Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Cavalry Cohorts
#1
FROM: The Roman Eastern Frontier and the Persian Wars (AD 226±363) A Documentary History Compiled and edited by Michael H.Dodgeon and Samuel N.C.Lieu
Cedrenus, i, p. 454, 3–6, CSHB: Valerian and Gallienus reigned for fifteen years. This Valerian made war against Shapur the Persian and was captured in Caesarea with a force of twenty thousand men. He was flayed by Shapur and died. His son Gallienus established the first cavalry cohorts: for the majority of Roman soldiers till then were infantry. (Lieu)
Are there any other Primary Sources that talk about cavalry cohorts?
How many soldiers does a cavalry cohort contain?
What could be the officer organization of a cavalry cohort?
Is “cohort” a right translation for the original Greek “tagmata”?
Let’s find out the answers to this enigmatic phrase.
Reply
#2
Cedrenus was an 11th-century Byzantine writer, and the phrase he uses in this passage is hippeon tagmata - 'units of cavalry'. Tagmata was a Byzantine military term, sometimes used inexactly to describe earlier formations like legions and numeri too. But it doesn't mean 'cohort', and we shouldn't assume that it always means 'legion' either.

This passage is the origin of the idea that Gallienus reformed the Roman army and added a new corps of cavalry - see Duncan Campbell's paper Coinage and Cavalry for an argument against this theory.

Clearly the Roman army did feature cavalry forces before Gallienus - both auxiliary alae and cohortes equitata, and legion cavalry. Post-Gallienus armies (or those few we have some idea about) do not seem to feature dramatically increased cavalry components. So it's not clear what Cedrenus means by this. It's perhaps a reference to the later Roman cavalry formations called vexillationes (maybe implying they were originally detached from some larger unit or other). These became the cavalry mainstay of the later Roman field armies, and seem to have first appeared around the mid to late 3rd century; perhaps Gallienus was indeed involved in their creation?

The numbers of a cavalry vexillatio are unknown. If we can base them on the apparent size of a unit of equites promoti in Egypt c.AD320 they may have been about 300 strong; officially they might perhaps have been about the strength of an old-style quingenary ala.

However, there does seem to be a large mounted component (of 275 veredarii) listed as part of the legion (or 'numerus legionum') of Perge, c.AD500. So perhaps that's connected in some way to Cedrenus's mysterious reference...
Nathan Ross
Reply
#3
Nathan Ross:
The numbers of a cavalry vexillatio are unknown. If we can base them on the apparent size of a unit of equites promoti in Egypt c.AD320 they may have been about 300 strong; officially they might perhaps have been about the strength of an old-style quingenary ala.

There are cavalry tagmata and infantry tagmata known for the Roman Army so let’s see if we can reverse engineer ourselves to the solution.
About the Ala, how do we know that the ala is still the same for this period of the roman army as for Augustus’ time?


From:
Treadgold Byzantium and its army 284-1081

A papyrus written for a commander of an ala in the mid fourth century already refers to a centurion (hecatontarch) who evidently belonged to that cavalry regiment
Abinnaeus-Archive no 80 page 163

I wonder, could this be one of the officers for our cavalry tagmata/cohort?

From:
http://www.academia.edu/3677031/_Drungus...004_96-130

Maurice defines these formations as follows: A meros is a grouping or drungus composed of three moirai; while a moira is a body composed of tagmata or numeri or banda.


Nathan Ross:
 
Tagmata was a Byzantine military term, sometimes used inexactly to describe earlier formations like legions and numeri too

The Historia Augusta contains the term drungus/drungi. This could mean that the other terms of Maurice are also earlier then the 6th century. So tagmata could be from the Roman Army Timeframe. Also the history of Zosimus contains the term tagmata.

Nathan Ross:
It's perhaps a reference to the later Roman cavalry formations called vexillationes (maybe implying they were originally detached from some larger unit or other).

Which vexillationes do you mean? Is it the Greek bexiillation or the uexillationes?
Reply
#4
(03-04-2017, 08:39 AM)Julian de Vries Wrote: let’s see if we can reverse engineer ourselves to the solution.

I'm not sure that 'reverse engineering' is very useful in this context. We know that later writers used the terminology of their own day to describe military formations of earlier centuries, and that this terminology was not used in a specific way: Zosimus (5.45.1) refers to six tagmata of auxilia; Sozomen (9.8) calls this same body of troops arithmoi - the latter word is probably closer to the truth, as the units in question appear to have been numeri (called arithmoi in Greek).

There is quite a bit of evidence for the official names of Roman military units of the 3rd-4th century, both literary and epigraphic, not least the Notitia Dignitatum. No contemporary source, as far as I know, calls any military unit a tagma (pl. tagmata). So we probably shouldn't start imagining that they existed because Greek writers centuries later used the word!


(03-04-2017, 08:39 AM)Julian de Vries Wrote: About the Ala, how do we know that the ala is still the same for this period of the roman army as for Augustus’ time?

We don't (my note above was intended to refer to the original strength of an 'old style' ala, i.e. in the principiate). Alae still existed as part of the frontier limitanei, but they may have numbered fewer than a couple of hundred men each (again based on Egyptian papyri).


(03-04-2017, 08:39 AM)Julian de Vries Wrote: a commander of an ala in the mid fourth century already refers to a centurion (hecatontarch)

Interesting. Abinnaeus (apparently) was the commander of one of these old alae. As far as I know, there's not much evidence of the ranks used by later alae (beneath praefectus as unit commander). 'Hecatontarch' could be a reference to the later rank of centenarius, used by new-style cavalry equites, the guard cavalry of the scholae and the palatine auxilia. Alternatively, perhaps these later alae could have adopted infantry ranks... (we have the equites promoti of II Traiana commanded by an ordinarius (centurion), I think, although the legion cavalry always did use an infantry rank system, it seems).


(03-04-2017, 08:39 AM)Julian de Vries Wrote: The Historia Augusta contains the term drungus/drungi.

And so does Vegetius. It does seem that the army of the later 4th century adopted several words of 'barbarian' origin around this time - fulcum would be another one, I suppose.

The Historia Augusta was probably written around the same time as Vegetius (although we don't have a definite date for either, or for the ND!) and contains a lot of anachronistic terminology.



(03-04-2017, 08:39 AM)Julian de Vries Wrote: Which vexillationes do you mean? Is it the Greek bexiillation or the uexillationes?

Latin vexillationes - or uexillationes if you prefer. Which may or may not be the same thing as a numerus equitum.
Nathan Ross
Reply
#5
The cohors Germanorum is a horse guard. The cohortes speculatores are probably also cavalry. I believe the stablesiani also have a cohort in some inscription.

Also there is Ammianus:

Ammianus 14,2,12 and 24,5,8.10 states cavalry cohorts.
29,5,20 states the cohors quarta equitum sagittariorum
21,11,2 states that a sagittariorum cohors (under a tribunus equitum turmae).
16,12,7 states that the Sagittarii are a equestris turmae.

One theory I am contemplating:
The Greek term speira is attested as cohort. Speira=cohort
From RAT:
However, the most direct account comes from Etymologicum Gordianum which at the "speiros" entry writes :
καὶ σύνταγμα στρατιωτικὸν ἀνδρῶν τριακοσίων· ἡ οὖν σπεῖρα καὶ ὁ χιλίαρχος ὡς ποῦ φησὶν, οὐχὶ πάντας εἶχεν ὁ χιλίαρχος τοῦ οἰκείου τάγματος, ἀλλὰ τοὺς τὸ τ μόνους

This translates as "A military unit of 300 men, known as speira and the chiliarch who, as they say, the chiliarch did not command all in the tagma (a term that can mean any unit but usually used as a synonym for legion when used for the time in question) but only the 300.
Could this perhaps be the 300 equites promoti of 320AD?
And mysteriously there is the tagma again.
Nathan Ross:
I'm not sure that 'reverse engineering' is very useful in this context. We know that later writers used the terminology of their own day to describe military formations of earlier centuries, and this this terminology was not used in a specific way: Zosimus (5.45.1) refers to six tagmata of auxilia; Sozomen (9.8) calls this same body of troops arithmoi - the latter word is probably closer to the truth, as the units in question appear to have been numeri (called arithmoi in Greek).
Who says the terms tagmata and arithmoi are not connected in some way.
Suda σ 933 and Photius:

Speirai’ plethe strateumaton, phialagges, noumera, legeon.
Numerus=arithmos=tagma=speira=cohort???
Reply
#6
(03-05-2017, 10:42 PM)Julian de Vries Wrote: The cohors Germanorum is a horse guard. The cohortes speculatores are probably also cavalry.

Were they? The 'Germanic bodyguard' might have included horsemen, the speculatores of Marcus Antonius and Octavian maybe likewise. But I don't think either could be called a 'horse guard'. Cohorts could include cavalry though - the auxiliary cohortes equitata did.


(03-05-2017, 10:42 PM)Julian de Vries Wrote: I believe the stablesiani also have a cohort in some inscription.

Not that I know of - do you have a reference? You might be thinking of the Deurne helmet inscription (AE 1927, 00153) - Stablesia(na) VI. This probably refers to Equites sexta Stablesiani, a new-style cavalry vexillatio.

A couple of other inscriptions, probably referring to a similar unit at around the same time, call it vexil(latione) eq(uitum) Stablesianorum (CIL 05, 04376) and n(umeri) eq(uitum) Stabl(e)s(ianorum) (AE 1974, 00342) - which I think underlines that late Roman unit terminology was often a lot more fluid than we might like!


(03-05-2017, 10:42 PM)Julian de Vries Wrote: Also there is Ammianus:

Yes, Ammianus is tricky. He does refer to cohorts, and implies that some of them are cavalry - but he uses a lot of words in an archaicising way, the 'grand style' that he mentions in his last line. Claudian does this too, referring at one point to a palatine legion as a 'cohort'. So it might be that, by the late 4th century, cohort was no longer a strict term for a particular military unit (aside from the old 'auxiliary' cohorts of the limitanei), but could be used more generally, as numerus and even legion were beginning to be used around the same time.


(03-05-2017, 10:42 PM)Julian de Vries Wrote: Ammianus 24,5,8.10

This one is particularly interesting. A 'cohort' led by a tribune is attacked and performs badly, so Julian "...reduced the surviving members of the cohort... to the infantry service (which is more burdensome) with loss of rank" (reliquos ex ea cohorte... ad pedestrem conpegit militiam, quae onerosior est, dignitatibus inminutis).

It could be that this was a cavalry unit, and Amm is using 'cohort' generally as he often appears to do. Alternatively, he could be referring to a mixed unit of cavalry and infantry, like the old cohors equitata. The different pay grades shown on the Perge inscription seem to show that horseman and infantry could coexist within the same units.


(03-05-2017, 10:42 PM)Julian de Vries Wrote: Could this perhaps be the 300 equites promoti of 320AD?

Unlikely, I think. The figure comes from Karl Strobel's interpretation of papyrus Columbia 7.188, and the number is 264 men. Strobel suggests that this is about half of the full cavalry complement of the legion, which he thinks should be about 518 - similar to a quingenary ala.


(03-05-2017, 10:42 PM)Julian de Vries Wrote: And mysteriously there is the tagma again.

Not so mysterious! The Etymologicum Gudianum (correct spelling) dates from the 10th century, so again the author is using the terminology of his own day. But 'speira' does seem to have been used, generally and perhaps not exclusively, by Greek writers to label what in latin would be called a cohort.


(03-05-2017, 10:42 PM)Julian de Vries Wrote: Who says the terms tagmata and arithmoi are not connected in some way.

Clearly they are, if Zosimus and Sozomen could use both words to refer to the same units. That does not mean that there were particular units in the 3rd-5th centuries called exclusively by these names.


(03-05-2017, 10:42 PM)Julian de Vries Wrote: Speirai’ plethe strateumaton, phialagges, noumera, legeon.

How would you translate this line?


(03-05-2017, 10:42 PM)Julian de Vries Wrote: Numerus=arithmos=tagma=speira=cohort???

Numerus=arithmos is pretty clear, I think; the words were synonymns, and both appear in official inscriptions.

Speira=cohort seems a safe enough translation too.

But tagma could refer to either a legion or a numerus, so had no certain definition in 3rd-5th century terms. I suppose it could have referred to a cohort, but since 'cohort' by this date also seems to be used generically (by Ammianus and Claudian), that doesn't help much!

So there's still no link between Cedrenus's hippeon tagmata and 'cavalry cohorts'.
Nathan Ross
Reply
#7
Some more Ammianus research.

Cavalry Tribunes

Ammianus 16,11,6 and 16,12,39 and 18,6,16 and 21,6,6 and 21,11,2 and 24,3,1 states that turma are commanded by a tribunus.


29.5.24 states the tribunus of a cohort.

Cataphractarii 16,12,63 sagittarii 29,5,24 promoti 15,4,10

Turmae 15,4,10 and 16,11,6 and 16,12,39 and 21,11,2 and 24,3,1

Ammianus 25,1,8 could be interpreted to state multiple cavalry tribunes.

Cohorts could include cavalry though - the auxiliary cohortes equitata did
Alternatively, he could be referring to a mixed unit of cavalry and infantry, like the old cohors equitata.

What is strange is that there are only 2 cohortes equitatae in the ND (Or. 34.43. ; 38.36) Secondly, it seems according to some secondary sources that certain numeri units were part-mounted.

Were they? The 'Germanic bodyguard' might have included horsemen, the speculatores of Marcus Antonius and Octavian maybe likewise. But I don't think either could be called a 'horse guard'.
Not that I know of - do you have a reference? You might be thinking of the Deurne helmet inscription (AE 1927, 00153) - Stablesia(na) VI. This probably refers to Equites sexta Stablesiani, a new-style cavalry vexillatio.

I try to state the primary sources when possible, but sometimes it is difficult to know the primary sources from the secondary Academic books.
Reply
#8
(03-07-2017, 12:44 PM)Julian de Vries Wrote: Ammianus 16,11,6 and 16,12,39 and 18,6,16 and 21,6,6 and 21,11,2 and 24,3,1 states that turma are commanded by a tribunus.

Yes, I noticed that in your last set of references. It is very strange - traditionally, of course, quingenary auxiliary cohorts were commanded by praefects, while tribunes commanded only the milliary ones. A turma was around 30 men, so certainly not a tribune's command! Whatever does Ammianus mean by this?

In the ND, it does seem that the limitanei cohorts (like the field army legions, equites, scholae and auxilia) are now led by tribunes. Praefecti command the limitanei legions, the alae and the numeri of 'milites'.

So 'tribune' seems to have become a rather variable sort of rank. Even so, having one command a 30-man turma (traditionally a decurion's role) seems a bit much.

I wonder, actually, whether the commanders of all these units were more usually called praepositii (this seems the most common designation in inscriptions), and 'tribune' was a sort of courtesy title, perhaps?

Turmae seem to have fallen out of use by the 4th century anyway - I'm not aware of any contemporary references to them outside Amm; do you know of any? - so maybe a bit like cohorts, in the field armies at least, this was an obsolete term used in a generic way.

Ammianus's 'grand style' I take to mean that he's writing in the manner of Tacitus. So perhaps he's using words like cohort and turma (and maniple, and Parthians!) because Tacitus did? We should remember that Amm himself was a Greek ('and a soldier', as he tells us at the end), so while he's writing his history in Latin he may have been 'translating' more accurate contemporary military nomenclature that he was familiar with in Greek (arithmoi, etc) into slightly archaic Latin.
Nathan Ross
Reply
#9
Meanwhile... I've been going through the Latin text of Ammianus to try and work out whether there's any connection between his uses of 'cohort' and 'turma' and other unit names.

Cohort is quite hard to work out - he seems to use the word fairly indiscrimately to refer to infantry units, and sometimes mounted units, that are not legions. It would seem to be a synonym for numerus, therefore (which could be either). Although he does (in Book 17) mention a 'Praetorian cohort' guarding Constantius II during the battle with the Limigantes - so I think in that case at least we can safely assume that he's using an anachronistic expression.

However, 25.1.7-9 gives an interesting example of the mutability of his military terminology:

Eodem die Tertiacorum a equestris numerus a legionibus incusatus est quod, cum ipsae hostium adversas inrumperent acies, illi paulatim dilapsi alacritatem paene totius minuissent exercitus. 8 Unde ad indignationem iustam imperator erectus ademptis signis hastisque diffractis omnes eos, qui fugisse arguebantur, inter inpedimenta et sarcinas et captivos agere iter inposuit, ductore eorum, qui solus fortiter decertarat, aliae turmae adposito, cuius tribunus turpiter proelium deseruisse convincebatur. 9 Abiecti sunt autem sacramento etiam alii quattuor ob flagitium simile vexillationum tribuni: hoc enim correctionis moderamine leniori inpendentium consideratione difficultatum contentus est imperator. 

Here we have a cavalry unit (the 'Tertiaci')* referred to as an equestris numerus; their leader (ductor), who had fought bravely, is given command of another (similar?) unit, referred to as a turma, whose cowardly previous leader was a tribune. We are then told that four other cavalry tribunes were also dismissed - this time their units are calledvexillationes. There is no indication, I think, that these were different types of unit.

So it would appear that Ammianus uses numerus equitum, turma and vexillatio interchangeably, and while the first and last terms would have been in official use, the second was either a colloquialism or deliberately archaic. I expect his use of 'cohort' follows a similar pattern.


* The 'Tertiaci' are possibly the numero tertio stablesiacorum / equites tertii stablesiani, listed in the ND under the command of the Magister Equitum per Orientem; Philip Rance has determined their existence into the 6th century, possibly based at Cyrrus in Syria ('The Third Equites Stablesiani at Cyrrus', Chiron 42, 2012)
Nathan Ross
Reply
#10
Nathan Ross:

I'm not sure that 'reverse engineering' is very useful in this context.
 
Road block.
 
Julian wrote:
Ammianus 14,2,12 and 24,5,8.10 states cavalry cohorts.
 
Livy uses the term cavalry cohort for 499 BC, and there are other references. My finding is for the republic a cavalry cohort appears to represent a unit that does not conform to the standard squadron organisation. This means the number of men cannot be divided into the standard squadron size.
 
Julian wrote:
In regard to Columbia 7 188. My conclusion is any cavalry numbers before 324 AD belongs to the Vegetius legion, so 264 equals 8 squadrons each 33 men, or two vexillations each of 132 men. I’ve been sitting on the fence with this one for some years, waiting for other information to support it or debunk it.
 
However, last week was a significant week in relation to my research. I’ve never seen the connection before, but the number of centuries in a tribe represents the number of Roman generations that have past. A Roman generation as stated by Dionysius lasts for 35 years, which is the same number of Roman tribes, and which both seem to match the Pythagorean 6:8:9:12 musical tetrachord (6 + 8 + 9 + 12 = 35). This made me look at the ratio of voting centuries to levy centuries in the tribal system and by changing the ratio from 4:1 to 2:1, the results were informative.
 
One tribe can produce 6 cavalry cohorts under the command of a tribune, and the reason why I have termed them tribune cavalry cohorts is each cavalry cohort has 10 centuries, and Livy’s cavalry cohorts of the early republic also had 10 centuries. Next these same tribune cavalry cohorts, by following the vexillation system are of the same size. Now these conclusions have not been pulled from thin air, what makes me excited is the data also conforms to the cavalry data from the primary sources, in which Ammianus is a gold mine. However I still do not understand how Ammianus describes 700 cavalry as being organised into two turmae. Maybe the term turmae no longer applied to a squadron, but I cannot see the reason as why they would do this.
 
Julian wrote:
Who says the terms tagmata and arithmoi are not connected in some way.
 
I would agree. However, my conclusion is Zosimus creates the confusion by terming units of various sizes a tagmata. So if a unit had 500 men, then Zosimus calls it a tagmata. If two units of 500 men are together than Zosimus calls that a tagmata of 1,000 men.
 
Julian wrote:
Numerus=arithmos=tagma=speira=cohort???
 
If a tagmata was 600 men, what would a 300 man unit be called?  

Also let's not forget we have two vital pieces of contradictory references in Zosimus and Sozomenos to the same event with Zosimos having 6 tagmata of 4 myriads and Sozomenos of 6 arithmoi of about 4,000 men. The difference between 4 myriads totalling 40,000 men and 4,000 men is a ratio of 10:1, and one can clearly see that Zosimus has made a mathematical mistake, in fact two mathematical mistakes to arrive at 40,000 men. Also Zosimus’ other reference of 5 tagmata numbering 6,000 has been rounded. That is the reason why Sozomenos states “about 4,000 men” and not 4,000 men. That is a major clue.
 
Reply
#11
(03-10-2017, 03:46 AM)Steven James Wrote: Zosimus creates the confusion by terming units of various sizes a tagmata.

Yes. But it's not just Zosimus - Cedrenus also uses inexact terminology, as does Ammianus and every other late antique writer. We need to distinguish the vague, colloquial or anachronistic terms used in literature from the official terminology of the day.

Luckily we have the evidence of inscriptions and the lists of the ND to tell us what Roman military units of the later period were actually called.

By the end of the 4th century, units of any size were called numeri (Greek arithmoi) - the term is near ubiquitous, and covers anything from a legion to a small unit of cavalry or frontier limitanei. Hence the apparent confusion in the literary sources: they are using the same words to describe units of variable types and sizes.


(03-10-2017, 03:46 AM)Steven James Wrote: If a tagmata was 600 men, what would a 300 man unit be called? 

From Maurikios's Strategikon:

"A moira is made up up tagmas, arithmoi or bandons. A count or tribune commands the tagma, arithmos or bandon." (Strat, 1.3)

So the terms were interchangeable in the 6th century - I suspect bandon was the same sort of cavalry unit as a vexillatione (named for its banner or standard); arithmos (numerus) was the same thing. Either could be referred to by later writers as a tagma, or (presumably, by Ammianus) a 'cohort' or 'turma'.

"Tagmas should be formed in strength from three hundred to four hundred at most, and counts, also called tribunes... should be placed over them. ... The tagma should not exceed four hundred men, except in the bandons of the Optimates... All of the tagmas should definitely not be the same strength. If they are, the enemy can easily determine the size of the army by counting standards. Still, the statement we have made above should be observed, that the tagma should not contain less than two hundred men or more than four hundred." (Strat, 1.4)

So here we have the system of the 6th century - tagmata aka arithmoi (aka numeri!) of 200-400 men. We know from Zosimus that the units he calls tagmata of earlier centuries could number over a thousand: these are numeri of the 4th century type. But there was, I think, no unit in the Roman army of the 4th century officially called a tagma.

So Ammianus's 'cohorts' and 'turmae' are numeri (either equitum/vexillationes or auxilia).

Zosimus's 'tagmata' and Sozomen's arithmoi are also numeri (legions or auxilia)

Cedrenus's hippeon tagmata are also numeri - the numeri equitum or vexillationes that formed the mainstay of the later Roman cavalry force.
Nathan Ross
Reply
#12
http://www.academia.edu/12008370/Cavalry...n_MA_2015_

while literary, epigraphic, and papyrological sources routinely blur particularities with generic Latin or Greek vocabulary for “regiment” or “unit”, such as numerus, ile, tagma or katalogos.

I do not agree with this. I think that these kind of terms are particular tactical unit formations and have particular unit sizes. Also why would the Ancient Sources consistently name peoples officer ranks and status. Rank and status could also be an indication of unit size.

Nathan Ross
Although he does (in Book 17) mention a 'Praetorian cohort' guarding Constantius II during the battle with the Limigantes

Yes, this is intriguing. The contemporary theory states that Constantine disbanded the praetorians. But maybe the praetorians were reformed? (Just as Caracalla recreated the equites extraordinarii.)

Another theory could be that the “Praetorian cohort” is actually the cohors Romana Palatina (AE 1934 157).
Reply
#13
(03-10-2017, 06:33 PM)Julian de Vries Wrote: http://www.academia.edu/12008370/Cavalry...n_MA_2015_

That's a good paper - thanks. I'd never heard of a bandwo! Another 'barbarian' term to join drungus and fulcum, it seems... Shame the author doesnt mention Ammianus and his 'cohorts'.

I do agree that literary sources in particular 'blur' things by using generic terms, but I don't believe that numerus was one of those terms. It was, I think, an entirely official term for late Roman military units of any size, near-ubiquitous by the end of the 4th century.

We have the explicit testimony of Maurikios (above) that the later tagmata were not of particular sizes (All of the tagmas should definitely not be the same strength) - not surprising, then, that later Greek writers used this term so broadly to describe earlier formations of varying sizes.


(03-10-2017, 06:33 PM)Julian de Vries Wrote: The contemporary theory states that Constantine disbanded the praetorians. But maybe the praetorians were reformed?

We have pretty good contemporary evidence for the disbanding of the Praetorians, and indeed they are never heard of again - except in this comment by Ammianus (and another from him about a formation called 'praetorian camp', which might have been a traditional term).

Very unlikely, I think, that they were ever reformed. Later Roman emperors had a number of equivalent guard formations that filled their role: scholae, protectores and candidati. No doubt it was one of these that Ammianus was referring to, using an archaic term in keeping with his 'grand style'.

The inscription is interesting, although too fragmentary to draw much from and lacking a date. I think Ammianus uses the term 'palace cohort' to refer to the functionaries and officials of the court, so perhaps this is the meaning here too.
Nathan Ross
Reply
#14
Nathan wrote:

By the end of the 4th century, units of any size were called numeri (Greek arithmoi) - the term is near ubiquitous, and covers anything from a legion to a small unit of cavalry or frontier limitanei. Hence the apparent confusion in the literary sources: they are using the same words to describe units of variable types and sizes.
 
I cannot accept that. That may be the case with some of the ancient historians, but I do not believe it applied to the army. It would create too much confusion.
 
Sozomen's 6 arithmoi are all of the same size. Zosimos’ 6 tagmata also have the same size but the difference between Sozomen and Zosimos is because Zosimos has calculated a century as being 100 men, while Sozomen has the right number of men in a century.
 
Zosimos refers to cavalry as a tagmata so he must relate this to a certain number or multiplications of that number as his basis for a tagmata. He’s working from a set organisation. I believe it is the number of centuries in an infantry unit and a cavalry unit that Zosimos has been working with. So for example, if an infantry tagma has the same number of centuries as a cavalry tagma, but the difference is the infantry century is larger in size than the cavalry century. In this manner, in Zosimos’ calculations, both the infantry unit and the cavalry unit are a tagma because both units have the same number of centuries.
 
Nathan posted:
"All of the tagmas should definitely not be the same strength. If they are, the enemy can easily determine the size of the army by counting standards." (Strat, 1.4)
 
I don’t think the author of the Strategicon knows what he is talking about in this matter. An ancient army not following a fixed size for their units would create confusion within the army. Why didn’t the Romans practice this if it concealed an army’s strength to the enemy? The simple fact of deploying and the frontage the army takes up can be an indicator of strength. Also the enemy has scouts and can get a good idea of an armies’ strength.
 
Nathan wrote:
But there was, I think, no unit in the Roman army of the 4th century officially called a tagma.
 
That makes sense.
 
Nathan wrote:
So Ammianus's 'cohorts' and 'turmae' are numeri (either equitum/vexillationes or auxilia). Zosimus's 'tagmata' and Sozomen's arithmoi are also numeri (legions or auxilia)
 
Then following this logic, Sozomen's 6 arithmoi (legions) amounting to about 4,000 men would mean a legion numbered around 666 men. This seems rather small to be a legion. Maybe the term numeri should be interpreted to mean part of a legion (the numeri of the legion). So for example, men could be selected from the numeri of the legion, and those men formed into another numeri that is then removed. To do this a numeri had a fixed size.
 
Therefore, in relation to the legion, I have the following units in order of the largest size to the smallest beginning with the vexillation, the tribune cohort, the numerus (which is a reduced vexillation), the standard cohort, the maniple and the century. As for auxiliary units they are based on the numeri, maniple and century.
 
For the cavalry, following Ammianus’ term: turmae, half turmae, vexillation, tribune cohort, and numerus. The infantry numerus has the same number of centuries as the numeri equitum but not the same number of men as the numeri equitum. The infantry vexillation has the same number of centuries as a cavalry vexillation but not the same number of men as the cavalry vexillation.
Reply
#15
(03-11-2017, 04:03 AM)Steven James Wrote: "All of the tagmas should definitely not be the same strength..."
 
I don’t think the author of the Strategicon knows what he is talking about in this matter.

If we accept that Maurikios was indeed the author, then he certainly knew what he was talking about - he was an experienced military commander turned Roman emperor!

The Strategikon is partly, I think, an attempt to regulate and organise what had become by that point a very ad-hoc army structure that seems quite incoherent to us (and perhaps even to Maurikios - the point about confusing the enemy does seem rather unconvincing!)

The process that created this structure began in the third century, accelerated under Diocletian and reached its definite form under Constantine and his sons. Essentially what we see is the breakup of larger formations (legions, milliarian cohorts, perhaps alae too) into smaller tactical units. Some of these tactical units kept the names of their parent formations, while others gained new names. Some were perhaps later subdivided, or recombined to make new units.

Thus we have the situation, by around the mid 4th century, of an army composed of many smaller units of varying sizes, some infantry and some cavalry, some perhaps a mix of both. The old 'building blocks' of the principiate army - legions, cohorts, alae - had been broken down, and in their place we have a much more flexible array of subunits and new formations, all of them referred to by the umbrella title 'numerus'. (c.f. Sozomen Ecc.Hist I.8 - "tagmata (ie legions) are now called arithmoi (ie numeri)")

This may seem to us incoherent and confusing - and it certainly leads to a lot of random-seeming nomenclature in historical sources - but to the military commanders of the day it may have been quite a practical solution to tactical necessities. When the late Romans needed to put a campaign force together, they could gather as many of these smaller units as they needed (or subdivide larger ones - as we see repeatedly in Ammianus with certain hundreds of men being sent here and there) and brigade them together under a commander of appropriate rank - dux, comes or magister - to respond to requirements.

As for the size of the units, I find Maurikios's statement that the 6th-century tagma had only an upper and lower size limit quite compelling; if this was drawing on earlier practice, we could well see an echo here of the (slightly larger) 4th-century numerus.


(03-11-2017, 04:03 AM)Steven James Wrote: Sozomen's 6 arithmoi (legions) amounting to about 4,000 men would mean a legion numbered around 666 men.

These particular numeri seem to have been auxilia. A numerus auxilium of between c.400 and c.800 men seems quite plausible. Apparently they were not all the same size, or they would produce a round number!



(03-11-2017, 04:03 AM)Steven James Wrote: the vexillation, the tribune cohort, the numerus (which is a reduced vexillation), the standard cohort, the maniple and the century.

This seems unnecessarily complex, and unsupported by evidence as far as I can see. A vexillation by this date was a cavalry formation, there is no such thing as a 'tribune cohort' (tribunus cohortis is the tribune of a cohort, i.e. the commander), a numerus was an independent unit not a subdivision, and it seems very unlikely (outside Ammianus and his 'grand style') that cohorts and maniples had any place in the internal structure of later military formations - although I accept that opinions differ on that last point!

I think a lot of our confusion and debate on this subject over the years stems from our attempts to force what was perhaps a very simple, flexible, practical and dynamic late Roman military doctrine into what we think the structure of a 'proper' Roman army should be.
Nathan Ross
Reply


Possibly Related Threads…
Thread Author Replies Views Last Post
  Cohorts, Cavalry and Whatever Goes antiochus 84 13,683 05-13-2013, 02:48 PM
Last Post: antiochus

Forum Jump: