Understanding threats: Language and genre
Recent years have seen an increase in threatening messages, not least through social media. Threats are meant to instil fear in their recipients; an effect produced largely by the use of language. While practitioners offer 'threat assessments' to determine the seriousness and intent of threats, basic research on threatening language and the genre of threats remains minimal.
The project 'Understanding threats' proposes an innovative combination of linguistics and genre studies to offer in-depth analyses of a text type that causes severe distress but lacks scientific research.
Given the pervasive nature of legislation upon any aspect of society, citizens cross paths with the legal system in myriad ways, leaving linguistic traces opportune for academic scrutiny. The combined forces of forensic linguistics and rhetorical genre studies promise new insights into threats and other illicit genres formed in the shadows of society.
Combining forensic linguistics (FL) and rhetorical genre studies (RGS) allows for integrative analyses of the language of threatening messages and their rhetorical structure when related to communicative purpose. RGS is eminently suited as a partner theory for FL, since the former explores the social actions performed in instantiations of genres, while the latter details the linguistic features used to attain communicative goals. Their research agendas are thus complementary and their empirical data overlapping.
Comparatively little is known about grammatical and stylistic traits of threatening language. Extending results for English data to Danish, a corpus of +250 authentic threatening letters will be analysed qualitatively and quantatively to discern the functions and distributional patterns of linguistic features related to expressions of author attitude ('stance').
Studying threatening messages as a genre is equally innovative. In contrast to ordinary institutionalised genres, threats constitute what we term an illicit genre, being socially and sometimes legally proscribed and belonging outside a coherent discourse community. The structure of threatening messages is expected to be more fluid than other genres, but nonetheless amenable to analysis by reference to sub-functions. Due to the varied linguistic and generic traits of threats, their 'uptake' in surrounding legal genres calls for further analysis (e.g. how does the indictment/verdict of a specific threat reflect definitions in legislation?).
This project yields synergistic effects by facilitating a continual exchange of theoretical and empirical advances between approaches, including the use of a shared, core data set in the form of the main corpus of threatening messages. The corpus builds upon the appr. 250 authentic threatening letters published by Engelhardt and Lund in 2008 and 2009 (permission obtained from the publisher). The project will increase the data set through a variety of sources, including the Police Museum and public figures who have been victims of threat crimes.
To aid the linguistic analyses, criteria for automatic coding of the data will be developed in dialogue with leading international experts and colleagues from Centre for Language Technology. The texts will be digitalised, annotated and prepared for corpus linguistic extraction in collaboration with Danish software developers WordMaps, a company specialised in supplying first-class database and annotation software for Digital Humanities. Statistical results on (correlations between) linguistic and situational factors will feed into in-depth qualitative analyses that again will inform the genre-based analyses. For these, a subset of the data will be analysed structurally and functionally in relation to the genre network in which they are societally and legally embedded. Additional material for this study includes relevant legislation, indictments and judicial verdicts from select cases.
The corpora of threatening and non-threatening messages are the building blocks for the rest of the project and data collection will begin even prior to project launch (1 February 2018) with expected completion by September 2018.
A workshop in March 2018 with leading international scholars will form the basis for linguistic and situational annotation criteria for the set of Danish threatening messages. A postdoc will be recruited for 2019 to control and correct automatic annotations.
Corpus linguistic analyses will establish recurrent as well as deviant patterns in the data and correlate them with available information on situational factors.
The project will include the following international colloquia:
- A workshop in 2018 invited leading forensic linguistics experts in corpus studies of threatening language to discuss best-practice linguistic annotation of threatening texts.
- A symposium in 2019 will gather scholars in rhetorical genre studies and forensic linguistics to discuss the theoretical implications integrative analyses of illicit genres and their uptake present.
- A conference in 2020 will present the most recent results from our group and its international partners on linguistic and generic traits of threatening messages and related genres.
The advisory board members will consult continually on the research group’s progress, co-author articles and teach master classes. Members:
Researching the linguistic traces of criminals
Read interview with Tanya Karoli Christensen (in Danish).