Re[5]: [tied] Gemination in Celtic

From: Brian M. Scott
Message: 56388
Date: 2008-04-01

At 2:17:32 PM on Tuesday, April 1, 2008, fournet.arnaud
wrote:

> From: Rick McCallister

>> I probably confused Tutbury, I did this from memory My
>> point is that if Norman French dropped s- before
>> consonants in Gmc topos, we should see something similar
>> in N. France

> I suppose it might be more complicated.
> Dutch-speaking areas in N. France kept s-C.
> French-speaking areas have e-C

Mostly. I know of at least one case in which the prothetic
vowel is now /a/, and I've found one French place-name whose
history shows /st-/ > /t-/ > /it-/.

Itteville (Yvelines) is <Steovilla> ~795, 850; <Theovilla>
1175; <Itevilla> 1268; <Ytevilla> ~1290. On the basis of
the early citations Morlet takes this to contain the Gmc.
name <Stadulf>; <Theovilla> then shows loss of /s-/ (with
<th> for /t/), though a prothetic vowel was added later.

> I suppose the drop of s- in some English topos must be a
> mutual reaction of English and Anglo-normand together.

In fact Domesday Book shows both treatments: <Esket, Eschet>
for ON <Skiótr>, <Esmellt, Esmeld> for OE <Smelt>, but
<Calp> for ON <Skálpr>, <Codreschelf> for ON <Skoðraskialf>.
Snellend appears both as <Esnelent> and as <Sneleslunt>.

> There's no reason English speakers should adopt
> Anglo-normand ways if they didn't want to.

They didn't always. For instance, the place recorded in DB
as <Codreschelf> is now <Skutterskelfe> and indeed already
appeared as <Skuðerschelf> in 1176; in this case, unlike
that of Nottingham, the (Anglo-)Norman form was not
retained. Another example is Sneinton: the OE name was
either <Snotingatu:n>, with the same first element as
<Snotingaha:m>, or <Snotantu:n>, with the genitive of a
masculine name <Snota>, a weak form of <Snot>. The name
appears as:

Notintone 1086
Notington 1263
Snotinton(e) 1165-97
Snottinton 1174
Sneinton 1194-
Snai/ynton 1227-1486
Snointu/on 1233

Ultimately the AN form without <S-> didn't survive, but it
did persist for a while. Conversely, Nottingham appears as
<Snotingham> as late as ~1275.

It appears that both options, adding /e-/ and dropping the
/s/, were available both in England and on the Continent; in
France the /e-/ option became standard, while in England the
/sC-/ combinations usually prevailed. <Nottingham> is
exceptional, perhaps because it was a more important town.
I don't know whether any AN innovations of the /esC-/ type
survived, but I'm not aware of any.

Brian