This is not a project (yet) but something I thought about while walking through my neighborhood (Capitol Hill in Washington DC), which I love-fear in similar parts, because it is beautiful, walkable, friendly and inviting… if only they’d stop shooting out behind our house every half year and breaking into houses, snatching purses from pregnant women, or treating our car as a neighborhood car share (we had it stolen and returned twice).
We have hard working, dedicated beat cops who are out there walking the neighborhood rain (snow) or shine and I know that each and everyone of them has a wealth of knowledge about the good, the bad and the ugly around here. One problem though is that everyone only has a part of the puzzle and this kind of knowledge is difficult to share in it’s whole complexity.
So, what I was thinking: Imagine we sat down for a Net-Mapping session with everyone responsible for the area or issue when there is the next string of car thefts or burglaries. We would write down the names of everyone we could think of having even the remotest connection to the issue, victims, potential perpetrators and anyone who might be helping in the background. We could color code the actors, having one color for victims and on the side of potential criminals one color for those who have been convicted of crime before and those who have not. Or, if it is a case of, e.g. gang related violence, we could have different colors for the members of different gangs.
To get the general structure of this network, we would start with one general link of “social or family connection”. Then the more interesting part would be the links for known and suspected criminal dealings between the actors. I think for getting the whole picture and digging as deep as possible into the areas where there is suspicion but no proof yet, it is important to have different kinds of links for “criminal interaction that has been proven and prosecuted”, “criminal interaction that we are sure of but couldn’t prosecute yet” and “suspected criminal interaction”. It would make sense to have them in the same color, a strong line for the prosecuted interactions, broken line for the ones we are sure of and dotted for suspected ones.
If the Net-Map is about some specific crime or string of crimes, the influence towers could be used to indicate how likely the police officers think that each actor is involved in this crime. If it is more about a general longer term situation (e.g. gang violence), we could have two colors of towers, one for influencing that the bad situation continues/escalates and one for influencing that it gets better.
The biggest benefit of doing this, would be that it allows the officers to share their in-depth system knowledge of the community with each other in a structured way and that they would together draw the bigger picture that would allow them to focus on the core actors.
I know there has been quite a bit of social network analysis applied to tracking international terrorists. But to my knowledge most of that has been done with quantitative analysis and by tracking directly traceable interactions such as money flows or emails. Now imagine you could sit down with some of the top anti terrorism experts and make them discuss the links as they draw them… But for now I’d be absolutely fine if we could do something about the car share situation in our own back yard. If someone were willing to try this out in our own neighborhood, I would even be willing to facilitate it for free…