nal juvenile gangs, and illegal gun carrying. But there is almost no research on violence among suburban and rural youth inside or outside schools. Longitudinal research is needed to follow up the perpetrators of school shootings, both those who remain in prison and those who are released. In addition, case studies of other kinds of rampages involving adults are needed as a point of comparison with student rampage shootings. Finally, while some research has followed children who have been victims of school rampage shootings in the past, further longitudinal research is needed to understand the impact of school shootings on both perpetrators and victims, including secondary victims.
As the committee worked with both the cases and the data, it seemed that there might be two quite different strands of lethal violence. In one strand, the lethal violence seemed to emerge from a fairly specific conflict among individuals, in which all the parties to the conflict were aware of the problem and all the parties felt vulnerable to a physical threat or attack. It also seems clear that others outside the conflict—the audience— knew about it and understood its logic and patterns. Because this kind of violence seemed most typical of the inner-city school cases—as one would expect in communities in which violence had become highly prevalent— we began calling this the “inner-city form of violence.” In a second strand, the lethal violence also emerged from a conflict in which the persons who committed the offenses felt that they were under threat and needed to protect themselves. But there are some important differences in the nature of the conflict. In this second strand of violence, the conflict was much less specific and differentiated than in the first strand. The “beef” was not with a particular person over a particular issue; it involved some abstract idea of “the school” or “the other kids who don’t respect me.” There can be many particular individuals who embody this abstract enemy. A second important distinction between this kind of lethal violence and the first kind is that not all those who were potential targets of the violence knew they were involved in a conflict with the person who did the shooting. Because this seemed to be the pattern that was characteristic of the suburban and rural shootings we examined closely, we came to call this the suburban-rural pattern.
It is the committee’s strong view that additional research is necessary to investigate the question of whether there are two different strands of lethal school violence and, if so, whether they are correlated with urban on one hand, and suburban and rural on the other. If these two different patterns of lethal school violence exist, and if they are in fact correlated with the characteristics of the communities in which the violence occurred,