strategy. But Julian lacked Shapur II's long experience, and paid the price for it. Ammianus says nothing about it, but it's probable that he also decided to dismiss the advice of the generals of Constantius II's army who had experience in this theater (and who were mostly Christian).

And there's worse. Ammianus was a pagan and an admirer of Julian. So, when he criticizes Julian for believing too much in oracles and divinations (he traveled with all his "philosopher" friends like Maximus of Ephesus), we should take the criticism seriously (other critics like Gregory of Nazianzus and John Chrisostomus said the same, but as they were Christian bishops, their opinions are often dismissed as sectarian). There seems to have been quite a bit (or even a lot) of messianism (in the pagan version) in Julian's enterprise.

 

[Acheron]

A bit of the former, and definitely 100% of the latter.

Though in what was differed the environments? Since the army marched along or between rivers, I would assume that water was not a critical issue, compared to Germania? As for food, I would assume that the sparsely populated tribal areas also didn't offer much in the way of foraging.

 

[demanvanwezel]

Though in what was differed the environments? Since the army marched along or between rivers, I would assume that water was not a critical issue, compared to Germania? As for food, I would assume that the sparsely populated tribal areas also didn't offer much in the way of foraging.

I think the issue with food is that you couldn't establish supply depots since the sasanids could easily move in from persia proper

with hindsight I think the only way you could ever properly invade ctesiphon is by either smashing the sassanid army beforehand or have 3 armies each of them large enough to defeat the sassanid army by itself and move 2 of them on ctesiphon on either side of the river while the 3th guards the flank with persia

 

[Semper Victor]

Though in what was differed the environments? Since the army marched along or between rivers, I would assume that water was not a critical issue, compared to Germania? As for food, I would assume that the sparsely populated tribal areas also didn't offer much in the way of foraging.

• Longer distances. The effect of the climate on the troops he'd brought from Europe was also noticeable (they were not used to the heat of Mesopotamis in summer).
• Heavily fortified cities. The army did well against smaller towns / cities but did not even try against Veh-Ardashir or Ctesiphon.
• The impossibility of achieving strategic surprise (not when you mobilize almost 100,000 men and have to gather huge amounts of supplies for them).
• An enemy that is mostly cavalry and that can and will force asymmetrical warfare upon you.
• Dealing with a large army that's far less cohesive than the small forces you've commanded in Gaul, after having poisoned the air with your behavior the last year in Antioch (see Ammianus' criticism about that). It applies both to the soldiers and the officers.
• Not knowing how to deal with allies or with secondary forces. Julian's mismanagement of Procopius and Sebastianus' army and especially of the Arab foederati cost him dearly.
• Mismanagement of supplies (very clear when the Roman army ran out of them in the retreat).

 

[Semper Victor]

I think the issue with food is that you couldn't establish supply depots since the sasanids could easily move in from persia proper

with hindsight I think the only way you could ever properly invade ctesiphon is by either smashing the sassanid army beforehand or have 3 armies each of them large enough to defeat the sassanid army by itself and move 2 of them on ctesiphon on either side of the river while the 3th guards the flank with persia

Exactly. Although the second alternative would have meant to mobilize far more men than the total sum of troops the Romans had deployed along all the borders of the empire.

 

[Acheron]

I think the issue with food is that you couldn't establish supply depots since the sasanids could easily move in from persia proper

with hindsight I think the only way you could ever properly invade ctesiphon is by either smashing the sassanid army beforehand or have 3 armies each of them large enough to defeat the sassanid army by itself and move 2 of them on ctesiphon on either side of the river while the 3th guards the flank with persia

What about what the Sassanids did beforehand against Constantinus II, taking over the border fortresses first and then going for the city? or did that only work due to the Roman civil war?

 

[Semper Victor]

What about what the Sassanids did beforehand against Constantinus II, taking over the border fortresses first and then going for the city? or did that only work due to the Roman civil war?

That would have been a viable strategy too, as long as there were no other emergencies in other borders. Julian had it quite good in this respect, as he and Constantius II had cowed the Franks, Alamanni, Quadi and Sarmatians into peace, and the Goths still respected the peace they'd made with Constantine I the Great. But he decided otherwise. First he would¡ve needed to recover and fortify again Singara and Bezade, and then choose one of the two invasion routes: either the Euphrates or the Tigris, taking all the fortresses in his way and building up a viable supply line behind him. The Euphrates would have been more suitable to this task.

 

[demanvanwezel]

Exactly. Although the second alternative would have meant to mobilize far more men than the total sum of troops the Romans had deployed along all the borders of the empire.

well my analysis is mostly based on 2 roman generals to take ctesiphon, eudinatus took option A and trajan took option B and seeing as how the sassanids only ever came out when the romans were very weak I can see why the romans often attempted to imitate option B

what I want to know is how shapur managed to survive 2 disastrous campaigns to have a succesfull one while julian got stabbed in one

 

[Avernite]

well my analysis is mostly based on 2 roman generals to take ctesiphon, eudinatus took option A and trajan took option B and seeing as how the sassanids only ever came out when the romans were very weak I can see why the romans often attempted to imitate option B

what I want to know is how shapur managed to survive 2 disastrous campaigns to have a succesfull one while julian got stabbed in one

I think the key is dynastic politics; Persia had them, Rome didn't. So Julian's list of competitors was very long, while any Persian conspiracy needed to enlist one of a very short short-list to be sure of success.

And of course, as the probable Persian successor would be a close family member of Shapur, there might come a time when said successor was morally obligated to punish the murderers of his kin (even while secretly thanking them for their service).

 

[Semper Victor]

I think the key is dynastic politics; Persia had them, Rome didn't. So Julian's list of competitors was very long, while any Persian conspiracy needed to enlist one of a very short short-list to be sure of success.

And of course, as the probable Persian successor would be a close family member of Shapur, there might come a time when said successor was morally obligated to punish the murderers of his kin (even while secretly thanking them for their service).

I share most of your opinion, and in following chapters I'll try to delve a little more into the royal ideology of the Sasanian empire and its basis on Zoroastrianism, and I will also compare it to Hunnish practice. The dynastic policies of the Arsacid and Sasanians were a legacy of the steppe (the Arsacid family was part of the Parni tribe, itself a part of the Dahae confederation of Scythian peoples), and it was a system very different to the ones that existed in the Middle East and the Mediterranean Basin in Antiquity. Some scholars believe that it was Hunnic influence that spread it among the Germanic peoples that succeeded the Western Roman Empire, and so it became widesoread in Europe during the Middle Ages.

 

[Semper Victor]

well my analysis is mostly based on 2 roman generals to take ctesiphon, eudinatus took option A and trajan took option B and seeing as how the sassanids only ever came out when the romans were very weak I can see why the romans often attempted to imitate option

Severus Alexander tried a similar strategy for his campaign against Ardashir I in 233-234 CE, and things also ended badly for the Romans; their armies coordinated badly and Ardashir I concentrated all his forces against the Euphrates army and massacred it.

what I want to know is how shapur managed to survive 2 disastrous campaigns to have a succesfull one while julian got stabbed in one

That's a good question indeed, but one that must be answered looking at the social and political structures of both empires (see my previous answer to Avernite).

 

[Avernite]

I share most of your opinion, and in following chapters I'll try to delve a little more into the royal ideology of the Sasanian empire and its basis on Zoroastrianism, and I will also compare it to Hunnish practice. The dynastic policies of the Arsacid and Sasanians were a legacy of the steppe (the Arsacid family was part of the Parni tribe, itself a part of the Dahae confederation of Scythian peoples), and it was a system very different to the ones that existed in the Middle East and the Mediterranean Basin in Antiquity. Some scholars believe that it was Hunnic influence that spread it among the Germanic peoples that succeeded the Western Roman Empire, and so it became widesoread in Europe during the Middle Ages.

I haven't heard that theory before - I always linked Germanic dynasticism more to its equivalent in, for example, Macedonia or Ptolemaic Egypt or Achaemenid Persia - i.e. the norm everywhere, and non-dynastic Classical Antiquity was the outlier.