Zur Prosodie in der Typologie

ACTA LINGUISTICA LITHUANICA L (2004), 127 - 134 On prosody in typology BOHUMIL VYKYPEL Ustav pro jazyk česky AV CR In the present paper an attempt is made to incorporate prosody into Vladimir Skalicka’s typology. Examples are taken from Lithuanian, English, and Chinese. 1. In our typological description of Lithuanian, Markus Giger and I have found that in the question of homonymy (polyfunctionality) of grammatical formants (phonological forms) there is a fundamental difference depending on whether the prosody is included in the description or not (cf. Giger, Vykypél 2001: 50f., 52, 71). This can be demonstrated here briefly (again) on the basis of the declination of the nouns. At the segmental level, homonymy is observed in Lithuanian, indicating the flexible type: the 67 Lithuanian morphs (morphological forms) correspond to 47 phonological forms; the homonymy is therefore on average 1: 1.43. However, if we also include the suprasegmental level, it must be assumed that each ending is characterized not only by its phonological form (i.e. by its segmentation or by a certain phonematic composition), but also by its inherent prosodic property, i.e. by the interaction it enters into with respect to accent with the stem. So if we include the suprasegmental level and also take into account the prosodic characteristics of each ending, there is no homonymy in Lithuanian, because each ending behaves differently. Thus, the ending -os on the suprasegmental level corresponds to the endings -os, and -05,; -0S, accepts the accent offered by the stem, but is not active itself, i.e. does not draw it to itself and expresses genitive singular, while -os, does not accept the accent offered by the stem, does not draw it to itself and expresses nominative plural. The situation is still different if one also takes into account the prosodic properties of the individual lexical stems. The corresponding accentuation of a word form is the result of the prosodic properties of the stem and the ending: the decisive factor is to which accent class the lexem belongs; the differentiating (i.e. the eliminating the segmental homonymy) power of the ending can then be neutralized by the properties of the stem, and segmentally identical forms like rankos (genitive singular) and rankos (nominative plural) remain homonymous (both are accentuated on the stem: rankos), while equally segmentally identical forms like galvos (genitive singular) and galvos (nominative plural) are distinguished by the accent: galvös vs. gälvos. 128 | BOHUMIL VYKYPEL We have found that an analysis limited to the segmental level is obviously incomplete for Lithuanian and that, for example, Lithuanian and Latvian are not comparable in terms of the homonymy of their formants on the suprasegmental level, because they have different prosodic characteristics that have different morphological effects. 2. Horst Geckeler (2001: 114) also encountered a similar problem in his typological description of English: “The frequent conversion between noun and verb in English (e.g. (the) fish > to fish) is stronger on the graphic level than on the phonic level, since on the latter many word pairs differ at least suprasegmentally, i.e. by the different accent, e.g. (the) compound /to compound, (the) consort | to consort, (the) increase /to increase.” Conversion represents a certain lexical counterpart to the homonymy of the grammatical formants: Similarly as the function of homonymous grammatical formants can only be identified by their connection with a certain class of lexical formants, so also the function of homonymous lexical formants occurring in conversion can only be identified by their connection with a certain class of grammatical formants (or at least with a certain class of content elements that are not expressed explicitly). And similarly as the segmentally homonymous grammatical formants can be distinguished by their prosodic properties, so the segmentally homonymous lexical formants occurring in the conversion can also be distinguished by their prosodic properties; The former is the case in Lithuanian, the latter in English!. 3. Vladimir Skali¢ka did not deal in his theory with a problem of this kind, i.e. with the effects of prosody in morphology. This is not surprising if one considers that Skalicka’s language typology was initially built on the “experiential data” from the “prosodic blunt” languages such as Turkish, Hungarian or Czech. Tilman Berger (1990) made an attempt to create an accent typology within the framework of the Prague typology. His explanations, however, represent a phonological implication of the basic “morphological” characteristics of the individual language types and thus concern the question of what functions the accent has in the different language types or —more generally— how the phonological system is constructed in the individual language types or in languages with the dominance of a certain type?. However, he does not deal with the role of prosody in the construction of the signifiants and their typological! Many examples of the grammatical function of prosody can already be found in Sapir (1921: 82-85). ? Skalicka himself has also devoted himself to the phonological implications of his typological theory (cf. Skalicka 1967; 1979: 238-244, 307-311). PROSODY IN TYPOLOGY | 129 Evaluation. Jifi V. Neustupny (1959) has contributed somewhat to this problem in his comments on accent in Japanese and Russian (see below). Thus, the question arises as to how the morphological effect of the prosodic elements, which eliminate or reduce homonymy and thus problematize it as a typological criterion, is to be assessed. One solution consists in completely deleting homonymy from the typological consideration as a typologically relevant phenomenon. This would also correspond to another fact that seems to make homonymy problematic, namely the need to distinguish between an intraparadigmatic and an extraparadigmatic homonymy (see here Giger, Vykypėl 2001: 55f., 71). The better solution, however, is another (which also seems more natural): homonymy is retained as a typological criterion, at a lower position in the implication hierarchy of type constructs (i.e. it is derived only from the basic properties of the types), and the prosodic properties are incorporated directly into the types of the construction of the grammatical formants, i.e. at the same level as the basic properties of the types. The basic characteristics of the types, in the formulation of Petr Sgall, consist in how the grammatical functions are expressed. While the lexical functions are expressed the same in all language types, and the grammatical functions are expressed differently: In the polysynthetic type by the arrangement of the lexical formants; in the isolative type the same as the lexical functions, i.e. by zwar durch eine Phonemkette, d.h. durch phonologische Formen oder Formanten, formants, whereby these are independent and appear as words; in the agglutinative type the same as the lexical formants, i.e. by formants, whereby these are non-independent and appear as affixes; in the flexible type by a modification or alternation of the lexical formants, whereby this occurs at the end of the formants; in the introflexive type by a modification or alternation of the lexical formants, whereby this occurs inside the formants (cf. e.g. Sgall 1992: 204f.). These differences can also be understood as differences in the degree or degree of discrete character (separability) of the grammatical and lexical formants (Jaroslav Popela speaks about the degree of fusion between lexical and grammatical signs; cf. e.g. Popela 1991). In the polysynthetic type, the discretion of both the grammatical and lexical formants is zero (the grammatical functions concern complexes of formants, not the individual formant components). In the insulative type, discretion is absolute. In the agglutinative type it is also absolute in its component aspect, but in its linear aspect it is relativized by the affixal character of the grammatical formants. The flexible type is characterized by reduced discretion of the grammatical formants; this can be connected with Skalička's (1938-39) concept of fluctuation in the function of the individual phonemes that form the formant in question as its components: In flexible type press 130 | BOHUMIL VYKYPEL some parts of both the grammatical and the lexical formants exclusively grammatical resp. lexical functions (i.e. they are discrete or separable), while other parts simultaneously express the grammatical and lexical function. At first glance, the introflexive type could be defined as such that some parts of the lexical formants express exclusively the lexical function, while all components of the grammatical formants fulfill the lexical function in addition to the grammatical function, i.e. that the discretion of the grammatical formants is zero. Such a formulation could be justified by such cases as German Vater —Väter, Bruder —Brüder, Mutter —Mütter or English foot —feet, man —men. However, among the forms of expression called introflexive, one can also find those whose parts (components) may well be considered as discrete, i.e., which express only the grammatical meaning: cf. Old Babylonian aprus ‘I separated’, i.e. a-‘1. Person Singular’, -u- ‘preteritum’, p-r-s ‘separate’, or German du beginnst —du begannst, i.e. du & -st “2. Person Singular’, -i-‘present’, -a-‘preteritum’, beg-nn-‘begin’. At the same time, we find expressions in which there is a partial fusion as in the flexible type (in other words, in which the boundary between the lexical and the grammatical formant is difficult to determine): cf. Old Babylonian aparras ‘I separate’, where a-‘1. Person Singular’ and the preteritum is expressed by the two vowels a and the reduplication of the second root consonant. The difference of the introflexive type with respect to the isolative and agglutinative type and the flexible type, with which it shares the alternative discretion and the partial fusion, is therefore rather to be formulated in such a way that the expression of the grammatical meaning may be discrete or partially fused with the expression of the lexical meaning, but that it always proceeds within the limits determined by the expression of the lexical meaning (this can possibly also be considered as a common denominator or connecting member to the polysynthetic type in the “typological spiral”). Schematically, the penetration, the fusion of the grammatical and lexical formants can be represented as follows (x, y = the individual phonological components; 1 = lexical function, g = grammatical function): the polysynthetic Type: (and (xy) the insulative Type: xy' x®y® (or (xy) (xy)?) the agglutinative Type: x'yIxeye (or (xy)'(xy)?) der flexive Typ: x'ylexleye der introflexive Typ: x'ygylgy! 3 3 In passing, it should be noted that the role of the individual phonological components in the construction of the formants explains the correspondence that can be established between the dominance of a type in a language and the graphic system of the language in question (cf. Unrelenting 1997: 36f. ). TO PROSODY IN TYPOLOGY | 131 Similar conditions apply to word formation. The grammatical function corresponds to the relationship between the base word and the derived word, and the degree of fusion between the expression of this relationship and the expression of the base word is observed: the polysynthetic type prefers composition, the isolative the creation of entirely new words or the borrowing, the agglutinative the word formation by affixes, the flexive by conversion and the isolative by infixes or conversion. The prosodic properties can then be easily incorporated into the above scheme by simply considering them as one of the phonological components of the given formants and observing whether they express only lexematic functions or only grammatical functions, or both at the same time. Neustupny (1959) proceeds similarly: in his view, the accent represents a component of the formants as well as the phonemes; the agglutinative type differs from the flexible type in that in the agglutinative type the accent belongs to the morpheme, whereas in the flexible type it belongs to the word. This means that in the agglutinative type the accent represents a part of either the grammatical or the lexical formant, whereas in the flexible type it forms both formants. 4.1. English word formation by accent shift is considered as an introflexive It is difficult to evaluate this characteristic, because the accent shift occurs within the framework determined by the lexical formant (from what was said above in $4 it also follows that the pure expression through prosodeme is ex definitione introflexive). However, this introflexive feature is combined with features of other types (such as the isolative in the compound or I compound, the agglutinative in compounds or compounded); it is clear that this is a combination of typologically different formants, and not formants that have been partially merged. 4.1.1. An interesting combination of the introflexive with the polysynthetic Word formation by composition is found in Chinese: cf. e.g. xiäojie ‘lady’, i.e. xiūo ‘small, young’ + jiė ‘older sister’, whereby the second component becomes atonic. At first glance, a similar combination is found in the case of “verbs” in the function of prepositions: some verbal formants appear in the function of prepositions, which is to be evaluated as a characteristic of the polysynthetic type, in which according to Skalicka lexemes can also perform grammatical functions, and this non-verbal function can be accompanied by the loss of the tonal characteristics of the formant: cf. gängbi zai zhuözishang ‘The ballpoint pen is on the table’ vs. wo zai beijing zhu ‘I live in Beijing’, i.e. zai ‘to be (somewhere)’ vs. zai ‘in’. The expression of the grammatical functions by lexemes, however, differs fundamentally from the expression of the grammatical functions by the arrangement of the lexemes (i.e. from the basic property of the polysynthetic type) from a segmental point of view: whereas in the latter case the grammatical and lexical functions in question are completely fused in their expression, the expression of the grammatical and lexical functions in question is 132 | BOHUMIL VYKYPEL in the first case reduces the lexical function of the given formant de facto to zero, since not the lexical function, which is expressed by this formant in another position, by the function of another formant in the position concerned. The grammatical function and the lexical function determined by it are thus discrete in their expression. The identification of a formant in a grammatical function with a formant in a lexical ausgedrückte grammatische Funktion determiniert wird, sondern eine lexikalische function occurs only on the basis of the identity of their phonological components. Daniel Weiss (orally) has even proposed to distinguish two types of language here: the amorphous, in which the expression of grammatical functions occurs through the arrangement of lexemes, and the polysynthetic, in which the expression of grammatical functions occurs through the appearance of lexemes in the grammatical functions. This distinction would be justified by what has been said above, but I do not believe that it is absolutely necessary. First of all, a good common denominator can be found between these two modes of expression, namely the role of the (sentence) context, because the grammatical function of a lexical formant is only recognizable or given by its position, just as in the case of the expression of grammatical functions by the arrangement of the lexems, the grammatical function that determines a lexical function is only recognizable by the position of the corresponding lexical formant; the role of context can also be formulated with Skalicka’s slogan, the polysynthetic type builds its grammar on semantics. Furthermore, the above-mentioned identification of a formant in a grammatical function with a formant in a lexical function can also be interpreted as meaning that the lexical function in the position in question becomes potential (is not actual), but is nevertheless contained in the given formant, due to the identity of its phonological components; Thus, in this case, the absolute fusion of a grammatical and a lexical function would also exist in its expression, even if the grammatical function is not merged with the lexical function currently determined by it. Finally, a general argument is to be cited, Ockham's razor: entia non sunt multiplicanda sine necessitate. Also the types of language should not be multiplied if it is not necessary, i.e. if there is a reason to combine the two mentioned ways of expression. If, however, the phonological components are not completely identical — even if only on the suprasegmental level, as in the Chinese example cited with zai and zai — then these formants are indeed to be evaluated as isolative or agglutinative. 4.2.The Lithuanian endings and lexical stems mentioned above ($1) with their prosodic characteristics represent a feature of the flexible type: a part of the formant refers only to the grammatical function respectively. the lexical function (i.e. the phonematic composition of the formants), another part refers to both (i.e. the prosodic PROSODY IN TYPOLOGY | 133 Characteristic of the word, which is a result of the interplay between the prosodic power of the root and that of the ending). 5. Finally, however, a general remark must be made. It is evident that Skalicka’s typology can indeed be called “word typology”, where “word” is to be understood in a broader sense as a minimal connection of lexical and grammatical elements (for example, as the ‘syntagma’ of early Hjelmslev, not as a sign of a certain kind defined by means of the signifiant- ): the core of this typology is indeed “morphological”, if morphology is to be understood as the study of the minimal connections of lexical and grammatical elements. However, it is about what the “results” of this connection look like and not what is connected. An (intro)flexible grammatical or an (intro)flexible lexical sign (a formant) could also be described (defined) as a complex of certain phonemes and a complex of certain rules for changing each other part of the minimal connection: Thus the plural in German Brüder would have a formant, which would be formed by zero and an ability to change the lexical formant in a certain way. The Lithuanian formant -os, would be formed by the phonemes o and s and by its ability to influence the prosodic characteristics of the lexical formants, and vice versa, the Lithuanian formant galv would be formed by the phonemes g, a, /and v and by the ability to influence the prosodic characteristics of, for example, the grammatical formant -os. If things were understood in this way, however, everything would be discrete (in the other sense of the word and therefore also fictional), and Skalicka would be a Chomskyan generativist ante litteras been’. BERGER, T. 1990: On the typological classification of the Russian accent. The World of Slavs 35, 370-380. GECKELER, H. 2001: English and French: a typological comparison according to the criteria of V. Skalička. Linguistics and Languages. Festschrift for Clemens-Peter Herbermann on the 60th anniversary Birthday. Ed. by U. H. Waßner, Aachen: Shaker, 111-124. GIGER, M., VYKYPEL, B. 2001: The Typology of Lithuanian and Latvian (with a general outlook on the perspectives of the Prague Typology). in: VYKYPEL, B., GIGER, M., VYKYPEL, B. 2001: The Typology of Lithuanian and Latvian (with a general outlook on the perspectives of the Prague Typology). NEUSTUPNY, J. V. 1959: Accent in Japanese and Russian (A Typological Study). Archiv orientälni 27, 12245-86. 142. NEUSTUPNY, J. V. 1997: Typology in Prague: The legacy of Vladimir Skalicka. In: B. Palek (Hrsg.): Typology: Prototypes, Item Orderings and Universals (Proceedings of LP'96). Prague: Karolinum, 29-40. (Acta Universitatis Carolinae. Philologica 1996, 3-4) 4 However, the segmentation mentioned above (§4) also represents a segmentation of an actual and individual word form, not a segmentation of a connection between the category of the lexematic and that of the grammatical signs, because from the point of view of such a potential and categorical connection, each component of the grammatical formants in the flexible type also possesses the lexical function and vice versa, since lexemes and grammes here form an inseparable (synthetic) unit —the word (a group of lexical signs has only certain grammatical signs and vice versa, whereby the grammatical signs also express the lexical function and vice versa); thus the flexible type would not be distinguishable from the polysynthetic. 134 | The 134th POPELA, J. 1991: Prospects of V. Skali¢ka- ’s Linguistic Typology. In: B. Palek & P. Janota (eds.): Proceedings of LP'90. Prague: Karolinum, 237-243. (Acta Universitatis Carolinae. Philologica 1992, 3-4.) SAPIR, E. 1921: Language. New York: Harcourt, Brace and Company. SGALL, P. 1992: Valence and Typology. Journal of Phonetics, Linguistics and Communication Research 45, 200-206. SKALICKA, V. 1938-39: On the concept of morpheme. Journal of the American Medical Association 16-17, 4-12. SKALICKA, V. 1967: Phonological Typology. Phonetica Pragensia. Edited by M. Romportl & V. Skalička. Cambridge: Cambridge University Press, 73-78. (Acta Universitatis Carolinae 1967. Philologica, 6.) SKALICKA, V. 1979: Typological Studies. Edited by P. Hartmann. Braunschweig —Wiesbaden, Vieweg. (Schriften zur Linguistik, 11.) Bohumil Vykypel Gauta 2004 04 06 Ustav pro jazyk česky AV CR etymological separation Veveri 97 CZ-60200 Brno ([email protected])