Murder by Structure
Source: http://www.hsph.harvard.edu/departments/society-human-development-and-health/files/a.p.art.2009.pdf
The above link goes to a paper titled, “Murder by Structure: Dominance Relations and the Social Structure of Gang Homicide.” I apologize for the length of the article, but definitely try and read the abstract, intro and scan through the article for the accompanying diagrams of the network structure that results from gang homicides.
This paper explores the relations between gangs in Chicago and the network of murders that result from their jockeying for dominance over one another. Because the paper is so lengthy, it is not possible to cover every facet of the paper in this one post, but I will concentrate on some interesting traits about the network structure of gang homicides in Chicago.
In class we studied basic networks and how edges and nodes can be many different things. Nodes can be companies that are connected through business interactions, people connected by friendship, or gangs connected by murders. In this paper Andrew Papachristos (the author) has built a network with the nodes being Chicago gangs and the edges being the murders between them. This would be an example (albeit an extreme one) of a negative edge between two nodes. By constructing this network, Papachristos is able to understand, at the incident level, the structural motivations for the murders. Instead of attributing “blame” for the murder to social conditions including poverty, stratification, etc… (as is done in other sociological papers), Papachristos examines the network to understand murder in the context of action and reaction dominated by violent social norms that dictate a need to display dominance and react strongly to all challenges to territory and prestige. When looking at the network structure, different aspects are striking. Some components of the network are dominated by single entities. The Latin Kings, a hispanic gang, assumes a central role in gang violence both committing and falling victim to a large portion of the murders taking place in the network. In this sense the Latin Kings are a hub because of the large amount of ties that run to and from it in the network. It is also a cut point because if one were able to take it out of the network, many other gangs (or nodes) would become disconnected from the murders of actors (gangs) in the network. This gives insight not only into the structure of gang warfare, but also into the possible ways that law enforcement could go about combatting gang violence. Rather than targeting all gang violence as one aggregate problem, it might make more sense for law enforcement officials to focus on just one more central gang (such as the Latin Kings) because if they can successfully dismantle this gang, it would isolate many of the other smaller gangs in the network and therefore not only stop murders committed by Latin King members, but also murders that would have been directed at Latin King members.
This is also a directed network (murders are not mutual, one person murders another person.) When looking at the structure another important trait is that many edges depart from and arrive at the same node. This indicates that a member of a gang murdered another member of the same gang. The prevalence of this situation (it can be seen with several gangs, and with some gangs it occurs at an alarmingly high rate) shows the danger that members should feel not only from rival gang members, but from fellow gang members as well. The amount of in-violence (as opposed to out-violence, which would be a murder directed at a rival gang member) within a gang is striking because one would think that in joining a gang, you would only have to protect yourself against rival gangs. Instead, a gang member must not only keep a wary eye on other gangs, but also on his/her own gang members. The laws of loyalty and brotherhood that usually accompany such stringent groups do not seem to protect their members from the murderous violence of one another.
The study is not a static depiction of gang violence in Chicago. The data collected spans across time, showing different networks at different intervals of time, Papachristos shows that the network largely retains its structure, even when the gang members change. This is interesting because it shows that it is not just individual motivations that fuel murder, but the actual structure of the network itself. Gang membership is very temporal due to the fact that many members are either arrested, killed, or move away from the gang as they get older. Despite the fact that gang membership duration is short, the relations between whole gangs remain largely intact over long spans of time. This perpetuation of the network demonstrates the power a network can have over individual agency.
Lastly, the way in which murders spread in the network paints a picture far different from just one actor killing another. The spread of murder takes on the traits of an epidemic. As one murder occurs, it sets of a chain reaction in the network that results in many more murders across the murder network. Murder, therefore, spreads like a social contagion, but rather than a fashion trend changing what people may wear on their feet, this contagion results in the deaths of many young gang members.