Disappearance of the auxiliary verb esti in the Lithuanian passive

PAWEŁ BRUDZYŃSKI University of Warsaw ORCID id: orcid.org/0000-0002-1850-0818 Fields of research: History of Baltic languages, morphology, syntax, semantics, voice, modality. DOI: doi.org/10.35321/all91-05 DISAPPEARANCE OF THE AUXILIARY VERB ESTI IN THE LITHUANIAN PASSIVE Pagalbinio veiksmažodžio esti išnykimas iš lietuvių kalbos neveikiamosios rūšies ANNOTATION The prototypical passive voice in Lithuanian is an analytical construction consisting of a passive past participle and the auxiliary verb būti ‘to be’. Nowadays, the conjugation of this verb in the present tense is regular, except for the suppletive form of the third person yra, an East Baltic innovation, and historically, its original form was esti. Although the development and uses of the verb yra are well described, its development as a passive auxiliary verb has not received enough attention. The paper aims to trace the disappearance of the form esti from the function of an auxiliary verb in the passive constructions based on Old Lithuanian texts from 1547–1816, represented by prints from three historical dialects of Lithuanian. KEYWORDS: Lithuanian (language), diachrony, passive, participle, auxiliary verb. ANOTACIJA Lietuvių kalbos neveikiamosios rūšies prototipas yra analitinė konstrukcija, sudaryta iš neveikiamosios rūšies būtojo laiko dalyvio ir pagalbinio veiksmažodžio būti. Dabartinėje lietuvių kalboje šio veiksmažodžio esamojo laiko asmenavimas yra taisyklingas, išskyrus supletyvinę 3-iojo asmens formą yra, kuri – rytų baltų kalbų inovacija, o istoriškai pirminė jos buvusi forma – esti. Nors veiksmažodžio yra raida ir vartosena gerai aprašyta, jo, kaip neveikiamosios rūšies pagalbinio veiksmažodžio, raidai skirta nepakankamai dėmesio. Straipsnio tikslas – remiantis 1547–1816 metų senosios lietuvių kalbos tekstais, kuriuos Straipsniai / Articles 117 PAWEŁ BRUDZYŃSKI reprezentuoja trijų istorinių lietuvių kalbos tarminių spaudiniai, atsek­ti formos esti, kaip atliekančios pagalbinio veiksmažodžio funkciją, išnykimą neveikiamosiose konstrukcijose. ESMINIAI ŽODŽIAI: lietuvių kalba, diachronija, neveikiamoji nuosaka, dalyvis, pagalbinis veiksmažodis. 1. INTRODUCTION In formulating universals related to the nature of the passive, one must be careful because it may differ in structure and function from language to language (Siewierska 1984; Shibatani 1985; Geniušienė 2016). Of course, it is possible to identify several common features constituting the canonical passive voice in European languages (Holvoet 2001: 366). However, it is not easy to find universals that are common to all languages. The concept of the passive voice itself is usually associated with its prototypical, personal use (Siewierska 1984: 28), where the sentence has a corresponding construction in the active voice, it is possible to determine the subject and the subject of the passive construction corresponds to the patient of the active construction. However, in this paper, the passive is understood broadly as the so-called passive family (Nau et al. 2020), i.e., several different predicative constructions with a passive participle, which can be grouped based on formal and functional parameters, including personal, impersonal, actional, resultative or modal constructions. The prototypical passive voice in Lithuanian is an analytical construction consisting of a so-called passive past participle (t-participle) or passive present participle (m-participle) and the auxiliary verb būti ‘to be’. Nowadays, the conjugation of this verb in the present tense is regular, except for the suppletive form of the third person yra, an East Baltic innovation, and historically, its original form was esti. The history of the new, suppletive form has been extensively documented (Stang 1947: 1970; Ford 1967; Ostrowski 2017a; 2017b; 2021). The verb esti, originally an existential verb, was replaced by the verb yra and now serves only a habitual function, similar to the form būna ‘usually is’ (Genialiausios mintys esti paprasčiausios. ‘The most genius ideas usually are the simplest ones.’). The original form of yra should be reconstructed as *i̯-r-ã, as indicated by the Old Baltic and dialectal forms (Endzelin 1922: 556; Ostrowski 2017b: 167, 2021 66–67). Etymologically, it is a formation based on the stem of the anaphoric pronoun i- (to be precise instr. sg. *h₁i-h₁), and two locative postpositions –r and *-ā́. Initially, it was a local verb, which then acquired an existential meaning (as evidenced by texts from the 16th and 17th centuries). This development path aligns with the typological data on the relationship between locational 118 Acta Linguistica Lithuanica XCI Disappearance of the auxiliary verb esti in the Lithuanian passive and existential sentences (Lyons 1967: 390–396; Clark 1978: 85–126; Yong, Jie 2013; Ostrowski 2017b: 168; 2021: 67). While the evolution and uses of the verb yra have been extensively documented, its evolution as a passive auxiliary verb has been a relatively unexplored area. This study fills this gap by meticulously tracing the disappearance of the form esti from the role of an auxiliary verb in passive constructions. The research is based on Old Lithuanian texts from 1547–1816, which represent three historical dialects of Lithuanian. The paper will delve into frequency data, examining the distribution of individual forms of the third person of the verb būti ‘to be’ in combination with passive participles. This comprehensive analysis aims to present a detailed use of these constructions and, crucially, to explain the vanishing of the form esti in the auxiliary function. 2. METHODOLOGY The results presented in the paper were based on a corpus created for the study, consisting of Old Lithuanian texts from the territories of various historical dialects of the Lithuanian language and written over almost 300 years. The texts were selected based on the publications of Zigmas Zinkevičius (1987; 1988; 1990; 1998) as the most representative monuments of the Old Lithuanian language. Zinkevičius writes about three historical dialects: the Western dialect used on the territory of Prussia and dialects spoken in the territory of the Grand Duchy of Lithuania (lith. Lietuvos Didžioji Kunigaikštystė (LDK)) – the Central and the Eastern.1 The period in which these dialects can be said to have existed lasted approximately until the end of the 17th century. In addition to texts from this period, some later texts have also been included. Even though it is a kind of anachronism, for the study, these translations were included in the dialects mentioned above, based on the origin of the authors, and therefore as a natural way of developing the language they used. This simplification was intended to facilitate the description and comparison of texts and, thus, to trace the development of the studied construction. The table below shows the sizes of individual texts 1 Full list of writings is attached in the abbreviation section. Straipsniai / Articles 119 PAWEŁ BRUDZYŃSKI TABLE 1: corpus size As can be seen, the number of texts from individual dialects is not uniform. The main goal in creating the corpus was to diversify texts so that the results obtained were as reliable as possible and to exclude possible individual linguistic characteristics of authors. The only text from the 19th century represents the Central variant because the greatest diversity distinguished this dialect during the study. The last problem concerns the representation of texts from the Eastern dialect. This is because, as Zinkevičius states (1998: 252–255), this dialect is the least represented and developed. For this reason, although the data from both texts provide interesting information, they are the least reliable and important for the study. To eliminate significant numerical discrepancies further, percentage data will also be presented in the analysis in addition to numbers. To find in the corpus described in the paper constructions, a formula was used consisting of: 1) third-person auxiliary verb yra or esti; 2) endings of the nominative singular and plural for both genders: a. past passive participles (t-participle), b. present passive participles (m-participle). In theory, the auxiliary verb combined with the passive participle is the basis of the passive. All formulas took into account the orthography used by the 120 Acta Linguistica Lithuanica XCI Disappearance of the auxiliary verb esti in the Lithuanian passive authors and the fact that the auxiliary verb could be in both postposition and preposition to the participles. Unfortunately, the results do not consider forms where the auxiliary verb could be in a further position relative to the participle. The obtained results were then manually filtered: TABLE 2: Distribution of auxiliaries in passive constructions Participle m-participle t-participle Auxiliary yra esti yra esti total 232 314 2251 210 546 2461 As can be seen, the results for individual constructions differ significantly. Constructions with t-participles dominate, but the distribution of the auxiliary verb in constructions with m-participles attracts more attention. The following sections will discuss a detailed analysis of individual auxiliaries in the passive. 3. ANALYSIS The tables below present numerical data on the distribution of yra and esti. Table 3 shows the absolute values of the occurrence of individual verbs in combination with m-participle and t-participle. These data are grouped into larger entities according to the century and dialect they represent. Table 4 shows the same data expressed as a percentage value. TABLE 3: Total number of auxiliaries by dialect DIALECT Century m-participle t-participle Prussian (Western) XVI 38 1 39 437 55 492 XVII 0 0 0 11 0 11 XVIII 45 0 45 243 3 246 LDK (Central) XVI 58 15 73 559 32 591 XVII 8 263 271 490 67 557 XVIII 41 34 75 216 46 262 XIX 42 1 43 295 2 297 LDK (Eastern) XVII 0 0 0 0 5 5 Straipsniai / Articles 121 PAWEŁ BRUDZYŃSKI TABLE 4: Total percentage of auxiliaries by text DIALECT | TEXT | m-participle | t-participle | | yra | est | = | yra | est | = Prussian (Western) | XVI | 97 | 3 | 100 | 89 | 11 | 100 | XVII | 0 | 0 | 0 | 100 | 0 | 100 | XVIII | 100 | 0 | 100 | 99 | 1 | 100 LDK (Central) | XVI | 79 | 21 | 100 | 95 | 5 | 100 | XVII | 3 | 97 | 100 | 88 | 12 | 100 | XVIII | 55 | 45 | 100 | 82 | 18 | 100 | XIX | 98 | 2 | 100 | 99 | 1 | 100 LDK (Eastern) | XVII | 0 | 0 | 0 | 0 | 100 | 100 A low number of occurrences of examples from the Eastern dialect of the Grand Duchy of Lithuania will be discussed in the next section. However, for obvious reasons, they are too statistically insignificant. The general picture that may emerge from the above data suggests that esti as an auxiliary was used marginally already in the times of the first written texts, especially in the case of the passive with the t-participle. However, a deeper analysis of individual texts may show interesting relationships between both forms. The following paragraphs will discuss the use of both verbs in combination with the passive based on the t-participle and the m-participle. 3.1. t-participle According to Vytautas Ambrazas (2006: 353), initially, passive was based on the so-called passive past participle formed by adding the suffix –t– and the appropriate ending to the infinitive stem. This participle was initially used to create deverbal adjectives (e.g., baltas ‘white’ < balti ‘to whiten’), and eventually, it became the basis for the passive and has probably served this function since pre-literate times. In the oldest Lithuanian texts from the 16th century, the form of the third person of the auxiliary is mostly yra. However, some constructions with esti still can be found: (1) 1579VEE: ... ghiffai yra paſkirtas nũgi 3.sg.m.det aux.prs.3 appoint.pst.pp.nom.sg.m from Diéwa fudžia God.gen.sg judge.gen.sg ‘...he is appointed by God-judge.’ 122 Acta Linguistica Lithuanica XCI Disappearance of the auxiliary verb esti in the Lithuanian passive (2) 1579VEE: Pagirtas eft waifus ðiwata tawa bless.pst.pp.nom.sg.m aux.prs.3 fruit.nom.sg womb.gen.sg poss.2.sg ‘Blessed is the fruit of thy womb.’ Observations related to the distribution of the third person forms of the verb būti are given in the tables below, where table 5 presents absolute numbers while table 6 – percentage data regarding the distribution in individual texts: TABLE 5: Total number of auxiliaries by text DIALECT | TEXT | t-participle | yra | est | TOTAL Prussian (Western) | 1547MžK | 0 | 3 | 3 Prussian (Western) | 1573WP | 237 | 27 | 264 Prussian (Western) | 1579VEE | 61 | 20 | 81 Prussian (Western) | 1579BB | 139 | 5 | 144 Prussian (Western) | 1612ZE | 11 | 0 | 11 Prussian (Western) | 1727KNT | 243 | 3 | 246 LDK (Central) | 1599DP | 414 | 6 | 420 LDK (Central) | 1660MP | 145 | 26 | 171 LDK (Central) | 1653KN | 0 | 12 | 12 LDK (Central) | 1660ChB | 490 | 56 | 546 LDK (Central) | 1701BtNT | 216 | 46 | 262 LDK (Central) | 1816GNT | 295 | 2 | 297 LDK (Eastern) | 1644ŠPS | 0 | 4 | 8 LDK (Eastern) | 1647JE | 0 | 1 | 1 Straipsniai / Articles 123 PAWEŁ BRUDZYŃSKI TABLE 6: Total percentage of auxiliaries by text The overwhelming majority of examples use the auxiliary yra as the default one. As expected, a gradual disappearance of the passive with the verb esti can be noticed. However, this process differed slightly between the dialects of Prussia and the Grand Duchy of Lithuania. Based on the data, it can be assumed that the construction with esti as an auxiliary verb finally went out of use in Prussia in the 18th century and the LDK in the 19th century. According to Ambrazas (2001: 15), this is probably related to the development of the passive itself, which was first fully grammaticalized in Western dialects and then gradually covered subsequent dialects to the East. This process in some eastern dialects was not fully completed until modern times (Ambrazas 2001: 15–16; 2006: 365–366). Below, the characteristics of the use of the mentioned verb forms in individual dialects will be discussed. 3.1.1. Prussian dialect The oldest studied text represents the Prussian dialect, and at the same time, the oldest Lithuanian book is the Martynas Mažvydas’ Catechism. During the corpus search, three examples of the passive with an auxiliary in the third person were found in the text mentioned above, and interestingly, all these examples are constructions with the verb esti: 124 Acta Linguistica Lithuanica XCI Disappearance of the auxiliary verb esti in the Lithuanian passive (3) 1547MžK ... Bažniczie padota esti Christui church.nom.sg give.pst.pp. aux.pst.3 pn.dat.sg (4) 1547MžK Szadis Dewą [...] aprašchitas esti word.nom.sg God.gen.sg describe.pst.pp.nom.sg aux.prs.3 Tais zadeis this.ins.pl.m word.ins.pl ‘Word of God is described with these words.’ Of course, the examples found are not numerous enough to be considered statistically significant. However, it is worth noting that in the text of Postilla from 1573, only 28 years after the Catechism, the dominant form of the auxiliary is yra, and esti occurs in only 10% of total examples (27 out of 264). However, despite this overwhelming discrepancy, there is no apparent difference in the use of the two forms. They can be used both in resultative meanings (5a) and (5b) and in actional meanings with an expressed agent (6a) and (6b): (5) Resultative: a. 1573WP Priliginta est karaliste dangaus βmagu equate.pst.pp.nom.sg.f aux.pst.3 kingdom.nom.sg heaven.gen.sg man.dat.sg kareliu king.dat.sg ‘The kingdom of heaven is like unto a certain king...’ b. 1573WP Priliginta ira karaliste Dangaus βmagu equate.pst.pp.nom.sg.f aux.pst.3 kingdom.nom.sg heaven.gen.sg man.gen.pl seiančiam gieru sėklų sow.prs.ap.dat.sg.m good.gen.pl.m seed.gen.pl ‘The kingdom of heaven is likened unto a man which sowed good seed...’ (6) Actional: a. 1573WP Die was [...] regetas est nog God.nom.sg see.pst.ap.nom.sg.m aux.prs.3 from angelų angel.gen.pl ‘...God was [...] seen of Angels.’ Straipsniai / Articles 125 PAWEŁ BRUDZYŃSKI b. 1573WP Jesus dotas ira mumis nog pn.nom.sg give.pst.pp.nom.sg.m aux.prs.3 1.pl.dat from Diėwa God.gen.sg ‘...Jesus is given to us by God...’ It is worth mentioning, however, that although both types of construction occur with both verb forms, the constructions with esti seem more actional than those with yra. Of course, this is difficult to state unambiguously because the level of grammaticalization of the Lithuanian passive is much lower than, for example, English (Brudzyński, Spraunienė 2021: 179–180); however, this can be proven by the fact that the agent in this type of constructions is expressed 6 times out of 27, while in constructions with yra only 5 out of 237. This could suggest that the fully developed passive, which has both resultative and actional uses, initially used the auxiliary esti. With the rise of yra, the verb gradually took over the initial role of esti, starting with attributive and resultative uses, thus retracing the path of passive development (Bybee et al. 1994: 63–69). A similar situation is still visible in Baltramiejus Vilentas’ 6 years older text, where constructions with esti account for 25% of all uses and play the role of a more actional passive. In contrast, constructions with yra mainly refer to the attributive or resultative meaning, often using mostly lexicalized forms such as pagirtas ‘praised,’ prilygintas ‘compared,’ or pašlovintas ‘blessed’. (7) 1579VEE: Ir regietas eßt nūg Cefaſcha and see.pst.pp.nom.sg.m aux.prs.3 from pn.gen.sg ‘And that he was seen of Cephas...’ (8) 1579VE: Paſchlowinti yra vbagai Dwaffeje bless.pst.pp.nom.pl.m aux.prs.3 poor.nom.pl spirit.loc.sg ‘Blessed are the poor in spirit...’ Interestingly, the proportions between both auxiliary verbs are different in the text of Jonas Bretkūnas’ Bible, which was translated from 1579 to 1590. In this text, the passive with esti appears only five times out of 144, and all examples with the expressed agent are limited only to the construction with yra. 126 Acta Linguistica Lithuanica XCI Disappearance of the auxiliary verb esti in the Lithuanian passive (9) 1579BB IEsus regetas ira nūg Cephoschaus pn.nom.sg seen.pst.pp.nom.sg.m aux.prs.3 from pn.gen.sg ‘Jesus was seen of Cephas.’ Therefore, it should be recognized that the end of the 16th century was a critical moment in the disappearance of the passive based on the auxiliary esti. In addition to Bretkūnas’ text, in which this verb appears sporadically, this fact may be proven by the Lozorius Zengštokas’ Enchiridion, which is several years older, in which this construction does not appear at all. It is worth noting that 100 years later, in the text of the Quandt’s Bible from 1727, the passive with esti appears 3 times out of 246: (10) 1727KNT Man [...] dūta esti βi 1.sg.dat give.pst.pp.nom.sg.f aux.prs.3 this.nom.sg.f Mallon grace.nom.sg ‘Unto me [...] is this grace given.’ (11) 1727KNT Bet kiekvienam išβ musu dūta but every_one.dat.sg from 1.pl.gen give.pst.pp.nom.sg.f esti Mallone aux.prs.3 grace.nom.sg ‘But unto every one of us is given grace...’ (12) 1727KNT Surinkimas padūtas esti Kristui taipo congregation.nom.sg give.pst.pp.nom.sg.m aux.prs.3 pn.dat.sg so ir Moteres savo Wyrams and woman.nom.pl rposs man.dat.pl ‘...the church is subject unto Christ, so let the wives be to their own husbands...’ It is worth adding that all these uses are rather resultative, which may indicate that the 17th century was a critical period in which esti finally fell out of use as the primary auxiliary verb in passive constructions of the Prussian dialect. Straipsniai / Articles 127 PAWEŁ BRUDZYŃSKI 3.1.2. Dialects of LDK The development of the third-person auxiliary in the case of LDK dialects is somewhat similar, but some chronological differences can be noticed. Similarly to Prussia, based on the data presented, we can assume a gradual disappearance of the auxiliary esti. However, the first visible difference is that these forms have been preserved in the territory of the LDK, both in the Central and Eastern dialects, longer than in Prussia. Looking at the data, this thesis could be contradicted by the distribution of auxiliary verbs in Mikalojus Daukša’s Postill (1599), who is considered a leading representative of the Central dialect and one of the most influential authors in the context of the development of the literary language (Zinkevičius 1998: 246). In the text, it can be noticed that out of 420 constructions with a t-participle, only six use esti as an auxiliary, while in Jokūbas Morkūnas’ text, which is only a year younger, this proportion is 26 examples out of 171: (13) 1599DP Prámintas est tieġ wârdas io call.pst.pp.nom.sg.m aux.prs.3 therefore name.nom.sg 3.sg.gen.m Iésus kurìs prámintas ẽst’ nùg pn.nom.sg which.nom.sg.m call.pst.pp.nom.sg.m aux.prs.3 from Angelo angel.gen.sg ‘He was given the name Jesus, which was given by the Angel.’ (14) 1599DP Ir pramintas yra wârdas io and call.pst.pp.nom.sg.m aux.prs.3 name.nom.sg 3.sg.gen.m Iesus pn.nom.sg ‘He was given the name Jesus.’ Zinkevičius (1998: 247) accuses Morkūnas of numerous errors, but this discrepancy seems too large to be a simple linguistic error. The answer to this significant difference in distribution may lie in the origin of the Daukša. The author came from Samogitia, which is geographically close to the territory of Prussia. Therefore, it can be assumed that the influence of the Western Dialect, and thus the development of the passive, on the author’s language was much more significant than on the language of other authors using the Central and Eastern Dialects. This relationship is particularly evident in the case of the distribution of auxiliary verbs with the m-participle, which will be discussed in the next section. This fact may, therefore, be evidence of the gradual development of 128 Acta Linguistica Lithuanica XCI Disappearance of the auxiliary verb esti in the Lithuanian passive the passive, which was territorially dependent, and at the same time, confirms the analysis presented in the previous chapter, which states that the turn of the 16th and the 17th centuries was the moment when this construction finally fell out of use in the territory of Prussia, and Daukša himself was potentially one of the reasons for the spread of the new construction into the territory of LDK. Therefore, considering the assumptions presented above, it can be concluded that esti has been preserved as an auxiliary verb in the territory of the LDK for longer than in Prussia. In the text mentioned above by Morkūnas, as many as 15% of passive constructions are still passives with esti, which are mostly actional (14). Passives with the form yra, although they can also be actional and can have an expressed agent, are mostly resultatives or used attributively: (15) Actional (1660MP) a Kás mumus tobùlay irã iš̀guldita what 1.pl.dat perfectly aux.prs.3 explain.pst.pp.n nuog Powila from pn.gen.sg b Mumus est duota nuog Pono 1.pl.dat aux.prs.3 give.pst.pp.n from lord.gen.sg musu 1pl.gen ‘It is given by our Lord.’ (16) Resultative (1660MP) Tas krauiàs ira prãlietas this.nom.sg.m blood.nom.sg aux.prs.3 spill.pst.pp.nom.sg.m ‘This blood is spilled...’ Interestingly, in the approximately half-a-century older text of the Knyga nobažnystės, similarly to the text of Mažvydas, the only auxiliary verb in the passive constructions with a t-participle was esti. (17) 1654KN Numiria teypagi ir bágočius ir pakastas esti die.pst.3 therefore and rich_man.nom.sg and bury.pst.pp.nom.sg.m aux.prs.3 ‘...the rich man also died and was buried.’ It is worth noting, however, that the examples in the text convey the meaning of the result, and there is no attested construction with an overtly expressed Straipsniai / Articles 129 PAWEŁ BRUDZYŃSKI agent. At this point, it is also worth noting that the above example, unlike the modern construction, has an obvious past tense meaning, which indicates an earlier stage of development of the passive. Similar examples can also be found in a number of other texts described in this study (Brudzyński, Sprauniėnė 2021). In the text of the Samuelis Boguslavas Chylinskis’ Bible published in 1660, the status of yra as the default form of the third person in constructions with the t-participle is already well established. However, a significant part (10%) are still constructions with esti. Moreover, Chylinskis’ text is characterized by several intriguing features. A general observation is the position of the auxiliary verb. In the case of the form yra, 386 out of 490 examples (79%) put the verb in postposition, while in the case of esti, it is only 8 out of 55 (15%). This is perhaps related to the use of the verb esti itself. Most of the examples found are negations (18) or hortatives (19) (44 out of 54 examples have a prefix next to the auxiliary verb, and among the remaining ten, 4 have a prefixed participle): (18) negation a. 1660ChB ... dar n’est apireiškita ko however, aux.prs.3.neg reveal.pst.pp.nom.n what.ins busime be.fut.1.pl ‘...yet it is not revealed what we will become. b. 1660ChB Nepažįnta est galibe jo known.pst.pp.nom.f.neg aux.prs.3 might.nom.sg 3.sg.m.gen ‘Indeed his might is not known.’ (19) hortatives a. 1660ChB ... t’est prakiaykta žiame del tawęs 2.sg.gen ‘...cursed is the ground for thy sake...’ b. 1660ChB Totmesta est nog jusu wisokia throw_away.pst.pp.nom.sg.f.hort aux.prs.3 from 2.pl.gen all.nom.sg kartbe bitterness.nom.sg ‘Let all bitterness [...] be put away from you...’ 130 Acta Linguistica Lithuanica XCI Disappearance of the auxiliary verb esti in the Lithuanian passive Among the remaining examples, one can also find actional passives with an expressed agent: (20) 1660ChB a… est aux.prs.3 pamusztas defeat.pst.pp.nom.sg.m Izrael pn.nom.sg nog from Filistynu pn.gen.pl ‘…Israel was defeated by Philistins…’ b [Abimelach] pn.nom.sg pakorotas punish.pst.pp.nom.sg.m žēklingey significantly nog by Diewa god.gen.sg est aux.prs.3 ‘[Abimelach] significantly was punished by the God.’ These examples, although few, suggest that the passive with esti was still used. However, the question remains whether the negation and hortative forms of the verb esti developed naturally as specialized constructions or whether they were an independent attempt by Chylinski​s to systematize the grammar. Due to the fact that it was not possible to find similar constructions in other authors, the second option seems more likely. Moving on to the 18th century, it can be noticed that while the construction with esti practically disappeared in Prussia, it was still used in LDK dialects, an example of which is the text of Samuelis Bitneris’ New Testament (1701). In the text, as many as 46 out of 295 constructions with t-particle (18%) use esti as an auxiliary, and they can be both actional and resultative: (21) 1701BNT a O exclam esti aux.prs.3 jau already βaknięsβ root.all ir and médio tree.gen.sg kirwis axe.nom.sg pridetas laid_unto.pst.pp.nom.sg.m ‘And now also the axe is laid unto the root of the tree.’ b Nes for per by netikis believe.prs.ap.nom.sg.m.neg moteri woman.acc.sg wijras man.nom.sg paβwestas sanctify.pst.pp.nom.sg.m esti aux.prs.3 ‘For the unbelieving husband is sanctified by the wife…’ Among the constructions with yra, there are more attributive, resultative, and impersonal uses, just like in Morkūnas’ text. This may suggest that at the Straipsniai / Articles 131 PAWEŁ BRUDZYŃSKI beginning of the 18th century, esti was still a productive auxiliary used in a fully-fledged passive, while the passive with yra, although numerous, was less developed. The status of the form esti has changed over 100 years, and in Juozapas Arnulfas Giedraitis’ text, only two uses of it in the passive construction can be found. Interestingly, both examples, similarly to Chylinskis’, are negations: (22) 1816GNI a [Jog jums duota ira zinot paslėptinės karalistes dangaus:] o jemis ne est duota exclam 3.pl.m.dat neg aux.prs.3 give.pst.pp.n ‘[Because it is given unto you to know the mysteries of the kingdom of heaven,] but to them it is not given.’ b [Daug teypogi ir kitu stebuklu padare Jezus po akimis mokitiniu savo,] kurie ne esti parasziti tose which.pl.m neg aux.prs.3 write.pst.pp.nom.pl.m this.loc.pl knigose book.loc.pl ‘[And many other signs truly did Jesus in the presence of his disciples,] which are not written in these book.’ The very low frequency of esti (1%) may suggest that the form is probably used for stylistic reasons, and most likely somewhere between the 18th and 19th centuries, esti as an auxiliary verb in the territory of the LDK finally fell into disuse. At this point, a small sample of the eastern dialect is worth mentioning – namely texts by Konstantinas Sirvydas and Jonas Jaknavičius. In both texts, only five examples of the passive with the third person were found, and all use the verb esti. (23) 1644ŠPS kayp as efti aux.prs.3 Perf. person.gen.pl kiti other.nom.sg.m dárbay work.nom.pl Diewo god.gen.sg [...] nuog from dariti made.pst.pp.nom.pl.m wifų all.gen.pl.m triiu three.gen.m Travyces Trinity.gen.sg S[ventosios] holy.gen.pl.f ‘As other works of the God are made by all Holy Trinity.’ 132 Acta Linguistica Lithuanica XCI Disappearance of the auxiliary verb esti in the Lithuanian passive (24) 1647JE A vienas ig iu kad exclam one.nom.sg from 3.pl.gen when į̨wido apčįstitas esti [sugrio su baltsu didziu garbindamas Diewu.] realize.pst.3 cleanse.pst.pp.nom.sg.m aux.prs.3 ‘And one of them when realized that is cleansed, collapsed on earth and praised loudly the Lord.’ The absence of yra forms in the passive could suggest that the process of developing forms of the third person of the passive voice started in the West and moved to the East and the Eastern dialect was the last to fully develop the passive. This would confirm Ambrazas’ (2001: 15–16) observations on the development of the passive, but unfortunately, the sample is too small to be fully reliable. Nevertheless, it shows a certain tendency and is a starting point for further analysis. 3.2. m-participle In addition to the t-participle, the Lithuanian passive may be based on the so-called present passive participle (m-participle), formed by adding the suffix -mand an ending to the present tense stem. Historically, the participle was used to create adjectives (cf. lat. firmus, en. warm), and nowadays, constructions using it are characterized by a durative aspect. In this function, it is a Lithuanian innovation. The same participle is present in Latvian. However, it has not developed as the basis for the passive and instead has a modal meaning. Interestingly, in the oldest texts from the territory of Prussia, the default auxiliary verb was yra, and esti appeared later, mainly in the territory of the LDK. The distribution of these forms is presented in the tables below: Straipsniai / Articles 133 PAWEŁ BRUDZYŃSKI TABLE 7: Total number of auxiliaries by text DIALECT | TEXT | m-particle | yra | est | = Prussian (Western) | 1547MŽK | 0 | 0 | 0 Prussian (Western) | 1573WP | 1 | 0 | 0 Prussian (Western) | 1579VE | 7 | 1 | 8 Prussian (Western) | 1579BB | 30 | 0 | 30 Prussian (Western) | 1612ZE | 0 | 0 | 0 Prussian (Western) | 1727KNT | 45 | 0 | 45 LDK (Central) | 1599DP | 58 | 2 | 60 LDK (Central) | 1660MP | 0 | 13 | 13 LDK (Central) | 1653KN | 0 | 5 | 5 LDK (Central) | 1660ChB | 7 | 257 | 264 LDK (Central) | 1701BtNT | 41 | 34 | 75 LDK (Central) | 1816GNT | 42 | 1 | 43 LDK (Eastern) | 1644ŠPS | 0 | 0 | 0 LDK (Eastern) | 1647JĘ | 0 | 0 | 0 TABLE 8: Total percentage of auxiliaries by text DIALECT | TEXT | m-particle | yra | est | = Prussian (Western) | 1547MŽK | 0 | 0 | 0 Prussian (Western) | 1573WP | 100 | 0 | 100 Prussian (Western) | 1579VE | 88 | 12 | 100 Prussian (Western) | 1579BB | 100 | 0 | 100 Prussian (Western) | 1612ZE | 0 | 0 | 0 Prussian (Western) | 1727KNT | 100 | 0 | 100 LDK (Central) | 1599DP | 97 | 3 | 100 LDK (Central) | 1660MP | 0 | 100 | 100 LDK (Central) | 1653KN | 0 | 100 | 100 LDK (Central) | 1660ChB | 3 | 97 | 100 LDK (Central) | 1701BtNT | 55 | 45 | 100 LDK (Central) | 1816GNT | 98 | 2 | 100 LDK (Eastern) | 1644ŠPS | 0 | 0 | 0 LDK (Eastern) | 1647JĘ | 0 | 0 | 0 134 Acta Linguistica Lithuanica XCI Disappearance of the auxiliary verb esti in the Lithuanian passive The distribution mentioned above seems puzzling. As already mentioned, passive was first grammaticalized in the territory of Prussia. This is particularly important taking into account the passive with the m-participle because it first emerged in Prussia and only later spread to the territory of LDK, and according to Ambrazas (2001: 16), a passive of this type has not developed in some Eastern dialects to this day, which seem to be confirmed by the data collected for this study. Below, an analysis of individual examples found in the Prussian and LDK dialects will be discussed. 3.2.1. Prussian dialect As already mentioned, the passive based on the m-participle is a Lithuanian innovation that first appeared on the territory of Prussia. In the oldest text – Mažvydas’ Catechism, there were no examples of this type of use with an auxiliary. It is difficult to say how developed this construction was during this period, and it may not be attested. However, it can be assumed that it was still not fully grammaticalized as a passive, and the oldest attested example comes from Vilentas’ text from 1579, where some different uses of the participle can be found: (25) 1579VEE: Nėla Agar wandinamas yra Arabioje for pn call.prs.pp.nom.sg.m aux.prs.3 pn.loc.pl kalnas Sinai mountain.nom.sg pn ‘For Mount Sinai is called Agar in Arabia.’ (26) 1579VEE: Wịssi daiktai taw yra galimi all.nom.pl.m thing.nom.pl 2.sg.dat aux.prs.3 can.prs.pp.nom.pl.m ‘All things are possible to you...’ (27) 1579VEE: ... ubagams apfakoma yra Euangelia poor.dat.pl preach.prs.pp.nom.sg.f aux.prs.3 gospel.nom.sg ‘...to the poor the gospel is preached.’ Straipsniai / Articles 135 PAWEŁ BRUDZYŃSKI (28) 1579VEE: … kurie which.nom.pl.m nîg from iô [Dievo] 3.sg.gen.m jra aux.prs.3 Įjunt emi send.prs.pp.nom.pl.m ‘…which are sent by him [the God]…’ Example (25) illustrates the attributive or resultative use, which typically marks the initial stage of the passive’s evolution (Bybee et al. 1994: 64; Ambrazas 2001: 12). Example (26) demonstrates a modal use. Examples (27) and (28) represent the actional passive, which shares the same meaning as the contemporary construction with the m-participle – namely, durative or habitual. Crucially, as shown in example (28), it is possible to express an agent in the construction, indicating that it is already fully developed, much like the passive based on the t-participle. It is worth noting that Vilentas’ text is the only examined Prussian Dialect text to feature construction with esti. (29) 1579VEE: … dukte daughter.nom.sg mana 1.sg.poss.nom.f ƒunke’ grievously nûgi from Welna devil.gen.sg efti aux.prs.3 warginama vex.prs.pp.nom.sg.f ‘… my daughter is grievously vexed with a devil.’ As can be seen, it is a full-fledged actional construction with an expressed agent, but it is the only example of this type in the examined texts. The texts by Bretkūnas and Quandt (the only ones from the territory of Prussia where the construction occurs) only use the combination yra + m-participle. Please consider below examples: (30) BB1579: 30 Mana 1.sg.poss.nom.f ira aux.prs.3 Duktie daughter.nom.sg muczijma vex.prs.pp.nom.sg.f nûg from Welino devil.gen.sg piktai grievously ‘… my daughter is grievously vexed with a devil.’ (31) KNT1727: Mano 1.sg.poss.nom.f Dukte daughter.nom.sg ƒunkiey grievously Wėlno devil.gen.sg wargin nama vex.prs.pp.nom.sg.f ‘… my daughter is grievously vexed with a devil.’ 136 Acta Linguistica Lithuanica XCI Disappearance of the auxiliary verb esti in the Lithuanian passive Due to the scarcity of data, it is challenging to pinpoint when the grammaticalization of the passive based on the m-particle occurred. However, it is undeniably plausible that by the 16th century, in the Prussian territory, it was already fully developed, and from the outset, its default auxiliary verb was yra. This is likely linked to the fact that when this type of construction emerged in the language, the passive based on the t-participle was in the final phase of grammaticalization. In the 16th century, yra was the dominant auxiliary. Most likely, the newly emerged passive construction with the m-particle automatically adopted the most frequent auxiliary during this period. 3.2.2. Central dialect Based on the previously presented data regarding the t-participle, it could be assumed that the situation in LDK was analogous to that in Prussia. However, the data presented in Tables (7) and (8) seem to contradict this. The oldest of the examined texts – Daukša’s Postyll – has a similar distribution to the texts of Prussian authors, and the only two examples out of 60 uses of the m-particle are constructions with esti: (32) DP1599: … numìrelei kàrtais giwúmpiumpi šiunczemi dead.nom.pl sometimes life.all send.prs.pp.nom.pl.m est aux.prs.3 ‘…dead sometimes are sent to life.’ (33) DP1599: … anié kurié lûdintiemus dâromi 3.pl.nom.m which.nom.pl.m witness.ins.pl made.prs.pp.nom.pl.m est’ aux.prs.3 …those who are made witnesses…’ However, already a year later, in Morkūnas’ text, the results are surprising, and the dominant auxiliary is esti, while yra with the m-particle does not appear at all: Straipsniai / Articles 137 PAWEŁ BRUDZYŃSKI (34) 1600MP: a. [Postilla LIETWISZKA Tátų est Ižgulimâs prástas Ewângeliu ant kožnos Nedelios ir Szwentês per wisus metus] kurios pagal budá sená Bãžnicžioy which.nom.sg.m according why.acc.sg old.acc.sg.m church.loc.sg Diewá est skáitomos God.gen.sg aux.prs.3 read.prs.pp.nom.pl.f ‘[This Lithuanian Postil is the simple explanation of the Gospel for every Sunday and holiday through the year], which according to the old tradition is read in the Congregation of the God.’ Zinkevičius (1998: 247) accuses Morkūnas of many mistakes, but the same cannot be said about Bitneris or Steponas Jaugelis Telega – renowned language experts of their time (Zinkevičius 1988: 216–223). In the 50 years older text of Knyga nobažnystės, the only construction expressing this type of passive is still esti + m-participle. (35) 1653KN 35... kita other.nom.sg.f irà be.prs.3 dángus heaven.nom.sg kurion which.ill.sg wisi all.nom.pl.m wierni faithful.nom.pl.m ir and pábani devout.nom.pl.m 3mones people.nom.pl siunciąmi send.prs.pp.nom.pl.m esti aux.prs.3 ‘...different is the Heaven to which all faithful and devout people are sent.’ What this type of distribution of construction with esti can indicate? Of course, it can be assumed that somewhere between the 16th and 17th centuries, the verb esti reappeared once again as the default auxiliary for the m-participle, but this seems unlikely. The reason for such a large discrepancy between Daukša and the other authors from the LDK, as already mentioned, should be sought in the origin of the author of the Postille. Therefore, it can be assumed that Daukša’s text, which is a representative of the Central dialect, was at least influenced by the Western dialect in this respect, which seems to be confirmed by both the data regarding the tand mparticiples. While the examples cited so far have not been numerous, the text of Chylinski’s Bible seems to confirm the thesis that in the territory of the LDK, the construction of yra + mparticiple has not yet been developed. In the mentioned translation, out of 266 examples, only seven constructions are combined with yra. Four are used with the participle zynoma ‘known’ and one 138 Acta Linguistica Lithuanica XCI Disappearance of the auxiliary verb esti in the Lithuanian passive with wadynama ‘called’ and notszaukiamas ‘not called’ – so one should consider them as lexicalized attributive or modal uses rather than the actual passive. (36) 1660ChB a… ana dirwa ju liezuwe that.nom.sg.f field.nom.sg 3.pl.gen.m tongue.loc.sg wadynama ira Akeldama call.prs.pp.nom.sg.f aux.prs.3 pn.nom ‘…that field is called in their proper tongue, Aceldama…’ b [Jog nes zymus zianklas padarytas ira per jos,] zynoma ira wisjemus kurie giwena Jeruzale] know.prs.pp.n aux.prs.3 all.dat.pl.m which.nom.pl.m live.prs.3 pn.loc.sg ‘[For that indeed a notable miracle hath been done by them] is manifest to all them that dwell in Jerusalem.’ Only one example can be interpreted as the actional passive: (37) 1660ChB [Bet tepraszо jos wieroy, tas nes kursej abejoja ligus ira] Wilniey mariu, kursej waromas ira wave.dat.sg sea.gen.pl which.dat.sg move.prs.pp.nom.sg.m aux.prs.3 nog from wejaus wind.gen.sg ‘[But let him ask in faith, nothing wavering. For he that wavereth is like] a wave of the sea driven with the wind…’ Most examples (259) with a passive based on the m-particle in Chylinskis’ text use the verb esti. Among them, one can also find constructions with the participles mentioned in examples (36) and (37), but there are only 6 of them. The overwhelming majority are passive constructions corresponding to contemporary durative and habitual use of the m-particle. (38) 1660ChB a…[teywaniste] nog Izaoka paszlawinime jo pažadama ir pastypŋnama est promise.prs.pp.nom.sg.f and reinforce.prs.pp.nom.sg.f aux.prs.3 ‘…[heritage] is promised and reinforced by Isaac’s blessing.’ Straipsniai / Articles 139 PAWEŁ BRUDZYŃSKI b Izraeliety pamuszami est nog Filistynu pn.nom.pl defeat.prs.pp.nom.pl.m aux.prs.3 from pn.gen.pl ‘Izraelites was defeated by Philistines.’ c Pastypyrnamas est asabliwu apireyszkimu reinforce.prs.pp.nom.sg.m aux.prs. 3 peculiar.ins.sg.m revelation.ins.sg ‘Was reinforced by peculiar revelation.’ As in the case of the t–participle, there are constructions with a hortative prefix (39a) and a negation prefix (39b). However, these are just five such examples: (39) 1660ChB a Wiezliviuste jusu t’est pažystama wisien[us] moderation.nom.sg 2.pl.gen aux.prs.3.hort know.prs.pp.nom.sg.f all.acc.pl.m zmone m[us] man.acc.pl ‘Let your moderation be known unto all men.’ b… jeygu wel n’est wadojama tada bus pagal apwertynima then be.fut.3 sell.pst.pp.n according to estimation.acc.sg tąwo 2.sg.poss ‘…if it is not redeemed, then it shall be sold according to thy estimation.’ This may indicate that the verb esti, which could have specialized functions in constructions with a t–participle, is rather a default auxiliary in the case of an m–participle. This hypothesis may also be supported by the fact that many of the examples come not from the text of the Bible itself but from introductions to chapters that are not inspired by the source of the translation and reflect the idiolect of the author himself – and therefore the actual language of the Central dialect of the second half of the 17th century. Bitneris’ text, which is 40 years older, the situation changes, and the distribution of both auxiliaries are almost equal (45% – esti and 55% – yra). There is no apparent difference between the use of both verbs, but the passive with esti seems more actional. At the same time, in combination with yra, there are more lexicalized or attributive forms. 140 Acta Linguistica Lithuanica XCI Disappearance of the auxiliary verb esti in the Lithuanian passive (40) 1701BNT a… dukte mano sunkey muczijama esti nūg Welńio aux.prs.3 daughter.nom.sg 1.sg.poss.nom.f grievously vex.prs.pp.nom.sg.f aux.prs.3 from devil.gen.sg ‘… my daughter is grievously vexed with a devil.’ b [Grōmata] skaitoma ira nūg wissū ȝmonū book.nom.sg read.prs.pp.n aux.prs.3 from all.gen.pl man.gen.pl ‘Book is read by all men.’ (41) 1701BNT a Tegul jumus wissiemus žinoma esti shall 2.pl.dat all.dat.pl.m know.prs.pp.n aux.prs.3 ‘Be it known unto you all…’ b Nes žinoma ira jog Wießpats musu iβėjo iβ plemės Iudos because know.prs.pp.n aux.prs.3 that Lord.nom.sg 1.pl.gen go.pst.3 from tribe.gen.sg pn.gen ‘For it is evident that our Lord sprang out of Juda…’ This may indicate that the turn of the 17th and 18th centuries was when the esti construction began to fall into disuse. Giedraitis’ text has only one construction with the verb esti out of 43. It can, therefore, be assumed that this construction was no longer productive in the 19th century. Although, as in the case of the Prussian dialect, the exact date of the grammaticalization of the construction with yra cannot be given, it must have occurred at the turn of the 18th and 19th centuries at the latest, i.e., over 100 years later. The data from the Eastern dialect are indeed fragmentary. However, the absence of the discussed construction in the examined texts may be a reason to claim that this construction must have developed and been grammaticalized even later. 4. CONCLUSIONS Historically, the third person of the Lithuanian verb būti ‘to be’ used the form esti, corresponding to other Baltic and Indo-European languages (cf. Prussian ast; Polish jest, Latin est, German ist, etc.). However, due to the semantic shift, Straipsniai / Articles 141 PAWEŁ BRUDZYŃSKI the place of the third person form was taken by the originally local verb yra, and this process must have started in the pre-literate period, as evidenced by the oldest Lithuanian texts and Latvian third person form ir. The original third person form esti already in the 16th century in the Western dialect served only as an auxiliary, and its original function was taken over by yra. However, esti as an auxiliary existed in the Central and probably Eastern dialects until the 18th century, which indicates apparent differences in grammaticalization timeline between individual dialects. In Prussia, the passive, based on the verb yra, was grammaticalized earlier and was widely used in the 16th century. Accordingly, the m-participle-based passive, which was a Lithuanian innovation that first appeared in Prussia, implicitly took the well-developed and widely used form yra as its auxiliary as soon as when it first appeared. It can, therefore, be assumed that both types of passive were grammaticalized at approximately the same time, i.e., at the turn of the 16th and 17th centuries, because from that period, the productivity of esti decreased dramatically. A completely different situation occurs in the central dialect. The process of grammaticalization of the passive began there later. Esti, as an auxiliary verb with the t-participle, began to disappear from this function at the beginning of the 17th century. However, it was still present and seemed to be productive. What is worth mentioning is that the passive with esti seems more actional than the forms based on yra. The origin of both verbs may explain this. By origin, the local yra, which had a more stative meaning, initially appeared in attributive and resultative constructions. The genesis of the Baltic passive assumes a reanalysis of the originally stative meaning as a dynamic one and a semantic extension of meaning. This reanalysis develops the meaning of the past tense or perfect next to the original resultative meaning of the present tense (Brudzyński, Spraunienė 2021: 167–168). The passive based on the m-participle, which mainly has a durative and, therefore, actional meaning, has taken the auxiliary esti, which was more dynamic due to its original existential meaning, as a default. Therefore, the grammaticalization of both constructions did not co-occur. The t-participle began to be grammaticalized in the 17th century, and the m-participle was only in the 18th century, but the process for both ended in the 19th century. One of the main evidence for differences in the grammaticalization of the passive in LDK and Prussia may be the text of Daukša, which can be considered a missing piece. The leading representative of the Central dialect has a distribution similar to the Western dialect, which may be due to geography, i.e., the location of Samogitia on the border of both dialects. Therefore, the passive form found in the Postill corresponds to the Western rather than the Central dialect and thus indicates that the process took place gradually from West to East. 142 Acta Linguistica Lithuanica XCI Disappearance of the auxiliary verb esti in the Lithuanian passive It is also worth mentioning the admittedly poorly represented texts of the Eastern dialect, which, in any case, may provide interesting hypotheses. Despite the small number of occurrences in the 17th century, the only form of the auxiliary verb with the t-participle was esti. At the same time, none of them appeared with the m-participle. Of course, this may be a coincidence, but it is likely a trace of the fact that this process was even more delayed in the eastern dialect compared to the other two dialects. However, more data is required to confirm this. The departure from one of the forms of the third person present tense of the verb būti was probably caused by the redundancy and common ontology of both verbs. The relationship between a local sentence and an existential sentence is easy to understand, remembering that with an unmarked sentence order, the thematic–rhematic structure of one is a mirror reflection of the other. Existence presupposes location (is where?), and location presupposes existence (where is it?) (Watkins 1967; 1994; Mikulskas 2009: 128). For this reason, the verb yra replaced the third person of the present tense as the default one, while esti initially moved to the function of auxiliary and ultimately by a semantic shift to the function of a habitual verb. ABBREVIATIONS 1 – first person; 2 – second person; 3 – third person; ACC – accusative; AP – active participle; AUX – auxiliary; DAT – dative; DET – determined form; F – feminine; GEN – genitive; INS – instrumental; LOC – locative; M – masculine; NOM – nominative; PL – plural; PN – personal name; POSS – possessive pronoun; PP – past participle; PRS – present; PST – past; REFL – reflexive; RPOSS – reflective possessive pronoun; SG – singular; SUP – superlative; VOC – vocative SOURCES 1547 MŽK – Mažvydas Martynas. Katechismusо prasti žodžiai, 1547. 1573 WP – Wolfenbüttelio postilė, 1573. 1579 BB – Bretkunas Jonas. Biblija, 1579–1590. 1579 VEE – Vilentas Baltramiejus. Evangelijos bei Epistolos, 1579. 1599 DP – Daukša Mikalojus. Postilė, 1599. 1612 ZE – Zengštokas Lozorius. Enchiridionas, 1612. 1644 ŠPS – Sirvydas Konstantinas. Punktai sakymų, 1644. Straipsniai / Articles 143 PAWEŁ BRUDZYŃSKI 1647JE – Jaknavičius Jonas. Evangelie Polskie y Litewskie, 1647. 1653KN – Telega Steponas Jaug(e)lis, Minvydas Samuelis, Bożymovskis Jonas. Knyga nobażnystės, 1653. 1660ChB – Chylińskis Samuelis Boguslavas. Biblija, 1660. 1660MP – Morkūnas Jokūbas. Postilė lietuviška 1600. 1701BtNT – Bitneris Samuelis. Naujasis Testamentas, 1701. 1727KNT – Kvantas Jonas, publisher. Naujasis Testamentas, 1727. 1816GNT – Giedraitis Juozapas Arnulfas. Naujas įstatymas, 1816. REFERENCES Ambrazas Vytautas 2001: Lietuvių kalbos pasyvo raidos bruožai. – Acta Linguistica Lithuanica 45, 11–38. Ambrazas Vytautas 2006: Lithuanian Grammar, Vilnius: Baltos lankos. Brudzyński Paweł, Spraunienė Birutė 2021: The Lithuanian passive perfect and its history. – Baltic Linguistics 12(12), 167–207. DOI: doi.org/10.32798/bl.923. Bybee Joan, Perkins Revere, Pagliuca William 1994: The Evolution of Grammar Tense, Aspect, and Modality in the Languages of the World, Chicago, London: The University of Chicago Press. Clark Eve V. 1978: Locationals. Existential, locative, and possessive constructions. – Universals of Human Language, vol. 4: Syntax, eds. J. H. Greenberg, Ch. A. Ferguson, E. A. Moravcsik, Stanford: Stanford University Press, 85–126. Endzelīns Jānis 1922: Lettische Grammatik, Riga: Lettisches Bildungsministerium. Ford Gordon B. 1967: estų und yra in Vilent’as Enchiridion. – Zeitschrift für Slavische Philologie 30, 353–357. Geniušienė Emma 2016: Passive Construction in Lithuanian: Selected works of Emma Geniušienė, Amsterdam, Philadelphia: John Benjamins Publishing Company. Holvoet Axel 2001: Impersonals and passives in Baltic and Finnic. – The Circum-Baltic Languages: Typology and Contact, vol. 2: Grammar and Typology, eds. Ö. Dalhl, M. Koptjewskaja-Tamm, Amsterdam, Philadelphia: John Benjamins Publishing Company, 363–389. DOI: doi.org/10.1075/slcs.55.04hol. Lyons John 1967: A note on possessive, existential, and locative sentences. – Foundations of Language 3, 390–396. 144 Acta Linguistica Lithuanica XCI Disappearance of the auxiliary verb esti in the Lithuanian passive Mikulskas Rolandas 2009: Jungties konstrukcijos ir jų gramatinis kontekstas. – Acta linguistica Lituanica 61, 113–156. Nau Nicole, Spraunienė Birutė, Žeimantienė Vaiva 2020: The Passive Family in Baltic. – Baltic Linguistics 11, 27–128. DOI: doi.org/10.32798/bl.699. Ostrowski Norbert 2017a: Non-verbal Predication in Baltic. Lithuanian yra. – Essays in the History of Languages and Linguistics. Dedicated to Marek Stachowski on the Occasion of his 60th Birthday, eds. M. Nemeth, B. Podolak, M. Urban, Cracow, 463–471. DOI: doi.org/10.12797/9788376388618.27. Ostrowski Norbert 2017b: Od comitativu do spójnika. O pochodzeniu słowiańskiego spójnika i oraz litewskiego ir ‘i’, też’. – LingVaria 12(24), 165–176. DOI: doi.org/10.12797/LV.12.2017.24.10. Ostrowski Norbert 2021: The Baltic *-á-illative. – Indogermanische Forschungen 126, 65–84. DOI: doi.org/10.1515/if-2021-004. Shibatani Masayoshi 1985: Passives and Related Constructions: A Prototype Analysis. – Language 61(4), 821–848. Siewierska Anna 1984: The Passive A Comparative Linguistic Analysis, London, Sydney, Wolfeboro, New Hampshire: Croom Helm. Stang Christian Sch. 1947–1970: Esti et yra dans les Punktay Sakimu de Szyrwid. – Norsk Tidskrift for Sprogvidenskap 14, 87–97 [reprint.: idem, 1970, Opuscula linguistica. Ausgewählte Aufsatze und Abhandlungen, Oslo, 172–180]. Yong Wang, Jie Xu 2013: A Systemic Typology of Existential and Possessive Constructions. – Functions of Language 20(1), 1–30. DOI: fol.20.1.01wan. Watkins Calvert 1967–1994: Remarks on the Genitive, in To Honor Roman Jakobson: Essays on the Occasion of his 70th Birthday 3: 2191–2198, Hague, Paris [reprint.: idem, Oliver, L. (ed.) 1994, Selected Writings, Language and Linguistics, Innsbruck]. Zinkevičius Zigmas 1987: Lietuvių kalbos istorija 2: Iki pirmųjų raštų, Vilnius: Mokslas. Zinkevičius Zigmas 1988: Lietuvių kalbos istorija 3: Senųjų raštų kalba, Vilnius: Mokslas. Zinkevičius Zigmas 1990: Lietuvių kalbos istorija 4: Lietuvių kalba XVIII–XIX a., Vilnius: Mokslas. Zinkevičius Zigmas 1998: The History of the Lithuanian Language, Vilnius: Mokslo ir enciklopedijų leidybos institutas. Straipsniai / Articles 145 PAWEŁ BRUDZYŃSKI Pagalbinio veiksmažodžio esti išnykimas iš lietuvių kalbos neveikiamosios rūšies SANTRAUKA Remiantis 1547–1816 metų tekstais, parašytais trijomis istorinėmis lietuvių kalbos tarmėmis, straipsnyje nagrinėjamas pagalbinio veiksmažodžio esti istorinis nykimas lietuvių neveikiamosios rūšies konstrukcijose. Straipsnį sudaro keturios dalys: įvadas, metodologija, analizė ir išvados. Istoriškai veiksmažodžio būti 3-iojo asmens forma buvo esti. Tačiau dėl semantinių poslinkių yra, iš pradžių buvęs vietos veiksmažodis, palaipsniui pakeitė esti šioje pozicijoje. Toks perėjimas greičiausiai prasidėjo dar prieš raštijos pradžią. Iki XVI amžiaus vakarų tarmėje esti buvo vartojamas tik kaip pagalbinis veiksmažodis, o jo pirminę funkciją perėmė yra. Iki XVIII amžiaus vidurio ir galbūt rytinėje tarmėje esti buvo vartojamas kaip pagalbinis veiksmažodis, atspindintis regioninius kalbos raidos skirtumus. Perėjimą nuo esti prie yra lėmė perteklius ir panašus abiejų veiksmažodžių vaidmuo vietos ir egzistenciniuose sakiniuose. Todėl yra tapo tipine 3-iojo asmens forma, o esti ėmė atlikti įprastą veiksmažodžio funkciją. Įteikta 2024 m. liepos 3 d. PAWEŁ BRUDZYŃSKI University of Warsaw Wydział Polonistyki, Katedra Językoznawstwa Ogólnego, Migowego i Bałtystyki Krakowskie Przedmieście 26/28 00–927 Warszawa, Poland [email protected] 146 Acta Linguistica Lithuanica XCI