Language variation in electronic messaging: Stylistic parameters that make a 'linguistic fingerprint' problematic.
Lead Research Organisation:
Bangor University
Department Name: Sch of Linguistics and English Language
Abstract
This project is an opportunity to undertake ground-breaking research in one of the newest and most exciting areas of linguistics, and the results will attract considerable, national and international attention; from both linguistics researchers as well as legal practitioners and the police. It involves a linguistic investigation into the types of variation that occur in electronic communications and will measure how the chosen forms vary according to a number of contextual factors. The combination of forensic and non-forensic corpus analyses will provide the necessary evidence to enhance the forensic science used in court, in particular, cases that require authorship attribution (i.e., the forensic linguist needs to determine if the suspect is or is not the author of a given legal text). These cases include: murder, (sexual) assault, malicious communications, terrorism and conspiracy (among others): in fact, any case where the authorship of key documents needs to be established. Authorship is established by isolating particular linguistic traits in those document(s) that can also be found in comparative documents where the authorship is known.
A number of researchers in this field argue for a writer's ideolect (or 'linguistic fingerprint') that will always make obvious the true author of a given text (cf., Coulthard 2003); thus enjoying the level of reliability associated with fingerprints. Despite this allure, however; it understates the extent to which a person's linguistic choices are the product of a concert of factors that need to be taken into account. Olsson (2009) has begun research into these factors and has so far shown that much more work is needed, to compare the same author across text types, and the same author in communication with different addresses (possibly contrasting gender and or status relationships) or with different talk objectives (e.g., favour vs. criticism).
Olsson's preliminary findings reveal that when more factors are taken into account, a more robust, reliable and individually identifying account of authorship can be established. This project will gather a substantial corpus of email and telephone SMS (short message service) text messages, in order to investigate which contextual features are most productive for forensic linguistics analysis. An improved set of strategies will improve police accountability; help secure convictions and offer a better protection to the innocent.
The data to be used in this investigation includes: (a) a new corpus of English language, SMS text data, to be collected from a cross-section of discourse contexts as provided by volunteer participants; and (b) a new corpus of email linguistic data. The College Ethics Research Committee approval will be sought before any research commences. In the case that Dr Olsson needs access to the data, he will be required to sign a confidentiality agreement prepared by Bangor University.
A 2nd dimension of this research is to include a Welsh and German medium pilots. The student will follow Welsh classes to develop Welsh language skills in the first year of this project, and the data collection will commence in the second year. This part of the project will be supervised in part by Ms Prys (Welsh) and Prof Tully (German) whose expertise will prove invaluable for these pilots. The pilots are to demonstrate the easiness (or difficulty) of applying the attribution strategies identified in the main project, to other language data, which will enhance the international impact of this project.
In summary, this project is a linguistic study designed to evidence the extent to which sociological and contextual factors must be taken into account to substantiate the plausibility of a speakers' 'linguistic fingerprint'. With this ultimate target, this research is timely, significant and internationally relevant. The results will c
A number of researchers in this field argue for a writer's ideolect (or 'linguistic fingerprint') that will always make obvious the true author of a given text (cf., Coulthard 2003); thus enjoying the level of reliability associated with fingerprints. Despite this allure, however; it understates the extent to which a person's linguistic choices are the product of a concert of factors that need to be taken into account. Olsson (2009) has begun research into these factors and has so far shown that much more work is needed, to compare the same author across text types, and the same author in communication with different addresses (possibly contrasting gender and or status relationships) or with different talk objectives (e.g., favour vs. criticism).
Olsson's preliminary findings reveal that when more factors are taken into account, a more robust, reliable and individually identifying account of authorship can be established. This project will gather a substantial corpus of email and telephone SMS (short message service) text messages, in order to investigate which contextual features are most productive for forensic linguistics analysis. An improved set of strategies will improve police accountability; help secure convictions and offer a better protection to the innocent.
The data to be used in this investigation includes: (a) a new corpus of English language, SMS text data, to be collected from a cross-section of discourse contexts as provided by volunteer participants; and (b) a new corpus of email linguistic data. The College Ethics Research Committee approval will be sought before any research commences. In the case that Dr Olsson needs access to the data, he will be required to sign a confidentiality agreement prepared by Bangor University.
A 2nd dimension of this research is to include a Welsh and German medium pilots. The student will follow Welsh classes to develop Welsh language skills in the first year of this project, and the data collection will commence in the second year. This part of the project will be supervised in part by Ms Prys (Welsh) and Prof Tully (German) whose expertise will prove invaluable for these pilots. The pilots are to demonstrate the easiness (or difficulty) of applying the attribution strategies identified in the main project, to other language data, which will enhance the international impact of this project.
In summary, this project is a linguistic study designed to evidence the extent to which sociological and contextual factors must be taken into account to substantiate the plausibility of a speakers' 'linguistic fingerprint'. With this ultimate target, this research is timely, significant and internationally relevant. The results will c
Planned Impact
The beneficiaries of this research project go way beyond academia: the target beneficiaries include most obviously the legal system, but also the general public at large. The impact will be wide ranging, in that a more robust 'linguistic fingerprint' which is the aim of this project, will help to ensure that no innocent person will be convicted of a crime where written evidence is taken into account, and vice versa, that guilty persons will be convicted. Because this project is collaborative with a professional forensic linguist, the insights gathered from this research would be quickly implemented in the field; and because of our publishing histories and growing reputations in this field of research, research outputs will quickly gain the attention of researchers from both the linguistic and forensic fields, but also among legal professionals and the police. As a result, the ultimate beneficiaries of this research will be the public at large.
To ensure all interested parties can stay abreast of findings from this project, a web page devoted to the project will be established; advertising publication outputs and presentations. In this way the outcomes from the research will be identified, communicated, transferred, exchanged, and exploited as appropriate. And to this end, the project developments will also be followed on the partner's professional website, which currently attracts hits from students and fellow academics and legal professionals all over the world. The conferences and publication venues to be targeted will be those that are frequented by linguists and legal professionals, such as the IAFL (International Association of Forensic Linguists) and their journal. Future development opportunities with the partner will also become viable, such as joint research, joint events and possible work placements as demand increases.
Furthermore, Bangor University is very committed and experienced in Impact/ knowledge transfer, and our research team member, Ms Prys, has in fact been personally involved with much of the University's success in this matter. Schools in the College of Arts, Education and Humanities have a history of collaborative working, and our School of Music was awarded the very first arts & humanities' KTP in the UK, which won a national award. This has been followed up with other similar collaborations in arts & humanities which have been award winning. Bangor has also recently won the first short KTP in the UK. Bangor University is very committed to these projects and the results so far show an improved environment for collaborative projects, and that progress with these projects is keenly monitored.
To ensure all interested parties can stay abreast of findings from this project, a web page devoted to the project will be established; advertising publication outputs and presentations. In this way the outcomes from the research will be identified, communicated, transferred, exchanged, and exploited as appropriate. And to this end, the project developments will also be followed on the partner's professional website, which currently attracts hits from students and fellow academics and legal professionals all over the world. The conferences and publication venues to be targeted will be those that are frequented by linguists and legal professionals, such as the IAFL (International Association of Forensic Linguists) and their journal. Future development opportunities with the partner will also become viable, such as joint research, joint events and possible work placements as demand increases.
Furthermore, Bangor University is very committed and experienced in Impact/ knowledge transfer, and our research team member, Ms Prys, has in fact been personally involved with much of the University's success in this matter. Schools in the College of Arts, Education and Humanities have a history of collaborative working, and our School of Music was awarded the very first arts & humanities' KTP in the UK, which won a national award. This has been followed up with other similar collaborations in arts & humanities which have been award winning. Bangor has also recently won the first short KTP in the UK. Bangor University is very committed to these projects and the results so far show an improved environment for collaborative projects, and that progress with these projects is keenly monitored.
Organisations
People |
ORCID iD |
| June Luchjenbroers (Principal Investigator) |