Thread Rating:
  • 3 Vote(s) - 4.33 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Late Roman Unit Sizes
In a previous thread on the Late Roman Army I cited the example given in Ammianus where the Batavi and the Heruli Auxilia Palatina units were paired to the extent that they even shared the same standard.

We also discussed the strange passage in Ammianus where he appears to mistakenly called the Iovii and Victores 'legions'. However, we must be very cautious here. We know that the surviving copies of Ammianus were damaged in parts, we have no idea if the copy all the surviving copies were based on was damaged in this particular part and the person copying it put 'legion' at that paoint because they did not know there were Auxilia units with the same name.

Correct me if I am wrong but doesn't Ammianus give the reason why the Regii were called the Regii? I don't read Italian but this person put the case forward for King Crocus as being the founder- Maurizio Colombo, "Constantinus rerum nouator: dal comitatus dioclezianeo ai palatini di Valentiniano I", Klio, 90, 2008, p. 124–161
Adrian Coombs-Hoar
Reply
(11-10-2016, 11:23 AM)ValentinianVictrix Wrote: doesn't Ammianus give the reason why the Regii were called the Regii?

I don't think so - he just calls them a 'formidable band' (formidabilis manus - 16.12.45). Unless there's some other reference to them... The Colombo paper might give some extra info, but as it's later than the Speidel one I suspect it just repeats the same idea. They do seem to be paired with the Batavi at this point though, so would have been auxilia.

Thanks for the reminder about the Batavi and Heruli - I recalled there was something about shared standards, but didn't have the reference!
Nathan Ross
Reply
(11-10-2016, 11:23 AM)ValentinianVictrix Wrote: We know that the surviving copies of Ammianus were damaged in parts,

Although you could be right here, and reading Amminus cautiously is appropriate, I cannot see any reason for such an error from the context of the description of the two units. The words legio and auxilia are different enough not be confused if just an excerpt is visible. A huge part of the text have to be blank or damaged to explain such a mistake. But then again, the content of the chapter here provides little evidence of such an accident.
We should also be very cautious not to declare all deviations as errors or as a mistake of the copyist.

We know that a whole series of old bararian numeri have been raised to cunei, alae, some of them probably to milites exploratores - which was of course a kind of promotional event for this units. The milites by the way belong to the auxiliares as well - clearly described in the ND Or. XXXIX - in the same way like the classical named "auxilium" does (ref. to ND Or. XLI).

In the late antiquity several military organizational forms appear as a very variable factor - a kind of adhocracy. And one must consider the Late-Antiquity-Legion as a kind of organizational form as many other military formations. In its exclusivity the formation "legio" has been surpassed by other unit-types - an exception is just the legio palatina. The time in which the legion was the backbone and all other troop-formations were "helping-hands" (like auxilia, cohors, ala etc) was over. This is also part of a cultural change. In the course of many centuries the legion was no longer the center of the classical roman "citizen". In the classical sense the identification decreased.
I consider the Legion to be a self-contained institution for a long time. I think that reports of a drastic barbarization of the legions in the West are pure exaggerations. However, the organizational form of the Legion should be considered differently regarding the late antiquity.

The old classical auxiliae, auxillares or auxiliarii were traditionally attached to the frontiers and they partially replaced the old cohors (compare command of Dux Daciae ripensis).
Speidel writes, the auxilia palatina was a new effort to strengthen the field army, it was however no new unit. And that the unit was called from the beginning "palatina" is also unlikely. Here is no dissent between my work and the phantastic article of Speidel. I also see that the vast majority of the 152 auxiliae were deployed years after 300AD.
So most of the units were actually newly deployed - this leaves the false impression that we are dealing here with a new institution. 
Compared to Speidel I place the origin of some of these regiments much earlier by considering the transformation of some numeri and old auxiliarii. Old numeri and old auxilarii, both groups appear as units of consistently barbaric character in the same way as the new auxilium does.


There is basically just a small difference between the new auxilia compared to the old auxilia or numeri of the pre-diocletianic time. The old units were sometimes mixed with cavalry, probably with a similar rating as the cohors equitata (and/or miliaria). The new auxilia (incl. palatina) appears as an infantry-troop only.
Furthermore a good ammount of units of numeri and auxiliae were shifted from the frontiers to the fieldarmies as pseudocomitatenses - and meanwhile I have collected several good examples showing that those "barbarian units" were classified - automatically so to speak - as legio when they served as pseudocomitatensian unit. Then, and this is the next exciting topic, some pseudocomitatensian legions in turn were also promoted to an auxilium of the field army (regardless of what they were before). Then, finally called with the additional epithton "palatina".
And this is perhaps the reason why some detachments appear as legio, while the sister unit appears as an auxilium. But a promotion from legio to an auxilium is not just limited to pseudocomitatensian units. Insofar Ammianus could be right here when referring both units in question to the legions.

So, an hypothetical career of a military unit could be (just an example to describe my point):
1. deployed as a numerus of 1000 men and 500 cavalry in the west (~240AD)
2. the unit is splitted and send to different locations, later on parts of the parent regiment is filled up again (~300)
3. the smaller detachment (500 strong) is taken immediately to the imperial field army and now called auxilia palatina. The other unit is stationed near the danube and responsible to man several towers and to ensure communication with riders.
4. after the loss of a field battle our numerus near the danube is taken to the remaining part of the field army (~378)
5. 425/430AD the Notitia lists 2 units of the same name - the unit in the west is an auxilium palatinum - the other unit in the east is classified as pseudocomitatenses and therefore most likely already referred to as "legio".
6. 60 years later the unit appears as αριθμος των λεγεώνων (numerus legionum) with +/-1000 infantry and 4-500 cavalry. The second unit is epigraphically no longer detectable.

(11-09-2016, 10:13 PM)Nathan Ross Wrote: Is there any support in 5th-6th C sources for this 'doubling' of units? It seems very common in the 4th century, especially with auxilia numeri.

If we follow the idea that the 'numerus legionum' (if you'll pardon the anachronism) might originally have been based on a two-cohort legion vexillation, as we see in the tetrarchic II Herculia inscription from Mauretania, and other places, might a 'numerus auxilium' have originally have been half the size, i.e. around the strength of an old cohort, and this is why they seem so often to operate in pairs?
A doubling or pairing of units is, as Evan said, no longer visible in the 6th century. Even in the course of the 5th century it is difficult to see if this concept has survived a long time. All signs indicate that members of the field armies were quartered in fixed garrisons before or during the reign of Justinian. Exceptions are the flexible units of foederatii and bucellarii, furthermore units of the 2 praesental-armies may have had changing wintering quarters. So, at least those units were still moving around. A doubling is possible here, but this must remain speculation.
Concerning your idea I must say that I belong to the group who considers a legionary detachment indeed as a group of round about 1000 men, and that the secondments are mirrored by the ammount of the tribuni, wherein the parent-unit (and the eagle) stays with the praefectus. If a cohortal system in a unit 1000 strong is still maintained is debatable. I don't believe it.
If we assume 500 men for an auxilia then we need 2 of them to match the number of the numerus legionis. I guess you're right here. In any case, a very interesting view.

by the way
barbarians catched the vexillum of the heruli et batavi and danced around it :-)
Amm. XXVII, 7,6: post cuius interitum Erulorum Batavorumque vexillum direptum, quod insultando tripudiantes barbari crebro sublatum altius ostendebant, post certamina receptum est magna.
Reply
(11-11-2016, 12:38 AM)Marcel Frederik Schwarze Wrote: So, an hypothetical career of a military unit could be (just an example to describe my point):
1. deployed as a numerus of 1000 men and 500 cavalry in the west (~240AD)
2. the unit is splitted and send to different locations, later on parts of the parent regiment is filled up again (~300)
3. the smaller detachment (500 strong) is taken immediately to the imperial field army and now called auxilia palatina. The other unit is stationed near the danube and responsible to man several towers and to ensure communication with riders.
4. after the loss of a field battle our numerus near the danube is taken to the remaining part of the field army (~378)
5. 425/430AD the Notitia lists 2 units of the same name - the unit in the west is an auxilium palatinum - the other unit in the east is classified as pseudocomitatenses as therefore most likely already referred to as legio.
6. 60 years later the unit appears as αριθμος των λεγεώνων (numerus legionum) with +/-1000 infantry and 4-500 cavalry. 

I Always had my doubts about the origins of the auxilia palatina being among the old auxilia cohorts of the army serving along the borders.
Somehow I can't imagine how an old unit, once raised among foreign tribes but probably long-since filled with Roman replacements, is suddenly rised not only to a position apparently above all the other army regiments, but given a new name as well and a 'palatial' association to boot!
If (as you pointed out) the training of the auxilia was below the level of the training of the legions, why are these few units picked and upgraded while others of the same name go the 'normal' way of upgraded limitanei units and only get to be listed as pseudocomitatenses?
For me, the origin of the auxilia palatinae is far more easily explained as totally newly-founded elite units.
Robert Vermaat
MODERATOR
FECTIO Late Romans
THE CAUSE OF WAR MUST BE JUST
(Maurikios-Strategikon, book VIII.2: Maxim 12)
Reply
(11-11-2016, 12:38 AM)Marcel Frederik Schwarze Wrote: If a cohortal system in a unit 1000 strong is still maintained is debatable. I don't believe it.

Me neither. The original two cohorts that may have made up a 'mobile' legion vexillation would probably have been blended into one integral unit ('numerus legionium') fairly soon. Whether that unit had any set internal structure (beyond, presumably, the century) is unknown...


(11-11-2016, 12:38 AM)Marcel Frederik Schwarze Wrote: If we assume 500 men for an auxilia then we need 2 of them to match the number of the numerus legionis. I guess you're right here.

Alternatively, if we give the original 'cohort' a paper strength of c.600 (based on the 6000-man 'traditional' legion of Vegetius, Lydus etc), then a 'numerus legionum' could have numbered c.1200, which fits with various previous theories. The 'numerus auxilium' could therefore be c.600 men - and the 300-man detachments we read about in Ammianus would be no more or less than half a numerus or quarter of a new legion!
Nathan Ross
Reply
(11-11-2016, 08:53 AM)Robert Vermaat Wrote: I Always had my doubts about the origins of the auxilia palatina being among the old auxilia cohorts
I know that some authors claim, falsely to my opinion, that some or many of the new styled aux. palatina were created out of the old cohors. (for example Goldsworthy, Roman Warfare, 2000, p.174). Against this theory we see that cohorts were still deployed by Aurelianus, at least 15 by Diocletianus and some others at a later time. For me a clear indication that this troops were still alive and had its place in the current hierarchy of the army.
Beside those old classical cohohrts, which were as you suggest indeed very romanized in our time, there were other units just called auxiliae, auxillares or auxiliarii. A chapter which is not too much explored. Also uncertain is the legal position of these additional auxiliares of the 3rd century. Traces of these (late?) 3rd-century auxiliaries are found at the Danube-ducats. Here they simply replaced the old cohorts. Also the Franki taken by Aurelianus to fight against Zenobia were neither cohorts nor foederati. The fact is, however, that they were a barbarian group like the old numeri. But as said, the vast maiority of the new auxiliae were surely deployed during and after the reign of Maximinus. But they were most likely no entire new troop.
We can be sure that some of those existing auxilia or numeri were taken immediately to the retinue of the emperor (or one of the co-emperors), others were promoted to a later date. Most of them were of course newly deployed and also taken immediately to the field army. Calling pseudocomitatenses to the field armies must be seen as a pure need out of necessity. I'm sure that there was also an official way for a legion (or one of its detachments) or an auxilium to get access to the field army.

In my understanding the epitheton "palatina" is referred to units marching close to the presence of the emperor in comitatu and praesenti. They were no new type of troop, just a preferred group within the mobile field army. We also wouldn't call the "legio palatina" a new type of troop. And we see that also as part of the frontier armies these auxilaries are listed first - the legions next.
The auxilia appears in an elevated position at the limes as part of the limitanei. First the barbarian cavalry is indicated in the Notitia - followed by the auxiliaries - then appears the legion. Thus we find the special position not only in the field armies.

Few Romans had the wish to serve in those barbarian troops called "auxilia" (Veg. II 3) - but even during the life time of Vegetius, the auxilia seems to be still of strong barbarian origin and character. Subordinated as well as allied peoples made troops available to the empire - often led by Princes or high nobles. This led to the fact that the best warriors often entered these units from the right side of the Rhine, following the example of their leader. The warriors who arrived there were already fully trained - certainly different from the Romans. On the other hand, formations such as the shield-wall were also known to the germans and gauls. Their ability and fighting power were feared by their opponents. And Constantius asked Julianus for providing 4 auxiliae for his persian campagin, showing the importance of those troops. Concerning the equipment and the used weapons I must say that I'm no specialist on this level and in this forum are dozens of people much better for giving some useful answers. It seems, however, that by the end of the fifth century all differences to other units regarding this question are largely eliminated.
Reply
(11-11-2016, 12:52 PM)Marcel Frederik Schwarze Wrote:
(11-11-2016, 08:53 AM)Robert Vermaat Wrote: I Always had my doubts about the origins of the auxilia palatina being among the old auxilia cohorts

...even during the life time of Vegetius, the auxilia seems to be still of strong barbarian origin and character...

This is one of these questions that we lack the evidence to answer adequately, but I think I'd take a middle course.

It does seem likely to me that the roots of the various later Roman auxiliae lie in the assorted 'barbarian' contingents that seem to have frequently accompanied imperial expeditions from the later 3rd century onward. From Aurelian and Probus, certain of these contingents may have been given regular status as old-style cohorts and alae (like the Cohors Gotthorum and Ala Iuthungorum in Egypt in the ND), while others may have been used as garrison forces on the frontiers (the limitanei 'auxilia' of the ND).

At some later point, either under Diocletian or Constantine, this mass of semi-irregular 'barbarian' formations, many of whom may have been serving (under their own leaders?) with the imperial forces for many years, were given a more regular status in the Roman army as auxilia numeri, with a Romanised rank structure, state-supplied equipment and training (and presumably some sort of standard size too!).

After this date, while these units may have preserved aspects of their 'barbarian' origins, and may have relied heavily on manpower from across the frontiers and from settled laeti communities, they also took in Roman citizen recruits, perhaps volunteers. In time this 'Romanisation' of the auxilia came to dissolve the boundaries between them and the legions; quite possibly by this point both sorts of unit had come to have a generally similar size anyway.
Nathan Ross
Reply
(11-11-2016, 02:15 PM)Nathan Ross Wrote: After this date, while these units may have preserved aspects of their 'barbarian' origins, and may have relied heavily on manpower from across the frontiers and from settled laeti communities, they also took in Roman citizen recruits, perhaps volunteers. In time this 'Romanisation' of the auxilia came to dissolve the boundaries between them and the legions; quite possibly by this point both sorts of unit had come to have a generally similar size anyway.
Well then. I have nothing more to add. Looks like a fluent process.
In their development those units tried to be similar to the cohort or legion (or it was imposed to them) resulting in a similar (!) ranking scheme with centenarii, ducenarii, senator (+primicerius?) and tribunus. And what about their unit size? What happened if a tribe has sent 800 men, another tribe just 350? Was the great unit (800 strong) halved? Or did the Romans always demand the same number of soldiers?
Reply
(11-11-2016, 03:33 PM)Marcel Frederik Schwarze Wrote: a similar (!) ranking scheme with centenarii, ducenarii, senator (+primicerius?) and tribunus.

I still haven't come across any theories as to why the auxilia adopted these ranks, rather than the old legion or auxiliary ones - perhaps to 'synchronise' them with the other new units of the imperial retinue - equites, scholae, etc? I'm not sure if we have any certain evidence for ranks within the palatine and comitatenses legions, but as far as I know they seem to have held onto their old ones, with ordinarii etc.


(11-11-2016, 03:33 PM)Marcel Frederik Schwarze Wrote: What happened if a tribe has sent 800 men, another tribe just 350? Was the great unit (800 strong) halved? Or did the Romans always demand the same number of soldiers?

My guess is that the institution of the auxilia was an attempt to regularise these previously rather ad-hoc contingents into into units of a set size and composition.

Perhaps men of different tribal (and Roman!) origin were mingled together in some units, and this is why we have these odd titles like Cornuti or Petulantes, or the names of archaic tribes being used?

Some did have current tribal names - Salii, Heruli, Bucinobantes, etc - which might suggest that those peoples, at least initially, were able to supply more men to fill out the entire units, perhaps?
Nathan Ross
Reply
Where are the Turmae in the Notitia Dignitatum?

According to Ammianus, Lydus and others the Turmae should have been a genuine Late Roman Unit, but why the N.D. have no mention of it.

Is there any evidence of the cavalry formation known as Turmae having been created in the Tetrarchy timeframe, or is it a Constantine creation?
Reply
(12-06-2017, 08:58 PM)Julian de Vries Wrote: Where are the Turmae in the Notitia Dignitatum?

They're contained within the Alae in the Notitia Dignitatum!

Turmae (singular turma) were troops, probably numbering 32 men, in the alae (squadrons) of the Principiate auxiliary cavalry. Each turma was commanded by a decurion.

A papyrus from Egypt (Chla 18.660) lists the officers of Ala III Assyriorum under Constantine - the list mentions both turmae (turmhs) and decurions. So it seems that the old-style cavalry alae still existing on the frontiers in the later period (and listed in the ND) were still organised in the traditional manner, whatever their numbers might have been.


(12-06-2017, 08:58 PM)Julian de Vries Wrote: According to Ammianus, Lydus and others the Turmae should have been a genuine Late Roman Unit

I believe Ammianus is using the word turma generically, to mean a cavalry unit. The units he's describing would perhaps more officially have been called numeri equitum. The prevalence of the rather colourless 'numerus' to refer to any sort of military unit in the later period perhaps led writers (in the 'grand style') to use older terms quite indiscriminately to describe them - so Claudian and his 'cohorts', for example.

Lydus does have a 'turma' of 500 horse archers in his list of units (De Magist.1.46). Although he seems to be describing the composition of the army in a much earlier period ("since the entire populace from the beginning served as soldiers, they decided to set up a definite, action-ready force") - so the reference is quite obscure.
Nathan Ross
Reply
THE LATER ROMAN EMPIRE 284-602 A SOCIAL ECONOMIC AND ADMINISTRATIVE SURVEY By A. H.  M.  ]ONES VOLUME  Ill (Page 210):

Libanius speaks  of  an ίλη όπλιτϖν (a cohort?) in Diocletian's day  as  numbering  500  men  (Or.  XI.  159  ff.,  xx. I8).

*

According to Dositheus, Hermeneumata Leidensia  εἴλη could translate to turma.
Suddenly  Lydus'  turma of 500 horse archers sounds reasonable.
The problem though, is that the proof rests on linguistics, and I don’t like that.
Something additional is needed in order for it to be agreeable.
Reply
(01-12-2018, 07:46 PM)Julian de Vries Wrote: an ίλη όπλιτϖν (a cohort?) in Diocletian's day  as  numbering  500  men

What an ilê of hoplites might be in official parlance is anybody's guess! But a cohort seems close enough. We know that cohorts existed at this point, as we have inscriptions dated only five years previously left by Maximian's troops in North Africa.

This reference is to the men commanded by Eugenius working on the harbour at Seleucia in AD303, who revolted after being required to do extra fatigues, proclaimed their commander emperor, got drunk and marched to Antioch, where they were defeated by the citizens.

So Libanius is using the word ilê to refer to an infantry unit. Another example of the vagueness of Greek-Latin-Greek translations of military terminology!
Nathan Ross
Reply
(11-11-2016, 12:21 PM)Nathan Ross Wrote: Alternatively, if we give the original 'cohort' a paper strength of c.600 (based on the 6000-man 'traditional' legion of Vegetius, Lydus etc), then a 'numerus legionum' could have numbered c.1200, which fits with various previous theories.

Nathan,

This cohort strength of c.600 fits in well with your diagrams over in the Perge legion thread!  If the Slab C truly does represent Vegetius' additions of Augustiales and Flaviales to a century, then many things fall into place, like a two-cohort vexillation morphing over time into a 1200 man independent legion.

Austin
Reply
I've just tracked down one of the references from Coello's Late Roman Unit Sizes.

On p.100 he writes: "An Egyptian papyrus, apparently dating from 18/19 March AD 399 records payments of annonae to a detachment of legio V Macedonica at Memphis. 835 rations of wine (each of one sextatius or 0.54 litres 10.9 pints]) and meat (0.45 kilogrammes [one pound] each) were to be paid apparently for two days, giving a detachment size of about 340-400 men. (This calculation is incidentally dependent on assuming that a fraction of soldiers received multiple annonae.)"

Coello gives no reference for this in his footnotes, but the papyrus appears to be HGV SB 20 14675 / ChLA 45 1328:

AD 399 Mar 19 

Est vino mil(itum) leg(ionis) Maced(onicae) sub c(ura) Gaioli trib(uni) d(iurnarum) an(nonarum) octingentas triginta cinque, ex d(ie) quintum decimo Kal(endas) April(es) [i]n d(iem) qu[art]um decimo Kal(endas) easd(em), duo cene. Post cons(ulatum) [dom(ini) n(ostri)] H[on]orii p(erpetui) Aug(usti) qua[ter] et Eytychiani v(iri) c(larissimi), p(er) Sergio actuario, ind. XII. [ -ca.?- ] γ(ίνεται) ὁ̣(μοῦ) ξ(έσται) ωλε.

Est carne mil(itum) leg(ionis) Maced(onicae) sub c(ura) Gaioli trib(uni) d(iurnarum) an(nonarum) octingentas trigint[a] ci[nqu]e, ex d(ie) qu[int]um decimo K[al](endas) [Apri]l(es) in d(iem) quart[u]m decimo Kal(endas) easd(em), [d]uo cene. Pos[t] c[ons(ulatum)] d[o]m(ini) n(ostri) Honorii [p(erpetui) Aug(usti) quater et E[yt]ychiani v(iri) c(larissima), p(er) Sergio a[ctuario, ind. XII]. [ -ca.?- ] γ(ίνεται) ὁ(μοῦ) λ(ίτραι) ωλ̣ε̣.


835 annonae over two days would give 417.5 per day, so obviously some soldiers must have received 1.5 annonae. Ordinarii alone may have had 8 annonae each at Perge levels, and Augustales 6.

But might a figure of 340-400 men be about right, and if so how did Coello work it out?

A figure of only 300-400 men for a 'frontier legion' of c.AD400 would rather alter our estimates for the Notitia Dignitatum...
Nathan Ross
Reply


Possibly Related Threads…
Thread Author Replies Views Last Post
  Roman unit transfers Jason Micallef 3 930 01-04-2019, 10:35 PM
Last Post: Jason Micallef
  Ile or ala? : the unit size of a Roman ile Julian de Vries 3 2,586 05-18-2017, 09:36 AM
Last Post: Julian de Vries
  Late Roman Unit Titles - By Weapon Mithras 2 3,299 03-16-2007, 11:28 PM
Last Post: Robert Vermaat

Forum Jump: