Assuming identities online: How linguistics is helping the policing of online grooming and the distribution of abusive images

In this chapter we seek to elucidate the potential of linguistic analysis in the undercover pursuit of criminals online. We examine the relationship between language and online identity performance and address the question of what linguistic analysis is necessary and sufficient to describe an online linguistic persona to the extent it could be successfully assumed by another individual. We compare the performance of trainee undercover officers (UCOs) at online impersonation of a specific individual before receiving any linguistic training to their performance after such input. We also report on a series of experiments in which participants engaged with each other over Instant Messaging (IM), before an impersonation situation was engineered. Information was collected about what level of accuracy and confidence individuals can detect the substitution of one interlocutor with another, what linguistic criteria give rise to such suspicions, and how individuals prepare for impersonation tasks. We also draw on other data sets including from police training and operational settings where identity disguise is employed online in the context of investigations into child grooming and the distribution of abusive images.