Linked to my questioning the unit size of the 'Imperial' Roman Cavalry units...
I have seen the phrase of the title applied when discussing the distribution of auxiliary cohors and ala around the empire, particularly (of course) where it relates to the fixed lines drawn around Hadrian's time and thus including Hadrian's Wall and the 'limes' in Germany.
In books and references, such as Dr Campbell's 'Roman Auxiliary Forts' and others, I will also see forts identified as 'Cavalry Forts'. I understand that this is mostly based upon an archaeological find such as an inscription, a grave marker or documentary like the Notitia Dignitatum (which is, of course, quite late in the chronology) and even lead seals. These often have some dating reference too where available.
In the case of Cavalry units, however, I have wondered why the Romans would have done this (if indeed they did) and whether it's entirely logical, or indeed sensible?
As is noted in many of the same references, the Roman Army was primarily offensive in nature and the forts in question were not supposed to be a 'Maginot Line' defensive work, but were permanent places to be based and had defences to match and effectively delay an attack until other troops could come up. In addition, those forts identified as 'Cavalry Forts' on Hadrian's Wall are those which lie astride the wall, so that 3 of the gates are available to sally from out to the North. However, even a 'protected place's defences are useless if there's no one manning them.
We also have, not unrelated, the forts occupied by the Cohors Equitata units (which has to be the simplest example of tweaking a military organisational system to produce garrison units that I know of).
My thought here to query the idea of a Cavalry Fort is simple: who guards the fort when the cavalry sally forth, go on patrol, conduct offensive missions? The Turmae, whilst occupying the same space as Centuriae, have far less men; thus forts containing cavalry units are much bigger than the manpower would imply, with a corresponding increase in wall space to man.
With a Coh Eq you can leave a century or two behind to 'hold the fort' if you go off to the aid of another sector; let alone (in the case of the wall defences) continue to patrol the walls and man the mile-castles and towers.
When it comes to those forts identified as having a Cavalry unit based, would it not be more likely that there's an infantry unit there also? A cavalry-man off his horse and left on the walls is a complete waste of mobile fighting power. Cavalry have never really been used like that, part from the experiments with Dragoons and the original US Cavalry of the American Plains perhaps, but not as garrison units such as these.
So, is there any real proof that 'Cavalry Forts' were the home of just the Alae on its own? Or is it just inferred?
It is interesting that you should mention this about Coh/Eq for I was many years ago a custodian working at Chesters (Cilurnum) Fort on Hadrian's Wall.
What I found to be interesting was that the Notitia has the II Ala Asturia at Chesters, as also the Ist Asturia at Condercum.
However when ever I used to clean the finger prints off the glass cases in the museum at Chesters, it would puzzle me as to why there are roof tiles in a case there that say very clearly Coh II Ast.
This might imply that the II Asturia may well have been a Coh/Eq. and not a pure Cavalry unit as we might think.
There was also in the Hadrian Wall area the only Ala Milliaria cavalry unit in Britain and this was the Ala Petriana.
In the book ALA 2 by John Spaul we find evidence of this unit found at Hexham where we have references to a funerary stone with invocation ALA PETR [Dis manibus, nominative] however it is listed as being Hexham (ie Corbridge)
Then we have again another reference Personal to Candidus, dec., listed yet again Hexham (ie Corbridge)
Then we have the famous tomb stone of Flavinus listed as Hexham (ie Corbridge)
However in a post Epiacum I have stated that the Ala Petriana were infact based at Hexham and not Corbridge as historians would have us believe.
The late Raymond Selkirk did suggest that this regiment of Cavalry were not only based there but that there may well be a fort beneath Hexham Abbey which stands upon his 255 Roman frontier line that pre-dates Hadrian's Wall and runs from east to west coast.
Roman fortresses were never based on the one unit, one fort principle.
Forts were built by either one legion or several army units, roof tiles came from a lot of places, for instance in the Netherlands we find a lot of tiling made by the Legionary furnaces in Nijmegen, along the entire Limes area as well as in other parts of the country. That does not mean that the LXG was posted along our entire stretch of the Limes.
Romans used the forts as semi-permanent as well as permanent bases, and a lot of units were moved wherever they were needed. Sometimes one fort could have as many as 5 to 10 units posted there in a span of time.
I suppose that the guards left behind will act more for "police" agains thieves than to guard a fort against enemies. As you said Roman tactic was offensive so the enemy will be engaged before arriving at an Ala fort behind the lines.
There must have been a significant number of servants (calones) in each Ala in addition to the ordinary soldiers. If the Ala did fight close their own fort they could stay behind to guard it, when whole units of a provincial army were on campaign, elements of the remaining units were probably redistributed to guard the forts. According to Josephus (Book 3 4.2) they had some military training:
Quote:... when all were united together, amounted to sixty thousand, besides the servants, who, as they followed in vast numbers, so because they had been trained up in war with the rest, ought not to be distinguished from the fighting men; for as they were in their masters' service in times of peace, so did they undergo the like dangers with them in times of war, insomuch that they were inferior to none, either in skill or in strength, only they were subject to their masters.
On the tombstones of Auxiliary cavalrymen they are often depicted standing behind the fighting eques, carrying extra spears and sometimes wearing some armor. Just the first page in the imagebase wields a number of such examples: Andes, Annauso or Caius Romanius Capito
Quote:In books and references, such as Dr Campbell's 'Roman Auxiliary Forts' and others, I will also see forts identified as 'Cavalry Forts'.
I'm not aware of having used the phrase "cavalry fort", but you've probably read my book more recently than I have. :wink: I know only too well that a fort built for one unit can easily end up being remodelled for a different one.
Quote:I understand that this is mostly based upon an archaeological find such as an inscription, a grave marker or documentary like the Notitia Dignitatum (which is, of course, quite late in the chronology) and even lead seals.
The evidence that you mention can certainly give us a clue. Of course, tombstones might be erected to people who were only "passing through", like the tombstone of the eques singularis found at Carberry, in East Lothian (near the Antonine fort of Inveresk): clearly, the man was not based there, but must have died while on official business with the governor.
Building inscriptions can be a better indicator, although it's clear that some troops were "helped out" by their colleagues from elsewhere during the construction process, so any unit named as the builders may not necessarily have lived there. Tile-stamps are even more tricky, as certain units appear to have utilised the fruits of their colleagues labours!
The clearest indicator -- if you're lucky enough to have one -- will be an altar, erected by the troops in garrison, or (more often, I think) by their commanding officer.
Quote:In addition, those forts identified as 'Cavalry Forts' on Hadrian's Wall are those which lie astride the wall, so that 3 of the gates are available to sally from out to the North.
Scholars of an earlier era were forced -- by lack of evidence -- to make such sweeping statements. Nevertheless, it must be said that we now know that Chesters (one of your "three-north-gates" forts) was built for an ala, and Wallsend (another of your "three-north-gates" forts) had a cavalry component in garrison and appears to have been designed for a cohors equitata. (It's late and I can't remember which other HW forts have securely identified primary garrisons.)
Quote:The Turmae, whilst occupying the same space as Centuriae, have far less men; thus forts containing cavalry units are much bigger than the manpower would imply, with a corresponding increase in wall space to man.
I'm not sure that your premise ("cavalry forts" are always bigger) is correct, but I haven't collated the data.
More worrying is your supposition that the garrison were supposed to line the wall à la Beau Geste! I don't think that's how it worked at all.
Quote:So, is there any real proof that 'Cavalry Forts' were the home of just the Alae on its own? Or is it just inferred?
Everything in ancient history is inferred. :wink: It's just that some things are more securely inferred than others!
We think that your "cavalry forts" (your phrase, not mine!) were the home of single units because there's only one commanding officer's house. And there's only one headquarters building.
In truth, not a single one has been completely excavated to ascertain whether all the accommodation was cavalry accommodation. But remember that the cohortes equitatae -- who fit the bill for your preferred garrisoning scenario -- vastly outnumbered cavalry alae, in any case.
I do begin to think there is a thought going around in our more modern times that the Roman garrisons in all areas were constantly on their toes, and peering out over the battlements looking for an enemy who are about to come dashing out of a wood or over a hill in the near distance.
This I am sure was not the case for in a situation such as the Hadrian Wall frontier area, there was to the north and indeed the south very many signal stations along with other spaced out forts.
This was the Roman armies inteligence system where on the wall all of the commanders would have had all the information about any nonsence going on at least 100 miles away to the north and indeed the south. In fact on many of the forts on Hadrians Wall we find that at least one if not both portals of gateways were blocked up shortly after their construction. The fort of Chesters had the east gate completely blocked and with no evidence it was ever used at all and the reason for this is it is so near to a river, then the west gate also had its north tower converted into a workshop so realy I think life on the wall was not so bad after all.
Therefore leaving a small amount of troops to look after the fort was no great problem at all, and any cavalry could trot all over the local area without fear of not knowing what was going on in their patch.
Quote:However when ever I used to clean the finger prints off the glass cases in the museum at Chesters, it would puzzle me as to why there are roof tiles in a case there that say very clearly Coh II Ast.
The COH.II.ASTVRV tiles (RIB 2467) were found at Great Chesters (Aesica) and stored in the museum at Chesters. Mystery solved!
Philus' & Marcus'
Thanks, but particularly for Marcus'; please forgive me if I confused that issue, but I am not asking about 'one unit having built the fort and then stayed there forever...' I am aware that 'any' unit(s) may have built the fort, often legionary detachments based in the province elsewhere - it's the idea, that I have seen espoused, that, in the main, only one unit occupied it at any one time. This is backed up by the point that only one Headquarters building is ever found (as Dr Campbell's point later). I would also not be surprised if things like roof tiles were not made in bulk elsewhere and shipped in; that said, they could be marked for the maker, or potentially for the destination!
MD
Yes, on campaign, all armies are followed by vast quantities of hangers on, be they: servants; baggage-train handlers & guards; merchants; and the ubiquitous 'camp-followers'. Most forts appear to have a 'Vicus' outside too and, even if there doesn't seem to be accommodation inside forts for many additional 'servants'; then it is more than likely that the 'civilians' would be allowed into the fortified space when they were in danger - then they could have assisted most certainly, so the idea is sound.
Dr Campbell
I'm so glad I used apostrophes rather than quotes for my best critic! I was particularly thinking of the phrase "the cavalry fort at Chesters", but you certainly didn't seem to use that in in my frantic rummage through said book. I did, thank goodness, find on pp29-31 the phrase "characteristic of a cavalry garrison", which I would ask to be considered essentially similar - ish :wink:
Please may I ask what the evidence is that specifies "that we now know that Chesters... was built for an ala"?
And no, I don't mean to imply that 'one man at every crenellation gap' lining the wall is the idea. hock: Although it could be that departing units left straw dummies in place..... :lol:
What I do wish to imply is 3 things: firstly that the auxiliary forts I am thinking of have ditches, ramparts and towers at corners, gates and even in-between; and if the defence is to be functional and at least delaying then it will take a reasonable amount of men; secondly that, simply by virtue of mathematics and if we do ascribe a centuriae and a turmae to occupy similar size barracks blocks, then, for any given number of sub-units within the fort, a cavalry unit would have a lot less men to man the equivalent perimeter - even if none left; and lastly, if we concentrate on a barrier like Hadrian's Wall, it is reasonable to expect that each fort (approximately, given the number of forts and length of the wall) along the wall had to provide a normal standing patrol of one centuriae (or equivalent) to the 2 mile-castles and 4 towers to the East and a similar amount to the West, then, even if a proportion of the unit leaves to another sector, this patrolling still has to be carried out - simply to allow for diversions, if nothing else.
Lastly, yes, the 'one headquarters building' is a very good argument. From my side, however, I would argue that a single headquarters building in a legionary fortress is enough to manage 10 cohors, so why couldn't a similar one in an auxiliary fort cope with 2-4 units. In addition, there could be separate smaller Tribune/Prefect houses with even a shared 'praetorium' facility to be found. However, not surprisingly, I will jump at the example of your Auxiliary Forts booklet on p15 of the fort at Drobeta with its potential 2 Praetoria. What you and others, however, have neatly presented throughout my research on the genuine hard archaeology evidence, is that many sites are not all, poorly or are simply no longer possible to excavate and identify the results more accurately.
Quote:I was particularly thinking of the phrase "the cavalry fort at Chesters", but you certainly didn't seem to use that in in my frantic rummage through said book. I did, thank goodness, find on pp29-31 the phrase "characteristic of a cavalry garrison", which I would ask to be considered essentially similar - ish
Fair's fair. If you read on to the next couple of paragraphs, you'll see that I allow for the possibility that "units could be divided or brigaded".
The phrase that you quote refers to the (now well-known) cavalry-style barracks, discovered (amongst other places) at Wallsend, which (please note) isn't one of your "cavalry forts"! mile: It has been interpreted as the base for a part-mounted cohors equitata. (Just such a cohort is attested there in the third/fourth centuries.)
Quote:Please may I ask what the evidence is that specifies "that we now know that Chesters... was built for an ala"?
A Hadrianic inscription naming the ala II Asturum was discovered about twenty years (?) ago. The earlier literature (e.g. Richmond) could only speculate on the identity of the primary garrison. Sadly, the visible barrack blocks were (a) excavated in the 19th century, and (b) not completely excavated, so it cannot be confirmed whether they resembled the Wallsend cavalry-style barracks. Nine-room barracks also occur at Benwell, another of your "three-north-gates" forts, where both a cohors equitata and an ala are attested at different times.
Quote:the auxiliary forts I am thinking of have ditches, ramparts and towers at corners, gates and even in-between; and if the defence is to be functional and at least delaying then it will take a reasonable amount of men;
Sorry, I don't see that as a Roman priority.
Quote:simply by virtue of mathematics and if we do ascribe a centuria and a turma to occupy similar size barracks blocks, then, for any given number of sub-units within the fort, a cavalry unit would have a lot less men to man the equivalent perimeter
... and it's hard work getting those horses up onto the ramparts! :wink:
Quote:yes, the 'one headquarters building' is a very good argument. ... a single headquarters building in a legionary fortress is enough to manage 10 cohors, so why couldn't a similar one in an auxiliary fort cope with 2-4 units. ... there could be separate smaller Tribune/Prefect houses with even a shared 'praetorium' facility to be found.
Legionary HQs are bigger. But I don't think that's the point. It's the commander's house that clinches it for me. The commander is appointed to a particular unit, not to a particular fort, and he wouldn't take kindly to sharing his nice Mediterranean-style villa with another commander! (He might allow some low-status riff-raff scouts inside the ramparts to sleep in a barrack well away from his "boys", but that's stretching things!)
Quote:However, not surprisingly, I will jump at the example of your Auxiliary Forts booklet on p15 of the fort at Drobeta with its potential 2 Praetoria.
It would have been naughty of me not to include an "untidy" fort -- the exception that proves the rule. The vast majority of forts seem to be occupied by a single unit (or part thereof, if a vexillation has been outposted) under a single commander. There are even instances of two forts built close together, as if to underline the point that each commander needs his own space.
Quote:What you and others, however, have neatly presented throughout my research on the genuine hard archaeology evidence, is that many sites are not all, poorly or are simply no longer possible to excavate and identify the results more accurately.
Quote:Marcus'
This is backed up by the point that only one Headquarters building is ever found (as Dr Campbell's point later). I would also not be surprised if things like roof tiles were not made in bulk elsewhere and shipped in; that said, they could be marked for the maker, or potentially for the destination!
Lastly, yes, the 'one headquarters building' is a very good argument. From my side, however, I would argue that a single headquarters building in a legionary fortress is enough to manage 10 cohors, so why couldn't a similar one in an auxiliary fort cope with 2-4 units. In addition, there could be separate smaller Tribune/Prefect houses with even a shared 'praetorium' facility to be found. However, not surprisingly, I will jump at the example of your Auxiliary Forts booklet on p15 of the fort at Drobeta with its potential 2 Praetoria. What you and others, however, have neatly presented throughout my research on the genuine hard archaeology evidence, is that many sites are not all, poorly or are simply no longer possible to excavate and identify the results more accurately.
I am sorry but I must say I find the one headquarter's building (Principia) for each unit ludicrous.
Here along our Limes a giant encampment for two Legions was found, not even having a headquarter's building, and what could have been it was not that enormous...(The Drusus/Tiberius encampment). In the LXG castra there was one large Principia. I think more along the lines that one building might have served several units in some instances.
The idea that only one building/house is found in some forts which is classified as a Centurio's mansion/house or officers house is also highly debatable. There are some large castra which have at the end of each barrack block a building that most of the times is explained as having been the officer's housing. Partly because the houses look more Romanesque with a courtyard. Only in Vindolanda we know fairly exact where the commander of Equestrian status lived and in what buildings his administration probably was located.
Also I do not agree with the idea that roof tiles were not transported in bulk along the line. We have more than enough proof in Germania Inferior that they were mass produced as well as shipped all along the line.
To state that there would only have been one officer's house, just like one headquarters building for each unit and basing your theories on that assumption alone is not proof in itself.
Quote:I must say I find the one headquarter's building (Principia) for each unit ludicrous. ... I think more along the lines that one building might have served several units in some instances.
This is the point that Mark and I have made about Drobeta, which has the usual HQ centrally located, exactly where we'd expect it, but the rest of the central range is unlike the usual pattern. First, the HQ is flanked by granaries, whereas the granaries are usually placed together (presumably to facilitate loading); and second, these in turn have a building beside them which resembles a simplified Mediterranean-style house. (The remainder of the fort -- the SW corner isn't available for excavation -- is filled with the usual barrack blocks.) This arrangement looks suspiciously like a fort halved down the middle. (Think Castra Vetera on a very much smaller scale.) The HQ can easily be shared, provided each unit has space for its own sacred standards, but the commanders' accommodation cannot.
Quote:The idea that only one building/house is found in some forts which is classified as a Centurio's mansion/house or officers house is also highly debatable. There are some large castra which have at the end of each barrack block a building that most of the times is explained as having been the officer's housing. Partly because the houses look more Romanesque with a courtyard. Only in Vindolanda we know fairly exact where the commander of Equestrian status lived and in what buildings his administration probably was located.
You seem to be thinking "legionary", when this thread is "auxiliary".
Duncan, I am thinking Roman military. The Roman military, including Auxilia stood under command of Romans of Equestrian rank if not higher in case of Legionary legates. There is no reason to believe that Auxilia would have built or layed out their fortresses in a different manner than the standard one which was not only generally accepted, but also decided upon top-down. That there were local differences in layout is fairly normal, we also see that in modern day army barracks. It depends on function more than on troop type. I do not think that Auxilia did things much different than Roman legionaries. There is a theory that these troops differed in style, food, living and rituals but I do not buy that. I think that there was far more acculturation or even Romanisation than people tend to grant the society along the borders of the Empire...
Quote:Only in Vindolanda we know fairly exact where the commander of Equestrian status lived and in what buildings his administration probably was located. ... To state that there would only have been one officer's house, just like one headquarters building for each unit and basing your theories on that assumption alone is not proof in itself.
I'm not sure why you think that Vindolanda is the only fort where the HQ and commander's house are known. (The exposed stone structures there are from the Severan fort.) It's pretty standard that the HQ always lies in the middle at the junction of the roads. The commanding officer's house is then positioned nearby, normally alongside, occasionally behind.
Here's Flavian Elginhaugh (another fort plan to make us feel uncomfortable, with its 10+3 arrangement of "cavalry" rooms, rather than the Hadrianic 9).
[attachment=3252]Elginhaugh.jpg[/attachment]
Notice that the HQ is slap-bang in the middle, with a Mediterranean-style villa on its righthand side (and a pair of granaries on its left). (Left and Right, when standing outside the HQ door, admiring what a tidy fort we command.)
btw I'm intrigued by your large Roman camp with no HQ -- where is that?
I was particularly thinking of the phrase "the cavalry fort at Chesters", but you certainly didn't seem to use that in in my frantic rummage through said book. I did, thank goodness, find on pp29-31 the phrase "characteristic of a cavalry garrison", which I would ask to be considered essentially similar - 
hock: Although it could be that departing units left straw dummies in place..... :lol: 