Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Roman Legion in late Antiquity
#1
Hello. Can someone help me with some numbers?
In 4th century Rome, I read that a legion numbed about 1200 men, a century about 80 men,. So what was the size of a cohort during that time, and how many in a legion (if there were only 1200 in a legion)?

I got this information from The Roman Army by Adrian Goldsworthy but the njmbers dont seem to add up

Thank you
Reply
#2
You may have mis-read. A century likely numbered between 60-100 men in the 4th century, depending on variable factors (understrength, etc). A legion varied. For Limitanei it is possible it numberd at 3000. For comitatenses it is possible it numbered at 800-1200, and for Legio Palatinae it is possible it numbered at around 800-1200.

A cohort probably still existed in some units, and didn't in others. Many Palatinae and Limitanei were not organized into Legions, and were cohort sized units. Overall it is believed that the 800-1200 man Legion was divided either directly into Centuries, or into maniple-sized units and then into centuries (The ducenarius is still debatable).
Reply
#3
Quote:what was the size of a cohort during that time, and how many in a legion (if there were only 1200 in a legion)?

There's very little hard evidence to go on, and a great deal of theorising!

It seems that the full strength of a legion in the 4th century was still officially 5-6000 men (Vegetius (Epitoma Rei Militaris I.17) claims that the Herculiani and Ioviani legions (perhaps actually I Iovia and II Herculia) had 6000 men - these seem to have been new formations of the late 3rd century). So this was perhaps the official number for the post-Severan period, even if most legions were very understrength.

However, many legions of this period sent detachments on active service, and these detachments seem to have comprised two cohorts, probably each under a tribune. The Oxyrynchus papyrii give details of Diocletian's field force in Egypt c.298, with these legion detachments listed in pairs brigaded together under a praepositus (perhaps one of the tribunes given a superior command role). If we assume that the cohorts still numbered around 4-600 men, this would make each detachment about 800-1200, and the brigaded pairs around 1600-2400 men.

Several of these detachments could be operating at one time away from their 'parent' legion - an inscription from Mauretania lists two cohorts of II Herculia, probably in the field army of Maximian, while other units of the Herculiani, perhaps from the same legion, were apparently serving in the east under Galerius and Diocletian, and perhaps also as garrison troops in the Crimea.

I suspect that what might have happened was that these mobile detachments in time acquired a semi-independent status as part of the imperial field armies, losing their connection with the parent legion. They became referred to as 'legions' in their own right, some of them using variations on the old legion name, while other new formations of the same size were raised alongside them. Thus we get the comitatensis 'legion' of c1200 men. Whether they maintained the two cohort split (perhaps this is the source of the 'maniple' mentioned by, I think, Ammianus and Vegetius?), or functioned as collections of centuries I don't think we know. Interestingly, the habit of brigading these 'legions' in pairs seems to have continued, and carried over into other units of the comitatus.

The parent legions, left back on the borders, would necessarily have been depleted by the departure of these detached cohorts, and further divided amongst various different garrisons. This might be why later legion fortresses (Troesmis on the Danube, home base of II Herculia, for example, or some of the forts in the east) appear to be smaller than the traditional bases. The legion itself was about the same size (perhaps), but it was spilt up into several smaller parts. These, then, would be the limitanei legions - a number of 3-4000 has been suggested, and seems right (although I read an article recently suggesting that numbers were very much lower... It probably varied over time.)

Alongside the legions, there were new formations altogether - the independent cohorts on the frontiers were surely the descendents of the old auxiliaries, still inferior to the legions it seems although now all citizens. There were also the 'numeri' of the auxilia, new-style auxiliary formations, some of which were taken into the imperial comitatus as auxilia palatina. Numbers for them are pretty much guesswork, I think. 800-1000 seems about right.

Evan mentioned the ducenarius - this rank is still not fully understood, but from what I can gather it was used only in the new-style formations - numeri of auxilia, cavalry equites and the bodyguard protectores.

If you can find it, take a look at Nicasie's Twilight of Empire - he provides about the most agreed-upon figures for later military units. Although I prefer A.D. Lee's War in Late Antiquity as a more readable alternative.

I've probably made quite a number of mistakes with the above, but hopefully others can chip in and expand further.

Confusedmile:
Nathan Ross
Reply
#4
So would the following be accurate?

The Roman centuria, or century, numbered between eighty and one hundred men. Some sources suggest that in the fourth century a legion numbered about twelve hundred men, though at full strength a legion could be as large as six thousand. A cohort consisted of around five hundred soldiers. A legion contained two or more cohorts, though in the latter empire some units no longer used the cohort system.
Reply
#5
One of the few bits of written evidence we have for the numbers of a Roman legion comes from a book called De munitionibus castrorum, by an unknown writer referred to as Pseudo-Hyginus and dating (probably) to some time in the 2nd century. He gives the century as 80 men, divided into ten groups of eight. This contradicts Vegetius, who describes a 'contubernium' of ten men.

Excavations of legionary and auxiliary forts and fortresses appear to bear out Hyginus to an extent - a barrack block seems to have housed a single century, and most had rooms for ten groups, often with a few extra. So the 80-man century seems a good bet for the principiate period, at least on paper.

The later legion may have been different, or different in some places. Some later forts (El Lejjun and Dionysius in the eastern provinces, possibly also Troesmis although I'm not sure about that one) appear to have had less rooms in a barrack block, and Nicasie suggests that this might mean a reduced size century. There are barrack blocks of between five and nine rooms, for example, which could imply a century of as few as 40 men.

As you can guess, there's a lot of uncertainty about all this, and many theories flying about. I would estimate a limitanei legion probably kept the old-style organisation, perhaps with fewer cohorts to account for missing detachments. Say eight cohorts each of six centuries numbering 60-80 men, which would give a legion size of around 3000. The field army legions could have had a totally different organisation - perhaps indeed Vegetius's ten-man contubernium derived from there?
Nathan Ross
Reply
#6
Nathan is probably right. I suggested 80-100 because I had heard it's possible Septimus Severus tried to increase the size of the Century.

As for the Legions... well 6000 is true and untrue. Many Comitatenses/Limitanei units had formed their own identities, especially by the time of the Notitia, and were no longer associated with whatever parent unit they may have originated from. There were still a handful of units, notably V Macedonica and XI Claudia, who maintained their identity across the empire. But I doubt they still numbered 6000 men, e.g. V Macedonica had 3 detachments - one Limitanei Ripensis garrison in Oescus, numbering approximately 500, one Legio Comitatenses in Syria, numbering between 800 and 1200 men, and one Legio Comitatenses in Egypt, numbering about the same as Syria. There may be another garrison I forgot to account for.

New units, or old legions that had lost their identity, remained at about 1000 men in strength individually. A 40-60 man century could mean that it was divded into 3-4 cohorts, not 2. Newer ones may have used a Manipular unit size with a Ducenarius in command, so instead of 3 cohorts you had more or less 8 maniples.

Legio Palatinae were usually units from the 3rd century reclassified into the new system, they seemed to have numbered about 1000 men and as Nathan said, oftentimes had detatchments. However, the total size of these units still varied. Newer ones probably used a Manipular-like system, while the ones granfathered in were likely using a cohortal system of some kind.

Auxilia Palatinae were a different story. Some, like the Cornuti, were brought into the new system during the reign of Constantine and were made out of the old style Auxilia. They somehow became Elite units instead of Limitanei (I have no Idea how) and were probably organized in the same way Auxilia Quingenaria cohorts were. Some more notable units may have been Milliaria cohorts, as they were then divided into Junior and Senior (Cornuti Iuniores and Cornuti Seniores). Furthermore, the new units seemed to be Milliaria. Take for example the Placidi Valentiniani, raised in 419-425, with two detatchments. The Placidi Valentiniani Felices Seniores in Illyria (listed simply as the "Valentinianenses Felices") and the Placidi Valentiniani Felices Iuniores in Gaul (listed simply as the "Valentinianenses Iuniores). Total they numbered 1000, but operated independently of each other like most late-era units. Raised in the new style they probably had a Ducenarius system with Maniple-sized regiments on top of their centuries. So they were 1 cohort of 3 maniples (or something like that).

Cavalry... actually I don't know much about late era cavalry organization. It was different than Principate or Dominate though.

I will note my reference to "Maniples" is due to lack of a better term, although Numerus could be suggested in its place based on later manuals like the Strategikon.
Reply
#7
Quote:one Limitanei Ripensis garrison in Oescus, numbering approximately 500, one Legio Comitatenses in Syria, numbering between 800 and 1200 men... New units, or old legions that had lost their identity, remained at about 1000 men in strength individually.

Probably, although I believe all these figures are modern estimates, and not based on any primary source.

One possible scrap of evidence for numbers, at least under the tetrarchy, comes from the Beatty papyrii from Panopolis, which give pay and provisions lists for various units based in Egypt c300. Richard Duncan-Jones (Structure and Scale in the Roman Economy) has analysed these figures to estimate numbers in the various units mentioned. The lanciarii (light infantry) detachment of Legio II Traiana numbers 439 men, and combined with another vexillation elsewhere of 554 men and a detachment of equites promoti (legion cavalry) of 77 men, the complete legion at this date, Duncan-Jones suggests, may have numbered around 1000. However, II Traiana was an old principiate legion; I don't know if we should draw from this that the old legions had been so dramatically downsized. A detachment of legionaries from the new III Diocletiana serving with the governor of the Thebaid, meanwhile, numbers 572. At least we can assume that the average legion vexillation of this date was c500 men, very close to the size of a traditional cohort.

The old auxiliary formations in the province do appear to have lost a lot of their strength though: Ala I Hiberorum based at Thmou has only 118 men (although this is only one interpretation - the figure could also be 439!). An auxiliary cohort apparently has only 164 men, while a unit of equites sagittarii numbers 121.


Quote:Some... were brought into the new system during the reign of Constantine and were made out of the old style Auxilia. They somehow became Elite units instead of Limitanei (I have no Idea how) and were probably organized in the same way Auxilia Quingenaria cohorts were.

I don't think we should draw a link between the auxiliaries of the principiate and the later Roman auxilia. The old auxiliary cohorts and alae still existed as border garrison troops, alongside but still inferior to the legions, and several turn up under their old names in the Notitia Dignitatum. The new auxilia were apparently organised as numeri (an irregular term just meaning 'number' or 'unit'), as stated on the many late Roman tombstones to field army troops at the cemetary at Concordia. Numeri had existed since the late 1st century (various Numeri Brittonum are attested from the Rhine frontier) as larger (perhaps) units of irregular troops. The auxilia of the late empire possibly derive from attempts in the late 3rd century to organise the various bodies of barbarian troops we find accompanying imperial field armies on a more regular footing, using the old numerus system. If so, their 'barbarian' origin was quickly eclipsed, and by the mid 4th century they included citizen soldiers armed and equipped as well, if not better, than the legions. That doesn't help us in guessing how many men were in a numerus or auxilium though!


Quote:Raised in the new style they probably had a Ducenarius system with Maniple-sized regiments on top of their centuries..

Although there's still no evidence, I think, for the ducenarius as a legion officer. They seem to be restricted to non-legion units of auxilia and protectores of the 'new' field armies.
Nathan Ross
Reply
#8
You're correct, I'm using modern estimates. Thank you for the references to actual numbers though.

As for the Ducenarius, I agree that its up to debate. I personally think that units raised in the 5th century were the first to begin using that officer, as it would make a perfect lead up in the evolution of the Army of the Strategikon, along with the introduction of Lance-and-Bow warfare by the Alans and Huns. There is limited evidence to suggest (but not to confirm) that they commanded a maniple sized unit, which could possibly have been called a numerus.

Thank you for your criticism though. I'm still learning as well.
Reply
#9
Quote:As for the Ducenarius... I personally think that units raised in the 5th century were the first to begin using that officer.

They were around earlier than that - an inscription (AE 1977, 00806) from Nakolea in Asia and dated to AD356 (first consulate of Julian, eighth of Constantius) describes Flavius Aemilianus as ducenarius of the numerus Ioviorum cornutorum seniorum...
Nathan Ross
Reply
#10
I wish I'd had the time to finish that 'little thesis' of mine (soon!) as defending it would have covered all of this, but I am minded to suggest the following:

The 4th/5thC legions were of two types:

- the 'old' legions of the principate, still serving on the borders as limitanei, many in the locations they had held for years, still had the theoretical structure of the 10 cohorts (the first being 5 double centuries, but with the legionary cavalry withdrawn); however they were more immobile, were effectively static garrison troops and, most likely, not kept up to strength (like the matching old auxiliary units too). Some units still garrisoned several locations, however, so must still have been fairly significant.

- the 'new' field army legio's of the Palatina, Comitatenses and Pseudo-Comitatenses, however are, in the main, the new name for the standard legion detachment/vexilatio of earlier times each the size, and probably the same organisation, or 2 cohorts (still 480+ men each). The limitanei legions were still certainly of decent quality and still supplied Pseudo-Comitatenses new legio's as necessary; some of which still never rejoined and became permanently detached.

In short I am theorizing that the Roman Army Organisation, certainly in respect of the legions, evolved very slowly over 800 or so years, and didn't really change that much, with a very consistent and conservative approach - more to follow when I've finished.

With - a century being a standard 80+3 throughout (at full strength); a cohort being of 6 centuries (the smaller version) and a vexilation/detachment and eventually field army legio being of 2 cohorts. I am hoping that in trying to defend that theory against all comers will help us get closer to the truth and see if there's real evidence that the theory doesn't hold up. Inherent in that defence is that century, guingenaria and milliaria do not mean 100, 500 and 1,000 respectively, but ancient authors may also have not unreasonably thought it.
Reply
#11
Quote:I wish I'd had the time to finish that 'little thesis' of mine (soon!) as defending it would have covered all of this... I am hoping that in trying to defend that theory against all comers will help us get closer to the truth and see if there's real evidence that the theory doesn't hold up.


While I'm intrigued by your thesis and wish you all the best in completing it, I'd be rather dubious of any theory that claims to cover the whole matter of late army organisation without some pretty spectacular new evidence to back it up!

People have been crunching numbers over this for years and years, and we have fairly believable general estimates, but as I've been suggesting there's so much that we simply don't know about the later Roman army (or any Roman army, for that matter) that unified theories are always going to be... hypothetical.

It might be better to look for evidence that the theory does stand up, rather than trying to refute evidence that it doesn't - that seems a rather legalistic than scientific approach (although generations of Roman orators would applaud you for it...)

But, as I say, I hope you get the thing done soon and then we can pick over it - I look forward to it!

:-)
Nathan Ross
Reply
#12
It's like String Theory (now called Membrane Theory). It's a pretty good guess but there is no way to test it.
Reply
#13
Quote:With - a century being a standard 80+3 throughout (at full strength);

Which is as theroretical as your guess about the strength of the late roman legion. 8+)
Nice guess. Who is your 2nd and 3rd supernumerarius beside the centurio himself? And WHY?

But I like your idea, that the new legion was simply the usual vexillatio of 2 cohorts. Makes a lot of sense, in the tradition and the usual practice during 2nd and 3rd century. IIRC there are also hints, about 2 ranks of tribunes in late roman army. That would match with brigaded cohorts lead by the higher tribune, too.

However I doubt, that the legions of the limitanei in the 4th century were still about 3000 guys. They usually lost more than one vexillatio over the decades and they often never came back. So why should the romans refill old legions up to 8 cohorts accidentally? Makes no sense to me. I am afraid limitanei legions were much weaker.

But as with almost everything about late roman legions, we have no proof.
Ut desint vires, tamen est laudanda voluntas
Reply
#14
Quote:
Mark Hygate post=342091 Wrote:With - a century being a standard 80+3 throughout (at full strength);

Which is as theroretical as your guess about the strength of the late roman legion. 8+)
Nice guess. Who is your 2nd and 3rd supernumerarius beside the centurio himself? And WHY?

Of course it is - because, apart from the Polybius construct we have nothing else of comparable detail. A century of 80 men (10 x contubernia of 8) with a Centurion, Optio and Signifer. My question back therefore, is why would this have then changed? There's no reason it should that we are aware of and therefore, without any countering information, why not suggest that it was still the same over 700 years later.
.................

Quote:However I doubt, that the legions of the limitanei in the 4th century were still about 3000 guys...................

But as with almost everything about late roman legions, we have no proof.

I do not suggest that the limitanei/border legions were all about 3,000 guys, but it is not an unreasonable figure and I can understand why it was and could be used - for many of them seem to have garrisoned multiple forts. What I will suggest, however, is that they were still the same old legions (in theory), just as the auxiliary alae and cohorts were still often the same left in place and now a 2nd Class to the Field Armies. They did still sometimes provide new temporary and occasionally permanent field army legio/detachments, so their abilities were not that far reduced and some may have still contained significant numbers.

Overall, however, the point is true - we have little or no information. So, is it not reasonable to theorize a reasonable construct and then see if it stands. It may not, it may have holes, but is it wrong to try? :errr:
Reply
#15
Quote:Overall, however, the point is true - we have little or no information. So, is it not reasonable to theorize a reasonable construct and then see if it stands. It may not, it may have holes, but is it wrong to try? :errr:

In some cases, yes, it is possible to theorize and it needs to be done. The only problem is that the information is so sparse that several theories can probably all be postulated: and all may 'stand'.

The old (now outmoded) claim was that the Roman army of the fourth century onwards was a monolithic structure with all units of a given type adhering to a strict internal format according to 'government' guidelines. However, the swift turnover of emperors and the divergent needs of the different regions probably meant that units in different areas had a different composition.

In addition, it is possible that the reforms promulgated by, for example, Diocletian, were not fully implemented. Or they might have been fully implemented, but the archaoeological and historical records of these orders has been lost and so what actually happened is poorly understood.

Further, the major problem may be that most of the surviving sources may represent government beliefs as to what should be happening. In the field, and so reflected in the archaeology, the truth may have been somewhat different.

As to why would the army change over the previous 600 years when it had been so successful and it therefore had no need to, the nature of the opposition had changed, and the empire had gone from largely offensive, terrritory-grabbing campaigns, to campaigns aimed at maintaining the current emperor in power against both internal and external opposition. As a result, he army itself would change in response.

Finally, the economic problems of the third century onwards demanded that the army change to armour and equipment that was cheaper to make and almost, if not as, effective as earlier styles. The change in equipment may have forced a change in tactics and hence a change in unit structure in different parts of the Empire.

Just a few suggestions. :-)
Ian (Sonic) Hughes
"I have described nothing but what I saw myself, or learned from others" - Thucydides, Peloponnesian War
"I have just jazzed mine up a little" - Spike Milligan, World War II
Reply


Possibly Related Threads…
Thread Author Replies Views Last Post
  Late Roman Legion size based on the Perge Inscription FlaviusB 10 237 5 hours ago
Last Post: Steven James
  The distribution of Roman military units in Dardania in late antiquity Horeum Margi 0 112 01-12-2023, 10:40 PM
Last Post: Horeum Margi
  Casualty counts in late antiquity. Flavivs Aetivs 16 2,961 10-15-2013, 03:46 PM
Last Post: AMELIANVS

Forum Jump: