I’ve just spent a very pleasant week in East Kent, and evidently didn’t manage to switch off entirely during our holiday. Alone of the family I trudged around Richborough Castle, readily imagining the daunting quadrifrons arch topped with a triumphal statue that welcomed visitors to the province of Britannia and marked the start of Watling Street; and its demolition a couple of centuries later when the current structure, a fortress against Saxon raiding parties, replaced the previously bustling town in the troubled Third Century.

Richborough, in antiquity Rutupiae and variants, could stand for Britain as a whole (Lucan 6.67), and was famous in its own right for oysters (Juvenal 4.141), as Whitstable just along the Kent coast is today. It was probably where part of the invading army in AD 43 originally came ashore, an event that would have fixed its status as the official gateway to Britannia.

But there had been earlier Roman invasions of Britain, of course, those undertaken by Julius Caesar in 55 and 54 BC, and since we happened to be staying close to the front at Deal and Walmer, historically leading candidates for Caesar’s landing spots, I found myself also pondering where they had actually been. In particular, I found THIS, an article in Current Archaeology from a couple of years ago that got some attention in the newspapers at the time. Its essential claim is that interesting archaeological discoveries at Ebbsfleet, some way north of Deal, point to that location as Caesar’s landing place in 54 (it expresses no opinion about 55). This sent me back to Caesar’s account of his expeditions in Books 4 and 5 of his Commentarii de Bello Gallico, and I found myself unconvinced. I’ll set out here what I found unpersuasive about the Ebbsfleet theory, a lot of it relating to Caesar’s own account of things, and maybe also to the importance of studying texts as texts.

Let’s start with that account. Caesar landed in Cantium, Kent, twice, in August 55 and a year later in July 54; his accounts of both invasions are at BG 4.20-36 and 5.5-23. The first invasion, little more than a reconnaissance mission, involved fierce conflict at the point of landing without any significant penetration beyond the shore, while the second saw the Roman forces marching as far as Cassivellaunus’ capital beyond the Thames somewhere in Hertfordshire.

The descriptions of the actual landings are what are at issue here, though, and they are as follows:

In 55 Caesar’s warships and transport craft cross without difficulty from the Pas de Calais toward the White Cliffs, but Caesar realises that a landing there would leave them badly exposed to attack (British warriors were seen massed on the heights), so they move up the coast (“about seven miles”, 4.23.6) and land where the coast is apertum et planum (4.23.6), “open and flat”, both adjectives implicitly in contrast with what he had faced at the cliffs. The landing is opposed by the British and there is fighting on the shore before the Romans can establish themselves. In 54 Caesar sets off with a much larger fleet (over 800 boats in total, he claims, at 5.8.6), but the wind drops and he is carried north with the tide. When the tide turns, and by dint of hard rowing, an unopposed landing is achieved (the British apparently daunted by the enormity of the incoming fleet) at “that part of the island where he had learned the previous summer that disembarkation was best” (5.8.3). The shore at this point is described as molle atque apertum (5.9.1).

The archaeological discovery at Ebbsfleet, posited as Caesar’s landing point in 54, was a large enclosure (encompassing more than 20 hectares), bounded by a defensive ditch. Aside from similarities to Roman camps found elsewhere, and conclusions drawn from a quite speculative reconstruction of Caesar’s indirect route to the British shore in 54, the key find was the tip of a Roman pilum located among pottery of a mid-first-century BC date. That, combined with Caesar’s account of the local topography, including his landing at a “sandy, open shore” (the shore at Deal and southwards is certainly not sandy), makes the case for a landing at Ebbsfleet, I think, but do please read the piece for yourselves.

We probably need a map, and here are two: one lifted from Tony Wilmott’s excellent English Heritage guide to Richborough and Reculver, and after that (since, as the first indicates, the landscape has altered dramatically since Roman times) a contemporary snippet from Google Earth for comparison.

On Tony Wilmott’s map, Ebbsfleet is the red dot furthest south on the Isle of Thanet (on the other map, it’s roughly where Cliffsend is); on the first map the east coast extends only as far south as Deal, whereas the second takes in Deal, Walmer, Kingsdown and the northern edge of the White Cliffs. In Caesar’s time modern Deal would mark the top of the long shingle coastline (though extended by spits to the north) that stretched down to Kingsdown and the White Cliffs (again, alone of my family, I walked the length of this ancient coastline a couple of days ago…).

My feeling is that the positive case for Ebbsfleet (always bearing in mind that it is only the 54 invasion at issue) is not especially strong, but I’ll concentrate on my negative thoughts. One is that a landing at Ebbsfleet would place the Roman troops on the Isle of Thanet when it was still an island. To access the interior (as he subsequently does), Caesar would have had to get his forces across a significant water barrier, the Wantsum Channel, but there is no reference to such a thing in the Commentaries, and it would be most unlike Caesar to fail to mention such a singular achievement. Another consideration is that Caesar’s account strongly implies that the two landings took place in essentially the same location, both explicitly (5.8.3) and by the almost identical language he uses to describe the nature of the shoreline in both instances. If they were at essentially the same spot, that rules out Ebbsfleet as the landing place in 54, since the location in 55, seven Roman miles or so from a point off the White Cliffs, places us somewhere between Deal and Kingsdown and nowhere near Ebbsfleet.

The third point concerns the translation of Caesar’s description of the shoreline. “Sandy, open shore” is the Loeb translation of litus molle atque apertum (5.9.1), with which we can combine the apertum et planum litus of 4.23.6. The Ebbsfleet theory sees this as a good description of Pegwell Bay, the little that remains of the Wantsum Channel. But the word mollis here is less likely to mean “sandy”, “soft underfoot”, than “easy”, “gentle” (i.e. “not steep”), “accessible”. The “traditional” location for the landings, somewhere on the long shingle beach that now extends from Kingsdown to beyond Sandwich (and in Caesar’s day from Kingsdown as far as Deal), is admirably “open” (apertum) but also molle in the sense of “easy of access” and planum in the sense of “level” (especially in comparison to the cliffs further along the coast). For me all of this makes it overwhelmingly likely that this stretch of coast is the real location of Caesar’s landing point.

Here are some images of that shingle coastline south of Deal, the shoreline running south as seen from Deal pier on the left, and the view from Kingsdown toward the White Cliffs on the right. In Caesar’s day the coastline would be further west, but essentially similar in character, we must assume:

An incidental consideration is that the stretch of water from Kingsdown to Deal, known as The Downs, has historically been a place for ships to shelter in the relative protection of the Goodwin Sands a few miles offshore. The unusually calm character of the sea along this coast is one of its most appealing features today, I can add, but it may also possibly be part of what Caesar was pointing to in molle. In any case, its general calmness does not preclude severe storms at times, and Caesar’s fleet was seriously damaged in both 55 (4.28-9) and 54 (5.10). The Goodwin Sands are more familiar as a menace to seafarers than a boon, of course.

All in all, then, I think Walmer is justified in having this memorial on its beach. The inscription is eroded, and a couple walking past when I was there were undecided whether it was Caesar or St. Augustine or “some other Roman”, but it reads, “THE FIRST ROMAN INVASION OF BRITAIN LED BY JULIUS CAESAR LANDED NEAR HERE LV BC.”

I think we can confidently extend that to LIV BC, too.