By Crasterfarian
Have you ever been to Bowness on Solway?
It’s a lonely bleak road west from Carlisle to the end of Hadrian’s Wall and its other Wallsend, Maia fort at Bowness.
As you head along the single-track road there are wide open marshlands and hints of a wild untamed frontier edge as you look north across the Solway Firth towards the coastline of southern Dumfriesshire.
How did the Romans manage this area and why does there appear to be a three-mile gap in the Wall and its associated features between Milecastle 73 at Burgh by Sands and Milecastle 76 at Drumburgh?
The Missing Piece
The accepted theory is neat, tidy and Roman. Too neat perhaps…
For generations antiquarians and archaeologists attempted to reconstruct the missing western sector of Hadrian’s Wall through mathematical spacing.
Since milecastles elsewhere on the frontier generally occur at regular intervals of one Roman mile apart, roughly 1480 metres in modern measurements, it was naturally assumed that the same system continued westward across Burgh Marsh.
But did it?
Were the ghostly positions between Milecastles 73 and 76 born from fact or assumption? What if the assumptions themselves are wrong?
As we move west from the confirmed site of Milecastle 73, the frontier system begins to dissolve into uncertainty. The Wall itself disappears into the marsh. The Vallum appears to drift south onto firmer ground. Turrets vanish completely. The Military Way becomes conjecture and even the supposed site of Milecastle 76 remains uncertain and little more than a faint platform east of Concavata (Drumburgh).
Perhaps we should ask a different question entirely: What if there never were 80 milecastles?
The Python Theory
“The other emperors said I was daft to build a wall in a swamp, but I built it all the same, just to show them. It sank into the swamp.”
The western marsh sector beyond Milecastle 73 is unlike anywhere else on the frontier. Here the Romans encountered unstable tidal terrain where the landscape constantly shifted beneath water, peat and salt marsh.
It may simply have been impractical or unnecessary to maintain the rigid mathematical spacing system used elsewhere.
The original western Wall was built not in stone but turf, possibly with timber strengthening or palisading above a stone foundation. This suggests speed of construction and adaptation were already taking precedence over the more robust engineering seen further east.
Timber structures and turf ramparts may have crossed the marsh in a lighter and more flexible form than the mighty stone curtain wall beyond the River Irthing.
Or perhaps they did not cross the marsh fully at all.
The Romans may initially have attempted to continue the frontier westward only to discover that the constantly shifting wet ground could not reliably support large scale stone construction.
Maybe they simply stopped at Milecastle 73: the edge of firm ground and last solid footing before the marsh swallowed the frontier. Rather than waste enormous resources battling the landscape, they may instead have adapted their frontier system to the terrain.
Rome was practical above all else.
The marsh already formed a formidable natural obstacle. Thick mud, hidden tidal channels and unstable ground would have funnelled movement onto narrow crossing points of firmer land. In effect, nature had already created its own defensive barrier. Perhaps the Romans simply accepted this.
Rather than forcing a continuous monumental wall across the marsh, they may have left a partially open frontier zone controlled instead by forts at each end.
During Roman times there were two ancient fordable crossing routes from the north. These crossing points, now long gone beneath marsh and tide, dictated the positioning of the forts in this area: Aballava to the east. Concavata to the west. A military full stop guarding the fords.
Economics and Trade
Perhaps the marsh was too valuable to bury beneath endless masonry and barriers of turf and ditch.
The Solway coast is known to have supported salt production for centuries and it is entirely possible that primitive Roman salt pans existed within these tidal flats.
Salt was enormously important to Rome for preserving food, feeding livestock and supplying the army. Entire taxation systems existed around its production and movement. Let us also not forget that part of a Roman soldier’s pay was a ration of salt, or sal. It is where the word salary originates.
If salt extraction was already taking place here then the marsh was not empty wasteland at the edge of empire. It was productive frontier land.
This may explain why the Vallum also seems to disappear. Perhaps access was needed to the fertile and productive salt flats, pans and fish traps. The Romans may have recognised that attempting to force every frontier component through unstable wetlands was simply impractical.
Such a theory would explain several long standing anomalies.
Firstly, the strange absence of entire frontier structures. Between Milecastle 73 and the presumed position of Milecastle 76 not merely the wall but two milecastles, six turrets, the ditch and military way all vanish almost completely from the archaeological record.
Secondly, the accepted numbering becomes questionable. The supposed location of Milecastle 76 near Concavata (Drumburgh) is based largely upon projected spacing rather than definitive evidence. Yet if the marsh frontier originally contained fewer installations than modern antiquarians expected, then perhaps the structure identified as Milecastle 76 was actually Milecastle 74 all along.
In such a scenario Hadrian’s Wall may never have possessed 80 milecastles at all but perhaps only 78, or even fewer if Maia (Bowness on Solway) was incorporated into the frontier system from the outset and did not replace a traditional milecastle.
The antiquarians of the eighteenth and nineteenth centuries may simply have imposed ideal Roman mathematical order upon what was in reality a far more adaptive frontier landscape.
The Rudge Cup Clue
This famous Roman vessel adds further intrigue.
The Rudge Cup lists the western forts as MAIS (Bowness), ABALLAVA (Burgh by Sands), UXELODUNUM (Stanwix), CAMBOGLANNA (Drumburgh) and BANNA (Birdoswald).
The cup’s exact purpose has long been debated but most modern interpretations lean towards it being a commemorative gift or souvenir of some kind, perhaps even a retirement or demobilisation gift for a soldier or centurion stationed upon the Wall.
The vessel is generally thought to date from the later second century AD, perhaps around 160 onwards, meaning it may preserve the names of forts known during an earlier phase of the frontier before later modifications and additions in the western marsh sector.
Yet Concavata is absent entirely despite lying between Aballava and Maia.
However, the later Amiens Skillet does include Concavata amongst the western frontier names.
Why?
Perhaps because Concavata did not yet exist when the Rudge Cup was produced.
If you look at the remains of the known fort, it is the smallest fort upon the Wall system and appears little more than a large milecastle.
An afterthought perhaps.
Or was it added later to bolster a weak point and provide permanently garrisoned troops at each end of an open frontier gap?
It is entirely possible that the original turf and timber frontier was completed first with the marsh sector relying largely upon the natural barrier of the Solway itself.
Or perhaps, as Monty Python so eloquently put it, “it sank into the swamp.”
Only later, during the rebuilding of the frontier in stone and the increasing militarisation of the western sector, with the famed Ala Petriana stationed at Uxelodunum and Septimius Severus himself active in the north, was Concavata added to strengthen control of the crossings, or perhaps even to impress the visiting emperor.
If so then the western frontier was evolving even while Rome occupied it.
Perhaps this strange unstable marshland frontier even inspired the later Roman coastal defences beyond Maia itself.
The milefortlets of the Cumbrian coast may represent lessons first learned here at Burgh Marsh.
Rather than constructing one continuous monumental barrier across unstable coastal terrain, the Romans instead developed a looser system of linked forts, fortlets, towers and patrol routes adapted to tides, coastlines and shifting ground.
In that sense, Burgh Marsh may represent the moment Rome realised that nature itself could become part of the frontier.
Not every boundary needed a wall. Sometimes clarts, watta, pila, gladii and a rampaging Ala cohort were all that was needed.
With love from the Crasterfarian xx
Never miss an article. Enter your email or give our FB page a like
