Re: 3rd Slavic palatalization

From: Grzegorz Jagodzinski
Message: 41107
Date: 2005-10-07

willemvermeer wrote:
> --- In cybalist@yahoogroups.com, "Grzegorz Jagodzinski"
> <grzegorj2000@...> wrote:
>
> [On the Progressive Palatalization]
>
>> But why this sound rule is so difficult? Because there are plenty of
>> proposed conditions (and counterexamples). We should answer the
>> following questions:
>
>
>> 1) how could have existed and functioned so multi-limited phonetic
> rule?
>
>
> In the traditional formulation it is a very ordinary phonetic rule,
> to wit a progressive palatalization of velars preceded by high front
> vowels unless followed by a consonant or a high rounded vowel.

The traditional formulation is a little different: the process occured after
a front vowel (including 'r.) and before a vowel, a more precise formulation
is that the process occured after i, I, e, perhaps 'r. (in the dialects
where such a sonant existed), and before a vowel. I believe the latter is
the original Jan Badouin de Courtenay's formulation (btw., the author was a
Pole despite of his French name). Such a formulation is commonly given in
the literature (e.g. Z. Stieber "Zarys gramatyki porównawczej je,zyków
sl/owian'skich" [Outline of comparative grammar of Slavic languages], all my
academic books on OCS, etc.). So, I am convinced that your formulation
cannot be termed "traditional".

Anyway, none of those three formulations describe a regular process, and
this fact is also emphasized in all sources I know.

> Note in this context that progressive palatalizations are less
> frequent than the common or garden type of palatalization of velars
> followed by front vowels. For that reason beginning linguists tend to
> be unaware of them and uneasy with them. Rules that strongly resemble
> the Slavic Progressive Palatalization have been reported for some
> Andi languages of the Caucasus, notably Tindi and Godoberi.

Oh, not only. In Old English a final k > ch after i, i:, final g > y after
any front vowels, and also sk > sh after any front vowel. Moreover, medial g
> y after a front vowel, if followed by another front vowel (note that OE
regressive palatalization of k, g took place only before i, i:).

A progressive palatalization occured also in Old French, where [k], [g] >
(IPA) [j] after a front vowel (including [a] which must have been front in
Gallo-Romance) like in doyen < de:ca:num, payer < pa:ca:re, OF lai < lacum
(the modern French <lac> is borrowed from Latin) or another OF lai < illa:c
(shortened in the modern French to <là> due to freq.), plaie < plagam,
flairer < flagrare. The French progressive palatalization occurred even
before back vowels and before some consonants.

A progressive palatalization is known from German as well where the "ich
laut" [ç] developed mainly after a front vowel.

>> 2) why analogy did not originally occur in 1st or 2nd
> palatalization but
>> appeared so strong in 3rd palatalization?
>
>
> It is easy to see why if you reconstruct the paradigms the way they
> looked immediately after palatalization had taken place.

OK, let's reconstruct. Let's take two Slavic words, *vIlkU 'wolf' (OCS
vlIkU) and *otIkU 'father', and let's apply the rule of the 3rd
palatalization in your shape. I will use OCS endings

WOLF (no 3rd palat.)
sg. N vlIkU, G vlIka, D vlIku, A vlIkU, I vlIkomI, L vlIce^, V vlIc^e
pl. NV vlIci, G vlIkU, D vlIkomU, A vlIky, I vlIky, L vlIce^xU
du. NAV vlIka, GL vlIku, DI vlIkoma

Despite of lack of 3rd palatalization here (because of no conditions), we
can observe the triple alternation k : c (loc.sg., loc.pl., nom.pl) : c^
(voc.sg.)

FATHER (hypothetical forms expected if the 3rd palatalization occured
strictly after your rule)
sg. N *otIkU, G otIca, D otIcu (*otIku), A *otIkU, I otIcemI, L *otIce^
(otIci), V otIc^e
pl. NV otIci, G *otIkU, D otIcemU, A *otIky, I *otIky, L *otIce^xU (otIcixU)
du. NAV otIca, GL otIcu (*otIku), DI otIcema

Real locative forms in brackets are analogic to the ones of soft-stemmed
declension.

> In all
> versions you get paradigms with extremely capricious alternations
> crying out for morphological levelling.

Extremelly capricious? No way, we can observe the same triple alternation k
: c : c^ as in the type without the 3rd palatalization, and the only
difference is a little changed distribution of k : c. And the number of
examples with -IkU and the like was rather large as -IkU became a diminutive
suffix, extremely popular in many Slavic languages Alternations are not a
problem since they have preserved in most modern Slavic languages for
hundreds of years. And in Polish they became even more complex, like in
Kozak 'Cossack' [k] - instrum.sg. Kozakiem [k'] - nom.pl. Kozacy [c] -
voc.sg. Kozacze [c^]. You can call this quadruple alteration capricious
because the number of examples is limited (because of using analogous
vocative forms without [c^]). Capricious but actual.

> And there are several
> contributing factors, such as the fact that the total number of
> morphemes involved was tiny and the fact that the regressive
> palatalizations were remained synchronically predictable for a long
> time after they had taken place, whereas the progressive
> palatalization became opaque almost immediately.

It is not the problem of opacity. You _assume_ that phonetic processes are
regular. You _assume_ that the 3rd palatalization was regular. And you
_conclude_ that it must have existed "regular" forms even if they are not
attested.

>> 3) what factors decided that analogy occured and what factors
> decided that
>> it did not, in particular instances?

> It always did, there are no extant examples of alternating paradigms.

It is not correct, the actual paradigm of otIcI has the alternation c : c^
(the latter in vocative sg.). And this model of alternation is unknown
outside the "tiny" group of substantives with the 3rd palatalization of the
final consonant. In other words, the 3rd palatalization caused the origin of
a *new model* of alternation, in such or another way (c^ in voc. sg., c
elsewhere). I see no reason for which the new model could not be different.

>> 4) what factors caused palatalization in a given word in some
> dialects but
>> not in other dialects?

> The factors that caused palatalization were exactly the same in all
> dialects, but the subsequent analogical levelling was carried through
> in different areas in different ways, for instance:

> (1) In iteratives the modified velar was generalized and even
> extended in South Slavic, but eliminated or virtually eliminated
> elsewhere.

We both agree that the South Slavic generalization was a result of analogy,
as OCS shows, as a rule, the "regular", original state. At the same time we
have no evidence that the forms with 3rd palatalization ever existed in West
and East Slavic. Which is more, expanding OCS as a language of church on
eastern Slavic area should prevent the original, changed forms (if they had
existed) from being eliminated. So, we have no evidence that such forms as
klicati, dvi3ati have ever existed outside South Slavic.

> (2) In a-stem nouns with a stem in *g, the modified velar was
> generalized everywhere except in the Russian/Belorussian area.

> (3) In all types of nouns, the modified velar was generalized if it
> was a *k.

It's a pity that you have not given any examples.

> A precise answer is not always possible because not all types of
> cases happen to be represented in the material, which, as I said in
> an earlier posting, has serious limitations.
>
> Then Grzegor wrote:
>
>> The answers are very simple: the 3rd, or progressive,
> palatalization, was a
>> tendency rather than rule.

> I'm not against tendencies on principle, although I can't help
> worrying a bit about their theoretical status, but let us recall that
> this discussion is not about the Progressve Palatalization per se,
> but about the question whether or not it is an _obvious_ example of
> an irregular or partial change. I should be clear by now that it is
> by no means an obvious example.

I disagree. What of it that some scholars still see here a regular change
since the process was so much affected by analogy and other "irregularizing"
factors. In this respect the 3rd palatalization differs from other Slavic
palatalizations, and this is the enough reason to call it "weak" change (or
tendency) while the other palatalizations - regular changes. I also state
that it is one of the best instances of "weak phonetic rules" (if you don't
like the word "tendency"). If you are not against tendencies, perhaps you
can point a better example which you will be able to call "obvious"?

The problem with tendencies is not in their theoretical status (we can
imagine how they rise) but in something else. If we agree that such "weak"
changes really occur, all the etymology will become less certain and less
scientific.

W. Man'czak has counted that ca. 30% forms in a text are simplified due to
frequency, modified by analogy, borrowed from dialects or irregular for
other reasons (sometimes unknown). Such countings are not very hard if you
have a Romance language - but we know the Proto-Romance language which was
Latin. The number of irregular forms in a dictionary is much less because
analogical and irregular forms are, as a rule, more frequent in texts.
Nevertheless they make a quite noticeable percentage of the lexicon. And if
we added "weak rules", the proportion of uncertainty would grow.

In fact, tendencies - or weak rules - appear to be known even if
(fortunately) not very frequent. Their weakness can also be various. For
example, the Greek change s- > h- was very regular except perhaps only one
example of sy:s 'pig, swine' - the alternative, regular form hy:s is also
known. A little bit less regular was rhotacism in Latin which left some
forms unchanged, like na:sus 'nose' (cf. also na:ris 'nostril', regular),
rosa 'rose' (we cannot say that inhibition by "r-" was regular, cf. ro:s,
gen. ro:ris 'dew', and even auro:ra < *auso:sa), miser 'poor', etc. Yet
weaker are various dissimilations, e.g. Lat. cavilla, taberna < *calvilla,
*traberna (*l, *r > 0), caeruleus < *caeluleus (typical dissimilation),
ro:ra:re, marmor, arbor etc. without dissimilation (even if the process was
alive in Romance: Sp. marmol, arbol). Sometimes it seems that two opposed
processes function at the same time, like in proto-Slavic *a- > je- ~ o- and
*e- > o- ~ je- (cf. Pol. jesiotr < *asetr- 'sturgeon', liter. jez.yna, dial.
oz.yna < *ez^- 'blackberry'). And finally, there are processes which spread
to single words only and thus we call them irregular (cf. Polish pokrzywa
'nettle' together with the town name Koprzywnica, Czech kopr^iva, Russ.
krapiva, Ukr. kropyva).

> GrzJ:
>
>> Of course we could multiply detailed conditions
>> to so large number that even all known instances would appear
> regular (but
>> the number of instances is finished even if it may be large). I do
> not think
>> this is the right way as it violates Occam's Razor. Instead we
> should seek
>> the simplest explanation, and it means we should agree that the 3rd
>> palatalization was a tendency (or a "weak change"), thus it was so
>> inconsequent and irregular because many factors were able to
> prevent that
>> process.
>
>
>
> Once again you make a caricature of the situation. I would urge you
> to read the literature more attentively and pay attention to what
> scholars have been saying and why.

Once again, all authors I know emphasize irregular character of the 3rd
palatalization. You should read their works as well.

"Trzecia palatalizacja nigdzie nie zachodzil/a bezwyja,tkowo" [the 3rd
palatalization occured unexceptionally anywhere] (Z. Stieber, "Zarys
gramatyki porównawczej je,zyków sl/owian'skich", § 63, p. 74).

"W przeciwien'stwie do palatalizacji I i II, palatalizacja progresywna
zaszl/a niekonsekwentnie" [Contrary to the 1st and the 2nd palatalizations,
the 3rd palatalization occurred inconsequently] (L. Moszyn'ski, "Wste,p do
filologii sl/owian'skiej", p. 205)

"Stan pos'wiadczony w je,zyku scs. oraz w innych je,zykach sl/owian'skich
wykazuje, z.e palatalizacja III nie zostal/a przeprowadzona konsekwentnie we
wszystkich pozycjach" [the state attested in OCS as well as in other Slavic
languages shows that the 3rd palatalization was not realized consequently in
all positions] (C. Bartula, "Podstawowe wiadomos'ci z gramatyki
staro-cerkiewno-sl/owian'skiej na tle porównawczym", p. 51)

And similarly in many more sources.

I hope you know these books as you are interested in Slavistics. So, please,
do not tell me that I should read the literature more attentively. Once
again, all authors emphasized irregular character of the process we are
talking about. I see no reason to disagree with them.

>
> I'd written:
>
>>> Unfortunately there is not a great amount of suitable material
>>> around. This is caused by the fact that the Progressive
>>> Palatalization is progressive: as a consequence most instances of
>>> velars potentially subject to it stand in stem-final position.
>>> Informative examples or counterexamples are very few and very far
>>> between. But they aren't completely absent either.
>
> Then GrzJ wrote:
>
>> Once again, the explanation that examples of stem-final consonants
> are not
>> informative because of the analogy - cannot be treated seriously.
> Other
>> palatalizations affected all consonants, also those stem-final
> ones, and
>> examples of such consonants are as good as any others.
>>
>> In other words: self-protection with analogy is only a desperate
> trial to
>> save neogrammarian ideas, nothing more. In order to admit that the
> 3rd
>> palatalization was a regular process we would have to know that it
>> occured in all instances where it should occur, and only next some
>> of those instances were retracted because of analogy. But we have
>> not such knowledge.
>
>
> It is normal methodology to regard cases that can easily be the
> consequence of analogical levelling as uninformative.

It would have been normal methodology if the number of exceptions hadn't
have exceeded the reasonable level.

>> Which is more, according to formulated rules, the instances like
> otIcI <
>> *otIkU should not show the 3rd palatalization (before -U). Why did
> it happen
>> in otIcI but not in numerous formations with -ikU?

> Because the *i in -ikU reflects earlier *ei, which does not trigger
> palatalization. This is standard knowledge. The examples are in the
> second edition of Meillet's "Le slave commun" (1934).

Sorry, it seems to me that you should also study some newer works. Have you
based only on works written a century ago?

Of course you are wrong. Together with masculine -ikU, it goes -ica in many
instances. Should I understand that -ikU < *-eiko- but -ica < *-i:ka:? It
looks highly improbable.

>> According to your
>> argumentation, analogy (as if) caused the 3rd palatalization to
> occur
>> despite of the presence of -U.

> No, first it is not "my" argumentation, but mainstream argumentation
> since 1910 or earlier.

The mainstream which was reflected in the literature printed a hundred years
ago... but which have been replaced with "inconsequence" in newer sources.

> But what is much more important, the formulation distorts the
> traditional view of what constitutes morphological analogy. It is not
> a matter of the Progressive Palatalization "occurring" in examples
> like *otIkU, but of the stem form *otIc-, which was found in such
> forms as Gsg *otIca, being subsequently (quite possibly several
> generations after the Progressive Palatalization had taken place)
> generalized to those cases in which the stem form *otIk- was
> phonologically regular, it is a matter of replacement resulting in
> simplification of the grammar.

Very strange indeed why such simplification did not comprise 1st and 2nd
palatalizations (as in vlIkU : vlIci : vlIc^e).

>
> Willem
>

Grzegorz J.



___________________________________________________________
To help you stay safe and secure online, we've developed the all new Yahoo! Security Centre. http://uk.security.yahoo.com