One of my long-term research endeavors concerns developing a computerized conflict early warning or alert system (CAS). An approach and design for such a system was published in Conflict Management and Peace Science CMPS 1998 paper-Finding Harbingers. However, in the process of making the design, I came to the conclusion that a lot more basic research has to be done before my proposed system could be implemented. My research since that realization thus focuses on answering basic research questions. The projects below should be viewed in that light.
Taxonomy of Violent Conflicts
An important roadblock to a CAS in my opinion is that we need to better distinguish between different types of conflict so that we can talk about the causes or early warning signs for a particular type or class of conflict. I think the situation we are in is roughly analogous to the situation in cancer research in the 1960s. At that time, different types of cancers were only crudely distinguished from each other if they were distinguished at all. An effort to classify cancers has subsequently contributed to advances in finding causes and treatments for cancer. Hoping to help conflict research in a similar manner, my largest research project entails creating a taxonomy of violent conflicts roughly analogous to Kingdom, Phylum, Class, Order… in zoology and botany. At the core of that project is building a Conflict Catalog, a database of all recorded violent conflicts since 1400 AD in which 32 or more people were killed. A sample of that scale is necessary to create a robust taxonomy.
Conflict Catalog Data
The Conflict Catalog is available as an Excel (.xlsx) file. A short document describing the Conflict Catalog can be found here. An additional spreadsheet provides an extension of the Conflict Catalog that goes back to 900 AD for Europe.
Findings from the Conflict Catalog
The long time scale and large sample size of the Conflict Catalog has made it possible to discern patterns in the nature of violent conflict that were not feasible before. Two papers, one delivered at the 1999 Peace Science Society Meeting and one at the 2001 Uppsala Conference on Conflict Data present findings from the catalog.
Power, International System Structure, and Violent Conflict
In the course of creating my Conflict Catalog, a colleague challenged me with, “Peter, this all sounds nice and impressive, but I would like to see how this work can be applied to advancing theory.” That, and the desire to have a good example for showing how to relate evidence to theory in my Empirical Research Methods and Causes of War classes, led me to examine the question of whether the structure of a system of states (countries) indeed affects the amount of violent conflict in the system, as is argued by the structural realist school of international relations theory. Namely, does it matter if you have a system with one dominant country, with two contending major powers (countries), with three major powers, or with more than three major powers? To answer that question, I ended up having to develop a dataset containing a power index (a measure of the relative power) of the most powerful countries in Europe (and to a lesser extent elsewhere) from 1494 to 1945. Now that the dataset has been developed, I am applying it to other concepts of relative country power such as power parity, power transitions, and power cycles and how they relate to violent conflict.