The Saxon Shore: Evolution of a Coastal Defence System

 In AD 43 the Roman Emperor Claudius’s invasion force of 40,000 men landed in Southern Britain. A century before, in 55 and 54 BC, Julius Caesar had invaded Britain with the aim of conquest but on the first occasion suffered losses to his fleet due to the unpredictable Channel weather. He returned a year later with a much larger invasion force of some 800 ships but was drawn away to troubles in Gaul. Hence, Claudius’s arrival is often termed the third invasion of Britain.

Debate continues as to where Claudius’s invasion force landed, some argue for the south coast, perhaps Chichester in Sussex, yet archaeological evidence points to Richborough in Kent. After landing, the army established a fortified bridgehead defended by a single gateway within a double ditch and rampart, running almost parallel to the shoreline for 600 metres, sections of these can still be seen today. The Claudian ditches were crossed by a causeway which would adjoin the main Roman road, the later Watling Street.

Wantsum Channel (English Heritage)

In those days, nearly two thousand years ago, Richborough stood on a small island just offshore, separated from the mainland by a narrow tidal channel named by the Romans as Portus Rutupis. The Wantsum Channel, as the Saxons called it, was up to 2 miles wide, 10 miles long and separated the Isle of Thanet from mainland Britain. 

Richborough island was situated at the southern end of this natural waterway, close to the confluence of the Wantsum and the open sea. At this point extensive shingle banks sheltered the Wantsum from the sea effectively forming a breakwater enclosing a sheltered anchorage beside Richborough island, making it an ideal harbour where the Roman port would develop. The Romans constructed a causeway to the island which would lead onto the Watling Street, through the Midlands to Chester.1

Richborough (Rutupiae) was therefore the site of the earliest Roman settlement in Britain, however a series of archaeological investigations by Bushe-Fox (1922-38) has revealed that the initial military earthworks were not extensive with no evidence of a heavy military presence, more a temporary defence for the disembarkation of troops during the initial phase of the invasion. No doubt the large invasion force quickly spread out to secure the Wantsum Channel controlling the northern end where it met the Thames estuary and the Isle of Thanet. 

Archaeological research at Regulbium (Reculver) at the northern mouth of the Wantsum has revealed pottery and pairs of ditches similar to those at Richborough, suggesting a small garrison relating to the invasion period.

This coastline has changed considerably since Roman times, the Wantsum Channel has silted up over the years and now reduced to a small stream with Thanet now joined to the mainland. Consequentially the two Roman sites associated with this waterway guarding both ends of the Wantsum, Reculver and Richborough, have changed over the centuries. Richborough is no longer an island and now lies 2 miles (3.2km) from the sea with the River Stour, all that remains of the Channel, having devoured the sandy cliff which forms the eastern edge of the monument and destroying the south western and north eastern ends of the defences.

At the time of the Roman invasion Reculver occupied the southern tip of a promontory at the north-western end of the Wantsum Channel with the sea almost a mile (1.4km) to the north. Coastal erosion has severely damaged the west and north east of the site so that only the southern half of the later stone fort survives as ruined walls, earthworks and below ground features, the rest now lost to the sea.

The Gateway to Roman Britain
Less than ten years after the invasion of AD 43 the site at Richborough was levelled to make way for the construction of a military and naval supply base.

Around AD 90 the Romans built a huge monumental arch at the centre of this base overlooking the shore at Richborough, one of the largest such monuments in the Roman empire. This monumental arch would have been seen by ships from miles away, a navigational aid to ships, and on landing the travellers crossed through the Richborough arch as the official gateway to Britannia.

The major Roman road, Watling Street, now started at the foot of the arch and proceeded through the West Gate across the causeway and continued for about 250 miles through the Province. 

This huge monumental arch at Richborough was of a particularly rare and elaborate type with four faces and four openings, an architectural style known as a ‘quadrifons’. Beneath the archways was a raised cross-shaped pavement accessed by steps on each side.  The arch was underpinned by stone foundations around 10 metres deep, estimated as necessary to support a structure some 25 metres high consisting of around 20,000 tonnes of material. The Richborough quadrifons was clad in Carrara marble, an exotic bright white stone quarried in Italy which could only be used by authority of the emperor. 

Yet the decision to build a quadrifons arch at Richborough is somewhat mysterious. Because of their four openings, quadrifonic arches typically had a crossroads at their centre, and were often positioned at important intersections. No doubt the east-west line which joined Watling Street was the road that linked the Empire to Britannia via the sea, yet the north-south archways had no apparent alignment, it may have symbolised the division between land and sea.

The Arch of Janus

The only quadrifrons or four-faced, triumphal arch surviving in Rome is The Arch of Janus. This unique piece of architecture,16 metres (52ft) high and 12 metres (40ft) wide, marked a crucial crossroads in the ancient city. Estimated to have been constructed in the early 4th century AD, using  material from redundant earlier structures that were comprehensively demolished so that the materials could be re-used in the construction of new buildings. The ancient and widespread practice of ‘Spoliation’ (from the Latin for 'spoils') was common in Roman architecture.

The town at Richborough developed around this massive arch, with new roads laid out and stone buildings constructed including shops and metal workshops around the monument. As the Roman army spread out into the Province the town and port boomed with trade from Gaul. A mansio complex and an amphitheatre had been erected before the town reached its peak around AD 125. However, by the early 3rd century the town was in decline, probably due to the growth of the cross-channel port at Dover (Dubris).

Classis Britannica
Being an island the Roman navy played a significant part in the invasion of Britain in AD 43. In 55 BC Julius Caesar’s invasion plans suffered owing to losses to the fleet caused by bad weather in the Channel. The next year he was back with a huge fleet that would not only transport the invasion force but also provide logistics support. Claudius had no doubt learnt from Caesar’s accounts of his trips to Britain a century earlier and came well prepared with the navy providing not only a means of transport but also essential support to the shore defences.

A massive fleet, the classis Britannica, was built for the Claudian invasion and was operating in British waters since at least AD 43, if not before. Initially the fleet’s headquarters was most likely at Boulogne on the north coast of Gaul.2

Accordingly, the fleet’s earliest British base was at Richborough, the beachhead of the Claudian invasion. Around AD 130-140 a new fortified port was built at Dover which became the main base for the fleet in Britain. This was thought to be the Novus Portus (New Port) as recorded in Ptolemy's 2nd century ‘Geography’. The presence of the fleet at Dover is attested by a large number of tiles stamped ‘CL BR’, found at the site and also across the Channel at Boulogne which continued as the main operating base of the classis Britannica. Tiles found at other sites indicates their association with the fleet but none have been found, to date, north of London.

The Hadrianic Period (c.122-136 AD)
The 1st and 2nd centuries witnessed the wider development of coastal installations around Britain.

Legionary fortresses were constructed at Caerleon and Chester, with smaller forts at Cardiff and Lancaster, all with access to the sea on the west of the country. Also on the western shore a series of forts, supply bases and signal stations were constructed on the Cumbrian coast guarding the Solway Firth, effectively forming a southern extension to Hadrian's Wall.

On the otherside of the country at the eastern terminus of the Wall a fort was constructed at South Shields on the River Tyne. This would have certainly operated as a supply base for the classis Britannica during northern campaigns of the Roman army in the 1st and 2nd centuries, yet no CL BR tiles have been found at the site.

Further south down the eastern coast it has been argued that a military supply base was established at Brough-on-Humber during the Hadrianic period. This base was on the northern shore of the Humber, the terminal of the ferry from Lincoln in the south and serving the legionary fort at York. Around AD 200 a new earth and timber rampart was constructed when the earlier defensive circuit was extended. But again, although the classis Britannica certainly sailed into the Humber no fleet tiles have been found at the site.

At Lympne in southern Kent over 30 ‘CL BR’ tiles have been found suggesting the existence of a port or supply depot during the 2nd century. Although evidence for this early port has not been found, the amount of re-used building materials in the 3rd-century Roman Shore Fort at Lympne (Portus Lemanis) suggest previous activity at the site or nearby.

At Reculver on the north Kent coast tiles have been found made from the same clay as that used by the classis Britannica re-used in the masonry of the later St Mary's church. Other tiles were also used in the construction of the east gate of the later Shore fort but it is not possible to determine if these had been stamped 'CL BR' as only the sides of the tiles remain exposed, the top and bottom surfaces mortared into masonry layers and cannot be examined.

Classis Britannica tile

CL BR tiles have been found at Richborough and Pevensey where there is little evidence for 2nd century military installations. Fleet tiles found at Folkstone villa suggest it may have been the residence of the prefect commanding the classis Britannica.

From the initial invasion in AD 43 we see evidence of the spread of Roman military operations by land across the Province, from the mouth of the Tyne on the north-east coast to the south-east coast in Kent and the north-west coast in Cumbria to south Wales. These coastal defences must have been served by the British fleet of the classis Britannica, providing essential logistics support, the bases the fleet operated from seemingly identified by the presence of tiles stamped ‘CL BR’.3

The Severan Period (193-235 AD)
Significant reorganisation of the south and east coast military forts was carried out towards the end of the 2nd century and the early 3rd century.

The Caledonian campaign of Septimius Severus ended in 211 when the emperor died at York with the Roman forces pulling back the northern frontier to Hadrian's Wall. Following the death of Severus his son Caracalla seems to have concentrated on reorganising defences along the east coast. The fort at South Shields was substantially rebuilt with the base continuing to function as the northern limit of maritime operations along the east coast until the 4th century. 

Further south along the east coast military installations were remodelled and new forts built during this period. This includes three forts south of the Wash at Brancaster and Caister in Norfolk and Reculver on the north coast of Kent which would all later become part of the Saxon Shore.

Brancaster has been dated to the early 3rd century based purely on the early architecture of the site but coin finds indicate activity in the late 3rd century. The installations at Brancaster, Caister and Reculver were no doubt part of the defence system established on the east coast, such as Brough-on-Humber and the site at Skegness, now lost to coastal erosion, intended to operate with Brancaster on the opposite side of the Wash.

The South Coast
Whereas the east coast of Britain witnessed renewed military activity during the Severan period there is a distinct lack of evidence for new installations on the south coast, which appears to be in decline rather than renewed growth.4

The Severan campaigns in northern Britain relied heavily on logistical support from the classis Britannica. The concentration of naval operations in the north seems to be associated with the abandonment of the fort at Dover which was demolished between 200-210.5 

15 miles further along the south coast occupation at the port at Lympne continued during the early 3rd century; however, it is uncertain whether this installation was civilian or military. 'Lemanis' is included in the Antonine Itineray, a register of the stations and distances between them on the Roman road network. The Itinerary is usually ascribed to the Roman emperor Antoninus Pius (reigned AD 138 – 161), however it seems unlikely that the British section was compiled at this time. It has been suggested that the Itinerary was compiled over two centuries with the British section titled “Iter Britanniarum” being plural, indicating that the British section was assembled after Britain was divided into two provinces in AD 211 AD, either by Septimius Severus or his son Caracalla.6 But as with Dover, the port as Lympne seems to have fallen out of use in the first half of the 3rd century.

The Later 3rd Century
When Alexander Severus, the last emperor from the Severan dynasty, was assassinated by his own troops in AD 235 it marked the beginning of the Crisis of the Third Century.

The next 50 years (AD 235–284) would see over 20 emperors rise compared with the 26 emperors who reigned from the first Roman emperor Gaius Julius Caesar Augustus in 27 BC to the death of Alexander Severus in AD 235, a period of over 250 years. During this period of social turmoil and chaos the empire experienced economic disintegration, repeated foreign invasions, and civil wars. Roman commanders in the field became increasingly independent of Rome's central authority resulting in the Roman Empire splintering into three political entities: the Gallic Empire, the Roman Empire, and the Palmyrene Empire.7

The Roman Shore Forts in the late 3rd Century AD

Following the anarchy of the 3rd century a new emperor was declared by his troops in AD 284 while on campaign in Persia who finally bring order back to the Empire. Diocletian's reign stabilized the empire and ended the Imperial Crisis. Around this time a series of substantial fortifications appeared on the coast of Britain from Brancaster on the Wash to Portchester on the Solent.

During the later 3rd century activity in Britain focused on the South and East Anglian coasts with relatively little new military building beyond the geographical area between the Wash and the Solent. Existing fortifications elsewhere must have been considered adequate or irrelevant to the current threat.

Brough-on-Humber was one of exceptions; bastions were added to the perimeter wall, a characteristic in Britain unique to the Shore Forts and the refortified London Wall. A large fort was constructed at Cardiff in South Wales, all but square in plan with projecting bastions on the perimeter walls, being very similar in design to Portchester.

The construction of the new fort at Cardiff has been dated to after 260, however it is suggested that the similarities with Lympne and Portchester tend to indicate a later 3rd century origin. It is reasonable to see Cardiff as contemporary with, rather than earlier than, the development of defences on the south coast of England.8

At Brancaster a coin of Tetricus, emperor of the Gallic Empire from AD 271 to 274, found in the rampart indicates this was not built before c.270. Yet, coins from the Carausian period, usurper AD 286-293 self-proclaimed “Emperor of the North”, make up the bulk of the finds at Brancaster,  indicating activity at the site was greatest during the late 3rd century.

These new coastal forts in Britain indicate a continued naval presence in British waters, although the classis Britannica had disappeared from the archaeological record by the mid-2nd century.9

We have seen how new military installations were built and the defences of existing fortifications augmented during the Severan period. During the 3rd century defensive emphasis moved away from the south coast, with defences reorganised along the eastern coast. Events at Richborough provide a snapshot of developments.

Return to Richborough
The huge monumental four-fronted arch constructed around AD 90 stood at the heart of the town of Richorough for almost 200 years. At 25m high and clad in white marble the arch was visible for miles. For whatever reason the Roman army moved back to Richborough in the middle of the 3rd century. Around AD 250 part of the town around the arch was demolished and replaced with a small fortlet. The arch appears to have been repurposed as a watchtower or signal station, protected by three ditches and a rampart complete with palisade.

Richborough Roman fort

During the last decades of the 3rd century the fortlet was levelled and a large stone fort was built at Richborough. Further parts of the town were demolished to provide material for the new fort. Even the arch, a ceremonial and symbolic gateway to the province of Britannia awarded the rare privilege of being faced with Carrara marble from the Imperial quarries was torn down and the triple ditches filled in, all to provide building material for the new fort. Much of the marble facing stones were burned to make lime to be reused in the mortar of the fort’s walls. Other marble slabs from the arch were used as packing in the walls. The substantial new stone fort at Richborough seems to have been built very quickly, by a man in a hurry. 

Notes & References
1. John Peddie, Conquest: The Roman Invasion of Britain, Sutton, 1997.
2. Andrew Pearson, The Roman Shore Forts, History Press, 2010, p.49. Chp 3, The Development of a Coastal System, pp.47-66.
3. Pearson, p.49
4. Pearson, p.55
5. Pearson, p.55
6. Rivet & Smith, Place Names of Roman Britain, Batsford, 1979, p. 154
7. Pearson, p.56
8. Pearson, p.63
9. Andrew Pearson, The Roman Shore Forts, History Press, 2010, Chp 3, The Development of a Coastal System, pp.47-66.

* * *