Reciprocal Destabilization: A Two-Level Security Dilemma Involving Rebellions, Refugees, and Regional Conflict

Why do governments so often provide support to rebels fighting the government of a neighboring state? Also, once a pattern of “reciprocal destabilization” has been established between neighboring states, their leaders often find it difficult to cease such assistance. This paper uses examples of reciprocal destabilization from recent conflicts in the Horn of Africa and Kurdistan to illustrate a few stylized facts about this phenomenon. An informal model is then used to motivate a series of hypotheses which specify the conditions under which reciprocal destabilization should be expected to occur and persist. The paper concludes with a discussion of two ways in which these hypotheses might be tested in future research. This paper was prepared for presentation at the Annual Meeting of the International Studies Association, February 21-24, 2001, Chicago, Illinois. Any observer of contemporary conflicts is likely to have noticed instances in which two neighboring governments support rebel groups that are fighting against the other government. For example, both Iran and Iraq supported rival Kurdish factions throughout their 1980-88 war. Cross-border support for rebels can persist over many years and may even survive dramatic changes in regime. For example, successive Sudanese governments provided support for Eritrean (ELF and EPLF) and Tigrean (TPLF) rebels fighting the Ethiopian government, and diverse Ethiopian regimes returned the favor by allowing southern Sudanese rebels (first the Anyanya and later the SPLA) to operate from bases in their own territory. This paper reports my initial efforts to come to terms with this puzzling phenomenon of reciprocal destabilization. By supporting rebels fighting each other, each government ends up facing a more pronounced domestic security threat (than would be the case if its own rebels lacked any external support). Since both governments would be better off if each stopped this practice, agreements to this effect would seem to be an obvious next step. However, anecdotal evidence suggests that bilateral agreements to cease this practice are rarely effective. For example, Iran agreed to stop supporting Iraqi Kurds in 1988 but resumed supporting Kurdish operations (especially the KDP) during the 1990-91 Gulf War. An earlier regime (the Shah’s) had made a similar agreement with Iraq in 1975, but support resumed soon after the Islamic Revolution in Iran. Throughout this period, Turkish forces attacked PKK rebel bases in Iraq, Iran, or Syria. I strongly suspect that if one were to look at this conflict at any specific time period in the last several decades, one will see that, despite the shifting alignments among governments and Kurdish rebels, at least one Kurdish faction will be obtaining significant support from at least one neighboring state in that time period. As will be detailed below, neighboring countries in the Horn of Africa seem to have had especially difficult problems in managing the cross-border operations of rebel groups. Reciprocal destabilization can be observed as early as 1967, when the leaders of Ethiopia and Sudan signed an agreement to stop supporting each other’s rebels but then promptly violated that agreement (Woodward 1996:121). When the leaders of Ethiopia (Mengistu) and Somalia (Siad Barre) signed an agreement in 1988 to stop supporting rebels in each other's territory, both were facing severe security problems from other sources. Nonetheless, Mengistu's regime continued to support the USC faction in southern Somalia (Woodward 1996:128). According to recent reports, the current regime in Ethiopia continues to support several factions in Somalia’s complex inter-clan conflict (IRIN 2001). One sequence of events suggests that this phenomenon of reciprocal destabilization (RD) might have a structural basis that transcends changes in regime types, that RD constitutes, in effect, a relatively stable equilibrium in certain conflict situations. Consider what happened in the Horn of Africa around the pivotal year of 1991. Before that time, the Mengistu regime in Ethiopia had been allowing SPLA forces to use its territory as bases in its fight against Sudan’s government, which in turn was providing similar support to EPLA and TPLA rebels fighting the central Ethiopian regime. In 1991 these rebels (and affiliated factions) took power in Addis Ababa, and, in gratitude to Sudan’s long-standing support, forced the SPLA out of its Ethiopian bases. Not long afterwards, however, Sudan was again supporting rebels operating against both of the new regimes in Ethiopia and Eritrea (which became independent in 1993), and the SPLA