Re: USAGE: Yet another few questions about Welsh.

On Monday, July 5, 2004, at 10:47 , Joe wrote:

[snip]
> Firstly - when did the [G]>0 sound change happen?  Since the Welsh
> started writing things down, or before?
When I lived in South Wales, I could just go along to the university
library in Cardiff & do a bit of research. Alas, that's too long a treck
from leafy Surrey   ;)

But from what I remember the 'soft mutation' it's generally assumed that
it happened very early on and in parallel to the same process that was
going on in the Vulgar Latin of the western provinces, i.e. it's likely to
go back to the Romano-British period.  In both sets of languages we find
that between vowels (in in similar 'voicing' environments) the voicless
plosives become voiced, and the voiced plosive become voiced fricatives.
this is exactly what happened with Bittonic lenition with, of course, the
lenition of /m/ to [v] or [w] with nasalization of preceding vowel.

The change [g] --> [G] --> 0 is likely to have occurred in Old Welsh. In
Old French, we find VL medial /g/ had become silent, e.g. augustu(m) -->
aoust (August).

In other words,  lenition in the Brittonic langs were contemporary with
those in western VL. The big difference, of course, is that the Brittonic
langs also mutated word initial sounds within phrases. These initial
mutations, as I'm sure you know, were at first conditioned phonologically.
  But as other sound changes took place, these initial mutations sometimes
lost the original phonological environment that triggered them, but the
mutations remained.

As for writing, I do know that in old Welsh the soft mutations were _not_
noted. You had to read the letter the correct way for the environment in
which it occurred.  This makes it difficult to know exactlt when the
changes began. But then, written language does tend to lag behind spoken
language. In Middle Welsh, the changes start to be reflected in spelling -
but two of the Old Welsh spellings remain to the present day, i.e. _ac_
/ag/ (and) & _nac_ /nag/ (nor).

> I probably shouldn't be
> treating you as a font of all knowledge, but you're doing pretty well so
> far ;-)
Thank you - but, alas, I'm not a font of all knowledge. It just that I've
had a interest in languages and especially the structure of languages for
almost as long as I can remember, and I did live in S. Wales for 22 years.
  But i'm by no means an expert on things Brittonic.

> It seems to have also occured in Cornish, but not, I am led to
> believe, Breton.
That's right. Cornish behaves like Welsh in this respect, but in Breton
initial [g] mutates to [h\] (voiced glottal fricative, IPA 'hooktop h'
ɦ ), spelled _c'h_ in the Peurunvan ('super-unified') orthography used in
some 80% of published Breton. The Skolveurieg ('universitaire')
orthography spells it simply as _h_ (reserving _c'h_ solely for /X/). The
Etrerannyezhal ('interdialectal') spelling is IIRC _c'h_.

However, Breton initial [gw] mutates simply to [w] - _gw_ --> _w_ in all
spellings  :)

>
> Also, how did all the Celtic mutations move around?  You have two in the
> Goidelic languages, four in Breton and Cornish, and three in Welsh, one
> of which is unique.  In Irish, Eclipsis seems to be a combination of the
> soft and nasal mutations, Lenition the soft and aspirate ones.
It is IMO quite misleading to think of Irish mutations in terms of the
Welsh ones. They are not related. The Gaelic mutations developed
distinctly from the Brittonic ones. The two systems probably had little if
any effect on the other. (The ancient Brits & Irish did not recognize
their languages as being related). It is IMHO a mistake to suppose there
was a common 'Celtic mutation' system.

The lenition or 'softening' of the Brittonic langs is, as I've noted above,
  parallel with similar intervocalic lenition in western Vulgar Romance.
The Gaelic lenition was quite different and remarkably similar to the
lenition we find, e.g. in Biblical Hebrew - the plosives became aspirated.
  Old Irish also extended this to /s/, /w/ and /m/.

Later, as in Hebrew, the aspirated plosives became fricatives (the modern
Ashkenazi, Sephardi & Israeli pronunciations preserve only some fricatives;
  the Yemini Jews, I'm told, retained all the older fricatives). In Irish
they continue to be so, although the alveolar fricatives are no longer
alveolar, but have merged with the velar/glottal fricatives (which, of
course, are palatal before front vowels); so, e.g.
soft [p] --> [p_h] --> [f]
soft [t] --> [t_h] --> [T] --> [h]

The Irish 'elipsis' is its own nasal mutation - it has nothing whatsoever
to do with Brittonic soft mutation. The voicing of of /p/, /t/, /k/ after
a nasal is quite widespread. I first came across it in Tamil as a young
teenager. It occurs closer to home in modern Greek, e.g.
/opat'era/ (the father [nom]) ~ /tomba'tera/ (the father [acc.]) <-- /ton/
  + /pa'tera/.
The pronunciation of /mb/ varies dialectally between [mb] and [b].

As I said, the modern Irish t- --> td- /d/ is nothing to do with Welsh
mutation; but it does have a close parallel in modern Greek  :)

The changes [mb] --> [m], [nd] --> [n] and [Ng] --> [N] are attested in
many languages besides Gaelic.

> Breton
> and Cornish don't have any kind of nasal mutation, but add a hard
> mutation, and a 'mixed' one.  It all seems rather, well, confusing.
OK - to return to the Brittonic langs. Besides the soft mutation or
lenition, all three have the 'spirant mutation' which /k/, /p/ and /t/
which become fricatives. This arose partly from a syllable final /h/ (from
earlier /s/) before the consonant.

Yes, Cornish & Breton do have a more complicated system: soft, spirant,
hard & mixed. I'm not au_fait with all the details, but I'm sure
nasalization played a part somewhere (in the mixed?) - IIRC the familar
three of Welsh  are due to simplification of the older, more complicated
system. the simplification still continues. The soft mutation is alive and
well in all dialects. The nasal mutation occurs only after 'fy' (my),
usually pronounced either as [@] or zero, and 'yn' (in). But the latter
triggers nasal mutation only in Welsh place names, not with foreign ones.
In many spoken dialects, the spirant mutation has disappeared, replaced by
the soft mutation, except in certain fixed phrases (e.g. tri chant 300)
and after 'ei' (her).

Sorry not to be more helpful with the history of the Cornish & Breton
mutations.

Ray
===============================================
http://home.freeuk.com/ray.brown
ray.brown@freeuk.com    (home)
raymond.brown@kingston-college.ac.uk   (work)
===============================================
"A mind which thinks at its own expense will always
interfere with language."         J.G. Hamann, 1760

Reply