From: Grzegorz Jagodziński
Message: 46221
Date: 2006-09-28
----- Original Message -----
From: "Mate Kapović" <mkapovic@...>
To: <cybalist@yahoogroups.com>
Sent: Monday, September 11, 2006 2:32 PM
Subject: Re: [tied] Re: Slavic endings
> Are you sure that 3rd palatalization would indeed work here even if it
> were already *i? I ask this because there are man examples where we get
> double forms, so-called "exceptions" etc. Many people consider 3rd
> palatalization as "sporadic", but I think you once said that that is not
> really true...
>
> Mate
The problem of the sporadic character of the 3rd palatalization was already
discussed here. Anyway, I am still NOT convinced to any of the
Neo-Grammarians' ideas concerning to some phonetic processes, and I still
believe that the 3rd Slavic palatalization is one of the best example for
such "semi-regular" processes.
The classic formulation of the process is: k, g, x --> c', 3', s' after I
(=soft yer), i, eN (= nasal e), R' (= palatalized sonantic r), without
saying a word on the origin of the i or eN. Only then some obstinate
followers of neo-grammarian ideas tried to make the condition of the process
more restrictive in order to make the facts fitting to their insane views.
All this sounds like a kind of medieval scholastic, not science. But the
facts tells us that the process took place only sometimes, and undependently
on those restrictions formulated by those Neo-Grammarian fans.
Let's even assume that once it really was so that the process happened
unexceptionally everywhere when it "should" have occured, and next STRONG
analogy (stronger than in any other phonetic process in Slavic!) caused both
retracting the process in some instances and spreading the palatalized forms
in other instances. If yes, what use do we have from even best theory, hence
it does not allow us to foresee virtually nothing!? Such a theory has not
the right to exist at all, at least in science (it may exist of course in
scholastics, metaphysics, para-science, quasi-science etc.).
But let's return to the matter. They say that the effects of the 3rd
palatalization were the weakest in the NE part of the Slavic territory (see
the Novgorod form vIxU "all"), and the strongest in the southern part (esp.
in OCS).
It is not so, as it appears. In fact, Polish has many forms
with -k-, -g-, -x- preserved in these instances where there
are -c-, -3-, -s- in S Slavic lngs, but there are also inverse instances.
Let's go back to the famous example of Slavic _lice_ "face". Old Polish had
lice, now lico, so the palatalized form. Neo-Grammarians, the followers of
the restrictive formulation of the phonetic rule, have problem here as the
word comes probably from *leik-, not *li:k- (see Vasmer). Yet bigger
problems is caused by the Bulgarian and Macedonian dialectal form liko (not
quoted by Vasmer but see Wiesl/aw Borys', "Sl/ownik etymologiczny je,zyka
polskiego", p. 287). This state of affairs "should not" exist for at least
two reasons:
1) _liko_ would be expected in N Slavic rather than in S Slavic (there is
inversely),
2) the palatalization would not be expected here at all (if lice|liko indeed
from *leik-)
They also say that any instances of the 3rd palatalization after R' comes
from southern Slavic. It is false, either. Czech and probably Polish has
preserved at least one instance for the process R'k > R'c', namely in one of
the Slavic words for "mirror", *zIrkadlo. Slovene has preserved both zrcalo
and zrkalo, East Slavic lngs have zerkalo but Old Ruthenian had zercalo
(OCS-ism?). Slovak has only zrkadlo but Czech both zrcadlo and zrkadlo. Old
Polish had only wierciadl/o ~ wiarciadl/o (now zwierciadl/o, a little
bookish), the product of the old *ziercadl/o transformed by purely phonetic
(not semantic) analogy to wiercic' "to bore, to drill" (btw., another
example versus Neo-Grammarian ideas of exceptionless phonetic rules).
According to prof. Borys', the form *ziercadl/o is more probable than
**zierkadl/o in pre-written Old Polish.
Indeed, Old Polish "c" was palatalized and acoustically not very distant
from c' < t' (like in wiercic'). Which is more, 1 sg., Present Tense, is
_wierce,_ with the same sound "c" like in *zIrcadlo, not "c' ". If there had
not been *zIrcadlo but *zIrkadlo in Old Polish, there would not have been
any base for the analogy and for the further change of *ziercadl/o into
*wierciadl/o. The change was possible because the word-formation base of
this word, *zierkac' "to peep at sb." (< "look"), also changed irregularily
into zerkac'. And if really both *zierkac' and *ziercadl/o existed in Old
Polish, no explanation in Neo-Grammarian style is possible at all.
The only sensible explanation is that the 3rd Slavic palatalization was
really a sporadic process, and all quests for more and more suitable
conditions for the "unexceptional process" are nothing but wasting of time.
Grzegorz Jagodzin'ski
___________________________________________________________
All new Yahoo! Mail "The new Interface is stunning in its simplicity and ease of use." - PC Magazine
http://uk.docs.yahoo.com/nowyoucan.html