Thesis AMD Ardon
Supervisor: Els de Busser
Second Reader: Vlad Niculescu-Dinca Word count: 14 172 ex annexes
Crowdsourced investigation, distributed
agency & vigilantism: citizens’ assistance
at police investigations
Table of contents 1. Abstract ... 4 2. Introduction ... 5 3. Research question ... 5 3.1 Sub questions ... 5 3.2 Academic relevance ... 5 3.3 Societal relevance ... 6 3.4 Reading guide ... 6
4. Theory and literature ... 7
4.2 Changes in the security field: Security networks ... 8
4.3 Security networks, citizens and technology ... 9
4.4 Actor Network Theory ... 10
4.5 Crowdsourced investigation ... 11 4.6 Crowdsourced policing ... 13 4.7 Vigilantism ... 14 4.8 Distributed agency ... 15 5. Methodology ... 16 5.1 Expectations ... 17 5.2 Type of research... 17
5.3 Reliability & Validity ... 17
5.4 Sampling ... 18
5.5 What will be researched ... 20
5.6 Coding preparation... 23
5.7 Unit of analysis ... 25
5.8 Coding rules ... 25
5.9 Elaborate coding scheme ... 25
5.10 Coding: Type of Analysis ... 29
6. Analysis ... 29
6.1 Crowdsourced investigation: why these projects? ... 29
6.2 About the data analysis ... 30
6.3 Distributed agency ... 33
6.4 Vigilantism ... 35
7. Conclusion, discussion and recommendations ... 41
7.1 Conclusion ... 41
7.3 Recommendations ... 43 8. Bibliography ... 45
Table of Figures
1. Abstract
Crowdsourced investigation is an emerging concept within the security field, as it is one of the intersections between security networks, police investigations and citizens’ assistance. Within this study, the theoretical context is explored in which crowdsourced investigation came into existence, and tries to answer what the incentives are for police institutions to initiate these crowdsourced investigations. This is done by analyzing document data about child sexual exploitation material investigations and interviews conducted with police officials in this field. Within this analysis, distributed agency theory and vigilante theory are used to interpret the views of police institutions of the citizens assisting them in the
investigation process. The results showed that two indicators of distributed agency, having a shared goal and cooperation with citizens, are seen by the police institutions as incentives to initiate crowdsourced investigation. There was not enough evidence to believe that police institutions view the users as vigilantes and are hindered to initiate crowdsourced
investigations by this. What should be noted is that the police institutions do see vigilantism and especially the risk of the use of illegal means by actors. They view this as an unintended outcome which can exist next to distributed agency. Further research into the risks of
2. Introduction
Linus’s law:
“given enough eyeballs, all bugs are shallow”
In May 2017, Europol started a project called “Stop child abuse” which shows small details of pictures of child sexual exploitation material (Europol, 2017a). For instance: a bottle of shampoo, a distinct shirt or the cover of a book. Europol then asks the website’s visitors to analyze the picture and try to recognize anything that might give a clue to where the picture was taken or other possible crucial information about the offender or victim. The idea fits Linus’s law as stated above; a motto used often during software development, that more eyeballs (more people), know more and are thus able to solve cases more quickly. This way, Europol has found a new possibility to gather intelligence about child sexual exploitation material cases. This raises new questions about why Europol turned to this way of doing investigations.
3. Research question
What are the incentives for police institutions to initiate crowdsourced investigation?
3.1 Sub questions
• What are the changes in the security field that made the rise of crowdsourced investigation possible?
• What is crowdsourced investigation?
• What is vigilante theory and what is distributed agency theory?
• How do police institutions see the citizens participating in crowdsourced investigation processes with regard to the distributed agency and vigilantism theory?
• What are the risks and benefits of crowdsourced investigation for police institutions?
3.2 Academic relevance
With the rise of the internet, police institutions have discovered new ways of investigations. This fits the general trend of private actors becoming more involved in security related issues (Spearin, 2010; Feigenbaum, Henig, 1994 p 185). These private actors are now not only private companies, but also private persons (citizens) helping to solve cases. Instead of a team
research in the field of private security and security networks (Spearin, 2010; Dupont 2004), there has not been much research done on crowdsourced investigations: most of the research is focused on the involvement of private companies instead of private citizens. There has been some research about how online groups tried to solve the Boston Bombing case, although this was mostly from a crisis response perspective (Tapia et al, 2014). This means there is an interesting gap to explore between the fields of security networks, online crowdsourced investigation and the institutional view on this new investigation technique by the police.
3.3 Societal relevance
In 2017, a record number of child sexual abuse cases were reported to the Police in the Netherlands, it was so much the police warned they might not have enough staff to invest all the cases (NOS, 2018). This means that citizens’ assistance at these investigations can possibly balance this workload. Although crowdsourced investigation can seem like an elegant solution to solve these child sexual exploitation material cases, especially the ones that might have been stuck for a long time, it can also have negative effects. In August 2017, the City of Rotterdam had a terrorist threat, which later turned out to be caused by online messages of an amateur investigator. He was trying to lure real terrorists into believing he was one of them, in an attempt to get information. This phenomenon doesn’t limit itself to the field of amateur anti-terrorism actions. Online investigations conducted by amateurs can lead to false accusations, media storms and sometimes even to violence when conclusions are made too fast, people are being threatened online and media outlets start to believe the “fast” online investigation as more reliable than the police investigation (see Nhan et al, 2017; Schneider and Trottier, 2012; Tapia et al, 2014). This means that the involvement of private persons within investigation projects when done wrong, can actually be harmful to the cause. A thorough understanding of the phenomenon of is thus necessary to conduct these kinds of investigations in an effective manner.
3.4 Reading guide
This thesis starts with an exploration of the changes in the security field in the last decades to answer the question “why now” when it comes to crowdsourced investigation and the factors making these cooperative investigations possible. This will funnel down to what is known about the topic of crowdsourced investigations itself. This will be followed with an in-depth exploration of the theories of vigilantism and distributed agency as possible ways to view the
citizens’ participating in these crowdsourced investigations. In the analytical part, there will be an exploration of how police institutions view these citizens in the light of these
aforementioned theories and try to explain why this might lead to incentives to cooperate with citizens. For this, two expectations based on vigilantism and distributed agency are used. This will lead to the conclusion where there will be formulated an answer to the main research question.
This research will use multiple cases in the field of child sexual exploitation material investigations and citizen’s involvement in these projects. Within the media and the general public, child sexual exploitation material or child sexual abuse is often called “child porn”, “kiddie porn” or “child sex tourism”. Both Interpol and Europol strongly recommend using the terms “child sexual abuse” and “child sexual exploitation”, focusing on the fact that minors can’t consent to sex, and the footage of sexual abuse should thus not be called
pornography (Interpol, 2017a). This research will follow these language guidelines provided by among others Interpol and Europol in the Luxembourg Guidelines (see Luxembourg Guidelines, 2016). For the users that provide information to the police institutions the words “users”, “individuals”, “citizens” and “actors” will be used interchangeably.
4. Theory and literature
Within the theory and literature section, the contextual changes within the security field that made the rise of crowdsourced investigation possible will be discussed. This will include a comparison between crowdsourced investigation and crowdsourced policing and an
explanation of the theoretical concepts used for this study: vigilantism and distributed agency.
4.1 Introduction
When looking at the changes in the security field in the past decade and their context, Europol is an example of an organization that came into existence because of these contextual
changes. Europol came to life during the midst of the 1990s (Schalken and Pronk, 2002). The EU saw a growing threat of organized criminal activities that intersected traditional forms of crime, next to a growing threat of transboundary crime involving more than two EU countries (European Union, 2016). This was the time police cooperation on a European level was discussed, and quite soon it became clear that if there were to be joint police investigation teams, there should be an institution coordinating it. In 1997 the Treaty of Amsterdam set the guidelines for the mandate Europol had. This meant Europol received a mandate for joint
exchange officers and equipment (European Union, 1997). Two years later the mandate changed in Tampere 1999. the earlier perceived threats about transboundary crimes grew, as when Schengen became effective, next to citizens it became also easier for criminals to travel freely within the Schengen zone (European Council, 1999). There was decided that Europol could be trusted to call for investigation procedures themselves, and Schengen gave a mandate to Europol for assisting member states with interrogations, infiltration and
intercepting of intelligence. This changed the role of Europol. Where it was at its beginning mainly the organ gathering, sharing and propagating of intelligence when asked by member states, at the beginning of the new millennium the institution could request form joint
investigation teams themselves with far reaching mandates on how to gather this intelligence (European Council, 1999). Although member states that are involved in cross border
investigations still do the organizing part (Schalken and Pronk, 2002). In practice, this doesn’t happen as easily as it does in the treaty documents. National police forces are tepid about sharing their intelligence with other countries. Although Europol can ask governments for an explanation why they don’t share the asked intelligence, Europol can’t coerce national police to give up their information (Schalken and Pronk, 2002). This means that Europol has to supplement the information that they get with their own intelligence, which is sometimes gathered via open source channels as well (Schalken and Pronk, 2002).
4.2 Changes in the security field: Security networks
The creation of Europol fits the contemporary observations that can be made about the organizational changes in security-focused institutions. From the 1980s on, security has become less an objective that is reached through solely state means, and more a goal that can be reached through horizontal network cooperation (Whelan, 2012). After the collapse of the USSR, western governments realized the biggest threat wasn’t foreign countries any more (Whelan, 2012). These were the threats that could be kept at bay with military measures. Suddenly, the focus became on transnational crime. These states realized that to fight these forms of international highly coordinated crime, there was a need of international horizontal networks cooperating in the different countries that were affected (Whelan, 2012). For instance, when creating measures about drug trafficking, the countries where the drug is made, shipped from and to and the main areas where the drugs are sold are all important (often transnational) nodes in the crime process. If the police actors in these countries work together and for instance, combine intelligence, the bigger picture about how these criminals
work can be created and combatted as well. From the end of the 80s, security networks came into existence. Europol is one of the prime examples of these kinds of cooperation. Next to horizontal participation of institutions, the involved actors changed as well. Before security networks, security was a topic with a state monopoly (Whelan, 2012). Only governments were involved with creating a safe society. With the financial reforms that came with New Public Management in the 1980s, the government realized that they were not capable of providing a secure state of their own. When security networks started to emerge, this created also a space for private actors to join the security field (Whelan, 2012). This can lead to different forms of security actors. Next to the traditional public police, intelligence and military actors, private forms of security can deal with tasks as well. This can be building security, private police, and also private investigation companies (Whelan, 2012). The power and decisions about security have been replaced from a state-centered mandate to a form which also includes involvement of non-state actors.
4.3 Security networks, citizens and technology
Private security companies are not the only new actor within the security field. Citizens can also participate in the security field (Terpstra, 2005). Terpstra writes about the typology of security networks and notices that especially in preventative security networks, there is room for citizens to participate (Terpstra, 2005). Terpstra specifically made this notion for
participating citizens on a local level. For instance, networks about neighborhood watch and shop owners collaborating to create safer streets. This can be applied to a more international level as well if we look at the way internet can be involved within these networks. As the internet can connect people from all over the world, the local aspect can become less
important and create space for citizens to discuss safety issues that can happen outside or their local neighborhood. Dupont writes about these types of networks where the exchange of information is key, and calls them informational networks. Within these networks, technology is often key as modern digital technology drives the exchange and storage of data and
information (Dupont, 2004). Some academics have even called this transition to information centered police work the information era of policing, linking to Castells theory about how technology created the age of information capitalism (Nhan, 2017). This theory from the 1970s has Marxist roots, and starts from the point that information(structures) have become a new type of capitalism, whereas information is a new way of gaining money. The way of organizing, structuring and store information has become much more important, which
explain the big boom of technological developments in the past decades (Webster, 2016). Information has become more of a central part of people’s lives, and everybody is more connected within these information systems as well. One of the examples of this is social media, which can be seen as big informational networks in which people voluntarily exchange information, and other actors (for instance, Facebook) collecting user data to gain money with. Castells argues that the world has become smaller because of this: in the past, it took weeks to deliver letters to another continent, now people can reach each other in seconds via email. This, Castells argues, is not just a change in the world, but also a change how humans connect to their environment. Before the information age, people highly valued their direct surroundings and the local dimension. Castells argues that because of the information revolution, the world feels smaller and people can more easily connect to people and events outside their local neighborhood. He calls this new dimension the space of flows: a concept of a (non-physical) space that would gain importance (Webster, 2006). If we connect this to Terpstra’s theory of local security networks, we could see these networks shift from a more local network out of necessity to the possibility of more international ones. Whereas people can still try to create more safety not only on their local level, but also in a wider world. Castells theory also connects to the way police officers do their job. Castells set the
expectation that all people would engage more with information in their jobs. All work would gain an information component (Webster, 2006). One way to see this could be the
digitalization that is currently happening in many workplaces. Nhan et al use this concept of the network society to explain why information has become a more central aspect of police work. Police officers have become “knowledge workers” that have to assess, analyze and collect more information than ever before (Nhan et al, 2017). The network society creates thus both circumstances in which citizens can work together in borderless networks, and a context in which the police can actually use help from citizens, using these networks.
4.4 Actor Network Theory
The way information networks are key to the functioning of the network could be seen through Actor Network Theory (ANT), a theory by Bruno Latour about the interactions between human actors and technological actants. Latour writes about security networks and states that technology facilitating the network can contain political design, but should also be seen as important as human actors (Latour in Niculescu Dinca, 2016). When analyzing these networks, these technological actants should be regarded with equal importance for the
network process as humans, as they both have the same agency. This was a criticism on social- and technological determinism that had a tendency to separate both the technical and social (human) parts of a network without seeing the ways technology and humans worked together (Latour, 1996). Latour notes that networks don’t have hierarchical positions, and that for instance thinking about networks this way pulls them away from their geographic location, and argues that international networks are not “bigger” in the traditional sense, but just more densely connected networks (Latour, 1996 p. 372). The lack of hierarchy can make the network exist both in local spaces but reach out internationally, connecting actors all over the world. This is consistent with Castells view of the world becoming “smaller” and more interconnected. ANT could be the bridge between Castells and Terpstra’s theory of local security networks, and thus be part of the context why participative international security networks appear now. Local networks and international ones can thus exist at the same time, and even being integrated in one and the same network. One way Bruno Latour shows the importance of technology in networks is by a thought experiment in which all technology gets taken away from the network process (Latour in Niculescu Dinca, 2016). For citizens
congregating online to discuss evidence pictures and deciding where these pictures have been taken, this would mean some actor would go to Europol (which one could argue would hardly exist in a world without networks) to get a description of the object, or perhaps a drawing of it. This would be taken to an offline space in which a select group of actors could join, take a look at the object, think if they have ever seen it before, and if they have they could go back to Europol to tell them about their clues. This would mean the international aspect would be lost as well, unless people travelled great distances to observe and discuss the evidence. What is interesting to note here is that the actual process of citizens interacting with the picture of the object evidence isn’t dependent on technology. The process of getting the picture of the object to the citizens, is. The process of interacting with the picture of the object could be described as crowdsourced investigation.
4.5 Crowdsourced investigation
Tapia et al wrote about how in 2013 citizens participated in a crowdsourced investigation network to identify the possible suspects of the Boston Bombings. After the Boston
Bombings, the FBI asked bystanders to upload media evidence they shot on the scene. Online groups started organizing themselves to analyze footage to create online mechanisms to do an online investigation process on the open source information. This was mainly the footage
made by people that were in the affected area at the time of the explosion, or footage made by traditional media (Tapia et al, 2014). They started analyzing this data frame by frame to determine what kind of bomb was used, and also if the bombers themselves could be spotted (Tapia et al, 2014). “The participants, both bystander and media, are participating in a policing investigation without the same training, constraints or awareness as officials have”, Tapia notes (Tapia et al, 2014 p. 61). The crowdsourced investigation done after the Boston Bombings are seen as the first in its kind, and differs from traditional help from citizens to the police. The researchers illustrate this with the traditional WANTED-posters, where police stated they needed help finding someone, and citizens could contact the police to give any information to them. This “early idea” of police and citizen cooperation made the police center of the “network” or cooperation, information flows were one-way, to the police and citizens could that way contribute (Tapia et al, 2014, p.61). Terpstra also notes that police are the central hub in networks with citizens involved (Terpstra, 2005). This is not always the case for the crowdsourced investigations. When the Boston Bombing happened, the FBI specifically asked witnesses to send them any footage that could contribute to the
investigation, and at the same time released intelligence as well to let the crowd help identify the suspects (Tapia et al, 2014). This all happened mainly online, which made the internet the central intelligence hub for analyzing the footage, instead of the police (Tapia et al, 2014). The authors find that crowdsourced investigation can contain four different components:
• Intelligence gathering: this can be police institutions setting up web pages where individuals can submit data, or individuals creating platforms (like boards on a forum) to gather and share data amongst each other.
• Crime solving: when there is crime solving, individuals don’t only share information with each other but analyze the data as well. In case of the Boston Bombings this lead to innocent civilians being “identified” as terrorists (Tapia et al, 2014).
• Group based or group filtered: this is not as much about the actual content of the investigation, but more the “behind-the-scenes” organization of the investigation. This can be people ordering all the information or coordinating who’s looking at which data.
• Media filtered: media filtering is the process of media (which can be journalistic media or social media) accidentally or on purpose using information or hypotheses created within these crowdsourcing processes, making it seem like the information is confirmed without actual confirmation from police institutions (Tapia et al, 2014).
4.6 Crowdsourced policing
Next to crowdsourced investigation, there is a similar concept called crowdsourced policing. Trottier et al describe this as the “the utilization by social media users of narratives consistent with criminal justice discourse, even as these activities stand apart from modern conventional policing” (Schneider and Trottier, 2012 p.62). They describe that the use of crowdsourced policing in 2011 after ice hockey riots lead to a “collective governance among social media users” (Schneider and Trottier, 2012 p.57) and also an increase of police presence on different types of social media (Schneider and Trottier, 2012). This is also about citizens not related to law enforcement working together on police work, or work related to police work, often through social media. The concepts of crowdsourced policing and crowdsourced
investigations have a lot of overlap. Crowdsourced policing focuses on online activity, and also has a big police involvement and interaction with the police. Next to this, within the articles about crowdsourced policing there is a big focus on the possible negative effects from vigilantism which is lacking in the crowdsourced investigation literature. Crowdsourced investigation on the other hand, focuses on the power that individual citizens can have that the police might lack (Tapia et al, 2014 p.61). This power can be seen as resources, as within these projects individuals are asked to provide (ordinary) knowledge and time on a voluntary basis, especially ordinary local knowledge the police may lack. This suits the projects of this research, because the police institutions are asking for information they can’t gather
themselves. This will be discussed in more detail within the methodology and analysis part. We do have to note here that both the concepts of crowdsourced investigation and
crowdsourced policing have not been thoroughly researched yet, which leads to some ambiguity where the one concept stops and the other one begins.
Both crowdsourced investigation and crowdsourced policing emphasize the role technology plays within the project. Cyberspace creates an environment in which citizens can find each other and collect data in such a manner that it makes the crowdsourcing of information and data analytics fairly easy (Schneider and Trottier, 2012 p. 62). This enables cooperation between citizens and police institutions as well. For instance, crowdsourced policing is made possible because surveillance techniques like cameras (CCTV) and online databases are not only available for the police, according to the authors. Individuals have access to smartphones with cameras and digital information hubs like Facebook, providing them with their own set of private investigating techniques (Schneider, Trottier, 2012).
4.7 Vigilantism
Within crowdsourced policing, vigilantism is mentioned as a possible risk. The authors note that the access individual citizens have to the aforementioned private investigation techniques give way to pseudo-police work conducted by citizens (Schneider, Trottier, 2012 p.62). This was also noted as a possible risk in the aftermath of the Boston Bombings, the primary case used for crowdsourced investigation research. This is backed up with the context the internet operates in: the internet is not a centrally governed structure. It was partially built and governed by individuals (if it’s governed at all) and the institutional values of the internet contain equality, libertarianism and a non-hierarchical community (Nhan et al, 2017). Users are used to a do-it-yourself-mentality and see themselves as independent, even separate from legal institutions (Nhan et al, 2017 p. 345). This combined with a “backbone” of open-source-based working, crowdsourcing and “doing it together” is not a strange concept within internet communities. These investigations often happen within ad-hoc formed groups after a big public event happen. The authors even coin this Digilantism, because of the particularly great fit of vigilantism in digital spaces (Nhan et al, 2017). When looking at definitions of
vigilantism, it can be observed that planned organization in response to crime of events, done by non-state agents are all three characteristics of vigilantism (Johnston, 1996). Next to this, the actors are not involved with state institutions and are independent from those, want personal or collective security for the group they are part of and, are willing to use force or illegal means to reach their goals (Johnston, 1996). This last part about using violence is also mentioned as a possible risk, Nhan et al observe cases in which the internet collective
Anonymous were so sure they “identified” a criminal in a sexual exploitation case, they published his personal information online (doxing), although they found out later he had nothing to do with the actual crime (Nhan et al, 2017 p. 345). This should also be taken into account when talking about violence or illegal means in vigilante theory in online spaces. In traditional vigilante theory this use of illegal means would be illustrated with illegal gun ownership or threats of the use of violence (Johnston, 1996 p. 233). If we apply this theory to digital spaces these illegal means get shaped by the online environment as well. This means that the “force” is not that much physical based, but can be the doxing of personal
information, hacking or threatening individuals. This leads to the following definition of vigilantism: “Vigilantism is a social movement giving rise to premeditated acts of force-or threatened force--by autonomous citizens. It arises as a reaction to the transgression of institutionalized norms by individuals or groups--or to their potential or imputed
offer assurances (or 'guarantees') of security both to participants and to other members of a given established order.” (Johnston, 1996 p.232).
4.8 Distributed agency
What we can see is that vigilante theory starts with a premise that individuals have political goals, or can even feel superior to the law enforcement. Distributed agency, on the other hand has a premise of different kind of actors with different backgrounds and motivations working together to reach a goal that cannot be attributed to one single actor (Garud, Karnøe, 2005 p. 94). Distributed agency is a concept developed within institutional thinking to explain
changes within institutions and between institutions and other actors (Whittle et al, 2011). The concept is about a possible variety of individuals and actors participating within or with institutions, hence distributing agency over multiple bodies. This thinking fits the 4th wave of institutional thinking (Whittle et al, 2011). In earlier waves of institutional thinking (around the 1980s), institutions were seen as fixed structures that gave meaning and functioned because of their continuous stability and aversion of change. Recent academic thinking about institutions do agree with this, but do believe that agency of individuals within or outside of the institutions can influence, work with or interact with institutions (Whittle et al, 2011, Abdelnour et al, 2017). This gives individuals a central role, but only in reaction to already existing institutions. A note should be made here that often, the research about this is about individuals that are within the institution itself. Whittle et al use institutional entrepreneurship for individuals within institutions that aim for deliberate institution building, while all the other actors involved in “interaction and conjoint activity of multiple actors and reflect social structure and context as they affect the distribution of agency through the system” including individuals outside the institution distributed agency (Whittle et al, p. 6). They also note that the spatial environment and the context play a role in this. For instance, Garod and Karnøe note in their research about distributed agency within technological projects that “users become involves based on the meanings they attach (…)” to projects, so individuals need an incentive to participate (Garod, Karnøe 2003 p.280). Distributed agency has a close
connection to the abovementioned Actor Network Theory because it describes action as a cooperation between a diverse set of actors and actants: “Actions emerge out of complicated constellations that are made of a hybrid mix of agencies like people, machines, and programs and that are embedded in coherent frames of action.” (Rammert, 2008 p.2). Distributed agency thus becomes a way of cooperation within networks, especially networks with
different kind of actors (institutions, individuals) and actants (technology). The definition used for distributed agency will thus be “The situation where the actions or operations of a range of different individuals often with different motivations, interests and in different places combine to create an outcome they all wanted.” (Oxford Committee, 2017).
Figure 1: Visualization of the theoretic funnel
5. Methodology
The research methodology section will discuss the expectations of this research, the used methods to test these expectations, a reflection on the reliability and validity of this study and the sample used for testing. Next to this, the concepts of distributed agency and vigilantism will be operationalized so they can be used for coding the collected data.
This research will not so much analyze the motivations of the individuals participating in crowdsourced investigations. More so, it will analyze how police institutions view these individuals. For this the two categories distributed agency and vigilantism will be used. Do police institutions perceive these individuals as self-organizing vigilantes that sees law
enforcement as inferior to their private investigation skills, or do police institutions view them as cooperative partners working to reach the same goal? For this, two hypothesis or
expectations will be used throughout this research:
Traditional investigation processes
+ security as a public good
Privatization of security: security
networks + participation of
private (individual) bodies
Theories about individual actors participating within
the security field
5.1 Expectations
Expectation 1: Police institutions will have a high incentive to initiate crowdsourced investigation projects if they view citizens participating in crowdsourced investigations as part of a distributed agency mechanism.
Expectation 2: Police institutions will have a low incentive to initiate crowdsourced investigation projects if they view citizens participating in crowdsourced investigation as vigilantes.
These expectations are chosen because as described above in the theoretical part, within distributed agency theory, individuals may have different backgrounds than police, they do work towards the same goals. Next to this, actors play a central role, although only within the context of the institution. Vigilante theory on the other hand sees actors as having their own (political) motivations and self-organization, sometimes even willing to use violence. The underlying assumption is then that this vigilantism can create risks for the police institutions, which makes them more skeptical to initiate these kinds of projects.
5.2 Type of research
This research is combination of content analysis and interviews. The content analysis part is the analysis of policy documents of various projects in the realm of child sexual exploitation material investigations (in the Netherlands, and international projects the Netherlands is involved in). The interviews are small in number, so they should be seen as a supporting method and not the main research method. The interviews are conducted with police officials involved in the actual child sexual exploitation material investigations. The document data that will be analyzed, is not specifically written for this research, or with the research question of this study in mind. This means the data should be analyzed in their proper context, to get the information supporting the research question in this project. The collected data (both the interviews and documents) will be analyzed with operationalized indicators for distributed agency and vigilantism.
5.3 Reliability & Validity
To measure reliability is to reflect on the repeatability of this study. One of the strengths of this research is that it uses clear indicators and a big part of the analysis is done on documents that will be available over time, making it thus easier to replicate. A downside of the use of
documents in this study is that not all documents are publicly available due to restrictions at Europol. This makes replication more difficult. The interviews will be harder to replicate, because answers can shift over time due to priming and new developments as the project is very new. To make the interviews as reliable as possible, transcripts are available.
When discussing validity, there must be a reflection on the generalization of this research (to other populations, for instance) but also on possible other causes creating a certain outcome (internal validity). This research comes from a starting point where the view of police institutions on individuals interacting with these aforementioned police institutions,
determines the incentive to work together with these actors. This doesn’t mean that their view of individuals is the only incentive for crowdsourced investigation. There can also be
constraints formed by judicial boundaries, financial stimuli or political imperatives. This research thus looks into one area that might have some effects on these incentives for crowdsourced investigation. External validity is about generalization to other universes or fields. Although the operationalization of the used theories and the interview questions can be used in a field that is not related to child sexual exploitation, this field does contain its own institutional context. Child sexual exploitation is something that is condemned universally for moral reasons, while drug trafficking (or even closer to child sexual exploitation: human trafficking) for instance has more political stances which can also determine the willingness of individuals to participate in projects like this. Factors like these should be considered when replicating similar studies outside the realm of child sexual exploitation.
5.4 Sampling
The universe that this research will take place in is child sexual exploitation material investigation projects where individual citizens can make online contributions to. This is because there have been several projects within this realm in the past years, in which
Europol’s Trace an Object campaign is the most prominent right now (Europol, 2017a). Next to this, there are a few other projects from which (document) data can be used for analysis. A more in-depth argument why the chosen projects are suitable can be found within the analysis chapter. A short overview of the projects:
- Stop Child Abuse (Europol): in Spring 2017 Europol launched the Stop Child Abuse project, in which they uploaded small details of child sexual exploitation material to their website. For instance, a pair of boots, a type of plastic bag, or a type of toy. The police institution asks the public to take a look at the objects and see if they can
identify more specific information about the object (Europol, 2017a). In response to this, several online media including Reddit and Bellingcat created online platforms to discuss the pictures together to identify them (Reddit, 2017, Bellingcat, 2017). As this is a clear request for the public to help investigate, this is an example of crowdsourced investigation.
- Meldpunt Kinderporno (EOKB): This is a central information hub in the
Netherlands. The foundation is funded by the government and individuals can submit (links to) possible child sexual exploitation material. The employees of the foundation will then analyze the data. If they deem the data to be illegal, they work together with Dutch ISPs to get the information removed from the internet (Meldpunt Kinderporno, 2017). This project is chosen for the cooperation between citizens within the Meldpunt that do the actual analysis of the data and the cooperation and privileges they get from the police institutions for doing so.
- Operation VICO (Interpol): In 2007, Interpol released a picture of an individual caught on child sexual assault material where he was visibly assaulting children. The individual altered the photos, creating a ‘swirl’ on his face so he couldn’t be identified. At a certain point Interpol succeeded “unswirling” his face and asked the public to help identify this man (Interpol, 2007). This is a clear case of intelligence gathering by the police institutions while the public is involved in crime solving.
Outside scope:
- International Child Sexual Exploitation image database (Interpol): This is a platform (based on I24/7, the international police database) used by Interpol member states to analyze child sexual exploitation material. The database gives alerts if has already been put in the database. It has identified over 10.000 victims in the past years (Interpol, 2017). Because access is restricted to national police institutions and
Europol only, the project doesn’t fall into the scope of crowdsourced investigations, as no untrained civilians are involved in the identification process.
- Sweetie (Terre des Hommes): The Dutch NGO Terre des Hommes created a video animation of a ten year old child, to trace child sexual exploitation via webcam
contacts. This lead to the organization gathering information of about 1000 individuals that offered money in trade for the performance of sexual acts. This collected data was then sent on to the police (Terre des Hommes, nd). This does not fit into the scope of
NGO and just sent on the police, the actual coordination between police institutions and the non-government actors is minimal. Next to that, it’s not really “crowdsourced” as it’s just one NGO involved and not a public group of actors.
- Financial Coalition Against Commercial Sexual Exploitation of Children Online: This network consists of different private actors fighting against child sexual
exploitation online. The network is open to new members that feel like they can help with this. It falls outside of the scope of this research because it focuses on the financial backbone of child sexual exploitation. For instance by following money flows and working together with credit card companies by shutting down accounts (European Financial Coalition, nd). Although it does contain a variety of non-police actors, no investigative work is done.
5.5 What will be researched
My sampling will be policy documents of Europol, Meldpunt Kinderporno and Interpol about their respective investigative projects projects with the public. Per project this will be:
Trace An Object (Europol)
Document name Why this
document?
Type of document Number of pages
Policy on the use of Stop Child Abuse – Trace an Object (Europol, 2017b)
This is the main policy document on the Trace an Object-campaign. Policy document 4 Europol Strategy 2016 – 2020 (Europol, 2016) This document talks about Europol’s goals and how to reach these, it also mentions their strategy with regards to cooperation with private actors. Policy document 28
Europol launches public appeal to help identify victims of child sexual
exploitation (Europol, 2017c)
This press release talks about the cooperation with individuals within the project. Press Release 2 10 new objects uploaded, continue to help stop child abuse (Europol, 2017d)
This is a follow-up press release about the project.
Press Release 2 Europol’s new regulation (Europol, 2017e) Within the Europol Strategy, this document was discussed as containing the judicial framework for possibilities of cooperation with private actors. European Regulation 62
Table 1: Documents used for the Trace an Object-project
Next to the analysis of these documents, there have been two interviews with people about the Trace an Object campaign for even more insight about this specific project. These interviews were with:
• Cathal Delaney, Europol Operations Directorate (team lead TWINS); • Dave Jansen, team lead “Team ter bestrijding van Kinderpornografie en
Kindersekstoerisme” (TBKK, the team against child sexual exploitation material and child sexual tourism).
Meldpunt Kinderporno (EOBK)
Document name Why this
document?
Type of document Number of pages
Op goede grond: de aanpak van seksueel
This document writes about all
geweld tegen kinderen. Chapter 4 about reporting. (Nationaal Rapporteur, 2014) the possibilities for professionals and private individuals to report child sexual exploitation to the police and government. Verslag van een
algemeen overleg op 25 mei 2016 (Tweede Kamer, 2016) This is a report from a conversation between
politicians and the minister who is responsible for the Dutch police about reporting child sexual exploitation. Government report 37 Meldpunt Kinderporno Jaarverslag 2016 (EOKM, 2017b)
This is the year report of the meldpunt that deals with the reports from individuals about child sexual exploitation. Year report 15 Persbericht 2016 EOKM (EOKM, 2017c)
This press release is about the reports the EOKM got that year.
Press Release 1
Operation Vico (Interpol)
Document name Why this
document?
Type of document Number of pages
Media release: INTERPOL seeks public’s help to identify man photographed sexually abusing children (Interpol. 2007a)
This is the initial press release of Interpol reaching out to the public for assistance.
Press release 1
INTERPOL
Operation Vico 2007 (Interpol, 2011)
In this video, the Secretary General of Interpol talks about the project and the cooperation with individuals. Video 6 minute-video Remarks by INTERPOL Secretary General Ronald K. Noble during the First Plenary Session, G8 Justice and Interior Ministers Meeting (Interpol, 2007b) Within this speech, the Secretary General of Interpol talks about the cooperation with individuals. Speech 3
Table 3: Documents used for the Operation Vico-project
5.6 Coding preparation
The used definition for distributed agency is: “The situation where the actions or operations of a range of different individuals often with different motivations, interests and in different places combine to create an outcome they all wanted.” (Oxford Committee, 2017). The source is theoretical.
This leads to the following indicators:
• Police institutions perceive actors to be diverse range of individuals;
• Police institutions perceive different motivations and interests of these actors; • Police institutions perceive that actors are working together;
• Police institutions perceive shared (wanted) outcome with these actors
The used definition for vigilante theory is: “Vigilantism is a social movement giving rise to premeditated acts of force-or threatened force--by autonomous citizens. It arises as a reaction to the transgression of institutionalized norms by individuals or groups--or to their potential or imputed transgression. Such acts are focused upon crime control and/ or social control and aim to offer assurances (or 'guarantees') of security both to participants and to other members of a given established order.” (Johnston, 1996 p.232). The source is a literature review about how the definition of vigilante changed over time.
This leads to for following indicators:
• Police institutions perceive actors as social movement or group with crime control as one of the goals;
• Police institutions perceive that the group is willing to use force outside of the power monopoly of the state;
• Police institutions perceive that the group assembled itself as a reaction to crime; • Police institutions perceive direct group acts are done to provide social – or crime
control for group participants and society
These categories are appropriate as the research is about how police institutions view these civilians helping them. These two theories look from a more cooperative and a more critical angle at the participating individuals. These categories capture not only the reasons why citizens cooperate with institutions, but also in which way, at the same time the categories can be used to capture how in policy papers civilians are described.
5.7 Unit of analysis
In this study there will be an analysis how police institutions view citizens participating in crowdsourced investigation projects. This will be done via content analysis of these police institutions and a limited number of interviews. This will mean the research will analyze these views on the level of the police institute in general. The interviews will be with individuals, but will be used to interpret the views of the police institution as well. This will be used to take measure of the way these institutions are open to similar crowdsourced investigation projects and the factors that play a role in the decision to run projects like these.
5.8 Coding rules
- If a sentence contains two codes, the sentence will be assigned to both codes - If the same information is mentioned twice in the same text (for instance, main text
and conclusion), it’s coded once.
- If a sentence can contain both a code from distributed agency and vigilante theory, it will be coded for both.
- Although documents will be used to say something about the police institution, to analyze these documents in more detail, codes will be assigned on a sentence level. - If a sentence is saying something about users, but the data doesn’t fit into one of the
existing codes, it will be assigned code 9 (Other).
5.9 Elaborate coding scheme
Within the coding scheme, both distributed agency and vigilante theory are given four
indicators that get multiple codes each. With these codes, the aforementioned documents from the three different projects and the interviews will be analyzed. Next to this, there will be a code 9.1 for data that is about the users but doesn’t fit within the existing codes.
Theory Definition Indicator Code
Distributed agency
“The situation where the actions or operations of a range of different individuals often with different motivations, interests and in different places combine to
Police institutions perceive actors to be diverse range of individuals
1.1 actors are seen as individuals 1.2 actors are seen as people with different
create an outcome they all wanted.” (Oxford Committee, 2017). backgrounds, characteristics etc. Police institutions perceive different motivations and interests of these actors 2.1 police institutions recognize actors may have different motivations 2.2 police institutions recognize these actors may have different interests Police institutions perceive that actors are working together 3.1 police institutions sense cooperation between these individuals (and/or with the police) 3.2 police
institutions sense the actors help each other to reach a certain goal Police institutions
perceive shared (wanted) outcome with these actors
4.1 police institutions have the same goal in mind as individual actors
Vigilante theory
“Vigilantism is a social movement giving rise to premeditated acts of force-or threatened force--by
Police institutions perceive actors as social movement
5.1 police institutions see
autonomous citizens. It arises as a reaction to the transgression of institutionalized norms by individuals or groups--or to their potential or imputed transgression. Such acts are focused upon crime control and/ or social control and aim to offer assurances (or 'guarantees') of security both to participants and to other members of a given established order.” (Johnston, 1996 p.232).
or group with crime control as one of the goals
actors as a coherent group 5.2 police institutions feel like the group has social or political goals (about crime control)
5.3 police institutions interpret that all actors share this common goal Police institutions
perceive that the group is willing to use force outside of the power monopoly of the state 6.1 police institutions perceive actors to be dissatisfied by the law enforcement’s use of power 6.2 police institutions perceive that actors are willing to use illegal means to reach goals
Police institutions perceive that the group assembled itself as a reaction to crime 7.1 police institutions perceive the groups to have
assembled themselves
Police institutions perceive direct group acts are done to provide social – or crime control for group participants and society
8.1 police institutions feel like actors or group want to act upon crime themselves 8.2 police institutions perceive group actors are wanting to provide social control
8.3 police institutions perceive group to act for themselves 8.4 police
institutions perceive group to act for society in general
Other This code is mainly if useful
information about the users (and a police institutional view of them) is mentioned within the
documents, but the data doesn’t fit within distributed agency or vigilante theory.
Data mentions users, but the data can’t be coded within one of existing codes
Table 4: coding scheme
5.10 Coding: Type of Analysis
The type of analysis is qualitative analysis, where the codes found in the data are used to say something deeper and explaining about the data and view in general. Counting may be used to say something in general but the content analysis is done to explain why and not how much. The interviews will be coded as well using the same coding scheme above.
6. Analysis
This analysis will start with a closer look on the chosen data samples and why they can be considered crowdsourced investigation projects. This will also contain a general overview per project on what kind of codes were found. Afterwards, the analyzed document and interview data will be structured per theory which will funnel to the expectation which is connected to the respective theory. In the final part, there will be an answer to the subquestion on how police institutions see the citizens participating in crowdsourced investigation processes and the benefits and risks of crowdsourced investigation processes.
6.1 Crowdsourced investigation: why these projects?
Firstly, we need to analyze why the chosen projects are indeed crowdsourced investigation. As discussed in the theoretical framework, crowdsourced investigation is a project between police and citizens wherein “The participants, both bystander and media, are participating in a policing investigation without the same training, constraints or awareness as officials have” (Tapia et al, 2014 p. 61). There are four mechanisms or processes that can be observed within these kind of projects: Intelligence gathering (both on the police site where it’s setting up websites for citizens to upload information to, or on the citizen side where individuals create platforms themselves), crime solving (where individuals try to analyze the data
themselves), group based work or group based filtering (about “behind the scenes” managing tasks to make the investigation process run smoothly, for instance flagging the data that has been identified or still needs to be analyzed) and media-filtered (about the involvement of traditional and online media in the process, for instance “confirming” theories or bringing attention to those on news platforms). From these four, the first three will be getting the
attention of this research, as we are focusing on individuals and groups involved in this process and not as much on the media that can be part of this process as well.
6.2 About the data analysis
During the data collection and analysis, a few (minor) issues were encountered. Firstly, we can observe that the way users are described differs per project. The Europol project, for instance, mentions the users in the biggest detail, while the Meldpunt Kinderporno barely mention the users at all, and mainly talk about the overall output (“we received an x number of reports this year”). During the analysis, there were also mentions of users that couldn’t be categorized in one of the two theories, for this a separate code (9, other) was created. These will be further discussed in the discussion part of this paper, but as they do provide important context it’s important to mention it here as well. During the data collection, the observation was made that within the policy papers the individuals helping the police institution aren’t extensively discussed. When conducting interviews, police officials do have a lot of thoughts about these individuals, but these views aren’t reflected in detail in the analyzed documents. This means there is a limited amount of data to work with, and thus that the amount of data that could be interpreted is lower than expected. This says something about the assumptions in the development of this research. There will be further examination of this in the discussion chapter of this study.
Europol Stop Child abuse
This project is a clear case of online crowdsourced investigation, as the police institution put up a website to submit data and individuals were creating online groups to discuss the data (intelligence gathering). Next to this, groups of individuals come together to analyze raw data (crime solving). This can be seen for instance in the different Reddit groups where individuals from all over the world participated in the process of working together to identify objects or locations (Reddit, 2017). As the groups were organized with moderators keeping in check which pictures were solved and giving regular updates, there was group filtering of the data as well (group based). This can also be observed in the same Reddit groups. Next to these online groups that can be observed by outsiders, there’s of course also regular citizens browsing the Europol page with the objects and reporting back to Europol if they recognize an object. This is also a process we can consider crime solving and intelligence gathering, but as this is more one-on-one it’s less visible for outsiders.
Within this sample, both interviews were conducted and document data was analyzed. This logically means that this sample or project has the most detailed data from all the projects, but also without the interviews the documents used for this were the most detailed for all three projects. For instance, within the policy papers from Europol about the Stop Child Abuse project, there was a mention of Europol’s responsibility of the safekeeping of the data against “accidental or unlawful destruction, accidental loss or unauthorised disclosure” (Europol, 2017b, p.2) referring to the possibility that users would unlawfully destruct or disclose data which could be seen as vigilantism (the use of illegal means). Also the general role of the citizens and the goal the police and citizens were working towards was described with some detail (Europol, 2017b; 2017c; 2017d). The biggest part of the mentions was about citizens “assisting” or “helping” Europol, which all fits more into the distributed agency theory. This will be discussed in more detail below.
Meldpunt Kinderporno
Within this project, individuals browsing the internet and spotting media that might contain child sexual exploitation, can report it to get it removed. Employees of the “meldpunt” (which can be translated as “reporting hub”) get special privileges from the police for watching and analyzing the media that gets forwarded. According to the vacancies on the website, this is a parttime job preferably for people with some legal background or knowledge of the internet, but it’s not a requirement (EOKM, 2017a). This means this is a project where individuals are supplying data to EOKM professionals, that do the analysis as a job, but don’t have the same constraints and power as police officers, next to the individuals who supply the data which don’t necessarily have a specific background or training at all. This makes it a crowdsourced investigation case by definition. We do have to note here that employees of the Meldpunt get special privileges for watching and analyzing the media, as they otherwise would be
committing criminal offenses (watching child sexual exploitation material) themselves. The actual hub and connection between individuals and the Meldpunt can be seen as intelligence gathering, while the employees analyzing the data can be seen as crime solving.
Within the documents used for Meldpunt Kinderporno, the least information was found about the users of the Meldpunt. As mentioned above, individuals report media to the Meldpunt’s employees for analysis. When the Meldpunt was mentioned, or cooperation on combatting
important actor, often with a lot of gratefulness towards the organization (see for instance Tweede Kamer, 2016 p.16). The individuals reporting are not mentioned as actors
cooperating or contributing to decreasing the amount of child sexual exploitation material on the internet at all, and the mentions over all are thus scarce. If they are mentioned, it’s often only measured as a statistic in the output they provided, for instance how many exploitive material reports the Meldpunt got (EOKM, 2017b). In a lot of the policy documents about reporting child sexual abuse material, the public was mainly mentioned as an actor next to professional healthcare facilities and for instance school teachers. The National Report on child abuse also admits that additional steps have to be taken when it comes to reporting digital child sexual exploitation (Nationaal Rapporteur, 2014 p.93).
Operation Vico
Within Operation Vico, the face of a known abuser was published so the public could help identify a man that was an active child abuser. The original webpage doesn’t exist any more, but a cached version from two days after the launch of the project shows that the website contained five different pictures of the abuser, with a public call to identify the man. Any information could be sent to the Interpol Trafficking for Human Beings Unit via email (Interpol via Wayback Machine, 2017). This means that citizens were called upon to analyze these pictured and identify the man in it. The results could be submitted via email. Although email isn’t a “separate platform”, this is a case of intelligence gathering as well where police institutions work with data submitted by citizens. The citizens trying to identify the offender can be seen as crime solving.
Where the Europol project does relatively often mention the users and the Meldpunt mainly is mentioned as an organization itself, the Interpol project is a bit in between these two. The individuals are mentioned, but not that much. Within a video about the project, Interpol’s Secretary General Roald K. Noble does mention the users, but also mentions how other police institutions and local police officers contributed to the project (Interpol, 2011). At the end of the video he also says “We use all the technology, all the resources, private, public, in order to make sure he or they [offender(s), AA] are identified, located and brought to justice.” (Interpol, 2011). Which could imply that Interpol sees public assistance mainly as one of the tools to combat child sexual assault images and not as distributed agents, or vigilantes. What can be said about all three projects is that there is some differentiation about the attention and
mention of individuals in their projects, which makes it hard to generalize over all three of the projects.
6.3 Distributed agency
When we look at distributed agency, four indicators were used:
- Police institutions perceive actors to be diverse range of individuals
Within the policy documents there was no mention of the diversity of users at all, this was too specific to be mentioned within the policy documents. The diversity of users was discussed in both the interviews, and both the interviewees mentioned to perceive (at least a part) of the actors as individuals. Cathal Delaney (Europol) for instance said:
“there’s quite a broad range of individuals that are contributing information. We have people with technical knowledge, with local knowledge, people that have just recognized something from the general knowledge that they have…” (Interview with
Cathal Delaney, 2017).
Although Dave Jansen from the Dutch police agrees on this view as well, as there is no mention to this in the policy documents, it’s thus hard to triangulate this
information.
- Police institutions perceive different motivations and interests of these actors. There is not enough mention of the motivations of the individuals for analysis.
- Police institutions perceive that actors are working together
This can be about cooperation between the individuals themselves, as well as cooperation between citizens and police institutions. What we can see from the interviews and the document data that the police does seem to feel like actors are working together with them, but also that the individuals share the workload together (for instance in Reddit groups) to provide better outcome. We can see this as in both the interviews and the document data the words “assisting” and “helping” are used often to mention the cooperation project. For instance, Europol refers to the interaction with individuals directly as “Remember, your help can provide the one piece of
General from Interpol notes in his speech “we have a mechanism and process in place through which we can seek help from the public (…)” in which he notes that the public can indeed help and work together with the police institution (Interpol, 2007b). Within the interviews, both the police officials saw it as a cooperation, but admitted the cooperation is one-sided, as the police doesn’t give much feedback to the individuals about what happens with the information they provided.
“we are looking for their cooperation, so we view them as cooperative partners but to a larger extent, that cooperation goes one way. Their cooperation is in the information that they provide to us. We’re not able to provide something back to them
as a result.” (Interview with Cathal Delaney)
This means that although there is cooperation, but the police are not contributing anything to the citizens. It’s not clear (and outside the scope of this research) if the citizens do expect any feedback from the police institution. As both policy documents and the interviews do view citizens as cooperating or cooperative partners, this
indicator can be considered true.
- Police institutions perceive shared (wanted) outcome with these actors
Both in the interviews as in the Europol document data, a shared outcome is perceived by the involved police institutions. The idea is that both the individuals and the police have a negative view of the committed crime, and strive to prosecute the offenders. This can clearly be seen in the interviews, where for instance Dave Jansen mentions that he sees the individuals helping like “his mother”, just people that want the same as him but helping in their own way.
“I would like to think their motivation is the same as ours, that they want a child safe as the result of their efforts.” – (Interview with Cathal Delaney)
Europol also mentions on their website that the project “seeks to use the global reach of the internet to allow the public to contribute to this investigation process and to play a real part in the global prevention of child abuse.” (Europol, 2017c), stating that the civilians have the same goal (global prevention of child abuse) as the police. What can be noted is that this is in close connection to the earlier mentioned working
together, this goal of global prevention is closely connected to the fact that the public can reach this shared outcome by providing assistance. As both policy documents and interviews do view citizens as having the same goal in mind as the police, this
indicator can be considered true.
6.4 Vigilantism
For vigilantism, four indicators were used as well.
- Police institutions perceive actors as social movement or group with crime control as one of the goals
There was not much information in the policy documents about this. When this phenomenon was mentioned, it was solely in the Europol project. During the
interviews, it got clear that at least the interviewed police officials felt like there were groups of people working with the information provided by Europol as well. Even more: there was a distinction between individual inputs and the input from people in online groups like Reddit and Bellingcat. They were often very positive about the input of individuals, but the groups were not something they expected when starting these projects, and it was seen as something unregulated as well. “They are completely outside of our control”, Dave Jansen said. As the interviews are supporting
documents, it’s difficult to use this as hard evidence of the actors as social group, but it does seem like the police officials make a distinction between individual
crowdsourced investigation and group based crowdsourced investigation. On the other hand, they didn’t perceive these groups to have alternative sociopolitical goals. One of the explanations for this can be given by Mr. Delaney:
“(…) in general, child sexual exploitation and abuse is a very politically neutral topic, everybody is against it.” (Interview with Cathal Delaney)
One could argue that within this specific universe of child sexual exploitation,
according to the police officials everybody has the same goal as it’s a topic everybody more-or-less has the same opinion on. But this is again only based on the interviews, as the documents didn’t discuss possible goals of users.
