Since the the CSTPV at St Andrews is organizing a symposium on "Terrorism Research: Past, Present and Future’", I thought I would make public my views on the topic. My main argument is not that provocative: throughout terrorism studies, the theories, models, concepts, and data used should be improved. My second argument, however, is that the best way to accomplish this might be through a research program -- a topic I will develop in a subsequent post.
Instead of binary thinking (rich/poor, religious/ideological, etc.), there should be more interest in variation within a concept and comparison between similar concepts. In other words, concepts such as religion, ideology, poverty, violence, etc., should be used in a more refined and precise manner and placed within a larger classification or typology. Similarly, any modeling of causation needs to include actors such as competing terrorist organizations, insurgent groups, criminal organizations, but also international organizations, states and their various agencies, civil society (NGOs, religious institutions, the private sector, the media, arts and entertainment), tribes, and especially the family and friendships.
These models could then be improved with several levels of analysis (international, regional, state, metropolitan, organizational, group, psychological, neurological) and the related emergent properties and theories thereof. For example, think of the effects on terrorism of factors such as armed conflict or peace; climate; education, health and income; constitutional and criminal law and regime type; social norms and values; collective memory and family history; personal religious and political beliefs, all of which vary across time and space.
Once all of these concepts and models are clearly established, it would be useful to seek to understand what types of social relationships and behaviors are stable and reoccurring and which are changing (either deliberately or not). This would require focusing on specific regions of the world, specific historical periods or specific types of terrorism before combining this knowledge at a higher geographical or conceptual level or in a larger historical period and theorizing interaction effects and path-dependency. For example, once all of this is done, research could try to model the spread of global trends and fashions in terrorism tactics (highjacking, suicide bombing, kidnapping, beheading, etc.) across space and time.
One way of organizing such progress would be through a research program, possibly based on scientific/critical realism and complexity theory, but I think that would not be very popular, for a number of reasons. Without such a program, however, the progress in terrorism studies is likely to remain piecemeal.