The Anglo-Ethiopian Society

Nationalism, Nationality and Regionalism in Ethiopia

C. Clapham

published 1995

Professor Christopher Clapham whose studies of Ethiopian politics began in the early 1960's was the author of 'Haile Selassie's Government'. More recently - in 1988 - his much acclaimed work 'Revolutionary Ethiopia' was published by the Cambridge University Press.

A former Visiting Professor at the Addis Ababa University, he is Professor of Politics and International Relations at the University of Lancaster.

This Occasional Paper is closely based on the John Hack Memorial Lecture which he gave to the Society in 1995.

In recent years, one issue in particular has come to preoccupy, not only those like myself who try to follow current Ethiopian affairs, but also the present Ethiopian Government and Ethiopians of every political persuasion. This is what is commonly described in Ethiopia as the question of 'nationalities', or in other words the relationships between the different peoples or ethnic groups - Afars, Amharas, Gurages, Oromos, Sidamas, Somalis, Tigrayans, Wollaytas and many more - who live within the frontiers of Ethiopia, and Ethiopia itself, viewed not simply as a geographical expression, but as a nation which has over a long period developed its own distinctive culture, identity and history.

This issue has been brought to the centre of Ethiopian politics, since the present Ethiopian Peoples Revolutionary Democratic Front regime came to power in 1991, through the division of the country into regions defined by the principal nationalities which inhabit them. The new Constitution of the Federal Democratic Republic of Ethiopia, introduced in 1995, goes so far as to guarantee to any 'nation, nationality or people' in Ethiopia the right to self-determination, up to and including secession. Any group capable of meeting the requirements laid down in that Constitution thus has the right to constitute itself as a sovereign and independent state. The problem of nationalities has not, however, simply been invented by the present government. Though muted or even suppressed under the imperial regime up to 1974, it became increasingly prominent from the start of the 1974 revolution, and was recognised by the government of Mengistu Haile Mariam, which established an Institute for the Study of Ethiopian Nationalities, the recommendations of which were in turn reflected in the Constitution of the short-lived People's Democratic Republic of Ethiopia (PDRE) established in 1987. Issues of nationality must equally be accorded a prominent place among the reasons for the collapse of the PDRE in 1991.

In this talk, I shall attempt to place these issues in a perspective drawn from Ethiopian history, and in the process try to unravel the pressures both for unity and for division in Ethiopia, and to show how the country has reached its present situation. This inevitably means that I shall have to venture into some sensitive areas, about which Ethiopians of different political persuasions are - quite rightly - intensely concerned. It is certainly not my intention to take sides in any of the resulting disputes, and I can only apologise to those Ethiopians who may feel that I have gone seriously astray. The subject is nonetheless so central to present-day Ethiopia that no-one who seeks to understand either its current problems or its recent history can afford to ignore it.

The place to start must be the fact that, over a recorded history that stretches back over many centuries, the Ethiopian state has never been restricted to any single nationality or people. It has been a multiethnic state, if not from the start, then at least for a very long time. It would indeed have been difficult if not impossible for Ethiopia to maintain one of the most enduring and effective political structures in Africa (and perhaps even in the world), without a capacity to incorporate peoples of many different nationalities. These have included not only Amharas and Tigrayans, the two Christian highland peoples with whom the history of the Ethiopian state has been most closely associated, but also Agaws, Gurages and many more. Oromos, for example, were playing an important role in the imperial government more than three hundred years ago.

This characteristic Ethiopian ability to incorporate different peoples into political life was encouraged, not just by intermarriage and a high level of social mobility, but also by one feature of the social structure of highland Ethiopia that renders it unique among the peoples of sub-Saharan Africa. Almost throughout Africa, people trace their primary social identity through either their father's or their mother's line, thus creating social identities based on descent, which have commonly (though the term is now often frowned on) been described as 'tribal'. Northern highland Ethiopia, by contrast, was not a 'tribal' society in this sense. Instead, both social status and rights in land could be derived from both one's father and one's mother. The effect was to blur ethnic identities and to facilitate integration, since the child of, say, a Gurage father and Amhara mother could very readily 'become' an Amhara. This process is evident in the ancestry of the most renowned Ethiopian of modern times, Emperor Haile-Selassie. He derived his Shoan royal blood and hence his claim to the throne from the mother of his father, Ras Makonnen. His father's father, on the other hand, was an Oromo, as also was his mother's father; his mother's mother was a Gurage. Had highland Ethiopian society been governed by the principles of patrineal descent which apply for example to the Somali, he would have been classed as an Oromo; had the matrilineal principle adopted by the Akan peoples of southern Ghana applied, he would have become a Gurage. As it was, he may be regarded as a classic 'Ethiopian'.

That this disregard for the ethnic origins even of the most prominent Ethiopians was nothing personal to Haile-Selassie can be demonstrated by identifying the 'nationality' of the country's rulers over a period of nearly a century and a half. Only the first of these, Tewodros, can be regarded as a reasonably 'pure-blooded' Amhara. His successor Yohannes was Tigrayan, while Menilek and his daughter Zawditu were of mixed ancestry. Lij Yasu was the son of the Oromo Negus Mikael, while since the revolution, the central government has been led in turn by the Eritrean Aman Andom, the Oromo Teferi Benti, the Kullo Konta (by most reports) Mengistu Haile-Mariam, the Tigrayan Meles Zenawi, and most recently the Oromo Negasso Gidada. Ethiopia did indeed have a politically dominant culture, and this was reflected in the dominant role of the Amharic language and of the Ethiopian Orthodox Church; and this in turn certainly acted as a barrier to effective political participation by those of other faiths, and notably Moslems. It could well however be argued that any viable form of government must be based on some uniting elements that its people hold in common, and that what matters is that people should be able to acquire those elements, on terms that do not evidently disadvantage them. Though people of many different national origins have been prominent, for example, in the government of the United States, it is hard to imagine that they could have reached that prominence without a thorough command of English, while the presidency has as yet been restricted to Christians.

If, then, one had looked at Ethiopia some thirty years ago, one would have been struck (as indeed I was) by the absence of ethnic elements in national politics, especially by contrast with those numerous states in formerly colonial Africa (such as Nigeria or Uganda) where ethnicity had emerged as the major force in defining political identities. One could even discern a sense of Ethiopian identity, to a large extent transcending the personal origins of particular individuals, and might plausibly have supposed that a unifying Ethiopian nationalism was in the making. Looking back from 1995, however, it is all too clear that this movement towards Ethiopian nationalism has not been sustained. With the benefit of hindsight (which I, among others, certainly lacked at the time), it is not difficult to discern the origins of the 'national question' behind the facade of relative tranquillity presented by the Haile-Selassie regime. These origins lay in three major problems which first imperial and then revolutionary Ethiopia failed to surmount, and which in turn laid the basis for the recent mobilisation of 'nationality' as a key constituent of Ethiopian political life.

The first of these problems was that of centralisation: the progressive squeezing of local autonomy and initiative by increasingly authoritarian central governments, which subjected the diverse regions and peoples of Ethiopia to a level of control from Addis Ababa which eventually prompted dissent and then revolt. In historic Ethiopia, this control had been rendered impossible by sheer physical separation. Especially during the rainy seasons, but to a large extent at all times, the central government was too distant, and the military and technological resources at its disposal were too slight, for it to be enforced. Despite the unlimited power which emperors enjoyed in principle, they always had to recognise its practical limitations, and to govern in large part through local officials who retained considerable autonomy. In effect, then, Ethiopia was until recent times governed through a federal system, with a constant tension between the national government represented by the emperor and his court, and regional governors who to a large extent ruled as they saw fit. In the mid-1920s, when Dejazmatch Balcha (a Gurage, as it happened) ruled Sidamo, the future Haile-Selassie (then of course Ras Tafari) sought to recall him to Addis Ababa. Messenger after messenger was sent to Yirgalem, until at last the old Dejazmatch returned his reply: "If Tafari wants me, let him come down here into Sidamo, and get me".

Though Tafari eventually got his way, as he usually did, by guile rather than force, it was not until after the Italian occupation of 1936-41 that the restored Emperor Haile-Selassie was able to carry centralisation to the point at which it was eventually in danger of becoming self-defeating; and even then, he recognised the need to respect a measure of local autonomy, especially in Tigray. On the other hand, he completely failed to recognise the need to allow Eritrea the level of self-government that could have reconciled Eritreans to the federation with Ethiopia. It was however left to the revolutionary government to test the idea of a centralised Ethiopian government to destruction, once Mengistu Haile-Mariam had replaced the initial movement towards a measure of regional devolution in the early months of the revolution, by his own ruthless brand of Jacobin nationalism. It is still far too soon to attempt any balanced appraisal of Mengistu's role in Ethiopian history, and some credit him with commitment to a vision of Ethiopia as a modern and united nation-state that would at least place him in a category above that of purely personal dictators such as Idi Amin or Samuel Doe. It is nonetheless clear that once this vision had failed, at an enormous cost in death and suffering to a very large number of people throughout Ethiopia, the possibility of maintaining any powerful centralised system of government had been destroyed beyond the possibility of immediate restoration.

The second major problem which neither the imperial nor, eventually, the revolutionary government was able to resolve was that of exploitation. There were inbuilt inequalities in the relationship between the central government and many of its peripheral peoples which placed enormous obstacles in the way of their incorporation into a single Ethiopian nationhood. Many of these problems go back to the allocation of land in the south and west of the country to settlers from Shoa and other northern regions under Menilek's neftenya system in the late nineteenth century. This in turn provided a mechanism for the exploitation of the peoples who produced Ethiopia's most important marketable crops, especially coffee, for the benefit of the central government and those who had influence with it. In addition, the Moslem peoples of Ethiopia were always largely excluded from state power, unless like Negus Mikael they converted to Christianity - an option which might be open to a few leading individuals, but which whole communities were unlikely to accept.

In its early years, the revolutionary government after 1974 sought to resolve this problem through land reform, a policy which was initially highly successful in reconciling those regions which had been forcibly incorporated into Ethiopia in the late nineteenth century to the new regime. One of the most striking facts about Mengistu's Ethiopia was that the major opposition movements did not arise in the 'new' Ethiopian regions of the south and west, from which indeed Mengistu himself originated, but rather from northern and Christian regions, notably Tigray, which had been associated with the Ethiopian state since its formation. Nonetheless, the attempt at reconciliation eventually failed. This was partly because of resentments built up over previous generations; but it was also because the exaction of resources by the revolutionary government, both in the quest to build a modern, socialist and centrally planned economy, and still more because of the costs of its endless wars, eventually outweighed the benefits of land reform. The imposition of the villagisation programme in the mid 1980s probably marked the definitive turning point in the alienation of southern Ethiopia from the Mengistu regime.

The third problem which prevented the creation of a single Ethiopian nationalism was that of participation. The basic difficulty here was just the same as in the former Soviet Union or Yugoslavia: it is much easier to keep a union of different peoples together when not too many of them need to take part in it. You can easily recruit a few selected members of any given ethnic group or nationality into the government, and use them either as token representatives, or at most as intermediaries between the central government and their own peoples. Once peoples as a whole demand a stake in government, however, they are likely to emphasise their own separate identities, creating in the process an ethnic or tribal politics which is very difficult to control. Where, as in Ethiopia (or indeed the Soviet Union), mass participation in politics was forcibly suppressed until it was beyond the central government's ability to suppress it, the resulting upsurge in ethnic identities is liable to be especially difficult to manage.

Even if the attempted imposition of increasingly centralised control was always eventually going to challenge the survival, first of the Haile-Selassie and then of the Mengistu regimes, it did not necessarily follow that the opposition to these regimes had to rely on mobilising the potentially explosive power of nationality. The great exception was in Eritrea, where the victorious Eritrean People's Liberation Front (EPLF) was strongly opposed to using either ethnicity or religion as a basis for fomenting opposition to the government in Addis Ababa, and did its best instead to create a multi-ethnic and multi-confessional sense of Eritrean nationhood. There were however reasons why this option was an attractive one in Eritrea, but was much more difficult to follow elsewhere. First, Eritrea is, in both ethnic and religious terms, a highly diverse territory, and any effective resistance movement had to reflect this fact. In particular, the emphasis placed by the first major resistance movement, the Eritrean Liberation Front (ELF), on the supposedly Moslem and even Arabic identity of Eritrea was inherently divisive in a territory whose people were almost evenly distributed between Islam and Christianity. Secondly, Eritrean resistance to the central Ethiopian government could always draw on the special status which Eritrea had possessed as a former Italian colony, and make a case for sovereign independent statehood. This in turn made Eritrea very much like other African nationalist movements, which sought independence within the boundaries established by the colonial partition, and which therefore had to do everything they could to bridge the divisions between the different ethnic sections of their own populations. It was, paradoxically, the very artificiality of Eritrea that imposed on the EPLF the urgent need to maintain a sense of national unity.

For the Tigray People's Liberation Front (TPLF) in Tigray, which led the opposition to the Mengistu regime in the rest of Ethiopia, and which now forms the core of the present Ethiopian People's Revolutionary Democratic Front (EPRDF) government, these factors worked the other way round. Indeed, they positively encouraged the emphasis on ethnic identities. For a start, the region within which the TPLF launched its long and ultimately successful struggle against the government in Addis Ababa had a distinctive identity of its own, which provided an obvious rallying point to encourage resistance. Its separate language in particular, coupled with its resentment of Shoan domination and its long and distinguished history, marked Tigray as a region apart. Furthermore, there was no plausible prospect of achieving separate independence, of the kind that beckoned to Eritrea: Tigray had formed part of Ethiopia for as long as any recognisable Ethiopian state had existed, and the most that the TPLF could hope for was a special status within Ethiopia. It also had to build alliances with other groups within the country, if it was ever going to win its war against the central government, and it could do this most effectively by encouraging the mobilisation of ethnic identities among these peoples too, and by promising them their own autonomous regions within a decentralised Ethiopia.

For Tigrayans, moreover, much more than for other nationalities within Ethiopia, there was no inherent conflict between their separate regional identity, which with their language they had maintained over a very long period, and their national identity as Ethiopians, which they had likewise sustained since the earliest times. It was easy for them to assume that this balance between regional and national loyalties could also be maintained by other Ethiopian peoples. In fact, the relationship of Tigray to the Ethiopian state, as part of it yet separate from it, was unique; and the TPLF's answer to the problem of nationalities, which seemed so obvious in Tigray, was much harder to implement elsewhere. On the one hand, for many Amhara, and especially those of Shoa, their own ethnic identity was virtually coterminous with an Ethiopian national one; the idea of distinguishing between being 'Amhara' on the one hand, and being 'Ethiopian' on the other, was virtually meaningless. For many of the other peoples of Ethiopia, and especially those incorporated through conquest in the later nineteenth century, their own historically suppressed identities as separate nationalities stood in opposition to the idea of being 'Ethiopian' at all. The mobilisation of their ethnic identities potentially posed a threat to the survival of Ethiopia, in the same way that Kazakh or Armenian nationalism posed a threat to the survival of the Soviet Union.

When the TPLF seized control of the central government with the capture of Addis Ababa in May 1991, the system that they established - based on their own idea of the proper relationship between the Ethiopian state and Tigray - thus seemed logical to them but was very strange to everyone else. In effect, people who wanted to take part in politics were permitted to be everything except Ethiopian. Or rather, they could only participate in politics as hyphenated Ethiopians - as Gurage-Ethiopians, Wallayta-Ethiopians, Oromo-Ethiopians, or whatever. Virtually all of the members of the transitional assembly established by the new regime were designated as the representatives of different nationalities, though whether they actually represented the people of their areas of origin in any meaningful sense was, in the absence of elections, uncertain. Any attempt to represent the identity of Ethiopia as a whole was greeted with intense suspicion by the new regime, which readily associated any idea of Ethiopian nationalism with Amhara supremacy and the abuses inflicted by the Mengistu government. Institutions which identified themselves as Ethiopian, such as Addis Ababa University, were targets for this suspicion.

To most of the rest of Africa, likewise, this policy seemed bizarre. Struggling to create viable states and nations within the generally artificial territories bequeathed to them by colonialism, the governments of other African states have been especially anxious to discourage what they saw as 'tribalism'. That such tribalism should, as they saw it, be positively fostered within Africa's most historic state was not only incomprehensible in itself, but filled with dangers for the rest of Africa. This view was shared by the new government in Eritrea, which despite its alliance with the TPLF during most of the struggle against the Mengistu regime, was as already noted much closer in its attitude to ethnicity to other post-colonial African states than to the EPRDF. The representative from Addis Ababa who recommended an Ethiopian solution to the national question to the Eritrean constitutional conference in January 1995 was listened to with barely concealed incredulity.

Nonetheless, just four and a half years after the TPLF triumph, the major elements of the new structure have been put in place. Ethiopia has been divided into regions defined by nationality, and the new constitution of the Federal Democratic Republic of Ethiopia (FDRE), based essentially on the principle of a confederation of nationalities, has been put in place. The immediate future of Ethiopia depends on how this system works.

It must be acknowledged, firstly, that this system represents a unique and subtle attempt to restructure the central-regional balance, combining the maintenance of a single Ethiopia on the one hand, with the recognition of the rights and identities of its constituent peoples on the other. In some degree, indeed, it may be regarded as an attempt to restore a modern equivalent to the federal system that Ethiopia once possessed, before the centralising ethos of the twentieth century, backed by firearms, imposed an ultimately unsustainable central autocracy.

This is nonetheless a highly problematic venture, and is subject to very considerable dangers. For one thing, federal systems of all kinds are difficult and expensive to run. They create bureaucratic duplication, of a kind that a very poor state such as Ethiopia can ill afford, and they involve an in-built tension between central and regional governments. In order to work, they depend on bargaining and compromise, a willingness to give and take, which have not been prominent virtues in recent Ethiopian politics.

Ethnic federal systems are particularly hard to run, since the logic of ethnicity is inherently divisive. Politicians entrenched within their own ethnic regions are under constant pressure to emphasise the distinctiveness and separateness of their own nationality, rather than the interests that the peoples of the whole country have in common. The emphasis on ethnicity as the basis for citizenship within each region readily leads to discrimination against those who are defined as outsiders, while representation at the central level characteristically turns into a numbers game, in which each ethnic region seeks to inflate its own population in order to claim a larger share of central benefits. For all these reasons, once a multi-ethnic political system has been split apart, it is very difficult to put it back together again.

The economics of federalism in Ethiopia is also sensitive, since the different regions have very different economic endowments. By far the greater part of Ethiopia's marketable produce, including not only coffee and other export crops such as chat, but also surplus food, comes from the Oromo and the combined Southern regions, while industry is heavily concentrated in and around Addis Ababa. The major food deficit regions are the Amhara and Tigrayan ones, though there are some Amhara areas of marketable production, especially in Gojjam. Paradoxically, given its central role in the new political order, Tigray is the most dependent region in the whole of Ethiopia. It cannot feed its population, and produces very little that can be sold either abroad or in other parts of the country. This in turn increases its reliance on external aid, and hence on its influence on the central government through which that aid is mostly channelled.

Historically, the economic discrepancies between different parts of Ethiopia, which are of very long standing, were dealt with through the political domination of some regions over others. One such mechanism was migration, normally from the north of the country towards the south and west, which helped to compensate for the over-population and ecological degradation of the most famine-prone areas of the country, from the time of Menilek's neftenya through to that of Mengistu's resettlement policy. That outlet has now been effectively excluded by the new regional system. Another mechanism was discriminatory taxation and redistribution, especially by using revenues generated by coffee exports in order to maintain a central government in which Shoans were disproportionately represented. This will also be much harder to manage within the new federal system, in which regional governments will tend to demand the retention of locally generated revenues.

Maintaining the kind of balancing act that the new federal system required is therefore bound to be an extremely delicate operation, and we now have some indication of the means by which the EPRDF government is attempting to achieve it. In essence, this involves a recognition of ethnic regionalism on the one hand, combined with a continued commitment to central control on the other. Regardless of the formal recognition of ethnic self-determination in the new constitution, federalism is only being permitted in practice within strictly controlled limits.

These limits do allow for an appreciable level of regional self-government, which goes well beyond the merely cosmetic. For example, the great majority of government officials at regional level now appear to be drawn from the region itself, and the old system in which local government was visibly manned by outsiders (and especially Shoans) is rapidly disappearing. The considerable latitude allowed to regional governments is most strikingly illustrated by the use of Oromifa, written in the Roman rather than the Ethiopic script, in the new Oromo region. The representation of nationalities also extends to the distribution of ministerial posts in the central government, which broadly corresponds to the respective populations of the different nationalities in the country as a whole.

On the other hand, the implicit rules of the game require all of these regional governments to be drawn from the EPRDF coalition. They do not extend to allowing opposition parties to function freely, let alone to the possibility that such parties might be permitted to take control of any regional government. I can see no good reason to dispute the truth of complaints, widely publicised by the Ethiopian Human Rights Council and by organisations outside the country, regarding the suppression of opposition movements (including those claiming to represent the Oromo, Amhara, Somali, Sidamo and other nationalities), or the killing, imprisonment or disappearance of individuals associated with them. A relatively high number of the most critical posts are also held by members of the TPLF.

Here, clearly, I am getting onto very sensitive political ground. While seeking so far as possible to distance myself from overt political commitment, to either one side or the other, I am concerned about what is actually likely to work, in the sense of providing an extremely complex and often violent country with the reasonably stable and peaceful government that all of its people so desperately need. In this context, I am reminded of an Akan proverb from southern Ghana, that power is like an egg: hold it too tightly and it crushes in your hand; hold it too loosely and it falls and breaks on the ground. Mengistu's government held the egg too tightly, crushing Ethiopia in unending wars; the danger in ethnic federalism is that it may fall, subjecting Ethiopia to a fragmentation in which the common interests of its peoples will likewise be destroyed. Some balance between the two is needed.

During a visit to Ethiopia earlier this year, I found myself - to my astonishment - invited to meet Meles Zenawi, who was then President and is now Prime Minister. Quite regardless of any views that one may hold about either his government or his political opinions, Ato Meles possesses not only great intelligence and charm, but a directness that I have found in no other politician, whether in Africa or elsewhere. When I asked him whether this new federal system would work, he replied: "I don't know; but we have tried everything else, and that didn't work". And if the head of the Ethiopian government does not know, I hope that I may be forgiven for concluding that I do not know either.

The events of recent years, in Europe as well as Africa, have shown how powerful is the pressure of ethnicity; and if even the mighty Soviet Union split apart under that pressure, then we certainly cannot be confident about the ability of Ethiopia to withstand it. We can however say that the attempt to compress the diversity of Ethiopia by force into a single centralised state was doomed to failure; and the more intensely that attempt was made, the more intensely it was resisted, at a terrible cost to a very large number of Ethiopians.

But if the diversity of Ethiopia is an enduring fact, then so also - at least in my view - is its identity. It could not have survived through so many centuries, and through such great upheavals, unless there were powerful elements holding its different parts together, to set against those which split them apart. Even when apparently closest to collapse, as for example in the early sixteenth and nineteenth centuries, it has shown a remarkable capacity for survival, and I am sure that these elements too will continue and eventually reassert themselves, despite the current pressures for fragmentation. Whether it will prove possible to combine these conflicting tendencies, towards unity and towards diversity, into political arrangements that will enable all Ethiopians to live peacefully and productively together, I do not know - but I certainly hope so.

First published as an Occasional Paper in 1995

Opinions expressed in the articles are those of the authors and not necessarily the views of the Society.
Information is offered in good faith but the Society does not warrant the status or reliability of the information contained.

© The Anglo-Ethiopian Society and Contributors 2003 - 2017