Lietuvių „dativus cum infinitivo“ ir latvių infinityviniai santykiniai sakiniai

ACTA LINGUISTICA LITHUANICA LITHUANIAN LANGUAGE QUESTIONS XLII (2000) Axel HOLVOET Institute of Lithuanian Language, Vilnius LITHUANIAN DATIVUS CUM INFINITIVO AND LATVIAN INFINITIVES IN RELATIVE SECTIONS This article discusses the evolution of some infinitive constructions in Lithuanian and Latvian languages, especially with regard to the Lithuanian dativus cum infinitivo construction and its genetic counterparts in Latvian. In Lithuanian, there are two variants of the dative cum infinitive construction, denoting the purpose of the object: in one case, the noun noun performs the function of the object, e.g., I bought a horse near the fields, in the other — the function of the subject, e.g., I poured milk to latch the puppies (examples according to Vytautas Ambrazas 1995). However, these variants are derived from one design. It seems that the dativus cum infinitivo arose in times when the communicative was still a nominal form, probably a noun of the action name [nomen actionis]*, and can therefore be described as a “dual noun of purpose (purpose) dependent on the nucleus of the sentence” (Ambrazas 1995: 79). Therefore, only later, when the communicative became a purely verb form, a secondary connection between the communicative and the noun as between the predicate and the object or subject dependent on it could be formed. Now, when the communicative has lost its original nominal character, we can already talk about the function of the noun as an object or subject in relation to the communicative. Changes in the nature of the communication may lead to further development of the dativus cum infinitive construction and eventually to its disappearance. This process occurred in the Slavic languages, where the dativus cum infinitivo is palidified, but is no longer used anywhere (Ambraz 1995: 83 cites evidence in Old Church Slavic and Old Czech and literature). In Latvian, there are equivalents only for the second of the above mentioned variants: as a relative phenomenon, the dativus cum infinitivo is paliuded, in which the noun performs the function of the subject of the communicative verb (see Endzelins 1951: 990): lai paliek galotnite putniniem uzmeties “so that the top remains for the birds to land". ! The author is grateful to Prof. Vytautas Ambrazas, who kindly agreed to read and comment on the first draft of this article. The author is solely responsible for any inaccuracies or errors in the article. ? Plg. Stang (1966: 447-448), Ambrazas (1995: 74-75). 105 It seems that in the common Latvian language, in ir most Latvian dialects, the first variant of su the object-user has no equivalent- .” The sentence is complemeir su nted by the relative pronoun kur before the communicative construction (see section 4.4). Endzelins 1951: 990): will stay in the end, where the bird will nest "to leave a tip for the bird to land where | the bird could land". In this way, an infinitive relative sentence is formed, a syntactic — type very characteristic of Latvian, but almost unknown in Lithuanian. More precisely, in Lithuanian (similar to Slavic languages) only one variant is used, in which the relative sentence is jo dependent on the subject of the existential sentence not formally expressed (e.g., here is not to eat) or from the | possessive ką construction of the verb to have a formally unexpressed su object (e.g., I don't have to eat in | the morning); ką for these constructions, see Ambrazas ir kt. 1994: 670. This is the case in the Old Church Slavonic and in the New Slavonic languages. s. b. sl. ne imąte česo ėsti, (nie) 1. mają ješč pan. co ir In Latvian, such relative sentences can also denote nouns, e.g. paņmu gramatas, lasit atpūtas laika ir Es ko “I took books to read to read) (| during the rest’. Such infinitive relative sentences differ from simple relative sentences (in which the personal form of the verb performs the function of the taring) in that they express the real object-related state of the defining noun ne su tu (e.g., Grammar, ko es lasu "Books I read"), only aš the o possibility su related to that object, the purpose of to the object (Gramatas, lasit ko *books to read, books that can be read"). Since the Latvian infinitive relative sentences have no equivalents in either Lithuanian or Slavic languages, it is possible to assume that such sentences are modernizations of this language. It is obvious that they are very close in function to the utility. su Both constructions are characterized by an intermediate function between the circumstance of the test and the object. Vytautas Ambrazas writes about the communicative noun that this construction only occasionally has a purely indicative function (e.g., in such sentences as That axe for wood chopping got quite sharp), usually o ji „related su to the verb and to the noun or pronoun it controls”, e.g., Atvedžiau karčių fence to weigh (Ambrazas 1995: 77), so its function is close to the circumstance of the object. The Latvian infinitive relative sentences, as relative (noun) sentences, are formally related to the noun, but can also be understood as circumstance sentences, e.g.: panėmu gramatas, ko lasit atpūtas laika | panėmu gramatas, lai tas lasitu * According to information provided by Lydia Leikuma, dative cum infinitive constructions with an object-using noun are also found in Latgalian dialects. Until data on the nature and prevalence of this construction in Latgalian dialects are published, it is difficult to assess their significance for the interpretation of the development of the dativus cum infinitive construction in Latvian. 106 rest time. Infinitive relative sentences in Latvian belong to the type of sentences that F. R. Palmeris (Palmer 1986: 178-179) called ‘relative purpose clauses’. In this type of sentences, which occupy an intermediate position between the adverbial object sentences of the indicative ir (relative) sentences, the verb forms characteristic of adverbial object sentences are used: often this is the general, but there is also the conjunctive or periphrastic constructions in the optative meaning." There are also differences between the communicative su noun and Latvian infinitive relative sentences: in the first case, the noun can only denote the subject or object of the action, in the second there are no such restrictions. 0 Since the infinitive relative sentences dependent on the subject of the existential construction (there are what-to-eat- -type sentences) are extinct in Lithuanian and Slavic languages, it is possible that such sentences spread further in Latvian and were used also to denote nouns. In this case, however, infinitive relative sentences were supposed to take the place of other, older constructions, and it seems quite likely that they were used precisely in place of infinitive constructions. The subject of the communicative action in Latvian infinitive relative sentences, if formally expressed, is always expressed by the noun as in any other infinitive sentence. Therefore, the communicative with the user is replaced by an infinitive relative sentence without additional morphosyntactic changes: it is enough to insert a relative pronoun or adverb before the communicative (galotnite putnam uzmeties —galotnite kur putnam uzmeties). In cases where the noun performs the function of an object, the difference between the two constructions is more pronounced, since the communicative object in the infinitive relative sentence is expressed not by the noun but by the adjective: Nopirku žirgu, ar ko laukus art. Latvian infinitive relative sentences must be a relatively old phenomenon, as they are probably also present in the structure of the present tense. As is known, the debt comes from the type of compound sentence with a relative sentence, and the prefix of the debt ja-(in some dialects ju-, derived from the original form of the ending or noun *juo) is a trace of a relative pronoun. The construction wait and see must have originated from * In Latin, the conjunctive was used here, e.g., Scribebat... orationes, guas alii dicerent "Wrote speeches for others to say" (Cicero, Brutus, see Palmer 1986: 178). In contemporary Latvian, the direct pronoun with lai is used in relational usage sentences in addition to the communicative pronoun, and is often considered to be a substitute for the 3rd person imperative pronoun, e.g., nolėma iecelt kasieri, kas lai to naudu uzglabatu (A. Deglavas) “notarė paskirti kasininką saugoti pinigus’. The distribution of communication and constructions with lai in Latvian relative purpose sentences has not been investigated. 107 previous structures *and wait for him near where *ju corresponds to the current di relative inward. True, in the present language the prefix is added to the communicane tive form (except for the form jūbūt “must be', which is based precisely on communicativeneo ss), to the present tense 3rd person singular forms. Ne all the researchers agree Prellwitz's ir Endzelyno raised the idea that The 3rd person form here takes the place of the previous communication. Mühlenbach (Mūhlenbach 1907: 322) thought that the debit prefix was iš originally added to the impersonal meaning of the 3rd person forms, ir man (and) jubrauc explained how “y MeHs1 I'm sorry, Ha I'm sorry. It seems that ir Fraenkel was inclined To consider the 3rd person form as the primary basis of the debitive, because it brings closer to Latvian su debitive constructions occurring in Lithuanian dialects su 3rd-person form in place of communication, such as him no longer, who eats “he will have no more to eat, jam zu nebebus ėsti' ko (Fraenkel 1928); iš to may have inferred that there jų was a genetic connection between them. However, it has not yet been proved that the aforementioned constructions of Lithuanian dialects are old. Iš Fraenkelis also quotes the construction nėr kam ažtaria rūsciuos žūdeliuos ‘nicht einer ist da, der... For me could insert pronoun at angry Worten' (Fraenkel 1936), similar design provided ir "In the Grammar of Contemporary Lithuanian Language": No one is going to the market (Ambrazas ir et al. 1994: 671). Utility is understood here only if we accept that The 3rd person form has taken the place of communication: there is no one to go to the market, no one to intercede for. The 3rd person form probably became established here because it was used in constructions where the ji relative pronoun meant the subject of the action, e.g., is not who works ‘non est, qui laboret' (cf. Ambrazas et al. 1994: 671). Thus, the aforementioned Lithuanian constructions are not sufficient grounds to reject Prellwitz's ir Endzelynas- 's idea that the prefix of the debitive was originally added to iš the communicative, although it seems that Endzelynas himself later abandoned his earlier opinion, probably due to the influence of Fraenkel (cf. Endzelins 1951: 885-886; the author here does not repeat the previous opinion, but quotes Fraenkelio iš lietuvių tarmių pateiktas konstrukcijas). Change of the Latvian debit form The reasons for the 3rd person form have not yet been explained. It is possible that the reasons for this change were similar to those that led to the inclusion of the 3rd person in the abovementioned Lithuanian language constructions, but certain arguments contradict this. As mentioned earlier, Lithuanian constructions nėra kam turgun važiuoja, nėra kam užtaria had to have arisen due to the influence of such constructions, in which the relative pronoun meant the actor of the action (e.g., nėra kas dirba), and such constructions could not have influenced the development of the later debitive, since they are not based on any variant of the debitive. Firstly, as the prepositional forms (*juo, *ja) show, the debitive originated from constructions with a finite, a possessive or a noun in a relative sentence; traces of a noun or a noun form (corresponding to similar forms in Lithuanian constructions there is no one to defend / there is no one to defend)) are not found here. Secondly, the phenomenon of the object nominative, which is still quite alive in Lithuanian, could have contributed to the emergence of a special dialectal Lithuanian construction: the use of the nominative what here probably has to be compared with what to do, what was done type constructions. 108 su (see section 4.4). Ambrazas 1995: 91-93). Such an object nominative could have appeared in relative sentences communicatisu vely (there is nothing to eat = there is ką / ko eat), į which were then included 3rd person form (there is no one eating) “there is ko no being’). In Latvian, such an influence of the object of the noun seems unlikely. On the other hand, the fact that the original basis of the debtor was precisely the relative sentence communicativ, can be su supported by several arguments: in the relative 1) form jūbūr" ; 2) the fact that now in ir Latvian infinitive relative sentences are used, which can be directly related to the primary, preserved in the relative form in the su debt, only the relative root *yobuvo was replaced by the noun root *k"o-; that 3) relative usage sentences, to which the relative sentences discussed here belong, are more characterized by the use of direct reference, other forms of the verb; in ne that in some cases the o relative object sentences in Lithuanian, Slavic 4) languages are paused in the communicative. If Prellwitz’s and Endzelynas’s opinion about infinitive su construction as the basis of Latvian debts ir is correct, ir then one can assume that infinitive relative sentences in Latvian are a relatively old phenomenon, dating back to the prehistoric period of the development of this language. This is shown by the archaic, relatively preserved use of the noun root *yo as a relative pronoun instead of the later *k*o-". However, they probably did not exist in the Baltic protolanguage: ten komunikatis 3 More of these relict forms could appear, but since the old Latvian texts contain many morphological errors, the interpretation will remain controversial. jų Uncertain are, for example, the forms used by Elger daud3e man no iums iarunnat vnd iatėsot (Evangelia et epistolae 19r), if we find next to it also mums nenek wairak iarunna (ibid. 12r, see Evangelien und Episteln ins Lettische iibersetzt von Georg Elger... Bd. 1. Texte, hrsg. von Karlis Dravinš, Lund 1961). ® V. Ambrazas also mentions constructions with the present tense participant, which he correctly compares with the dativus cum infinitivo and with Latvian constructions from which the debitive arose: nebuvo kas bedarbas, nėra kas valgą (see Ambrazas 1990: 115-116). In general, the use of participants in relative sentences can be considered an archaic phenomenon inherited from protolanguage (see Ambrazas 1989; 1990: 121-122), so a priori it seems unlikely that the participants have recently penetrated into the mentioned constructions, occupying the place of communication. It would be more likely to expect a change of participants in other forms than their further spread. Nevertheless, the object of the nominative of the participle construction is not who eats “there is nothing to eat' is probably related to the initial communicative use instead of the later participant, because it would be difficult to interpret it otherwise than as a trace of nominativus cum infinitivo. In those cases where the relative pronoun performs the function of the subject (there was nobody unemployed), the use of the participant seems to be an archaism. In addition to the 3-person form, a similar process may have occurred in infinitive relative sentences at the same time. "Although the relationship between the relative root *ye/and the demonstrative root *in the Baltic languages is not quite clear (see Brugmann 1911: 331, 347-348, Stang 1966: 233-234), there is no doubt that the relative use of the root *ye/is inherited from the proto-language, and therefore the archaicness of the noun root used in the composition of the debitive is beyond doubt. A. Rosinas (1975: 169) denies the relative usage of the pronouns he, she (which he does not distinguish from the relative *ye/o-) in the Baltic proto-language, but he does not take into account the origin of the Latvian debt. 109 was probably still a nominal form. The primary nominal nature of the relation is evidenced by the Lithuanian dativus cum infinitivo. The above structural type *ir lauks juo art eventually acquired a modal function: in modern language, lauks jaar simply means ir ‘need close to the field’. However, in some dialects some of the original iš meaning of these constructions has been preserved; this is illustrated by the previously mentioned variant of the sentence su in the debitive used instead of the communicative (see section 4.4). Endzelins 1901: 73 = 1971: 149-150): lai stūvēja virsūnite putniniem jauzmets. Very interesting are Miihlenbacho (1907: 316-317) iš Local dialects cite mi examples of similar use of debitive. In this dialect there was a differentiation of two variants of the debitive: one ju, the prefix form iš ja-, has the su modal function of the debitive of the present Latvian common language: man nau jabraūc ‘st He I'm sorry. The second variant, the prepositionsu al form ju-, has retained the original function of an infinitive relative sentence. In the examples cited by Mūhlenbach, the prefix jua performs the function of a noun: man nau jubraūc 'MHe He Ha YeM eXATb'; bral'am nau zirga juecé 'y Gpara HeT JIONNAaJIM, YTOObl OOpOHHTb'; Ienas aquatica ‘IPUHECHU BOJIbI JŲISI MbITbA'; we would have had the powder you say ‘ecm Obl y HAC GUT TIOpoXx, YTOOBI CTpeJISTb'. This function also corresponds to the prepositional form, since the prefix ju-, shortened from *juo in the pronoun position, can be considered either a suffix or a suffix (for the suffix form *juo, which formally corresponds to the feminine form ją, see Stang 1966: 437, Endzelins 1951: 518). Among the examples cited by Mühlenbach, we do not find any in which the prefix is understood as a suffix — perhaps because the suffix of the object has been displaced by the adjective, which has penetrated from negative constructions (for a generalization of the prefix and its form, see Holvoet’s paper submitted to the press). The variant with the relative pronoun performing the function of the object became the basis of the debt as a modal construction: this is shown by the example from the Vietalva dialect man nait jabratic “st He 10JKeH exaTr'. In this variant, the modal meaning eventually replaced the original meaning, which remained in the construction of old Latvian texts teems ne bij ja-ehd ‘they had nothing to eat' (Glucks Bible, Mark VIII.1, see Mūhlenbach 1907: 319). The original meaning was retained in the paliuded variant with the pronoun in the Vietalva dialect, later turned into the prefix *juo-> ju-. It is this variant that can be compared with the Lithuanian variant of the dativus cum infinitivo, where the noun denotes the object of the verb. Since the prefix *juojučia is a conjunct, the final of the object, the position of the object at this adnominal debitive remains free, 110 > o ne ir ją can take the final part of a noun: pariam kač pavadu zirgu jusapin ‘BO3bMH XOTb IIOBOJĮ, YTOOBI MM 3aIlyTaTh It's not a word. Get a leash for the horse. A similar example (from the lyrics) quotes Endzelynas (1901: 73 =1971: 149; 1951: 973) iš in other dialects (here generalized prefix form ja-): devu sav kumelinu ruožu darzu jūecė liet. I gave my little horse to the rose garden; lai stav [naudal ... guovi janopérk liet. Let there be money to buy a cow. Similar debts with the prefix “ja” in the meaning of “owner” are also found in Old Latvian texts, such as Elgerium In the Gospel ir Epistles: vnd kad tems ne by iamakfa, abbems pedewwe 'not havetibusillis, unde redderent, donavit utrisgue' (64v = Lk 7, 42). It seems quite likely that the relative pronoun of the construction was originally su one of the main infinitive variants of relative sentences; This explains their grammatalization, i.e. the appearance of a particular grammatical form of agglutinative nature when the pronoun and the verb are mature. This type is probably related to the Lithuanian dativus cum infinitivo as its genetic counterpart. Both constructions originated from proto-language constructions with a communicative purpose (purpose) meaning, which were replaced in Latvian by inserting a relative pronoun. The Lithuanian construction with pure communicative purpose (without subject or object) corresponds in the Localva dialect to the non-objective debitive with the prefix ju-: brolis neturi arklio akėti = *bral'am nav zirga juo ecėt > bral'am nau zirga juecé. The infinitive construction of the pronoun has been changed here by inserting a relative noun; as in the constructions with the noun performing the function of the communicative subject, there was no need to change here. The Lithuanian communicate with the object-functioning participle corresponds to the variant of the Vietalva verb with the object-finalis: pasiimk pavadį arkliui supančioti = *panem pavadu žirgu juo sapit > panem pavadu žirgu jusapin. It was not only necessary to insert a noun here, but also to change the verb, because the noun became the object of relational sentence communication. It is obvious that the absence of the dative cum infinitive construction with the object of the noun in Latvian is related to this process, it is only unclear at which stage of this process the object end was included. There are two possibilities: then infinitive relative sentences displaced the old dativus cum infinitivo 1) the object ending was included in infinitive relative sentences and only 111 (the connection between the appearance of infinitiviniy relative ir sentences with the disappearance of dativus cum infinitive in this case would be indirect); or the relative 2) pronoun was included in the simple dative cum infinitive construction, only after the į noun was replaced by the adjective. o This last assumption is supported by the position of the inserted pronoun: the pronoun could probably grow into a verb in object constructions su if the original word order was su jo *panem leash horse him sapit, o ne *Lead him to the horse. If in the above examples the order of the pallets is not ta random object ir (zirgu, ruožu darzu, guovi) always stood before the pronoun communicative complex, so the ir subsequent communicative object initially had to be an object, was dependent on iš the noun itself, which before ne inserting o jvardi simply to su (as the beneficiary of the purpose or purpose) was associatir ed with communicating. In this case, your current verb form (galininka) it could have acquired only later, when the infinitive relative sentence appeared, the original form of the verb must have o been jo different: it was probably a noun. This means that the construction panem pavadu zirgu jusapin may have originated from the simple dative cum infinitive construction. The fact that both dativus cum infinitivo iš variants had to exist already in the Baltic protolanguage, and that later in Latvian they were replaced by other constructions, taking into account the data of Lithuanian and Slavic languages, is a priori probable. It seems that the data of the o Latvian language itself allows to trace the traces of this construction in the development of infinitive constructions of the Latvian language. Received 25 May 1999 Ambrazas V. 1989: Baltic languages and the reconstruction of the Indo-European complex sentence. — General Linguistics 29, 52-59. Ambrazas V. 1990: Sravnitel'nyj sintaksis pricastij baltijskich jazykov /Comparative syntax of the Baltic languages, Vilnius: Mokslas. Ambrazas V. 1995: Development of communicative constructions in Lithuanian language. —Lietuvių kalbotyros klausimai 33, 74-109. 1994: The Grammar of the Contemporary Lithuanian Language, Vilnius: Mokslo ir enciklopedileidyklju a. Brugmann K. 1911: Comparative Grammar of the Indo-European Languages, Vol. II, 2. T,, Strasbourg: Karl J. Triibner. Endzelins [Endzelin] J. 1901/1971: Origin and Use of the Latvian Debitive. —Bezzenberger's Beitrige 26, 66-74. (Perspausdinta: idem, Darbu izlase 1, Riga: Zinatne, Endzelins [Endzelin] J. 1905/1971: On the Origin of 143-150.) the Latvian Debitive. —Bezzenberger's Beitrige 29, 320-321. (Reprinted: idem, Selected Works 1, Riga: Zinatne, 288289.) Endzelins J. 1951: Grammar of the Latvian language, Riga: Latvijas Valsts izdevniecība. 112 Fraenkel E. 1928: Lithuanians Contributions. — Indo-European languages Research 46, 44-57. Fraenkel E. 1936: Der Status of the Research of the im Vilnius areas spoke Lithuanian. — Balticoslavica 2, 14—107. Bolvoet A. (for press) the rise and On grammaticalisation the of Latvian debitive. — Linguistics Baltica. Mihlenbach [Mijulenbachs] K. 1907: O Debit card. — Russian Academy of Sciences Russian i Academy of Sciences Naukv 12/3, 313-333. Palmer FE R. 1986: Mood and modality, Cambridge: University Press. Prellwitz W. 1904: Zur Origin of the Latvian Debts. — Bezzenbergers Beirut 28, 319. Rosinas A. 1975: Ar balty *I-, *ioresp. *i, *iakamienai were relative? — Baltistica 11 (2). 165-170. Stang Chr. S. 1966: Comparative Grammar of the Baltic Languages, Oslo etc. : University Press. THE LITHUANIAN DATIVUS CUM INFINITIVO AND THE LATVIAN NON-FINITE RELATIVE CLAUSES Summary The dativus cum infinitivo, archaic an infinitive construction which probably arose from double dative purpose a of (the dative being originally the dative of a verbal noun), is well retained in Lithuanian but has hardly left any traces in Latvian. Of the two varieties attested in Lithuanian, with the dative subject as the of infinitive (jpyliau pieno šuniukams uzlakti ‘1 poured out some milk for the dogs lap’) and with the dative its to object as (nupirkau arklį laukams arti ‘I bought a horse to plough the fields with’) only the former vestigially attested is in Latvian. In the paper attempt an made bring is this fact to connection with the development in non-finite relative clauses of in Latvian. In Lithuanian (as in Slavonic) the occurrence of such clauses is restricted to but few a types, and in Latvian their scope was probably extended relative pronouns as were inserted constructions with infinitives expressing in purpose. In reconstructing this process, evidence may drawn from some dialectal be varieties the debitive, which, of in a fossilised shape, represents original an non-finite relative clause type with the relative pronominal stem *ye/o-. In the dialect of Vietalva an adnominal variety of the debitive with the prefix ju- (an original instrumental) attested, is which object in may occur an (pariam kač leash horse jusapin ‘take with you lead a shackle to the horse with’). This seems to to correspond exactly the to Lithuanian dativus cum infinitivo with the dative as an object (take the leash to supancioti the horse), not only functionally, but also genetically. Though the original dative was replaced with an accusative as the original infinitival construction developed into an infinitival relative clause, a direct connection with the dativus cum infinitivo is suggested, among other things, by word order, as the incorporation of the relative pronoun into the verbal forms (*juo sapit > jusapin) suggests an original word order *panem pavadu zirgu juo sapit, where the second noun, before being reinterpreted as an object in the relative clause, must have been in adnominal position, and must therefore have been in the dative rather than in the accusative. 113