LIETUVIŲ KALBOTYROS KLAUSIMAI XXXVI (1996)

LIETUVIŲ KALBA: TYRĖJAI IR TYRIMAI

Danguolė MIKULĖNIENĖ

ON THE ACCENTUATION VARIANTS OF DECLINABLE WORDS IN THE SOUTHERN HIGHLANDER DIALECTS

So far, accentuation variants of declinable words have more often been discussed from the perspective of the standard language—considering which of them is preferable in the standard language. All kinds of accentuation patterns are possible in dialects, but there is generally a tendency toward ending accentuation (Būga 1961: 79; Skardžius 1935: 35—42, 144—155; 1968: 58—59, Iliič-Svityč 1963: 29—30, 41—42, 55; Girdenis 1967: 6 et al.).

The spoken language’s tendency toward oxytone accentuation has been substantiated statistically: according to B. Stundžia’s data (1984: 88), disyllabic variants with root-versus-ending accentuation account for 86% of all cases of variation. Most commonly, accentual paradigms differing in the intonation of the same root vary—those of the first and third types (cf. beržas 1 / beržas 3, tėvas 1 / tėvas 3), and the second and fourth types (cf. aušra 2 / aušra 4, bala 2 / bala 4). Words whose accentuation changes together with the intonation of the root account for half as many—31.6% (cf. audra 1 / audra 4, kaina 1 / kaina 4) (Stundžia 1984: 87, Table 1).

§1. The southern Highlander dialects differ in the abundance of accentual variants: they contain many words and forms accented differently from those in the standard language; cf. in Zietela ir pas_t o k ū s ž m o n i s; vieni v i e n ū s mėgo; mės pastatėm itus p ė d ų s (Vidugiris 1959: 207—209), or examples recorded in the vicinity of Dieveniškės: ir dar blizga k a u l a i; o kur čieį r a k t a i; pinki t v a r t a i, vienos duri˙s, and so on (Lipskienė, Vidugiris 1967: 195—196).

In these dialects, the accentuation of the singular instrumental and plural accusative fluctuates particularly: words with acute intonation belonging to the third (and sometimes also the first) accentual paradigm are often accented on the ending. Thus, in the Dieveniškės subdialect one may hear arklūis, galvàs, grūdūs (Lipskienė, Vidugiris 1967: 196); in Švendubrė—karštų, nuogų, pėsčių; arklius, karštus (Naktinienė, Paulauskienė, Vitkauskas 1988: 8); in Punskas—pėdų, pėdūs; gyvų, gyvūs; ilgų, ilgūs; šaltų, šaltūs (Grumadienė 1992: 11).

§2. Since accentuation variants are usually recorded in the most endangered—and therefore intensively studied—southern borderland Highlander dialects, the presence of accentual variants is often linked to the influence of Slavic languages (cf. Zinkevičius 1966: 31). Under their constant influence, the relationship between dialectal intonations and stress placement in certain word forms is disrupted, and there is a tendency toward morphological leveling of the accentual paradigm—toward columnar or ending accentuation (cf. Grumadienė 1988: 144; 1994: 104).

148

As the dialect’s accentuation system disintegrates, Slavic languages undoubtedly act as a catalyst, but the mixing of accentual paradigms cannot always be regarded as a new phenomenon. The variants under discussion are found not only in endangered borderland subdialects but throughout the area of the southern Highlander (Dzūkian) dialects and neighboring subdialects¹. They are also found farther north—for example, in the southern part of the eastern Highlander Širvintos dialect, where tendencies toward columnar accentuation are also observed (Morkūnas 1969: 117). Thus, they extend up to the retraction-of-stress zone.

§3. Attempts have been made to seek phonetic causes for this phenomenon: it has been explained by the less distinct differentiation of long-vowel intonations in the eastern and southern Highlander dialects (Stundžia 1984: 86), by the convergence in tone of circumflex and acute intonations (Naktinienė, Paulauskienė, Vitkauskas 1988: 8), or by adapting the circumflex intonation of a word stem to the acute intonation of that dialect (Grinaveckis 1991: 110—111). However, the presence of forms retaining acute intonation—for example, the singular accusative kárštą (trm. ká.rštu.) alongside karštù, karštùs, or árklį (trm. á.rkl'i.) alongside arkliùs (and árklius) (Naktinienė, Paulauskienė, Vitkauskas 1988: 8)—shows that the stem acute has not disappeared completely in the southern Highlander dialects. Moreover, the phonological opposition between intonations in diphthongs is well preserved².

In Z. Zinkevičius’s view, the use of several alternating forms in the same locality (and even by the same informant—author’s note) indicates that accentuation variants are non-phonetic in origin and should therefore first be explained as resulting from “forgetting the old accent and subsequently leveling it by analogy” (Zinkevičius 1979: 91). It is therefore logical to suppose that the “leveled” variants, regarded synchronically as hypercorrect (Grumadienė 1988: 144—148), may also have historical grounds for their existence.

§4. That such variants existed is already attested in the writings of M. Daukša; cf. M. Daukša’s *daiktùs (daiktús) alongside dáiktus (dáiktus), *darbùs (darbùs) alongside dárbus (dárbus, dárbus), *pėdàs (pedás); *gardžiàs (gardžiás) alongside gárdžius (gárdžius, gárdžius), *gyvù, gyvùs (giwú, giwùs) alongside gývus (giwus), *pilnù (pilnù), as well as *aiškùmi, *patogùmi (aiškùmi, patogùmi) alongside aiškius, (ne)patógius (aiškius, acc. pl. patógus, nepatógus).

It is doubtful that these nouns with ending accentuation are merely simple typographical errors—the plural accusative form *daiktùs was found by P. Skardžius five times (dáiktus—six times), and *darbùs three times (dárbus—also three times). It is entirely plausible that M. Daukša may have known them from the southern Highlander—

¹ This is evidenced by material from the manuscript archive of the Language and Dialectology Department of the Institute of the Lithuanian Language and from recent dialectological field expeditions (1991—Žasliai, Pelesė; 1992—Valkininkai, Dieveniškės, Balbieriškis; 1993—Leipalingis, Mielagėnai; 1994—Dieveniškės; 1995—Vištytis).

² The intonations of long vowels (respectively, ie, uo) have not yet been sufficiently investigated. Although it is said of the eastern kausai, neighbors of the western Dzūkians, that they too “almost fail to distinguish /.../ the acute and circumflex intonations of stressed long vowels and diphthongs ie, uo” (Lietuvių 1970: 29), it seems premature to speak of their neutralization in the word stem (cf. Grinaveckis 1991: 118—119). All the more so since the southern western Highlanders continue to distinguish acute from circumflex even in unstressed syllables (Girdenis 1973: 74—75; Zinkevičius 1983: 121—122).

149

čių tarnių arba, pasak P. Skardžiaus (1935: 40), ”tai galėjo būti ir vilniečio korektoriaus padaras”. Nors galūnėje sukirčiuotas formas (bent jau būdvardžius) P. Skardžius ir priskyręs prie korektūros riktų, tačiau jo nuomonės neparemia senųjų raštų įvardžiai, skaitvardžiai, nė patys būdvardžiai.

That words with ending stress cannot be regarded as either accidental or mistaken is shown by M. Daukša’s pronouns kokìs, vìs, tokìs, ~ìs: the consistently end-stressed singular instrumental *kokìa (kokía, kokė), *tokìa (tokía) or plural accusative *kokìas (kokės, kokìas, kokėjg), *tokìas (tokės) alongside tókias (tókias), *tokìas (tokìas), alongside tókùs (tókùs), together with other case forms, attests that they were oxytones (Skardžius 1935: 198). The baritone pronouns had no corresponding forms with ending stress.

Numerals, like pronouns, also seem to have retained a paradigm with ending stress. This is shown by the plural accusative forms in M. Daukša’s writings, *keturìas (kieᵃ turìas, keturés), *septynìas (septinés) (Skardžius 1935: 183), as well as by the non-neuter forms of numerals in the 1605 anonymous catechism, *pirmà (pirmá), *ant(a)rà (ontárá), *trečià (Trecìá) alongside trẽčia (Trẽcia), *ketvirtà (Ketwirtá) (Zinkevicius 1975: 63).

§5. Given the small number of examples, it is difficult to recognize *gardžius, *gyvùs, *pilnù as relics of ending stress—especially since K. Būga is inclined to explain the form *gardžius as a form from another dialect (Būga 1961:77). Nevertheless, there is indirect evidence.

First, there is the masculine nominative plural of disyllabic acute-stem adjectives, baltì (< *bált-ie), stressed on the ending—exactly like pirmì (< *pirm-ie) or kokì (< *kák-ie)—and the feminine nominative singular baltà (< *bált-ā), like pirmà (< *pirm-ā) or kokià (< *kák-(j)ā). Thus, before Saussure’s and Fortunatov’s laws took effect, these forms were indistinguishable in their accentuation—with two acute accents—from the singular instrumental and plural accusative; cf. báltu/baltù (< *bált-uo), báltus/baltùs (< *bált-uos) (Mikulienė 1993: 22).

By M. Daukša’s time, Saussure’s and Fortunatov’s laws had already affected baritones with circumflex stems; cf. *drãsus (drãsus), drãsu (drãsu), but drąsiùs (draṡiùs), *graudùs (graudùs), graudùmi (gráudumi), but graudžià (i.sg. graudžiá), *sunkùs (súkus), sunkù (súku), but sunkìus (sùkius) (Skardžius 1935: 143–144). Then what, if not a trace of root stress, should we consider *sveikì (ſweikì), sveikùs (ſweikùs) alongside sveikà (ſweiká), sveikàs (ſweikás), sveikù (i.sg. [weikù) (Skardžius 1935: 151)?

Apparently, neuter adjectives should also be regarded as relics of baritone accentuation; cf. *drãsu, *sunkù, *teísu (téis[u), *vãrgu (wãrgu) alongside the considerably rarer form vargù (wargù) (Skardžius 1935: 143–144). The same accentuation—bjaũru (biaũru), daílu, graũdu, patógu, sunkù (ſunkù)—is found in F. Kuršaitis’s dictionary (Kurschat 1883).

Thus, it is plausible that disyllabic adjectives, like pronouns or numerals, had two accentual paradigms—root-stressed and ending-stressed.

§6. The 1605 catechism is mentioned for a reason: it is conjectured that its translator came from the southern part of the Vilnius dialect area (Zinkevicius 1968: 109 ff., 1975:

150

4). In terms of accentuation features, the language of the anonymous catechism is close to the writings of M. Daukša; therefore, we must acknowledge either that the aforementioned “Vilnius proofreader” had a strong hand in them (cf. also K. Būga’s (1961: 77) statement: “Daukša’s Postil is a translation by several people”), or that the accentuation of the catechism translator’s dialect had not at that time diverged greatly from the dialect of M. Daukša’s native region. For all the accentuation variants found in old writings have survived to this day in the southern Aukštaitian area.

First, these are the universally used pronouns with ending accentuation. In various variants — tokìs, ~ìà; tokiàs, ~ià; tokỹs, ~ià (alongside tókis, ~ia or tókias, ~ia) and kokìs, ~ià; kokiàs, ~ià; kokỹs, ~ià (alongside kókis, ~ia or kókias, ~ia) — they have been recorded in the environs of Gervėčiai (Kardelytė 1975: 69), Lazūnai (Vidugiris 1959:208; Petrauskas, Vidugiris 1985: 124, 268; 1987: 5, 6 ff.), Zietela (Vidugiris 1959: 208, 209; 1960: 122; Zinkevičius 1966: 318), Varanavas (Grinaveckienė 1994: 96, 106), Dieveniškės (Grinaveckienė 1993: 166), Druskininkai (Naktinienė, Paulauskienė, Vitkauskas 1988: 161, 410), Punskas (Mikalonienė, Morkūnas 1988), and elsewhere.

Second, the neuter forms widespread in the southern Aukštaitian dialects, which have retained root stress to this day — adjectives of the sunkù type (Vidugiris 1959: 206; Kardelytė 1975:54; Petrauskas, Vidugiris 1985: 250; 1987: 26, 31 ff.). Since even in the Kapsai–Zanavykai area they have two types of accentuation (on both the root and the ending)—for example, baĩsù, baĩgù, bjáurù, graũdù, nýkù, skaũdù, svárbù, sunkù... (Senkus 1960: 142)—it can be argued that the ending-stressed variant sunkù is newer.

Third, the numerals of the Lazūnai dialect vienù, vienùs; aštuonìas (alongside aštuóni), septynìas (alongside septýnis) (Petrauskas, Vidugiris 1987: 10, 19, 20, 33, 34, 38, 57), confirm the authenticity of the numeral forms in M. Daukša’s writings.

Thus, it would seem that the Dzūkians have “confused” only the accentuation of nouns, adjectives, and numerals inflected like adjectives?

§7. A morphological (phonological) leveling of endings then seems entirely plausible (cf. Girdenis 1991): once Saussure’s and Fortunatov’s law had taken hold, the endings of these cases began to be stressed even in words with acute roots. This applies first of all to adjectives; cf. sveĩkas: sveĩkù (<*sveĩk-úo), sveikùs (<*sveĩk-úos) and báltas: báltu/baltù (<*bált-úo), báltus/baltùs (<*bált-úos). By analogy with them, exactly as in the case of plural nominative accentuation—cf. sveĩkì (<*sveĩk-ìe) and báltì (<*bált-ìe)—the fixed-initial-accent endings ie and uo may have played a role; their acute accent may have been determined by the strong position of the ending (Mikulėnienė 1992: 14).

The presence of homonymous inflections in the southern Aukštaitian dialects probably helped acute endings become established (Mikulėnienė 1993: 22); cf. the singular instrumental forms of masculine and feminine adjectives — báltu/baltù (<*bált-úo) báltu/baltù (<*bált-ú <*bált-ą(n)).

Generalized acute accentuation of endings may also have been applied to nouns—

151

nouns. Moreover, since only the position of the accent distinguishes, for example, the singular instrumental vilkù of an ǒ-stem noun from the singular dative vìlkų, whose ending -u is characteristic of the southern Aukštaitian dialects (Zinkevičius 1966: 208, map no. 75; Kazlauskas 1968: 148; Lietuvių 1991: 40, map no. 36).

§8. The tendency to accentuate the accusative plural of root-accented nouns on the ending may have been prompted by the collective plural, which remains alive to this day in the Dzūkian dialects (Mikulėnienė 1994: 18). Formally, the plural forms of ǒ-stem nouns and the forms of nomina collectiva with -ai coincided (Stundžia 1981: 191 ff.; 1994: 21; Ambrazas 1992: 36–37; Kortlandt 1993: 46–47). Thus, forms with ending stress such as beržùs, grūdùs, kelmùs, pėdùs, tvartùs... (Vidugiris 1959: 207; Petrauskas, Vidugiris 1985: 71,90; 1987: 9, 13 ff.) can be treated as cases of paradigm leveling (cf. beržai, grūdai, kelmai, pėdai, tvartai...).

This hypothesis is supported by nouns of pluralia tantum, which are stressed according to the same pattern; cf. the Lazūnai dialect forms akėčiàs, pavalkùs (Senkus 1959: 217, 222), marškiniùs, pelėnùs (Petrauskas, Vidugiris 1987: 5 tt). In general, it seems that in the Lazūnai dialect, final-stress accentuation of polysyllabic nouns was generalized; cf. audeklùs, pjautuvùs, sėmenùs, vailokùs (Senkus 1959: 217, 225), avižùs, riešutùs, obuoliùs, samanàs, valgymùs, voveràs... (Petrauskas, Vidugiris 1987: 5, 7 tt).

It is true that the root-accented variants recorded alongside these forms—for example, márškinius (Senkus 1959: 221), áudeklus (Petrauskas, Vidugiris 1985: 20), or avìžas (Petrauskas, Vidugiris 1987: 42)—do not allow us to claim unreservedly that ending accentuation was older for these nouns, but neither do they rule out this possibility (Kazlauskas 1968: 35–36).

§9. Generalized ending accentuation in the plural may also have affected singular forms that retained stem accentuation. In these forms, a circumflex appears in the root or suffix as a reflection of this influence; circumflex accentuation is generally characteristic of retracted accent (Zinkevičius 1966: 40–41). Cf. krėslas (3)/krėslas (4), mókslas (4), ǒras (4), žìedas (4)... (Naktinienė, Paulauskienė, Vitkauskas 1988), or derivatives with the suffix -ėjas, ~à: dė́jėjas, ~à (4), dengė́jas (4) mezgė́ja (4), siuvė́jas, ~à (4)... (Naktinienė, Paulauskienė, Vitkauskas 1988; see also Kurschat 1883).

Galbūt taip paaiškinami ir P. Skardžiaus (1935: 40) minėtieji tarminiai variantai — daiktas, veidas, žiedas. Plg.:

singular dáiktas: plural dáiktai: collective plural daiktaĩ → singular daĩktas daiktùs

If the circumflex accentuation of these singular forms is regarded as secondary, the possibility of circumflex metatony is acknowledged. The unchanged root intonation of certain words in the dialects (e.g., káulas, šáukštas ar tvártas) shows, however, that this process was not complete in every case; cf.:

singular káulas: plural káulai: collective plural kaulaĩ → singular... kaulùs

§10. The accentuation variants mentioned above continually compete in spoken language, so linguists sometimes find it difficult to determine the accentual pa-

152

paradigm: singular forms may exhibit baritone accentuation, while plural forms may exhibit mobile stress; cf. the Lazūnai dialect tvārtas (1): tvārti, tvārtan (Senkus 1959: 216—217) and tvartai (3): tvartóp, tvartai (Petrauskas, Vidugiris 1985: 273). Sometimes a mobile paradigm with an acute root is recorded with an exception; cf. kaulai (3): acc. sg. kaũlų, kaulai, but kaulùs: Ānas kaulùs palāužė džirbdamõs (Petrauskas, Vidugiris 1985: 112).

Thus, in the southern Aukštaitian dialects, the generalized “strong” ending position of the singular instrumental and plural accusative is determined diachronically more by morphological than by phonetic circumstances. Therefore, the alternating relationship between the root tone and stress position in certain forms of disyllabic words, caused by the influence of the paradigm, cannot be regarded as the most important and sole indicator of an accent paradigm.

REFERENCES AND SOURCES

A m b r a z a s S. 1992: On the development of nomina collectiva in the Baltic languages. — Linguistica Baltica 1, 35—48.

B ū g a K. 1961: Lietuvių kalbos žodynas. — Būga K., Rinktiniai raštai 3, Vilnius: Valstybinė politinės ir mokslinės literatūros leidykla, 9—483.

G i r d e n i s A. 1967: Kodėl įsigali ”galūninis” vardažodžio kirčiavimas. — Mokslinė aukšt. m-klų dėstytojų konferencija lietuvių kalbos klausimais. Programa ir pranešimų tezės. VU, Vilnius, 6.

G i r d e n i s A. 1973: Zapadnolitovskije i drevneprusskije slogovyje akcenty. — Baltijskije jazyki i ich vzaimosvjazi so slavjanskimi, finougorskimi i germanskimi jazykami. Tezisy dokladov naučnoj konferencii, posvjaščennoj 100-letiju so dnja roždenija akad. Ja. Endzelina, Riga, 71—75.

G i r d e n i s A. 1991: Žodžio galo priegaidžių kilmė ir raida lietuvių kalboje. — VI tarptautinis baltistų kongresas, 1991 m. spalio 2—4 d. Pranešimų tezės, Vilnius, 33—34.

G r i n a v e c k i e n ė E. 1993: Šaltinių šnektos tekstai. — Periferinės lietuvių kalbos tarmės (Lietuvių kalbotyros klausimai 30), 134—180.

G r i n a v e c k i e n ė E. 1994: Lietuvių tarmių tekstai 1. Balatna (Varanavas), Vilnius: Mokslo ir enciklopedijų leidykla.

G r i n a v e c k i s V. 1991: Apie vakarų dzūkų tarmės kirčiavimą. — Grinaveckis V., Lietuvių kalbos tarmių kirčiavimo klausimai. VPI, Vilnius, 110—119.

G r u m a d i e n ė L. 1988: Sociolingvistinis vilniečių lietuvių kalbos tyrimas: konsonantizmas ir akcentuacija. — Lietuvių kalba ir bilingvizmas (Lietuvių kalbotyros klausimai 27), 132—149.

G r u m a d i e n ė L. 1992: Kirčiuočių mišimas Punsko jaunimo kalboje. — Lietuvių kalbos tarmės ir jų tyrinėjimai. Praeitis ir dabartis. Antanui Saliui (1902 07 21—1972) paminėti. Konferencijos pranešimų tezės. 1992 m. lapkričio 5—6 d., Vilnius, 11.

G r u m a d i e n ė L. 1994: Keletas naujesnių Punsko šnektos linksniavimo ir kirčiavimo ypatybių. — Lietuvių kalbos tarmės ir jų tyrinėjimai (Lietuvių kalbotyros klausimai 34), 97—106.

I l l i č - S v i t y č V. M. 1963: Imennaja akcentuacija v baltijskom i slavjanskom, Moscow: Nauka.

K a r d e l y t ė J. 1975: Gervėčių tarmė (fonetika ir morfologija), Vilnius: Mintis.

K a z l a u s k a s J. 1968: Lietuvių kalbos istorinė gramatika, Vilnius: Mintis.

K o r t l a n d t F. 1993: Such cold mornings. — Baltistica 28(1), 45—48.

K u r s c h a t F. 1883: Wörterbuch der Litauischen Sprache, Halle.

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

Lietuvių 1991: Lietuvių kalbos atlasas 3, Morfologija, Vilnius: Mokslas.

153

L i p s k i e n ė J., V i d u g i r i s A. 1967: The Dieveniškės dialect. — Lietuvių kalbos gramatinė sandara (Lietuvių kalbos kultūros klausimai 9), 183—222.

M i k a l o n i e n ė B., M o r k ū n a s K. 1988: Punsko valsč. Kreivėnų k. tarminiai tekstai. (Rankraštis saugomas Lietuvos kalbos instituto Kalbos istorijos ir dialektologijos skyriuje. Reg. Nr. 63).

M i k u l ė n i e n ė D. 1992: Priegaidžių raidos klausimu. — Jono Kazlausko diena: istorinės gramatikos dalykai. Konferencijos programa ir tezės, Vilnius, 13—14.

M i k u l ė n i e n ė D. 1993: Dėl pietų aukštaičių ir rytų aukštaičių vilniškių akcentinių paradigmų mišimo. — Lietuvių kalba: tyrėjai ir tyrimai. Jurgui Geruliui (1883 08 13— apie 1945) paminėti. Konferencijos pranešimų tezės, Vilnius, 21—22.

M i k u l ė n i e n ė D. 1994: Dėl ō-kamienių daiktavardžių raidos. — Lietuvių kalba: tyrėjai ir tyrimai. Pranui Skardžiui (1899 03 26—1975 12 18) paminėti. Konferencijos pranešimų tezės. Vilnius, 17—18.

M o r k ū n a s K. 1969: Rytų aukštaičių pietinės tarmės morfologija. — Lietuvių kalbos gramatikos tyrinėjimai (Lietuvių kalbotyros klausimai 11), 107—146.

N a k t i n i e n ė G., P a u l a u s k i e n ė A., V i t k a u s k a s V. 1988: Dictionary of the Druskininkai Dialect, Vilnius: Mokslas.

P e t r a u s k a s J., V i d u g i r i s A. 1985: Dictionary of the Lazūnai Dialect, Vilnius: Mokslas.

P e t r a u s k a s J., V i d u g i r i s A. 1987: Texts of the Lazūnai Dialect, Vilnius: Mokslas.

S e n k u s J. 1959: Lazūnų tarmės tekstai. — Lietuvių kalbotyros klausimai 2, 215—230.

S e n k u s J. 1960: Kapsų-zanavykų tarmių būdvardžio ir skaitvardžio kait ybos bruožai. — Lietuvių kalbotyros klausimai 3, 133—159.

S k a r d ž i u s P. 1935: Daukša’s Accentology, Kaunas.

S k a r d ž i u s P. 1968: Lithuanian Accentuation, Chicago.

S t u n d ž i a B. 1984: Lietuvių kalbos dviskiemenių daiktavardžių kirčiavimo variantai. — Kalbotyra 35(1), 86—92.

S t u n d ž i a B. 1981: Daugiskaitiniai asmenvardiniai oikonimai ir jų reikšmė linksniavimo bei kirčiavimo sistemų rekonstrukcijai. — Lietuvių onomastikos tyrinėjimai (Lietuvių kalbotyros klausimai 21), 185—198.

S t u n d ž i a B. 1994: Daiktavardžio kamienų bei giminių variantai baltų kalbose (ide. ir bendroji baltų bei slavų kalbų leksika). — Baltistica 27(2), 13—30.

V i d u g i r i s A. 1959: Kai kurios Zietelos tarmės ypatybės. — Lietuvių kalbotyros klausimai 2, Vilnius, 195—213.

V i d u g i r i s A. 1960: Dictionary of the Zietela dialect. — Questions of Lithuanian Linguistics 3, 113—133.

Z i n k e v i č i u s Z. 1966: Lithuanian Dialectology, Vilnius: Mintis.

Z i n k e v i č i u s Z. 1968: The dialect of the 1605 catechism. — Baltistica 4(1), 109—116.

Z i n k e v i č i u s Z. 1975: From the History of Lithuanian Accentology: The Accentuation of the 1605 Catechism, Vilnius.

Z i n k e v i č i u s Z. 1979: Retraction of the accent and accentuation paradigms. — Kalbotyra 30(1), 90—93.

Z i n k e v i č i u s Z. 1983: On the development of Lithuanian accent and pitch accents. — Baltistica 19(2), 120—122.

154

ON THE QUESTION OF ACCENTUATION VARIANTS OF NOUNS IN THE SOUTHERN AUKŠTAITIAN DIALECTS

Summary

The Southern Aukštaitian dialects exhibit a relatively large number of accentuation variants. Many words and forms are accented differently here than in the standard language, cf.: vienùs / vienùs, pedùs / pėdùs, kaulai / kàulai, among others. Accentuation varies particularly in the singular instrumental and plural accusative forms.

Since most accentuation variants occur in the Southern Aukštaitian dialects, in the border region of Lithuanian, these variants are often attributed to the influence of the Slavic languages. Although Slavic influence is also apparent, the mixing of accentuation paradigms should not always be regarded as a new phenomenon.

This article advances the view that diachronic conditions, too, may have led to the emergence of accentuation variants in the Southern Aukštaitian dialects. A comparison of the relevant material from old writings (M. Daukša’s Postil and Catechism, the anonymous Catechism of 1606, and F. Kurschat’s dictionaries) with later research data from the Aukštaitian dialects (Lazūnai, Gervėčiai, Dieveniškės, Druskininkai, Punskas) shows that all accentuation variants found in the old writings have been preserved in these dialects. Consequently, they are not new.

Several possible causes for the emergence of accentuation variants are identified. First is the morpho(pho)nological leveling of endings, cf.: sveìkas: sveikù (<*sveik-úo), sveikùs (<*sveik-úos) and báltas: báltu / baltù (<*bált-úo), báltus / baltùs (<*bált-úos). The accentuation of these endings is analogous to that of plural nominative forms (cf.: sveikì and baltì). In addition, homonymous inflections may have contributed to the stabilization of ending stress; cf. in these dialects the identical singular instrumental forms of masculine and feminine nouns báltu / baltù and báltu / baltù, as well as the singular dative of masculine o-stem nouns vìlkù and the singular instrumental vilkù. Third, the accentuation of plural accusative forms of nouns must have been influenced by the forms of Nomina Collectiva that are still current in these dialects. Once the collective nouns in -ai became established in usage, they coincided completely with the plural of o-stem nouns.

It is plausible that the plural’s ending stress could have affected the stem-syllable accent of singular forms, cf.: Sing. dáiktas: Pl. dáiktai: Nom. Coll. Pl. daiktai → Sing. daiktas

Thus, from a diachronic perspective, the unified “strong” position of the ending in the singular instrumental and plural accusative in the Southern Aukštaitian dialects may have been determined predominantly by morphological conditions, and less by phonetic ones.

155