--- In cybalist@..., Miguel Carrasquer Vidal <mcv@...> wrote:
> I don't know if I have changed my mind on Baltic. The reason why I
> thought Baltic *sW > s was given above (the loc.pl.). Also the
> general laxness of Baltic in observing Pedersen's RUKI Law (e.g. the
> -e,s , -usio participle).What's the evidence again for RUKI in
> Baltic?
Why Pedersen's (with all my respect to him)? As for at least Slavic,
the rule is mentioned already in Miklosich's Etymologisches
Wцrterbuch der slavischen Sprachen (1886, by the way, it's in that
source where the sandhi explanation of *xod-:s^Id- appears for the
first time). Holger Pedersen was just 21 at the moment.
I agree that remarks on RUKI in Baltic in primers like Beekes'
('s^ after i,u,r,k (often s analogically [what makes him think so? -
S. T.] restored)')
and Semerйnyi
('A peculiarity of the eastern languages (Aryan, Slavic, and in part
Baltic) is a change of s to s^ after i u r k' and examples from
4.6.1),
as well as in such (a bit excentric and too 'global') monographs
like that of Gamqrelidze-Ivanov
('nuz^no polagat', c^to distribucija samoj spirantnoj nemarkirovannoj
fonemy *s xarakterizovalas' nekotorymi ogranic^enijami,
obuslovlennymi nalic^ijem nejtralizacii protivopostavlenija *s~*s^ v
opredel'onnyx pozicijax. V kac^estve takix pozicij nejtralizacii dl'a
protivopostavlenija *s~s^ vystupajut pozicii posle i,u,r,k, v kotoryx
dolz^en byl projavl'atsia imenno palatalizovannyj (kompaktnyj) c^len
fonemnoj oppozicii. Dal'nejs^eje razvitije sibil'antnyx spirantov v
jazykax gruppy satem legko objasnimo pri dopus^c^enii obs^c^ej
tendencii diffuznosti, a takz^e isxodnoj sxemy raspredelenija
sibil'antnyx fonem: v istoric^eskix satemnyx dialektax
postulirujemyje sibil'antnyje fonemy predstajut v vide diffuznogo
spiranta s, za iskl'uc^enijem pozicii posle i,u,r,k, gde oni dajut
kompaktnuju sibil'antnuju fonemu s', otraz^ajus^c^uju neposredstvenno
indojevropejskoje s^ (c^astic^no *s'O [so they introduce something
like sW as well, and even explain B-S 6 through it, though in general
treat it in a different manner than you - S. T.])' and examples, also
the scheme on p. 130 [PIE *s^ > Baltic s^ before r,k])
are not enough to clarify the matter. Nevertheless, in a more
specialized literature (first of all _Stang Chr. S.: Vergleichende
Grammatik der baltischen Sprachen, Oslo-Bergen-Tromsц,
Universitetsforlaget, 1966, 94-100_ and especially Karaliu_nas S. "K
voprosu ob i.-e. *s posle i,u v litovskom jazyke", Baltistica 1,
Vilnius, 1966, 113-126__, optionally _Hamp E. P.: "On IE *s after i,u
in Baltic", Baltistica 3, 7-11_) the folowing insights can be found:
1. *{r,k}s > s^ is normal and nearly doesn't have exceptions; I'm not
likely to copy a long and boring list of examples 'pro' (like Lith.
virs^u`s (Acc. sg. vir~s^u,)), since this is plausible, and waiting
of your counter-examples.
2. *{i,u}s > {s,s^} is more trickier. UI-part fof the RUKI fails in
the contexts {i,u}-sS and {i,u}s-S, where '-' is morphemic border and
S stand for any sonorant (of course some minor irregularities can be
found).
Your exapmle matches this Karaliu_nas-rule conditions: -us-jo-.
Have you changed your mind now?
Sergei