C. seeks to demonstrate that modern academic theories presenting Gallienus as a cavalry reformer are based on insufficient evidence. In doing so, he follows in the footsteps of a long-neglected article by M. Springer (‘Die angebliche Heeresreform des Kaisers Gallienus’, in: Krise – Krisenbewuβtsein – Krisenbewältigung [1988], pp. 97–100). But unlike Springer’s piece, C.’s book details the history of scholarship on this issue. This review focuses on the arguments that seem most pertinent to the question of whether we accept or reject the idea that Gallienus made changes to the cavalry, however we envision those changes.
Chapter 1 discusses scholarship on Gallienus in the eighteenth and nineteenth centuries, and Chapter 2 focuses on Hermann Schiller, who in 1883 became the first to suggest that Gallienus made a change to the cavalry. Schiller noted a passage of Cedrenus, which derives from the Byzantine historian Symeon Logothetes. According to Symeon (80.3), Gallienus ‘first established cavalry units (ἱππικὰ τάγματα). For the soldiers of the Romans were generally infantry’. C. suggests that Symeon misinterpreted an earlier account describing Gallienus’ efforts in favour of the equestrian order (ἱππικόν τάγμα), noting that Byzantine authors usually use καθίστημι to mean ‘ordain’/‘appoint’ to an office. This is an intriguing idea that should be carefully considered, but Symeon on three other occasions uses καθίστημι to mean ‘make’ or ‘establish as’ (42.17, 44.10, 49.2), and I find ‘ordained the equestrian order’ an awkward alternative.
Chapter 3 critiques the ideas of Emil Ritterling and Alfred von Domaszewski, who discussed this topic in the 1900s. In 1903 Ritterling noticed that Dalmatian cavalry are first securely attested and became a prominent military unit during the reign of Gallienus: they were participants in his assassination (e.g. Zos. 1.40.2), distinguished themselves in Claudius’ Gothic war (HA, Claud. 11.9; Zos. 1.43.2) and participated in Aurelian’s Palmyrene war (Zos. 1.52.3). The 48 units attested in the Notitia Dignitatum are likely relevant, but C. expresses reasonable reservations about Ritterling’s attempt to use the Notitia to make more specific conclusions about Gallienus’ cavalry.
Chapter 4 considers the ideas of Léon Homo, who in 1913 revealed the biases of Latin accounts against Gallienus and promoted an image of him as a reformer. C. rejects the Greek claim (Zos. 1.40.1; Zon. 12.25) that the general Aureolus was ‘commander of all the cavalry’ and was stationed in or had seized Milan when he rebelled against Gallienus in 268. C. notes that Aurelius Victor (HAb 33.17) instead has Aureolus commanding legions in Raetia, and he suggests that the Greek authors have duplicated Aureolus’ cavalry command at the battle of Mursa in 260, narrated by Zonaras (12.24). However, as regards this battle, Zonaras does not speak of ‘all the cavalry’ (Gallienus led the Mauri), and the Historia Augusta (Aur. 18.1) likewise claims that Aurelian, during Claudius’ reign and before becoming emperor, commanded all the cavalry. Neither Aureolus’ command over all the cavalry nor Aurelian’s are associated with a specific battle (cf. p. 152). Aureolus may have held this and the Raetian command at different times, leading to confusion in the sources, but it is possible that they overlapped, as Milan was located on one of the roads connecting Raetia and Italy, and Gallienus was absent, fighting the Goths in the Balkans (cf. Marcus Aurelius’ praetentura Italiae et Alpium [ILS 8977]; L. de Blois, Image and Reality of Roman Imperial Power in the Third Century ad [2019], p. 80). For overlapping commands in the third century, note for example Domitius Philippus (praefectus uigilum, σρατηλάτης), Octavianus (dux per Africam Numidiam Mauretaniamque), Placidianus (who as praefectus uigilum and seemingly then as praetorian prefect commanded vexillations and cavalry in Narbonensis) and Marcellinus (praefectus Mesopotamiae, rector Orientis).
Chapter 5 critiques the ideas of Robert Grosse, who in 1920 credited Gallienus with a variety of military reforms in addition to reorganising the cavalry. C. rejects Aurelius Victor’s claim (HAb 33.33–4, 37.6) that Gallienus excluded senators from military commands. However, while probably there was no edict to this effect, C. Davenport has shown that in the 260s Gallienus ceased to appoint tribuni laticlauii and legati legionis, preferring to appoint equestrians as praefecti legionis and as the commanders of expeditionary forces (The History of the Equestrian Order [2019], pp. 534–45). Against this, C. notes Aurelius Marcianus, a stator legati legionis in 270 (CIL III 8117). However, the fact that Marcianus had been serving since 244/5 problematises the chronological significance of the title, and one may also ask whether we should expect to see an attendant designated as ‘of the praefectus legionis’, since attendants for the Parthica legions, always led by praefecti, are regardless designated as ‘of the legatus legionis’ (e.g. AE 1971.469, 1993.1586–1587; I thank C. Davenport for pointing this out to me).
C. also argues against the idea that the title protector Aug(g) was created under Gallienus. The title is first securely attested during the reign of Valerian and Gallienus, and there is considerable epigraphic testimony for its use during Gallienus’ reign (Davenport, pp. 533–4, 538–9). C. notes that CIL XI 4082 may date to the reign of Gordian III, based on the description of the unknown emperor as sanctissimus and inuictus. Although Gallienus also received those titles, their combination indeed better suits Gordian, Decius and Gallus, which may support the Byzantine claim (Georgios Monachus 461; Cedrenus 276 [ed. Tartaglia]) that ‘Iunior’ (Gordian?) formed the protectores (M. Emion, Les protectores Augusti [III e–VI e s. p.C.] [2023], pp. 42–51, adduces further possible evidence for a Gordianic origin, including CIL VI 1645). C. suggests that a protector Auggg (CIL III 3126) dates to 209–211, when there were three Severan Augusti, but Gallienus’ sons, although Caesares, seemingly could be included in Auggg formulations (CIL VIII 2634). Although C. notes other protectores possibly or certainly dating to the Severan period, it is uncertain whether protectores Augusti, rather than protectores of governors and prefects, can be dated this early.
Chapter 6 contends against Andreas Alföldi’s argument, published in 1927, that Aureolus, during his rebellion against Gallienus in 268, minted coins in Milan for the ‘Gallic’ emperor Postumus. Crucially, these coins, using multiple legends, appear to have honoured the cavalry under Aureolus’ command. The identification of Milan derives from the M that appears alongside officina marks on issues from the same mint struck under Gallienus, and the interpretation of Aureolus as a defector may explain why there are no authentic coins minted in his name and why Zonaras (12.24) claims that he was earlier unwilling to pursue Postumus. C. instead argues that the coins were minted in Raetia when Postumus temporarily held the province, either between 261 (Gallienus’ fourth consulship) and 262/3 (Gallienus’ decennalia), or after 262/3. The latter possibility is undermined by the fact that the crisis years of 260–261 provided the best opportunity for Raetia to have joined Postumus. Indeed, sources suggest that Postumus was on the defensive after this time (Petr. Patr. De Sententiis 165; Zon. 12.24). But the brief window of 261–262/3 is equally difficult to accept, since the sequence of erasures on the Augsburg Altar indicates at least two phases of ‘Gallic’ occupation, with Genialis and Honoratianus first in favour and then in disgrace (K. Dietz, ‘Zum Kampf zwischen Gallienus und Postumus’, in: T. Fischer [ed.], Die Krise des 3. Jahrhunderts n. Chr. und das Gallische Sonderreich [2012], pp. 29–62). The possibility that Valerian, who in 253 usurped power in Raetia and marched into Italy, produced coins at this mint before making Gallienus his colleague (p. 104; R.A.G. Carson, ‘Mints in the Mid-Third Century’, in: R.A.G. Carson and C.M. Kraay [edd.], Scripta nummaria Romana: Essays Presented to Humphrey Sutherland [1978], pp. 65–74) suits both Milan and Raetia as potential locations.
Chapters 7 and 8 critique the hypotheses that followed in Alföldi’s wake concerning a cavalry army, increased cavalry numbers or the development of mobile units consisting of both cavalry and infantry. Chapter 9 suggests that the third-century Dalmatian cavalry were the equites singulares Augusti, which was a small unit (pp. 148–9), but this small size sits uneasily with the Dalmatians killing 3,000 Goths in an engagement (Zos. 1.43.2).
In this reviewer’s opinion we should consider the indications that various cavalry commanders in this period eventually became emperors or usurpers. In addition to Aureolus and Aurelian, Claudius was said to have been a cavalry commander at some point (Zon. 12.26), and according to the Epitome de Caesaribus (36.2) Probus was named Equitius, seemingly a nickname to which coinage appears to allude (cf. p. 93 n. 20); the Epitome is precise on nomenclature (J.A. Stover and G. Woudhuysen, The Lost History of Sextus Aurelius Victor [2023], pp. 124–6). It may be relevant that Diocletian was a Dalmatian career soldier. Moreover, cavalry ducenarii are first attested in 269 (CIL XII 2228; Davenport, pp. 540–1), and R. Hedlund (“… Achieved Nothing Worthy of Memory” [2008], pp. 108–13) has argued that numismatic legends and iconography attest to periods in the late third century when cavalry was especially important.
C. laments that neither Springer’s paper nor his own previous articles on this topic have provoked discussion (p. 130). This is unfortunate, for while it seems to me that Gallienus did make changes to the cavalry, this iconoclastic monograph demands engagement. I have focused on several of its key arguments, but C. repeatedly draws attention to the uncritical use of previous scholarship, imprecise and insufficient argumentation, and overly imaginative reconstructions of other scholars. Researchers on the third century should be aware of these issues in the scholarly record. C. reminds us that much of what we think we know about the third century rests on slender evidence, and that new information could lead to major revisions.
