Vindolanda

Vindolanda was a Roman auxiliary fort on a hill just south of Hadrian's Wall, near the modern villages of Bardon Mill and Henshaw in Northumberland. It guarded the Stanegate, the Roman road that ran from the River Tyne to the Solway Firth, and it was already occupied before the Wall was built. The name is probably Celtic and means something like "white lawn". An inscribed altar names a group called the Textoverdi, who may have been the people living in the district.

There is no evidence of a settlement on the fortress hill itself before the Roman period. An iron-age hillfort stood on Barcombe Hill, about three quarters of a mile, or 1.3 kilometres, to the northeast, and in the Roman period that hill was used as a quarry and a signal station. Two pre-Roman farmsteads are known nearby. About three miles, or five kilometres, to the west, the standing stones called the Mare and Foal are what remains of a stone circle, which might have been a local religious centre.

The hill has steep slopes to the north, east, and south. A deep dip once ran north to south through the middle of the hill and was filled, layer by layer, by successive occupations. The visible stone fort stands on a platform above the Chainley Burn to the east and the Doe Sike to the south. Excavation has divided the sequence into nine periods.

The first five were forts of timber and turf, each with a different footprint. Later stone forts were built over them. An account of the site says it was demolished and completely rebuilt no fewer than nine times. The earliest timberwork lies as much as 13 feet, or 4 metres, down, and some deposits reach 20 feet, or 6 metres.

Elsewhere the primary fort is described as lying 6 metres beneath the later stone forts. The soil at those depths is waterlogged and starved of oxygen, so wood, leather, cloth, and bracken survived where they would have rotted on a dry site.

The fort belonged to the Roman Empire, the state that held the Mediterranean and much of Europe, western Asia, and North Africa. Single-person rule followed the settlement of 27 BC, and the western half of the empire ended in 476 while the eastern half lasted until 1453. In Britain the army's northern edge moved more than once. The Stanegate, whose name means "stone road" in Northumbrian dialect, linked Corbridge on the Tyne with Carlisle on the Eden and ran on toward the Solway.

It is thought to have been laid out under Gnaeus Julius Agricola, between 77 and 85, while the fighting edge was still farther north, on the Forth-Clyde line, and the forts along it were then a day's march apart, about 14 Roman miles, or 13 modern miles, or 21 kilometres. Unlike many Roman roads it followed easier gradients and wandered. A large stretch between Fourstones and the fort is still a minor road. When the army began to leave Scotland from around 87, the Stanegate became a frontier in its own right and forts were added at a half-day interval, among them Newbrough, Carvoran, and Brampton Old Church.

The road remained the main east-west route behind Hadrian's Wall until a new Military Way was built in the second half of the second century.

That British road-and-fort frontier was one local answer. Far to the southeast, the Norican Limes was another. It guarded part of the province of Noricum from the first century to the fifth, along the Danube from Passau, or Boiodurum, to Zeiselmauer, or Cannabiaca, in what are now Upper and Lower Austria. It was a river border, held by a loose chain of cohort forts, and its main road was the via iuxta amnem Danuvium.

Wood and earth camps were converted into stone under Hadrian, strengthened again in the fourth century, and watched by towers and, later, burgi. In the Wachau, between Favianis and Melk, towers were few because the forested cliffs already made the bank hard to reach. Each camp had a landing stage, because the river was the trade route as well as the border, and civilian vici grew up beside the camps, with walled towns such as Aelium Cetium and Ovilava, now Wels, behind them. In late antiquity Noricum was split into an upper and a lower part.

The whole Roman frontier is measured in two ways that do not match: one circuit, from the Atlantic coast of northern Britain to the Black Sea, the Red Sea, and back across North Africa, is given as more than 5,000 kilometres, and another as more than 10,000 kilometres, or 6,214 miles. The Upper German-Raetian stretch is given as about 550 kilometres in both.

Gnaeus Julius Agricola, born on 13 June 40 at Forum Julii in Gaul and dead on 23 August 93, commanded much of the conquest of Britain. He had served there as a tribune under Gaius Suetonius Paulinus, including the suppression of Boudica in 61, and later commanded the Twentieth Legion against the Brigantes. He was suffect consul in 76 or 77, and he is also said to have been made consul and governor in 77, arriving in the middle of that summer. He defeated the Ordovices in north Wales, took Anglesey, and pushed through northern England into Scotland.

In the summer of 83 he fought the Caledonians under Calgacus at Mons Graupius. The history written by his son-in-law puts the Caledonian strength above 30,000, the Caledonian dead at about 10,000, and the Roman dead at 360, and says two thirds of the Caledonian army escaped into the highlands. The battle is also placed in 83, probably near Aberdeen. Agricola was recalled in 85, after an unusually long governorship, and he received triumphal decorations but no further command.

He died in 93.

The first unit at Vindolanda was not a legion. It was the First Cohort of Tungrians, infantry raised among the Tungri of Gallia Belgica, the people of the civitas whose capital was Tongeren. Tacitus treated the Tungri as the group first called Germani, the name that was later extended to peoples east of the Rhine. Their first cohort was commanded by Julius Verecundus, also written Iulius Verecundus.

A writing tablet, a troop roster, records 746 men and 6 centurions in the unit. Of those, 295 men and 1 centurion were at the fort, 335 men and 2 centurions were at Corbridge, and 116 men and 3 centurions were away on other duties. Of the men on the spot, 15 were sick, 6 were wounded, and 10 were unfit with conjunctivitis. The date of this first timber fort is not given as a single year.

One account puts the forts at Vindolanda, Nether Denton, Corbridge, and Carlisle together in the 70s and 80s. Another places the first fort after Agricola's campaigns of the mid-80s, about 85, on the evidence of unused samian pottery and a coin of Vespasian, and says it had been demolished by 92 at the latest, when the second fort was commissioned. Only parts of the outer ditch, on the west and south, have been dug. The fill held wood, leather, metal, writing tablets, a toy wooden sword, an iron stylus, and the crest of a centurion's helmet made from local hair moss.

Around 92 the Tungrians rebuilt the site as a second timber fort, about twice the size of the first and, unlike it, aligned to the cardinal directions. Another account says this second fort was built between 90 and 92, extended 100 metres farther west, and covered about 2.8 hectares. Dendrochronology points to 92 or shortly after. The western part of a wattle-and-daub praetorium, the south gate, and the adjoining rampart have been excavated.

That western wing is also described as 45.75 metres long and at least 11.75 metres wide, built over the filled ditch of the first fort. The south rampart was turf, 15 feet or 4.5 metres wide. The gate was a gap 10 feet and 5/8, or 3.23 metres, wide, revetted with alder boards and four large oak posts on each side, with a softwood lintel. It sat in a dip that carried water off the hill.

Clay, branches, and split logs laid to the north of the gate did not cure the instability. A water tank in the courtyard was 35 feet and 1/4 long, 10 feet and 1/4 wide, and 4 feet deep, about 10.75 by 3.125 by 1.26 metres, with wattle-and-daub sides, fed by a conduit from the north. Floors of hard clay were covered by a layer of bracken about 0.7 to 0.8 metres thick, probably renewed every year. Tablets and pieces of tent were preserved in it.

Outside the fort, to the southeast of the gate and above the drop to the Doe Sike, the cohort built a bathhouse of the row type, with cold rooms at the west end and hot rooms and a furnace at the east. A further heated room was added on the north and a cold plunge on the south, with a separate latrine just south of the plunge. A tablet places the building in this second period. The bricks and tiles were probably made on site.

The work shows that an auxiliary cohort, not only a legion, could carry out a complicated building. The garrison later shrank, the baths became too large to run, and they were demolished by the middle of the second century at the latest.

In the 90s the Tungrians were replaced by the Ninth Cohort of Batavians, a mixed infantry and cavalry unit of 1,000 men from the Rhine delta in Germania Inferior, the region of the modern Netherlands. The Batavi had an old arrangement with Rome under which they supplied soldiers and arms rather than tribute, and their cohorts were commanded by their own nobles. They had provided eight or nine or ten cohorts in the first century. Eight of them served in the conquest of Britain.

In 69, during the Year of the Four Emperors, Julius Civilis led the Batavian revolt. After it, Batavian units were posted to Britain again, and in the second century many were moved to the Danube. At Vindolanda they were commanded first by Flavius Genialis and then by Flavius Cerialis, whose name is also written Cerealis. Soon after they arrived they rebuilt the fort again, on the same lines as the second fort but to a higher standard, using oak rather than the unseasoned ash and alder of the previous build.

The tablets put this third period at 97 to 105. The explored praetorium was 166 feet and 1/8, or 50.65 metres, from north to south, and at least 102 feet and 1/2, or 31.25 metres, from east to west, one of the largest known in Britain. Interior walls stood on oak beams, with alder and birch uprights, wattle and daub, and a plaster wash. Bracken, cut in July or August, dried for a month, and then used for a year, covered the floors.

Nuts, shells, stable-fly pupae, and traces of excrement and urine were lodged in it. The rooms would have been warm, and they would have smelled.

Nineteen rooms have been identified, with a southwest yard and an internal courtyard. A large room in the southwest corner, 31 by 18 feet, or 9.5 by 5.5 metres, with beams for an upper storey, was probably the prefect's audience room. A long narrow room to the west held many tablets and is read as a writing office. A small kitchen, about 2.95 metres square, had two bread ovens and a rubbish pit.

Its store used a recycled tabletop as a floor. A room that was not fully cleared held a stack of oak roof tiles. Another held a sock and a nearly complete chamfron, the face armour for a horse. The northwest room was a forge, its floor covered by 40 millimetres of coal dust, charcoal, and ash and hundreds of bronze, iron, and lead scraps, mostly military.

Rooms east of a corridor are read as workspaces for the prefect's enslaved household. Finds there included a comb in a leather case, a crowbar left in the wall, gutting knives, and barrel staves branded as imports from the Rhone valley. The cobbled courtyard had three long oak fittings whose purpose is unknown. A wooden pipe of bored alder trunks, jointed with oak boxes, brought water from wells at the western edge of the plateau.

The trees were felled in 97. The south gate was moved 6 feet and 7/8, or 2.08 metres, west of its predecessor, and another account says 2 metres west, to keep runoff out of the passage. Eight modern posts now mark it on the surface.

Most of the ink tablets, 128 of them from this layer, come from bonfires of official papers and rubbish set west of the praetorium. Rain put the fires out and mud sealed the wood before it burned. The archive is largely that of Cerialis. It includes his wife's correspondence and school exercises belonging to his children.

Sulpicia Lepidina, his wife, received two letters from Claudia Severa, wife of Aelius Brocchus, who commanded a nearby fort. One is an invitation to a birthday. A scribe wrote most of it, and Severa added a closing in her own hand, calling Lepidina her sister. With another fragment in the same hand, the invitation is taken to be the oldest writing by a Roman woman yet found in Britain, or perhaps anywhere.

A further letter, from a woman whose name is hard to read and who is thought to be called Paterna, is too damaged to decide between two readings: either she was bringing remedies, or she was sending two enslaved girls, free of fever, one for Lepidina and one for a person whose name is lost. The letters were found in the 1970s in the bracken and in filled ditches. A selection of the tablets has been shown at the fort museum since 2011, lent by the museum that holds the collection.

The cohort was then ordered to leave. One account says that in 104 the men abandoned the fort to join Trajan's second invasion of Dacia, that valuables were removed, and that rubbish, including the tablets, was thrown on bonfires. Another says the garrison left hurriedly, probably for the Danube, in 102 or 103. Trajan's Dacian Wars were two campaigns, in 101 to 102 and in 105 to 106, against King Decebalus.

The first ended in a treaty. Decebalus broke it, and the second war began in 105, crossed the Danube on a stone bridge at Drobeta, and ended in 106 with the fall of Sarmizegetusa and the king's suicide. After 105 the fort stood empty for a short time. It was then levelled and covered with clay and turf up to 85 centimetres, or 33 inches, deep.

Dendrochronology puts the new build within about a year of 105.

The Tungrians came back in or after 105 and built a fourth timber fort on the same alignment, extended 200 feet, or 60 metres, farther west, the largest fort ever built on the hill. A contingent of Varduli cavalry, from northern Spain, joined them. A large tombstone of Titus Annius, a centurion of the cohort, says he was killed in the war. That is read as evidence of heavy fighting in northern Britain in 117, after which the cohort was reduced from 800 men to 500.

The fighting is also the context given for the decision to build Hadrian's Wall a little to the north. Hadrian became emperor in 117. Britain had rebelled in 118 and 119. Inscriptions record a British expedition and the movement of about 3,000 men in a detachment, and coins of 119 and 120 mark the sending of Quintus Pompeius Falco to restore order.

In 122 Hadrian ordered a wall "to separate Romans from barbarians". Whether the wall answered a present threat, or a wish to stop expanding, or a wish to hold the edge at a lower cost than a massed field army, is not settled. He had probably planned it before he arrived. Sandstone fragments from Jarrow, dated 118 or 119, speak of his wish to keep the empire intact.

The wall runs from Wallsend on the Tyne to Bowness-on-Solway. Its length is given as 80 Roman miles, as 73 modern miles, as 117 kilometres, and as 117.5 kilometres. A wider description of the frontier gives the wall as 118 kilometres, and also as 130 kilometres from the mouth of the Tyne to the Solway, with a further 36 kilometres of installations down the Cumbrian coast. It was begun in 122.

Most of it was built in six years by three legions, the Second Augusta, the Sixth Victrix, and the Twentieth Valeria Victrix, about 15,000 soldiers, with help from the fleet. The planned thickness of 10 feet, or 10 Roman feet, was reduced to about 8 feet, or to about 7.5 feet on the narrow sections, while broad sections are about 9.5 feet. The western stretch was first built of turf and later replaced in stone, the first phase under Hadrian and the second after the return from the Antonine Wall, though that second phase has also been placed under Septimius Severus. Forts along the wall are described as each holding 600 men, with milecastles held by 12 to 20.

A ditch and banks, the Vallum, run behind the curtain. The original scheme kept the main garrisons on the Stanegate, so the fifth timber fort at Vindolanda may have been built for that role. New forts were then added to the wall itself.

A very large wooden building north of the earlier praetorium, with opus signinum floors, plastered walls, and a private bath, has been read as a house for Hadrian and his party while he watched the first stages of the wall in 122. Tablet 344, and an appeal addressed to "your majesty" from a centurion's rooms in the barracks, are part of that reading. It remains a suggestion, not a recorded stay. Hadrian wintered in 122 to 123 at Tarraco in Spain, so any stay on the frontier was not the whole year.

The same fourth-fort layer has a junior officers' mess that burned, a barracks of fourteen single rooms in two rows of seven, later with the eastern verandah turned into extra rooms, and two larger apartments at the north end for the centurions. Each small room may have held up to ten soldiers. Small shoes may have belonged to children and wives, or to enslaved boys. One room held more than a thousand leather fragments from a tent, and the building also produced tablets, shoes, pens, combs, ballista bolts, and tools.

About thirty metres farther west, a courtyard building with probes, tweezers, spoons, knives, razors, and needles is read as an infirmary. Oak posts of a western defence, 2 feet square and set every 21 feet for more than 115 feet, were felled between 102 and 112.

Those ballista bolts belong with the artillery the army called Roman siege engines. The machines were mostly adapted from Hellenistic designs. Until the late Republic they were used when a siege required them, and a ballista had no permanent place on a legion's books. By the first century AD every century, a group of 60 to 100 men, is said to have had one, under a chief of the ballistae, with specialists and artillerymen.

The weapon looked more like a large crossbow than a catapult. Two arms sat in vertical skeins of rope, were hauled back with a winch, and threw either stones or bolts. Early machines were wood, clamped with iron. They were accurate enough that single men are said to have been picked off, though light bolts did not carry the force of the later stone-throwers, whose shot could weigh 90 to 135 kilograms.

Nothing in the fort's record describes a siege of Vindolanda itself. The bolts in the barrack, and an ox skull later shown by three-dimensional imaging to have been used for target practice, are evidence of weapons kept and trained with, not of an assault on the hill.

The fifth timber fort is the least known, because the foundations of the first stone fort cut it up. On the site of the earlier headquarters a workshop measured 157 feet, or 48 metres, north to south and about 66 feet, or 20 metres, east to west, with flagstone floors, alder and oak posts, coal dust, iron slag, a forge, and deep pits. Another courtyard workshop farther west, at least 84 by 52 feet, had a hearth in every room. This fort is placed in the 120s, and also in the span from after 112 to 150, overlapping Hadrian's visit.

Around 128 the Tungrians left for Housesteads, the new wall fort about two miles, or three kilometres, to the northeast. A later list of commands puts the First Cohort of Tungrians at Housesteads from 205 or 208, so the move around 128 and the later stationing are both on record. The Second Cohort of Tungrians, nominally a thousand strong and part-mounted, was at Birrens from 159 to about 184, and a fourth cohort was at Abusina in the second century.

What happened on the hill between the move to the wall and the return of the frontier is unclear. There are signs of a new timber fort of late Hadrianic or early Antonine date. In 142 Antoninus Pius moved the frontier north to the Antonine Wall. He was emperor from 138 to 161, the fourth of the adoptive emperors, and he died of illness on 7 March 161.

He never visited Britain. A campaign in southern Scotland early in his reign produced the new wall, supervised at first by the governor Quintus Lollius Urbicus. The wall is a turf bank on a stone base, about 3 metres high and 5 metres wide, with a ditch to the north, from the Forth to the Clyde. Its length is given as about 60 kilometres and as about 63 kilometres, or 39 miles.

Sixteen forts are one count, and nineteen another, after the spacing was tightened. It was abandoned only eight years after completion, and the garrisons are also said to have gone back to Hadrian's Wall in 162, within two decades of the finish. Around 162 the Second Cohort of Nervians, infantry from a Belgic people of Gallia Belgica, was at Vindolanda and turned the timber fort into the first stone fort.

In the early third century, attacks in the north led Septimius Severus to invade Caledonia in person. He was emperor from 193 to 211, and he came to Britain in 208 with his sons. One account gives him an army of 50,000, and another an army of more than 40,000, judged from the size of camps. He strengthened Hadrian's Wall.

He is also said to have reoccupied and enhanced the Antonine Wall, while another account finds substantial work on Hadrian's Wall in 208 to 210 and no evidence of attention to the Antonine Wall in those years. A Roman history written long afterwards says he fought no pitched battle, that the army suffered in forest, marsh, and river, and that a full fifty thousand died. That figure is a claim about losses, not a third measure of the army he brought. By 210 the Caledonians had asked for peace.

They revolted again with the Maeatae. Severus prepared another campaign and fell ill. He died at York on 4 February 211. His sons Caracalla and Geta paid off the rebels and left for Rome.

Caracalla, born Lucius Septimius Bassianus on 4 April 188 and emperor, at first jointly, from 198 until his death on 8 April 217, had been made co-emperor on 28 January 198. Geta was made Augustus in 209. The brothers ended the northern war by a peace that put the border back on Hadrian's Wall. On 26 December 211, at a meeting arranged by their mother Julia Domna, Geta was killed by praetorians loyal to Caracalla.

Caracalla then had his brother's memory condemned.

At the fort, the Severan years produced a short-lived stone complex with as many as two hundred circular structures. They may have housed irregular troops, or British civilians working the quarries, or hostages taken on the campaign, or local people sheltering from the invasion. The record does not decide among those uses. A skull from the south ditch was probably displayed in front of the fort as a trophy.

Sometime in the 210s the complex was demolished. A second stone fort replaced it, and a civilian vicus was built to the west. Most of what a visitor sees belongs to this phase or later. The garrison was the Fourth Cavalry Cohort of Gauls.

A dedication to Caracalla shows they were here by 213. An inscription of 223 records that they rebuilt the west gate. Quintus Petronius Urbicus commanded them under Severus Alexander, who reigned from 222 to 235. Marcus Caecilius Celer, Pituanius Secundus, and Gaius Sulpicius Pudens are also named as commanders, without dates.

The last text for the cohort is a fragment of building work under Probus, who reigned from 276 to 282.

It had been assumed that by then the title "Gauls" was only a name and that the men were recruited locally. An inscription found in 2005 says otherwise. It reads, in the Latin, that citizens of Gaul, in agreement with the Britons, dedicated a statue to the goddess Gallia. Native Gauls were still in the regiment and still marked themselves off from the British-born soldiers.

The same years left a crowded religious record. Seventeen inscribed dedications are to Roman gods: seven to Jupiter, two to the genius of the headquarters, three to Mercury, and one each to the Fortune of the Roman People, Vulcan, Silvanus, Mars Victor, and Hercules. There are statues of Diana, Hercules, and Priapus. Twenty dedications belong to gods of northwestern Europe: twelve to the Veteres, the "old ones", five to mother goddesses, and one each to Cocidius, Mogons, and Maponus, with statues of the mothers and of Maponus.

One dedication is to the goddess of Hama in Syria. A detachment from that city served at Carvoran, to the west.

The visible stone fort, the second, is placed in the first quarter of the third century. Its platform measures 154 by 93 metres, about 1.43 hectares. The first stone fort was larger, about 1.61 hectares, and its north wall is estimated to have stood 16 metres beyond the later north wall. The later north wall was built over four demolished circular structures, which are consolidated and shown, and three more are known as buried remains from digs in 1934 to 1936 and 1979 to 1980.

The north, east, and west gates have been uncovered. The south gate shows as an earthwork. The headquarters that faces north is on show and covers an earlier stone headquarters that faced south. A latrine in the northeast corner is consolidated.

Back-to-back barracks in the northeast quarter, found in 1980, are not on display.

West of the fort, a road lined with masonry belongs to a military annexe of the first stone fort, at first mistaken for the civilian village. The masonry covers a timber annexe of about 180. The grandest building, with a bath suite around a courtyard, was once called an inn for official travellers. It is now thought it may have been the commanding officer's house in the late second and early third centuries.

A clay rampart enclosed the annexe. In 1991 it appeared that a length of the first stone fort's west wall had been taken down to join annexe and fort. In the third century, probably when the second stone fort was built, the annexe gave way to the civil settlement, with large stone blocks under timber walls. The military bath house stood in that settlement.

A coin hoard in the destruction of one of the buildings, and the absence of coins of Tetricus I, who reigned from 270 to 273, are read as evidence that the settlement was destroyed and abandoned before 270. Other accounts place the abandonment of the vicus around the end of the third century, or about 285, for reasons that are not known, and say the fort itself was largely abandoned then. The fort's pottery and coins continue. One reading takes occupation at least to the later fourth century, and the fort, unlike the other Stanegate forts, appears in a late Roman list of military installations.

A stone altar found in 1914, and kept in the museum, shows that the settlement was formally a vicus and that its Roman name was Vindolanda. Early modern visitors had called the ruins Chesters on Caudley, Little Chesters, the Bower, and Chesterholm, and one early source spelled the name Vindolana. The 1914 altar settled the spelling. It was set up by the civilians for the Divine House and for Vulcan.

South of the fort a large bath building served people from the whole site. The later stone fort and the village are also said to have remained in use until about 285.

Around the end of the third century the fort was briefly abandoned and then rebuilt. That rebuild may belong with the reforms of Diocletian, who reorganised the provinces, separated and enlarged the civil and military services, and placed administrative centres closer to the frontiers than Rome. The garrison grew smaller through the fourth century, and the vicus was not lived in again. One explanation is that life outside the walls was no longer safe.

In the second half of the fourth century the fort was rebuilt once more, perhaps after Count Theodosius defeated the Great Conspiracy in 368, and an apsed building, possibly a church, went up with that work. About 370 the defences were roughly repaired, perhaps by irregular soldiers. Another account places a rebuild about 300, without the vicus, and a rough repair about 370.

The Great Conspiracy was a year of war near the end of Roman rule in Britain. In the winter of 367 the garrison of Hadrian's Wall is said to have rebelled and let the Picts through, while Attacotti, Irish raiders, and Saxons landed elsewhere, and Franks and Saxons landed in northern Gaul. The count of the coastal region, Nectaridus, was killed, and the duke Fullofaudes was besieged or captured. The places of those defeats may have been in Britain or in Gaul.

Valentinian I was fighting the Alamanni and did not come himself. In the spring of 368 a relief force under Count Theodosius gathered at Boulogne. It included four units, among them the Batavi, and also his son, the later emperor Theodosius, and probably his nephew Magnus Maximus. Count Theodosius crossed to Richborough, found troops unpaid, unwilling, or gone, and then marched on London, broke up raiding bands, offered deserters an amnesty, and reoccupied abandoned forts.

By the end of the year the raiders had been driven back, mutineers had been executed, and Hadrian's Wall had been retaken. A new duke, Dulcitius, and a civilian head, Civilis, were appointed. The local intelligence agents who had taken bribes were removed. Count Theodosius also defeated Valentinus, a Pannonian exile who had joined the invaders.

He died in 376. He was the father of the emperor Theodosius.

The last Roman troops left Britain in 407, but people stayed at Vindolanda for an uncertain time. There is no evidence for a sudden end in 410, and occupation may have thinned slowly. A recent account of the site puts Roman occupation from the late first century until the early fifth. A small church was built in the courtyard of the old headquarters sometime after 400.

A Christian dedication of about 600, found at and near the site, may mean a monastic house. A tombstone of a man named Brigomaglos, also of about 600, was found nearby. His name suggests a Welsh origin. He might have been a priest, or a commander for one of the kings of the Old North.

Some Anglo-Saxon brooches have been found on the hill. In 2020 a chalice of the fifth century was found in a collapsed church. It carries crosses, angels, a smiling priestly figure with a crook, fish, a whale, ships, and the Greek letters chi and rho, with writing in Latin, Greek, and possibly Ogham.

The first post-Roman notice of the ruins is William Camden's, in 1586. Travellers over the next two centuries described stone that was later robbed. The military baths were still partly roofed when Christopher Hunter came in 1702. About 1715 an excise officer, John Warburton, removed an altar.

In 1814 the Reverend Anthony Hedley began the first real digging. He died in 1835 without writing it up. In 1929 Eric Birley, born on 12 January 1906 and dead on 20 October 1995, bought the Clayton estate at Chesterholm, the house that had belonged to Hedley. The purchase is also placed in the 1930s.

He could not afford Housesteads as well. He had dug on the Wall since 1927, first at Birdoswald under F. G. Simpson, and in 1929 two inscription stones there led him to redate the Wall periods, a scheme later work has followed.

In 1931 he became a lecturer at Durham. In 1934 he married Margaret "Peggy" Birley, born Goodlet, a student of his who dug at Housesteads and at Vindolanda. They published a report together. Their sons were Robin Birley, born on 19 January 1935 and dead on 29 August 2018, and Anthony Birley, born on 8 October 1937 and dead on 19 December 2020.

The family left Chesterholm for Durham in 1950. Robin Birley became Director of Excavations and head of the research committee. Anthony Birley, a historian of the second-century emperors, took part in the digs from boyhood and wrote on Roman Britain. His doctorate, finished in 1966, treated the Roman high command from the death of Hadrian to the death of Caracalla.

Robin's son Andrew Birley, born on 28 October 1974, is Director of Excavations. A 2025 account also calls him chief executive of the charity that cares for the site. He graduated from Leicester in 1996 and finished a doctorate in 2010 on the extramural settlement. His wife Barbara Birley, born on 15 February 1976, is the curator.

She works on the conservation, cataloguing, and display of the finds, has published on the jewellery and the combs, and was elected a fellow of the Society of Antiquaries on 17 June 2021. Digging still runs each summer, from April to September, with professional staff and volunteers. The volunteer places ask for people over 16 who can give two consecutive weeks. More than 400 volunteers have joined in a year, and the 2026 season is also described as drawing hundreds.

In 1970 a registered charity was founded to administer the site and the museum. Since then more than 5.3 million people have visited the fort and the army museum the charity also runs at Carvoran, which it acquired in 1972 and took in hand in 1997. As of 2009 the charity was the largest employer in Bardon Mill. The fort was scheduled on 12 December 1928, and the scheduling was amended on 14 July 1997.

It lies in the parishes of Henshaw and Bardon Mill, in the Northumberland National Park. In 1987 the fort and Carvoran joined the other Hadrian's Wall sites in a World Heritage property. In 2005 the German frontier was added, and in 2008 the Antonine Wall, so that the property became the Frontiers of the Roman Empire. The monument also takes in the civil settlement, the cemeteries, a length of the Stanegate, and two milestones.

One milestone, uninscribed, still stands at Codley Gate, 120 metres northeast of the fort, 1.7 metres high, and is taken to mark the 15th Roman mile west of Corbridge. The stump of a second, about 1,500 metres west of the fort, marked the 16th. It was intact when Horsley saw it in 1725 and carried the words "to him who was born for the good of the state". The top was later split for gateposts.

A cemetery north of the road held cremation pots and the tombstone of Ingenuus, who lived 24 years, 4 months, and 7 days. Another cemetery lay west of the settlement. Hugh Ridley dug up burial urns in his garden in the eighteenth century at Archy's Flat, about 400 metres west of the fort. A parade ground and a possible construction camp north of the road are known from the air and do not show on the ground.

The museum, also called Chesterholm Museum, stands in gardens with full-size reconstructions of a temple, a shop, a house, and a Northumbrian croft. In 1973 a full-size replica of two sections of Hadrian's Wall, one of turf and one of stone, was built on the site. The museum shows boots, shoes, armour, jewellery, coins, and infrared photographs of the tablets. It reopened in 2011.

In 2018 a gallery of the wooden underworld was added for about two thousand years of wooden finds, including a toilet seat and a toy sword. Excavation is visible from close to the trenches between April and September.

The organic finds are the reason the fort changed what can be known about ordinary life on the northern frontier. About 850 ink tablets are on record. They show writing by the prefect's household, by soldiers, and by families who sent parcels with notes of the contents. Other finds include more than 160 boxwood combs, most of a military tent, and the largest known collection of Roman shoes.

Spindle whorls from the northwestern quarter indicate spinners of both low and high status in the third and fourth centuries. A study of a stylus tablet found in 2014, and published in 2025, reads the surviving lower leaf as part of a deed of sale for an enslaved person, only the second such document from Britain. It lay in material used to level ground for the buildings of the second and third periods, about 9 metres west of the outermost ditch of the first fort, so the fill was dumped late in the first period, about 85 to 90, or possibly early in the second, about 90 to 95. The purchaser was probably Verecundus, who is otherwise known to have owned enslaved people.

The text shows a legal formula from Rome in use on this frontier.

Later seasons have kept producing objects with dates that do not always match the layer they were found in. In 2006 a bronze and silver brooch in the form of Mars, punched with the name Quintus Sollonius, was found. In 2010 the remains of a child of about eight to ten years were uncovered in a shallow pit in a barrack, in a position that suggests the arms may have been bound. Further analysis indicated that the child could have been a girl, and the death is placed about 1,800 years ago.

In 2014 a wooden toilet seat, one of the few known from the empire, was found, and so was the only gold coin from the site, minted in 64 or 65 and lying in a fourth-century layer. In 2017 a cavalry barrack dated around 105 produced swords, tablets, textiles, and arrowheads, and a separate trench in a first-century layer produced 25 ink tablets, the second-largest cache yet found, after a cache from the same fort in 1992. In 2017 two leather boxing gloves, dated to 120, were also found. They are not a pair.

The larger was cut from one piece of leather, folded into a pouch, and packed with plant material as a shock absorber. They are read as sports gloves rather than the heavier gloves of the arena, and they are on show in the museum. They have also been described as nearly 2,000 years old and as 1,900 years old. In 2021 a carved sandstone of a naked warrior or god in front of a horse, or a similar animal, was found a few inches under a floor.

Early readings point to Mars or to Mercury. In February 2023 a wooden phallus 160 millimetres long, and nearly 2,000 years old, was published. One account calls it a toy. The study that published it left the purpose open among luck, protection, power, and pleasure.

On 16 June 2026, during that year's season, Andrew Birley was digging a fourth-century barrack when a rounded flagstone caught his eye. Turned over, it was a sandstone relief of a Genius, the protective spirit of a household or a place, 44 centimetres high, 23 wide, and 8 deep. The figure holds a cornucopia and a patera, the shallow offering dish. It is thought to have come from a domestic shrine and to have been reused as building stone.

The carving is believed to be local, either by a mason at the fort or in a tradition shared with the fort at Lanchester in County Durham. Inscriptions to a Genius are common in Roman Britain. Carved reliefs are comparatively rare. The stone had lain under the barrack floor for more than 1,600 years, a span also given as more than sixteen centuries.

It was being conserved before display in the museum on the site.

The dogs of the fort have been studied in their own right, in a series running from 2016 to 2026, separating domestic from wild canids and describing sighthounds, guard dogs, terriers, and dwarf dogs. Together with the shoes, the combs, the letters, and the shrine stone, they belong to the same buried town: a place where auxiliary soldiers, their families, enslaved people, and, later, a civilian village and a shrinking late garrison lived on one hill, rebuilding it each time the frontier moved.