Online appendix for "When Have Violent Civil Conflicts Spread? Introducing a Dataset of Substate Conflict Contagion"



Download 1.24 Mb.
Page12/18
Date30.04.2017
Size1.24 Mb.
#16926
1   ...   8   9   10   11   12   13   14   15   ...   18
no contagion from Chad or Uganda to Ethiopia in 1976, pending confirmation from Tareke. Tareke confirms in 5/24/10 e-mail.


Ethiopia, 1977 (vs. OLF – State A could be Chad, Democratic Republic of the Congo, Sudan, or Uganda)

  • Sudan  Ethiopia, 1977. UCDP conflict summary: “Apart from the national context, also the regional context provided opportunities for the rebel group. Relations between Ethiopia and neighboring Sudan were tense during the Dergue-period and the two countries attempted to destabilize each other by supporting rebel groups opposing the neighboring state’s regime. Thus, while Ethiopia funneled Soviet weapons and Libyan money to Sudanese rebel group SPLM/A (Sudan People’s Liberation Movement/Army) and allowed them to have bases on Ethiopian territory, the Khartoum government let OLF train its soldiers in southwestern Sudan.” No mention of Chad, Zaire, or Uganda.

    • SPLA didn’t exist until early 1980s, but Ethiopian support for southern resistance movements (leftovers from 1963-1972 Sudanese civil war) can be dated to 1976 – see Sudan 1983 below.

  • No mention of potential State As in Alex De Waal, Evil Days: Thirty Years of War and Famine in Ethiopia (New York: Human Rights Watch, 1991), pp. 65-70, 81-82.

  • No mention of potential State As in Dawit Wolde Giorgis, Red Tears: War, Famine and Revolution in Ethiopia (Trenton, N.J.: Red Sea Press, 1989), p. 117.

  • No mention of potential State As in Edmond J. Keller, Revolutionary Ethiopia: From Empire to People’s Republic (Bloomington: Indiana University Press, 1988), pp. 158-163.

  • Hence I will code no contagion from Chad, Democratic Republic of the Congo, or Uganda to Ethiopia in 1977, pending confirmation from Keller. Keller confirms in 5/20/10 e-mail. Says Sudan’s support for OLF was “minimal,” but I’ll favor the UCDP account.


Ethiopia, 1989 (vs. ALF – State A could be Chad, Somalia, Sudan, or Uganda)

  • There is no actual evidence of the conflict’s re-onset in 1989. There was a re-onset in 1996, when the ARDUF undertook armed struggle (UCDP conflict summary). See that conflict below.


Ethiopia, 1994 (vs. ONLF – State A could be Burundi, Chad, Djibouti, Republic of the Congo, Rwanda, Somalia, Sudan, or Uganda)

  • Somalia  Ethiopia, 1994. “Via the massive wave of returnees to the Ogaden following Somalia’s [1991] disintegration, the ONLF managed to establish a presence within the territory it sought to liberate.” Tobias Hagmann and Mohamud H. Khalif, “State and Politics in Ethiopia’s Somali Region Since 1991,” Bildhaan: An International Journal of Somali Studies, Vol. 6 (2006): 25-49, p. 40. No mention of other potential State As.

  • No mention of potential State As’ contribution to onset in UCDP conflict summary.

  • No mention of potential State As (besides Somalia) in John Markakis, “The Somali in Ethiopia,” Review of African Political Economy, Vol. 23, No. 70 (1996): 567-570.

  • No mention of potential State As (besides Somalia) in Theodros Dagne, “Africa and the War on Terrorism: The Case of Somalia,” Mediterranean Quarterly, Vol. 13, No. 4 (2002): 62-73, pp. 68-69.

  • Hence I will code no contagion from Burundi, Chad, Djibouti, Republic of the Congo, Rwanda, Sudan, or Uganda to Ethiopia in 1994, pending confirmation from Hagmann. Hagmann confirms in 5/27/10 e-mail.


Ethiopia, 1995 (vs. al-Itahad al-Islami – State A could be Burundi, Chad, Djibouti, Republic of the Congo, Rwanda, Somalia, Sudan, or Uganda)

  • Somalia  Ethiopia, 1995. “Ethiopia conducted occasional cross-border raids against Islamist terrorists in the country [Somalia] whom Addis Ababa blamed for several attacks in Ethiopia itself.” Daniel L. Byman and Kenneth M. Pollack, Things Fall Apart: Containing the Spillover from an Iraqi Civil War (Washington: Brookings Institution Press, 2007), p. 167.

  • Sudan  Ethiopia, 1995. “Sudan’s attitude towards terrorism in Kenya was complicated by the fact that its government provided support to the Islamist group Al-Ittihad al-Islami (Islamic Union). Al- Islami was founded around 1991, at the end of Siyyad Barre’s regime, with the objective of finding a minimal element of ‘national’ cohesion premised on Islamist ideology. Unfortunately, Al-Ittihad became an instrument of Sudanese foreign policy, declaring a jihad against ‘infidels’ in the region. The organization conducted activities in Ethiopia through the intermediary of its large resident Somali population.” (Eric E. Otenyo, “New Terrorism: Toward an Explanation of Cases in Kenya,” African Security Review, Vol. 13, No. 3 (2004): 75-84, p. 78) This is Sudan  Ethiopia contagion because the rise of the Islamist NIF regime in Sudan in 1989 was related to the conflict in Southern Sudan (see “Eritrea” below). No mention of other potential State As (besides Somalia).

  • No mention of potential State As (besides Somalia) in David H. Shinn, “Al Qaeda in East Africa and the Horn,” Journal of Conflict Studies (Summer, 2007).

  • No mention of potential State As (besides Somalia) in Tobias Hagmann and Mohamud H. Khalif, “State and Politics in Ethiopia’s Somali Region Since 1991,” Bildhaan, Vol. 6 (2006): 25-49, p. 37.

  • No mention of potential State As (besides Somalia) in Ken Menkhaus, “Political Islam in Somalia,” Middle East Policy, Vol. 9, No. 1 (2002): 109-123.

  • Hence I will code no contagion from Burundi, Chad, Djibouti, Republic of the Congo, Rwanda, and Uganda to Ethiopia in 1995, pending confirmation from Menkhaus.


Ethiopia, 1996 (vs. ARDUF – State A could be Burundi, Central African Republic, Chad, Democratic Republic of the Congo, Djibouti, Republic of the Congo, Rwanda, Somalia, Sudan, or Uganda)

  • No mention of potential State As in UCDP conflict summary.

  • No mention of potential State As’ contribution to onset in Patrick Gilkes, Ethiopia: Perspectives on Conflict, 1991-1999 (Swiss Peace Foundation, 1999).

  • No mention of potential State As’ contribution to onset in T. Berhe and Y. Adaye, “The Impact of Local Conflict on Regional Stability” (no further information on reference available).

  • No mention of potential State As’ contribution to onset in Yasin Mohammed Yasin, “Political History of the Afar in Ethiopia and Eritrea,” Afrika Spectrum, Vol. 42, No. 1 (2008): 39-65.

  • Hence I will code no contagion from Burundi, Central African Republic, Chad, Democratic Republic of the Congo, Djibouti, Republic of the Congo, Rwanda, Somalia, Sudan, or Uganda to Ethiopia in 1996, pending confirmation from Schwab. Schwab confirms in 5/19/10 e-mail.


Eritrea, 1997 (vs. EIJM – AS – State A could be Burundi, Central African Republic, Chad, Democratic Republic of the Congo, Djibouti, Ethiopia, Republic of the Congo, Rwanda, Somalia, Sudan, or Uganda)

  • Ethiopia  Eritrea, 1997. The “EJM” not only split off from the ELF in the 1970s (UCDP conflict summary); it was also apparently supported by the ELF in Sudan during the 1990s. (Ruth Iyob, “The Eritrean Experiment: A Cautious Pragmatism?” Journal of Modern African Studies, Vol. 35, No. 4 (1997): 647-673, pp. 665-667. (ELF was last active in the Ethiopian conflict in 1980, according to UCDP, but it remained active in Sudan thereafter.) No mention of other potential State As (besides Sudan).

    • Also a refugee spillover effect: “During the prolonged stay in Sudan, many refugees [from Ethiopia], the great majority of whom were Muslim, became increasingly integrated through residence and education into northern Sudanese Arabo-Islamic culture. Saudi Arabia had initially been a major supporter of refugee education and of a strong Islamic emphasis in the curriculum. After the Sudanese National Islamic Front government took power it supported Eritrean Islamic Jihad (EIJ) and provided sanctuary for factions of the defeated ELF. After independence, … EIJ … began guerilla activities in western Eritrea.” (David Pool, From Guerillas to Government: The Eritrean People’s Liberation Front (Oxford, U.K.: James Currey, 2001), p. 192) No mention of other potential State As.

  • Sudan  Eritrea, 1997. “The seeds of the Eritrean-Sudanese quarrel were sown in 1989, when the NIF reportedly sponsored the creation of the Eritrean Islamic Jihad Movement (EIJM).” (Amare Tekle, “International Relations in the Horn of Africa (1991-1996), Review of African Political Economy, Vol. 23, No, 70 (1996): 499-509, p. 505) This is contagion because the civil war in Southern Sudan was one of the causes of the rise of Omar al-Bashir and the NIF to power in 1989.

    • “[In early 1989] Sadiq [previous ruler] claimed his new government was committed to ending the southern civil war by implementing the November 1988 DUP-SPLM agreement. He also promised to mobilize government resources to bring food relief to famine areas, reduce the government's international debt, and build a national political consensus. Sadiq’s inability to live up to these promises eventually caused his downfall. On June 30, 1989, Colonel (later Lieutenant General) Umar Hassan Ahmad al Bashir overthrew Sadiq and established the Revolutionary Command Council for National Salvation to rule Sudan.” (Helen Chapin Metz, ed., Sudan: A Country Study (Washington: GPO for the Library of Congress, 1991), http://countrystudies.us/sudan/28.htm)

  • No mention of potential State As (besides Ethiopia and Sudan) in UCDP conflict summary.

  • No mention of potential State As (besides Ethiopia and Sudan) in Gaim Kibreab, Eritrea: A Dream Deferred (Woodbridge, U.K.: James Currey, 2009), pp. 90-98, 319-322, 380. Does note that refugees from Ethiopia were among those who founded EIJM in eastern Sudan in the 1980s (p. 211).

  • No mention of potential State As (besides Ethiopia and Sudan) in Dan Connell, Rethinking Revolution: New Strategies for Democracy and Social Justice: The Experiences of Eritrea, South Africa, Palestine, and Nicaragua (Lawrenceville, N.J.: Red Sea Press, 2002), pp. 18, 123.

  • Hence I will code no contagion from Burundi, Central African Republic, Chad, Democratic Republic of the Congo, Djibouti, Republic of the Congo, Rwanda, Somalia, or Uganda to Eritrea, pending confirmation from Iyob. Iyob confirms in 5/20/10 e-mail.


Angola, 1975 (vs. FNLA/UNITA – State A could be Madagascar, South Africa, or Zimbabwe)

  • China  Angola, 1975. See below.

  • South Africa  Angola, 1975. UCDP conflict summary: “At the outset of the civil war, there were substantial elements of foreign involvement, as the domestic conflict was drawn into both the southern African regional conflict and the global power struggle between the superpowers USA and the USSR. … The FNLA [and UNITA]… received … significant military and logistic support … from South Africa. Vehemently anti-communist, South Africa felt threatened by MPLA’s close relationship with the Soviet Union, as well as by the presence of Cuban troops so close to its borders. Pretoria also wanted to retaliate for MPLA’s support for ANC and SWAPO.” No mention of Madagascar or Zimbabwe.

  • Democratic Republic of the Congo  Angola, 1975. “[Among the reasons Mobutu supported the FNLA militarily,] fourth, Mobutu had been angry for some time at MPLA for its close association with the Katanga gendarmes.” (W. Martin James III, A Political History of the Civil War in Angola 1974-1990 (New Brunswick, N.J.: Transaction Publishers, 1992, p. 60) No mention of Madagascar. In November 1975 “Savimbi … circumspectly acknowledged South African, Rhodesian, and French military support.” (James 1992, 66) Not clear if this was related to the conflict in Rhodesia/Zimbabwe; the contribution seems quite marginal in any case and is barely mentioned in the rest of the book. The footnote (p. 84, no. 74) also mentions Biafra (Nigeria), but again in a very marginal sense. On p. 180, James lists 31 foreign supporters of UNITA (most prominently China), though actual levels of support (with the exception of China) seem pretty minimal.

  • No mention of Madagascar or Rhodesia/Zimbabwe (with respect to Angola) in James Ciment, Angola and Mozambique: Postcolonial Wars in Southern Africa (New York: Facts on File, Inc., 1997), pp. 44-63.

  • No mention of Madagascar or Rhodesia/Zimbabwe in Assis Malaquias, Rebels and Robbers: Violence in Post-Colonial Angola (Uppsala: Nordiska Afrikainstitutet, 2007), pp. 57-80.

  • Hence I will code no contagion from Madagascar or Rhodesia/Zimbabwe to Angola in 1975, pending confirmation from Malaquias.


Angola, 1991 (vs. FLEC-R – State A could be Comoros, Mozambique, or South Africa)

  • No mention of potential State As in UCDP conflict summary. Does note: “The various FLEC factions initially received logistical support from Zaire, which had its eyes on Cabinda’s oil, but after President Neto of Angola and President Mobutu of Zaire signed an agreement to improve the two states’ relations, this source of support dried up. No new foreign help was forthcoming until in the late 1980s, when Zaire once more initiated contact, supplying FLEC-R with both weapons and rear bases. FLEC-FAC, for its part, received similar support from Congo-Brazzaville. This support remained constant until 1997, when the regimes in both Zaire and Congo-Brazzaville were overthrown, with Angolan support.” Neither Congo constitutes a conflict link, since one was well before 1991, one was after, and Congo-Kinshasa was free of conflict in the 1980s and early 1990s.

  • No mention of potential State As (with respect to Angola) in James Ciment, Angola and Mozambique: Postcolonial Wars in Southern Africa (New York: Facts on File, Inc., 1997), p. 171.

  • No mention of potential State As in Joao Gomes Porto, “Cabinda: Notes on a Soon to be Forgotten War,” African Security Analysis Program Occasional Paper, Institute for Security Studies (2003).

  • Hence I will code no contagion from Comoros, Mozambique, or South Africa to Angola in 1991, pending confirmation from Porto.


Mozambique, 1977 (vs. RENAMO – State A could be Angola, South Africa, or Zimbabwe)

  • Zimbabwe  Mozambique, 1977. UCDP conflict summary: “On the regional level, Mozambique’s independence threatened the white regimes in both Rhodesia and South Africa, since the new, Marxist, government openly supported Rhodesian rebel group ZANU (Zimbabwe African National Union) and South Africa’s ANC (African National Congress). Subsequently, the white regime in Rhodesia set about to assist the creation of a counter-revolutionary force, and was in this endeavor able to exploit the public disaffection with Frelimo, which was rampant through much of the Mozambican countryside.” Also South Africa  Mozambique if (1) South Africa helped Mozambique rebels too in 1977 (not sure it did) and (2) ANC conflict was active in 1977 (not sure it was). Surprisingly, no mention of Angola.

    • UCDP conflict summary: “Following Rhodesia’s independence in 1980 and the rise to power of a government friendly towards Frelimo, sponsorship of Renamo was taken over by South Africa. South Africa, as had been the case with Rhodesia, felt threatened by black left-wing rule in Mozambique and other southern African states, and saw Renamo as a potential vehicle for destabilization of Mozambique and a tool for fighting anti-apartheid groups. Consequently, Renamo bases were relocated to Transvaal in northeast South Africa, and Pretoria took over the role of training and equipping the rebels.” Sounds like this was post-onset only.

  • South Africa’s involvement with RENAMO seems entirely post-onset in James Ciment, Angola and Mozambique: Postcolonial Wars in Southern Africa (New York: Facts on File, Inc., 1997), pp. 110-116. No mention of Angola (with respect to Mozambique), pp. 64-70.

  • No mention of Angola, and no mention of South Africa before 1979, in Glenda Morgan, “Violence in Mozambique: Towards an Understanding of Renamo,” Journal of Modern African Studies, Vol. 28, No. 4 (1990): 603-619.

  • No mention of Angola’s contribution to onset, and no mention of South Africa before 1979, in Margaret Hall, “The Mozambican National Resistance Movement (Renamo): A Study in the Destruction of an African Country,” Africa, Vol. 60, No. 1 (1990): 39-68, pp. 39-40.

  • Hence I will code no contagion from Angola or South Africa to Mozambique, pending confirmation from Ciment. E-mail bounces; Morgan and Hall can’t be located either.


South Africa, 1981 (vs. ANC – State A could be Angola, Mozambique, or Zimbabwe)

  • Zimbabwe  South Africa, 1981. “By 1980, a combination of guerilla warfare and negotiation had installed an African government in Zimbabwe. … The militantly anti-colonial bent of the new rulers throughout southern Africa created a hostile environment for the South African government. Because the country’s buffer with independent Africa was gone, these ‘frontline states’ became potential launching grounds for guerilla activity in South Africa.” (Iris Berger, South Africa in World History (Oxford, U.K.: Oxford University Press (2009), p. 140) No mention of the conflicts in Angola and Mozambique, though notes that the ANC studied Vietnam (p. 142).

  • No mention of State As’ conflicts in terms of contribution to onset in UCDP conflict summary. Does note: “In the mid-1970s, the conditions for guerrilla struggle against South Africa improved. Political upheaval in Portugal in 1974 paved the way for the independence of Angola and Mozambique, under governments led by the old ANC allies MPLA and Frelimo, respectively. This opened up the possibility for establishing camps in those countries, and launching cross-border attacks from Mozambique.” But this is not contagion because the external supporters of ANC were friendly state governments, not rebel groups in Angola or Mozambique. Also notes: “Attempts were made in 1967-68 to create a corridor through Zimbabwe’s Matabeleland through joint operations with ZAPU.” This is too far from 1981 to count as contagion (the Zimbabwe conflict is not even coded as starting until 1972).

  • No mention of potential State A conflicts’ contribution to onset in Nigel Worden, The Making of Modern South Africa: Conquest, Apartheid, Democracy, Fourth Edition (Malden, Mass.: Blackwell Publishing, 2007), pp. 126-148. (I believe Mozambique supported the ANC before South Africa supported RENAMO, meaning there was no contagion of the RENAMO conflict.)

  • No mention of potential State A conflicts’ contribution to onset in Leonard Thompson, A History of South Africa, Third Edition (New Haven: Yale University Press, 2001), pp. 221-235.

  • Hence I will code no contagion from Angola or Mozambique to South Africa, pending confirmation from Berger. Berger confirms in 5/24/10 e-mail.


Lesotho, 1998 (vs. military faction – State A could be Angola or Comoros)

  • No mention of either potential State A in UCDP conflict summary.

  • No mention of either potential State A in Khabele Matlosa and Neville W. Pule, “Civil-Military Relations in Lesotho, 1966–1998: Problems and Prospects,” in Rocky Williams, Gavin Cawthra, and Diane Abrahams, Ourselves to Know: Civil-Military Relations and Defense Transformation in Southern Africa (Pretoria: Institute for Security Studies, 2002).

  • No mention of either potential State A in terms of contribution to onset in Fako Johnson Likoti, “The 1998 Military Intervention in Lesotho: SADC Peace Mission or Resource War?” International Peacekeeping, Vol. 14, No. 2 (2007): 251-263.

  • Hence I will code no contagion from Angola or Comoros to Lesotho, pending confirmation from Matlosa.


Madagascar, 1971 (vs. Monima – State A could be South Africa)

  • No mention of South Africa in (very brief) UCDP conflict summary. It’s mentioned that MONIMA was Maoist.

  • No mention of South Africa in Solofo Randrianja and Stephen Ellis, Madagascar: A Short History (London: Hurst & Company, 2009), pp. 184-185.

  • No mention of South African (Namibian) conflict contributing to onset in Philip M. Allen, Madagascar: Conflicts of Authority in the Great Island (Boulder: Westview Press, 1995), pp. 59-64.

  • Hence I will code no contagion between South Africa and Madagascar, pending confirmation from Allen. Allen confirms in 5/24/10 e-mail.


Comoros, 1989 (vs. Presidential Guard – State A could be Angola, Mozambique, or South Africa)

  • South Africa  Comoros, 1989. UCDP conflict summary: “France and South Africa - previous backers of Denard and the Presidential Guard, but now with new reform-minded governments, embarrassed by the excesses of the Guard - pressured Abdallah to dispense with the mercenaries. Abdallah agreed to both expel Denard and his fellow officers by the end of the year and to absorb the GP into the regular army. It is widely believed that it was Denard who, faced with deportation, shot the president to death on the night of 26-27 November 1989.” This is contagion if the “reform-minded government” in South Africa came to power as a result of ANC violence there (surely it did). No mention of Angola or Mozambique.

    • “The end of the Cold War, the persistent calls from Western donors for democratization, and the striking turn of events in South Africa following the release of Nelson Mandela were the harbingers for a new political era in the Comoros. A shift in foreign policy in France and South Africa resulted in the two countries combining their leverage and set the ground for ousting Denard from the Comoros in December 1989, against the backdrop of President Abdallah’s assassination.” (Ali Y. Alwahti, “Prevention of Secessionist Movements in a Micro-State: The International Mediation in the Comoros Islands,” Swords & Ploughshares, American University (date unknown), p. 2) No mention of Angola or of Mozambique’s contribution to onset.

  • No mention of Angola or Mozambique’s conflicts’ contributions to 1989 Comoros onset in Eliphas G. Mukonoweshuro, “The Politics of Squalor and Dependency: Chronic Political Instability and Economic Collapse in the Comoro Islands,” African Affairs, Vol. 89, No. 357 (1990): 555-577. Does note (p. 563): “In May 1978, a group of 50 European mercenaries and a German Shepherd dog, led by ex-Congo mercenary Colonel Bob Denard, overthrew the regime.” This could suggest DRCComoros contagion, but Denard was a French national who had numerous mercenary experiences in his career; Katanga was just one of them.

  • Hence I will code no contagion from Angola or Mozambique to Comoros in 1989, pending confirmation from Alwahti.


Comoros, 1997 (vs. MPA/Republic of Anjouan – State A could be Angola or Mozambique)
1   ...   8   9   10   11   12   13   14   15   ...   18



The database is protected by copyright ©ininet.org 2024
send message
    Main page
Guide
Instructions
Report
Request
Review