Organised crime and gangs are contested concepts notoriously difficult to define and a source of long-standing, often fierce, debate among scholars and practitioners alike (see, among others, Levi 1998; von Lampe 2001; Paoli 2002; Varese 2010; Campana 2011; Decker and Pyrooz 2014; Kleemans 2014; Hobbs and Antonopoulos 2014; Antonopoulos and Papanicolaou 2018; Campana and Varese 2018). Very few scholars, however, would disagree on the fact that organised crime – and gangs – possess some degree of coordination both among its members and with non-members, regardless of how the concept may be defined. Coordination is an inherently relational phenomenon as it involves establishing a relation between at least two individuals with the intent of achieving certain goals. Similarly, few scholars would disagree on the fact that gangs and organised crime groups can be involved in violence, be it for exerting control over a territory, keeping competitors at bay or punishing non-compliant customers or accomplices. No matter the motivation, violence is inherently relational as it involves establishing a (violent) relation between at least one offender and one victim (see Campana 2023a for further discussion; also Black 1990 and 2004 for a relational intuition in the study of violence). Furthermore, as organised crime groups – and to some extent gangs – are involved in illicit economic activities, they generate flows of money that often need to be laundered before being reinvested in the legal economy. Flows of dirty money are inherently relational as they are made of relations connecting, for example, individuals, bank accounts, companies or countries. Flows of illegal commodities, e.g. drugs, the sharing of resources or the movement of organised crime members across places are further examples of inherently relational phenomena in the organised crime space – and beyond. Bringing relations to the fore, then, can greatly enhance our understanding of organised crime, gangs and illicit markets more generally.

Relations can be conceptualised, operationalised and systematically analysed through the lenses of social network analysis (SNA). As Borgatti et al. (2013: 1) put it, networks are “a way of thinking about social systems that focuses our attention on the relationships among the entities that make up the system”. This special issue on crime and networks seeks to advance our relational understanding of organised crime, gangs and crime more generally by presenting six works exploring a diverse set of questions related to organised crime, street gangs, outlaw motorcycle gangs, money laundering and groups operating in cyberspace.

The first set of three works explores issues related to gangs and organised crime in three different settings: the United States, Australia and the United Kingdom. In ‘Crossing lines: Structural advantages of inter-racial criminal gang violence’, Bichler and Norris leverage publicly available court files to study street gang violence in Los Angeles. They focus on a yet under-explored aspect of gang-on-gang violence, namely conflict patterns across ethnicities. They show that such conflicts are driven by a desire to increase market “efficiency”, meaning an attempt by gangs to place themselves in positions that offer competitive advantages vis-à-vis the other gangs operating in the same economic environment. They nicely draw comparisons – and borrow concepts – from theories of organisational competition within commercial settings, most notably the work of Ronald Burt on the social structure of competition.

Bright, Sadewo, Cubitt, Dowling and Morgan explore ‘Co-offending networks among members of outlaw motorcycle gangs across types of crime’ in New South Wales (NSW), Australia. They use data obtained from the NSW Police linked with data from the NSW Bureau of Crime and Statistics Research and NSW Corrective Services to create an integrated dataset on the relational (and non-relational) characteristics of 5,513 individuals affiliated with 23 outlaw motorcycle gangs (OMCGs). As OMCGs are rather unique in possessing both a formal structure and an informal structure linked to criminal activities, they explore the former vis-à-vis the co-offending network structure exhibited by (some of) their affiliated members. They show that illegal activities tend to be conducted in small cliques as clubs within clubs. At the core of such networks, they found a greater prevalence of non-office bearer affiliates, pointing to a divergence between criminal cooperation and formal structures in most – but not all – outlaw motorcycle gangs. As they rightly note, it remains to be ascertained whether the less central roles played by office bearers is the result of a strategic decision to minimise the risk of detection by placing themselves at arm’s length from criminal operations or they are genuinely marginal in the club’s illegal operations (while occupying a central role in its social operations). Bright et al. nicely show the importance of keeping in mind the limits of each relational source of data, in this case co-offending records, when formulating research questions and interpreting results.

In the third paper on the topic, ‘Organised Crime Movement across Local Communities: A Network Approach’, Campana and Meneghini propose a novel network-based methodology to study the movement of organised crime group members across locales based on the idea that two areas (communities) are linked if an offender has committed a crime in both. They apply this approach to police records from Cambridgeshire, UK, to identify ‘turf’ and ‘target’ areas and then explore the determinants of movement between such areas. They show that organised crime group members target urban communities with higher-than-average illegal market opportunities as well as the effect of socio-demographic homophily between turf and target communities, suggesting that organised crime members might target communities that are similar to their own. This novel methodology can offer a way to identify communities at risk of being targeted by organised crime groups; it is scalable and portable across contexts and can be extended to study long-distance movement and the movement of offenders beyond organised crime groups.

This special issue then continues with two papers looking at money laundering from two different angles. In ‘Money laundering as a service: Investigating business-like behaviour in money-laundering networks in the Netherlands’, Kramer, Blokland, Kleemans and Soudijn leverage a novel impressive dataset encompassing police intelligence-driven contacts of 198 financial facilitators involved in drug-related money laundering. They show that contrary to what is commonly assumed, money launderers are indeed linked through extensive networks. The largest connected component (the so-called ‘giant component’ in network terminology) is surprisingly large, encompassing 60% of financial facilitators involved in drug-related money laundering. While individual money launderers might appear peripheral in specific police investigations, the integration of several police investigations over a longer period – Kramer et al. looked at 628 police cases between 2016 and 2020 – shows that (i) a relational structure connecting launderers exists, and (ii) not all money launderers are peripheral, nor they occupy the same position. Crucially, by integrating tasks and structure, the authors were also able to show that degree centrality, i.e. the number of connections money launderers have established with other money launderers, vary depending also on the type of financial services provided with underground banking, financial advice and real estate possessing the highest degree. As money laundering is made of flows of money as well as information on laundering opportunities, Kramer et al. compellingly explored the betweenness centrality of launderers, i.e. their potential for being a broker. They discovered that not all high-degree launderers are also important brokers – the latter tend to be involved in financial advice.

Costa looks at the ‘Nexus between corruption and money laundering: deconstructing the Toledo-Odebrecht network in Peru’ by leveraging publicly available data from two connected high-profile investigations conducted in Brazil and Peru (Lava Jato and Ecoteva, respectively). This work reconstructs the ‘financial infrastructure’ that sustains corruption by extracting and systematising a very long list of different types of relations ranging from kinship to financial flows and co-participation in meetings, parties, associations as well as structures of management and control of legal entities (e.g., owners, boards of directors or shareholders). The paper nicely brings together a network picture of the financial infrastructure with an analysis of roles, rules and mechanisms of informal governance underpinning such structure – the latter derived from what Costa defines as “network ethnography”. The paper maintains that the informal governance system is “well-functioning” thanks to a balanced mix between centralisation and decentralisation: a core group of central actors sets rules, inputs and outputs while delegating the day-to-day operations to a decentralised set of managers. This study convincingly shows that, when adopting a network approach, we can move beyond “the traditional idea of a bribe as a form of dyadic exchange between two individuals” and identify more complex structures underpinning – and sustaining – cases of ‘grand’ corruption.

Finally, in ‘Cybercriminal networks in the UK and Beyond: Network structure, criminal cooperation and external interactions’ Lusthaus, Kleemans, Leukfeldt, Levi and Holt use information from ten, non-publicly available, law enforcement closed investigations into groups of offenders with a financial motivation operating in cyberspace. Based on unique data from cybercrime investigations in the UK, this article explores how cybercriminal networks are organized, how cybercriminals operate, and how they interact with their environment. The authors show how network concepts can fruitfully inform a qualitative analysis of official records to tease out organisational arrangements that give rise to specific network patterns, e.g. clusterisation, internal cohesion (or lack thereof) and the presence of specialised components. The comparison of these ten cases by Lusthaus and colleagues resulted in a wide range of interesting findings on cybercriminal networks, including: a common division between the scam/attack components and the money components; the presence of offline/local elements; and a broad spectrum of cybercriminal behaviour and organisation. Importantly, these cases indicate that the boundary between cyber-dependent crimes and cyber-enabled crimes is often blurred.

This special issue advances the current scholarship on crime and networks in at least five different ways. Firstly, it enriches the body of works exploring areas in which a relational approach has so far received very limited attention. These include the study of large-scale corruptive practices, the structure of money launderers and the organisation of groups operating in cyberspace. It also offers a novel network-based empirical approach to model the movement of organised crime members (and groups).

Secondly, it shows the versatility of a network approach as it can be fruitfully applied to a variety of different sources of evidence ranging from openly accessible court files to police intelligence reports and investigations as well as large-scale administrative records, including police records and correctional services records. Each source of data comes, of course, with its strengths and limitations that are discussed separately in each paper.

Thirdly, it shows how networks can be integrated into mechanism-based approaches and be subject to hypothesis testing by relying on procedures that take into account the specificity of relational data, e.g. Quadratic Assignment Procedure (QAP), Exponential Random Graph Models (ERGMs) or Clustered Standard Errors (CSE).

Fourthly, it speaks to the importance of integrating quantitative and qualitative evidence to fully understand the context behind a network, the mechanisms underpinning it and the interplay between specific roles/tasks and network positions. The works presented in this special issue have shown the importance of combining structure and attributes to fully understand relational phenomena. As pointed out by Campana and Varese (2020: 157), “being in a position to add qualitative information in the form of actors’ attributes, as well as information about the mechanisms underpinning a given network, plays a crucial role in avoiding potential misinterpretations of the structure”.

Finally, it provides further evidence in support of the view that the most fruitful way of conceptualising networks is by following a network science approach, namely considering networks as “a finite set or sets of actors and the relation or relations defined on them” (Wasserman and Faust 1994: 20). This is what Campana (2016) terms as ‘instrumental’ approach to networks in opposition to a ‘substantive’ approach that views networks only as a specific form of organisation adopted by organised crime (as, e.g., in Williams 2001: 62, “there is a growing recognition that organized crime is increasingly operating through fluid network structures rather than more formal hierarchies”). This latter view is too limiting as it reserves the term ‘network’ to only one possible network configuration and limits the scope of a network approach to the study of organisations. The works presented in this special issue show that networks go well beyond the study of organisations and that empirical structures need to be derived rather than assumed (in line with Morselli 2009; Varese 2010; Campana 2016; Campana and Varese 2020; for further discussion, see Campana 2016, 2023a, b).

Fifteen years have passed since Klaus von Lampe edited an influential special issue in this same journal on Human capital and social capital in criminal networks (Von Lampe 2009) – a pioneering exploration into the then-emerging field of criminal networks. Since then, there has been a burgeoning interest in this field, which now looks more established and more mature theoretically, empirically and analytically. This special issue constitutes a further step into the development of a relational approach to the study of organised crime – and crime more generally. With the continuing increase in computing capabilities, in the availability of (large-scale) datasets as well as of new statistical models specifically developed for relational data, the evolution and expansion of the field is likely to continue – and we look forward to seeing what a special issue on crime and networks will look like fifteen years from now.