A system dynamics model of insurgencies is built using the U. S. Army and Marine Counterinsurgency Manual (FM 3-24) as a basis. It must, however, be supplemented by additional theory from outside sources to enable calibration to a historical dataset. Parameter and policy analyses are conducted. These highlight the criticality of some features of insurgencies described by FM 3-24 such as the importance of obtaining and maintaining popular support and employing sufficient counterinsurgency forces to gather intelligence. Other features, not highlighted by FM 3-24, are also shown to be important such as the rate at which the number of violent incidents committed per insurgent increases during the insurgency and how easily detained/eliminated insurgents can be replaced. A number of sensitivity and policy tests are conducted. One finding, which supports conventional wisdom, is that the timing of withdrawal of counterinsurgency forces is critical. In particular, if the withdrawal is too early, the insurgency may end up being worse than if no counterinsurgency forces had ever been deployed. A second result is support for the contention that counterinsurgency policies are synergistic, which implies that policies are most effective if deployed together. Finally, contraindicating the expectations of FM 3-24 and other conventional wisdom, results from optimization simulations suggest that counterinsurgency forces should in many cases initially emphasize intelligence-gathering efforts and minimize direct combat action against the insurgents. Only later in the counterinsurgency effort should counterinsurgency forces reverse this policy and ramp up direct combat action against the insurgents. Copyright (C) 2011 System Dynamics Society.
