Zum Schicksal der Dualformen (ein tschechisch-lettisch-litauisch-sorbischer Vergleich mit einigen allgemeinen Bemerkungen)

ACTA LINGUISTICA LITHUANICA XLVII (2002), 103–107 On the fate of the dual forms (a Czech-Latvian-Lithuanian-Sorbian comparison with some general remarks)* Bohumil Vykypěl Institute for the Czech Language of the Czech Academy of Sciences In the present paper, Czech, Latvian, Lithuanian, and Sorbian morphological forms are compared. A few general observations are made concerning (a) the difference between the diachronic viewpoint of historical linguistics and the achronic viewpoint of comparative linguistics, (b) the projection of diachrony into synchrony, and (c) the difference between two forms of loss of morphological categories. To Prof. Radoslav Večerka on his 75th birthday 1. As is well known, Old Czech had the morphological category of the dual (whereby we understand a morphological category to be the set of morphemes with a particular Skalicka sem, in the case of the dual category, the dual sem). During the 15th and 16th centuries, however, the dual was lost. From the perspective of the diachronic formal morphology of Czech, the question of what happened to the original dual forms is of interest. As far as the written language is concerned, these partly assumed the plural function with some lexemes (for details, see Berger 1999). In some substandard varieties of Czech, the dative-instrumental dual ending became established in the function of the instrumental plural without lexical restriction. More frequent attestations of this are found from the end of the 16th century onward (see Gebauer 1896: 20; Oberpfalcer 1935: 173). This assumption of a plural function by an originally dual ending finds an instructive parallel in the history of Latvian declension. As is well known, modern Latvian has the following dative plural endings: -iem, -’iem, -ām, -ēm, -īm. If the vocalic bases of these endings, i.e. ie, īe, ā, ē and ī, are each interpreted as a stem-forming formant (theme), then -m can be regarded as the actual form of the dative plural morpheme, common to all declensions. In older language monuments, however, one also finds * I thank Markus Giger (Prague) for linguistic corrections. 104 | BOHUMIL VYKYPĚL -iems, -’iems, -āms, -ēms, -īms, i.e. without a vocalic base, -ms. These endings—with the exception of insignificant cases that may also be regarded as misprints—are the only ones found from the first Latvian-written language monuments of the 1580s until the 1630s. From about the middle of the 17th century, variation between -ms and -m begins. From the end of the 17th century, the ending -m predominates, eventually completely displacing the ending -ms during the 18th century. The ending -ms occurs only in West Courland dialects. (See Endzelin 1923: 296f., §231, 307, §255, 311, §270, 320, §283; Vanags 1994: 123.) As is well known, the ending -ms is also found in modern Standard Latvian, namely in the paradigm of the plural designation of the speaker and the addressee mu-ms ‘us’, ju-ms ‘you; you (formal)’. The only means of determining the chronologically primary function of these Latvian forms is the comparative method, that is, in this case, comparison with Lithuanian. Lithuanian has three number sems: singular, plural, and dual. In the modern language, the dual morphemes are, as is well known, peripheral in their use (they are hardly used); nevertheless, they exist as a fully-fledged and paradigmatically complete category. As regards the dative, the following endings occur: in the plural -ams, -oms, -ēms, -ims, -ums, -iems; in the dual -am, -om, -ēm, -im, -um, -iem; that is, without a vocalic base as theme, -ms and -m. On the basis of the comparison of Latvian with Lithuanian, the following account can therefore be offered (see also Rūķe-Draviņa 1977: 98): In the original system, the form -ms had the function of the dative plural and the form -m that of the dative dual; after the loss of the dual, the ending -m came to be used in Latvian with the plural function, while the ending -ms remained only in the designation of the speaker and the addressee (i.e. in categories sharply distinguished from the other designations, since the latter occur in the third person). It is clear that there is an EPISTEMIC difference between the description of the development of the Czech instrumental plural and the description of the development of the Latvian dative plural: the knowledge provided by the former description has a different status from that provided by the latter. While in the first case one describes something that exists, this is only partly so in the second case. This example also makes it possible to observe very clearly the difference between the character of the method of HISTORICAL linguistics and that of COMPARATIVE linguistics. While the former may rightly be called DIACHRONIC, the latter is rather ACHRONIC: at the decisive moment, the historical method cannot be applied here, for, in the words of Antoine Meillet (1934: 49), „on ne peut expliquer historiquement une forme que par une forme plus ancienne“. If, however, no older form precedes it, one can only construct an explanatory model on the basis of comparison with certain other language systems, a model which in itself is bound neither to time nor to space (see Meillet 1925: 15; Hjelmslev ZUM SCHICKSAL DER DUALFORMEN | 105 1968: 34). The difference between the diachronic standpoint of historical linguistics and the achronic one of comparative linguistics can be compared with the difference between history and evolution: the former places the objects under investigation in time and culture and presents a non-necessary sequence of changes, whereas the latter instead establishes the relations between the objects and forms a necessary order of changes. Paul Diderichsen (1966: 307, 338) wrote on this subject: „We may go even further and maintain that evolution has nothing to do with ‘true history’ or ‘time’ in the ordinary sense. The primary task is to arrange variants in a structural hierarchy; it is secondary thing that we can formulate hypotheses about manifestations of these structures following after each other in time and space“. […] „The confusion of the concepts of evolution and history has led to a one-sided view of linguistics. The latter term should be used only when language is seen as part of cultural history (it thus has to do with the geographical distribution of dialects and the like). The former is much more difficult. It implies that we can find general laws that apply to the change of language“. Undoubtedly, it would be possible to construct other explanatory models for the Latvian dative plural endings. We believe, however, that the one we have presented above (and which is indeed self-evident) has the advantage of being the simplest, i.e. it leaves the fewest unanswered questions. 2. As mentioned, modern Standard Lithuanian has no longer used the dual morphemes since approximately the interwar period (cf. Zinkevicius 1980: 180). Most Lithuanian dialects also have no dual. As regards the dative plural, different situations are found in individual dialects, corresponding to those in the various periods of Latvian linguistic history. The Samogitian dialects have -ms (as in the oldest Latvian linguistic monuments). In some Central Lithuanian dialects, there is variation between the endings -ms and -m (as in Latvian linguistic monuments from the 1630s through the 18th century). Finally, the Aukštaitian dialects have the ending -m (as in Latvian since the 18th century). The distinction between the dual ending -m and the plural ending -ms is made only in dialects around Klaipėda (as in older Lithuanian written usage). (Cf. Zinkevicius 1966: 205, 234f.; 1980: 181.) It can therefore be said that the Aukštaitian dialects have treated the dual ending -m in the same way as Latvian. This Aukštaitian–Samogitian dichotomy in the form of the dative plural morpheme has a counterpart in the Old Lithuanian period. There is, in fact, another ending here, namely -mus. The endings -mus and -ms compete with one another, and in the course of the 18th century the ending -ms gradually prevails. This process proceeds more quickly among authors from the Samogitian-speaking region: among them, -ms already predominates in the oldest texts, whereas this occurs only later among Aukštaitian authors (cf. Senn 1966: 95; Zinkevicius 1980: 193f.; 1987: 180). 106 | BOHUMIL VYKYpĚL From a general point of view, it should be noted that the comparison of the dative plural endings in Lithuanian dialects with the history of Latvian is clearly an excellent illustration of the old thesis of the Prague School, according to which diachrony is projected into synchrony either vertically, through the social and stylistic stratification of the language, or horizontally, in dialects. The latter was the case in our Lithuanian example. 3. The fate of the morphological dual category in the languages mentioned is instructive in another respect as well: it can clearly illustrate that the loss of a morphological category can take two forms. One type of loss may be called disintegration, the other disappearance. In the former case, there is an actual disintegration of the category, or of its morphemes—namely, a separation of morphemic forms from their functions. However, the relevant forms, or at least some of them, do not disappear from the language; rather, they are assigned other functions. In the case of the disappearance of a morphological category, the morphemic forms are simply abandoned; people stop using them. As far as the dual category is concerned, we have seen that it has disintegrated in Czech and in most Lithuanian dialects, and we have also assumed its disintegration in Latvian. Disappearance has occurred in modern Standard Lithuanian (if this process may be considered complete here) and in some western Lithuanian dialects. The situation in Sorbian is also interesting and instructive in this respect. The dual exists in the written language (in both of its variants). In spoken language, however, one can observe the waning of the dual (the process of loss is not yet complete; cf. Faßke, Michalk 1981: 428–430). The dialects can be divided into three groups with regard to the dual (cf. Faßke 1975: 19–36). In some dialects the dual exists and is used (the situation varies with regard to the number of positions in which the dual can be replaced by the plural). In other dialects the dual exists, but can be replaced by the plural in all positions (this is therefore a waning of the category). Finally, in the group of southern dialects, the morphological dual category has been lost through disintegration: former dual forms occur here as optional variants of plural forms or as plural forms of certain lexemes. It is obvious that historical linguistics can capture both forms of loss of a morphological category, whereas comparative linguistics can capture only the first of them, disintegration. In other words, if a morphological category is lost in the prehistoric period of a language, this can be established only if the loss took the form of disintegration. Thus, no amount of ingenuity in the world can reconstruct the Latin ending -ibus on the basis of the Romance languages. Zum Schicksal der Dualformen 107 LITERATURVERZEICHNIS Berger, T. 1999: Zur scheinbaren Produktivität des Duals im Tschechischen. Girke, W., Guski, A., Kretschner, A., Hrsg., Vertragdbinnogenetiv. Festschrift für Helmut Jachnow, München: Otto Sagner, 9–20. (Specimina philologiae Slavicae, Supplementband 64.) Diderichsen, P. 1966: Helhed og struktur, Köbenhavn: C. E. G. Gads Forlag. Endzelin, J. 1923: Lettische Grammatik, Heidelberg: Carl Winter. Fasske, H. 1975: Sorbischer Sprachatlas 11. Morphologie. Die grammatischen Kategorien – Die Paradigmatik des Substantivs, Bautzen: Domowina. Fasske, H., Michalk, S. 1981: Grammatik der obersorbischen Sprache der Gegenwart. Morphologie, Bautzen: Domowina. Gebauer, J. 1896: Historická mluvnice jazyka českého III/1, Praha–Vídeň: Nákladem F. Tempského. Hjelmslev, L. 1968: Die Sprache. Übers. von O. Werner, Darmstadt: Wissenschaftliche Buchgesellschaft. Meillet, A. 1925: La méthode comparative en linguistique historique, Oslo: H. Aschehoug & Co. Meillet, A. 1934: Introduction à l’étude comparative des langues indo-européennes. 7ᵐᵉ édition, Paris: Hachette. Oberpfalcer, F. 1935: Jazyk knih černých, jinak smolných, Praha: Nákladem České akademie věd a umění. (Rozpravy České akademie věd a umění. Třída III. Číslo 73.) Rūke-Dravina, V. 1977: The Standardization Process in Latvian, Stockholm: Almqvist & Wiksell. (Stockholm Slavic Studies, 11.) Vanags, P. 1994: Die Entwicklung der Kasusendungen in den ältesten lettischen Sprachdenkmälern. Linguistica Baltica 3, 121–130. Senn, A. 1966: Handbuch der litauischen Sprache I, Heidelberg: Carl Winter. Zinkevičius, Z. 1966: Lietuvių dialektologija, Vilnius: Mintis. Zinkevičius, Z. 1980: Lietuvių kalbos istorinė gramatika I, Vilnius: Mokslas. Zinkevičius, Z. 1987: Lietuvių kalbos istorija II, Vilnius: Mokslas. Bohumil Vykypěl Ústav pro jazyk český AV ČR etymologické oddělení Veveří 97 CZ-60200 Brno [email protected] Gauta 2002 12 05