Linguistic/Dialectal Profiling of Dialect Speakers: The Method Presented on the Idiolect from BerCinovac, Eastern Serbia

The aim of the undertaken research is to develop a statistical method that allows to estimate frequency and regularity for the distribution of distinctive dialect features in the spontaneous speech of speakers of non-standardized idioms, and thus to discover intrasystem implications between different dialect features creating linguistic profiles of dialect speakers. At a later stage of research, the achievement of this aim would mean obtaining a quantitative tool for assessing intrasystem coherence and degree of dialect preservation for any Slavic or Balkan dialect. The object of study is Draginja MikiC's (b. 1906) idiolect. The narrative she produced was recorded in the village of BerCinovac, situated near Knjazevac in the ZajeCar county (Eastern Serbia). This narrative of approximately 5,300 word tokens was published by A. Sobolev in the third volume of his work called Sprachatlas Ostserbiens und Westbulgariens [Dialectological Atlas of Eastern Serbia and Western Bulgaria]. The analyzed data were extracted from the dialect text using the continuous sampling method, and subsequently subjected to etymological, formal and functional semantic analysis. The comparison procedure itself consisted in counting the number of uses of the “correct” (i.e. corresponding to the conventional dialectal reference point) options for dialectal difference implementation. As a result, the authors could determine as a percentage how close this idiom is to the ideal model of Timok dialect, how “dialectal” it is in general, and how much different it is from the other Serbian dialects and from the standard language. In order to obtain statistically relevant results based on a sufficient amount of text, the authors selected twelve most significant dialectal features, which allowed them to oppose all the dialects of the studied region of Eastern Serbia with the Neo-Stokavian dialects underlying the standard language. Dependence of the experiment results on the frequency of certain lexemes used in the informant's speech can be observed. A dialect Timok word form often comes into opposition with a non-dialect (standard), according to not just one, but to a number of parameters. Thus, the form maceja ‘stepmother' corresponds to macehu-Acc.Sg not only with /c/, but also with the phoneme /h/, which is not typical of the Timok dialects. Such an opposition is quite often implemented not in the same lexical material (cf. maceja ~ maceha, dsn ~ dan ‘day'), but in different groups of tokens (cf. vijuvali ‘(they)saw' ~ roj;ena ‘(she was) born'); and not in the same word form of the lexeme, but in different segments of its paradigm (the dative forms of pronouns in sg. mene, nemu against the background of analytical forms pl. na nas). The data reflect different relative frequency of correlative members pairs of different dialectal features and, therefore, a different degree of regularity of their implementation, regardless of the language level. The authors hope that applying various modern research methods to a sufficiently large amount of material, including linguogeographic, corpus and psycholinguistic methods, can provide them with the desired result (including in the cases of negative statements concerning features that do not exist in a dialect).

Keywords

Authors

References