1. In some cases I have left sources unmentioned in the footnote because their information (with a reference to the source) has been included in the secondary source mentioned in the footnote.
1. See Syvänne MHLR vols.1–2.
2. Syvänne, Caracalla, 35ff.; with Syvänne, Gallienus (still forthcoming at the time of writing).
3. Syvänne, 2011; Syvänne, Gallienus.
4. Greco-Roman military theory had several names for the duplex acies depending on which direction the phalanxes faced. The diphalangia amphistomos had two phalanxes facing opposite directions. The diphalangia antistomos had two phalanxes facing each other. The diphalangia homoistomos had two phalanxes facing the same direction. The diphalangia heterostomos had two phalanxes facing different directions. As is obvious, the terms allow different interpretations and have also been differently interpreted in the treatises, but it is not necessary to analyse these in this case because the main point here is that the two phalanxes could be used to face different directions as needed.
5. The following discussion is based on: Zahariade (2011), D’Amato (2009b), Rankov, Reddé (highly recommended), Pitassi (2009, 2011), Bounegru and Zahariade (Danubian Fleets), Starr, and Syvänne (2004, 2013–14, MHLR vol.1, Caracalla).
6. Commodus is credited with the founding of the fleets Classis Nova Libyca, and Classis Africana to protect the grain shipments from Egypt, but some historians suspect their existence. The existence of the Classis Mauretanica is also contested because some historians believe these ships to have been detachments drawn from other fleets.
7. Research paper/presentation held at Norfolk in 2014; Herodian 2.14. 6–7; Starr (1943 with 1960); Paulus in Digest 49.18.5 (Levick 2, p.75).
8. This is based on Syvänne, 2004 and my presentations 2013–14 with MHLR Vol.1; Pitassi.
9. A fuller discussion with sources in Syvänne, 2016 and Caracalla.
10. Syvänne, 2011; Gallienus.
1. Sources collected in PLRE1, Aurelianus 6; Cizek, 11–12.
2. Sources, analysis and comments in: the PLRE1 Aurelianus 6, Cizek, 12–13.
3. HA Aurel. 5.3ff., 6.1ff., 7.3ff.,
4. HA Aurel. 8.1ff.
5. No such legion is known, but since the evidence for the third century is so poor one cannot exclude the possibility that such a legion existed. As tribune Aurelian would have been the acting commander of the legion (i.e. praepositus). He would have been the tribunus maior of Modestus and Vegetius. See Appendix 1. It is also likely that Aurelian would have been a member of the imperial bodyguards when assigned to such command.
6. Magie (HA 3, p.214) notes that none of the five third legions is known to have possessed the cognomen Felix, but it is nevertheless entirely plausible that such a legion existed. One of the existing legions could have obtained this title or one of the newly recruited third century legions (we do not know the names of all of these) could have held the title.
7. It is more probable that the reconnaissance was conducted with cavalry which could flee if necessary.
8. Maximinus Thrax is claimed to have held the same position just before he murdered Alexander Severus in 235.
9. HA Aurelian 8–15. The author claims that Aurelian acted as deputy (vicar) for generals (dukes i.e. duces) and tribunes on about forty occasions and in addition held very many commands as a general and tribune. The HA also claims that Aurelian saw service both in Gaul and Illyricum. In the latter case, he is supposed to have acted as a deputy to Ulpius Crinitus dux of Thrace and Illyricum during his medical leave. In addition, the HA contains a letter supposedly sent by Valerian to Gallus in which he states that he had left his (grand)son (Saloninus) Gallienus to the care of Postumus, because he considered Aurelian much too stern and harsh and therefore not suited to accompany the merry making Gallienus. If there is any truth in this, this sequence of events would suggest that Valerian had left Postumus with Gallienus to act as his advisor. It would also suggest that Valerian appointed Aurelian as a deputy general of Crinitus in 255 or 256. During Crinitus’s absence in 255 Aurelian distinguished himself with the result that when Valerian returned to Byzantium in 255, he ordered Crinitus to adopt Aurelian as his son and appointed Aurelian to the positions of consul and inspector of all army camps, the latter being the same position that Maximinus Thrax is also supposed to have held as a reward and from which position he rose to become the emperor after the murder of Alexander Severus. White (50ff.) also believes that Crinitus was a real person, but dates the events differently.
10. When the consules ordinarii were asked to resign after six months by the emperor, he replaced them with the consules suffecti. This was a means which the emperors used to bribe people with the consular honour.
11. Full argumentation for this and the offices can be found in Syvänne, MHLR vol.1 and Gallienus.
12. Magie in his edition of the HA (Probus 3, Loeb ed. p.341, n. 4) suggests that the name of the sister is fictitious on the basis of the name or that she was a half-sister. Considering the period practice of taking new names when assuming imperial office, I would not preclude the possibility that she was a full sister, but it is not impossible that she could have been a half-sister. After all, it was not uncommon for women to die when giving birth to a child.
13. Sources collected in the PLRE1, Delmatius 1, Probus 3.
14. Magie (HA, p.342, n.1) notes in his edition/translation that it is possible that Artabassis Syrus is an error for the Armenian Artavasdes, but in my opinion this is unlikely on the grounds that Artavasdes/Artavazd the Mandakuni or Mamikonean fled to the Roman territory only in about 260/1. See my arguments in MHLR Vol.1, Gallienus and Syvänne & Maksymiuk, 2018. On the basis of Kerdir’s inscription (the so-called KKZ) it is clear that Shapur I needed to conquer Armenia, Iberia/Georgia and Albania in about 260. This is also confirmed by the HA (Val. 1.1.ff.), which lists these areas as independent from Persia. My educated guess is that Khosrov the Great had been murdered in about 256/7 by Persian special operatives after which the new regent Artavazd allied himself with Shapur I only to change his alliance back to the Roman side by about 259.
15. HA, Probus 5.1ff.
16. HA, Probus 6.1ff. Syvänne, Gallienus.
17. The following chapter on enemies is based on Syvänne MHLR Vol.1 and Caracalla.
18. The Georgian Chronicles (pp.80–1 see with Syvänne MHLR Vol.1) specifically claim that the Khazars (anachronistic term which either means some nomadic group or the Alans) were in the habit of attacking the Derbend Pass at this time, which may suggest some Roman diplomatic efforts in the area against the Persians.
19. The evidence is collected in Banchich (Zonaras 12.25, pp.114–16), Bray (239ff.), Cizek (22–3) and Geiger (173ff.). The following account is based on these and Syvänne, Gallienus.
1. In general for the reign of Claudius see Syvänne Gallienus and Southern (2001, 108–10). Note, however, that my reconstruction of Claudius’ reign differs from all other accounts because I date the events differently. The account provided here is also slightly different from the one that I give in Gallienus.
2. Based on my biography of Gallienus.
3. The HA Claud. 4.3 includes a letter to the Senate. The letter is probably fabricated, but it is still clear that Claudius initially sent only a letter and his henchmen to Rome to secure the backing of the Senate in a situation in which he still needed to defeat Aureolus.
4. Zonaras 2.25, HA (Gall., esp. 14–15); Aur. Vict. 33. See the summary of the various later versions in Zonaras/Banchich pp.114–16.
5. A complete analysis of the sources and the campaigns are offered by Syvänne, Gallienus. In other words, I go even further in the reconciliation of Gallienus than Alföldi.
6. Aurelius Victor 33, Eutropius 9.9; HA Tr. 3–8, 24; PLRE1 for other sources. For other interpretations, see Potter (263–4) and Drinkwater (1987, esp. 23–36, 148ff.) with Syvänne, Gallienus.
7. Zonaras (12.26) has preserved us an incident which took place when Claudius was in Rome. It was then that Claudius gave the order that nobody was entitled to receive the property of another as a gift from the emperor. As a result of this, an old woman approached the emperor and noted that Gallienus had confiscated her property and had given it as a present to Claudius. Claudius duly returned her property. The incident proves two things: Claudius was not incorruptible; Claudius had some sense of justice left in him and it is quite possible that he sought to correct the abuse of power when he was in the position to do so.
8. Potter, 265.
9. HA Claud.; Aurel.; Probus esp. 5.5ff. ; Syvänne, MHLR vol.1 (165–7, 179–81, 189), Caracalla (Index under Aulici), and Gallienus.
10. If there is any truth to this story, one could argue that it was actually this Claudius who stopped the purge of Gallienus’s relatives and friends and not the real Claudius! However, it is still more likely that it was the real Claudius who did this because the soldiers demanded it. On the basis of this, it would also be possible to argue that the HA has dated the usurper to the wrong reign so that Censorinus and Urbanus would have been usurpers during Aurelian’s reign in 271. Censorinus’s use of the name Claudius (Claudus) could be seen as a gesture to seek support from those who had supported Claudius and Quintillus.
11. The claimed career of Censorinus in the HA (TR 33) is of interest because it gives a possibility of making one additional speculation, which is to suggest a possible collusion with Aurelian. It claims that Censorinus had not only served as praetorian prefect twice, but also that he had held the post of extraordinary envoy to the Persians. When we remember that Aurelian had also served as an envoy (Aurel. 5.5–6) to the Persians, it becomes possible to think that these two men knew each other and could have cooperated against Claudius. The likeliest date for such a collusion is the year 270 when the sources make it clear that Aurelian was not doing anything worthy of note against the invaders in the Balkans. It should be noted, however, that this is mere speculation based on the uncertain evidence presented by the very unreliable source.
12. The letter from Claudius to Aurelian included in the HA (Aurel. 16.1–17.4, letter in 17.1–4).
13. Dexippus fr. 6; See the analysis in Drinkwater, 2007, 73–5.
14. Capture of Narbonensis and Spain in Drinkwater, 1987, 36, 120. The name Gratianopolis is later and dates from the period of the reign of Gratian in the fourth century.
15. HA Tr. 6; Aurelius Victor 33.2.
16. For a full analysis, see Syvänne, Gallienus.
17. On the basis of earlier research, which in its turn is based on the Arabic traditions preserved by al-Tabari, Al-Yaqubi. Ibn al-Athir amd al-Nuwairi, Stoneman (156–7) suggests the possibility that Zenobia (Zabba/Zebba) could have used a ruse to kill the sheik of the Tanukh Arabs called Jadhima in order to gain control of the desert region up to al-Hira. The whole account is completely ludicrious and is solely based on the fact that Zabba/Zebba of Palmyra is the Arabic version of Zenobia. It has long been recognized that the story of Zabba/Zebba of Palmyra contains fabulous legends, but several previous researchers have still quite mistakenly placed the events mentioned in this tradition occurred during the reign of the famous Zenobia. The account in question is preserved in detail in al-Tabari (i.750–71), but for some unknown reason the previous researchers who accept the above have failed to take into account the timeline given by al-Tabari. He states (i.768–71) in no uncertain terms that Zebba and Jadhima both lived just before the rise of the Sasanians under Ardashir I and Shapur I. In other words, this Zenobia/Zebba of Palmyra lived during the period before the year 224/226.
18. Details of the minting of various coins here and below are referred to in Watson, 221–3.
19. Legio VI Ferrata was posted at Legio-Kefar Otnay (Caparcotna), but it disappears from the sources in ca. 260 and it is possible that it was destroyed together with Valerian’s army in ca. 259/60 (or during the usurpation of the Macriani, my comment). See Farnum, 20. However, in my opinion, it is possible that a new legion was raised to replace it during the 260s which could have been one of the newly raised third century legions which we know nothing about unless it is included in the Notitia Dignitatum (turn of the fifth century).
20. HA Claud. 11.1; Zosimus 1.44; Malalas 12.28; Watson, 60–4; Southern, 2008, 106–7.
21. For the trade and trade routes, see Southern (2008, 24–33); Stoneman, 31ff., 53.
22. This is my interpretation for what happened.
23. The terms I have used come from Greek/Byzantine military theory and their inclusion in this discussion is entirely my own. For other analyses of the Palmyran military, see Southern (2008, 24–6), Nicolle (22–41). Nicolle considers the role of the mounted archers to have been of greater importance than that of the cataphracts, but I disagree because the sources make it quite clear that most of the combat cavalry consisted of these and that they were the most fearsome force wielded by the Palmyran state.
24. Stoneman, 121ff. with indexes on the men named.
25. Stoneman, 129ff.
26. Stoneman (146–51) collects the evidence. He does not accept the traditional interpretation, but may be correct in stating that the appeal to Aurelian was already done in 270.
27. Stoneman, 123ff.
28. See e.g. PLRE1 Probus 3, Probus 5.
29. For a fuller analysis of the Moorish tribes and their ways of fighting, see Syvänne MHLR vol.1.
30. The official marriage status was probably that of concubine/mistress with the emperor so as not to upset Roman customs, but opinions differ on this point and it is also probable that the Marcomanni saw the marriage as real.
31. Dexippus fr. 6; See the analysis in Drinkwater, 2007, 73–5.
32. Epitome de Caesaribus 34.2; Watson, 43.
33. Potter, 263–4; Watson 44.
1. The HA Aur. 18.1 states that Aurelian served as commander of all cavalry under Claudius. The evidence for the position of Aurelian under Claudius is collected by Saunders p.134ff. For the titles, see my MHLR Vol.1 and Caracalla. However, as noted above, it is actually possible that Aurelian had the title tribunus et magister officiorum.
2. This chapter is based on Saunders (137ff.), but unlike him I accept the traditional sequence of events and would not consider it impossible for Quintillus to have ruled for 77 days.
3. The HA (Claud. 12) claims that some of the surviving Goths tried to seize Anchialus and Nicopolis when they retreated northwards, but were then defeated by the valour of the provincials. In other words, Vopiscus claims that the locals managed to defeat the retreating Goths at a time when Aurelian had withdrawn his forces and marched to Sirmium to declare his bid for power. This is a good example of the tendency of Historia Augusta to whitewash all of the failures that took place under Claudius and Aurelian. Ammianus (31.5.16) states in no uncertain terms that the Goths pillaged both Anchialus and Nicopolis at this time. The other possibility is that this failure took place in 269 and was the reason why Claudius had urged Aurelian to improve his performance and why he had intended to send his brother with reinforcements to the Balkans only to change his mind later so that he marched there in person. See HA, Claud. 17 with Syvänne, Gallienus.
4. For the coins and papyri, see Watson 60–4, 221–3. Note, however, that my reconstruction of the events differs completely from the one adopted by him and others in suggesting an alliance between Aurelian and Zenobia against Quintillus.
5. The information concerning the dating, minting and papyri is in Watson, 60–4, 221–3. See also Southern 2008, 112–15.
6. The date of November is from Watson, 62.
7. For a completely different reconstruction of the early wars of Aurelian against the Iuthungi, Vandals, Alamanni and others, see Watson, 48ff. We reconstruct the campaigns in a completely different order and manner. In my opinion, he places the campaign of the Iuthungi mentioned by Dexippus’s fragment in the wrong context by connecting it with the campaign in which Aurelian suffered a defeat at Placentia in 271.
8. This emendation was made for the first time by Valesianus.
9. See the forthcoming Syvänne (Gallienus) with Zon. 12.26.
10. Saunders (169ff.) reconstructs Aurelian’s march in such a manner that he would not have reached Aquileia before turning northwards to catch the retreating invaders. This is a distinct possibility.
11. At this time, after the reign of Gallienus, the field army was divided into separate cavalry and infantry corps. The most important reasons for this conclusion are: 1) The existence of a separate cavalry commander implies the existence of a separate infantry commander; 2) the sources (e.g. Zos. 1.45) show the cavalry and infantry operating independently of each other, with the implication that these had separate command structures. For additional info, see Syvänne (MHLR Vol.1 with the forthcoming Gallienus).
12. Müller FHG 682ff.; Dindorf HGM 190ff. As noted, Watson interprets the evidence differently.
13. For the various views, see: Drinkwater (2007, 73–5); Saunders 163ff; White 65 ff.
14. The connection between the myriads of killed Alamanni with this campaign was first made by Saunders (p.171). The reason for my educated guess that Aurelian used cavalry for the ambush is that I interpret the use of the word peraioû (to carry over, convey to the opposite bank) in this context not to mean the carrying of the Roman attack over to the other side of the Danube, but to mean the throwing of the Iuthungi back over the river with a cavalry attack. The best evidence for this is that the Roman army was deployed on one side of the river as a crescent when the Iuthungi envoys came to meet Aurelian. The other reasons for this conclusion are: 1) Aurelian had previously served as hipparchos; 2) the Roman army was deployed as a single crescent on both sides of Aurelian when the Iuthungi envoys came to meet him and all of the leading officers are claimed to have stood beside their horses; 3) the Iuthungi envoys stressed the mounted fighting qualities of their horsemen (hippomachia); 4) it would have been difficult to overtake the retreating Iuthungi with infantry forces and then bypass them to place an ambush on the opposite bank of the Danube, whereas it would have been relatively easy for a cavalry force to mount a pursuit during which they would have been able to bypass the enemy which was encumbered with infantry and spoils of war and place an ambush on the opposite bank.
15. For the use of similar practices to receive enemy embassies under Constantius II, see Syvänne MHLR Vol.1.
16. For a fuller discussion, see the forthcoming Syvänne, Gallienus with Syvänne MHLR Vol.1.
17. I do accept Drinkwater’s (2007, 73–5) suggestion that the Iuthungi had previously been allowed to settle inside the Roman Empire by Gallienus and had received subsidies in return for defence of their lands somewhere in Pannonia.
18. Drinkwater interprets this to mean undiluted members of the warrior band, but I am inclined to agree with those historians who think that this refers to racial purity. This would certainly not have been the only instance of such behaviour among the Germanic peoples. It is not without reason that there was a distinction between the Alamanni (all-men) and ‘katharos’ Iuthungi (young ones), and between the Bastarnae (bastards with mixed Celtic and German lines) and Sciri (pure ones, the pure-bred Germans).
19. In short, I do not accept Drinkwater’s (2007, 77) claim that the Iuthungi army would have consisted simply of bands of young warriors. Would the emperor himself have taken the field with his field army, if the enemy consisted only of a few thousand raiders? Wouldn’t it have sufficed for him to let his generals handle such minor threats? Furthermore, could the numerous generals of the third century (namely those that were given command of entire sections of frontiers with the title of dux) have usurped power if they had only a few thousand men at their disposal? On the contrary, it is clear that the generals had large armies that allowed them to face the even larger enemy forces; the downside was that this also enabled them to usurp the throne. Most importantly, would there have been any need for the large- scale fortification programme (including the rebuilding of the walls of Rome), if the enemy consisted only of small numbers of raiders? Obviously not, because if this had been the case then even citizen militias would have been able to deal with them.
20. The timing of the events in Watson, 48–9.
21. Zonaras 12.26–7; most of the sources collected in Bancich, Zonaras, 59–60, 116–25; Banchich, Peter, 127.
22. Translations of the original texts conveniently in Banchich, 60, 124.
23. Saunders 173–5.
24. For Aurelian’s anti-corruption policies and other policies meant for the upkeep of the public morals, see e.g. HA Aurel. 36.1ff., 37.7, 38.2ff., 39.3ff. 49.3ff. The sequence in which these measures were introduced is not clear, but it is likely that Aurelian would have started his legislative reforms early on with those measures which corrected the abuse of power under Gallienus. Gallienus’s colossal project was cancelled by Aurelian: HA Gall. 18.2ff.
25. This is the strategy that was later recommended by Vegetius, Maurice’s Strategikon and Phocas’ De Velitatione. See Syvänne (2004) with forthcoming Nikephoros II Phokas.
26. Dexippus, Skythika fr. 7; Zosimus 1.48.
27. Dexippus, Skythika fr. 7; Zosimus 1.48–9. For the mistakes of Gallienus, see the forthcoming Syvänne, Gallienus.
28. The discussion is based on HA Prob. 6.5–6 and my biography of Gallienus.
29. Dexippus, Skythika fr. 7; Zosimus 1.48–9.
30. Peter the Patrician Banchich frg. 193 = Anon. Cont. Dio, Müller FHG 4, p.197.
31. Epitome 35.2 claims that the battle of Placentia was a victory for Aurelian, but this is clearly a mistake.
32. See my Gallienus.
33. Epitome 35.3; Llewellyn; Watson, 52; White, 76. Domitian the Second was probably the gifted general of Gallienus who contributed greatly to the victory of Aureolus over the Macriani in 261.
34. See the discussion in Llewellyn. Note however that my interpretation differs from that proposed by Llewellyn.
35. Since the two other usurpers, Septimius and Domitianus, can be located to Dalmatia and Gaul repectively, it is clear that Urbanus must have been at Rome.
36. The fact that Aurelian subsequently tried to establish the sun god as supreme god and used the Apollonius of Tyana as a Christ-figure for this religion suggests a cynical readiness to use religious concepts to promote unity within the empire. On the basis of this, it might be plausible that he could have sent the abovementioned letter to appease the traditionalists at Rome. On the other hand, the punishment Aurelianus unleashed against the senators at Rome when he finally arrived and the subsequent establishment of the sun god and Apollonius of Tyana might also be seen as his reaction against the old religion with its foolish consultings of the Sibylline Book. It would also have represented his rather desperate attempt to subvert the growing power of Christianity with a competing god and new Christ-figure acceptable to most pagans.
37. HA Aurel. 18.3–6.
38. Dio Cont. fr.10.4; Epitome 35.2; HA Aurelian 18.6ff, 21.5, 38.2–4, 39.8; Watson, 50–2; White, 81–2.
39. Zosimus 1.49.2; Epitome 53.3; Watson, 52; Llewelyn.
40. For Aurelian’s supposed cruelty, see Watson, 159ff.; Allard 2006b; White, 75–7.
41. Most of the Praetorians would have been with Aurelian, but it is still clear that a detachment of the guardsmen would have remained in the city itself. These would have been quite ready to join their comrades in Aurelian’s army and fight against the usurpers.
42. HA 21.5ff.; Zosimus 1.49.2; Victor 35; Epitome 35.4; Eutropius 9.14; Amm., 30.8.8. Watson 162, 223.
43. Watson, 133.
44. In this case, the militia suggests the troops in the emperor’s presence (i.e. Praetorians); riparienses frontier troops posted alongside the rivers (i.e. the Danube); lembariarii (after lembos - boat/ship) suggests either mariners/marines posted alongside the rivers (i.e. the Danube) or the mariners/marines of the regular sea-going navy, which in this case would have presumably meant detachments from the fleets of Ravenna and Misenum posted in Rome (note the location of their headquarters on the accompanying map); the castrensiani suggests troops posted in frontier forts; the Dacians were either regular troops posted in the province of Dacia or Dacian conscripts, the former being the more likely alternative. The list of troops suggest that Aurelian deployed specialist infantry corps used to fighting on the streets, rivers and hills. In other words, troops most suited to fighting in the streets of Rome that included all of these variants.
45. HA 21.5ff.; Zosimus 1.49.2; Victor 35; Epitome 35.4; Eutropius 9.14; John of Antioch Roberto ed. frg. 236.
46. The following discussion and the images of the wall are based on Nick Field’s study of the walls (esp. 23ff.)
47. HA Aur. 21.9, 35.1–2, 45.4–5, 48.1;
48. In other words, the effect would have been akin to a tax reduction in a modern society. Another good example would be the US system in which bankrupt individuals would regain their credit after some years have passed since their bankruptcy. This is one of the reasons for American economic dynamism which is missing from Europe, where the debts often stay on the records until paid in full, the latter of which makes most individuals quite reluctant to even attempt to improve their position in society.
49. HA Aurel. 39.3–4.
550. Watson, 127–35.
1. HA Aur. 22, Prob 9 (esp. 9.5);
2. HA Aur. 22; Amm. 31.5.17.
3. It was probably then that Constantius, the father of Constantine the Great, met Helena, the mother of Constantine, at Drepanum in Bithynia. Constantine was born at Naissus on February 273. See Syvänne, 2015, 169–70 after Barnes.
4. Vopiscus’s (HA Prob. 9.1–5) account of Probus’ exploits in North Africa and Egypt is confused and it is probable that he has confused the two Probi with each other. Therefore it is possible that M. Aurelius Probus actually fought against the rebels in Carthage including the duel against Aradio in 271 and advanced from there to Egypt, but since Vopiscus claims that Probus advanced from Libya, where he had fought against the Marmaridae, to Carthage, I have here adopted the view that these events probably took place in 269, just as the instance in which Probus was almost taken prisoner by the Palmyrenes would have taken place in 270 so that the return with reinforcements would have taken place in 272. Regardless, the state of the evidence is such that it is possible that the campaign in Carthage could have taken place in 271 and the reconquest of Egypt in 272.
5. Peter the Patrician, Banchich frg. 194 = Anon. Cont. Dio FHG 4, p.197.
6. Southern 2008, 134–5.
7. Saunders, 208ff.
8. Saunders (214ff.) suggests that Egypt surrendered when the news of the Palmyrene defeat at Immae was brought there and that the Romans reached it via land. He also does not accept that M. Aurelius Probus would have had any role. I agree with the usual view that Probus was in charge of the operation and that the conquest was probably an amphibious operation while it is not impossible that Probus was actually advancing from Carthage.
9. HA Probus 9.5; Cizek, 106; Watson 69–71, 223–4; Southern 2008, 115, 131–3. My interpretation of the campaign of M. Aurelius Probus in Egypt in 272 is that it is only the last mentioned part of the text of the HA, Probus 9.5 that belongs to the reign of Aurelian and that the incident in which Probus fought so carelessly actually belongs to the period when he fought under Tenagino Probus. I do not accept the claims put forward by some historians that there would not be any shred of evidence for fighting in Egypt (e.g. Southern p. 132 refers to Hartmann’s similar arguments). The shred of evidence is in the Historia Augusta, but of course these historians do not accept this as a valid source to be used.
10. Saunders (212ff.) places the battle of Immae to late April/May, but prefers late April. Either way it was easily warm enough if the wind blew in the right direction because the climate was warmer during the Imperial period – even in our less warm climate the weather can get hot if the wind blows from the right direction.
11. See e.g. Southern 2008,135–8; Watson, 73–4; White, 93–5. Stoneman includes a neutral comment regarding Downey’s reconstruction (170) but does not endorse it.
12. See above with White, 95–6.
13. For an analysis, see Syvänne 2004 with Syvänne MHLR Vol.1, 314–5.
14. For a fuller analysis, see my Caracalla. Caracalla kept the southern portion of the conquests, but apparently gave the northern portion of it back in return for formal recognition of Rome’s superior position so that these tribes became Roman allies.
15. Frontinus (Stratagems 1.9.8ff., 2.4.18) lists a number of instances of the use of the supernatural which had been used by commanders in the past and this is only one example. The religious practices of others could also be used. See e.g. Frontinus 2.1.16–17. Frontinus (1.12) wisely also included examples of how to dispel fears inspired by omens. It is likely that Aurelian was familiar with similar methods.
16. This referred to the barbarian troops of Sertorius who manipulated the barbarians by using a white deer as a means of obtaining ‘prophecies’.
17. See the units listed in Pseudo-Hyginus’s treatise.
18. Aelian (Devine ed. 37.3); Byzantine Interpolation of Aelian (Devine ed. 40.1–6, Dain ed. E1– 5); Strategikon 12.1.7.307ff., 12.2.13.
19. Saunders puts the end of the siege at August 272, but it is quite probable that it ended before as suggested by Cizek (111–114).
20. A good description of these can be found in Stoneman (172–3) and in Saunders (222ff., esp. 229ff.). Stoneman, however, puts the distance between Emesa and Palmyra at 140 km, while Watson puts the distance about 160 km. Consequently, there is no agreement regarding the distance: all are within the range 140–160 km, which is a march of four to six days in desert conditions for infantry and baggage train. The reason for the differences in the distances is that they were measured differently in different geographical works. These differences are listed in Saunders p. 230 n. 260. The following list emends the km figures, but it is possible that his miles should be emended. The shortest old estimate for the distance from Emesa to Palmyra is 80 miles, while the modern estimates vary from 88 miles to c.97/8 miles (155–157 km, 144 km in Saunders) up to 111–112 miles (178–180 km, 165 km in Saunders). The differences in estimates are therefore significant. White (98–9) states that the distance was 80–90 miles (130–145 km). For a general discussion of how the Romans organized their water supplies for their campaigns, see Syvänne, 2006.
21. E.g. Stoneman 173; Southern 2008, 140–1; Watson, 76; White, 98.
22. Tabari i.768ff., i.821ff.
23. Banchich Frg.195, Anon. Cont. of Dio, Müller FHG 4, p.197.
24. Gawlikowski; Saunders, 224ff.; Southern 140ff.; White, 99ff. On the basis of the finds, Southern (140ff.) claims that the walls did not surround the entire city, so she claims that there was no real siege at all.
25. See Saunders, 216ff.
26. Moses 2.78. Of note is also the claim that the Seres (i.e. the Chinese) revered Aurelian as god in the HA Aurel. 41.10. This is obviously suspect, but all the same it is quite probable that the fame of Aurelian reached China either via Persia or via the Red Sea route, or rather that the Seres in question were Chinese soldiers now in Roman custody.
27. Agathangelos (37) and Moses (2.79) with Syvänne MHLR vol.1. According to Moses (2.81), a Chinese refugee called Mamgon, who is claimed to have belonged to the imperial family of China, was sent to Armenia by Shapur I at the same time as Tiridates was returning to Armenia in 277. Mamgon chose not to fight against Tiridates, but Tiridates did not use him against the Persians. However, he gave Mamgon and his entourage a place to settle and a stipend to support his followers. For Mamgon and the Chinese connection see Moses 2.81, and for an analysis of the circumstances of his arrival, see Syvänne and Maksymiuk, 104. Did these Chinese bring with them the knowhow to build trebuchets, which probably made an appearance in Roman armies soon after this if we take fourth century heavy ammunition as evidence? There is obviously no need for this assumption because the Greeks and Romans certainly knew the mechanical principle as they used cranes. In sum, the evidence that we have suggests the likelihood that there were Chinese units in the Persian army at this time and that then as a result there were also some Chinese prisoners in the Roman army. The State of Wei and its successor Western Jin were neighbours of Persia in Central Asia and it is quite easy to see that it was possible for the Chinese to flee to Persia and vice versa. In fact, the son of the last Sasanian ruler fled to China. Conditions prevailing in China after the fall of the Han Dynasty in the third century would certainly have been such that there could have been a continuous string of refugees fleeing towards the west and the safety provided by the Persian Empire. A good summary of Chinese history at this time can be found e.g. in Lewis (31ff.).
28. HA Aurel. 33; Saunders, 239ff. PLRE1 Marcellinus 1.
29. HA Aurel. 28.4, 29.2, 33.2; al-Thalibi pp. 498–9. For the circumstances, see Syvänne MHLR vol.1 169ff. with Syvänne & Maksymiuk (108–9). The time of the death of Shapur I is not known with certainty. Estimates vary from 270 until 273. I and Maksymiuk prefer 272.
30. Banchich in The Lost History of Peter the Patrician (p.131) and Watson (186–7).
31. Examples of ‘cruelty’: HA Aurel. 6.3–8, 8.1–4, 16, 23.4–5, 31.1ff., 32.1ff., 36.1ff., 37.4, 37.7, 38.2ff., 39.2ff., 39.8ff., 40.2, 44.2, 45.1, 47.1ff, 49.3ff.; Epitome 35.4; Eutropius 14; Sextus Aurelius Victor 35.7ff. My interpretation of the evidence is basically the same as that adopted by White (75–7). That Aurelian liked to be present when people were tortured presages similar instances that took place under the Tetrarchs when they and their guests could be having banquets while people (e.g. Christians) were tortured in their presence. For this, see Syvänne, MHLR vol.1. The times were violent.
1. Cizek, 113–17; Homo, 108–15; Saunders 243ff.; Southern 2008, 152ff.; Watson 80ff.; White, 105ff.
2. The inscription (CIL III 12456) which mentions a battle along the lower Danube in Dobruja between Carsium and Sucidava reads as follows: [I(ovi O(optimo)… gratum referens quod i] mp(erator Aure(ianus) vicit [reginam Ze]nobiam inviso[sque tyrannos et Carpos inter Ca]rsium et Sucid[avam delevit] Duros[torum] Aurel[ianum]. Saunders notes that most historians associate this inscription with Aurelian’s Carpic war, but at the same time he correctly notes that the crucial word Carpos is a restoration and the evidence is therefore uncertain.
3. Who exactly were the laeti is not known, but it is probable they were defeated enemies handed over to the Romans by the defeated tribe who were then settled on Roman soil as farmers and soldiers, while the foederati (allies) were defeated tribes settled on Roman soil in their entirety.
4. Saunders 243–4.
5. Saunders (p.242) calculates that the journey of ca. 916 miles from Emesa to Byzantium took about 70 days.
6. In the early fifth century the courier (agens in rebus) Palladius (Socrates 7.18–19 with the PLRE1 Palladius 5) was able to carry the most important messages from the eastern front to the capital in three days and then back to the eastern front also in three days. See Syvänne, 2016.
7. Saunders, 250–1, Watson 82; White, 107–8. Watson claims that a person who swims among crocodiles, rides ostriches, drinks huge amounts of wine and gets people to strike an anvil balanced on his chest while he was in the crab position is scarcely a credible historical figure. In my opinion this is too restricted a view of what people do to get the attention and adoration that they crave. I see no reason to disbelieve the account because of this. People have done far stranger things to obtain followers. Contrary to majority opinion Stoneman (178–9) considers Firmus to have been a historical figure, but his interpretation of the papyri which call Firmus epanorthotes (corrector) is challenged by others. Southern (2008, 155–6) remains non-committal about the historicity of Firmus presumably in an effort to avoid criticism, but she at least notes that the HA did separate three different men with the same name Firmus.
8. The fact that the HA separates the Aksumites/Axomitae from the Indians means that the latter signifies the real India. The sudden emergence of the Blemmyes and Aksumites at this stage as supporters of Zenobia’s party should be connected with the events described by the so-called Monumentum Adulitum II (translation in Burstein 101–4). Hatke’s (37–66) analysis proves that it describes the conquest of the western shore of the Red Sea (the eastern desert with the Blemmye tribes of Atalmo, Beja and Tangaites) up to Roman Egypt, and the conquest of the eastern shoreline from the Gulf of Aqaba up to the land of the Sabaeans (this included the defeating of the Arabitae, which I interpret to be Arabs of Arabia Felix, and the defeating of the Cinaedocolpitae, which I interpret to be the Kinda) by some unknown Aksumite ruler during the period c.200–270. Hatke quite correctly notes that at this stage the Aksumites did not invade the territory held by the Meroites/Kushans, which took place in the following century. I would date the Aksumite conquest of the eastern desert roughly to the period 259– 270 on the basis of the information included in the HA. The earliest date for the Aksumite advance up to the borders of Egypt would be around 259–61 because it was in about 261 that the Prefect of Egypt Aemilianus usurped power and then advanced against the barbarians whom he forced to retreat (HA Thirty 22.6). The problem with this is that the text does not identify who the barbarians were. It is possible that they were the Blemmyes or some other barbarian tribe. If they were the Blemmyes then one can think that the Aksumites subdued them on behalf of Rome. We are on firmer ground for the period after 270. It was then that the Aksumites and Blemmyes were fighting on behalf of Zenobia (HA Aurel. 33.4–5; FSPB 3ff.). On the basis of this I would suggest that the Aksumites probably conquered the eastern desert regions with the western and eastern shorelines of the Red Sea during the 260s possibly on behalf of Rome. It was thanks to this that the Palmyrenes and their supporters in Egypt sought and obtained help from both the Blemmyes and the Aksumites.
9. For a detailed analysis of these different types of enemies, see Syvänne MHLR vols. 1–2 with Caracalla.
10. Sailing/rowing speeds are based on Casson’s estimates (281–99).
11. HA Aurel. 45, 47; Ammianus 22.16.15.
12. For Roman strategy when dealing with Parthians/Persians, see Syvänne MHLR vol.1 290–1, 337–9, Caracalla. The Romans operated in Indian waters also later, see Syvänne MHLR vol.2.
13. Syvänne Caracalla 227–31 with Dio 76.13.1. It is of course possible to think that Aurelian could have sought to outdo Septimius Severus and did indeed conduct the campaign against the Blemmyes, Saracens, Yemenis and Aksumites in person because Aurelian sought a return to the golden age of the Severans at least in his monetary policies. Regardless, a campaign by some general is more likely.
14. Based on Watson 128–42.
1. For an analysis of the term restitutor orbis, see Valérie Allard 1.
2. Strictly speaking Victor mentions only Tetricus present in the triumph.
3. This brings up the methodological question: Why do those who dismiss the HA as worthless not treat Zonaras or any other texts which they claim to include falsifications or errors in the same manner? If the same problems are present in these other sources, shouldn’t these too be dismissed as worthless. And vice versa if these other sources are accepted as valid for argumentation, why not the HA?
4. See the various theories in: Saunders, 262ff.; Southern 2008, 160–1; Watson, 113; White, 117–18. The PLRE1, however, has Severina as Augusta from 270 until 275.
5. Southern (2008, 160–1) lists the sources which mention the descendants of Zenobia. Southern, however, does not accept the claim that Aurelian would have married Zenobia’s daughter.
6. In other words, I do not agree with Watson’s analysis (93–5) but agree with White (112–3) that Tetricus and his staff betrayed their army to be butchered.
7. For the sources, see PLRE1 Tetricus1 and Watson 95, 246.
8. I therefore disagree with White (113) who thinks that the massacre was unintentional because Aurelian would have wanted to pardon those men just like Tetricus because they were needed. In my opinion the massacre was intentional because Aurelian did not need such men.
9. White 112–14.
10. See Syvänne, Gallienus.
11. Bernard (p.232) translates opes regiae as train of the court, while Magie translates it as wealth of the kings on the basis of the meaning of Ops as goddess of plenty. Both are possible.
12. Epitome 35.5; HA Aurel. 35.4–36, 46, 49.7ff., HA Tac. 10.4, 11.6.
13. HA Aurel. 45, 47–8, 50.
14. HA Aurel. 49–50.
15. HA Aurel. 36.3, 37.4, 39.8–9, 42.1–2, 50.5; Eutropius 9.14; Epit. 35.9.
16. Watson (101–2) suggests that the HA refers loosely to Gaul when it describes the barbarian invasion of Raetia, or alternatively that the invaders reached Gaul too, or that Aurelian marched to Raetia via Gaul. However, I am inclinded to agree with Saunders’ (258, 265–7) speculation that the rebels in Gaul could have been supporters of Faustinus in Trier while the attackers in Raetia consisted of barbarians (Alamanni). However, in other details my reconstruction differs from that adopted by Saunders. Cizek (190–2) and White (142–3) suggest that there was a rebellion at Lugdunum because the silver coins were not up to the new standard, but offers other alternatives. For my view, see the text.
17. Cizek, 192–6; Saunders 267ff.; Watson, 102–3; White, 143. It should be noted that the argument regarding the order of the cities in different sources regarding the place where Aurelian was subsequently murdered is not relevant because those that name Byzantium/ Constantinople first wrote in Constantinople so they naturally saw things from their own perspective.
18. One would like to know whether Aurelian had had a role in this fight between the kings in the same manner as Caracalla had had when he encouraged Artabanus to begin his civil war against Vologaesus. See Syvänne, Caracalla.
19. See Syvänne MHLR vol.1 with Syvänne and Makzymiuk (2018), 109ff.
20. Saunders’ claims (269ff.) that this should be dismissed on the grounds that the HA states that the Goths in question were the Maeotidae (Goths of the Sea of Azov) is false. He is mistaken here because the fact that the Bosporans later engaged the retreating Goths proves that the HA is correct.
21. See Syvänne, Caracalla. Caracalla, however, was not even the first emperor to employ Goths in the army because Maximinus Thrax (half-Goth, half-Alan) was already enrolled in the bodyguards of his father Septimius Severus.
22. See Syvänne Gallienus for the use of Goths by Valerian. See Syvänne MHLR vols.1–2 for the use of Goths under Galerius, Constantine and Constantius II. Constantine the Great actually concluded an alliance treaty with the Tervingi Goths which required them to contribute soldiers to the Romans when this was required. Julian was the first Roman ruler who did not understand the importance of this. He dismissed the barbarian helpers, although he still appears to have had some left in his army thanks to the policies of Constantius. The treaty relationship between the Romans and Goths was broken by Valens in the most foolish manner.
23. Christians could be motivated to fight with tricks, as Constantine the Great was to prove, but they could not be motivated with sights of the Sol or Apollonius of Tyana.
24. Jordanes Romana 290 states that Aurelian launched a persecution, but Eusebius is a period source and his view is to be preferred in this case because he did not have any reason to lie about this.
25. Eusebius HE 7.20–22; Lactantius 2.14–21; Orosius 7.23; Syncellus AM 5764; Homo 375–7, Watson, 198–202.
26. This would have been the general view of the Roman authorities towards what at the time was considered a strange cult which was alien to Roman culture.
27. The analysis of the murder is based on the following sources. Greek sources: Zosimus 1.62; Zonaras 12.27; Leo Grammaticus (also known as Symeon Magister), p.79; Consularia Constantinopolitana p.229; John of Antioch Roberto ed. frg. 238; Syncellus AM5765; Malalas 12.30; Cedrenus, p.455; Scutariotes/Synopsis Sathas p.39. Latin sources: Victor 35.8; Eutropius 9.15.2, Epitome 35.8, Lactantius 6; Chronica Urbis Romae p.146; Jordanes Romana 291; Jerome 263 Olympiad 260/2290/223c; HA Aurel 35.5, 36.4–6. Secondary sources: Homo, 322–6; Saunders, 273–80; White, 144–7, 153–5; 195–207; Watson, 104–7; Banchich, 122–5.
28. It is not surprising that the corrupt persons would have consisted of the officers of the bodyguards because Zonaras’s text (12.26) proves that this corps was already corrupt under Gallienus because Claudius (who was the commander of this unit, see my Gallienus) had been guilty of at least one such instance.
29. Vopiscus HA Aurel. 37.1–4.
30. Saunders, 276–280; White, 153–155; Cizek, 203–206; Watson, 104–116.
31. Ibid.
32. HA Prob. 6.5–7.
1. These are Orosius’s words (7.23).
2. It should be kept in mind that it was and is very rare even for great military commanders to have an unblemished war record with no losses. It is because of this that we should not judge Aurelian too harshly by the one defeat that he had. After all, he was able to rectify the situation after his mistake.
3. This and the following chapters build upon my earlier study of the reigns of Tacitus, Florianus and Probus in the MHLR (vol.1, 171ff.), and include new material.
4. Aurelius Victor 36; HA Tac. 1.1–7.7; Zonaras 12.28.
5. E.g. McMahon, Tacitus. Translations of the terms are by McMahon.
6. See e.g. McMahon, Tacitus.
7. HA Prob. 7.1–5.
8. Syncellus AM5765; John of Antioch Roberto ed. frg. 239: Tacitus defeated the Maeotidae and was on his way back to Europe when he was killed in a conspiracy. The reason for this was the nomination of Maximinus as governor of Syria. The assassins of Aurelian killed Maximinus and then Tacitus.
9. See Syvänne, MHLR vol.1, 335–6. The man murdered was Silvanus.
1. This and subsequent chapters are based on my earlier study of the reign of Probus and his successors (MHLR vol.1, 171ff.), but includes plenty of new material.
2. For a fuller discussion, see Syvänne, 2004 with Syvänne, Water and Caracalla.
3. John of Antioch, Roberto ed. frg. 240, Zonaras 12.28; Aur. Vict. 37.1.
4. See e.g. the list of opinions in McMahon, Tacitus n.29.
5. See later. Kreucher (202–12) lists the known governors of the provinces including Saturninus (p.207).
6. Kennedy, 231–6; Kreucher, 187ff.
1. Kreucher, 136–7. Kennedy (219) suggests in his thesis that the date in the Justinian Code is likely to be mistaken as are so many other dates in it, but on the basis of the Armenian and Georgian sources I agree with Kreucher that it is more than likely that the date is correct.
2. See Kazanski & Mastykova, 52–3.
3. Agathangelos 46–7.
4. We know that Probus had achieved some military success against the Persians by 21 October 279, because the title Persicus Maximus has been found on a papyrus bearing that date. The date and analysis of the role of Tiridates in these events can be found in Kreucher (158–61) and in Syvänne MHLR vol.1, 173.
5. Kreucher, 136–9.
6. For example by Kennedy, 220.
7. According to Julius Caesar (Gallic War 1.29), his description of the size of the army of the Helvetii was based on the register of the soldiers written in Greek. The list separated the children, old men, women and warriors from each other so that out of the entire force of 368,000 persons about 92,000 were able to bear arms. When one remembers that the Helvetii were just a small force in comparison with the large third century tribal confederations (note e.g. the size of the Iuthungi force), it is clear that a combination of these could easily put to the field as many as 400,000 warriors so that e.g. the Franks would have invaded with ca. 100,000 men, the Alamanni with 100,000 men, the Iuthungi with 100,000 men, the Vandals with ca. 50,000 men, the Burgundians with 80,000 and so forth.
8. The figure of 60 men is consistent with the six ranks formation included in Modestus (see Appendix 1), while the figure of 50 is consistent with the cavalry units (e.g. 10 files, 5 ranks). See the discussion in Appendix 2.
9. Syvänne, 2015, 175.
10. Syvänne, 2015, 175.
11. Syvänne, 2015, 175; Kreucher, 144–5; PLRE1 Clementius Valerius Marcellinus, Julius Nuffuzis, Julius Matif.
12. At the heart of the dating problems is the dating of the revolt of Saturninus the Moor, because the different ancient historians and chroniclers provide us with conflicting evidence about its date. Zosimus dates the revolt to the beginning of Probus’ reign, but as already discussed he has placed the events in the wrong order. Jerome (Hieronymus) dates the revolt to 281 (but his dates are one year too late) while Syncellus dates it to the sixth year of Probus’ reign (282). Vopiscus (HA Prob. 18.4–5, 19.1–20.1) also clearly implies that the revolt of Saturninus took place before the revolts in Gaul. If the revolt is to be dated early to the reign of Probus then it would probably have taken place in about 278 with the implication that Probus marched east to crush the rebel in 278/9. But if it is to be dated later it is likely that it would have taken place in about 279–280 or 280–281. In the former case Probus would have been forced to abandon his planned campaign in Germania or against Proculus and Bonosus and then campaign against Saturninus in about 280. The third alternative is that Saturninus revolted at about the same time as the barbarians in the Balkans so that Probus continued his campaign from the Balkans to the east in 281. In this study I have decided to adopt the view that Saturninus revolted in about 279–280 and was the reason why Probus did not continue from the Balkans to Gaul in 279, but turned back to the east and only then marched against Proculus and Bonosus in 280. The analysis of the evidence can be found in: Kreucher, 172ff. (dates the revolt to the year 281); Kennedy, 224 (dates it to 279); Syvänne, 2015, 175 (dates it to 279, but places the campaign in different context) and the sources and discussion in this book. Kreucher notes that Jerome has misplaced the events by one year so that the actual date he referred to would be 280.
13. This chapter is entirely based on Mitchell, 177–217, and the original sources mentioned. The other studies that I have used in this chapter are Kennedy (223–4) and Kreucher (150–5).
14. I suggested in 2004 that it is possible that the Romans were already using the trebuchet by the late Roman period but with the caveat that it was also possible to shoot these extra heavy pieces of ammunition with the older artillery pieces. Regardless, it is clear that the references to super heavy ammunition multiply after the third century, which means that it is possible that the Chinese prisoners captured from the Persian army (see the narrative) could have been responsible for the introduction of the trebuchet at this time.
15. Mitchell, 217.
16. The Georgian Chronicles falsely refers to the man who attacked Tiridates with the name Mirian (the King of Georgia/Iberia) whereas in truth Mirian would have been a subordinate of the Great King of Armenia Narses. Narses was the second most important man in Persia right after the king of kings Bahram II who was fighting against another brother of his in the east called Hormizd. I have here adopted the solution of calling Mirian of the Chronicles almost always with the name of Narses. However, it is still clear that the real Mirian did participate in these operations, but only as a subordinate of Narses.
17. Agathangelos (122–132).
18. In the former case Vopiscus places the revolt of Saturninus to take place before the revolts in Gaul while in the latter case Vopiscus states that the next project of Probus after the Triumph at Rome was to march against the Persians, which means that he no longer had to deal with Saturninus in 282. However, if the revolt of Saturninus was ended in the manner as described by Zosimus (see later) even the second of Vopiscus’s statements can be explained away.
19. According to Vopiscus (HA FSPB 7.2), it had been Aurelian who had nominated Saturninus as supreme commander of the eastern forces and that he had forbade Saturninus from visiting Egypt, but from the other sources it becomes apparent that Saturninus was also in this position under Probus.
20. It is of course possible that Zosimus has created a doublet between the names in 1.66.1–2, but in light of the fact that Probus was definitely the commander of the Moorish bodyguard cavalry at one point in time of his career makes it more than probable that Probus promoted his former comrades and friends to high positions.
21. The titles in Kennedy (225–6) and Kreucher (158–161).
22. The reason for this conclusion is that according to the extant sources four generations separate Odin from Horsa and Hengist, and that it was during this period that we see the mass migration of the Goths and others towards the Roman territories in 267 onwards and then the mass migration of the Franks, Alamanni, Burgundians, and Vandals. It is also at the end of this period that we see the so-called Saxons starting their raids at about the same time as Odin and his folk would have reached the north of Germany and Denmark. The first version of this appendix has been available online since 2016 at academia.edu so that everyone can check up why I have suggested this. The monograph Britain in the Age of Arthur has not yet been released by the publisher (Pen & Sword) at the time of writing.
23. See Kennedy, 224–5; Kreucher, 172–7; and above. Aurelius Victor (37.3) dates the revolts of Saturninus and Bonosus and Proculus to take place simultaneously, which would make it possible that the revolt of Saturninus took place in about 279–280.
24. All that we know is that there exist two coins which prove that at some unknown point in time in the third century there existed an otherwise unknown usurper Silbannacus who minted coins. See my biographies of The Reign of Emperor Gallienus (2019) and Gordian III and Philip the Arab (2021).
25. The sources are collected and translated by Banchich/Zonaras, pp.128–9.
26. Kennedy, 228–9.
27. Peter the Patrician Banchich frg. 197–8 = Anon. Cont. Dio Müller FHG 4, p.198 with the comments of Banchich p.131–2. For good examples of the effective use of disinformation and propaganda in the past see Syvänne, 2011, Caesar vs. Pompey, and Caracalla. Carus was certainly not the first Roman to use these.
28. Peter the Patrician Banchich frg. 197 = Anon. Cont. Dio Müller FHG 4, p.198 with the comments of Banchich p.131.
1. Kennedy, 235–6.
1. As stated in the beginning of the book, I will use the HA despite its general unreliability. In other words, I do not follow the absurd practice of either dismissing its evidence completely or the practice of frowning on it as a source after which it is still used because it is the only source for many events.
2. The sources are collected in the PLRE1 Carus. See also Leadbetter 1–3.
3. Leadbetter 1–3.
4. Carinus made Augustus early in 283 in Leadbetter 1 and Leadbetter 2.
5. Leadbetter 1–2; Comment on the coins of Numerianus by Magie in the HA Carus pp.426–7.
6. HA Carus 8.1ff., 9.4.
7. HA Carus 8.1ff., 9.4; Leadbetter 1–2.
8. Leadbetter 2 with Nemesianus.
9. The sources are usefully collected in REF1, 111–121 with endnotes that add references to other sources.
10. See my Caracalla for the effects of this behaviour.
11. It is for the same reason that modern day officers do not wear signs of their position in the field and why their subordinates are forbidden to salute them while on a campaign. The idea is to make it difficult for enemy special forces to target the commanders.
12. See also Syvänne, Caracalla.
13. See Syvänne and Maksymiuk, 2018, 124–6.
14. See the various versions preserved in REF1, 112ff.
15. For a similar situation facing Julian, see Syvänne Desperta Ferro 2015, MHLR vol.2, 94–112.
16. I have discussed this matter with Katarzyna Maksymiuk in the Military History of Third Century Iran (115–6) in which we decided to adopt a compromise position. However, since this book is a monograph, I do not need to make compromises in this case. In my opinion, it is quite clear that these sources have confused the reigns of Numerianus and Valerian and that we should not include any information drawn from them.
17. Numerianus at Emesa in March 284 in Leadbetter 3.
18. This chapter is based on my Military History of Late Rome vol.1 (177–90), but adds some new material which is missing from that book. The Military History of Late Rome also includes material which I have not included in this book.
19. Eutropius 9.19; Epitome 38.7; Zonaras 12.30.
1. See e.g. Rolan and Casas, esp. 306ff.
2. The text of Vegetius used here is: Flavius Vegetius Renatus, Epitoma Rei Militaris, edited with an English translation by Leo F. Stelten (New York 1990).
3. The late Roman Notitia Dignitatum includes among the vexillationes units which have tribal names such as the Mauri, Taifali, Marcomanni, Parthi, Dalmatae, Persae etc. The problem with this is that the vexillationes also include units that cannot be securely placed in the ‘tribal’ category, but it is of course possible that the units were promoted or transferred from one section to another in the late Roman hierarchy because the cavalry vexillationes ranked highest.
4. This is my interpretation of the problematic sentence: ‘Sic decem centurionibus regebatur quibus…’ which has usually been emended on the basis of Vegetius 2.8 as follows: ‘Sic decem centuriae cohortis primae a quinque ordinariis regebatur,’ meaning that the ten centuries of the first cohort were commanded by five men called ordinarii.
5. This is the reason why I accept the edition of Rolan and Casas rather than the emended text.
6. The light-armed levis armatura/ferentarii were at the time of Vegetius called exculcatores. The armaturae were men equipped with plumbatae, gladii and missilia like most of the forces of Vegetius’s day. The sagittarii were archers. The funditores were slingers equipped with slings or fustibali (staff-sling). The tragularii were equipped with crossbows that were either called manuballistae (hand-ballistae, presumably torsion-powered crossbows) or arcuballistae (bow- ballistae, presumably regular crossbows).
7. Alternative translations would be: 1) of the lightly-equipped shield-bearers; 2) of the men equipped with a light shield. Scutati expeditissimus allows many interpretations because soldiers described as expediti in the sources could mean regular legionaries without baggage or regular legionaries equipped as light infantry, and one can also think that the scutati expeditissimus would refer to the weight of the scutum-shield.
8. The Rolan & Casas edition (p.318) has destinatis illos, but the destinatis Indos of Vegetius is more likely.
9. Vegetius has magistri but not magistri militum.
10. Saw (serra) was a formation in which units advanced and retreated like a saw. The globus/ drungus was a separate detachment operating independently. The orbis was a unit or units facing two or all directions, the equivalent of the amfistomos difalangia of Greek military theory. The wedge (cuneus) was used to break through enemy arrays, while the pincer (forfex) was used to counter the wedge.
11. There is evidence for the use of legionaries as light infantry from the Republican era so the inclusion of these in the battle array was not new. Furthermore, there is also evidence for the use of extra-large legions like this one from that era too. Regardless, it is clear that Septimius Severus (if he created these legions) increased the overall size of the legion with additional light-armed men because before this the legionaries had just been seconded as light-armed when there was a need for this. For Republican-era use of legionaries as light infantry, see Syvänne, 2011.
1. Leo, Taktika, 4.43–9; Sylloge 35 (esp.35.2); Syvänne, 2004.
2. See e.g. Syvänne, 2004, 2011.