Part Two: The Case for Democratic Union
By Worku Aberra
I. The Question the Statement Leaves Open
Part One of this commentary found that the statement issued by eight Oromo intellectuals on 17 October 2026 under the name of the Council of Oromo Scholars marks an advance toward cooperation for a common political future, but leaves the constitutional options open. The authors call for a process by which “the peoples of Ethiopia can freely determine their political future, without coercion or the imposition of predetermined outcomes” and name “the right to self-determination” among the aims of an inclusive transition (Council of Oromo Scholars 2026, 2). Their commitment is to free consent; it does not settle whether Ethiopia should remain united.
I support democratic union. Part Two distinguishes that preference from my objection to a premature decision on secession. I favour self-government within a united and democratic Ethiopia because I believe it offers greater social, economic, and political advantages than separation. That is my longer-term position. The absence of conditions for free and informed consent, together with present armed rivalry, provides an additional reason to oppose a premature decision.
My argument proceeds from the necessary conditions for a legitimate decision on secession to the merits of the alternatives. It first sets out safeguards for a referendum and explains why an interval of democratic self-government would give citizens a firmer basis for their choice. It then compares union and separation by three criteria: the credibility of their commitments, their economic costs, and their risks of armed conflict. In a referendum held after the safeguards and the democratic interval are in place, I would campaign for union and accept a legitimate decision to separate.
II. Conditions for a Legitimate Decision
A legitimate decision on secession requires safeguards that Ethiopia does not yet have:
- an independent electoral commission to administer the referendum;
- eligibility rules set beforehand and applied impartially by that commission;
- freedom to advocate union or separation without intimidation;
- effective participation by displaced people;
- a threshold for a binding result, fixed before the vote;
- access to competing arguments and adequate time for public deliberation.
Eritrea’s referendum of 1993 shows why the first safeguard matters. The provisional government established the referendum framework and appointed the commissioners, a concentration of authority that warrants scrutiny of the safeguards for impartial administration. United Nations observers nevertheless described the commission as independent, found registration impartial, and judged the referendum free and fair at every stage. No organization registered to oppose independence, and the tenor of the campaign left little ground for one; neither fact by itself establishes that opposition was suppressed (United Nations 1993; Aberra 2026). A favourable verdict from observers confirms the conduct of a vote; the independence of the authority that organizes it, and the freedom to argue the other side, require safeguards of their own.
The fifth safeguard, a threshold fixed before the vote, sets the measure of consent. Article 39 requires a two-thirds majority in the relevant legislative council to initiate a demand for secession, followed by majority support in a referendum (Ethiopia 1995, art. 39). In my view, the referendum threshold provides too narrow a basis for dividing a country. Secession ranks among the gravest decisions a political community can make, and comparative experience offers several approaches to the consent it requires. Canada’s Clarity Act of 2000 requires a “clear majority” before the federal government can enter negotiations on secession. The House of Commons must weigh the size of the majority and the share of eligible voters who participate; the Act does not fix a numerical threshold (Canada 2000).
Other referendum frameworks specify numerical thresholds. Montenegro’s referendum law required 55 per cent of valid votes in favour and turnout of at least half the registered electorate, under terms negotiated with European Union involvement (OSCE/ODIHR 2006). The constitution of Saint Kitts and Nevis requires a two-thirds majority of valid referendum votes in Nevis; although a majority voted for independence in 1998, support fell short of the required threshold (Saint Kitts and Nevis 1983, s. 113; IFES n.d.). Scotland’s referendum legislation provided for a simple majority (Scottish Independence Referendum Act 2013). Catalonia shows the cost of a vote without agreed rules. Its government held a referendum on 1 October 2017 after Spain’s Constitutional Court had suspended the law that authorized it; opponents of secession largely boycotted the vote, and the Catalan government reported 90 per cent in favour on a turnout of 43 per cent (Hernández and Closa 2022; Reuters 2017). The declaration of independence that followed brought direct rule from Madrid. A vote held without legal authority or agreed rules settles nothing, whatever its majority.
For Ethiopia, I propose a qualified majority overall and, as the final part of this section argues, an agreed procedure for contested territories. A decision with consequences so difficult to reverse warrants broader consent than an ordinary electoral choice.
The safeguards protect the vote; a democratic interval would inform it. The interval would give citizens experience of the union whose future they must decide and allow them to judge democratic self-government within Ethiopia by its results. That experience would provide a firmer basis for comparison, though it would carry no obligation to favour union.
Where the safeguards govern the vote, five indicators would judge the interval: elections allow a real possibility of a peaceful transfer of power; opposition parties and the press operate without intimidation; courts and the electoral commission have ruled against the government and seen their rulings obeyed; access to land, jobs, and public services no longer depends on political loyalty; and an impartial civil service has put into effect the freedoms and autonomy promised in the transition.
These indicators are my adaptation of the five arenas Linz and Stepan (1996) identify as supports for democratic consolidation, one indicator for each arena: political society, civil society, the rule of law, economic society, and a usable state bureaucracy. The specific tests are mine; their concept of economic society extends beyond freedom from political discrimination.
Each indicator also bears on the freedom of the eventual vote. A government that controls land and employment can reward or punish voters; in Ethiopia, the Constitution vests ownership of land “exclusively” in “the State and in the peoples of Ethiopia” (Ethiopia 1995, art. 40). Election officials loyal to one side can compile the electoral roll and count the ballots in their favour. An impartial civil service would help protect the vote and demonstrate whether the government had honoured the commitments to self-government made during the transition.
An independent body would assess whether these conditions had been met against criteria agreed in advance. Because Article 39 grants “an unconditional right to self-determination, including the right to secession” and requires a referendum within three years of the federal government’s receipt of the demand, both the higher threshold and the proposed democratic interval would require constitutional amendment (Ethiopia 1995, arts. 39, 105). Amendment under Article 105(1) requires approval by a majority in every state council, not only that of the region seeking secession, and by a two-thirds majority in each federal house separately.
Nationalist critics may object that these conditions hand the centre a veto, since a government that refuses to democratize could postpone self-determination indefinitely. That objection has force. An agreed backstop would limit the delay: after a specified period, incomplete democratic consolidation would no longer suffice to defer a referendum. The independent body would still have to establish that the basic safeguards for a free vote were in place. Citizens should not have to wait indefinitely for a fully consolidated democracy, but a deadline cannot make consent free where intimidation, exclusion, or partisan administration determines the result.
Territorial disputes complicate both the conduct of a referendum and acceptance of its outcome. Addis Ababa is a federal city surrounded by Oromia, its status contested by Oromo nationalists. Disputes also affect the boundaries of Oromia, and clashes along them have uprooted large numbers: cumulative displacement associated with conflict along the Oromia–Somali regional boundary had reached approximately one million people by early 2018 (AFP 2018). Elsewhere, as in Welkait, jurisdiction itself is contested.
A credible procedure for contested territories must ensure participation and establish rules that the parties accept. Displaced residents may be excluded from the electoral roll, minorities on the losing side may need protection, and either party may reject the authority that organizes the vote. The procedure must therefore establish, before the referendum, who may vote, what majority decides the outcome, and how the rights of those on the losing side will be protected.
III. Self-Government or Secession
Self-government and secession both rest on commitments, and each endures only where institutions make those commitments credible (Fearon 1995; Powell 2006). Self-government within Ethiopia requires constitutional guarantees, fiscal arrangements, protection against unilateral central intervention, and safeguards for minorities within regions. A democratic union would provide procedures for citizens and their representatives to challenge breaches and revise these arrangements. Such procedures offer a means of redress, although their effectiveness depends on whether public authorities obey them.
Secession would place relations between the successor states under agreements across sovereign jurisdictions. Those states would have to settle borders, citizenship, assets, debt, and access to shared infrastructure, while each would be responsible for the protection of minorities under its domestic institutions and international commitments. These arrangements could also be revised, but changes to relations between the states would require their agreement. The advantage I claim for democratic union lies in the common institutions available to address disputes; neither union nor separation guarantees that commitments will be honoured.
Separation can also carry economic costs. A smaller state may forgo economies of scale in public goods—defence, infrastructure, and administration—and spread their costs over fewer taxpayers (Alesina and Spolaore 1997). After Czechoslovakia dissolved, trade between its successor states fell sharply relative to the level predicted by a gravity model: from approximately 43 times that level around dissolution to seven times by 1998 (Fidrmuc and Fidrmuc 2003).
Disruption of established economic ties can contribute to a fall in output, although its effects are difficult to separate from those of other changes. The successor states of the Soviet Union and Yugoslavia suffered severe recessions in the 1990s. The end of central planning accounts for much of the decline, but disruption of supply chains compounded the difficulties; output fell further in the former Soviet Union than in central Europe (Blanchard and Kremer 1997; Campos and Coricelli 2002). South Sudan illustrates the vulnerability of dependence on infrastructure beyond a new border. Its oil exports halted by a dispute with Sudan over pipeline fees, the country saw real GDP contract by nearly half in 2012 (World Bank 2017).
Separation need not produce economic decline. In January 2008, the International Monetary Fund (IMF) estimated growth in Montenegro at 7.5 per cent for 2007, although it also identified signs of overheating and financial vulnerability (IMF 2008). The case cautions against a general prediction of decline; the costs of separation depend on the economic ties at stake and the terms on which the parties divide. For Ethiopia, the extent of those ties and the prospect of contested separation warrant particular attention.
The geography of Oromia makes economic ties across regional boundaries especially important. The region surrounds Addis Ababa and borders every other regional state except Tigray. Many of its towns lie closer to centres in neighbouring regions than to one another. Shashemene lies some 25 kilometres from Hawassa, in Sidama, and several hundred from Nekemte, in Wollega; its residents look to Hawassa for employment and services. A border around Oromia would place such ties under the jurisdiction of separate states, and their continuity would depend on agreements over trade, transit, and access to infrastructure. The size of the population would preserve substantial advantages of scale, but those advantages would also depend on the strength of internal connections and continued access to neighbouring markets.
Peace after separation is not assured. Czechoslovakia in 1993 and Montenegro in 2006 separated peacefully, without rival armed movements in active contention for control of the successor states. Where such movements survive independence, the new state can inherit their rivalry. South Sudan won independence in 2011 after decades of war against the Sudanese government; civil war broke out within the new state two years later. By one estimate, the crisis caused approximately 383,000 excess deaths between December 2013 and April 2018, roughly half of them from violence (Checchi et al. 2018).
Separation can leave disputes over territory and resources unresolved and turn them into conflicts between sovereign states. Eritrea, independent since 1993, fought a border war with Ethiopia from 1998 to 2000 that killed between 50,000 and 100,000 people by a contemporary estimate (Dagne 2000). The war followed the collapse of a common market whose benefits had flowed mainly to Eritrea (Aberra 2016). On my reading, the border served as the pretext and the loss of those benefits as the main cause, although most accounts treat the conflict as a territorial dispute. After the Soviet Union dissolved, the dispute over Nagorno-Karabakh became a war between Armenia and Azerbaijan; a ceasefire in 1994 froze it, fighting resumed in 2020, and Azerbaijan brought the territory under its control in 2023 (Britannica Editors 2026). A cross-national study of partitions after ethnic wars found no statistically significant reduction, on average, in the probability of renewed violence (Sambanis 1999).
The present armed rivalry strengthens the case against a premature decision on separation. Armed organizations already contest power in Oromia. The Oromo Liberation Army (OLA) wages an insurgency against the government, and a Reuters investigation reported that a secret security committee, led by senior officials of the regional government, ordered dozens of killings and hundreds of arrests in the campaign against it (Reuters 2024). The federal and regional governments did not respond to Reuters’ questions; the head of the Ethiopian Human Rights Commission confirmed that the committee exists. Rival organizations could contest control of a new state while its institutions remained weak, and unresolved territorial claims could become disputes between sovereign governments. These dangers concern separation under present conditions; their persistence after a democratic interval would require a separate assessment.
External powers would raise the stakes of that rivalry. States across the Red Sea and in the Nile basin pursue interests in ports, water, and regional influence through alliances with governments and rival factions. Gulf states, an International Crisis Group report concludes, have been “injecting resources and exporting rivalries in ways that could further destabilise fragile local politics” (International Crisis Group 2019). In Sudan, United Nations monitors reported “credible” evidence that the United Arab Emirates supplied the Rapid Support Forces, a charge the Emirates deny (Al Jazeera 2024). After Ethiopia reached an agreement with Somaliland in January 2024 on access to the coast, Egypt signed a defence pact with Somalia seven months later (Sadek 2024). Secession is no prerequisite for intervention: Ethiopia already faces Egyptian opposition over the Nile, and Eritrean forces have intervened in its wars. State fragmentation could nevertheless reduce the cost of external influence. A patron’s resources would carry greater weight in a small state whose authorities face internal rivals and depend on foreign support. Rival factions would have incentives to seek competing sponsors, and those sponsors would carry external rivalries into the struggle for domestic power.
The longer-term comparison must assess union and separation under equivalent democratic conditions. If democratic institutions had taken root, the dangers associated with armed rivalry might diminish under either arrangement. Separation would nevertheless require durable agreements on borders, citizenship, assets, debt, and economic relations, while democratic union would retain common institutions to address disputes over these matters. My preference for union rests on the value of those institutions and the advantages of continued economic integration, alongside effective self-government.
Advocates of independence read the contrast between autonomy and secession differently. For them, the revisability of autonomy is its weakness, because a centre that can revise self-government can also hollow it out. Ethiopia’s federal record offers evidence: a dominant party and a centralizing state have overshadowed regional self-rule (Fiseha 2019). Independence, on that view, places self-government beyond the power of the centre to revoke it. It would also give a sovereign Oromia control over its land, revenue, and security, as well as international standing and alliances in its own right. Its population would rank it among Africa’s most populous states, large enough to retain much of the scale that smaller successors forgo.
Advocates may also argue that armed conflict arises from central domination and would ease under independence. Union itself has provided no assurance of peace: the war in Tigray from 2020 to 2022 was fought within a common state, and conflict in Oromia continues. What I regard as a costly loss of common institutions, they regard as protection against renewed domination.
Their argument has force, but independence also depends on credible commitments. Size would ease the loss of scale without removing the potential costs of new borders. Under present conditions, a sovereign Oromia could inherit rival armed organizations, internal divisions, and contested borders, although a democratic interval might reduce those risks. Independence would not guarantee democracy either. Whoever prevailed in the contest for a new state would have strong incentives to entrench its rule, and the security committee reported above suggests that the present regional authorities already resort to extrajudicial means. Eritrea shows how a liberation movement can become a lasting autocracy: it has never held national elections, has never put its 1997 constitution into force, and its transitional assembly has not met since 2002 (Freedom House 2024).
Their remaining claims concern rights, resources, and allies. The protection of rights would still depend on domestic institutions, just as it would under regional autonomy. Effective self-government within a democratic union can secure substantial control over land and revenue. The value of foreign alliances depends on interests that may diverge from those of the people they promise to support. Union under authoritarian rule has brought war; union under democratic governance has yet to be tried. The proposed interval would allow citizens to judge whether democratic union fulfils its promises.
IV. Conclusion
The case for democratic union rests on three advantages: effective self-government, the gains of economic integration, and common institutions through which citizens can seek redress and revise constitutional arrangements. The absence of conditions for free and informed consent, compounded by present armed rivalry and the risk that independence would entrench a new autocracy, provides an additional reason to oppose a premature decision on separation. My preference for union would hold after a democratic interval; citizens would then judge it from experience of democratic self-government.
I accept a legitimate decision to separate under rules agreed in advance. The authors’ call for a process “without coercion or the imposition of predetermined outcomes” permits advocacy for union while excluding its imposition. I defend a qualified majority because a decision of such consequence warrants broad consent, and a democratic interval because citizens should have an opportunity to judge union by its performance under democracy. Neither proposal should give the centre an indefinite veto over the citizens’ choice; hence the backstop.
The authors’ commitments to peaceful coexistence and concern for state survival provide grounds for a democratic union, though they do not amount to an explicit endorsement of it. Their call for free consent and my case for union can meet in a common project: a democratic Ethiopia whose citizens choose union because it has served them.
References
Aberra, Worku. 2016. “Asymmetric Benefits: The Ethio-Eritrea Common Market (1991 to 1998).” International Journal of African Development 4 (1): article 6. https://scholarworks.wmich.edu/ijad/vol4/iss1/6
Aberra, Worku. 2026. “Sovereignty and Institutions: A Reply to Messay Kebede.” Borkena, 10 October. https://borkena.com/ethiopia-sovereignty-and-institutions-a-reply-to-messay-kebede/
AFP (Agence France-Presse). 2018. “Ethiopia Ethnic Fighting Displaces One Million.” Arab News, 5 February. https://www.arabnews.com/world/ethiopia-ethnic-fighting-displaces-one-million-1240011
Al Jazeera. 2024. “UAE Denies Sending Weapons to Sudan’s RSF Paramilitary: Report.” 24 January. https://www.aljazeera.com/news/2024/1/24/uae-denies-sending-weapons-to-sudans-rsf-paramilitary-report
Alesina, Alberto, and Enrico Spolaore. 1997. “On the Number and Size of Nations.” Quarterly Journal of Economics 112 (4): 1027–1056.
Blanchard, Olivier, and Michael Kremer. 1997. “Disorganization.” Quarterly Journal of Economics 112 (4): 1091–1126. https://doi.org/10.1162/003355300555439
Britannica Editors. 2026. “Nagorno-Karabakh.” Encyclopaedia Britannica, last updated 26 September. https://www.britannica.com/place/Nagorno-Karabakh
Campos, Nauro F., and Fabrizio Coricelli. 2002. “Growth in Transition: What We Know, What We Don’t, and What We Should.” Journal of Economic Literature 40 (3): 793–836.
Canada. 2000. An Act to Give Effect to the Requirement for Clarity as Set Out in the Opinion of the Supreme Court of Canada in the Quebec Secession Reference (Clarity Act). S.C. 2000, c. 26. https://laws-lois.justice.gc.ca/eng/acts/C-31.8/FullText.html
Checchi, Francesco, Adrienne Testa, Abdihamid Warsame, Le Quach, and Rachel Burns. 2018. Estimates of Crisis-Attributable Mortality in South Sudan, December 2013–April 2018: A Statistical Analysis. London: London School of Hygiene and Tropical Medicine, September. https://www.lshtm.ac.uk/south-sudan-full-report
Council of Oromo Scholars. 2026. “Statement by Oromo Intellectuals on Ethiopia’s Political Crisis and the Need for an Inclusive Transition.” 17 October. Reproduced by Qeerroo Bilisummaa Oromoo. https://qbo-abo-wbo.org/wp-content/uploads/2026/10/OromoIntelR-128129-2.pdf
Dagne, Ted. 2000. The Ethiopia–Eritrea Conflict. CRS Report RL30598. Washington, DC: Congressional Research Service, 6 July.
Ethiopia. 1995. Constitution of the Federal Democratic Republic of Ethiopia. Proclamation No. 1/1995. https://www.unesco.org/education/edurights/media/docs/225808f58dec4e22af73b8b7cfbcb181b9039e76.pdf
Fearon, James D. 1995. “Rationalist Explanations for War.” International Organization 49 (3): 379–414. https://doi.org/10.1017/S0020818300033324
Fidrmuc, Jan, and Jarko Fidrmuc. 2003. “Disintegration and Trade.” Review of International Economics 11 (5): 811–829. https://doi.org/10.1046/j.1467-9396.2003.00419.x
Fiseha, Assefa. 2019. “Federalism, Development and the Changing Political Dynamics in Ethiopia.” International Journal of Constitutional Law 17 (1): 151–176. https://doi.org/10.1093/icon/moz008
Freedom House. 2024. “Eritrea.” Freedom in the World 2024. Washington, DC: Freedom House. https://freedomhouse.org/country/eritrea/freedom-world/2024
Hernández, Gonzalo, and Carlos Closa. 2022. “The Challenge of Catalan Secessionism to the European Model of the Rule of Law.” Hague Journal on the Rule of Law 14 (2): 257–285. https://doi.org/10.1007/s40803-022-00177-7
IFES (International Foundation for Electoral Systems). n.d. “Elections: Saint Kitts and Nevis Referendum 1998.” ElectionGuide. https://electionguide.org/elections/id/97/
IMF (International Monetary Fund). 2008. “IMF Executive Board Concludes 2007 Article IV Consultation with Montenegro.” Public Information Notice 08/06, 22 January. https://www.imf.org/en/News/Articles/2015/09/28/04/53/pn0806
International Crisis Group. 2019. Intra-Gulf Competition in Africa’s Horn: Lessening the Impact. Middle East Report 206. Brussels: International Crisis Group, 19 September. https://www.crisisgroup.org/node/11321
Linz, Juan J., and Alfred Stepan. 1996. Problems of Democratic Transition and Consolidation: Southern Europe, South America, and Post-Communist Europe. Baltimore: Johns Hopkins University Press.
OSCE/ODIHR. 2006. Republic of Montenegro: Referendum on State-Status, 21 May 2006. Final Report. Warsaw, 4 August. https://www.osce.org/files/f/documents/f/0/20098.pdf
Powell, Robert. 2006. “War as a Commitment Problem.” International Organization 60 (1): 169–203. https://doi.org/10.1017/S0020818306060061
Reuters. 2017. “Spain Set to Impose Direct Rule in Catalonia as Crisis Spirals.” Arab News, 27 October. https://www.arabnews.com/node/1184001
Reuters. 2024. “In Ethiopia, a Secret Committee Orders Killings and Arrests to Crush Rebels.” Citizen Digital, 25 February. https://citizen.digital/news/in-ethiopia-a-secret-committee-orders-killings-and-arrests-to-crush-rebels-n337336
Sadek, George. 2024. “Egypt/Somalia: Military Cooperation Agreement Signed.” Global Legal Monitor. Washington, DC: Law Library of Congress, 26 September. https://www.loc.gov/item/global-legal-monitor/2024-09-25/egypt-somalia-military-cooperation-agreement-signed/
Saint Kitts and Nevis. 1983. Saint Christopher and Nevis Constitution Order 1983. Section 113. https://lawcommission.gov.kn/wp-content/documents/Annual-Laws/The-Constitution-of-St-Christopher-and-Nevis.pdf
Sambanis, Nicholas. 1999. “Ethnic Partition as a Solution to Ethnic War: An Empirical Critique of the Theoretical Literature.” Policy Research Working Paper 2208. Washington, DC: World Bank. https://ideas.repec.org/p/wbk/wbrwps/2208.html
Scottish Independence Referendum Act 2013. asp 14. https://www.legislation.gov.uk/asp/2013/14/enacted
United Nations. 1993. “United Nations Observer Mission to Verify the Referendum in Eritrea: Report of the Secretary-General.” A/48/283, 11 August. https://docs.un.org/en/A/48/283
World Bank. 2017. South Sudan Economic Update: Taming the Tides of High Inflation; Policy Options for South Sudan. Washington, DC: World Bank. https://documents1.worldbank.org/curated/en/806291508505062484/pdf/120563-WP-SSEUUpdatefinal-PUBLIC.pdf
Worku Aberra is a professor and scholar in residence in the Department of Economics at Dawson College in Montreal, Canada.