On Wednesday, October 13, 2004, at 01:10 , Elliott Lash wrote:

> --- Rodlox <[log in to unmask]> wrote:
>>> He could have added that in Welsh focused elements
>> are fronted.
>>> Prynodd Huw  gar ddoe.
>>> (bought Hugh car yesterday).
>>>     English (cleft): It was Hugh who bought a car
>> yesterday.
>>>               Welsh: Huw (a) brynodd gar ddoe.
>> [The (a) is normally
>>> omitted in spoken Welsh]
>>>     English (cleft): It was a car (that) Hugh
>> bought yesterday.
>>>               Welsh: Car (a) brynodd Huw ddoe.
>>>     English (cleft): It was yesterday that Hugh
>> bought a car.
>>>               Welsh: Ddoe (y) prynodd Huw gar.
>> [The (y) is normally
>>> omitted in spoken Welsh]
>>>               Welsh: Prynu gar ddoe (a) wnaeth
>> Huw.
>>>                      (Buy  car yesterday did
>> Hugh).

OOPS!!!!! *blushes deeply*

I got carried away with those mutations. There is no mutation after the
verb-noun _prynu_ & yesterday should go with the finite verb:
Welsh: Prynu  car      (a) wnaeth Huw ddoe.
       (buying of-a-car     did   Hugh yesterday)

= Hugh *bought* a car yesterday/ What Hugh did yesterday was to buy a car.

>>> (The Welsh examples have the added interest of
>> initial consonant mutation
>>  um, if I might ask, how/where did it mutate?

Where? In Britain. - probably as early as the Romano-British period
because similar changes to the Brittonic 'soft mutation' were going on in
western Vulgar Latin at the same time.

How? By normal process of one sound affecting the pronunciation of an
adjacent sound or of a sound being pronounced differently in different
environments. It goes on all the time in languages. For example, in
colloquial British English /t/ between vowels or as syllables finals are
'automatically' pronounced as [?]. In American English, intervocalic /t/
is, I understand, a dental or alveolar flap.

The thing that distinguish Welsh and other modern Celtic langs is that in
_initial_ position, some of these changes that were originally 'automatic'
  later became fixed and grammaticalized.

>> as
>> far as I can see,
>> |prynodd| & |brynodd| became |prynu|...but wouldn't
>> that be a suffix or
>> affix to the opening word?

Well -odd and -u are suffixes, not an initial consonants. I did not
mention suffixes; nor did I mention initial _words_. In any case |prynodd|
  & |brynodd| are not by any means always the opening word.

>>  apologies if I'm misunderstanding what you mean by
>> 'initial'.

Probably you understand the same as me by 'initial' but you do not seem to
understand the same as me by "initial consonant", which is what I actually

By _initial consonant_  I mean "the consonant at the beginning (of a word)

In other words, "initial consonant mutation" has nothing whatsoever to do
with suffixes like -u and -odd. But it has a lot to do with the initial
consonants in:
prynodd ~ brynodd
car ~ gar

> Well, what happened was this:
> prynu "buy"  (stem pryn-)
Yep - the stem is pryn- and the verb-noun* suffix is -u.

> becomes:  pryn + odd   (3rd singular past tense)

Yep - spot on. Just like, say, German _kauf-en_ and _kauf-te_  - nothing
unusual there.

> Then in certain contexts, the "p" mutates to "b".
> Since the "p" is at the beginning of the word, it is
> called "initial consonant mutation".

Yep - after the particle 'a' (even if omitted in speech) soft mutation
takes place, so
Prynodd Huw ~ Huw (a) brynodd

But after the particle 'y' there is no mutation, so:
Ddoe y prynodd Huw gar.

Similarly, the Welsh for _car_ is 'car' (plural: ceir). So we have:
Car a brynodd Huw.

Its placement before the connecting particle _a_ prevents it from mutating.
  But normally the direct object of a synthetically conjugated tense (Welsh
tends to use analytical constructions for most tenses, moods etc)
automatically has soft mutation, so:
Prynodd Huw gar.

*the verb-noun (berfenw) is roughly the equivalent of the infinitive or
the gerund in other languages. But it is strictly a noun; unlike
infinitives & gerunds, it does not take the same object arguments as a
finite verb. Thus in the correct version of the last Welsh example we have
_prynu car_ with no mutation of _car_ since it is *not* the direct object
of _prynu_ even though it it is the direct object of _prynodd_ in _prynodd

_prynu car_ is the Welsh genitive construction: the two nouns are merely
juxtaposed, the possessed bying followed by the possessor, e.g.
llyfr bachgen
book  boy      = a boy's book

So: prynu car = buying of a car

Fascinating things, these mutations - but I must be careful in checking
them next time   :)

[log in to unmask]
Anything is possible in the fabulous Celtic twilight,
which is not so much a twilight of the gods
as of the reason."      [JRRT, "English and Welsh" ]