When Good Fences Make Bad Neighbors - Belfer Center for ...
When Good Fences Make Bad Neighbors - Belfer Center for ...
When Good Fences Make Bad Neighbors - Belfer Center for ...
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong><br />
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong><br />
<strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong><br />
Fixed Borders, State Weakness, and<br />
International Conºict<br />
Boaz Atzili<br />
Since the end of World<br />
War II, the norm of ªxed borders—the proscription against <strong>for</strong>eign conquest<br />
and annexation of homeland territory—has gained prevalence in world politics.<br />
But have ªxed borders made international conºict less frequent? Observers<br />
might assume they have, given that territorial issues have historically<br />
been a major cause of war. 1 However, among sociopolitically weak states (i.e.,<br />
states that lack legitimate and effective governmental institutions), ªxed borders<br />
can actually increase instability and conºict. <strong>Good</strong> fences can make bad<br />
neighbors.<br />
Until the late 1980s, the scholarly literature had devoted little attention to<br />
theories regarding the role of territory and borders in international relations. 2<br />
Since then, however, a growing body of work on this subject has emerged. 3<br />
One promising line of inquiry has focused on international norms concerning<br />
changes in borders. Mark Zacher and Tanisha Fazal, <strong>for</strong> example, have found<br />
that post–World War II cases of <strong>for</strong>eign conquest and annexation are a rarity<br />
and that the norm of ªxed borders has grown stronger over the years. 4 The ef-<br />
Boaz Atzili is a Research Fellow in the <strong>Belfer</strong> <strong>Center</strong> <strong>for</strong> Science and International Affairs at the John F. Kennedy<br />
School of Government at Harvard University.<br />
The author thanks Ericka Albaugh, Michal Ben-Josef Hirsch, Patrick Boyd, Naomi Chazan,<br />
Thomas Christensen, Tanisha Fazal, Rachel Gisselquist, Arie Kacowicz, Sarah Lischer, Galia Press-<br />
Bar Nathan, Kenneth Oye, Roger Petersen, Jessica Piombo, Jeremy Pressman, Stephen Van Evera,<br />
and two anonymous reviewers <strong>for</strong> helpful comments on earlier drafts of this article.<br />
1. John A. Vasquez, The War Puzzle (New York: Cambridge University Press, 1993), pp. 123–152;<br />
and Kalevi J. Holsti, Peace and War: Armed Conºicts and International Order, 1648–1989 (New York:<br />
Cambridge University Press, 1991), pp. 306–311.<br />
2. Noting this is John G. Ruggie, “Territoriality and Beyond: Problematizing Modernity in International<br />
Relations,” International Organization, Vol. 47, No. 1 (Winter 1993), pp. 139–174.<br />
3. See, <strong>for</strong> example, Friedrich Kratochwil, “Of Systems, Boundaries, and Territoriality: An Inquiry<br />
into the Formation of the State System,” World Politics, Vol. 39, No. 1 (October 1986), pp. 27–52;<br />
Brendan O’Leary, Ian S. Lustick, and Thomas Callaghy, eds., Right-sizing the State: The Politics of<br />
Moving Borders (New York: Ox<strong>for</strong>d University Press, 2002); and Monica Duffy Toft, The Geography<br />
of Ethnic Violence: Identity, Interests, and the Indivisibility of Territory (Princeton, N.J.: Princeton University<br />
Press, 2003).<br />
4. Mark W. Zacher, “The Territorial Integrity Norm: International Boundaries and the Use of<br />
Force,” International Organization, Vol. 55, No. 2 (Spring 2001), pp. 215–250; and Tanisha M. Fazal,<br />
“The Origins and Implications of the Territorial Sovereignty Norm,” paper presented at the annual<br />
convention of the American Political Science Association, San Francisco, Cali<strong>for</strong>nia, August 29–<br />
September 2, 2001. Rather than referring to this norm as “territorial integrity” or “territorial sovereignty,”<br />
I use “ªxed borders” to distinguish between territorial and nonterritorial intervention.<br />
International Security, Vol. 31, No. 3 (Winter 2006/07), pp. 139–173<br />
© 2007 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.<br />
139
International Security 31:3 140<br />
fects of this norm on interstate relations, however, have yet to be analyzed—an<br />
omission this article aims to address.<br />
The article posits that, in many regions of the world, adherence to the norm<br />
of ªxed borders has led to international conºicts and growing instability by<br />
perpetuating and exacerbating state weakness. Three factors account <strong>for</strong> these<br />
negative effects. First, an international system of states with ªxed borders deprives<br />
states of what were historically their greatest incentives to develop<br />
strong institutions: external threats to their territorial integrity and opportunities<br />
<strong>for</strong> territorial expansion. Second, without such territorial threats, a coherent<br />
in-group identity and loyalty to the state are difªcult to establish. Third,<br />
without a mechanism through which weak states can be overtaken by stronger<br />
ones, the <strong>for</strong>mer may persist and perhaps become even weaker.<br />
Sociopolitically weak states in a world of ªxed borders may be more prone<br />
to internal conºict or even civil war because the incentives <strong>for</strong> excluding<br />
whole groups of citizens are greater, and because there is a higher likelihood of<br />
the emergence of an internal security dilemma. Such conºict may spill over as<br />
neighboring states feel obliged to come to the assistance of threatened<br />
coethnics within the weak state, and as refugee ºows create breeding grounds<br />
<strong>for</strong> insurgency and counterinsurgency, which in turn can lead to <strong>for</strong>eign intervention<br />
or even full-ºedged war. <strong>Neighbors</strong> might also view the state’s weakness<br />
as providing an opportunity to seek economic gains or political inºuence,<br />
including regime change. 5<br />
The argument that a norm that seeks to make the world a more peaceful<br />
place may instead cause it to become more conºict prone is both counterintuitive<br />
and theoretically new. In addition, given that the phenomenon of<br />
weak and failed states is widespread, the argument potentially has important<br />
empirical implications. In 2006 the Failed States Index listed twenty-eight<br />
countries as being in a state of “alert” and seventy-eight more as in a state of<br />
“warning” with regard to their prospects <strong>for</strong> becoming failed states. These include<br />
countries in Africa, the Middle East, Asia, the <strong>for</strong>mer Soviet Union, Latin<br />
America, and the Balkans. 6<br />
This article employs a single case study—the war in Congo, a country that<br />
was known as Zaire from 1971 to 1997 and since then as the Democratic<br />
5. Although outside the scope of the article, it is useful to note that states in an international system<br />
of ªxed borders create more favorable conditions <strong>for</strong> global terrorist organizations to emerge<br />
(e.g., in Lebanon and Somalia). See Robert I. Rotberg, ed., “Failed States, Collapsed States, Weak<br />
States: Causes and Indicators,” in Rotberg, ed., State Failure and State Weakness in a Time of Terror<br />
(Cambridge, Mass.: World Peace Foundation, 2003), pp. 1–25.<br />
6. Fund <strong>for</strong> Peace, “Failed States Index,” http://www.fund<strong>for</strong>peace.org/programs/fsi/fsindex<br />
.php. See also Robert I. Rotberg, “The New Nature of Nation-State Failure,” Washington Quarterly,<br />
Vol. 25, No. 3 (Summer 2002), p. 85.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 141<br />
Republic of Congo (DRC)—as a plausibility probe to determine the theory’s<br />
validity and applicability. 7 The conºict produced tremendous carnage: as<br />
many as 3.8 million dead and many more injured or displaced. Both phases<br />
of the war (1996–97 and 1998–2002) involved domestic militias, a massive <strong>for</strong>eign<br />
invasion, and shifting alliances—with Angola, Rwanda, Uganda, and<br />
Zimbabwe playing major roles. Even though the war has ofªcially ended,<br />
peace remains elusive.<br />
I selected the Congo war as my case study <strong>for</strong> two reasons. First, Congo is an<br />
extreme (or most likely) case, because as a country in Africa—a continent<br />
where the norm of ªxed borders is strongest and where the states are<br />
weakest—it presents high values on both the independent variable (the norm<br />
of ªxed borders) and the intervening variable (a very weak state). A probe into<br />
the dynamics relating ªxed borders to state weakness and to international<br />
conºict should thus offer clear, discernable results. 8 Second, the Congo conºict<br />
is multifaceted; that is, many observation points can be generated from what is<br />
ostensibly a single case. Because this study is an initial probe into a newly<br />
identiªed causal mechanism, however, and because it involves only one case,<br />
the ability to generalize from its ªndings should not be exaggerated.<br />
The ªrst section of the article deªnes the norm of ªxed borders and discusses<br />
its development and growing inºuence. The second section explicates<br />
the effects of this norm on the strength or weakness of states. The third section<br />
examines the effects of the combination of state weakness and the norm of<br />
ªxed borders on relations between neighboring countries. The fourth section<br />
considers the case of the war in Congo as a preliminary probe of the plausibility<br />
of the article’s theoretical model. The article concludes with suggestions <strong>for</strong><br />
future research.<br />
The Norm of Fixed Borders<br />
State borders are social constructs. Different international systems have historically<br />
maintained different types of international borders. This section deªnes<br />
and analyzes the inºuence of one such construct: the international norm of<br />
7. I use “Congo” when referring to the country in general, and Zaire and Democratic Republic of<br />
Congo (DRC) when it was ofªcially termed so. Plausibility probes are “preliminary studies on relatively<br />
untested theories and hypotheses to determine whether more intensive and laborious testing<br />
is warranted.” See Alexander L. George and Andrew Bennett, Case Studies and Theory<br />
Development in the Social Sciences (Cambridge, Mass.: MIT Press, 2005), pp. 75–76.<br />
8. On the justiªcation and logic of using an extreme case, see ibid., pp. 120–123. For an example of<br />
a study that uses a similar approach, see Stephen Van Evera, “The Cult of the Offensive and the<br />
Origins of the First World War,” International Security, Vol. 9, No. 1 (Summer 1984), pp. 58–107.
International Security 31:3 142<br />
ªxed borders, 9 or the notion that state borders should not be changed and that<br />
annexation of a neighbor’s territory should be viewed as illegitimate. 10<br />
Historically, conquest and annexation were common. (See Figure 1.) <strong>When</strong><br />
Frederick the Great of Prussia invaded the Austrian province of Silesia in 1740,<br />
<strong>for</strong> example, he knew that as long as he maintained control over it, other states<br />
would not challenge his rule. Today, however, as Charles Tilly writes, “With a<br />
few signiªcant exceptions, military conquest across borders has ended, states<br />
have ceased ªghting each other over disputed territory.” 11 Indeed, there have<br />
been only ten cases of <strong>for</strong>eign military conquest of homeland territory since<br />
1950. 12 In a similar vein, Donald Horowitz has shown that cases of irredentism<br />
have become rare, although the potential <strong>for</strong> irredentist claims based on ethnicity<br />
have greatly increased. 13<br />
The norm of ªxed borders is the product of several material and ideational<br />
factors. Material factors include a decline in the importance of land as a means<br />
of production and the rise of technology in the developing world, which<br />
together have greatly decreased the incentives <strong>for</strong> territorial conquest; 14 the<br />
potential costs of conquest, which have risen exponentially with the development<br />
of nuclear weapons; 15 and the inability of weak central governments to<br />
project power over vast, scarcely populated tracts of land. 16<br />
Without the ideational “glue,” however, these material factors would have<br />
9. For most actors, a norm would mean “collective expectations <strong>for</strong> the proper behavior of actors<br />
with a given identity.” See Peter J. Katzenstein, “Introduction,” in Katzenstein, ed., The Culture of<br />
National Security: Norms and Identity in World Politics (New York: Columbia University Press, 1996),<br />
p. 5. Some actors, however, adhere to a norm simply because it is common practice, or out of fear<br />
of retribution should they breach it. See Robert Axelrod, “An Evolutionary Approach to Norms,”<br />
American Political Science Review, Vol. 80, No. 4 (December 1986), pp. 1095–1111.<br />
10. For similar arguments, see Zacher, “The Territorial Integrity Norm”; and Fazal, “The Origins<br />
and Implications of the Territorial Sovereignty Norm.”<br />
11. Charles Tilly, Coercion, Capital, and European States, 990–1992 (Cambridge, Mass.: Blackwell,<br />
1992), p. 203.<br />
12. I use the term “homeland territory” to control <strong>for</strong> cases of decolonization and the breakup of<br />
empires, which involve territorial changes but are not relevant to the norm of ªxed borders. The<br />
ten cases are Israel from Syria, 1967; Israel from Jordan, 1967; Israel from Egypt, 1967; Iran from<br />
the United Arab Emirates, 1971; India from Pakistan, 1971; Libya from Chad, 1973; Israel from<br />
Syria, 1973; Turkey from Cyprus, 1973; China from South Vietnam, 1974; and Armenia from<br />
Azerbaijan, 1991–94 (a border case between <strong>for</strong>eign conquest and secession). I do not consider the<br />
U.S. occupation in Iraq as one of these cases, because it is not intended to be permanent occupation<br />
or annexation. Should the occupation continue much longer, however, it might be considered as<br />
such.<br />
13. Donald L. Horowitz, Ethnic Groups in Conºict (Berkeley: University of Cali<strong>for</strong>nia Press, 1985),<br />
pp. 281–288.<br />
14. Carl Kaysen, “Is War Obsolete? A Review Essay,” International Security, Vol. 14, No. 4 (Spring<br />
1990), pp. 42–64.<br />
15. Moreover, John E. Mueller argues that even without nuclear weapons, war has become unthinkable<br />
because of its devastating costs <strong>for</strong> advanced industrial societies. See Mueller, Retreat<br />
from Doomsday: The Obsolescence of Major War (Rochester, N.Y.: University of Rochester Press, 1996).<br />
16. Jeffrey Herbst, States and Power in Africa: Comparative Lessons in Authority and Control (Princeton,<br />
N.J.: Princeton University Press, 2000).
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 143<br />
Figure 1. Conquest and Annexation of Homeland Territory per Contiguous Dyad,<br />
1820–2000<br />
SOURCES: For data on the number of conquests and annexations, see Jaroslav Tir, Philip<br />
Schafer, Paul F. Diehl, and Gary Goertz, “Territorial Changes, 1816–1996,” Conflict Management<br />
and Peace Science, Vol. 16, No. 1 (Spring 1998), pp. 89–97 (adjusted to include<br />
1997–2000 data and to fit author’s definitions). For data on contiguous dyads, see Douglas<br />
M. Stinnett, Jaroslav Tir, Philip Schafer, Paul F. Diehl, and Charles Gochman, “The Correlates<br />
of War Project Direct Contiguity Data, Version 3,” Conflict Management and Peace<br />
Science, Vol. 19, No. 2 (Summer 2002), pp. 58–66.<br />
been unlikely to produce the now internationally accepted norm of ªxed borders.<br />
The idea that conquest does not entitle one state to annex the territory of<br />
another has its roots in the eighteenth century, with the notion of popular sovereign<br />
rights. But only at the conclusion World War I, with U.S. President<br />
Woodrow Wilson’s promotion of the concept of self-determination, did this<br />
idea begin to show tangible results. 17 The 1919 Covenant of the League of<br />
Nations and the Kellogg-Briand Pact of 1928 both sought to lend strength<br />
to the norm of ªxed borders, as did changes in the laws of war, which in-<br />
17. Wilson’s words in 1919 to the Senate Committee on Foreign Relations expressed precisely the<br />
idea of ªxed borders: “I understand that article [article 10 of the League covenant] to mean that no<br />
nation is at liberty to invade the territorial integrity of another. Its territorial integrity is not destroyed<br />
by armed intervention; it is destroyed by the retention of territory, by taking territory<br />
away from it.” “A Conversation with Members of the Foreign Relations Committee,” August 19,<br />
1919, in Arthur S. Link, The Papers of Woodrow Wilson, quoted in Fazal, “The Origins and Implications<br />
of the Territorial Sovereignty Norm,” p. 25.
International Security 31:3 144<br />
creasingly viewed belligerent occupation to be acceptable only if it was<br />
temporary. 18<br />
The 1945 United Nations charter and many of its subsequent resolutions<br />
gave the norm powerful legal backing. As General Assembly resolution<br />
2625 of 1970 declares, “No territorial acquisition resulting from the threat<br />
of use of <strong>for</strong>ce shall be recognized as legal.” 19 The charter does not prohibit<br />
war or the use of <strong>for</strong>ce per se: under certain circumstances, notably <strong>for</strong> purposes<br />
of self-defense and collective security, war is permissible. It does,<br />
though, prohibit territorial annexation, even as a result of a just war (i.e., a war<br />
of self-defense). 20<br />
During the Cold War, U.S. hegemony over the Western world played an important<br />
role in the institutionalization of the norm of ªxed borders, as well as<br />
in its promotion and en<strong>for</strong>cement. 21 The Soviet Union played a similar role in<br />
its sphere of inºuence. Despite frequently intervening in other states’ affairs<br />
and sending troops overseas, both superpowers refrained from territorial aggrandizement<br />
and did not annex any territory after 1945. In essence, both<br />
found it beneªcial to safeguard the international territorial status quo. 22 In<br />
some cases, such as the 1956 Suez crisis, they even intervened to prevent other<br />
states from violating the norm of ªxed borders. The norm, however, is not<br />
merely a Cold War artifact. 23 Since the collapse of the Soviet Union, no state<br />
has acquired territory through <strong>for</strong>ce (Armenia’s role in Nagorno-Karabakh is<br />
ambiguous). The response of the international community to violations of the<br />
norm, such as the nonrecognition of Israel’s 1967 conquests, and the military<br />
rollback of Iraq following its invasion of Kuwait underscore the strength of the<br />
norm. Indeed, as U.S. President Bill Clinton noted, “[The current] era does not<br />
reward people who struggle in vain to redraw borders with blood.” 24<br />
18. On the development of these ideas, see Sharon Korman, The Right of Conquest: The Acquisition<br />
of Territory by Force in International Law and Practice (Ox<strong>for</strong>d: Clarendon, 1996), pp. 133–199; and<br />
Zacher, “The Territorial Integrity Norm,” pp. 216–221.<br />
19. United Nations General Assembly, “Declaration on Principles of International Law Concerning<br />
Friendly Relations and Cooperation among States in Accordance with the Charter of the<br />
United Nations,” Resolution 2625, October 24, 1970. In addition, some regional organizations emphasize<br />
the sanctity of borders in their conventions and resolutions. For the institutionalization of<br />
the norm in the United Nations and regional organizations, see Korman, The Right of Conquest,<br />
pp. 199–214; and Zacher, “The Territorial Integrity Norm,” pp. 221–223.<br />
20. For this interpretation, see Korman, The Right of Conquest, pp. 209–214.<br />
21. Fazal, “The Origins and Implications of the Territorial Sovereignty Norm.”<br />
22. John Lewis Gaddis, “The Long Peace: Elements of Stability in the Postwar International System,”<br />
International Security, Vol. 10, No. 4 (Spring 1986), pp. 99–142.<br />
23. Michael C. Desch, “War and Strong States, Peace and Weak States?” International Organization,<br />
Vol. 50, No. 2 (Spring 1996), pp. 237–268, could be interpreted this way.<br />
24. William Jefferson Clinton, “Remarks by the President in Greeting to the Peoples of Pakistan,”<br />
Islamabad, Pakistan, March 25, 2000.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 145<br />
Africa’s borders are particularly intriguing. Despite the arbitrariness with<br />
which many state borders in Africa were drawn, they have remained largely<br />
ªxed. 25 From its inception in 1963, the Organization of African Unity (OAU)<br />
has endorsed the norm in accordance with the principle of preserving the colonial<br />
territorial status quo. 26 In practice, as Jeffery Herbst notes, “the vast majority<br />
of [African borders] have remained virtually untouched since the late<br />
1800s, when they were ªrst demarcated.” The OAU’s determination to uphold<br />
the norm was demonstrated, <strong>for</strong> instance, in the 1967–70 civil war in Nigeria,<br />
when the organization sought to prevent Biafra’s attempts to secede. 27<br />
Common sense might suggest that the norm of ªxed borders should increase<br />
the stability and security of states. In the next two sections, however, I<br />
discuss how it can have the opposite effect, particularly in regions that comprise<br />
mainly weak states. The norm is likely to perpetuate state weakness, and<br />
weak states with ªxed borders are often a source of international conºict.<br />
Fixed Borders and State Weakness<br />
Below are four hypotheses regarding the role of ªxed borders in promoting<br />
conºict between neighboring states. Figure 2 offers a graphic illustration of<br />
this process.<br />
Hypothesis 1: Fixed borders can perpetuate or exacerbate the weakness of<br />
already weak states.<br />
Hypothesis 2: Weak states in a ªxed-borders world can create conditions<br />
that can give rise to violent internal conºicts.<br />
Hypothesis 3: Refugee movements, insurgencies, and kin connections<br />
across international borders can cause civil conºicts in weak states in a<br />
25. Saadia Touval, The Boundary Politics of Independent Africa (Cambridge, Mass.: Harvard University<br />
Press, 1972).<br />
26. Michel-Cyr Djiena Wembou, “The OAU and International Law,” in Yassin El-Ayouty, ed., The<br />
Organization of African Unity after Thirty Years (Westport, Conn.: Praeger, 1994), p. 16.<br />
27. Jeffery Herbst, “The Creation and Maintenance of National Boundaries in Africa,” International<br />
Organization, Vol. 43, No. 4 (Autumn 1989), pp. 683–687, at p. 673. See also Herbst, States and Power<br />
in Africa; Jeremy M. Weinstein, “Africa’s ‘Scramble <strong>for</strong> Africa’: Lessons of a Continental War,”<br />
World Policy, Vol. 17, No. 2 (Summer 2000), pp. 11–20; and Robert H. Jackson, Quasi-states: Sovereignty,<br />
International Relations, and the Third World (New York: Cambridge University Press, 1990),<br />
pp. 87–90. The only case that possibly contradicts the norm of ªxed borders is Morocco’s annexation<br />
of the Western Sahara in 1976. Note that the African norm against changing borders goes beyond<br />
even that of the wider international community by prohibiting secessions, as well as<br />
conquests. I consider this only a variant of the international norm of ªxed borders, not a separate<br />
phenomenon. From the point of view of the individual state, moreover, this norm is still “international,”<br />
regardless of whether it originates in the region or beyond it.
International Security 31:3 146<br />
Figure 2. Potential Effects of Fixed Borders on Weak States<br />
ªxed-borders world to spill over their borders and become international<br />
conºicts and possibly full-ºedged wars.<br />
Hypothesis 4: State weakness promotes the possibility of international<br />
conºict by creating opportunities <strong>for</strong> neighbors to intervene to exploit the<br />
weak states economically or politically.<br />
In regions where states lack sociopolitical strength, the norm of ªxed borders<br />
may perpetuate or exacerbate state weakness because it deprives the state<br />
of a key historical factor in state building: an external threat to its borders and,<br />
in some instances, to its very survival. I deªne “sociopolitical strength” as the<br />
state’s capacity to maintain a monopoly over the legitimate use of <strong>for</strong>ce, its<br />
ability to govern the population (including extracting revenues through taxes<br />
and providing public goods), and its ability to maintain a reasonable level of<br />
social cohesiveness and identiªcation of its residents with the state. 28<br />
Fixed borders do not make states weak, but they can make those that are already<br />
weak even weaker by (1) denying them the incentives and coercive capabilities<br />
that have traditionally accompanied the threat of territorial war and<br />
28. For an example of a similar use of state strength, see Kalevi J. Holsti, The State, War, and the<br />
State of War (Cambridge: Cambridge University Press, 1996), pp. 82–98. See also Barry Buzan, People,<br />
States, and Fear: An Agenda <strong>for</strong> International Security Studies in the Post–Cold War Era, 2d ed.<br />
(Boulder, Colo.: Lynne Rienner, 1991), pp. 57–111.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 147<br />
opportunities <strong>for</strong> territorial expansion; (2) supplying counterincentives to state<br />
building; and (3) denying them the mechanism that weeded out weak states<br />
and bolstered stronger ones.<br />
territorial war and state building<br />
Like states in early modern Europe, many countries in the developing world<br />
are confronting the daunting challenge of building strong, viable states. The<br />
<strong>for</strong>mer, however, faced different external pressures and incentives than do<br />
states today. 29 It is these differences that help to explain the perpetuation of<br />
state weakness in the contemporary international system. Contrary to the conventional<br />
wisdom, many European states in the past were confronted with<br />
conditions similar to those faced by weak states today, including a lack of effective<br />
institutions and a socially cohesive citizenry. The relative ethnic and<br />
linguistic homogeneity of contemporary European states is a product, rather<br />
than a precondition, of the state-building process. In 1640, <strong>for</strong> instance,<br />
Brandenburg-Prussia was a disjointed collection of territories with deep religious<br />
and regional cleavages; it could claim no collective identity; and it was<br />
loosely controlled by a weak center. A century later, its successor—the kingdom<br />
of Prussia—had become a stronger, socially cohesive state. 30<br />
International war, and territorial war in particular, <strong>for</strong>ced this trans<strong>for</strong>mation<br />
<strong>for</strong> two reasons—both of which involved territory. First, the development<br />
of large, centrally supplied and ªnanced standing armies profoundly increased<br />
the need <strong>for</strong> states to extract resources from their societies, and at the<br />
same time greatly enhanced their ability to coerce their populations into providing<br />
these resources. This need, in turn, required the creation of new, more<br />
efªcient bureaucracies to manage the collection of taxes and the distribution of<br />
resources. Where these ef<strong>for</strong>ts succeeded, they strengthened the state and, in<br />
the long run, served much more than the narrow military function <strong>for</strong> which<br />
they were created. 31<br />
29. Mohammed Ayoob, The Third World Security Predicament: State Making, Regional Conºict, and<br />
the International System (Boulder, Colo.: Lynne Rienner, 1995), pp. 21–46, 193–194.<br />
30. See, <strong>for</strong> example, Samuel E. Finer, “State- and Nation-Building in Europe: The Role of the Military,”<br />
in Charles Tilly, ed., The Formation of National States in Western Europe (Princeton, N.J.: Princeton<br />
University Press, 1975), pp. 134–144. France is another example. See Eugen Weber, Peasants into<br />
Frenchmen: The Modernization of Rural France, 1870–1914 (Stan<strong>for</strong>d, Calif.: Stan<strong>for</strong>d University<br />
Press, 1979).<br />
31. Otto Hintze, The Historical Essays of Otto Hintze, Felix Gilbert, ed. (New York: Ox<strong>for</strong>d University<br />
Press, 1975); Charles Tilly, “Reºections on the History of European State-Making,” in Tilly, The<br />
Formation of National States in Western Europe, pp. 3–83; Tilly, Coercion, Capital, and European States,<br />
990–1992; Brian M. Downing, The Military Revolution and Political Change: Origins of Democracy<br />
and Autocracy in Early Modern Europe (Princeton, N.J.: Princeton University Press, 1992); and<br />
Thomas Ertman, Birth of the Leviathan: Building States and Regimes in Medieval and Early Modern Europe<br />
(New York: Cambridge University Press, 1997).
International Security 31:3 148<br />
Second, a common external threat <strong>for</strong>ged a common internal identity. Belonging<br />
to a particular group (the in-group) can help individuals fulªll some<br />
of their psychological needs (e.g., the need <strong>for</strong> self-esteem and affection). Yet<br />
membership in one group may mean nonmembership in another (the outgroup).<br />
Hostility toward the out-group often increases in-group cohesiveness.<br />
32 The more threats that are posed to both its existence and its boundaries,<br />
the more likely the group is to develop internal cohesion. 33 Opportunities <strong>for</strong><br />
territorial expansion through warfare can also strengthen society’s bonds and<br />
enhance its self-perception. 34<br />
Wars in early modern Europe were effective in promoting state building precisely<br />
because they threatened the seizure of territory. Wars that do not seek<br />
territorial gains do not endanger the survival of the state, nor do they endanger<br />
the population as a whole. Portraying an attack as an assault on the ingroup<br />
is much harder when only some segments of the population feel threatened.<br />
Nonterritorial wars are less likely to result in public and elite acceptance<br />
of a heavier ªscal burden, increased central control, and greater social cohesion.<br />
35 For the same reasons, civil war is unlikely to signiªcantly strengthen<br />
the state. 36 Developing states today thus lack the very incentives that allowed<br />
most European states to succeed in the state-building project. 37<br />
To be sure, the mechanism that causes territorial pressures to result in stronger<br />
states is not deterministic. Although wars over territory provide states<br />
32. See, <strong>for</strong> example, Henry Tajfel and John C. Turner, “The Social Identity Theory of Intergroup<br />
Behavior,” in Stephen Worchel and William G. Austin, eds., Psychology of Intergroup Relations, 2d<br />
ed. (Chicago: Nelson-Hall, 1986), pp. 7–24. For a sociological viewpoint, see Lewis A. Coser, The<br />
Functions of Social Conºict (New York: Free Press, 1956), pp. 87–110.<br />
33. Coser, The Functions of Social Conºict, pp. 87–110; Janice Gross Stein, “Image, Identity, and the<br />
Resolution of Violent Conºict,” in Chester A. Crocker, Fen Osler Hampson, and Pamela Aall, eds.,<br />
Turbulent Peace: The Challenges of Managing International Conºict (Washington, D.C.: United States<br />
Institute of Peace Press, 2001), pp. 189–208; and Desch, “War and Strong States, Peace and Weak<br />
States?” pp. 247–248.<br />
34. See Jack S. Levy, “Domestic Politics and War,” Journal of Interdisciplinary History, Vol. 18, No. 4<br />
(Spring 1988), pp. 653–673; and Max Weber, “The Prestige and Power of the Great Powers,” in<br />
H.H. Gerth and C. Wright Mills, eds., From Max Weber: Essays in Sociology (New York: Ox<strong>for</strong>d University<br />
Press, 1946), pp. 159–162.<br />
35. Here I diverge from the argument of Herbst, “War and the State in Africa”; and Desch, “War<br />
and Strong States, Peace and Weak States?” Both authors argue that the absence of international<br />
wars (of any kind) is what weakens third world states, whereas I maintain that only the absence of<br />
a particular kind of war (i.e., wars of territorial conquest) has this effect.<br />
36. See Cameron G. Thies, “State Building, Interstate and Intrastate Rivalry: A Study of Postcolonial<br />
Developing Country Extractive Ef<strong>for</strong>ts, 1975–2000,” International Studies Quarterly, Vol. 48,<br />
No. 1 (March 2004), pp. 53–72.<br />
37. In most cases in which a state in the post–World War II era faced a real threat to its territorial<br />
integrity and its survival—such as in Israel, South Korea, and Taiwan (though not ofªcially a<br />
state)—it engaged relatively successfully in state building. See, <strong>for</strong> instance, Joel S. Migdal, Strong<br />
Societies and Weak States: State-Society Relations and State Capabilities in the Third World (Princeton,<br />
N.J.: Princeton University Press, 1988), pp. 142–205. Although some states that faced territorial<br />
threats did not grow stronger (e.g., Jordan and Pakistan), they were not wiped off the map, as they<br />
probably would have been in previous eras.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 149<br />
with incentives to become stronger, the decision to engage in state building<br />
rests with policymakers. These individuals often do not correctly interpret<br />
their strategic environment, or they are unable to engage in the costly project<br />
of state building due to objections from powerful actors in their societies. Still,<br />
war is often the ªnal arbiter between correct (self-empowering) decisions and<br />
wrong (self-defeating) ones. In the past, states that built strong institutions and<br />
mobilized their societies not only survived but thrived. Those that did not<br />
eventually crumbled. 38 By contrast, states today do not “die,” because this selection<br />
mechanism no longer operates, thus practically guaranteeing the survival<br />
of weak states. 39<br />
juridical statehood, moral hazard, and state building<br />
By embracing the idea of ªxed borders, the international community agreed,<br />
in essence, to preserve the shell of the state regardless of its overall weakness—<br />
a phenomenon that Robert Jackson called “juridical statehood.” 40 Juridical<br />
statehood provides rulers in already weak states with strong incentives to<br />
abandon investment in and control of their peripheries, thus making the state<br />
even weaker. The logic of Jackson’s argument is simple: to achieve a minimally<br />
strong state, rulers must invest resources and take signiªcant political risks by<br />
threatening the interests of entrenched elites. This includes creating a monopoly<br />
over the means of violence (i.e., policing) and developing investment strategies<br />
to meet the population’s basic needs (e.g., building infrastructure and<br />
establishing viable educational, judicial, and bureaucratic institutions). In the<br />
absence of such steps, rulers will have great difªculty convincing their citizens<br />
of the legitimacy and authority of the state.<br />
Yet juridical statehood guarantees states membership in the community of<br />
nations, with all the advantages this entails. It also guarantees that as long as a<br />
ruler is able to control the capital, his status as head of state is assured. It is,<br />
Jackson concluded, “like an insurance policy: The policy holders, and consequently<br />
the main beneªciaries, are the rulers and regimes—not the people”—<br />
hence, the moral hazard. In the absence of a juridical or military threat to the<br />
existence of a state and its territory, a ruler of a weak state has few incentives<br />
to engage in the costly and often dangerous enterprise of controlling the state’s<br />
38. Tilly, Coercion, Capital, and European States; and Hendrik Spruyt, The Sovereign State and Its Competitors<br />
(Princeton, N.J.: Princeton University Press, 1994), pp. 32–33. For discussion of a similar<br />
process in ancient China, see Victoria Tin-bor Hui, “Toward a Dynamic Theory of International<br />
Politics: Insights from Comparing Ancient China and Early Modern Europe,” International Organization,<br />
Vol. 58, No. 1 (January 2004), pp. 175–205.<br />
39. See Tanisha M. Fazal, “State Death in the International System,” International Organization, Vol.<br />
58, No. 2 (April 2004), p. 320.<br />
40. Robert H. Jackson, “Juridical Statehood in Sub-Saharan Africa,” Journal of International Affairs,<br />
Vol. 46, No. 1 (Summer 1992), pp. 1–16.
International Security 31:3 150<br />
periphery. 41 Moreover, well-trained armies and efªcient bureaucracies—two<br />
central institutions of state building—are not only expensive to create and<br />
maintain, but given their potential to compete <strong>for</strong> power, they may be perceived<br />
as potential threats to the ruler. In addition, rulers of weak states are<br />
likely to marginalize peripheral regions both politically and economically<br />
without fear of their seceding or being annexed. Neither secession nor annexation<br />
by a neighbor would gain support from an international community that<br />
highly values the territorial status quo. 42 The consequences, again, are a further<br />
weakening of the state and, in some cases, its complete disintegration—<br />
the ultimate <strong>for</strong>m of which is a collapsed state. Examples include, at various<br />
times, Afghanistan, Cambodia, Chad, Lebanon, Liberia, Sierra Leone, Somalia,<br />
and Zaire/DRC. 43<br />
The next section details the mechanisms that make many weak states in a<br />
world of ªxed borders constant sources of conºict and instability.<br />
Fixed Borders, Weak States, and International Conºict<br />
Insecurity and opportunism remain important factors in the instigation of<br />
conºict in regions with weak states. The mechanisms through which these two<br />
factors operate, however, are profoundly different from those typically discussed<br />
in the international relations literature, which has concentrated instead<br />
on the developed world, in general, and on great power relations, in particular.<br />
In the case of weak states, insecurity can lead to the spillover of civil conºict<br />
into neighboring states, while opportunism may be manifested through nonterritorial<br />
intervention. 44<br />
weak states, civil conflict, and spillover<br />
The spillover of internal strife is the ªrst mechanism through which weak<br />
states can become a source of international conºict. Since the end of World<br />
41. Jackson, Quasi-states, pp. 1–12, 40–59, 74–81.<br />
42. William Reno, Warlord Politics and African States (Boulder, Colo.: Lynne Rienner, 1998), pp. 1–<br />
40, at pp. 8–9.<br />
43. I. William Zartman, “Introduction: Posing the Problem of State Collapse,” in Zartman, ed., Collapsed<br />
States: The Disintegration and Restoration of Legitimate Authority (Boulder, Colo.: Lynne<br />
Rienner, 1995), pp. 1–11; René Lemarchand, “Patterns of State Collapse and Reconstruction in Central<br />
Africa: Reºections on the Crisis in the Great Lakes,” African Studies Quarterly, Vol. 1, No. 3<br />
(1997), http://web.africa.uº.edu/asq/v1/3/2.htm; and Reno, Warlord Politics and African States,<br />
pp. 45–79.<br />
44. Brian L. Job, “The Insecurity Dilemma: National, Regime, and State Securities in the Third<br />
World,” in Job, ed., The Insecurity Dilemma: National Security of Third World States (Boulder, Colo.:<br />
Lynne Rienner, 1992), pp. 17–19; and Steven R. David, “Explaining Third World Alignment,” World<br />
Politics, Vol. 43, No. 2 (January 1991), pp. 233–256.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 151<br />
War II, internal wars have been much more common than wars between<br />
states. 45 One study, <strong>for</strong> example, identiªed 126 such conºicts out of a total of<br />
164 wars fought from 1945 to 1995. 46 Another found that 92 of the 108 armed<br />
conºicts in the 1990s were wars between organized communal groups or between<br />
such groups and their governments. 47 Communal wars are a common<br />
feature in regions with relatively young states, such as Africa, Asia, and the<br />
Middle East. 48 That some regions are more prone to civil and communal war<br />
than others cannot be explained solely by the presence of ethnic or other minority<br />
groups. Although many states contain such groups, the distribution of<br />
civil/communal conºict remains uneven. The strength of the state often accounts<br />
<strong>for</strong> this disparity: the weaker the state, the more likely it is to be involved<br />
in a civil or communal war.<br />
The outbreak of communal ªghting can be the result of one of two processes,<br />
both of which are characteristic of weak states. The ªrst is “emerging<br />
anarchy”: that is, as a state’s ability to en<strong>for</strong>ce order and provide internal security<br />
decreases, the potential danger that one group poses to another increases.<br />
49 At the same time, the weakness of state authority encourages some<br />
groups to think they can achieve victory in a civil war or otherwise gain from<br />
predatory behavior. 50<br />
Second, the lack of institutional means <strong>for</strong> gaining legitimacy is inherent in<br />
weak states. To compensate <strong>for</strong> this deªciency without having to resort to the<br />
expensive practice of state building, leaders might choose to incite or promote<br />
internal conºict in hopes of riding an ethnic wave to political and economic<br />
gain. They might engage in divide-and-rule tactics or side explicitly or implicitly<br />
with one group against another—<strong>for</strong> instance, by branding members of a<br />
group as “<strong>for</strong>eign agents.” This practice is likely to be more common in an era<br />
45. Ted Robert Gurr, “Minorities and Nationalists: Managing Ethnopolitical Conºict in the New<br />
Century,” in Crocker, Hampson, and Aall, Turbulent Peace, pp. 163–188.<br />
46. Holsti, The State, War, and the State of War, p. 22.<br />
47. Peter Wallensteen and Margareta Sollenberg, “Armed Conºict, 1989–98,” Journal of Peace Research,<br />
Vol. 36, No. 5 (September 1999), pp. 593–606.<br />
48. Holsti, The State, War, and the State of War, pp. 19–21. Civil wars were common in Latin America<br />
as well, but their motives—at least during the Cold War era—were more often ideological.<br />
49. See Barry R. Posen, “The Security Dilemma and Ethnic Conºict,” Survival, Vol. 35, No. 1<br />
(Spring 1993), pp. 27–47; James D. Fearon, “Commitment Problems and the Spread of Ethnic<br />
Conºict,” in David A. Lake and Donald Rothchild, eds., The International Spread of Ethnic Conºict:<br />
Fear, Diffusion, and Escalation (Princeton, N.J.: Princeton University Press, 1998), pp. 115–125; David<br />
A. Lake and Donald Rothchild, “Containing Fear: The Origins and Management of Ethnic<br />
Conºict,” International Security, Vol. 21, No. 2 (Fall 1996), pp. 48–52; and Sarah K. Lischer, “Causes<br />
of Communal War: Fear and Feasibility,” Studies in Conºict and Terrorism, Vol. 22, No. 4 (November<br />
1999), pp. 331–335.<br />
50. On greed as a motive <strong>for</strong> violence in failed states, see Nelson Kasªr, “Domestic Anarchy, Security<br />
Dilemmas, and Violent Predation: Causes and Failures,” in Robert I. Rotberg, ed., <strong>When</strong> States<br />
Fail: Causes and Consequences (Princeton, N.J.: Princeton University Press, 2004), pp. 53–76.
International Security 31:3 152<br />
of ªxed borders because targeted groups have no exit option (i.e., secession or<br />
irredentism). It is also likely to further increase communal fear and communal<br />
opportunities <strong>for</strong> predation. Thus, the weaker the state, the greater is the likelihood<br />
of civil violence. 51<br />
Civil conºict has the potential to lead to international conºict in two ways, 52<br />
both of which are related to fear and insecurity. The ªrst involves refugee<br />
ºows and cross-border insurgency. Refugee ºows are often the vehicle through<br />
which internal ªghting spreads to neighboring countries. Internal wars, most<br />
notably ethnic conºicts, tend to produce more refugees than do interstate<br />
conºicts. 53 These refugees, especially those living in camps close to the borders<br />
of their homeland who enjoyed relatively high group cohesion prior to ºeeing,<br />
may engage in cross-border inªltrations and attacks. <strong>When</strong> their home country<br />
retaliates, the conºict can escalate into a full-scale international war.<br />
Refugee ºows are more likely to produce international conºict when weak<br />
states are involved. The governments of such states lack the capacity either to<br />
resettle the refugees (and thus provide them with incentives to integrate into<br />
local populations rather than to continue to ªght against their home countries)<br />
or to <strong>for</strong>ce them to abandon their armed struggle to avoid retaliation by their<br />
state of origin. 54 The case of Palestinian refugees in Lebanon is a classic example:<br />
the Lebanese government, too weak to integrate the refugees, to <strong>for</strong>ce its<br />
authority over them, or to effectively counter Israeli retaliation against attacks<br />
by the Palestinian Liberation Organization, could do little to prevent Israel’s<br />
1982 invasion.<br />
The second way in which civil strife can lead to international war is through<br />
the “kin-country syndrome”: that is, when ethnic afªliations do not corre-<br />
51. Stein, “Image, Identity, and the Resolution of Violent Conºict,” p. 193; and Lake and<br />
Rothchild, “Containing Fear,” pp. 53–56. Of course, protracted civil war might further magnify the<br />
state’s weakness. Yet cases in which a strong state becomes embroiled in civil war and emerges<br />
from it as a weak state are few. Yugoslavia is perhaps one, but it was considerably weak even be<strong>for</strong>e<br />
the war.<br />
52. Michael E. Brown, “Introduction,” in Brown, ed., The International Dimensions of Internal<br />
Conºict (Cambridge, Mass.: MIT Press, 1996), pp. 1–32.<br />
53. Myron Weiner, “<strong>Bad</strong> <strong>Neighbors</strong>, <strong>Bad</strong> Neighborhoods: An Inquiry into the Causes of Refugee<br />
Flows,” International Security, Vol. 21, No. 1 (Summer 1996), pp. 5–42. See also Job, “The Insecurity<br />
Dilemma,” pp. 3–7; and Kathleen Newland, “Ethnic Conºict and Refugees,” in Michael E. Brown,<br />
ed., Ethnic Conºict and International Security (Princeton, N.J.: Princeton University Press, 1993),<br />
pp. 143–163.<br />
54. See especially Sarah K. Lischer, Dangerous Sanctuaries: Refugee Camps, Civil War, and the Dilemmas<br />
of Humanitarian Aid (Ithaca, N.Y.: Cornell University Press, 2005), pp. 28–31; but also Ted<br />
Robert Gurr, “The Internationalization of Protracted Communal Conºicts since 1945: Which<br />
Groups, Where, and How?” in Manus I. Midlarsky, ed., The Internationalization of Communal Strife<br />
(London: Routledge, 1992), pp. 4–5; and I. William Zartman, “Internationalization of Communal<br />
Strife: Temptations and Opportunities of Triangulation,” in Midlarsky, The Internationalization of<br />
Communal Strife, pp. 27–42.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 153<br />
spond to international borders, and when ethnic groups in one country become<br />
alarmed by the grievances of their brethren across the border. 55 This<br />
syndrome can increase tensions between neighboring states and ultimately<br />
lead to an international war or intervention. 56 Turkey’s 1974 intervention in<br />
Cyprus, <strong>for</strong> instance, could be considered a consequence of the kin-country<br />
syndrome.<br />
The potential <strong>for</strong> this syndrome to occur exists everywhere. Yet it is much<br />
more likely to emerge in weak states, because the people have less afªnity <strong>for</strong><br />
those states, and because these states are unlikely to achieve greater social cohesiveness<br />
in the absence of external territorial pressures.<br />
opportunism and greed<br />
The second mechanism through which weak states can spur international<br />
conºict is opportunity and greed. In a world of ªxed borders, opportunities <strong>for</strong><br />
territorial revisionism are greatly reduced. Still, in some cases greed can play a<br />
signiªcant role. Although states cannot legally annex a neighbor’s territory,<br />
they may be able to exploit this territory <strong>for</strong> economic or political gain, including<br />
regime change. 57 Weak states are more likely to be victims of neighboring<br />
states’ greed <strong>for</strong> two reasons. First, neighbors are more likely to ªnd potential<br />
allies in weak states than in strong ones. Second, many sociopolitically weak<br />
states are also militarily weak, because maintaining a strong, cohesive military<br />
and collecting taxes <strong>for</strong> this purpose represent huge challenges. 58 Examples of<br />
states that have pursued predatory policies at the expense of a weak neighbor<br />
include Israel and Syria. Since the 1970s, both countries have sought (sometimes<br />
successfully) to exert control over Lebanese politics and (in Syria’s case)<br />
to exploit Lebanon’s economy through direct military intervention. Never,<br />
however, has either country sought to annex Lebanese territory.<br />
The effect of these nonterritorial threats differs signiªcantly from that of territorial<br />
threats because the <strong>for</strong>mer are often perceived as threats to a particular<br />
55. Samuel P. Huntington, “The Clash of Civilizations,” Foreign Affairs, Vol. 72, No. 3 (Summer<br />
1993), pp. 35–39; Lemarchand, “Patterns of State Collapse and Reconstruction in Central Africa,”<br />
pp. 5–6; and Stuart J. Kaufman, “An ‘International’ Theory of Inter-ethnic War,” Review of International<br />
Studies, Vol. 22, No. 2 (April 1996), p. 153.<br />
56. Stephen M. Saideman, The Ties That Divide: Ethnic Politics, Foreign Policy, and International<br />
Conºict (New York: Columbia University Press, 2001), pp. 1–35, 203–222. For a counterargument,<br />
see Michael E. Brown, “The Causes and Regional Dimensions of Internal Conºict,” in Brown, The<br />
International Dimensions of Internal Conºict, pp. 603–627.<br />
57. On the proposition that territorial conquest was replaced by intervention <strong>for</strong> regime change,<br />
see Tanisha M. Fazal, “From Conquest to Intervention: State, Regime, and Leader Exit,” paper presented<br />
at the annual meeting of the International Studies Association, Honolulu, Hawaii, March 1–<br />
5, 2005.<br />
58. Buzan, People, States, and Fear, p. 113.
group within the weak state, not to the state as a whole. There<strong>for</strong>e they do not<br />
generate the usual in-group/out-group dynamics. As Lewis Coser notes, “The<br />
relations between outer conºict and inner cohesion does not hold true where<br />
internal cohesion be<strong>for</strong>e the outbreak of the conºict is so low that the group<br />
members have ceased to regard preservation of the group as worthwhile, or<br />
actually see the outside threat to concern ‘them’ rather than ‘us.’” 59 The next<br />
section analyzes the case of the war in Congo in light of the theory suggested<br />
here and the hypotheses derived from it.<br />
The Case of Congo<br />
International Security 31:3 154<br />
The war in Congo illustrates the mechanism through which adherence to the<br />
norm of ªxed borders can lead to international conºict. After offering a brief<br />
review of Congo’s postindependence history, I describe how the norm perpetuated<br />
and exacerbated the state’s weakness. I then show how this weakness<br />
triggered mechanisms that expanded the conºict between Congo and its<br />
neighbors.<br />
from independence to war<br />
<strong>When</strong> Congo achieved independence from Belgium in 1960, it displayed all<br />
the signs of a weak state: it suffered from regional fragmentation, a crisis in<br />
central authority, and a high level of political violence. 60 Political divisions<br />
along ethnic lines were prevalent, though this fact could be misleading. Ethnicity<br />
became the primary source of political mobilization in Congo not because<br />
of the salience of rigid ethnic identities, but because the state lacked legitimacy,<br />
a unifying idea around which the new nation could rally, and the institutions<br />
to promote it. 61 In 1965 Chief of the Army Mobutu Sese Seko seized power in a<br />
coup and established, in the state he renamed Zaire, what became one of<br />
Africa’s most enduring regimes. 62<br />
Throughout Mobutu’s rule, Zaire frequently interfered in the internal affairs<br />
of neighboring states, backing guerrilla movements in, <strong>for</strong> example, Angola,<br />
Burundi, Rwanda, and Uganda. Such actions often produced counterinterventions,<br />
including Angolan support of the Katangese rebellions in Shaba<br />
Province in the 1970s.<br />
59. Coser, The Functions of Social Conºict, pp. 93–95, at p. 93.<br />
60. Craw<strong>for</strong>d Young and Thomas Turner, The Rise and Decline of the Zairian State (Madison: University<br />
of Wisconsin Press, 1985), pp. 41–42.<br />
61. In fact, scholars of Congo note the considerable degree of ºuidity of Congolese ethnic identities<br />
at the time. See ibid., pp. 40–42; and John Breuilly, Nationalism and the State, 2d ed. (Chicago:<br />
University of Chicago Press, 1994), pp. 200–204.<br />
62. Young and Turner, The Rise and Decline of the Zairian State, pp. 42–43.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 155<br />
In 1996 an armed rebellion against Mobutu’s regime, led by Laurent Désiré<br />
Kabila of the Alliance of the Democratic Forces <strong>for</strong> the Liberation of Congo-<br />
Zaire (ADFL), erupted in Kivu Province. The ADFL, backed by Rwandan<br />
<strong>for</strong>ces and supported (ªnancially and politically) by most states in the region,<br />
rapidly defeated Congo’s armed <strong>for</strong>ces. In May 1997 Mobutu ºed into exile, allowing<br />
Kabila to take power and change the name of the country from Zaire to<br />
the Democratic Republic of Congo.<br />
The rebellion represented the ªrst phase in a protracted war that ended only<br />
in 2002. Soon after taking Kinshasa, Kabila began to alienate many of his <strong>for</strong>mer<br />
allies, both domestic and <strong>for</strong>eign. Lending support to a new rebellion in<br />
eastern Congo, Burundian, Rwandan, and Ugandan <strong>for</strong>ces undertook a massive<br />
intervention in August 1998 and pushed rapidly toward the Congolese<br />
capital. To save his collapsing regime, Kabila requested military support from<br />
Angola, Chad, Namibia, Sudan, and Zimbabwe. Their counterintervention<br />
prevented Kabila’s downfall, but the war raged on as <strong>for</strong>eign troops and local<br />
rebels established control over much of Congo’s vast territory. Casualties ran<br />
as high as 3.8 million dead as a direct or indirect result of the war. 63 The country’s<br />
already fragile economic situation and poor health conditions deteriorated<br />
even further.<br />
Laurent Kabila was assassinated by his bodyguard in January 2001.<br />
Replacing him was his son, Joseph Kabila. In July 2002 the DRC and Rwanda<br />
signed the Pretoria accord, which led to the withdrawal of all <strong>for</strong>eign troops<br />
from Congo and the creation of a transitional government that included rebels<br />
and opposition parties. What has been labeled “Africa’s Great War” reached its<br />
end. The ªghting between the Congolese army and various local militias<br />
continues, however, while armed clashes between political parties in the aftermath<br />
of the August 2006 elections illustrate the country’s ongoing vulnerability.<br />
Moreover, Rwanda has threatened renewed intervention should the<br />
Congolese government continue its support of the anti-Rwandan Hutu insurgency<br />
mounted from the DRC. The disastrous consequences of Africa’s Great<br />
War—including the huge death toll, tremendous human suffering, massive<br />
displacement, and signiªcant damage to Congo’s infrastructure will probably<br />
take decades to overcome. In addition, the country’s ongoing instability serves<br />
as a powerful impediment to prospects <strong>for</strong> a brighter future. 64<br />
63. This ªgure comes from a survey conducted by the International Rescue Committee in 2004; it<br />
includes military and civilian casualties, as well as war-related deaths caused directly by violence<br />
and indirectly as a consequence of famine and epidemics.<br />
64. Comprehensive sources on the war are still scarce. For a decent beginning, see John F. Clark,<br />
ed., The African Stakes of the Congo War (New York: Palgrave Macmillan, 2002).
International Security 31:3 156<br />
fixed borders and state weakness in congo<br />
Congo’s emergence from Belgian rule as a weak state is not surprising; most<br />
states are weak at this early stage. Like many other states in Africa and Asia,<br />
however, Congo has remained so; indeed, it is weaker today than it was in the<br />
1960s. The explanation <strong>for</strong> this ongoing weakness lies largely with the norm of<br />
ªxed borders as stipulated in hypothesis 1. This section examines Congo’s<br />
state-building ef<strong>for</strong>ts (or lack thereof) in ªve core areas: (1) monopoly over the<br />
legitimate use of violence; (2) revenue extraction; (3) state bureaucracy and<br />
institutions; (4) distribution of public goods; and (5) social cohesion and identiªcation<br />
with the state.<br />
monopoly over the legitimate use of <strong>for</strong>ce. Absent an external territorial<br />
threat, governments can af<strong>for</strong>d to have inefªcient, weak, and corrupt<br />
militaries. Writing about Zaire under President Mobutu, Edgar O’Ballance<br />
noted, “The main role of the armed <strong>for</strong>ces remained internal security, with little<br />
thought or energy going to national defense.” 65 Militaries of this kind require<br />
little investment or sacriªce by either the state or its citizens. Moreover, a<br />
weak military reduces the threat of a coup. In Zaire’s case, Mobutu systematically<br />
rechanneled resources intended <strong>for</strong> the army to his Presidential Guard,<br />
whose members were largely recruited from his home province, Equator. 66<br />
“The very importance of the military in Zairian politics,” argued Michael<br />
Shafer, resulted “in Mobutu’s extraordinary ef<strong>for</strong>ts to divide, control, manipulate,<br />
politicize, and otherwise deinstitutionalize and de-professionalize it.” 67<br />
<strong>When</strong> faced with internal rebellion, as <strong>for</strong> example in the two Shaba (Katanga)<br />
rebellions in the 1970s, Mobutu had to rely on troops from Morocco and<br />
France to compensate <strong>for</strong> Zaire’s deªcient army. 68 By the 1990s, rather than<br />
paying Zaire’s military <strong>for</strong>ces with state funds, Mobutu encouraged them to<br />
seek compensation through illegal activities, such as looting and kidnapping<br />
<strong>for</strong> ransom. 69<br />
Laurent Kabila’s record in this respect was no different. Despite the country<br />
being torn by civil and interstate war, the DRC’s military expenditures in 2003<br />
stood at only 1.4 percent of the country’s gross domestic product (GDP). 70 Yet<br />
65. Edgar O’Ballance, The Congo-Zaire Experience, 1960–98 (New York: St. Martin’s, 2000), pp. 114–<br />
115.<br />
66. David Shearer, “Africa’s Great War,” Survival, Vol. 41, No. 2 (June 1999), pp. 89–106.<br />
67. Quoted in Young and Turner, The Rise and Decline of the Zairian State, p. 274.<br />
68. Ibid., pp. 74–75, 248–275; and Thomas M. Callaghy, The State-Society Struggle: Zaire in Comparative<br />
Perspective (New York: Columbia University Press, 1984), pp. 206–209.<br />
69. Reno, Warlord Politics and African States, pp. 159–162; and René Lemarchand, “The Democratic<br />
Republic of Congo: From Failure to Potential Reconstruction,” in Rotberg, State Failure and State<br />
Weakness in a Time of Terror, pp. 40–41.<br />
70. Central Intelligence Agency, “The World Factbook, 2004.”
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 157<br />
even when <strong>for</strong>eign troops controlled much of Congo’s territory, Kabila’s government<br />
was still the recognized sovereign of the entire country. The norm of<br />
ªxed borders prevented external challenges to Congo’s juridical sovereignty,<br />
and no internal secession attempts could muster the international support<br />
needed to succeed. 71<br />
revenue extraction. The ability of the state to tax its citizenry is a sign of a<br />
strong state, and the revenues derived from taxation help to sustain it. Under<br />
Mobutu, Zaire’s tax collection system was inefªcient and corrupt. 72 Tax revenues<br />
throughout the 1970s and 1980s amounted to a mere 6–11 percent of the<br />
country’s GDP, of which only 25–35 percent consisted of taxes on income and<br />
capital revenues. In 1995 tax revenue plunged to about 5 percent of GDP; in<br />
2000 it fell to 4 percent. 73 Given that by the 1990s many economic transactions<br />
were being conducted outside the ofªcial Zairian economy, the actual tax burden<br />
on Zaire’s citizens was probably much smaller.<br />
Instead of tax revenues, Mobutu relied on short-term policies, external support,<br />
and Zaire’s increasing <strong>for</strong>eign debt to sustain the patrimonial network<br />
through which he controlled the state. In 1973, <strong>for</strong> example, he oversaw the<br />
state’s seizure of vast tracts of land and commercial enterprises owned by <strong>for</strong>eign<br />
nationals in a process called “Zairianization.” As Craw<strong>for</strong>d Young and<br />
Thomas Turner write, “The cement of clientage was access to resources. The<br />
sudden takeover of this huge zone of the economy offered a vast new pool of<br />
goods <strong>for</strong> patrimonial distribution to deserving members of the political<br />
class.” 74<br />
state bureaucracy and institutions. Early in his reign, Mobutu did<br />
take some steps toward building Zaire’s institutions. He founded the Popular<br />
Movement of the Revolution (known by the French acronym MPR) as Zaire’s<br />
sole political party in an attempt to generate legitimacy <strong>for</strong> his regime and provide<br />
it with the organizational tools needed to penetrate the civil society. In the<br />
late 1960s, Mobutu sought to use the MPR to expand his institutional control<br />
over the country, including the education system (hitherto controlled by the<br />
Church), the military, and the regional authorities. 75<br />
71. Jeffrey Herbst, “Let Them Fail: State Failure in Theory and Practice—Implications <strong>for</strong> Policy,”<br />
in Rotberg, <strong>When</strong> States Fail, pp. 306–308; and Reno, Warlord Politics and African States, pp. 172–173.<br />
72. Callaghy, The State-Society Struggle, pp. 194–204.<br />
73. For these data, see World Bank, Development Indicators, http://devdata.worldbank.org.ezp2<br />
.harvard.edu/dataonline. Compare this, <strong>for</strong> example, with tax revenues of 30–38 percent <strong>for</strong> the<br />
same years in France and 16–19 percent in the United States, which in the latter case are considered<br />
to be very low levels of taxation by Western standards.<br />
74. Young and Turner, The Rise and Decline of the Zairian State, p. 328. See also Reno, Warlord Politics<br />
and African States, p. 152; and Callaghy, The State-Society Struggle, pp. 191–192.<br />
75. It is at least plausible to assume, though, that it took Mobutu and his regime a few years to recognize<br />
the extent of the new norm of ªxed borders, and that only after the unequivocal position of
International Security 31:3 158<br />
The appearance of state building, though, is misleading. In fact, Mobutu’s<br />
policies and rhetoric were geared primarily toward achieving three aims. First,<br />
Mobutu aspired to be an African—or even a third world—leader, an aspiration<br />
that sometimes conºicted with his pro-Western policies and dependence on<br />
Western support and loans. His nationalist rhetoric and extravagant projects<br />
were instead meant to cast him in a more positive light in the eyes of other<br />
African leaders. Two public relations campaigns—the “Authenticity” campaign<br />
of 1971 and “Mobutism” in 1974—attempted to achieve this objective<br />
through the use of nothing more than cheap rhetoric. “Zairianization” in 1973<br />
and “Radicalization” in 1974 sought to fulªll the same aim through economic<br />
means. In this way, Mobutu believed that he could claim African leadership<br />
without abandoning his dependence on Western assistance. 76<br />
Second, Mubutu’s policies aimed at sustaining and enhancing his patrimonial<br />
network. Zaire, as Young and Turner write, possessed “a dual character:<br />
<strong>for</strong>mally institutionalized, in party and administration, but in<strong>for</strong>mally patrimonial<br />
and personal.” 77 By the mid-1980s, Zaire had become “an early<br />
modern leviathan, but a lame one.” 78 On the one hand, the state was highly<br />
centralized and authoritarian, much like the absolutist states of seventeenthand<br />
eighteenth-century Europe. On the other hand, its reach was extremely<br />
limited. The state of Zaire, as an entity separate from its ruler, did not exist in<br />
any meaningful sense. 79 By 1975 Mobutu had abandoned even the pretense of<br />
state building. Faced with a growing <strong>for</strong>eign debt and a declining revenue<br />
base (due to the falling price of copper and the colossal economic failure of<br />
Zairianization), “Mobutu rejected the pursuit of policies or the building of institutions<br />
that would have served the collective good. Instead, he consolidated<br />
the international community on the Nigerian civil war (in 1970) became clear did they fully grasp<br />
the new norm’s implications.<br />
76. “Authenticity” emphasized the need to replace colonial and Western names and institutions<br />
with authentic African ones. “Mobutism” was, in essence, the elevation of Mobutu’s policies and<br />
words over all others. “Zairianization” and “Radicalization” were economic policies that allowed<br />
the state to seize private assets belonging to <strong>for</strong>eign nationals and redistribute them among its<br />
supporters. See Young and Turner, The Rise and Decline of the Zairian State, pp. 54–71, 185–247, 326–<br />
362, at p. 327; Callaghy, The State-Society Struggle, pp. 233–276; Kevin C. Dunn, “Imagining<br />
Mobutu’s Zaire: The Production and Consumption of Identity in International Relations,” Millennium:<br />
Journal of International Studies, Vol. 30, No. 2 (June 2001), pp. 245–258; and Reno, Warlord Politics<br />
and African States, pp. 151–153.<br />
77. Young and Turner, The Rise and Decline of the Zairian State, p. 397. Max Weber uses the term<br />
“patrimonial” to denote a regime that is, <strong>for</strong> most practical purposes, an extension of the ruler’s<br />
household. See Weber, Economy and Society: An Outline of Interpretive Sociology (New York:<br />
Bedminster, 1968), pp. 231–232, 1010–1044.<br />
78. Callaghy, The State-Society Struggle, p. 409. See also Jean-Claude Willame, Patrimonialism and<br />
Political Change in the Congo (Stan<strong>for</strong>d, Calif.: Stan<strong>for</strong>d University Press, 1972), p. 2.<br />
79. Young and Turner, The Rise and Decline of the Zairian State, pp. 164–184.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 159<br />
his own authority by monopolizing resources,” ushering in an era of state decay<br />
and collapse. 80<br />
Third, Mobutu deferred from creating a stronger state apparatus because an<br />
efªcient bureaucracy, much like a strong army, could have served as an independent<br />
power base that might have threatened his rule, particularly in times<br />
of crisis. Instead, he chose to shrink Zaire’s state bureaucracy and rely on a network<br />
of clients (especially regional strongmen) to fulªll state functions. 81<br />
<strong>When</strong> Laurent Kabila took power in 1997, he “inherited less a state than a<br />
ªefdom. Normal state functions had been replaced by patronage.” 82 Like<br />
Mobutu, Kabila faced structural disincentives to state building, including the<br />
international community’s support <strong>for</strong> the territorial status quo and the moral<br />
hazard that attended its uncritical recognition of Congo’s juridical statehood.<br />
Indeed, Kabila’s brief reign revealed many of the same tendencies discussed<br />
above, such as constructing a clientele network instead of efªcient bureaucracies<br />
and playing ethnic groups against each other. 83 As a result, Laurent<br />
Kabila’s Congo remained “an institutional clone of its predecessor.” 84<br />
distribution of public goods. The ability of the state to provide its citizens<br />
with public goods is an important indicator of its overall strength.<br />
Mobutu’s government spent little on public goods, and when it did the focus<br />
was mostly on extravagant, highly visible, and often misguided projects—<strong>for</strong><br />
example, the building of the Ingha-Shaba Dam and construction of a highpower<br />
grid to transfer the energy produced by the dam to the Shaba mines,<br />
which actually could have relied on much cheaper energy sources. At the same<br />
time, state spending on social services shrank from 17.5 percent in 1972 to virtually<br />
nothing in 1992. This pattern of state expenditure points again to a policy<br />
aimed at gaining cheap legitimacy (both internal and external) at the<br />
expense of creating real, though politically less beneªcial, public goods. 85 In<br />
addition, an ever increasing portion of the state’s budget was put under the<br />
sole discretion of the president. These funds, which in 1992 amounted to 95<br />
percent of Zaire’s budget, were funneled either to Mobutu’s personal accounts<br />
or to his cronies.<br />
During this period, government spending targeted some parts of the coun-<br />
80. Ibid., pp. 71–77; and Weinstein, “Africa’s ‘Scramble <strong>for</strong> Africa,’” pp. 14–15.<br />
81. Reno, Warlord Politics and African States, pp. 149–152.<br />
82. Shearer, “Africa’s Great War,” p. 92.<br />
83. Reno, Warlord Politics and African States, pp. 172–173; and Lemarchand, “The Democratic Republic<br />
of Congo,” pp. 44–45.<br />
84. Lemarchand, “The Democratic Republic of Congo,” p. 52.<br />
85. Ibid., p. 154; and Callaghy, The State-Society Struggle, pp. 184–194.
International Security 31:3 160<br />
try while neglecting others. Because control of the capital city is the only requirement<br />
the international community maintains <strong>for</strong> recognizing state<br />
sovereignty, it was the only asset that Mobutu’s regime needed to invest in<br />
(apart from sustaining the stream of cash from Zaire’s natural resources<br />
through local strongmen). Between 1969 and 1972, <strong>for</strong> example, Kinshasa—<br />
home to about 6 percent of Zaire’s population—was allocated 31 percent of the<br />
budget <strong>for</strong> government-approved investment projects. Shaba Province received<br />
about 47 percent (all of which went to the mineral mining industry), despite<br />
being home to less than 13 percent of Zaire’s population. In contrast,<br />
Kivu Province, with approximately 15 percent of the population, was allocated<br />
only 1.5 percent of the budget. And Kasai Province, with almost 20 percent of<br />
the state’s population, received virtually nothing. 86 Had the government<br />
needed to enhance the legitimacy of the state and assure, in particular, the allegiance<br />
of inhabitants along Zaire’s borders (so as to prevent irredentism), the<br />
distribution of funds would have looked quite different.<br />
social cohesion and identification with the state. In many African<br />
states, social cohesion and identiªcation of the population with the state are<br />
tenuous at best. 87 Congo is no different, although determining the level of loyalty<br />
to the state is difªcult in the absence of multiparty elections or reliable<br />
polling. Moreover, with authoritarian regimes such as Mobutu’s Zaire, ascertaining<br />
whether resistance is directed at the dictator or at the state more generally<br />
can be extremely difªcult. Still, one can glimpse some signs of the low<br />
level of legitimacy the population of Zaire accorded their state. One was the<br />
outbreak of rebellions that explicitly sought secession from the state. Three<br />
such rebellions occurred in the Shaba region in the early 1960s and in 1977 and<br />
1978. A second indicator was the size of the in<strong>for</strong>mal (or parallel) economy.<br />
Smuggling, tax evasion, a robust black market, unregulated trade, and withdrawal<br />
to noncommercial agriculture can all be considered <strong>for</strong>ms of social disengagement<br />
from and disapproval of the state—and in Zaire, all were<br />
common practices that became more popular over time. 88 “The in<strong>for</strong>mal sector,”<br />
a U.S. State Department analysis concluded in 2005, “now dominates the<br />
86. For the marginalization of provinces beyond Kinshasa and the Copper Belt region in Shaba,<br />
see Young and Turner, The Rise and Decline of the Zairian State, pp. 78–91. For data, see ibid., pp. 82–<br />
83.<br />
87. See Naomi Chazan, “Patterns of State-Society Incorporation and Disengagement in Africa,” in<br />
Donald Rothchild and Chazan, eds., The Precarious Balance: State and Society in Africa (Boulder,<br />
Colo.: Westview, 1980), pp. 121–148.<br />
88. For the general argument, see Victor Azarya, “Civil Society and Disengagement in Africa,” in<br />
John W. Harbeson, Donald Rothchild, and Naomi Chazan, eds., Civil Society and the State in Africa<br />
(Boulder, Colo.: Lynne Rienner, 1994), pp. 83–100. For the case of Congo, see Janet MacGaffey,<br />
“Civil Society in Zaire: Hidden Resistance and the Use of Personal Ties in Class Struggle,” in<br />
Harbeson, Rothchild, and Chazan, Civil Society and the State in Africa, pp. 169–189.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 161<br />
[Congolese] economy.” 89 Writing in 1984, Thomas Callaghy noted a third sign:<br />
Mobutu had “occupied just about every possible ideological position in a<br />
whirl of legitimating eclecticism.” 90 Mobutu’s decision to focus much of his ef<strong>for</strong>ts<br />
on the campaigns described above highlights a basic lack of state legitimacy<br />
and low levels of public loyalty to the state. Although these signs are<br />
indirect measures, they should sufªce in cases where direct measures are<br />
unavailable.<br />
Kabila’s DRC, like Mobutu’s Zaire, was confronted with threats of external<br />
intervention, but none produced ef<strong>for</strong>ts at increasing social cohesion because<br />
none represented a territorial threat, let alone a threat to the survival of the<br />
state. As Coser notes, <strong>for</strong> an external threat to produce internal cohesiveness,<br />
that threat must be perceived as a threat to the entire country. 91 In the DRC, the<br />
external threat was viewed in ethnic terms: a threat to some, but an opportunity<br />
<strong>for</strong> others to change the internal balance of power, though not the DRC’s<br />
borders. There<strong>for</strong>e the 1998 invasion by <strong>for</strong>eign <strong>for</strong>ces, much like previous interventions,<br />
did not generate greater in-group cohesion. 92<br />
Mobutu’s Zaire was a weak state at its inception and remained so throughout<br />
his reign; the same was true of Laurent Kabila’s DRC. Although Mobutu’s<br />
and Kabila’s decisionmaking may not directly prove the relationship between<br />
state weakness and the norm of ªxed borders, it does point to the absence of<br />
concern <strong>for</strong> the survival of the state (as opposed to the regime) and to the viability<br />
of their country’s international boundaries. With the norm of ªxed borders<br />
in place, Congo was able to endure and remain territorially intact.<br />
state weakness, civil conflict, and spillover in central africa<br />
If the norm of ªxed borders perpetuates and exacerbates weakness in already<br />
weak states, then one should at least be able to expect that it would curb instances<br />
of external conºict. In the case of Congo, however, it had encouraged<br />
them. This section explores the proposition in hypotheses 2 and 3 that state<br />
weakness can lead to international conºict through the spillover of internal<br />
conºict into neighboring states. It examines the spread of ethnic war across the<br />
borders of Congo and its neighbors and explains why state weakness and<br />
ªxed borders facilitated this process and brought about intervention and international<br />
war.<br />
89. U.S. Department of State country background proªle at http://www.state.gov/r/pa/ei/bgn/<br />
2823.htm.<br />
90. Callaghy, The State-Society Struggle, pp. 410–414, at p. 412.<br />
91. Coser, The Functions of Social Conºict, pp. 93–95.<br />
92. On the way various ethnic groups in Congo allied themselves with the different external interveners,<br />
see Thomas Turner, “The Kabilas’ Congo,” Current History, May 2001, pp. 215–218.
International Security 31:3 162<br />
state weakness and civil conflict in congo. Emerging anarchy and<br />
the exclusion of certain ethnic groups <strong>for</strong> political purposes increase the likelihood<br />
that weak states will become involved in civil conºicts and internal wars.<br />
Both factors played crucial roles in Congo. Low-level, violent civil conºicts<br />
simmered throughout Mobutu’s rule, though usually out of the public eye<br />
(e.g., the Katangese rebellions in Shaba Province in the 1970s, which were part<br />
of an Angolan-assisted secessionist drive that had to be quelled by Belgian and<br />
Tunisian troops when it became clear that Zaire’s armed <strong>for</strong>ces were not up to<br />
the task).<br />
In the early 1990s, tensions among Zaire’s multiple ethnic groups erupted<br />
into full-ºedged internal wars. In Shaba Province, pro-Mobutu gangs drove<br />
hundreds of thousands of Kasai people from their homes. In North Kivu<br />
Province, approximately 10,000 Banyarwanda were killed and another 250,000<br />
became refugees. 93 In a strong state, the government and the military would<br />
have sought to deter such wars or at least step in once they had begun. But in<br />
Zaire, “three months and several thousand deaths [after the breakup of violence,]<br />
the authorities had done nothing.” 94 The internal wars, the Economist<br />
thus concluded, were “a symptom of Zaire’s chronic ungovernment.” 95<br />
This, however, is only part of the explanation. Another factor was the use of<br />
diversionary political tactics. The ethnic dimension of Zairian politics gained<br />
increasing salience as Mobutu lost popular legitimacy and as external powers<br />
repeatedly intervened to prevent secessionist attempts. Although Mobutu’s<br />
rhetoric played down Zaire’s ethnic divisions, in practice he came to rely increasingly<br />
on an inner circle of Ngbandi from Equator Province and Lingala<br />
speakers, like himself. This was especially true of the security apparatus and<br />
the government’s most sensitive ªnancial posts. 96 Throughout his reign,<br />
Mobutu employed divide-and-rule tactics, particularly in Kivu Province,<br />
where in the 1980s and 1990s he sought to portray the Banyarwanda as “<strong>for</strong>eigners”<br />
in an attempt to gain the support of ethnic groups that competed with<br />
the Banyarwanda <strong>for</strong> land and resources. In 1991 he signed a law that stripped<br />
the Banyarwanda people of their citizenship and right to hold ofªce. Jermaine<br />
McCalpin attributes this action to Mobutu’s desire “to create a scapegoat <strong>for</strong><br />
93. Lemarchand, “The Democratic Republic of Congo,” pp. 38–43; and Chris McGreal, “Zaire<br />
Buries the Victims of Hatred and Expediency,” Guardian, September 24, 1993. Banyarwanda is the<br />
collective term <strong>for</strong> groups that migrated from Rwanda and Burundi (ethnically both Hutu and<br />
Tutsi) to Congo, mostly in the precolonial era.<br />
94. McGreal, “Zaire Buries the Victims of Hatred and Expediency.”<br />
95. “Zaire: Folly by the Numbers,” Economist, August 7, 1993, p. 41.<br />
96. Young and Turner, The Rise and Decline of the Zairian State, pp. 152–157; and Lemarchand, “The<br />
Democratic Republic of Congo,” pp. 40–41.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 163<br />
Zaire’s many problems and to distract the attention of the populace from their<br />
real source of misery.” 97<br />
Laurent Kabila engaged in similar tactics, targeting in particular the<br />
Banyamulenge (i.e., ethnic Tutsi who are part of the larger Banyarwanda<br />
group) and accusing them of being spies <strong>for</strong> Rwanda. Kabila’s policy of exclusion<br />
also originated from a lack of alternatives to gain legitimacy in a weak<br />
state. By the end of 1997, wrote René Lemarchand, “the choice [Kabila] faced<br />
was either to hang on to his Rwandan protectors, and suffer an even greater<br />
loss of legitimacy, or to free himself of their embrace and face the consequences.”<br />
98 Still, Kabila chose not only to order all <strong>for</strong>eign troops out of the<br />
DRC, but also to encourage his army to take part in a massacre of hundreds of<br />
Tutsi living in Kinshasa. 99 He then tried to ride this wave of hatred to regain legitimacy.<br />
The strategy was successful because, at least in the short run,<br />
Kabila’s popularity did rise. 100 These new exclusionary policies, however, had<br />
tremendous consequences <strong>for</strong> the second Rwandan invasion.<br />
spillover of civil conflict: refugees and insurgencies. Congo was not<br />
the only weak state in the neighborhood, and its civil conºicts combined with<br />
those of neighboring states to produce an international conºagration in the<br />
mid-1990s. Chief among these states were Rwanda, Burundi, Uganda, and<br />
Angola. 101<br />
In the aftermath of the 1994 genocide in Rwanda—in which roughly 800,000<br />
Tutsi and moderate Hutu were slaughtered—massive numbers of Hutus ºed<br />
the country, fearing retribution at the hands of the government, which had<br />
come under the control of the Tutsi-dominated Rwandan Patriotic Front. The<br />
leadership of the Hutu militants fed this fear and encouraged the exodus. 102<br />
The Rwandan refugees settled in camps in Burundi, Tanzania, Uganda, and<br />
Zaire. In Zaire, 1 million to 1.5 million Rwandan Hutus settled in huge refugee<br />
camps in South and North Kivu, the easternmost provinces of the country, in<br />
close proximity to the Rwandan border. Among them were 50,000–100,000 <strong>for</strong>-<br />
97. Jermaine O. McCalpin, “Historicity of a Crisis: The Origins of the Congo War,” in Clark, The<br />
African Stakes in the Congo War, p. 46. See also Kisangani N.F. Emizat, “The Massacre of Refugees in<br />
Congo: A Case of UN Peacekeeping Failure and International Law,” Journal of Modern African<br />
Studies, Vol. 38, No. 2 (June 2002), pp. 166–167; and Reno, Warlord Politics and African States, p. 161.<br />
98. Lemarchand, “The Democratic Republic of Congo,” p. 45.<br />
99. Ibid.<br />
100. David Shearer, “Lines on a Map,” World Today, Vol. 45, No. 11 (November 1998), p. 295.<br />
101. All of these states are ranked among the weakest in the Failed States Index; their rankings<br />
would have been even worse in the early 1900s. See Fund <strong>for</strong> Peace, “Failed States Index”; and<br />
Lemarchand, “Patterns of State Collapse and Reconstruction in Central Africa,” pp. 7–10.<br />
102. Sarah K. Lischer, “Internal Conºict and International Contagion: Refugees, Rebels, and Humanitarian<br />
Aid,” paper presented at the annual meeting of the American Political Science Association,<br />
Boston, Massachusetts, August 29–September 1, 2002, pp. 13–14.
International Security 31:3 164<br />
mer soldiers from the Rwandan Armed Forces (known by its French acronym<br />
FAR) and members of the militant Hutu militia known as the Interahamwe,<br />
which was largely responsible <strong>for</strong> the genocide. These groups were highly organized<br />
and well armed, making them a classic example of a “state in exile.”<br />
The presence of the Hutu refugees in eastern Zaire and their domination by<br />
militant groups were likely the primary cause of both phases of the international<br />
intervention in this country. Under the leadership of the Interahamwe,<br />
“the refugee camps were turned into military bases from which regular crossborder<br />
incursions were launched in order to destabilize the new Rwandan<br />
government.” 103<br />
The exodus of refugees after a military defeat in a civil war is common, as is<br />
the presence among them of guerrillas and <strong>for</strong>mer soldiers from the losing<br />
side. These factors do not always result in border wars of incursion and retaliation,<br />
however. If the host state can prevent refugee camps from becoming<br />
bases <strong>for</strong> insurgency, a border war may be avoidable. The Rwandan Hutu refugees<br />
who ªlled the camps of western Tanzania, <strong>for</strong> example, were prevented<br />
from using their camps as a base <strong>for</strong> launching attacks on Rwanda by the<br />
Tanzanian military’s <strong>for</strong>ceful policing and sealing off of the border. Moreover,<br />
Tanzania extended citizenship to the refugees, thus reducing their level of resentment<br />
and increasing their chances of peacefully integrating into the local<br />
population. 104<br />
In sharp contrast, and despite similarities in the composition of the population<br />
and the prevailing economic conditions, 105 Zaire could not control its borders<br />
with Rwanda or the refugee camps. Despite repeated Rwandan requests,<br />
Mobutu not only refused to confront the Hutu militants but allowed (and possibly<br />
even directed) his army to assist them. Zaire’s army facilitated the arms<br />
trade of the Interahamwe, allowed ex-FAR <strong>for</strong>ces to make use of its military<br />
camps and headquarters, and was heavily involved in the militants’ extortion<br />
and suppression of the camps’ civilian refugees. 106 Mobutu met regularly<br />
with senior elements of the Hutu militants, including the <strong>for</strong>mer Rwandan<br />
army chief of staff. 107 Although Mobutu never tried seriously to disarm the<br />
103. Christian R. Manahl, “From Genocide to Regional War: The Breakdown of International<br />
Order in Central Africa,” African Studies Quarterly, Vol. 4, No. 1 (Spring 2000), http://web<br />
.africa.uº.edu/asq/v4/v4i1a2.htm. See also Lischer, Dangerous Sanctuaries, p. 82; Shearer, “Africa’s<br />
Great War,” pp. 90–92; and Emizet, “The Massacre of Refugees in Congo,” p. 165.<br />
104. Lischer, Dangerous Sanctuaries, pp. 91–111; and Herbst, States and Power in Africa, p. 238.<br />
105. See Lischer, Dangerous Sanctuaries, pp. 108–111.<br />
106. For more details, see Human Right Watch Arms Project, “Rwanda/Zaire: Rearming with Impunity:<br />
International Support of the Perpetrators of the Rwandan Genocide” (Washington, D.C.:<br />
Human Rights Watch, May 1995).<br />
107. Lischer, Dangerous Sanctuaries, p. 85.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 165<br />
Rwandan Hutu militia, the rapid crumbling of his army in the face of the 1996<br />
rebellion made clear that he would not have succeeded even if this had been<br />
an objective. Because the goal of the Hutu militants was to replace Rwanda’s<br />
Tutsi regime (and perhaps also continue with their genocide), the Rwandan<br />
government viewed the Hutu insurgency in Zaire as a grave threat. 108 Thus,<br />
Rwanda’s vice president, Paul Kagame, told diplomats in early 1996 that “if<br />
the international community was unable to stop the delivery of weapons to the<br />
ex-FAR and Interahamwe and the military training in the refugee camps, the<br />
Rwandan government could decide to take preventive military action.” 109<br />
Similar dynamics, though perhaps on a smaller scale, played out between<br />
Mobutu’s Zaire and Burundi, Uganda, and Angola. Refugee camps around<br />
Uvira in South Kivu Province contained 150,000 Burundian Hutu refugees and<br />
had been a breeding ground <strong>for</strong> rebels seeking to overthrow Burundi’s Tutsi<br />
minority government. Beginning in 1995, the rebel group National Council <strong>for</strong><br />
the Defense of Democracy mounted attacks using these camps as a rear base.<br />
Burundi’s backing of the 1996–97 AFDL war against Mobutu was largely a response<br />
to this threat. 110 Uganda shared similar concerns. Three of its armed<br />
opposition groups, most notably the Allied Democratic Forces (ADF), had also<br />
found safe haven in eastern Zaire. 111 Again, regardless of whether Mobutu was<br />
just allowing these groups to operate from Zairian soil or was actively assisting<br />
them, Uganda’s interests in sponsoring a friendlier regime in Kinshasa and<br />
the possibility <strong>for</strong> direct action in eastern Zaire against the rebels were clear.<br />
Uganda, which supported Kabila’s war to oust Mobutu, later received permission<br />
from Kabila to enter the DRC in pursuit of rebels threatening its borders<br />
and population. 112<br />
The case of Angola offers another example of intervention in a neighboring<br />
state that assisted a domestic foe. The National Union <strong>for</strong> the Total Independence<br />
of Angola (known by the Portuguese acronym UNITA) fought against the<br />
Angolan government from independence, in 1975, until 2002. Mobutu supplied<br />
the primary lifeline <strong>for</strong> UNITA both by allowing the smuggling of dia-<br />
108. Manahl, “From Genocide to Regional War,” pp. 1–4; Shearer, “Africa’s Great War,” pp. 90–92;<br />
Bruce Baker, “Going to War Democratically: The Case of the Second Congo War (1998–2000),” Contemporary<br />
Politics, Vol. 6, No. 3 (September 2000), p. 266; and Lemarchand, “Patterns of State Collapse<br />
and Reconstruction in Central Africa,” p. 2.<br />
109. Manahl, “From Genocide to Regional War,” p. 3. See also Shearer, “Africa’s Great War”; and<br />
Emizet, “The Massacre of Refugees in Congo,” p. 168.<br />
110. Lemarchand, “Patterns of State Collapse and Reconstruction in Central Africa,” p. 2; and<br />
Shearer, “Africa’s Great War,” p. 95.<br />
111. Lemarchand, “Patterns of State Collapse and Reconstruction in Central Africa ,” p. 2; and<br />
John F. Clark, “Explaining Ugandan Intervention in Congo: Evidence and Interpretations,” Journal<br />
of Modern African Studies, Vol. 39, No. 2 (June 2001), pp. 271–272.<br />
112. Clark, “Explaining Ugandan Intervention in Congo,” pp. 271–273.
International Security 31:3 166<br />
monds from Angola to pass through Zaire (estimated at $500 million annually)<br />
as well as the transfer of weapons bought with this money, and by permitting<br />
UNITA to establish its rear bases on Zairian territory. Zaire also served as a<br />
conduit <strong>for</strong> the movement of money from the United States and South Africa<br />
to be used by UNITA (until the early 1990s) against Angola’s socialist government.<br />
Mobutu himself amassed huge sums from these transactions by playing<br />
the middleman and extracting fees <strong>for</strong> his services. Not surprisingly, the<br />
Angolan government supported Kabila’s ef<strong>for</strong>t to oust the friend of its worst<br />
enemy—Mobutu. 113<br />
Laurent Kabila’s accession to power in 1996 changed the players but not the<br />
rules of the game. Congo (now the DRC) was still an extremely weak state, unable<br />
to prevent most of the attacks on its neighbors by rebels operating from its<br />
territory, an issue that Kabila chose not make a priority of his regime. Kabila’s<br />
lack of ef<strong>for</strong>t to disarm the Hutu militias operating in the DRC’s eastern provinces<br />
remained Rwanda’s largest concern and led to its decision to replace<br />
him—just as it had done with Mobutu. Rwanda’s second invasion of Congo in<br />
1998 and its selection of a local ally to lead the march on Kinshasa bore a striking<br />
resemblance to the 1996 invasion. 114 And, driven by a desire to eradicate<br />
Burundian Hutu rebels operating along Lake Tanganyika, Burundi participated<br />
in the invasion as well. 115<br />
Uganda’s ofªcial explanation of its decision to join the 1998 intervention<br />
stressed the need to ºush out guerrilla bases in eastern Congo that served the<br />
ADF in its raids across the Ugandan border. At the minimum, Kabila had done<br />
nothing to rid his country of these rebels. At the maximum, as the Ugandan<br />
government alleged, he allowed Sudan to assist in the arming and training of<br />
the ADF, with the hope of securing aid <strong>for</strong> his army as well. Uganda’s claim of<br />
ªnding a “smoking gun” pointing to Sudanese involvement in such activities<br />
is thus plausible, given the Sudanese leadership’s policy that any enemy of<br />
Uganda was a friend of Sudan and because Uganda was assisting the rebels of<br />
southern Sudan in their ªght against the Khartoum government. 116 Uganda,<br />
though, was not content with conducting limited operations against the ADF<br />
or even with establishing a security zone along its border with Congo. It participated<br />
in Rwanda’s initial airlift of troops that was intended to take control<br />
of Kinshasa, and it supported rebel groups in Equator Province (more than<br />
113. Thomas Turner, “Angola’s Role in the Congo War,” in Clark, The African Stakes of the Congo<br />
War, pp. 75–83; Reno, Warlord Politics and African States, pp. 151–162; Weinstein, “Africa’s ‘Scramble<br />
<strong>for</strong> Africa,’” p. 14; and Shearer, “Africa’s Great War,” pp. 96–97.<br />
114. Shearer, “Africa’s Great War,” pp. 93–95; Weinstein, “Africa’s ‘Scramble <strong>for</strong> Africa,’” p. 12;<br />
and Lemarchand, “The Democratic Republic of Congo,” pp. 45–48.<br />
115. Baker, “Going to War Democratically,” p. 269.<br />
116. Shearer, “Africa’s Great War,” pp. 95–96; Baker, “Going to War Democratically,” pp. 268–269,<br />
at p. 269; and Clark, “Explaining Ugandan Intervention in Congo,” pp. 271–273.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 167<br />
1,000 miles away from the Ugandan border). Uganda’s president, Yoweri<br />
Museveni, apparently concluded that a long-term solution to his country’s security<br />
problems required a friendlier regime in Kinshasa. 117<br />
In 1998 Angola, an ally of Rwanda and Uganda in their 1996 drive to oust<br />
Mobutu, found itself supporting Kabila against them. Angolan troops, with<br />
Zimbabwean assistance, saved Kinshasa from falling once again into the<br />
hands of a group of Rwandan-backed rebels known as the Congolese Rally <strong>for</strong><br />
Democracy (known by the French acronym RCD). Considerations regarding<br />
Congo’s internal war account <strong>for</strong> Angola’s actions this time as well. Kabila disrupted<br />
the ºow of UNITA’s weaponry and diamonds through Congo. There<strong>for</strong>e<br />
Luanda, though not satisªed with Kabila’s handling of his government,<br />
was unwilling to risk creation of a regime in Kinshasa that might renew its<br />
support <strong>for</strong> the Angolan rebels. 118 Rumors of UNITA and Mobutu sympathizers’<br />
connections with the RCD and with Kigali fueled Angolan suspicions,<br />
eventually bringing about the decision to heed Kabila’s request and send in<br />
troops. 119 That curbing UNITA’s activity was the primary factor in prompting<br />
Angola’s intervention on behalf of Kabila’s government is evident from Angola’s<br />
quick withdrawal from Congo following UNITA’s defeat and the death<br />
of its leader, Jonas Savimbi.<br />
spillover of civil conflict: the kin-country syndrome. Like Burundi,<br />
Rwanda, and western Tanzania, eastern Congo is home to both Hutus and<br />
Tutsis. Yet unlike in Burundi and Rwanda, few tensions existed between these<br />
two groups until the 1990s. Rather, the Hutus and Tutsis in eastern Congo<br />
(again, known collectively as Banyarwanda) were resented by some ethnic<br />
groups, which considered them aliens (even though many of the Banyarwanda<br />
immigrated to this region more than 200 years ago). 120<br />
The Rwandan genocide and the ensuing refugee ºows had consequences <strong>for</strong><br />
all of Central Africa. Hutu Zairians and so-called indigenous Zairians allied<br />
with the Rwandan Interahamwe against the Tutsi, including those native to<br />
Zaire as well as relative newcomers from Rwanda (known as Banyamulenge,<br />
a term that eventually came to describe all Tutsis in Zaire). In 1996 the<br />
117. Clark, “Explaining Ugandan Intervention in Congo,” pp. 272–273, 278–281; and Shearer, “Africa’s<br />
Great War,” pp. 95–96. This is consistent with Fazal’s argument regarding the replacement of<br />
the practice of territorial conquest with one of regime change. See Fazal, “From Conquest to<br />
Intervention.”<br />
118. Weinstein, “Africa’s ‘Scramble <strong>for</strong> Africa,’” p. 14; and Shearer, “Africa’s Great War,” p. 95.<br />
119. Shearer, “Lines on a Map,” p. 295; and Manahl, “From Genocide to Regional War,” p. 6. On<br />
Kabila’s appeal to Angola to intervene, which explicitly promised to continue pursuing anti-<br />
UNITA policies, see François Misser and Alan Rake, “An African World War?” New African, Vol.<br />
367 (October 1998), p. 14.<br />
120. Emizet, “The Massacre of Refugees in Congo,” pp. 166–167; Lemarchand, “The Democratic<br />
Republic of Congo,” pp. 49–52; and Lemarchand, “Patterns of State Collapse and Reconstruction<br />
in Central Africa,” p. 6.
International Security 31:3 168<br />
Interahamwe and ex-FAR soldiers in the refugee camps attacked and killed<br />
thousands of Tutsis in North and South Kivu Provinces. Given that the Hutu<br />
and indigenous Zairian groups were supported by Mobutu’s troops and<br />
money, the Tutsi sought assistance from the Tutsi-ruled Rwandan government.<br />
In the words of Lemarchand, “The kin-country syndrome ...asserted itself<br />
with a vengeance, driving Hutu and Tutsi, irrespective of other distinctions, to<br />
opposing camps.” 121<br />
Mobutu’s exclusion of the Banyamulenge was a primary cause of the rebellion<br />
against him and, because of the Banyamulenge’s kin relations to Rwanda,<br />
of its intervention in Congo. 122 In October 1996, after being ordered to “return”<br />
to Rwanda within six days, the Banyamulenge looked to Kigali <strong>for</strong> military<br />
support. This situation only exacerbated the fears of the Tutsi in Rwanda and<br />
led to their massive attack on the refugee camps in eastern Zaire (and to<br />
the massive slaughter of Hutu refugees), which precipitated the rapid crumbling<br />
of Mobutu’s regime. 123 Similarly, the DRC’s ethnic policies regarding the<br />
Banyamulenge were an important factor in propelling the 1998 Rwandan invasion.<br />
First, Kabila failed to fulªll his promise to Rwanda to bestow citizenship<br />
on the Banyamulenge. Second, in early 1998 he tried to boost his popularity in<br />
Kinshasa by ousting the Tutsi ministers in his government and issuing a decree<br />
that all <strong>for</strong>eign (i.e., Rwandan) troops leave the country. Third, his army<br />
then encouraged a massacre of the Banyamulenge (and anyone else with Tutsilike<br />
physical features) in Kinshasa. 124<br />
In sum, spillover of the civil conºicts between Zaire and many of its neighbors<br />
played a primary role in causing both phases of the war in Congo. The<br />
spillover was a result of two factors: refugees and insurgencies, and neighboring<br />
states coming to assist their excluded and endangered coethnics across the<br />
border. Both factors were directly related to state weakness, and both would<br />
have been less likely had the norm of ªxed borders not existed.<br />
state weakness, opportunism, and greed in the congo war<br />
“Africa’s ‘scramble <strong>for</strong> Africa,’” Jeremy Weinstein argues, “is a primary cause<br />
of the rise of the interstate war on the continent.” 125 To assess the validity of<br />
121. Lemarchand, “Patterns of State Collapse and Reconstruction in Central Africa,” p. 6.<br />
122. Herbst, States and Power in Africa, p. 238.<br />
123. Emizet, “The Massacre of Refugees in Congo,” pp. 167–179; Lemarchand, “Patterns of State<br />
Collapse and Reconstruction in Central Africa,” pp. 6–7; Manahl, “From Genocide to Regional<br />
War,” pp. 2–4; and Weinstein, “Africa’s ‘Scramble <strong>for</strong> Africa,’” p. 13.<br />
124. Lemarchand, “The Democratic Republic of Congo,” pp. 45–46; and Misser and Rake, “An African<br />
World War?” p. 15.<br />
125. Weinstein, “Africa’s ‘Scramble <strong>for</strong> Africa,’” p. 18. For a similar explanation of the war, see<br />
Marc Lacey, “Congo Tires of War, but the End Is Not in Sight,” New York Times, July 15, 2002.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 169<br />
hypothesis 4, this section explores the exploitation of the collapsed Congolese<br />
state by its neighbors as a cause of the Congo war.<br />
Three states involved in the war—Namibia, Chad, and Zimbabwe—had few<br />
reasons to feel directly threatened by the situation in Congo. Yet all three sent<br />
troops to back Kabila’s regime, despite not sharing borders with Congo. Their<br />
motives <strong>for</strong> intervening, there<strong>for</strong>e, must have laid elsewhere. Namibia’s motivations<br />
are unclear. While the government claimed to be enraged by Rwanda’s<br />
and Uganda’s violation of DRC sovereignty, the political opposition accused<br />
the Namibian president of sending troops into Congo to defend his family’s<br />
mining interests there. 126 The case of Chad is also ambiguous, but there are<br />
two possible explanations <strong>for</strong> its troop deployments. First, Chad supposedly<br />
received ªnancial assistance from Libyan President Muammar Gaddhaª, who<br />
was eager to exert his inºuence in sub-Saharan Africa. Second, Chad might<br />
have been interested in the gold mines of northern Congo. 127 Zimbabwe, on<br />
the other hand, had three explicit reasons <strong>for</strong> intervening in the DRC, all having<br />
to do with opportunism. First, Kabila and his ADFL movement owed the<br />
Zimbabwean government large sums of money that had been lent to them<br />
during the rebellion against Mobutu. Zimbabwe feared that if Kabila’s regime<br />
were ousted, it would not get its money back. 128 Second, although the state of<br />
Zimbabwe probably lost money from the Congolese adventure, the extension<br />
of patronage politics from Zimbabwe to Congo hugely beneªted some private<br />
interests. Zimbabwe’s president, Robert Mugabe, openly discussed the potential<br />
economic proªt to be reaped in the DRC as a chief reason <strong>for</strong> Zimbabwean<br />
involvement and, there<strong>for</strong>e, encouraged businesses to exploit the intervention.<br />
Having secured lucrative deals with Kabila’s government, both the army as an<br />
organization and his commanders as private businessmen were heavily involved<br />
in the mining and trade of diamonds, gold, and copper from parts of<br />
the DRC controlled by Zimbabwean troops. Business interests associated with<br />
the ruling Zimbabwe African National Union–Patriotic Front and cronies of<br />
Mugabe reaped their share of the proªts as well. 129 The concentration of<br />
Zimbabwean troops around important mining towns in the DRC testiªed to<br />
these interests. 130 Third, Mugabe hoped to exploit the Congolese conºict to<br />
promote his ambition of becoming a regional leader in southern Africa. 131<br />
126. Baker, “Going to War Democratically,” p. 274.<br />
127. Ibid., p. 269; and Shearer, “Lines on a Map,” p. 295.<br />
128. Misser and Rake, “An African World War?” p. 14; and Shearer, “Africa’s Great War,” p. 98.<br />
129. Baker, “Going to War Democratically,” pp. 268, 274–275; and Shearer, “Africa’s Great War,”<br />
p. 98.<br />
130. Weinstein, “Africa’s ‘Scramble <strong>for</strong> Africa,’” pp. 15–16.<br />
131. Shearer, “Lines on a Map,” p. 295; and Misser and Rake, “An African World War?” p. 14.
International Security 31:3 170<br />
The backers of Kabila’s government, though, did not have a monopoly on<br />
the economic exploitation of Congo through military intervention. Rwanda<br />
and Uganda also had economic interests in Congo, although <strong>for</strong> them economics<br />
seems to have played a secondary role to security. Sufªce it to say that since<br />
1999 Rwanda and Uganda, like Zimbabwe, have been exporting diamonds,<br />
despite the absence of diamond mines on their territory. 132 A 2001 UN report<br />
documents the great extent to which Rwandan and Ugandan military <strong>for</strong>ces<br />
and private companies exploited the territory under their occupation. In addition,<br />
the document traces the relationship between these elements and the<br />
highest echelons in Kigali and Kampala. 133 Although greed played an important<br />
role in the escalation of the war and the prolongation of the Rwandan and<br />
Ugandan presence in Congo, it was not the main reason <strong>for</strong> the two countries’<br />
intervention. Their agreement to remove their <strong>for</strong>ces from the DRC in 2002<br />
supports this contention. Congo’s weakness clearly enabled this economic exploitation.<br />
And as a 2003 UN panel of experts noted, “In the absence of a<br />
strong, central, and democratically elected government that is in control of its<br />
territory, illegal exploitation will continue and serve as the motivation and the<br />
fuel <strong>for</strong> continued conºicts in the region.” 134<br />
Congo’s weakness also encouraged political predation and played a<br />
signiªcant role in facilitating the war. The likelihood of war (in both 1996 and<br />
1998) increased because regime change was a real possibility and because all of<br />
the protagonists had political allies within Congo. The aim of Rwanda’s 1996<br />
invasion, writes Lemarchand, was to “wrestle the Mobutist monster to the<br />
ground and make the whole of Congo safe <strong>for</strong> Rwanda.” 135 Rwanda succeeded<br />
in removing Mobutu and almost succeeded in ousting Kabila precisely because<br />
Congo was an extremely weak state. Similarly, Angola’s apparent involvement<br />
in the assassination of Laurent Kabila could be viewed as an<br />
example of political opportunism—in this case, replacing a disliked leader of a<br />
weak state with a more favorable one. 136 Of course, absent the security motivations,<br />
these predatory actions would not have been necessary. But once a secu-<br />
132. Baker, “Going to War Democratically,” pp. 275, 278. For more evidence of exploitation, see<br />
Clark, “Explaining Ugandan Intervention in Congo,” pp. 275–278; Weinstein, “Africa’s ‘Scramble<br />
<strong>for</strong> Africa,’” p. 17; and Turner, “The Kabilas’ Congo,” p. 217.<br />
133. United Nations Security Council, “Report of the Panel of Experts on the Illegal Exploitation<br />
of Natural Resources and Other Forms of Wealth in the Democratic Republic of Congo,” S/2001/<br />
357 (New York: United Nations, April 12, 2001).<br />
134. United Nations Security Council, “Report of the Panel of Experts on the Illegal Exploitation<br />
of Natural Resources and Other Forms of Wealth in the Democratic Republic of Congo,” S/2003/<br />
1027 (New York: United Nations, October 23, 2003).<br />
135. Lemarchand, “The Democratic Republic of Congo,” p. 43.<br />
136. Turner, “The Kabilas’ Congo,” pp. 216–217; and Lemarchand, “The Democratic Republic of<br />
Congo,” pp. 52–53.
ity threat exists, the weakness of the state provides relatively cheap<br />
opportunities <strong>for</strong> political predation as a way to offset it. Table 1 summarizes<br />
the ªndings in the case of Congo.<br />
Conclusion<br />
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 171<br />
Table 1. The Norm of Fixed Borders and Weak States: A Summary of Findings in the<br />
Congo Case<br />
Hypotheses<br />
Hypothesis 1: Fixed borders can perpetuate or<br />
exacerbate the weakness of already weak states.<br />
Hypothesis 2: Weak states in a fixed-borders world<br />
can create conditions that can give rise to violent<br />
internal conflicts.<br />
Hypothesis 3: Refugee movements, insurgencies, and<br />
kin connections across international borders can<br />
cause civil conflicts in weak states in a fixed-borders<br />
world to spill over their borders and become<br />
international conflicts and possible full-fledged wars.<br />
Hypothesis 4: State weakness promotes the<br />
possibility of international conflict because it creates<br />
opportunities <strong>for</strong> neighbors to intervene to exploit<br />
the weak state economically and politically.<br />
Findings in the Case of the Congo<br />
War<br />
Confirmed, though indirect<br />
Strongly confirmed<br />
Strongly confirmed <strong>for</strong> both the<br />
1996–97 and 1998–2002<br />
interventions<br />
Confirmed <strong>for</strong> Zimbabwe’s<br />
intervention<br />
Confirmed as an important reason<br />
<strong>for</strong> continuation of involvement<br />
of other actors, though not as a<br />
primary reason <strong>for</strong> going to war<br />
“The greatest threats to our security,” U.S. Secretary of State Condoleezza Rice<br />
remarked in a 2005 op-ed, “are deªned more by dynamics within weak states<br />
than by the borders between strong and aggressive ones.” 137 This article offers<br />
an explanation <strong>for</strong> this phenomenon. In regions where states are sociopolitically<br />
weak, the norm of ªxed international borders can increase the likelihood of<br />
international conºict. In such situations, ªxed borders perpetuate and exacerbate<br />
the weakness of the state, which, in turn, is a major cause of internal wars.<br />
These internal wars can then spill over and become international conºicts. The<br />
137. Condoleezza Rice, “The Promise of Democratic Peace: Why Promoting Freedom Is the Only<br />
Realistic Path to Security,” Washington Post, December 11, 2005. Rice’s remarks were in line with<br />
the 2002 U.S. National Security Strategy. See President George W. Bush, National Security Strategy of<br />
the United States of America (Washington, D.C.: White House, 2002).
International Security 31:3 172<br />
Congo war illustrates what can happen when the norm of ªxed border exacerbates<br />
state weakness.<br />
The norm of ªxed borders creates constraints and incentives that compel<br />
weak states to remain weak or that make them even weaker. With no territorial<br />
threats and their juridical statehood assured by the international community,<br />
rulers are not pressured to assume the risks and to pay the costs that are involved<br />
in the process of state building. And in contrast to the experience of<br />
states prior to World War II, failed or collapsed states are no longer weeded<br />
out by the system.<br />
Scholars widely agree that state weakness can lead to internal strife and civil<br />
conºicts. What is less often acknowledged is that such weakness, especially<br />
when borders are unchangeable, can also precipitate international wars. This<br />
article explored two paths leading from state weakness to international<br />
conºict. In the ªrst, weak states create conditions that are rife <strong>for</strong> internal<br />
conºict by giving way to emerging anarchy and by allowing leaders to exploit<br />
ethnic divisions to compensate <strong>for</strong> their own lack of legitimacy. Civil conºicts<br />
in weak states under the norm of ªxed borders, in turn, are more likely to spill<br />
over and involve neighboring states than are other kinds of civil conºicts because<br />
of both the inability of the state to prevent insurgencies and the existence<br />
of the kin-country syndrome. In the second path, the weakness of the state creates<br />
opportunities <strong>for</strong> its neighbors to engage in political or economic (though<br />
not territorial) predation.<br />
The case study of Congo largely conªrms these arguments. The behavior of<br />
Congo’s leaders suggests that the norm of ªxed borders was a signiªcant factor<br />
in the weakening of the Congolese state. Congo’s growing weakness, in<br />
turn, was a major cause of its internal wars, especially during the 1990s. The<br />
spillover of these wars (as well as the civil conºicts in neighboring weak<br />
states) contributed signiªcantly to the outbreak of war in the Great Lakes region<br />
in 1996 and again in 1998. Greed was also an important factor: it spurred<br />
Zimbabwe’s decision to go to war, and it contributed to other actors’ prolonged<br />
involvement in the conºict. Yet it was not Rwanda’s, Burundi’s,<br />
Uganda’s, or Angola’s primary motivation <strong>for</strong> going to war.<br />
Caution is needed when attempting to draw general conclusions from a single<br />
case study. A carefully designed study that compares the Congo war with<br />
other cases, past and present, is needed to conªrm the conclusions presented<br />
here. 138 A study that corroborates my hypotheses on the relationship between<br />
the norm of ªxed borders and the likelihood of international conºict would<br />
138. See, <strong>for</strong> example, Boaz Atzili, “Border Fixity: <strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong>,”<br />
Ph.D. dissertation, Massachusetts Institute of Technology, 2006.
<strong>When</strong> <strong>Good</strong> <strong>Fences</strong> <strong>Make</strong> <strong>Bad</strong> <strong>Neighbors</strong> 173<br />
represent an important step in understanding conºict in regions where sociopolitically<br />
weak states predominate, as is the case in most of the developing<br />
world. That an international norm designed to enhance stability and peace<br />
might actually produce the opposite effect is a subject worthy of scholarly attention.<br />
That good fences can make bad neighbors may be bad news, but it is<br />
news that we might well have to reckon with.