Degree Project in Criminology Malmö University
Master 30 Credit Two-year master Faculty of Health and Society
Masters in Criminology 205 06 Malmö
June 2021
THE PREVENTION OF ONLINE
EXTREMISM
A QUALITATIVE STUDY ON PROFESSIONALS
APPROACH WHEN PREVENTING ONLINE
EXTREMISM IN A DANISH CONTEXT
THE PREVENTION OF ONLINE
EXTREMISM
A QUALITATIVE STUDY ON PROFESSIONAL’S
APPROACH WHEN PREVENTING ONLINE
EXTREMISM IN A DANISH CONTEXT
NIKOLAJ MØLLER THOMSEN
Thomsen Møller, N. Prevention of online extremism. A qualitative study on professional’s approach when preventing online extremism in a Danish context. Degree project in criminology 30 Credits. Malmö University: Faculty of Health and Society, Department of Criminology, 2021.
Online crime, in general, is increasing these years, crimes committed through technology that is almost impossible for the authorities to solve and prevent. The average Danish teenager spends six hours a day on online forums, and studies on how content affects the user show that the users do get affected by the content. How do practitioners prevent teenage users from being embedded in extremist online communities when there are no national guidelines or research to support an evidence-based approach to prevention? Practitioners give presentations on this topic to teenagers and other professionals and seek to create stabile strong social bonds among peers and talk openly about pitfalls of the internet with the
teenagers.
This thesis applies Social Learning Theory and Social Bond theory in the
understanding of why some users get radicalized through online forums. Applying these theories is relevant because mirroring propaganda, creating norm deviating social bonds and a “them and us” detachment from the surrounding norm society can be explained by these theoretical perspectives – and can all be precursors towards extremist behavior.
The practitioners believe that they are navigating in a field where the need for knowledge is huge, and the methodological approach highly relies independently on the practitioner themselves. The main theoretical discussion is based on the founding hypothesis that being exposed to extremist content either leads to nothing or an increase in deviant behavior. The theoretical perspective in this thesis and previous research suggests that there is a causal link between traditional real-life risk factors and online crime, but also attachment to deviant online
communities can lead to extremism – just as it is in a real-life setting.
Keywords: Denmark, extremism, methodology, online, propaganda, qualitative, radicalization.
Table of content
Table of content ... 3
1: Introduction ... 4
2: Aim and objectives ... 5
3: Theory ... 5
3.1 Social Learning Theory ... 6
3.2 Social Bonds ... 7
3.3 Propaganda ... 8
3.3.1 But why is online propaganda so different? ... 8
3.4 Current Knowledge ... 9
3.4.1 Reinforcement of criminal behavior ... 9
3.4.2 The role of the parents ...10
4: Methods and ethics ...11
4.1 Methodology ...11 4.2 Bias ...11 4.3 Validation ...12 4.4 Ethics ...12 5: Collected data ...13 5.1: The interviewees ...13
5.2: Current prevention methods ...13
6: Limitations ...18
6.1: Theoretical discussion ...19
6.2: Practical discussion ...20
7: Conclusion ...20
1: Introduction
Humans are social creatures and will always seek communities where they have a sense of belonging and a sense of being accepted by the majority of the
community. This thesis is about how professionals prevent the Danish youth from being radicalized through online media.
In general crime rates are lowering across western countries. On the other hand, reported online crime in mostly all categories has increased from 2014 towards 2020 (Schulze-Larsen, 2020). A fair question to ask would be if the crime, in general, is lowering, or is it because of dark numbers - we just do not know they occur? Until 2014 Denmark did not have a department for investigating online crime, and the department was established to put a focus on this new world, that currently reminds more of the early 1900’s wild west than a modern society run by rule of law.
Radicalization has for many years been subject for research and is in a sense a field that is well described. But over the past many years online media has continually increased its effect on radicalization. Where in the early beginning extreme online content was something, you would have to look for, or something you would have to be invited to be a part of, it is now easy and far more available than before.
The Danish Crime Prevention Council (Det Kriminalpræventive Råd) conducted a study on Danish students and their online behavior (N=49,000) where it showed a statistically significant correlation between conducting online and offline crime (Demant et al., 2018)
In 2016 Denmark got the European record for highest online media usage. Danish teenagers spent on average 6 hours and 2 minutes every day on online media (Christensen, 2016). If an average teenager sleeps 9 hours a day, roughly 40 % of their awake hours are spent unsupervised, and leave them potentially exposed to unregulated content.
This project is based on a Danish perspective, and how the Danish municipalities have developed their prevention strategies to be preventing radicalization in physical environments, and how that potentially harmonizes with the criminological knowledge on online radicalization.
Most practitioners are trained to look for signs of extremism, such as changed clothing, appearance, or sudden change of political standpoints – and they are not trained to understand the dynamics on online social media and potential warning signs.
It seems like crime and radicalistic environments are developing on the internet. It can be through social media, blog pages, or as what this project is about,
subgroups that by demoralized means demand extremist behavior, norm deviant viewpoints, and accept violence either physically or online as means to reach the goals of the group.
By interviewing practitioners in the field, this project sheds light on the
2: Aim and objectives
There are two aims in this thesis, the first is to investigate how online forums can affect the development of extremism in an online and physical setting. The second aim is to look into how professionals prevent this from happening by taking a look at the knowledge that is available on the field.
By interviewing members of the prevention departments countrywide this thesis asks them if they believe that they are capable to solve this problem. This thesis also investigates the connection between theory and practice – and how does practice correspond to previous research?
The research questions are, therefore:
1. Are practitioners equipped with the right knowledge and tools to prevent young people from being online radicalized?
2. Can criminological theories be used into understanding the issues of the online radicalization process, and
3. What does previous research show regarding the extent of online radicalism?
3: Theory
By the virtue of our lives are constantly digitalized, it is fair to assume the processes behind radicalization are also adapting to this new technology.
Traditionally a radicalization process would have to be something that took place on a street corner or peer to peer. The comprehensive use of social media changes the entire way we today understand and describe communication and social interaction. This new way of perceiving social interactions calls for new ways of preventing breach of norms and the formation of a subculture.
At this point, many professionals working in local municipality prevention departments still struggle with the perseverance of the two-basic hypothesis on online radicalization.
The debate is about whether being exposed to extremist content online leads to more or less political violence and terror. The first hypothesis is based on a social learning theory perspective, where it is expected that the exposure, and taking a part of an online community that glorifies violence and terror as legitimate means, can lead to an increase in political violence and terror (Gemmerli, 2011). On the other hand, the catharsis hypothesis suggests a decrease in political violence and terror, because the individual can vent for already existing tendencies through online exposure (Gemmerli, 2011).
The problem with these two hypotheses is the scientific foundation and previous research which has a hard time proving any causalities because the complexity of the individual cases are high and there is limited knowledge on the subjects included in studies on the background of radicalized individuals (Gemmerli, 2011).
3.1 Social Learning Theory
The theoretical standpoint on the social learning approach is in a way
indeterministic and explains individuals’ pathways to radicalization explained by one's effect on each other. This theory explains the reward of following group norm behavior and deterrence if individuals break it (Akers, 1990). Akers describes conforming behavior as something which is formed over time (and a dynamic that is crucial to the crowd, to secure the norms are being followed) (Akers, 1990). In the online world that dynamic is also seen as a “groomer”1 that
influences an individual into conforming to group norms either by exposing the individual with propaganda or by social control.
The two main theories that describe crime at a micro level are the Social Control and Social Learning theories2, social learning theory has its focus on
communication as the precursor of crime in general, and it is expected that the change of morality and norms are related to online communication and the creation of online relations (Pritchett & Moeller, 2021). But the Social Control Theory has the more deterministic approach that will always be motivated
offenders in the society, but refrain due to formal and informal control (Holt et al., 2018). Social Learning theories are arguing that criminality as a whole is
something acquired through action and informed in the individuals’ social relations (Akers, 2009)
In sum, the Social Control theory explains the development of delinquency as something that is acquired through social relations to other delinquent peers – on the other hand, the social learning theory is relevant for understanding how individuals mirror delinquent behavior, and detachment to the surrounding norm society. Using social media is mostly about interacting superficially with other users and does often not lead to deep social bonds. It seems most relevant to use the social learning theory to describe the effect of being exposed to extremist content and propaganda (Holt et al., 2018).
For a person to be radicalized someone or something needs to take advantage of the cognitive opening3 at the individual. The cognitive opening could be loss of
employment, drastic change in existing family structure, or a personal loss – in sum something that put the individual on the pathway towards disconnecting the surrounding norm society (Trip et al., 2019). A cognitive opening from a social learning perspective is the beginning of a normative conflict that challenges the normative consensus (Trip et al., 2019). The normative consensus is a shared agreement among the majority of the group which is where aspects such as morality are based, constructed, and maintain (Trip et al., 2019). The normative conflict describes the process of the commotion by an individual that is going through a change in the set of norms – such as being radicalized (Trip et al., 2019).
1 A Groomer is a person that uses positive feedback to affect a victim over a long period
of time to ensure the victim adapt to deviant norms.
2 Social Control by Hirschi and Sampson & Laub.
3 Cognitive opening describes that condition where an individual gets receptive towards
Case 1 – from schoolgirl to bombmaker
16-year-old Natascha Colding-Olsen was in 2016 charged and later convicted for planning a terror attack on two schools where one of them is a Jewish school.
In the beginning, she was just an introverted teenager like everybody else, but after a period of school bullying, she ended up being radicalized in just a few months.
In her example, the bullying at school created a cognitive opening where she over time got receptive towards new groups. The cognitive opening was not something that was caused in a short period, more likely was it strung over several years. Colding-Olsen had a childhood influenced by severe various social problems, and besides school bullying, she had problems with her relationship with her parents. She also had a long period where she had a self-harming cutting4 behavior for at least six months.
During that period, she was institutionalized but ended up going back to live with her parents. During that tumultuous period, her friends described her as an introverted girl, with self-esteem issues, and they did not suspect any
underlying radicalistic tendencies.
Before she got radicalized she was a very active user on various social media, and it was on a social media platform she came across Islamic State propaganda and other material that glorifies Salafism and extremism. During her transaction from being a member of the surrounding norm society towards being a part of a subculture, her normative conflict was supported by a groomer, who supported her and gave her confidence as a member of this new subculture.
The transaction is visible on her social media platforms because in the beginning she posted content like any other teenage girl and she ended up posting content that sympathized with Islamic State.
In sum, the feeling of being ruled out of the norm society, the inclusion in a subculture, exposure to extremist propaganda, and the sensation of belonging leads her to plan a deadly terror attack. At the time she was caught she already bought the ingredients needed to make a bomb (Pigen, 2017)
The girl in the case example had social problems, but nobody suspected her to be on the pathway towards being radicalized. In hindsight, she did show signs, such as an increased interest in the Islamic world, political interest in the situation with Syria and Islamic State – something she never expressed being interested in before. Via the online media, she could interact, read, and get influenced by the online content without nobody knew.
The pathway towards extremism could in this case be explained by Social Learning Theory.
3.2 Social Bonds
Most of the preventative work in local municipalities is based on the idea of creating strong social bonds because it reduces the inclination to commit a crime. But normative functioning social bonds can be in various settings, and not only limited to real-life relations among peers. In reality, we need to look into the
impact of Social Bonds when they are based in an online context. Is it possible to obtain and maintain strong social bonds strictly limited to online media? It is fair to assume that the quality of online social bonds can be just as meaningful and constructive as in real life – at least this project has not been able to find any documentation otherwise (Holt et al., 2018).
The established society in Denmark is built upon a conception that social bonds are created through schools and maintained through free-time activities such as football, handball youth club, or other public facilitated activities. As previously stated the amount spent on online media has increased dramatically in the past 20 years, and we need more research on how positive online social bonds can refrain from committing a real-life crime.
3.3 Propaganda
Propaganda is not something new, but there is a difference in its effectiveness and availability when it is online. There are overall three target groups when we are talking propaganda. (1) Those who already believe and need to be restrained, (2) those who are hesitant and need to be convinced, and finally (3) the opponents that need to be discredited (Nissen, 2018). Especially the second target group contains those with a cognitive opening, that is especially receptive towards the propaganda –those in the third group who needs to be discredited – can be the precursor for the later “us and them” ideology.
3.3.1 But why is online propaganda so different?
The traditional view on propaganda is something that is top-down controlled, for instance when Hitler created door-to-door distributed propaganda. But on the internet, the propaganda can be orchestrated centrally, but the target groups play a role in distributing the propaganda (Nissen, 2018). Modern online propaganda works best in the shades, based on the edge of facts. The means used are often fractions of facts, that the users can relate to, or recognize and then twisted in a way that benefits the cause of the propaganda (Nissen, 2018)
The technology behind it is very intelligent and is based on modern algorithms that boost the exposure to the users. By viewing just a few videos with extremist content, algorithms and botnets will show you more similar content (Nissen, 2018). This dynamic is described as an echo chamber because the user gets exposed to more of the same content – and not any contradicting content that could create the nuances of the given topic (Nissen, 2018). The use of botnets can also give the illusion of strong support to the case, because botnets are
automatically generated profiles, that create messages and comments, and by that gives the illusion of a big community that strictly speaking could consist of one single person. It can be super hard to prevent this, because most of it is automated and how the internet is built. But being educated in spotting that dynamic and improve critical thinking could be highly relevant (Nissen, 2018).
Online propaganda is highly relevant when talking about Social Learning Theory because propaganda affects users into accepting a new norm, which can lead to the users imitating criminal behavior such as violence.
3.4 Current Knowledge
The impact on being exposed to online content was subject to research in a large-scale Facebook experiment (N=689,003) in 2014 (Kramer et al., 2014). The study was conducted among Facebook users, the idea was to understand what
emotionally consequences it has for individuals when they are exposed to
manipulated news and content (Kramer et al., 2014) The results of the study was significant, and it showed emotional states can be transferred between users – making users experience similar emotions without their awareness (Kramer et al., 2014). In that study, they call it “emotional contagion” when emotions spread among a group of users, and it can happen without a direct interaction – for instance when users share the content (Kramer et al., 2014). In the study they manipulated the content, so randomly users would either be exposed to overly positive or overly negative content. They later discovered that the user's responses were correlated with the type of content they were exposed to – so those
overexposed to negative content showed signs of acting more negatively and vice versa (Kramer et al., 2014)(King & Wincup, 2010).
3.4.1 Reinforcement of criminal behavior
The increase of conspiracy-minded online environments is still seen across the world (Lane et al., 2021). A study conducted in the United States shows that 60 % of Americans get their news via social media, and that news flow is expected to be influenced by the echo chamber as previously accounted for (Lane et al., 2021). A great present example of how social media can reinforce radicalistic standpoints is how QAnon5 can lead individuals and/or groups towards engaging in criminal and violent behavior (Lane et al., 2021), which is what happened on the 6th of January 2021 in the US Capitol attack. It is believed that QAnon prior to a public speech held by President Trump organized themselves via social media – which later caused a full-scale invasion of Capitol Hill which led to human injury and death. The internet is a still fast-growing network where a still-growing part of our lives is carried out, whether it is online banking or communication with friends and family. Despite that, there is no consensus among researchers regarding the internet’s actual effect on human interaction. It is almost like an entire continuum of opinions that are more or less supported by scientific studies – ranging from a rapport made by the European Parliament in 2014 where they warn about
overestimating the impact and role of new online media (Bigo et al., 2014) to the former CIA operations officer psychiatrist Marc Sageman in his article “The next generation of terror” where he states that face-to-face radicalization has been replaced by online radicalization (Sageman, 2009). Benson (2014) rationalized the matter because he considers it odd if the terrorist did not use online media, as it would be odd if terrorists in the past did not use a telephone or printed material (Benson, 2014).
There is a general problem regarding how we view online lives because today's adults seem to look at online lives and physical lives as something that cannot be reconciled. Several of the professionals interviewed for this thesis point towards another viewpoint which is namely that online lives and physical lives are not two different things but are two worlds that co-exist and merged – especially for those born from the year 2000.
5 QAnon is a right wing conspiracy theory describing an alleged plot by the deep state
3.4.2 The role of the parents
In 2021 The City of Copenhagen surveyed how parents perceived their kids' online lives. The survey (N=1,162) was conducted by random selection from a public register and focused on parents to kids in the age 12-17 years old (Epinion, 2021) The results were quite interesting because 42% of the parents stated that they are aware of their kids experiencing degrading content on other people’s gender, sexuality, ethnicity, religion or social status (Epinion, 2021), 42% experienced misinformation, 36% experienced conspiracy theories and finally 17% have experienced content that encourages violence or crime (Epinion, 2021). 62% of the parents state that they fully or partially
know who their kids talk to online, and a little less, 42%, feel secure regarding the type of information their kids are exposed to (Epinion, 2021). Another study conducted by the Danish Crime Prevention Council states similar results (N=20), and they also asked about parents and how they surveil – most parents do not have surveillance on their kids’ phone, so their knowledge is based on sporadic comments from their kids (Insight, 2021)
The reason why parents are important is that they are the professional's most important collaborators because they help set the stage for supporting the framing of positive online lives.
There is one huge bias between these two studies and quantitative studies in general. The study from the City of Copenhagen had in total 1,162 respondents and the respondents with high social-economic statuses were overrepresented (84% of respondents) as well as married parents (12 % overrepresented) (Epinion, 2021). A general tendency among studying behavior in socially disadvantaged residential areas is the low response rate, and it is hard to collect data based on individuals with low social-economic status (SES) and create incitement for them to participate. As in the study by the City of Copenhagen to compensate for the lack of data, they conducted a post-stratification, but that lift another issue which is the sample selection bias, and some of the surveys conducted within this field have some founding issues with the internal validity.
Gladsaxe Municipality has recently conducted three studies on collective efficacy among adults in disadvantaged residential areas, and then they experienced the same issues as in Copenhagen – the overrepresentation of adults with a high SES, and their conclusion points towards especially a language barrier as one of the reasons to the low response rate (Thomsen, 2020).
To compensate for that Gladsaxe Municipality started using a method where the surveys are distributed as random selection mail and also by using snowball sampling to target specific underrepresented ethnic groups, and groups with low SES (Thomsen, 2020).
In sum, the issue is when teenagers who are susceptible to outer influence, spend time with online media, that because of the personalized algorithms, place the individuals in a filter bubble, echo chamber, and hyper expose the user to conforming content – and that is expected to be conducive to deviance or conformity (Wolfowicz, 2015). This field does need further research, based on specifically the consequences of being exposed to extremist content online – and also how can parents, governments, and the kids work together to prevent it to happen.
4: Methods and ethics
This section will account for the methods used in the genesis of this thesis.
Finally, this section also accounts for the ethical aspects and considerations of this project.
4.1 Methodology
The thesis investigates the potential gap between how professionals view online radicalization and prevention, and how criminology perceives that. The main empirical foundation of this thesis is qualitative. Furthermore, the thesis seeks to concatenate previous studies and the conducted interviews.
The interviews are conducted as semi-structured interviews with interviewees that directly or indirectly work with online radicalism in their professional life.
The theoretical foundation of the methodology in this thesis is based on a grounded theory perspective – as opposed to the hypothetico-deductive perspective, it seems more relevant to this project to approach the topic
inductively (King & Wincup, 2010). From the grounded theory perspective this project meets the interviewees with an open mind and theorizes the analyses of the gathered material from the interviews. This is a strategy for conceptualizing, describing, and explaining what is at stake for the professionals (Shah, 2019).
4.2 Bias
Even though conducting qualitative in-depth interviews is often criticized for being biased and have a low level of validation, the qualitative interviews gather information on topics that has a high level of complexity and are normally difficult to deduct into controlled variables in a quantitative survey such as pathways to online radicalism.
In this project data from both primary and secondary sources are included. The primary sources are the interviewees, and the secondary sources are the literature gathered. The interviewees are selected through at least the second link in the network of the author. The reason for that is to avoid any biases regarding fear of employment issues, or personal knowledge bias.
In sum, this thesis looks into the thinking of the practitioners and how they relate their working methods to already known criminological theory. By making a comparative analysis the flaws and strengths of preventing online radicalization will be clearer.
This project is overall focused on the two types of bias, participant bias and researcher bias (Shah, 2019). Under participant bias the aspect of acquiescence bias where the participant will agree just to complete the interview – especially if it is a long interview and the participant experience fatigue (Shah, 2019). To avoid that the interviews are kept within a timeframe of 30-60 minutes. Another aspect is the social desirability bias, when the respondent answer whatever they think, the interviewer would like to hear, with the purpose of feeling liked (Shah, 2019). To avoid that it is made very clear at the beginning of the interview that the outcome of the interview has no influence on the connection between the
In the formulation of the interview questions, there is a risk of a habituation bias that can occur, when the respondent answers the same to questions with similar words (Shah, 2019). The questions are there for formulated in different wording and with down-to-earth phrasing.
The risk of sponsor bias is not considered relevant in this project, since it is an independent project.
On the other hand, because of the research bias risk, this project is focused on three aspects; confirmation bias, questions bias, and leading question bias. In this project, the design of the questions is constructed as open questions, with no leading elements in them. The main focus point is to start a dialog and ask follow-up questions (Denscombe et al., 2009; Shah, 2019). Another reason why the questions are constructed open is to avoid asking leading questions and to ask only questions that are not prejudiced.
4.3 Validation
It is hard to strengthen the validation of the information obtained in qualitative interviews, but this project seeks towards using triangulation to validate the statements obtained in the interviews. By combining the interview data with previous quantitative studies on internet users, a part of the triangulation in this project is to generalize aspects of relevance.
4.4 Ethics
Ethical considerations when studying human behavior are highly important because the act of studying and the following analysis can have tremendous effects on human lives. The study method that might have the lowest risk of affecting human lives could be observation studies because an actual interaction is not present.
In general, every time two people communicate it has some sort of impact on each other, and it is especially that impact that can cause ethical implications.
Another criterion of collecting data in qualitative interviews is the importance of the interviewee to exactly know the context of the questions. The researcher Descombe made a study on taxi driver's experience on crime. The researcher had preliminary informal talks with a random taxi driver, where the taxi driver’s answers ended up being a foundation of a later study (Aadland, 2008). When respondents are not aware of in which context their answer will be used, they can end up giving answers that do not necessarily represent their believes or
standpoints.
Since this project is focused on the professionals and their work with deviant teenagers and adolescence a part of it is focused on ethical aspects of their approach to the target group and the discrepancy between the professionals' ethical perspectives and the target groups' ethical perspective.
This project is operating within the three ethical main groups; Action
deontological, rule deontological, and teleological theories (Aadland, 2008). The theoretical dilemma in this project is that the professionals work within both of the deontological perspectives, which is focused on the rule of action, and what you are supposed to do – but also focused on the set of rules and law by which the professionals are imposed to work within.
could be obtaining stronger social bonds, and the means could be engaging in online criminal activities to gain acceptance. In the analyses of the interviews, these dilemmas and dynamics will be clear.
This project was presented to the Ethical Board at Malmö University, and they had no objections.
5: Collected data
For this thesis, there has been conducted five semi-structured interviews with professionals that either work in a municipality setting or a government setting. Despite the different angles they all work towards the same endgame – which is preventing the Danish youth from being radicalized and embedded into extremist environments online. All of the interviewees participated in this thesis
anonymously.
The data is presented in three major themes; the first is “Current prevention methods” which describes what the professionals already, and the second theme is “Fundamental issues” which describes the issues that stand in the way of
conducting best practice and the last theme is the “Suggestions from practitioners” which describes the professional’s ideas on how to improve this field.
5.1: The interviewees
- The first interviewee is called Bob and he is working in a medium-sized municipality in the Central Region of Denmark. He works as a consultant at the department of crime prevention and has been working there for almost 8 years.
- The second interviewee is called Alice and she is working in a suburban municipality outside Copenhagen. She works with the feeling of safety and crime prevention – she is also a former member of a media council.
- The third interviewee is called Steven and he is working within the National Police department of prevention. He has been working there for about 20 years.
- The fourth interviewee is called Christy and she is working at a local police department in the Southern Region of Denmark, and she works with the prevention of petty crimes. She has been working there for almost 10 years.
- The fifth interviewee is called Anthony and he is an assistant professor at a Danish University College, and he works with teaching pedagogue classes and has been doing that for three years.
5.2: Current prevention methods
All of the interviewees express that there are no standardized working methods to prevent teenagers from being radicalized in online communities. In this following section, the interviewees present how they developed their methods.
5.2.1: Bob
among youth groups, and at the same time get an overview of who participates in the groups and especially how the young people participate. A lot of the analysis is about hierarchy and what is legitimate in the group. We then pass on the knowledge to the social workers so they can initiate relevant endeavors.” (Bob, 2021)
At Bob’s workplace, they especially work with their own risk assessment tools; they are inspired by models such as PATRIARK and SARA:SV6. Seen from a
criminological perspective their prevention efforts are based on sociological theories, where professionals not only look for weak social bonds to peers but also social bonds among family. Since one of their focus points is the connection to criminal peers it can be argued that they have an interactionist perspective. They are also inspired by the Danish Centre for Digital Youth Care and their presentations on “how to teach kids about the pitfalls of the internet”, is something they do because they prepare them for making educated decisions when navigating in online forums.
“We hold presentations on digital education where the content is very much centered around how to interact online and in social forums, and especially also source criticism, what rights you have and what significance algorithms have.” (Bob, 2021)
Since their working methods are not standardized, it could be interesting to know how they evaluate their preventative work. At Bob’s work, they do not evaluate, but they take an educated guess on the effects of their work.
“It is difficult to say anything about the effect of our preventative work, but I believe that it provides a common language that can be used in the daily conversations you have in a school class. The teacher could do that, too, but it just gives a different dynamic that someone from the outside can talk about these things - it can otherwise quickly drown in so much else” (Bob, 2021)
The problem with not systematically evaluate endeavors initiated by local agencies is the lack of efficiency. And it does not give an opportunity to ask the most relevant two questions “how it works” and “why it works”.
5.2.2: Alice
Alice believes that the problem within the existing prevention of online radicalism is about how the teachers and other professionals approach the matter. She
believes that the problems lie in the methodology of how the teachers address these issues verbally during a school class.
“Actually, super good material has been developed, there are super talented teachers, but they are really afraid of what people say if you discuss a sensitive topic such as extremist opinions, and then does not have the time to follow up on it…” (Alice, 2021)
Their approach is pragmatic, and they have this open-minded idea that the kids should be provided a safe zone to discuss and challenge opinions.
6 PATRIARK is a risk assessment tool within honor-based crimes – and SARA:SV is an
“There are no standardized risk assessment tools for online extremism, but if, for example, you take an article from Rasmus Paludan7 and read it through, you will be able to see through some of the words he uses that he has definitely been on 4Chan because there are some of the same words that recur.” (Alice, 2021) Their empirical foundation is in other words based on intuition when spotting risk behavior, and their working methods reminds mostly of an approach inspired by symbolic interactionism.
5.2.3: Steven and Christy
They work both at the police and work with the prevention of crime. Despite their police background they experience both a more human approach towards
prevention and do not use deterrence as a prevention method.
“There is sometimes a discrepancy between the knowledge obtained by the professionals and school pupils. In the beginning, our presentation to schools was about online forums and how they worked, and how criminal organizations
worked. It was quickly visible that the pupils knew quite a lot about how these dynamics work, and quite a few of them are mini experts and have quite extensive knowledge on how and where to look for online debate forums. On the other hand, a lot of the professionals have this old traditional perspective on online forums – and do not necessarily see the potential force online forums can have as a criminogenic factor. It is my experience, that these professionals often see the youths' time spent on online forums as something juxtaposed to a free time thing, instead of what it is – an unsupervised arena for complex social interactions”. (Steven, 2021)
Christy agrees with the approach Alice is explaining regarding creating an open space and challenge the discussions.
“There is not much else to do than telling teachers and parents about the pitfalls of being embedded in online deviant communities, and to avoid parent panic and create a safe space to openly discuss concerns.” (Christy, 2021) This way of working preventatively is based on traditional ideas of approaching the preventative field. It is like in the old days where local schools held
presentations on how to perform safe sex, and how to avoid unwanted pregnancy – and the critique here is we do not know the effect of giving presentations which mostly reminds us of a deterrence approach even though a professional as Christy states she wants so avoid parent panic.
5.3: Fundamental issues
When preventing an issue which you do not necessarily know the extent of, and more importantly what it consists of, puts the professionals somewhere on the continuum between advocating the problem and deny the problem, mostly navigated with the previous knowledge-based in sociology and psychology. Transferring current knowledge into an online version seems to be complicated.
5.3.1: Alice
“… The question is really whether you notice those with different opinions or other issues. Do you notice those who go with the thought that “the reason I have been stopped 15 times by the police must be due to my skin color” It could also be that some young people go with problems regarding sexuality, climate or that you
7 Rasmus Paludan is a right-wing racism convicted politician who is active on social
are a white quite nerdy boy and believe that “the reason girls don’t talk to me is because the girls just think I’m disgusting”. Do we notice them? Or do we in the Danish society more often notice people like us, the theory could at least point to that. Therefore, as an average teacher, a middle-aged white woman notices what they themselves look like, that is what they have learned. So the whole discussion about what is right and wrong really reflects who we are - our opinions are formed and built up from what we have learned from our parents and in particular, the network we are in. That perspective is missing!” (Alice, 2021) In sum, Alice believes that there is a lot of great working material, such as
pamphlets and guidelines – but she believes that the professionals need training in what problematic online behavior looks like in real life.
Alice is implying that in the example with the boys who believe they are stopped because of their skin color and the boys who believe that all girls think they are disgusting might have false narratives that dictate them as non-members of the surrounding community. In these two examples issues as racism and misogyny can work as a precursor for extremist behavior because Alice thinks that if those narratives are not challenged constructively then there is a chance of these kids turning to a community where they can talk to likeminded and root these narratives as a part of their identity.
5.3.2: Bob
Bob sees it as a problem that they do not use standardized tools because standardization levels up the quality of the professional’s work.
“We do not use standardized tools because nothing is really available.” (Bob, 2021)
And not using empirically proven methods can have unintended consequences. “When it comes to prevention, one assumes things about the citizens and can quickly end up stigmatizing the citizens. It is, therefore, important to qualify the efforts and methods we work with.” (Bob, 2021)
The issue with how the established prevention initiatives deal with the transaction from real-life extremist radicalization towards online radicalization is something Bob recognizes fully.
“Usually, street-level employees can easily overhear some conversations in, for example, a youth club, and in that way get a sense of what is happening out there with the young people. You just cannot do that online” “…It is a challenge that we do not have access to that world (online forums)”. (Bob, 2021)
Bob also recognizes the issues related to these well-known dynamics and how they are developing online.
“It can be a challenge to deal with these complex issues, where the professionals basically have a good grasp of the sociological models that describe the
standardized presentation based on something that potentially evolves overnight. To be fair some of the ground principles such as echo chamber, propaganda, and hate speech remain the same.
5.3.3: Steven and Christy
Christy and Steven point towards the issues of working with preventing online crime in general by mentioning the resources.
“The main issue in preventing online radicalization is resources. It is really comprehensive to adjust our way of thinking prevention into something that could take place online. Am I ending up doing police work from a computer – I am not sure that I want to do that.” (Steven, 2021)
“I am not sure that just making presentations on the pitfalls of online forums does the job. It makes me think of the outcome of years of touring public schools with a presentation of drugs, and the dangers thereof. The abuse on drugs remains the same, and has been for years.”(Christy, 2021)
Steven and Christy point towards another issue which is the professionals. They question whether their kind of professional is even the right profession for
preventing radicalism online. If it is then they currently do not have the right tools to do that effectively.
5.3.4: Anthony
Anthony states that there is a knowledge gap, and he esoterically implies that we maybe need to rethink risk factors – because being radicalized might once have shown itself by growing a beard and isolating – but signs of being on the pathway towards extremism could be anti-social opinions on ethnic minorities or misogyny – and not the traditional symptoms.
“As professionals, we are trained to be able to spot risk factors in all possible contexts, but are we suited to be able to spot the quiet type who withdraws from social interactions, or if some teenagers have talked about identity changing topics, can we distinguish between constructive or destructive conversations among teenagers on e.g., homosexuality? We have the material and all this, but if we look at what is beyond statements about racism and extremism, do we see those risk factors? I do not think we do”. (Anthony, 2021)
The risk of overseeing risk factors because of the lack of a systematic approach can be high. On the other hand, giving definitive answers to complex and dynamic problems can be troublesome, because it almost seems like utopia to make a risk model that includes all aspects of human online interaction.
It seems like a lot of this comes down to understanding the contexts in which the kids are engaging in. Because it is not necessarily a sign of extremism that someone is introverted, does not like homosexuals, or starts to play shooting computer games – it is more what else is present, these cases need to be approached holistically.
5.4: Suggestions from practitioners
Online radicalization is an underestimated problem, and assistant professor Anthony thinks that we do not have adequate knowledge on online forums and recruitment to extremist environments.
familiar with. And the municipalities need modernization to be able to follow the youth and approach it on their field – the municipality is here to serve its people, not the other way around” (Anthony, 2021)
And Alice also suggests what could almost be described as a paradigm change into thinking resources and strong social bonds alternatively.
“One should inquire into the young person's online life, not just how he or she is in school and leisure club. Because in reality, the young person may well appear as someone with few resources in school or leisure club, but at the same time been captain of a Fortnite game or well-liked on Discord - so we must be aware that resources can look different and come from different fronts”. (Alice, 2021)
Christy from the police also explains that she sees online media as a platform to host and inspire real-life crime, but points towards the lack of knowledge. “In professional settings, we tend to focus on obvious problems and think
solutions we know works. I am not aware of any scientific research that explains the extent of this problem – and most of all how to prevent it. Because of that, it is hard to deal with these issues when there are no guidelines to lean on”. (Christy, 2021)
Alice points out that we also need to be cautious that we are not afraid of our youth spending time online. When she talks about stability, she metaphorically explains it as “legs on the chair” which refers to the need for stability in all social arenas (family, friends, work, and school) to keep seated.
“In a world where parents keep an eye on where their kids are on the phone and where there is rubber under the swing, so you do not hit yourself when you fall. Where do you go as a young person to test boundaries and develop your identity? As long as you just pay attention to the legs of the chair”
Christy also agrees with Alice’s point “… and I believe that we need to create a foundation for strong positive relations on online forums and avoid extremist organizations to take advantage of the cognitive opening of a vulnerable teenager.” (Christy, 2021)
Alice ends her interview by mentioning a great point “Few people want to fight for their own ideology, most are just attracted to the community.” “… And it is our job to make sure they get attracted to the right communities”. (Alice, 2021) In sum, Alice believes that the preventative work did not follow the development of the internet and internet use. She also believes that society should change its view on internet forums, and train professionals to be better at looking for resources that could be a force into preventing online radicalism.
But one of the biggest obstacles is the fact that there is a lack of knowledge and guidelines on how to prevent online crime – and they want more research to be carried out and follow up on relevant recommendations on how to approach it.
6: Limitations
research is not fulfilling at all. Also, these interviews offer detailed reflections on best practices, challenges, and areas to improve.
Alternatively, it would have been ideal to conduct quantitative studies based on professionals on a larger scale, but another question arises here; Are there enough practitioners to secure validity in a quantitative study in this field.
6.1: Theoretical discussion
Hence the fact that there have not been conducted studies on how to prevent online extremism in a Danish setting it has not been possible to carry out a comparative study on the theoretical framework. Most of the theoretical
framework in criminology, in general, is typically founded in an American setting such as Social Learning Theory and Social Bonds Theory, and whether that is even applicable in a Scandinavian context is in some aspects still a debate. Another general critique in studies on crime is the gender difference. Most criminological theories are based on male offenders, and just because we assume males are more likely into getting involved in online crime, it does not mean that it is the truth and females are less likely to get involved in online crime.
To apply Social Learning Theory into understanding the dynamics of being online radicalized seems reasonable, and especially how Alice is talking about being aware of a cognitive opening and a normative conflict. Because in some way the normative conflict is a normal state in a teenager's brain, in a world full of outer influence – where it was the social realistic tunes from Beatles in the 60’ies it is today the inequality and hunger easily broadcasted through the internet as videos or blog posts. And the cognitive opening is also a thing that could easily be created by the break from a high school sweetheart, not having the right fashionable clothing, low self-esteem, or wrong self-narratives. In both Social Bonds and Social Learning Theory seems to explain how an individual who does not feel like a member of the surrounding norm society feels, and how
technological solutions and forums can both aid and abets the development of extremist behavior.
In this debate on online extremism, part of the research field believes that being exposed to extremist content online will not lead to real-life radicalism explained by the catharsis hypothesis. It seems more likely that catharsis could apply to some younger people – especially those with strong positive social bonds to the surrounding norm society. But in the Facebook study where they examined how users got affected by content, did show a causality, so that people did experience an effect on their mood based on whether they were exposed to overly negative or positive content. It is therefore fair to assume, that it could be applicable in cases where we discuss being exposed to online extremist content. But what we do not know is the predictive validity of using already existing risk assessment models in an online setting – therefore further research would be needed to assess that but also to establish the effect on being exposed to online extremist content.
Further research is also needed to establish whether the theoretical framework can be applied in an online setting.
6.2: Practical discussion
Roughly all of the practitioners use the same old tool from the toolbox – which is giving presentations. Some are more science-based than others, thinking of Bob who has his analytical approach to analyzing dynamics, and Christy and Steven who just give presentations on how these forums work. But most important of all is that they all point towards the same issue – the lack of knowledge on the field. It is important to use evidence-based methods in the work with humans – and also assessing and evaluating methods to secure the most efficient outcome. A study in the Nordic Police Research searched for the effect of using evidence-based risk assessment models by Police officers in a Swedish context. The study showed that using evidence-based methods is useful in predicting violent behavior among male offenders (Svalin et al., 2017).
There is also an ethical issue with not having a systematic approach in the prevention of online extremism because the practical resources can be allocated based on false criteria and therefore the endeavors initiated can be without effect or be aimed at the wrong group of kids.
The study from The City of Copenhagen also showed that the parents are highly interested in following their kid's online lives – but might need more tools and knowledge to be ready to do that constructively.
The empirical foundation of that study, but also other studies made on online behavior, has a slight skewness, and it could enlighten knowledge on this field if research within individuals who have a low social economic status was included. The majority of offenders, in general, have their origin in this group, so it would be expected to look similar in an online context.
The interviews show a field with no regulations or national guidelines, and despite the Police have had a department for online crime since 2014 it is objectionable that practitioners do not have any official methods that are empirical-based and supported by science.
7: Conclusion
8: Reference
Digital risikoadfærd. (2021). Det Kriminalpræventive Råd and Forbrugerrådet Tænk.
Adams, W. C. (2015). Conducting Semi-Structured Interviews. Handbook of Practical Program Evaluation: Fourth Edition, August 2015, 492–505. https://doi.org/10.1002/9781119171386.ch19
Akers, R. L. (1990). Rational Choice, Deterrence, and Social Learning Theory in Criminology: The Path Not Taken*. Journal of Criminal Law and
Criminology, 81(3), 653–676. https://doi.org/10.4324/9781315095301-14 Akers, R. L. (2009). Social Learning and Social Structure - a General Theory of
Crime and Deviance. Transaction Publishers, Rutgers. Alice. (2021). Interview no. 2.
Anthony. (2021). Interview no. 5. Anthony. (2021). Interview no. 5.
Bakshy, E., Messing, S., & Adamic, L. A. (2015). Exposure to ideologically diverse news and opinion on Facebook. Science, 348(6239), 1130–1132. https://doi.org/10.1126/science.aaa1160
Benson, D. C. (2014). Why the Internet Is Not Increasing Terrorism. Security Studies, 23(2), 293–328. https://doi.org/10.1080/09636412.2014.905353 Bigo, D., Laurent, B., Guittet, E.-P., & Ragazzi, F. (2014). Preventing and
countering youth radicalization in the EU. Pigen, (2017).
Bob. (2021). Interview no. 1.
Borum, R. (2011). Radicalization into Violent Extremism I: A Review of Social Science Theories. Journal of Strategic Security, 4(4), 7–36.
https://doi.org/10.5038/1944-0472.4.4.1
Bryman, A. (2016). Social Research Methods. Oxford University Press, USA. Christensen, K. D. (2016). Danske børn har Europarekord i skærmtid. Forstå
hvad de bruger deres tid på – og bliv overrasket undervejs. Zetland. Christy. (2021). Interview no. 4.
Demant, J., Jørgensen, K. E., & Harder, S. K. (2018). Unges kriminelle adfærd på nettet.
drivers’ experiences of crime. International Review of Victimology, 16(3), 301–308. https://doi.org/10.1177/026975800901600304
Epinion. (2021). Undersøgelse af forældre i Københavns Kommunes oplevelser af børn og unges onlineliv.
Gemmerli, T. (2011). Online-radikalisering: et uafklaret begreb.
http://pure.diis.dk/ws/files/58165/diisreport2014_07_FINAL_for_web.pdf Golfashani, N. (2003). Understanding Reliability and Validity in Qualitative
Research. The Qualitative Report, 8(4), 597–607.
http://nsuworks.nova.edu/tqr%0Ahttp://nsuworks.nova.edu/tqr/vol8/iss4/6%0 Ahttps://nsuworks.nova.edu/tqr/vol8/iss4/6
Hemmingsen, A.-S. (2015). THE DANISH APPROACH TO COUNTERING AND PREVENTING EXTREMISM AND RADICALISATION THE DANISH APPROACH TO COUNTERING AND PREVENTING EXTREMISM AND RADICALIZATION Retsudvalget 2015-16 REU Alm.del Bilag 248 Offentligt THE DANISH APPROACH TO COUNTERING AND PREVENT.
www.diis.dk
Holt, T. J., Freilich, J. D., Chermak, S. M., & LaFree, G. (2018). Examining the utility of social control and social learning in the radicalization of violent and non-violent extremists. In Dynamics of Asymmetric Conflict: Pathways toward Terrorism and Genocide (Vol. 11, Issue 3, pp. 125–148). https://doi.org/10.1080/17467586.2018.1470661
Insight, L. G. (2021). Forældres rolle i børns digitale liv Forebyggelse af børn og unges risikoadfærd på nettet.
King, R. D., & Wincup, E. (2010). Doing research on crime and justice. Oxford University Press, USA.
Kramer, A. D. I., Guillory, J. E., & Hancock, J. T. (2014). Experimental evidence of massive-scale emotional contagion through social networks. Proceedings of the National Academy of Sciences of the United States of America, 111(29), 10779. https://doi.org/10.1073/pnas.1412469111
Lane, J. E., Mccaffree, K., & Shults, F. L. (2021). Is radicalization reinforced by social media censorship ? Computers and Society, 1–16.
Lemieux, A. F., Brachman, J. M., Levitt, J., & Wood, J. (2014). Inspire Magazine: A Critical Analysis of its Significance and Potential Impact Through the Lens of the Information, Motivation, and Behavioral Skills Model. Terrorism and Political Violence, 26(2), 354–371.
https://doi.org/10.1080/09546553.2013.828604
NCFE. (2020). Nye fænomener inden for ekstremisme online En pixi af ’
Kortlægning af viden om forebyggelse af ekstremisme online blandt børn og unge ’. 12.
Nissen, T. E. (2018). Nettets Vildveje. August.
Pritchett, S., & Moeller, K. (2021). Can social bonds and social learning theories help explain radical violent extremism? Nordic Journal of Criminology, 00(00), 1–19. https://doi.org/10.1080/2578983X.2021.1889133
Sageman, M. (2009). The next generation of terror. Foreign Policy. https://foreignpolicy.com/2009/10/08/the-next-generation-of-terror/ Schulze-Larsen, A. (2020). It kriminalitet i tal. Det Kriminalpræventive Råd.
https://dkr.dk/it/it-kriminalitet-i-tal/
Shah, S. (2019). 7 Biases to avoid in qualitative research. Editage.
https://www.editage.com/insights/7-biases-to-avoid-in-qualitative-research Steven. (2021). Interview no. 3.
Svalin, K., Mellgren, C., Levander, M. T., & Levander, S. (2017). The Inter-Rater Reliability of Violence Risk Assessment Tools Used by Police Employees in Swedish Police Settings. Nordisk Politiforskning, 4(01), 9–28.
https://doi.org/10.18261/issn.1894-8693-2017-01-03 Thomsen, N. (2020). Kortlægning i udsatte boligområder.
Trip, S., Bora, C. H., Marian, M., Halmajan, A., & Drugas, M. I. (2019). Psychological mechanisms involved in radicalization and extremism. A rational emotive behavioral conceptualization. Frontiers in Psychology, 10(MAR). https://doi.org/10.3389/fpsyg.2019.00437
Wolfowicz, M. (2015). A social learning theory examination of the complicity of personalization algorithms in the creation of echo chambers of online radicalization to violent extremism. 1–14.
