Re: [tied] Re: ka and k^a

On Wed, 28 Sep 2005 15:07:25 +0000, Rob
<magwich78@...> wrote:

>--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:
>> On Tue, 27 Sep 2005 14:08:46 +0000, Rob
>> <magwich78@...> wrote:
>
>> >> I reconstruct a 2x3 vowel system for pre-PIE. Under the
>> >> stress, the normal developments were:
>> >>
>> >> *a > *e
>> >> *i > *e
>> >> *u > *e
>> >>
>> >> *a: > *o:
>> >> *i: > *e:
>> >> *u: > *o:
>> >
>> >Are there any attested languages in the world that reflect such
>> >changes?
>>
>> If we assume that the merger of the short vowels passed
>> through a stage *& (*y&, *w&), there are plenty of attested
>> languages that show evidence of such a phenomenon. The
>> merger of /i/ and /u/ to /&/ is particularly common (e.g.
>> Slavic), but a lower vowel may also be affected (e.g.
>> Tocharian /i/, /u/, /e/ > /&/). And /&/ > /e/ is also
>> rather common (e.g. Catalan: Mallorcan k&d&na > Mainland
>> East Catalan k&dEn&).
>
>Are you using '&' to mean schwa, or the 'a' in English _cat_? I
>prefer to use '@' for the former and '&' for the latter.

But @, used in a word, is interpreted by Yahoo groups as an
email address, and is mangled.

>Forgive my ignorance, but where did Slavic merge /i/ and /u/ to /&/?

The jers.

>If /i/, /u/, and /e/ > /&/ in Tocharian, why not /o/ also (unless /u/
>had already undergone changes without /o/)?

I mentioned that yesterday. /o/ was not a short vowel. In
Tocharian, /o/ merged with /e:/ (palatalization aside).

>Could you provide some of the attested languages that show short
>vowels (particularly stressed ones) passing through a stage *& (*y&,
>*w&)?

The NW Caucasian 2-vowel languages, where all short vowels
became /&/ (with or without palatalization and labialization
of the preceding consonant), and long vowels /a(:)/.

Diakonoff proposed it for Afro-Asiatic.

>> >> Because *o: had no short counterpart, the length was not
>> >> contrastive and could be lost (but *ó is still long in open
>> >> syllables in Indo-Iranian, and does not get reduced to /ä/
>> >> in Tocharian [as are /e/, /i/ an /u/]).
>> >
>> >Unstressed */o/ is also long in InIr, though: e.g. Vedic
>> >_ma:náyati_ < *monéyeti.
>>
>> I didn't say it wasn't.
>>
>> I limited my diagram above to stressed position not to
>> complicate matters. But if you insist:
>>
>> In pretonic position, the normal developments are:
>>
>> *a > 0
>> *i > 0
>> *u > 0
>>
>> *a: > *e
>> *i: > 0
>> *u: > 0
>>
>> In posttonic position, short vowels were lengthened unless
>> the stressed and/or posttonic syllable was already
>> long/heavy, and we have:
>>
>> *a > 0 / *o
>> *i > 0 / *e(:)
>> *u > 0 / *o
>>
>> *a: > *o
>> *i: > *e(:)
>> *u: > *o
>
>Is there often a distinction made between pretonic and posttonic
>positions in languages?

Yes.

>How does such a distinction come about?

Different factors. Initial syllables tend to be more
prominent than final syllables. Posttonic syllables have a
distinctive falling intonation (some accented Vedic texts
only mark the svarita).

>> Immediately adjacent to a morpheme boundary *i(:) and *u(:)
>> usually keep the glide, so we have *yé (~ *í), *y (~ *i) and
>> *ye(:); *wé (~ *ú), *w (~ *u) and *wo.
>
>Why "immediately adjacent to a morpheme boundary" and nowhere else?

Because after a homomorphemic consonant, the palatalization
and labialization were carried over (and usually absorbed at
a later stage) by the consonant (*Ci > *Cj& > *C'& > *C& >
*Ce ; *Cu > *Cw& > *CW& [> *C& > *Ce]).

=======================
Miguel Carrasquer Vidal
mcv@...