Zodiac and Late Roman Army Organisation

[Nathan Ross]

I find Nicasie's book to be frustratingly unreliable.

The first adverse comment I've ever read about Nicasie! Interesting - I'm currently waiting for this one on interloan, so any specific caveats would be helpful.

I tried to cut through some of the factoids in a recent issue of Ancient Warfare magazine

And I notice that a forthcoming issue of AW is devoted to the Army of Diocletian :grin:

[ValentinianVictrix]

I'd also be interested on why Duncan is not too keen on Nicasie as I have the book and find it much more useful than Dixon & Southern or even Elton's book.

[ValentinianVictrix]

I'm sure I've read, possibly in Vegetius, that the legion size of 6000 was orginally based on tribal organisation?

[ValentinianVictrix]

once I can find a copy of Zosimus in the original Greek

Adrian,

Try this:

http://www.archive.org/details/novahistoriaed00zosiuoft

Thank you very much Michael and I will be in touch as I may need your help on this one.

[D B Campbell]

The first adverse comment I've ever read about Nicasie! Interesting - I'm currently waiting for this one on interloan, so any specific caveats would be helpful.

I should clarify that I was thinking specifically of unit strengths (the topic of this thread). He doesn't (imho) differentiate adequately between the Diocletianic and Constantinian state of affairs.

He seems to suggest that, because many ND legions are found still occupying their Severan stations, their internal organization can't have changed; but he also talks about repeated withdrawal of vexillations as a cause of shrinkage. (Where are these vexillations going, and why aren't they coming back?) There seems to be some confusion over the distribution of full-strength legions across a province (e.g. III Augusta in Africa; V Macedonica in "Dacia") and the de novo creation of self-contained mini-legions. Just my impression. :?

[antiochus]

Caballo wrote: I'd recommend Nicasie's book, Twilight of the Empire. An excellent analysis of the Late Roman Army. I have to confess, I don't follow your theory though.

Thanks for the recommendation. Sorry about not clarifying my theory. I’m still trying to work it out. However, I have put this together since my last posting. Ammianus gives units of 300 men or 500 men, as does Lydus (300 man cohort and 500 vexillations). For this exercise, I have split the 6000 men into two bodies of 3000 men. To have both 300 and 500 man units, a 3000 man unit consists of three lines (A B and C) of 1000 men amounting to 10 centuries per line.

A1 A2 A3 A4 A5 A6 A7 A8 A9 A10

B1 B2 B3 B4 B5 B6 B7 B8 B9 B10

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10

Each line is divided into two vexilliationes of 500 men. For example centuries A1 to A5 make up one vexillation numbering 500 men and centuries A6 to A10 make up another 500 man vexilliation. In this manner the 3000 men are organised into six vexillationes of 500 men. Now if you add the six vexillations from the another 3000 men, the total is 12 vexillations of 500 men. Isidore claims a legion consisting of 12 cohorts numbering 500 men. Many academics have considered Isidore to be in error but he is dead right. So a 500 man vexillation is also referred to as a 500 man cohort. To form cohorts of 300 men, the 3000 men are vertically organised into 10 cohorts of 300 men with a cohort consisting of one century from each line. For example, a cohort would consist of centuries A1, B1 and C1. Ammianus has references to 800 men to man the fleets, so this could possibly work out to one 500 man vexillation and one 300 man cohort: example a vexillation consisting of centuries A1 to A5 plus cohort six (centuries A6, B6 and C6).

Ok I did that by using the primary sources and in the process I hope I have successfully vindicated Isidore’s good name.

Nathan Ross wrote: Anyway, as mentioned on the thread above, if there was some grand mathematical plan for troop numbers within army organisation, we don't have any record of it and it certainly isn't supported by the evidence we do possess.

Actually there is evidence of the existence of a grand mathematical plan and it is discussed in snippets by the Romans in various primary sources. Some say things like the changing of the intervals, or if the republic doesn’t change. There is also one reference explaining why the Romans do not talk about the system. They are embarrassed it does not belong to them and it should. To understand the grand mathematical plan as you call it requires the reader to investigate more disciplines than just the military. The size of a cohort can tell a lot if you know how to use the information. Many ancient authorities tell us a zodiac amounts to 30 degrees and Strabo remarks that one degree equals 700 stadia. Therefore, one zodiac equals 21,000 stadia. By taking the 480 man cohort as our example, when divided by 30 degrees the result is 16 zodiacs. In this manner one man in a cohort equals one degree, so a 480 man cohort when multiplied by 700 stadia per degree equals 336,000 stadia. This number means nothing to most readers at this point, but it is the number of men in the 35 tribes. So 336,000 men divided by 35 tribes equals 9600 men per tribe. Now by following Livy claiming the number of iuniores to seniores is 50/50, the 9600 men divides into 4800 iuniores and 4800 seniores. In short, the tribe dictates the size of the legion! Next rule is the number of cavalry squadrons must represent the number of zodiacs for that time period, so an alae of 512 men divided by 16 zodiacs equals 32 men. Therefore, a 512 alae contains 16 squadrons. The number 16 represents the Etruscan system of dividing the heavens into 16 parts. At this point in their history, which is during the reign of Augustus, the Romans have theoretically run out of heaven. However, I have found two references to how they extend it.

Mr Campbell wrote: Why are your units of 1,200 men gathered into sixes, then?

The number six is based on the six property classes cited by Dionysius.

Mr Campbell wrote: What is the point of the 7,200-man unit? And why have you decided that the units should comprise 1,200 men, in any case?

The number 7200 comes from the tribal system made up of six classes of 1200 men. This is based on my research following the Roman cosmos system. I am not asking anyone to believe it, nor accept it, but from Augustus the tribal system produces the number 4800 men, then 6000 men and in 400 AD, 7200 men. The tribes advance by 2400 men (1200 iuniores and 1200 seniores. The 2400 men when multiplied by 35 tribes equals 84,000 stadia which when divided by 21,000 stadia per zodiac, equates to four zodiacs or 120 degrees. So every time the fourth zodiac passes the zenith the Romans increase the 35 tribes by a total of 84,000 men, which equates to 2400 men per tribe. However, this system only started with Augustus. Before that they used a similar but awkward system (one of three).

Robert Vermaat wrote: I have never, EVER, come across a structure such as this hypothesis, in any Late Roman unit. IF this would have been the case, we would at some point have found a description of a soldier being either this or that, serving in the same unit. We haven’t.

Do you mean we have no idea of what troop types the units in the Late Roman army are composed of?

Steven

[D B Campbell]

Why are your units of 1,200 men gathered into sixes, then? What is the point of the 7,200-man unit? And why have you decided that the units should comprise 1,200 men, in any case? :?

The number six is based on the six property classes cited by Dionysius. The number 7200 comes from the tribal system made up of six classes of 1200 men.

Would I be correct to assume that your theory requires every late Roman soldier to be enrolled in one of 35 tribes? :?

[Robert Vermaat]

Robert, could I have some examples of sources from 400 to 600 CE which talk about an ideal organization, not the strength of particular units at particular times?

Well, that's the point, it's debatable if there ever was such a thing as an 'ideal organization' of Roman units. When you read Vegetius or Maurikios (and i think the latter was far more experienced then the former, who may have been writing in theory only), they are clearly advocating ideas. Maurikios is clear on the point: you may have this number, but that number would be ideal, while this number is too many, etc. There is no 'ideal cohort' after which all cohorts are modelled. I think there is a 'practical number', like e.g. 300 or perhaps 500, but the Roman army had no 'mold' after which every unit was formed. The contubernium was 8 men, and everything above that would depend on practical things.

Even today we don't have this in black & white: how large is a division? How large a regiment? How many men in a batallion? The logical answer would be another question - what sort of batallion?

[Robert Vermaat]

Robert Vermaat wrote: I have never, EVER, come across a structure such as this hypothesis, in any Late Roman unit. IF this would have been the case, we would at some point have found a description of a soldier being either this or that, serving in the same unit. We haven’t.Do you mean we have no idea of what troop types the units in the Late Roman army are composed of?

To the contrary, we do.
I meant that, while we see inscriptions from all sorts of ranks and grades within one unit, we have never found anything like your subdivision WITHIN ONE UNIT of a subdivision into iuniores and seniores, or a vertical organisation into troop types named exculatores, lanciarii and more lanciarii. Seniores are always in another unit from the iuniores (it's not even clear what exactly they represent). Exculcatores are never a specialism within one unit. If they were, we'd need at least several soldiers from one unit describing themselves as either an exculcator, or a lanciarius, or what have you. They don't. We have a number of specialisms, and Vegetius, the Perge inscription and Lydus mention a great deal of them, but exculcatores are not among them.

[Sean Manning]

Wouldn't births, deaths, and the enrolment of those granted citizenship spoil any system that required the number of men in the tribes to be a multiple of a particular value? Every civis was supposed to be enrolled in a tribus and a (voting) centuria, and nobody controlled how many cives there were (or knew how many of the citizens the next census would record, since the error could be in tens of percent). For example, the census population almost doubled from 194/3 to 189/8 BCE (Brunt gives 143,704 and 258,318 cives for those years) but clearly the Romans didn't double the strength of their legions.

Eratosthenes' estimate of the circumference of the earth as 252,000 = 700 x 360 stadia appeared around 200 BCE. If the Romans linked their population to the circumference of the earth, what figure did they use before then?

As always, I will be interested to see you explain your theory and its supporting evidence in your book.

[Flavivs Aetivs]

Well I've read in Modern and Ancient sources and this is what I've deduced:

The Legiones Comitatenses numbered about 800-1200
The Auxilia Palatina and LEgiones Palatinae numbered 400-600
The Limitanei numbered around 400-600 garrisoned, and probably operated 800-1200 as pseudocomitatenses.
Vexillationes numbered 300 for cavalry, and varied for infantry depending on the size of the detatchment

And units were structured as such:
Contubernium - 8
Century - 60-80
Maniple (4th Century) or Numerus (5th Century) - 2 centuries (120-160 men)
Cohors - 3-4 Maniples/Numerii (360 - 640 men)
Legion - 2 Cohors (Or 8 Numerii)(800-1200 men)
Excercitus - Multiple LEgions and Cohorts of Comitatensian, Limitaneii, and Palatina troops

I deduced this, but there's not much evidence supporting any actual structural organization or unit sizes, and the distributio numerorum is just paper strength...

[antiochus]

Mr Campbell wrote: Would I be correct to assume that your theory requires every late Roman soldier to be enrolled in one of 35 tribes?

No, the tribes are a template or governing body for determining the size of a legion (via the cosmos).

Robert Vermaat wrote: Seniores are always in another unit from the iuniores (it's not even clear what exactly they represent). Exculcatores are never a specialism within one unit. If they were, we'd need at least several soldiers from one unit describing themselves as either an exculcator, or a lanciarius, or what have you. They don't. We have a number of specialisms, and Vegetius, the Perge inscription and Lydus mention a great deal of them, but exculcatores are not among them.

I found the term seniores exculcatores in the ND taken from the internet so I thought it must have been a separate unit. Now I am really confused about the ND. Do you know which year Zosimus is relating to when he cites the 6000 men is organised into five regiments?

Sean Manning wrote: Wouldn't births, deaths, and the enrolment of those granted citizenship spoil any system that required the number of men in the tribes to be a multiple of a particular value?

The number of those registered in the census is far greater than the numbers permitted by the tribal system. In the Augustine period, the census has over 5 million people, but only 336,000 people are when required chosen possibly by lot to vote in the tribes. I think it could work in the same manner as jury selection. There was a time when the number of men in the tribes did not capitalise on their available manpower, so the Romans accelerate the creation of the tribes from its cosmological timetable. The 35 tribes are created earlier than they should.

Sean Manning wrote: Eratosthenes' estimate of the circumference of the earth as 252,000 = 700 x 360 stadia appeared around 200 BCE. If the Romans linked their population to the circumference of the earth, what figure did they use before.

The Romans do not link the population to the circumference of the earth. They link it to the orbit of the fixed stars also termed the zodiac or the abode of the gods. But the system undergoes three transformations. This was the most difficult passage of the research, especially when the cosmos system does not adhere to the foundation date of Rome as its starting point. Thankfully the cosmos system no longer throws up any more surprises. There was a period there when I thought it would never end.

My focus lately has been finishing constructing the Latin levy system. I found a reference that links the Latins to using the same cosmos system as the Romans. There are bits and pieces of mathematical data on the Latins and if you take Livy’s comment that each community had to supply 120 cavalry. I multiplied this by the 30 cities on the Latin league which suggest they can levy 3600 cavalry. As Livy reports Tarquin merged Latin and Roman centuries to make maniples, this indicates they must be similar, so I have used the Roman century structure to see if I can model the Latin century structure for the 30 cities. What I have found on a century bases, the Latins produce double the number of cavalry as the Romans but the output of the 30 cities they produce three times the number of cavalry. The ratio of iuniores to seniores is different in the Latin structure and has more iuniores to seniores than the Roman system. I’ve taken the system to the Cannae levy and it shows the Latins are abiding by the same changes to the Romans system.

In 402 AD the tribal system my theory shows the tribes will produce six units of 1200 men. If no sources can back up units of 1200 men then I am happy to accept with the rise of Christianity, the traditional cosmos system that the Romans had adhered for so long has finally been abandoned.

Steven

[Macedon]

Could you make a clear, brief description of the system you propose and the evidence you base it on? I always found such theories very impractical and far-fetched albeit intruiging and interesting at a philosophical level. What eras of the Roman history do you try to encompass in it? I read the posts thus far, but I would appreciate a complete formulation of the main points of your proposal, if one is possible, before I can say anything about it.

Thanks

George

[Robert Vermaat]

The Limitanei numbered around 400-600 garrisoned, and probably operated 800-1200 as pseudocomitatenses.

Limitanei what exactly? Legiones? Cohortes?

Vexillationes numbered 300 for cavalry, and varied for infantry depending on the size of the detatchment

Well, vexillationes can be any number, of course. So if you say "varied for infantry depending on the size of the detatchment" that's of course right!
Julian sent detachments of 300 men from each unit to Constantine.
Gratian sent detachments of 500 men to storm the position of the Lentienses.
Valens sent detachments of 300 from each legion to meet the Goths at Adrianople.

Maniple (4th Century) or Numerus (5th Century) - 2 centuries (120-160 men)
Cohors - 3-4 Maniples/Numerii (360 - 640 men)
Legion - 2 Cohors (Or 8 Numerii)(800-1200 men)

What do you base that on? Maniples did ot exist in the new model Roman army, and 'numerus' just means 'unit' (generic) and can mean just about anything. There's no evidence for a 'numerus' being 2 centuries. A numerus could be an ala, a cuneus, a cohort and what have you.

[Robert Vermaat]

Robert Vermaat wrote: Seniores are always in another unit from the iuniores (it's not even clear what exactly they represent). Exculcatores are never a specialism within one unit. If they were, we'd need at least several soldiers from one unit describing themselves as either an exculcator, or a lanciarius, or what have you. They don't. We have a number of specialisms, and Vegetius, the Perge inscription and Lydus mention a great deal of them, but exculcatores are not among them.
I found the term seniores exculcatores in the ND taken from the internet so I thought it must have been a separate unit. Now I am really confused about the ND.

You did not find 'seniores exculcatores', you found 'exculcatores seniores'. As the exculcatores iuniores, they are part of the auxilia palatina, commanded by the Magister Peditum. The may hark back to one single unit of that name, later split into a seniores as well as a iniuores unit at some date, as happened with quite a number of units.

Do you know which year Zosimus is relating to when he cites the 6000 men is organised into five regiments?

Sadly, no. The publishing date of Zosimus is somewhere in the 5th century, but we don't know exactly when. He may have lived under Anastasios but his work does not continue beyond the year 410.