The present study by William Lewis, his revised and extended PhD thesis, seeks to investigate the history and nature of the division of the Roman empire as it occurred during the rules of the sons of Constantine from 337 to 353 c.e. In focusing on the reigns of Constantine II and in particular Constans, L. reappraises the administrative history of the western Roman empire and highlights how it was impacted by their rule.

The discussion is laid out in eight chapters. The main chapters examine several key instances of political and administrative events that can illuminate the possible unity or division of empire. Ch. 2, ‘Constantius, Divider of Empire’, revisits the succession of the empire after the death of Constantine in 337. L. is interested in the power dynamics between the three brothers Constantine II, Constans and Constantius II before and after the funeral of their father. The relationship between the governments of the three brothers and the character of their governance from 337 to 340 is examined in ch. 3, ‘The Constantinian Triarchy’. L. identifies flexibility, cooperation and certain degrees of unity in a period that is usually considered to have been marked by animosity and conflict. Ch. 4, ‘The Civil War of 340’, advances a convincing, revisionist reading of the civil war between Constantine II and Constans which blames the latter for conspiring of assassinating the former. The next two chapters form the core of the argument. Ch. 5, ‘Ulpius Limenius and the West’, reconstructs the administration of the empire between 340 to 350 based on prosopographical studies of the praetorian prefects and consuls of that decade. L. reappraises joint consular appointments as evidence of the respect of two brothers had for each other’s sovereignty rather than as reflection of a unified empire under dual sovereignty. Ch. 6 opens in similar fashion with a study of the urban prefects, before analysing their role in the overthrow of Constans. L. sees the division of empire as an established fact in 350. He blames dynastic consideration and ‘general stubbornness’ (15) on the part of Constantius II as reasons for the reunification of the empire in 353. Ch. 7 then offers a discussion of events under Constantius and subsequent emperors, especially the Valentinianic emperors who fell back on the model of the 340s, up to the Theodosian rulers when the division became permanent.

L. makes many useful and important observations. He argues that it was between 340 and 350 that, imperial collegiality having failed, the first true division of the empire was put in place and the empire ‘began to ossify into separate halves’ (4). While it did not last, L. considers this division to be a model for similar divisions under later emperors. By contrast, he is a fervent defender of the idea that the Diocletian’s empire was not a divided empire. His rule is rather seen as a period of the modularisation of the empire that facilitated and led to division in the 340s.

While this is helpful, there are several problems. First, L.’s argument that the division of the empire under Constans and Constantius II was structurally (or qualitatively) more thoroughly ‘divided’ (or partitioned?) than other arrangements between fourth-century emperors is impeded by a lack of comparison. For instance, the failure to discuss the situation of the empire under Constantine and Licinius in any detail (other than noting that the empire was ‘functionally divided’ then, 13) means that it remains unclear why their empire was less well ‘divided’ or offering less of a model for later emperors. One reason for this lacuna may be that much of that discussion has occurred in scholarship in French (Feissel) and Italian (Porena), which is considered rather scantily here (as is scholarship in German). Secondly, in his analysis L. forgoes discussing the role of the Senate of Rome in this period, even if to some readers the fact that the empire remained undivided in this respect (the Senate in Constantinople only began to function as the seat of the senators of the East in the 350s) may speak against a division of empire in the 340s. There is also no engagement with messages of unity in imperial propaganda (coinage, the ‘twin-city’ motif, or panegyric, notably Lib., Or. 59, as well as public dedications). L. argues that this unity was only perceived or ideological (10), which is precisely why it matters to this study.

But perhaps the main issue is terminology. As the Introduction highlights, ‘division’ can cover a range of things. Hence L.’s criteria for a divided empire (10–11) usefully outline what needs to be considered, from territorial sovereignty, the independence and autonomy of rulers and governments to the separation of executive, administration, judiciary and military authorities. Equally, dual sovereignty, a sign of unity, is differentiated from divided rule, and the separation of empire during civil war (during the late Republican period) from consensual division. However, overall they are considered as different shades of ‘division’. A more stringent terminology would have furthered L.’s aim to ‘highlighting approaches and avenues of investigation to the question of identifying divisions and distinguishing between them’ (13).

Overall, however, L. has furnished an insightful and much-needed study of the administrative development of the western half of the empire under emperor Constans (and his brothers). May his calls for further exploration of these emperors and their ignominiously understudied reigns and policies not go unheard.