Ethiopia’s Long War in Historical Perspective Part 3

The episode of 1967-68 is telling in many respects. For Hishal, who triggered the factional infighting, Jebha could not go on as it once did. Whereas he and others like him had been willing to cooperate with their Christian comrades in better times, the debacle of 1967 broke down the credible commitments between these groups. The response of the Christians and their allies was in part a direct reaction to these killings, but also a recognition that the organization was failing. Their criticisms of the decentralized zonal system were a telling indication of the latter. In any case, once the initial breakdown of cooperation occurred, factional infighting spiraled out of control, diffusing and incorporating broader cross-sections of the organization. Fragmentation was the end result. While this factional moment was complex, involving a range of decisions and strategic calculations that should be examined more closely, one thing is clear: Jebha’s performance failures were a key driver of factional infighting, and one cannot understand the events of 1967 without accounting for the context of territorial loss in which they occurred. Slow Recovery, Stalemate, and Unity: Jebha 1969-1973 After the military failures of 1967-68, and the torrent of factional infighting that it produced, Jebha began a slow, halting recovery. Though the Ethiopian offensive had drastically curtailed Jebha’s operations in many parts of the country, in the long term, it had done little to permanently damage the guerilla army’s structure. In addition, supplies from Arab supporters resumed at higher levels in the years after the 1967Arab-Israeli War, serving to sustain and enhance Jebha’s operations (Lefebvre, Middle East Conflicts and Middle Level Power Intevention in the Horn of Africa 1996). Under the Palestine Liberation Organization’s (PLO) advisement, Jebha began to expand its tactical repertoire. Intent on capturing headlines, Jebha destroyed Ethiopian commercial jetliners as far afield as Karachi, Rome, and Frankfurt in 1969 and early 1970. Vital transport routes like the Asmara-Massawa and the Asmara-Tessenei roads became hazardous to travel, as Jebha did its utmost to exact an economic price for Ethiopia’s occupation. Foreigners were kidnapped or detained, including the US counselor in Asmara, and in November 1970 the most senior military commander in the province – Teshome Ergetu – was killed in ambush on the Asmara-Keren road.49 The United States, which had a large signals installation in Asmara, began to worry about the security of over 2000 American personnel who operated the installation.50 Surprised by Jebha’s staying power, the Ethiopians declared martial law over the province in 1970, and in 1971 replaced the civilian Gov. General with a military man and a known hardliner, General Debebe Haile-Mariam. After the First National Congress at Arr in 1971, comprised of 500 delegates elected by the field units themselves, Jebha’s re-organized the zonal commands into a more centralized administrative structure.51 With a new, more rational system of command and organization, and between 1500 -2000 trained fighters, Jebha seemed rejuvenated. Yet despite these new realities, Jebha faced far from a rosy scenario: the splinter groups of 1967-68, despite numerous conciliatory gestures, refused to rejoin. With their hand forced, Jebha leaders attempted to liquidate these splinter elements in 1972, declaring in the First Congress Political Programme that the “the Eritrean field can only tolerate one 49 Because Jebha had split by this point, it is unclear that they were responsible for all of these attacks. For example, from the American documents, we know that Ali Said Abdella, Mussie Tesfamichael, and Yohannes Sebahtu, were all involved in the airplane bombings- however, it is widely known that these individuals would later become part of the leadership of the splinter elements that coalesce into Shaebia in 1972. Furthermore, (Erlich 1983) claims that it was dissident factions that killed Teshome Eregtu in 1971. Press clippings from the period reveal that Jebha was often unwilling to claim credit for these attacks. In any case, it was obvious by this point that Jebha, as the major rebel organization in the province, had recovered from the low point of summer 1967 50 This was in fact, the reason the Americans were writing so many reports about insurgent activities. Scarcely a month went by during this period where dispatches did not indicate fear at the prospect of direct Jebha attack on America assets in the province. However, such attacks never materialized 51 This is an important point that should be discussed more. Jebha essentially went from a loose coalitional structure to a highly centralized Leninist organizational structure nationalist organization.” 52 Despite its resource advantages Jebha was unsuccessful, and instead of directing the balance of its forces against the Ethiopians, began a costly war with its co-nationalists.53 Also of critical importance during this period was the Sudanese closure of overland routes into western Eritrea, and Jebha offices in Kassala and Khartoum. Prompted by Ethiopia’s support of a peace agreement that ended the first phase of the Sudanese civil war in 1972, Sudan’s actions caused great hardship for Jebha cadres, as Sudan had been a key conduit of Arab arms into Eritrea, a rear base for Jebha’s military operations, and home to the now tens of thousands of Eritrean refugees that served as Jebha’s core supporters. Indeed, Sudan’s actions were symptomatic of its wider ambivalence about Jebha, and presaged the rocky relationship it would have with the Eritrean nationalist movement throughout the late 1970s and 80s. In any case, the position of the Ethiopian military in the province was simply too strong to be challenged. Backed by their erstwhile American and Israeli supporters, the chance that the Ethiopian military would incur significant losses at the hands of an increasingly bold, but outgunned band of rebels was remote.54 At the same time, the Ethiopian’s seemed unable to make any major military breakthrough, if in fact there was one to be had. American intelligence assessments from the period, in a telling series of memos, describe what can only be interpreted as a stalemated military situation in the early 1970’s, noting that the insurgents “lack the ability to win a decisive victory” but that the 52 ELF 1st National Congress Political Programme See (Pool 2001) for the dynamics of this conflict, as well as (Markakis 1987). Also, authors interview, Shaebia Commander, May 28, 2009- commander noted that during the 1972-74 war between Jebha and Shaebia, the Ethiopian’s barely launched any attacks against the rebels, and were largely content to watch the rebels fight amongst themselves 54 See (Lefebvre, Arms for the Horn: US Security Policy in Ethiopia and Somalia 1953-1991 1991) for the best and most complete discussion of this support 53 Ethiopian army had no chance to “break the stalemate without substantially expanding and improving its counter insurgency efforts.”55 Such assessments were reflected in territorial terms. While Jebha had made some initial, very slow recoveries in Western Eritrea, and was able operate in Barka, Gash-Setit and parts of Semhar along the Asmara-Massawa road by 1970, these shifts were a pittance compared to the losses it had incurred in 1967-68. Furthermore, between 1970 and 1973, there was no perceptible change in the territory over which Jebha was able to operate. Interestingly, this period was characterized by an absence of factional infighting or fragmentation within Jebha ranks, a fact that is more striking when one considers the potentially divisive organizational elections held at Arr in 1971. In this context, it is clear that territorial stalemate served to underpin cooperation within Jebha ranks, providing the basis of what I have identified as a “cohesive stalemate.” While Jebha cadres – and the elites who led them – had reason to believe that the organizational status quo was sustainable, the threat to their collective security loomed large. Such fertile conditions for cooperation, however, were about to quickly change. Gains and Fragmentation: Jebha 1974-77 The Ethiopian Revolution of 1974 substantially altered the dynamics between Jebha and the Ethiopian government. The literature on the Ethiopian Revolution is vast, and I do not wish to recount all of it here.56 In short, following a set of army mutinies (the first in Eritrea), a series of strikes in Addis Ababa, and student protests, the monarchy of Haile Selassie was overthrown. A cabal of junior military officers called the Derg, ostensibly Marxists, took power in the chaos of mass 55 INR Research Study, T.Murphy, May 10, 1973, RG 59, Central Files, “Ethiopia: Eritrean Liberation Front Persists Despite Weakness.”- the report also went on to say that “The ELF is not likely soon to jeopardize Ethiopian control over Eritrea, but the movement is not dead.” See also INR Research Study, A. Palmer, Feb 24, 1972, RG 59, Central Files, “Ethiopia: Status of Eritrean Liberation Front.” 56 See (Clapham 1990) and (Keller 1991) upheaval.57 While the causes of the revolution were deep and complex, and the insurgency in Eritrea was only a minor factor. Somewhat ironically, the military’s takeover exacerbated political instability in Addis Ababa, sparking a series of battles between the Derg and the student movement for control of the government, as well as a sequence of major confrontations within the Derg. Under siege by its external and internal opponents, the military became paralyzed and unable to prosecute the war in Eritrea. As the central authorities in Addis began to devour themselves, Jebha managed to strike a tactical alliance with its splinter groups which by then had been consolidated into an organization called “Shaebia.” The insurgents took the initiative, and began to expand from one district to the next. In early 1975 the nationalist organizations made real breakthroughs, not only penetrating the Christian highland plateau, but attacking the provincial capital of Asmara—Eritrea’s largest city and the second the largest urban center in Ethiopia. For three days at the end of January, the city was on fire. In the streets, the rebels fought running battles with government forces. While Jebha failed to seize the city, the Asmara raids signaled a major transformation in the direction and scale of the war. The Ethiopian response in the aftermath of the fighting was as tragic as it was consequential. Angry government soldiers went on a rampage, dragging people out of their homes and murdering them. Youths were garroted with piano wire and disappeared. Stores and businesses were looted. The social and political order in Eritrea’s cities and towns was beginning to break down.58 57 58 Derg means “committee” in Amharic For a systematic accounting of these atrocities, see (de Waal 1991) In the months ahead, terror became the mainstay of the Ethiopian Army in Eritrea. The degree of brutality that the Marxist military regime was willing to apply – particularly on the Christian highland plateau – was like nothing Eritrean civilians had seen during the Haile Selassie era. 59 That many of Eritrea’s youths faced pervasive insecurity in government controlled urban areas at precisely the time that the government seemed to be teetering on the brink of collapse triggered mass defection to the rebels.60 While Muslims had always gravitated towards the secessionist movement, Christian youths flocked into Jebha in ways that had never been seen before. In 1975, an organization that was no more than 2000 people took in nearly 10,000 recruits. 1976-77 saw a similar spike in recruitment. Jebha’s training camps and infrastructure could barely keep pace with the expansion in numbers.61 In his memoirs, former Derg Governor of Eritrea Dawit Wolde Giorgis captures the sentiment of the time quite clearly when he writes: “More men and women fled to join the movements. With indiscriminate death all around, it was either stay and die, or flee and fight. Anywhere from 300 to 800 people slipped out of the cities each month to join the rebels. Only the aged, the sick, the weak, and those with connections to the military in Asmara were left behind. Out of a total population of over 200,000 in 1974, Asmara dwindled to less than 90,000 people in 1977. Many of those remaining spent their nights in the churches for fear that they would be dragged out of their beds and killed by the death squads. The rest of the major towns were also virtually abandoned. Another individual, who actually made the decision to join Jebha during this tumultuous period, puts it even more succinctly when he says: 59 Authors interview, Provincial Bureaucrat, Oct 11th, 2009 (Goodwin 2001) comparative analysis of revolutions makes exactly this sort of argument 61 Authors interviews, Jebha fighters, April 22, 23, and May 3, 2009. Many of these fighters indicated that Jebha officials often asked new recruits to wait before leaving home to join Jebha camps, because the organization simply lacked the capacity to absorb all of them so quickly 60 “The way the Ethiopians were treating the public provided a good ingredient to leave and join the revolution. They were torturing people; younger people were the main targets, killing them in the streets. So I decided to join the revolution and fight for my country, instead of dying in the streets like my other friends.” 62 Many of the urbanites who joined Jebha during this period were relatively well educated and brought with them valuable technical skills— clerks, accountants, doctors, and mechanics. Heavy equipment and trucks were captured from retreating government forces, a shift in logistics that allowed Jebha to seamlessly transition from guerilla warfare to conventional combat. By 1977, nationalists had captured nearly 90 percent of Eritrea and every major town except Asmara, Barentu, and Massawa, although Jebha only controlled a little over half of the province (see Figures 7 & 8). Combined, Jebha and Shaebia outnumbered the 20,000 Ethiopian combat troops stationed in Eritrea (Connell, Against All Odds: A Chronicle of the Eritrean Revolution 1993). Fortuitously, or perhaps by design, the Somali regime of Mohammed Siad Biarre took the opportunity to wage an offensive of his own against the Ethiopian government, penetrating deep into the Ethiopia’s southeastern province of Ogaden. Meanwhile, separate insurgencies had spontaneously appeared in nearly all of Ethiopia’s other 13 provinces. 62 Authors interview, Jebha fighter, April 22, 2009 Figure 7: Jebha’s Area of Operation, by District, January 1974 It was ironic that as Jebha was achieving significant territorial gains, it descended into bloody factional infighting. The trigger was the 2nd National Congress in May 1975, where Vice-Secretary General Hirui Tedla Bairu – an individual popular in Christian circles whose father was a prominent unionist and former Ethiopian Ambassador to Sweden 63 - lost his post in an election marked by rancor and recriminations. Herui had earned the disfavor of many of the more hard line conservative Muslim elements of the leadership,64 many of whom were linked to the Iraqi Baathist party, for his overtures to Shaebia on issues of nationalist unity. With the prodding (and perhaps threats) of the Iraqi representative at the 2nd National Congress, Abu Ala, members of the leadership 63 64 Authors interview, Jebha fighter, June 10, 2009 Abdullah Idris is often listed as one of these people voted Herui out of power.65 Herui himself claimed that the whole process was rigged, arguing that his ballots were thrown and that the elections should have been direct instead of tiered— as had been the case during the election of 1971. Interestingly, many of the actors that sought to oust Herui from the leadership in 1975 as Jebha was surging, had been among those who supported his candidacy in 1971 in the hope that he could attract Christian support66. The incentives to cooperate seemed to have changed. Added to this combustible mix, was the issue of the political status of the thousands of new, mainly Christian recruits who were swelling Jebha’s ranks. By this point, most of these recruits were in three major training camps in Jebha rear bases in Barka, and were demanding that they be given equal representation at the Congress and allowed to elect delegates. Yet many in the leadership argued that as newcomers, these recruits should not be granted equal voting status with older, mostly Muslim members. In the end, the votes of these new recruits were counted at 1/3rd of those of older members, and while substantially affecting the outcome of the 1975 elections, the feeling that Christian influence was being deliberately limited was widespread. After all, Christians were now the majority within Jebha. 65 66 Authors interview, Jebha intelligence officer, Sept 11, 2009. Also see (Medhanie 1986) Authors interview, Jebha intelligence officer, May 19 and Sept 11, 2009 Figure 8: Jebha’s Area of Operation, by District, March 1975 Voted out power, Herui was upset. Had he let the issue rest, the conflagration that was to come may have been averted. Instead, he went to the camps and sought to mobilize support, accusing the leadership of willfully trying to suppress Christian aspirations. Facing severe sanctions from the Jebha’s Central Committee, Herui went to Sudan and founded a new group called the Eritrean Democratic Movement (EDM) or Ma Ma Gu. Responding to these events, Christian fighters in the field began to challenge the leadership on a whole range of issues. Chief among these was the relationship between Jebha and Shaebia and the issue of nationalist unity, which many Christian members deeply desired.67 The agitation, with the support of several Christians in the leadership organs of Jebha, rapidly spread through several units, 67 Shaebia had a large Christian base, and many, if not most, of its core leaders were Christian forcing the leadership to hold open seminars to defend their policies. This was to no avail, as truculent fighters demanded a fresh Congress to debate many of the major policy issues Jebha now faced, and published a manifesto formally declaring themselves the “Revolutionary Democrats.” Exemplifying the wide division that was now emerging within Jebha ranks, opponents of the insurrectionists responded by giving them the pejorative title of “Falool.”68 By the summer of 1977, as Jebha seized major towns like Adi Keih, Agordat, and Mendefera, Falool had become open insubordination. With few alternatives, Chairman Ahmed Nasser and military chief Abdullah Idris decided to disarm the rowdy elements, launching an attack on a unit dominated by Falool members, Brigade 149. However, the attack failed to resolve the situation, as by this late hour Falool had mushroomed into a much larger phenomenon. In response, Falool killed two members of the leadership in an ambush in Danakil— Abdul Kadir Ramadan and Ali Mohammed Ibrahim. All-out war ensued. Yet outmaneuvered and unable to mobilize a critical mass of support, Falool’s play for power largely failed.69 Facing complete destruction at the hands of the leadership and its supporters, Falool elements fled to the Ala plains. Surrounded by both Jebha and Shaebia forces, they defected to Shaebia in the summer of 1977 largely out of desperation, nearly 2000 in all. 70 The EDM and Falool episode is instructive. That Herui Tedla Bairu – and by extension some of the Christian interests he represented – was deliberately excluded from the leadership in 1975 by those who had been eager to include him in 1971 raises many questions. What changed? As Jebha began 68 69 “Falool” is the Tigrinya word for “anarchists” Part of Falool’s problem was that key Christian leaders, like Seyoum Ogbamichael, mobilized against them. In fact, it was Ogbamichael who coined the well known jibe against Fallool “ ha hu b el faloolay ab seretna metse,” which basically means “Learn your ABC’s Falool, and return to the battlefield.” 70 Most of the information in the preceding 5 paragraphs is drawn from Authors interview, Jebha Intelligence officer, May 19 and Sept 11, 2009 and Saleh AA Younis Interview with H. Tedla Bairu, Jan 1, 2001, www.awate.com. See also (Markakis 1987) and (Medhanie 1986) to make serious gains in 1975, the pro-Baathist elements in the leadership saw little incentive to maintain the cooperative status quo. At the same time, Herui was becoming increasingly aggressive on the issue of unity with Shaebia and took steps (including holding private meetings with Shaebia leaders)71 that many hardliners could not countenance. For others who were dragged into the dispute, the battlefield victories seemed to generate brinkmanship and provocation. One wonders whether, if it had been 1971 in the midst of military stalemate, Falool ideologues would have rose up in open insubordination against their superiors. Nearly 25 years later, Herui himself summed up the dynamic best, commenting that “The first error was made, I believe, when we did not assess our situation properly. We were slightly blinded; we were victims of our own success.” 72 As the war went, so went cooperation. Alternative Arguments Of course, it is difficult to ascertain whether the arguments made in this paper are valid without carefully weighing alternative hypotheses. In this section, I evaluate several competing explanations that could plausibly account for variation on the dependent variable of interest, fragmentation. In addition to the two moments of fragmentation discussed in this paper, Jebha also fragments in 1982 after a bitter military confrontation with Shaebia— a case of fragmentation I include in this discussion for analytical leverage, and that led to Jebha’s final expulsion from the Eritrean theatre. It should also be noted that while some of the alternative variables discussed may affect fragmentation by shaping the possibilities of gains and losses, taken alone they cannot explain the variation on the dependent variable we observe. 71 72 Authors interview, Jebha intelligence officer, May 19 and Sept 11, 2009 Saleh AA Younis Interview with H. Tedla Bairu, Jan 1, 2001, www.awate.com State Repression and Counter-Insurgency Strategies A number of researchers have argued that state repression and counter-insurgency operations impact the internal cohesion of rebel organizations, although there is a difference of opinion regarding the exact nature of this causal relationship. Shapiro (2008) argues that high levels of state repression force insurgents to decentralize their operations, and thus reduce a leader’s control over subordinates. More intuitively, it also seems as though increases in state repression increase the extent to which potential dissident factions believe that the rebellion cannot succeed, and therefore encourage fragmentation. If these theories are correct, we would expect increases in the coercive resources that the Ethiopian government deployed against Jebha to be correlated with their fragmentation. Using standard, but very blunt measure of state repression – measures of numerical preponderance – we can see that significant increases in the coercive capacity of the Ethiopian state do not explain the three incidences of cooperation breakdown cited in this paper (Figure 9 &10). While troop levels and military expenditures dramatically increased in 1977-78 – with the first real effect of these increases being felt in June of 1978 during the Derg’s First Offensive – these increases cannot explain much of the variation on the dependent variable. Jebha’s fragmentation in 1967-68, and the emergence of EDM/Fallool during 1975-1977, largely predated the huge increases in the coercive resources of the Ethiopian state in 1977-78. Furthermore, Jebha’s disintegration into three groups in 1982 occurred a full 4 years after this increase. Figure 9: Size of Ethiopian Armed Forces 1960-1991 (in tens of thousands) 350 300 250 200 150 100 50 0 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 Size of Ethiopian Armed Forces (In Tens of Thousands) Interestingly, many have argued that battlefield outcomes and state power have less to do with numerical preponderance and more to do with military technology.73 Therefore major increases in the technology employed by the Ethiopian military, rather than numerical preponderance, should roughly correlate with Jebha’s fragmentation. Figure 10: Ethiopian Military Expenditures in USD Dollars (in hundreds of millions) 1400000 1200000 1000000 800000 Ethiopian Military Expenditures (In Tens of Millions) 600000 400000 200000 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 0 73 (Biddle 2004, 16-17) collectively calls these explanations “Dyadic Technology Theory” Although a full discussion of the equipment employed by the Ethiopian military is beyond the scope of this paper, a good proxy for variation in technology employed by the Ethiopian state over time would seem to be arms transfers from the major producers of military technology during the Cold War, the US and the Soviets.74 Figure 11 charts this variation. As one can see, the huge spike in Soviet arms transfers to Ethiopia in the late 1970’s explains only one episode of Jebha’s fragmentation, which is 1982. Figure 11: US and Soviet Arms Transfers to Ethiopia 1960-1991 (in millions) 1200 1000 800 600 400 US Arms Transfers Soviet Arms Transfers 200 0 However, there are strong reasons to think we should take a closer look at state repression. Biddle (Biddle 2004) persuasively argues that numerical preponderance and technology are poor predictors of war outcomes because they tell us nothing about how effectively coercive resources are used. Furthermore, it seems self-evident that numerical preponderance is endogenous to the performance of the government’s counter-insurgency operations: if the state has failed to curb the activities of the rebels, or the rebels are surging, state authorities are forced to ramp up their coercive capacities. 74 See (Westad 2007) for what is by far the most well sourced account of superpower involvement in Ethiopia Lastly, the Ethiopian state has historically had many internal and external opponents, so increases in numerical preponderance may be an artifact of developments unrelated to the insurgency in Eritrea. A better, albeit imperfect measure of state repression would be to look at major Ethiopian government offensives against Jebha between 1960-1982, and examine the extent to which fragmentation occurred during, or directly following, such offensives. Of course, not all offensives achieve a degree of success, but if theories about state repression are true, fragmentation should be preceded by a major government offensive that has achieved some success. My research suggests that there were roughly four major offensives against Jebha in Eritrea, with the first occurring in 1967, and the last occurring in 1979. Figure 12 (appendix) lists these offensives, provides details on each, roughly approximates their outcomes, and lists whether they were followed by fragmentation. One can see from the table that the occurrence of major government offensives does not explain much— particularly the events of 1975-77. Of course, successful counter-insurgency efforts are not solely comprised of coercion. Some governments have successfully split rebel organizations by offering political or economic incentives that persuade particular segments to defect or provide an issue of contention within rebel ranks. Such arguments are closely related to theories about “spoilers” and peace negotiations that split rebel organizations between those that seek to cooperate with the government, and those who want to continue war (Stedman 1997). Yet in the Eritrean context, these sorts of explanations are not convincing. First, the Ethiopian government had virtually no official contacts of any consequence with Jebha until the monarchy was overthrown in 1974. Until that point, the Ethiopian government largely communicated with Jebha indirectly through community elders who would relay messages between the two sides.75 In fact, until the mid-1970’s, the Ethiopian government largely denied there was even an insurgency in the province of Eritrea. Therefore, it does not seem likely that serious government efforts to split the rebels could have been made before 1974, and thus such efforts could not explain the splits of 1967-68. Even after the precipitous demise of the monarchy in 1974, and despite the rhetoric of the new Marxist military government in Addis Ababa, few actual efforts were made to encourage defection from Jebha through the provision of political or economic incentives. Figure 13, which is data from Minorities at Risk’s Discrimination data-set, shows there was little variation in the degree to which the political institutions of the Ethiopian state discriminated against Eritreans (see appendix for explanation of criteria). However, an honest history should note that the Ethiopian government periodically offered amnesty to enemy combatants.76 Yet in practice, these efforts were open-ended instead of time bound, and as such, cannot explain temporal variation in fragmentation. Furthermore, it is doubtful whether amnesty, in the absence of extra political or economic benefits, would be sufficient to trigger defection amongst those who have taken such high-risks to obtain political or economic benefits through war. 75 For instance, in cases where the government and Jebha were seeking to negotiate the release of foreigners whom Jebha had detained 76 Airgram, Asmara to State, May 4, 1970, RG 59, Central Files, “ELF Defections”. Also, Interview with Jebha fighter, May 3, 2009- Described incessant Ethiopian radio broadcasts and loudspeakers across the front lines designed to encourage defection. Not surprisingly, the tactic was ineffective Figure 13: Ethiopian Political and Economic Discrimination against Eritreans 19601991(Minorities at Risk) 4 3 2 Political Discrimination Economic Discrimination 1 1990 1988 1986 1984 1982 1980 1978 1976 1974 1972 1970 1968 1966 1964 1962 1960 0 Another point to consider, and perhaps the most compelling rebuttal against arguments about state efforts to politically co-opt the rebels, is that if these sorts of theories are true, we would expect all splits of rebel organizations to result in defection to the government. However, we know from the historical analyses that this is not the case. In fact, of the eleven identified groups that that split from Jebha, only one group defected to the government. External Support and Control Another explanation for the occurrence of factional infighting and fragmentation within rebel organizations is the presence of a strong external backer who can enforce cooperation and provide leaders with the material tools necessary to forge (and sometimes coerce) organizational unity. If such a theory were true, we would expect shocks in external support to roughly correlate with the fragmentation of Jebha. Yet the reality is that this is a weak explanation in the Eritrean context. While it is difficult to accurately assess the degree to which states materially supported Jebha, a brief historical commentary is illuminating. There were a broad collection of state and non-state actors that provided material support to Jebha over its 22 year armed confrontation with the Ethiopian government.77 The historiography of the war in Eritrea often understates the degree of external support that the rebels received, yet given the extent to which the revolution became an extension of Arab-Israeli conflicts, intra-Arab rivalries, and Cold War politics, the external dimension is hard to ignore. Early support for Jebha came from Egypt, which allowed radio broadcasts and the leadership to organize openly in Cairo. By 1963, the Egyptians, who for political reasons opposed the dismemberment of Ethiopia, withdrew their support, only to be replaced by the recently transformed radical Baathist states of Iraq and Syria (Erlich 1983).78 Throughout the 1960’s, the Iraqis and Syrians were Jebha’s most committed supporters, channeling much of their financial and military aid to the field through Jebha’s external leadership and providing Jebha members with valuable training in their military academies. During the 1960’s, Jebha radio broadcasts from Damascus and Bagdad railing against imperial Ethiopia’s oppression of its Eritrean minority were the primary mediums through which the Eritrean struggle was publicized on the international stage. For the Arab Baathist states, the war in Eritrea pitted a Muslim minority against a Christian dominated empire, and thus on grounds of religious affinity, was deserving of Arab support. If this were not reason enough to support Jebha, the Arabs had the strong suspicion – 77 See INR Research Study, Alison Palmer, Feb 22, 1971, RG 59, Central Files “Ethiopia- Foreign Aid to the Eritrean Liberation Front”- report implicates nearly 16 countries in supporting ELF at one point in time or another 78 Because the official position of the Organization of African Unity (OAU) was that Africa’s colonial borders were sacrosanct, Egypt feared alienating its African partners by supporting Eritrean nationalists. Furthermore, because of concerns over the Nile, Egypt did not want to encourage South Sudan’s secession by legitimizing the practice which was correct – that the Ethiopians had a secret, albeit very close security partnership with the Israeli’s (Lefebvre, Middle East Conflicts and Middle Level Power Intevention in the Horn of Africa 1996) . For its part, neighboring Sudan, by virtue of ethnic kinship with Eritrea’s western lowlanders particularly the Beja and Beni Amir tribes that occupy eastern Sudan and much of adjoining parts of Eritrea – border issues, and pressure from its Arab neighbors, also supported the cause of Eritrean nationalism, allowing Jebha to establish its executive in Khartoum and a field command in Kassala. Despite the fact that Ethiopia devoted substantial time and energy to stopping the flow of personnel and arms from radical Arab states into Eritrea, such efforts were largely unsuccessful. Recognizing that Eritrea’s 600 mile Red Sea coast was simply too long to effectively police, Jebha’s Arab supporters would drop shipments of arms along the coast – with the friendly regime of South Yemen as the conduit – where they would be picked up by Jebha militants who were waiting at predetermined locations. On many occasions, Jebha would use local fishermen to ferry weapons to the coast and successfully avoid the interdiction efforts of the Imperial Ethiopian Navy. 79 This was by no means the only way that men and material were moved by the rebels into Eritrea in the 1960’s and early 1970’s. Overland routes through Kassala were often utilized, as were direct overland routes from Port Sudan, where much of the arms and munitions from the Arab world destined for the rebels arrived80. Furthermore, at the primary Eritrean port of Massawa, Jebha leader Osman Saleh Sabbe had constructed a clandestine network of customs officials who allowed crucial supplies to slip through to the rebels (Pool 2001). 79 Airgram, Addis to State, June 14, 1971, RG 59, Central files, “Dhow Activity in Eritrean Waters” Authors interviews, Jebha intelligence officer, May 19 and Sept 11, 2009 and Shaebia intelligence officer, August 11, 2009 80