Vespasian\'s Auxiliaries

[Jona Lendering]

After the Batavian Revolt, Vespasian decreed that auxiliaries serving along the Lower Rhine were no longer to be recruited from Gallia Belgica. So, Tungrian and Batavian units were transferred to the Danube and England, while the new auxiliaries of Germania Inferior were from Spain and the Balkans. This measure is understandable after the Batavian Revolt.

However, is there evidence for similar transfers in other provinces?

[Peroni]

Certainly most of the British cohort levies were sent overseas (Dacia Porolisensis).

Coh.I Ulpia Britonnum milliaria
Coh. I Britannica equitata
Coh II Nervia Brittonum milliaria
Coh. II Britannica milliaria

Reference Military diploma CIL 16, 00185

Coh. III Britannorum (sent to Raetia)
Reference Military diploma RMD-02, 00112


Diploma discharging British soldiers in Dacia (pedites Britannici et sunt in Dacia)

Ala I Britannica civium Romanorum
Coh. I Brittonum milliaria Ulpia torquata civium Romanorum
Coh. I Britannica milliaria civium Romanorum
Coh. II Britannorum milliaria civium Romanorum Pia Fidelis

Reference Military diploma CIL 16, 00163

[Rumo]

Also Dacian soldiers were serving all over the empire but Dacia (Egypt, Mauretania, Syria, Britain, Germany). If I'm not mistaken a similar situation happened to Thracians and probably also in other provinces where the natives were notoriously warlike and rebellious.

[Jona Lendering]

Thanks!

[Matthew Amt]

Isn't there an inscription in Britain (near the Humber?) noting the presence of bargemen from the Euphrates River? Don't know what those poor slobs did to get posted there! Especially since there must have been perfectly competent bargemen in Britain itself, or Gaul, or Spain, or...

Oh, the Hamian archers on Hadrian's Wall spring to mind, too.

Matthew

[PhilusEstilius]

Matt.
There were also the Bacari Tigrinensis who were at the fort on the mouth of the river Tyne which very well ties in with the late Raymond Selkirks' evidence of Dams along the Tyne to supply Hadrians' Wall.

[Matthew Amt]

There were also the Bacari Tigrinensis who were at the fort on the mouth of the river Tyne

Ah, thanks, Brian, those may be exactly the guys I was trying to remember! Tigris, Euphrates--okay, SOMEwhere in Mesopotamia...

Matthew

[M. Demetrius]

Heh. It doesn't get any more Mesopotamian than between the Tigris and the Euphrates, does it? :lol: :roll:

[D B Campbell]

After the Batavian Revolt, Vespasian decreed that auxiliaries serving along the Lower Rhine were no longer to be recruited from Gallia Belgica.

I have heard a similar theory (perhaps in Cambridge Ancient History?) but cannot find any evidence to support it. The ancient authors are happy to mention the legions (no doubt, for complex reasons), but are notoriously reticent when it comes to the auxilia. It seems unlikely that any ancient author ever attributed such a decree to Vespasian. (But if someone knows otherwise, please tell us!)

So, Tungrian and Batavian units were transferred to the Danube and England, while the new auxiliaries of Germania Inferior were from Spain and the Balkans. This measure is understandable after the Batavian Revolt.

Indeed, that is a plausible explanation of an epigraphically attested phenomenon. But Vespasian was not the first to enact such a policy. (And there may not have been a "policy", in any case -- see below.) Cheesman noted that, after the Pannonian Revolt of AD 6-9, Pannonia and Dalmatia were garrisoned by regiments from elsewhere (notably Spain and the Alps), while Pannonian and Dalmatian regiments were deployed elsewhere. Had Augustus evolved the very policy that we are attributing to Vespasian (but applied to Illyricum rather than the Rhineland)?

We must be careful not to see "policy" where none existed. Wherever regiments (even legions) had been involved in unrest, it was only sensible to remove them elsewhere -- as Jona has noted. But wherever troops were required, they had to come from somewhere. In the aftermath of the Batavian revolt, troops were required in Britain to replenish the depleted garrison. Would it have been possible officially to deny German auxiliaries to the governor of Lower Germany? (Possibly.) Or Spanish auxiliaries to the governor of Tarraconensis? (Hmmm ... difficult.) Or Syrian auxiliaries to the governor of Syria? More importantly, was it policy?

To prove that a "policy" existed, we would need to demonstrate not only (a) that regiments were never deployed near their homeland (difficult, as the bulk of auxiliaries in the West seem always to have come from Gaul, Spain and Thrace, because these areas were fertile recruiting grounds), but also (b) that newly-raised regiments were always removed far from their homeland, expressly to avoid trouble.

The last bit is important. Others have pointed to Dacia as supporting evidence. But were the newly-raised Dacian regiments deployed far from home as part of a "policy" (to avoid trouble), or because they were not required in Dacia? It is a difficult issue to decide (and one worthy of Ancient Warfare magazine's "Debate"!).

[Rumo]

It is hard to argue for a deployment policy as a "must" for the entire empire and over centuries. But certainly there's a tendency which needs to be explained.

I know several scholars already argued for it (e.g. M. P. Speidel, "The Rise of Ethnic Units in the Roman Imperial Army" in ANRW II.3/1975: "Those tribal and regional contingents that during second century A.D. had become permanent and almost-regular or regular units, generally served far away from their homelands in distant corners of the Empire, garrisoning their assigned frontier districts often for over a century.") and I might be able to shape a particular case like the Dacian one.

In Dacia besides legions (XIII Gemina at Apulum and since the Marcomanic wars also V Macedonica at Potaissa) we find many irregular (numeri, gentiles) and regular (alae, cohortes) auxiliaries of Hispani, Galli, Brittani, Tungri, Batavi, Germani, Pannoni, Illyri, Thraci, Bessi, Bosporani, Palmyreni, Syri, Mauri, Gaetuli, etc. At the same time several Dacian units were located somewhere else: ala I Ulpia Dacorum in Cappadocia, cohors I Ulpia Dacorum in Syria, cohors I Aelia Dacorum in Britannia.

Of course the native soldiers were not recruited only in "ethnic" units. There was also some general replacement of the ranks.
In the inscriptions from Dacia we have a very small number of confirmed natives (by names or origin). It's nevertheless true that not all the natives preserved their ancestral names or mentioned their origin. We have a diploma from year 144 from Lambaesis (Mauretania Caesarensis) and among those soldiers of Legio III Augusta having mostly Roman names we find several recruited in Napoca under Hadrian (Publii Aelii). Perhaps not all of them were Dacians, but probably at least a part of them were.

Another interesting case comes from Roman Egypt. At Krokodilô (Hélène Cuvigny, Ostraca de Krokodilô. La correspondence militaire et sa circulation (O. Krok. 1-151). Praesidia du désert de Bérénice II, 2005) we find several Dacian soldiers writing letters on ostraca or being mentioned in them. In OKrok 98 (dated in 109), the horseman Dekinais wrote to his colleague Kaigiza that he heard that Sulpicius Similis, the prefect of Egypt, ordered all the Dacian soldiers to be gathered in Alexandria. We know no Dacian "ethnic" units here but ala Vocontiorum in the nearby Koptos. However I don't know if this unit participated in the Dacian Wars.

[Jona Lendering]

Thanks!

[D B Campbell]

Hello, Drago?. I am grateful for your specialised knowledge of Dacia.

I know several scholars already argued for it (e.g. M. P. Speidel, "The Rise of Ethnic Units in the Roman Imperial Army" in ANRW II.3/1975: "Those tribal and regional contingents that during second century A.D. had become permanent and almost-regular or regular units, generally served far away from their homelands in distant corners of the Empire, garrisoning their assigned frontier districts often for over a century.") and I might be able to shape a particular case like the Dacian one.

I think you have misunderstood Speidel's thesis. By "ethnic units", he means those numeri which appeared as as adjunct to the alae and cohortes. They were characterised by broadly ethnic titulature -- e.g. the numerus Brittonum which pops up in the forts of the Odenwald limes. The Germans have even designated an entire class of fortifications as "Numeruskastelle". (Thus a cohors Dacorum would not qualify as one of Speidel's "ethnic units".)

As usual, Speidel has spun out an entire theory from very little hard evidence. When he writes that they "generally served far away from their homelands", he is generalising from the Brittones on the German frontier, and very little else. The Palmyreni in Dacia and the Raeti in Britain are usually lumped into the same category.

In Dacia besides legions (XIII Gemina at Apulum and since the Marcomanic wars also V Macedonica at Potaissa) we find many irregular (numeri, gentiles) and regular (alae, cohortes) auxiliaries of Hispani, Galli, Brittani, Tungri, Batavi, Germani, Pannoni, Illyri, Thraci, Bessi, Bosporani, Palmyreni, Syri, Mauri, Gaetuli, etc. At the same time several Dacian units were located somewhere else: ala I Ulpia Dacorum in Cappadocia, cohors I Ulpia Dacorum in Syria, cohors I Aelia Dacorum in Britannia. Of course the native soldiers were not recruited only in "ethnic" units. There was also some general replacement of the ranks.

So this is exactly the same picture we find in other provinces. The army of invasion, assembled from a variety of units, becomes the garrison. It eventually begins to recruit locally. At the same time, any entire units subsequently conscripted in the province are sent elsewhere, as and when required.

[Jona Lendering]

The army of invasion, assembled from a variety of units, becomes the garrison. It eventually begins to recruit locally. At the same time, any entire units subsequently conscripted in the province are sent elsewhere, as and when required.

I indeed get the impression this is the correct scenario.

[Rumo]

Hello, Drago?. I am grateful for your specialised knowledge of Dacia.

Hello and thank you for your kind appreciation.

I think you have misunderstood Speidel's thesis. By "ethnic units", he means those numeri which appeared as as adjunct to the alae and cohortes. They were characterised by broadly ethnic titulature -- e.g. the numerus Brittonum which pops up in the forts of the Odenwald limes. The Germans have even designated an entire class of fortifications as "Numeruskastelle". (Thus a cohors Dacorum would not qualify as one of Speidel's "ethnic units".)

As usual, Speidel has spun out an entire theory from very little hard evidence. When he writes that they "generally served far away from their homelands", he is generalising from the Brittones on the German frontier, and very little else. The Palmyreni in Dacia and the Raeti in Britain are usually lumped into the same category.

My fault for being imprecise and I apologize for that. I know Spiedel is only making a point about what I called irregular units (that very paragraph I quoted from ends with "...many of these units are likely to have been romanized as the neighbouring alae and cohorts."). I realize now the confusion is due to my similar wording (ethnic, regular).

However I cannot agree with your assessment of Spiedel's thesis. In the same study he covers a larger variety of auxiliaries as nationes (numeri, vexillationes, equites), adducing both literary (Caracalla's barbarian guard and allies mentioned by Cassius Dio) and epigraphic (Mauri, Dalmatae, Marcomanni, etc.) evidence for their presence in garrisons, field army and personal guards.

So this is exactly the same picture we find in other provinces. The army of invasion, assembled from a variety of units, becomes the garrison. It eventually begins to recruit locally. At the same time, any entire units subsequently conscripted in the province are sent elsewhere, as and when required.

That may explain this phenomenon to some extent, but I don't think it's the only cause. The units were not changing their bases only when new provinces were created. Casualties but also new or enlarged fortifications required new men. Dacia was a heavily defended province and not all its guarding troops arrived there with Trajan. I already mentioned Legio V Macedonica being moved to Potaissa.
Also, most of the Dacian names and people of Dacian origins we know are not from Roman Dacia (as we would expect) but from other provinces. In Dacia there's a striking absence of native names. In the 60s-70s some scholars enthusiastically counted ~60 "natives" in an epigraphic corpus of ~3000. About that many Dacians (~50 distinct names) we find only in those ostraca from Egypt (Dan Dana, "Les daces dans les ostraca du désert oriental de l'Égypte. Morphologie des noms daces" in ZPE, 143/2003, 166-86).

[D B Campbell]

As usual, Speidel has spun out an entire theory from very little hard evidence.

However I cannot agree with your assessment of Spiedel's thesis. In the same study he covers a larger variety of auxiliaries as nationes (numeri, vexillationes, equites), adducing both literary (Caracalla's barbarian guard and allies mentioned by Cassius Dio) and epigraphic (Mauri, Dalmatae, Marcomanni, etc.) evidence for their presence in garrisons, field army and personal guards.

I am grateful that you raised this subject, Drago?, as it made me go back and re-read Speidel's 1975 paper. (And I apologise to Jona for derailing his thread slightly!) I have great respect for Professor Speidel, but I find this to be one of his weaker papers, characterised by special pleading from limited evidence and selective citation of evidence to suit his theory.

(1) He calls for the use of the term "natio" as the technical name for "ethnic units" like the numeri Brittonum. I see no justification for this. Romans used the word numerus in two ways: (a) as a general term referring back to any "unit" which had already been mentioned or whose identity was otherwise clear (so, an inscription referring to a particular cohors may subsequently name [e.g.] the commander of that "unit", numerus, rather than use the term cohors again); and (b) as a technical term for those units which were not alae, cohortes or legiones (in other words, units which had no other technical designation). The fact that the equites singulares (e.g.) were not organised as an ala causes no confusion; we all know, as the Romans themselves knew, that we must refer to them as a "unit" (numerus). Similarly, the "ethnic units" that appear on the Odenwald and elsewhere called themselves numeri, because they were not organised as alae or cohortes.

(2) He claims that his "nationes" (i.e. the "ethnic units" like the Brittones on the Odenwald) "gradually became alae or cohorts, or were lost from sight" (p. 138). He offers no proof, and I have never encountered any. Similarly, he does not elaborate on the "gradual process of romanization that in time made them as reliable units as any other auxilia" (p. 145). Caveat lector!

(3) Finally, he jumps forward in time to discuss the equites and vexillationes of the later army. These have no connection whatsoever with the "ethnic units". Neither do Caracalla's German bodyguard.

The units were not changing their bases only when new provinces were created. Casualties but also new or enlarged fortifications required new men.

Interesting subject. It's my impression that a provincial garrison only changed during periods of warfare or annexation (there or elsewhere, by domino effect). But it would probably bear detailed study.

In Dacia there's a striking absence of native names. In the 60s-70s some scholars enthusiastically counted ~60 "natives" in an epigraphic corpus of ~3000.

Again, an interesting phenomenon. Of course, if natives were banned from serving in their own province, we'd expect 0 Dacian names in Dacia! But, of course, as elsewhere, Roman units were recruiting locally.