In the Contemporary Central Cassubian Dialect

LECHOSŁAW JOCZ

University of Leipzig

THEOPPOSITION */ʃ, ʒ/↔*/r̝/

IN THE CONTEMPORARY CENTRAL CASSUBIAN DIALECT

Keywords: phonology, continuants, Kassubian continuants, the Kassubian language, central Kassubian

Abstract

The article concerns the opposition */ʃ, ʒ/↔*/r̝/ in central Cassubian dialects. The existing literature does not answer the question whether the opposition has been retained. Descriptions of the continuants */ʃ, ӡ, r̝/ contradict one another and the disappearance of the vibration of */r̝/ is regarded, unjustifiably, as a phonological identification of */r̝/ with */ʃ, ʒ/. Even when synchronic differences are identified, the existing phonological interpretations are unsatisfactory. Contemporary central Cassubian data prove that the opposition continues to exist. */r̝/ is consistently realised as [ʂ, ʐ], while */ʃ, ʒ/ is realised optionally as [ʃ, ʒ] or [ʂ, ʐ]. The shift of the continuants */ʃ, ʒ/ towards clear palatalisation is most probably a result of the transfer [r̝] → [ʂ, ʐ].

The disapearance of the final t presents the greatest problem of all. It can hardly be ascribed to frequency (as in e.g. istis ‘is’ already in Middle High German, Yiddish iz; also MHG und(e) / unt > Yiddish un ‘and’) for the word is quite rare; neither is there any consonantal cluster to be held responsible for this reduction (cf. e.g. Yiddish mark ‘market’ < MHG mark(e)t, kunts ‘trick’ < kunst). Thus the only possibility seems the anticipatory voicing assimilation and the subsequent simpfilication of the geminate appearing in the combination with the definite article: *pázet dem / der / di
*[ˈpazǝd‿ d…] > [ˈpazǝ‿ d…]. This could be supported by such prefixed Yiddish forms as e.g. antdrémlt [anˈdʀɛml̩t] ‘dozed-off’, avékgebn [aˈvɛˌɡɛbm̩] ‘to give away’, ópbershtn [ˈɔˌbɛʀʃtn̩] ‘to brush off’ &c.

I think the formal development can be compared with that of ā́yu- n. *h2ok̑- in PAnat. *ʔṓku- ‘point(?)’ of its medial stop to *ʔokkú- ‘pointed’ > *ʔokú-.

yalıntık-tın kooš altun yartmak bolmıš täg ulatı beš-tin on y(i)g(i)rmi

ru zhi shuang jinqian ji wu shi ershi

如隻雙金錢及五十二十

“so wie von einem (durch Hinzufügen eines zweiten) zwei Goldbarren, weiter aus fünf (Münzen durch Hinzufügen) zehn (oder) zwanzig werden”.

Il vaut mieux rejeter du dossier la rasine *deh2- ‘decouper en parts’[1] car un élargissement inorganique *-p- est fort rare en indo-européen, comme de Vaan le concède («*-p- is rarely found as a root extension»). De plus, sémantiquement, on attend plutôt une racine ‘bouffer, dévorer’ qu’une racine ‘répartir’[2]: à preuve, les emplois du verbe δάπτω et la désignation i.-e. du ‘festinde magnificence’(i.-e. *gʰós-t-) formée sur la racine i.-e. *gʰes- ‘bouffer, ronger, dévorer’(av. √GAH-), et qui est à l’origine du dérivé secondaire *gʰós-t-o- ‘tablée’pluralisé en *gʰós-t-o-es, *-i-ns ‘hôtes’(Garnier 2013: 62).

m1 {sa} L1

m3 {sa}L2

m2 {sa} L2

Figure 1. Hybridisation of a phrase (or clause)

What makes the language of the manuscript heterogeneous is the distribution of the vocalisation signs standing, on the one hand, for e, i.e. tzere (ֵ) and the seldom used seghol (ֶ), and, on the other hand, for a, i.e. pattāḥ (ַ) and qāmātz (ָ). The use of these four vowel points is crucial in view of the description of the harmony shift. Labial ö and 'o were written in the same way, with the letters yodh and waw combined with a ḥōlām (word-initially supplemented with aleph), i.e. ‹יוֹ(א)›. Mutatis mutandis, labial ü and 'u were not distinguished in writing, either, both were written with the letters yodh and wawwith a shūrūq – word-initially introduced with the letter aleph, i.e. ‹יוּ(א)›.

Fig. 2. The distribution of macaroni /maccaroni (Ngram Viewer 1760-2000)

Searches for other possible word-forms in Google Books did not bring any results, so one has to state that OED3 includes an exhaustive range of variants for macaroni.

A0x0 [ʻz0ʼ] : Ai1x̄i1 [ʻz̄i1ʼ] Ai2x̄i2 [ʻz̄i2ʼ] … Aimx̄im [ʻz̄imʼ]

It is clear that the verb *ἴπτομαι (which is not attested in the present) meant ‘attack, hurt, inflict pain’.

Thirdly, a good question is what the phonological status of the palatalised consonants was (of , , ǵ, ĺ, ḿ, ń, ŕ, ś, , and also , ,,ṕ,, , notattested in the sample material) in the transitional period. Should these consonants be treated as separate phonemes or rather as allophones of their non-palatalised equivalents? If the latter is true, then another question arises: should they be classified as free-variant allophones (see the e ~ 'a alternation in the same phonetic environment) or rather as complementary ones (for they occur only in front of certain vowels)? Fortunately for us, however, we can leave the latter question to language theoreticians.

zu amhar. šəräṭ (ሽረጥ) und širṭ (ሺርጥ) „apron” (L., p. 253), das ← syr.-arab. šerṭa (ሽረጥ) „piece of cloth”

hergleitet wird, ist amhar. širəṭ nachzutragen;

This stage is attested in the manuscript mentioned above; in III-78 from 1750. On the one hand, it contains words in which the harmony shift did not occur or it occurred only in the final syllable (in a suffix; this is what we see in III-73), e.g. tügelטיוּגֵיל(243 vo), beslegenaבֵיסְלֵיגֵינְדָא(243 vo), körgüzgeninaכיוֹרְגיוּזְגֵינִינְדָא(243 vo), but, on the other hand, we find there words in which the e'a change took place in non-final syllables or/and in the stems themselves, too, cf. e.g. kĺajdĺarכְלַײְדְלַיר(244 ro), čearצֵיבָר(244 ro), bijančbylaבִײָנְצְבִילָא(244 ro). This means that further researches confirmed our initial conclusions.

Clauson’un ‘gentle, mild, humble’ anlamını vererek arkaik *altçak biçimini düşündüğü alçak kelimesi için Sevortyan al+çak > alçak (ESTY: 143-144), Banguoğlu ise aluçak > alçak (TG: 110) etimolojilerini yaparlar. Röhrborn, Sevortyan’ın görüşnü ‘?’ işaretiyle aktarırken (UW: 92); Nişanyan, kelimeyi *al- ‘yumuşamak’ fiiliyle ilgili görür (ÇTES: 16). Tietze, ‘burnu havada olmayan, üstünlük taslamayan’ şekilinde anlamlandırdığı kelimenin, sıfat yapan {-(I)k} ekiyle ilişkisini düşünse (TETL: 71-72) de Räsänen kelimeyi al ‘das Untere’ kelimesinden türetir. Gülensoy da *al ‘alt, aşağı’ + çak görüşündedir. Starostin ve ekibi ise, kelimeyle ilgili, *ale ‘below, lower’ maddesinde Türkçe *ăl, Japonca *ǝ̀rǝ́-, Korece *àrái biçimlerini tasarlayarak Sevortyan ve Räsänen’in görüşünü aktarırlar (EDAL: 285-286).

Vine’s proposal (2012a) that *-e-, when unaccented according to the PIE mobile accentuation rule, was raised to *-i- before -i̯V-, in order to explain forms such as causative sōpiō ‘cause to sleep’ (< *su̯ṓp-ei̯e-) and denominative serviō ‘wait on’ (< *seru̯e-i̯é-), may be regarded as a very early type of vowel reduction.

NNS subcorpus[3] / NS corpus
lexical item / number of occurrences / frequency per 1000 words / number of occurrences / frequency per 1000 words
crush / 0 / 0 / 12 / 1.5
girlfriend / 34 / 2.9 / 3 / 0.4
love / 47 / 4 / 7 / 0.9

Kent (1932: 66) likewise assumes that the change was internally motivated. Ballester (1990: 38–39) also, based on his own accentual rules for Pre-Literary Latin (see §3.5), illustrates the process as *dḗperī̀redḕperī́re ‘to perish’. Leumann (1977: 248) elaborates the mechanism, by assuming that secondary accent first occurred in heavy antepenults, and then replaced initial primary accent in terms of status: *sápièntiasàpiéntia.

In the Proto-Tocharian period the palatal consonants developed before front vowels (mainly before e, less frequently before ä, i, Ï), just as in Uralic, Turkic and other languages of northern, north-eastern and Central Asia. As a result of later phonetic processes (reductions of vowels and other changes) in the historical period the correlation of palatalisation was phonologised. As far as the Indo-European languages are concerned, this phenomenon is known only in the Balto-Slavic languages which are suspected of Finno-Ugric influence: y TB  + E, : TA wkä, TB yaknePIE *eh-no-; TA want, TB yente ‘wind’ < PIE *ēnt-. As has been pointed out by R. Jakobson (1931), palatal consonants are known in most of the languages of Northern Eurasia (Uralic, Altaic, Chukchi-Kamchadal languages).

Figure 3: Dynamic space of the Praeteritum

Fig. 4. A sample entry from Calepino’s 1590 edition

[1]Bien reflétée en grec par δη-μός m. *‘division territoriale’(< i.-e. *deh2-mó-), et, sous une forme élargie *deh2-i- «partager, répartir », par l’hom. δαί-νῡμι ‘servir’ vs δαί-νῠμαι ‘être reçu’, par l’hom. δαῖ-τες f. pl. ‘festin’ (<
*déh2-i-t-es), et par le nom d’agent *δαί-τωρ (reflété par l’anthroponyme hom. Δαίτωρ) ‘écuyer tranchant’(< *déh2-i-tor-), qui se renouvelle dès Homère en un doublet thématique δαιτρός.

[2]Le rattachement du hitt. LUtappala- ‘cuisinier royal’admis par Matasović (2010: 6) est douteux.

[3]

Nişanyan, çevrimiçi etimolojik sözlüğünde daha ayrıntılı bir etimoloji yapmaktadır: alçak ETüalçak/alşakETüalış-‘yumuşamak’+Ak (ÇTES-@).