Understanding individual behaviors within covert networks: the interplay of individual qualities, psychological predispositions, and network effects

Abstract

This article theorizes about how individual factors and network effects interact with each other in ways relevant to the study of networks generally, but in particular of criminal networks. In modern network analysis, careful technical descriptions that involve important graph-theory measures are entirely sensible, but they often ignore specific details about the individuals within the network. For study of a human social system, to ignore qualities of the actors is to risk an incomplete, possibly spurious, explanation, so individual-level factors may be important for a more complete understanding of the system. In covert and criminal networks, actors have motivations to keep some activities from public view, so it is impossible to understand such networks without appreciating at least that individual-level intention. This article describes five different levels of effects, both individual and relational, relevant to network-based social systems, and explains how these effects may interact. Important implications for the study of criminal networks include the formation of trust within networks, the exercise of control, and the identification of network brokers. A richer description of individual action within a complex social system will require better knowledge about how personality, social identity and other psychological factors are distinct from, and yet may interact with, self organizing network processes.

Appendix: Glossary of some common network terms used in this article

Actor: typically, an individual who “acts” within a social environment; more generally, a social entity that comprises the nodes of a network. In a criminal network, the actors typically are the criminals.

Broker: A network broker connects otherwise disconnected parts of the network. In a criminal communication network, an intermediary between two groups who do not otherwise communicate is a network broker. Brokers occupy structural holes, the “gaps” in the network where certain parts of the network are disconnected.

Centrality of an actor: The “importance” of an actor to a network. There are various types of centralities, two of which are degree centrality (with the most central actors being those with the highest degrees) and betweenness centrality (where the most central actors are those who keep the network connected). In a criminal communication network, criminals with the highest degree centrality are those who have the most communication partners, whereas those who have the highest betweenness centrality are those whose absence would most likely disconnect communications across the network.

Clique: a subset of actors, all of whom are tied to each other.

Connected: Two actors are connected if there is a path from one to the other. A network is disconnected when some actors are not connected to other actors.

Degree: In a nondirected network, the degree of an actor is the number of ties in which that actor is involved; in a directed network, the indegree of the actor is the number of ties received, and the outdegree, the number of ties sent. In a nondirected criminal network relating to communication, the degree of an actor is the number of other criminals with whom the actor communicates. In a criminal trust network, the indegree of an actor represents popularity in terms of how many others trust that actor, whereas the outdegree represents activity (sometimes, termed expansiveness) in the sense of how many others the actor trusts.

Degree distribution: The distribution across the whole network of the number of actors with given degrees. In a criminal communication network that was highly centralized, with some criminals having many communication partners and many with few communication partners, the distribution would be both skewed and bimodal, with a few high degree nodes, and many low degree nodes. For directed networks, there are both indegree and outdegree distributions.

Density of a network is the proportion of observed ties to possible ties.

Directed/nondirected network: A network may be directed in that an actor expresses (or sends) a tie towards another actor (who receives it), or nondirected (or undirected) where there is no directionality in the tie. In criminal network studies, examples of nondirected networks might include alliance (i.e. two criminals might be allied to each other); examples of directed networks might include threat (i.e. one criminal might threaten another.)

Dyad: a pair of actors and the relations between them.

Geodesic: The shortest path between two actors is a geodesic, the length of which is the geodesic distance (taken to be infinite if the pair of actors is disconnected, i.e. without a path between them.) In a criminal communication network, the geodesic distance between two criminals i and j is the smallest number of communications by which i can communicate with j. If i and j are tied, then this distance is 1; if not, but they can communicate through one intermediary k, then the geodesic distance is 2.

Geodesic distribution: The distribution across the whole network of geodesic distances.

Graph: a mathematical object used to represent a network, comprising a set of nodes or vertices, representing actors, and edges (lines) representing ties. A graph can be drawn as a network visualization.

Network: comprises a set of actors (individuals) and a set of relational ties among them.

Path is a sequence of connected ties from one actor to another; the length of the path is the number of ties in it.

Reciprocity: In a directed network, a reciprocated (or mutual) tie occurs when ties both from i to j and from j to i are present in the network.

Relational tie: a social connection between actors. Different types of relational ties express different types of social connections (e.g. advice, communication, trust, acquaintanceship, friendship, hatred.)

Structural hole: See broker.

Triangle: a clique of three actors.