ACTA LINGUISTICA LITHUANICA

LIETUVIŲ KALBOTYROS KLAUSIMAI XLIII (2000)

Kazimieras GARŠVA

Lietuvių kalbos institutas, Vilnius

THE “MURMURED VOWELS” OF THE NORTHWESTERN PANEVEZYS DIALECT

The northern Panevėžys dialects can be divided into four parts: the largest, central Northwestern Panevėžys subdialect (covering the Pakruojis and Pasvalys districts) and the peripheral Biržai, Panevėžys, and Joniškis varieties. The Northwestern Panevėžys subdialect is distinguished by three phonetic features: the so-called retraction of stress, complete reduction of former short ending vowels, and the change of vowels i, u to ę, ǫ (except in the position of primary stress), as well as of ę, ą to ę̄, ǭ, and of circumflex ām, ān, ēm, ēn to ōm, ōn, ēm, ēn (cf. Garšva 1977: 76–77; Urbanavičiūtė-Markevičienė 1994: 85–86).

The subdialect boundaries are indicated on the map.

Fig. 1. Area of the Northwestern Panevėžys subdialect

165

1. K. Jaunius’s “Short Vowels”

The archive of the Lygumai parish shows that the vowels of the Northwestern Panevėžys subdialect were already unquestionably reduced in the second half of the 18th century. Wiejas / Wiejs was recorded in 1764, and Peledziuns in 1765 and 1766, etc. From 1777 to 1781, approximately 10 proper names with reduced short vowels in closed endings were recorded annually in the parish, from various (including considerably more distant) locations (Moteiuns, Jokans, Preymans, Szlondans, Stankuns, Szymkuns, Balczuns, Bajors, Maczuns, Motiejuns, Walantyns, Waytekuns, Zaks); from 1782 to 1785, the annual number ranged from 2 to 6, so these spellings cannot be considered accidental. The form Rekšs, “Regžtys,” was also recorded in 1781; it shows that even then the scribe perceived the final consonant cluster ts as the consonant c (without an intervening “murmured vowel”).

On August 28, 1860, in Lauksodis, the priest P. Gomalevskis carefully prepared “Kniga kurioj ira surasziti wisi brolej ir seseris brostwinninkaj prigulitis ir dobrawalnaj isiraszusius ing brotswa blajwistes…” The names of 866 men and 906 women, as well as family surnames, are recorded here. Surnames with reduced endings make up 16%, and given names 30%—almost twice as many. Most are given names and surnames with a reduced short vowel a, especially in three- and four-syllable words (Kurliuks, Morkuns, Marcions, Nagions, Wilips, Lukusziuns, etc.), as well as with the vowel i (Laszins, Nieks, Noreiks) and u (Burks). The form Ferens, “Ferensas,” shows that the scribe perceived the cluster of two consonants ss as a single consonant. People from the same village, and sometimes even the same family, were recorded both with the short final vowel retained and with it reduced (children more often than parents): Jonas / Jons, Motiejis, “Motiejus,” / Motiejis, etc. Thus, even at that time the short final vowel had no phonological significance.

In 1929, Kazimieras Gasparavičius recorded folklore texts in the village of Puodžiūnai (8 km south of Lauksodis) in which not 30 percent but about 50 percent of short endings had been reduced. It may be assumed that over 70 years, 20 percent of short endings were reduced in the central part of the subdialect, and that this process accelerated, since over the following 30 years another 40 percent or so of endings were reduced. In 1960, E. Grinaveckienė found very few short vowels in the village of Steigviliai (1 km from Puodžiūnai toward Lauksodis), but, based on consonant and vowel features, she noted that here “the loss of word-final vowels […] is a relatively recent phenomenon” (Grinaveckienė 1961:168).

Judging by the reported rates of reduction, it may have been transmitted to the Lithuanians by the Semigallians in the 13th–14th centuries, but it had little effect until the 19th century.

The “murmured” (“short”) vowels of the Northwestern Panevėžys subdialect were first described in 1898 by the Samogitian Kazimieras Jaunius (cf. 1898 (1970): 127). Drawing on data collected in 1892 in the peripheral Northwestern Panevėžys varieties of Pãmpėnai and Pãsvalys, as well as on Eduardas Volteris, he generalized the “short” ending vowels to many words in the subdialect. At the end of the 19th century, remnants of short vowels were pronounced along the Northwestern Panevėžys border with the Biržai dialect—

166

more often than now, but not as systematically as K. Jaunius indicated. Having surveyed few dialects and possessing few accurate texts, he did not notice that the same word could be pronounced both with a “short” vowel (a facultative variant) and without it.

A text from Pāmpėnai in Antanas Baranauskas’s collection was recorded a year earlier than K. Jaunius’s, but in it only 20% of the remnants of ending vowels remained¹, while in a text from the Pasvalys area the figure was 1% (Specht 1920: 162–164, 171–174). Petras Būtenas, born in 1896 near Gružiai, wrote in 1932: “although a kymograph might still sometimes record remnants of word endings (emphasis everywhere in the original – K. G.), they are almost imperceptible to the human ear” (Būtenas 1932: 169).

In 1936, the aforementioned linguist K. Gasparavičius, born in 1908 in Puodžiūnai, 10 km north of Linkuva, noted that ending vowels “turn into very short little sounds and may later disappear altogether”², having recorded about half of the ending-vowel remnants in a collection of accent-marked folklore (1929: 2–123) and characterized the dialect on the basis of the speech of informants born in the mid-nineteenth century (texts by people born in the second half of the previous century were mostly recorded without “murmured vowels” by later researchers as well). K. Gasparavičius also added in 1936 that those “barely audible whispering sounds […] are only still pronounced slowly by older people.”

Dabar teigiama panašiai (plg. Zinkevičius 1978: 60), nors per 40 metų seniais pavirto jaunimas, pagal K. Gasparavičių „murmamojo balsio“ neturėjęs.

K. Jaunius’s new theory had no followers for nearly 70 years (during that period, remnants of short vowels were more frequent than they are now). These vowels were rejected not only by A. Baranauskas (1898: 76), who published his description in the same year as K. Jaunius, but also by K. Būga (1924: 82–83), K. Gasparavičius (1929), A. Salys (1931, cf. footnote 2), and P. Būtenas (1932: 168, 169), who was born in Dovydai village in the Gružiai area. K. Būga also relied on him as a professional Lithuanian linguist with a good command of his native dialect, with the exception of J. Gerulis (1930: 41, 61–65). This feature was not accepted at the beginning of the second half of the twentieth century either, when more detailed works on the dialect began to appear (Jašinskaitė 1957: 189–194; Grinaveckienė 1961: 167–183; 1962: 147–168; Šukys 1962: 176–183, etc.). The same view was maintained in student theses written at the time and in A. Salys’s description “Lietuvių kalbos tarmės,” prepared in 1935 and published in exile (1992: 304–305). “Short,” completely reduced vowels had also existed about 300 years before K. Jaunius, but no one marked them; cf. Petrošiūns (1600), Gulbins (1678)³.

¹About half of the words have these remnants only in absolute word-final position after a palatalized consonant, for example, pirtì, mažì, and so on. ²Antanas Salys made a similar point in a lecture on 1931-12-01 (K. Gasparavičius’s notes in the manuscript collection of the LLTI library). ³Joniškis baptism register, p. 10^b; Linkuva baptism register, p. 17^a.

167

2. “Murmured vowels”

The concept (term) “murmured vowel” was revived and disseminated from 1966 onward (Girdenis, Zinkevičius 1966: 143, cf. LDS 1999: 126, 130, 142–145; Girdenis 1981: 61–62, etc.). Authoritative dialectologists, having described the new classification of Lithuanian dialects as an improved version of A. Baranauskas’s classification, rejected his view concerning the short vowels of the Panevėžys dialect, even though, as an Eastern Aukštaitian, he had investigated the matter more deeply than the Samogitian K. Jaunius (whose dialect classification linguists rejected).

E. Grinaveckienė had already described the short vowels and the consonants adjacent to them quite accurately (1961: 168). The chrestomathy “Lietuvių kalbos tarmės” (LKT: 33, 35, 237–259), prepared by the Institute of the Lithuanian Language and Literature, noted that most Eastern Aukštaitians who retract the stress drop ending vowels entirely, and these were not artificially marked in the texts. But following the publication of reviews (Girdenis 1971: 206–208; Zinkevičius 1975: 85–86; 1976: 129, etc.), the phonetics volume of the “Atlas of the Lithuanian Language” (LKA 1982: 18, 22–26) acknowledged that northern dialects generally drop the short vowel at the end of a word, except in three types of consonant clusters, where “certain traces of a murmured sound or breath” may occur. In the “Encyclopedia of the Lithuanian Language” (LKE: 459–461), an unpronounced “murmured sound” is already marked without exception.

J. Šliavas, who believed that “murmured vowels” in practical speech “are still noticeable almost only when voiced and voiceless consonants meet at the end of a word” (1972: 399), began marking those “vowels” in 1975 (cf. also Zinkevičius 1976: 129). Much confusion about the “murmured vowel” arose from imposing the rules of other dialects on the Panevėžys dialect, adopting a one-sided theory, conflating an unpronounced “vowel” with remnants of ending vowels, confusing phonology with phonetics, understanding phonology in different ways, conducting experimental commutation, and relying on dialect speakers of different ages.

Linguists are usually divided into those who do not distinguish phenomena associated with the “murmured vowel” (cf. I. Jašinskaitė, E. Grinaveckienė, J. Šukys, etc. – see Zinkevičius 1975: 84⁴) and those who do. In fact, the latter group further divides into three: those who explain the aforementioned differences in word endings 1) phonetically and phonologically (G. Kačiuškienė, I. Ručinskaitė), 2) phonologically only (A. Girdenis, Z. Zinkevičius), and 3) prosodically, that is, non-segmentally, assigning no greater significance to these phenomena and treating them only as potentially available auxiliary means of the phonological system (the earlier interpretation; cf. Garšva 1984: 128–137). Phonetically, one can speak of “murmured vowels”; phonologically, of a “murmured vowel.”

The first view among those who distinguish “murmured vowels” is the least reliable; it has not even been demonstrated experimentally. Of the 14 word pairs examined by G. Kačiuškienė (1983: 29), only 4 (29%) could show a “murmured vowel” (10 differ in primary versus retracted stress), but even those 4 “pairs” at least in the western part of the dialect

168

in this part, first of all, differ in the voicing–softness of the penultimate consonant in the clusters ks’ – k’s’, ps’ – p’s’. Thus, these words too do not form minimal pairs and cannot prove the existence of any “murmured vowel”.

After much broader auditory experiments were conducted with three pairs of the type lipės – lipsi (lėp’s’) under more natural conditions (repeating each once, without emphasizing anything), 55–48% of the pairs were distinguished (n=191; Garšva 1984: 132–135). Under artificial conditions (repeating the pairs as many as five times), 64–58% of the pairs were distinguished (n=305; Bielskytė 1984: 18–20). The latter result was obtained not at once, but after an additional experiment: the earlier listeners (n=215) distinguished only 58–48% of the pairs. Thus, even auditory experiments do not prove the existence of a “murmured vowel.”

Words pešiu – peši (pėš’) were distinguished under natural conditions by 41% of listeners (and by 23% of those tested using tests), whereas in I. Ručinskaitė’s (1984: 73) experiment, the difference between similar homophones rose to 77.7%. It is well known that the results of auditory experiments tend to be “better” the more artificially the experiment is conducted and when the researcher wants to “accept with justification the hypothesis formulated at the beginning of the study” (forgetting that the refutation of that hypothesis is equally valuable to science). Until at least some pairs are distinguished in those experiments (e.g., tò rito kàs’ – tò dabà kàs’, vaike gèr’ – vànden’ gèr’ – piėno gèr’, etc.), nothing of greater importance will have been proved. Pairs of the Pešiu – peši type could hardly be distinguished by “murmured vowels” of “two qualitative types,” since those vowels had often merged even before final reduction; compare: k’āli “I raise,” dòmis, gaspadòris, mėlynczi (Specht 1920: 85, 147, 164, 173), lėnki, vėrki, kòjis (‘<a), nòr, prasėkėr’s’, lýg’s (‘<u), and so on (Gasparavičius 1929: 25–92; see also Bilevičius 1980: 6, etc.). In discussing homonymous forms of the nėš’ type (present tense, 2nd person singular; future tense, 1st and 2nd person singular), K. Gasparavičius (1936: 2) and J. Bilevičius⁴ rightly pointed out that their distinction can be detected only syntactically.

The second (phonological) interpretation is theoretically quite interesting, but the Panevėžys dialect is viewed largely from the perspective of the North Žemaitian and other dialects, and of the standard language, drawing on historical grammar and phonological theory. It is argued, for example, that in the word blógs the consonant g does not become voiceless because a usually unpronounced and inaudible “murmured vowel” is theoretically posited after it, and that this vowel is needed because g does not become voiceless. But balts, viens, libdams, alkans, aug, duod and the like are also said by Šiauliai speakers, Latvians, Kaunas speakers (the “striukiai” of Lithuania Minor), and Southern Žemaitians (cf. Zinkevičius 1978:34, 36, 91, 94, 100, 154), whereas a “murmured vowel” is posited only for the Northern Panevėžys dialect.

The area of ending reduction extended from Riga to Tilžė. Northern Šiauliai speakers say not only sauss (like Latvians), but also káušs, daržs, lėds (or làds; Zinkevičius 1978: 38), and so on. It is no surprise that the Panevėžys dialect, too, preserved final consonant clusters b, d, g, ž, š, s+s, especially since the vowel of the ending was gradually and finally reduced over the last 100 years. The aforementioned consonantal features of the word

⁴ 1982 06 10 and subsequent information.

169

are explained in detail by the preserved voicing–voicelessness opposition, the cluster of two consonants ss, and so on; therefore, there is no point in marking the same feature with an additional phoneme (“murmured vowel”).

The greatest flaw of this interpretation is that, practically everywhere and without exception, an attempt is made to represent the standard language—even in the infinitive and other verb forms (e.g., KačiUškienė 1984: 15¹⁵), where K. Jaunius (1898 (1970): 115–160) heard no “murmured vowel”: láistet’ ‘to water’, vălgem’ ‘we ate’, and so on (over 90 years, even the former remnants of reduced ending vowels have almost entirely disappeared). There is no reliable evidence as to which consonant clusters or stem-vowel features indicate that a “murmured vowel” has survived, for example, in the words lėken ‘slender, slenderly’, rág’n ‘I wind, he/she winds’, šárk ‘pike’, táms ‘dark’, and so on. Z. Zinkevičius rightly writes that “young people and even the middle generation often do not pronounce the murmured sound” (1978: 60). Northern Panevėžys speakers have completely levelled that sound; in many places it can no longer be heard (Zinkevičius 2000: 8–9). And “sound features that are inaudible, imperceptible to hearing, and indistinguishable cannot be phonological elements of a language—they simply do not exist in the language system” (cf. Girdenis 1981: 189). Unpronounced and inaudible sounds need not be marked in transcription in any way (with an apostrophe, etc.).

The third interpretation arose in an effort to account more appropriately for possible differences between minimal pairs (without a fictitious “murmured vowel”). The prosodic interpretation has faced six counterarguments (Girdenis 1981: 61–62; KačiUškienė 1983: 33–34; Ručinskaitė 1984: 76).

One of them would help reformulate the hypothesis of two syllable types. According to it, Type I and Type II syllables differ in that, when the former occur before an unreduced vowel, no new (voiceless) consonant is added to them in any form of the same word; that is, the syllable structure does not change. Syllables of the two types consist of four kinds of sounds: long vowels (e.g., Type I vóks ‘vókas’, cf. gen. vó-ka, and so on – Type II vóks ‘vógs’, cf. vóg-dava, vó-k-sem’, though also văgem’, vó-ge); diphthongs (laūks ‘laūkas’, laū-ka – laūk, laūk-šem’); mixed diphthongs (v’ŏlks ‘vilkas’, v’ŏl-ka – v’ŏlks ‘vilk’s’, v’ŏlg-dava, and so on); and short vowels (lęp’s ‘lipės (tails)’, li.-pės’ –lęp’s, li.p’-kèm’, and so on). If the two syllable types are not accepted, one can posit a “murmured vowel” at the level of auxiliary phonological elements, or two further secondary accents with restricted distribution. In long syllables, these would occur only in the position of the main stress (cf. Garšva 1977: 122–128; KačiUškienė 1984: 7–8), and in short syllables—in the position of retracted stress (in final syllables). Otherwise, two quantities of diphthongs, long vowels, and short vowels would have to be marked in those positions.

Prosodic syllable type and the segmentally pronounced remnant of an ending vowel were and are independent phenomena; they do not contradict one another. Syllable type was conditioned by whether the syllable was formerly open (somewhat longer) or closed (shorter); the vowel remnant is pronounced exceptionally rarely and, in connected text, usually does not alter the information either.

170

A substantial article by A. Girdenis and G. Židonytė, “The Vowel System of the Northern Panevėžys Subdialects (the Rozalimas Dialect)” (Girdenis, Židonytė 1994), has been prepared. It illustrates vowel distribution and attempts to integrate the dialect into the systems of other dialects. An attempt is made to revive K. Jaunius’s view that in the position of weak (retracted or ancient final) stress, vowels of positional length e, a are “a hair’s breadth” shorter than long vowels.

Jau anksčiau buvo nurodyta (plg. Garšva 1977, 1984, 1986 ir t. t.), kad ne tik Rozalimo, bet ir Pakruojo, Pašvitinio, Linkuvõs, Joniškėlio, Pasvalio ir kitų vietų gyventojai jokių trumpųjų balsių nemurma, poras dù.r’s’ „dursi“ – dù.r’s’ „durys“, soṇ.ks’ „sunksi“ – soṇ’k’s’ „sunkūs“, már’ges’ „Margės“ – már’ges’ „margysi“, n’à.š „nešu“ – n’àš „neš“, kàps „kapsū“ – kà.ps „kapūs“, cèp’s’ „cipės“ (kalėjimus) – cèp’s’ „cipsi“ retkarčiais skirdami ne pagal „murmamuosius balsius“, o pagal kirčiuoto balsio ilgumą, priebalsių kietumą – minkštumą, skiemens ribą ir t. t. Dirbtinėmis sąlygomis (eksperimentuose) retkarčiais maksimaliai galbūt įžiūrimus skirtumėlius buvo siūlyta laikyti fonologinės sistemos pagalbinėmis priemonėmis ir variantais.

Of 100 informants, three seem still to have retained a trace of a short vowel in a closed ending (soṇ.’k’k’s’ “heavy”). In the central parts of the dialect, traces of such vowels are still pronounced in about 1–3% of cases, and in the peripheral areas in up to 70%; therefore, there is no need to look for murmured vowels in nineteenth-century cemetery inscriptions and use them to draw conclusions about the present-day phonological system (Girdenis, Židonytė 1994: 115).

The newly introduced, highly reduced, colorless front- and back-series vowels b, ъ are marked “only when the precise vowel quality cannot be determined” (Skirmantas, Girdenis 1998: 113). When reduction is especially pronounced—not only does the vowel quality change, but the vowel also fails to form a syllable—the symbol is additionally placed in parentheses. It is unclear whether anyone could demonstrate differences between the “murmured vowels” in the forms žmón’n’s’ “people” and v’ė˙(i)b’s “wind,” kò.k’b “what kind” and v’òšt(ə) “hen,” v’èr.gdamə “while working as a slave” and apsẹkabi.n’əs “having embraced one another” (Skirmantas, Girdenis 1998: 112–114, 120). The fantastic form kàkəl(ə) “neck” is still marked, although the vowel was inserted precisely because there is no vowel after -l. The numerous contradictions in examples of “murmured vowels” show that, lacking clear criteria for these “vowels,” even their proponents became confused.

3. Summary

Church archive records show that endings in the northwestern Panevėžys subdialect underwent slight reduction in the seventeenth century. By the beginning of the twentieth century, many traces of short ending vowels remained. In A. Baranauskas’s texts, more short vowels were retained closer to the stressed syllable, in formerly stressed endings, in open syllables, and after a palatalized consonant (Kačiuškienė 1983: 28). When stress was retracted and the diphthongs am, an, em, en were narrowed, the ending vowels had not yet been reduced and formed a separate syllable; cf.: álkans “hungry,” n’agàl “cannot—

171

—cannot” (retracted stress always emphasizes the second component of the old diphthong; cf. n’akal.t “innocent”).

Some traces of short vowels are still pronounced today (especially in peripheral parts of the subdialect). They show that the unstressed ending vowels were reduced first; cf.: laũ.ks “field,” ló.u “bed,” but ròn.kas “hands,” šaka “branch” (Norvaišiai, Mikniūnai, Lygumai). Before vowel reduction, all ending vowels after a palatalized consonant had merged into a single vowel, i. This reflex remains in A. Baranauskas’s texts, in the peripheral areas of the dialect, and in songs; e.g., karā.li “kings,” vėlnis “devil.”

Word-final consonants are palatalized before the reduced vowels i, e or the fronted u, a (>i) and after front vowels or palatalized consonants: gàl’ “may,” galì “may,” kár’ves “cows,” stàːč’s “upright.” Exceptions are consonants following back vowels a, o, ɔ (išeinam, ži.not, gai.l’əs “I regret”) and following front vowels i, e in the plural dative and instrumental (naktím “to/with the nights”), and in the second person of the future tense of verbs (sidės “will sit”).

At least twelve better-known linguists (A. Baranauskas; J. Bilevičius 1978; 1999: 95; K. Būga; P. Būtėnas; A. Butkus 1989: 56–60; K. Garšva; K. Gasparavičius; E. Grinaveckienė; I. Jašinskaitė; A. Salys; J. Šukys; S. Tumėnas 1987: 171–175, cf. LDS 1999: 328–330) marked traces of short vowels only where they were actually audible.

Perhaps seven linguists held a different view (J. Gerullis, A. Girdenis, V. Grinaveckis, K. Jaunius, G. Kačiūškienė, I. Ručinskaitė, Z. Zinkevičius). If we compared only linguists who grew up in the Panevėžys dialect area, the difference would be even more pronounced—eight to two. It is therefore inappropriate for dialectology textbooks, encyclopedias, and lectures to present only the second view while completely omitting the first, especially since the linguists in the second group do not entirely agree among themselves either: some insist that “murmured vowels” should represent all short ending vowels of the standard language, while others acknowledged exceptions.

An interpretation using auxiliary phonological notation for syllables with reduced ending vowels may still have a certain role. It could help linguists holding various views reach a compromise. From Kazimieras Jaunius’s 1898 (1970: 127) formulation—“speakers of other dialects think that the shortest vowels of the Panevėžys dialect are not pronounced at all, but […] the existence of a vocalic element […], when one listens attentively, is easy to hear” (that is, without listening specially, the sound is barely noticeable)—it is clear that even almost 100 years ago the “murmured vowel” had no more than auxiliary significance.

While viewing with caution K. Jaunius’s account of the five-way vowel quantity distinction in the Panevėžys dialect and of certain pitch accents, we cannot blindly accept the “short vowels” either.

“Murmured vowels” are now generally absent in most of the northern Panevėžys dialect area. These vowels should be marked only when they are actually pronounced. This was also the practice when recording and publishing dialect texts (cf. Bilevičius 1980; JAŠT; LRD).

172

ABBREVIATIONS

APh – Archivum Philologicum 1–8, Kaunas, 1930–1939.

JAŠT – Joniškėlio apylinkių šnektos tekstai, sudarė K. Garšva, Vilnius, 1981.

LDS – Lithuanian Dialectology Readings 1, compiled by Z. Babickienė, B. Jasiūnaitė, Vilnius: Department of General Linguistics, Vilnius University, 1999.

LKA – Atlas of the Lithuanian Language 2, Phonetics, Vilnius: Mokslas, 1982.

LKE – Encyclopedia of the Lithuanian Language, prepared by K. Morkūnas, Vilnius: Institute of Science and Encyclopaedia Publishing, 1999.

LKI – Institute of the Lithuanian Language.

LKT – Lietuvių kalbos tarmės. Chrestomatija, Vilnius: Mintis, 1970.

LLTI – Institute of Lithuanian Literature and Folklore.

LRD – I Was Pulling Flax, I Was Singing, compiled by B. Stundžienė, Z. Puteikienė, K. Garšva, Vilnius: Vaga, 1990.

MB – Scientific Library.

LTR – Lithuanian Folklore Manuscript Repository.

ŠPU – Šiauliai Pedagogical University.

VU – Vilnius University.

LITERATŪRA

Baranauskas A. 1898: Барановский А. Заметки о литовском языке и словаре, Санктпетербург. Plg. LDS: 34–57.

Bielskytė J. 1984: Keletas šiaurės vakarų panevėžiškių tarmės fonetikos ypatybių (homoformų audicinis nagrinėjimas) (Kursinis darbas), Šiauliai. ŠPU lietuvių kalbos katedra.

Bilevičius J. 1978: Pašvitinio šnektos žodynas (rankraštis), Joniškis. LKI Didžiojo „Lietuvių kalbos žodyno“ skyrius.

Bilevičius J. 1980: Pašvitinio šnektos tekstai (rankraštis), Joniškis. LKI Kalbos istorijos ir dialektologijos skyrius.

Bilevičius J. 1999: Pašvitinio šnektos dažninis žodynas. – Žiemgala 2, Kaunas: „Žiemgalos“ leidykla, 93–95.

Būga K. 1924: Kirčio ir priegaidės mokslas. – Rinktiniai raštai 3, Vilnius: Valstybinė politinės ir mokslinės literatūros leidykla, 1961, 19–84.

Būtėnas P. 1932: Aukštaičių tarmės okuojančios pašnektės sienos. – APh 3, 168–193.

Butkus A. 1989: Latvių kalbos įtaka Skyronių lietuvių šnektai. – Lietuvių kalbotyros klausimai 28 (Kalbų ryšiai ir sąveikos), 55–61.

Garšva K. 1973: Šiaurės vakarų panevėžiškių prozodija ir vokalizmas (Diplominis darbas), Vilnius: VU Bendrosios kalbotyros katedra (D 434).

Garšva K. 1977: Akcentuacijos ir vokalizmo sąryšis šiaurės vakarų panevėžiškių tarmėje. – Lietuvių kalbotyros klausimai 17 (Lietuvių arealinės lingvistikos klausimai), 76–87.

Garšva K. 1982: Svarbesnės šiaurės vakarų panevėžiškių fonologijos ypatybės. – Baltistica 18 (1), 65–74.

Garšva K. 1984: Šiaurės vakarų panevėžiškių fonologinės sistemos pagalbinės priemonės ir variantai. – LTSR MA darbai, A serija, 4 (89), 128–137.

Garšva K. 1986: Šiaurės panevėžiškių „murmamasis balsis“ ir penkios jo interpretacijos. – Jaunųjų filologų darbai 2, 110–117.

Gasparavičius K. 1929: Puodžiūnų sodyiaus tautosaka (LLTI LTR 5373).

Gasparavičius K. 1936: Morfologinės linkuvių šnektos ypatybės (rankraštis) LLTI MB.

173

Gerullis G. 1930: Litauische Dialektstudien, Leipzig.

Girdenis A., Zinkevičius Z. 1966: Dėl lietuvių kalbos tarmių klasifikacijos. – Kalbotyra 14, 139–147. LDS 125–135.

Girdenis A. 1971: Lietuvių kalbos tarmės (rec.). – Baltistica 7 (2), 201–209.

Girdenis A., Židon ytė G. 1994: Šiaurės panevėžiškių (Rozalimo šnektos) balsių sistema. – Baltistica 29 (2), 115–154.

Grinaveckienė E. 1961: Prebalsių minkštėjimo ryšys su galūnių trumpėjimu Stegvilų tarmėje. – Lietuvių kalbotyros klausimai 4, 167–183.

Grinaveckienė E. 1962: Kai kurios lietuvių kalbos tarmių ypatybės (iš 1961 metų dialektologinių ekspedicijų). – Lietuvių kalbotyros klausimai 5, 147–169.

Grinaveckis V. 1973: Žemaičių tarmių istorija, Vilnius: Mintis.

Jaunius K. 1898 (1970): Panevėžiškių tarmė. – Drotvinas V., Grinaveckis V. Kalbininkas Kazimieras Jaunius, Vilnius: Mintis, 1970.

Jašinskaitė I. 1957: Kirtis, priegaidė ir jų poveikis vokalizmui Biržų tarmėje. – Lietuvių kalbotyros klausimai 1, 189–194.

Kačiūškienė G. 1983: Šiaurės panevėžiškių murm amųjų balsių fonetinės ypatybės, distribucija ir fonologinė interpretacija. – Kalbotyra 34 (1), 24–38.

Kačiūškienė G. 1984: Šiaurės panevėžiškių tarmės balsių psychoakustinė klasifikacija. – Kalbotyra 35 (1), 42–53.

Ru činskaitė I. 1984: „Murmamųjų“ balsių kokybinė opozicija šiaurės panevėžiškių tarmėje. – Kalbotyra 35 (1), 72–76.

Salys A. 1992: Raštai 4, Roma.

Skirmantas P., Girdenis A. 1998: Naujesni tarmių fonetinės transkripcijos rašmenys. – Kalbotyra 47 (1), 107–121.

Specht F. 1920: Litauische Mundarten, gesammelt von A. Baranowski 1, Leipzig.

Šliavas J. 1972: Žeimelis, Žeimelis (rankraštis) VU MB.

Šukys J. 1962: Rytų aukštaičių pontininkų ir vakarinų pantininkų tarmių būdingosios ypatybės. – Lietuvių kalbotyros klausimai 5, 175–184.

Tumėnas S. 1987: Dėl prielinksnių konstrukcijų išplitimo šiaurinėse aukštaičių tarmėse. – Lietuvių kalbotyros klausimai 26, 171–175.

Urbanavičiūtė-Mark evičienė Ž. Dar kartą apie rytų aukštaičių tarmių ribas. – Kalbotyra 43 (1), 85–87.

Zinkevičius Z. 1966: Lietuvių dialektologija, Vilnius: Mintis.

Zinkevičius Z. 1969: Apie lietuvių kalbos tarmių skirstymą. – LTSR MA darbai, A serija, 137–151. LDS 136–152.

Zinkevičius Z. 1975: Smulkmenos. XII. – Baltistica 11 (1), 85–86.

Zinkevičius Z. 1976: Smulkmenos XXX. – Baltistica 12 (2), 165–160.

Zinkevičius Z. 1978: Lietuvių kalbos dialektologija, Vilnius: Mokslo ir enciklopedijų leidykla.

Zinkevičius Z. 2000: Juozas Balčikonis ir Pranas Skardžius. Pastabos apie panevėžiškių tarmę. – Panevėžio krašto kalbininkai, Panevėžys, 4–8.

WHISPERED VOWELS IN THE NORTH-WESTERN PANEVĖŽIŠKIAI DIALECT

Summary

The loss of whispered vowels in the dialect of the North-Western Panevėžiškiai set in at the beginning of the 19th century and their reduction was completed in the early 20th century. It is suggested here that, in dialect texts, whispered vowels should be marked only in those rare instances where they can really be pronounced.

174

The interpretation of the whispered vowels as a subsidiary phonological means in those cases where reduced vowels in inflectional endings may retain a certain distinctive value could contribute to a compromise among linguists advocating different interpretations of these vowels.

СЕВЕРО-ЗАПАДНО ПАНЕВЕЖСКИЕ “ШЕПОТНЫЕ ГЛАСНЫЕ”

Резюме

Краткие гласные в северо-западном паневежском говоре начали исчезать в начале XVII в. и процесс их редукции практически завершился к началу XX в. В настоящее время как сверхкраткие, так и "шепотные (бормотаемые) гласные" в говоре обычно не произносятся, а встречаются лишь спорадически, и, разумеется, только в таких случаях их и следует обозначать в диалектных текстах.

Kazimieras GARŠVA Lietuvių kalbos institutas Antakalnio g. 6 LT-2055, Vilnius, Lietuva El. paštas [email protected]

Gauta 2000 04 10

175