Thinking Folklore: Approaching Dialect from Folkoristic, Ethnological, and Computational Perspectives

The project investigates the nature of a dialect from folkloristic and ethnological, semantic, and technological perspective. Specifically, it examines the transitioning between three types of language codes present in folklore – a dialect, a literary language, and a supra-dialect as an intermediate form of the two. It involves basic concepts of constructing local identities that are clearly associated with the use of dialect expressions in songs, whereas the transitioning relates to the structure of linguistic expression that is moving away from a pure dialect towards literary code. This duality is pronounced when the bearer from a particular dialect group transfers the dialect used in speech to the non-dialect (supra-dialect) when singing. In both, we see the expression of collective identity and the preservation of collective memory (Halbwachs 2001).

The folkloristic and ethnological perspective tackles the dialect as a textological problem associated with the texture (melody) of the text and influenced by the bearer's interpretation, as well as local and wider social and cultural context. It further addresses the choice of a particular language code with respect to the social situation in which the bearer enters. The collective consciousness of the local stems from its identification with the dialect (conversations, memories, reflections), whereas the collective consciousness of the national stems from its identification with the supra-dialect (singing).

The semantic challenge stems from transitioning between the language codes. As each code is subjected to a specific semantics, the transitioning always generates a semantic problem. The coupling of the code to the conceptual structure of the bearer and the wider semantic structure, which provide for the collective use of the code in its (local) environment, cannot be completely preserved in any kind of transitioning where either the bearer, the meaning, or the context (situational, social, and cultural) are being altered.

The technological perspective will examine novel audio analysis and music information retrieval methods to automate the search for words in their different dialectal forms in field recordings and simultaneously quantify the phonetic differences between their pronunciation. It will be necessary to develop new approaches that address the specifics of the analysed materials, in particular the use of dialect, which can significantly deviate from written transcriptions. The purpose of the technological research is twofold: alignment can enable word searching in large collections of sound recordings, and thus substantiate the analysis conducted in the course of the project. At the same time, alignment forms the basis for development of algorithms for assessment of phonetic parameters of pronunciation of individual phonemes (e.g. f0 movement, position of formats, length of vocals, etc.), which will enable quantification of the differences between dialect and supra-dialect in speech and singing.

The proposed researches are original as they are the first in Slovenia to examine the link between dialect and supra-dialect and the contents of the folklore text or event. In this way, we also highlight the importance of dialects for the identity of individuals and communities, local and national. At the same time the project recognizes the spontaneous and conscious use of dialect or supra-dialect, depending on the social situation. In technological terms, the proposed research is original since it will develop new approaches to alignment of texts and audio recordings, which will allow for a broader and more detailed analysis of dialect and supra-dialect. Connecting textual, contextual and semantic aspects with the proposed computational approaches to analyze large amounts recordings can form the basis for the development of context oriented and text oriented folkloristics.