Trajan Decius (obverse) with Dacia holding a draco (reverse)
From Tony Wilmott's Cohors I Aelia Dacorum: A Dacian Unit on Hadrian's Wall (ACTA MVSEI
NAPOCENSIS 38/1 CLVJ-NAPOCA 2001):
There has been much discussion on the origin of the cohort. Holder22 , Petolescu23
and Jarrett24 have all suggested that cohors I Aelia Dacorum was originally raised very
13 F. Haverfield, /nscriptions Preserved At Birdoswa/d, Transactions of the Cumberland Westmorland
Antiquarian Archaeological Society, O Ser 15, 1898, 197-200.
14 F. Haverfield (n. 2).
15 J. Horsley (n. 5), 253.
16 RIB 1981-83.
17 J. Horsley (n. 5), 256.
18 R. P. Wright, E. J. Phillips, Roman lnscribed and Sculpted Stones in Carii sie Museum, Carlisle 1975, 21.
19 T. Wilmott, Birdoswald Excavtions of a Roman Fort on Hadrian's Wali and its Successor Settlements;
Excavations 1987-92, English Heritage Archaeological Reports 14, London 1997, 189-91.
20 J. C. Bruce, The Roman Wall3, Newcastle 1867, 297.
21 I. A. Richmond, E. B. Birley, Excavations on Hadrian's Wali in the Birdoswa/d-Pike Hi/I Sector, 7 929,
Transactions of the Cumberland Westmorland Antiquarian Archaeological Society, N Ser, 30, 1 930,
169-205.
22 P. A. Holder, The Roman Army in Britain, London 1 982, 70.
23 C. C. Petolescu, Dacii în armata romana, Revista de Istorie 33, 1 980.
24 M. G. Jarrett, Non-Legionary Troops in Roman Britain: Part One, The Units, Britannia 25, 1994, 35-77.
106 Tony Wilmott
soon after the conquest of Dacia by Trajan, and was probably transferred almost
immediately to Britain. Spaull disagrees, believing that if this was the case the cohort
would have been designated Vipia, on the model of the cohort attested in Syria, the
cohors I Vipia Dacorum. He favours the idea that the cohort was raised under Domitian
as a quingenary unit, after his Dacian wars25 • The cohort worked on the building of the
Vallum of Hadrian's Wall under Hadrian, as attested by a building stane recording the
century of Aelius Dida of cohors I Dacorum (No. 34 ), found near Benwell. Their first
known garrison was at Bewcastle where they are again attested simply as cohors I
Dacorum (No. 35). The editors of RIB suggest that the Bewcastle inscription is
Hadrianic because Aelia îs omitted. Jarrett26 suggests that the title was a battle
honour, though it has been suggested that units may have received such honorifics
when milliary status was granted27 • Spaull's idea is that this occurred during the move
of the cohort from Bewcastle to Birdoswald but, as we will see, it is highly unlikely that
this took place under Hadrian. lf the fort at Bewcastle was built for this cohort28 , then
the large size of the fort clearly suggests that the unit already had a milliary
establishment, despite the lack of any milliary indication on No. 35. Holder29 has
demonstrated that the issue has been solved by the diploma of 20th August AD 1 27,
which gives the style coh(ors) I Ael(ia) Dac(orum) (milliaria), including the milliary
symbol. He suggests that the men discharged in AD 1 27 were recruited in AD 1 02 or
a little earlier, perhaps from Dacians settled within the Empire. Those discharged would
have been part of the cadre around which the unit was originally formed. They would
have included the Dacian soldier who received the diploma, one ltaxa, son of Stamilla
(ITAXAE STAMILLAE F DAC0)30• The award by Hadrian of the epithet Aelia might have
been a battle honour awarded on the lower Danube, alternatively the unit may have
originated as a numerus Dacorum under Trajan, and been made a cohors by his
successor. Though Holder further suggests31 that the cohort accompanied Hadrian to
Britain on his visit in AD 1 22, the tact that the cohort achieved the title and milliary
status before AD 1 27 at the latest, does not automatically mean that either or both
of these were granted before the unit came to Britain.
lt is not known how long the cohort served at Bewcastle, though it is likely that
it did not appear at Birdoswald until the late second or early third century, as we will
see. For the middle and later second century it is referred two on two diplomas and
an inscription. On the diplomas of AD 14632 and AD 15833 , no milliary symbol is given.
However, the diploma of AD 1 46 does not give a milliary symbol for other units which
are known to have been of this strength, such as cohors I Fida Vardullorum milliaria.
This does not mean that these units were not milliary, and Birley34 has noted that only
25 J. Spaull, Cohors2 : The Evidence for, and a Short History of the Auxilliary lnfantry Units Of The
Imperial Roman Army, BAR lnt. Ser. 84 1 , Oxford 2000, 345-346.
26 C. C. Petolescu (n. 23 ).
27 V. A. Maxfield, The Military Oecorations Of The Roman Army, London 1981, 234; also by J. Spaull
(n. 25).
26 P. S. Austen, Bewcastle and Old Penrith, a Roman Outpost Fort and a Frontier Vicus: Excavations
1977-78, Cumberland Westmorland Antiq Archaeol Soc Res Ser. 6, Kendal 1991, 43-44.
29 P. A. Holder, Auxiliary Units Entitled Ae/ia, ZPE 122, 1998, 255-257.
30 J. Noile, Militărdiplom fur einen in Britannien entlassenen "Daker", ZPE 117, 1997, 269-276.
31 J. Noile, op. cit., 262.
32 CIL XVI 93.
33 P. A. Holder, A Roman Military Diploma From Ravenglass, Cumbria, Buii. John Rylands Library
Manchester 79, 3-41.
34 E. B. Birley, Alae and cohortes milliariae, in Corolla Memoriae Erich Swoboda Dedicata, Romische
Forschungen in Niederosterreich, Graz 1 966, 61.
COHORS I AELIA DACORUM: A DACIAN UNIT ON HADRIAN'S WALL 107
in diplomata where some units are referred to specifically as milliary should the
absence of such a reference for other units recorded as milliary elsewhere be
considered significant. Holder suggests that the cohort was reduced in size by the
detachment of a vexillation, and that this is alsa the reason for the lack of any milliary
reference on the two diplomas. This author reached a similar conclusion to account for
the tact that no milliary symbol appears on any of the third century /OM altars35 , and
yet the commanders are almost always tribuni. Margaret Roxan has suggested that the
symbol might be omitted simply because the status of the unit was well enough known
and established for its use to be unnecessary. The only epigraphic source to accord
the cohort its tuli style, coh(ortis) mil(liaria) Ael(ia) Dacor(um) is from Lambaesis in
Libya. This gives the career of the equestrian officer Ti. Claudius Proculus Cornelianus,
who served his militia secunda as tribunus of the cohort, probably during the reign of
Antoninus Pius36• The peculiarity of this inscription is that it omits the numeral, though
there is little doubt that aur cohort is referred to.
The earliest of the inscriptions of cohors I Aelia Dacorum from Birdoswald may be
No. 1 9, an altar which was rediscovered in 1 990. This records a tribunus called Domitius
Honoratus. Roger Tomlin37 has suggested that if this was L. Domitius Honoratus,
prefect of Egypt under Severus Alexander, his tribunate would have been around the
reign of Septimius Severus. This can, however, nat be much earlier than the horreum
building inscription found during 1 929 (No. 1 ), which dates to AD 205-208. The
construction of the horrea within the fort was part of a very major late first and early
second century building programme at Birdoswald, which alsa saw the alteration of the
troops' accommodation. The north-western barrack was divided into two buildings. The
men's quarters, divided into contubernia, remained largely unaltered, but the
centurion's quarters in the projecting section at the west end was detached to form a
separate square suite of rooms, including a private privy, and at least one room which
was equipped with a hypocaust38• lt seems likely that these widespread changes might
have been occasioned by the arrival of a new unit, and it is difficult to avoid the
conclusion that the unit concerned was the Dacian cohort. Spaull's39 argument that the
cohort took part in the initial construction of the stane fort is based on his
interpretation of No. 30, a centuria! stane. The tact that this was found ex situ, reused
in the wall of Birdoswald tarmhouse, invalidates Spaull's interpretation.
The unit, and the names of many of its commanders (all either tribuni or
centurions in temporary command), appear on a large series of altars dedicated to
Jupiter Optimus Maximus. lt has long been suggested40 that they originate from a
ceremonial burial deposit on the parade ground. The group seems to be a dispersed
example of a deposit of altars of a kind found in situ at Maryport41 • lt was originally
thought that the altars were dedicated as part of the official religious observances of
35 T. Wilmott (n. 19), 196.
36 H.- G. Pflaum, Deux carrieres equestres de Lambese et de Zana, Lybica 3, 1955, 123-54; idem, Les
carrieres procuratoriennes equestres sous le Haut-Empire Romain, Paris 1961; AE 1956, 123.
37 R. S. O. Tomlin, //. /nscriptions, in S. S. Frere, R. S. O. Tomlin, Roman Britain in 1990, Britannia 22,
1991, 309.
38 T. Wilmott, Birdoswald, in P. Bidwell (Ed.), Hadrian's Wali, 1989-1999, Newcastle Upon Tyne 1999,
1 51-1 54.
39 M. G. Jarrett (n. 24 ).
40 R. W. Davies, The Training Grounds of the Roman Cavalry, AJ 125, 1 968, 79.
41 RIB 815-837; L. P. Wenham, The Garrisoning Of Maryport, Transactions of the Cumberland
Westmorland Antiquarian Archaeological Society, N Ser 39, 1939, 19-30; M. G. Jarrett, G. R.
Stephens, The Roman Garrisons Of Maryport, Transactions of the Cumberland Westmorland
Antiquarian Archaeological Society, N Ser 87, 1987, 62.
108 Tony Wilmott
the garrison, on the fort's parade ground at the votorum nuncupatio each year on
January 3rd. Honourable burial would either take place on the dedication of the new
altar, or at a periodica! lustrum. Of the Birdoswald group, many altars were found
either reused in the fort or further afield, but five were found to the east of the fort
at Wiliowford (Nas. 4, 13, 20, 21, 22), and one at nearby Underheugh (No. 23) (PI.
3, 1 ). Two of these altars, one (No. 1 7) found in the cliff above Underheugh, and the
other ploughed up 'nearer to milecastle 49 than to Birdoswald fort' (No. 11 ), may
have been found in their original positions42 • The Maryport paraliel has been invoked
to suggest that the Birdoswald altars came from a parade ground, and that this lay
between the Wali and the river on the east side of the fort43 • This area, however, was
limited in extent by the river and the Wali, and the interpretation of a parade ground
site here has never been very satisfactory. Recent geophysical survey has nat
supported the idea but has, on the contrary, suggested the existence of a parade
ground to the north of the Wali, near the north gate of the fort44 • The context of the
location of the Maryport altars, which were found in 1 870 buried in pits, has recently
been questioned45 as there is no evidence for a parade ground in the vicinity of their
discovery. Again a parade ground has been identified in a different place, this time to
the south of the fort. Peter Hili46 has invoked the paraliels of rows of altars found in
military shrines at Osterburken and Sirmium to suggest a similar context for the
Maryport altars. David Breeze47 has concurred, concluding that the Maryport altars
were located near a shrine to Jupiter, burial being practised when the available space
had been filied, in order to aliow for the erection of new altars. A similar context is
more logical for the Birdoswald group alsa. The location of the reused altars, focused
on Underheugh and Wiliowford (PI. 1, 2), together with the discovery of two examples
in situ near to milecastle 49, and in the cliff above Underheugh is suggestive. Firstly
it is likely that shrine was located near milecastle 49, and secondly that the altar
deposit has been severely eroded by the retreating river cliff, aliowing altars to fali
down slope and to be reused in the nearby farm. The possibility that further altars
remain to be found in this area, however, is a strong one.
The relationship of the Maryport altars to official „State Religion" has alsa been
questioned by John Mann48 • He emphasises that the dedication of the altars by army
units and their commanders were individual or corporate personal actions, and that
this explains the examples of altars which were either dedicated by the unit without
mentioning the commander or vice versa. Of the other two religious inscriptions from
Birdoswald, which are nat dedicated to /OM, the inscription on a statue base to the
Signis et Numini Augusti (No. 28) may have come from a statue in the aedes of the
principia. The dedication to Mars ·(No. 27) by a military unit requires no comment.
The altars (Table 1) cover the whole of the third century. As discussed, the
earliest may be No. 19, which gives the name of Domitius Honoratus. lf so, there is
evidence for two Severan tribuni, the other being Aurelius Julianus, who built the
42 C. M. Daniels (ed.), The Eleventh Pilgrimage of Hadrian's Wali, Newcastle Upon Tyne 1978, 198.
43 P. Bidweli, N. Holbrook, Hadrian's Wali Bridges, Hbmce Archaeol Reps 9, London 1989, 85, 95.
44 J. A. Biggins, J. Robinson, D. J. A. Taylor, Birdoswald Geophysical Survey, in P. Bidweli (ed.),
Hadrian's Wali 1989-1999, Newcastle Upon Tyne 1999, 1 55-157.
45 D. J. Breeze, The Regiments Stationed at Maryport and Their Commanders, in J. R. A. Wilson (ed.),
Roman Maryport and lts Setting: Essays in Memory of Michael Jarrett, Kendal 1 997, 68-69.
46 P. R. Hili, The Maryport A/tars; Some First Thoughts, in J. R. A. Wilson (ed.), Roman Maryport And
lts Setting: Essays ln Memory Of Michael Jarrett, Kendal 1997, 92-104.
47 M. G. Jarrett, G. R. Stephens (n. 41 ), 70.
48 J. C. Mann, A Note On The Maryport A/tars, in J. R. A. Wilson (ed.), Roman Maryport And lts Setting:
Essays ln Memory Of Michael Jarrett, Kendal 1997, 90-92.
COHORS I AELIA DACORUM. A DACIAN UNIT ON HADRIAN'S WALL 109
horreum between AD 205-208 (No. 1 ). During Julianus' tribunate at Birdoswald his
son, Aurelius Concordius, died at the age of one year and five days and was buried in
the fort cemetery (No. 31 )49 • Subsequent altars can be dated by the imperial honorific
titles granted to the unit, and one (No. 5) by reference to a consulship (Table 1 ). ln
addition to the altars, the building inscription from the porta principalis dextra (No. 2)
records the cohort working in AD 21 9 under the tribunus Claudius Menander. The
latest inscription is an altar dated by the imperial honorific title Probiana to AD 276-
282 (No. 11 ).
AII but one of the datable inscriptions recording the name of the cohort give
tribuni as commanding officers, and the names of seventeen of these are now known.
The latest datable inscription from the site (RIB 1912), a building dedication of AD
297-305 which does nat specifically name the cohort, gives one Flavius Martinus, a
centurion, as the praepositus in temporary command. The single inscription of the
cohort from its previous base at Bewcastle (No. 3 5) records that it was under the
temporary command of an un-named centurion of legio li Augusta. Two other
centurions in such positions are listed an the Birdoswald cohors I Aelia Dacorum altars;
Aurelius Saturninus (No. 1 3 ), and Iulius Marcellinus of legio li Augusta (No. 17). A
fragmentary altar (RIB 1907) upon which the cohort name is nat apparent (and which
therefore does nat appear in the catalogue) records L. Vereius Fortunatus of legio VI
Victrix Pia Fidelis who may be in an analogous position to Saturninus and Marcellinus.
Marcellinus' appointment affords one of two known examples of centurions from
Britannia Superior being employed in the northern province50• During the reign of
Maximinus Thrax (AD 235-238), the cohort was under the command of a former
evocatus of the First Praetorian Cohort, Flavius Maximianus (No. 4). Command of an
auxiliary unit for a praetorian is by no means unprecedented in Britain as well as
elsewhere, and three such commanders were stationed at various times at
Bewcastle51 • Such postings were an alternative to promotion to the position of
legionary centurion52• Although mast of the inscriptions are to /OM alone, the exPraetorian
dedicates to lovi Optimo Maximo Dolicheno. This may reflect a personal
devotion to this eastern aspect of Jupiter, and it is nat stretching the evidence to
suggest that Maximianus might have formed such devotion during his service in Rome.
He does, however, dedicate the altar an behalf of the cohort, and this seems to
demonstrate a personal, but corporate, dedication of the kind suggested by Mann. The
precise date of AD 237 for No. 5, with its consular date, demonstrates that the
tribunus Aurelius Faustus was probably Maximianus' direct successor, and that the
former praetorian thus commanded in AD 235-236. The un-named commander who
dedicated for cohors I Aelia Dacorum Gordiana (AD 238-244) (No. 6) would surely
have been Faustus' direct successor. Another group of more closely dateable altars
appear some two decades later. Marcius Gallicus, the only case where we have two
dedications by one commander, dedicates once to /OM alone (No. 9), and once to /OM
et numini Augusti (No. 8). lt is interesting that the first of these uses the honorific
title Postumiana for the cohort, while the second does nat. The lack of the title an an
altar which is alsa, uniquely in this series, dedicated to the numini Augusti, is surely
significant. lt seems likely that Gallicus' command spanned the period of Postumus'
49 T. Wilmott, The Roman Cremation Cemetery in New Field, Birdoswald, Transactions of the
Cumberland Westmorland Antiquarian Archaeological Society N Ser 93, 79-85.
50 P. A. Holder (n. 22).
51 C. C. Petolescu (n. 23), 47.
52 D. J. Breeze, The Organisation of the Career Structure of lmmunes and Principa/es of the Roman
Army, BJ 174, 1974, 251-254.
11 O Tony Wilmott
usurpation and that No. 8 should be dated to shortly before AD 259, when the
succession was in doubt, and No. 9 shortly after. lf this was the case, then Gallicus'
successor, might have been Probius Augendus (No. 7), who alsa dedicates for cohors
I Aelia Dacorum Postumiana. lt may be significant in terms of context that the two
Postumi an altars were found together '1 00 yds east of Birdoswald', though this
context remains obscure. Spaull53 suggests that the only altar which gives the cohort
the title Augusta (No. 14) may have been dedicated at a similar time of dispute over
the succession, possibly around AD 268-270.
Catalogue No Commanders Honorific/ dating Date
3 ANTONINIANA AD 213-222
4 Flavius Maximianus MAXIMINIANA AD 235-238
5 Aurelius Faustus cos Perpetuus [ +Cornelianus] AD 237
6 GORDIANA AD 238-244
8 Marcius Gallicus 7c. AD 258-259
9 Marcius Gallicus POSTUMIANA AD 259-268
7 Probius Augendus POSTUMIANA AD 259-268
10 Pomponius Desideratus TETRICIANORUM AD 270-273
11 Aurelius Verinus PROBIANA AD 276-282
Table 1. Summary of dated /OM altars from Birdoswald in probable chronological order.
lf Holder's view of the origin of the cohort is correct, it had been in Britain for
several generations by the time it is first recorded at Birdoswald, and it would have
been normal practice to make up numbers from the local British population 54 • Despite
the probability that this occured within cohors I Aelia Dacorum, there is clear
evidence that the Dacian character was not entirely dissipated, and was certainly not
forgotten. The evidence for this is twofold. Firstly the appearance of falces on the
two building inscriptions (Figs. 2, 3), and secondly the use of Dacian personal names.
The use of the falx in this epigrahic context is unique to the Birdoswald inscriptions,
and this writer is not aware of any other case where ethnic weaponry is displayed on
inscriptions of auxiliary units. Whether this was merely a regimentai badge, or
whether it depicts part of the actual equipment of the cohort is not known. John
Coulston's55 discussions of this issue come to no clear conclusion, though he favours
the idea of a regimentai badge. He does, however, raise an interesting potential
parallel; that of the Gurkha troops whose regiments form an integral part of the
modern British army. These soldiers have been recruited into regiments under British
officers from the Himalayan kingdom of Nepal since 181656 , and their mast recent
combat role was as part of the task force in the Anglo-Argentinian conflict over the
Falkland lslands (Malvinas) in 1982. The Gurkhas have a traditional edged weapon - a
fearsome curved knife known as the kukri - with which they are still equipped on
parade, in combat, and in their martial dances which are performed at military
displays. Every Gurkha regiment has as its badge a pair of crossed kukris. To press
53 M. G. Jarrett (n. 24 ).
54 B. Dobson, J. C. Mann, The Roman Army in Britain and Britons in the Roman Army, Britannia 4, 1973,
191-205.
55 J. C. N. Coulston, A Scu/ptured Dacian Falx from Birdoswald, Archaeol Aeliana 5 Ser, 9, 1981, 348-
350; J. C. N. Coulston, E. J. Phillips, Corpus Signorum Imperii Romani: Great Britain 16. Hadrian's Wali
West of the North Tyne, and Carlisle, Oxford1987, 106.
56 P. Mason, A Matter of Honour. An Account of the Indian Army, lts Officers and Men, London 1974, 137.
COHORS I AELIA DACORUM: A DACIAN UNIT ON HADRIAN'S WALL 111
this parallel îs, however, probably taking the comparison between British India and the
Roman Empire too far57 •
The second piece of evidence for an ethnic Dacian character at Birdoswald îs the
presence of at least one Dacian name. ln 1752 a tombstone was found to the west of
the fort, where the fort cemetery lies (No. 32). This was set up to the memories of
two children, Decibalus, who lived only a few days, and Blaesus. The Dacian origin of
the name Decibalus îs obvious, and it seems likely that the father of the two boys was,
therefore, a Dacian soldier. Unfortunately the tombstone îs very plain, and shows no
design which might be dated stylistically58• The presence of this tombstone at
Birdoswald shows that it îs likely to have been late second or early third century în
date at the earliest, as that îs when the Dacian cohort probably arrived. The fact that
Decibalus died after only days of life further shows that the name was given at this
time. lt was tenuously suggested59 at the seminar that two further names were Dacian
în origin. These were centurions; Aelius Dida (No. 34) was în charge of the century
building the Valium at Benwell. This was a Hadrianic building operation, and the use of
the name Aelius suggests a centurion who had been granted citizenship rather
recently, though this may fit happily with the notion that the cohort came to Britain
with Hadrian. The second îs Decius Saxa (No. 30), who appears on a centuria! stane
at Birdoswald60•
The falx inscriptions and the Decibalus tombstone are the sole evidence which
definitely indicates some kind of ethnic Dacian influence at Birdoswald. The falx
inscriptions are internally dated to the early third century, while the date of the
Decibalus tombstone îs inferred from the probable arrival date of the cohort at this
time. The inscriptions show Dacian identity being asserted about a century after the
unit was raised, and therefore after one might assume several drafts of new, local
recruits had taken place. AII of the original Dacian recruits must have been discharged
by the time of the diploma of 1 27, the generation of the veteran ltaxa. lf the cohort
arrived at Birdoswald around 200, then Decibalus' father can hardly have been one of
the original, Trajanic, founder-recruits of the cohort. This îs the best of evidence we
have for the possibility of continued recruitment from the Dacian homeland, but it
must be treated with caution. lt remains possible that the name was given by a local
recruit as a nod to the heritage of the unit.
There îs no other evidence for Dacian recruitment to this cohort into the third
century. Excavation of around 21 % of the fort interior has provided no evidence of
any sort of characteristic material culture, which might have identified the cohort as
„different". One place where one might expect this to be apparent îs în the fort
cemetery, but excavation here has hitherto been extremely small în scale61 • ln the
third century, the fort at Brougham în north-western England was the base of the
Numerus equitum Stratonicianorum62• The report on the excavation of its cemetery îs
57 Cf. R. Hingley, Roman Officers and English Gentlemen: The Imperial Origins of Roman Archaeology,
London 2000, 1 59-160.
58 J. C. N. Coulston, E. J. Phillips (n. 55), 92.
59 By Prof. Nantris.
60 This author is no linguist. Whether or not these names are Dacian in origin awaits expert comment.
A. Mocsy et al., Nomenclator Provinciarum Europae Latinarum et Gallia Cisalpinae, Dissertationes
Pannonicae, Ser 3, 1, Budapest 1983, lists one other example of the name Saxa, from Gallia
Lugdunensis, and Dida twice in Gallia Cisalpina, and once each in Hispania, Gallia Belgica or Germania
and Moesia.
61 T. Wilmott (n. 49), 79-85; T. Wilmott, forthcoming, Excavation and Survey at Birdoswald, 1996-
2000, English Heritage Archaeological Reports, London.
62 RIB 780; C. C. Petolescu (n. 23 ).
112 Tony Wilmott
approaching publication, and Hilary Cool has demonstrated that the foreign, în this
case Danubian, character of these troops and their dependants would have been very
clear from their dress and personal adornment63• There îs no evidence that this was
true of cohors I Aelia Dacorum. Having said that, the very uniqueness of the device of
the falx and the palm frond îs characteristic enough, and the possibility cannot be
dismissed that the father of Decibalus was a Dacian recruited to a specialist unit armed
with the falx6 4 • The only way of proving this would be to find such a weapon în
excavation.
lf Dacian recruitment really was the case up to the early third century, it probably
stopped very soon after, as the falx appears on none of the /OM altars throughout the
third century. The final reference to cohors I Aelia Dacorum îs în the Notitia
Dignitatum. This later fourth century list of offices and units în the empire shows at
least that a unit called cohors I Aelia Dacorum was still în garrison at Birdoswald.
However, even if Dacian recruitment had survived into the early third century, there îs
no evidence at all that this continued throughout the century. lt would not, în any
case, have survived the abandonment of the tria Daciae at the end of the third
century. One must conclude that during the fourth century cohort was certainly a
locally recruited unit of limitanei which was similar to the others on Hadrian's Wali: a
unit whose ethnic heritage was only discernible în its name. [1]
From Dan Ruscu's The Supposed Extermination of the Dacians: The Literary Evidence:
And Roger Tomlin's thoughts on the Dacian unit at Birdoswald:
"So many other auxiliary units are recruited from recently-conquered peoples – and then carefully sent to other provinces to get them out of the way – that I can't think this didn't happen in Dacia as well. The diploma of AD 127 (RMD IV.240) is very helpful, in showing that cohors I Aelia Dacorum was already part of the British garrison, and in being issued to a 'Dacian', even if he belonged to another unit. The Vallum inscription (RIB 1365) surely confirms it was there then, even if the Aelia title is omitted for lack of space.
It is surely right to suppose that the cohort was not recruited exclusively from Dacians, but not right to exaggerate the literary evidence into supposing that it contained no Dacians. The falces at Birdoswald are a good argument to the contrary.
An argument ad absurdum. If the cohort consisted only of aggressive Roman colonists, why wasn't it simply kept in Dacia to keep an eye on the three Dacian teenagers who survived the conquest?
I would have been surprised to find Sarmatians in a 'Dacian' unit. [2] Not surprised to find Thracians, though, on the assumption that some had settled across the Danube. For what it's worth, the centurion of RIB 1365, Dida, has either a Thracian or Dacian name and he belonged to the Cohors I Aelia Dacorum [https://romaninscriptionsofbritain.org/inscriptions/search?qv=dida&submit=]."
[1]
While it may be true that a speficially Dacian cultural identity had been lost by the late period at Birdoswald, we must remember that the Dacian draco (https://mistshadows.blogspot.com/2019/03/the-draco-standard-by-jcn-coulston.html) was incorporated into the Roman army and so this particular Dacian artefact may have continued to be of significant value to the inhabitants of Birdoswald - including the sub-Roman inheritors of the site.
It may even be that the draco was so sacred to the Dacian unit garrisoned at Birdoswald that the fort was referred to as the 'Aelian Dragon', a distinction due entirely to the unit's affinity with the draco standard. See
[2]
From Julian Bennett's Trajan: Optimus Princeps, P. 167:
The Roxolani, of Wallachia and Moldavia, on the other hand, had proved consistently
hostile to Rome. They had supported Decebalus in the insurrection of 105, and now
found as much as half of their traditional southern grazing grounds incorporated into the
Roman province of Dacia. In their case a similar confrontation over land rights was
averted by diplomacy. While the Romans had needed to occupy the vast and featureless
Wallachian and Moldavian steppes in order to attack Dacia from the east, the territory
was not necessary for the subsequent development or protection of the province. True,
the steppes were useful in their own right, as prime grazing land, but it was also realized
that they could not sustain settled occupation, and would be almost impossible to hold and
control permanently without the deployment of several cavalry units, backed by a
network of established forts and linear obstacles.16 Consequently, as archaeological
excavations have shown, most of the forts in Wallachia and Moldavia may well have been
quickly abandoned after the province was formed, a few being retained as outposts,
indicating that these territories, the transdanubian extension of Moesia Inferior, were
restored to the Roxolani. Oltenia, however, was retained, to secure the right flank of the
province, in return for which Trajan granted a subsidy to the tribe.17
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.