sehepunkte 26 (2026), Nr. 1

William Lewis: Division of Empire

In recent years the sons of Constantine have begun to receive sustained scholarly attention. William Lewis' monograph Division of Empire: The Reign of the Son of Constantine is a fine addition to this welcome scholarly trend. The book is a mature piece of scholarship, the end result of a long-standing scholarly interest which began as an appendix to Lewis' masters thesis at Cardiff University, became his PhD thesis, and was published (in part) as a chapter in an edited volume. This is a fine scholarly achievement, which reflects the concentration of expertise for the study of late antiquity at Cardiff university (and brings inevitable scrutiny to bear on the university's recent decision to reduce its ancient history program). Its eight chapters span from the death of Constantine to the reign of Theodosius, but the majority are focused on the 340s and 350s. The first six chapters explore the carving up of the Roman empire into domains of influence following the death of Constantine and the political tensions that ensued, both between Constantine's sons, and involving other, leading Roman politicians such as Ulpius Limenius (Chapter 5) and Fabius Titianus (Chapter 6). The last two chapters respectively apply the knowledge gained to the remainder of the fourth century and then reflect on what the division of the Roman empire after Constantine may tell us about the nature of such political divisions in the Roman world. Another review has admirably summarised each chapter, this review will instead focus on discussing a few, specific points. [1]

Lewis makes a compelling argument that "the years 337 to 340 must be recognized as an individual period with a distinct character" but to some extent our understanding of the political machinations of Constantine's sons (Constantine II, Constantius II, and Constans) hinges on what happened in the immediate aftermath of their father's death. As Lewis acknowledges, the evidence for who did what remains unclear. This is surely a partial consequence of the political sensitivities that go hand-in-hand with implicating a still living, imperial figure in the bloody reprisals against nearly every potential rival to imperial power in Constantine's extended family, what Burgess famously labelled "The Summer of Blood". [2] Lewis' shifting of the focus to 340 is justified on the historical ground that the events of 340 are pivotal, but this approach also benefits from the availability of the primary evidence, certainly compared to the relative dearth of evidence for 337. To put it differently, was 337 critical in establishing a pecking order between Constantine's sons, or a cabal of interests? In the power vacuum that followed Constantine's death, did they all stand-up equally to assume the mantle of imperial power, or had the divisions that became manifest in 340 already started to show their fault lines?

A real strength of Lewis' approach is the wide net of evidence that he analyses, from detailed epigraphic analysis of the Troesmis Inscription and careful scrutineering of consular lists to the rich array of literary sources including both historiography and panegyric. In large part these sources are used with care and attention, but at times the historical claims of panegyrics could do with further interrogation. On p. 35, for example, Lewis analyses the mutiny of Constantius' army in Syria which is described by Julian in his first panegyric for Constantius, Oration 1. Lewis quotes from Jul. Or. 1.19a (only in translation, the Greek is not printed in the text nor the footnotes) and then offers his own assessment ""in this crisis there seemed to be but one hope of safety, that you [Constantius] should take charge of affairs and plan the campaign." It seems likely, then, that the troops wanted to begin the stalled campaign [...] and they were agitating for Constantius to assume the title of Augustus and return to command the army." This explanation is possible, but some discussion of Julian's agenda in making these comments is warranted. Was he trying to induce Constantius to busy himself with military campaigning? Or is this the well-worked panegyrical line that there are major political and military problems that only the honorand can resolve? Or does this reflect Julian's own conception of military leadership, that potentially mutinous soldiers must be busied and led in an initiative, rather than be left alone to become restless and insubordinate? All of these are possibilities, especially given there is no evidence (other than Julian's remarks, which fall well short of being conclusive) that Constantius' men wanted him to assume the title of Augustus.

Lewis' use of historiographical evidence is insightful, especially for historical events that lack the coverage that one would hope. This means that at times Lewis must infer evidence from other pieces of evidence. This is fraught but necessary ground given the scope of the inquiry. On pp. 67-68, for example, Lewis use an anecdote in Ammianus regarding Constantius' treatment of Amphilochius, who had been a tribune under Constans. Upon Constantius' arrival in Antioch, Amphilochius attempted to gain access to his court was barred "under well-founded suspicion of having sown the seeds of discord between the deceased brother [i.e. Constans and Constantine II)." This enables Lewis to infer that Ammianus' narrative of 340 in one of his now lost books alleged that "discord originated in the court of Constans." This argument may have been pressed further; Amphilochius' untimely death is suggestive of some sort of delayed, divine justice, and Constantius' claim that he had predicted Amphilochius' imminent demise (explored insightfully by Woods, whose work is uncited) further suggests a pronounced hostility on Constantius' part towards Amphilochius. [3] To substantiate his interpretation of Ammianus' likely criticism of Constans Lewis turns to a number of other historiographical sources, including Photius' Epitome of Philostorgius, Zosimus' account, and Symeon Magister's version from the tenth century. These sources all have something in common in that they are free from the compulsion to cast Constans in a positive light through a combination of historical distance and disdain for the 'official narrative' as championed by the majority of sources, including the ecclesiastical historiographical tradition of Rufinus, Socrates Scholasticus, and Sozomen. It would have been useful to see more discussion of how these two narratives, 'the official narrative' and the 'counter narrative' interacted with each other. Ammianus presumably knew Libanius' panegyrics well; Zosimus may well have known Ammianus' lost account; similarly the fifth century Greek historiographers knew Rufinus, and so it is not simply that these two narrative traditions operate in parallel, but that they responded to and critiqued each other, both across and within their narrative bounds.

The prosopographical analyses are well executed and reasoned and enable Lewis to create a rich characterisation of the dynamics between the East and West courts throughout the 340s, as involving both tensions but also negotiation and cooperation (p. 116). This approach necessitates correcting several assumptions that have permeated into the scholarly tradition, especially around how and why Constans and Constantius II appointed specific men to the consulship. By and large, such appointments were not controversial and ensured the ongoing stability of the Roman empire through a transitional decade. This detailed historical work enables Lewis to underline the contrast with the 350s where instability and uncertainty came to the fore, including a rash of usurpations. This prosopographical approach is fruitfully continued, as Lewis turns his attention to the prefecture of Rome and the operation and struggles of the western court in the early 350s. Ultimately Lewis concludes that Constans' was removal from office arose from "a consensus among his senior officials in Gaul" who threw their support behind Magnentius, providing the leading military commander with "administration, arms, supplies, legitimacy, and the city of Rome itself." (p. 152).

Lewis has done the field a tremendous service by offering nuance and depth to our understanding of a key decade in the fourth century. This book will surely exert an ongoing influence from the undergraduate lecture theatre to the kind of scholarly claims that are made about the political relationships between Constantine's sons, and how and when these relationships deteriorated.

Notes:

[1] T. Kaçar in: Bryn Mawr Classical Review (BMCR) 2025, https://bmcr.brynmawr.edu/2025/2025.11.27/.

[2] R. W. Burgess: The Summer of Blood. The "Great Massacre" of 337 and the Promotion of the Sons of Constantine, in: Dumbarton Oaks Papers 62 (2008), 5-51.

[3] D. Woods: Notes and Discussions: Ammianus Marcellinus 21.6.3: A Misunderstood Omen, in: Classical Philology (CPhil) 99 (2004), 163-168.