The Efficacy of Violent Resistance

Jadegold raised some interesting questions in a previous post. To summarize, the key points Jade made seemed to be:

• Armed resistance generally does not produce meaningful change; often it produces negative backlash resulting in greater harm to the insurgent/victim.
• Violent insurgencies almost never achieve their political goals because they fracture the base of support which they rely on. Non-violent insurgencies are more successful because they impact institutions which brings about reconciliation.

First, there are two levels to consider this question on: the individual and the group (or nation-state vs. non-state actor).

THE INDIVIDUAL

The best strategy is to foil a violent attack in the target selection or planning process. You don't want to be the softest most inviting target, and you want to complicate planning. But assume the worst has happened and you are a victim of violent crime. You have four basic options:

• Avoid the kill zone. If you can spot the ambuscade and avoid the encounter altogether this is preferable.
• Evacuate. You can find an exit and run for it.
• Dominate. You can aggressively attack, create pain/distance, and then seek to break contact with the attacker if they are still mobile.
• Submit. You can comply with the aggressor's demands.

Those are the basic options. Given that the attacker holds most of the cards because they chose the time, place, and method of the attack -- and probably have the benefit of surprise, as well -- the defender may be in a pretty bad tactical situation. Furthermore, unless you have armor (like being in a vehicle) or standoff (distance between you and the attacker), you have very limited time to make a quick decision.

The British Home Office looked at robberies where the victim was injured that most of the folks who were injured in robberies tried non-violent resistance:

• Resisting with a gun 6%
• Did nothing at all 25%
• Resisted with a knife 40%
• Non-violent resistance 45%

The DOJ National Crime Victimization survey found that "The probability of serious injury from an attack is 2.5 times greater for women offering no resistance than for women resisting with a gun. Men also benefit from using a gun, but the benefits are smaller at 1.4 times more likely to receive a serious injury."

Gary Kleck -- reprinted by the Nat'l Academy of Science -- found the following probabilities of being injured:

TABLE 5-2 Probability of Injury and Loss Among Victims by Means of Self-Protection

Method	Robbery		Assault
	Injury	Loss	Injury
With gun	12.8	15.2	27.9
All self-protection	34.0	52.8	58.1
No self-protection	23.6	83.6	55.2
All incidents	30.2	69.9	57.4
SOURCE: Adapted from Kleck (2001b:289, Table 7.1).

Using violence in self-defense is not always the best option. If there is a possibility to avoid the attack or give the attacker what they want, that's probably best. I'd prefer to give up my wallet than pull out a gun. However, I think there are certainly some situations in which submission has ceased to be a viable option.

• If an attacker wants to put you into a car and take you somewhere else, or take you out of a public place, your odds of survival will drop significantly.
• If you are on a hijacked airplane in the post 9/11 world, then your odds of survival with passive compliance are not great either.
• If you are put into an execution position (on your knees) then your options have just narrowed substantially.
• If you are bound then you will lose many options as well. It also implies that the perp has further (nefarious) intentions.
  

Seriously; if anyone argues in the post-9/11 world that one should comply passively with hijackers on an airplane then you need to think hard about what possible outcomes are next; these days you aren't likely to take a trip to Cuba until someone hands the hijackers a suitcase full of cash. The plane is going to be crashed into a target or you're going to be shot down once you get close to a city.

Clearly there is a time and place for the use of sudden violence in self defense. I have not seen any convincing evidence that suggests that women who passively submit to rapists are less likely to be injured by their attackers, for example. And looking at the list of options, there aren't many choices; what if you can't avoid the attack (already in the kill zone) and can't run away (no exits/mobility)? Violent resistance -- preferably armed -- may be the best remaining option on a short menu of bad choices.

THE GROUP

There seems to be some interesting research that suggests that non-violent insurgencies have enjoyed greater success in the last century or so than violent uprisings. I am unsure about the causal link, however.

For example, it is possible that insurgents turn to violence as a last resort, just like individuals. That is, they don't break out violence until the situation is already near hopeless. An example of this would be the Warsaw Jews. They couldn't run (how are a quarter of a million civilians including women, children, elderly, and sick people going to escape in Nazi Germany?), they couldn't avoid the situation (they were already being shipped off to death camps, literally), and submission wasn't really a viable option (time wasn't on their side; every day of compliance was another day of slow starvation and another train of military aged males put on trains to a labor or concentration camp). In fact, back in 1939, under the "Pabst Plan" the Nazis were planning on utterly razing the city, eliminating almost all of its inhabitants, and replacing it with a quaint German town. By 1943 it had become clear to Jewish resistance leaders that a genocide was in process, with over a quarter of a million Jews already executed. The odds of a successful violent uprising were not good, but they were arguably better than any other option.

Even though the Warsaw rebellion was crushed, it had some positive effects. First, 100-300 Nazi troops were killed and wounded (almost 10%), vehicles were destroyed, and a month's worth of fuel, food, and other valuable military equipment was consumed. The German troops were also tied up for a month; that includes around a thousand SS, along with the logistics backbone that army motorized formation requires. Less directly, the insurgent's ability to fly Jewish and Polish flags -- powerful symbols of resistance -- for days inspired other resistance fighters, including a much more succesful uprising in 1944. The uprising also delayed deportations, bought time for civilians to arm themselves or run for the forests, and executed collaborators and gestapo agents.

Violence was not the first choice in Warsaw. Jewish resistance leaders tried to comply until it was clear that compliance was a suicide pact. The insurgency didn't go violent until the odds of success were already very low. If other violent insurgencies share this trait, we shouldn't be surprised to see that many of them fail.

Types of Violent Insurgency

Likewise, there is a relevant qualitative aspect. I think you can come up with four categories of violence:

• Indiscriminate, near random violence: This category is actually rather rare. Even terrorists usually select targets for a reason. On 9/11, AQ selected financial, political, and military targets. Insurgencies which have very poor command and control or mixed objectives may feature spasmodic violence which appears to be near-random.
  
• Violence used to facilitate criminal ends: Often insurgencies are associated with criminal violence. For example, the Afghan Taliban insurgency is often in bed with violent narco-traffickers who happen to share a goal of destabilizing legitimate governance. This sort of violence likely weakens the insurgency by fracturing support.
• Violence directed at non-combatants for purposes of intimidation: Night letters, targeted assassinations of "cooperators," and other threatening violence intended to intimidate can be highly effective at at least keeping the population from supporting an insurgency's adversaries. In Vietnam, the Vietcong were able to constantly keep the government of the south off balance through intimidation and assassination in villages in the south. Of course, such thuggish tactics can also backfire; this is likely what happened when the Sunni insurgency in Iraq threw out Al Qaeda, exhausted by AQ's violent tactics.
• Violence directed at legitimate targets in self defense: The example of a black woman arming herself with a rifle to protect her children from the Klan is an example of this type of violent resistance. Afghan insurgents that manage to portray themselves as freedom fighters and direct their violence against "occupying crusaders" might also claim this mantle (at least in the eyes of the public).
  

The first two categories almost certainly undermine the characteristics of non-violent insurgency that seem to be effective at reaching political settlements. The third category may not lead to victory, but may be a way to stave off defeat. By denying access to the population to the legitimate government by means of targeted violence, the insurgency can try to run out the clock.

The final category seems like it has some benefits. It certainly won't alienate supporters, and it may deter attackers. Of course, it also creates a security dilemma, which could result in uncontrollable violence escalation. If the defenders are unprepared to climb the escalation ladder, then yes, they might get rolled. An example would be a tiny country picking a fight with a big one. However, if the defenders have a credible deterrent capability then it could create conditions for political settlement; the example here would be two nuclear armed states who have the capability to destroy each other (or in a conventional setting, a stalemate where neither side is able to gain a decisive military advantage over the other). The insurgency may also benefit if their opposition has its hands tied; for example, maybe the government does have overwhelming forces but can't bring them to bear due to ROEs or fear of provoking the population because the insurgents melt into the civilian population, who sees them as legitimate defenders.

The challenge, of course, is that it is hard to stay in the last category. If you start climbing an escalation ladder, then you might be willing to use the "enemy of my enemy is my friend" strategy. When the Taliban ally themselves with drug runners because they need as many shooters as they can get, it fractures their support. If the insurgency needs to start "requisitioning" supplies at gunpoint from civilians then they'll risk their base as well. And in any insurgency, there is a risk of poor C2 leading to spasms of uncontrolled violence which hurt the cause.

At the least, I think there is a serious research question here. I haven't done a thorough literature review or anything, but it seems that there are certainly several possibilities out there which are probably based on circumstances. If non-violent insurgency was always the best option, and insurgents are rational actors, then why are the Taliban planting IEDs? Clearly there is at least some perceived benefit to employing a strategy of violent resistance.