|
RECONSTRUCTING PRUSSIAN |
| Today 2 main trends
in reconstructing Old Prussian can be distinguished. One group of
researchers treat Prussian as a language which essentially differs from
Lithuanian and Latvian. Vytautas Mažiulis, Wolfgang P. Schmid, Vladimir
Toporov, Victor Martynov represent this group. Other scholars try to
research Prussian on the basis of Lithuanian data. Mostly outstanding
representatives of this group are William Schmalstieg and Wojciech
Smoczyński. According to V. Mažiulis, disintegration of Common Baltic began in ca. 5th c. B.C. and meant formation of a peripheral area of Proto-Baltic dialects. Future East-Baltic idioms remained in the central area. The peripheral area of Proto-Baltic dialects, in which West-Baltic originated, was also close to area in which Proto-Slavic took its initial shape. Therefore the Prussian language not only resembles Lithuanian and Latvian, but differs from these two languages and often reminds of a kind of Proto-Slavic reconstruction. Nevertheless, although being much more archaic than the East-Baltic languages, Prussian shares with them such typically Baltic features as e.g. non-differentiation of number in the 3rd person of the verb. The latter possesses no formant of the 3rd person at all, if not being an athematic verb. With no doubt, such features are Baltic archaism and not an innovation. This is still argued by those linguists who are afraid to overestimate the archaicity of Baltic in order not to be suspected transgressing principles of classical Indoeuropean linguistics. Part of them are Balts Lithuanians whose main field is research of Lithuanian and Latvian. Others are Slavists, who once have learned the single Lithuanian very well and are proud of this all their life. Therefore, such people try to find in Prussian what is characteristic of Lithuanian or Latvian dialects. What they are doing with Prussian is nothing else but a Lithuanisation of this West-Baltic language. On this site almost all Prussian reconstructions are based on the works of V. Mažiulis, but the following postulates have been developed by Letas Palmaitis: 1. Prussian vocalism reflects a further stage of the Common-Baltic vocalism with its back short /a/ on the place of “Indoeuropean” /o/, /a/ and with its stressed closed and unstressed open variants of the long /ō/ (J.Kazlauskas–V.Mažiulis). The latter became to be generalized on the stressed positions, too, in Prussian of the 13th c., and coincided with the back open /ā/ tending to be diphthongized under the stress parallelly to the diphthongized pronunciation of the stressed long /ē/ (Klusis, cf. spellings in E: the historically stressed allophone of the long /ō/ in the barytone glossis, the generalized unstressed allophone of the long /ō/ in the stressed position in woaltis, the same of the long /ā/ in noatis, parallel to geasnis and contrary to the unstressed stem in sosto, wosee). 2. Spellings of toponyms in German documents demonstrate the same situation in Samland of the 14th c. (Wosispile) (Mažiulis), therefore the “restitution” of the “Baltic” /ā/ in Samlandian catechisms of the 16th c. more likely points out to influence of another Baltic language (Sudovian in Samland) than to a diachronic change (Palmaitis). 3. Diachronic was obviously shortening of the unstressed length in the 14th/15th c. 4. The development of /ā, ō/ > /ū/ after the labials and gutturals with the parallel process of /ē/ > /ī/ (in all positions except the stressed ending where the process was stopped due to the alternation -ē / -ēi, see further) iniciated the process [ū] > [ou], [ī] > [ei] in the middle of the 16th c. 5. The process
/ē/ > /ī/ > (in the unstressed position which often was end of
the word) > /i/ caused merger of the ē- and i-types of
declension. This merger, as well as the merger of the i- and ja-types
of declension, was propped by PALATALIZATION of consonants, the evidence
of which may be seen in such variations of spelling, as acc. etwerpsennian
III 45 / etwerpsennien III 71 < acc. *-(n)jan,
corresponding to etwerpsennin III 65 of the mixed ja-/i-type
of the 16th c. nom. (busennis - acc. gulsennin) and
showing no kind of any ending *-ijan which W.Smoczyński tries to
find here. The same evidence of the palatalization, as etwerpsennian / etwerpsennien / etwerpsennin, are spellings of the acc. masc. schan I 5, III 1034, 10328 = schian III 999 = schien III 1317, 1332 and even fem. schan, schien, schen III 7913, 8116 (the spelling schan reflecting palatal character of this consonant < *sj). 6. Similar to schian, schien are spellings of the ending -an after the undoubtedly palatal j in such instances as twaian 2917 etc. = twaien I 522, twayien II 512, 522. All this demonstrates one no less significant feature more: the distinction between Prussian short /a/ and short /e/ was neutralized after the palatals in the same way as in Lithuanian! The Prussia short /e/ was an open vowel. 7. Prussian shared with Prussian Lithuanian and Samogithian the same feature as absorption of the weak second component of the diphthong by its first strong component (Samog. vākā < vāikāi, Pr. Lith. Kurschat < Kursch'aitis), cf. in Prussian circumflex diphthongs seamis < *zēimīs, moasis < *māisas. A lot of parallel spellings as 3 praes. niswintina III 49 / niswintinai III 51cannot be explained if not assuming the alternations -āi/-ā, -ai/-a, -ēi/-ē, -ei/-e explained by Klusis already in 1989. Such alternations came into being first in secondary circumflexized diphthongs of the shortened -āja, -ēja stem-endings: *-āja > -āi > -ā, *-ēja > -ēi > -ē with occasional preservation of -āi, -ēi partly due to such systemic reasons as existence of forms with the preserved j in one syllable (bēi) or non-suffixal (ettrāi) forms, cf. enwackē, seggē, maitā, billā, billai, enwackēimai, enwackēmai, seggēmai, waitiāmai, waitiaintins. This alternation spread in analogous unstressed forms, too, while the thematic root-verbs already had the long (later shorted) unstressed ending *-ā in the 3rd person of the present (sic!) tense (see further). The form nom. sg. f. giwei III 7521 is an example of the supercorrection of *giwē (oxytone as seen in Latvian dzīve) due to this alternation. THIS SOLVES TWO FAMOUS PROBLEMS: first - that of the impossibility of the later process *-ē > *-ī in the stressed ending where the parallelism with the *-ē/ -ēi < *-ēja was still alive (the process *ē > *i took place in monophthongs but not in diphthongs); secondly - it explains the “mysterious” athematic form 1st sg. asmai as if coinsiding with the 1st pl. asmai. The 1st sg. is a typical sample of *asmi thematized as Lith. asmu, Latv. asmu and testified as asmu < *asmā in III 6723 which had coincided with the 1st pl. asmai in the unstressed position due to the alternation -āi/-ā, -ai/-a before the process *mā > *mū started. 8. Due to the coincidence of the unstressed allophone of the Baltic long /ō/ with the Baltic back long /ā/ and later generalization of the unstressed allophone on unstressed positions in analogous grammatical forms in Prussian, the verbal ending of the 1st sg. *-ō > *-ā > -a (as crixtia tien III 129, V.Mažiulis) coincided with the 3rd person of the ā-stem verbs in the present as well as: with their 1st sg. -a < *-ā < *-āā < *-āō, with the analogous â-stem forms of the thematic root-verbs in the past. This caused GENERALIZATION OF THE ENDING -a first on the 1st person singular and the 3rd person of the root and the ā-stem verbs in the present and in the past tense (a phonetic process which had no morphological alternative), then – on the 2nd person singular with the subsequent generalizing of any form of the 3rd person to this and to the 1st person singular as well ([t]u gīwu < *-ā, quoi tu, as quoi) with the alternative to use ending -sei of the athematic berbs for the 2nd person singular: giwassi, postāsei [not a FUTURE form because Prussian analytical future forms (“become” + the participle) had no origin neither in German (“become” + the infinitive), nor in Polish (“be” + the participle)]. 9. Back character of the short and shortened (in the diphthongs either) a in Prussian caused appearance of the two allomorphs of the genitive ending in plural: of the -un < the stressed *-ōn and of the *-an < the unstressed *-ōn (V.Mažiulis). The latter coinciding with the accusative ending in singular, the so-called CASUS GENERALIS came into being. To avoid misunderstanding caused by this coincidence, artroidal syntactical construction with the “article” stas spread, because the forms of the accusative singular (stan) and of the genitive plural (steisan, steisun) of this pronoun apparently differed from each other. 10. Thus the Prussian “articles” were not mistakes of “German translators” but real phenomenon of the living speech. German propped up development of some purely Baltic processes in Prussian, although its influense also resulted in such barbarisms which were authentic in their turn and not “mistakes of translators”, as e.g. use of the particle prei (German zu) with the infinitive. Although there are enough mistakes in translations of the catechisms, their language reflects REAL LANGUAGE of Samlandian Prussians in the 16th c. (V.Mažiulis). 11. Four cases of the Prussian declension are not an innovation but an archaical feature uniting West Baltic with Germanic and Greek. Only nominative, genitive, dative and accusative forms have constant inintercrossing functions in various Indoeuropean languages, while formants used for the instrumental or locative cases and traditionally declared to be Common Indoeuropean, have intercrossing functions: e.g. the IE *-ois may appear in the instrumental in one languages and in the locative in other ones, or *-ō/ (apohonically) -ē appears as -āt in Indo-Iranian Ablative and as -it in Hittite instrumental. Such intercrossing elements were used for semi-paradigmatical adverbial forms, differently paradigmatized in different languages (V.Toporov, V.Mažiulis). A.Rosinas belongs to scholars who are trying to show that Prussian has lost the instrumental and the locative cases. In his thourough study of Baltic pronouns (Vilnius, 1988) he defends the inveterated opinion as if the form maim 'me', which is twice used in the instrumental and once – in the dative meaning in the 3rd Catechism, is a Common-Baltic instrumental form *manimi equal to Lith. manimi, with the letter n substituted with the dash over the letter a. But of the 3 instances of maim one is without any dash which in most cases is a regular sign of the length/ tone in the 3rd Catechism. The instrumental meaning is expressed in Prussian only and consequently by using the dative form, similarly to its expression in German or in Greek. Of 3 instances of maim one is undoubtedly dative. Oblique cases of the 1st personal pronoun are testified only with the stem men- = Slavic men-, therefore the supposition of Rosinas as if the vocalism has been later changed in this form (but why not in the dative mennei?) is ungrounded. Another typical attempt to Lithuanize Prussian is a reconstruction of the demonstrative stas as of a composite *sitas = Lith., Latv. šitas. A.Rosinas' systematic analysis is amazing, nevertheless it has one minus: not all systematical reasons always become realized in the history of a language. In this concrete reconstruction by A.Rosinas one sees a premise and a conclusion but no proof, because all facts of testified Prussian show this language to be very poor in syncopes, especially when compared with Lithuanian (cf. e.g. Lith. kelnės < *kelinēs etc.). It looks fantastical that this pronoun should have been syncopated in Prussian but not syncopated in Lithuanian or at least in Latvian (with the same *sis as in Prussian!). 12. Forms of the neutral gender have not sufficiently developed in Baltic. In Prussian their development stopped on the stage of singular forms. For the plural either masculine forms or the feminine singular form with the collective meaning were used. >
Back |