Introduction
It is a well-established theme in African International Relations research that African leaders are generally opposed to territorial revisions. This theme goes back to the 1960s, with the construction of the Organization of African Unity (OAU), and first saw expression in print with Saadia Touval’s The Boundary Politics of Independent Africa (Touval Reference Touval1972). However, the question of why the bias of African leaders against secession movements has persisted even as the continental and international contexts have changed remains relatively under-researched. While the arbitrariness of African country borders may have been a compelling reason for this bias in the early post-colonial period because of the high potential for interstate conflict, and because the success of any secession movement could encourage subsequent secession movements, nevertheless it seems that fifty-plus years past the end of colonial rule, these borders should have achieved some stability for most countries. Figure 1 below shows that the number of interstate territorial claims in Africa on the basis of an ethnic, religious, linguistic, or other identity dropped from around seventeen claims in a given year in the early 1960s to around six in a given year in the late 1990s (Frederick, Hensel & Macaulay Reference Frederick, Hensel and Macaulay2017).
Figure 1: Annual interstate territorial claims in Africa based on an ethnic, religious, linguistic, or other identity, 1960-2001
Similarly, while the Cold War context may have also played a role in the first three decades of independence, the demise of the Eastern Bloc might otherwise be expected to have altered the international context significantly. In short, the persistence of the African commitment to the territorial integrity norm over such a long period of time is far more puzzling than its creation. This article aims to fill this gap in research.
Opposition to territorial revisions in post-colonial Africa was initially driven primarily by a collective interest in maintaining peace and security in spite of weak state institutions, ethnic heterogeneity, and questionably legitimate borders. This soon led to the social construction of the territorial integrity norm. The formation of the territorial integrity norm, in turn, encouraged African leaders to conceive or reconceive their interests according to this bias. The perpetuation of the territorial integrity norm is thus a result of this internalization of the norm among African leaders. However, since most African borders have gained some stability, and because of the end of the Cold War, African leaders have significantly fewer incentives to apply the territorial integrity norm strictly across the continent. Instead of solely being a result of state interests, the African commitment to the territorial integrity norm beyond the early post-colonial period is for a large part a result of what can be described as “identity interests” (Jepperson, Wendt & Katzenstein Reference Jepperson, Wendt, Katzenstein and Katzenstein1996). The opposition to territorial revisions has from the very early post-colonial period been part of African identity, which explains why, even if the territorial integrity norm has become less utilitarian over time, a strong commitment to the norm will likely remain.
This article focuses on two of the most visible implications of the perpetual robustness of the territorial integrity norm in Africa: (1) the propensity of mediators working with African situations to promote the territorial status quo; and relatedly, (2) the extremely low probability of the conclusion of a negotiated settlement that paves the way for independence.
The territorial integrity norm is not a uniquely African phenomenon. Dating back to the beginning of the Westphalian system, commitment to the sanctity of borders was globally enshrined by the UN Charter of 1945, which affirmed that borders should not be altered through force (Zacher Reference Zacher2003). Hence, non-African third parties are often also reluctant to advocate for territorial changes when they are involved in territorial conflicts in Africa, having a general propensity to promote the territorial status quo. However, because the potentially negative spillover effects of territorial changes were particularly strong in Africa, the identity of African leaders in general has been strongly tied to maintaining the territorial status quo.
Moreover, in those rare instances in which the mediation effort is actually aimed at a territorial revision, non-African third parties are more likely to be successful in accomplishing this objective than African third parties. Non-African negotiators have far greater military and economic resources which they can use to coerce or induce conflict parties, which increases their effectiveness in negotiating territorial revisions.
This article makes two major contributions. First, it contributes to the scholarly understanding of norm perpetuation with respect to the international relations of African states. The question of how norms and self-interest intermingle in African international relations is too rarely addressed. Over time the territorial integrity norm has become less utilitarian but it nevertheless remains robust because of the strong identity interests of African leaders.
Second, this article contributes to the conflict resolution literature by examining how a given third party’s self-interest and norms affect the solutions this third party promotes. Little attention has been given to factors influencing whether third parties decide to advocate for a solution based on the territorial integrity norm or not. This reflects a broader gap in research regarding the reasons behind the solutions third parties promote in the peacemaking process (Duursma Reference Duursma2014).
This article proceeds as follows. The next section will outline how the territorial integrity norm appeared in the late 1950s and early 1960, as well as how this norm has remained relevant for both African and non-African third parties. This theoretical argument is followed by some descriptive statistics about international mediation in civil wars with a territorial dimension in Africa between 1960 and 2014. The subsequent part of the article presents seven short case studies on mediation efforts in the civil war in Africa that support the main argument. The final section concludes that the status-quo bias makes third parties often unwilling to promote a territorial revision as part of the mediation process. In addition, major power leverage seems to be a critical factor in producing territorial revisions in Africa in those cases in which a territorial change is sought.
The Role of the Territorial Integrity Norm in African Peace Processes: A Theory of Third Party Willingness and Opportunity
This section first explains how the territorial integrity norm has shaped the commitment of third parties to push for solutions to civil wars that do not lead to territorial revisions. In addition, it discusses how non-African third parties, and especially external major power mediators, have resources with which to compensate African governments for making territorial concessions to end their civil wars.
The Construction and Perpetuation of the Territorial Integrity Norm in Africa
Norms can be defined as “collective expectations about proper behavior for a given identity” (Jepperson, Wendt & Katzenstein Reference Jepperson, Wendt, Katzenstein and Katzenstein1996:54). Since norms are collective expectations, they require intersubjective agreement between at least two actors. When there is a level of intersubjectivity, norms can regulate and constrain the behavior for an already-constituted identity and generate shared assessments about standards of behavior in varying circumstances. Since the regulative element of a norm involves a collective expectation, it presupposes a community able to pass judgements on appropriateness (Finnemore & Sikkink 1998).
In the case of peace processes in Africa, the African community of states passes judgements on appropriateness regarding upholding the territorial integrity of the civil war country. The African commitment to the territorial integrity norm was initially constructed on the basis of the self-interest of African states. When most African states gained independence in the late 1950s and early 1960s, African leaders realized that calling boundaries into question would make them equally vulnerable to such actions against themselves (Clapham Reference Clapham1996:35; Zartman Reference Zartman, El-Ayouty and Zartman1984:39). Respecting the sovereignty of other African states and their inherited borders from the colonial era was thus the best solution to achieving a peaceful coexistence on the continent, and by extension, the best option for safeguarding regime survival (Herbst Reference Herbst1989; Jackson & Rosberg Reference Jackson and Rosberg1982).
The institutionalization of the territorial status quo as an instrument for maintaining peace in Africa is reflected in the 1963 Charter of the OAU, in which respect for the member states’ territorial integrity was heavily emphasized (OAU 1963:preamble). In addition, a resolution adopted by the Assembly of Heads of State and Government of the OAU on July 21, 1964, incorporated the principle of uti possidetis, meaning that all states should respect the borders existing at the time of their independence. The rationale for this resolution was succinctly summarized by President Migan Apithy of Benin, who stated that if countries were allowed to contest their boundaries, “the tidal wave of territorial claims would plunge our continent into apocryphal chaos which would sweep our Organization away like a wisp of straw” (Amate Reference Amate1986:404).
Furthermore, while the commitment to territorial integrity initially emerged as a result of the need to regulate inter-African relations in post-colonial Africa, the norm was soon extended to also rule out territorial conflicts internal to individual states. Territorial changes in Africa either on an international level or domestically were both perceived as a priori detrimental to Africa, since these changes could set a precedent that could lead to the further balkanization of the continent (Amate Reference Amate1986:404; Červenka Reference Červenka1977:70). Consequently, African peace efforts in civil wars in Africa have been guided by a strong commitment to territorial integrity from the early post-colonial period onward. Samuel G. Amoo and William I. Zartman observe in this regard that principles of legitimate behavior defined in the Charter of the OAU, including respect for the territorial integrity of other member states, provide the basis for searching for a mediated solution (Amoo & Zartman Reference Amoo and Zartman1992:135).
It follows from the above that African commitment to the territorial integrity norm was constructed on the basis of the self-interest of African states. Yet, the territorial integrity norm has persisted in spite of the borders in Africa gaining some stability over time because of its internalization among African leaders (on the internalization of norms, see: Kratochwil Reference Kratochwil1989:97; March & Olsen 1998:54). Martha Finnemore and Kathryn Sikkink explain that norms are internalized through habit and institutionalization—and because actors want to conform to norms. As a result, “norms may become so widely accepted that they are internalized by actors and achieve a ‘taken-for- granted’ quality that makes conformance with the norm almost automatic” (Finnemore & Sikkink 1998:904). Telling in this regard is that, just like the OAU, the 2002 Constitutive Act of the African Union (AU) underscores the territorial integrity of its member states and emphasizes the AU’s respect of the “borders existing on achievement of independence” (African Union 2002).Footnote 1
Indeed, commitment to the territorial integrity norm is part of Africa’s security culture. An international security culture—or strategic culture as Alastair Johnston coins it—refers to an integrated system of norms among states that “acts to establish and maintain pervasive and long-lasting preferences by formulating concepts on the role and efficacy” of particular approaches to protecting cherished values (Johnston Reference Johnston and Katzenstein1996). Ronald L. Jepperson et al. use the term “identity interests” to explain how norms are often connected to a given identity, which makes actors with this identity perceive their self-interest in line with these norms (Jepperson, Wendt & Katzenstein Reference Jepperson, Wendt, Katzenstein and Katzenstein1996). To put it differently, international identities are shaped by the normative context in which actors operate; these identities, in turn, shape the interest and behavior of actors (Finnemore Reference Finnemore1996; Katzenstein Reference Katzenstein and Katzenstein1996:902).
Africa’s security culture is the product of a common African interest in maintaining peace and stability; it did not appear out of thin air, but was built by African leaders and subsequently became a social fact. The formation of Africa’s security culture brought with it a set of pervasive and long-lasting preferences, of which the territorial integrity norm is a major component. As a result, even while the initial collective interest of committing to the territorial integrity norm is arguably much less pronounced now than it was in the early post-colonial period, African leaders are still conforming to this norm.
Why Third Parties Refrain from Mediating Territorial Revisions
A potential third party has three options when a civil war is ongoing: (1) not offer mediation or decline a request to mediate; (2) mediate without an agenda; and (3) mediate with an agenda. Potential African and non-African third parties frequently make a conscious decision not to mediate because of reasons related to the territorial integrity norm. A telling example in this regard is that while the war between the Sudan and the Anyanya movement had been ongoing since 1963, it was not until 1972 that mediation took place for the first time. Francis D. Deng and Khalid M. Medani explain that African countries did not want to offer mediation, since this would recognize the southern secessionists: “The prevalent Pan-Africanist philosophy emphasized the unity and brotherhood of all Africans, Arabs included, and African leaders held fast to the principle of the sacrosanctity of the existing African boundaries borrowed from their former colonizers” (Deng & Medani Reference Deng, Medani, Keller and Rothchild1996:106). Christopher S. Clapham explains that major powers have also been historically reluctant to recognize secessionist movements in Africa (Clapham Reference Clapham1996:64; Zacher Reference Zacher2003)
While potential third parties committed to the territorial integrity norm may refrain from offering to mediate, another common response is to mediate with an agenda. Indeed, due to Africa’s security culture—and specifically the commitment to the territorial integrity norm—African third parties in particular commonly mediate with an agenda aimed at concluding a peace agreement that preserves the territorial integrity of the civil war country. The idea that third parties would have a preference about the outcome of the civil war in which they are mediating is not new. Philip Gulliver has pointed out that third parties often seek settlements that are in line with their own ideas and interests (Gulliver Reference Gulliver1979:213). An example of an African mediation effort aimed preserving the territorial integrity of a civil war country is the mediation effort by Ethiopia in 1972 in Sudan’s first civil war. At the start of his mediation effort, Ethiopian Emperor Haile Selassie heavily emphasized that African nations had the duty to prevent secession and promote unity in other African countries (Alier Reference Alier1990:111).
Mediating Territorial Solutions through Manipulation
There have, of course, been instances in which third parties pushed for territorial revisions in spite of the strong commitment to territorial integrity. Yet, these cases could be seen as the exceptions that prove the rule. Moreover, only in a few instances have these mediation efforts led to an actual territorial change. Non-African third parties have played a crucial role in these mediation efforts. Due to a high degree of leverage, non-African third parties—particularly major powers involved in mediation in Africa—are often in a better position to push incumbent governments in civil wars towards making costly territorial concessions. Indeed, merely having an agenda is not sufficient. Third parties also need the means to pursue their agendas (Sisk Reference Sisk2009).
A high degree of third-party leverage is necessary to move incumbent governments toward territorial concessions, because territorial intrastate conflicts are relatively difficult to resolve (Fearon Reference Fearon2004). Part of this intractability is because territorial conflicts are usually between different ethnic groups concentrated in clearly defined territories (Toft Reference Toft2003). Three other major explanations for why territorial conflicts are difficult to resolve have been identified in previous research. First, governments are unwilling to give up a piece of territory because of the geostrategic and material value of territory (Sprout & Sprout Reference Sprout and Sprout1968). Second, governments are often unwilling to agree to secession because of the symbolic value of territory (Hassner Reference Hassner2003). Finally, several studies have shown that governments are often hesitant to strike a territorial deal with rebel movement because this potentially creates a precedent for other groups that might wish autonomy or independence (Toft Reference Toft2003; Walter Reference Walter2006). The fear of creating a precedent for other potential secessionist groups, coupled with the material and symbolic value of territory, sheds light on why incumbent governments in civil wars are extremely reluctant to agree to secession or to a referendum on secession.
As a consequence of this reluctance, incumbent governments experiencing civil wars will usually try to resist third-party efforts to make them accept secession. Consequently, if a mediator wants to reach a negotiated settlement that stipulates independence or a referendum on independence, it needs to provide strong incentives to the government side to make such a costly concession more attractive. While third-party legitimacy in general may be an effective source of mediation success (Duursma Reference Duursma2017a), it is arguably less useful for mediating territorial revisions. Unlike most of the non-African third parties involved in civil wars in Africa, African third parties have much less leverage in the form of material incentives they can provide to persuade the conflict parties to accept secession (Jackson Reference Jackson2000; Khadiagala Reference Khadiagala2007). By contrast, the ability of non-African parties, particularly the major powers, to move conflict parties toward compromise based on their high degree of economic and military resources is well documented in the literature (Smock & Gregorian Reference Smock, Gregorian and Smock1993:12; Touval Reference Touval, Bercovitch and Rubin1992).
In sum, a willingness to mediate a peace agreement that stipulates secession to resolve the civil war is not sufficient; the third party also needs to be able to either persuade or coerce the incumbent government of a civil war country to make an extremely costly territorial concession. Non-African third parties, particularly major powers, have more capacity to move conflict parties toward making territorial concessions. Consequently, for reasons of motivation, African and non-African third parties are unlikely to push for a negotiated settlement that paves the way for independence. Yet, when they do, non-African third parties are generally more effective because of their greater third-party capacity.
An Overview of International Mediation in Africa’s Territorial Civil Wars
The Uppsala Conflict Data Program (UCDP) identifies thirty-one civil wars fought over a piece of territory and 158 armed conflicts over control of the government in Africa between 1960 and 2014 (Pettersson & Wallensteen Reference Pettersson and Wallensteen2015). The probability of a territorial conflict is thus much lower than that of non-territorial conflicts in Africa. Previous research shows that this low proportion of territorial conflict is fairly unique to Africa: African insurgents are less likely to make secessionist claims when compared to non-African insurgents (Englebert & Hummel Reference Englebert and Hummel2005).
Of the thirty-one territorial civil wars, thirteen experienced no mediation. Table 1 in the online appendix lists these thirteen civil wars. International mediation did take place in the remaining eighteen civil wars fought over a piece of territory, using analytical summaries of this dataset on international mediation in civil wars in Africa to determine whether mediation took place in a given civil war and whether African and/or non-African third parties were involved in mediation (Duursma Reference Duursma2017a).
It is acknowledged that the mediation efforts in these civil wars were not always united and did not have a single aim throughout. Moreover, each of these mediation processes had many participants, and strategies shifted over time. Nevertheless, it is possible to identify whether the mediators proposed or even advocated for a territorial revision at some point during each of these eighteen peace processes. Taking this as a benchmark, I find that all the peace efforts in thirteen of these eighteen civil wars were solely aimed at preserving the territorial integrity of the civil war country. Table 2 in the online appendix lists these thirteen civil wars, including a short analytical summary of the peace efforts in each of these wars. Hence, there were only five civil wars fought over a piece of territory in Africa between 1960 and 2014 in which a territorial revision was proposed or recommended by a third party at some point over the course of the war. These five civil wars are listed in Table 3 in the online appendix.
African third parties were involved in twelve of the thirteen civil wars in which the mediation that took place was aimed solely at maintaining the territorial integrity of the civil war state. Non-African third parties were involved in five of these thirteen civil wars at some point. In some conflicts, African and non-African third parties both mediated jointly or successively in order to reach an agreement based on the territorial integrity principle.
Out of the five conflicts that experienced mediation aimed at independence or a referendum on independence, three experienced the involvement of African third parties, while five experienced the involvement of non-African third parties. In some rare instances, mediation efforts in Africa between 1960 and 2014 have thus been aimed at a secession rather than at preserving the territorial integrity of the civil war country. Yet, a closer look at these mediation efforts shows that these cases are the exceptions that prove the rule. Two of these five efforts took place in the Western Sahara dispute between Morocco and Mauritania on the one side and Polisario on the other side. The Western Sahara conflict was widely considered as a decolonization conflict. When Spain granted independence to the Western Sahara in 1975, Morocco and Mauritania annexed it. This explains why the African third parties involved in mediation in these two conflicts recommended self-determination of the Sahrawi people rather than the territorial integrity of Morocco and Mauritania (Naldi Reference Naldi1982:152–59).
In addition, it is noteworthy that three out of four cases with redrawn boundaries are based on colonial boundaries previously demarcated by Europeans: Eritrea, Western Sahara, and Namibia. South Sudan is the only case in which there was no pre-existing colonial boundary. One interpretation of this fact would be that the international community is more likely to become engaged in mediating conflicts that relate to boundaries that the major powers themselves played a role in drawing in the first place.Footnote 2
A closer look at the mediation efforts aimed at a territorial revision further suggests that major power mediation is crucial in order to reach an agreement that can lead to secession. The pressure put on the Government of South Africa by the Western Contact Group paved the way for an independent Namibia, and the U.S. played a major role from 2002 onwards in the north-south Sudan peace process. This is, however, not the case for Eritrea. The Tigrayan People’s Liberation Front (TPLF) and Eritrean People’s Liberation Front (EPLF) had a pre-existing agreement that Eritrea would seek its independence after they defeated the Derg regime (Young Reference Young and Clapham1998). So, even though the U.S. was involved in mediation prior to Eritrea’s referendum on independence, this mediation was not crucial for Eritrea’s gaining independence. In short, major power involvement seems to have relatively often led to the conclusion of an agreement that stipulates independence. However, it should be noted that the cases of actual territorial change is so small, that one should be careful to draw any conclusions solely based on these figures. The Eritrea case illustrates that correlation is not causation: non-African third parties were involved in mediation, but they did not play a crucial role in the lead-up to Eritrea’s referendum on independence.
Case Studies
This section employs seven brief case studies focusing on five civil wars. These cases illustrate that the territorial integrity norm has had a profound effect on the nature of mediation efforts since the early 1960s. They were selected to illustrate the propensity of mediators involved in mediation in Africa to promote the territorial status quo (Nigeria, the Comoros, and Sudan 1992–1993), as well as to illustrate how non-African third parties have played a crucial role in persuading conflict parties to conclude a negotiated settlement that paves the way for independence (Namibia and Sudan 2001–2005) and to illustrate that African third parties without the help of major powers are unlikely to mediate territorial changes (Western Sahara and Sudan 1994–2001). Because of the extremely limited number of territorial revisions in Africa, all cases in which a territorial revision took place following negotiations are considered (Namibia and Sudan 2001–2005), with the exception of the Eritrea case. It would not be meaningful to examine Eritrea, because the TPLF and EPLF had a pre-existing agreement that Eritrea would seek its independence after they defeated the Derg. In this case, the outcome was thus already pre-determined before the mediation process started (Young Reference Young and Clapham1998). Focusing on these seven cases thus essentially follows the logic of a diverse case study selection methodology (Seawright & Gerring Reference Seawright and Gerring2008), making it possible to illustrate the propensity of mediators involved in mediation in Africa to promote the territorial status quo, as well as to illustrate the crucial role of non-African third parties in those cases in which the mediation effort is aimed at territorial change. In addition, the three cases on Sudan make it possible to leverage within-case variation (Zartman Reference Zartman2005), essentially showing how the involvement of the U.S. between 2001 and 20015 moved the peace process toward the independence of Southern Sudan.
Map 1. The political map of Africa of 2018 is used, so rather than highlighting only Namibia, Eritrea, South Sudan, I highlight also the “rump state,” since the territorial civil war occurred in these countries. A star approximates where the territorial civil war was fought.
Nigeria, 1967–1968
In 1967, three states of the Eastern region in Nigeria declared their secession, together forming the self-proclaimed Republic of Biafra. As a response to the civil war in Nigeria, the Assembly of Heads of State and Government of the OAU created a special commission of six heads of state at a summit in Kinshasa in September 1967. Careful to avoid any implicit recognition of Biafra through its third-party involvement, the authority of what was referred to as the OAU Consultative Mission was limited to assisting in the preservation of Nigeria’s territorial integrity (Meyers Reference Meyers1974:35; Stremlau Reference Stremlau1977:92–93). The Resolution stated: “Desirous of exploring the possibilities of placing the services of the Assembly at the disposal of the Federal Government of Nigeria, Resolves to send a consultative mission of 6 Heads of State, (Cameroon, Congo [Kinshasa], Ethiopia, Ghana, Liberia and Niger) to the Head of the Federal Government of Nigeria to assure him of the Assembly’s desire for the territorial integrity, unity and peace of Nigeria” (OAU 1967).
Indeed, when the OAU mission arrived in Lagos on November 23, 1967, Emperor Haile Selassie of Ethiopia assured the Nigerian Head of State, General Yakubu Gowon, that the OAU “unreserved supports Nigerian national unity.” Selassie also stated that through their involvement, the African heads of states hoped “to persuade the secessionist group that the consequences of their actions inconsistent, not only with the interest of the Nigerian people but the interest of all of Africa.” Selassie further elaborated that “secessionist tendencies are to be found in almost all African states. This is the legacy we inherited from the colonialists who tried ruling Africans by dividing them. We know that secession, if it were to be tolerated on our continent, would lead to the destruction of what Africans hold in high esteem—their independence and their progress” (Stremlau Reference Stremlau1977:100).
In the concluding session of the meeting at Lagos on November 23, 1967, several members of the OAU mission emphasized Nigeria’s territorial integrity. President Diori of Niger stated that the most important guiding principle of future peace talks would be the territorial integrity of Nigeria. Emperor Selassie stated that the task of the OAU should be to end secession, as a successful secession would set a dangerous precedent for Africa. It was then finally agreed that if Biafra would agree to be reintegrated into Nigeria, then official mediated negotiations could take place (De St. Jorre Reference De St. Jorre1972:180–202; Stremlau Reference Stremlau1977:101–3). This was unacceptable to the Biafran leadership. In fact, since the start of the war, it had constantly emphasized that “Biafra’s sovereignty should not be negotiable and that Biafra would be invited to the peace talks as a sovereign State and not as part of Nigeria” (Červenka Reference Červenka1977:99). This explains why it took until May 23, 1968, before the parties finally entered into negotiations (Rothchild Reference Rothchild1997a:51–52).
The peace talks that started in May 1968 were mediated by President Milton Obote of Uganda and focused on the question of foreign observers, the conditions for ending hostilities, and the arrangements of a permanent settlement; yet, it was clear throughout the negotiations that neither the government of Nigeria nor the African mediators considered the possibility of the right of self-determination for Biafra, while this this was an absolute demand of the Biafran negotiators. On May 31, 1968, peace talks broke down (Červenka Reference Červenka1977:100; Rothchild Reference Rothchild1997a:52).
Subsequent negotiations took place in Niger in July 1968 and Ethiopia in August and December 1968, but none of these rounds of peace talks resulted in the conclusion of a mutually acceptable agreement. The rebel side continued to insist on a territorial solution to the civil war, which was unacceptable to both the government of Nigeria and the African third parties involved in the peace process (Červenka Reference Červenka1977:106–7; Rothchild Reference Rothchild1997a).
Namibia, 1976–1988
In 1966, the South West African People’s Organization (SWAPO) took up arms against the government of South Africa. SWAPO sought total independence of South West Africa, which is present-day Namibia. No significant efforts to resolve the conflict were undertaken between 1966 and 1975, but during the second half of 1976, U.S. Secretary of State Henry Kissinger held several separate talks with South African Prime Minister John Vorster and SWAPO leader Sam Nujoma on the future of Namibia. As soon as the U.S. became involved, Washington started advocating for a territorial solution to the civil war. Kissinger even discussed the time frame for Namibia’s independence with Nujoma (Melber & Saunders Reference Melber and Saunders2007:78; Zartman & Touval 1985).
Kissinger’s shuttle diplomacy laid the basis for further diplomatic involvement by the Western Contact Group (WCG), which was formed in 1977 by the U.S., the U.K., Canada, France, and West Germany. In May 1977, the WCG held negotiations with the South African government. At these talks, the WCG threatened to drop their veto on UN economic sanctions if South Africa continued to resist Namibian independence (Melber & Saunders Reference Melber and Saunders2007:78–79; Zartman Reference Zartman1985:171–207). As a result, South Africa agreed to negotiations. Several rounds of separate talks with the conflict parties were held during the rest of the year (Zartman Reference Zartman1985:173). It soon became clear during these talks that South Africa was willing to concede to Namibia’s gaining independence, although it preferred to transfer power to the white minority in Namibia, based on South Africa’s apartheid model (Tome Reference Tome1992:274–75). Consequently, the focal point of the negotiations shifted toward the modalities surrounding Namibia’s gaining independence, including the status of the white minority in Namibia, the status of SWAPO in the first elections, and the details of a possible UN peacekeeping mission (Melber & Saunders Reference Melber and Saunders2007:79–80).
On February 11–12, 1978, shuttle talks were held in New York, but the conflict parties could not agree on whether South Africa should keep a military presence in Namibia until the formation of a new government. In order to break the deadlock, the WCG put forward a peace proposal in March, which called for independence of Namibia through free and fair elections under the supervision and control of the United Nations. The government of South Africa agreed to this proposal in April 1978, while SWAPO agreed to the Western Contact Group proposal on July 12, 1978 (Melber & Saunders Reference Melber and Saunders2007:80–81; Zartman Reference Zartman1985:173–76). The proposals were subsequently incorporated into UN Security Council Resolution 435, which was accepted on September 29, 1978 (United Nations Security Council).
However, concerned about the status of the whites living in Namibia following independence, the government of South Africa repeatedly backtracked its commitment to the WCG proposal in the peace process that followed. It took years of diplomatic pressure from the WCG—and particularly from the U.S.—to push the Government of South Africa toward accepting an independent Namibia (Papp Reference Papp and Smock1993:185–86; Rothchild & Hartzell Reference Rothchild, Hartzell and Zartman1995:187). For example, U.S. Secretary of State Alexander Haig told South African Foreign Minister Pik Botha in early 1981 that the stalemate in Namibia was the main obstacle to the development of a new relationship between the U.S. and South Africa (Rothchild Reference Rothchild1997b:101). Crucially, the U.S. also promised to ensure that Namibian independence would result in the withdrawal of Cuban troops from Angola (Crocker Reference Crocker1994). Consequently, the Namibian peace process was linked to negotiations between South Africa, Cuba, and Angola. While it would take until December 1988 until these countries finally reached an agreement, South Africa agreed to implement the WCG proposal by mid-1982.
Yet, negotiations on Cuban withdrawal would drag on until December 22, 1988, when South Africa, Angola, and Cuba finally concluded the Angola/Namibia Accords (Papp Reference Papp and Smock1993:185–86; Rothchild & Hartzell Reference Rothchild, Hartzell and Zartman1995:187). The Angola/Namibia Accords paved the way for Namibia’s independence on March 21, 1990, following elections won by SWAPO. It was the power brokerage of the U.S. that made the mediation of Namibia’s independence possible (Melber & Saunders Reference Melber and Saunders2007).
Western Sahara, 1980–1984
African third parties were initially hesitant to become involved in mediation in the Western Sahara conflict, since the conflict was perceived as an issue of decolonization and self-determination, which was considered the domain of the UN. Yet, from 1980 onward, African third parties became increasingly involved. From September 9–12, 1980, an OAU “committee of wise men” hosted a meeting in Freetown, in which representatives of Morocco, Algeria, Mauritania, and Polisario participated (Naldi Reference Naldi1982:152–59). Sierra Leonean President Siaka Stevens acted as the chief mediator at this meeting, at the conclusion of which the committee presented a proposal for a ceasefire, a UN peacekeeping force, and, crucially, the holding of a referendum on independence. African third-party involvement in the Western Sahara conflict thus contradicts the argument that the strong commitment of African third parties to territorial integrity translates to the mediation of non-territorial solutions to civil wars in Africa. However, African leaders perceived the Western Sahara conflict as different from other conflicts in Africa; the conflict was seen as involving the matter of self-determination rather than secession from an established member state of the OAU. Spain had granted independence to the Western Sahara in 1975; its subsequent annexation by Morocco was seen as illegal by most African states (Clapham Reference Clapham1996:114; Naldi Reference Naldi1982:159; Zartman Reference Zartman, El-Ayouty and Zartman1984:40). This explains why African leaders called for the implementation of a territorial solution to the conflict. The Western Sahara case thus also illustrates how solutions to conflicts can bring norms into competition. African leaders value territorial integrity, but they also value de-colonization. In the case of the Western Sahara, de-colonization was privileged.Footnote 3
The OAU proposal to organize a referendum on independence was accepted by Polisario, but the government of Morocco rejected it. Following this refusal by the government of Morocco, the OAU continued to push for the OAU peace plan, mediating another round of peace talks between the conflict parties in August 1981 (Zartman Reference Zartman1985:53). However, Morocco continued reject the idea of independence being an option in the referendum. Indeed, prior to the OAU summit in June, King Hassan stated that “we will not renounce a single grain of sand of this Moroccan Sahara for which so many of us have scarified our blood and which has cost us so much money” (Sater Reference Sater2009:128).
Endorsing the African effort aimed at concluding a territorial solution to the Western Sahara conflict, the Assembly of Heads of State and Government of the OAU called upon Morocco and the Polisario in December 1983 to “undertake direct negotiations with a view to bringing about a ceasefire to create the necessary condition for a peaceful and fair referendum for self-determination of the people of Western Sahara, a referendum without any administrative or military constraints, under the auspices of the OAU and the UN” (OAU 1983). The African insistence on a referendum on independence as a solution to the conflict significantly contributed to Moroccan resistance against third-party involvements (Durch Reference Durch1993; Layachi Reference Layachi and El-Ayouty1994).
What is more, the OAU could not mitigate the growing intransigence of the government of Morocco through providing negative or positive incentives, as it lacked any leverage to coerce the government of Morocco to make a territorial concession. Amadu Sesay argues that the paralysis of the OAU’s conflict resolution efforts in the Western Sahara dispute in the early 1980s was due to a “lack of material resources with which it could embark upon an effective mediatory role” (Sesay, Ojo & Fasehun Reference Sesay, Ojo and Fasehun1984:56). When in November 1984 a majority of the OAU member states recognized the Western Sahara as an official member of the OAU, Morocco immediately suspended its membership of the OAU, effectively ending the African mediation effort to end the conflict (Labella Reference Labella2003:78; Zoubir Reference Zoubir1996:190).
The Comoros, 1997
On August 3, 1997, the leader of the separatist organization the Republic of Anjouan, Abdallah Ibrahim, unilaterally declared the Anjouan island independent from the rest of the Comoros (Cornwell Reference Cornwell1998:51; Hassan Reference Hassan2009; Naldi Reference Naldi1998:247). In response to this declaration, the OAU crisis committee declared “the unequivocal support of the OAU for the unity, territorial integrity and cohesion of the Comoros” (IPS News, August 21, 1997). Naldi notes that the OAU could not accept the break-up of the Comoros “without undermining its core values and establishing a dangerous precedent for separatism in Africa” (Naldi Reference Naldi1998:256).
Even though it had immediately condemned the unilateral declaration of independence, the OAU dispatched Pierre Yere as a special envoy to the Comoros to serve as mediator and shuttle between the government of the Comoros and Anjouan. As soon as peace negotiations started, Yere stated that according to the OAU, secession was “totally unacceptable” (Naldi Reference Naldi1999:45). Moreover, prior to the negotiations, OAU Secretary General Salim had publicly stated that the secessionist movement in Anjouan represented a dangerous precedent for the neighbors of the Comoros: “Anjouan and Moheli are not far from Mozambique, Tanzania and Kenya. The consequences of their moves will be felt beyond their borders” (IPS News, August 21, 1997). Richard Cornwell describes how the emphasis on the territorial integrity of the Comoros in the OAU mediation effort “confirmed suspicions among some of the secessionists that the Organization [of African Unity] was in no position to play the role of honest broker” (Cornwell Reference Cornwell1998:58). Yere proved indeed unable to formulate an agreement that the rebels were also willing to accept.
Since the shuttle talks conducted by Yere in August failed to produce a way out of the conflict, the OAU invited the conflict parties to a reconciliation conference in Addis Ababa. The conflict parties accepted the invitation, after which the conference was held from December 10 to 13, 1997. A public statement was drawn up at the end of the conference, calling for a “new institutional framework meeting the aspiration of the Comorian people.” However, the document did not specify what this institutional framework would look like. With the conflict remaining unresolved, the secessionists in Anjouan led by Abdallah Ibrahim held a constitutional referendum in February 1998, after which they once more declared the island a sovereign and democratic state. This triggered the OAU to send a ministerial delegation to Anjouan. Representatives from Tanzania, Madagascar, Zimbabwe, Kenya, Mauritius, South Africa, and Burkina Faso traveled to Anjouan to persuade Ibrahim Abdallah to refrain from forming an independent government. Yet, Abdallah remained committed to the secession of Anjouan. Abdallah’s refusal to give in to the collective pressure of these regional representatives marked the end of the mediation effort by the OAU to resolve the conflict (Cornwell Reference Cornwell1998:60).
Sudan, 1992–1993
At the annual OAU summit in 1991, African heads of state mandated their new chairman, Nigerian President Ibrahim Babangida, to mediate the armed conflict between Sudan and the SPLM/A that had been ongoing since 1983 (Khadiagala Reference Khadiagala2007:189; Wöndu & Lesch Reference Wöndu and Lesch2000:21–22). Sudanese government officials anticipated that mediation by Nigeria could restrain the rebels’ claim for self-determination, since Nigeria had overcome the secessionist claims of the Biafran rebels (Iyob & Khadiagala Reference Iyob and Khadiagala2006:94). Indeed, the government of Sudan negotiation team obtained a confirmation from Nigerian president Babangida prior to the negotiations in which he promised that “Sudan will never be dismembered” (Khalifa Reference Khalifa1999:102).
When Nigerian Foreign Minister Ike Nwachukwu officially opened the Abuja negotiations on May 26, 1992, he told the conflict parties to “make your country a nation in which all its citizens have a strong sense of belonging irrespective of ethnic or religious differences.” The chief mediator of the conference, Minister of Internal Affairs Tunji Olagunju, added that the peace talks should focus on finding a formula that could safeguard the diversities of Sudan. Several government negotiators highlighted during the peace talks that they were committed to finding such a formula. Angelo Beda stated that due to the artificial borders resulting from colonialism, Sudan needed to “weld a nation out of this collection of people who have different backgrounds” (Wöndu & Lesch Reference Wöndu and Lesch2000:30–31).
Nevertheless, talks soon focused on the issue of self-determination. In addition to the SPLM/A negotiators demanding a high degree of autonomy for southern Sudan, a breakaway faction of the SPLM/A expressed its wish to put the possibility of a referendum on the independence of Southern Sudan on the agenda. However, insisting that the unity of Sudan would be non-negotiable, Mohamed al-Amin Khalifa, the head negotiator of the government delegation, immediately responded by stating that “separation comes through the barrel of the gun and not by debate.” That secession was something that could not be negotiated was also advocated by the Nigerian mediation team. Chief mediator Tunji Olagunju told the conflict parties on the second day of the talks that Nigeria was “interested and committed to the unity of Sudan.” Another member of the mediation team, Lawan Gwadabe, reminded the adversaries on the sixth day of the peace talks that “it is not the intention of Nigeria to preside over the balkanization or break-up of Sudan.” As a result of the unwillingness of both the government of Sudan and the Nigerian mediators to consider a territorial solution to the civil war, peace talks broke down on June 4, 1992 (Wöndu & Lesch Reference Wöndu and Lesch2000:50–51).
On April 26, 1993, a second round of peace talks was held at Abuja, but this round of talks encountered the same difficulties as the first round of talks in Abuja. The Nigerian mediation team and the Sudanese government negotiators delegation kept insisting that talks on self-determination were a non-starter. As a result, the second round of the Abuja talks broke down on May 7, 1993, without any tangible results (Maundi et al. Reference Maundi, Zartman, Khadiagala and Nuamah2006:146).
Sudan, 1994–2001
Concerned with the negative consequences of Sudan’s civil war for the region, the Intergovernmental Authority on Development (IGAD) started to mediate between the government of Sudan and the SPLM/A in January 1994 (Iyob & Khadiagala Reference Iyob and Khadiagala2006:104). Since the first two rounds had yielded no results, not even on an agenda for negotiations, the IGAD mediation team drafted a Declaration of Principles (DoP) which the conflict parties were meant to review in preparation for a third round of negotiations in July 1994. The compromise envisioned in the DoP was that while the south would have the right to self-determination because of the high level of polarization in Sudan, priority should be given to unity to be created through a national consensus (El-Affendi Reference El-Affendi2001:585).
The government of Sudan refused to accept the DoP, as this would put the issue of self-determination on the agenda (El-Affendi Reference El-Affendi2001:586; Iyob & Khadiagala Reference Iyob and Khadiagala2006:106). Indeed, the leader of the government delegation, Ghazi Salah al-Din al-Atabani, criticized the DoP on the basis that “Self-determination-alias-separation of southern Sudan is bound to elicit a chain-reaction afflicting the rest of Africa. This is an eventuality that the founding fathers of the OAU consciously tried to avoid” (Brosché & Duursma 2017; El-Affendi Reference El-Affendi2001:588).
In spite of this initial unequivocal refusal to accept the DoP, the government of Sudan agreed to resume negotiations on the basis of the DoP at an IGAD summit in Nairobi in July 1997. The reason for the government of Sudan’s acceptance of the DoP is that, in contrast to 1994, the SPLM/A was in strong military position in 1997. Indeed, by early 1997, it was even conceivable that Juba would be captured by the SPLM/A. While the government of Sudan agreed to discuss the issue of self-determination, discussions on this matter remained deadlocked (El-Affendi Reference El-Affendi2001:588; Iyob & Khadiagala Reference Iyob and Khadiagala2006:112). It would not be until the U.S. became directly involved in the mediation process in 2001 that a territorial solution to the civil war became feasible.
Sudan, 2001–2005
While the U.S. pursued a confrontational approach in relation to Sudan throughout the 1990s, Washington shifted its policy toward constructive engagement from January 2001 onward, when the new American administration of President George W. Bush took office and formulated peace in Sudan as a foreign policy goal (Woodward Reference Woodward2006:94–95). At a secret meeting in Nairobi in May 2001, Sudanese and U.S. officials agreed that in return for a normalization of relations, Khartoum would make peace with the SPLM/A, as well as provide Washington with intelligence on terrorist networks that had been operating in Sudan throughout the 1990s. The U.S. commitment to this deal grew even stronger after the attack on the Twin Towers on September 11, 2001 (Cockett Reference Cockett2010:156–61; Woodward Reference Woodward2006:104–9).
In October 2001, the U.S., the U.K., and Norway agreed to form a Troika to coordinate their involvement in the IGAD-led mediation process. The Troika representatives agreed to put pressure on the conflict parties in order to move them toward compromise. Reflecting on the U.S. strategy to move Khartoum toward compromise, U.S. Deputy Assistant Secretary of State for Africa Charles Snyder notes how the incentive for the government of Sudan to make peace was clear: “Sudan would be allowed back into the international community” (Cockett Reference Cockett2010:161). These strong incentives to make peace lay at the heart of the government of Sudan’s decision to sign the Machakos Protocol on July 20, 2002. The Machakos Protocol stipulated that following a six-year interim period, southern Sudan was allowed to hold a referendum on whether to secede from northern Sudan.
Many other issues had to be negotiated in addition to the issue of self-determination for southern Sudan, but the conclusion of the Machakos protocol was arguably the most crucial milestone on the path toward the conclusion of the Comprehensive Peace Agreement between the government of Sudan and the SPLM/A in Naivasha on January 9, 2005. The Sudanese leaders anticipated that in return for the territorial concession made in the CPA, the U.S. would lift the economic sanctions and Sudan would be taken off the list of countries supporting terrorism (Cockett Reference Cockett2010; Duursma Reference Duursma2017b).
Conclusion
That African leaders are generally opposed to territorial revisions is a well-established theme in African International Relations research (Jackson & Rosberg Reference Jackson and Rosberg1982; Jackson Reference Jackson1990). Yet, few studies have addressed why the territorial integrity norm has persisted over such a long period of time. African leaders have internalized the territorial integrity norm, which is why an opposition to territorial revisions in Africa has become part of the identity interests of African leaders; Africa’s continent-wide opposition to territorial revisions means that mediation processes in which a territorial revision is proposed are relatively unlikely in Africa. In those processes in which a territorial revision is advocated for, non-African third parties will generally be more successful. Non-African mediators, and especially external major power mediators, have resources with which to compensate African governments for making territorial concessions to end their civil wars. Hence, secession as a means to resolve a civil war is more likely when mediators are not African, compared to when the mediators are Africans.
The cases cited here as examples illustrate that third parties involved in civil wars in Africa have acted in accordance with the territorial integrity norm for more than fifty years. African third parties put normative pressure on the conflict parties to respect the territorial integrity of the civil war country. This is evident in the cases of Nigeria and the Comoros, as well as in the Nigerian mediation effort in the north-south Sudan conflict. Once African third parties decide to push for secession—as in the Western Sahara conflict and in the north-south Sudan civil war—African mediation is not effective in securing settlements that stipulate independence. Finally, the U.S. involvement in the Namibia and the north-south Sudan conflicts show that major powers generally have more leverage than African third parties to move the government side toward agreeing to secession.
These examples confirm that third-party mediators involved in African civil wars are generally reluctant to press for secession. This reluctance originated from a belief that condoning secession might set a bad precedent that might incentivize secessionist groups in other African states to push for concessions. Yet, as a result of the internalization of the territorial integrity norm, African leaders have remained opposed to territorial revisions, even though the borders have become much more stable. This shows that third parties are not black boxes; every third party has a set of interests and norms when it becomes involved in mediation. These interests and norms matter, since third parties will try to find solutions to civil wars that are in line with these interests and norms.
Supplementary Material
To view supplementary material for this article, please visit https://doi.org/10.1017/asr.2018.103
Acknowledgments
For comments on earlier versions of this article, I would like to thank the participants of the Annual Meeting of the American Political Science Association in Washington DC, August 28–31, 2014, as well as the participants at the Conflict Research Society Annual Conference in Canterbury, September 14–15, 2015. I would particularly like to thank Caroline Hartzell, Govinda Clayton, Joakim Kreutz, Mathilda Lingren, Monica Toft, Neil MacFarlane, and Scott Sigmund Gartner for helpful comments and suggestions. Any mistakes are my own.
