Why democracies don't lose insurgencies

[Rex Brynen]

What's the dataset coding definition for insurgency, Cavguy? Does it depend on battle deaths (and if so, is the threshold set in absolute numbers, or relative to population size)? How does it differentiate from terrorism?

I ask this because, arguably, the great success of democracies might be that groups never make the transition from protest movement > small terrorist group > full-blown insurgency, and that the "success" of democracies lies rather earlier than their ability to engage in full-scale COIN.

A critic might argue that it is rather like trying to measure the effectiveness of a bug-zapper atop Mount Everest (the body count isn't really getting at the issue of why there are so few mosquitos around...)

[AmericanPride]

He employs violence. He employs terror, but only in phase III as in Vietnam or China (sorry Ken) does he employ "military means" Certainly the counterinsurgent, possessing a military is free to, and often does, "employ military means" in responding to such violent popular uprisings.

In your opinion, what is the significance and consequences of differentiating between violence as "military means" and other forms of politically-driven violence?

[jmm99]

because of the differences you cite, seem to engage using the statistical approach (vs. the "matched" units approach) only in their Model 7, where their criteria do still differ from yours, as you point out:

(Mech article, p.25 in .pdf)
Interestingly, REGIME only reaches statistical significance in the post-1945 era ~Model 7! and is negatively associated with incumbent victory, suggesting that as states become more democratic their vulnerability to defeat increases. This at once supports and qualifies existing arguments about presumed ineffectiveness of democratic states in COIN wars. To be sure, it suggests that as political systems become more open, the likelihood of defeat is also increased. Yet these shifts in regime type could occur at lower ends of the 21-point Polity2 values—that is, a shift from 5 to a 0—and therefore one should not conclude that stable democracies are especially vulnerable. [71] Moreover, the time-dependent nature of this effect is at least partially inconsistent with claims that audience costs or educated middle classes are hobbling democratic war efforts. Indeed, neither mechanism is exclusive to democracies, nor is it clear why their effects should only be observed after 1945.

[71] We re-estimated Model 7 with a dummy DEMOC variable that denoted whether a state’s regime score was .7, the conventional threshold for being coded a stable democracy, DEMOC was not significant.

This part of the discourse reminds me of many "battles of the experts" in which I've participated - not as an expert, but as the guy who had to translate the expert to the judge or jury. As you know, many of these battles degenerate (from the non-expert's standpoint) into inside baseball arguments between the experts.

What I've gleaned from my experience is that the practical examples weigh more with the non-expert than anything else. E.g.,

from Cavguy
... but bottom line if mech hurts COIN forces than how come 3ACR (has least number of dismounted troops of any BCT in the Army), and 1/1 AD (heavy legacy BCT) were hugely COIN successful ?

because ... da, da, da - and you're off to the races. After reading through Lyall's articles, I can see why you are considering the topic. Bonne Chance.

[Ken White]

Whoops -- already used that line once this month...

I'll change it; I love inexperienced experts...

...because ... da, da, da - and you're off to the races. After reading through Lyall's articles, I can see why you are considering the topic. Bonne Chance.

See what I mean about the number crunchers getting it wrong. They almost always do.

Always hated it when my Commander was one of them; the intuitive guys just do it a lot better. No numbers or machinery involved. Ain't at, right Niel...

[Cavguy]

Because my degree is in Security Studies, papers tend to focus on the quantitative. I am less than impressed with the way IR "codes" history. I think Stephen Walt captured most of the issues well in "Rigor or Rigor Mortis?"

That said, quant has its uses but can be carried too far. It does tend to delve into "inside baseball" between datasets, coding issues, and regression. It's also why most policymakers pay little attention to "hard" IR theory.

[Bob's World]

In your opinion, what is the significance and consequences of differentiating between violence as "military means" and other forms of politically-driven violence?

First to preface this, the core concept being developed here is that:

"In order to resolve a problem one must first have a good understanding of what the nature of the problem is."


If an infantry squad is moving down a country lane and enters into an L-shaped ambush, it pretty much looks the same and has the same effect if it is executed by a state-opponent's military force or if it is executed by a bunch of local men who have joined an insurgent movement.

Similarly, the tactical actions to counter that ambushing unit are much the same as well.

To successfully address and resolve WHY the ambush was executed, however, are two very different things altogether. This is why I immediately become guarded when things that appear very similar on their face are immediately lumped into convenient bins based not on the purpose for action, but more by the physical nature of the action itself.

Examples of this currently are those who lump many diverse groups with equally diverse purpose for action under the deliberately inflammatory banner of "Terrorist." Similarly, the U.S. conventional force has recently trended to lumping actions against a similarly diversely motivated set of organizations as being "COIN." Taken to an even larger generalization, because we are using the military to defeat violent activities we lump it again under the even broader umbrella of "war."

Such lack of sophistication in seeking to understand ones opponent's purpose for action prior to defining and shaping your own counter to resolve such problems can only lead to long, drawn out affairs that will tend to be too heavily weighted toward military action directed at the symptoms of the problem, as opposed to a holistic program of engagement directed with laser precision at the sources of the problem.

So just as every fist fight is not a boxing match (both look similar, but begin for different reasons, and are won by different measures) all violence is not warfare. The boxer jumped in the alley on the way home will use all of his boxing skills and have a much greater chance of prevailing over a street fighter than an untrained individual encountering the same problem. But if the boxer thinks he's going to get a break every 3 minutes, receive no punches below the belt, or not get hit on the back his head by a pipe swung by his mugger's girlfriend, he is going to be in trouble. We all can grasp this, but for some reason when 200,000 of us get on planes and fly to a foreign country to wage "war", we can't seem to look past the "threat" we are there to defeat and really do the frontend analysis to figure out not just HOW he fights, but WHY he fights and then correspondingly, how to get him to stand down.

Many previous operations chalked up as COIN "victories" were really little more than a temporary set back for the insurgent movement. Insurgencies are internal violence for political change by a populace against its own government. If beaten down from phase II to phase I or even 0, so long as the conditions giving rise to the insurgency remain unaddressed, the populace will move back up into higher phase insurgency as soon as they have the capacity to do so. Algeria and the Philippines are both great examples of this. You can kill Berbers and Moros every generation for the next 1000 years; or you can take the time to figure out why they are so willing to keep getting back up to fight, and attempt to shape a solution that provides some sense of justice and good governance for the affected populace.

Intervening powers are like police officers responding to domestic violence calls. Dangerous crazy stuff. You can't really hate someone unless you love them first. Similarly, a populace driven to violence against itself has a whole other degree of motivation than a professional army sent against another state for King and country. The latter is just business. The Former is personal.

So, in a democracy, a populace is less likely to revolt as they typically can resolve their differences with their government by legitimate means short of violence. Democracy is the greatest COIN tool ever devised for this very reason. The other side of this though, is that when a democracy intervenes in someone else’s insurgency their actions are tempered by their own populaceslack of will for harsh violence, which prevents them from waging the type of insurgent crushing operations that will put an insurgency into phase 0 for an extended period of time. Insurgency was not a big problem in Yugoslavia until Tito's iron hand was removed from back of the populace’s neck.

So I guess here is a take-away: Insurgency is not apt to erupt in either a strong democracy, or a strong dictatorship. The first because the conditions giving rise to insurgency can be addressed peacefully, the second because any effort to protest those conditions can be immediately and brutally crushed. Insurgency will then tend to occur in those governments that lie in between these two poles. Worst case is a democracy too insecure to allow true popular input to effect change, but not heavy handed enough to crush those who chose to act out. When you intervene in such a country (think, oh, how about Iraq or Afghanistan) you are in for a challenge. But remember what pole we are there to lead them toward, and shape your engagement accordingly.

[AmericanPride]

Bob,

I agree with much of what you said, though I would go further and argue that violence is at the foundation of all political activity, and only that mitigating factors such as time, resources, the strength of the opposition, etc prevent distributing power based solely on the capacity to impose one's will on another through violence. That said, selective and subjective definitions of 'democracy' and 'dictatorship' (i.e. modern, Western versions of either) seems to suggest that insurgencies are unlikely to erupt in either. But I would dispute any correlation on the basis that every democracy, dictatorship, and insurgency are different, and that any conclusion drawn from such comparisons would be incomplete. There is no absolute democracy or dictatorship that we can use as an appropriate measure. I would argue that why democracies (and dictatorships) are successful in either mitigating insurgency or do not generally face the threat in the first place is a function of class factionalism and not the organization of government (which, IMO, reflects the distribution of power among classes; hence the different shapes of governments that are nonetheless characterized as the same or similar). Whatever the agenda of post-WW2 insurgencies, the common attributes are often a dispossessed concentrated ethnic majority led by an educated, relatively well-off (whether in wealth, prestige, power, etc) ideological elite. Looking through Cav's examples in his first post, I have to ask how many of those insurgencies "lost" as a direct result of the system of government in place? Or in contrast, how many "lost" because of violence, resettlement, etc but they happened to have been defeated by a democratic government (Chechnya, Lebanon, Turkey)? Another question: given that the minority of states were democratic through the latter half of the 20th century, what is the possibility that democratic states appear statistically more likely to defeat insurgencies because there are fewer examples of 'democratic insurgencies'?

The common factor, IMO, seems to be that democratic governments are more willing to use violence to defeat insurgencies that are considered 'illegitimate' according to the political climate/context of the time. Given that democracies are generally ruled by the ethnic/ideological majority, are democracies more likely to pursue violence until the 'bitter end' so to speak? And since the democratic state is often at war with a dispossessed minority, how does the advantage in resources and mobilization for the state impact the outcome of the conflict? Last question: if democratic governments are more successful in mitigating insurgency because of ease of accommodation for rival political factions, why do 'democratic insurgencies' occur in the first place?

Just my thoughts.

[Cavguy]

What's the dataset coding definition for insurgency, Cavguy? Does it depend on battle deaths (and if so, is the threshold set in absolute numbers, or relative to population size)? How does it differentiate from terrorism?

I ask this because, arguably, the great success of democracies might be that groups never make the transition from protest movement > small terrorist group > full-blown insurgency, and that the "success" of democracies lies rather earlier than their ability to engage in full-scale COIN.

A critic might argue that it is rather like trying to measure the effectiveness of a bug-zapper atop Mount Everest (the body count isn't really getting at the issue of why there are so few mosquitos around...)

Agreed. The RAND Terrorist post-1945 dataset has over 600 listed entries. The Insurgency dataset has 89. The coding is based off of a number of factors, including size, goals, casualties, and other factors. Adding terrorist data actually reinforces the hypothesis, as they are often defeated and rarely successful.

Intuitively, I think you have hit on the reason - the structure of democracy makes it hard for terror groups to metastasize into full blown rebellions. There would have to be a huge reservoir of non-reconcilable discontent unresolvable through political means - which in turn suggests the contest is hardcore ideological (religious-like or separatist) in nature, rather than grievance based.