The 1955 Revised Constitution: American Advisors, Bicameral Parliament, Imperial Sovereignty

Thesis

The 1955 Revised Constitution is the most carefully drafted constitutional document in Ethiopian history. It was prepared over four years by a constitutional commission that included American constitutional lawyers, Ethiopian jurists trained in European and American law schools, and senior officials of the imperial government, working under the personal supervision of John H. Spencer — the American advisor who had served as Haile Selassie’s chief legal counsellor since 1936 — and Ato Aklilu Habtewold, who became foreign minister and later prime minister.¹ It contained a justiciable bill of rights, a bicameral parliament with an elected lower chamber, an independent judiciary on paper, and the formal machinery of a constitutional monarchy. It was promulgated on 4 November 1955, the silver jubilee of Haile Selassie’s coronation, in deliberate ceremonial association with the emperor’s twenty-five years of rule.² It is the high-water mark of pre-revolutionary Ethiopian constitutionalism in textual ambition.

It is also the clearest demonstration in Ethiopian constitutional history that careful drafting, expert assistance, and elaborate textual provision do not, by themselves, produce constitutionalism. The 1955 Constitution contained the form of every protection a modern constitutional document of its era was expected to contain, and in nineteen years of operation it constrained imperial decision-making on no consequential question. Its suspension by Proclamation No. 1 of 1974 was effected by a colonels’ communiqué without resistance from any constitutional institution because no constitutional institution had ever, in nineteen years, established itself as a centre of independent authority capable of resisting.³ The text was real. The constitutionalism was not.

The drafting

The drafting history matters because it answers, in advance, the obvious objection — that the failure of the 1955 Constitution was a failure of drafting that better drafting could have cured. Better drafting was attempted. The constitutional commission worked from 1951 to 1955, consulted American, British, French, and Indian constitutional precedents, and produced multiple drafts. John H. Spencer, the principal American advisor, was an experienced constitutional lawyer who had worked on Ethiopian legal matters since the 1930s and who later wrote a detailed account of the process.⁴ Aklilu Habtewold and other senior Ethiopian officials brought practical political experience. The text was reviewed by the emperor personally and revised in his presence.

The drafters knew what a working constitution required. They knew about judicial review (the United States had it; India had just adopted it; the Federal Republic of Germany had just adopted it in stronger form). They knew about parliamentary supremacy in lawmaking (the Westminster model was familiar). They knew about bills of rights as judicially enforceable constraints on executive power. They could have written any of these into the 1955 Constitution. The fact that they did not write them in operationally effective form was not an oversight. It was a constraint imposed by the only client that mattered: Haile Selassie himself, whose interest in the constitution was its capacity to demonstrate Ethiopian modernity to foreign audiences and to manage the federation with Eritrea, not its capacity to constrain his own decisions.⁵

The text

The 1955 Constitution contained 131 articles in seven chapters, considerably more elaborate than the 1931 instrument. Its key features:

Sovereignty and the emperor. Article 1 declared the empire indivisible. Article 2 declared the emperor’s person sacred and his dignity inviolable. Article 26 vested supreme authority over all the affairs of the empire in the emperor. Articles 27 through 36 enumerated his prerogatives: command of the armed forces, declaration of war, conclusion of treaties, appointment of all senior officials, dissolution of parliament, grant of pardons, conferment of titles and honours, sanctioning of legislation.⁶ In short, the entire framework of imperial absolutism was preserved.

Parliament. Articles 76 through 100 established a bicameral parliament. The Chamber of Senators (76 members) was appointed by the emperor. The Chamber of Deputies (210 members) was directly elected by Ethiopian citizens — a major change from 1931, when even the deputies had been selected by appointed notables.⁷ Direct election of the lower chamber was a real institutional innovation. The constitution required elections every four years and provided for universal suffrage of Ethiopian citizens over 21.

Rights. Articles 37 through 65 contained an extensive enumeration of civil rights: equality before the law, freedom from arbitrary arrest, freedom of religion, freedom of speech and press, freedom of assembly and association, the right to property, the right to fair trial.⁸ This was the bill of rights the 1931 instrument had lacked, drafted in language drawn substantially from the Universal Declaration of Human Rights and from American constitutional precedent.

Judiciary. Articles 109 through 122 established a judiciary “independent” of the executive and legislative branches, with the Supreme Imperial Court at its apex. Judges were to be appointed by the emperor and to serve during good behaviour.⁹

That is what the text said. The crucial questions are what each of these provisions did in operation and why each failed to operate as written.

What it did in operation

Parliament. Direct elections to the Chamber of Deputies were held in 1957, 1961, 1965, 1969, and 1973. Turnout was modest, candidate selection was heavily managed by local notables and the imperial bureaucracy, and the chamber’s legislative output was almost entirely on matters the emperor or his cabinet referred to it.¹⁰ Genuinely consequential legislation — the agricultural reforms that became central to the late-imperial crisis, the land reform proposals that did not happen, foreign-policy decisions, military expenditures, the management of the Eritrean federation — moved through the imperial cabinet and the emperor personally, not through parliament. The chamber’s most significant act of independent assertion came in early 1974, when it briefly resisted Endalkachew Makonnen’s cabinet during the unrest that preceded the revolution. By then it was too late: within months the constitution was suspended and the parliament dissolved.¹¹

Rights. The enumerated rights in Articles 37–65 were never the subject of significant judicial enforcement against the imperial government. Ethiopian courts did not develop a body of constitutional rights jurisprudence comparable to American, Indian, or German jurisprudence in the same period. Two reasons: there was no provision for direct constitutional review by courts, and the Supreme Imperial Court was constituted by the emperor’s appointees who had not been selected for independence. The student protests of the 1960s, the workers’ demonstrations of the late 1960s, the famine reports suppressed in 1972–73 — none of these produced cases in which constitutional rights were vindicated against the state.¹²

Judiciary. The judiciary of the 1955 Constitution era was professionally competent in civil and commercial matters and produced a real body of Ethiopian civil law, much of it drafted under the influence of the French legal scholar René David, who prepared the 1960 Civil Code.¹³ What it did not do, because the constitutional framework did not require or permit it to do, was develop a tradition of judicial review of executive action on constitutional grounds. Ethiopian judges did not strike down imperial decrees. They could not have done so without losing their commissions; the structural incentives for judicial independence did not exist.

The Eritrean federation

The 1955 Constitution operated in an Ethiopia that was, between 1952 and 1962, federally united with Eritrea under a UN-supervised arrangement.¹⁴ Eritrea had its own constitution, its own elected assembly, its own working languages (Tigrinya and Arabic), and its own administrative structures. The two-tier system was a real federal arrangement.

It was also dismantled by stages. The Eritrean assembly was pressured into reducing its own competences in 1955 and again in 1956. The use of Tigrinya and Arabic in Eritrean administration was curtailed in favour of Amharic. The Eritrean flag was abolished in 1959. In 1962, the federation was formally dissolved by an Eritrean assembly vote held under conditions that the UN’s own subsequent assessment characterised as coercive, and Eritrea was incorporated as the fourteenth province of Ethiopia.¹⁵ This sequence — the dismantling of a federal arrangement by the centralizing actions of an executive while the constitutional structure that nominally protected the federation looked on — would be repeated, in different form, in the 1995 federation’s post-2018 trajectory. The pattern of executive consolidation against federal structure is not new to Ethiopian constitutional history.

Why the form did not produce substance

Three structural features of the 1955 Constitution explain why its impressive textual provisions did not produce binding constraint.

First, there was no machinery of constitutional review. The constitution did not provide for a constitutional court, did not provide for judicial review of legislation, and did not provide a mechanism by which a citizen could challenge an executive act as unconstitutional. Article 122 vested some interpretive authority in the Supreme Imperial Court, but the court did not develop the practice of constitutional review and was not, by its composition, capable of doing so against the executive.¹⁶

Second, the parliament’s legislative power was conditional on imperial sanction (Article 88). All legislation required the emperor’s signature; the emperor could refuse to sign; and the constitution provided no override. Effective lawmaking authority therefore remained where it had been in 1931.

Third, the political conditions for civil-society demand for constitutional enforcement were absent. The constitution did not permit organised political parties outside imperial sanction. The press operated under licensing arrangements that excluded substantial criticism of the imperial government. The university — Haile Selassie I University, founded in 1961 — eventually became a centre of opposition, but its students were too few and their organisational reach too limited to constitute a constitutional counterweight to the imperial state.¹⁷ The constituency that would have demanded enforcement of the 1955 Constitution against the emperor did not exist in 1955 and was only beginning to exist by 1974, by which time the constitution itself was about to be suspended.

Steelman of the opposing reading

The defenders of the 1955 Constitution make three serious arguments that this article must engage.

The first is that the 1955 Constitution was a genuine step toward constitutional government, and that its failure to produce binding constraint reflects the limits of what was politically possible in 1955, not a flaw in its drafting or intent. This is partly right. The 1955 Constitution was more constraining than the 1931 instrument. It did provide direct election of the lower house. It did contain a bill of rights. To dismiss it as merely cosmetic is unfair to its drafters, several of whom — including the Ethiopian members of the constitutional commission — believed they were building toward a working constitutional monarchy and would have continued to do so had the revolution not interrupted them. The reply is that “would have continued” is the operative phrase. Nineteen years is a long time. The constitutional commission’s apparent intent to continue building was not matched by actual building. The institutions that would have made the text constraining were not constructed.

The second is that the 1955 Constitution operated in an extraordinarily difficult context — the federation with Eritrea, the Cold War, the famine of 1972–73, the student movement, the army’s growing politicization — and that no constitutional document could have survived intact under those pressures. This is also partly right, but it points in the wrong direction. The pressures of the late 1960s and early 1970s were exactly the kind of pressures a working constitution is supposed to mediate. India absorbed the 1971 war, the Bangladesh refugee crisis, and the emergency of 1975–77 and emerged with its constitutional structure damaged but intact. The 1955 Constitution did not survive its first major test (the 1974 unrest) because it had never been institutionally tested before. The institutions that would have absorbed shock did not exist to absorb it.

The third is that the 1955 Constitution did establish principles — written law, parliamentary representation, enumerated rights — that have continued to shape Ethiopian constitutional discourse, including the 1995 Constitution’s much more elaborate human-rights chapter.¹⁸ This is correct and important. The 1955 Constitution is not nothing. It established Ethiopian constitutional vocabulary, trained a first generation of Ethiopian constitutional lawyers, and produced a textual model that subsequent drafters could build on. The criticism is not that the 1955 Constitution accomplished nothing. The criticism is that what it accomplished was textual rather than institutional, and that the institutional gap it left has not been filled by any subsequent regime.

Multi-causal accounting

The failure of the 1955 Constitution to produce binding constraint reflects (a) the deliberate choice of the emperor and his immediate advisors to retain unrestricted prerogatives, (b) the absence of social forces with sufficient organization to demand enforcement, (c) the international context in which formal constitutional modernity was rewarded more than substantive constraint, (d) the civil-law tradition Ethiopia inherited via the French legal advisors, which structurally produces less judicial creativity in constitutional interpretation than common-law systems, and (e) the historical absence of any indigenous tradition of structural constraint on the sovereign that the constitutional drafters could have built on. None of these factors excuses the failure; together they explain why the failure ran as deep as it did.

What the article establishes

The 1955 Constitution is the most carefully drafted Ethiopian constitution before 1995, and possibly after. Its failure to produce binding constitutional constraint is therefore the most informative case in the Ethiopian record. Better drafting was tried. American advisors were employed. Extensive textual provision was made. Direct elections were held five times. None of this produced the rule of law in the Hayekian sense, the separation of powers in the Madisonian sense, or the inclusive institutions in the Acemoglu–Robinson sense. The conclusion is uncomfortable but unavoidable: the missing element in Ethiopian constitutionalism is not drafting expertise. It is the political-economy configuration that would make the drafted text binding on the sovereign. That configuration was not present in 1955 and has not been present since.

The 1955 Constitution was suspended on the morning of 12 September 1974. The next article takes up what replaced it: the constitutional suspension itself as a founding act of a new regime, and the decade of rule by decree that followed.

---

References

Footnotes

1. John H. Spencer, Ethiopia at Bay: A Personal Account of the Haile Sellassie Years (Algonac, MI: Reference Publications, 1984), ch. 18; Christopher Clapham, Haile-Selassie’s Government (London: Longmans, 1969), 24–48. ↩

2. Constitution of Ethiopia (Revised), 4 November 1955, Negarit Gazeta, Proclamation No. 149/1955. ↩

3. “Proclamation No. 1 of 1974,” Negarit Gazeta, 12 September 1974. ↩

4. Spencer, Ethiopia at Bay, 273–285 (drafting history). ↩

5. On the emperor’s instrumental purposes for the 1955 Constitution, see Bahru Zewde, A History of Modern Ethiopia, 1855–1991, 2nd ed. (Oxford: James Currey, 2001), 200–203, and Marcus, A History of Ethiopia, 162–166. ↩

6. 1955 Constitution, articles as cited; see Clapham, Haile-Selassie’s Government, 28–48, for structural summary. ↩

7. Articles 76–100 of the 1955 Constitution; Bahru Zewde, A History of Modern Ethiopia, 201–203. ↩

8. Articles 37–65; see Spencer, Ethiopia at Bay, 280–283, on the bill of rights. ↩

9. Articles 109–122; on the judicial structure in operation, see Aberra Jembere, An Introduction to the Legal History of Ethiopia 1434–1974 (Münster: LIT Verlag, 2000). ↩

10. On the operation of the imperial parliament, see Clapham, Haile-Selassie’s Government, 49–93, and Markakis, Ethiopia: Anatomy of a Traditional Polity, 251–270. ↩

11. Andargachew Tiruneh, The Ethiopian Revolution 1974–1987, 45–55. ↩

12. On the failure of the imperial judiciary to develop constitutional rights jurisprudence, see Aberra Jembere, Introduction to the Legal History, and Bahru Zewde, A History of Modern Ethiopia, 220–225. ↩

13. René David, Le droit de l’Éthiopie (Paris: Dalloz, 1962), on the drafting of the 1960 Ethiopian Civil Code. See also Norman J. Singer, “The Ethiopian Civil Code and the Recognition of Customary Law,” Houston Law Review 9 (1971): 460–93. ↩

14. UN General Assembly Resolution 390 (V), 2 December 1950; Tekeste Negash, Eritrea and Ethiopia: The Federal Experience (Uppsala: Nordic Africa Institute, 1997). ↩

15. Tekeste Negash, Eritrea and Ethiopia, chs. 4–6; Ruth Iyob, The Eritrean Struggle for Independence (Cambridge: Cambridge University Press, 1995), 82–96. ↩

16. On the absence of constitutional review in the 1955 system, see Getachew Assefa, “All About Words: Discovering the Intention of the Makers of the Ethiopian Constitution on the Scope and Meaning of Constitutional Interpretation,” Journal of Ethiopian Law 24, no. 2 (2010). ↩

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

18. This is the position taken by some scholars sympathetic to Ethiopian constitutional continuity, e.g. Tsegaye Regassa, “The Making and Legitimacy of the Ethiopian Constitution,” Afrika Focus 23, no. 1 (2010). ↩