Conflict, Self-Interest, and Disaggregating the Pointy End of the State

The following is a guest post by Jonathan Powell. Dr. Powell is an Assistant Professor of International Relations at Nazarbayev University in Astana, Kazakhstan.  His research focuses on conflict processes, particularly aspects related to civil-military relations. He recently received his Ph.D. in Political Science from the University of Kentucky.

James Fearon’s recent response to Anjan Sundaram’s Foreign Policy article on the M23 rebellion leads to several points of relevance to civil war scholars. Fearon asks “Why don’t these rulers, in their own self-interest, take some of that money and use it to build crack units, presidential guards, or strong and loyal army divisions that would protect their hold on power against two dozen putschists, or a hundred or couple thousand rebels armed with rifles and maybe some mortars?” He further claims that the “relevant poli sci literatures” fail to adequately explain these trends. Though perhaps in its infancy, there is a growing body of literature that attempts to explain the dynamics that Fearon has noted.

The ongoing inability of the DRC to combat M23 is paralleled by the country’s prior history, specifically the poor performance of the Forces Armées Zaïrois (FAZ) during the Shaba conflicts and the first Congo War. These failures are not a coincidence, and they are not due to an absence of a self-interested policy on the part of leaders such as Mobutu or Kabila. To the contrary, a self-serving policy of political survival would seem to be a suitable explanation for continued strife in the eastern DRC and beyond.

Leaders can suffer from many types of threats that are manifested by different types of actors, including foreign invaders, a popular uprising, military coups, insurgents, and even an electorate. One of the challenges for these “self-interested” leaders is to determine the most efficient policy that will keep them in power. Think of it as a Downsian theory of ­non-democracy. Leaders would, of course, prefer to have a blanket policy that will stamp out any threat to their continued tenure. Unfortunately, the multitude of threats and the different actors that a leader must consider will force him or her to make trade-offs. In the end, a policy that increases security against one actor might decrease security against another. Think of it as a broadening of the civil-military problematique.

To answer the first part of Fearon’s question: leaders do take money and build “crack units, presidential guards, or strong and loyal army divisions.” See the work of Belkin & Schofer, Pilster and Bohmelt, and my work in the current issue of JCR on this point. Leaders cannot suddenly implement a coup-proofing apparatus if one is lacking; such an effort might actually precipitate a coup, as Lt. Col Samsudeen Sarr has claimed in the 1994 Gambian coup. Jun Koga has a very promising working paper on this point, suggesting that leaders wishing to build up a presidential guard will largely have to limit themselves to doing so when coup risk is low.

To answer the second part of Fearon’s question, it is important to clarify that efforts to address the first concern (coups) often increase vulnerability to the second concern (rebellion). Fearon points to the excellent work of Roessler as the best manifestation of this paradox. Roessler shows that leaders increase the mobilizational capacity of non-state actors when they purge disloyal elements from the government. A growing literature shows coup-proofing also decreases the counterinsurgency capabilities of the state.

The idea that coup-proofing reduces military capability has been suggested in the interstate conflict literature by a number of qualitative efforts (e.g., Pollack, Biddle & Zirkle, Quinlivan). More recent quantitative work shows more heavily coup-proofed regimes both perform poorly in interstate combat and seem to be conflict-avoidant even when other incentives for a transnational dispute are present. The detrimental influences of coup-proofing on military effectiveness are too broad to discuss here in detail. The creation of parallel chains of command and armed counterweights undermine the ability of armed forces to conduct coordinated maneuvers; military funds and materiel are frequently re-directed from the regular armed forces to the coup-proofing apparatus; non-merit promotion, retention and recruitment reduces overall soldier quality and morale; and command shuffles reduce unit cohesion and competence. Perhaps even more problematic is that many leaders simply appear to be disinterested in preventing or ending civil war.

My recently-defended dissertation closely investigates Jeffrey Herbst’s observation that African militaries show a “remarkable” failure to mobilize at the onset of civil war. This is followed by an equilibrium that sees leadership happily maintain a secure capital while an insurgency might be in control of a far-off territory. Crushing the insurgency will prove too costly for the leader, while taking the capital from the better-trained coup-proofing apparatus will prove too costly for the rebels.

This calculus has been illustrated in the DRC/Zaire for decades. Both manifestations of the state have combined a weak army with a strong coup-proofing apparatus. Armed counterweights such as the Special Presidential Division, the Civil Guard, and the gendarmerie possessed better training and resources than the regular army under Mobutu. However, he was simply not willing to weaken his control of Kinshasa by deploying his coup-proofing apparatus to fight a far-off insurgency.

Peter Mangold has noted Mobutu’s refusal to mobilize the paramilitary during Shaba, while Michela Wrong’s journalistic account pointed to a similar outcome during Laurent Kabila’s westward march: “Zaireans waited for Mobutu to send to send the elite forces they had heard so much about to the east. No one, after all, could expect the Forces Armées Zaïrois (FAZ) to stand up…They waited and waited.” Howard French’s coverage for the New York Times described army chief of staff General Monga lamenting that “Zaire’s ill-equipped and untrained army had not been given the means to fight a war” while the “well-equipped” Special Presidential Division was seeing “no action.” Monga was promptly suspended.

Just prior to Mali’s March 2012 coup, its soldiers openly complained of President Touré’s desire to “fight a war against the rebels in return for staying in power.” This equilibrium was ultimately disrupted when 2000-3000 Tuaregs, trained and armed as mercenaries in Libya, returned home after the fall of the Qadhafi regime in Libya. This tipped the balance of capabilities in favor of the rebels, and the regular rank-and-file of the Malian army calculated that Touré would do little to better their situation. The expected utility of a coup dramatically increased and the putschists acted.

Though civil wars will continue, the self-interested leaders can—and usually will—maintain power. The existence of a coup seems to lead Fearon to conclude that Touré would have been served well to invest in a “loyal” armed backing. To be clear, Mali’s 7300 man army is counterbalanced by a 4800-strong paramilitary apparatus, spearheaded by the “Red Berets” Presidential Guard. The fact that this was Mali’s first coup since 1991 should indicate the success of coup-proofing. Jay Ufelder (and my models in the current JCR) has consistently ranked Mali amongst the most likely to suffer a coup. In contrast to a coup being “easy,” the skill of the Red Berets was demonstrated by Mali’s nearly two coup-free decades, and their loyalty illustrated with their effort at a failed countercoup.

The countercoup clearly wasn’t easy. Neither was the 1985 Liberian effort that saw its leader publicly tortured, castrated, dismembered, and (according to some) cannibalized, with hundreds of his ethnic brethren slaughtered. Nor was the 2002 effort in Côte d’Ivoire that spiraled into a civil war that resulted in the deaths of thousands. The difficulty of these efforts is clear from these cases as well as the growing rareness of coups. While the work of John Clark (here and here) largely points to the influence of democratization, my own work and countless qualitative offerings also point to the importance of “crack units, presidential guards, or strong and loyal army divisions.” Building these units takes time and resources, but their existence and effectiveness demonstrates that leaders act in a self-interested manner with regard to military organization.