The Roman Invasion of AD 43:
The Sussex Case

David Bird discusses the historical sources and archaeological evidence for the Roman invasion of Britain by Claudius in AD 43 and proposes a case for a Sussex landing

John Hind’s 1989 Britannia article was a timely reminder that it is actually no more than a hypothesis that the Roman invasion of Britain was carried out by a landing in Kent. He makes a good case for a landing in the harbours behind the Isle of Wight instead. If we then consider what would have happened after such a landing, it can be seen that the western route actually makes more sense. There is only sufficient space to develop the arguments briefly here (some of them are treated more fully in my paper for the Oxford Journal of Archaeology 19.1, 2000), and it will be necessary to take somewhat for granted a knowledge of the standard theory of the invasion and the available evidence.

Sources and Evidence for the Invasion of AD 43
Historical sources are limited to a short narrative by Cassius Dio with a few passing references by other authors; archaeological evidence is not much help because of the need for very precise dating. Using standard methods we cannot hope to be closer than five or ten years in our dating, unless we have the bonus of finding waterlogged wood able to provide secure dating by dendrochronology, as at Alchester.

These are the key points from Dio (quotations are from Hind’s translation): Aulus Plautius leads an expedition to Britain. Someone called Berikos (we assume this is Verica of the Atrebates) has been expelled and has persuaded Claudius to send a force. The troops are divided into three squadrons ‘to avoid having an opposed landing, which might hold up a single force’. They take heart from a shooting star travelling east-west in the direction they wish to travel. They land without opposition, as the Britons have not gathered their forces. Plautius has difficulty searching out the British forces, but when he does so he defeats ‘first Caratacus and then Togodumnus’. After their flight he secures the alliance of part of the Bodunni (we assume that this is the Dobunni), leaves a garrison and moves on. He comes to a river which the Britons think the Romans cannot cross without a bridge, but the Romans get across and win a battle. The Britons retreat to the Thames ‘where it empties into the Ocean and at flood-tide forms a lake’. They know how to cross but the Romans get into difficulties. However, some get across and others cross ‘some way upstream by a bridge’ and the Romans win again. Togodumnus is said to have died about this time. Plautius now stops and waits for Claudius, who turns up and ‘joins the troops who were awaiting him by the Thames’. He crosses, defeats the enemy and takes Camulodunum (Colchester), which is described as having been ‘the capital of Cunobelinus’.

The Roman Army
One useful starting point in considering the events of AD 43 is to remember the normal behaviour of a Roman army. There is a tendency to talk of this army as some sort of machine, operating in the same way as a modern force, but this is misleading. For a start, a considerable part of the force, including the cavalry, would be allies, sometimes very recently recruited as a tribal group. A better analogy might be with 18th-century armies, but of course without firearms — a major difference. The usual Roman plan of action was to seek out the enemy’s forces and bring them to battle, which was done by striking for the enemy’s main stronghold and/or food supply. One battle was usually sufficient to bring the war — or at least that year’s phase of the
war — to an end. There were no ‘front lines’ in campaigns; the well-known system of road grids and forts was for areas accepted as longer-term frontiers. Caesar even set out for Britain, in 55 BC, from a port in the territory of a tribe he had not yet conquered. In the west, the war was between the Roman army and a tribal group and it would usually end in that group surrendering and handing over hostages and tribute. Some tribes would ally themselves with the Romans from the start — British tribes sent submissions to Caesar even before he set out in 55 BC. Divisions between tribes and within tribes were all exploited.

Supplies and Forts
Much is made of supplies in discussions of the campaign, but we know very little about how Roman armies were supplied. Caesar relied on local supplies, delivered by his allies or taken as tribute from conquered tribes. He actually gave as a reason for abandoning a campaign in Germany his belief that the Suebi did not grow crops and so he would not have sufficient food. There is also a tendency among modern writers to expect standard ‘Agricolan’ type forts. But some conquest-period forts in Germany are far from the standard rectangles. In Gallia Belgica the area conquered by Caesar was held for many years afterwards by a mixture of allied chieftains and carefully placed garrisons, sometimes within the native strongholds. It is quite clear that we should not expect standard forts as part of the initial campaign.

Sea Crossings
Some modern writers make much of the need to use a short sea crossing, but other Roman campaigns show considerable use of the sea: examples include Caesar against the Veneti; Augustus using a fleet along the northern Spanish coast; Germanicus sailing round to northern Germany. The Seine–Solent crossing was one of four crossings in regular use, according to Strabo. It is entirely acceptable to think in terms of the Romans aiming for three good known harbours in an area likely to provide a friendly base with a good food supply, all of which implies the Solent (the arguments are well developed by Hind).

Landings in the harbours behind the Isle of Wight also better explain what Suetonius tells us about Vespasian. He is said to have conquered the island itself and defeated two tribes and captured 20 ‘oppida’. Although this is usually held to have taken place after the capture of Camulodunum, Suetonius implies that Vespasian took part in Claudius’ triumph, to do which he would have needed to leave Britain with the emperor. There is good evidence for early Roman military activity both at Fishbourne and Chichester, in the centre of an important corn-producing area, among allies (the area had favoured status for many years after AD 43) and well-placed for campaigns throughout southern England. It might be noted that even if Camulodunum was seen as the ultimate objective, the distance from Chichester to Colchester is not much further than that from Richborough.

Richborough is an unlikely spot for the initial landings: it was not then a known port and appears to have been on an island of dry land divided from the mainland by marsh. A causeway was needed for the road exit. Such a landing place would be potentially dangerous and was a very unhelpful place to start any land campaign. The well-known defences have a semi-permanent gate, which indicates that it was stronger than a temporary bridgehead. Also, the defences preceded the granaries of the so-called supply base, which cannot therefore be contemporary.

Dio’s Account
The story of the river battle usually supposed to be at the Medway relies entirely on Dio and it is therefore unfair to ignore what he says. We are specifically told that Togodumnus and Caratacus were not ready, and had to gather their forces after the Roman army arrived. It will have taken them several days to reach Kent. If the Romans landed at Richborough then Plautius’ infantry should have been over the Medway before the opposing forces arrived — they had no reason to wait around. Guerrilla actions will not have delayed them any more than they did Caesar, who easily reached the Verulamium area with a smaller force. He was 12 miles inland by daybreak after landing in 54 BC. Even if Plautius was uncharacteristically slow, we are told that he had put both enemy leaders to flight in actions before the river battle — after doing this, how could he have failed to cross the Medway before they had time to regather their forces?

A landing in the Chichester area makes better sense of Dio’s story. He tells us that Caratacus was beaten first before the same happened to Togodumnus. Caratacus seems to have been based at Silchester, while Togodumnus probably had to come from further east, so separate defeats would make sense. A move north from Chichester to Silchester would be natural, following well-established routes, and would avoid any problem of crossing the Weald (which would certainly have been bypassed). This scenario also better explains the absence of Caratacus from the rest of the campaign, for there is no mention of him as leading after Togodumnus dies. Tacitus tells us Caratacus was famous even at Rome, and after his capture in later campaigns gives him a speech there, directed to Claudius in person. There is no mention of Camulodunum, yet if Caratacus had been there, he would have been directly opposed to Claudius himself, giving Tacitus a fine ‘angle’ for the speech.

If Plautius based himself in the Silchester area he would be very well placed in a position to control the whole of southern England (it was later served by as many Roman roads as London). The surrender of the Dobunni and the placing of a garrison with them, probably somewhere near Cirencester, makes good sense in this context. It is also likely that a deal was struck with those who controlled the Verulamium area, as something of the sort is needed to explain their subsequently favoured treatment. There is good evidence for an early base at Silchester and it may be that Plautius actually used it himself in AD 43. Alchester, where Eberhard Sauer has obtained a date for a fort at least as early as AD 44, is due north from Silchester and fits well in this context.

The River Battle
Dio’s story suggests that the river battle took place not long before the struggle at the crossing of the Thames near ‘the tidal lake’, which presumably was near the future site of London. The action cannot have been much further downstream because the bridge referred to should not be far away (the Putney area would be a reasonable location). The narrative reads as though the Roman army was in pursuit, which is why the soldiers got into difficulties. On this reading of events therefore the river battle must have been somewhere on a route between Silchester and London, and quite near the latter. A number of rivers are candidates, such as the Wey, Mole or Wandle. (It is not necessary to find a wide, swift-flowing river; Dio merely tells us that the units we assume to be Batavians were able to cross such rivers. The only requirements for this river are that the Britons should have expected the Romans to need a bridge, and that chariot warfare should be possible on the far bank.)

There is no evidence that Plautius’ army remained near the tidal lake to await Claudius or that they established a base there. The crossing of the Thames under Claudius when he turned up to join the army is more likely to have been further upstream, perhaps even at Staines by a newly created bridge. (The site has a Latin name, Pontibus, which might be contrasted with the British name for Rochester, Durobrivae: ‘fort at the bridge’. This last name suggests a pre-Roman bridge with a controlled crossing, as might reasonably be expected in the late Iron Age — there is no need to assume that this is a reference to an early Roman fort.)

Plautius and his army were very experienced. It was a very large army, much larger than William the Conqueror’s forces, whose initial campaign after a single victory we accept quite readily. The size of the Roman invasion army proved big enough in due course to control almost all the British tribes. In AD 43 it faced tribes divided from one another and divided internally by old enmities; some of the tribes will have regarded themselves as Roman allies. We must rid our minds of front lines and the gradual conquest of territory. A far better model is Caesar’s actions against the Gauls, often fought by actions hundreds of miles apart in the same year. The Romans conquered tribes, or accepted their voluntary ‘alliance’; they did not conquer territory as such. The tribes indeed usually then continued to have an existence, forming the administrative unit within the north-western provincial system. The Roman ‘invasion’ was probably more like a procession punctuated by occasional outbreaks of violence, and on present evidence that procession is more likely to have started near Chichester.

Professors Frere and Fulford have restated the case a Kent landing (S. Frere and M. Fulford, ‘The Roman invasion of AD 43’, Britannia 32, 2001). Their version can be challenged at key points and I have attempted to do this in a note (D. Bird, ‘The events of AD 43: further reflections’ Britannia 33, 2002).