History Podcasts

Ancient Warfare Vol VII, Issue 2: Struggle for control: Wars in ancient Sicily

Ancient Warfare Vol VII, Issue 2: Struggle for control: Wars in ancient Sicily

Ancient Warfare Vol VII, Issue 2: Struggle for control: Wars in ancient Sicily

Ancient Warfare Vol VII, Issue 2: Struggle for control: Wars in ancient Sicily

The main theme here is the long series of wars in ancient Sicily. This was at least a five-sided conflict, involving the native Sicilians, Ionian Greeks, Dorian Greeks, Phoenicians and Carthaginians, and eventually the Romans, who ended the series of ancient wars. Sicily's position in the centre of the Mediterranean was especially important in the Ancient World as control of its coasts gave one control of major shipping lanes.

The theme begins with a useful introduction, with an overview of the various players on Sicily and the main wars, supported by a nice map showing the main ancient cities on the island. Next comes a look at the Greek fortifications of the island, with material on the type of construction used and some examples taken from the surviving remains. The disastrous Athenian attack on Syracuse is covered with a look at the performance of the Athenian cavalry during this campaign. The wars against Carthage are represented by articles on Timoleon of Corinth, a general who came to the island to help during a crisis, Dionysius's siege of Motya and the role of Agathocles in the conflict. Finally the role of mercenaries and specialists in ancient warfare is examined in an article on Cretan mercenary archers.

Away from the main theme there is an interesting article on Roman ownership inscriptions - essentially name tags carved onto metal gear (at least that is what survives), that often contain useful snippets of information about the organisation of the Roman army. The cost of service in the Legions is examined in an article that looks at the reasons why some men cut their own thumbs off to avoid service. The final article, on Alexander's victory at the battle of the Granicus, provides an interesting take on this cavalry battle, taking into account the nature of the terrain and of the Persian cavalry to produce a coherent version of the battle.

Struggle for control: Historical introduction
Sicilian Greek fortifications: Military architecture as source
Timoleon of Corinth: Saviour of Sicily
An underestimation of the enemy's cavalry: Athenian cavalry in Sicily
In the service of Syracusan tyrants: A regiment of Cretan mercenary archers
Dionysius I's Siege of Motya, 379 BC: Sow the wind and reap the whirlwind
Treachery, tyrant and terror: Agathocles of Syracuse and the Third Greco-Punic War
'Keep your grubby paws off my stuff!' - Roman ownership inscriptions
'I would rather cut off my thumb' - Refusal of military service in ancient Rome
Alexander's great cavalry battle - What really happened at the River Granicus



Institutional Login
Log in to Wiley Online Library

If you have previously obtained access with your personal account, please log in.

Purchase Instant Access
  • View the article PDF and any associated supplements and figures for a period of 48 hours.
  • Article can not be printed.
  • Article can not be downloaded.
  • Article can not be redistributed.
  • Unlimited viewing of the article PDF and any associated supplements and figures.
  • Article can not be printed.
  • Article can not be downloaded.
  • Article can not be redistributed.
  • Unlimited viewing of the article/chapter PDF and any associated supplements and figures.
  • Article/chapter can be printed.
  • Article/chapter can be downloaded.
  • Article/chapter can not be redistributed.

Abstract

Siege warfare in antiquity was bloodier than other forms of combat, usually involving urban centers rather than purely military installations. Though sieges, in contrast to open battles, required complicated logistics and employed high technology, there was remarkably little development of fundamental designs. The two significant advances were the invention of torsion artillery around 400 bc and the introduction of traction artillery in the sixth century ad . The besieger could attempt one of three ways to get into a town: under the walls (by mining), over the walls (by scaling), or through the walls (by rams, artillery, or subterfuge). A complex variety of motives encouraged men to run the enormous risks involved in being first through the breach or over the wall once in the town itself, they faced the bloody prospect of house-to-house combat. Following a successful assault, the mental and physical strains imposed on the besiegers often led to savage reprisals. Other sieges, however, ended not with storm and sack but with the capitulation of the starving defenders. Sieges upset societal norms exceptionally in the ancient world, women might be found taking an active part in combat. As urban centers were often religious centers, it is unsurprising that the gods often feature prominently in literary accounts. High literature not only responded to the excitement and heroism of sieges but also could actively shape the ways in which sieges were prosecuted.


Ancient Warfare magazine Vol IX.4 - Clash of the Colossi

The First Punic War (264 to 241 BC ) was the longest uninterrupted war in antiquity and the beginning of a series of military conflicts between Carthage and Rome. During the struggle, these ancient powers fought for the control of Sicily, a strategic point in the central Mediterranean. In the end, Rome was victorious and Carthage lost Sicily.

The source: Tilman Moritz, "The fragments of Fabius Pictor - Puzzling history"

Allegedly it all started with Polybius. Striving for a balanced view of the wars that had given rise to Roman supremacy, the Greek historian was - by his own account - the first to compile and evaluate the contradictory evidence. He drew on sources ranging from literary and documentary texts to monuments as well as oral tradition and, not the least, personal experience as an eyewitness to the latest events.Thus it was Polybius who, following in the footsteps of Thucydides and Herodotus, raised Roman historiography to universal Hellenistic standards - or perhaps not?

Theme: Christa Steinby, "The escalation of war at sea - Ship builders wanted"

The First Punic War (264–241 BC ) is best known for the big fleets and sea battles, which first took place at Mylae in 260, and continued until the Battle of the Aegates islands in 241. The conflict escalated gradually, requiring more men and ships and also the introduction of a completely new line of ship in the Roman navy, the quinquereme. The purpose of this article is to shed light on the Roman operations in Sicily in 264–260, the gradual increase of their ambitions at sea, and the process that led to the building of the first big fleet.

The reenactor: Jean-Luc Féraud, "A Carthaginian war elephant - Charge!"

War elephants were made famous when Hannibal used them during the invasion of Italy in the Second Punic War. However, these magnificent animals were used in far larger numbers by the Carthaginians during the First Punic War.

Theme: Seán Hußmann, "Elephants in war - Behemoths of the battlefield"

Throughout history, man has waged war on his fellow man. And throughout history, he made use of animals to reach his military goals. Be it horses, oxen, dogs, mules, even cats – there are few domesticated animals that were not in some way or other used for warfare. Perhaps the most magnificent beast that was employed in a military function was the elephant. Furthermore, this mighty animal is the only one that was not only used in a logistical role, but as an active combatant.

Theme: Sidney E. Dean, "Agrigentum, 262–261 BC - Besiegers besieged"

The siege and subsequent battle of Agrigentum on Sicily was the first major land engagement of the First Punic War. For both sides it was an educational experience, providing insight into the military capabilities and tactics of the opposing side. Rome ultimately won the confrontation at Agrigentum, but it was, in many ways, a Pyrrhic victory.

The find: Michael Taylor, "Archaeology from the Battle of the Aegates Islands - From the bottom of the sea floor"

In 241 BC , the First Punic War was stalemated, with the Romans maintaining a desultory siege of Punic forces near Mount Eryx. The Carthaginians, believing they enjoyed naval superiority after their smashing victory at Drepanna seven years before, did not maintain a major naval presence, even as they supplied their forces in Sicily by sea. When the Romans surprised the Carthaginians by deploying a new fleet, the latter hastily dispatched a naval relief force. As the Punic fleet prepared to sail on its last leg to the Sicilian coast, Roman warships, lying in wait behind the Aegates (now Egadi) Islands, pounced. After a sharp naval engagement, the Romans won a decisive victory, effectively ending the war.

Theme: Robert Holmes, "Battle speeches of the First Punic War - Taciturn Romans, loquacious Carthaginians"

Although the concept of the battle speech as a distinct historical genre can be traced back to Homer and Herodotus, it was Thucydides who set forth the conventions future historians would seek to emulate. According to these conventions, the battle speech had to both keep to what was said and what was probably said given the situation they also drew heavily on such themes as the nobility of giving one’s life for their country, earning the favour of the gods, and crossroads where one must vanquish or die. The Thucydidean battle speech conventions were later employed to good effect by Polybius in his account of the First Punic War.

Theme: Joseph Hall, "The Battle of Cape Ecnomus - War on the waves"

In 256 BC , Rome’s strategic plan for how the war was to be won involved a direct landing in Africa itself. To this end, scores of legionaries embarked upon ships laid on for the voyage. When the enormous fleet finally set sail each ship was crowded with men and materiel for the coming campaign, and included both consuls themselves. En-route to Africa however, the fleet encountered the vast navy of Carthage near Cape Ecnomus, south of Sicily. With over a quarter of a million men reputedly aboard close to seven hundred ships, the titanic engagement that followed has gone down in history as the largest naval battle ever fought.

Theme: Arnold Blumberg, "The Battle of Tunis, 255 BC - Rome’s disastrous North-African interlude"

The war between Rome and Carthage for the mastery of the western Mediterranean had arrived at a tipping point by the year 255 BC . Since 264 BC, the conflict on land between the two republics had revolved around control of the island of Sicily, the wealthiest and most strategically placed spot in the entire region due to its geographical position in the centre of the Mediterranean Sea, and the fact that it served as a bridge between Africa and Europe. A plan by the Romans to switch the centre of gravity of the ground war from Sicily to Africa resulted in an unanticipated Roman battlefield defeat and added years of fighting to the First Punic War.

The find: Josho Brouwers, "A gilded bronze cuirass from a Punic grave - The beautiful body"

Earlier in 2015, the National Museum of Antiquities in Leiden, the Netherlands, organized an exhibition on Carthage. A large number of beautiful artefacts were transported from Tunisia to be exhibited in the museum, including a beautiful example of a cuirass that not only highlights the splendour of the Carthaginian war machine, but also its cosmopolitan nature.

Fiction: Marcus Pailing, "We left our dead at the Polytimetos"

We watched the barbarians from the walls of Marakanda. They preened and postured, making their sturdy mounts prance and wheel as if in a dance- it was like a savage mating ritual, even. Yet this was no mere entertainment. Instead it was a challenge, and we all knew it.

Special: Wassilis Stephan Linidis, "Recreating an extra-heavy Archaic Greek hoplite - Like a bronze statue of Ares"

When we think of Greek hoplites, most think about the typical Classical hoplite with a Corinthian helmet, a cuirass, greaves, and shield. While the hoplite dominated Greek infantry warfare, this type of soldier underwent significant equipment changes over time. However, most reenactors choose the Persian Wars period for their impressions, with only a handful going for a later Classical kit and even fewer exploring earlier phases of hoplite warfare. But exploring these earlier stages in particular offers a fascinating journey.

The debate: Owen Rees and Jason Crowley, "Was there mental trauma in ancient warfare - PTSD in ancient Greece"

For over fifteen years it has been an accepted historical practice to refer to episodes within ancient Greek history, and portrayals of characters in Greek drama, as exhibiting behaviours akin to modern veterans suffering with Post-traumatic Stress Disorder (PTSD ). Indeed, it is becoming more and more common that figures from ancient history are being retrospectively diagnosed with PTSD. Yet, to date no book length treatment exists to debate the validity of this implied universalism: that PTSD can be found and equated in history, with little regard for the social and chronological boundaries that separate the present from the past.

Hollywood Romans: Graham Sumner, "The Roman army on screen, part 2 - Quo Vadis (1951)"Caesar and Cleopatra (1945) had been a vain attempt by the British film industry to take on Hollywood. However, in the next decade Hollywood itself was indeed under threat, not from Britain but from the rapid rise in popularity of television. Hollywood believed that big budget blockbuster films set in ancient Rome made with the very latest film and sound technology, shown on extra large screens, were one answer to win back audiences. Consequently, the 1950s ushered in a golden age of epic movies.


Ancient Warfare Vol VII, Issue 2: Struggle for control: Wars in ancient Sicily - History

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to upgrade your browser.

Supervisors: Andrew B. Gallia

104-100) and the Cilician campaign of Marcus Antonius Orator (102 onward. more The Second Servile War (

104-100) and the Cilician campaign of Marcus Antonius Orator (102 onwards) are rarely thought to have anything in common. In this paper, I suggest they were actually tightly connected—specifically, that there was some link, real or imagined, between the slaves in Sicily and the Cilicians in Asia Minor that caused the Romans to suspect that the slave revolt had professional backing from a foreign power (the freelance mercenaries/pirates of Cilicia) and furthermore, that they believed in order to crush the slaves, the Cilicians had to be dealt with as well. Thus the eastern command of Marcus Antonius Orator was believed to be (or promoted as) a useful complement to the ongoing Sicilian affair.
Discussions of the campaign of Antonius (e.g. Freeman 1986, Keaveney 1982, de Souza 1999, 102-108) are decidedly limited, and the question of what prompted this action at this particular time is unanswered. Historically, the Romans had been content to ignore the southern coast of Anatolia from Apamea (189) onward and to allow Rhodes and Lycia a free hand, despite piracy in the area. At the time of Antonius gaining the Cilician command, Marius was fighting the Teutones in the north and the Second Servile War was in full swing in the south. With such unfavorable timing, I suggest this move on Cilicia was not, at the time, considered a separate affair but was somehow linked to Sicilian matters.
In the ancient accounts of the First and Second Servile Wars (aptly compiled in Shaw 2001), we can note the clear prominence of Syrian and Cilician slaves, both as combatants and as leaders. The self-declared kings Eunus and Salvius were Syrians and the generals Athenion and Cleon Cilicians. While this may simply be a love of symmetry on the part of Diodorus Siculus (our principal source), it is hardly an unlikely situation. In the second half of the second century, Rome had fought fewer large wars and needed alternate sources for slaves, while at the same time Seleucid Syria had been beset by numerous civil wars. It is reasonable to suppose that amidst the chaos in Cilicia and Syria, numerous slaves were taken and sold in the Aegean. The Syrian and Cilician slaves thus almost certainly had the most military experience among the slave population of Sicily. Furthermore, the slaves themselves emphasized links with the Hellenistic East. Eunus renamed himself Antiochus, while later, Salvius took on the name Tryphon, invoking Diodotus Tryphon, who, as Strabo (14.5.2) informs us, had organized the Cilician pirates to work together.
Despite the lack of evidence concerning the campaign of Antonius, (known through such sources as Livy, Ep. 68 Obsequens, Prodig. 44 and ILLRP 1.342), I argue that Antonius was given the command not as a separate campaign, but as a supplement to the Sicilian. This hypothesis is further supported by the so-called ‘pirate provisions’ in an unusual law: the lex de provinciis praetoriis (described by Crawford 1996 in Roman Statutes). This law (dated to 101/100) makes the consuls write letters to most of the Hellenistic rulers and insists that they not allow any pirates from Cilicia access to their lands. Based on Roman depictions of earlier pirates (albeit from historians writing later), it is reasonable to predict any and all forces still at large in Cilicia would have been considered pirates regardless of their initial status. I argue that the Romans are not concerned with piracy per se, but rather with sending a message. The fugitive Cilicians are regarded similarly to fugitive slaves. Nevertheless, the kingdoms of the east would readily have noted the punitive invasion of Cilicia. The LdPP, in talking about Cilicia, can only be fully understood in reference to Sicily. In justifying the Roman invasion of Cilicia, it also serves as a warning to not repeat the (supposed) Cilician mistake in Sicily.

This project examines the role that pirates, bandits and other ‘freelance men of violence’ played. more This project examines the role that pirates, bandits and other ‘freelance men of violence’ played in the second and first centuries BC. In this examination, I argue that the role of the ‘freelancer’ needs to be understood as a unified whole, not compartmentalized as three (or more) groups: pirates, bandits and mercenaries. Throughout the work, I contend that their portrayal by the ancient writers dramatically affects the perceived legitimacy of their actions. Most modern studies of banditry or piracy (e.g. de Souza 1999, Grünewald 1999, Pohl 1993, Knapp 2011) focus on the ‘real’ bandits and set aside the clear polemic. I turn to the description and ask what semantic baggage words like latro or leistes carried that they were commonly used as part of invectives. How was freelance plundering considered dishonorable while blatant triumph-hunting was deemed honorable?

The freelancer was, in a form of ancient Realpolitik, generally more acceptable to states than our occasionally-stuffy aristocratic historians would prefer that we believe. Moreover, states were far more concerned with control of these ‘freelancers’ than in their elimination. Overall, this dissertation shows that the chief semantic burden of terms like latro and leistes are to convey (il)legitimacy: individuals that possess power that they should not. Condemnation of these figures is not rooted in their actions of plundering (rarely dissimilar from official acts of war) but rather their holding the power to do so in the first place. In short, this study reveals that the ‘at-large’ soldier was a far more complex and influential character than is normally shown by either ancient or modern historians.

In Thursday's talk, I will focus on the depiction of Illyria and the Illyrians' failure to comply with a changing set of Roman rules. I argue that the wars in Illyria provide some of the earliest evidence that Roman attitudes regarding the normal conduct of war were changing, despite later historians' insistence that no such thing was happening.


Ancient Warfare Vol VII, Issue 2: Struggle for control: Wars in ancient Sicily - History

MA Hons. in History (Ancient/Mediaeval) and Classics (1st class) University of St. Andrews.
PhD University of St. Andrews, with a thesis on late Roman military literature (supervisor Michael Whitby).

PHILIP RANCE has taught ancient and medieval history and/or Greek language and literature at universities in the United Kingdom, Germany and Belgium.
He has been a visiting scholar at universities in Heidelberg (2005-6), Thessaloniki (2011-12) and Berlin (2016-19), and has been awarded senior research fellowships at
universities and research institutes in Munich (2009-11), Istanbul (2013-14), Berlin (2015), Wolfenbüttel (2016), Gotha/Erfurt (2017) and Sofia (2019-20).

His research interests include late Roman and Byzantine history, historiography and literature warfare and martial culture the manuscript transmission and reception of
Greco-Roman technical and scientific literature Byzantine codicology and book culture Vulgar and Late Latin and Byzantine Greek philology.

Some Recent Projects: https://fu-berlin.academia.edu/PhilipRance/RESEARCH-PROJECTS

(Proceedings of the International Conference on Greek Taktika held at the University of Toruń, 7-. more (Proceedings of the International Conference on Greek Taktika held at the University of Toruń, 7-11 April 2005)

Contributors: Wojciech Brillowski (Adam Mickiewicz University, Poznań) Bogdan Burliga (University of Gdańsk) Radosław A. Gawroński (Cardinal Stefan Wyszyński University, Warsaw) Pierre O. Juhel Burkhard Meißner (Helmut Schmidt-Universität / Universität der Bundeswehr Hamburg) Alexander Nefedkin (St. Petersburg State University) Philip Rance (Freie Universität Berlin) Keith Roberts Jacek Rzepka (University of Warsaw) Hans Michael Schellenberg (Heinrich Heine-Universität Düsseldorf) Nicholas Sekunda (University of Gdańsk) Sławomir Sprawski (Jagiellonian University, Kraków)

Manchester University Press: https://manchesteruniversitypress.co.uk/9781526138620/ FU-Berlin P. more Manchester University Press: https://manchesteruniversitypress.co.uk/9781526138620/

ABSTRACT: Although the medical services of the Roman army have long attracted scholarly interest. more ABSTRACT: Although the medical services of the Roman army have long attracted scholarly interest, all previous studies terminate in the mid/late third century, partly owing to conventional periodizations of Roman military history, but primarily in response to a drastic diminution in the epigraphic and archaeological record. This chapter assembles the evidence relating to health and medicine in the late Roman army (AD 250-600) and identifies interpretive challenges posed by different categories of literary and documentary source material. Where possible, analysis extends beyond medical personnel, facilities and procedures to examine broader medico-historical perspectives, including vulnerability to disease, cultural attitudes to soldiers’ health, especially combat wounds, and arrangements for invalided servicemen. Contrary to traditional notions of ‘decline and fall’, the evidence points to substantial continuity in institutional approaches to safeguarding soldiers’ health and treating those who became sick, injured or wounded, despite changes in the nature and expectations of military service in Late Antiquity.

(Proceedings of the International Conference on Greek Taktika held at the University of Toruń, 7-. more (Proceedings of the International Conference on Greek Taktika held at the University of Toruń, 7-11 April 2005)

Contributors: Wojciech Brillowski (Adam Mickiewicz University, Poznań) Bogdan Burliga (University of Gdańsk) Radosław A. Gawroński (Cardinal Stefan Wyszyński University, Warsaw) Pierre O. Juhel Burkhard Meißner (Helmut Schmidt-Universität / Universität der Bundeswehr Hamburg) Alexander Nefedkin (St. Petersburg State University) Philip Rance (Freie Universität Berlin) Keith Roberts Jacek Rzepka (University of Warsaw) Hans Michael Schellenberg (Heinrich Heine-Universität Düsseldorf) Nicholas Sekunda (University of Gdańsk) Sławomir Sprawski (Jagiellonian University, Kraków)

Manchester University Press: https://manchesteruniversitypress.co.uk/9781526138620/ FU-Berlin P. more Manchester University Press: https://manchesteruniversitypress.co.uk/9781526138620/

ABSTRACT: Although the medical services of the Roman army have long attracted scholarly interest. more ABSTRACT: Although the medical services of the Roman army have long attracted scholarly interest, all previous studies terminate in the mid/late third century, partly owing to conventional periodizations of Roman military history, but primarily in response to a drastic diminution in the epigraphic and archaeological record. This chapter assembles the evidence relating to health and medicine in the late Roman army (AD 250-600) and identifies interpretive challenges posed by different categories of literary and documentary source material. Where possible, analysis extends beyond medical personnel, facilities and procedures to examine broader medico-historical perspectives, including vulnerability to disease, cultural attitudes to soldiers’ health, especially combat wounds, and arrangements for invalided servicemen. Contrary to traditional notions of ‘decline and fall’, the evidence points to substantial continuity in institutional approaches to safeguarding soldiers’ health and treating those who became sick, injured or wounded, despite changes in the nature and expectations of military service in Late Antiquity.

This paper examines Justinian’s Novel 130 and associated documents with a view to elucidating asp. more This paper examines Justinian’s Novel 130 and associated documents with a view to elucidating aspects of military food supply in the sixth century, particularly from the perspective of interaction between military institutions and civilian communities. Issued in 545, this enactment specifies comprehensive procedural regulations for provisioning troops in transit within the empire, principally by means of compulsory purchase (coemptio), recognizing that such transient circumstances posed peculiar challenges of control, scrutiny, documentation, and accountability. An assessment of procedures, personnel, and implementation, in light of the recent legislative background and operational practicalities, reveals remedial innovations designed to protect rural taxpayers and food-producers from loss, damage, and intimidation but also to safeguard soldiers against exploitation and corruption. Investigation of historical contexts — fiscal, military, and agrarian — in the early to mid-540s finds general and specific motives for government intervention in this sphere, while the fragmentary epigraphic record preserves imperial responses to appeals by agricultural communities in Asia Minor afflicted by the passage of soldiers in the 520s or 530s, illustrating processes of complaint and redress, and, more generally, modes of communication between periphery and center. Ultimately, principles and practices prescribed in Novel 130, even if products of a specific time and place, exercised enduring legislative force, inasmuch as military logistical arrangements of the Middle Byzantine period have a discernible Justinianic pedigree.

Byzantium was heir to a tradition of Greek and Roman military literature stretching back to the f. more Byzantium was heir to a tradition of Greek and Roman military literature stretching back to the fourth century BC, which was manifest both in the collection, editing and adaptation of surviving texts from classical antiquity and in the composition of numerous new treatises devoted to warfare on land and sea. This broad genre always exhibited a diversity of content, style, language and approach, reflective of different categories of author and reader. Originating in a research project to prepare a full critical edition of Nikephoros Ouranos’ Taktika (c.1000), conventionally acknowledged as the longest and last representative of this literary and intellectual tradition, this paper explores the subsequent and more obscure history of this genre in the Late Byzantine period. Aspects of continuity are discernible in isolated instances of military writing, specifically a tactical opusculum by the scholar-courtier Michael Psellos (c.1050s-70s) and evidence for a lost work by the general Michael Doukas Glabas Tarchaneiotes (c.1297-1305/8), and in a partially overlapping but distinct genre of advisory literature (Kekaumenos, c.1075-8 Theodore Palaiologos, c.1326/7). The investigation addresses the more difficult question of the Late Byzantine audience(s) of military treatises, as reflected in evidence for aristocratic education and literary culture and in what can be inferred from manuscript production and ownership. In particular, these criteria show the continued esteem accorded to Greco-Roman ‘classics’, notably Ailianos’ Taktika Theoria (c.106-13 AD). More generally, they highlight the socio-cultural function of this genre as a component of the schooling, identity and perspectives of Late Byzantine military and civilian elites, over and above whatever practical utility these texts might (or might not) have possessed as ‘technical’ or ‘scientific’ literature. The discussion introduces some hitherto unexploited manuscripts in the Topkapı Palace Library (TSMK G.İ. 19 and 36).

(This paper was originally presented at the First International Conference on the Military History of the Mediterranean Sea, Fatih Üniversitesi, Istanbul, 26-28 June, 2015).

ABSTRACT: Among the Greco-Roman texts that exercised a formative influence on the military cultur. more ABSTRACT: Among the Greco-Roman texts that exercised a formative influence on the military culture and literature of Early Modern Europe, by far the most popular Greek military “classic” was Onasander’s Strategicus (Στρατηγικός), an ethical-philosophical treatise on the qualities, education and conduct of an ideal general, written by a Platonic philosopher in the 1st century AD. One of the first Ancient Greek military handbooks to be made accessible to a western readership from c.1455, by the mid-18th century Onasander had become the most often printed, most widely translated and most extensively commented-upon Greek military author. The project examines the reception of Onasander’s Strategicus c.1500 – c.1750, primarily in Germany and France, but also in Italy, the Netherlands and England. Using the exceptional collection of early editions, translations and commentaries at the Forschungsbibliothek Gotha, the project aims to explain how and why this work gained such popularity and authority, and remained relevant to readers within diverse and shifting literary, intellectual, didactic, socio-cultural and military contexts of Late Humanism and the Enlightenment. The study encompasses the transmission and collecting of manuscripts printing and book culture translational methodologies and translations as a medium of cultural transfer strategies of patronage and promotion the interaction between classical texts and contemporary discourse on war.

KEY WORDS: Reception of ancient literature codicology printing and book culture in the Early Modern period Ancient Greek military literature Early Modern military culture(s)

ABSTRACT: Modern scholarship has long acknowledged the popularity and influence of Greco-Roman mi. more ABSTRACT: Modern scholarship has long acknowledged the popularity and influence of Greco-Roman military literature in the early modern period, both within the intellectual, educational and cultural currents of late humanist scholarship, and in relation to the “military revolution” in northern Europe. While previous research has mainly focused on better-known classical authors, the corresponding significance of Byzantine military texts, which first began to be studied, translated and printed from the mid-16th century, has been largely neglected. The research project examines the reception of Byzantine military literature c.1550-c.1700, primarily in the Dutch Republic, Germany, England and Scandinavia, but also in France, Italy and Russia. This multidimensional study encompasses the transmission and collection of Greek manuscripts developments in editorial principles, printing and book culture Latin and vernacular translations as a medium of cultural transfer the implications of the “barbarous” post-classical idiom, technical content and cultural provenance of Byzantine texts with respect to strategies of patronage and promotion their contribution to the creation of the first modern lexica of medieval Greek and to defining “Byzantium” as a newly emergent field of scholarly enquiry and the influence of this corpus of Byzantine military treatises on contemporary military theory and practice in northern Europe. The primary objective is a monograph that will be of interest across disciplinary specialisms, including early modern literary and intellectual history, Byzantine studies, military history and the history of ideas.

KEY WORDS: Classical and Byzantine technical literature, codicology, printing and book culture, literary reception, lexicography, warfare, Byzantine studies

ABSTRACT: This research project aims to assemble, study and publish the surviving fragments of a . more ABSTRACT: This research project aims to assemble, study and publish the surviving fragments of a Greek work on cryptography or specifically steganography, the science and practice of transmitting information in secret, primarily for the purposes of war and espionage, both crucial elements of ancient and medieval statecraft. These fragments are preserved as discrete collections of excerpts incorporated into two Byzantine military treatises, the so-called Sylloge Tacticorum (c.950) and the Taktika of Nikephoros Ouranos (c.1000). These two works derive in part from a common source, a lost military compendium or ‘Corpus Perditum’, in which this cryptographic material formed a self-contained component of undetermined date and authorship. The two partially overlapping excerpt collections broadly coincide in sequence, structure and content, but differ substantially in form, style, idiom and editorial methodology. The relevant chapters of the Taktika, the more authoritative and complete witness, have not previously been published. The original text prescribed diverse methods of secret correspondence, ranging from simple concealment to ingenious contrivances. The content encompasses modified extracts from extant classical military and historical literature and otherwise unknown material of unidentified provenance. The research objective is to prepare a first critical edition of all the fragments with an English translation to analyse their complex textual transmission, compositional history, sources and literary affinities and to attempt to reconstruct the multi-layered cultural milieu and historical contexts of these cryptographic fragments, both as a rare specimen of a poorly attested (sub)genre of Greek technical literature and in relation to the reception of Greco-Roman texts in Byzantium. The project builds on preliminary textual and codicological studies undertaken during a Humboldt-Forschungsstipendium für erfahrene Wissen-schaftler (Institut für Byzantinistik, Ludwig-Maximilians-Universität München, 2009-11) and subsequent research in the Greek manuscripts collection of the Topkapı Palace Library, Istanbul (Senior Research Fellowship, RCAC, Koç University, 2013-14).

KEYWORDS: Nikephoros Ouranos, Taktika, Sylloge Tacticorum, classical Greek and Byzantine technical literature, classical reception, military science, Byzantine codicology, Byzantine philology

ABSTRACT: The Taktika of Nikephoros Ouranos is a vast compendium of military science compiled by . more ABSTRACT: The Taktika of Nikephoros Ouranos is a vast compendium of military science compiled by a distinguished Byzantine general, courtier and man of letters in c.1000. The last and longest treatise written in a self-conscious genre of Greek tactical manuals that stretches back fourteen centuries, the Taktika incorporates material from classical Greek, Roman and Byzantine military literature, ranging from the fourth century BC to the tenth century AD, modified and/or supplemented by Nikephoros’ observations on contemporary practices. Today sections of the Taktika are preserved in three manuscript prototypes, in Munich, Istanbul and Oxford, none containing the whole work, but from which collectively the text can be reconstituted almost in its entirety. At the time of application, just 21 of 178 chapters are available in modern critical editions. The proposed research project has three main objectives:
1. to edit a substantial section of the Taktika that has never been published (chs. 75-178). The edition will be based on a collation of two codices, Monacensis gr. 452 (158 folios) (c.1350-60), in the Bayerische Staatsbibliothek, and Oxoniensis Baroccianus 131 (262r-288v) (c.1250-80), in the Bodleian Library, which represent two different recensions of the text.
2. to identify extant Greek, Roman and Byzantine sources used by Nikephoros for which he had access to better and/or fuller manuscripts than those that survive today, and thus where his Taktika potentially represents a more accurate and/or complete indirect textual tradition, overlooked in modern critical editions of those sources (e.g. Onasander, Aelian, Arrian).
3. to identify fragments of ‘lost’ works embedded in the Taktika that have not otherwise survived in a direct tradition (e.g. excerpts from a work on cryptography).
This study represents the first stage of a longer-term, multiphase project to produce an editio princeps of the complete text of the Taktika, which will be of intrinsic value and allow a better appreciation of this soldier-scholar, and his methodologies as a compilator, editor, paraphrast and author.

KEYWORDS: Nikephoros Ouranos, Taktika, classical Greek and Byzantine technical literature, classical reception, military science, Byzantine codicology, Byzantine philology

PHILIP RANCE (FREIE UNIVERSITÄT BERLIN) The “barbarica conspiratio” of 367-9: Barbarian Threats . more PHILIP RANCE (FREIE UNIVERSITÄT BERLIN)

The “barbarica conspiratio” of 367-9: Barbarian Threats to Britannia in the Reign of Valentinian I

ABSTRACT:
During the reign of Valentinian I, in 367-9, Britannia was beset by a period of turmoil conventionally termed the “barbarica conspiratio”, during which various hostile barbarians – Picts in northern Britain, Scotti and Atecotti from Ireland, and Saxons from the continent – appeared to co-ordinate assaults on several fronts and threatened to overwhelm the military and civil administration of the diocese. Although one of the better-documented episodes in fourth-century Britannia, scholarship has long acknowledged the shortcomings and possible distorting bias of the surviving sources, particularly inasmuch as all explicit testimony dates to or immediately after the reign of Theodosius I (379-95), whose father, comes Theodosius, had been responsible for restoring order in Britannia in 368-9. Accepting that barbarian invasion was one element of a multifaceted crisis, this paper aims to clarify the nature of the external threat to Britannia in terms of the scale and locations of incursions and the aims and identities of the perpetrators. It also assesses the imperial response and subsequent defensive measures implemented in Britannia in relation to Valentinian’s military policies on other frontiers of the western Roman Empire. In addition, a chronologically sensitive re-evaluation of the textual evidence attempts to discern how this emergency might have been perceived and portrayed during Valentinian’s reign, before it acquired enhanced significance in Theodosian dynastic image making after 379.

PHILIP RANCE (FREIE UNIVERSITÄT BERLIN) ‘Late Byzantine Elites and Military Literature: Authors. more PHILIP RANCE (FREIE UNIVERSITÄT BERLIN)

‘Late Byzantine Elites and Military Literature: Authors, Readers and Manuscripts (11th-15th Centuries)’


The Assassination Strategy

The Assassins did not enjoy a great military strength and so their strategy of targeting specific and powerful opponents was a good one. The weapon of choice for assassination was almost always the knife, and the mission was usually carried out by a small team, sometimes in disguise as beggars, ascetics, or monks. The assassination was often planned to be carried out in a crowded location to maximise the political and religious consequences of the act. The assassins were not expected to survive their mission and were known as fidain (sing. fidai) or ‘suicide commandos.’

That men were willing to die for the ‘Old Man of the Mountain’ is clear but the reasons why are not. Marco Polo (1254-1324 CE), the Venetian explorer, offers the following explanation in his Travels, an account of his adventures across Asia in the last quarter of the 13th century CE, information which may also explain the real use of hashish amongst the Assassins:

The Old Man was called in their language Al-eddin…In a beautiful valley enclosed between two lofty mountains, he had formed a luxurious garden, stored with every delicious fruit and every fragrant shrub that could be procured…Palaces of various sizes and forms were erected…The inhabitants of these palaces were elegant and beautiful damsels, accomplished in the art of singing, playing upon all sorts of musical instruments, dancing and especially those of dalliance and amorous allurement…At his court, likewise, this chief entertained a number of youths…To them he was in the daily practice of discoursing on the subject of the paradise announced by the prophet…and at certain times he caused opium to be administered to ten or a dozen of the youths and when half dead with sleep he had them conveyed to the several apartments of the palaces in the garden. Upon awakening….each perceived himself surrounded by lovely damsels, singing, playing, and attracting his regards by the most fascinating caresses, serving him also with delicate viands and exquisite wines until intoxicated with excess of enjoyment…he believed himself assuredly in Paradise…When four or five days had thus been passed, they were thrown once more into a state of somnolency, and carried out of the garden…Questioned by him [the Old Man] as to where they had been, their answer was, ‘in Paradise, through the favour of your highness.’ The chief thereupon addressing them, said: ‘We have the assurances of our prophet that he who defends his lord shall inherit Paradise, and if you show yourselves devoted to the obedience of my orders, that happy lot awaits you.’

(Bk 1, Ch. XXII)

There is a corroborating passage from a text called the Xishiji by Chang-de, the Chinese government official and traveller, written in 1263 CE. Here, Chang-de notes that the Assassins:

…spotted any strong man [and] they lured him with material goods…They intoxicated him, escorted him to a basement and entertained him with music and beauty. They let him indulge in sensual pleasure…At the time he woke up…they taught him that if he could die as an assassin, he would live in joy and comfort like that.

(quoted in Hillenbrand, 24)


Mao Tse-tung and the Art of War

2 Sun Tzu: The Art of War ( Oxford , 1963 )Google Scholar .

3 Mao Tse-tung on Guerrilla Warfare ( New York , 1962 )Google Scholar The Battle for Guadalcanal ( Philadelphia , 1963 )Google Scholar .

4 See “ The Glorious Military Thought of Comrade Mao Tse-tung' ,” Foreign Affairs , 42 . 4 (July, 1964 ) 669 –74CrossRefGoogle Scholar .

5 The Giles translation has been reprinted twice in the United States by the Military Service Publishing Company, Harrisburg, Pennsylvania: as one chapter in Major Thomas R. Phillips (ed.), Roots of Strategy (1941), and as a separate book entitled Sun Tzu Wu, The Art of War (1944).

6 Three Military Classics of China (Sydney, 1944).

7 Because of the different systems used in numbering sections of the original text, references given below to the two principal English translations (Griffith's volume under review and Giles , San Tzu on the Art of War , London , 1910 )Google Scholar are to pages. References to Mao's , writings are to Selected Military Writings of Mao Tse-tung ( Peking , 1963 )Google Scholar .

8 Kuo Hua-jo, “Sun-tzu ping-fa ch'u-pu yen-chiu,” appeared in the Chün-cheng tsa-chih (Military-Political Magazine) of the Eighth Route Army in 1939. See Hua-jo , Kuo , Sun-tzu ping-fa ( Peking , 1962 ), 5 .Google Scholar

9 Translation as given in Selected Military Writings, 86. Griffith's version, taken from Mao, Selected Works, I, 187, differs slightly. Still another reference to this axiom of Sun Tzu as a “scientific truth” appears in Mao's On Protracted War of May 1938 (Selected Military Writings, 238).

10 In English, the most convenient body of material is that edited by the Foreign Languages Press, Selected Military Writings of Mao Tse-tung (Peking, 1963), which contains twenty-nine important speeches, articles, and directives covering the period from October 1928 to April 1949. In Chinese, a useful selection of passages from Mao's writings dealing with military affairs has been prepared by the Honan branch of the China Historical Society in its periodical, Shixue yuekan (Shih-hsueh yueh-k'an), published at Kaifeng by the Honan People's Publishing Company. References on “war” appear in part 4 of the compilation, Shixue yuekan. No. 2, February 1959, 19–33.

11 See the two papers by Bobrow , Davis B. , “ Peking's Military Calculus ,” World Politics , XVI . 2 (January 1964 )Google Scholar and “Mao's Military Model” (Princeton: Center of International Studies, mimeographed, n.d.).

12 This review article does not attempt to survey the growing literature in English on Peking's nuclear doctrine.

13 (New York, 1963). Major O'Ballance is also the author of “The Armed Might of Red China,” Military Review, XL. 8 (November 1960), 33–42.

15 (New York, 1940). Carlson also wrote The Chinese Army: Its Organization and Military Efficiency ( New York , 1940 )Google Scholar .

16 Clubb , O. Edmund , 20th Century China ( New York , 1964 )Google Scholar Johnson , Chalmers A. , Peasant Nationalism and Communist Power: the Emergence of Revolutionary China, 1937–1945 ( Stanford , 1962 )Google Scholar Liu , F. F. , A Military History of Modern China, 1924–1949 ( Princeton , 1956 )Google Scholar . The volume by Colonel Rigg , Robert B. , Red China's Fighting Hordes ( Harrisburg, Pennsylvania , 1951 )Google Scholar , is flashy and journalistic, reflecting limited understanding of China and less of the Chinese Communists.

17 Aside from the works cited below, specific mention should be made of the interesting paper by Bondurant , Joan V. , “Paraguerrilla Strategy: a New Concept in Arms Control” in Singer , J. David (ed.). Weapons Management in World Politics Google Scholar , proceedings of the International Arms Control Symposium held in Ann Arbor, Michigan, December 17–20, 1962, published in the joint issue of The Journal of Conflict Resolution , VII . 3 (September 1963 )Google Scholar and the Journal of Arms Control, I. 4 (October 1963), 235CR–245CR.

18 The first and in some respects still the best theoretical analysis is the paper by Katzenbach , Edward L. Jr. and Hanrahan , Gene Z. , “ The Revolutionary Strategy of Mao Tse-tung ,” which appeared in the Political Science Quarterly , LXX . 3 (September 1955 ), 321 –40CrossRefGoogle Scholar . Others who approach the subject with understanding of the Chinese environment are Hanrahan , Gene Z. (ed.), Chinese Communist Guerrilla Tactics ( Washington : Department of the Army , 1952 , mimeographed)Google Scholar Chiu , S. M. , Chinese Communist Revolutionary Strategy, 1945–1949 ( Princeton : Center of International Studies , Research Monograph No. 13, 1961 )Google Scholar and Hinton , Harold C. , “Political Aspects of Military Power and Policy in Communist China,” in Coles , Harry L. (ed.), Total War and Cold War ( Columbus, Ohio , 1962 ), 266 –92Google Scholar .

19 Though the only point of which the bibliographer of guerrilla warfare may be certain is that his listing will be outdated before it is published, the following are representatives of the genre: Dixon , Brigadier C. Aubrey , and Heilbrunn , Otto , Communist Guerrilla Warfare ( New York , 1955 )Google Scholar Fall , Bernard B. , Street without Joy: Insurgency in Indochina, 1949–1963 , third rev. ed. ( Harrisburg, Pennsylvania , 1963 )Google Scholar Garthoff , Raymond L. , “ Unconventional Warfare in Communist Strategy ,” Foreign Affairs , XL . 4 (July 1962 ), 566 – 575 CrossRefGoogle Scholar Giap , Vo Nguyen , People's War, People's Army ( New York , 1962 )CrossRefGoogle Scholar Lt. Col., T. N. Greene, USMC (ed.), The Guerrillaand How to Fight Him, selections from the Marine Corps Gazette ( New York , 1962 )Google Scholar Che Guevara on Guerrilla Warfare , with introduction by Peterson , Harries-Clichy , Major, USMCR ( New York , 1961 )Google Scholar Heilbrunn , Otto , Partisan Warfare ( New York , 1962 )Google Scholar Johnson , Chalmers A. , “ Civilian Loyalties and Guerrilla Conflict ,” World Politics , XIV . 4 (July 1962 ), 646 – 661 CrossRefGoogle Scholar Knorr , Klaus , “Unconventional Warfare: Strategy and Tactics in Internal Strife,” in Zawodny , J. K. (ed.), “Unconventional Warfare,” The Annals of the American Academy of Political and Social Science , Vol. 341 (May 1962 ), 53 – 64 Google Scholar Lindsay , Frank A. , “ Unconventional Warfare ,” Foreign Affairs , XL . 2 (January 1962 ), 264 – 274 CrossRefGoogle Scholar Marine Corps Gazette, special issue on guerrilla warfare, Vol. 46. 1 (January 1962) Franklin Mark Osanka, (ed.), Modern Guerrilla Warfare, with introduction by Huntington , Samuel P. , “ Guerrilla Warfare in Theory and Practice ” ( New York , 1962 )Google Scholar Paret , Peter , and Shy , John W. , Guerrillas in the 1960's ( New York , 1962 )Google Scholar Tanham , George K. , Communist Revolutionary Warfare: the Vietminh in Indochina ( New York , 1961 )Google Scholar United States Naval Institute , Studies in Guerrilla Warfare ( Annapolis , 1963 )Google Scholar .

20 See, for example, Dinerstein , Herbert S. , War and the Soviet Union , rev. ed. ( New York , 1962 )Google Scholar Erickson , John , The Soviet High Command ( London , 1962 )Google Scholar Fedotoff-White , D. , The Growth of the Red Army ( Princeton , 1944 )Google Scholar Gardioff , Raymond L. , Soviet Military Doctrine ( Glencoe, Illinois , 1953 )Google Scholar and Soviet Strategy in the Nuclear Age , rev. ed. ( New York , 1962 )Google Scholar Hart , B. H. Liddell (ed.), The Red Army ( New York , 1956 )Google Scholar Sokolovskii , V. D. (ed.), Soviet Military Strategy , translated and edited by Dinerstein , Herbert S. , Goure , Leon , and Wolfe , Thomas W. of the RAND Corporation ( Englewood Cliffs, N. J. , 1963 )Google Scholar and other works.


Heyday and Fall of Ērānšahr

Subscribe to the CK II Expansion and enjoy unlimited access to 13 major expansions and more!

Paradox Midsummer Sale has arrived! Up to 75% off!

Enjoy some sun and song this Midsummer, but when the sun goes down, the fun doesn't have to stop! Paradox has a festive sale on plenty of games to keep your summer nights going!

Crusader Kings III Available Now!

The realm rejoices as Paradox Interactive announces the launch of Crusader Kings III, the latest entry in the publisher’s grand strategy role-playing game franchise. Advisors may now jockey for positions of influence and adversaries should save their schemes for another day, because on this day Crusader Kings III can be purchased on Steam, the Paradox Store, and other major online retailers.

Recent threadmarks

Semper Victor

Šahān Šāh Ērān ud Anērān

4.13. THE IBERIAN WAR (IV). THE BATTLE OF SATALA AND RENEWED PEACE TALKS.


In the summer of 530 CE, at the same time that his armies were invading Roman Mesopotamia, Kawād I also sent a second invasion army against Roman Armenia, thus launching a coordinated offensive in both parts of the common border. As with the battle of Dara, Procopius offers a detailed account of the events (and in this case, he is the only source too):

Procopius of Caesarea, History of the Wars – Book I: The Persian War, XV:


This opening passage of chapter XV of Procopius’ History of the Wars (entirely dedicated to this campaign), offers a lot of information, and is worth a detailed commentary. First, about the commanding officers. Modern scholars consider Procopius’ Mermeroes (Μερμερόης) to be a Greek corruption of the Middle Persian name Mihr-Mihrōē. Procopius also informs us that by this time Sittas had risen in rank to Magister Militum Præsentalis, i.e. commander of one of the two præsentalis armies based around Constantinople, and that he had been replaced as Magister Militum per Armeniam by Dorotheus. His presence in this theater implies that the Romans were not surprised in Armenia like they were in Mesopotamia by the Sasanian invasion, or that maybe they were preparing an attack of their own. Procopius also makes it clear that Sittas outranked Dorotheus, and so there was no split command issues, as the former commanded all the Roman forced in the theater, that he must have known well, as he had been Magister Militum per Armeniam until recently.

Another important information provided by Procopius is that the alliance of the Sabirs with the Romans had been short-lived, as 3,000 of their warriors had joined the army led by Mihr-Mihrōē. Obviously, Kawād I had managed to counteract Justinian I’s trans Caucasian diplomacy and had outbid him, thus convincing the Sabirs to join his cause. Obviously, as the Sasanians controlled the main Caucasian passes, there were no physical or logistic obstacles for the displacement of this body of Sabir troops to Sasanian-controlled Armenia. Other than the Sabirs, Procopius also informs us that Mihr-Mihrōē’s army also included another contingent of trans Caucasian allies, the “Sunitae”, who some modern scholars consider to be a Hunnic people. This is also interesting, for since the immediate aftermath of the Anastasian War, the Caucasian Huns had been generally favorable to the Roman cause and had kept up the raiding and pressure against the Sasanian border on the Caucasus, but now this array of peoples in the Sasanian army of Armenia seems to paint a different picture. After the excavations of the last twenty years, archaeologists have determined that the Darband Wall was rebuilt in stone during the VI c. CE, and that most probably Kawād I was responsible for this project. As we have seen in a previous post, he also reinforced and rebuilt the fortresses of the Darial Pass further to the west, so this shift in alliances among the steppe nomads of the Caucasus and beyond might be the result of this strengthening of the border, that dissuaded these peoples from their traditional raiding activities into Sasanian territory south of the Caucasus Mountains. Now, if they wanted to obtain riches from their southern neighbor, the only way left was to join its armies as “allies” (i.e. mercenaries or vassals entitled to subsidies).

From his actions at the start of the war, it is clear that Sittas was a proactive and daring commander, willing to take risks, and so he decided to attack first by means of a surprise raid against the enemy camp, although in Procopius’ text it is not made clear if this camp was the main Sasanian camp, or the camp of their Sabir and Hunnic allies. As we have seen in the previous thread “Ērān against Tūrān”, Sasanian camps in the V c. CE were formidable compounds, and scholars believe that this practice continued during the VI c. CE a surprise attack against one of such camps would not have achieved much. And indeed, Procopius imply so in the next passage, as Mihr-Mihrōē’s army invaded Roman territory soon afterwards, unopposed by Sittas’ army:

Procopius of Caesarea, History of the Wars – Book I: The Persian War, XV:


Procopius carefully leaves some things out of his account here. Satala (modern Sadak, in Turkey) had been during the Principate the base of Legio XV Apollinaris. It was then located at the border between the Roman province of Cappadocia and the independent Kingdom of Armenia. But after the partition of Armenian between the Roman and Sasanian empires in the late IV c. CE, the Roman border moved considerably to the East. The new main Roman fortified border city became Theodosiopolis (modern Erzurum, in Turkey), located 143 km to the east in a straight line.

Location of Theodosiopolis and Satala. You can also see the location of the fortresses of Bolum and Pharangium.

This means that the Sasanian army must have enjoyed a considerable numerical superiority, as the Romans retreated quite far into their own territory without fighting it. Procopius also does not mention what happened with Theodosiopolis, but we can infer from the lack of mentions to it that the city was bypassed by the invaders. It is quite unconceivable that the Sasanians, who were usually careful about such things, would have left such an important fortified city (with an accordingly large garrison within its walls) unguarded behind their lines, so Mihr-Mihrōē must have left part of his army blocking the city. If despite this Sittas still decided not to fight him and to retreat, this means that Sasanian numerical advantage must have been quite substantial indeed, both magistri left all of Roman Armenia to the Sasanians before putting up a fight at Satala. According to Procopius, at the time of the final encounter the Sasanian forces amounted to 30,000 men against 15,000 Romans, which seems to me quite a low number for two joined field armies, even if Sittas had decided to leave most of the Field Army of Armenia ensconced within the walls of Theodosiopolis.

Given that Theodosiopolis’ fortifications had been reinforced by Anastasius I and Justinian I and that in Armenia the invaders did not enjoy the advantage of surprise, the decision by Mihr-Mihrōē to bypass it is not surprising clearly his objective was not to conquer territory, but to either plunder the Roman provinces or to engage and defeat the Roman field army in open battle.

Practically nothing remains today of ancient Satala above ground, and the site is still largely unexcavated. View of the remains of the eastern wall.

Satala is located in broken country, in a plain surrounded by mountains, a bit north of the upper Euphrates Valley, on the crossroads of two Roman roads: one that led from Samosata in the southwest to Trabzon to the north, and another that led from Amaseia and Bithynia in the West to Theodosiopolis and Armenia. Probably, the Sasanian invaders followed this road on an east-west direction. Upon reaching the environs of Satala, the army of Mihr-Mihrōē built a camp according to Sasanian custom, for apparently the retiring Roman army had taken refuge within the walls of Satala. They probably used for this the old camp of Legio XV Apollinaris, that this unit (according to archaeologists) had occupied until the V c. CE, and which was probably large enough to accommodate Sittas’ 15,000 soldiers. According to Procopius, the Sasanians pitched their camp in a place called Octava, located fifty-six stadia from Satala, that is, about eight to nine km.

The only visible remains of Satala today above ground is this fragment of the aqueduct that once brought water to the settlement from the surrounding mountains.

Sittas, who had demonstrated before that he was a dynamic and proactive leader, refused to simply wait for his force to be besieged at Satala, and instead he decided to lead a small part of his force (about 1,000 men according to Procopius) and take advantage of the hills and mountains that surrounded the plain of Satala to surprise his enemy. From Procopius’ account it seems that Sittas attacked with the Sasanians with this small force when they were already closing upon the walls of Satala, and that in that very moment the remaining Roman forces carried out a sally from the walls, thus taking the Sasanians between two fires and creating much confusion. Interestingly, Procopius also states clearly that both armies were formed entirely by cavalry, which is not surprising in the case of the Sasanians, but is more uncommon for the Romans this could explain why Sittas and Dorotheus only had 15,000 men with them (i.e. they only took the field with the combined cavalry of both the Field Army of Armenia and Sittas’ Præsentalis Army). Still, the Romans were unable to prevail in the fight until in a lucky coup the Romans were able to cut down the standard bearer of Mihr-Mihrōē, which caused the Sasanian army to retreat.

But still, and like it happened in Dara, the Sasanians were able to retreat in good order and undisturbed all the 143 km to the Roman-Sasanian border, and Sittas and Dorotheus decided not to pursue them. In this way, Justinian I’s commanders managed to beat the double invasion planned by Kawād I for the season campaign of 530 CE. The Roman victory in the north though had more consequences than in the south, especially among the always problematic nobility of Sasanian Armenia:

Procopius of Caesarea, History of the Wars – Book I: The Persian War, XV:


The significance of the fall of these two Armenian fortresses in Roman hands is bigger than what it may seem, for as Procopius wrote, there were gold mines in the vicinity of Pharangium, and these were important enough to have been one of the causes in one of the two Roman-Sasanian wars of the V c. CE. We will see this in more detail when we address the issue of the territorial administration of the Sasanian Empire, but there were gold mines in the trans-Caucasian territories of Armenia, Iberia and Albania, enough so that the Sasanian court created the post of zarrbed (lit. “master of the gold” in Middle Persian) to oversee the mines in these territories, a post that is unattested elsewhere in the Empire. Procopius goes into more detail about this event:

Procopius of Caesarea, History of the Wars – Book I: The Persian War, XV:


The Roman successes also caused some defections among the Armenian nobility:

Procopius of Caesarea, History of the Wars – Book I: The Persian War, XV:


So, Pharangium was handed over to the Romans by Symeon, a member of the Tzanni people, and Bolum by an Armenian noble called Isaac, who two elder brothers Narses and Aratius had already deserted to the Romans soon after their victory at Satala.

Seeing that the military situation had taken a turn in his favor, Justinian I decided to resume peace talks in the fall of 530 CE, and sent his ambassadors Rufinus and Hermogenes to the court of the Šāhān-šāh:

Procopius of Caesarea, History of the Wars – Book I: The Persian War, XVI:


In his response to the Roman ambassador as reported by Procopius, Kawād I kept insisting about the need that the Romans involved themselves in the defense of the Caucasian passes, i.e. that they contributed with money to the upkeep of the fortifications and the garrisons there. This is the same response that Kawād I had given to the Roman ambassadors during the previous attempts at peace talks and by now it was a standard Sasanian demand (also demanded by Pērōz, and by Kawād I during the Anastasian War). But now the Šāhān-šāh offered another possibility: he would agree to a peace between both empires if the Romans either contributed to the maintenance of the Caucasian fortresses, or if the Romans dismantled the fortifications of Daras, that he considered a Roman aggressive move against Ērānšahr.

John Malalas also mentions these peace talks in his work, and he included in it details absent from Procopius’ account:

John Malalas, Chronographia, XVIII, 53-:

At the end of the month of September the Roman ambassadors who had been sent to Persian territory returned, having made a treaty. The emperor Justinian, on learning that he had won peace for the Romans, was filled with joy. When he received the letter accompanying the treaty and read it, he found that it was as follows:

“Our ambassadors who had been sent to your Clemency have now returned and have announced to us the good intention of your paternal disposition. we have rendered thanks for all things to the Lord God in that an event befitting his goodness has taken place and that peace has been made with the help of God to the benefit of the two states and the credit of us both. It is clear that great glory and credit is due in all the earth before God and men to the fact that peace has been established between the two worlds under the reign of your Clemency and of us who truly love you. The enemies of both our states will be destroyed when with God’s help this peace is established. Our ambassadors then will arrive with all speed, for they must complete what is necessary to secure the peace. We pray indeed that your paternal disposition be preserved for many years.”

Rufinus was sent once more by the Romans, whence a second letter was dispatched to Persian territory he found the Persian emperor had withdrawn from the peace agreement they had made between them. For news had come that the Samaritans in Roman territory, incurring the anger of the emperor Justinian, as was described above, had fled and gone over to Koades, the Persian emperor, from their own territory in Palestine, and had promised to fight for him. They numbered 50,000. They promised to hand over to the Persian emperor their own land, all Palestine and the Holy Places, a city which possessed donations from various emperors, both a large sum of gold and an untold quantity of precious stones. When the Persian emperor heard this and had been convinced by their statements, he withdrew from the agreement to make the treaty. He made his excuse the question of the gold-bearing area that had been discovered formerly in the time of the emperor Anastasios and was under Roman jurisdiction these mountains had formerly been part of the Persian state. the gold-bearing mountains lie on the border between Roman Armenia and Persarmenia, as the experts say. These mountains produce much gold, for when rain and storms occur the soil of these mountains is washed away and pours out flakes of gold. Previously certain people leased these mountains from the Romans and Persians for 200 “litrai” of gold, but from the time the mountains were taken over by the most sacred Anastasios only the Romans were in receipt of the revenue that had been decreed. This was what upset negotiations over the treaty.
The Romans learnt of the Samaritan betrayal when certain of their men of substance were captured on their return from Persian territory, and were recognized after their journey to Koades, the emperor of the Persians, and after their agreement with him to betray their land as was mentioned above. There were five Samaritans who were recognized. On being captured, these were taken before the Magister Militum per Orientem and were examined in his presence. They confessed to the treachery which they were planning. the report on them was read to the emperor Justinian.
(…)
At this time, an ambassador was sent by the Persian emperor to the Roman emperor and, having handed over the letter he was carrying, he was sent away bearing gifts.
When the Roman emperor heard from the ambassador Rufinus about the transgression of the emperor of the Persians, Koades, he composed and despatched sacred commands to the emperor of the Axoumitai.


Malalas continues this passage by stating that Justinian I’s ambassadors convinced the Negus of Axum to invade and conquer Ḥimyar, but this seems to be a chronological mistake by Malalas, as the Axumite invasion of Ḥimyar had taken place the preceding decade, against the Jewish king Dhū Nuwās. What seems more interesting from the end of Malalas’ account is that allegedly the Axumite monarch sent “his Saracens” (i.e. the south Arabian tribes under his control) against the “Persian Saracens”, that is, against the Laḵmids and the tribes controlled and to or controlled by them in eastern and central Arabia.

Practically nothing is known about Sasanian heraldry and vexillology. This is a fragment of a late Sasanian wool and linen textile preserved at the National Museum in Athens, where a standard bearer can be seen carrying a standard behind a king or general. The scholar Matteo Compareti believes that the sixteen-rays rosette might have been a symbol linked to the goddess Anāhīd, while other scholars link it to Mihr/Mithra, the Sun God.

But the rest of Malalas’ account is really interesting, as the reasons mentioned in it for the failure of the peace talks are completely different from those described by Procopius. I am not versed enough in the study of these Greek sources to be able to guess the reason for such huge differences, but they are quite puzzling. Procopius was a strict contemporary of the facts, and a member of the imperial bureaucracy he clearly knew people in the imperial administration and was generally well informed this is why the absence of the story about the Samaritans and Justinian I’s appeal to the Axumites from his account is so puzzling, as well as the very different account about the Armenian gold mines: according to Procopius, they were in the Persian side of the border, and according to Malalas, they were in Roman territory.

Malalas wrote his account in the second half of the VI c. CE, so he was not a strict contemporary of the events, and he was a lawyer of Syrian origin (probably from Antioch) who moved to Constantinople and developed his legal career in the capital, and wrote his Chronographia in Greek. In this respect, he was not an “insider” of the imperial administration as Procopius had been, and these two points would have been enough to discard his account in favor of Procopius’ one, if it were not by two facts: first, that it makes evident sense in itself, and second, that it is quoted almost exactly by an even later source, Theophanes the Confessor (IX c. CE), but with some details and sentences absent from Malalas this strongly suggests that both Malalas and Theophanes resorted to another, now lost source, that may have been contemporary to the events, and independent from Procopius. What this source may have been, I do not know, for as I have said above I am not versed enough in the Greek sources of the VI c. CE.

The fantastic creature that can be seen in this Sasanian silver dish (kept in the Hermitage Museum in St. Petersburg) is often described as a “sēnmurw”. The sēnmurw, evolved into New Persian as “simorgh”, appears in Ferdowsī’s Šāh-nāma as a magical bird associated to the hero Rostām. But the scholars Matteo Compareti and Touraj Daryaee believe that the griffin-like creature depicted in this plate and in many other examples of late Sasanian art was not a sēnmurw (it doesn’t fit at all with Ferdowsī’s description), but merely a symbol or embodiment of the old Iranian concept of “xwarrah” (“fārr” in New Persian) the kingly glory that symbolized the right to rule bestowed by the gods into kings or nobles. As such, some historians associate it closely with the House of Sāsān itself, but that is far from clear, as the only direct link appears in a late rock relief of Xusrō II at Taq-ē Bostān in western Iran.

That the rebel Samaritans were in contact with Kawād I evidently makes sense. Even if they were motivated by Messianic hopes, the promise (or even the expectation) of military help by the Sasanian superpower must have seemed to them as a “once-in-a-lifetime” opportunity this would also explain why after the uprising was crushed in Samaria the Samaritans moved across the Jordan River into the Roman province of Arabia in this place they probably would have hoped that Sasanian help would have been better able to reach them, probably in the form of the Laḵmids of al-Munḏir III. What seems more surprising to me is that 50,000 Samaritans might have been able to reach the safety of Sasanian territory it seems quite improbable that they could have reached it crossing across Roman held territory, as this would have implied crossing across several heavily militarized provinces, so the only open way would have been the wastes of the Syrian desert, probably with the help of the Laḵmids. But evacuating 50,000 refugees (Malalas seems to imply they were all men capable to bear arms, but this seems quite impossible to me, they would have been 50,000 in total, including women, children, and the elderly) across such a large extension of desert would have been a real logistic achievement, so I have my doubts about the total numbers. As for offering them asylum within Ērānšahr, that would not have been a problem, as the Sasanians had a long history of resettlements, deportations and taking in persecuted minorities within their Empire. What is also surprising is that Justinian I and the court of Constantinople took so long to realize this and reveals that (once more) the Sasanians seemed to be more capable at deception, espionage, and cover operations than their Roman foes.

Both accounts may be complementary to some point, although the reasons for the disagreements are so wildly different in them, that I have some difficulty putting them together. Personally, I would put more credibility on Procopius’ account in this case, if only because they fall more in line with the usual bones of contention between the Roman and Sasanian empire since the start of the VI c. CE, the influence of the Samaritan refugees and the issue of the gold mines may have been secondary or even “tactical” considerations that pushed Kawād I towards trying to keep the military pression up against the Roman Empire in the hopes of extracting more favorable peace terms from the court of Constantinople.

Pzt_Kami

Second Lieutenant

I personally do not believe Romano-Greek accounts especially when it comes to numbers. Many of them proved to be false and for propaganda purpose.

Still another swesome read from you @Semper Victor

Semper Victor

Šahān Šāh Ērān ud Anērān

I personally do not believe Romano-Greek accounts especially when it comes to numbers. Many of them proved to be false and for propaganda purpose.

Still another swesome read from you @Semper Victor

First of all, thank you for the appreciation.

As for the sources, I try to engage in some criticism within my very limited qualifications and abilities the proper study of these ancient texts and the provenance of the informations they offer is a highly specialized and demanding task for which high levels of skill in ancient languages and scripts (and having read a lot of primary and secondary literature) are a must, and I lack all of this.

But Procopius is quite a clear and uncomplicated case, compared to many other ancient sources. The bulk of what he wrote happened in his lifetime and he was a contemporary, and in some cases a direct witness, so his data in these cases are usually reliable. Another matter is when he writes sbout events that happened before his lifetime or in places far removed from his location a good example is his account of Peroz's final defeat and death. It is evident that he based his account on an older source, but which one? Ancient historians rarely mentioned where they took their information from, and this is a serious problem for modern historians when they find contradictory accounts of a same event in different ancient texts.

Generally speaking, Procopius is a reliable source when he offers data about the Roman army of his time, like the numbers of men that took part in a certain campaign or battle, and even more so if he was present. As Belisarius' secretary, he would have had access to all (hell, he probably redacted them himself) the official dispatches and reports that Belisarius sent to the imperial court of to other fellow generals or officials. That doesn't mean he wasn't manipulative or tendentious when it suited him, all the contrary. He is probably the ancient historian from whom a larger part of his work has survived to our times and that has allowed us to get a very good perspective of his many philies and phobies, but as far as I am aware he is a reliable authority when dealing with numbers related to the military. Not so when dealing with the personal life of Justinian and Theodora.

Pzt_Kami

Second Lieutenant

As I said dear @Semper Victor I'm not criticizing you as you should logically be loyal to the sources you chosen for your analysis (and apparently you have a good reason because Procopius is much more reliable than other historians according to you)

I just wanted to remind others to generally be wary of numbers in history books. Herodotus is a prime and maybe among the earliest example of a western historian who tries to manipulate facts/Numbers for the Propaganda purposes. Let's assume Procopius to be a very honest guy still Mistakes can happen. I've heard somewhere that even nowadays there can be misyakes when it comes to numbers in historical accounts/records of battles ,Like WWII that is less than a centuary away from us.

The reason I'm stongly disagree here and especially on the numbers in your last post is it totally seem illogical to me is that a Force twice the number of opposing side (30000 to 15000) easily beaten while both even have using nearly the same type of units (both are Cavalry armies and probably mostly heavy Cavalry) . I'm neither historian nor military expert but we both have massive interest in military history and persoanally read alot of books and other materials about it ,so to know that while it is not impossible but extremely difficult to achieve victory against a foe twice stronger

Semper Victor

Šahān Šāh Ērān ud Anērān

As I said dear @Semper Victor I'm not criticizing you as you should logically be loyal to the sources you chosen for your analysis (and apparently you have a good reason because Procopius is much more reliable than other historians according to you)

I just wanted to remind others to generally be wary of numbers in history books. Herodotus is a prime and maybe among the earliest example of a western historian who tries to manipulate facts/Numbers for the Propaganda purposes. Let's assume Procopius to be a very honest guy still Mistakes can happen. I've heard somewhere that even nowadays there can be misyakes when it comes to numbers in historical accounts/records of battles ,Like WWII that is less than a centuary away from us.

The reason I'm stongly disagree here and especially on the numbers in your last post is it totally seem illogical to me is that a Force twice the number of opposing side (30000 to 15000) easily beaten while both even have using nearly the same type of units (both are Cavalry armies and probably mostly heavy Cavalry) . I'm neither historian nor military expert but we both have massive interest in military history and persoanally read alot of books and other materials about it ,so to know that while it is not impossible but extremely difficult to achieve victory against a foe twice stronger

Satala was not a frontal battle the Sasanian army fell into a trap and was attacked unexpectedly from the front and rear, and the Romans were lucky and struck down the standard-bearer of the enemy general in most ancient armies, such a situation would have caused the dissolution of the army into a panicked mass of men and animals trying to retreat, which clearly did not happen here, or at Dara. Procopius does not say it explicitly but it is quite clear from the context otherwise Sittas (or Belisarius) would have pursued their foes. That the Sasanian army was able to react in such a professional and disciplined manner and did not lose its cohesion anytime is testimony to its high military standards. Indeed, Procopius makes it clear in his account of the battle of Dara that a Roman victory in open field against the Sasanians was a rare feat.

As for superiority in numbers, the next post will deal with the battle of Callinicum in 531 CE, in which the numbers were reversed: 15,000 Sasanians and Lakhmids against a Roman army of +30,000 men (including their own Arab foederati), and Procopius' account does not hide what happened: it was a complete Sasanian victory, and the Romans were saved from a disaster because of the terrain and because the Roman infantry stood firm at the end with their backs to the Euphrates. There is little or no manipulation of numbers in Procopius, although he of course tries to manipulate some things in his account of Callinicum he tried to shift the blame for the Roman defeat against a much smaller enemy onto the Arab foederati, in order to protect the reputation of his patron Belisarius. But emperor Justinian was not fooled and had Belisarius cashiered he only recovered the augustus' favor because he led his private army against the Nika rioteers in the Hippodrome of Constantinople a few years later and organized an obscene bloodbath among the populace of the capital. This saved Justinian's throne, and the grateful emperor appointed him to lead the expedition against the Vandals, which turned him into the most prestigious Roman general of his generation, thus restoring his military career.

Pzt_Kami

Second Lieutenant

Semper Victor

Šahān Šāh Ērān ud Anērān

4.14. THE IBERIAN WAR (V). THE BATTLE OF CALLINICUM.


For this post, I will follow mainly the account by Procopius of Cæsarea in his History of the Wars, while I will also resort to the (much shorter) account by the Pseudo-Zacharias of Mytilene and the Chronicle of John Malalas, whose account differs substantially from Procopius. As a secondary source, I will also use the paper by the British historian Ian Hughes about the battle of Callinicum (published in Ancient Warfare Vol. 5, Issue 3). But there is a problem in all this: Hughes and other historians usually follow Procopius’ account closely, although it clashes considerably with Malalas. Again, this happens because Procopius was not only a strict contemporary of the events, but that as Belisarius’ secretary, he may have even been present in the battle. For the sake of legibility, I will follow first Procopius’ account and Hughes’ reconstruction of the events, and at the end I will quote the two other accounts in full and try to address on the differences between them.

After the failure of the peace talks during the previous fall and winter seasons, Kawād I renewed the attacks against Oriens in the spring of 531 CE, as described by Procopius:

Procopius of Cæsarea, History of the Wars – Book I: The Persian War, XVII:


The Sasanians attacked this time along the Euphrates route, instead of more to the north in the north Mesopotamian plain the same route that had been used by Julian in his 363 CE invasion or by Šābuhr I in his second campaign that culminated in his victory at Barbalissos and the first Sasanian sack of Antioch. Once they crossed the border into Roman territory, the Romans would have indeed entered the Roman province of Euphratesia, which had been not targeted by attacks of the Sasanian army in this war, but which had been attacked several times by their Laḵmid allies. So, Procopius statement that until that point in the war Euphratesia had not been attacked by the Sasanians is right, but only if we ignore the raids by al-Munḏir III. And indeed Procopius offers an elaborate explanation about why this route was chosen this time, and he attributes it directly to the advice given by the Laḵmid king to the Šāhān-šāh:

Procopius of Cæsarea, History of the Wars – Book I: The Persian War, XVII:


Pērōz Mihrān, the defeated commander at Dara, was a member of the wuzurgān, the upper crust of the Iranian nobility, so any punishment that Kawād I may have wished to inflict on him would have been more symbolic than physical. Procopius’ account that he retired from him the right to wear an ornament that he displayed on his hair fits with what is known about displays of elite status in Sasanian Iranian: as I have commented in previous posts, numismatist Rika Gyselen thinks that the kolāh displayed by certain high officials in Sasanian seals might have been a bejeweled hat (other scholars think that it might have actually been not a hat, but a helmet, maybe made of precious metals) reserved exclusively to the upper nobility and members of the House of Sāsān. Moreover, in almost all cases these kolāhs also displayed a sort of emblem on the sides that can be seen clearly in seals and in the great rock reliefs of the III c. CE that depicted the kings and their courts. So, maybe Kawād I withheld the right to wear the kolāh from the unfortunate Pērōz Mihrān, or maybe the right to display an emblem on it.

Two personal Sasanian seals depicting high-ranking dignitaries (the one on the right is a mowbed) wearing the kolāh with personal emblems on it. In some cases, scholars have been possible to establish that these emblems are actually example of Pahlavi text with abbreviations of their names or posts, but in other cases their meaning remains obscure.

The gold cap above was found in Korea and is dated to the V c. CE, to the Kingdom of Silla. It is remarkably similar to a Sasanian kolāh, and it even displays the “feather motive” known from Sasanian swords and helmets. Objects like this one, highly influenced by the steppe cultures, have been found in abundance in archaeological sites that belong to the Kingdom of Silla.

What the Laḵmid king al-Munḏir III proposed to the Šāhān-šāh was the sort of deep incursion into the Roman rearguard that he had undertaken several times in past years, and according to Procopius it aimed at outflanking the heavy Roman defensive deployment that existed in Mesopotamia to the north and in Syria to the south (the Limes Arabicus, now guarded by the Arab fœderati under the command of the Ghassānid king Ḥārith ibn Jabala). The Euphrates Valley, located in between both border sectors, was lightly guarded in comparison, and offered a direct venue of approach to the rich Roman provinces in northern Syria and to Antioch itself. Of course, it was a risky operation, and everything would depend on speed and surprise, for if Belisarius’ army at Dara and the Ghassānids to the south were given time to react, regroup their forces and maneuver, they could easily close the Sasanian force’s path of retreat, and even trap it between superior forces. The invasion would be a major raid aimed at forcing Justinian I to accept Kawād I’s peace terms, and to make it potent enough to achieve this diplomatic goal, it would be mainly formed by Sasanian regular forces with the Laḵmids, led by their king, as auxiliaries. But the overall commander would be an Iranian general:

Procopius of Cæsarea, History of the Wars – Book I: The Persian War, XVIII:


So, the army would be guided by al-Munḏir III in its path of invasion, but the overall commander would be Azarethes, whom Procopius calls “an exceptionally able warrior”. As usual, scholars are fairly sure that Azarethes was not his name, and that once more Procopius is rendering here, in a Hellenized form, a title or rank rather than a personal name. The Pseudo-Zacharias of Mytilene calls him “the Asthebid”, which could be a corruption (through Syriac) of Middle Persian Aspbed, i.e. “Master of the Horse” (“Commander of the Cavalry”), an ancient military rank that existed already under the Arsacids, while the British scholar Geoffrey Greatrex thought it could be a Greek corruption of the Middle Persian office Hazāraft (also rendered as Hazāruft and which was perhaps the same office as Hazārbed). This title means in Middle Persian “Commander of the Thousand”, and some scholars believe that it may have designated the commander of the “royal guard”, i.e. the elite troops of the royal household (what western historians anachronistically called “the 10,000 Immortals” following the lead of Herodotus and Thucydides). Any one of those two hypothesis imply that he was a man of extremely high rank, and that he belonged to the highest circles of the Sasanian court immediately below the Šāhān-šāh. He reappears in western texts only once more, during the account of the siege of Edessa in 544 CE by Procopius (then serving under Xusrō I, Kawād I’s son and successor). In Procopius account, it seems that this Sasanian invasion came as a complete surprise for the Romans:

Procopius of Cæsarea, History of the Wars – Book I: The Persian War, XVIII:


The first sentence in this passage informs us that the invaders advanced along the right bank of the Euphrates, probably in order to avoid the dangerous operation of having to cross a major river within Roman territory. It was the same route used by Šābuhr I in the disastrous (for the Romans) campaign that led to the battle of Barbalissos and the fall of Antioch in 252-253 CE, so Procopius was wrong when he stated that the Sasanians had never used this route. Belisarius (who was presumably based at Dara) was taken by surprise, and he had to split his forces: part of them were left to garrison the cities of Mesopotamia in case this was just a Sasanian feint, and he moved with the rest against Azarethes and Al-Munḏir III. Procopius states that he crossed the Euphrates “in great haste” presumably, he did so to the north of the Sasanians’ line of advance, in the vicinity of Edessa/Zeugma, and from there they reached Chalcis in Syria (modern Qinnasrin), where they encamped. In Chalcis, Belisarius’ army was joined by the Ghassānid king Ḥārith ibn Jabala. We do not know how large this contingent of Arab fœderati was Ian Hughes assumed that it amounted to 5,000 men, probably all of them cavalry, based on the account by John Malalas (that we will address later) thus raising the overall numbers of the Roman army under Belisarius to about 25,000 men.

At this point, Belisarius’ army was located (according to Procopius) at about 120 km of the invading force, presumably blocking its path of advance towards Antioch. Upon learning this, the invaders decided to retreat along the same path they had followed in their approach (“with the Euphrates to their left” according to Procopius), and Belisarius pursued them. The army that Belisarius took with him to Chalcis amounted to twenty thousand men, including infantry and cavalry, and it also included the forces of the Ghassānid king Ḥārith ibn Jabala.

Map of the Roman provinces of the northern part of the Levant. The Euphrates divided Euphratesia to the south from Osrhoene to the north notice also how at Callinicum (modern Raqqa in Syria) the Euphrates flows practically on an eastward direction.

According to Procopius, Belisarius was satisfied only with having cut the advance of the invaders and forced them to retreat, and he did not intend to fight a battle with them, despite the fact that he enjoyed numerical superiority, and his men were aware of it. But the course of events would soon take a turn against Belisarius’ intentions:

Procopius of Cæsarea, History of the Wars – Book I: The Persian War, XVIII:


This is a long passage, and in here Procopius’ defends Belisarius, trying to relieve him of any responsibility in the defeat: he did not want to fight the battle, but his men forced it upon him, even disrespectful to God Himself because they disrespected Easter (Procopius knew very well that this argument would go down well with the pious Emperor). According to Malalas, the battle happened on Easter Saturday, 19th April 531 CE, and at the time it was customary for Christians for fast during Holy Week before Easter Sunday. Both sources agree on this, but it seems strange to me that Belisarius would risk a major battle this way (which could happen if he approached the Sasanian army) if his men were so weakened by a week of fasting. Also, according to Malalas’ account, Belisarius and his generals sought the battle and did not shrink from it.

The Sasanian army was formed entirely by cavalry and so it should have been able to flee without trouble from Belisarius’ mixed infantry-cavalry army. If the latter was able to catch his foes, the most probable explanation for it is that the forces of Azarethes and al-Munḏir III were loaded with loot and prisoners captured in Syria (and indeed, Malalas’ account is explicit in this respect). It would have been exceedingly difficult for Belisarius to explain to Justinian I why he had allowed an inferior enemy to escape from his reach when it was loaded with Roman plunder and prisoners. The only reason that could make this reluctance believable is that Belisarius had secret instructions from Constantinople not to engage in field battle if it could be avoided, otherwise it would have been a decision impossible to justify, but Procopius says nothing of the sort (and if this were the case, he would just have said so, as it would have saved him the effort of having to resort to the “religious” justification). The interception had also happened relatively far from the border, which was located at Circesium, 161 km downstream the Euphrates from Callinicum.

Belisarius’ speech to his troops is nothing more than the usual rhetorical device so beloved by ancient historians, especially those of the classicizing variety like Procopius. Callinicum was located on the left bank of the Euphrates, so the Romans would not have the benefit of a walled city to their rearguard where they could have taken refuge in case of defeat, so both armies deployed for battle:

Procopius of Cæsarea, History of the Wars – Book I: The Persian War, XVIII:


It is difficult, or even impossible, to overcome a mistake in an initial deployment for battle, and according to Ian Hughes Belisarius made several of them in this occasion. According to Procopius, the Roman right wing (i.e. the Sasanian left one) was located on a series of low hills that rose sharply from the river valley, while the Roman left wing and center were indeed on relatively flat ground that rose towards the hills to their right. So, this seems to imply (and Hughes stated so) that the opposing armies positioned themselves in a perpendicular direction to the Euphrates. The problem is that this “reconstruction” by Hughes as based in Procopius’ account, goes explicitly against the initial deployment of the Roman army as stated by Malalas, who wrote that the Roman army was deployed “with their backs to the river”, that is, parallel to the Euphrates, and that Belisarius had ordered a large number of boats to be assembled on the river to his rearguard, to keep communications open with Callinicum and so that the army would be able to retreat if necessary. I will return to this point later.

Belisarius stationed himself in the center, and thus (according to Hughes) he lost the opportunity of having a clear view over the entirety of the battlefield. This was to be a serious mistake. He also put the Roman infantry (led by Peter, a member of Justinian I’s guard, another example of how these units acted as a recruiting pool for Roman officers) to his left, anchored on the riverbank. The regular Roman infantry led by Peter was to avoid a complete disaster in the end (in Procopius’ version). Procopius does not mention them here, but with the information he furnished in the passages that follow this one, we can reconstruct the rest of the Roman line of battle. In the center of the Roman line, to the right of Peter’s infantry, there was the Roman cavalry commanded by Belisarius himself, and to his right a group of Huns (high-quality troops that had won single-handedly the battle at Daras) commanded by Simmas and Sunicas. To their right was a further force of Roman cavalry under Ascan, and to the right of this latter force there was a force of 2,000 Isaurian infantrymen under the command of Longinus and Stephanacius. And finally, the Arab fœderati, commanded by Ḥārith ibn Jabala, were located at the extreme right of the Roman line. From Malalas’ text, we know that there were other “Saracen phylarchs” in that wing, and so that most probably not all of these 5,000 men were Ghassānids.

View of the Euphrates near Raqqa in the 1920s. In recent times the landscape of this part of the Euphrates Valley has been considerably altered by the building of the Tabqa Dam between 1968 and 1973.

It was quite clear that the Sasanians would not be able to try to outflank the Roman left flank, especially as, being spring, the Euphrates probably carried even more water than usual, so the decisive fight had to happen either in the center or the Roman right flank. But for the Sasanians, who were inferior in overall numbers but probably superior in cavalry, attacking in the center when the Romans had an open right flank would have been quite a bad choice. The logical course of action (and what the Sasanians followed in Procopius’ account) would be to try to concentrate their forces (using their superior tactical mobility) against the Roman right flank to try to envelop the Roman center and left wing, and this is exactly what Azarethes did. Until now, Belisarius had only fought static frontal battles at Thannuris and Dara against the Sasanians he was unused to the danger posed by their superior cavalry, and so he had to learn it the hard way. Also, as we will see in the passages that follow, Belisarius made another mistake: he put the recently recruited and inexperienced body of 2,000 Isaurian infantrymen in a key position, linking the Roman troops with the Arab fœderati, and these untested troops collapsed quickly when they were attacked by the Sasanians in their flank and rear.

Procopius of Cæsarea, History of the Wars – Book I: The Persian War, XVIII:


From this passage, historians have assumed that 2/3 of the Sasanian army was formed by Iranian savārān, and 1/3 by their Laḵmid allies, assuming that the three parts in which Azarethes divided his army were of equal strength. That would mean 5,000 savārān in each of the Sasanian right wing (anchored on the Euphrates’ riverside) and center, and 5,000 Laḵmid horsemen in the Sasanian left wing, which would prove to be the decisive sector of the battlefield, presumably under the command of al-Munḏir III. Hughes guessed that the Sasanian force was deployed in a single line, due to the fact that they had to cover the same front the Romans did with less men, but this would go against customary Iranian battle deployments, as described in later Islamic treatises. In my opinion, given Sasanian custom and how the battle developed, it is more probable that (like at Daras) the three bodies of the Sasanian army (including the Laḵmids) deployed in two lines. This would have allowed them to keep a constant rotation of men in the front shooting arrows constantly and following the traditional Arsacid and Sasanian “wave tactics” while providing a reserve able to exploit any weakness in the enemy line or to perform maneuvers behind the front line, which would be harder to see by the enemy in the middle of the dust and confusion of the battle.

The battle began with the usual mixture of individual duels and the interchange of arrows. According to Procopius, Roman archery was more effective than the Sasanian one, which seems quite unlikely. Based on this single passage by Procopius, some modern scholars have hypothesized that by this time the Romans had fully adopted the larger Hunnic composite bow while the Sasanians still used the older and smaller Parthian bow (a variant of the Scythian bow). The problem is that this is not only unattested by archaeological or iconographic findings, but that it is quite unlikely. Out of the two empires, it was the Sasanians who not only had always given more importance to archery (as Procopius admits) but they also had to fight far longer and bitter wars against the Huns and other Inner Asian nomadic peoples, and so it seems quite likely that (if the Hunnic bow was superior) they would not have hesitated to adopt it. The Sasanian employ of bows in battle was based on mass archery tactics, by deploying rapid concentrated fire against dense enemy formations, and for this tactic, Hunnic bows would have been as suitable as smaller Parthian ones. Hughes accepted Procopius’ account and considered that maybe Belisarius hoped to break the morale of the numerically inferior Sasanian army through attrition if so, he was sorely mistaken.

Aerial view of the remains of the wall of Raqqa as it was before the Syrian civil war. This wall was built in Abbasid time during the 780s CE, when the city underwent a period of great splendor, and cover an area probably much larger than that of the previous Roman city, of which nothing remains above ground (and as far as I am aware of, no excavations have been undertaken either).

As Procopius admits, after a long exchange of arrows between the Romans and the numerically weaker Sasanian force, the battle was still undecided, so Azarethes decided to launch his attack:

Procopius of Cæsarea, History of the Wars – Book I: The Persian War, XVIII:


This was the decisive moment of the battle, and if Procopius account is true to the events, this was a true stroke of military genius by Azarethes, at the level of Frederick the Great or Napoleon. What happened? According to Hughes, the rear lines of the Sasanian right and center moved to the left and together with the Laḵmids, launched a concentrated assault against the Arab fœderati. Personally, I think more probable that the Sasanian general ordered the second lines of his right and center to carry out this maneuver. The Sasanians probably took advantage of the lack of visibility in the battlefield caused by more than 20,000 horses running and galloping across it the Sasanian front lines in the right and center kept launching their “wave” attacks against the Romans and showering them with arrows, fixing them in place, and the second line could carry on this displacement towards the left undetected. In this way, Azarethes concentrated his numerically inferior forces against the Roman right wing, thus achieving local numerical superiority. Most probably, if Belisarius had stationed himself in his right flank on higher ground, he could have detected the enemy maneuver, but as he was in the center, he was unable to see it.

Initial deployment of both armies at the battle of Callinicum, movement of part of the Sasanian cavalry towards the left wing of the Iranian army and collapse of the Roman right wing (according to Procopius). Source: Wikipedia.

Procopius blamed the Ghassānids for fleeing the Sasanian onslaught without opposing resistance, and Hughes believed his account and excused Ḥārith for it stating that unlike Belisarius the Ghassānid king would have seen what was happening and that his men were greatly outnumbered and decided to preserve his forces. The problem is again that Malalas wrote that despite the fact that “some Saracen phylarchs” fled, Ḥārith kept on fighting. As I have written in previous posts, Procopius’ dislike against Ḥārith ibn Jabala is obvious in his works and he did not disguise it, so Malalas account seems more credible. Perhaps the other Arab phylarchs resented having been put recently under the suzerainty of the Ghassānid basileus by the Roman augustus. It is quite revealing in my opinion that, according to Malalas, after the battle Justinian I ordered an official investigation to be carried on, and Belisarius was replaced as Magister Militum per Orientem, but Ḥārith did not suffer any reprisals or punishments.

Having destroyed the Roman right wing, now the Sasanian and Laḵmid cavalry turned to their right and began to systematically roll the Roman line up, attacking it from the flank and rear, and with the added benefit that they were attacking from the higher ground, which would have greatly helped their visibility, archery, and charges. Now, the other mistake by Belisarius (placing his untrained Isaurian infantry in this flank) would become painfully obvious:

Procopius of Cæsarea, History of the Wars – Book I: The Persian War, XVIII:


Procopius displays here the issue of the battle taking place on Easter Day and the Romans being weakened by fasting (which makes me wonder why the men bullied their general into battle in the first place, according to Procopius, and what had happened to the energy with which they had been peppering the enemy with arrows until a few moments earlier). Ascan was the commander of the Roman cavalry stationed on the right of the Roman line, and apparently he put up an energic resistance, until he fell in battle, and then his cavalry collapsed. The name “Ascan” suggests he was not a native Roman (perhaps a Goth or an Alan) and that his cavalry was perhaps a “barbarian” unit in Roman service in these cases, the death of their tribal chief/commander almost always led to the rout of the unit. Although in Malalas’ account, he talks of “Phrygians” fighting in this wing together with Ḥārith’s Arabs, so they could have been Roman cavalry after all. As for the Isaurian infantry, it broke down completely and was massacred by the enemy Procopius explains it by informing us they were in fact green recruits and by the rather surprising news that the majority of them were not Isaurians but Lycaonians (as if that explained their lack of martial spirit, I guess).

The Sasanian and Laḵmid cavalry rolls up the Roman line collapse of the Roman center and last stand of the Roman infantry and the survivors by the riverbank (according to Procopius). Source: Wikipedia.

After the utter collapse of what remained of the Roman right, the Sasanians kept rolling the Roman line up, and fell upon the Roman center (where Belisarius had located himself) on its front, back and flank:

Procopius of Cæsarea, History of the Wars – Book I: The Persian War, XVIII:


The Roman cavalry in the center collapsed swiftly under the onslaught, and Belisarius fled to the security offered by what remained of the Roman infantry led by Peter in the Roman left flank (according to Procopius, many of the infantrymen had fled too), which was probably formed in close order (in a “phalanx”). There, Belisarius dismounted, and his cavalrymen followed his example, to join the infantry in fending off the enemy. It was their last opportunity. If what remained of the Roman army also collapsed and routed, the Sasanian cavalry would butcher them, and the defeat would become a complete disaster. Procopius does not mention the two Hun commanders in Roman service Sunicas and Simmas at all, and in my opinion this can be quite telling if we look at Malalas’ account, which we will do at the end of this post.

Procopius of Cæsarea, History of the Wars – Book I: The Persian War, XVIII:


The Roman infantry and the dismounted cavalry formed some sort of closed formation with their backs to the river and managed to withstand the Sasanian attack. It is noticeable the discipline of the Sasanian force, for Procopius states that the cavalrymen who were pursuing the feeling Romans returned to take part in the attack against the last resisting group of Romans. This gives some credibility to the claims by some historians that for this campaign Azarethes had been entrusted by Kawād I with elite forces, perhaps some gunds of the “royal guard”. The exceptional performance of the Sasanian savārān during the battle also seems to support this assumption. Hughes hypothesized that the Romans may have deployed in the Late Roman formation known as fulcum, forming a triangle with one of the sides unoccupied (the one that aligned with the riverside), spearmen in the other two sides and bowmen in the center, allowing them to fight in a very compact formation impossible to outflank, and offering maximum protection to their archers.

If the Roman foot soldiers stood firm, this formation would have been impossible to break for the enemy cavalry, and although the Sasanians kept trying to do so until nightfall, finally they had to retreat without having achieved so, and the Romans also abandoned the battlefield under cover of darkness. Procopius also offers an account of the outcome after the battle:

Procopius of Cæsarea, History of the Wars – Book I: The Persian War, XVIII:


Procopius does not mention Belisarius’ fate in his coda to the battle of Callinicum, and only mentions it, in passing, at the start of chapter XXI of this same book:

Procopius of Cæsarea, History of the Wars – Book I: The Persian War, XXI:


The reason alleged by Procopius for Belisarius’ removal from his post as Magister Militum per Orientem is quite unconvincing, as the expedition against the Vandals was not launched until late June 533 CE, two years later. Let us now see the short account of the battle by the Pseudo-Zacharias of Mytilene:

The Syriac Chronicle of the Pseudo-Zacharias of Mytilene, Book IX, Chapter IV:


And finally, the longer account by John Malalas. After it, I will make a comparison between the three accounts:

John Malalas, Chronographia, XVIII, 60-61:


The three accounts have some points of agreement and many points of disagreement. As the longest and most detailed account and written by someone who may have even been present at the battle is the one by Procopius, I will treat it as the “main” or “canonical” account, and I will compare the other two to it.

I will also remind the readers here that Procopius wrote as Belisarius “advisor” and personal secretary and so he was strictly contemporary to the events and personally involved in them. The so-called Pseudo-Zacharias of Mytilene was an anonymous Miaphysite cleric (probably from Amida) who was also a contemporary of the events and who wrote in Syriac. And that John Malalas was a lawyer native from Antioch who moved later in life to Constantinople to carry on his trade and where he wrote his Chronicle in Greek, during the second half of the VI c. CE, that is, three or four decades after the events. Thus, while Procopius and the Pseudo-Zacharias wrote either as eyewitnesses, from what they had learnt from people who were present in the battle, or from news that spread soon after it, Malalas must have resorted to some written source now lost to us, but one that lacked the personal involvement of Procopius and his biases (i.e. his attempt to protect Belisarius’ reputation, his antipathy towards Ḥārith ibn Jabala, etc.) so although he is the most removed in time and space from the events, he was probably more impartial than Procopius, or at least as impartial as his source was. The great amount of disagreements between his account and the one by Procopius makes it clear that he did not use the latter as source.

The three accounts are in agreement about the date when the battle happened: towards the end of the Easter fast, and Malalas precises it happened on Easter Saturday, April 19th 531 CE. The three accounts also agree that it took place by the Euphrates, although the Pseudo-Zacharias of Mytilene does not detail the precise location, but the other two accounts agree that it happened on the right riverbank, opposite to the walled Roman city of Callinicum. Other than this, they are in disagreement about everything else.

Belisarius’ reluctance about fighting during the Easter fast is omitted altogether from Malalas’ account, while in the text of the Pseudo-Zacharias, it is the Sasanian commander who invokes it in order to avoid battle with the superior Roman force, and after some deliberation Belisarius inclines to agree with his proposal, but his officers (not the men at large, as in Procopius’ account) oppose it strongly, and he has to fight the battle.

Malalas’ disagreements with Procopius begin with very start of the campaign. According to Procopius, this was a marauding expedition that was aimed at taking Antioch itself in a coup de main, while in Malalas’ account as soon as it crossed the border the Sasanian force built a fortified encampment and began looting the province of Euphratesia (that Malalas calls anachronistically “Osrhoene”). The description of the thorough fortification work done by the Sasanian army agrees strongly with what (as we saw in the previous thread) the Sasanian army did in Gorgān, the southeastern Caucasus and northern Iran during the V c. CE and so is in itself perfectly credible. The problem is that such behavior is completely absurd in case of a surprise raid, so both versions are incompatible with each other. Malalas adds that they had encamped in front of the fortress of Gabboula (Gabula in Latin, presumed to correspond to the modern village of al-Jabbūl in Syria, near the salt marsh of the same name). Furthermore, Malalas states that the Sasanian army formally besieged the fortress with war machines and managed to take it, sacking it, and enslaving its dwellers. This further contradicts Procopius’ description of the campaign as a surprise raid and seems to imply that the Sasanian army might have also included infantry in order to carry out the siege works.

Possible portrait of Belisarius standing to the right of Emperor Justinian I in the mosaics of the main apse of the church of San Vitale in Ravenna, Italy.

There is also disagreement about the number of men in the Roman army and their concentration place. According to Procopius, Belisarius marched to Chalcis with an army of 20,000 from Dara, and in Chalcis (modern Qinnasrin, in Syria) he was joined by the Arab fœderati led by Ḥārith ibn Jabala, who probably would have journeyed there from the south. According to Malalas, Belisarius came to this menaced part of Oriens with 8,000 men, and it is unclear if the 5,000 led by Ḥārith were included in this total or not when he reached Euphratesia he reinforced the local duces (probably the duces of this province and the neighboring provinces). This again makes sense in itself the Romans did not need to move the whole Field Army of the East to meet a limited thread like this one, as they had already the provincial armies lead by their respective duces. Belisarius would have moved directly towards Hierapolis Bambyce (modern Manbij in Syria), with his men (either 8,000 or 5,000 men, depending if the Arab fœderati are included in the initial total or not). But the first one to react according to Malalas had been the dux Sunicas, who with 4,000 men had harassed the Sasanian marauders. The fact that he is called a dux and that he was the first to react probably means that Sunicas (who was present at Dara, was a Hun by birth and had performed brilliantly there) wa probably the dux of the province of Euphratesia (according to Malalas report). Belisarius, as overall commander of the Roman forces in Oriens, clearly resented this show of independence by Sunicas, and it was necessary for a higher official to arrive to settle this issue.

This is another disagreement between Malalas and Procopius. In the latter’s account, Hermogenes is completely absent, while according to Malalas, he reinforced Belisarius at Hierapolis with 4,000 further men, including the commanders Stephanos and Apakal and the dux Simmas (the other Hun commander who had played such a brilliant role at Dara, and who also appears here also elevated to the dignity of dux). As an ex-Magister Officiorum and the Emperor’s personal envoy, Hermogenes outranked Belisarios and forced a reconciliation between him and Sunicas. Malalas’ prose is quite unclear, and so it is difficult to get an idea of the total numbers of the Roman army. If we take a “maximalist” approach, when Hermogenes reached Hierapolis the total numbers of the Roman army must have reached 21,000 men, quite close to Procopius’ numbers, but divided quite differently: 8,000 men had come with Belisarius from Dara, 5,000 were Arab fœderati, 4,000 were the provincial army of Euphratesia under Sunicas and another 4,000 were the reinforcements arrived with Hermogenes that included the other Hunnic dux, Simmas.

This methodic concentration of forces must have taken its time and so it would have furnished the Sasanians with plenty of time to besiege and take Gabboula, and then to start their retreat back to Ērānšahr, and it was while they were already on their way back that the Romans overcame them. In Malalas’ account, it is stated clearly that they were laden with loot and captives, so this must have slowed their march enough to allow the Romans to reach them.

The battle deployment according to Malalas seems at first view different to the one by Procopius, but a detailed look at it shows that is not the case. According to the former, the Romans deployed “with the river to their backs”, and with some sort of provisional bridge or ferry service made with boats covering the width of the Euphrates and linking their rearguard to Callinicum. According to him, Ḥārith and the Isaurians were deployed to the south while the two Hunnic duces Sunicas and Simmas were deployed to the north. As in this part of its valley the Euphrates flows in an eastward direction, this would imply that the Roman army was deployed perpendicularly to the river (otherwise, the distinctions between “south” and “north” would be absurd) and that Belisarius was located in the center, with the Arab fœderati and the Isaurians to his right (i.e. to the south) and the forces of Sunicas and Simmas to his left (i.e. to the north) that is, a deployment quite similar to the one described by Procopius, except that according to the latter Sunicas and Simmas were posted in the center an the Roman left wing was led by Peter.

There is no fancy maneuvering in Malalas’ account of the battle, just a frontal assault by the Sasanians against the Roman line. Initially the fight was undecided, with many casualties on both sides (according to Malalas, one of them was al-Nu’man, son of the Laḵmid king al-Munḏir III). According to this version, Apakal, one of the Roman commanders, charged into the middle of the Sasanians and was killed, which led to the flight of the “Phrygians” (Apakal is described as their “exarch”). This figure is quite similar to Procopius’ Ascan, and in both cases his death causes the flight of his troops, that according to Malalas were Phrygians. It is noticeable also that in Malalas’ account it is the death of this character that leads to the collapse of the Roman right wing (the Phrygians rout and this in turn causes part of the “Saracens” to flee) while in Procopius’ account the succession of events is reversed: the Arab fœderati rout first and then Ascan’s troops resist until he is killed. And here we arrive to another of the obvious disagreements between Procopius and Malalas: according to the former Ḥārith ibn Jabala fled, while according to the latter, he carried on fighting even after other Arab phylarchs had fled as we will see, Malala’s version is quite more credible in this respect. As in Procopius’ version, in Malalas text the Isaurians also rout and try to cross the Euphrates swimming. I should precise here that in the Euphrates riverbed just opposite Callinicum (modern Raqqa in Syria) there are several islands, so the crossing of the river is less difficult in this point that could be thought otherwise.

And then we reach the main point of disagreement between both sources: the role played by Belisarius himself. While in Procopius’ detailed account Belisarius can only be found guilty of committing deployment mistakes (i.e. of bad generalship) in Malalas account he is guilty of fleeing the battlefield with his men after the right wing crumbled, crossing the Euphrates to the safety of the left bank and the walls of Callinicum. And obviously, this behavior would have been unacceptable in a general. Again, both accounts are incompatible, and given Procopius’ obvious interest in upholding the reputation of his patron Belisarius, there is reason to suspect that he might have embellished his account, although it is impossible for us to prove it beyond doubt.

Two members of the reenactment group Numerus Invictorum in the garb of Roman infantry soldiers of the late VI – early VII centuries CE.

The role played by the Roman infantry of the left wing, its commander Peter, and Belisarius in Procopius’ account as the focus of the heroic last resistance against the Sasanians in Procopius’ account is attributed to the troops of the Roman right wing under Sunicas and Simmas. According to Malalas, these two Hun commanders dismounted and them and his troops fought as infantry fending off the Sasanian attacks, and remained in the battlefield after the Sasanians retreated, to the point that in this account it is the Romans who appear as the victors, having been left in possession of the battlefield. Of course, this part of Malalas’ account is also incompatible with Procopius. But there is something that makes me suspicious that again Malalas might have been right here: Sunicas and Simmas appear in the initial battle deployment by Procopius, but later they disappear completely from his narration, and given their excellent performance at Dara it is clear that they were good field commanders.

The outcome after the battle also makes more sense in Malalas’ version. Both sources agree that Belisarius was summoned to Constantinople, but Procopius’ statement that this was so that Belisarius could prepare the campaign against the Vandals is quite hard to believe. The invasion fleet towards North Africa would not leave the Golden Horn until late June 533 CE, more than two years later, and if Belisarius were a successful general Justinian I would have never taken him out from the East while there was still a war against the Sasanians going on. Malalas’ account that Justinian I sent Constantiolus to carry out an official enquiry makes more sense and is completely in line with Justinian’s way of acting, and that the conclusions of this enquiry were not favorable to Belisarius seems a more plausible reason for his being recalled to the capital and his replacement for Mundus as Magister Militum per Orientem. Notice also that Justinian I did not punish Ḥārith ibn Jabala either and this seems to imply again that Malalas’ account is correct i.e. that he was not guilty of fleeing the battlefield. If Procopius’ account were true, he would have been most probably dismissed as overall commander of all the Arab fœderati of Rome in the East, especially considering that at this point in time his appointment was still quite recent.

Still, both sides paint the outcome of the battle as either a Pyrrhic victory for the Sasanians with many casualties or as a Roman semi-victory as the Roman infantry remained in the battlefield. But the fact that Justinian I was forced to dispatch to Oriens Sittas with his præsentalis army (which had been in Armenia until then) as attested by both Procopius and Malalas, seems to suggest otherwise.


References

Primary sources

  • Population of Vindolanda (100 AD). "(the Tablets)" (shtml) . Vindolanda Tablets Online: The Roman Army: Activities. Centre for the Study of Ancient Documents, Academic Computing Development Team at Oxford University.
  • pseudo-Hyginus. "De Munitionibus Castrorum". The Latin Library. Ad fontes Academy. (Latin text.)
  • site. Polybius Web publication on Bill Thayer's
  • Roman government (160 AD). "(Military Diploma)". Military Diploma of Discharge and Roman Citizenship. Metz, George W. Legion xxiv website.
  • Unknown inscriber (3rd century AD). "(the Tombstone)". Tombstone of Anicius Ingenuus, Museum of Antiquities Website. Newcastle University.
  • Vegetius. "Flavius Vegetius Renatus Epitoma Rei Militaris Book I". Armamentarium. Selections, Latin and English juxtaposed by paragraph. Translator unknown.
  • Books I-III only. The unknown editor altered the translation "to conform to modern usage" and abbreviated the text. Access is by subtitle. Search only within subsection.

Secondary sources

  • Johnson, Anne (1983). Roman Forts of the 1st and 2nd centuries AD in Britain and the German Provinces,. London: Adam & Charles Black.
  • Keppie, Lawrence (1994). The Making of the Roman Army from Republic to Empire. New York: Barnes and Noble Books.


Watch the video: Wars In Ancient Sicily (December 2021).