There exist several conditions under which indiscriminate victimization of civilians fulfils this interdiction/denial function. First, indiscriminate violence is more effective the smaller the underlying civilian population. Why, for example, was the indiscriminate violence employed by the British Army in South Africa so effective at defeating the Boer insurgency, whereas German victimization of non-combatants in the Soviet Union backfired and increased the size and strength of the partisans? One answer is that indiscriminate violence is more effective when the civilian population that the insurgents draw their support from is small[...]The smaller the population, therefore, the more plausible it is to kill or completely isolate it, and thus the more effective large-scale civilian victimization can be.
Second, indiscriminate violence in guerrilla wars is likely to be more effective the smaller the size of the geographic area comprising the theatre of battle. The logic is simple: If people have nowhere to run, then it is easier to hunt them down and kill or intern them. [...] It also helped that the population in question only consisted of about 100,000 people. Similarly, in the Philippines (1899–1902), American troops relied on concentration camps to dry up the water in which the Filipino insurgents were swimming, a task aided in no small measure by the fact that the fighting occurred on small, isolated islands.
Third, indiscriminate violence is more likely to be effective – or less likely to backfire – when the insurgency is denied sanctuary and is cut off from external sources of supply. Population concentration and/or large-scale killing is best conceived of as a brutal method of ‘draining the sea,’ preventing the insurgents from obtaining need supplies from civilians. If the rebels have sanctuary across a border in another state, or an open supply line from a third party, draining the domestic sea can only partially sever the insurgents’ ability to obtain supplies and continue to fight. [...].
Finally,
indiscriminate violence is likely to be more effective than selective violence when the population is solidly committed to the insurgents’ cause[my emphasis]. Kalyvas argues that
all civilians – no matter what their prior ideological commitments, if any – give their support to the actor whose coercive threats are most credible [my emphasis], that is, the actor who controls the area militarily. Yet we know this to be untrue empirically, at least in some cases.[...] When a group of civilians is strongly committed by virtue of ethnicity or ideology to an insurgency, targeted violence may be insufficient to deter it from supporting the guerrillas. Indeed,
selective violence may only infuriate group members and make them more determined to resist. In such situations, rare though they may be, massive applications of violence in support of a strategy of rendering the group incapable of supporting the insurgents might prove more effective.[my emphasis] Indiscriminate violence against civilians can be effective in defeating guerrilla insurgencies under certain –
relatively restrictive – conditions[my emphasis]. When the population from which the guerrillas draw support is relatively small, the land area in which the insurgents operate is similarly constricted, and external sanctuary and supply is not available, governments have been able to strangle rebel movements with indiscriminate violence. [...] Killing (or imprisoning) the population is not the only way – and certainly not the most desirable way from a normative point of view – to defeat a guerrilla insurgency. Targeted violence, as noted above, is more efficacious in circumstances where the incumbent is not able to control the whole population physically. Less violent methods can also be successful when an insurgency is very weak. (culled from pp.438-440)
Bookmarks