On 21/02/2018 18:22, And Rosta wrote:
> On 21 February 2018 at 17:49, Raymond Brown wrote:
>> On 21/02/2018 16:47, And Rosta wrote:
>>> On 21/02/2018 16:55, And Rosta wrote:

>>> But it's also highly likely that subsequently /h/ and /θ, ð/
>>> were lost again. But alas that cannot spare you the labour of 
>>> documenting their life span.
>> Why?  / /h/ remained in French till the 15th century at least, and 
>> still remains in Picard and Walloon.
>> Most Germanic languages have now lost /θ, ð/ - but Icelandic and 
>> English have retained them.  Breton has lost them (except one 
>> dialect that retains /ð/), but Welsh has retained them.   Why 
>> should not the inhabitants of Southern Britain (i.e. south of 
>> Gaelic speakers) likewise retain them in BART.
> My thinking is that /h/ and  /θ, ð/ rank high on the scale of 
> susceptibility to loss. In English, many dialects lack /θ, ð/ and 
> (not quibbling about the details) /h/: what keeps them alive, I 
> suggest, is that their loss has been stigmatized and hasn't occurred 
> in the prestige dialects.

...and that they continue to be pronounced in some areas.

The shift of [θ] -> [f] is, indeed, widespread in colloqiual British
speech, as is the shift of [ð] -> [v] when medial or final.  When
initial [ð] seems either to shift to [d] or fall silent.  But these seem
to have spread from London and, probably, other urban areas during the
last century.  There is no parallel for such shifting in Romance
languages.  The question of [ð] is more complicated (Portuguese and
Spanish have it, for example), but the evidence is surely that if
Germanic [θ] was not retained it would have shifted to [t] *at an early

As for /h/, yes it true that it is widely dropped in colloquial English
and, indeed, in south Wales, but it remains in north Wales and, I think,
largely in Scotland.  I was going to say that [h] came in with Germanic
borrowing but did not maintain itself and was dropped just like the
Latin /h/.   But if it held on till the 15th cent. in France and still
holds on in Picardy and Wallonia and in the Romansh dialects, that seems
to me a week cop-out.

On 21/02/2018 19:09, Pete Bleackley wrote:
> Tolkien believed that the retention of the dental fricatives in 
> English was an areal feature, due to contact with Welsh.

I agree that this is an areal feature, but I would not add 'due to
contact with Welsh'.  If it was due to such contact, then surely we
would expect Old English <hl> to be a voiceless alveolar lateral
fricative [ɬ], as it is, indeed, in modern Icelandic.

What I have written on my 'Preliminary Considerations' page is:
I shall be aware of areal features such as the development of
progressive tenses and the retention of [θ] and [ð], since the peoples
of Britain will not differ very much in BART from those in our own
timeline. I shall not, however, apply either early Welsh or English
sound changes to the language in a so-called 'bogolang' manner. I shall
from time to time refer to Welsh and English in our timeline, but only
for considering common features, which might have, therefore, been
common to our British Romance.

While [ɬ] is confined to Welsh and does not occur in Cornish or English,
I do not consider it to be an areal feature.  On the other hand, [θ] and
[ð] are found in Welsh, Cornish and English and I maintain this is
evidence of its being an areal feature, just like, for example, analytic
progressive tenses.

n 21/02/2018 22:04, Melroch wrote:
> All the Iberian languages, some varieties of Occitan and some South 
> Italian varieties have [ð] as an allophone of /d/.

Or some would hold they have [d] as an allophone of /ð/.   :)

Yes, it seems [ð] is more widespread and resistant than [θ].  I observed
in an earlier email that while Welsh and Cornish have /θ/ and /ð/,
Breton does not.  There /θ/ has gone entirely, having apparently merged
with /ð/ before the latter shifted to [z] or [ɦ], according to dialect
(hence the Breton spelling <zh> for the sound).  But the older sound [ð]
has persisted in in some varieties of Haut-Vannetais, where it may be
written <z>, <ẓ> or <dd>.

> Also if Britain suffered as much Norse invasion and settlement as it
>  did here Britainese will have at least as much Norse borrowings as 
> Norman, and possibly more varied.

As I wrote on the 'Preliminary Considerations' page:
Vikings were as energetic in BART as they were in our own timeline. But
it should be recalled that although there were Vikings settlements in
coastal regions around Ireland, the Scottish isles and coastal areas of
Britain itself, nowhere did the Norse language take root, except in
Orkney and Shetland where Norn dialects seem to have held on among an
ever-diminishing number of speakers until the 19th century. There is no
reason to suppose that this was not so in BART as well.

But there is no Norman invasion in 1066.

> English would surely have more sure Norse loans if Old Norse and Old
>  English hadn't been so alike! Additionally I would Google "Norse 
> loanwords in Irish".

I shall.

> FWIW a lot of personal names in all Western Romance languages are 
> Germanic, and make sure to look at Old French specifically. Many 
> military and administrative terms of Gmc. origin are now obsolete.
> I would certainly look at Meyer-Lübke's Germanic index verborum: