Plagiarism : Evidence-based detection and analysis in forensic contexts

The question of plagiarism in forensic contexts has attracted much attention worldwide in recent years, especially as a result of high-profile cases involving politicians and famous journalists. The perception that plagiarism is a widespread phenomenon led to the development of plagiarism detection software, which operates by establishing a comparison between a suspect text and other previously produced texts. However, at most, so-called plagiarism detection software is able to detect textual overlap (i.e. similar or identical strings of text), which is not necessarily plagiarism. To evade computational systems, plagiarists have devised sophisticated text manipulation techniques that are easily missed by automatic systems. This chapter discusses how, although computational systems can assist the forensic linguist in the detection procedure, an appropriate assessment of plagiarism requires a thorough forensic linguistic analysis. Examples from real academic and forensic cases are presented to describe and explain the strategies most commonly used by plagiarists, and subsequently show how such strategies can be identified and justified linguistically, and how the linguistic analysis can prove or disprove the hypothesis that the lifting is inadvertent. The chapter concludes with a discussion of contract cheating and how the problem of essay purchase can be investigated from a forensic linguistic perspective.