Crime through a corpus : the linguistic construction of offenders, victims and crimes in the German and UK press

Abstract

In this thesis I analyse and compare the linguistic construction of offenders, victims and crimes in the British and German press. I have collected a corpus of British and German newspaper articles reporting on crime and criminal trials and carried out a corpus linguistic analysis of this data using the software package Wordsmith Tools (Scott, 2004). Reports on crime do not construct a neutral representation of offenders. By employing the tools offered by Critical Stylistics (Jeffries, 2010a) and combining them with Corpus Linguistics I identify the linguistic features used to pre convict offenders and to invoke a feeling of insecurity and fear in the public. The negative associations assigned to crime are transferred to the offenders and thus construct them as being evil and label them as deviant (Becker, 1966: 31). The linguistic construction of the victim ultimately impacts on the construction of offenders because the two are placed at opposite ends of a morality scale. It is through language that such ideologically motivated representations of offenders are constructed and reinforced. The image of the evil-perpetrating monster constructed in the media as part of societal discourse on crime is based on ideologies which my research aims to reveal. I argue that the underlying ideologies for the construction of offenders, victims and crimes in the British and German press are comparable and that the linguistic triggers for these in the texts are similar. I found no distinction between the persona of the offender and his or her crime because offenders only gain a celebrity-like status following the crime they have committed. This fascination with crime in the media has roots in the ‘backstage nature of crime’ (Surette, 2009: 240) which satisfies the voyeuristic desire of the audience.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 27 citations exist upstream.