Camulodunum
Camulodunum was Britain's first capital, the base of king Cunobelinus and then a Roman colonia. All the main ancient sources supply the name, among which CAMVLODVNO on coins seems pretty definitive. Some spellings are weird, such as Manulodulo Colonia at or Ravenna Cosmography 111 or Καμουδολανον at Ptolemy 2.3.22. The location at modern Colchester is beyond doubt, thanks to copious archaeological remains. So there ought to be no mystery about the name.
Unfortunately, the meaning of Camulus has never been settled, despite much discussion. See particularly Evans (1967). Camulus is commonly claimed as a Celtic war god, but that idea rests mainly on a small number of inscriptions referring to Mars Camulus associated with a region in Belgica, particular around the Montagne de Reims. Olmsted (1994:334-5) argued for Camulus as a free-standing deity name of equal status to Mars but in a different language. It may be better to see Camulus as more of an adjective-like attribute, like most of the “Celtic deities” identified with Mars (listed here), which tend to be geographical more than religious.
It makes best sense to translate camulus as ‘small hill’, a meaning that may previously have been missed because East Anglia is so archetypally flat. Actually the land drops away by ten to twenty metres all around the historic walls of Colchester, which a pedestrian there really notices. The Latin diminutive ending -ulus shows up on several Latin words referring to types of hill (tumulus, colliculus, monticulus) and in ancient personal names such as Andecamulus that were analogous to modern names such as Hill. A particularly fine physical realisation of a *camulus may be seen at Redhill, SK492305, where the river Soar joins the river Trent and a curse tablet has been found that mentions Camulorix.
The root *cam shows up in Latin camera ‘vault’ and in the northern English word kame or kaim ‘ridge, hillock’. However, words for curves in place names are controversial. See, for example, the discussion of camb ‘comb’ and of cumb and cwm ‘valley’ by Gelling and Cole (2003). Celticists tend to see an element cam-, in at least eight ancient place names in Britain and on the Continent, as referring to curves in a horizontal plane (like bends in a river), whereas in most cases the landscape suggests vertical curves (small hills). The nearest parallel may be the southern French word camelle, used for a heap of salt crystals, possibly derived from Latin cumulus ‘heap’.
About -dunum see here. An element camulus also shows up in Camulossesa.
Ptolemy's Καμουλοδουνον πολις of the Brigantes was probably some kind of hillfort, or oppidum on a low hill, with coordinates that point in the general area of Huddersfield. The most likely location is the Adel Roman settlement around SE277410 north of Leeds, discussed by Wilson (2015). Candidate places that can be firmly ruled out include: Slack Roman fort, suggested by Rivet & Smith (actually Camboduno); various viewpoints without obvious fortifications; and hillforts at Birstall and Barwick-in-Elmet. Castle Hill, at Almondbury, SE153141, is notably large and high and served as the central place for a substantial population, a role later filled by Huddersfield market town, but its archaeology does not fit. Another candidate east of there is South Kirby Camp, at SE435105, possibly more of a defended settlement than a hillfort, probably an indigenous power centre for a lot of good farmland. It was shadowed by Burghwallis Roman fort, 8 km away at SE519120, where the modern A1 main road (= the Roman road from Doncaster to Tadcaster, number 28b of Margary, 1973) crosses the river Skell.
A third place with a similar name occurs in the Ravenna Cosmography's sequence of names Mantio - Alunna - Camulodono - Caluvio, which appears to be heading generally north out of Manchester, but not towards the Adel Καμουλοδουνον. On that road the first good candidate is the promontory fort at Castle Steads, but that probably belongs to Alunna. So Camulodono was most likely the Hawksclough hillfort at SD57402405, near Clayton-le-Woods, Lancashire, which is just 4 km from the Roman site (possibly a supply depot rather than a fort) at Walton-le-Dale, at SD55132812, on the river Ribble.
You may copy this text freely, provided you acknowledge its source as www.romaneranames.uk, recognise that it is liable to human error, and try to offer suggestions for improvement.
Last edited 14 October 2020 &nsp; To main Menu