Presentation
This project aims to study the speakers’ idiolect in its application to forensic linguistics (in particular to forensic phonetics, in order to identify speakers and establish linguistic profiles, on the one hand, and to authorship determination/attribution of written texts, on the other). A speaker’s idiolect can be defined as the set of options that he/she takes from the linguistic repertoire (phonological, morphosyntactic, pragmatics forms) available to him/her as a speaker/writer of a specific language (Nolan 1994:331). Thus, a speaker’s idiolect is individual and unique. Idiolectometry is the emerging discipline which studies the idiolect. So, far this discipline has measured the linguistic distance between speakers and has established the borderline between different idiolects; a borderline which by definition keeps a person separated from the rest of speakers. By contrary, what this project will explore and develop is the possibility of measuring the linguistic differences existing between idiolects and each individual’s idiolectal distance, so that an Index of Idiolectal Similitude (Índice de Similitud Idiolectal (ISI)), which will compare several linguistic samples and calculate the linguistic distance between them, can be obtained. More specifically, it is a question of being able to establish what kind of idiolectal similitude one needs to observe before one can say that two linguistic samples (spoken or written) have been produced by the same person. The application of the study of the idiolect to forensic linguistics is fundamental since this application will allow linguists acting as expert witnesses in court to unmistakably identify speakers or writers by comparing a disputed recording or written text and a set of non-disputed spoken or written texts from the set of options chosen by each individual speaker or writer. A protocol will be devised which will set up the above mentioned ISI from which, once different linguistic parameters (phonological, morphosyntactic, discourse-pragmatic) are analysed, it will be possible to decide whether or not two recordings or two written texts have been produced by the same person. In this exploratory project we will design the technique to be applied to any language, the overall protocol and the statistical methods to be used, and finally the phonological module for Spanish and Catalan willl be evaluated.
 |
 |
 |
 |