Skip to main content Accessibility help

Ammianus and some tribuni scholarum palatinarum c. A.D. 53–364

  • David Woods (a1)


The Res Gestae of Ammianus Marcellinus is a major source of our knowledge of the late Roman army. However, although himself a former army officer, it was not the intention of Ammianus to explain the institutions and organization of the late Roman army to his readers. We learn about these only from the incidental pieces of information which are scattered throughout his text. It was not his intention either to present us with the regimental histories of any individual units, yet repeated references to the more prominent and prestigious units were inevitable in a history such as his. This was particularly true in the case of the scholae palatinae because of their role as the bodyguard units of the emperors. It is my intention here to draw together such information as Ammianus provides about the scholae in order to demonstrate how, in a number of cases in particular, it is possible to reconstruct an almost complete list of their commanders for the period c. 353–364



Hide All

1 No treatment of any aspect of the late Roman army can fail to utilize this work as a primary source. More detailed treatments of Ammianus′ role as a military historian include Crump, G. A., Ammianus Marcellinus as a Military Historian (Historia, Einzelschriften 27; Wiesbaden, 1975), and Austin, N. J. E., Ammianus on Warfare: An Investigation into Ammianus′ Military Knowledge (Collection Latomus 165; Brussels, 1979). See also Matthews, J. F., The Roman Empire of Ammianus (London, 1989), pp. 279303. The most comprehensive treatment of the late Roman army remains that by D. Hoffmann, Das Spdtromische Bewegungsheer unddie Notitia Dignitatum (Epigraphische Studien 7; 2 vols, Diisseldorf, 19691970) (DSB henceforth). More recent general works includeSouthern, P. and Dixon, K. R., The Late Roman Army (London, 1996), and Elton, H., Warfare in Roman Europe AD. 350425 (Oxford, 1996).

2 On the origin and career of Ammianus, see Thompson, E. A., The Historical Work of Ammianus Marcellinus (Cambridge, 1947), pp. 119; Matthews, op. cit. (n. 1), pp. 6780. Contra Matthews see the review of this work by Bowersock, G. W., JRS 80 (1990), 244250, esp. 247, esp. 57–61;Fornara, C. W., ‘Studies in Ammianus Marcellinus, I: The Letter of Libanius and Ammianus Connections with Antioch’, Historia 41 (1992), 328–44. In reply again, see Matthews, J. F., ‘The Origin of Ammianus’, CQ 44 (1994), 252269.

3 For a description of the early history and organization of the scholae palatinae, see Frank, R. I., Scholae Palatinae: The Palace Guards of the Later Roman Empire (Papers and Monographs of the American Academy in Rome 23: Rome, 1969); also, more authoritatively, Hoffmann, DSB I (n. 1), pp. 279–303, and DSB II, pp. 117–128.

4 On the dating of this document see most recently Siebt, W., ‘Wurde die “Notitia Dignitatum” 408 von Stilicho in Auftrag gegeben?’, MIOEG 90 (1982), 339–346; also, Mann, J. C., ‘The Notitia Dignitatum–Dating and Survival’, Britannia 22 (1991), 215219.

5 ND Or. 11.4–10; ND Oc 9.4–8.See Woods, D., ‘The Scholae Palatinae and the Notitia Dignitatum, JRMES 7 (1996), forthcoming, for a fuller account of how individual scholae were transferred between the two halves of the empire at the turn of the fourth and fifth centuries, and how their total number rose from ten to thirteen during the same period. On the role of the magister officiorum see Clauss, M., Der Magister Officiorum in der Spdtantike (4.–6. Jahrhundert): Das Amt undsein Einfluss aus der kaiserliche Politik (Vestigia 32: Munich, 1980), pp. 4045.

6 So the existence of the eastern schola scutariorum prima (ND Or. 11.4) and its western homonym (ND Oc 9.4) point to the division of a single original schola scutariorum prima, while the existence of the eastern schola scutariorum secunda (ND Or. 11.5) and its western homonym (ND Oc 9.5) point to the division of a schola scutariorum secunda. In the other cases the products of these divisions seem to have been distinguished by the addition of the titles seniores or iuniores. So the eastern schola armaturarum iuniorum (ND Or. 11.9) and the western schola armaturarum seniorum (ND Oc 9.6) point to the division of an original schola armaturarum, while the eastern schola gentilium iuniorum (ND Or. 11.10) and the western schola gentilium seniorum (ND Oc 9.7) point to the division of an original schola gentilium. The anomalous eastern schola gentilium seniorum (ND Or. 11.6) represents a transfer for a period to the east of the western schola gentilium seniorum. The original schola scutariorum clibanariorum remained undivided in the east (NDOr. 11.8).

7 See Tomlin, R. S. O., ‘Seniores-iuniores in the Late-Roman Field Army’, AJPh 93 (1972),253278; Hoffmann, DSB I (n. 1), pp. 117–30. A more recent discovery reveals that undue importance has been attached to the division of the army between Valens and Valentinian I in 364, and that some seniores-iuniores pairs of units probably existed before this date. See Drew-Bear, T., ‘A Fourth-century Latin Soldier′s Epitaph at Nakolea’, HSCPh 81 (1977), 257–74. This does not prove, though, that all seniores-iuniores pairs must have existed before 364, and in the case of the scholae palatinae at least the evidence seems as compelling as ever that it was that particular year which saw their division.

8 On the origin of the schola scutariorum sagittariorum (ND Or. 11.7), see Woods, D., ‘Subarmachius, Bacurius, and the Schola Scutariorum Sagittariorum’, CPh 91 (1996), 365371.

9 In the absence of evidence to the contrary, I assume that when Ammianus refers to a tribunus scutariorum he means by this a tribune of either the schola scutariorum prima or the schola scutariorum secunda. Likewise, whenever I mention the scholae scutariorum as such, I refer to these last two units in particular, excluding the schola scutariorum clibanariorum.

10 Zos. HN2.50.2–3; Paschoud, F. (ed.), Zosime Histoire Nouvelle (Paris, 1971), p. 258

11 See Barnes, T. D., ‘Structure and Chronology in Ammianus, Book 14’, HSCPh 92 (1989), 413422.

12 On Julian′s scutarii: Amm. 16.4.1 (356); 20.4.3 (360).

13 Amm. 20.2.5: ‘Agilone ad eius locum immodico saltu promoto, ex gentilium et scutariorum tribuno’. The reading and interpretation of the text here is much disputed. See Hoffmann, DSBI (n. 1), p. 294, and DSB II, pp. 122–23, n. 850; also den J. Boeft, den D. Hengst, H.C. Teitler, Philological and Historical Commentary on Ammianus Marcellinus XX (Groningen, 1987), p. 21. The problem centres on the introduction of the word et between gentilium and scutariorum, an amendment to the manuscript reading. Three different interpretations are possible depending whether one retains this et. If one retains et, then it is possible to interpret this phrase in reference to the successive command of these scholae, i.e. command of the schola gentilium followed by command of a schola scutariorum, as I do here, or in reference to the simultaneous command of both, as Elton, op. cit. (n. 1), p. 91, n. 7, tentatively suggests. Those who reject the et prefer to interpret the tribunus gentilium scutariorum as that officer more usually described simply as the tribunus gentilium.

14 On Julian′s gentiles: Amm. 16.4.1 (356); 20.4.3 (360).

15 For a detailed discussion of the circumstances of this revolt, see Drinkwater, J. F., ‘The “Pagan Underground”, Constantius IPs “Secret Service”, and the Survival, and the Usurpation of Julian the Apostate’, in Deroux, C. (ed.), Studies in Latin Literature and Roman History III (Collection Latomus 180; Brussels, 1983), pp. 348387.

16 Amm. 20.4.10,4.16, 8.7; Lib. Or. 18.95

17 See J. den Boeft et al., op. cit. (n. 13), p. 20, where this remark is also explained as a result of Ammianus′ dislike of Alamanns and his admiration of Agilo′s predecessor Ursicinus. All three factors were probably at play.

18 The magister equitum Iovinus was originally charged with reducing Aquileia to surrender (Amm. 21.12.2). He may not have learned of the death of Constantius until his assignment elsewhere, but the order for his replacement by the comes Immo came from Julian in Constantinople after the death of Constantius (Amm. 21.12.3).

19 Ammianus′ description of Agilo seems to hint that his name had a certain news value at that particular point in time such as might well have been earned by his defection from Julian to Constantius (Amm. 21.12.16): ‘Agilonem magistrum peditum ea tempestate probe cognitum’.

20 Pace Bowersock, G. W., Julian the Apostate (London, 1978), p. 68, who claims that Agilo ‘was a person who accepted with equanimity the fact of Julian′s elevation’.

21 This disappearance has been commented upon, for example, by Kaegi W E., ‘Domestic Military Problems of Julian the Apostate’, Byzantinische Forschungen 2 (1967), 247–264, esp. 254, where displeasure at his conduct during the commission of Chalcedon, or fear of his growing power, are mooted as possible reasons for his dismissal by Julian. In general on the trials at Chalcedon, see Thompson, op. cit. (n. 2), pp. 73–79.

22 On the growing resistance to Julian′s religious policies among the palatine regiments, and on Ammianus′ concealment of this problem, see Woods, D., ‘Ammianus Marcellinus and the Deaths of Bonosus and Maximilianus’, Hagiographica 2 (1995), 25–55; idem, ‘Valens, Valentinian I and; the Ioviani Cornut’, in Deroux, C. (ed.), Studies in Latin Literature and Roman History VIII (Collection Latomus: Brussels, 1997), forthcoming.

23 Theod. HE 3.12–3; Soz. HE5M; Greg. Naz. Or. 4.82–4. On the general background to this incident, see Gleason, M., ‘Festive Satire: Julian′s Misopogon and the New Year at Antioch’, JRS 76(1986), 106119;

24 See Boeft, den J., Drijvers, J. W., Hengst, den D., Teitler, H.C, Philological and Historical Commentary on Ammianus Marcellinus XXII (Groningen, 1995), p. 201; Woods, D., “The Final Commission of Artemius the Former Dux Aegypti”, forthcoming.

25 See Woods, D., ‘The Martyrdom of the Priest Basil of Ancyra’, Vigiliae Christianae 46 (1992), 3139; also, for an opposed view, Teitler, H. C., ‘History and Hagiography: the Passio of Basil of Ancyra as a Historical Source’, Vigiliae Christianae 50 (1996), 7380.I will only say here that we should not require of the author or editor of the surviving passio standards any difFerent to those of the surviving historians of late antiquity. Fictitious speeches or miracles no more invitiate the passio of Basil than do the speeches and omens the Res Gestae of Ammianus.

26 CTh. 6.13.1 (413); PLREII, p. 582, Jacobus 2 (431).

27 In Woods, art. cit. (n. 25), 35–6,1 was unwisely tempted to identify the comes scutariorum Frumentinus with the comes domesticorum Dagalaifus. In doing so I placed too much emphasis, firstly upon the occurrence of the title comes, and secondly upon that phenomenon of late antique literature where an individual is preserved by one name in one source, but by another in a different source (e.g. Gintonius/Sintula of Jul. Ep. adAth. 282d/Amm. 20.4.3).

28 See Warmington, B. H., ‘The Career of Romanus, Comes Africae’, Byzantinische Zeitschrift 49 (1956), 5564, esp. 63.

29 Victores (Amm. 24.4.23, 25.6.3); Tertiaci (Amm. 25.1.7); Zianni (Amm. 25.1.19); loviani(Amm. 25.5.8, 6.2); Herculiani (Amm. 25.6.2); Iovii (Amm. 25.6.3).

30 The relationship between the similar but often divergent accounts in Ammianus and Zosimus of Julian′s campaign has been discussed in detail, most recently by C. W Fornara, ‘Julian′s Persian Expedition in Ammianus and Zosimus’, JHS 111 (1991), 1–15.

31 On the depiction of Julian as the ideal emperor, see Blockley, R. C., Ammianus Marcellinus:A Study of His Historiography and Political Thought (Collection Latomus 141: Brussels, 1975), pp.73103; also, Elliott, T. G., Ammianus Marcellinus and Fourth Century History (Florida, 1983), pp.69134.

32 Libanius agrees that Julian wore no armour when he was attacked, claiming also that he wasattended by only one bodyguard (Or. 18.269). Elsewhere, however, he tacitly admits the presence of more than one bodyguard because this suits his argument at that time: that the wounding of Julian alone proved that he had been deliberately targeted in an assassination attempt (Or. 24.17). The latter admission that Julian had in fact been accompanied by more than one bodyguard, which agrees also with Ammianus′ claim that his candidati had been momentarily separated fromJulian (Amm. 25.3.6), reinforces one′s suspicion that more people had actually accompaniedJulian on his last ride than either author wishes us to realize.

33 See Liebeschuetz, J. H. W. G., ‘Ammianus, Julian, and Divination’, in Wisseman, M. (ed.),Roma Renascens: Beitrage zur Spatantike und Rezeptionsgeschichte. Ilona Opelt von Freunden undSchulern zum 9.7.1988 gewidmet (Frankfurt, 1988), pp. 198213.

34 Against two tribunes: Amm. 24.3.2; against five: Amm. 25.1.8–9.

35 For a more detailed discussion of this matter, see Woods, D., ‘A Note Concerning the Early Career of Valentinian I’, Ancient Society 26 (1995), 273288.

36 Sulp. Sev. Vit. Mart. 2.2. For an excellent discussion of the circumstances and timing ofMartin′s resignation, see Stancliffe, C., St Martin and his Hagiographer (Oxford, 1983), pp.134–48.

37 Martin was from Sabaria in Pannonia Prima (Sulp. Sev. Vit. Mart. 2.1); Valentinian wasfrom Cibalae in Pannonia Secunda (Amm. 30.7.2).

38 In addition to the tribuni scutariorum whom I have already mentioned, Ammianus namesalso Barzimeres (Amm. 30.1.11), and Cassio (Amm. 31.12.16), tribunes in 374–377 and 378 respectively.

39 I disagree here with PLREI, p. 200, where it is said of Charietto at the time of this missionthat ‘he evidently held no official rank’, apparently because Ammianus refers to Charietto simplyas a vir with no indication of rank (Amm. 17.10.5). Yet Ammianus often uses the term vir to describe various officers where a more precise term of rank would be much more welcome to themodern reader. He can avoid unclassical technical terms and still denote a more senior commandrelatively easily (e.g. Amm. 25.1.19), but such circumlocutions would have been much more difficult in the case of junior officers.

40 Jon Chrys. In Iuventinum el Maximinum (PG 50, 571–578); Theod. HE 3.15. On the date, see Peeters, P, ‘La date de la fete des saints Juventin et Maximin’, Analecta Bollandiana 42 (1924), 7782.

41 Jeffreys E. et al., The Chronicle of John Malalas (Byzantina Australiensia 4; Melbourne, 1986), 178. Malalas records local Antiochene tradition, a testimony that deserves seriousconsideration, therefore.

42 See I.van de, Gheyn, ‘Passio Antiquior S$. Sergii et Bacchi Graece nunc primum edita’, Analecta Bollandiana 14 (1895), 373395. An English translation is now available in Boswell, J., The Marriage of Likeness: Same-sex Unions in Pre-modern Europe (London, 1995), pp. 375390, although the schola gentilium was not a school in the modern educational sense as Boswellassumes, p. 377. That the passion of Sergius and Bacchus relates more to the era of Julian than ofGalerius Maximianus was noted by Franchi, P. de′ Cavalieri, ‘Dei ss. Gioventino e Massimino’, inhis Note agiograflche 9 (Studi e Testi 175: Rome, 1953), 169–200, esp. 194–199, although he arguedthat it was derived from a lost account of the martyrs Iuventinus and Maximinus rather than anindependent account of two otherwise unknown confessors. See now Woods, D, ‘The EmperorJulian and the Passion of Sergius and Bacchus’, JECS 6 (1998), forthcoming.

43 Amm. 24.1.2, 7.2; 25.5.2, 7.7.

44 Zos. HN 3.13.3; 4.2.4. In general, see Ridley, R. T., ‘The Fourth and Fifth Century MilitaryHierarchy in Zosimus’, Byzantion 40 (1970), 91104. However, I remain unconvinced by Ridley, R. T., Zosimus′ New History: A Translation with Commentary (Byzantina Australiensia 2:Canberra, 1982) where the of the above passages is rendered as magister.

45 Victor as dux: Amm. 24.4.13, 6.13; Victor as comes: Amm. 24.4.31, 6.4. It is arguable thatVictor was comes of two auxilia palatina, the Iovii and the Victores. See Woods, D., ‘The Role ofthe Comes Lucillianus during Julian′s Persian Expedition’, L′Antiquite Classique 67 (1998), forthcoming.

46 In addition to the forces which were requested of him in early 360, Julian had alreadysupplied Constantius with seven infantry and two cavalry units (Jul. Ep. adAth. 280d).

47 Jul. Ep. ad Ath. 285b: Lib. Or. 18.102:.

48 Very little is known about Helena′s movements. However, she had returned from Julian in Gaul to the court of her brother Constantius on one occasion at least (Amm. 16.10.18), and a mote prolonged visit is not out of the question. In general, seeAujoulat, N., ‘Eusebie, Helene et Julien I: Le temoignage de Julien’, Byzantion 53 (1983), 78103, and ‘II: Le temoignage des historiens’, 421422.

49 See Woods, D., ‘The Fate of the Magister Equitum Marcellus’, CQ 45 (1995), 266268.

50 On his Christianity: Theod. HE4.33.3; Bas. Ep.152–153; Greg. Naz. Ep. 133–134.

51 In general, see Maio, di M., ‘The Transfer of the Remains of the Emperor Julian from Tarsus to Constantinople’, Byzantion 48 (1978), 4350.

52 Ammianus treats the burial of Julian as if it were of interest solely because of the presence there of the future usurper Procopius (Amm. 25.9.12–13). Yet the two brief notices do not necessarily contradict one another, as Merobaudes was present in an official military capacity only, while Procopius was a relative of the deceased, present in a personal capacity.

53 C.Th. 14.17.9 (389); ND Or. 11.8.

54 See Eadie, J. W., ‘The Development of Roman Mailed Cavalry’, JRS 57 (1967), 161173; also, Hoffmann, DSBI(n. 1), pp. 265–278, and DSBII, pp. 110–117.

55 The equiles Persae clibanarii (ND Or. 6.32), and the comites [Persae] clibanarii (ND Or. 5.29). The full title of the latter has to be restored from hagiographical sources, an anonymous Life of Constantine (BHG 364), and the Passio Eusignii (BHG 639). See Woods, D., ‘Constantine I and the Persocomites’, Maynooth Review 15 (1997), forthcoming.

56 On his defection, see Zos. HN 2.27; Zon. 13.5. The latter describes his command of cavalry under Constantius II:

57 Not once does Ammianus refer to Hormisdas by a Roman title, despite frequent referencesto his exploits during the Persian campaign (Amm. 24.1.2, 1.8, 2.4, 2.11, 2.20, 5.4). The onlysource for his rank as comes is the Passio SS. Bonosi et Maximiliani (BHL 1427) which is probablyto be trusted in this matter. On the value of this text, see Woods, ‘Ammianus Marcellinus and theDeaths of Bonosus and Maximilianus’, art. cit. (n. 22). Zosimus describes Hormisdas (HN 3.11.3) or (HN 3.13.3), which confirms that Hormisdaswas a cavalry commander, but does not allow us to distinguish whether his correct Latin title wasdux, comes, or magister even.

58 Amm. 16.10.8–16. In general, see Cameron, A., ‘Biondo′s Ammianus: Constantius andHormisdas at Rome’, HSCPh 92 (1989), 423436.

59 Since Hormisdas′ last recorded appearance was at that same battle which saw the death ofthe magister officiorum Anatolius, and the lucky escape of the praetorian prefect Salutius, onesuspects that he himself was a victim of this same disaster (Zos. HN 3.29.2).


Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed