In this book, the author explores the changing roles of the most senior military officers of the late Roman empire, the magistri militum, from the initial creation of this rank probably under Constantine I in the 320s to the death of Theodosius I in 395. The main body of the text consists of four chapters of approximately equal length. The first two investigate the appointment of and roles played by the magistri militum as part of the larger military and political history of the fourth century while the final two perform different analyses of the information concerning the number and roles of the magistri militum established during these earlier narrative chapters. There is no overarching thesis.
After introductory sections including a survey of the state of research into the origin and development of the fourth-century military hierarchy and a survey of the key sources for this work, chapter one, ‘The Magistri Militum from 341 to 363 CE’, identifies all the magistri militum known to have been appointed during this period and investigates the various functions that they performed, with a strong emphasis on the nature of their relationships with different emperors. Particular attention is paid to the revolts by Vetranio in 350 and Silvanus in 355. Chapter Two, ‘The Magistri Militum from 364 to 395 CE’, repeats the process for the subsequent period, with particular attention on the rise and fall of the general Arbogast. In Chapter Three, ‘The Networks of the Magistri Militum’, Bendle introduces the reader to social network theory before using this to analyse first the network surrounding the rebel magister Silvanus in 355, then the networks surrounding Valentinian I during his reign. In Chapter Four, ‘The Prosopography of the Magistri Militum’, Bendle analyses the career paths of the fifty-two magistri militum that he has identified in order to determine what the typical sequence of office was before reaching this rank and whether this changed over time. He then analyses how long men normally held this rank. Finally, after a lengthy introduction examining what it meant to be a ‘barbarian’ in the fourth century, he attempts to classify these fifty-two magistri militum as either Roman or ‘barbarian’ to investigate the extent of the barbarization of the Roman military hierarchy by the end of the fourth century.
It is unfortunate that this book was published almost simultaneously with that by Kaldellis and Kruse on the development of the Roman military hierarchy in the eastern part of the empire during the period 361-630, this after almost half a century since the last systematic examination of this topic by Demandt.[1] Despite their different focuses, there is a substantial overlap between the present volume and that by Kaldellis and Kruse, and it is instructive to note their different approaches and conclusions. For example, given the potential relevance of the Notitia Dignitatum to any investigation of the structure of the late Roman military hierarchy, one would naturally expect any discussion of this topic to include detailed arguments concerning the date and value of this document. Kaldellis and Kruse do not disappoint here, as they offer a detailed argument in support of redating the composition of the eastern Notitia Dignitatum to c. 450, but Bendle offers only a few short lines on this document as he accepts the traditional dating of that part of it dealing with the eastern empire to c. 395 (pp. 25-26). It is not surprising, therefore, that he argues that Theodosius had five magistri in c. 384, since their number and locations suggest that Bendle has identified them as the holders of the five eastern magisterial posts listed in the Notitia Dignitatum (p. 111).
Bendle reaches different conclusions to those presented by Kaldellis and Kruse on numerous smaller matters resulting in different understandings of the nature of the military hierarchy at any point during the late fourth century. For example, Bendle argues that there were five eastern magistri militum at the time of the battle of Adrianople in 378 (Traianus, Victor, Sebastianus, Saturninus, and Julius; p. 96), while Kaldellis and Kruse argue that there were only two (Julius, Victor; p. 105). These differences are the result of contrasting interpretations of the same evidence rather than of the use by one side of evidence overlooked by the other. For example, Bendle seems to accept the traditional interpretation of Ammianus 31.12.1 to mean that emperor Valens restored the former magister militum Traianus to active service at the same rank again shortly before the battle. In contrast, Kaldellis and Kruse point out the text states only that Traianus was restored to military service. It does not specify the rank to which he was restored. As a result, they assume that he was restored instead to some lesser rank. But would a Roman officer really have accepted restoration to military service at a lower rank? Kaldellis and Kruse fail to prove this. Hence Bendle, even if he fails to subject Ammianus 31.12.1 to the same degree of scrutiny as Kaldellis and Kruse, need not be wrong on this matter.
The most serious criticism that may be made of this volume is that there is a tendency to avoid detailed analysis of the relevant sections of ancient text. For example, he reports that emperor Constans ordered his general Salia to escort two western bishops to the court of Constantius II in the east in 344 without revealing that the sole source for this event, bishop Theodoret, refers to Salianus rather than to Salia (p. 34). Consequently, he fails to explain why one should correct the text to refer to Salia instead. This tendency to avoid detailed textual analysis helps explain why Bendle does not cite the standard commentary on the history of Ammianus Marcellinus even once in his footnotes even though this history serves as the main ancient source for his work.[2] There is also a tendency to digress into descriptions of modern debates that are of tangential relevance at best to the topic under discussion. For example, he digresses at length on the controversy concerning the recognition by the western empire of the magister Bonosus as consul for the first half of 344 before it recognized Sallustius as the consul in his stead for the rest of that year (pp. 34-35), where this is of dubious relevance to his wider discussion of the roles of the magistri militum. One could perhaps explain Bendle’s desire to investigate this problem as he does if he had any new insights to offer, but he does not, despite his firm rejection of the two most recent attempts to explain this problem.[3]
Bendle attempts to offer new arguments on occasion, but they do not always work. For example, he argues that Ursicinus, the commanding officer of Ammianus Marcellinus, was probably a barbarian based on a rumour that he was “seeking the throne, not for himself, but for his children,” as described at Ammianus 14.11.2-3 (p. 187; see also pp. 194-95). But this misrepresents Ammianus’ text which, far from denying that Ursicinus was aiming at the throne, specifically states that he would aim at “higher things” (altiora) if the opportunity offered. Furthermore, the reason why Ursicinus’ enemies mentioned his sons to Constantius II, as they spread rumours against him, was not because he could not himself hope to gain the throne, but because of the effect that the mere mention of these sons would have on Constantius II, who was childless and painfully conscious of the fact. The reminder of Ursicinus’ fertility would be enough to arouse the jealousy of Constantius II against him, even if he did not accept any of the specific allegations against Ursicinus and his sons. Another apparent novelty lies in the suggestion that Julian Caesar may have played some part in the downfall and execution of the magister militum Barbatio in 359 as part of his larger plot to manoeuvre himself into a position from which to launch his rebellion against Constantius II (p. 63). However, little effort is made to support this suggestion, and Bendle shows no knowledge of a recent argument that the involvement of Barbatio’s wife in the death of empress Eusebia, the wife of Constantius II, may have been the real cause of the execution of both.[4]
This book is well-written and well-structured, but is probably not a book for the educated public or an undergraduate readership. It contains much of interest to specialists in the history of the Roman army and deserves a place of honour on the bookshelf of anyone interested in the command structure of the late Roman army. One may disagree with much that is written here, as I do, and still benefit from reading an attempt to come to grips with some very difficult evidence. Those who read it in conjunction with the recent volume by Kaldellis and Kruse will probably derive the most benefit from it. Finally, one notes that the standard of production is high, although some errors have slipped through (e.g., equites Romanus, p. 27) and two key diagrams (Figures Two and Three) are difficult to read on account of their colour schemes and small font.
Notes
[1] A. Kaldellis and M. Kruse, The Field Armies of the East Roman Empire, 361-630 (Cambridge, 2023); A. Demandt, “Magister Militum,” Real-Encyclopädie (Pauly-Wissowa) Suppl. 12: 553–790.
[2] J. den Boeft et al., Philological and Historical Commentary on Ammianus Marcellinus, 12 vols. (Groningen-Leiden, 1987-2018).
[3] B. Salway, ‘Roman Consuls, Imperial Politics, and Egyptian Papyri: The Consulates of 325 and 344 CE,’ Journal of Late Antiquity 1.2 (2008): 278-310; and D. Woods, ‘Flavius Bonosus and the Consuls of AD 344,’ Classical Quarterly 62.2 (2012): 895-898.
[4] D. Woods, “Chrysostom, Ammianus, and the Death of the Empress Eusebia”, L’Antiquité Classique 87 (2018): 177-192.