Political history

Goths in the Roman Army in 3rd Century

Goths in the Roman Army in 3rd  Century

Kalin Stoev

Institute for Balkan Studies & Center of Thracology

Bulgarian Academy of Sciences


University of Library Studies and Information Technologies, Bulgaria

Fuculty of Library Studies and Cultural Heritage

20 June 2020
This email address is being protected from spambots. You need JavaScript enabled to view it. core article


Abstract: Long before the official establishment of the foederati institute in the Roman Army, some Goths or other Transdanubians penetrated into its lines. Two inscriptions are particularly revealing because together with the military cursus of Maximinus Thrax, they could give us an insight for the general Roman policy to the barbarians as early as 3 rd century AD.

Key words: Еarly Goths, Roman Army, Mobile Army, Inscriptions.


The entry of the Goths in the Roman society as federates – massive in the 4th century, has its analogues in both – the early empire and the turbulent 3rd century. Reasons for this varie – the need for loyalty to the emperor, sought among peoples with a developed hierarchical clan sense of belonging; policy of assimilation of the barbaric aristocracy; the changed tactical fighting system of the Roman army, etc.

The outer political statics that gripped the empire after Trajan's victorious wars and the challenge of dealing with both internal and external enemies caused the wear and tear of the main unit in the Roman army – the Legion. The role of other specific troops, performing a wide range of tasks, started to grow. As early as the middle of the 2nd century, the active use of autonomous (ie without subordination to a certain legion or provincial garrison) cavalry garrisons began to be marked. In specific cases, if necessary, they were united into a mobile squadron.

They are especially well attested in the "outer front" in the Balkans – the province of Dacia. There we find equites Mauri, located in several places in Dacia, and perhaps also in Ratiaria. The Mauri must have come from tribes that made foedus with Rome and undertook military service.[1] They can be considered as elite, aristocratic in nature tribal cavalry, with tactical significance. Originally they were Mauri Gentiles[2] or vexillatio equitum Maurorum, that constituted the strike force of Trajan's general Lucius Quiet, who himself was of Moorish descent. They, like the Illyrians, were later turned into a numeri, parts of which were located at various garrisons in Dacia: some of them had defensive functions: for example, in salt quarries (numerus Maurorum S [alinensium?]). Or by the gold deposits at Ampelum (numerus Maurorum H [ispanensium?]).[3] In Tibiscum, by a castle near a thermal spring, honored by the frequent presence of governors, there is also a Moorish detachment, with ex singularibus consularis, i.a special detachment of bodyguards of the governor (AE 1978, 662 = AE 1994, 1490 = IDR III.5.219).[4] It is important to say that this is another evidence of how part of the Guards troops are ethnically differentiated (in this case, Moors or Africans) into a separate irregular army.

The "Germans" are well-known military units used as bodyguards in the court of the Roman emperor since the time of the Juliu-Claudian dynasty. It is believed that the trusted guards of Caracalla were recruited from the Alemanni after the expedition against them. To such conclusion leads the only source of this guard (another source: Herodian confirms the statement only  in regard to German guards) Cassius Dio (79, 6: 80 et seq.) : …καὶ γὰρ Σκύθας καὶ Κελτούς, οὐ μόνον ἐλευθέρους ἀλλὰ καὶ δούλους, καὶ ἀνδρῶν καὶ γυναικῶν ἀφελόμενος, ὡπλίκει καὶ περὶ αὑτὸν εἶχεν, ὡς καὶ μᾶλλον αὐτοῖς ἢ τοῖς  στρατιώταις θαρσῶν: τά τε γὰρ ἄλλα καὶ ἑκατονταρχίαις σφᾶς ἐτίμα.., where it is clear that Caracalla has recruited Goths (?) and Germans,[5] not only freemen but also slaves, for his guard and promoted them to the centurion rang. The latter information casts doubt on Dio's assertion that these were entirely barbaric troops – holding the command post of a centurion required a considerable degree of romanitas, including a command of the language for effective command of legionary units, as the mentioned centurions undoubtedly were those in the city and Praetorian armies.  It is impossible for "barbaric" officers from gentiles to became part of such a Romanized environment. That means that among the "lions" there were also mixobarbarians,[6] residents of the border regions of Germany and Scythia, and most of them are electi from the German garrison, ie. selected guardsmen[7]or, as in the case of Maximinus, from the garrison of Lower Moesia, or in general federations of tribes inhabiting the above mentioned regions. It can be said with great certainty and on the basis of a wide range of information that Maximinus the Thracian, the future emperor, was part of these "lions".

Information about Gothi (!) Gentiles , i.e. the probable "Scythians", exists from earlier times – 208 AD. Most likely their recruitment began in the time of Septimius Severus as part of the cavalry against the Parths for the campaign of 197–199 AD.[8] As evidence we have an inscription from the province of Arabia, which mentions a certain Guta (...), son of Hermanarius, who is indicated by the position praepositus of Gentiles, which probably should be identified with the famous from the 4th century Gothi gentiles:[9]

 μνημεῖον Γουθθα, υἱοῦ Ἑρμιναρίου πραιποσίτου γεντιλίων ἐν Μοθανοῖς ἀναφερομένων ἀπογεν<ομέν>ου ἐτῶν ιδʹ. ἔτι {²⁶ἔτει}²⁶ ρβʹ Περιτίου καʹ.

 Tombstone of Gutta, son of Ermanarius, the prepositus of Gentiles (ethnic armies , author's note) in Motanis. He died at the age of 14. In the year 102, on the 21st day of the month of Peritius (= February 28, 208)

The aforementioned Guta must undoubtedly be identified as part of the "Scythians" preparing to invade the empire in the time of Septimius Severus (Dio, 75, 3). It is possible that relations with them were settled by a special federal treaty, under which they undertook to provide contingents for the Roman army. An example of "barbaric" people who gave soldiers for the Roman army in 2nd – 3rd century, are the Tauroscythae, which were mentioned as defeated by Antoninus Pius after besieging Olbia. They were obliged to give hostages to the Romans (HA, Ant. P., IX, 9). This information is hapax, but that does not detract from its credibility. The name of the tribe is result of archaization, in which the name of one long-extinct people (Taurus) is transferred to another (in this case the amorphous term Scythians), i.e. it means nothing.

However, the Scythians were an opponent of the Bosporus kings (according to inscriptions) in 193 AD. On the Crimean peninsula, in the second half of the 2nd century, there was a "German" influence in the custom of funeral cremation. Therefore, it is possible that the turmoil at Pius’s time was caused by Sarmatian invasions, which were supported by German troops looking for prey and a field to settle in the south.[10] This may explain also the presence of Goths in the Roman army, under whose tutelage was located Bosporus Kingdom.  Another possibility is to accept that the Tauroscythae are in fact the Dacians, and so the main Roman opponent and donor of troops of the 2nd–3rd century are the "free" Dacians. Moreover, Antonin Pius is also known as Dacicus,[11] which name must be given for the defeat of "barbarians" near the borders of Dacia and the former Goth territories .

Caracalla commence an active foreign policy and by all means a diplomacy, including allute young aristocrats against the Dacians (Dio, 79, 27). Maximin the Thracian was the last emperor to  officially  bear the title of Dacicus (Maximus), after which the Carps and Goths left the coalition of the Free Dacians and acted alone. Guta's inscription can therefore be considered as an argument in favor of the claim that the "Goths" were part of the "Scythians," i.e. the Dacians.

Maximinus the Thracian himself was probably part of the "lions" of Caracalla. His origin of "mixobarbarian" is confirmed by several sources.[12] He subsequently held centuriate posts. Maximinus' career between his Praetorian tribunal (approximately between 219 and 222 AD) and his appointment to the Fourth Legion and the "mobile army" of the East (232–234) remain a mystery. The additions proposed so far in the literature seem either too low for a Praetorian or urban tribune at all, or too high (procurator in Mauritania Tingitana (IAM 02, 298),[13] cf. e.g. Licinius Hierocles or Ignotus[14]). In practice, for the time until the mid-220s, there is only one possible position, that of primus pilus bis, ie. second time primipil – a position that can be either a rank in Rome and without a specific army, or a certificate of occupation of two more primipili one after another.[15]

It must be assumed that Maximinus served again in the Roman garrison, in the Guards legio II Parthica: К. Lorio believes that Maximinus was a praepositus/ dux of the vexillations, which included one of the Parthian legions. Legio II Parthica is very suitable in regard of Maximin’s previous career, between 226–228, for this Imperial Guard entourage legion is composed mainly of Illyrians or Thracians. This means that Maximin is suitable for an officer position in it.[16] 

Now this assumption is confirmed by another inscription, which is probably related to the Goths or other Transdanubian population, and could also speak in favour  of the policy of Septimius Severus or Caracalla for surround themself with barbarian guards (i.e. “Scythians”):

D(is) M(anibus)/ Magni Mattonis/ discent(is) phalang(ariorum)/leg (ionis) II Part (hicae)|(centuria) II pil(i) pos(terioris)/ qui vixit ann(os) XXXX/ mil(itavit) ann(os) XX Sollemni(us) / Victorinus imagin(ifer) / heres bene merenti/ faciundum cu/ravit.[17] 

Shows that the aforementioned is generally followed the same career ladder as Maximin and probably started his career and his Roman army in the time of Septimius Severus and Caracalla, then, like Maximin, has found a place in the prestigious phalanx composed of Severus Alexander for his struggle with Persia (HA, Sev. Alex, 55), which, incidentally, confirms its historicity.[18] Magnus Matto is probably also of Germanic/Gothic descent – his cognomen is for now unparalleled, but is probably a nick-name derived from the place of the garrison of the Gothi gentiles in the East – the mentioned in the previous inscription Motanis.


[1] Onofrei 2007, 271–272.

[2] Christescu 1937, 196–197.

[3] Onofrei 2007, 272–274.

[4] AE 1978, 662 = AE 1994, 1490 = IDR III.5.219, Iulius Gracilis, [eques numeri] M(aurorum) Tib(iscensium) ex singulari [consularis] For the other Moorish detachments in Dacia, see Petulescu 2002: 134 ff.

[5] This is how the ethnonyms Scythians and Celts should be understood: Speidel 1984, 256.

[6] Ibid., 257.

[7] Domaszewski 1902, 509.

[8] Speidel 1984, 254 (712).

[9] Syria. Publications of the Princeton University Archaeological Expeditions to Syria in 1904–1905 and 1909. Division III, Greek and Latin Inscriptions in Syria, Section A, Parts 1-7. Southern Syria.

[10] Yartsev 2011, 144–164.

[11] Kienast 1995, 135.

[12] Stoev 2019.

[13] An inscription mentioning the praeses pro legato of Mauritania: IAM 02, 298 : [[[Iu]liae Mamaeae Augustae]]/ [[[matri Imperatoris Caes(aris) M(arci) Aure]]] / [[[li Severi] Alexand[ri]]] / [[P(ii) F(elicis) Aug(usti)]] cohors IIII / Gallorum devota / Numini maiestati / que eorum aram (?) C(aio) Iu/lio Maximino v(iro) e(gregio) / [p]raeside pro legato / [V]alerio Salviano [pr]aefecto posuit . The opinion is imposed in the literature by Speidel 1994, 68, without giving a specific argument.

[14] Dobson, 1978, No. 118a.

[15] Dobson 1978, 238–239.

[16] Stoev 2019 , 137 ff.

[17] Livius. org: " Apamea, Tombstone of Magnus Matto, soldier of II Parthica"

[18] Cf. Wheeler 1979.



HA, Ant. P. = Iuli Capitolini Antoninus Pius

HA, Sev. Alex = Historia Augusta, Alexander Severus Aelii Lampridii

AE  = Annee epigraphique

IAM  = Inscriptions antiques du Maroc

IDR  = Inscriptiones Daciae Romanae

Syria = Syria. Publications of the Princeton University Archaeological Expeditions to Syria in 1904–1905 and 1909. Division III, Greek and Latin Inscriptions in Syria, Section A, Parts 1–7. Southern Syria. Late E.J. Brill, 1907–1921.


Secondary sources:

Стоев, К. (2019), Providentia imperatoris. Биография на император Максимин Тракиеца. (Ямбол)/ Stoev, K.  (2019), Providentia imperatoris. Biografia na imperator Mauirmin Trakietsa (Yambol).

Ярцев, С.В. (2011), Проблема этнической идентификации варваров, напавших на Ольвию во время правления Антонина Пия. – Вестник древней истории, 2011, 2, 144–164/ Yartsev, S.V. (2011), Preblema etnicheskoy identifikatsii varvarov, napabshih na Olviyu vo vremja pravlenia Antonina Piya. – Vestnik Drevney Istorii 2011, 2, 144–164

Christescu, V. (1937), Istoria militară a Diaciei Romane (Bucureşti).

Dobson, B. (1978), Die Primipilares. Entwicklung und Bedeutung, Laufbahnen und Persönlichkeiten eines römischen Offizierranges (Bonn).

Domaszewski, A. v. (1902), „Untersuchungen zur römischen Kaisergeschichte I. Die Ermordung Caracallas.”  Rheinisches Museum 57, 506–516.

Kienast, D. (1995), Kienast, D. Römische Kaisertabelle. Grundzüge einer Kaiserchronologie


Livius. org: „Apamea, Tombstone of Magnus Matto, soldier of II Parthica” https://www.livius.org/pictures/syria/qalat-al-mudiq-apamea/apamea-military-tombstones/apamea-tombstone-of-magnus-matto-soldier-of-ii-parthica/

(visited 18.11.2018)

Onofrei, C. (2007), “Maurii în Dacia.Revista Bistriţei, 21,1, 269–282.

Petolescu (2002), Petolescu, C.C. Auxilia Daciae. Contribuţii la istoria militară a Daciei Romane (București).

Speidel, M. (1984), Roman Army Studies I (Amsterdam).

Speidel, M. P. (1994), Riding for Caesar. The Roman Emperors’ Horse Guards (London).

Wheeler, E. (1979), Legion as Phalanx. Chiron 9, 303–318.


This page is part of the project LABedia: Еncyclopedia of Late Antique Balkans, 4th-5th c.,
financed by the National Science Fund, contract КП-06-Н30/6, 13.12.2018