Vindolanda was a Roman auxiliary fort (castrum) just south of Hadrian's Wall in northern England, which it pre-dated. Archaeological excavations of the site show it was under Roman occupation from roughly 85 AD to 370 AD. Located near the modern village of Bardon Mill in Northumberland, it guarded the Stanegate, the Roman road from the River Tyne to the Solway Firth. It is noted for the Vindolanda tablets, a set of wooden leaf-tablets that were, at the time of their discovery, the oldest surviving handwritten documents in Britain.
History and garrison
The site is a hill on the Stanegate road, with steep slopes on the north, east, and south sides. It originally had a deep dip running north-south through the centre of the hill, which was gradually filled up by successive layers of occupation. There is currently no evidence for settlement on the fortress site before the Roman period, but there was an iron-age hillfort at Barcombe Hill, 3⁄4 mile (1.3 km) to the northeast (which served as a quarry and signal station in the Roman period). Two pre-Roman farmsteads have been found in the area and two standing stones called the Mare and Foal, about 3 miles (5 km) to the west, are the remnants of a stone circle, which might have been the local religious centre. The name "Vindolanda" is probably Celtic, meaning "white lawn" or similar. An inscribed altar records a group called the Textoverdi, who may have been the native inhabitants of the area.
Late first century AD (Periods I–III)
Archaeological excavations, inscriptions, and the tablets indicate that Vindolanda was under Roman occupation from roughly 85 AD to 370 AD. The garrison consisted of a series of different infantry and cavalry auxilia units, not components of Roman legions. The first unit to be based at the site was the Cohors I Tungrorum (First Cohort of Tungrians), composed of Tungri from Gallia Belgica, led by Julius Verecundus, who seem to have arrived after Gnaeus Julius Agricola's conquest of the north and built the First Timber Fort (Period I). Vindolanda Tablet 154, a troop roster, records that the unit contained 746 men and 6 centurions, of which 295 men and 1 centurion were then at Vindolanda itself, while 335 men and 2 centurions were away at Coria (Corbridge), and 116 men and 3 centurions were away on other missions. Of the troops at Vindolanda, 15 were sick, 6 were wounded, and 10 were out of action with conjunctivitis. Around 92 AD, they rebuilt the site as the Second Timber Fort (Period II).
In the 90s AD, the Cohors I Tungrorum were replaced by the Cohors IX Batavorum (ninth Cohort of Batavians), a combined infantry and cavalry cohort of 1,000 men, composed of Batavi from the Rhine delta in Germania Inferior (modern-day Netherlands). They had the privilege of being commanded by one of their own noblemen, first Flavius Genialis and later Flavius Cerialis. Shortly after their arrival, they rebuilt the site again as the Third Timber Fort (Period III). The majority of the tablets found at Vindolanda belonged to Cerealis' archives. These include the correspondence of his wife Sulpicia Lepidina and his children's school exercises. In 104 AD, the cohort was ordered to abandon the fort in order to join Trajan's second invasion of Dacia. The fort was demolished, valuables were removed, and rubbish was burnt on bonfires. The tablets were among the items thrown on the bonfire, but a chance rain storm seems to have saved them from destruction.
Second century AD (Periods IV–VIb)
The Cohors I Tungrorum returned to Vindolanda in or after 105 and built the Fourth Timber Fort (Period IV). They were joined by a contingent of Varduli cavalry, a Basque-speaking group from northern Spain. A large tombstone for Titus Annius, a centurion of Cohors I Tungrorum, which records that he was "killed in the war", probably indicates that the cohort was involved in heavy fighting that broke out in northern Britain in 117 AD. This seems to have led to the cohort's reduction from 800 to 500 men. This fighting prompted the construction of Hadrian's Wall slightly to the north and the fort probably served as an administrative hub during the construction of Hadrian's Wall. Vindolanda Tablet 344 and a very large wooden building may indicate that Hadrian resided at the fort for a period, as he oversaw the initial stages of construction in 122 AD. In the 120s, the cohort rebuilt the fortress again as the Fifth Timber Fort (Period V). The original plan for Hadrian's Wall was for it to consist of a turf wall with a series of milecastles and watchtowers along its length, but the main garrison remaining at Vindolanda and the other existing fortresses along the Stanegate road, so the construction of the Fifth Fort may have been in preparation for it to serve in this role. Later, it was decided to build new fortresses on the wall itself. Around 128 AD, the Cohors I Tungrorum abandoned Vindolanda and moved into the nearest of these new fortresses, Vercovicium (Housesteads Roman Fort, about two miles (three kilometres) to the north-east of Vindolanda).
In 142, Antoninus Pius moved the frontier north from Hadrian's Wall to the new Antonine Wall, but around 162 this was abandoned and the troops returned to Hadrian's Wall. It is unclear what happened to Vindolanda in this period. There are signs of a new timber fortress of late Hadrianic or early Antonine date. By 162, Vindolanda was occupied by the Cohors II Nerviorum (Second Cohort of Nervians), an infantry force composed of Nervii, a Belgic tribe from Gallia Belgica, who converted the timber fort into Stone Fort I (Period VIa).
In the early third century AD, there was an increase of attacks on the Romans in northern Britain and Emperor Septimius Severus decided to resolve the matter by personally leading an invasion of Caledonia (modern-day Scotland) in 208 AD. As part of these campaigns, a short-lived stone fortress, the Severan complex (Period VIb), was built at Vindolanda. The identity of the garrison in this period is uncertain. The Severan complex incorporated up to two hundred circular structures, which may have housed irregular auxiliary troops, native British civilians employed at the nearby quarries, hostages taken by Severus during the campaign, or pro-Roman British civilians seeking shelter from the disorder produced by the invasion. A skull found in the south ditch of the fortress was probably displayed before the fortress as a war trophy. Septimius Severus died at York in 211 AD; his sons Caracalla and Geta paid off the rebels and left for Rome.
Third and fourth centuries AD (Periods VII–IX)
Sometime in the 210s, the Severan complex was demolished and replaced with Stone Fort II (Period VII) and a vicus (civilian settlement) was built to the west of the fortress. The majority of the structures that are visible today belong to this period or later. The occupants of the site at this time were the Cohors IV Gallorum equitata (Fourth cavalry Cohort of Gauls). A dedication to Caracalla shows that they were settled at Vindolanda by 213 AD. It had been presumed that the title of the cohort was, by this time, purely nominal, with auxiliary troops being recruited locally but an inscription found in 2005 suggests that native Gauls were still to be found in the regiment and that they liked to distinguish themselves from British soldiers. The inscription reads:
CIVES GALLI
DE GALLIAE
CONCORDES-
QUE BRITANNI
Translation:
The troops from Gaul dedicate this statue
to the goddess Gallia
with the full support of
the British-born troops
Another inscription records that the cohort rebuilt the western gate of the fortress in 223 AD. The cohort was commanded by Quintus Petronius Urbicus under Severus Alexander (222–235 AD). Marcus Caecilius Celer, Pituanius Secundus, and Gaius Sulpicius Pudens are also recorded as commanders, but their tenures cannot be dated. The final evidence for the cohort at Vindolanda is a fragmentary inscription recording renovations or construction under the emperor Probus (276–282 AD). Religious dedications survive from this period for a range of gods. Seventeen inscribed dedications have been found for Roman deities: seven for Jupiter, two for the genius (protective spirit) of the fortress headquarters, three for Mercury, one for the Fortune of the Roman People, one for Vulcan, one for Silvanus, one for Mars Victor, and one for Hercules. There are also statues of Diana, Hercules, and Priapus. Twenty dedications are attested for deities of northwestern Europe: twelve dedications to the Veteres ('old ones'), five for mother goddesses, and individual dedications for Cocidius, Mogons, and Maponus. There are also statues of the mothers and of Maponus. One dedication was erected for the goddess of Hama in Syria (a detachment from that city was based at Carvoran to the west).
Around the end of the third century, the fortress was briefly abandoned and then reconstructed (Period VIII). This may have been the result of reforms to the border defences of the Roman Empire by Diocletian. Over the fourth century, the size of the garrison declined and the vicus was abandoned. The fortress was once again rebuilt in the second half of the fourth century (Period IX), perhaps following Count Theodosius' defeat of the Great Conspiracy in 368. This rebuild was accompanied by the construction of an apsed building, possibly a church.
The last Roman troops left Britain in 407 AD, but occupation at Vindolanda continued for an uncertain amount of time. A small church was built inside the courtyard of the old headquarters sometime after 400 AD. A Christian dedication dating to around 600 AD has been found at and near the site, which may indicate that there was a monastic establishment at Vindolanda in this period. A tombstone for one Brigomaglos, whose name suggests Welsh origins, was found near Vindolanda and also dates to around 600 AD. He might have been a priest or a garrison commander for one of the kings of the Old North. Some Anglo-Saxon brooches have been found on the site.
Description
The site consists of a series of fortresses built one on top of the other, which have been divided by archaeologists into nine periods. The first five periods are successive forts built of wood and turf with different footprints (Periods I–V). They are buried as much as 13 feet (4 m) deep in the anoxic waterlogged soil and are covered by the remains of later periods, which makes excavation difficult, but has also led to the preservation of large amounts of organic matter. The fifth fort was converted to stone in period VIa, replaced with a new stone fort (the Severan complex) in Period VIb, and replaced with a second stone fort (Periods VII-IX), which accounts for most of the remains visible on the site today. Outside the fortress itself, a large bathhouse was built to the south in Period II, a smaller bathhouse and two temples were built to the west in Period VIa, and a vicus (self-governing village) existed to the west of the fort in Period VII.
First timber fort (Period I)
The first timber fortress covers largely the same area as the stone fortress which is visible today, although on a slightly different alignment, and only a few sections of the outer ditch that surrounded the fortress have been excavated, at the western and southern sides. These contained various items of rubbish which were dumped there towards the end of the fort's life, including various wooden, leather, and metal objects, notably writing tablets, a toy wooden sword, the crest for a centurion's helmet, made from local hair moss, and an iron stylus. The date of this fort is placed ca. 85 AD by a cache of Samian ware and a coin of Vespasian.
Second timber fort (Period II)
The second timber fort was about twice the size of the first fort and, unlike the first fort, it was aligned with the cardinal directions. The remains of this layer are up to 13 feet (4 m) below the modern ground level. The western portion of the wattle and daub timber praetorium (headquarters), the fortress' south gate, and adjacent parts of the south rampart have been excavated. Dendrochronology suggests that the date of construction was 92 AD or shortly thereafter.