Contents

I.  The Problem
II.  Test of Hypothesis
III.  Explanation of Findings
A.  Concepts with Variable Meanings
1.  “Federalism”
2.  “Ethnic Problems”
B.  Additional Variables too Numerous for the Cases
C.  Cases and Causes which Vary Over Time
IV.  Implications of Findings
A.  What is the Genus?
1.  Of the Species Federalism
a. Identities are assumed by individuals when it is in their self-interest
b.  Individuals have multiple identities
c. Individuals may change the type, intensity and rank order of their identities
d. Identity groups aggregate a variety of traits which link individuals
e. Political leaders use identity groups to extend the number of their followers
f. To survive political units must create an identity group of their subjects
B.  Are the Genera Tools?
1.  Distribution of power
2.  Identity groups
V.  Conclusion

I.  The Problem

 In a seminal work, S. Rufus Davis argued that there was no causal relationship between federalism and anything else:

The truth of the matter is--and experience has been the teacher--that some ‘federal’ systems fail, some do not;...some inhibit economic growth, some do not;...some promote a great measure of civil liberty, some do not; some are highly adaptive, some are not...whatever their condition at any one time...it is rarely clear that it is so because of their federalness, or the particular character of their federal institutions, or the special way they practice federalism, or in spite of their federalness. [i]

If Davis is right, then federalism may be associated in some cases with a rise in the frequency and intensity of ethnic problems and in other cases with a decline in the frequency and intensity of such problems.  That is, no consistent relationship would exist between federalism and the rise or decline of ethnic problems.

The preponderance of scholarly work on this issue in Africa and elsewhere supports the Davis thesis, i.e., it suggests that federalism is not consistently related to the promotion or settlement of ethnic problems. [ii]  Yet, federalism continues to be viewed by some leaders of minority groups in Africa as a solution to, and by some leaders of majority groups as a cause of, such problems. Why would they advocate or oppose something which has not proved to consistently cause or solve ethnic problems?

Our purpose in this paper is three-fold:  First, to test briefly the Davis thesis that federalism does not consistently cause or solve ethnic problems in Africa;  second, to account for both the inconsistent impact of federalism and for the persistence of its advocates and detractors; and, third, to explore the implications of the test and its explanation for the building of general knowledge about the relationship between federalism and ethnic problems.

Back to Top

II.  Test of Hypothesis

The simplest way to determine the impact of federalism on ethnic problems is to examine what happened to such problems in federal systems.  Of course, a multiplicity of factors other than federalism affect ethnic problems.  Nevertheless, federalism can be said to cause an intensification or diminution of ethnic problems if a consistent relationship with one or the other recurs in most cases of federalism.

The most commonly identified cases of federalism in Africa are listed in Table 1:

Table 1: Post-Independence Federal Systems in Africa

Cameroon: The Cameroonian experience with federalism was not intended by its sponsors to be a means of overcoming ethnic problems, but rather a means of easing the integration of the southern part of the British Trust Territory and the French Trust Territory at the time of independence.   It’s demise appears little-related to ethnic problems.  According to Martin Dent, in 1972 “...President Ahidjou...unilaterally abolished the federal status to establish a closer union of a unitary kind.  The change was ‘legitimized’ in a plebiscite in which the Government did not hesitate to use its powers to produce a ‘Yes” vote.  Although West Cameroon has tolerated the unitary status there is little doubt that if they were given the choice they would gladly go back to federalism.” [iii]  Although such a preference may have existed, there is no clear evidence to show that an increase or decrease in ethnic problems resulted from the establishment or demise of the federation.

Congo (Zaire):   The early years of Congo (Kinshasa)’s existence involved a serious struggle between those who wanted a federal system and those who were opposed to it. The adoption of federalism, Crawford Young argued, was a reasonable response to the weakness of the central government for it placed power where it could be “responsibly exercised.” [iv]  Benyamin Neuberger saw it as an attempt to find “a solution to the problem of internal pluralism,” though he recognized that some advocates saw it as a step toward secession. [v]  Rather than solve ethnic problems, federalism became more of an adjustment to them.  When Mobutu seized power in 1965 and ended the early federal experiment an intensification of ethnic problems did not come as an immediate consequence.

Ethiopia: Federalism in Ethiopia normally refers to the arrangements during two periods: the decade of formal federation between Eritrea and Ethiopia (1952-1962) and the period since the adoption of the Charter in 1991.  The formal federation did not solve the Eritrean problem, nor did the abolition of the federation. The return of federalism in 1991, however partial its implementation, led to the grant of some autonomy to the 13 regions left after Eritrea’s independence. Like the early Congolese federalism, that in Ethiopia was more a recognition of reality than an attempt to cause or solve ethnic problems.

The form federalism took may have promoted the resistance of the Oromo Liberation Front (OLF), the most prominent organization claiming to speak for the Oromo.  Leaders of  that organization characterize the new federation as camouflage for the rule of Tigreans.  Federalism in Ethiopia has not brought an end to ethnic problems, though it may have assuaged the concerns of some of Ethiopia’s ethnic groups.

Kenya:  Kenya came to independence in 1963 with a federal “majimbo” constitution.  Benyamin Neuberger claims that the federalism was not adopted to solve ethnic problems, but as “the only way to get independence...and to prevent the secession of ...the Northern Frontier District (N.F.D.) of Kenya.” [vi]  Empowerment of the seven regional governments established by the constitution was delayed by the KANU government.  Within a year the main opposition party and supporter of the “majimbo” constitution, KADU, was essentially absorbed into KANU.  In a sense, the federal system may have contributed to a transition in which leaders of small ethnic groups were able to become a part of a patron-client system which worked to solve evolving ethnic problems.  Federalism, though, appears to have had little other impact.

Nigeria:  Although the Federal Republic of Nigeria has persisted since 1960, its “federalness” has undergone many changes.  Since the periods of its decline correspond with periods of military rule, changes in the nature of ethnic problems during such times may have little to do with federalism.  Yet, even during periods of civilian rule, federalism’s impact appears ambiguous.

A few incidents are indicative:  During the First Republic, the Northern Region’s dominance over the two and then three southern regions produced resentment against the Hausa/Fulani.  Political struggle was organized during that period around ethnically based political parties, partly as a result of the shape of the federal system. The 1966 killing of Igbo people in the North and the Biafran secession resulting in a great loss of life were not prevented by federalism. The creation of new states brought to the fore many grievances minority ethnic groups had suppressed under the domination of the three large ethnic groups whose domination was assured by the size of the initial federal units. The effort to introduce the notion of “federal character” as a means of overcoming an aspect of the ethnic problem appears to have done more to exacerbate than solve it. [vii] Roberta McKown made a similar observation, suggesting that “federalism has served only to provide a structure for the exacerbation of ethnic and regional conflict.  It may be that federalism is appropriate for moderate amounts of diversity, but not the extremes to be found in Africa.” [viii]

Donald Horowitz contends that federalism’s impact was not consistently harmful: “the Nigerian evidence shows that federalism can either exacerbate or mitigate ethnic conflict.  Much depends on the number of component states in a federation, their boundaries, and their ethnic composition.” [ix]  At the start of the Second Republic he effused:

The new Nigerian federal framework...utilized all five mechanisms of conflict reduction....  First, the proliferation of states /dispersed some of the conflict into more parochial forums.  Second, the new states provided arenas in which intraethnic conflict might also occur.  Third, a result of this was to enhance the position of some political parties at the expense of others, especially in the North, paving the way for greater interethnic cooperation in the all-Nigerian arena.  Fourth, as the new states fought to advance their interests, a few nonethnic issues and actors were also introduced.  And, fifth, the separate state bureaucracies provided career opportunities for groups not well represented in the federal civil service.  Using all these tools, the nineteen states readjusted, realigned, and complexified the Nigerian political system. [x]

The military overthrow of the Second Republic at the end of 1983 led Horowitz to reconsider what he thought was a success:  “As the ingenious federal engineering of the Nigerian second republic went down before a military coup the jury must remain out on the success or otherwise of democratic federalism in resolving Nigeria’s ethnic dilemmas.” [xi]

Senegal/Soudan:  The Mali Federation was so short-lived that it’s impact on ethnic problems was virtually nil.  In addition, the motivation for its founding was not the control of ethnicity or the desire to intensify it, but more the dream of pan-Africanism.

South Africa:  The implications of federalism embodied in South Africa’s new constitution for ethnic problems are gradually becoming apparent.  In order to gain the acceptance of a section of the Afrikaner opposition and Inkatha--and thereby foster ethnic harmony--the ANC agreed to give the nine provinces considerable powers.  Jenny Robinson argues that the constitution has “institutionalised and enhanced the political powers of certain ethnically based political groups and also set the stage for the emergence of new, powerful territorially-based groups.” [xii]  In other words, federalism may have helped solve an immediate problem, but it may give rise to ethnic problems in the future.

Sudan: A kind of federalism was introduced with the signing of the Addis Ababa Accord in March 1972 between representatives of the Sudanese government and of the South.  Martin Dent noted that “The result was a return of peace to the war ravaged South, and that peace lasted until President Numeiri, in his last despotic days of power, destroyed the autonomy he had himself created and civil war began again.” [xiii]  Although federalism may have diminished the North-South ethnic problem, Horowitz suggests that it produced other ethnic conflicts.  He observed, “Limited though Southern authority was, it was sufficient to set in motion a contest for control of the region that brought to the fore all the intra-Southern differences that had been manifested during the civil war....”[xiv]  And, these were intensified.  Thus, one could argue that federalism both diminished and increased ethnic problems in the Sudan.

Tanzania:  The federal arrangement established in March 1964 between Tanganyika, and Zanzibar was an asymmetrical one:  Zanzibar retained its own government while that of Tanganyika was merged into the union government of Tanzania and the population of Zanzibar was less than 5% of the population of Tanzania. The arrangement was not established to deal with ethnic problems and its impact on them has been complex. The federation has both fostered and restrained mainland-island conflict.  Politicians on the islands have used Zanzibari identity as opposed to Tanzanian identity to rally voters.  With the introduction of a multi-party system, some politicians on the mainland have done the reverse.  Consequent problems are restrained by the fact that from 1977 to the present the same political party, CCM, dominated both the mainland and the islands.  That fact has fostered animosity against the mainland on the part of the major Zanzibar opposition party, CUF [xv].  Thus, there is no consistent relationship between federalism and an increase or decrease in ethnic problems.

Uganda: Like Kenya, federalism in Uganda was viewed by Milton Obote as necessary to get independence--and to prevent the secession of Buganda.  It accomplished both, yet it did not prevent the intensification of ethnic problems.  In 1966 they reached such a serious level that Obote sent troops to seize the Kabaka and then abrogated the federal constitution.  Again, federalism was associated with a period of relative ethnic peace and violent ethnic confrontation.

This brief summary of ten cases is sufficient to show a very inconsistent relationship between federalism and ethnic problems, confirming the Davis hypothesis—and, the observations of most scholars of Africa.  Why is there no consistent relationship?

Back to Top

III.  Explanation of Findings

If one adheres to a non-science based epistemology, i.e., the view that the development of general knowledge is not feasible and that each case is relatively unique, then the absence of a consistent relation is not problematic.  If, on the other hand, one adheres to a science-based epistemology, i.e., the view that the development of general knowledge is feasible and that finding regularities is important, then such a finding needs to be “explained.”

Our explanation is that three requisites for a science-based study are absent.  First, the conceptions of federalism and ethnic problems do not have measurable and consistent meanings.  Second, a sufficient number of cases is not available to allow one to control for alternative causal factors.  Third, information about the variation of cases over time is not available.  Without these requisites, one can only expect ambiguous findings.

Back to Top

A.  Concepts with Variable Meanings

1.  “Federalism”

Daniel Elazar says: “If a political system is established by compact and has at/least two ‘arenas,’ ‘planes,’ ‘spheres,’ ‘tiers’ or ‘levels’ of government, each endowed with independent legitimacy and a constitutionally guaranteed place in the overall system, and possessing its own set of institutions, powers, and responsibilities, it is deemed to be federal.” [xvi]  To apply this definition to determine whether a state has a federal system we must be able to locate a compact.  Presumably, this is a constitution or a constitution-like document.  The problem in Africa is that such formal agreements may exist but may not guide behavior.  This was the case in Kenya in the immediate post-independence period.  Elazar seems to recognize this problem when he amends his definition by saying that “Only in those polities where the processes of government reflected federal principles is the structure of federalism meaningful.” [xvii]  Determining whether federal principles guide “the processes of government” may pose problems and lead to a variety of interpretations.  In the case of the Ethiopian-Eritrean federation, one might have expected Ethiopians to interpret reality in a much different way than Eritreans did.  If all that is required is for federal principles to be followed, the necessity of the “compact” is unclear.  Furthermore, the extent or scope of issues dealt with according to the federal principle, which is required to call a polity a federation, is not specified. [xviii] 

To accommodate the diversity of states encompassed by federalism, Elazar refers to “forms of self-rule and autonomy” [xix] and “political systems utilizing federal arrangements,” as well as “federal systems.” [xx]  He includes Southwest Africa (Namibia) among the first, but not the second category; and Equatorial Guinea and Zaire among the second, but not the first category.  Such disagreement over what is or is not a federal system is even more common when one looks at lists of different authors.

Benyamin Neuberger suggests that the extent of federalism in Africa can be increased “if we are ready to be flexible in defining federalism....the federal principle is used in Africa in non-democratic federations, in regional autonomy structures, in informal consociational arrangements and schemes of functional cooperation.” [xxi]  As Preston King has said “The variety of meanings associated with federation creates a genuine basis for misunderstanding.” [xxii]

Back to Top

2.  “Ethnic Problems”

The impreciseness in the concept of “ethnic problems” derives from the diversity of meanings given to both words in the expression.

Soren Bollerup and Christian Christensen note that the concept of conflict is contested.  “The main distinction is between those who advocate a broad definition of conflict, including both latent and manifest conflict, and those who advocate a narrow definition encompassing only manifest conflict….Our definition is…simply… ‘perceived divergence of interest’….A national conflict, then, involves conflict between two or more nation-groups.” [xxiii]  Our use of the word “problems” corresponds to Bollerup and Christensen’s broad definition of conflict, i.e., one including both manifest and latent forms.

Although such conflict involves many possible issues, the distinguishing feature is the adjective which precedes it.  Rudolfo Stavenhagen argues that

The term ‘ethnic conflict’ covers a wide range of situations.  In fact, it might be argued that ethnic conflict as such does not exist.  What does exist is social, political and economic conflict between groups of people who identify each other in ethnic terms:  color, race, religion, language, national origin.  Very often such ethnic characteristics may mask other distinguishing features, such as class interests and political power, which on analysis may turn out to be the more important elements in the conflict.  Still, when ethnic differences are used consciously or unconsciously to distinguish the opposing actors in a conflict situation—particularly when they become powerful mobilizing symbols, as is so often the case—then ethnicity does become a determining factor in the nature and dynamic of the conflict. [xxiv]

John Coakley makes a similar point when he argues that  “the phenomenon of ethnicity....has been used to cover a range of types of political conflict that are differentiated not merely by the dynamics of competition between rival groups but also by the very significance of ethnicity itself....The reality is that the same label is used here as an umbrella for a great diversity of types of conflict.” [xxv]

There is an extensive literature on the meaning of ethnicity, nationalism and associated concepts. Richard Caplan and John Feffer suggest that “The key to nationalism’s elusiveness seems to be its lack of clear objective criteria.  All such criteria--race, language, territoriality--can be shown to apply in some particulars, but not in all cases.  One method of avoiding this quandary has been to stress the subjective nature of nationalism.” [xxvi]  This tendency is often an outcome of defining the concepts in instrumental or situational terms, rather than in primordial ones.  That is, as a socially determined identity rather than a biologically determined one.

Still, the definitions differ over “markers,” i.e., over the set of traits defining the identity.  These may include history, language, caste, race...or a combination of them. Different terms are used for identities that vary in their inclusiveness.  For example, the political scientist James Coleman defined nationalism as “a consciousness of belonging to a nation (existent or in the realm of aspiration) or a nationality, and a desire, as manifest in sentiment and activity, to secure or maintain its welfare, prosperity, and integrity, and to maximize political autonomy.”  He suggested that four types of nationalism might be distinguished on the basis of the inclusiveness of the identities involved:  Pan-African nationalism, Nigerian nationalism, regional nationalism and group nationalism. [xxvii]  The politician Leopold Senghor suggested that there were only two types of identities distinguished by their inclusiveness, fatherlands and the Nation.  He argued in a manipulative and colorful way that

The Nation groups fatherlands in order to transcend them....Far from rejecting the realities of the Fatherland, the Nation will lean on them, or more precisely, will lean on their virtues, their realistic character, and therefore on their emotional strength.  It will unite the virtues of the fatherlands or most often will choose those virtues which, by reason of climate, history or race, have a common denominator or a universal value.  Once realised, the Nation forges a harmonious ensemble from its different provinces:  a single country for a single people, animated by the same faith and striving toward the same goal....the Nation is superior to the Fatherland on the level of humanity, and even of efficiency.  It distills the values of the latter, sublimates them by transcending them....If the Nation is a conscious determination to reconstruct, the State is its major means. [xxviii]

The academic Ivo Duchacek used different terms in describing what Senghor called the Nation and fatherland: “Devotion to one’s geographically determined area, its values and inhabitants, is called nationalism, and on a subnational level, local patriotism....People’s allegiance to subnational communities, territorial or functional, is usually less intensive and less unconditional than their devotion to the national community.” [xxix]

The variety of conceptions of “ethnic” and of “problems” is obvious.  John Breuilly asked himself why there is such confusion over the concept of nationalism and replied:  “Partly it is because some of the terms themselves are difficult to define precisely.  But above all it is because the term ‘nationalism’ is used too widely and covers many and different kinds of things.  It is used to refer to ideas, sentiments and politics, yet these are distinct kinds of things which do not stand in any necessary relationship to one another.” [xxx]

Back to Top

B.  Additional Variables too Numerous for the Cases

The variables which might cause ethnic problems to become greater or to become lesser are many.  These may include such factors as the number of ethnic groups; the nature of the leader who seeks to speak for each; the regional distribution of wealth; the degree to which income of groups fluctuates with world prices; the amount of foreign aid; the ability of the police and army to suppress dissent; the degree to which the state leaders have instilled a common ideology in the population;  the history of perceived grievances;  the degree to which people are engaged in a capitalist economy; the degree of urbanization; the threat from abroad; the skill of state leaders; and many others.  Yet, the number of cases listed in Table 1 is fewer than the number of variables other than federalism which might account for ethnic problems.  Because they are so numerous, they cannot be properly controlled.  That federalism, rather than one or more of these additional factors, caused the increase or decrease in ethnic problems can not be determined with any precision is the consequence.

Back to Top

C.  Cases and Causes which Vary Over Time

There are two problems related to the affect of the passage of time on the relationship between federalism and ethnic problems.

First, whether one defines federalism in formal constitutional terms or in informal division of powers terms, its actually existing character does not remain constant. Federalism in Nigeria, for example, has not been a constant since 1960.  It has altered its character due to the increase in the number of states from 3 to 30; the imposition of military rule; the effort to overcome the “faults” of the First Republic in the creation of the Second; and, the hesitant and so far abortive efforts to return to civilian rule since the late 1980s.  An assessment of federalism’s impact on ethnic problems is really an assessment of the impact of many different forms of federalism.  Even if federalism were causal, variations in the type of federalism would probably affect ethnic problems differently.  What is needed is an index of the degree of federalism at any point in time, like the indices of democracy—but, without the inaccuracies they embody.

Second, another temporal problem is that of a lag in the impact of federalism on ethnic problems.  In the short-term federalism may assuage ethnic demands thereby reducing ethnic problems, e.g., the Baganda’s federal relationship with Uganda in 1962;  in the long-term federalism may codify ethnic allegiance thereby increasing ethnic problems, e.g., the Baganda resistance to Milton Obote in February of 1966.  With a limited number of cases, with a bundle of possible causes besides federalism, with variation in the form of federalism over time federalism, determining the time-lag impact on ethnic problems is extremely difficult.

Back to Top

IV.  Implications of Findings

Some sense can be made of the relationship between federalism and ethnic problems if we focus on (a) the general phenomenon of which both federalism and ethnic groups are manifestations and (b) their use as political tools.  Federalism is one mechanism for the distribution of power within a political system;  ethnic problems are the outcome of competition among groups representing different types of identity.  The distribution of power can be used for a variety of purposes, as can identity.  Just as a carpenter can use tools to produce many different products, so can those who manipulate the distribution of power and employ identity.  The tool may marginally affect what can be built, but what is built depends more on the intentions and skills of those able to use the tool.

Back to Top

A.  What is the Genus?

1.  Of the Species Federalism

Federalism is really a set of ideas about how power should be distributed in a society.   The disputes over the concept’s meaning are irrelevant if one focuses on this broad idea, the genus of the species “federalism.”  All are types of power distributions.  The differences among the conceptualizations have to do with the relative balance and types of powers found in the hands of individuals and institutions with authority over parts of a society. These differences may interfere with understanding relationships with other concepts.  Politics is concerned with the struggle to gain access to power.  Much of the literature on federalism makes more sense when one moves from the species to the genus, distribution of power.

Jenny Robinson, writing on federalism in South Africa, argues similarly:  “...I will suggest that federal discourse is necessarily linked to, and constitutive of, power relations.” [xxxi]  She contends that federalism is “first and foremost a political project clearly associated with transforming the character of state power and with asserting certain types of political subjects as central....Groups identified on the basis of a wide variety of social cleavages, perhaps age, gender, class or sexuality, or groups dispersed across territories, find federalism of no use.” [xxxii]  She goes on to say:  “not only does a federal solution empower particular political identities and organisations and disempower others, but it is also likely to generate new forms of political identity, not least territorial ones.” [xxxiii]

Thus, federalism might be viewed as a mechanism for the distribution of power in a society.

Back to Top

2.  Of the Species Ethnic Group

An ethnic group is really a form of identity group.  The disputes over its meaning and how it differs from a nation are irrelevant if one focuses on this broad idea, the genus of the species “ethnic group.”  All are forms of identity groups.  The differences among the conceptualizations have to do with the cluster of traits that establish identity.  These differences may interfere with understanding relationships with other concepts.  Politicians construct and use identity groups in their struggle for political power.  Much of the literature on ethnic groups and ethnic problems makes more sense when one moves from the species to the genus, the identity group.  It can be examined by focusing on implicit claims.

Back to Top

a.  Identities are assumed by individuals when it is in their self-interest

The assumption of an identity appears closely related to self-interest.  Michael Keating argues that situationists claim that “individuals define themselves collectively in ‘ethnic’ terms in order to extract resources in systems where rewards are to be had from such self-identification; or where necessary to defend themselves from persecution or oppression.” [xxxiv]  And, Duchacek argues that “People usually support, and identify with, persons and institutions that largely satisfy their fundamental demands for identity, internal order, external security, progress, welfare, and culture.” [xxxv]

These identities are not necessarily conflictual. Dan Smith observes that “A safe and secure identity formation is not likely to generate conflict;  it is the threatened identity that tends to produce violence.  And once violent conflict has begun, the possibility of defeat is a further threat to social identity.” [xxxvi]  Self-interest in the form of self-preservation may account for the violence.

Back to Top

b.  Individuals have multiple identities

The idea that people have multiple identities is used to account for the proliferation of new nationalisms. Keating has observed that some minority or peripheral nationalisms “are prepared to credit citizens with dual or multiple loyalties and identities, and the capacity to act in different arenas.” [xxxvii]  He suggests that “Minority, or separatist nationalism, involves the denial of exclusive claims on the part of the state nationalism and the assertion of national rights of self-determination for groups within it.” [xxxviii]

The conflict among multiple identities was at the heart of the famous Nyerere-Nkrumah argument over how to build a United States of Africa.  Basically, Nkrumah’s opposition to East African federation was grounded on the view that there is a kind of battle going on within people between different identities.  An East African federation would introduce a new identity which was an additional barrier to the acquisition of a pan-African identity.  Nkrumah argued that “regional federations are a form of balkanization on a grand scale.” [xxxix]  He suggested that “such federations may even find objection to the notion of African unity....We must endeavour to eradicate quickly the forces that have kept us apart.  The best means of doing so is to begin to create a larger and all-embracing loyalty which will hold Africa together as a united people with one government and one destiny.” [xl] Don Rothchild has argued similarly:  “If nationalism is a force for integration on a continent subdivided by hundreds of traditional allegiances, it is also a disintegrative force with respect to supranational groupings.” [xli]

Nyerere’s view was that changing identities was not as difficult as Nkrumah contended.   He argued that “It is said, for one thing, that new local loyalties will be built up which will militate against loyalty to Africa.  Yet it is hard to take this argument seriously.  Over the past ten years our people have had to expand their tribal loyalties to encompass the nation.  It will be much easier for them to feel a loyalty to ‘Africa’ which does not divide their tribe, and which, in the sense of their own experience, is little more immense than their own nations frequently are.  The introduction, through a Federation, of another interim step to unity is no great complication;  it might indeed prevent the growth of an insular nationalism which would later prevent a wider loyalty.” [xlii]

Graham Smith has suggested that the success of secessionist movements is related to the nature of multiple identities.  He observed that “the ability to mobilise support behind the secessionist cause will depend upon whether a sense of common identity is strong enough to overcome identities that crosscut ethno-regional divisions or whether local identity is so diluted by stratification and segmentation as to undermine a shared interest in cultural survival.” [xliii]

Back to Top

c. Individuals may change the type, intensity and rank order of their identities

Central to identity ideas is the view that they are subjective.  Jenny Robinson suggests that “all political identities are constructed and are constantly being transformed....”[xliv]  Benyamin Neuberger expresses a similar view:  “The perception of nations as indivisible is based on the fallacy that nationhood is something permanent.  In reality, nations come and go, and national identity may expand, contract, or disappear.  Perhaps the reason for the sanctity of the indivisible nation is rooted in the hidden knowledge that all nations are divisible.” [xlv]  He argues that this notion is basic to an understanding of self-determination:

In Africa, as elsewhere, there is no permanent national self.  Someone’s primary identity may be Ethiopian, Eritrean, or Tigrinean in Eritrea or Nigerian, Southern, Eastern, or Ibo in Iboland.  It will be a function of time and context.  Ronen is right to say that the switch from one identity to another is possible because all identities are optional weapons in the quest for self determination.  The identity of the self, which is so crucial to the establishment of self-determination, may vary with time.  Different times and different conditions may lead to different identities and to a different perception of ‘us’ and ‘them.’ [xlvi]

Stuart Hall contends that identities appear to be changing simultaneously in global and local ways:  “on the one hand, the nation and all the identities that go with it appear to have gone upwards—reabsorbed into larger communities that overreach and interconnect national identities.  But at the same time there is also movement down below.  Peoples and groups and tribes who were previously harnessed together in the entities called the nation-states begin to rediscover identities that they had forgotten….So on the one hand, we have global identities because we have a stake in something global and, on the other hand, we can only know ourselves because we are part of some face-to-face communities.” [xlvii]

Back to Top

d. Identity groups aggregate a variety of traits which link individuals

Identities appear to be composed of aggregates of traits.  Human Rights Watch put it this way: “No group can be defined by a single racial, religious or ethnic characteristic, just as no human being is/ethnically ‘pure’ and no society exists that does not have its minorities.  In many situations, ethnic identity has traditionally been fluid, such that over time the issue of difference has been resolved by one group’s blending into another for accommodation.” [xlviii]

Dov Ronen suggests that over time the characteristics or traits of the major identity groups have changed:  “national, class, minorities, non-European/racial, and ethnic--succeeded each other as dominant types from the nineteenth century to the latter part of the twentieth.  However, a new type does not eliminate previously dominant types.  All five types exist today....”[xlix]  Obviously, though, a variety of different clusters of traits could form the bases of identity groups.

Back to Top

e. Political leaders use identity groups to extend the number of their followers

Identity groups are of central importance to politics.  Soren Bollerup and Christian Christensen argue that identities are mobilized by elites. [l]  In the struggle against colonial rule, the legitimacy of political leaders was determined by the size of their identity groups.  Patrick Chabal notes that:  “Competing nationalist organisations attempted to demonstrate their legitimacy as nationalist spokesmen by claiming to represent the colonial subjects.  This was tested by the colonial state through elections….Where there were no elections, the nationalists sought to prove before the UN and other international forums their claim to represent the colony’s population by (peaceful or violent) political mobilisation.” [li]  Political leaders seek to create and sustain identity groups which they use for instrumental purposes.

In competing for members, political leaders create identities which best serve leaders’ purposes.  John Sorenson has described various the Ethiopian and Eritrean “constructed nationalisms.”  They differ in their descriptions of the facts and in what they include and exclude: “Whereas Ethiopian discourse claims that Eritreans approved of federation and then voted for its abolition a decade later to rejoin Ethiopia, Eritrean nationalists argue that federation was first imposed against majority opinion and later illegally abrogated so that Ethiopia could exercise direct control....For the most part, Ethiopian nationalists have ignored points of international law and concentrated on producing versions of the past which argue that Eritreans favored unification or that no national identity existed among Eritreans.” [lii]

In a sense, politics is a struggle to find the combination of myths and ideas that will create the largest and most loyal identity group.

Back to Top

f. To survive political units must create an identity group of their subjects

A prerequisite for the longevity of political units appears to be the development of an identity group of those within its jurisdiction.  That is why “nation-building” occupied such a prominent place in post-independent African countries.  Leaders’ opposition to federalism often appears to be a consequence a fear that in a federal system citizens would transfer their loyalties to the constituent units. McKown asks “Why did federations fail?”  Her answer is that the “‘urge toward unity’ was conspicuously absent on the part of many partners in what appeared to be forced marriages.” [liii]  Thus, states actively seek to build a sense of identity among citizens.

In sum, it appears not only possible but also advantageous to focus on the genus “identity” in stead of the species “ethnicity” or “nationalism” in seeking to account for the impact of federalism on ethnic problems.

Back to Top

B.  Are the Genera Tools?

1.  Distribution of power

If the distribution of power is a tool in political struggle, then much of what is written about federalism in Africa becomes more understandable.  Such an interpretation appears to accord with the views of many scholars. Abeysinghe Navaratna-Bandara observes that “The Nigerian example shows how a central government may use devolution as a political instrument to weaken the political strength of ethno-regionalism....” [liv]  More than thirty years ago Don Rothchild described the way leaders used federalism for a variety of purposes. [lv] Jenny Robinson contends that federalism is “first and foremost a political project clearly associated with transforming the character of state power and with asserting certain types of political subjects as central.” [lvi]  Cynthia Enloe has argued that “Federalism is best understood if it is seen as one conflict management formula among several.  Typically, its utility is measured according to its functionality for central elites (which may or may not be ethnically exclusive).”[lvii] Benyamin Neuberger suggests that in the 1950s and early 1960s, “Federalism within the state was the outcome of devolution, and thus its function was more to mediate between the ethnic groups than to integrate them into one uniform whole.” [lviii]  And Ruth Lapidoth notes that autonomy, closely related to federalism, is “a tool or a framework that can constitute an adequate compromise if the parties are looking for one....Like any tool, it must be used in accordance with the special circumstances of each case.” [lix]

Two conclusions are obvious:  First, the distribution of power is widely viewed by observers as a tool; second, the tool may be used for a variety of purposes.

Back to Top

2.  Identity groups

Much the same appears true for identity groups.  Dov Ronen, writing about self-determination, notes that “The switch from one identity to another is possible because all identities are optional weapons in quests for self-determination.” [lx]  Earl Conteh-Morgan writes of the “strategic utility of exploiting ethnic sentiments as a political weapon by an ethnic power elite.” [lxi]  And, John Bruilley says “Many of the approaches I have so far discussed can be given a functional form.  Nationalism can ‘function’ as an instrument of class interest, or it can function to furnish an identity need.” [lxii]  Just as the distribution of power appears to be used as a tool for a variety of purposes, so has identity.  Viewed in this way, the critical questions lie not with the identity but with who is using it and for what purpose. 

Back to Top

V.  Conclusion

Thus, S. Rufus Davis’ justified frustration over the apparent impotence of federalism does not mean it plays an insignificant role in African political struggles.  We have seen that it may be associated with the solution to ethnic problems;  it may be associated with the intensification of ethnic problems;  and, it may be associated with neither.  The analytical problem is conceptual, having to do with both the meaning and use of the terms employed.  The meanings attributed to “federalism” and “ethnic problems” are inconsistent, i.e., the words are used by different scholars to refer to many different “things.” The use of both “federalism” and “ethnic problems” is inappropriate, i.e., the terms are assumed to be either a cause or an effect.  If we look at the “elephant” instead of its tail or ears or trunk or leg, our understanding of what we observe may be increased.  In other words, if we look at concepts of which federalism and ethnicity are aspects, more sense can be made of the roles they play.  These core concepts are the distribution of power and identities.  And, if we see the distribution of power and identity groups as tools in political struggles rather than causes and effects, the lack of a consistent relationship between federalism and ethnic problems becomes understandable--and, more fruitful analyses become possible.

Back to Top

ENDNOTES

[i] S. Rufus Davis, The Federal Principle (Berkeley:  University of California Press, 1978), pp. 211-212.

[ii] Roberta McKown contends that “neither a federal nor a unitary constitution is a solution to multiculturally based problems but a structural context within which they may be confronted.”  See Roberta McKown, “Federalism in Africa,” in C. Loyd Brown-John, ed., Centralizing and Decentralizing Trends in Federal States (Lanham, MD: University Press of America, 1988), p. 298.  John McGarry and Brendan O’Leary observe:  “Unfortunately federalism has a poor track-record as a conflict regulating device in multi-ethnic states….”  See John McGarry and Brendan O’Leary, “Introduction, The Macro-Political Regulation of Ethnic Conflict,” in McGarry and O’Leary, The Politics of Ethnic Conflict Regulation (London: Routledge, 1993), p. 34. Vincent Ostrom observed that “Whether federal structures provide an adequate basis for the government of diverse ethnic communities is an entirely problematic question.”  See Vincent Ostrom, “Federal Principles of Organization and Ethnic Communities,” in Daniel Elazar, ed., Federalism and Political Integration (Tel Aviv:  Turtledove Publishing for the Jerusalem Institute for Federal Studies, 1979), p. 81.  Eric Nordlinger noted that “federalism would seem to be especially appropriate in divided societies with territorially clustered segments....”  Yet, he noted also that “federalism may actually contribute to a conflict’s exacerbation and the failure of conflict regulation.”  See Eric Nordlinger, Conflict Regulation in Divided Societies (Boston:  Center for International Affairs, Harvard University, 1972), p. 31. Don Rothchild has similar ambivalence.  He suggests, on the one hand, that federalism in South Africa might work because “the existence of relatively autonomous subregions reassured ethnic minorities about their ability to participate in decisional matters of importance to them,” yet, on the other hand, that it might not work because the new government fears “the divisive effects of ethnically inspired federalism....”  See Donald Rothchild, Managing Ethnic Conflict in Africa (Washington, D.C.:  Brookings Institution Press, 1997), p. 57. T. Vincent Maphai argues that in the case of South Africa,  “If the primary idea behind federalism is to prevent ethnic conflict, then any possible success in this respect is likely to be limited….”  Yet, he suggests it may help minimize “possible conflict between African groups.”  See T. Vincent Maphai, “Liberal Democracy and Ethnic Conflict in South Africa,” in Harvey Glickman, ed., Ethnic Conflict and Democratization in Africa (Atlanta:  The African Studies Association Press, 1995), p. 106.  John Coakley is similarly pessimistic about the possibility that federalism might reduce ethnic tensions.  See John Coakley, “Introduction:  The Territorial Management of Ethnic Conflict,” in Coakley, ed., The Territorial Management of Ethnic Conflict (London:  Frank Cass, 1993), p. 19.  Nevertheless, Ivo Duchacek has noted that “a federal solution of polyethnicity is a relatively rare response to ethnic demands.” See Ivo Duchacek, “Federalist Responses to Ethnic Demands:  An Overview,” Elazar (1979), p. 66.

[iii] Martin Dent, “Federalism in Africa, with Special Reference to Nigeria,” in Murray Forsyth, ed., Federalism and Nationalism (NY:  St. Martin’s, 1989), p. 172.

[iv] Crawford Young, Politics in the Congo (Princeton:  Princeton University Press, 1965), p. 532.

[v] Benyamin Neuberger, “Federalism in Africa:  Experience and Prospects,” in Elazar (1979), pp. 178 and 174.

[vi] Neuberger, “Federalism in Africa:  Experience and Prospects,” p. 174.

[vii] A.E. Afigbo, “Federal Character:  Its Meaning and History,” in Peter Ekeh and Eghosa Osaghae, eds., Federal Character and Federalism in Nigeria (Ibadan:  Heinemann Educational Books (Nigeria) Ltd., 1989), p. 41.

[viii] McKown, p. 306.

[ix] Donald Horowitz, Ethnic Groups in Conflict (Berkeley:  University of California Press, 1985), p. 603.

[x] Horowitz (1985), pp. 612-613.

[xi] Horowitz (1985), p. 34.

[xii] Jenny Robinson, “Federalism and the Transformation of the South African State,” in Graham Smith, ed., Federalism, The Multiethnic Challenge (London:  Longman, 1995), p. 275.

[xiii] Martin Dent, pp. 171-172.

[xiv] Horowitz (1985), p. 615.

[xv] Dean E. McHenry, Jr., Limited Choices, The Political Struggle for Socialism in Tanzania (Boulder:  Lynne Rienner, 1994), pp. 189-213.

[xvi] Daniel Elazar, “The Role of Federalism in Political Integration....,” in Elazar (1979), pp. 29-30

[xvii] Elazar, “The Role of Federalism in Political Integration....,” p. 30.

[xviii] Martin Dent defines federalism in a parallel fashion.  Noting that “federalism is a somewhat elusive concept,” he says it “refers to both a constitutional dispensation and to a means of exercising power....one can have federal elements in the running of a state even where the constitution is not fully federal.”  See Dent, p. 169.

[xix] Elazar (1979), pp. 224-225.

[xx] Daniel Elazar, “Introduction:  Why Federalism?” in Elazar (1979), pp. 19 & 21.

[xxi] Neuberger, “Federalism in Africa:  Experience and Prospects,” in Elazar (1979), p. 185.

[xxii] Preston King, Federalism and Federation (Baltimore: Johns Hopkins University Press, 1982), p. 71.

[xxiii] Soren Rinder Bollerup and Christian Dons Christensen, Nationalism in Eastern Europe, Causes and Consequences of the National Revivals and Conflicts in Late-Twentieth-Century Eastern Europe (NY: St. Martin’s, 1997), p. 3

[xxiv] Rodolfo Stavenhagen, “Ethnic Conflict and Human Rights, Their Interrelationship,” in Kumar Rupesinghe, ed., Ethnic Conflict and Human Rights (Tokyo: The United Nations University and Oslo: Norwegian University Press, 1988), p. 17.

[xxv] John Coakley, p. 3.

[xxvi] Richard Caplan and John Feffer, “Introduction,” in Caplan and Feffer, eds., Europe’s New Nationalism, States and Minorities in Conflict (NY:  Oxford University Press, 1996), p. 9.

[xxvii] James Coleman, Nigeria, Background to Nationalism (Berkeley:  University of California Press, 1960), pp. 425-427.

[xxviii] Leopold Sedar Senghor, Nationhood and the African Road to Socialism (NY: Praeger, 1962), pp. 22-31, reprinted in Rupert Emerson and Martin Kilson, eds., The Political Awakening of Africa (Englewood Cliffs, NJ: Prentice-Hall, 1965), p. 153.

[xxix] Ivo Duchacek, Comparative Federalism, The Territorial Dimension of Politics (NY:  Holt, Rinehart and Winston, 1970), pp. 21-22.

[xxx] John Breuilly, Nationalism and the State, Second Edition (Chicago:  University of Chicago Press, 1994), p. 420.

[xxxi] Robinson, p. 256.

[xxxii] Robinson, p. 274.

[xxxiii] Robinson, p. 275.

[xxxiv] Michael Keating, Nations Against the State, The New Politics of Nationalism in Quebec, Catalonia and Scotland (NY: St. Martin’s, 1996), p. 4.

[xxxv] Duchacek (1970), p. 31.

[xxxvi] Dan Smith, “Reconciling Identities in Conflict,” in Caplan and Feffer, eds. (1996), p. 206.

[xxxvii] Keating, p. 19.

[xxxviii] Keating, p. 18.

[xxxix] Kwame Nkrumah, Africa Must Unite (London: Heinemann, 1963), p. 214.

[xl] Nkrumah, p. 215.

[xli] Donald Rothchild, “The Limits of Federalism:  an Examination of Political Institutional Transfer in Africa,” Journal of Modern African Studies, Vol. 4, No. 3 (November 1966), pp. 284-285.

[xlii] Julius Nyerere, “The Nature and Requirements of African Unity,” in Nyerere, Freedom and Unity (London:  Oxford University Press, 1967), p. 348.

[xliii] Graham Smith, “Mapping the Federal Condition:  Ideology, Political Practice and Social Justice,” in Smith, ed., Federalism:  The Multiethnic Challenge (London:  Longman, 1995), p. 11.

[xliv] Robinson, p. 273.

[xlv] Benyamin Neuberger, National Self-determination in Postcolonial Africa (Boulder:  Lynne Riener, 1986), p. 56.

[xlvi] Neuberger (1986), p. 59.

[xlvii] Stuart Hall, “Ethnicity:  Identity and Difference,” in Geoff Eley and Ronald Grigor Suny, eds., Becoming National, A Reader (NY:  Oxford University Press, 1996), p. 343.

[xlviii] Human Rights Watch, Playing the ‘Communal Card,’ Communal Violence and Human Rights (NY: Human Rights Watch, 1995), p. x-xi.

[xlix] Dov Ronen, The Quest for Self-Determination (New Haven:  Yale University Press, 1979), p. 13.

[l] Bollerup and Christensen, p. 34.

[li] Patrick Chabal, Power in Africa, An Essay in Political Interpretation (NY: St. Martin’s Press, 1992), p. 136.

[lii] John Sorenson, Imagining Ethiopia, Struggles for History and Identity in the Horn of Africa (New Brunswick, NJ: Rutgers University Press, 1993), pp. 47-48.

[liii] McKown, p. 298.

[liv] Abeysinghe Navaratna-Bandara, The Management of Ethnic Secessionist Conflict, The Big Neighbour Syndrome (Aldershot, U.K.:  Dartmouth Publishing Co. Ltd., 1995), p. 23.

[lv] Donald Rothchild (1966), p. 280.

[lvi] Robinson, p. 274.

[lvii] Cynthia Enloe, “Internal Colonialism, Federalism and Alternative State Development Strategies,” Publius, Vol. 7, No. 4 (Fall 1977), p. 159.

[lviii] Neuberger (1979), p. 173.

[lix] Ruth Lapidoth, Autonomy, Flexible Solutions to Ethnic Conflicts (Washington, D.C.:  United States Institute of Peace Press, 1997), p. 204.

[lx] Ronen, p. 13.

[lxi] Earl Conteh-Morgan, Democratization in Africa, The Theory and Dynamics of Political Transitions (Westport, CN:  Praeger, 1997), pp. 102-103.

[lxii] Breuilly (1994), p. 418.