By Lisa Kaati
One of the most challenging threats towards societal security is attacks from so called violent ‘lone offenders’. These individuals are said to act alone or with minimal help from others, without any economic incitements or direct orders from violent extremist organisations. Between late 2018 and early 2020, several violent attacks in New Zealand, the US, Germany, and Norway were committed by individuals with little or no connection to extremist organisations or terrorist groups.
However, what they all had in common was that they had been active in digital environments. Fundamentally, before acting ‘alone’, these individuals were influenced either by someone, or by something, and increasingly in the online digital space.
Digital environments play an important role in inspiring, motivating, and encouraging individuals to commit violent attacks. They provide opportunities to communicate with like-minded individuals all over the world, whilst the prevalence of toxic language and conspiracy theories creates conceptions around groups of people, often stigmatising them as enemies. These conceptions subsequently serve to justify contempt, discrimination, and violence. In some digital environments glorification of violence is not uncommon and previous offenders are portrayed as inspirational role models and referred to as ‘heroes’ or ‘saints’. Indeed, some of the recent violent attacks have been preceded by manifestos published online where the perpetrators had outlined their ideology, motivation, and tactical choices.
Research shows that language generally has communicative functions beyond what is said explicitly. The way of using language reflects the author’s attitudes, personality, and emotional state. What words we use and how we express ourselves reflect both who we are and what social relationships we have. Language is the most common way for people to express inner thoughts and feelings so that others can understand. By analysing the linguistic characteristics of communication, we can understand factors that the author does not necessarily explicitly express in their text.
Thus, by analysing the perpetrators’ own words, we can learn more about them, their personality, emotional state, and how they see themselves and others. This kind of information is useful when developing methods for threat assessment in the digital space. Threat assessment takes place before an attack has been committed and focuses on detecting and preventing attacks. Threat assessment is generally carried out by law enforcement, intelligence analysts, and other security professionals.
Whilst most research on threat assessment of individuals has focused on offline settings where there is accessible information about an individual or where the individual is present and can answer questions, there is very little research that has focused on threat assessment in digital environments. Indeed, several of the common psychological factors that are considered in threat assessment in an offline setting can also be assessed in the digital space.
One of the commonly used threat assessment protocols is The Terrorist Radicalization Assessment Protocol-18™ (TRAP-18). TRAP-18 consists of two sets of variables: eight warning behaviours to identify patterns of proximal risk for intended or targeted violence and ten characteristics of lone-actor terrorists. While TRAP-18 is designed for an offline setting, some of the proximal warning behaviours and characteristics can also be identified in digital communication. The warning behaviours that are most likely to be identified in digital communication are fixation, identification, leakage, and directly communicated threats. Among the characteristics, personal grievance and dependence on a virtual community are examples of two characteristics that can be observed in digital communication.
The large amount of data that is available makes it difficult for human analysts to manually assess the threat from individuals that are active in the digital space. To assist analysts in their work there is a need for automatic or semi-automatic technologies for threat assessment that can be used as one tool in a threat assessment toolbox. While there are tools that can be used to assess the threat of an individual base on their written communication, it is still a complex problem that requires more research before it can be fully operationalised. Regardless, since digital environments have played an important role in several deadly attacks, digital threat assessment is a promising tool that can assist in preventing targeted violence.
To be able to prevent future deadly attacks we need to be more active in monitoring the digital space and identify individuals with potential violent behaviour. This means that we need apply threat assessment methods that already exist in offline communication, particularly focusing on written communication, which requires knowledge of language and ways of communicating in the fast changing landscape of social media platforms and discussion forums.
Lisa Kaati is a Deputy Research Director at the Swedish Defence Research Agency (FOI).
This article is republished with permission from the September 2021 edition of Spotlight magazine, ‘Emerging Threats‘, originally published with the title ‘Lone Actors in Digital Environments’. Spotlight is a publication from the European Commission’s Radicalisation Awareness Network for RAN’s network of practitioners. Image credit: Unsplash.