Attested:  AI iter 13 Durocornovio

Where:  The Roman settlement at Nythe Farm, SU192854, on the outskirts of Swindon, where a Roman road crossed the river Cole.

Name originDuro ‘crossing’ of a river *Cornovium, analogous to *Coronavis, seen also in Purocoronavis.  There is plenty of evidence for *navis ‘river’ but the coro- part is more difficult.  Among its many possible meanings, the one that seems best is ‘bendy, winding’ from PIE *(s)ker- ‘to bend, turn’, whose descendants include κορωνος ‘curved’, Latin corona ‘garland, crown’, German Kehre ‘U-turn’, and a huge range of other words from whore to church.  Roots considered but rejected include: PIE *kar- ‘hard’, which Nicolaisen (2001:241-2) cited to explain rivers with “pre-Celtic” names like Carron; PIE *ka- ‘to like, desire’, which Ekwall indirectly invoked for Carrant, Carey, etc; and PIE *ker- ‘head, horn’, the likely root of Ptolemy’s Καρνονακαι.  A quick search for rivers that look a bit winding and have names possibly related to coro yielded Churn, Churnet, Cerne, Cherwell, Ceiriog, and Char, plus other lost names deduced by Ekwall.  Multiple instances of Cornovii people presumably lived in their river basins.

Notes:  AI’s iter 13 missed out one line near here, which R&S suggested belonged at Cirencester (Corinium).  Presumably this means the original names of the rivers Cole and Churn, at two consecutive crossings misled an ancient scribe.  The river Cole “ has been extensively modified by man for a variety of reasons over the past 900 years”, so it is hard to guess how bendy it used to be.

Standard terms of use: You may copy this text freely, provided you acknowledge its source, recognise that it is liable to human error, and try to offer suggestions for improvement.
Last Edited: 6 August 2016