Re: [tied] Some accentological thoughts...

On Tue, 27 Feb 2007 00:44:14 +0100 (CET), Mate Kapović
<mkapovic@...> wrote:

>On Pon, veljača 26, 2007 10:35 pm, Miguel Carrasquer Vidal reče:
>> The thing is that if I look at the (admittedly incomplete)
>> inventory in Derksen's list, there are 13 non-neuter
>> polysyllabic a.p. b nouns. It's a small sample, but it must
>> be significant that 11(!) of them have a yer in the first or
>> second syllable (arImU/arImo, orIlU, ostInU, osIlU, otIcI,
>> ovInU, ovIsU, pęsUkU, pI(c/k)UlU/pI(c/k)Ulo, bIc^ela,
>> vIdova).
>
>Sure, but what many adjectives in *-enU, *-elU, *-okU like Croat. ze`len,
>zele`na, zele`no, de`beo, debe`la, debe`lo, viso`k, viso`ka, viso`ko etc.
>There are tons of adjectives like this.
>Of course, these are not the only examples, there is a fairly common type
>pokróv, pokrová in East Slavic, cf. also Croat. dial. i`stok, isto`ka
>"east", toponym Za`gvozd, Zagvo`zda etc.

These are not problematical for my theory. We can have final
accentuation throughout (zelenÚ [> zelénU], zelenó, zelená),
or, in cases where a prefix is involved, we can have initial
accentuation, shifted right by Dybo's law (po-króvU,
po-króva: > pokróvU [c.q. pokrovÚ > pokróvU], pokrová).

In forms involving a suffix (such as *-enU, *-elU, *-okU),
all it takes is for the suffix to be dominant to produce
this pattern. As I have stated on previous occasions, I
think that "dominant" simply means "stressed" (in PIE). In
principle, dominant suffixes like the ones mentioned above
could in PIE have had the stress on either syllable: *-énos
or *-enós. Balto-Slavic seems to make no distinction between
the two types: we always have -énas, -élas in Lithuanian, we
have -énU, -ená; -élU, -elá in Slavic (except when there's
an acute: -i"nU, -i"na etc.) Slavic end-stress can be
original, or it can be due to Dybo's law. I think it's
original (the stress in -i"na is then simply due to Hirt's
law).

>> So what do you think of 3sg. -e: as a possible explanation
>> of thematic lengthening?
>
>I didn't quite get the argument but long -e: is a *-ě in Slavic. *-e is
>supposed to be short. The whole deal with the length in final open
>syllables is fishy since you cannot get any general rules, you have to
>suppose a lot of analogies in every possible theory.

The general rule is that the endings which were stressed in
a.p. c were lengthened. This happened after the original
long vowels had acquired different quality (which opened the
door for long vowel shortenings and short vowel
lengthenings). The transfer of this length to a.p. a and b
was of course analogical, so I would expect some degree of
irregularity in the overall reflexes.

>Thematic *-e- was lengthened in some Slavic dialects, perhaps only localy
>in a later period, perhaps already in Common Slavic period (after the
>re-establishment of distinctive length on all new vowels) in the stretch
>from Posavina till present-day Slovakia and Czech Republic.

It's also attested in Old Polish. It's attested in Modern
Polish, if you count ja biorę, oni biorą... I don't think
there's a way to tell if it once occurred in East Slavic as
well. Well, perhaps Nikolaev can find it :-)

=======================
Miguel Carrasquer Vidal
miguelc@...