Iseorganiseeruvad keele-elemendid eesti keele emotsioonisõnavara näitel
The self-organizing elements of language; the case study of the Estonian emotion terms
Author(s): Ene Vainik, Toomas KirtSubject(s): Language and Literature Studies
Published by: Eesti Rakenduslingvistika Ühing (ERÜ)
Keywords: leksikoloogia; semantika; psühholoogia; emotsioonisõnavara; iseorganiseerumine; semantilised tunnused ; self-organizing elements of language; the case study ; Estonian emotion terms ; estonian linguis
Summary/Abstract: In the present article it was claimed that one way to study the inherent structure of language and its elements is to use a self-organizing approach. In the reported study the method of self-organizing maps (SOM) was applied on the semantic data of Estonian emotion terms. One hundred informants had filled in the questionnaires about the presence and measure of seven specific semantic features in different emotion concepts. The purpose of the task was to find out about the reasons of the main controversy the matter of the proposed universal structure of the emotion lexicons has been handled in the psychological literature so far. As a result the self-organizing algorithm of the program generated a topological map of Estonian emotion terms that demonstrated a clear division of the concepts into two bilaterally symmetric clusters: positive and negative emotion terms. This kind of structure is in accord with the findings of the universal structure of the emotion lexicons in the psychological literature. The discussion is about whether the positivity and negativity should be treated as unipolar (and not mutually exclusive) dimensions or as the opposite ends of one bipolar dimension acting together with a crossing dimension of general activation. The role of the single features co acting in the generation of the self-organizing map were further investigated and discussed. It was found that four of the seven features were distinctive by nature causing the set of selected emotion terms to cluster clearly into two subsets (unpleasant vs. pleasant, increases vs. decreases action readiness, follows vs. precedes the event, felt in the mind vs. body). Three of the selected features were found to be indistinctive (strong vs. weak emotion, long vs. short in duration, depends mostly on oneself vs. others). The scale of hedonistic evaluations (pleasant vs. unpleasant) was suspected to act as a higher order abstraction among the specific evaluations given on the other scales, as it caused a far stronger division of the concepts, even extending the self-organizing map in one direction. It was concluded from the analyses that it might be reasonable to interpret the overall negativity of emotion terms as the unpleasantness of the referred states and the general positivity as the capacity of the referred state to increase one’s action readiness. In that way the controversy of the psychological literature will be resolved. It was further hypothesized in the end of the article that the hedonistic dualism of the emotion concepts could be understood as a projection of the evolutionary primary two-valenced evaluations on to the multiplicity of contexts and situations presented in the specific semantic content of every single emotion term. The contrary is also possible: the additional semantic features and distinctions in the situations and contexts have been added in the history of human cognition to interpret and to learn better form the two-valenced primitive feedba
Journal: Eesti Rakenduslingvistika Ühingu aastaraamat
- Issue Year: 2005
- Issue No: 1
- Page Range: 171-186
- Page Count: 16
- Language: Estonian