Revising the Territorial Structure of the Ethiopian Federal System: A Good Idea?

Introduction

Yonatan Fessha, a leading scholar on comparative federalism in Africa, maintains that “the original sin” that the framers of the Ethiopian Constitution committed, when designing the country’s federal system, was that they took ethnicity as the sole factor for demarcating the subnational unit of the federation.¹ He argues that they had another option: By breaking the large states, such as Oromia, Amhara, and Somali, they could have created smaller yet ethnically homogenous states. If that was the case, not only ethnic but also “historically and politically relevant” subethnic identities, such as provincial identities, would have been territorially accommodated. Moreover, the prospect of essentializing ethnic identity as the only relevant factor of political mobilization would have been avoided since such a territorial design would have led to intra-ethnic political competition. This, in turn, would have led to the federation experiencing “less strain as intra-ethnic divisions … are less emotionally charged and, as a result, more manageable than inter-ethnic division."² Yonatan’s argument seems to have found traction among some political parties and actors. There is now a clamor for revisiting the federal design, the focus being on re-demarcating the boundaries of the subnational units. Many are calling for the division of the states I mentioned earlier into smaller units wherein the relevant communities in these regions retain their majority status. This begs the question whether re-demarcating the boundaries of the states is doable and timely. In this paper I argue that, if successfully implemented, the re-demarcation of the boundaries of the subnational units could be self-sustaining. However, it is extremely dangerous and may lead to another round of civil war that the federal republic might not survive. Any attempt to alter the boundaries of the states should be undertaken carefully, only after a consideration of the danger that such a venture poses.

The paper begins with a brief discussion of the political evolution that led to the establishment of ethnicity-based states followed by a discussion of why the principle that requires the formation of one state for an ethnic community found its way into the federal Constitution. The paper then turns to discuss the danger of attempting to redraw state boundaries by dividing some of the existing states. It finally raises the issue of whether redrawing the state boundaries is an impossible mission.

1. The Root of the Ethnic States

The idea of organizing the subnational units of the country along ethnic lines and providing territorial autonomy to ethnic communities was first debated during the time of the Ethiopian Student Movement (ESM). Even Wallelign Mekonnen in his famous essay on “the question of nationalities” implied that there were defined or definable territorial areas that each ethnic community inhabited and within which each community could enjoy political and cultural autonomy.³ There were indeed disagreements within the ESM on how the “question of nationalities” should be handled. There was a general agreement in the movement that, at a minimum, territorial autonomy in the form of regional autonomy could be used for dealing with the ethnic question even though political groups, such as the Ethiopian People’s Revolutionary Party (EPRP), went as far as recognizing the right of ethnic communities to secede from the country.⁴

In any case, the Derg took the first practical step of demarcating the administrative boundaries of the country with the explicit purpose of responding to the ethnic issue. In the 1980s, it established the Institute of Nationalities with mandates that included the studying the ethnic composition of the country and the development of a proposal on how the internal boundaries of the country could be redesigned to respond to the ethnic question.⁵ Based on the proposal of the Institute, the Derg attempted to reorganize the provincial boundaries.⁶ It created 25 administrative regions and a few ethnically defined so-called “autonomous regions”—such as Tigray, Assab, and Ogaden, which were supposed to enjoy extra autonomy compared to the other provinces.⁷ The Derg did this with the intent to appease the ethnonationalist and secessionist insurgencies, such as the Tigray People’s Liberation Front (TPLF) and Eritrean People’s Liberation Movement (EPLF), which were at the time succeeding in their military campaign against the center.⁸ It was not, however, successful in this regard, because the creation of administrative units along ethnic lines was not accompanied by the devolution of real power to the newly created ethnicity-based provinces. The autonomous provinces had identical powers and competencies to the ordinary provinces.⁹

The most decisive measure toward restructuring the country’s subnational units along ethnic lines was taken after 1991 during the Transitional Period (1991-1995). The Transitional Period Charter (TPC) of 1991 laid the groundwork for ethnic federalism, which principally aimed at responding to the ethnic question. Not only did the TPC recognize the right to self-determination for “each nation, nationality and people” of Ethiopia, but it also made this right the “governing political principle” of the country.¹⁰ This principle was to find expression in the right of each ethnic community to administer its own affairs within the territory it inhabited, entrenching the logic of ethnicity-based states.¹¹ The TPC further provided that the boundaries of regional and sub-regional units would be demarcated based on the geographical settlement patterns of the ethnic communities of the country.¹² The Representative Council (the legislative organ of the Transitional Government of Ethiopia or TGE) also issued Proclamation No 7 (1992) which, mainly based on the studies of the Institute of Nationalities, identified 63 territorially concentrated ethnic communities, 47 of which could automatically establish their own governments, starting from the woreda level.¹³ Hence, woredas, most of whose residents belonged to a specific ethnic community, were brought together to form a subregional or regional self-governing area of the ethnic community.¹⁴ In this manner, 13 regions were established based on ethnic criteria. Addis Ababa, the 14^th^ region, was the exception to this logic. Under Proclamation 7 (1992) smaller ethnic communities could by agreement join hands to create a larger unit.¹⁵ Accordingly, five of the 13 ethnic regions “opted” to unite and create the region now known as the Southern Nations, Nationalities, and Peoples' Region. The other 17 ethnic communities, which were found to be too small to exercise self-governance even at the woreda level, were guaranteed fair representation in woreda councils.

Nothing major was introduced in terms of re-structuring the sub-national boundaries of Ethiopia after the adoption of the 1995 Constitution. The Constitution simply entrenched the ethnic regions which were created during the Transitional Period. There are three important points that need to be stressed here. First, the Constitution does not contain a list of ethnic communities of the country. It simply provides a broad definition of what constitutes an ethnic community.¹⁶ This implies that Ethiopia does not have a definite number of ethnic communities. The Constitution keeps the door open, as with the Silte, for new ethnic communities to be recognized. Several groups of people have also petitioned the House of Federations (HoF), including the Wollene, the Kebena, the Qimant, the Mareqo, and the like, demanding to be recognized as distinct ethnic communities even after the formation of the federation. Second, the Constitution initially recognized nine states, implying that territorial autonomy was to be exercised not only through the formation of a state but also with the establishment of self-government at kebele, woreda, and zone level.¹⁷ In any case, a community that seeks to establish its own state could do so by seceding from the state within which it is found. There are no economic, demographic, or other criteria that it needs to fulfil before demanding for state of its own, so long it is recognized as a distinct nation, nationality and people (NNP). Two additional states (the Sidama and the South-West) were recently added to the list of states. Based on this logic, Ethiopia may have as many states as the number of ethnic communities that are and could be recognized.

Finally, the right to self-determination and, therefore, territorial autonomy belongs to an ethnic community in its entirety, not to a section of it. Implicit in this is that an ethnic community can exercise its right to self-determination within a single territorial unit.

2. Why the One-Ethnic-Community-One-State Formula?

What were the reasons for adopting the one-ethnic-community-one-state logic? The argument in favor of recognizing the right of each ethnic community to establish its own state (at least in principle) was based on the notion of the equality of ethnic communities, regardless of the differences among them in population, size, territorial area they occupy, and economic status.¹⁸ Taking the latter factors as conditions for the formation of a state would have gone against this notion.

Besides, during the Peace and Democracy Conference of June 1991, some of the ethnicity-based political parties made clear that the division of the ethnic communities, whose cause they supposedly champion, into different administrative units was completely unacceptable. The Oromo Liberation Front (OLF) was especially explicit that it would not accept the division of the Oromo people into different subnational units, equating this to the divide-and-rule policies of previous regimes.

The formation of a single state for each ethnic community was also underpinned by the notion of vanguardism that the TPLF (Ethiopian People’s Revolutionary Democratic Front—EPRDF) espoused. Vanguardism is a Marxist-Leninist notion that is often linked to the exclusive right to exist of a party or a similar organization which is established by the most “class conscious” members of the working class to lead the latter in its “revolutionary struggle."¹⁹ Political contestations based on the notion of vanguardism in Ethiopia can be traced back to the 1960s when Marxism-Leninism was the dominant ideology in the ESM.²⁰ Some parties that sprang from this movement, including EPRP and All-Ethiopia Socialist Movement (AESM), better known by its native Amharic acronym መኢሶን (Me’ison), saw themselves as a national vanguard party and sought to eliminate their opponents. The Workers' Party of Ethiopia (WPE) was later constitutionally recognized as the only party with the right to exist in the country. The ethnicity-based rebel groups, especially the TPLF, espoused Marxist-Leninist political theories to fit their own purposes, and declared themselves vanguard parties of the ethnic groups they claimed to represent.²¹ Within the context of the country’s ethnic federal system, EPRDF’s vanguardist view was that every ethnic community should have a single vanguard party and that the ethnic community needs no other party to champion its cause.²² At the regional level, therefore, each of EPRDF’s constituent and ally parties was expected to play the role of a vanguard party with respect to the relevant ethnic community. For each ethnic party to effectively play the role of a vanguard party, the ethnic community of which the party is supposed to be a vanguard must be within a single territorial unit.

3. The Dangers of Breaking up the Regional States

Much has been said already about the dangers of creating subnational units based on the one-ethnic-community-one-state logic, some of which turned out to be true.²³ My focus rather is on the danger of the undoing of the already entrenched ethnicity-based subnational boundaries. Even Yonatan, who argues that the issue of remapping the boundaries of the subnational units should be on the table for political negotiation, cautions that such an endeavor may have various perilous consequences that those who enthusiastically espouse his arguments choose to ignore. I maintain there are five major reasons why the remapping of the internal boundaries of the country can be risky.

Indeed, in the early 1990s, the state boundaries could have been demarcated in such a way that an ethnic community becomes the majority in multiple states despite some challenges from certain political forces which at the time could have been overcome. Back then, the current states were non-existent and their maps and flags were largely unknown. However, after three decades of ethnic federalism, there are many now who identify themselves with the states. Members of the ethnic communities now know the maps and flags of the states and consider the latter as their own. Students are taught about the states' history and geography in schools. To undo the states now and create new ones is less likely to settle well among the ordinary members of the relevant ethnic community, let alone the political actors.

There is what seems to be an almost irreconcilable difference in the institutional choices of different political groups for responding to the ethnic issue. Some are for an institutional design that is geared towards integration, while others are in favor of the accommodationist approach.²⁴ Those in the first group support Yonatan’s proposal while those in the latter group oppose it. The remapping of the internal boundaries of the federation cannot be undertaken without angering those who are in favor of the accommodationist approach of managing the ethnic diversity of the Ethiopian people.

There are many who have political and economic stake in the current arrangement. There are political appointees, civil servants, and businessmen who have economic interests and would resist any attempt to divide the existing states—especially those which are under consideration.  

The political ground is not fertile for the implementation of the policy under consideration. Major political forces in Oromia (OLF and Oromo Federalist Congress-OFC, and even the Prosperity Party-Oromia wing) are not even willing to entertain the idea of breaking Oromia into smaller states. I am not aware of a single political party or political group, with the prefix “Oromo” or “Somali,” that supports the idea of breaking these states into smaller units. Given this reality, any suggestion to remap the internal boundaries of the country is likely to face immense resistance and eventually lead to violence  

It is almost taken for granted that the restructuring of the federal system along the lines discussed above will be enthusiastically supported by the Amhara people, political parties, and elites. I think this ignores the fact that there is a growing Amhara nationalism. Besides, there is a realization among many political actors of the Amhara that there is a need to maintain the state as is, if not territorially expand it, to overcome the political ascendency of the Oromo. Given these facts, Amhara’s support for the redrawing of the internal boundaries of the regions that results in the division of the Amhara state into its historical regions can no longer be taken for granted. I think the Amhara will not agree, at least, with their state being the first where the practicability of this policy is tested.

4. Remapping the Internal Boundaries: Mission Impossible?

There are two ways that remapping the internal boundaries of the states could be effected: through a political compromise that would eventually lead to a constitutional revision or through a top-down imposition of the project.

I am not aware of any case where an ethnically organized subregional unit was broken into smaller units as a result of political compromise. In countries such as Nigeria and Kenya (when Majimboism was abolished) ethnicity-based subnational units were divided into smaller units by the decisions of the central governments.²⁵ The Nigerian states were divided several times following military coups. In Ethiopia, considering the lack of a culture of political compromise, the ideological rigidity of the political actors in the country, and the lack of trust among them, I cannot imagine a political compromise resulting in the remapping of the internal boundaries of the country. A top-down decision to break up the larger regions, if it can withstand the initial onslaught of resistance, protest, and even violence, could indeed be self-sustaining for the same reasons that the current arrangement has been self-sustaining: new identities will be formed and a new group of people with an economic and political stake in the new arrangement will emerge to defend it. This has been the case everywhere subnational boundaries are altered by the decision of those holding power at the national level. However, this is too risky. Given the Ethiopian historical and political context, any attempt to do so may be catastrophic.  

Discussions

Dr. Christophe Van de Beken — Discussant

One important strength of Zemelak’s paper is the emphasis on the historical and political background of the ethno-territorial federal system that is now in place, which Yonatan Tesfaye rightly referred to as “the original sin.” Of course, the issue with the Constitution also has to do with the process of its adoption, which was not sufficiently inclusive, as it was dominated by one specific political party, which was a coalition of ethnicity-based parties. This also explains the institutional choices present in the Constitution. The lack of inclusiveness, and to some extent the flawed process of constitutional drafting and adoption, has from the very beginning affected the legitimacy of the Constitution. From the very beginning it has been contested, and this has contributed to the lack of constitutionalism we have observed. Therefore, when we now discuss a potential amendment of this Constitution, we should not repeat the same mistake. The process is very important; if we want to avoid contestations, we need to take due care and follow a proper process of constitutional amendment that is participatory—not only through lip service to participation, but effective participation as well as that is broadly inclusive. Having a proper constitutional amendment process would lead to a more legitimate constitution and would contribute to a stronger constitutionalism in the country. If and when the amendment takes place, it should not be a rushed process.

Additionally, I agree that restructuring the states along territorial lines is important. But even if you do that, the issue of accommodation and integration would still be there; the issues of ethnic pluralism, ethnic minorities, ethnic nationalism at the regional level would still be there. You may weaken ethnic nationalism to some extent, in a way that it may no longer threaten the existence of the state, but the issue of diversity and the need for accommodation will be there. So, I am not sure that the state restructuring would make much difference.

My last point is on the issue of participation; I agree with the Zemelak that it would be difficult to do restructuring of the subnational units of the Ethiopian Federation. On the other hand, we should not take that for granted; there may be an attachment to these states as this arrangement has existed for nearly 30 years. We cannot be sure, however; so we should also have some kind of public opinion poll or survey, because at this point we are only guessing that there is an attachment while we do not really know if that is the case. The constitutional amendment process should not be purely elitist.

Dr. Zelalem M. Teferra

You noted that it is possible to restructure the current federal arrangement, but it could be a risky endeavor. Why do you think that it is risky to introduce some nuances to the current system? We should also consider other options like multilingualism or multiculturalism, which I believe would be even more useful than restructuring the federal arrangement.

Dr. Sisay Alamahu

Zemelak, you proposed two possible formal solutions regarding the restructuring of regions: 1) political compromise and 2) a top-down approach. Is it possible to consider something in between the two approaches? For example, can we start with a practice whereby power is more devolved to subnational units?

Dr. Mohammed Dejen

Revising the territorial structure of the Ethiopian federal system may be a good idea, but the issue is how we should go about it. I think any restructuring exercise should not be imposed; the experiences of other countries may not be applicable to the situation in our country. For example, the restructuring in Nigeria took place during time of coup d’état. It might have been easier for EPRDF to attempt restructuring, but not as easy for Prosperity Party (PP). This is because PP is accused of unitarist tendencies, so it is making an attempt at restructuring will be suspicious and contestable. Therefore, the best way to go about it is through discussions and dialogue, and ultimately the people should be allowed to choose so that we can avoid making the same mistake that was made by TPLF. There is a general perception that the Amhara may support the restructuring of the states, but this may also not be the case since now there is growing Amhara nationalism.

Dr. Abadir M. Ibrahim

I think it needs to be clearly noted that we must be wary of the imposition of constitutional restructuring by any government not least because it is a non-democratic one. Constitution-making should follow a big-tent participatory approach. Making the process inclusive is important on its own terms, independent of whatever the outcomes of constitutional reform are. It is also important, of course, to make sure a constitutional reform process does not end in creating socially and politically significant “losers” thereby leaving behind the seeds of future instability. Anyone who ignores this point should know they are playing with fire; this lesson should have been learned from the recent past. Unfortunately, it is not just the regime, it is all of us who will face the consequences of taking constitution-making lightly. The negative impacts of taking the constitution-making process lightly will come to bear on the legitimacy of the constitution and eventually the stability of the state—it is a question of sooner or later, and not if. I wanted to raise this point because I have been hearing that the government is planning to go ahead with constitutional reform on its own by sidelining its opponents as decried elites and going to the public in a “direct democracy” type of approach. We have all become familiar with this type of discourse and practice over the last decades. Anyone who may be asked to advise this process should underline this point—you should get the process right—before talking about the merits of specific reform ideas.

Notes