The 1995 Constitution: Ethnic Federalism as a Theory of the State

Thesis

The Constitution of the Federal Democratic Republic of Ethiopia is the longest-serving Ethiopian constitution in modern history. It was ratified by a Constituent Assembly on 8 December 1994 and entered into force on 21 August 1995. As of mid-2026 it remains nominally in force, although its substantive operation has been deeply contested since 2018 and effectively suspended in important respects since the Tigray war.¹ The document contains 106 articles in 11 chapters, including the most elaborate enumerated rights chapter of any constitution in Africa (Articles 13–44, roughly a third of the text), a federal structure dividing the country into nine — now twelve — regional states, an upper house of parliament (the House of Federation) charged with constitutional interpretation, and the explicit identification of “the Nations, Nationalities and Peoples of Ethiopia” as the sovereign holders of constituent power (Article 8).²

This is, on paper, the most ambitious Ethiopian constitution. It is also the document that explicitly enshrines the central distinguishing feature of Ethiopian constitutional theory: the identification of corporate ethnic groups, rather than individual citizens, as the primary constitutional subjects. The Preamble begins “We, the Nations, Nationalities and Peoples of Ethiopia.” Article 8 vests sovereignty in them. Article 39 grants each of them an unconditional right to self-determination including secession. The 1995 Constitution is, in its own self-understanding, a treaty among ethnic groups rather than a contract among individuals. This is not a casual feature; it is the foundational theoretical commitment. Every subsequent constitutional disagreement in Ethiopia — about federalism, about citizenship, about minority rights within regions, about the legitimacy of secession movements, about the future of the federation — is downstream of this commitment.

The drafting

The Constitutional Commission was established by Proclamation No. 24/1992 with twenty-nine members chaired by Ato Kifle Wodajo, including representatives of the EPRDF’s constituent parties, some independent jurists, and a small number of representatives from organisations outside the EPRDF.³ The commission worked between 1993 and 1994. It consulted comparative constitutional material from federal states (the United States, Switzerland, Germany, India, Belgium, Canada), held public hearings in various regions, received written submissions, and produced multiple drafts. The drafting process was, in technical terms, careful.

The Constituent Assembly that ratified the draft was elected in June 1994. The election was boycotted by significant parts of the opposition — the OLF had withdrawn from the transitional government in 1992 and refused to participate; the All-Amhara People’s Organisation (AAPO) and several smaller parties also boycotted — and the EPRDF-affiliated parties won the overwhelming majority of seats. The assembly ratified the draft on 8 December 1994.⁴ The opposition boycotts and the EPRDF-dominated composition of the assembly are recurring elements in the legitimacy critique of the 1995 Constitution and continue to be cited by opposition movements as of 2026.

The text

The 1995 Constitution is structurally and stylistically more ambitious than its predecessors. Key features:

Sovereignty and the constitutional subject. The Preamble identifies the constitutional authors as “We, the Nations, Nationalities and Peoples of Ethiopia.” Article 8 declares: “Sovereign power resides in the Nations, Nationalities and Peoples of Ethiopia. This Constitution is an expression of their sovereignty.”⁵ This is the constitutional choice that distinguishes Ethiopian federalism from American, German, Indian, or Swiss federalism: the constituent subjects are ethnic groups, not citizens, regions, or states.

Federal structure. Article 47 originally established nine regional states: Tigray, Afar, Amhara, Oromia, Somali, Benishangul-Gumuz, Southern Nations Nationalities and Peoples, Gambella, and Harari. Article 47(2) provides for the creation of additional states by the constitutional self-determination process. Since 2018 this process has produced Sidama (2020), Southwest Ethiopia (2021), South Ethiopia (2023), and Central Ethiopia (2023), bringing the total to thirteen counting the two chartered cities (Addis Ababa and Dire Dawa) and twelve regional states.⁶

Rights chapter. Articles 13–44 enumerate civil, political, economic, social, cultural, and group rights at length. The chapter incorporates by reference international human-rights instruments to which Ethiopia is a party (Article 13.2). It contains rights to life, liberty, security of person, equality before the law, freedom of religion, freedom of expression, freedom of association and assembly, the right to property, the right to vote, the right to a fair trial, the rights of women and children, the right to development, and the rights of nations/nationalities/peoples.⁷ This is the most extensive rights chapter in any African constitution.

The House of Federation. Articles 61–68 establish the House of Federation as the upper house, composed of representatives of nations/nationalities/peoples elected by the regional councils. Its membership in 2026 is 153.⁸ Crucially — and this is the subject of Article 13 of this series — Article 62 vests constitutional interpretation in the House of Federation rather than in a constitutional court or the regular judiciary. This is the structural feature that, more than any other, has determined the operational weakness of constitutional review in Ethiopia.

Secession. Article 39 grants every nation/nationality/people the unconditional right to self-determination, including the right to secession through a defined procedure.⁹ Subject of Article 10 of this series.

Land. Article 40 vests all land in the state and prohibits its private alienation, granting only use-rights to peasants and pastoralists.¹⁰ Subject of Article 11.

Federal-state relations. Articles 50–52 divide competences between federal and state governments, allocating to the federal government foreign affairs, defence, foreign trade, currency, intellectual property, and a list of specified national matters, and allocating to the states everything not specifically reserved to the federal government.¹¹

The executive. Articles 72–77 establish a parliamentary system with the Prime Minister as head of government, selected from and accountable to the House of Peoples’ Representatives. The President is a largely ceremonial head of state. The Council of Ministers exercises executive power.¹²

The judiciary. Articles 78–82 establish the federal and state judicial systems, declaring judicial independence (Article 78.1), with the Federal Supreme Court at the apex of the federal judiciary and the regional supreme courts at the apex of the state judiciaries.¹³ As with previous Ethiopian constitutions, the text declares judicial independence; the operational reality is more complicated.

Ethnic federalism as a theory

The theoretical commitment to ethnic federalism is the constitution’s most distinctive and most contested feature. Why is it there?

Three reasons are commonly cited. First, the EPRDF’s intellectual framework — particularly the TPLF’s reading of Marxist-Leninist national-question theory — held that the “nationality question” was the central political problem of Ethiopia, that the imperial and Derg states had been “prison-houses of nationalities” that suppressed legitimate national aspirations, and that only a federation organised around national-territorial autonomy could resolve the question.¹⁴ This reading was not exclusively TPLF; it was shared by the OLF, by significant parts of the southern nationality organisations, and by the broader Ethiopian student movement of the 1960s and 1970s from which the EPRDF leadership emerged.¹⁵

Second, the practical political conditions of 1991–95 required some accommodation of the substantive demands that had powered the war. The TPLF had won by mobilising Tigrayan nationalism; the EPLF by mobilising Eritrean nationalism (and was being granted independence as a result); the OLF by mobilising Oromo nationalism. The various southern nationality movements had aligned with the EPRDF on the basis of expected nationality-based autonomy. To draft a constitution that did not deliver on these expectations was politically impossible.

Third, the international constitutional moment of 1991–94 was favourable to ethnic-federal arrangements. The post-Cold War transitions in Eastern Europe and the former Soviet Union, the Yugoslav crisis, and the South African transition were all unfolding in ways that put the management of ethnic and national difference at the centre of constitutional design. Ethiopian drafters consulted these comparators.

The result is a constitution that, more thoroughly than any other in the world, organises political life around corporate ethnic identity. The boundaries of the federal states are drawn primarily by ethnicity (with significant historical and topographical irregularities). The working languages of state and federal institutions are determined by ethnic considerations. Political party formation has, in practice, been overwhelmingly along ethnic lines. The House of Federation represents nationalities rather than territorial units. The most significant rights — including the right to secession — are conferred on nationalities rather than individuals.

What it has produced

In thirty-one years of operation (1995–2026), the 1995 Constitution has produced a particular set of consequences that the article must catalogue honestly.

Linguistic and cultural recognition. Regional languages — Tigrinya, Afaan Oromo, Somali, Afar, and others — have become the official languages of regional governments, of regional courts and administration, and of regional primary education. Cultural and religious institutions of various nationalities have received recognition and support. The pre-1991 dominance of Amharic as the sole language of the modern state has been substantially diminished. This is a real outcome and matters.¹⁶

Federal representation. Regional governments have, with varying degrees of autonomy, conducted their own affairs in many domains: education, health, agriculture, local administration, regional police. The principle that Ethiopian governance is layered between federal and regional levels has been institutionalised.

Ethnic political competition. Politics has been organised around ethnic identity. Inter-ethnic political competition, sometimes peaceful and sometimes violent, has become the operative pattern. The 2018 transition, the post-2018 fragmentation of the EPRDF, the formation of the Prosperity Party, the Tigray war, the Amhara-Fano conflict, the OLA insurgency in Oromia — these are all developments shaped by the constitutional framework’s organisation of politics by ethnicity.¹⁷

Subdivision pressure. The constitutional right of nations/nationalities/peoples to their own regional state has produced a continuing pressure to subdivide. The dissolution of SNNPR into Sidama (2020), Southwest Ethiopia (2021), South Ethiopia (2023), and Central Ethiopia (2023) is the most dramatic instance, but pressure for further subdivision continues — the Gurage demand for separate statehood, multiple zonal claims within the Amhara region, claims within Oromia.¹⁸

Failure of cross-ethnic political organisation. No major Ethiopian political party with significant electoral support has succeeded in organising across ethnic lines on a sustained basis. The CUD/Kinijit’s 2005 effort came closest and was suppressed. The Prosperity Party’s nominal pan-Ethiopianism is, in operation, an Oromo-led coalition with regional affiliates, and is widely understood as such by Ethiopian voters.¹⁹ The constitutional architecture has institutionalised ethnic political organisation; cross-ethnic alternatives operate against the structural grain.

Violence at federation boundaries. The boundaries between regional states have produced repeated violence: the Oromia-Somali border (2017–18), the Tigray-Amhara contested areas (Welkait, Tselemt, Raya, Humera), the Sidama-Wolayta boundary, the Benishangul-Gumuz-Oromia border, multiple smaller hotspots.²⁰ When political authority is organised around the corporate identity of titular ethnic groups, the boundaries between titular zones become high-stakes.

Steelman of the defenders

The federalist defence of the 1995 Constitution has serious advocates and serious arguments.

The first argument is that ethnic federalism is the constitutional architecture appropriate to Ethiopia’s actual demographic and historical reality. Ethiopia is a multinational state with seventy-plus identifiable nationality groups; no single ethnicity is a majority; the historical alternative — Amharic-language imperial-cultural domination — was illegitimate and unstable. A constitution that ignores nationality, on this view, would not be neutral; it would be a continuation of the pre-1991 settlement under different vocabulary. This is a serious argument that the unitary critics often fail to engage on its own terms.²¹

The second argument is that the 1995 Constitution has, despite its imperfections, produced more institutional output than any previous Ethiopian constitution: regional governments that exist and function, four general elections (1995, 2000, 2005, 2010, 2015, 2021, 2026), language rights actually exercised, federal-state revenue-sharing arrangements that work, and Afrobarometer survey data showing 54 percent of Ethiopians prefer federalism to a unitary state and 67 percent support keeping the constitution with some amendments (2023).²² The institutional output is real. To dismiss it is unfair to the document and to the regional governments that have operated under it.

The third argument is that the criticisms of the 1995 Constitution — that it has produced violence, that it has institutionalised ethnic politics, that it has weakened pan-Ethiopian identity — are confusions of cause and effect. The violence existed before 1995, the ethnic politics existed before 1995, the weak pan-Ethiopian identity existed before 1995. The 1995 Constitution did not cause these things; it acknowledged them. The alternative constitutional design that the critics imply — strong central state, individual citizenship, suppression of nationality politics — was the imperial and Derg model, which produced its own and worse violence.

These are not throwaway arguments. The series engages them at length in Articles 13, 14, and 22.

The reply

The reply to the federalist defence is not that ethnic federalism is wrong as an abstract design but that the specific way Ethiopia’s ethnic federalism was constructed produces particular failures that are not necessary consequences of federalism as such. South Africa’s 1996 Constitution recognises eleven official languages and contains group-rights provisions, but it does not vest sovereignty in ethnic groups, does not grant constitutional secession rights, and does not organise federal boundaries primarily by ethnicity. India’s constitution recognises multiple linguistic states without granting any of them the right of self-determination. Switzerland’s constitution recognises four official languages within a federation organised around historical-territorial cantons. Each of these federal systems acknowledges diversity without elevating corporate ethnic identity to the status of constitutional sovereign. Ethiopia’s choice was a specific choice, not a necessary consequence of taking diversity seriously.

The other element of the reply is that the operational record of the 1995 Constitution, examined honestly, shows that the corporate-ethnic constitutional logic has produced the federation’s worst pathologies: the inability to resolve boundary disputes peacefully, the inability to manage cross-ethnic political competition, the inability to provide stable status for minorities within regional states, and the inability to constrain the federal executive when it has aligned with one of the constitutional subjects against the others. These pathologies are taken up in the subsequent articles. The point for the present article is that they are downstream of the foundational choices made in 1995, not contingent failures that could be reformed without revisiting the foundations.

Multi-causal accounting

The 1995 Constitution is the product of (a) the EPRDF’s ideological commitments, particularly TPLF Marxism-Leninism’s reading of the nationality question; (b) the practical political requirements of governing post-civil-war Ethiopia under coalition pressure from formerly armed nationality movements; (c) the international constitutional context of the early 1990s; (d) the personal influence of specific constitutional drafters, including Andreas Eshete, Fasil Nahum, and others; and (e) the constraints imposed by the prior decisions of the 1991 Transitional Period Charter, which the 1995 Constituent Assembly was not in a position to fundamentally revisit. Each of these is a real cause. The 1995 Constitution is overdetermined; it was not going to be something else given the conditions of 1991–94.

What the article establishes

The 1995 Constitution is the document whose theoretical commitment to ethnic federalism shapes contemporary Ethiopian politics most heavily. Its rights chapter is the most extensive in Africa. Its institutional design is more elaborate than any previous Ethiopian instrument. Its operational record is mixed: real institutional output in some domains, persistent inability to constrain executive power against the constitutional design in others. The next two articles examine the two provisions whose operational consequences have been most decisive: Article 39 on secession and Article 40 on land.

---

References

Footnotes

1. Constitution of the Federal Democratic Republic of Ethiopia, Federal Negarit Gazeta, Proclamation No. 1/1995, entered into force 21 August 1995. On the contested current operation, see World Report 2026: Ethiopia, Human Rights Watch, https://www.hrw.org/world-report/2026/country-chapters/ethiopia. ↩

2. 1995 Constitution, Preamble and Article 8. ↩

3. Proclamation No. 24/1992; see Fasil Nahum, Constitution for a Nation of Nations, ch. 5, and Tsegaye Regassa, “The Making and Legitimacy of the Ethiopian Constitution,” Afrika Focus 23, no. 1 (2010). ↩

4. On the Constituent Assembly elections and ratification, see Marina Ottaway, “The Ethiopian Transition: Democratization or New Authoritarianism?,” Northeast African Studies 2, no. 3 (1995). ↩

5. 1995 Constitution, Preamble and Article 8. ↩

6. 1995 Constitution, Article 47; on the post-2018 region creations, see ACLED Ethiopia Peace Observatory, “SNNPR,” https://epo.acleddata.com/southern-nations-nationalities-and-peoples-region-snnpr/. ↩

7. 1995 Constitution, Articles 13–44. ↩

8. 1995 Constitution, Articles 61–68. ↩

9. 1995 Constitution, Article 39. ↩

10. 1995 Constitution, Article 40. ↩

11. 1995 Constitution, Articles 50–52. ↩

12. 1995 Constitution, Articles 72–77. ↩

13. 1995 Constitution, Articles 78–82. ↩

14. On the TPLF/EPRDF intellectual framework, see Aregawi Berhe, A Political History of the Tigray People’s Liberation Front (Los Angeles: Tsehai, 2009); John Young, Peasant Revolution in Ethiopia, ch. 5. ↩

15. Bahru Zewde, The Quest for Socialist Utopia: The Ethiopian Student Movement, c. 1960–1974 (Oxford: James Currey, 2014). ↩

16. On language rights outcomes, see Yonatan Fessha, Ethnic Diversity and Federalism: Constitution Making in South Africa and Ethiopia (Aldershot: Ashgate, 2010), and the Afrobarometer Round 9 (2023) data. ↩

17. For contemporary documentation, see Critical Threats Project Africa File series, 2025–2026, https://www.criticalthreats.org/. ↩

18. ACLED Ethiopia Peace Observatory data; see also “Splitting Ethiopia’s Southern Nations Region into four could promote peace,” The Africa Report, October 2020. ↩

19. On the Prosperity Party as effectively Oromo-led, see “Abiy Ahmed’s Hegemonistic Prosperity Party,” Borkena, June 2026, https://borkena.com/2026/06/24/abiy-ahmeds-hegemonistic-prosperity-party-a-liability-for-ethiopia/. ↩

20. On federation boundary violence, see ACLED Ethiopia data 2017–2026. ↩

21. For the federalist defence, see Yonatan Fessha, Ethnic Diversity and Federalism; Tsegaye Regassa, “Sub-National Constitutions in Ethiopia,” Mizan Law Review 3, no. 1 (2009); Solomon Negussie, Fiscal Federalism in the Ethiopian Ethnic-based Federal System (Nijmegen: Wolf Legal Publishers, 2006). ↩

22. Afrobarometer Round 9, Ethiopia, 2023, cited in https://50shadesoffederalism.com/case-studies/federalism-and-ethnic-accommodation-in-ethiopia-a-promised-solution-or-a-persistent-dilemma/. ↩