The Recurring Failure: A Framework for Reading Ethiopia's Constitutional History

Thesis

Ethiopia is not a country without constitutions. It is a country with too many of them — five constitutional instruments in nine decades, each drafted with apparent care, each ratified through apparent process, each setting out the structure of a state and the rights of its citizens with apparent confidence. Yet across those nine decades, no Ethiopian executive has been removed from power by constitutional means, no Ethiopian constitutional court has voided an executive act on rights grounds, no Ethiopian opposition party has taken office through an electoral defeat of an incumbent, and no Ethiopian citizen has been able to count on the constitutional text as a reliable shield against the state. The texts exist. The constitutionalism does not. This series exists to explain why.

The texts

The catalogue is short and easily stated. The 1931 Constitution of Haile Selassie I — Ethiopia’s first written constitution — was promulgated in July of that year and modelled loosely on the Meiji Constitution of 1889.¹ It was revised and substantially expanded in 1955, with the assistance of American advisors led by John H. Spencer, to produce a document with a bicameral parliament, a justiciable bill of rights, and an elected Chamber of Deputies.² Proclamation No. 1 of 1974 — the Provisional Military Administrative Council’s founding act — suspended the 1955 Constitution and ruled by decree until 1987.³ The 1987 Constitution of the People’s Democratic Republic of Ethiopia, modelled on the Soviet Constitution of 1977, established a one-party state under the Workers’ Party of Ethiopia.⁴ The Transitional Period Charter of July 1991 governed Ethiopia for four years after the EPRDF’s victory.⁵ The 1995 Constitution of the Federal Democratic Republic of Ethiopia, ratified December 1994 and entered into force August 1995, is the constitution still nominally in force today.⁶

That is the inheritance. Five formal documents, each ostensibly a fresh start, each in fact an exercise in re-legitimation by a new sovereign.

The pattern

The pattern that links them is not in their texts but in their lives. Each was drafted with seriousness — the 1955 by American constitutional lawyers, the 1987 by Soviet-trained jurists, the 1995 by a Constitutional Commission of fifteen members chaired by Kifle Wodajo and including some of Ethiopia’s most distinguished legal scholars. Each was promulgated with ceremony. Each contained, in its text, provisions that on their face would have made the executive answerable to law. The 1955 Constitution contained an elected lower chamber and a justiciable bill of rights. The 1987 Constitution contained provisions for citizen participation, the rule of law, and socialist legality. The 1995 Constitution contains the most elaborate human-rights chapter of any constitution in Africa — fully a third of its text — together with a federal structure, a House of Federation as constitutional adjudicator, and an unconditional right of secession under Article 39.

And yet. The 1955 Constitution was suspended on the morning of 12 September 1974 by a colonels’ communiqué.⁷ The 1987 Constitution was abrogated by the EPRDF in May 1991 by force of arms. The 1995 Constitution remains in force on paper and has, in operation, produced the Tigray war, the postponed elections of 2020 — extended by an act of constitutional interpretation that opposition parties and many constitutional lawyers regarded as fabricated — and a federation that as of mid-2026 contains three regions (Amhara, Tigray, Oromia) where federal authority is contested by force of arms.⁸ The texts are not the disease. The disease is what happens to them.

The three lenses

To diagnose the disease, the series uses three lenses, each from a thinker whose politics differed sharply from the others but whose constitutional conclusions converged.

Madison. James Madison’s argument in Federalist 10, 47, 48, and 51 is that the cure for the disease of faction in a republic is structural: ambition must be made to counteract ambition, power must be set against power, and “parchment barriers” — words on paper — are insufficient.⁹ A constitution that does not actually divide power, give each branch the means and the motive to resist encroachment by the others, and produce a working system of checks is a constitution only in name. The Madisonian test is institutional, not textual. By that test, every Ethiopian constitution has failed, for reasons specific to each but structurally identical: in each, the executive of the day has commanded such an overwhelming preponderance of power over the legislature and judiciary that the formal separations contained in the text have been operationally irrelevant.

Hayek. Friedrich Hayek’s argument in The Constitution of Liberty and Law, Legislation, and Liberty is that the rule of law means predictable general rules that apply equally to all and constrain official discretion.¹⁰ A regime in which the law means whatever the executive needs it to mean in the moment — what Carl Schmitt celebrated as sovereign decision on the exception — is the antithesis of constitutionalism, not a version of it. Hayek’s test is whether ordinary citizens can predict, in advance, how the state will treat them under the law, and whether the answer depends on their political affiliation. By that test, Ethiopian constitutional practice across all three regimes has been Schmittian, not Hayekian: the law has reached one set of citizens and not another, depending on who governed.

Acemoglu and Robinson. Daron Acemoglu and James Robinson’s argument in Why Nations Fail and The Narrow Corridor is that institutions are inclusive when they distribute power broadly and constrain incumbents, and extractive when they concentrate power and entrench incumbents.¹¹ Whether a written constitution becomes a binding constraint depends on the underlying political economy: on whether enough actors with enough power have reason to enforce the text against those who would violate it. By that test, Ethiopia has never had a domestic political coalition with sufficient scale, cross-ethnic reach, or durability to make constitutional enforcement the operative agenda of any major political actor. This is not because such coalitions never existed — the CUD/Kinijit platform of 2005 had real liberal-constitutional content, and parts of the 2015–18 protests did as well — but because each was either suppressed, co-opted, or fragmented before it could anchor a new equilibrium.

Schmitt as diagnosis

Throughout this series, Carl Schmitt appears as the most accurate descriptor of what Ethiopian constitutional practice has actually been — not as a thinker whose conclusions are endorsed, but as a diagnostician whose framework explains the facts on the ground.¹² When Mengistu suspended the 1955 Constitution by decree in 1974, that was Schmittian decision on the exception. When the House of Federation extended the terms of all federal and regional legislatures in June 2020 by an act of “interpretation” that found language in the constitution that was not there, that was Schmittian decision on the exception.¹³ When the federal government in February 2026 removed five constituencies from Tigrayan electoral oversight ahead of the June elections, citing unresolved territorial claims, that was Schmittian decision on the exception.¹⁴ The pattern is the same; only the regime changes.

This matters because the most common defence of these episodes — that they were necessary responses to crisis, that the alternative was state collapse, that constitutional letter must yield to political reality — is precisely Schmitt’s argument. To accept that argument is to accept that Ethiopia has not had a constitutional order in the substantive sense at any point in its modern history, only successive sovereigns making decisions about which rules will apply when. That conclusion is uncomfortable, and the series does not arrive at it lightly. But the alternative — to claim that Ethiopia has had functioning constitutionalism that was occasionally interrupted by crisis — does not survive the evidence.

Steelman of the opposing reading

The defenders of Ethiopian constitutional practice make three serious arguments that this series will engage repeatedly rather than dismiss.

The first is the federalist-defender argument: that the 1995 Constitution produced real institutional output for nearly three decades — language rights, regional self-government, four general elections, the Sidama referendum of 2019, the Southwest Ethiopia and South Ethiopia referendums of 2021 and 2023 — and that to say it “did not function as a constitution” ignores what it actually delivered.¹⁵ The reply is not to deny the output; the reply is to specify what kind of output. Each of the cited successes was an exercise of corporate ethnic self-determination consistent with regime preferences; none was a constitutional restraint imposed against regime preferences. The distinction matters.

The second is the sequencing argument: that constitutionalism in a country emerging from civil war is necessarily incremental, that consolidation must precede constraint, and that to demand Madisonian institutions in 1995 or 2018 was to demand the impossible.¹⁶ The reply is empirical: South Africa managed it in 1996, India in 1950, Indonesia between 1998 and 2004. Comparable countries have done what Ethiopia has not. The sequencing argument explains why it is hard; it does not explain why it has not happened across three decades.

The third is the cultural-fit argument: that liberal constitutionalism is a Western imposition and that Ethiopia has its own indigenous traditions of legitimate authority — the Fetha Nagast, the gada institutions, customary mediation — which serve the functions a Western constitution would serve.¹⁷ The reply is that these institutions have been displaced rather than incorporated by the modern Ethiopian state, that none of the three regimes has actually relied on them for restraint, and that to invoke them now as substitutes for the constitutional constraints that do not exist is to perform a sleight of hand.

What this series will do

Each subsequent article takes one constitutional instrument, one constitutional moment, or one structural feature of the Ethiopian constitutional order, examines what it claimed to do, what it actually did, and why those two things diverged. The arc runs from the Fetha Nagast through the 2024–26 crisis. The argument is cumulative: by the end, the claim that Ethiopia has had constitutional form without constitutionalism for ninety years should rest not on assertion but on the weight of twenty-five separate cases each pointing in the same direction.

The argument is also prescriptive. The final three articles compare Ethiopia to South Africa, India, Indonesia, and Nigeria; ask why each Ethiopian regime has repeated the pattern; and propose what a constitutional re-founding would actually require. The proposition is that the cure is known and has been implemented elsewhere. The obstacle has never been knowledge. It has been political will, and the political will has been absent because the political-economy configuration that requires unconstrained executive sovereignty has survived every change of regime. Until that configuration changes, the texts will continue to be drafted and the constitutions will continue not to function.

That is the disease. The series, like the Birr series before it, exists to name it.

---

References

Footnotes

1. Bahru Zewde, A History of Modern Ethiopia, 1855–1991, 2nd ed. (Oxford: James Currey, 2001), 110–11; Margery Perham, The Government of Ethiopia (London: Faber, 1948), 84–88. The 1931 Constitution is reproduced in Negarit Gazeta No. 1, Year 1 (1931). ↩

2. John H. Spencer, Ethiopia at Bay: A Personal Account of the Haile Sellassie Years (Algonac, MI: Reference Publications, 1984), 273–285; Christopher Clapham, Haile-Selassie’s Government (London: Longmans, 1969), 24–48. The 1955 Revised Constitution, Negarit Gazeta, Proclamation No. 149 of 1955. ↩

3. “Proclamation No. 1 of 1974,” Negarit Gazeta, 12 September 1974. See Andargachew Tiruneh, The Ethiopian Revolution 1974–1987: A Transformation from an Aristocratic to a Totalitarian Autocracy (Cambridge: Cambridge University Press, 1993), 56–62. ↩

4. Constitution of the People’s Democratic Republic of Ethiopia, Negarit Gazeta, Proclamation No. 1 of 1987. See Christopher Clapham, Transformation and Continuity in Revolutionary Ethiopia (Cambridge: Cambridge University Press, 1988), 53–67. ↩

5. Transitional Period Charter of Ethiopia, Negarit Gazeta, 22 July 1991. ↩

6. Constitution of the Federal Democratic Republic of Ethiopia, Federal Negarit Gazeta, Proclamation No. 1/1995, entered into force 21 August 1995. ↩

7. The 1955 Constitution was suspended by Proclamation No. 1 of 1974; the Provisional Military Administrative Council’s first communiqué, issued on the morning of 12 September 1974, deposed Haile Selassie. See Andargachew Tiruneh, The Ethiopian Revolution, 56–62. ↩

8. Critical Threats Project / Institute for the Study of War, “Africa File, May 28, 2026” and “Africa File, April 23, 2026,” documenting active conflict in Amhara, contested authority in Tigray, and ongoing OLA insurgency in Oromia. See https://www.criticalthreats.org/analysis/ethiopia-fano-tplf-abiy-somalia-isis-puntland-drc-m23-rwanda-rubaya-africa-file-may-28-2026 and https://www.criticalthreats.org/analysis/ethiopia-tigray-tplf-tia-mozambique-rwanda-ismp-sudan-rsf-saf-africa-file-april-23-2026. ↩

9. James Madison, Federalist 10, 47, 48, 51, in Alexander Hamilton, James Madison, and John Jay, The Federalist Papers, ed. Clinton Rossiter (New York: New American Library, 1961). The phrase “parchment barriers” is from Federalist 48. ↩

10. F.A. Hayek, The Constitution of Liberty (Chicago: University of Chicago Press, 1960), chs. 10–14; Law, Legislation, and Liberty, 3 vols. (Chicago: University of Chicago Press, 1973–79), vol. 1, ch. 4. ↩

11. Daron Acemoglu and James A. Robinson, Why Nations Fail (New York: Crown, 2012), chs. 3–4; The Narrow Corridor (New York: Penguin, 2019), chs. 1–2. ↩

12. Carl Schmitt, Political Theology: Four Chapters on the Concept of Sovereignty, trans. George Schwab (Cambridge, MA: MIT Press, 1985 [1922]), ch. 1: “Sovereign is he who decides on the exception.” Used in this series descriptively to characterize what Ethiopian constitutional practice has been, not normatively to endorse it. ↩

13. Adem Kassie Abebe and Solomon Negussie, critiques of the House of Federation’s June 2020 decision; Marew Abebe and Mekonnen Yoseph, “The Constitutionality of Election Postponement in Ethiopia amidst COVID-19 Pandemic,” Jimma University Journal of Law (2021), https://journals.ju.edu.et/index.php/jlaw/article/download/2667/1353/. ↩

14. “Ethiopia’s Abiy is using the Pretoria accord to strip Tigrayans of their rights,” The Africa Report, April 2026, https://www.theafricareport.com/414003/ethiopia-the-pretoria-agreement-as-a-weapon-of-war-against-tigray/. ↩

15. Yonatan Fessha, Ethnic Diversity and Federalism: Constitution Making in South Africa and Ethiopia (Aldershot: Ashgate, 2010); Tsegaye Regassa, “Sub-National Constitutions in Ethiopia: Towards Entrenching Constitutionalism at State Level,” Mizan Law Review 3, no. 1 (2009): 33–69; Afrobarometer Round 9 (2023): 54% of Ethiopians prefer federalism to a unitary state; 67% support keeping the constitution with some amendments, https://50shadesoffederalism.com/case-studies/federalism-and-ethnic-accommodation-in-ethiopia-a-promised-solution-or-a-persistent-dilemma/. ↩

16. This is broadly the position of Andreas Eshete in his earlier work, and of the constitutional defenders associated with the EPRDF intellectual tradition. See Andreas Eshete, “Ethnic Federalism: New Frontiers in Ethiopian Politics,” in Ethnic Federalism: The Ethiopian Experience in Comparative Perspective, ed. David Turton (Oxford: James Currey, 2006). ↩

17. This argument is made in various forms by Donald N. Levine, Greater Ethiopia: The Evolution of a Multiethnic Society, 2nd ed. (Chicago: University of Chicago Press, 2000), and by some contemporary defenders of the Ethiopian Orthodox traditionalist position. The classic statement of the Fetha Nagast tradition is Abba Paulos Tzadua, trans., The Fetha Nagast: The Law of the Kings (Addis Ababa: Haile Selassie I University, 1968). ↩