VILIIJA RAGAIŠIENĖ

Institute of the Lithuanian Language

ORCID iD: https://orcid.org/0000-0002-5026-4667

Research areas: accentology, dialectology, ethnolinguistics, lexicography.

DOI: doi.org/10.35321/all83-10

STRESS RETRACTION IN TEXTS OF THE EASTERN HIGHLANDER ANYKŠČIAI SUBDIALECT

Stress Retraction in the Texts of the Eastern Aukštaitian Subdialect of Anykščiai

ANNOTATION

Based on written sources—texts from the 1870s–1890s published in the German linguist Franz Specht’s Leipzig edition Litauische Mundarten gesammelt von A. Baranowski, descriptions of the Anykščiai dialects from the 1950s–1960s preserved in the Dialect Archive of the Geolinguistics Centre of the Institute of the Lithuanian Language, and dialectal texts prepared by Žaneta Markevičienė and other researchers in the second half of the twentieth and early twenty-first centuries—stress retraction phenomena in the Eastern Highlander Anykščiai subdialect are analysed. The article seeks to describe as comprehensively as possible the formation of stress retraction in the Anykščiai subdialect, determine which stress-retraction laws may have operated in the subdialect area during this period, reveal the areal distribution of the phenomenon, and discuss changes in stress retraction and their causes.

KEYWORDS: Eastern Highlander Anykščiai subdialect, stress, syllable, dialectal texts.

ANNOTATION

Based on the written sources – the texts from the 1860s–1880s published by German linguist Franz Specht in the publication Litauische Mundarten gesammelt von A. Baranowski in Leipzig, the descriptions of the local dialects of Anykščiai dating to the period of the 1950s–1960s, which are stored in the Dialect Archive of the Geolinguistic Centre at the Institute of the Lithuanian Language, the dialectal texts prepared by Žaneta Markevičienė and other researchers in the second half of the 20^th century and the beginning of the 21^st

Straipsniai / Articles

211

VILIJA RAGAIŠIENĖ

century – the article analyses the phenomena of stress retraction in the Eastern Aukštaitian subdialect of Anykščiai. It seeks to describe the formation of stress retraction in the subdialect of Anykščiai in the most detailed manner possible, to determine what laws of stress retraction could be in effect in the area of the subdialect during a given period, to reveal the areal distribution of the phenomenon and to discuss the changes in stress retraction and their causes.

KEYWORDS: Eastern Aukštaitian subdialect of Anykščiai, stress, syllable, dialectal texts.

INTRODUCTION

Of all the Eastern Highlander subdialects, the Anykščiai subdialect has probably been the least studied to date from the perspective of accentuation. In the scholarly literature, the accentuation of this subdialect has been discussed rather fragmentarily. Works containing data on Anykščiai accentuation usually provide only a brief description of the nature of stress retraction and intonation (Zinkevičius 1966: 36, 38, 41; LKT 35; LKTCH 144; Markevičienė 2001: 15–18; 2004: 39–45; RARA 36–37; AAT 18). From the perspective of accentuation, the texts collected in the 1870s–1890s in Antanas Baranauskas’s native Anykščiai region and published in the edition Litauische Mundarten gesammelt von A. Baranowski have been studied somewhat more thoroughly by linguists (Stundžia 1989: 434–442; Ragaišienė 2015: 166–210)^². Thus, there is a great lack of research on the accentuation system of the subdialect and of scholarly publications devoted to its accentual features and tendencies.

Sources, aim, methods. The material under investigation was collected from several written sources: texts of the Anykščiai subdialect published in the first volume of Litauische Mundarten gesammelt von A. Baranowski, a work published in Leipzig in 1920 by the German linguist Franz Specht (hereafter Litauische Mundarten)^³; descriptions of the linguistic features of the dialects of this subdialect prepared in the 1950s–1960s^⁴; dialectal texts published in the 1970 anthology Lietuvių kalbos tarmės; texts of the Anykščiai subdialect published in the second part of Ž. Markevičienė’s 2001 publication Aukštaičių tarmių tekstai and in the 2004 Lietuvių kalbos tarmių chrestomatija; and the manuscript of a publication prepared in 2013 by Ž. Markevičienė, Aurimas Markevičius

^¹ The manuscript page is indicated.

^² The analysis of these texts is supported by A. Baranauskas’s extensive correspondence with Hugo Weber (see Almanauskis 1930: 60–102).

^³ While serving as a professor at the Kaunas Priest Seminary, A. Baranauskas organised his students to collect dialectal texts around 1867–1884. He sent the texts they collected to the German linguist H. Weber.

^⁴ The manuscripts are preserved in the Dialect Archive of the Geolinguistics Centre of the Institute of the Lithuanian Language.

212

Acta Linguistica Lithuanica LXXXIII

Kirčio atitraukimas rytų aukštaičių anykštėnų patarmės tekstuose

and the manuscript of the instructional book Rytų aukštaičiai anykštėnai compiled by Vilija Salienė, and of the 2020 publication Anykščių apylinkių tekstai prepared by Regina Rinkauskienė.

The dialectal texts presented in the sources listed cover three periods:

1) The 1870s–1890s. The texts published in Litauische Mundarten come from two parishes: Anykščiai (305), assigned by A. Baranauskas to the fourth Eastern group (R4)^⁵, and the Subačius dialect (235), assigned to the third Eastern group (R3). Several short narratives were recorded from each of these localities (17 pages in total)^⁶. Geographically, the texts come from the northern area of the subdialect.

2) The 1950s–1960s. In collecting material for the Lietuvių kalbos atlasui, virtually all localities (points) belonging to the Anykščiai area were investigated and descriptions of their dialectal features were prepared (see Fig. 1 and the appendix)^⁷. The manuscript source for each point consists of a presentation of the locality investigated, the linguistic features of its dialect, and the informants, together with transcribed texts^⁸. The archive contains 23 sources (approximately 500 pages) from 18 points (see the appendix)^⁹. The manuscript-source material is supplemented by texts from almost all points of the subdialect published in the anthology Lietuvių kalbos tarmės from the same period. Texts recorded during the period under investigation can also be found in publications issued at the beginning of the twenty-first century.

3) The late twentieth to early twenty-first century. Texts from various localities of the subdialect under investigation were published in editions prepared by Ž. Markevičienė, R. Rinkauskienė, and other dialectologists. Admittedly, relatively few texts from the northern part of the area were published in editions from this period. For example, Ž. Markevičienė’s book Aukštaičių tarmių tekstų contains mostly narratives from the southern area of the subdialect, while the manuscript of R. Rinkauskienė’s publication Anykščių apylinkių tekstai contains texts from Anykščiai and neighbouring dialects.

To summarise the material published in the sources analysed, it can be stated that the data found in them are sufficient to assess the formation of stress retraction as a law in the Anykščiai subdialect area. This is particularly so because the sources discussed reflect the accentuation features of the subdialect under investigation relatively well.

^⁵ The numbers of the points (localities) of the Lietuvių kalbos atlaso are given in parentheses (see LKA I 21–30; see also Geržotaitė 2015: 147–149, 156–165).

^⁶ The Subačius dialect texts were provided by one informant, and the Anykščiai texts by two informants.

^⁷ The only point without a manuscript is Subačius (235). Aloyzas Vidugiris prepared the transcribed texts of this dialect, recorded in the 1950s–1960s, in 1984.

^⁸ None of the manuscripts of this subdialect contains answers to the lexical, phonetic, morphological, and syntactic questions in the Lietuvių kalbos atlaso medžiagos rinkimo programos.

^⁹ This material is supplemented by short transcribed texts of the Svedasai (239), Ažuožeriai (305), and Kniitiškiai (375) dialects recorded in 1976 by students of the Vilnius State Pedagogical Institute.

Straipsniai / Articles

213

VILIIJA RAGÃIŠIENĖ

Thus, this article analyses stress-retraction phenomena in the Eastern Highlander Anykščiai subdialect by comparing data from dialectal sources separated by more than a century. Its aim is to describe as comprehensively as possible the formation of stress retraction in the Anykščiai subdialect, discuss the stress-retraction phenomena operating from the second half of the nineteenth to the beginning of the twenty-first century, and reveal their areal distribution across the subdialect area. As is known, the Anykščiai subdialect now belongs to the area of conditional stress retraction (LKT 35; LKTCH 153). It exhibits the so-called stronger conditional stress retraction: stress is retracted from a short final syllable to the long syllable preceding it.

The study relied on geolinguistic, statistical, comparative, and internal-reconstruction methods.

THE FORMATION OF STRESS RETRACTION: CHANGE AND TENDENCIES

The texts provided by A. Baranauskas show that in the second half of the nineteenth century the law of stress retraction operated with varying intensity in the Anykščiai subdialect. In the texts from the Subačius parish, stress was regularly shifted from the short ending to the long syllable, for example: džiáũksmu, gýwi, kàr̃tus, aũkas, koply­tėlas, wìetas, ar̃ti, nãršus, tur̃tus, laũži, žmōna, kir̃wju, laũki, miẽstus, laũkus, wiẽni and so on (pp. 104–109). Only a few forms that did not retract stress were found in the texts; statistically, they are insignificant, for example: yrà (alongside ỹra), arbà, Alkà (pp. 104–105). The situation was different in the environs of Anykščiai in the 1870s–1890s. Here, stress was not retracted from the short final syllable to the preceding long syllable, for example: gwotū, yrà, žmogùs, tur̃tus, t́arpu, žõdžus, baisù, gàitu, patiekà, Szioulus, and so on (pp. 36–47). However, the marking of secondary stress with a special transcription symbol on the penultimate long syllable of some words indicates that retracted stress was already forming in the Anykščiai dialect, for example: tōkì, wãr̃ti, lĩngwà, knĩgù, kàr̃tus, szwĩntàs, trūmpà, pĩlnì, kàr̃ti, studinùtus, sūnùs, pīr̃ztus, griõusmų and so on (pp. 36–47)^¹⁰. It is assumed that the special symbols in these cases indicate a secondary-stress type, that is, “two-peaked” stress (Stundžia 1989: 435; Ragaišienė 2015: 174–175; 2018: 182–184)^¹¹. All the more so because in the southern part of the Anykščiai area and in the neighbouring

^¹⁰ If the law operated regularly in the dialects under investigation, stress in these words should be retracted to the penultimate long syllable, cf. gaspadorius → gaspadõrius. ^¹¹ For more on the term “two-peaked” stress, see: Girdenis, Rosinas 1976: 194, n. 20; Girdenis 1978: 75–76.

214

Acta Linguistica Lithuanica LXXXIII

Kirčio atitraukimas rytų aukštaičių anykštėnų patarmės tekstuose

Uteniškiai subdialect, “two-peaked” stress has been preserved to this day (Markevičienė 1999: 16; 2001: 15–16)^¹².

The texts presented probably indicate the tendencies toward the establishment of “two-peaked” stress in the Anykščiai dialect. Forms with “two-peaked” stress are more widespread in the texts under investigation, for example: gār̃di, rūbùs, krõmi, sūnkì and so on (pp. 36–37). It is likely that in the environs of Anykščiai in A. Baranauskas’s time this stress may have been sufficiently strong—that is, in some cases difficult to distinguish from words with two stresses and from incipient stress retraction^¹³. This assumption is partly supported by a thought A. Baranauskas expressed in one of his letters to H. Weber, namely that an unstressed syllable of a word is sometimes as if stressed, “as though kir̃wiù, kir̃wiùs were sounding” (Almanauskis 1930: 95; Stundžia 1989: 435)^¹⁴. The twofold marking of stress in the texts under investigation may indicate that the penultimate long syllable in the Anykščiai dialect—the first component of “two-peaked” stress—could be heard both more weakly and more strongly, cf. rõdas / ròd'as (cf. Stundžia 1989: 435; Girdenis 2003: 263). Strong “two-peaked” stress may eventually have provided an impetus for retracting stress from the short ending to the long penultimate syllable, cf. gaspadõrius → gaspadõriùs → gaspadõrius (cf. Ragaišienė 2018: 184).

The material provided by A. Baranauskas shows the uneven formation of stress retraction as a law in the Anykščiai area: the texts from the Subačius parish testify that this process had already ended, whereas the narratives recorded in the environs of Anykščiai show that it was only beginning (cf. Ragaišienė 2015: 180–183). This assumption is supported, albeit indirectly, by the analogous development of the phenomenon in the neighbouring Kupiškiai subdialect (Ragaišienė 2018: 181–184, see Fig. 2).

Manuscript sources of the Anykščiai dialects from the 1950s–1960s show that in the second half of the twentieth century stress was retracted throughout the subdialect area, though with varying intensity. Based on the descriptions, the Anykščiai area can be divided into two larger areas in which the phenomenon had different intensities. In the northern part of the area, except for the Verksnionys (208)–Svėdasai (239)–Mičiūnys (273) area, during the period under investigation stress was retracted from the short ending to the long syllable preceding it fairly regularly (see Fig. 1).

^¹² Ž. Markevičienė (2001: 15–16) calls the clearly audible stress on the penultimate long syllable in the Anykščiai dialects secondary stress. In Jolita Urbanavičienė’s (2006: 224) view, the first component of “two-peaked” stress can be classified as secondary stress, and the second as primary stress.

^¹³ An analogous situation is characteristic today of the neighbouring Uteniškiai subdialect (Urbanavičiūtė-Markevičienė, Grinaveckis 1992: 12; Markevičienė 1999: 16; Urbanavičienė 2006: 224).

^¹⁴ As is known, A. Baranauskas was convinced that a word could have only one stressed syllable (Alminauskis 1930: 95).

Straipsniai / Articles

215

VILIJA RAGAİŠIENĖ

FIG. 1. The nature of stress retraction in manuscript sources from the 1950s–1960s¹⁵

In the relatively extensive texts of these dialects, no cases of unretracted stress were found at all, for example: razum.ni, razum.nus (alongside razumnų), vaikus, vil.ku, su kariėtu, vietą, rūbus, dienas, kėliṅti, pìr.stus, gàl.va, spȧl.va, trūṁpi, ìlgi, aũ.kšti and so on (Rybōkai, 146); jaũna, ūnksti, mérga, ka̋rtu, nebegi.va, per laũkus, viéną, per dienas (Rūkiškis, 270); ìlga, ga̋lva, miñkšta, dùnkčius, ūnksti, diéna, viéni, viéną, lìntas, piẽtus, agūkus, blìnus, piñ̃kus (Andriōniškis, 272), and so on^¹⁶. Regular stress retraction

¹⁵ Attention should be drawn to points 236 and 237, which in recent works are assigned to the Anykščiai subdialect (RAA 63–70). The points discussed are not included on the map provided in the article. During the period under investigation, they were assigned to the Kupiškiai subdialect (LKA I 33, map 1). In the 2014 publication XXI a. pradžios lietuvių tarmės: geolingvistinis ir sociolingvistinis tyrimas, both points are likewise assigned to the Kupiškiai subdialect (LTT 187, Fig. 109).

¹⁶ The stress of the root forms žmogus / žmõgus / žmą̃gus may testify to the presence in the dialects of an adjacent variant of the second accentual paradigm. The cases discussed may also be associated with stress retraction, cf. žma.gaũs (Ragaišienė 2018: 186, n. 27). The stem-stressed form bai̇sus may likewise be associated with stress retraction, although the possibility cannot be excluded that it may also have been inherited from an earlier period.

216

Acta Linguistica Lithuanica LXXXIII

Kirčio atitraukimas rytų aukštaičių anykštėnų patarmės tekstuose

in this area is also attested by texts published in the anthology Lietuvių kalbos tarmės (LKT 295, 302–303); cf. also a text recorded at the Rūkiškis point in 1964 (LKTCH 149). Narratives recorded in the environs of Andríoniškis in 1965 likewise predominantly contain forms with retracted stress, for example: k̇leim̃u, pìr̃ki, mer̃.ga.s, žm̃õ.gus (alongside žmà.gaus), žò.džiu / žo·džiu (AAT 77–78).

In the 1950s–1960s, the situation was different in the Verksnió­nys (208)–Svedasaĩ (239)–Mičíonys (273) area. Manuscript-source material shows that unretracted stress predominated here, for example: pilnà, čigãnus, ràžu, liẽsvas, viẽdru, in būdàs, bizū­nù, galvà, žmãgus, būdà, u.lìča, grausmù, irà, dikà, lintù, sùnus, klaimì, valdžà, lìn­tàs, vaikùs, daržùs, saldù (Verksnió­nys, 208); varpùs, žmãgus, kardù, vienì, vartùs, kraujù, bėdà, dienà, dienàs, kėlìntà, navìnà, mergàs (Svedasaĩ, 239); rañdį, baltì, švintù, širmà, pìlnà, pìlvù, nuogà, pìlkà, gailù, kūmpì, vienà, žmãgus, bjãurùs, dienàs, šimtùs, su kãčelù, ràžu (Mićió­nys, 273). Only isolated examples of stress retraction were found in the dialects under investigation, for example: blãgi (Verksnió­nys, 208); liẽtus, pẽr pĩetus (Svedasaĩ, 239); rùnkas, var̃tus, dù šimtù (Mićió­nys, 273).

The situation appears to have been much more complex in the southern part of the Anykščiai area. During the period under investigation, stress was rarely retracted from the short ending to the penultimate long syllable in some dialects, but fairly regularly in others. Regular stress retraction was recorded in the Kurkliai (341)–Kni­tiškiai (375)–Juodáusiai (447) area. In the short manuscripts of these dialects, virtually no forms without stress retraction were found, for example: diẽvu, róžu, liẽtus, maišùs, irà, kartù, vienà, ùṅksti, suñki, baltì, baĩsi, tvarkà, vaikùs, juódì, sãisra (Kurkliai, 341); sūnus, lìṅgvi, pìṅkas, pìṅkus, jaũni, pas pónus, dienà, žmónà, ùntrì, baravìkus, maĩšus, ma.kìkla, kálba, liẽtus, užmìga miẽgu, vienì, i.ra (Kni­tiškiai, 375); róžu, ainù, dienà, piẽtus, kviẽtkus (Juodáusiai, 447). The narratives published in the anthology Lietuvių kalbos tarmės also demonstrate the regular stress retraction present in these dialects in the 1950s–1960s, for example: su viẽdru, vienà, pir̃tis, ainì, ainù, par piẽtus (LKT 305, 447). Stress also appears to have been retracted fairly regularly in the Kiaukliai (477) dialect, which borders the area of weak conditional stress retraction, for example: pir̃ką, kìrvu, mar̃ti, pìṅkus, rùnka, pìṅki, vaikùs, kapeikàs, žmógus, pas susiẽdus and so on.

The descriptive material shows that irregular stress retraction predominated in the rest of the southern area in the 1950s–1960s. It appears that stress was not retracted from the short ending to the preceding long syllable only in the Skiemoniai (343) dialect, for example: mergà, žiemà, sausà and so on¹⁷. In the other dialects of this area, the phenomenon had varying intensity. In the texts of the Anykščiai (305), Rubikiai (342), and Dapkūniškiai (376) dialects, forms with stress retraction account for almost 90 percent, for example: kiẽmus, rùnkas,

¹⁷ This is also confirmed by the text published in the anthology Lietuvių kalbos tarmės. Stress was not retracted from the short word ending to the preceding long syllable, for example: šeimà, jaũni, numirsù (LKT 306).

Straipsniai / Articles

217

VILИJA RAGAİŠIENĖ

ỹra, sū.nus, sʼvīr.ní, díenas, līetus, žēdna mēr.ga, su dū.ša (Anykščiai, 305)^¹⁸; žmogus (alongside žmogaũs), kār.tus, bėdas, rùnka, žiñgsnìs, díenas (Rubikiai, 342); žmogus, žmōna, rōzu, sūnus, rū.bus, bėda, su diédu (Dapkūniškiai, 376)^¹⁹. In the environs of Bālnin-kai, Lyduõkiai, Žélva, and Šešuõliai, they account for approximately two-thirds of the forms capable of retracting stress, for example: žmōgus (alongside žmogũmì), pìnki, sū.nus (alongside sū-naũs), kaìtra / šviesà, daržbì, gir̃džiù, pirmà, mergà (Bālninkai, 411); sū.nus, blō.gì, mō.tka, per rùnkas, vièna, žmō.na, skū.ra, pìņkis, šimtus, rū.bus, mergà, razũmni / ainù, ainì, sunkù, mergà, kartù, raitì (Lyduõkiai, 410); nùogi, rū.bus, līetus, rōzus, kaũlus, galvà slōbna, viéni / dungùs, minkà, nėrà (Žélva, 449); gìva, mōtka, vār-du, mėsa, andarõkus / sunkù, skalnà, galvù (Šešuõliai, 448). Analogous stress-retraction tendencies are also attested in these dialects in other texts from the period under investigation (LKT 307–309).

In summary, it can be stated that manuscript sources from the 1950s–1960s and other texts demonstrate the development of the law of stronger conditional stress retraction in the Anykščiai subdialect area. Thus, during this period, the phenomenon that had become established in the neighbouring Kupiškiai subdialect continued moving from north to south and spread throughout the Anykščiai area (for more detail, see Ragaišienė 2018: 185–187).

Narratives recorded in the area of the subdialect under investigation from the late twentieth to the early twenty-first century show ongoing, albeit minor, changes in the phenomenon (cf. Figs. 1 and 2). It is evident that stress retraction is becoming increasingly established in the northern part of the Anykščiai area, more precisely in the Verksnionys (208)–Svėdasai (239)–Mičiūnonys (273) area. At the Verksnionys point, more cases of stress retraction from the short final syllable to the preceding long syllable were found than in manuscript sources from the 1950s–1960s, cf.: vìžas, uñks’ti, jaũna, kavalier’ùs / k’aũlas²⁰, aždun.cù (LKT 147–148)²¹; laĩki, ìra, sà.di, mė.sà, grì.čė / laĩkù, laĩkì, žiemà, ì.ra, blɔ·gà and so on (RAA 52–55)²². In the environs of Svėdasai, stress retraction has not yet become stable. In this dialect, the law appears to operate very unevenly. In texts recorded in 2000, forms with stress retraction account for almost half of the possible cases, for example: užmirštù, sàu-gi vietà, vièna, laũkus, pìņki(s) šimtus, gārsus, grì.čė, su grìčų, vārpus²³ / laĩkùs, vièna, su draugù and so on (Markevičienė 2001: 63–70). The tendency to retract stress is indicated by

^¹⁸ Analogous usage is also attested in other texts recorded in the 1960s, for example: sɔ·di (alongside sɔ·di) vaĩkus, susiėdus, laĩkus, jùkus (AAT 67–68).

^¹⁹ In the texts of the Anykščiai dialect, slightly more forms without stress retraction were found than in the texts from the Rubikiai and Dapkūniškiai points. It is true that the dialectal texts from the latter dialects are shorter than those from the Anykščiai dialect.

^²⁰ The form is used three times in the narrative.

^²¹ The sources from which the examples are taken are given in parentheses.

^²² Both texts were recorded in the environs of Verksnionys in 1979.

^²³ The form is used seven times in the text.

218

Acta Linguistica Lithuanica LXXXIII

Kirčio atitraukimas rytų aukštaičių anykštėnų patarmės tekstuose

and the use of forms with “two-peaked” stress, for example: nugrįm̃si, laĩkūs, idãmų, lìnksmį, malaišą, duobės and so on (Markevičienė 2001: 63–67). In narratives recorded in 2003, forms without stress retraction predominate, for example: nelìgà, žiemà, vienì, karštà, ainù, baltà, lintà, gelsvà, gandrù and so on (RAA 56–62)^²⁴. Only isolated forms with “two-peaked” stress were found in these texts, for example, žiemà (RAA 56–62).

^² FIG. The nature of stress retraction in texts from the late twentieth to the early twenty-first century^²⁵

By the late twentieth to early twenty-first century, retracted stress had, so to speak, already become established in the environs of Anykščiai (305). Few word forms were found in the texts under investigation that did not retract stress from the short ending to the penultimate long syllable, for example: s³· dbl, slì·vas, gė·les, vìeni, jāuni, uņkšti, gaĩlta, žmò·na, nuveĩnu, sėrgi gėltu, var̃lės / miežùs, aušrà (RAA 71–79); jāuną, su klebš·nu, suprañti, sveĩka gīvà, peñkus, šìrdis, uņkšti / nebegì·và, vaikus (AAT 69–72).

Written sources show that changes in the phenomenon from the late twentieth to the early twenty-first century are also characteristic of the southern part of the Anykščiai area. According to the intensity of stress retraction, the area

^²⁴ The informants of the texts recorded at the Švedasai point are representatives of the oldest generation.

^²⁵ No texts have been published from the points not marked on the map.

Straipsniai / Articles

219

VILIJA RAGAIŠIENĖ

is divided into two larger areas. Stress is retracted fairly regularly on the western edge of this area—in the Kurkliai (341) and Lyduokiai (410) dialects. For example, in narratives recorded in the environs of Kurkliai from representatives of the oldest generation, word forms retaining stress in the ending account for approximately one-fifth of the possible cases, for example: aini, ąinu, viena, ųksti, laiku, jauni, supranti, žǫdžu, rąndi, šir̃dis, rųbus, pinkės / linksmì, žaizdžù, biednà, šienì, rųbus (Markevičienė 2001: 71–76); aini, žmõna, pįnki, įra, baigù / su dainù, gamtà (RAA 82–85); skąudi, cigž̨nus, pi-nigę̇lus, laiku, su męsa, mɔkįkla, pijɔ̨kus, penki, įra, vienà / įrà, vienà, rųbus, kartų, vardų, miestè (AAT 32–42). An analogous situation is also recorded in the localities of Lyduokiai, for example: su vir̃vi, dįki, ąinu, krųva, švięsa, ląuki, run̨ku, kįemu, žaizda, pęilu, spakąina, ąlga, dur̨na, jąuna, jauni, merga / pįnki, kartų, ąinu, un̨ksti, vardų, girtà, varksti, darbi (Markevičienė 2001: 77–86); dalįkus, ąlga, kartųs, sil̨pna, kraįčus, ver̨pju, pįnki, ąinu, bal̨sus, šeima nekalti / vaikųs, žmɔ̨gus (RAA 89–106).

The situation is much more complex on the eastern edge of the area under discussion. Here, the law of conditional stress retraction operates irregularly (cf. Markevičienė 2001: 15). A comparatively large number of cases without stress retraction were found in texts from the environs of Skiemoniai (343), Bálninkai (411), and Žėlva (449). Most were found in the Skiemoniai dialect, where forms retaining stress on the short ending constitute almost half of the possible cases, for example: viena, vieni, įra, nebę̇ra, supranti, žǫdžu, tięsa, dę̇sra.s, kįemus, du kąulu, penki, žmɔ̨gus (alongside žma.gąus) / ailės, marčės, ląuki, ląukus, jąuna, bąsius, vaikųs, męsa, laivù, draugės (AAT 48–61). Similar stress-retraction tendencies are characteristic of Bálninkai (411), for example: dur̃ni, valdžė, pįnkus, vardu, vienà, run̨kas, baržda / pukb̨l, pirmà, kartų, ląukus, linksmì, lintàs; Dapkūniškiai (376), for example: dur̨ni, valdžė, pįnkus / run̨ką; and Zibalai (448)^²⁶, for example: grųšė, vįšne, burš̨kus, svagųnus, nąujus, dienas, vienà, biedni / šal̨ta, viršųs, vaikųs, aini, tvarkà, kalbà, kartųs, penki, šnektoms (Markevičienė 2001: 86–93, 106–109)^²⁷. The texts of the dialects discussed contain many forms with “two-peaked” stress, which may testify both to the tendency for the phenomenon to become established and to its uneven operation according to the principle of phonological interaction, for example: kalbà, dainas, ląuki, šeimà; barždų, šeimų, sniegų, miegi, ląuki, šir̃dis, galvà, šienų; jilgà, kumpi, ląuki, ųksti, biedni and so on (Markevičienė 2001: 87–90, 106–109).

In terms of the intensity of stress retraction, the Kiaukliai (477) point, bordering the Širvintiškiai subdialect, also belongs with these dialects. Here, stress-retracting forms are more widespread

^²⁶ The dialect belongs to the Šešuoliai point.

^²⁷ A fairly regular stress retraction also occurs in some texts from the Balninkai point, cf.: vįšne, merga, burna, žaizda, dalįkus, rųbus (RAA 85–88). The person who provided the text belongs to the oldest generation, but the year of the recording is unknown. The possibility that it was recorded in the second half of the 20th century cannot be ruled out.

220

Acta Linguistica Lithuanica LXXXIII

Kirčio atitraukimas rytų aukštaičių anykštėnų patarmės tekstuose

stress-retracting forms, i.e. they constitute slightly more than half of the possible cases, e.g.: vieni, antri, vérki, maišù, viẽna, diẽnas, vìlkus / pirmà, aišù, gíržù, láukė (RAA 112–123). It should be noted that, during the period under investigation, fewer stress-retracting forms were found in texts recorded in these localities than in manuscript sources from the 1950s–1960s (see Fig. 1).

There is also a tendency to retract stress more regularly from a short ending to the immediately preceding long syllable in the vicinity of Želva (449). In this dialect, the tendencies toward stress retraction do not essentially differ from those of the neighboring Kiaukliai (477) point. In the settlements belonging to the Želva point, stress-retracting forms are likewise more widespread, e.g.: viẽni, biẽdni, žmõ·na, nelaĩku, su bizù·nu, paliẽku, kártù, diẽnas / vaĩkù(s), maišùs, durnà, jáunà, baĩgù (Markevičienė 2001: 98–102); pìnkì, diẽnas, dùrni, láuki / vaĩkùs, bandà (RAA 107–111).

Texts produced by representatives of the middle and youngest generations cast doubt on the persistence of the stress-retraction rule in the southern part of the Anykštėnai area. In narratives recorded at the beginning of the 21st century in the Skiemónys (343) and Kurkliai (341) dialects, tendencies toward columnar accentuation, rather than stress retraction, are probably more strongly attested, e.g.: mkì klà, žõ·džù, miẽsti, suvogù·nus, blì·nus, klõldù·nus, diẽna, šõ·kus / vaĩkùs, i rà, šáulùs, rankà, vė̃iklà, šventės, šeimàs, láukė, kalnė, penkì and others (AAT 43–47, 62–66). Especially since there is a stronger tendency to generalize stem stress on the penultimate syllable of words, whose nucleus consists of long vowels or the monophthongs ie, uo^²⁸. This tendency is also noticeable in the southern part of the Anykštėnai subdialect. Quite a few such usage cases were found, for example, in texts by representatives of the oldest generation of the Žibalai (448) dialect, cf.: grù·še, vì·šne, bur·kus, svagù·nus, naũjus, diẽnas, viẽna, biẽdni / šaltà, viršùs, vaĩkùs, aĩni, tvarkà, kalbà, kartùs, penkì. That stem-stressed forms may be associated with the morphophonological leveling of the accentual paradigm (columnar stress) is indicated by another fact. In the southern part of the Anykštėnai area, there is a tendency to stress the instrumental singular and accusative plural forms of acute stems on the ending, e.g.: vẽ·žù, vì·rù, lungùs, šõ·nùs, vì·rùs and others (Markevičienė 2001: 16). Their existence may indicate the formation of the second member of the accentual model—the ending type (Ragaišienė 2016: 163). Understandably, it is not possible to determine reliably which cases are associated with stress retraction and which arose through the formation of columnar stress. The stem stress of forms whose syllable nucleus consists of short vowels is probably primarily associated with the columnar accentuation type, e.g.: cepelìnus, su ma.mù·ti, sežù·ni and others (AAT 18). There is no doubt that the two phenomena are interrelated and probably operate according to the principle of overlap.

²⁸ A model of two accentuation types—columnar and ending—is becoming increasingly established in the non-stress-retracting Southern Aukštaitian, Eastern Aukštaitian Vilniškiai, and other Aukštaitian subdialects (Grumadienė 1994: 104; Mikulenienė 1994: 104; Urbanavičienė 2010: 177; Ragaišienė 2016: 163; 2019: 11).

Straipsniai / Articles

221

VILIJА RAGĂIŠIENĖ

In summary, it may be stated that source material from the late 20th and early 21st centuries attests to changes in stress retraction as a phenomenon occurring throughout the Anykštėnai subdialect area. At the western edge of the area (the Anykščiai–Kurkliai–Lyduokiai zone), stress is retracted fairly regularly from the short final syllable to the preceding long syllable; at the eastern edge (the Svėdasai–Skiemónys–Balninkai–Kiaukliai zone), forms retaining stress on the short ending are abundant.

CONCLUSIONS

A comparative analysis of A. Baranauskas’s texts from the second half of the 19th century, descriptions of the dialectal features of the dialects from the 1950s–1960s, and sources from the late 20th and early 21st centuries reveals the development of stress retraction in the Eastern Aukštaitian Anykštėnai subdialect and the processes of change affecting this phenomenon during the period under investigation. The more intensive conditional stress-retraction rule began to develop relatively late in the Anykštėnai subdialect and developed unevenly—it became established in the northern part of the area no earlier than the second half of the 20th century, while in the southern area it remains irregular and should be regarded as still developing. The study supports the current theory of the origin of Lithuanian dialects, according to which changes, including stress retraction, proceeded from north to south. The data of the Anykštėnai subdialect support the assumption that, in the second half of the 19th century, an innovative wave of stress retraction moved from the northern part, where it was fairly regular, toward the eastern edge. The possibility that the formation of stress retraction in the Anykštėnai subdialect was considerably influenced by a Selonian substratum cannot be ruled out (Zinkevičius 2006: 30, 73).

In the sources of the Anykštėnai subdialect examined, stress is retracted from a short ending onto the immediately preceding syllable, whether acute or circumflex. The intonation of the retracted stress of long syllables is falling (cf. Kazlauskas 1968: 26; Ragăišienė 2018: 187).

Stress retraction may also, indirectly, be associated with the morphophonological leveling of the accentual paradigm, that is, with the formation of the columnar accentuation type. Two facts indicate this: 1) stress tends to be retracted more often in the instrumental singular and accusative plural forms; 2) throughout the subdialect area, stress is fairly regularly retracted from a short ending onto the penultimate syllable, whose nucleus consists of long vowels and the monophthongs ie, uo. It is likely that the two phenomena—stress retraction and columnar stress—are interrelated and may operate according to the principle of overlap.

222

Acta Linguistica Lithuanica LXXXIII

Kirčio atitraukimas rytų aukštaičių anykštėnų patarmės tekstuose

LITERATURE AND SOURCES

AAT – Texts from the Anykščiai area, ed. R. Rinkauskienė, Vilnius: Institute of the Lithuanian Language, 2020.

Almanauskis Kazimieras 1930: Letters of Bishop Antanas Baranauskas to Hugo Weber. – Archivum Philologicum 1, 69–102.

Geržotaitė Laura 2015: A Geolinguistic Analysis of Antanas Baranauskas’s Classification of Lithuanian Dialects. – Problems of the History of Language and Dialectology 4, Vilnius: Institute of the Lithuanian Language, 133–165.

Girdenis Aleksas 1978: A Small Accentological Note. – Kalbotyra 14(1), 75–76.

Girdenis Aleksas 2003: Theoretical Foundations of Lithuanian Phonology, Vilnius: Institute of Science and Encyclopaedia Publishing.

Girdenis Aleksas, Rosinas Albertas 1976: Some Reflections on Issues of Dialectal Phonetics. – Baltistica 12(2), 188–197.

Grumadienė Laima 1994: Some More Recent Features of the Declension and Accentuation of the Punsk Dialect. – Questions of Lithuanian Linguistics 34, 97–106.

Kazlauskas Jonas 1968: Historical Grammar of the Lithuanian Language (Accentuation, Noun, Verb), Vilnius: Mintis.

LKA I – Atlas of the Lithuanian Language 1: Lexicon, responsible editor K. Morkūnas, Vilnius: Mokslas, 1977.

LKT – Lithuanian Dialects: A Chrestomathy, eds. E. Grinaveckienė, A. Jonait ytė, K. Morkūnas, B. Vanagienė, A. Vidugiris, Vilnius: Mintis, 1970.

LKTCH – Chrestomathy of Lithuanian Dialects, eds. R. Bacevičiūtė, A. Ivanauskienė, A. Leskauskaitė, E. Trumpa, Vilnius: Institute of the Lithuanian Language, 2004.

LTT – Lithuanian Dialects at the Beginning of the 21st Century: A Geolinguistic and Sociolinguistic Study (Maps and Their Commentary), eds. D. Mikulėnienė, V. Meiliūnaitė, Vilnius: Briedis, 2014.

Markevičienė Žaneta 1999: Texts of the Aukštaitian Dialects 1, Vilnius: Vilnius University Press.

Markevičienė Žaneta 2001: Texts of the Aukštaitian Dialects 2, Vilnius: Vilnius University Press.

Markevičienė Žaneta 2004: Changes in the Aukštaitian Dialects. – Kalbotyra 53(1), 39–45.

Mikulėnienė Danguolė 1996: On Accentuation Variants of Inflected Words in the Southern Aukštaitian Dialects. – Questions of Lithuanian Linguistics 36, 148–155.

Straipsniai / Articles

223

VILIJA RAGAIŠIENĖ

RAA – Eastern Aukštaitian Anykštėnai, eds. Ž. Markevičienė, A. Markevičius, V. Salienė, Vilnius: Institute of the Lithuanian Language, 2013.

Ragaišienė Vilija 2015: Accentuation Features of Antanas Baranauskas’s Texts. – Problems of the History of Language and Dialectology 4, Vilnius: Institute of the Lithuanian Language, 166–210.

Ragaišienė Vilija 2016: The Accentuation System and Its Changes. – The Change of Lithuanian Dialects at the Beginning of the 21st Century: Lithuanian Dialects in Poland, Vilnius: Institute of the Lithuanian Language, 120–204.

Ragaišienė Vilija 2018: Changes in the Accentuation System of the Eastern Aukštaitian Kupiškėnai Dialect from the 1870s–1890s to the 1950s–1960s. – The Eastern Aukštaitian Subdialect: Change and Changes, Vilnius: Institute of the Lithuanian Language, 176–203.

Specht Franz 1920: Lithuanian Dialects Collected by A. Baranowski 1. Texts, Leipzig: Verlag von K. F. Koehler.

Ragaišienė Vilija 2019: Accentuation Tendencies of Polysyllabic Nouns in Manuscript Sources of the Western Aukštaitian Kauniškiai Dialects from the 1950s–1960s. – Lietuvių kalba 13, 1–19. Available online: http://www.lietuviukalba.lt/index.php/lietuviu-kalba/article/view/268/19.

Stundžia Bonifacas 1989: Antanas Baranauskas the Accentologist. – Baltistica III (2) supplement, 434–442.

Urbanavičienė Jolita 2006: Accentuation Features of the Peripheral Dialects of the Eastern Aukštaitian Vilniškiai. – Baltistica 41(2), 219–231.

Urbanavičienė Jolita 2010: The Phonological System of the Svirkai Dialect: Vocalism and Prosody, Vilnius: Institute of the Lithuanian Language.

Urbanavičiūtė-Markevičienė Žaneta, Grinaveckis Vladas 1992: Lithuanian Dialects: Phonetics, Morphology 2, Vilnius: Vilnius Pedagogical University.

Zinkevičius Zigmas 1966: Dialectology of the Lithuanian Language, Vilnius: Mintis.

Zinkevičius Zigmas 2006: The Origin of Lithuanian Dialects, Vilnius: Institute of the Lithuanian Language.

224

Acta Linguistica Lithuanica LXXXIII

Kirčio atitraukimas rytų aukštaičių anykštėnų patarmės tekstuose

APPENDIX. Manuscript sources of the Anykštėnai subdialect in the Dialect Archive

Name of the LKA point (number) Name of the locality studied²⁹ Year of study, recorder, source registration number Rybokai (146) Rybokai, Miesteliškės, Rokiškis district 1956, E. Grinaveckienė, D. Gargaitė, reg. no. 279 Verksnionys (208) Verksnionys, Rokiškis district 1954, V. Valunta, reg. no. 208 Švėdasai (239) Švėdasai, Anykščiai district 1954, V. Valunta, reg. no. 195 Rukiškis (270) Surdegis, Troškūnai district 1957, A. Kratulis, reg. no. 393 Andrioniškis (272) Andrioniškis, Anykščiai district 1956, J. Subačiūtė, reg. no. 310 Andrioniškis (272a) Ramaškonys, Anykščiai district 1964, J. Kaušpėdas, A. Kairytė, reg. no. 827 Mičionys (273) Mičionys, Anykščiai district 1954, V. Valunta, reg. no. 205 Anykščiai (305) Liudiškiai, Anykščiai district 1956, A. Satkūnaitė, D. Vytenaitė, reg. no. 328 Anykščiai (305a) Dabužiai, Anykščiai district 1964, G. Jurkūnaitė, R. Venckutė, reg. no. 828 Kurkliai (341) Kurkliai, Kovarskas district 1956, S. Grytė, reg. no. 317 Kurkliai (341a) Kurkleliai, Anykščiai district 1964, N. Kačiವsyತė, L. Urbutytė, reg. no. 829 Rubikiai (342) Rubikiai, Anykščiai district 1956, D. Trumpickas, reg. no. 313 Rubikiai (342a) Mažeikiškiai, Anykščiai district 1956, M. Naginionytė, G. Kalinskaitė, reg. no. 314 Skiemоnys (343) Skiemоnys, Anykščiai district 1956, D. Drakickaitė, G. Drakickaitė, reg. no. 315 Knitiškiai (375) Knitiškiai, Kovarskas district 1957, S. Jurevičius, A. Vyšniauskas, reg. no. 396 Dapkūniškiai (376) Dapkūniškiai, Kovarskas district 1956, G. Leščūtė, reg. no. 318 Lyduokiai (410) Lyduokiai, Ukmergė district 1957, V. Gaškaitė, A. Trečiokaitė, reg. no. 374 Lyduokiai (410a) Kunigiškiai, Anykščiai district; Žemaitkiemis, Ukmergė district 1964, D. Mielytė, R. Stankaitė, V. Čelutkaitė, reg. no. 831 Balninkai (411) Balninkai, Kovarskas district 1956, Martikonytė, reg. no. 319 Juodausiai (447) Juodausiai, Ukmergė district 1959, Z. Zinkevičius, reg. no. 566 Šešuoliai (448) Šešuoliai, Ukmergė district 1957, O. Sedelskytė, A. Bagočiūnaitė, reg. no. 349 Želva (449) Želva, Ukmergė district 1957, O. Sedelskytė, A. Bagočiūnaitė, reg. no. 350 Kiaukliai (477) Kiaukliai, Širvintos district 1957, O. Sedelskytė, A. Bagočiūnaitė, reg. no. 348

²⁹ The name of the locality studied is given exactly as it appears in the source registration journal and in the source.

Straipsniai / Articles

225

VILIJA RAGAIŠIENĖ

Stress Retraction in the Texts of the Eastern Aukštaitian Subdialect of Anykščiai

SUMMARY

Based on the written sources dating from the 1860s–1880s to the early 21^st century, the article analyses the phenomena of stress retraction in the Eastern Aukštaitian subdialect of Anykščiai. The aim of the study is to describe the formation of stress retraction in the subdialect of Anykščiai in the most detailed manner possible, to discuss the phenomena of stress retraction which were in effect during a given period and to reveal their areal distribution in the area of the subdialect.

The comparative analysis of the sources dating to the periods, which are separated by more than a century, shows the formation of stress retraction in the Eastern Aukštaitian subdialect of Anykščiai and reveals the processes of change of this phenomenon taking place during the period in question. The available data show that a more intensive law of stress retraction started forming in the subdialect of Anykščiai comparatively late and its development was not even – it established itself in the northern part of the area not earlier than the second half of the 20^th century; meanwhile, it is still irregular in the southern area and is construed as still forming. This study substantiates the present theory of the origin of Lithuanian dialects, according to which the changes, including stress retraction, moved from the north to the south. The data from the subdialect of Anykščiai support the assumptions that in the second half of the 19^th century the innovative wave of stress retraction moved from the north, where it was rather regular, towards the eastern boundary.

Stress retraction may be related with the morphological levelling of accentual paradigm, i.e., the formation of columnar (fixed) type accentuation. It is shown by two aspects. First, stress is more commonly retracted in the instrumental singular and accusative plural forms. Second, stress is rather regularly retracted from the short ending into the penultimate syllable formed by long vowels and the monophthongs ie, uo in the whole area of the subdialect. It is likely that both phenomena – stress retraction and columnar stress – are interrelated and may demonstrate the principle of overlapping action.

Received 19 November 2020

VILIJA RAGAIŠIENĖ

Institute of the Lithuanian Language

5 Petro Vile išio St., LT–10308 Vilnius, Lithuania

[email protected]

226

Acta Linguistica Lithuanica LXXXIII