By Kateira Aryaeinejad and Thomas Leo Scherer

In 2012, the National Institute of Justice’s (NIJ) Domestic Radicalization to Terrorism program began funding research on issues related to domestic radicalization and terrorism in the United States. A component of this portfolio is focused specifically on evaluations and assessments of terrorism prevention programs implemented across different populations. This report synthesizes these efforts, reviewing shared and singular lessons from NIJ-sponsored evaluations of terrorism prevention and terrorism prevention-relevant programming and focusing on both evaluation findings and processes. The projects reviewed for this report vary notably in their timelines, methods, programmatic focuses, and audiences. As such, the lessons drawn from them are not necessarily scientifically comparable or generalizable. However, important lessons emerge from this research, underscoring shared insights from NIJ-sponsored evaluative efforts and illuminating important areas for consideration in future programmatic and evaluative endeavors. First, findings from NIJ-sponsored evaluations emphasize the importance of gaining community buy-in and assuring program relevance for participants before implementing terrorism prevention or terrorism prevention-relevant programs. Although this may vary based on the intended program audience (e.g., youth, communities, law enforcement), the findings overall suggest that to ensure program fidelity, utility, relevance, and buy-in, programs should be developed in close consultation and cooperation with the audiences for which they are constructed. Doing so can also help alleviate concerns regarding the perceived stigmatization or targeting of specific communities in which terrorism prevention efforts take place and help address concerns about and issues in associating specific communities or demographics with national security threats. Second, programs for terrorism prevention and preventing/countering violent extremism can yield benefits beyond meeting terrorism prevention goals. They can also be complementary to and even carried out within public health or community resilience initiatives. The utility of incorporating terrorism prevention into these frameworks may vary based on the intended program audience and outputs. Careful attention should be paid in framing program activities and language when they are implemented via these larger frameworks. Indeed, if violent extremism awareness and prevention efforts are not appropriately integrated into a public health or community resilience model, participants may be confused as to the purpose, goals, and nature of the activities in which they are participating, to the detriment of short- and long-term programmatic goals and awarenessraising activities. Third, evaluative efforts in general face programmatic and methodological challenges that limit their ability to assess the impact of a program and the generalizability and veracity of their findings. NIJ-sponsored evaluations went to commendable lengths to address issues associated with data availability, generalizability, validity, and determinations of impact; however, certain challenges — including participant attrition rates and lack of available data — limited their ability to do so. These challenges offer important lessons for future evaluations of terrorism prevention programs. These evaluation findings suggest that several activities should be considered in addressing challenges that may arise during the evaluation process. These activities could include implementing formative evaluations or evaluability assessments prior to program implementation and evaluation, implementing strategies to limit participant attrition in the interest of maintaining appropriate sample sizes, and ensuring adequate resourcing for continued engagement in terrorism prevention initiatives. In addition, NIJ-sponsored evaluation findings suggest that program evaluators should consider incorporating control or comparison groups to assess programmatic impact on individual attitudes, behaviors, and outcomes, along with comparative analysis focused on assessing the outcomes and impact of programs replicated in different communities and settings to increase the generalizability, validity, and utility of evaluation findings. More information about NIJ-sponsored efforts to incorporate these elements is detailed in this report. Looking forward, findings from this review suggest important gaps and considerations that should be addressed by and incorporated in future programmatic and evaluation decisions. These include evaluating the long-term impact of programs through follow-up assessments, incorporating additional indicators for measuring and assessing potential behavioral change and knowledge retention, and assessing the comparability of programs’ impact on different forms of terrorism across different ideological spectrums. Finally, further consideration should be given to determining the appropriate approach, strategy, and goals for terrorism prevention activities based on the audiences they are geared toward. Based on the evaluation findings, in some cases it may be beneficial to explore and test efforts aimed at incorporating terrorism prevention activities and programs within a broader portfolio of public health and violence reduction-focused efforts. Future programmatic, research, and policy-oriented activities might consider further exploration of the benefits and potential issues associated with doing so in addressing radicalization to terrorism within the United States

Washington, DC: U.S. National Institute of Justice, Office of Justice Programs, 2024. 33p.

By Arianna Vedaschi and Chiara Graziani

While states and terrorists have always used emerging technology in their endeavours, there has seldom been an emerging technology with the reach, implications, and possibilities of AI. In this masterful book, Vedaschi and Graziani skilfully merge law, computer science, psychology and more to provide the authoritative account of how AI enables terrorist actors, promises security, and challenges the rule of law.’

Cheltenham, UK; Northampton, MA: Edward Elgar, 2025. 168p.

Edited by Morten Bøås, Gilad Ben-Nun, Ulf Engel, and Kari Osland

Precarious living conditions across the Balkans, the Middle East, and North Africa create fertile ground for radical ideas. Yet, despite genuine grievances and legitimate grounds for anger, most people living in these regions are not radicalized and do not embrace ideas that lead to acts of violent extremism. Which raises the question ... why?

To answer this question, the authors of Resisting Radicalization investigate the nonoccurrence of violent extremism in what they term enabling environments. Their work, the result of a multiyear international project, has critical implications for the future of P/CVE (Preventing and Countering Violent Extremism) programs.

Boulder, CO: Lynne Rienner Publishers, Inc., 2025, 293p.

By Claudia Lanza, Abdelkader Lahmadi, and Jérôme François

This book presents the development of a classification scheme to organize and represent ransomware threat knowledge through the implementation of an innovative methodology centered around the semantic annotation of domain-specific source documentation. By combining principles from computer science, document management, and semantic data processing, the research establishes an innovative framework to organize ransomware data extracted from specialized source texts in a systematic classification system. Through detailed chapters, the book explores the process of applying semantic annotation to a specialized corpus comprising CVE prose descriptions linked to known ransomware threats. This approach not only organizes but also deeply analyzes these descriptions, uncovering patterns and vulnerabilities within ransomware operations. The book presents a pioneering methodology that integrates CVE descriptions with ATT&CK frameworks, significantly refining the granularity of threat intelligence. The insights gained from a pattern-based analysis of vulnerability-related documentation are structured into a hierarchical model within an ontology framework, enhancing the capability for predictive operations. This model prepares cybersecurity professionals to anticipate and mitigate risks associated with new vulnerabilities as they are cataloged in the CVE list, by identifying recurrent characteristics tied to specific ransomware and related vulnerabilities. With real-world examples, this book empowers its readers to implement these methodologies in their environments, leading to improved prediction and prevention strategies in the face of growing ransomware challenges.

Boca Raton, FL: CRC Press, 2024. 113p.

By Norman Abrams

During the protests that roiled campuses throughout the United States (2023-2025), many students and student organizations have extolled the Hamas organization and praised its violent attack on Israel on October 7, 2023.

Since 1997, Hamas has been officially designated under U.S. law as a “foreign terrorist organization.” Providing “material support” to a foreign terrorist organization is a serious crime under the federal criminal code.

The fact that Hamas is a designated foreign terrorist organization and occupies a central place in the student protests raises two key questions: Does praise and advocacy in support of a foreign terrorist organization amount to “providing material support” under the relevant federal criminal statutes? Would attaching criminal penalties to such utterances violate the First Amendment?

To date, the federal statutes that criminalize the providing of material-support-to-a-foreign-terrorist-organization have been reviewed by the U.S. Supreme Court only once, in Holder v. Humanitarian Law Project (2010). The main burden of this article is to examine those statutes and the Holder decision and its legal and practical implications for the student protest statements.

In light of the foregoing, recommendations are made: a) whether there is a legal basis and a need for a federal law enforcement investigation, and b) the general form it might take. These recommendations reflect concerns both about the possibility that a foreign terrorist organization is manipulating U.S. campus discourse and maintaining campuses as bastions of free speech, debate and tolerance of diverse opinions.

Keywords: Student protests, demonstrations, encampments, campus free speech, First Amendment, foreign terrorist organization, material support offenses

Journal of National Security Law and Policy (forthcoming), UCLA School of Law, Public Law Research Paper No. 25-18, 31p.

By Jessica Davis

Illicit actors in Afghanistan, including drug traffickers, warlords, terrorist groups, and even former government officials, exploit the country to achieve their own political and economic objectives. Historical and contemporary sources demonstrate that there are patterns in how these actors raise, manage, store, move, and obscure money. This paper provides a historical and contemporary overview of illicit financing activities in Afghanistan. It uses a terrorist financing framework to explain the various mechanisms involved in how illicit actors raise, use, move, store, manage, and obscure their funds. Specific jurisdictions used for illicit finance and global financial vulnerabilities that illicit actors with a nexus to Afghanistan exploit in their financial activities are discussed, outlining the threat-agnostic capabilities that could tackle some of these illicit financial challenges.

To raise funds, illicit actors engage in the production and trafficking of narcotics, taxation and extortion activities, illegal mining and timber production, and any other activities that can generate revenues. Depending on the actor in question, management of the acquired funds might be centralised, within the purview of a financial head, or decentralised, with control vested within several stakeholders. Illicit actors store some of their funds in cash within Afghanistan and use hawalas and banks to both store and transfer wealth out of the country. Since the Taliban takeover, international sanctions have largely ended the ability of banks to transact with most aspects of the global financial sector, with the exception of regional banking relationships that remain intact. Therefore, in addition to hawalas, common methods of moving money out of Afghanistan include bulk cash couriers, the transfer of precious metals and stones, and trade-based money laundering schemes. Wealth is moved out of Afghanistan to several key jurisdictions. Much of it makes its way to the United Arab Emirates (UAE) where funds are invested in businesses and real estate. Funds are also moved to Pakistan, Turkey, and to a lesser extent Iran and other neighbouring or proximate countries.

The stability of illicit financing mechanisms over time and across illicit actors presents opportunities for detection and disruption. However, this also applies to limitations of disruption opportunities. Since illicit actors in Afghanistan raise most of their funds within the country, there are few opportunities to disrupt internal Afghan revenue sources in the post-August 2021 context. As such, other detection and disruption opportunities that can be used to combat illicit financing with a nexus to Afghanistan must be examined. The policy options elaborated in this report include: enhancing the monitoring of aid and donor funding entering the country, facilitating the adoption of foreign asset and beneficial ownership tracing, introducing reforms to the hawala and banking sectors, and addressing deficiencies in international sanctions regimes.

A regionally coordinated policy approach to counter illicit financing from Afghanistan can reduce the ability of illicit actors to access their wealth and fund their activities. However, without access to a cooperative government in Afghanistan, there are limits to what can be achieved. Nonetheless, many of the policy and disruption opportunities available to counter illicit Afghan finance also serve to strengthen the integrity of the global financial system, address existing policy gaps in other jurisdictions, and can generally serve to enhance international cooperation on counter-illicit financing.

SOC ACE Research Paper No 11.

Birmingham, UK: University of Birmingham, 2022. 32p.

By The Counter Extremism Project, Global Strategies to Combat Extremism

In recent years, the U.S. sanctions regime has targeted the world’s most egregious human rights violators. It has done so based on the 2016 Global Magnitsky Human Rights Accountability Act (the Magnitsky Act) and the 2017 Executive Order (E.O.) 13818, which empower the U.S. government to impose “tangible and significant consequences on those who commit serious human rights abuse or engage in corruption, as well as to protect the financial system of the United States from abuse by these same persons.” Since then, the White House has used sanctions to target government organizations, officials, corporations, and private individuals engaged in the systematic violation of human rights around the globe. Such sanctions have been enacted against entities in Europe, the Middle East, Asia, and beyond. Unfortunately, the crimes committed against the Yemeni people by the Houthi rebels have gone largely overlooked. Fewer than 15 Houthi officials have been sanctioned in total, and the majority of those targeted are military officials who have been sanctioned for their role in destabilizing Yemen (E.O. 13611), with minimal attention paid to Houthi violations committed against the population under their rule. This report will focus on the Houthis’ systematic abuse of women and journalists. While these are tragically common phenomena in Houthi-controlled areas, they remain poorly documented. In addition to clearly documenting these abuses, the report will also map out the Houthi mechanics of repression: from the security forces responsible for unjustly arresting and harassing vulnerable individuals to those officials in the justice and prison systems responsible for subjecting the victims to torture and sham trials. The aim of this report is to raise awareness of Houthi human rights violations and provide informational support for efforts to penalize those who can and should be targeted with sanctions for their crimes.

New York: The Counter Extremism Project, Global Strategies to Combat Extremism, 2023. 16p.

By Laura Winkelmuller-Real, Kacper Rekawek, Thomas Renard

How do counter-terrorism services assess the threat in Europe? What trends do they notice? What are their most pressing concerns? Experts and observers regularly discuss the evolution of the terrorist threat, but what do counter-terrorism professionals have to say about this? This report aims to bring to the fore the perspectives of those that are in the first line to assess or respond to terrorism. Compared with some other initiatives focused on key terrorism trends (such as terrorism databases or indexes),1 this report focuses more strictly on a counter-terrorism perspective. While some reports on counter-terrorism already exist, they are largely drafted by and for governments. This includes notably the assessments conducted by the UN Counter-Terrorism Executive Directorate (UN CTED), which are unfortunately not in the public domain. In Europe, Europol’s TE-SAT annual reports constitute a very useful publicly available resource, although very much focused on a law-enforcement perspective. This report constitutes, therefore, an innovative attempt to monitor counter-terrorism efforts, with a focus on European Union (EU) Member States. It focuses on original sources, such as reports published by security services, a treasure of information largely neglected by scholars. It covers various dimensions, namely: 1. National threat levels (related to terrorism and/or extremism); 2. Discussion on the main types of terrorist and extremist threats perceived; 3. National security contexts in which the aforementioned threat levels and threat assessments were produced; 4. New legislative efforts designed and introduced to help CT and P/CVE efforts; 5. Main challenges to CT or P/CVE activities encountered by the EU MS; 6. Major CT operations conducted in Europe; 7. Main efforts in countering the financing of terrorism by different EU MS; 8. Key developments with regard to terrorism sanctions and proscriptions; 9. Communication and transparency efforts related to CT and P/CVE; To conduct this research, the International Centre for Counter-Terrorism (ICCT) compiled and analysed more than 70 official counter-terrorism and security reports published by the EU Member States (MS) in 2024 and 2023, which it combined with a survey distributed among European counter-terrorism officials. More details on the methodology of this report are available in the appendix. By combining practitioner perspectives gathered through the survey with detailed, structured narratives from national security documents, the report captures both operational realities and strategic priorities.  . These insights were used to develop specific themes for analysis, and to effectively describe the state of terrorism and counter-terrorism in Europe, as described by security services themselves. Overall, this report, while combining new material with data from the aforementioned secondary sources, offers original insights into the secretive world of counter-terrorism and highlights interesting trends. Furthermore, it could serve as a basis for further investigations into counterterrorism efforts, in Europe or elsewhere, including with a view to their being evaluated.   

The Hague: The International Centre for Counter-Terrorism (ICCT) 2025. 34p.

By Sydney Litterer, Ryan Scrivens,Thomas W. Wojciechowski &Richard Frank

Researchers, practitioners, and policymakers are increasingly interested in examining online posting behaviors in virtual communities known to facilitate violent racially and ethnically motivated extremism. However, little is empirically known about how such behaviors develop over time, and even less is known about how the content of posts is related to other posting behaviors. This study used group-based multi-trajectory modeling to explore how users’ online posting behaviors (i.e., posting frequency and use of offensive language) evolved as they engaged with other users on Stormfront, the largest and most well-known white supremacist forum, relative to typical user behaviors. Overall, several noteworthy posting behaviors were identified in the data. We conclude with a discussion of the implications of the analysis, its limitations, and avenues for future research.

Behavioral Sciences of Terrorism and Political Aggression, 2025, 21p.

By Arthur Bradley and Galuzzi, O.

The abuse of digital technologies by violent extremists is keeping pace with the exponential growth of new technologies, and poses multifaceted challenges to national and global security. Cyber-enabled threats manifest for example in terrorist-operated websites, the shift to alternative or fringe social media platforms, the use of the decentralised web, the exploitation of gaming and adjacent platforms, and the abuse of live-streaming technologies to amplify terrorist and violent extremist attacks. In addition to these online activities, there are concerns also around more disruptive or destructive cyber operations, such as Distributed Denial-of-Service attacks and the hacking of critical infrastructure to cause civilian casualties. In all the research on the diverse range of malicious actors behind these threats, there is comparatively little on the online activity of violent extremist movements, whether right-wing or left-wing, in the Global South. This report forms part of UNICRI’s effort to investigate the threats stemming from the complex interplay between terrorism, violent extremism and cybercriminality – threats that are often overlooked, owing to the difficulty of gathering evidence and attributing offensive cyber operations, and to the prioritisation of more pressing security threats in diverse geographic locations. UNICRI strives to shed light on the online presence, activities and trends of right- and left-wing violent extremist movements and the cyber-enabled threats they may consequently pose to global security. The report was compiled following a three-part research methodology consisting of a literature review, expert interviews and open-source investigations conducted in order to analyse the online activities of right- and left-wing violent extremist movements in South America, Africa and Asia, and examining both their intent and ability to mount offensive cyber-attacks. The report includes particular case studies within these regions, including in Brazil, South Africa, India and Maritime South-East Asia. The case studies were selected because of the availability of public information online, the known presence of active non-state violent extremist actors with right- and left-wing ideologies, the similarities and differences these actors present, and their geographic diversity. These factors, and consequently the choice of case studies, demonstrate the global nature of the phenomenon which still requires contextually relevant solutions. The selection of these case studies does not imply that similar threats in other geographies are not considered relevant to international peace or development, and conversely, the omission of any movements is merely the result of restricted resources and time. The groups and movements presented within this report are not necessarily referenced as violent extremists either by the United Nations or by the Member States mentioned, however, their alignment, proximity, and connection with right- and left-wing violent extremist ideologies, as well as their use of violent extremist tactics, justify mention in this report to ultimately reflect on the global dimension of the abuse of digital technologies by violent extremists . This report finds widespread exploitation of digital platforms by right- and left-wing violent extremists based in South America, Africa and Asia. Groups and their affiliated networks use a wide variety of platforms and services for a range of different purposes, and they often seem to face fewer restrictions in terms of content moderation by technology companies, many of which are based in the United States or European countries. In particular, it found: As in Europe, North America, and Australasia, the online activities of right- and leftwing violent extremist groups in South America, Africa and Asia are increasingly superseded by more disparate, horizontal online networks. In many of the case studies, for example in Brazil and India, physical attacks have increasingly been carried out by lone actors or small cells, some of which may have had previous engagement with organised groups. This dynamic has implications for the ability of technology companies and law enforcement agencies to counter the threat, as planned attacks and their perpetrators may be more difficult to prevent or identify. * Violent extremist networks and individuals are increasingly using a more diverse range of online platforms and services to further their goals. This is in line with the increase in the number of online platforms and services used by broader populations generally, but it may also be part of a concerted effort by these networks to reach a broad audience and mitigate the impact of the potential removal of their accounts or groups by technology companies. Violent extremist networks continue to exploit multiple platforms simultaneously, using outlinking between platforms to evade detection or enforcement by specific companies. Violent extremists comprise the minority of the perpetrators delivering cyber-attacks globally, most of which are believed to be carried out by state-backed actors, hacktivist collectives, or financially motivated criminals. Interviews with a group of 31 experts consulted as part of this research, however, indicate that the threat from cyber-attacks motivated by a belief system and delivered by individuals or groups affiliated with violent extremist movements is likely to increase in the coming years, and is likely to be particularly high in countries believed to have less developed cybersecurity defences. This report suggests that international technology companies are not adequately fulfilling their content moderation policies as consistently in South America, Africa and Asia as in other countries in Europe, North America, and Australasia. Also, they do not appear to be allocating sufficient resources to ensuring platform safety in these regions, where they face significant challenges in effectively countering the exploitation of their services by violent extremist movements. Practical challenges are compounded by definitional challenges regarding contentious terms such as “violent extremism” and “terrorism”, neither of which has an internationally agreed definition. Also, technology companies, it seems, still struggle to detect and understand violent extremist content or communications effectively in languages other than English. This task is made more difficult by the challenge of interpreting and understanding local dynamics and the community-specific slang found in content, and by the efforts of malevolent networks to evade detection or enforcement by moderation teams. Evidence suggests that, to date, this – together with an imbalance in resource allocation – means that the capability of many technology companies to moderate content in languages other than English is comparatively ineffective. Often, a splintered regulatory landscape also makes it difficult for technology companies to apply their policies consistently across multiple jurisdictions around the world. Technology companies operating globally are subject to a variety of differing and often contradictory regulatory requirements, including those relating to designations, hate speech legislation and Internet-related laws, and companies can be under pressure from the political or cultural contexts in particular countries. This can make it difficult for these companies consistently and effectively to maintain a balance between removing violative content and upholding human rights and fundamental freedoms. The report focuses on a set of case studies diving into the online activities of right- and left-wing violent extremist groups in South America, Africa and Asia, and the ways in which they abuse digital technologies

Turin, Italy: United Nations Interregional Crime and Justice Research Institute (UNICRI) and the VOX-Pol Institute. 2025. 88p.

By Mohamed Bin Ali and Rafillah Rapit

The increasing involvement of women in terrorism challenges traditional perceptions and necessitates a shift in counterterrorism strategies. Recent cases in Singapore illustrate how women are radicalised through online propaganda, familial influence, and ideological indoctrination, leading to their participation in extremist activities. Addressing this evolving threat demands gender-sensitive prevention and rehabilitation efforts, community engagement, digital counterterrorism, and tailored intervention programmes to counter radicalisation at its roots. COMMENTARY The recent detention of a 15-year-old self-radicalised female student in Singapore highlights the growing vulnerability of young individuals, especially women, to extremist influences online. Her case, the first involving a female minor under the Internal Security Act (ISA), signals an alarming shift in radicalisation trends, where even adolescents are being drawn to violent ideologies. This underscores the urgency of addressing radicalisation at its early stages, particularly in digital spaces where extremist propaganda thrives. Just months earlier, the case of a 56-year-old radicalised housewife demonstrated that women are no longer confined to passive roles in extremist networks. These developments challenge traditional assumptions about gender and radicalisation, reinforcing the need for targeted intervention efforts across different demographic groups.

In many cases, women were often involved as facilitators of terrorism, providing logistical support, acting as recruiters, or serving as symbols of ideological purity. However, recent trends show a growing number of women assuming leadership roles, engaging in combat, and even carrying out suicide attacks. Several factors drive this shift, including ideological indoctrination, social and economic disenfranchisement, and extremist organisations' strategic exploitation of gender norms. Terrorist groups such as ISIS, Boko Haram, and Al-Qaeda have actively sought to recruit women, recognising their potential to evade security measures and access targets that male operatives might struggle to reach. Extremist groups have used women as suicide bombers, assassins, and enforcers of their doctrine. In some instances, they have also taken on prominent roles in radicalisation efforts, using online platforms to spread extremist propaganda and recruit new members.

S. Rajaratnam School of International Studies, NTU Singapore 2025. 4p,

Edited by Gary LaFree and Joshua D. Freilich

Much has changed in the relationship between terrorism research and the field of criminology in recent years. Law and justice scholars such as Kittrie (1978) and Turk (1982) made significant contributions to the research literature on terrorism and responses to terrorism in the past half-century, but much of this early work was isolated, piecemeal, and done without federal research funding. It took the Murrah Federal Building bombing in Oklahoma City in 1995 and, more importantly, the September 11, 2001, attacks in New York City, Washington, DC, and Pennsylvania before major funding to support social science research on terrorism in the United States became available. After these events, funding through the National Memorial Institute for the Prevention of Terrorism; terrorism research solicitations by the National Science Foundation, the National Institute of Justice, and the Bureau of Justice Administration; and the creation of the Department of Homeland Security’s academic Centers of Excellence program all strengthened support for social scientific studies of terrorism and responses to terrorism (for a review, see LaFree & Dugan, 2016). More recently, the Minerva program, funded by the Department of Defense (DOD, 2015), and the Domestic Radicalization to Violent Extremism research program, funded by the National Institute of Justice (NIJ, 2014), have further increased levels of support.

Wiley Blackwell, 2017, 620p.

By Wesam Charkawi , Kevin Dunn ,  Ana-Maria Bliuc

Since the September 11 attacks, prevention and countering violent extremism (P/CVE) programs have rapidly increased worldwide, garnering significant interest among researchers. This paper is a systematic review focusing on the evaluations of primary, secondary and tertiary prevention programs from 2001 until 2020. The review identified 74 program evaluations that included satisfactory measures and metrics. Only 32% of the studies deemed the intervention successful, 55% described limited success, and 8% deemed the program had failed. Many of the programs evaluated failed to reach their objectives; some generated negative outcomes such as community disdain and an increase in the likelihood of alienation and stigma. Success was largely a self-assessed measure by the facilitators or stakeholders of the programs or the evaluators of the study. Success indicators can be operationalized as the degree of enhanced sense of belonging (connectedness to the community, social connection), trust and willingness to engage in programs, development of critical thinking skills (integrative complexity theory), and a strong sense of worth (quest for significance). Without a generally accepted set of metrics and no cohesive framework for conducting evaluations, this review offers an important addition to the field on the evidence suitable for program evaluations. An important aim of this systematic review was to identify what makes an effective and successful countering violent extremism program. The key findings indicate that enhancing belonging, identity, trust and community engagement, acknowledging perceptions of injustice, religious mentoring, and the promotion of critical thinking/self-reflection are associated with successful programs. The findings press upon policymakers, funders, and researchers the need to consider and support high-quality evaluations of programs.

International Journal of Law, Crime and Justice, 2024., 19p.

By By Stephane J. Baele,* Lewys Brace, and Elahe Naserian 

Scholars and practitioners investigating extremist and violent political actors’ online communications face increasingly large information environments containing ever-growing amounts of data to find, collect, organise, and analyse. In this context, this article encourages terrorism and extremism analysts to use computational visual methods, mirroring for images what is now routinely done for text. Specifically, we chart how computer vision methods can be successfully applied to strengthen the study of extremist and violent political actors’ online ecosystems. Deploying two such methods – unsupervised deep clustering and supervised object identification – on an illustrative case (an original corpus containing thousands of images collected from incel platforms) allows us to explain the logic of these tools, to identify their specific advantages (and limitations), and to subsequently propose a research workflow associating computational methods with the other visual analysis approaches traditionally leveraged  

Perspectives on Terrorism, Volume XIX, Issue 1 March 2025  

By Stephane J. Baele and & Lewys Brace

How to best chart and analyse extremist digital ecosystems? This paper proposes to complement standard mapping methods based on URL outlinks, which are typically deployed but present several critical flaws, with socio-semantic network analysis. Adapting bi-nodal socio-semantic network principles to the specificities of extremist and terrorist digital communications, we put forward a simple yet efficient method for generating informative networks based on ideological or thematic proximity rather than URL connections. We use three datasets of various sizes and nature (French far-right websites, British far-right websites, US far-right Telegram channels) to compare traditional URL-based vs. socio-semantic networks, demonstrating how the latter bypasses the flaws of the former and offers significant advantages in multi-methods research designs. This empirical validation of our methodological proposition unfolds several new observations concerning the contemporary Western far-right digital ecosystem, highlighting specificities and commonalities of its French, British, and American variants.

By Luke Baumgartner

On July 22, 2011, 32-year-old Anders Behring Breivik embarked on a threehour murder spree across Norway, killing seventy-seven in what would become the single largest loss of life in the country since the Second World War. Beginning that morning in central Oslo, Breivik’s terrorist attack claimed its first victims after detonating a van filled with nearly one ton of explosives outside of the Regjeringskvartalet complex, which housed the prime minister’s administrative offices. Disguised as a police officer, Breivik then boarded a ferry for the island of Utøya, where he proceeded to kill sixty-nine students attending a youth worker’s party summer camp in a mass shooting that lasted nearly ninety minutes. An hour before the bombing, Breivik emailed his manifesto, 2083: A European Declaration of Independence, to more than 1,000 contacts, asking for their assistance in disseminating his magnum opus, which, according to him, took more than three years to complete. The 1,500-page compendium contained a litany of grievances and rails against political correctness, feminists, and “multiculturalist/cultural Marxists,” ultimately accusing them of complicity in a grand conspiracy orchestrated by political elites of the European Union and Arab states to destroy Western civilization through the gradual replacement of white Christian Europeans by way of increased Muslim immigration—in other words, a Great Replacement. Breivik’s actions that day would inspire a cascade of copycat attacks in the United States, New Zealand, and Germany. As part of their attacks, each perpetrator wrote a manifesto to explain the rationale for their actions. In doing so, the manifestos can–and often do–serve multiple purposes: air the personal and societal grievances that led them to kill, provide a blueprint for future attackers, and perhaps, most importantly, gain the public notoriety and infamy they so desperately crave. At its core, terrorism is an act of violence that seeks validation through its very nature as a public spectacle, and thus, manifestos can provide answers to the ever-present question of why an event of this magnitude occurs. While these attacks are separated from one another in both time and space, the ideological glue binding them together is the belief in a racially and culturally homogenous dystopian future, one in which white people gradually cease to comprise demographic majorities in traditionally white dominions such as the United States and Europe. Or, in the most dire circumstances, cease to exist at all. These manifestos contain several common themes central to replacement theory–the conspiracy that motivated their attacks. This paper attempts to build on the current body of academic literature that focuses on the thematic elements of the manifestos and the historical and theoretical foundations upon which the attackers’ justifications for their actions lie. Beginning with a detailed history of replacement sentiment in the United States and Europe, this section seeks to provide the necessary background and context for where replacement theory comes from and how it has motivated actors across time and space. Within the context of the United States, replacement theory finds its ideological roots in late-19th and early-20th century race science, or eugenics, beginning with the works of sociologist Edward A. Ross and anthropologist Madison Grant, both of whom popularized the notion of “race suicide.” Subsequent generations of post-war white supremacists, such as David Duke and David Lane, transformed race suicide into the explicitly anti-semitic conspiracy theory of “white genocide,” laying the groundwork for contemporary militant far-right extremists to coalesce around the idea of replacement theory. Meanwhile, European manifestations of parallel racist conspiracies emerged during an intellectual movement of the 1960s in France known as the Nouvelle Droite, championed by the likes of Jean Raspail. Raspail’s work would influence leading public figures such as Italian author Oriana Fallaci and British Historian Bat Ye’or, who were instrumental in sparking Renaud Camus’ coining of the “Great Replacement.”

Washington, DC: Program on Extremism at George Washington University , 2025. 32p.

By Leland Lazarus and Alexander Gocso

This paper provides a comprehensive analysis of the Chinese individuals, gangs, and companies engaging in illicit activities in Latin America and the Caribbean. Our methodology was to research academic literature, news articles, press releases, official statements, and podcasts in Spanish, Portuguese, Mandarin and English, as well as conduct off-the-record interviews with U.S. and LAC intelligence and law enforcement officials to ascertain growing trends in Chinese criminal behavior in the region.

Miami: Florida International University, 2023. 33p.

By Phil Williams and Sandra Quincoses

The economic and political environments in Latin America have been advantageous for local, regional, and transnational threat networks. Specifically, technology, increased international trade and economic interdependence, heightened interest in natural resources for profit, synthetic drug production, economic disparities, corruption, impunity, and unstable political conditions have led to a complex web of opportunities that requires new, progressive ways to address criminal activities. The creativity of threat networks along with their entrepreneurial strategies have resulted in increasing power and influence. Despite efforts by the United States and some governments in Latin America to combat these networks, the everchanging global environment has worked in their favor. Indeed, some countries in Central and South America are in danger of transforming into what Jorge Chabat described as “criminally possessed states.” Furthermore, gangs in Central America, especially in Honduras where MS-13 has become more closely linked to drug trafficking, have reduced local extortion, become more aware of their nascent political power and have even engaged in rudimentary social welfare provision. Another major trend identified in this report is the strategic diversification of trafficking routes, activities, and markets. For instance, licitly established transpacific trade between East Asia and Latin America has been exploited by criminal groups involved in wildlife and drug trafficking. This, along with other activities such as illegal logging and mining, have undermined licit trade. Moreover, criminal groups have also conducted cyber-attacks against government agencies to obtain false documentation to illegally conduct logging activities in protected areas of Brazil’s Amazon rainforest. Such activities illustrate the growing relationship between technology, illicit behavior, and criminal groups’ diverse capabilities. An emerging nexus between state and non-state threat networks is also identified as a key trend in the region. Notable links between criminal groups and malign state actors such as Cuba, Venezuela, and Nicaragua are publicly known with indigenous and under-resourced communities violently confronting illicit organizations that sometimes have the tacit support of government authorities and agencies. Moreover, external state actors’ investments and economic interests in highly corrupt and unstable political environments pave the way for impunity, which enables threat networks to operate without inhibition. The most explicit example pertains to the Revolutionary Armed Forces of Colombia (Fuerzas Armadas Revolucionarias de Colombia—FARC)-linked mining operations in Venezuela. Undoubtedly, trends are becoming increasingly complex now that activities have reached a global scale. This is also evident in illicit financial flows into, through, and out of the region. In terms of money laundering, tried and tested methods still seem to be favored, although it is likely that new opportunities, such as those provided by digital currencies, will be increasingly exploited in the future. The paper concludes that, in spite of adverse trends, it is important to avoid worst-case analysis. It also suggests, however, that many of the problems in the region stem fundamentally from poor governance—and that taking steps to deal with this should be a priority.

Miami: Florida International University, Jack D. Gordon Institute for Public Policy, 2019. 45p.

By Lea Gerster, Francesca Arcostanzo, Nestor Prieto-Chavana, Dominik Hammer and Christian Schwieter

As part of its project on “Combating Radicalisation in Right-Wing Extremist Online Subcultures”, ISD is investigating smaller platforms to which the German-speaking far-right online scene is retreating. Their aim is to circumvent regulation and moderation on large platforms, for example as required by the German Network Enforcement Act (NetzDG, or 'Facebook Act'). Analyses contained in the report on “Telegram as a Buttress” have already made clear the importance of investigating PeerTube. While writing the previous report, the ISD research team came across multiple video platforms with almost identical layouts and functionalities. It was found that eight out of 19 video platforms identified were set up using the free software PeerTube. PeerTube is an example of a growing socio-technological movement that attempts to turn away from large, centralised platforms towards decentralised and mostly community-managed websites. Instead of a single platform with a monopoly on content, this movement is building a network based on servers that are maintained independently of one another. This is leading to a “hydra effect”: even if connected servers are cut off, the network itself survives, allowing for new servers to be added at will. While far-right extremists are not the ones driving this phenomenon, it does appear that they are exploiting these new-found possibilities. For example, various far-right figureheads have established platforms via PeerTube where users can in some cases create their own accounts. Some of these PeerTube platforms record millions of visits per month. But it is not just its reach but also its structure which makes PeerTube worth investigating. With PeerTube, individuals or organisations can create their own video platforms where they set the rules for content, moderation and user registration. This is essential for far-right and conspiracy groups and individuals who, according to previous ISD research, prefer audiovisual platforms to other types, such as micro-blogging services. PeerTube is a particularly valuable tool since it is more technologically demanding to host and access audiovisual materials than text files. In contrast to centralized platforms such as YouTube, PeerTube content is managed separately by so-called instances - this refers to the small-scale video platforms created with PeerTube software. Different instances can network with each other and form federations. This allows videos that have been uploaded on one instance to be played on another instance without having to change the website. PeerTube belongs to the so-called Fediverse, which will be discussed in more detail below. Another difference to centralized video platforms is that this software uses peer-to-peer technology (P2P), which presumably explains the name. The fact that instances are not managed by large companies, but by individuals or groups at their own expense and with the help of free software, also has implications for their regulation. Key Findings • There is no central moderating authority for managing content. PeerTube offers individuals or groups, whose content has been blocked on centrally managed video platforms for to violating the terms of service, an attractive way of continuing to share their content online. Where it is these groups or individuals who control moderation, content can only be removed from the network by switching off the server. • It is difficult to accurately map the size and connections of the Fediverse. The network is constantly in flux as the relationships between instances can change rapidly due to blocking and follow requests. Instances can also go offline from one day to the next. • The instances used by far-right and conspiracy actors make up only a small fraction of the Fediverse network. They primarily network among themselves. However, some of the instances that were investigated are connected to a wider Fediverse through their own highly networked servers. • The deletion of extremist YouTube channels is not necessarily reflected in the number of account registrations on the corresponding PeerTube instances. PeerTube is rather used as a back-up option for deplatforming. The frequently observed phenomenon that not all users migrate to the new platform can also be observed here. • Instances are often customised in different ways. For example, each instance varies in whether they permit third parties to register for accounts and upload videos. There is no clear correlation between the numbers of accounts, videos and views. However, the most watched videos on relevant instances were mostly created by prominent members of the milieu, which would appear to indicate that persons with a pre-established audience are particularly successful on PeerTube. • The instances selected for the five case studies hosted a lot of content that focused on the COVID-19 pandemic. Another frequent narrative was an alleged conspiracy perpetuated by elites who, according to conspiracy theorists, use events such as the pandemic or Putin's war against Ukraine to further their secret agenda. These findings suggest that PeerTube instances provide safe refuge for disinformation. • Because PeerTube is a piece of software that anyone can access and use in a variety of different ways, state regulation will do little to limit its use by farright extremists. While state agencies can enforce individual aspects of the NetzDG against PeerTube instances, many instances do not have the number of users required to impose reporting obligations or requirements to delete content. Moreover, most content is not hosted with the aim of generating profit, which limits the applicability of both the Facebook Act and the EU's Digital Services Act. • However, PeerTube's community moderation function does allow the community to moderate the use of PeerTube for promoting harmful content. One way of doing this is by isolating extremist instances. Efforts should be undertaken to work with the Fediverse community, i.e. with the server operators and their users, and to develop best practices for identifying and combating extremist activities. This could include further training on how to spot hate speech or setting up a body for reporting extremist instances.

London: Institute for Strategic Dialogue (2023). 44p.

By Gary LaFree

In recent years, nutrition researchers have found that ice cream may have as many health benefits as low-fat milk or yoghurt for those with diabetes or at risk of diabetes. Nonetheless, they have resisted reporting this finding to the media, the public, or other researchers. This observation got me thinking about how preconceived assumptions and biases affect social science in general and criminology in particular, are affected by the preconceived assumptions and biases of those who produce them. In this essay, I argue that the production of criminology is a cultural enterprise that reflects the attitudes and values of those who produce it. In my address for the Stockholm Prize, I summarize the main thesis of my Losing Legitimacy book and then discuss two recent projects that were influenced by the idea that strong social institutions reduce criminal behavior. The first examines the impact of the death of Freddie Gray in Baltimore in 2015 on crime and arrest rates. The second examines the impact of globalization on national homicide rates. In both cases, the results were unexpected.

Policy Implications

As scientists, it is critical that we evaluate research based on its theoretical soundness and methodological sophistication rather than whether it fits a currently hot topic or politically popular perspective. Science progresses by the constant process of evaluating theoretical propositions with empirical data—regardless of where those data lead us. Openness about crime and reactions to crime is no less important than honesty about the positive benefits of ice-cream consumption.

Criminology and Public Policy, 2025, 19p.