Historical phonology in Lithuanian and Balto-Slavic: relative chronology of Monosyllabic Circumflexion Yoko Yamazaki Historical phonology in Lithuanian and Balto-Slavic: relative chronology of Monosyllabic Circumflexion Yoko Yamazaki Baltic Languages, Stockholm university Phonology Colloquium, Dec. 4, 2015
40
Embed
Historical phonology in Lithuanian and Balto-Slavic ...yy/hp/documents/phon-coll.pdfHistorical phonology in Lithuanian and Balto-Slavic: relative chronology of Monosyllabic Circum
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Historical phonology in Lithuanian andBalto-Slavic: relative chronology of
Monosyllabic Circumflexion
Yoko Yamazaki
Baltic Languages, Stockholm university
Phonology Colloquium, Dec. 4, 2015
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Lithuanian as an Indo-European language
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Monosyllabic Circumflexion
In Lithuanian (also in other Balto-Slavic languages), longvowels in monosyllabic words exhibit a circumflex toneinstead of the expected acute.
cf. Hanssen (1885), Zinkevicius (1980–81: II, 161ff.),Rasmussen (1999)
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Acute and circumflex tones ILithuanian distinguishes two types of tones on bimoraicsyllable nuclei (i.e., long vowels, diphthongs, and mixeddiphthongs [tautosyllabic vowel + resonant]);
(i) acute (falling) 〈 ´V〉= /VV/
[μ μ]σ
H
(ii) circumflex (rising) 〈 ˜V〉= /VV/
[μ μ]σ
HBlevins (1993)
e.g., vyras ’man’ vynas ’wine’gauti ’to get’ naujas ’young’varna ’crow’ varnas ’raven’
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Acute and circumflex tones II
Historical Laws that affected acute vowels:
I Saussure’s Law: a high tone of a non-acute syllable isattracted to its immediately following acute syllable.
I geras (AP4; nom. sg. m.) ‘good’ ∼ geruosius(acc. pl. m. def.)dievas (AP4; nom. sg.) ‘god’ ∼ dievus (acc. pl.)
I vyras (AP1; nom. sg.) ‘man’ ∼ vyrus (acc. pl.)
I Leskien’s Law: a word-final acute long vowel isshortened.
I gerıe-ji (nom. pl. m.) ‘the good’ ∼ gerı (nom. pl. m.)I gero-ji (nom. sg. f. def.) ‘the good’ ∼ gera
(acc. pl. m.)I ger´u-ju (nom.-acc. du. m.) ‘the good’ ∼ geru
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Acute tone from a historical viewpoint IA systematic correspondence of Lithuanian acute tone andlong vowels in the cognate words in other Indo-Europeanlanguages:
de Saussure (1894: 492ff.) hypothesized that ’old’ longvowels received the acute tone in Lithuanian.Later, Kury lowicz (1948) showed that the same applies tolong diphthongs as well:
Lith. vilkas Skt. vr˙ka- ’wolf’ (< PIE *u
“r
˚kw o-)
Lith. vılna Skt. ´urna ’wool’ (< PIE *u“l
˚Hneh2)
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Acute tone from a historical viewpoint II
Kury lowicz (1948) assumed that long diphthongs, as well aslong syllabic sonants, received the acute accent when theywere shortened in the Proto-Balto-Slavic stage (Osthoff’sLaw). He proposed that those long diphthongs and longsyllabic sonants are generated as the result of compensatorylengthening caused by laryngeal loss: *R
˚H, *VRH > *R
˚,
*VR > Lith. ıR, VR:
I Lith. v ılna, Skt. ´urna- < PIE *u“l
˚H-neh2 ‘wool’;
Lith. g ırti , past. pass. part. g ırtas, Skt. g urtas‘pleasant,’ Lat. gratus ‘beloved’ < PIE *g r
˚tos <
*gr˚
Htos
I g erti ‘to drink’ < PIE *gw erH- vs. penkı, f. penkios,Skt. panca, Gk. πέντε ‘five’ < PIE *penkw e-.
Christian Stang (1957) further made a contribution to showthe systematic correspondence of Slavic and Baltic tones.
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Acute tone from a historical viewpoint IIII Lith. AP 1 (d ´umai ; Latv. d umi ‘smoke’ (nom.pl.)) ∼
I Lith. AP3 (galva, g alva (sg. acc.) ‘head’; Latv. g alva)∼ (c) mobile stress [SCr. glava, gl“avu (acc.sg.)]
I Lith. AP4 (draugas ‘friend’; Latv. dr augs) ∼ (c)mobile stress [SCr. dr“ug ]
Thus, the tonal contrast “acute vs. non-acute” isreconstructed in PBS. It is generally understood that theacute tone is a marked feature, while non-acute (orcircumflex) is unmarked. However, the phonetic reality ofsuch acute tone or acute feature is unknown. Stang (1966:137) and Jasanoff (2004: 251) speculate it to be a glottalicfeature comparable to present day Danish stød .
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
The categories where MC is observed
I the 3rd person future forms of monosyllabic root
I Distribution/condition of MC in each categoryNot all the monosyllabic forms have the circumflextone, i.e., there are exceptions: lıs ‘will rain (3p.)’ (∼lyti), bus ‘will be (3p.)’ (∼ b´uti)
I Relative Chronology of MCRasmussen (1999, 2007), Villanueva Svensson (2011)maintain that MC can be Proto-Balto-Slavic, whileKortlandt (2014) recognizes two chronological layers ofMC (Proto-Balto-Slavic, and Lithuanian Aukstaitiandialects).
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
3rd person future forms: irregularity I
future paradigm of duoti ‘give’
sg. 1. duosiu du. 1. duosiva pl. 1. duosime2. duosi 2. duosita 2. duosite3. duos — —
future paradigm of b´uti ‘be’
sg. 1. b´usiu du. 1. b´usiva pl. 1. b´usime2. b´usi 2. b´usita 2. b´usite3. bus — —
future paradigm of Latv. duot ‘give’
sg. 1. duosu pl. 1. duosim2. duosi 2. duosit3. duos —
I Senn (1966: 231ff.), Kazlauskas (1968: 104): the acutelong vowels y and ´u are regularly shortened by Leskien’sLaw in the word final position, including inmonosyllables. Therefore the shortening is regular inthose 3p. future forms. Some of them remain withcircumflex long vowels due to the expected homonymicclash, e.g., vys ‘will droop’ vs. vıs (vısti ‘to fall apart’),si˜us ‘will sew’ vs. sius (siusti ‘to rage’).
I Zinkevicius (1984–95: II, 161ff.): MC is regular outcomeof the 3p. future forms. Some gained the short vowelsthrough the analogy from their polysyllabic variants,e.g., bus ‘will be’ from nebus ‘will not be’ < *neb´us, inaddition to the avoidance of homonymic clash.
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
3rd person future forms: irregularity VI Petit (2002): Leskien’s Law did not shortened ıe and uo
but y and ´u in general. Therefore, the shortened futureforms are regular for the monosyllabic root in y and ´u.Some remained with a long circumflex root because oftheir preterit forms with a long vowel (e.g., gnybs ‘willpinch’ ← gnybo ‘he/they pinched’).
I Villanueva Svensson (2011: 19): MC was probablyregular among all the 3p. future forms. For thoseshortened, the acute root vocalism was restored forsome reason.
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
3rd person future forms: irregularity VI
Questions:
I Why some verbs like gnybs – gnybti ‘pinch,’ znybti –znybs have their future forms with long circumflexvowels, although there are no gnıbti or znıbti .
I Why copying the vocalism of preterit forms only to the3p. future forms? Motivation??
I If Villanuva Svensson’s opinion is right, what could bethe condition of the alleged “restoration of the acutetone” to the future forms?
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
3rd person future forms: distribution in thepresent and preterit formation I
The paradigms of the verbs with the shortened 3rd personfuture forms have:
I nasal-infix presentI a-preterit.
For example (infinitive, present, preterit – 3p. future):
3rd person future forms: distribution in thepresent and preterit formation II
the verbs which have circumflex tone in their 3p. futureforms do not have nasal-infix present.
I vyti, veja/vıja/vyna, vıjo – vys ‘to drive, wind’
I gr´usti, gr´uda, gr´udo – gr˜us ‘to crush’
I gnybti, gnyba, gnybo – gnybs ‘to pinch, bite’
I znybti, znybia, znybe – znybs ‘to pinch, to tweak’
I dygti, dygsta, dygo – dygs ‘to spring, shoot’
I klysti, klysta, klydo – klys ‘to be mistaken’
I slysti, slysta, slydo/slıdo – slys ‘to slide’
I lysti, lysta, lyso – lys ‘to become thin’
I lyzti, lyzta, lyzo – lys ‘to slacken’
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
3rd person future forms: distribution in thepresent and preterit formation III
Historical background:
I Introduction of a root-final semivowel to the nasal infixpresent: the expected nasal-infix form of the verbswould be 7lına- ‘rain’ (< *li-n-H-o-), 7puna- ‘rot’ (<*pu-n-H-o-), etc.Proportion of Analogy (Gorbachov 2007: 167):minga ‘sleep(s)’ : *mıga ‘slept’ = X ‘rain(s)’ : *lıja‘rained’X = *linja.
I regular nasal loss:
Vn > V > V /
r, l, m, n,j, v,s, s, z
.
*linja > lyja
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
3rd person future forms: distribution in thepresent and preterit formation IV
I Both nasal infix present and a-preterit are built to thezero-grade of the root, and the verbs which have anasa-infix present almost always have a-preterit as theirpreterit paradigm, cf. Stang (1942: 132ff.), Gorbachov(2007: 152ff.). Semantically, they are inchoatives.
3rd person future forms: distribution in thepresent and preterit formation V
I Their infinitives (dative or locative singular of ti-stem)kept their old zero-grade formation, e.g., *gw ih3-tei >Lith. gyti , *liH-tei > Lith. lyti , etc.Usually, the infinitive roots were remodeled afterpreterit stem (Stang 1942: 122).
I In addition, many of the verbs of this type have good IEroot etymology, and some of them belong to theintransitive inchoative thmatic verbs established forNorthern Indo-European languages (i.e., Germanic,Baltic and Slavic) in Grobachov (2007: 159ff.).
I Lith. b˜uva ∼ OCS bodo ‘will be’I Lith. p˜uva ∼ PG *f˘uni/a- (cf. ON funar ‘rots’)I Lith. slyja ∼ PG *hlini/a- (cf. OE hlinian, OHG hlinen
‘to lean’)
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
3rd person future forms: distribution in thepresent and preterit formation VI
The historical background of the future formation: PIEs-future
Skt.√
da ‘to give’
sg. 1. dasy´ami du. 1. dasy´avas pl. 1. dasy´amas2. dasyasi 2. dasyathas 2. dasyatha3. dasyati 3. dasyatas 3. dasyanti
Gk. δίδωμι ‘to give’
sg. 1. δώσω du. 1. — pl. 1. δώσομεν
2. δώσεις 2. δώσετον 2. δώσετε
3. δώσει 3. δώσετον 3. δώσουσι
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
3rd person future forms: distribution in thepresent and preterit formation VII
I The origin of s-future may be desiderative formation in-(h1)se/o-, -(h1)si
“e/o-, and -(h1)s-, the last one of
which is continued as Baltic future (VillanuevaSvensson 2010: 218ff.).
I Endzelıns (1971: 234), further advanced in Schmalstieg(1958: 120ff.) and Jasanoff (1978: 103ff.), hasproposed that the origin of the -i- element in the Balticfuture suffix in 1sg./du./pl., 2sg./du./pl. forms is thenow disappeared athematic 3pl. ending *-n
˚t(i) which
developed to PBS *int(i). The implication of this ideais that the 3pl. (weak) form had the accent on the root,i.e., the paradigm was probably in Narten type:S: R(´e)-S(∅)-E(∅), W: R(e)-S(∅)-E(∅).
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
3rd person future forms: distribution in thepresent and preterit formation VIII
I Therefore, the zero-grade vocalism of the shortened3p. future forms must be secondarily introduced whenthe Baltic future formation grammar, i.e., building thefuture stem to the infinitive stem, was introduced in theBaltic language.
I On the other hand, the full-grade vocalism of some 3rdperson future forms must be the inherited vocalism,e.g., duoti – duos ‘will give,’ d´eti – d˜es ‘will place,’ stoti– stos ‘will stand,’ etc., and their full-grade infinitivestem may be rather secondary.
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
3rd person future forms: distribution in thepresent and preterit formation IX
I the condition of the restoration of the acute tone wasthat the verbs belong to an inchoative thematic groupwith inherited zeso-grade infinitive, nasal-infix present,and a-preterit. It is not motivated by the avoidance ofhomonymic clash at Lithuanian stage.
I This means that the inherited future forms show theresult of MC, while the future forms with the secondaryvocalism show the shortening. This indicates that MC isan old sound change, which can be estimated no laterthan Proto-Baltic.
I Then, how old is MC...?At least, 3rd person future forms show that it was after”Winter’s Law,” a lengthening of a vowel before anon-aspirated voiced consonant, e.g.,Lith. b˜egs (<*b´egs < *bhegw s) ∼ Lith. b´egti ‘to run,’ OCS bezati ,Gk. φέβομαι ‘to flee,’ PIE *bhegw -.
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Two Latvian reflexes of PIE root nouns I
What’s a root noun?
I a category of nouns which have a structure of a rootand an ending, without any suffix (R-E), cf. Schindler(1972: 31ff.).
I Their possible ablaut patters were
I Acrostatic:S: R(o)-E(∅); W: R(e)-E(∅)
I Mobile:S: R(e)-E(∅); W: R(∅)-E(e)orS: R(a)-E(∅); W: R(∅)-E(e)
Why root nouns?
I For a monosyllabic root (which was quite a commoncase with Indo-European), the nominative singular andpossibly accusative singular case forms weremonosyllabic with a long vowel.
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Two Latvian reflexes of PIE root nouns II
I In PBS, PIE root nouns shifted to i-stems, due to theaccusative endings *-m
˚(sg.) and *-m
˚s (pl.) of animate
genders developing to *-in and *-ins, which wereidentical to the accusative case endings of i-stems(Vaillant 1958: 174ff.; Stang 1966: 219). As a result,they became disyllabic forms.
I By examining how root nouns obtained their tones, it ispossible to establish the relative chronology of MC andthe shift of root nouns to i-stems in PBS (a moredetailed relative chronology)
There are two noteworthy reflexes of PIE root nouns inLatvian with the circumflex tone.
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Two Latvian reflexes of PIE root nouns IIII Latv. guovs ‘cow’ f.
Two Latvian reflexes of PIE root nouns VIOther reflexes of root nouns have the acute tone in the root:e.g., Lith. zverıs ‘wild animal’
Cognates:Baltic: Latv. zvers m. < PB *zverisSlavic: OCSzverž, SCr. zv‚ıjer , Sln. zv“er < PS *zv“erž m.
(c)PBS *zveris ← *zver < PIE *ghu
“er
other IE: Gk. θήρ, θηρός m., Lat. fera f. ‘wild beast’PIE: nom.sg. *ghu
“er-s (> *ghu
“´er), gen.sg. *ghu
“r
˚-es
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Two Latvian reflexes of PIE root nouns VII
phonological/morphological shifts of the paradigmof ‘wild animal’ from PIE to late PBS
nom.sg. acc.sg. obl.PIE *ghu
“er-s (> *ghu
“er) *ghu
“er-m
˚*ghu
“r
˚-´
PBS loss of ablaut and palatalization of *gh
*zver *zver-m˚
*zver-´acute assignment to long vowels
and extension of syllabic resonant*zv´er *zv´er-im zver-´
MC*zv˜er *zv´er-im zver-´
generalization of accusative stem in i-stem*zv´eri-s *zv´eri-m *zveri-´
Osthoff’s Law: vacuous operation
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Conclusions
I The analysis of the 3p. future forms of monosyllabicacute stems indicates that the replacement of the futurestems with the infinitive stem in the zero-grade tookplace no later than Proto-Baltic.
I The analysis of the root nouns provides an indirectpiece of evidence for MC taoking place in PB.
I The relative Chronology:Winter’s Law (→ acute assignment) → MC →generalization of i-stem → Osthoff’s Law
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Bibliography I
Blevins, J. 1993. “A Tonal Analysis of Lithuanian NominalAccent” Language, 69(2), 237–273.
de Saussure, F. 1894. “A propos de l’accentuationlituanienne (intonations et accent proprement dit)”Memoires de la Societe de Linguistique de Paris, 8,425–446. Quoted from Ferdinand de Saussure Recueildes publications scientifiques de Ferdinand deSaussure, Geneve / Lausanne / Heidelberg: Societeanonyme des editions sonor, 1922, pp. 490–512.
Endzelıns, J. 1971. Comparative phonology and morphologyof the Baltic languages. Mouton, The Hague/Paris.Translated by William R. Schmalstieg and BenjaminsJegers from Janis Endzelıns Baltu valodu skanas unformas, Rıga 1948.
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Bibliography II
Gorbachov, Y. 2007. Indo-European Origins of the NasalInchoative Class in Germanic, Baltic and Slavic. Ph.D.thesis, Harvard University.
Hanssen, F. 1885. “Der Griechische circumflex stammt ausder ursprache” Zeitschrift fur vergleichendeSprachforschung, 27(Neue Folge 7), 612–617.
Jasanoff, J. 1978. Stative and middle in Indo-European.Institut fur Sprachwissenschaft, Innsbruck.
Jasanoff, J. 2004. “Acute vs. Circumflex: Some Notes onPIE and Post-PIE Prosodic Phonology” In Hyllested,A., Anders Richardt Jørgensen, Jenny Helena Larsson,& Thomas Olander (Eds.), Per Aspera ad Asteriscos,pp. 247–255. Institut fur Sprachen und Literaturen derUniversitat Innsbruck, Innsbruck.
Kazlauskas, J. 1968. Lietuviu kalbos istorine gramatika.Mintis, Vilnius.
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Bibliography III
Kortlandt, F. 2014. “Metatony in monosyllables” Baltistica,49(2), 217–224.
Kury lowicz, J. 1948. “Le degre long en balto-slave” Rocznikslawistyczny, 16, 1–14.
Petit, D. 2002. “Abregement et metatonie dans le futurlituanien: pour une reformulation de la loi de Leskien”Bulletin de la Societe de Linguistique de Paris, IIIC (I),245–282.
Rasmussen, J. E. 1999. “Die Vorgeschichte derbaltoslavischen Akzentuierung – Beitrage zu einervereinfachten Losung” In Selected papers onIndo-European linguistics, Part 2, pp. 469–489.Museum Tusculanum Press, University of Copenhagen.First published in: B. Barschel, M. Kozianka & K.Weber (Hrsgg.): Indogermanisch, Slawisch undBaltisch. Materialien des vom 21.–22. September in
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Bibliography IV
Jena in Zusammenarbeit mit der IndogermanischenGesellschaft durchgefuhrten Kolloquiums (=Slavistische Beitrage, Bd. 285), Munchen: Otto Sagner1992, pp.173–200.
Rasmussen, J. 2007. “The Accent on Balto-SlavicMonosyllables” In Kapovic, M., & Matasovic, R.(Eds.), Tones and theories: proceedings of theInternational Workshop on Balto-Slavic Accentology,pp. 29–38. Institut za hrvatski jezik i jezikoslovlje,Zagreb.
Rix, H., & et al. (Eds.). 2001. Lexikon der indogermanischenVerben. DR. Ludwig Reichert Verlag, Wiesbaden.
Schindler, J. 1972. “L’apophonie des noms-racinesIndo-europeens” Bulletin de la Societe de Linguistiquede Paris, 67, 31 – 38.
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Bibliography V
Schmalstieg, W. 1958. “The vocalism of the Lithuaniansigmatic future” The Slavic and East EuropeanJournal, 2(2), 120–129.
Senn, A. 1966. Handbuch der litauischen Sprache. CarlWinter Universitatverlag, Heidelberg.
Stang, C. S. 1942. Das slavische und baltische Verbum. IKommisjon hos Jakob Dybwad, Oslo.
Stang, C. S. 1966. Vergleichende Grammatik der baltischenSprachen. Universitetsforlagedt, Oslo.
Stang, C. 1957. Slavonic accentuation. Universitetsforlaget,Oslo.
Stang, C. 1965. “Indo-Europeen *gw om, *d(i)i“em” In
Symbolae linguisticae in honorem Georgii Kury lowicz,pp. 292–296. Zak lad Narodowy imienia Ossolinskich,Wroc law.
Historicalphonology inLithuanian andBalto-Slavic:
relative chronologyof MonosyllabicCircumflexion
Yoko Yamazaki
Bibliography VI
Vaillant, A. 1958. Grammaire comparee des langues slaves,Vol. II. Klincksieck, Paris.
Villanueva Svensson, M. 2010. “Baltic sta-presents and theIndo-European desiderative” IndogermanischeForschungen, 115, 204–233.
Villanueva Svensson, M. 2011. “Indo-European Long Vowelsin Balto-Slavic” Baltistica, 46(1), 5–38.
Zinkevicius, Z. 1984–1995. Lietuviu kalbos istorija, Vol.I–VII. Mokslas, Vilnius.