Attested: (1) Ptolemy 2,3,3 Ουξελλα
(2) Ptolemy 2,3,8 Ουξελλον or Ουζελλον a πολις of the Σελγοουαι; Uxela at position 169 in the Ravenna Cosmography
(3) Ptolemy 2,3,30 Ουξελλα or Ουζελα a πολις of the Δουμνονιοι; Uxelis at position 13 in the Ravenna Cosmography
Where: (1) Axe estuary mouth in north Somerset, by
Uphill at ST31795818 and Brean Down, formerly an important port.
(2) Ward Law Roman camp at NY024669, uphill from Caerlaverock, Dumfries.
(3) Cadson Bury hillfort, at SX343673 in Cornwall, by a crossing of the the river Lynher. This location outranks R&S's idea of a lost Roman fort under Launceston Castle mainly because it gives the best track across the map for RC's list of names, here following a Roman-style almost-straight route from Isca Dumnoniorum, the fortress at Exeter, to Duriarno, the fort near Bodmin.
Name Origin: *Uxela ‘high point’ has excellent parallels in Celtic words for ‘high’(Welsh uchel, Breton/Cornish uhel, Irish uasal ). That descends from PIE *aug- ‘to augment’ (from *h2eug-), which also led to Old English eacan, Dutch oken, Gothic aukan, and Latin augeo ‘to increase’, plus German auch ‘also’, Russian высокий‘high’, Greek αυξις ‘growth’, and English eke. A meaning ‘high’ and that extra L also appear in Greek υψσηλος. A suffix -el was used extensively in Germanic noun formation (Smith, 1956). Many later place names across England contain an element *ecels/*iecels ‘addition, land added to an estate’, not separately attested in a text but pronounced very like Welsh uchel, in which U in sounds very like an I in English. Conjugated forms of Old English eacan often contained vowel Y, presumably pronounced like Ü.
Notes: Latin ocellus ‘little eye’, from oculus plus diminutive suffix -lus (see here), and hence perhaps ‘lookout’ plus auxilium ‘helper’ could have influenced the uptake of this name element. Uxelodunum (Stanwix near Carlisle) and the Ochil Hills in Scotland preserve the L. Uxacona (Redhill, Staffordshire) has no L, as does icanhoe (in the Anglo-Saxon Chronicle for AD 653), probably modern Iken in Suffolk, where Yarm Hill shows that in a flat landscape even a 15-metre hillock can count as high. In France, at least 12 modern place names may derive from forms containing *uxelo-, presumed to be Celtic (Lacroix, 2003:122-3). Below site (2) (see here for a map facing its p104) the salt-marsh has grown since Roman times due to isostatic rebound, so it is possible that the site of Caerlaverock Castle was an inlet capable of being a port, in the niche later occupied by Glencaple. That area was settled by Friesians before AD 400, as explained by Skene (1862) (hence names such as Dumfries), which raises a suspician that Roman auxiliary troops from Frisia were encouraged to retire to farms north of Hadrian's Wall. Site (3) controlled a significant old crossing of the river Lynher indicated by the names Durnaford (‘hidden ford’) Farm and Newbridge, while higher up that river lay metal ores likely to have interested the Romans, plus a suggestive name Upton Cross. The lack of obvious Roman forts in this area possibly hints at mutually profitable relationships with the locals.
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 5 April 2020 To main Menu