On Mon, 08 Dec 2003 12:47:29 +0000, elmeras2000 <
jer@...> wrote:
>I still believe the system I arrived at has a very high degree of
>plausibility, although it may of course well be over-simplified. The
>older stages are only accessible via inspection of the traces they
>have left in the later language. I do not find myself in a position
>to evaluate Miguel's much more complicated (and, it did appear last
>time I took the trouble to go into it, less pricipled) solution.
I must of course protest against "much more complicated" (only slightly
so), and "less principled" (I don't think so).
In fact our two solutions are very similar in their basic principles:
jer mcv
person m / t / s m / t / s
stem w
number 0 / G / D 0 / ik / at
case 1 u
case 2 a (sg.) / i (du./pl)
case 3 m ma
Whether /w/ is a stem somehow meaning "person", or an ancient case marking
/u/ is academic. The small benefit my analysis gives me is that I can
explain the enclitic forms *me, *te, *se as derived from the other
(oblique) ancient case marker /a/ [I don't recall off-hand how Jens
explains them]. In the dual and plural, -u is in any case part of the stem
in both analyses.
The number markers are essentially equivalent, as is the acc(/gen) case
marker, except that my forms have built-in vowels.
One might argue that my choice of rect. vs. oblique du/.pl. markers (*-iku,
*-atu vs. obl. *-iki, *-ati) is less principled or more complicated
compared with Jens' soundlaw *-D > *-s or *-y depending on the preceding
vowel. On the other hand, I think it allows me to explain the "recessive"
variants *més and *swéy(-es) in an elegant manner (in the dual we only have
the "dominant" pattern *wéh1 < obl. and *yúh3 < rect.).
One aspect where I think Jens' theory is "less principled" than mine, is
the vocalization, where Jens needs a vowel /e/ in the 1st. person, but /u/
in the 2nd. person (non-oblique). Since the vowels are already built into
my algebra, I don't have that complication.
As to the required soundlaws, both Jens' and my sets are, as well as I can
determine, self-consistent and plausible in principle.
Of course Jens' analysis is cleaner with respect to the bone of contention
(what are the 2nd person plural forms about?) in that *y- and *w- simply
follow from the soundlaws (*twuD > *DwuD > *DuD > *jus, while *tweDmé >
*DweDmé > *weDmé > *usmé), while my analysis requires the irregular loss of
*t- in a sequence tu...tu (or tW..tW / cW..cW), and later the irregular
insertion of anti-hiatic *y- in the nominative. But I have my reasons for
sticking to my account and not adopting Jens' explanation (which, I might
maliciously add, only works because of the arbitrary ["non-principled"]
vocalization).
In short, there is not that much distance between the two theories.
=======================
Miguel Carrasquer Vidal
mcv@...