In previous blog posts, we discussed Jeremy Weinstein’s work on the internal dynamics of rebel groups, and how they lead some groups to commit indiscriminate violence against civilians. In a nutshell, groups with limited initial resources are forced to establish close relationships with civilian communities in order to survive, which forces them to discipline their forces and share significant power with communal leaders.

By contrast, groups with significant wealth at their founding (for example, due to state sponsors or involvement in the drug trade) have no strategic imperative to depend on civilians, and furthermore tend to recruit personnel who are in it for the money. As a result, personnel tend to abuse civilians, the leadership doesn’t want to incur the costs of disciplining them, and the groups tend not to share power with the civilian populace.

We are now at the crux of it. The foregoing processes tend to encourage “rich” rebel groups to use massive violence against civilians, mostly because their previous mistreatment of civilians leaves them with no other options. Let’s see why.

Rebel groups want civilians in (or near) their territory to cooperate with them—to provide food, tax revenue, recruits, and information about government troops and collaborators. Civilians, on the other hand, may or may not want to cooperate. Some might be government supporters or officials. Even if civilians oppose the government, they may not want to risk government reprisals. And they might view the rebels as worse than the government, and not want to cooperate with them even if they could do so safely. Therefore, rebel groups (and governments, for that matter) will sometimes want to harm civilians who cooperate with the enemy or refuse to cooperate with them—not least in order to frighten other civilians into complying with their demands.

A rebel group that has close ties with civilian populations (usually because it began its existence as “poor”) will have a much easier time using violence in a selective, targeted fashion. Because the populace trusts them, civilians are more willing to give information to the rebels. And because the rebels have close ties to the populace, they will be able to vet information they receive to make sure that their informants are telling the truth, so that they don’t harm an innocent party by mistake. Punishments are usually more graduated (such as kidnapping civilians and confining them for a time), giving rebels the chance to discover a mistake before harm becomes irreparable (i.e. the wrong person is shot). Finally, when mistakes are made, the rebels usually make amends to the populace and punish the offending personnel, reinforcing the trust that the population has in them.

As a result, “poor” rebel groups will tend to use violence selectively against civilians, seizing or assassinating government officials and collaborators and rarely harming the wrong people. The overall level of violence against civilians will be fairly low (at least from the rebel side; often government forces are less discriminate, for the same reasons we are about to discuss with reference to “rich” groups).

In contrast, we discussed how “rich” groups will tend to abuse civilians because they don’t bother disciplining their troops, and they will tend to exclude civilians from power arrangements. As a result, civilians will tend not to trust such rebel groups, even if they nominally support them over the government (and they may not). Rebel groups will thus receive less information from civilian sympathizers, making it harder for them to selectively target government collaborators or functionaries, or to punish civilians who are refusing to cooperate with them.

Worse, when they do receive information from civilians about potential targets, “rich” rebel groups will have a hard time verifying its accuracy (if they even care to). As a result, malicious civilians will frequently exploit the rebel groups to take revenge against their neighborhood enemies. Even without such deliberate deceit, rebels will frequently target the wrong people, ending up harming innocents. This will cause civilian trust to erode still further and causing information flows to slow or stop. In the end, even if rebel groups wanted to target civilians selectively, they will find it impossible.

But such groups still have a strategic need to force compliance by civilians. Unable to use violence selectively, they will instead resort to collective punishment, massacring people at random or even whole communities in order to frighten other communities.

Obviously this is a suboptimal outcome for the rebels, even setting aside moral concerns. Once you murder people indiscriminately, it becomes almost impossible to go back as no civilians will trust you or want to help you. “Rich” rebel groups are thus set on a path to continued massacre and bloodshed that ends only when they establish unchallenged control over a given community or population. (And even then, the pervasive acts of individual exploitation will continue.) Their ability to gain popular support will be very much hobbled, and their effectiveness in challenging the government and ruling the populace will be significantly less than it might have been.

Still, it’s not impossible to overthrow the government and rule a country while murdering indiscriminately. (Charles Taylor comes to mind in Liberia.) Less dramatically, in Mozambique, RENAMO managed to bring the ruling government to the bargaining table after a very long and bloody civil war.

In the final post of this sequence, we will discuss how external shocks can challenge rebel groups’ ability to operate, and how their responses to such shocks might change their pattens of behavior from “activist” to “rich” or (rarely) vice versa.

*****

(This post is part of Politics for Worldbuilders, an occasional series. Many of the previous posts in this series eventually became grist for my handbook for authors and game designers, Beyond Kings and Princesses: Governments for Worldbuilders. The topic of this post belongs in the planned fourth book in this series, working title War for Worldbuilders. No idea when it will be finished, but it should be fun!)