The Lesser Evil: When does it make sense to intervene on behalf of incumbents?

*This piece was first published in the SSR Resource Centre’s The Hub and is republished with permission from the Centre for Security Governance.

A few weeks ago, Edward H. Carpenter came out with two compelling posts (here and here) in the Duck of Minerva. In his first article, he notes that the Islamic State’s (IS) advance in Syria and Iraq is only one example of recent victories by mobile, non-state Islamic fundamentalist groups organized as networks. In his second, he argues that while the governments these insurgencies seek to topple may not meet international standards of good governance:

“No population has benefited from the long-running civil wars or instability that has resulted from an existing state government being overthrown by networked opposition elements; and few regional governments are able to beat these networks quickly and decisively on their own. Thus, the West should consider making short-term, limited intervention on the side of existing governmental bodies the norm.”

These interventions, Carpenter writes, would combine airstrikes and ground forces comprised of government and international troops. Interventions would only occur when a conflict meets a threshold of a few pre-determined conditions, “Such a threshold would probably need to specify level and pace of conflict, presence (or lack) of diplomatic avenues of resolution, and several other measures beyond a simple casualty count.”

In response to Carpenter’s article, Rachel Strohm wrote a response piece teasing out some of the argument’s problems. Strohm uses the Rwandan Genocide as an example of a case when a state’s violent overthrow led to an improvement in the lives of its citizens. Because it is hard not to improve on a regime that kills a huge percentage of its population, there is a good argument that when a state is that brutal, seeking to crush any armed resistance will only allow the state to continue attacks on civilians.

Strohm’s point is a good one, and gets at something Carpenter’s argument seems to imply, but does not come out and say. The issue is not really with network insurgencies in general, but their relative capability to create a new stable new government. The ability of violent groups to create positive and intentional change is frequently overstated, and only in rare cases of extreme states weakness or government brutality does a rebel group’s ascent to power potentially offer a less violent future.

Determining when this is the case is difficult, but Carpenter’s own analysis of the nature of insurgent groups offers one potential avenue. He notes that they tend to be organized in networks rather than hierarchies, allowing for battlefield success. However, networks are less effective in performing governance than hierarchies because they lack the centralization and chain of command necessary to perform activities like tax collection, consistent law enforcement, and paying civil servants. As Weinstein argues, when commanders lack control over their soldiers, these soldiers are more likely to abuse civilians. Therefore, one metric for determining whether or not to support a non-state actor is their level of hierarchal organization in comparison to the state’s. In Rwanda, the state’s devolution of violent power to the Interahamwe, a non-state actor, meant it more closely resembled a network than the Rwandan Patriotic Front (RPF) rebels.

Following up on Strohm’s post, I see five additional implications of Carpenter’s argument that are worth fleshing out. First, Carpenter’s proposed interventions would follow the example of the French intervention in Mali, where superior airpower and ground troops were able to halt the insurgents’ advance. Carpenter hints that a similar policy would have been the right choice in Libya. However, these two countries share two characteristics that most others lack. Libya and Mali contain huge swathes of desert between cities and a correspondingly low population density. Rebels wishing to overthrow the state must traverse these areas, and in the process, become easy targets for a competent air force.

In many other countries this isn’t the case. In Syria, which Carpenter also mentions, putting down the rebels early would have required significant numbers of international ground troops due to western Syria’s population density. Assad has already tried, and failed, to crush the rebellion with superior airpower. While there is a good argument that Carpenter’s template approach would work against IS, there are many other insurgencies around the world where it would not.

Second, Carpenter doesn’t fully account for the possibility of failed interventions, which could happen in several ways. Had an international coalition attempted to intervene in Syria for example, its probable failure to crush rebel forces would have likely entrenched the conflict even more rapidly. Additionally, even if responses are pre-planned in the way Carpenter envisions, they may suffer from poor coordination, or a lack of financial and human resources. If the intervention fails to crush the rebels quickly, they may end up prolonging the conflict and supporting an abusive government.

Third, a norm that provides for consistent international military support of incumbents would provide abusive regimes with perverse incentives. Leaders wishing to crush a rival or gain domestic support could provoke a violent confrontation with opponents, leading to an international intervention in the incumbent’s favor. For states like Sudan that chronically make war against internal opponents, consistent international support for incumbents against military challengers could encourage persistent aggressive behavior.

Fourth, Carpenter perhaps underestimates the enormity of the normative shift that he prescribes. Widespread international armed support for incumbent regimes would effectively de-legitimize armed resistance as a way to force concessions or overthrow oppressive authorities. Subsequently, it would bring states closer together by putting each one, regardless of its behavior, on equal footing. While decreasing the overall legitimacy of armed challenges to states would likely be a positive development, the few potential exceptions outlined above stand out clearly. It would also be very difficult to convince powerful states to work together to defeat all armed insurgents. Powerful states are not the ones that tend to face armed challengers, while various non-state armed groups often further their interests. Therefore, it is highly unlikely the norm would be enforced consistently, even if this isn’t necessarily a bad thing.

Finally, the de-legitimization of armed struggle that would occur through Carpenter’s proposal would mean a likely increase in the number of nonviolent insurrections against incumbents. As Maria Stephan and Erica Chenoweth have shown, nonviolent campaigns that overthrow the government lead to more stable and more democratic subsequent governments, so this change would be unquestionably positive.

Carpenter’s argument has its merits, and his somewhat controversial support for incumbents makes sense in some circumstances. However, before taking the proposal seriously, as I hope some policymakers will, it makes sense to give the argument a stronger theoretical background and identify exceptions. Doing so might lead to an exceptionally promising if somewhat unconventional way to think about international violence prevention.

Advertisement

It’s Not You, It’s Me

A few weeks ago, Rachel Strohm wrote two exploratory pieces on the similarities between violence and its prevention in Iraq and Chicago and Boston and Mexico, respectively. The pieces got me thinking, because comparing domestic and foreign violence is a topic I really like and one I’ve written about before (here and here), but it’s taken me awhile to sort out my thoughts. This post is a bit scattered, but if you stay with it, I promise I have a point at the end.

Replicable violence prevention programs

I just finished David Kennedy‘s Don’t Shoot, which details the author’s experiences over many years of working to prevent inner-city gun violence. In the book, Kennedy argues that gun violence in inner-cities is really all about gangs. To stop homicides, you make it clear to gangs that if they kill anyone, law enforcement will come down with everything they got. The first gang that does kill anyone then gets decimated by raids and tough prosecution. Gangs then get the message that while they can continue operating, they absolutely cannot use murder to achieve their aims.

Kennedy was one of the major players in Operation Ceasefire, which was enormously effective in reducing gun homicides in Boston in the 90’s, and Strohm cites Ceasefire via this Dan Snyder piece that recommends a similar approach against Mexican cartels. At the end of Strohm’s piece, she asks whether Ceasefire-style tactics might work against rebel groups. In Mexico, there is an argument for Ceasefire-style tactics, but there are also reasons that we could expect it to be much less effective. Ceasefire relies on the issuing of credible threats by law enforcement, and since cartels frequently possess superior firepower than the Mexican state, delivering enforcement promises may be difficult. Additionally, Ceasefire relied heavily on changing social norms within cities’ tight-knit gang networks, but it’s unclear to me after reading Kennedy’s book if a similar approach could work in Mexico, where cartels are spread over vast areas.

For rebel groups in general, I don’t see how Ceasefire-style tactics could be effective. The government would have to have the ability to militarily combat the rebel group without engendering resistance through excessive brutality. Perhaps, focusing on the most violent actors within a rebel group could send a message to other members, but finding the right balance of force and restraint would be very challenging for states lacking the relative unity, force, and intelligence-gathering apparatuses of America’s police departments.

Perceptions of the transformative power of violence

Strohm’s post on Iraq and Chicago, which I haven’t really mentioned yet, ponders leaders’ perception of the ability of American violence to change societies, illuminating a strange dichotomy. On the one hand, many prominent American pundits continue to promote the idea that Iraq can be/could have been fundamentally altered through US military might (despite all evidence to the contrary), “What does interest me about all of this is why foreign policy hawks continue to believe that this type of transformation is possible.  I don’t just mean ‘why haven’t they read their history’ or ‘why are they so arrogant,’ but rather ‘through which causal mechanisms do they believe that American money and troops can overcome sectarian divides, build a strong state and prevent violent rebellion?'” On the other, there is a consensus that the violence on Chicago’s South Side is natural, impervious to outside intervention. These dual approaches represent a paradox. Strohm again, “The place where the US government should in principle have the most policy leverage is where violence is depicted as uncontrollable, while the place where US leverage is limited by Iraqi sovereignty and, oh yes, the fact of being an invading force, is where American policy is expected to be most transformative.”

Without any specialist knowledge, it seems likely to me that a belief in the military abroad but not the police at home stems from cognitive biases. The more familiar we are with a situation, the more likely we are to understand why it exists and expect it to continue in perpetuity. However, the less familiar we are with a country’s history, the more likely we are to see any action our side takes as decisive, because we don’t understand the other important players.

Out of Iraq, Mexico, and Chicago, the latter would seem the place most likely for an armed intervention to positively shape the situation. However, Chicago’s homicide rate remains startlingly high. Surely, as Kennedy argues persuasively in Don’t Shoot, there is nothing inherent about certain cities that make violence high. While set in the general context of urban poverty, it is relatively flexible gang networks that determine the murder rate.

The problems of our own institutions

While Kennedy’s stories certainly gave me hope that there is a realistic way to combat inner-city gun violence (which could certainly teach us more lessons about violence prevention in general than the few I sketched out above), the often optimistic tone of the book is undercut by some serious and repeated failures. The original Operation Ceasefire, for example, fell apart due to the crumbling of the inter-agency cooperation required to do Ceasefire effectively. Consequently, gun homicides shot up again in Boston. Boston is not alone in failing to maintain a Ceasefire-style program. In my reading, Kennedy doesn’t fully acknowledge the systematic nature of the initiatives’ failings. Many of the Ceasefire-style programs were held together by exceptional amounts of interpersonal work that left Kennedy exhausted. Kennedy is adamant that these types of initiatives can work anywhere in America, but he doesn’t propose a strategy other than relentless mediation that can sustain the necessary cooperation.

Without denying the many successes of programs Kennedy participated in across the United States, his book also reads like a case study on the frequent dysfunction of American governmental institutions. It only gets mentioned once in the book, but nothing I’ve ever come across gets at this issue quite as well as The Wire. The show, while generous to its characters, is incredibly cynical about the possibilities for progressive change in America’s inner-cities. In season three, good-guy cop Colvin proposes his Hamsterdam plan to essentially legalize drug-dealing in a small area of Baltimore to stop the police wasting time and putting themselves in danger on low-level busts. The plan isn’t Ceasefire, but shares some similarities. The plan fails when the news reaches the media, forcing politicians to crack down in order to save face. Colvin’s plan was a sensible initiative that ran up against city politics institutionally incapable of supporting good drug policy.

Ceasefire suffers from some of the same vulnerabilities. It requires prosecutors and police to functionally crack down on some drug dealers and not others at particular times, and could feasibly be accused of containing Hamsterdam’s same fatal flaws. If politicians, cops, or prosecutors begin to doubt the program’s effectiveness, it can all fall apart. And while the Hamsterdam incident is revealing of The Wire‘s beliefs on political change, no episode is more crushing than the series finale. Without giving too much away, despite the best intentions of numerous individuals to break the negative cycle, it repeats itself, this time with new protagonists. For The Wire‘s creators, inner-city Baltimore is a machine that unceasingly produces poverty, corruption, and violence. Those wishing to change the cycle, including those coming from the outside, are quickly sucked in and co-opted by the machine. Fittingly, in the end of his Baltimore chapter, after suffering rampant in-fighting and ultimate failure in the city, Kennedy writes, “I don’t know how to control the good guys.”

This leads me to an important, if obvious point. Whether it be in Chicago or Iraq, the challenge of stopping violence is about both the intervener’s institutions and the difficulties presented by the target society. In Iraq, the problems presented by Iraqi society would have made it very difficult for any intervening force to make a difference, but American institutions also failed to create the internal conditions most conducive to making change. The same lesson holds true in Mexico. The cartel’s tactical superiority makes enforcement difficult, but the thorough corruption of the Mexican state makes even attempting enforcement challenging. In Chicago however, one could make a reasonable argument it is entirely about the intervener’s institutions. If what Kennedy proposes is true (and I’m strongly inclined to believe it is), then a fairly simple plan could drastically reduce gun violence in Chicago. Because that plan exists and is fairly well-known among US law enforcement, it is these formal institutions, as compared to the informal institutions of Chicago’s gangs, that carry the culpability for astronomical homicide rates. Now, obviously Chicago law enforcement is not the root cause of gun violence in Chicago. But if we’re talking in terms of making change, the breakdown is occurring with the intervener’s capacity to implement and carry-out a plan, rather than the target society’s pushback (even if the two are connected at some level).

When originally conceiving this post, I had thought about heavily citing James Scott’s theories on the limits of transformative state power. However, I realized Scott and Kennedy are coming at the same problem from different directions. For Scott, the inherent complexity of society makes it impossible for the state to control many things about society. For Kennedy, eliminating violence is impossible, but pulling certain levers within gang networks can lead to radical change. The challenges for the intervention are internal.

Scott’s work is certainly helpful for thinking about why states, or other organizations, fail to accomplish certain goals. However, without downplaying  the importance of Scott, his framework is a limited one. Scott defines failure in terms of programs that are implemented and do not accomplish their goals, but doesn’t examine why certain programs with transformative goals are sometimes never implemented at all (there’s also a case that Scott erroneously sees all government failure as due to external, and not internal, problems). The institutions themselves, rather than the society they serve, may be the primarily source of failure.

To fully understand why violence prevention initiatives fails, it’s necessary to combine the Kennedy approach and the Scott approach. However, academia seems to largely be on the right track. There are large bodies of work both on why societies are resistant to change, and literature like Young, Autesserre, and Ferguson on international institutions’ deficiencies.

Strohm’s post on Iraq focused on policymakers, and I think that the problems lie with this group. For example, policymakers regularly speak of a foreign policy toolbox, without examining the situation-specific effectiveness of those tools. Overcoming that barrier is just the first step, though. Policymakers and pundits, especially those recommending violence prevention plans or military interventions, should more strongly consider the possibility that “effectiveness” is a two-sided coin, requiring both effective institutions and societies conducive to deep change. After the wars in Iraq and Afghanistan, there is now a grudging acceptance of the latter, but I fear few take the former seriously. By seriously, I mean that it is not just that the American governmental bureaucracy has some problems, but that government bureaucracies in the 21st century state-system are unprepared to launch effective interventions to prevent or mitigate violence, at home or abroad.