thanks Martin C especially and Henry for comments on PWJ's swimming-fords.
The Berkshire Ock is usually/always seen as a "salmon" river. (KJ's British *esako quoted by MG in PN BRK, W. eog).
I'm going to say that I think this is an excessively phonological explanation of the name.
I don't think there are any other Celtic salmon RNs - though see the French toponomists on *salar- (the gen. view is that the fish is not the referent here despite appearing to be so).
There are of course ON salmon rivers (e.g. Laxey IOM and Leixlip in Leinster).
However, the source of the Ock is at the very foot of the former Aescesbyrig (Uffingdon) BRK, with the White Horse carved into the chalk of the top of the hill.
See MG's "on occene wyllas" at Woolstone BRK. (PN BRK p. 14)
(I see Aescesbyrig as - probably- a re-interpreted uxello- name, see previous posts).
The river runs through the Vale of the White Horse (this must be a modern name) and joins the Thames at Abbingdon.
I think the probability, using elephant theory, is that the White Horse must be the original referent of the RN.
Elephant theory is just a useful tool - it can give wrong solutions (e.g. you could see the Pool of London *lindon as the elephant referent of the London name, & this is pretty surely not right).
But this is a very beautiful big white elephant, to mix metaphors.
Nicholas Thomas dates the horse as "prob. 1st Cent BC ". (He does flag up that much later dates have been proposed). (NT - Guide to Prehistoric England Batsford/BCA 1960 and 1976. Photo of horse on p. 181)
For the forms of the RN, please see ERN p. 307 and PN BRK p. 14,15
For Old Irish "ech" please see Patricia Kelly's chapter "The Earliest Words for "Horse" in the Celtic Languages" in
"The Horse in Celtic Culture" ed. Sioned Davies & Nerys Ann Jones UWP 1997.
(an excellent work)
Following P. Kelly, is it a problem to posit a Q-Celtic form of IE *ekwo for this name?
(the archaic W survives in compounds and derivs e.g. ebol "foal", Mynydd Eppynt BRE like OE "horspath" OXF)
Not really: she compares what seems to be the "horse month" EQUOS in the Calendrier de Coligny (see P-Y Lambert on this artefact in "la Langue Gauloise"/Errance 1997.
She says: "the retention of the older Q-Celtic form in a P-Celtic area may be due to the archaic religious context".
There is a Spanish rock inscription "equeisui" which PK translates as "equine".
See the French toponomists for a possibly relevant discussion of the Sequana/Seine name. (might or might not be a -q- survival in a p-area).
thanks for reading this post.
nick
|