Place-names on the Saxon Shore [1]:

Example: Etymology of DOVER

This explanation of the name Dover is based on part of a paper submitted for publication
in an academic journal. On the web we can be a bit more chatty and can add more pictures.

Dover is “Exhibit A” whenever people argue the case that people spoke a Celtic language in Roman-era south-east Britain. Standard books about the history of English (by top authors such as David Crystal) explain that very few place names in or near Kent can be traced to a Celtic root, but Dover stands out as being unquestionably Celtic.

This argument is wrong, but it will take some space to explain why. The critical point is that something like 170 place names in England end in –over (or its relatives) where no one disputes an origin from an Old English word that meant something like ‘sea shore’ or ‘bank’. Why should Dover (and a few other names that end in –dover) be exceptions?

Let’s go back to the earliest evidence. A Latin document called the Antonine Itinerary, written around AD 300, but probably based on Roman army marching orders from before AD 100, twice mentions ad Portum Dubris. Then Dubris shows up in three more late Roman documents, but is absent from Ptolemy’s Geography of about AD 140. After that it shows up in Anglo-Saxon texts as Dofras, Dobrum, Doferum, Doferan, etc.

The standard book on Roman-era place names of Britain is by Rivet and Smith, who completely accepted the Celtic argument that Dubris was named from two small rivers that flowed down through the cliffs to reach the sea in its little port. Their exact words were: “the British name was *Dubras ‘waters, stream’ (perhaps ‘streams’), plural of *dubro- ‘water’ (Welsh dwfr, dwr, Cornish dofer, dour, Breton dour; Old Irish dobur) … all records of the name, even those of the Antonine Itinerary set in a grammatical structure, show it as a locative plural in -is”.

These authors are so highly respected that no one seems to have noticed their mistake in Latin grammar. The word dubris was far more likely to be a genitive singular than a locative plural. So its Latin nominative form could have been either duber or dubris. Of course it is not fatal to the Celtic argument to be knocked down from two rivers to one, but there is worse to come.

The argument that dubris came from Celtic is in fact circular and all boils down to a dogmatic presumption that Dubris existed before Dover (etc). Step one: decide that Celtic was the language most likely to have been spoken in ancient Dover. Step two: look for a Celtic best fit to dubris. Step three: construct grammatical and phonetic arguments why the hypothetical Celtic root is closer to dubris than its cognates in other languages, such as English dub, deep, dip and dimple.

According to Pokorny (who wrote the classic dictionary of the Proto-Indo-European base language) the exact word dubris existed in Illyrian. That sounds exotic, but Illyria was where Julius Caesar spent the winter between his two trips to Britain and where the Romans recruited troops to garrison Britain. Furthermore, Illyrian may be just one of a band of ancient languages that got squeezed out between Latin and Celtic.

To anyone who lives by the sea, calling a place ‘port of the waters’ sounds daft. Also, most English place names are two-part compounds (qualifier plus generic), so it is slightly curious that Anglo-Saxon Dofras etc dropped the Port part.

So the argument in favour of Celtic ‘waters’ looks a bit feeble, but can the Germanic languages do any better? A proto-Germanic form like obera can be reconstructed from modern German, Dutch, etc words for bank/shore/beach, which obviously seems appropriate for Caesar’s troops landing ‘on the beaches’ like D-day in 1944. But where could an initial D have come from? Modern Dutch aan de oever ‘on the beach’ or its Old English equivalent aet ofer are ruled out (if you believe standard linguistic doctrine) because definite articles evolved too late in the Germanic languages.

The story of English place names containing ofer is complex, but its essence was worked out by Eilert Ekwall, Margaret Gelling, and Ann Cole. In brief, there may have been two words ofer, of which one led to modern over, while the other one meaning bank/beach had variants ufer, yfre, and ora. Perhaps this variety reflects different dialects of Old English, with ora being a loan word into Saxon from Latin, where it originally meant ‘sea shore’ but came to mean ‘land ahoy’.

Whatever the precise etymology, ofer/ora place names are associated with a distinctive topographical feature – a flat-topped ridge with a convex shoulder, like the end of an upturned canoe, usable as a landmark by ancient travellers on land or water. All round the coast, from Exeter in the west to Maidenhead high up the Thames, every port of any significance in Roman times seems to have been marked by at least one ofer/ora place.

Imagine sailing without modern aids from the Solent (where there are more than 20 ofer/ora place names) around the coast and into the Thames Estuary. You could navigate past the following seamarks, all distinctively visible though often not particularly high:

The Owers near Bognor
Oreham inland from Shoreham
Ore near Hastings
Drellingore west of Dover
River inland from Dover
Stonar south-east end of Wantsum Channel
Oar Farm north end of Wantsum Channel
Oare near Faversham
The Nore near Sheerness
Upnor in the Medway

It is an obvious suggestion that Dover itself belongs as a seamark in this list. Its massively visible notch in the white cliffs constitutes a very definite double ofer. However, although this explanation is better than Celtic ‘waters’, it is far from perfect. For a start the ofer/ora expert, Ann Cole, did not run with the idea. And Dover’s two valleys in the white cliffs, plus Shakespeare Cliff to its left and St Margaret’s to its right add up to eight instances of an ofer/ora, or ten if one includes Drellingore and River.

St Margaret's bay: a gap in the cliffs.

North of it lies the historic sailing-ship anchorage off Deal known as the Downs, inshore of a complex of sandbanks known as the Goodwin Sands. South of Dover, deeper and further out, lies the Varne Bank. The Goodwin Sands are notorious for swallowing wrecked ships and are partially exposed at low tides. Their southern tip now lies off St. Margaret’s Bay (under the sea in the photograph above), about 5 miles from the historic centre of Dover.

Let’s turn now to that pesky initial D, which seems to be consistent in all forms of the name. The fact that it is a single letter is no problem, because lots of place names result from misdivision: picking up a single letter by transfer from a preceding word. They include several instances of The Nore (formerly atten ore) and River (probably built from atter ‘at there’), plus Hever.

A plausible source for a D is the numeral two, as in Latin dua orae ‘two shores’, for example. One manifestation of two is the prefix dis- (or di-) denoting separation or division in lots of modern English words, which have completely superseded lots of Old English words that began with a prefix to-, such as tofær ‘departure’.

Two small linguistic quibbles need to be addressed. The initial D in dubris/Dover hints that its parent word existed before Old English used T in to- and words like two. And the way that to- generally compounded with verbs not nouns hints at a verbal deep etymology of ofer.

However, the really critical question is whether anything at ancient Dover could justify a name that meant something like ‘double bank’ or ‘twin beach’.

A possible answer can be seen in Bruges, Belgium, where Dijver or Dyver is the name (with no known etymology) of a historic boat-unloading basin flanked by two beaches. Bruges or Brugge in Dutch had during the Middle Ages a direct canal to the sea, called sincfall.

The Dijver or Dyver in Bruges (Belgium) street lies next to a former double, v-shaped beach where ships could safely be beached. In blue, the existing canal.

Isle of Wight

And in the Isle of Wight, duver or dover is a generic local word for a low-lying piece of land along the coast, subject to occasional inundation by the sea. The etymology is unknown, but certainly goes back before AD 1774.  Nowadays there are four examples, at St Helens, Seaview, Ryde, and Hamstead.

It requires some local knowledge, or careful perusal of Ordnance Survey maps, to discover the distinctive common feature of all four duvers/dovers: they have (or had) water or marsh on their landward side and were more in the nature of a sandbar than a beach.

Right-click here to open this document in a separate tab of your browser, then look at these illustrations to see the four Isle of Wight dovers.
Plate 3 shows a view of where Ryde Dover used to be (before the Esplanade and Canoe Lake were built)
Plate 4 shows Seaview Duver
Plate 5 shows St.Helens Duver
Plate 28 and Figure 18 show Hamstead Duver



It is inherent in the nature of spits or sandbars across the mouth of an estuary (or a river confluence) to change over the centuries or through human interference, so recognising dover in other names is difficult. Nevertheless two clear examples from the mainland go back to Domesday Book in AD 1084.

Now the two mainland Dovers.

Doverhay (today a street in Porlock in Somerset) was Doveri in Domesday Book and its estuary has developed an alluvial fan:

The village of Dovercourt (next to Harwich in Essex) was Druvrecourt in Domesday Book and still shows a perfect estuary-closing spit:

 

Dover, duver = double beach

For sailors even now it can be a matter of life and death to avoid underwater shoals by recognising the changes in water colour and wave patterns that they produce. So, from the point of view of anyone who lived at or visited ancient Dover, its most noteworthy characteristic must have been the proximity of a massive offshore sandbank. Therefore we confidently translate ancient Dover/Dubris, like modern dover/duver in the Isle of Wight, as something like sandbar or ‘double beach’.  This is perfectly well illustrated with the Dubris of the Roman era: