\'Hyginus\' Cavalry Units

[Mark Hygate]

Mark Hygate:

The inscriptions mentioned above exclude the possibility that were three decurions, duplicarii and sesquiplicarii in this turmae. In CIL 06, 30720 we have one of each:

.......

Well, yes, it certainly would seem to argue against it - but I won't give up yet!

There is an interesting section in Maurikos Strategikon ...... "Reserve horses" .....

The extra horse for the higher ranks are frequently shown on their tombstones.....

Yes, my studies have never extended far into the 'formal' Byzantine period, but am not surprised that this does get done and noted by this time as the cavalry became the primary arm over the infantry and much more numerous than ever before. I don't disagree that the Romans earlier would also have needed spare horses, but that they don't seem to get mentioned as being with the troops much, if at all.

I agree that a fixed turma size makes more sense, however with 36 ordinary horsemen in an milliaria turma you get 864 horses which is closer to Hyginus 832 than 720 with 24 turmae of 30 men. Just because something sounds unfamilar to us shouldn't be sufficient to exclude the possibility that the Romans thought otherwise when archeology seemingly supports the ancient source.

Unlike the Greek Tactician or Hyginus, Arrian commanded a real Roman army with several Alae as governor in Cappodocia and inspected this units regularly when in garrison. He surely knew there strength and organization.

Oh indeed, and it's this wider knowledge that I've been seeking as counters to my thoughts. In fact a lot of my theory on the cavalry unit size does stem from early thoughts around the desperate search for clues as to what 'milliaria' meant and how each unit could be considered to be, at least close to, 1,000 and all the permutations that stem from it.

However, all the answers that increase sub-unit sizes to match (infantry included) are very counter-intuitive to a reflective view of how the Romans (and Ancients in general, although there are more modern parallels) fought. Sub-units had to be distributed around the battlefield, moved into position, rotating maniples (sic); let alone the beautifully simplistic (not Hyginus!) approach to making camps and even permanent forts. All those argue for standardising sub-unit size - and all I have learned helps me to believe the Romans were very bought into that approach.

The original maniple fought 20-men wide by 6 deep (with Velites in front and then behind). I believe the early Turmae was therefore 10 wide by 3 deep to match (and is indicated in that organisation that we see). That was the situation in ~300BC.

Come to 500-600AD and we have the later Roman and then early Byzantine organisations, which, for the infantry seem to lead towards the Numerous/Arithmoi of, quite possibly, 4 centuries of 320 deployed 20 x 16 (spears and archers) and the Cavalry, where there has been a need for tactical change (by Byzantine times) to increase the 3 ranks to 5 to incorporate 2 ranks of archer-cavalry and so, therefore the sub-unit has increased from 30 to 50, but still deploys 10-wide. There would seem to be a great deal of consistency, although I note that the Arithmoi has also been assumed to be 256 men to hark back to the earlier Greeks, but I am arguing for evolution rather than change.

More tellingly for me, we have, at the beginning of the period, the Early Republican operational unit size of cavalry as 300. 800+ years later we have the later Vexillationes and Byzantine Tagma at 300. I do wonder why there would necessarily have been any real change in the middle - what's the reason?

Now, yes, Arrian was on site at the time and we might reasonably expect him to know - I certainly can't refute that. What I seriously wonder at, however, is that application of management theory that points out that a 'leader' can only effective manage between 6-10 (possibly 12) subordinates at any one time; and is behind every incidence of hierarchy organisation with which I am familiar. This has never been more true than on the battlefield. It is no surprise to me that the legionary contubernia is 8-men and the British Army squad building block is still the same. The Roman army definitely seems to be built with this in mind.

I therefore, given I believe there to be too much cavalry in the Roman Army attested by adding together all the units and multiplying by 500 or ~800; do seriously question how a Praefect/Tribune could possibly control 16, or worse 24, sub-units on the field of battle, let alone use them operationally. Certainly I am given to understand that there is no evidence of any intermediate command hierarchy.

It just doesn't seem logical. Certainly I know of, after armies were once more regularly organised some 500 years ago, no cavalry units (Regiments would seem to be the most common term) that approaches such size. Even the largest at ~450 men had the intermediate level of the 'Squadron'.

So yes, there is indeed evidence that can be used to support what we see written in Hyginus and Arrian, et al, but that doesn't mean it makes much sense - at least to me. 8)

[D B Campbell]

As an aside, for anyone interested in the Latin text of Vegetius:

Epitoma Rei Militaris (ed. Carl Lang); published by Nabu Press; ISBN 1-142-06107-8; £16.53

Lang's first edition of the Epitoma was published in 1869, with a revised edition in 1885 (reprinted 1967). It is not clear which of them this is. The latter was the standard edition for over 100 years, until superseded by those of A. Önnerfors in 1995 and M.D. Reeve in 2004 (I say nothing of the heavily criticized edition of L.F. Stelten, published in 1990). The advantage that both of Lang's editions have over their successors is a splendid index verborum, listing and referencing by page and line number virtually every word in the text. I find this absolutely invaluable for navigating my way about the work. I would not be without it, although I use Reeve for the purposes of quotation. This volume seems a relatively cheap way of acquiring a still serviceable text with a valuable tool that can be used in conjunction with more modern editions.

Here's an even cheaper way: http://www.archive.org/details/flavivege...00vegeuoft

[Renatus]

And that way you can be certain of getting the 1885 edition. It seems to be well scanned, too.

[D B Campbell]

The idea of a larger ca. 40 - 42 strong turma is older than the excavations in Heidenheim, which seemingly provided proof.

I cannot lay my hands on the relevant report, but the plan of Heidenheim that I have (in Die Römer in Baden-Württemberg) makes it seem unlikely that it provided proof of a large-sized turma. I hope that someone will prove me wrong!

Basically the idea is that an Ala milliaria with 24 turmae can only have around 1000 soldiers with larger turmae. He quotes an inscription from Coptos, CIL III 6627, describing the composition of a vexilation. Among soldiers from other units there are 424 equites from 3 Alae with 5 decurions, 1 duplicarius and 4 sesquiplicarii sugessting that each of them commanded around 42 men.

The inscription was republished as CIL III 14147, where DEC(uriones) V ("five decurions") was re-read as DEC(uriones) VI ("six decurions") -- which sort of spoils the arithmetic.

This photo is no help at all, but at least it gives you an idea of how difficult the sources of evidence can be!
[attachment=3267]ILS2483-Alae.jpg[/attachment]

It's probably worth emphasizing that, in the grand scheme of things, the ala milliaria is a minor player. There were only ever seven or eight of them, each seemingly created by enlarging an existing ala. My thoughts are that, if you've decided to enlarge an ala, you probably start by deciding on the maximum manageable number of turmae to entrust to the one man (who could, thereby, become as powerful as a legionary commander, with similar opportunity to wreak havoc). So you decide on half as large again: 16 + 8 = 24. Then you start to enlarge each turma, as and when the manpower becomes available. Whether you ever reach a standard number of 40 men, 42 men, whatever, is a moot point. (Just my sixpence-worth! :wink: )

[D B Campbell]

So you decide on half as large again: 16 + 8 = 24. Then you start to enlarge each turma, as and when the manpower becomes available. Whether you ever reach a standard number of 40 men, 42 men, whatever, is a moot point.

It's maybe worth noting that Hyginus says that the ala milliaria has 1,000 horses, not 1,000 men. And also that the 1,000 horses includes the 96 remounts required for the officers. I don't know why he should make that distinction, unless he knew that the manpower fell short of the thousand implied by the name.

[Mark Hygate]

Duncan,

Many thanks for the extra information, from which I'd like to ask two further queries, if anyone can answer.....

Firstly, do we have any genuine lists of the Auxiliary units (like, for example, the inscription ILS2288, which I saw in your 'Roman Legionary Fortresses' book)? Or, is all the information that, eg Cheesman 1914 and others, put together from 'other' various evidence, mainly perhaps from archaeological finds near permanent bases?

If the latter and perhaps factoring in the possible likelihood that milliaria units would be the ones used for offensive operations, where smaller ones would be for garrison areas - is it possible that there were more milliaria units than those attested, simply because they left less evidence? Especially if these units were the first to transition to the later Cavalry/Vexillatio units in the case of the Ala?

Secondly, and returning simply to the infantry-based forerunner. It doesn't seem contentious that the evidence for the Coh M refers to ten centuries of 80 men, equalling 800. Is it therefore not quite possible that milliaria doesn't mean 1000 men, but 10 x '100'? The fact that, I believe, it commonly became assumed that it did, even to early historians, is a different factor.

[D B Campbell]

Many thanks for the extra information

Nagging at the back of my mind, I vaguely recalled an article by David Kennedy, in which he tried to analyze the numbers in CIL III, 6627/14147 -- the inscription we noted earlier. (It's in Journal of Egyptian Archaeology 71, 1985.) His numbers are skewed, because he believed that cavalrymen bunked 6 or 8 to a room. (We're now pretty sure that it's only 3 to a room.)

But his arithmetic is interesting, because he thinks that, to assemble the full work party (the inscription also lists legionaries from six cohorts and soldiers from a cohors equitata, all engaged on construction work at Coptos, Egypt) the authorities probably drew on fixed numbers of sub-units (centuriae and turmae), supplemented by contributions from the remaining sub-units.

In the case of the 424 equites from 3 alae, he suggests -- it's only a suggestion, but an interesting one -- that six full turmae were sent (192 men under the six listed decurions), supplemented by contributions from the remaining turmae (three alae of 16 turmae each, minus the six full turmae already withdrawn, leaves 42 turmae to contribute men). He goes on to suggest that the remaining 232 men were contributed as one unit of 6 men per turma (which he sees as a likely operational contubernium). His numbers could be right -- except that it will have been 2 bunk-squads of 3 men per remaining turma.

Either way, the totals would appear to have little relevance for the ala milliaria. :roll: Sorry for wasting your time!

Firstly, do we have any genuine lists of the Auxiliary units

In some ways, it's even more exciting. No, we don't have a complete list of units at any particular time -- although Paul Holder has tried hard to draw up a list from the time of Hadrian* --, but we have an ever-expanding pool of diplomas -- the so-called discharge certificates which an auxiliary could obtain to prove that he'd earned his citizenship by serving out his time with the colours. The brilliant bit is that each diploma lists all the units in a given provincial command which were all discharging men on the same day. It's like an ever-expanding jigsaw that will (hopefully) eventually give us something like the list you're looking for! Come back in 20 years.

* Paul, if you're lurking, please drop me an e-mail.

[Nathan Ross]

each diploma lists all the units in a given provincial command which were all discharging men on the same day.

And as an example, here's the relevant bit of CIL 16, 70, from Stannington, dated AD124:

equitib(us) et peditib(us) qui militaverunt in alis VI et coh(ortibus) XXI quae ap[p(ellantur) / I Hisp(anorum) Astur(um) et I et I Thrac(um) et Picentian(a) / et Petrian(a) et I Afr(orum) / c(ivium) R(omanorum) et I Hisp(anorum] et I Frisiavon(um) et I Hamior(um) / sagitt(aria) et I Sunuc(orum) et I Vang(ionum) (milliaria) et I B(a)etasior(um) / et I Delm(atarum) et I Aquit(anorum) et I Menap(iorum) et I Ulp(ia) Traiana / Cuger(norum) c(ivium) R(omanorum) et I Fida Vardull(orum) |(milliaria) c(ivium) R(omanorum) et I Batav(orum) / et I Tungr(orum) et II Ling(onum) et II Astur(um) et II Pannon(iorum) / et II Nerv(iorum) et III Brac(ar)augustanor(um) et III Nerv(iorum) et VI Nerv(iorum) / quae sunt in Brittannia sub Platorio Ne/pote...

The text mentions six alae and twenty-one cohorts, more than the usual complement of the British garrison; many of these would be part of Hadrian's expeditionary force, connected with the establishment of the wall, I presume. One of the ala and two (I think) of the cohort names are missing, but you should be able to make out the rest! :wink:

[D B Campbell]

The text mentions six alae and twenty-one cohorts, more than the usual complement of the British garrison

:?: Surely some mistake? The diploma of AD 122 lists 13 alae and 37 cohortes. Holder reckons that there might have been another ala and maybe five cohorts which didn't appear in the list (because they weren't discharging men). Britannia had a humongous garrison -- presumably because the proto-Glaswegians were so fierce. :wink:

[Mark Hygate]

..... His numbers are skewed, because he believed that cavalrymen bunked 6 or 8 to a room. (We're now pretty sure that it's only 3 to a room.)

But his arithmetic is interesting.....

In the case of the 424 equites from 3 alae.......

... Sorry for wasting your time!

Not a chance!

Given I've already noted that I think it's 4 to a room (as 'my' 30 man Turmae is still (re Polybius) made up of 3 x 10-man elements, each of a leader (it was a Decurion, might still be, but could be double-men & one-and-a-halfer's), a 'rank closer'/NCO and an otherwise 'normal' 8-man contubernia-equivalent). Then I too could mathematically suggest that:

424 /8 /3 = 17.666, for the near total of 18 turma. Now, if I happen to believe that an Ala D is 6 turma total, then could this possibly be that these are all the equites from 3 Ala? They might have thought that 5 or 6 Decurions and 5 'others' were all that were needed to manage them - given that 'cavalry officers' have always been averse to hard work (my own cynical and contentious view :lol.

..... Come back in 20 years. ]

Actually, given my interest and organisational proclivities, I would love to collate such! At the present I shall have to limit myself to a re-analysis of Cheesman.

[D B Campbell]

Now, if I happen to believe that an Ala D is 6 turmae total, then could this possibly be that these are all the equites from 3 alae? They might have thought that 5 or 6 Decurions and 5 'others' were all that were needed to manage them

We're getting awfully far away from the evidence, Mark. You've chucked Arrian ("a cavalry ala has 512 men") out the window. Hyginus ("an ala quingenaria has 16 turmae ... and 64 spare horses") is about to follow him, but he's hanging onto the ledge! :wink: Did you really mean that the 424 equites ("horsemen") from three alae are the total complement?! Or did you mean that, all together, they make up the equivalent of a single ala? :?

Edit: Also, re. your four cavalrymen to a room, I'm not sure that there would be space for four horses in the front room. They seem to be just the right size for three horses.

[Nathan Ross]

The diploma of AD 122 lists 13 alae and 37 cohortes... Britannia had a humongous garrison

I didn't know it was quite that humongous! mile:

Actually, I was mainly going on the figures given on other British diplomata, which seem to average 3-4 alae and around ten cohorts, rather than any deeper consideration. Surely, though, the number of units in the province around the time of Hadrian's visit could have been greater than normal? The AD122 one (AE 2008, 800, from Brompton) would also fall within this period. We know from Pontius Sabinus' inscription (CIL 10, 5829) that there was an expeditione Brittannica around then - Sabinus himself commanded legion vexillations, but there would surely have been auxiliaires as well.

Then again, it could be that the units in the province were just taking advantage of the imperial presence to lay on some grand honesta missio discharge ceremonies!

But I digress, sorry...

[Vindex]

Edit: Also, re. your four cavalrymen to a room, I'm not sure that there would be space for four horses in the front room. They seem to be just the right size for three horses.

I'd second that if I may be so bold...but it does depend a lot on the size of the horses!

[D B Campbell]

Then again, it could be that the units in the province were just taking advantage of the imperial presence to lay on some grand honesta missio discharge ceremonies!

Could be something like that. (If so, thank goodness they did!) The Trajanic garrison looks like 10 or 11 alae and 25 or 26 cohorts -- a bit smaller than the Hadrianic one --, so there might have been transfers accompanying the Sixth Victrix, if it arrived as early as AD 122. (I know Brian/Philus doesn't like "might" and "maybe" -- on the One Unit, One Fort thread, but you've just got to be honest!)

[Mark Hygate]

Now, if I happen to believe that an Ala D is 6 turmae total, then could this possibly be that these are all the equites from 3 alae? They might have thought that 5 or 6 Decurions and 5 'others' were all that were needed to manage them

We're getting awfully far away from the evidence, Mark. You've chucked Arrian ("a cavalry ala has 512 men") out the window. Hyginus ("an ala quingenaria has 16 turmae ... and 64 spare horses") is about to follow him, but he's hanging onto the ledge! :wink: Did you really mean that the 424 equites ("horsemen") from three alae are the total complement?! Or did you mean that, all together, they make up the equivalent of a single ala? :?

Edit: Also, re. your four cavalrymen to a room, I'm not sure that there would be space for four horses in the front room. They seem to be just the right size for three horses.

Actually, I'm fairly comfortable that we aren't too far from the evidence - but that's for me to argue and defend.

I must try to get hold of a copy of Arrian, I only wish one were online, but I've not found it yet. Cheesman, however, is and, whilst I had looked at a fair amount of the data already in it, I am reading carefully at present; I was most interested in his note that Arrian's reference to Ala and 512 men was as part of a digression and comment on earlier Greek organisations - so, not dismissed, but certainly queried.

I have read, and reviewed, Hyginus a few times already and noted 2 distinct things. Firstly his statement - "...on account of my inexperience of military affairs...": to which I can only comment - too right! :lol: Secondly, that, compared to Polybius' camp layouts and all we have seen in the 'standardised' playing-card camps and forts; that treatise is one of the most difficult and crazily complex systems I have ever seen. Nothing I have ever learned of the 'Roman systems' could support it. At best I would see it as a mathematical puzzle to set clever students. I simply don't believe that anyone would make something so complicated - when I genuinely do believe there is an alternate. hock:

So, yes, I do believe there were indeed 432 'equites' (regular soldiers) in 3 x 6 x 3 x 8 sections in 3 Ala D. It's consistent with my theory that Quingenary and Milliaria are nothing to do with 500 or 1000, except by fortuitous circumstance in a couple of cases.

As to 4 per room - yes, and disregarding any thoughts of modern animal welfare, knowing that horses then were generally smaller, looking at the picture in your book and feeling confident that, whilst fitting 30 horses into those 9 room sets, I would separate the 3 squads into 3 rooms each - simply because of the military need for 'room inspections' (sic). If you wanted to fit 32 or 36 or 42 horses in, whilst I don't argue for it, then it would get even tighter.

From Cheesman as well, I've even found another possible reason for the 2 pairs of 16 Decurions in those lists: if only 6 turma, then you need 18 'officers' - that would be the 16 Decurions; the praefectus alae; and the subpraefectus alae (perhaps known under another name). I'd never seen that last; has it been refuted/disclaimed in the last 94 years?