Use this URL to cite or link to this record in EThOS:
Title: Towards the modelling of forensic authorship methods in the light of aspects of idiolect, text, author and genre relations
Author: Olsson, John Gabriel
ISNI:       0000 0004 2682 2205
Awarding Body: University of Glamorgan
Current Institution: University of South Wales
Date of Award: 2009
Availability of Full Text:
Access from EThOS:
Currently, some prominent forensic linguists rely on notions of the idiolect to undertake forensic authorship comparisons. The first project questions the validity of this doctrine, firstly by examining how authorship and its subsidiary activity of attribution, based on concepts such as style and individual distinctiveness, came into existence. The writings and casework of several forensic linguists are quoted from: it is seen that the notion of distinctiveness has been embraced by these practitioners to the neglect of a balanced observation of variation. A mobile phone corpus was built in order to evaluate the degree of variation in this form of written language. It was found, as expected, that variation is the norm and not the exception - contrary to the claims of some forensic linguists. The second project studies the nature of variability in genre and non genre text, examining factors which can contribute to author or textual variation, including aspects of register and context. An experiment into genre variation is carried out in order to discover the sources and causes of variation. The experiment shows that variation is not author based, but is genre dependent, related mainly to questions of time and tropic factors. The final project outlines an approach to the authorship of mobile phone texts and provides a sample method of authorship comparison based on the notion of range of variation, a concept modified from sociolinguistics. Because all forensic work is carried out within the context of legal requirements and evidence rules these are placed at the heart of the approach and methodology. The research programme reported here demonstrates the importance of a principled, variable-based, selection procedure for test features and the need for an understanding of variation, and it highlights the inadequacy of the notion of the idiolect (as this concept is interpreted by some forensic linguists) for forensic authorship comparison. It underlines the importance of the sociolinguistic observation that style is more usually a property of context and text than it is of the author
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available