Organized crime is particularly challenging to define due to the multitude of groups, sectors of activity, and objectives pursued. In May 1988, at an international conference on organized crime, the INTERPOL international secretariat proposed the following definition for the first time: “Any association or group of individuals engaged in ongoing illegal activity with the objective of profit, regardless of borders.” The United Nations has defined organized crime as “a structured group of two or more individuals who act together over a limited period with the aim of committing one or more high-level offenses for the purpose of obtaining financial or material gain, directly or indirectly.”
The variety of these definitions, including one from the Council of Europe, highlights the difficulty in characterizing a dynamic and polymorphic concept. Therefore, the strategy for combating organized crime must consider adaptive and proactive analytical approaches. Criminal activity encompasses a diverse range of sectors, including human trafficking, extortion, arms trafficking, vehicle smuggling, cultural property theft, burglaries, environmental offenses, drug-related crimes, banking fraud, and various scams.
Addressing organized crime, which is in a constant state of change, requires adopting an offensive stance based on spatial, temporal, and structural analysis. Whereas organized crime groups were highly hierarchical just a decade ago, many now adhere to rules more akin to chaos, necessitating an adjustment from security forces. This chaos, characterized by a flattening of hierarchy, generates polysemic criminality that relies on both local delinquency and specialized crime, utilizing traditional methods alongside new communication technologies. The focus has shifted from executing a specific “hit” to undertaking mass attacks, which allow for lower individual profit while also decreasing risk. Profit is generated through the multiplication of offenses. This hierarchical flattening allows for more creatively isolated, sometimes violent, criminal forms that are harder to control. While law enforcement has particularly evolved in retrospective investigative techniques—especially through the advancement of technical and scientific police work—the current emphasis is on developing proactive methods via crime intelligence that are based on a scientific foundation, integrating criminological components and judicial aspects.
This triptych is essential for an analytical approach to mass treatment, interpreting data within a global criminogenic context, and appropriately judicializing intelligence. The aim of this article is to present various analytical forms that offer new perspectives in the fight against organized crime. The approach is grounded in anticipation-oriented intelligence, utilizing spatial and temporal crime analysis, as well as network analysis—the very foundation of criminal groups.
Spatial Analysis in Combating Organized Crime
Since crime is not a random phenomenon, the location of criminal acts meets criteria that can explain the notion of geographical concentration, the criminal trajectory of an offender or group, and potential future crime zones. The essence of spatial crime analysis is to fulfill this triple objective. This form of analysis is particularly vital in the context of organized crime, where transnational dimensions and the increasing mobility of groups necessitate an anticipatory approach to avoid constantly lagging behind and diminishing overall effectiveness.
Spatial analysis is nomothetic and relies on a number of observed facts that must be localized precisely and attributed to the same criminal group. Because organized crime disrupts the statistical cycle of local crime—which remains relatively constant over time and space—methods for detecting and delineating geographical areas encompassing changes in criminal forms involve supplementing common aggregation functions (sum, average) used in Spatial Online Analytical Processing (SOLAP) with other functions such as range and standard deviation. SOLAP techniques are particularly fitting, especially when considering discrete data. To enhance the geographical visualization and continuity of criminal trajectories, a kernel density estimation algorithm is employed, particularly in producing maps of criminal concentration (hot spots).
Such methods not only allow for the detection of changes in criminal forms and thus the mechanisms of organized crime but also enable the consideration of possible diffusion processes of criminal forms across broader territories. In addition to spatial analysis, temporal visualization also presents opportunities for anticipation.
From Temporal Analysis to Crime Anticipation
As organized crime increasingly exploits local delinquency to remain below detection thresholds, law enforcement must also direct their actions toward this delinquency to impact organized criminal groups. The reactive strategy has proven limited in the face of constantly evolving crime, making temporal analysis for anticipatory purposes particularly relevant in countering criminal threats.
“Forecasting is projecting into the future what we have perceived in the past,” stated Bergson over a century ago. This encapsulates the aim of the adopted predictive approach. By detecting trends of increase or decrease and/or seasonality from the years 2008 to 2013, it becomes possible to construct a predictive model validated against data from 2014 and projected into 2015. These methods allow for anticipating burglaries quantitatively as well as formally. While predicting the exact number of burglaries in March or June 2015 is undoubtedly utopian, forecasting a quantitative trend and an evolving direction is a highly relevant objective in the context of an anticipatory approach.
Complementing time series analysis, regression methods can explain and anticipate criminal phenomena based on external variables. Given that organized crime capitalizes on the dynamics of socio-economic degradation, it is useful to exploit external variables to describe correlations between socio-economic data and offenses, and to act upon these links to provide predictive elements. This allows for anticipating the consequences of rising or falling delinquency by varying certain factors, such as the average salary of residents in a region, the number of second homes, or the population between the ages of nineteen and twenty-five. Such analysis can prevent the emergence of specific criminal forms by addressing the socio-economic factors of a territory.
Predictive models founded on time series analysis or regression approaches bring real added value by enabling quicker, more objective, and cost-effective decision-making. They become genuine decision-making tools, particularly in the face of polysemic crime. Such analysis, exemplified through burglaries, can also be applied to armed robberies, assaults, or vendettas—essentially any offense from which lessons can be drawn from the past.
Combating Crime Through Structural Analysis of Criminal Networks
In addition to spatio-temporal analysis, structural analysis of networks serves as a decision-support tool. This approach is both graphic and analytical, consisting of extracting elements of understanding from a cartographic representation of interconnected entities. Entities can be of various types: individuals, organizations, groups, telephones, computers, etc. From these entities, the goal is to provide an overarching view of the network by highlighting the most influential elements within the structure. Thus, the focus is not on the individual entity but on the network as a whole, thereby enhancing the resilience of interventions over time.
Among the various methods of structural network analysis, graph theory offers promising prospects. M. K. Sparrow, a specialist in mathematical network analysis and a former member of the British Police Service for a decade, emphasized in 1991—when social media did not exist—the need to surpass mere description of criminal structures to focus on their analysis through the lessons of network science. Numerous possibilities exist to characterize a network (shape, density, links) and to identify the key individual within a structure (links, centrality).
Regardless of whether a network is star-shaped, circular, or anarchic, its form influences reactivity, robustness, and resilience capacity. The density of a network corresponds to the ratio of existing links among actors relative to the total possible links. High density can be extremely dangerous due to the multitude of links that may render the structure less controllable and also more vulnerable. Generally, such networks operate with low hierarchical weight but possess high reactivity. Conversely, a low-density network may indicate a structured group with an effective protection and safeguarding policy, leading to networks that are very difficult to penetrate deeply, such as transnational criminal groups.
Links within a network can be familial, cultural, environmental, virtual, or professional. Measures of centrality (L. C. Freeman) offer insights for explaining and predicting individuals’ roles, their distances, their leadership, and the network’s robustness following a disruption, facilitating objective targeting. Notable centrality measures include degree centrality, betweenness centrality (C. Morselli), eigenvector centrality, and articulation centrality (M. K. Sparrow).
In facing a new criminal order that encompasses both high and low spectrums of globalized, multifaceted crime—abandoning traditional mafia-like structures—the means of combat must undergo a significant transformation by emphasizing the notion of anticipation. Sharing information, analyzing structures, and predicting threats appear to be essential combinations in the fight against organized crime. Police cooperation, as well as collaboration with other armed forces, is now crucial to tackle transnational criminality that finds true opportunities for development within the current geopolitical context. The political instability of many countries, the challenging international economic situation, and the plethora of conflict zones contributing to the proliferation of firearms create particularly favorable conditions for criminal activities. Thus, combating organized crime, while leveraging scientific contributions, offers new perspectives for security forces, enabling a shift from a reactive to a proactive posture. Given the constantly increasing speed of communication methods and the advent of big data, it is inconceivable today not to adopt an anticipatory approach in the face of criminal groups that are ever more quick to shift their activities.
Bibliography
Justin Gosling: “The Global Response to Transnational Organized Environmental Crime: A Research Report,” The Global Initiative, Geneva, 2014. INTERPOL: Against Organized Crime: Interpol Trafficking and Counterfeiting Casebook, Lyon, 2014. Phil Williams and Roy Godson: “Anticipating Organized and Transnational Crime,” Crime, Law and Social Change, Kluwer Academic Publishers, 2002. Jeffrey Scott McIllwain: “Organized Crime: A Social Network Approach,” Crime, Law and Social Change, Kluwer Academic Publishers, 2000. Patrick Perrot: “Analyzing Criminal Risk: The Emergence of a New Approach,” Revue de l’Électricité et de l’Électronique, 2014. Patrick Perrot and Kader Tedj Achi: “Forecasting Analysis in a Criminal Intelligence Context,” The International Crime and Intelligence Analysis Conference, Manchester, 2015. Malcolm K. Sparrow: “The Application of Network Analysis to Criminal Intelligence: An Assessment of the Prospects,” Social Networks, 1991. Linton C. Freeman: “Centrality in Social Networks: Conceptual Clarification,” Social Networks, 1979. Carlo Morselli: “Inside Criminal Networks,” Springer Publications, New York, 2009.
