Multilingualism and Nation Building:

Language and Education in Eritrea

Chefena Hailemariam
Department of English, University of Asmara, PO Box 1220, Asmara, Eritrea
Sjaak Kroon
Babylon, Center for Studies of Multilingualism in the Multicultural Society, Faculty
of Arts, Tilburg University, PO Box 90153, 5000 LE Tilburg, The Netherlands
Joel Walters
Department of English, Language Policy Research Center, Bar-Ilan University,
Ramat Gan, Israel
This paper focuses on language choice inthe newest nationinthe Hornof Africawithin
a broader context of language policy in multilingual states. Pre-colonial and
post-colonial language policies inEritreaare surveyedinrelationtoevolving linguistic
and political nationalism. Language contact and its social consequences are discussed
inanattempt toshedlight on language policiespursuedduring different periods inthe
colonial history of Eritrea. Using descriptive frameworks provided by contemporary
sociolinguistics, post-independence language policy, with language and education at
the centre, is looked at fromthe perspectiveof the functional allocationof nine Eritrean
languages and the points at which they conflict and complement each other. Public
responses and evaluations are analysed and implications for further research are
advanced.
Language

and

Nation

in

Africa
When geography, political borders, religion, ethnicity and language all coin-
cide, there is generally very little conflict over language policy, and the minority
groups whose political aspirations and language practices differ fromthe mono-
lingual majority often find very little room to make a claim for plurilingualism.
When, however, categories such as geography, ethnicity, and religion cut across
language and literacy, the potential for controversyand even conflict multiplies.
If we add to this a long history of war and colonialism, we begin to get some
insight intothe complexityof the language situationin Eritrea, one of the worlds
most recently independent nations. Central highlanders are predominantly
Christianand speakers of Tigrigna, the same name serving as their ethnicity. But
there are also Tigrigna who live outside the central highlands as non-natives of
those areas, or who practice Islam and who speak Tigre or one or another of
Eritreas nine major languages. By the same token, there are Tigre, for the most
part Moslems who live in the lowlands and make up Eritreas second largest
ethnolinguistic group, who speak Tigrigna or are bilingual in Tigre and Tigrigna
and practice one of three major varieties of Christianity in Eritrea. While the
Tigrigna and Tigre account for approximately eighty% of the population of
Eritrea, there are seven other groups whose geographic, political, religious, and
ethnolinguistic identity cut across traditional sociolinguistic categories. This
0143-4632/99/06 0475-19 $10.00/0 1999 C. Hailemariam et al.
JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT Vol. 20, No. 6, 1999
4 7 5
paper is an attempt to describe some of this complexity and indicate some of the
major issues involved in Eritreas emerging language policy.
Among Africans, nationality is taken to mean a sub-national entity, and
nationalism, a related concept, refers to ethnic solidarity and related senti-
ments. Toaddtothe complexity, nation refers tothe politico-geographic notion,
which African states have inherited from colonialism. In this same vein,
nationism represents the sense of political community which accompanies this
geo-political notion. As a result of these terminological distinctions, national-
ism is understoodto represent a sociocultural identity that maynot have a corre-
sponding geo-political realisation. The interplay of these concepts on the
background of African ethno-cultural patterns which have evolved from
inter-group behaviour and solidarityis deemed important for an understanding
of contemporary language policy developments and decision-making. More-
over, such a conceptual and terminological distinction provides greater insight
into why social solidarity is not a precondition for the existence of a national
political communityandhowa national political communitycanattainsuchsoli-
darity in successive steps (Fishman, 1968: 39). The validity of this statement of
Fishman seems to have been corroborated in the evolution of Eritrean nation-
alism, which is one of the issues, treated in this paper.
When elites of (geo-political) nationalist movements took power from their
colonisers, they inherited various population groups with distinct languages
and cultures. According to Abdulaziz (1991), Somalia is an exception to this
pattern, being one of the fewAfricancountries, whichis largely monolingual and
ethnically uniform. While the elites in Somalia or in most African countries have
been successful in rallying their respective populations to independence, they
have been far less successful in attempts to create a strong sense of nationhood.
Selection of languages for national communication, a decision-making process
embodied with power and prestige, was at the centre of the problem. Another
unique example among African countries, Tanzania is cited for promoting what
was originally the language of one of its minority groups in order to unite its
multilingual population and to serve as the post-independence national
language. Kiswahili had been the language of the political elites during the
movement for independence. This fact, according to Rubagumya (1990: 9),
enhanced the status of the language after independence, for it was now rightly
regarded as the national language.
Historical Background
Eritrea is a small country in the Horn of Africa with approximately 119,000
square kilometres of landspace and a population of about 3.5 million (Habte
Selassie, 1980). The country is bordered on the north and west by the Sudan, and
on the South by Ethiopia and Djibouti. Since 1889, Eritrea has existed within the
same external boundaries since the treaty signed between Menelik, King of Ethi-
opia, and the Italians.
Eritreas history is one of conflict, revolving around Eritreas claimfor nation-
hood versus Ethiopias contention that Eritrea formed an organic part of Ethi-
opia. Like many post-colonial states, Eritrea is a territorial entity of the
nineteenth century created by European colonialism and thus entitled to the
4 7 6 Journal of Multilingual and Multicultural Development
right toself-determination. At the endof the SecondWorldWar, Italylost posses-
sion of all its colonies. Although the United Nations approved independence for
Italys former colonies, it decided in favour of federationof Eritrea withEthiopia.
The debate in the United Nations General Assembly and the subsequent resolu-
tion give some of the background and the role of the great powers. Space limita-
tions do not allow full details; we present only the highlights of major events.
Many factors militate against Eritreas right to self-determination. The coun-
trys multi-ethnic composition has been exaggerated and used to present the
country as too fragmented linguistically and religiously to sustain a single
nation. The British used this approach as early as 1943 to divide Eritrea into a
Christian portion that would join Ethiopia and a Moslem part that would form
part of the Sudan (Habte Selassie, 1980). That scheme was, however, rejected by
the Eritreans themselves (Tesfay, 1997).
The years between 1941 and 1952 were marked by major events and political
turbulence. Ethnic and religious politics, hitherto unknown, was introduced
during this period. By 1945 three political parties or groups had already been
formed. These were: The Eritrean Independence Party, The Islamic League
Party, and the Unionist Party. The present government, the successor of the
EritreanPeoples Liberation Front (EPLF), wouldargue that whether Eritrea was
divided (as conceived by Western writers) and whether the greater part of the
highlands formedanorganic part of Ethiopia is a debate whichshould be limited
to academics. The essential problem was defined as a matter of
self-identification. Ethiopian historians claim that Eritrea, known as Mereb
Melash, possesses no identity it can claim as its own other than the north Ethio-
pian identity (Abbay, 1998) is dismissed by Eritrean scholars.
According to Ghebre-Medhin (1989), a critic of Ethiopian studies, the idea of
Greater Ethiopia has dominated regional studies in the area for decades. Thus,
Ethiopia was studied as a distinct and unique entity, with the Abyssinians as
carriers of a superior culture, andthe rest of the peoples in the empire were to be
seen as inferior. Although Ethiopia, according to Clapham (1988: 195) is no less
ethnically varied than other African states, the political role of ethnicity has
differed significantly. Ethiopia was anempire with the Amharic language as the
core of the cultural area, and it gave little weight to issues of ethnic origin and
rights. As Hameso (1997) noted, in the 1880s when Emperor Menelik a Shoan
Amhara expanded his domain beyond the traditionally feudal Abyssinia, the
effect was soon matched by linguistic, political and cultural domination. It was
the Haileselassie regime which consolidated the rule of the Amhara by
promoting the Amharic language to a national status (Bloom & Tamrat, 1996).
The origin and construction of the Eritrean national identity needs, then, to be
looked at against this background.
After years of neglect, the role of ethnic and religious divisions in Eritrea is
gaining attention. There are scholars who hold that religious and ethnic cleav-
ages are stronger than nationalist feelings (Araya, 1990; Abbay, 1998; Negash,
1997). These writers base their observations mainly on the history of Eritrean
nationalist movements and the role of ethnic and religious movements, with
particular reference to the 1940s. The competing view interprets Eritrean nation-
alismas a widely sharedsentiment superseding narrowethnic and religious ties
with unity strengthened by the war against Ethiopia (Markakis, 1994; Ottaway,
Multilingualism and Language Education in Eritrea 4 7 7
1998). Athirdperspective views Eritreanidentity as a dynamic phenomenon in a
constant process of construction (Tesfay, 1997). On the whole, it can be argued
that the period of Italian rule contributed by its very length to the strengthening
of a sense of Eritrean national identity (Erlich, 1983; Patman, 1990). But that
national identity in subsequent periods (19411952; 19521961; 19611991) may
not be fully understood without proper analysis of the process of interaction,
competition, and conflict among the various cultural groups as well as between
the cultural groups and the nationalist movements in their efforts to gain polit-
ical power in Eritrea (Woldemikael, 1993). According to Tesfay (1997), the polit-
ical parties of the 1940s and 1950s were so loosely organised that members could
move fromone to the other, withamazingfluidity andfrequency. The politics of
the 1940s seem to have provided a somewhat distorted impression. Moreover,
the political and social historyneeds to be looked at in the context of regional and
international political alignment. A brief account of the events leading to the
disposal of Eritrea by the United Nations andthe role Great Britain, Ethiopia and
the United States should give a much clearer understanding of the origins of the
conflict.
The origin of the Eritrean conflict can be traced particularly to the period
between 1941 and 1952. That was when the first generation of anti-colonialists
failed to forma coalitioncapable of establishinga unitedEritreanstateto counter
Ethiopias claims. Eyob (1995) argues the de-colonisation of Eritrea was compli-
cated by six factors:
(1) The overriding interest of the liberating allied forces, which delayed the
process of de-colonisation; (2) an international consensus which favoured
Ethiopias territorial claims; (3) Ethiopias intervention in the internal
Eritrean political arena; (4) the absence of any cohesive political institution
capable of reconciling the divergent interests of multi-ethnic Eritrean
society; (5) the politicization of long standing religious and regional rival-
ries among the inhabitants; and (6) the absence of a neutral and effective
international de-colonising agency. (Eyob, 1995: 61)
While the further elaboration on the magnitude of each of these factors may
give better understanding, the united intervention of Ethiopia and the United
States is documented in the words John Foster Dulles, then US Secretary of State,
who stated in a speech before the UN Security Council in 1952, as quoted by
Habte Selassie (1980: 58):
From the point of view of justice, the opinions of the Eritrean people must
receive consideration. Nevertheless, the strategic intents of the UnitedStates
in the RedSea basin and considerationof the securityandworldpeace make
it necessary that the country has to be linked with our ally, Ethiopia.
Upon this background of conflict, Eritrea presents a case of neglected colonial
struggle. After its resolution to federate Eritrea with Ethiopia against the wishes
of its people, the UNnever raisedthe issue of the abrogationof the Federal consti-
tution. November 14, 1962 marked the day of complete annexation of Eritrea by
Ethiopia. But the clandestine political struggle, which had already started in the
towns, was transformed into an armed struggle in September 1961 by the
Eritrean Liberation Front (ELF).
4 7 8 Journal of Multilingual and Multicultural Development
Colonial Language Policy in Eritrea
European colonialism, in spite of some of its negative consequences, has
undoubtedly contributed to the social cohesion of the Eritrean peoples. This is
particularly true of the impact of the Italian colonial period (18891941). Despite
self-interest, the Italiancolonial administrationsecured the rule of lawand social
welfare for the Eritreans (Trevaskis, 1960: 10). Thus, the formation of Eritrea as a
political community by bringing together the various nationalities can be attrib-
uted in the first instance to the Italians. According to Tseggi (1981: 132):
They connected each other by extensive road systems, telephone, services
and railway systems connecting the economically significant cities in
Eritrea. For the first time a set of colonial rules andregulations applicable to
all nationalities was enforced.
This account, however, should not imply that the Italians had moulded a
cohesive political unity. But, as was apparent from the era to follow, this unity
did not prove to be too fragile to succumb to the newcolonisers policy of divide
and rule. Notwithstanding attempts of more recent colonisers, the British and
the Ethiopians, who incited divisions along religious lines, the Eritreans
remained united for over a century. As Pateman (1990: 67) writes:
Over the fifty years that the Italians ruled Eritrea, some of the linguistic,
religious and ethnic divisions had been broken down. Many saw them-
selves primarily as Eritreans and not Moslems, Tigrigna or Saho farmers.
Differences still remained, differences subsequently exploited cleverly by
Ethiopia and unionists within Eritrea.
The Italian period is not noted for its progressive educational policy, as it
curbed the native populations eagerness for knowledge by restrictingeducation
to the first four years of primary school. In addition, educational policy was both
discriminatory and colonialist. It was discriminatory in that it allowed the
natives to study for only four years. It was colonialist in that the main content of
the curriculumwas basedonItaliangeography, historyandculture, as education
was perceived as an instrument for enhancing colonial rule (Negash, 1987).
By comparison, the British policy for Eritrean education (19411952) afforded
far more access to education and advancement of literary traditions. The British
promoted Tigrigna and Arabic on the grounds that the two languages could
equally serve the majority Christian Tigrigna nationality, which constituted
approximately half of the total population of the country, the other half being
predominantly Moslem nationalities. This decision represented not only a push
forward for the development of the Tigrigna language, but also an important
step that ushered the way for use of this language as a language of political
debate onthe future of Eritrea in the newspapers founded by the BritishMinistry
of Information. Of particular relevance to this topic is that, when the weekly
newspaper in the Tigrigna language made its debut in 1942, the British Adminis-
tration encouraged the formation of the Tigrigna Language Council which was
responsible for corpus planning to help Tigrigna cope with its linguistic limita-
tions. These limitations revolved mainly around the twin questions of spelling
and terminology (Fellman, 1979: 25).
Multilingualism and Language Education in Eritrea 4 7 9
The rationale behind the British decision to promote only two languages i.e.
Tigrigna andArabic as co-official languages andlanguages of educationcannot
be accounted for as easily as the Italian policy, which promoted Italian to the
exclusion of all local languages. The British policy was based on both linguistic
and political grounds. From a linguistic point of view, Tigrigna and Arabic each
had a written tradition. Other Eritrean languages had no literary heritage, other
than the Bible translation available for Tigre and Kunama. The script most
familiar to Moslem Eritreans is Classical Arabic, which is learned via the Koran.
Teklehaimanot (1996: 13) describes British educational policy as follows:
Instruction was given in Tigrigna for Christians and Arabic for Moslems.
The textbooks in Arabic were procuredfromEgypt andSudan; andthose in
Tigrigna were prepared by the Department [of Education], with the first
textbook written by Ato Issac Tewoldemedhen.
The other motive behind British language policy was political, and the polit-
ical goals of language decisions are often obscured by their overt linguistic ends
(Cooper, 1989). The prevailing political conditions during the British Mandate
Period help shed light on the other motives for the language policy. The British
grand strategy to partitionEritrea, accordingto Tseggi (1988: 71), was drawn up
soas toinflame religious animositybetweenChristians andMoslems tothe point
that no unity among Eritreans seeking independence would become possible.
According to Trevaskis (1960), author of the most authoritative but not neces-
sarily the most objective chronicle of that period, religious differences which
were muted during the Italian period began to surface with the economic crisis.
Trevaskis reports that the Christian Highlanders were suspicious of British
intentions, since they felt that the British discriminated against themin favour of
other groups, viz. the highland trader community, Italians, Arabs and Sudanese.
The religious enmity fomented by the Britishculminatedin anarmedclashin the
capital Asmara at about the same time as the future of Eritrea was being
discussed at the United Nations (19501952). Meanwhile, the Ethiopians used
the situation to advance their own political ends, initiating a guerrilla war by
infiltratinginto Eritrea (Tseggi, 1988). The approachtakenhere is that overt polit-
ical decision and actions can be used to infer motives behind language policy
decisions.
Whatever the covert motive was at that time, Tigrigna and Arabic served as
the two official languages until 1956 when the Ethiopian government began
encroaching on the Federation Agreement, prior to its total annexationof Eritrea
in 1961. Following annexation, Ethiopia declared that both Tigrigna and Arabic
should cease to be official languages (Bokru, 1988). Fellman (1979: 26) notes the
political origins of this decision when he states that even the 1952 Eritrean
[Federal] Government resolution declaring Tigrigna as the official language,
along with Arabic, was more a prestige resolution than a linguistic one.
The development of the Tigrigna language declined with the growing inter-
ference of the Imperial government of Ethiopia in the affairs of Eritrea. With the
1950 constitution of the United Nations resolution to grant Eritrea only local
autonomy within the Ethiopian Empire, the status of Amharic, Ethiopias
national language began to rise at the expense of Tigrigna. Ethiopian rule was
marked by deliberate dismantling of Eritreas educational and cultural institu-
4 80 Journal of Multilingual and Multicultural Development
tions. All textbooks prepared bythe Department of Educationwere collectedand
burned in a big rubbish area in a suburb of Asmara (Bokru, 1988). Amharic
replaced both Tigrigna and Arabic in all official government and school
domains. As a consequence, these two languages remained limited to the home
environment during the Ethiopian colonial period. The educational conse-
quences are described by Habte Selassie (1980: 61):
The imposition of Amharic not only expressly violated a provision of the
UNresolution, but it imposeda potential obstacle inthe pathof the Eritrean
children the effects of which were felt when countless numbers failed
secondary and university entrance examinations and were effectively
denied higher education.
The last era of Eritrean colonial history (19621991) is of greatest
sociolinguistic interest. Ethiopian prohibition of Tigrigna, which had achieved
official status under the British, was tantamount to waging war on the national
identity of the Eritrean people. As Araya (1990: 95) puts it:
The emperor imposed Amharic as the official language in Eritrea, and the
resulting widespread resentment undoubtedly helped Eritrean nation-
alism to thrive thereafter in the highlands.
Indeed, in the long years of struggle for national liberation and self-assertion
of national identity, the symbolic function of Tigrigna plays a major role. The
liberation struggle, drawing upon a romantic view of the Eritrean past and
appealing to a collective memory of that society, succeeded in pulling together
all efforts to develop the language, both in the Diaspora and in the areas which
were liberated. Efforts exerted by the EritreanPeoples LiberationFront (EPLF) in
all forms of development were described by Leonard (1988: 131) as follows:
I am not aware of any liberation front which under the conditions of a
prolonged war of national liberation, has achieved the level of political,
social and economic development one observes today in Eritrea.
But also, the Social Affairs Bureauof the EritreanLiberationFront (ELF) which
directed the Education and Literacy Campaign Department, was successful in
wiping out illiteracy from among the ranks of its fighters (ELF Foreign Informa-
tion Centre, 1977).
Multilingualism and Language Planning
In the above an attempt was made to distinguish the term nation as used in
the African context fromthe nation-state of the European model, a monolingual
political entity with well-defined geographic boundaries. In this latter concep-
tion, the development of a sense of nation as an outcome of long ethno-cultural
processes is understood in light of the history of human social development
evolving a sense of nation(Mansour, 1993). Despite the fact that multilingualism
is a common phenomena shared by many countries in Africa, Asia and Latin
America, it has been regardedas a problemthat works against nationalismor the
sense of nation (Fasold, 1984: 4). It has been equated with chaos and associated
with underdevelopment, resulting from lack of proper societal communication.
Multilingualism and Language Education in Eritrea 4 81
No causal connection has yet demonstrated this assertion. In fact, community
and individual multilingualism in many African states facilitates communica-
tion processes in different settings at inter-ethnic, sub-national or national levels
of communication. As Mazrui and Mazrui (1992: 84) state:
Africas ethnic heterogeneity finds its differentiation in language. Per
capita there is a wider range of languages than in any other region of the
world. By strange twist of destiny, there are also more French-speaking,
English-speaking and Portuguese-speaking countries in Africa than
anywhere else in the world.
Many studies are handicapped by a bias that the achievement of national and
linguistic unification is a condition for the attainment of nationalism and prog-
ress in general. In the Eritrean perspective particular historical facts provide an
insight into how a strong sense of nation could be forged in the process of the
commonhistorythat binds peoples together. As Markakis (1994: 6) puts it: In the
course of the long struggle, regional solidarity evolved into identity that can be
called national, at least in the case of Eritrea.
Early recognition of societal multilingualism as a communication problem
was what induced sociolinguists to think of language manipulation as a solu-
tion to this problem (Stewart, 1968: 532). Language planning is limited in its
scope when it is considered an activity to solve language problems, as indicated
by its early definition as the organisedpursuit of solutions tolanguage problems
typically at the national level (Fishman, 1974: 79). With the evolution of
sociolinguistics over the last two decades, there are now models for analysis of
multilingual situations and policies available to deal with these situations. The
study of language policy, therefore, which constitutes an important component
in the studyof language planning, has recentlyattractedthe attentionof a variety
of social science disciplines. In its broadest sense, language planning can be
defined as deliberate efforts to influence the behaviour of others with respect to
acquisition, structure or functional allocation of their language codes (Cooper,
1989: 45).
Decision making with respect to the functional allocation of languages is
indeed anessential aspect of language planning and provides a commonground
of discussion for our present purpose. According to Bamgbose (1989) such deci-
sions may affect language status or the development of a corpus, and they may
occur at any point in the planning process. It is also pointedout that implementa-
tionof language planning oftenoccurs inthe absence of a coherent policy which
has been deliberated upon prior to its initiation. Moreover, manylanguage poli-
cies are carriedout locally without either coherent policy or planning initiatives
(Grabe, 1994). These phenomena imply that policy decisions ought not to be
considered absoluteformulas to produce specific results. As in social planningin
general, language planning is not only unpredictable but also may produce
unexpected consequences (Fishman, 1974).
The unpredictable nature of the projected outcomes of a social policy may,
however, serve as an excuse for an arbitrarypolicy adopted by planners without
due considerationto the inherent fluidity and variability of local situations. This
is particularly true of fundamental decisions taken at the governmental level
with respect to questions such as whether a language should be an official one or
4 82 Journal of Multilingual and Multicultural Development
whether it should be a medium of instruction. Such decisions may be received
positively or negatively and accompanied by sudden shifts in attitudes, influ-
enced by the prevailing social forces of change. For example, after the fall of the
Soviet Union, there were strong tendencies to abandon the Cyrillic alphabet in
Kazakhstanas well as in the neighbouring central Asianrepublics of Kirghizstan
and Turkmenistan (Coulmas, 1994). Similarly, in Eritrea the Geez script, a tradi-
tionpeculiar to the languages in Eritrea and Ethiopia, is received withless enthu-
siasmby some Tigre speakers onreligious andsociocultural grounds rather than
because of its phonological inadequacy (Mahmud, 1996).
In dealing with multilingualism and the processes of nation-building, devel-
oping countries have had to find ways of balancing the twin goals of national
unity and multilingual diversity. The success of some language policies does not
necessarily depend on the type of policy approach selected. Policies apparently
progressiveor democraticat the level of rhetoric have sometimes provedfailures
in practice. As Coulmas (1994: 39) shows, both the Soviet Union and Yugoslavia
had overtly liberal language policies, based on the principle of language
equality. () Nonetheless, Russianwas the language of power andadvancement
in the Soviet Union as was Serbo-Croatian in Yugoslavia. The former Soviet
Union was not necessarily a failure of pluralism, but perhaps a failure to apply
pluralist principles systematically, in language policy as in other domains
(Mansur, 1993). A dichotomy between pluralism and assimilation can be
misleading and cannot capture the inherent complexity of the local situation.
Eritrean Ethnolinguistic Diversity
Before addressing the current state of Eritreas language policy, the present
section offers a general sociolinguistic portrait of the nation with a focus on its
multilingualism. Among the sociolinguistic categories relevant for generating
this picture are geographic distribution, demographic size, clan and ethnic
origin. The descriptionof special status languages is alsogiven distinct treatment
because of its importance in status planning. In addition, information about
literary traditions and the influence of colonialismon the sociolinguistic portrait
are discussed. It is important to note that the resulting portrait is not a definitive
empirical sociolinguistic profile the Eritrean language situation. Due to a lack of
recent statistical data, the present portrait will rely on speculative estimates
basedonsecondarysources as well as a varietyof as yet unpublished sources and
reports. Although many reports, e.g. the UNICEF (1994) document, still describe
the relative size of ethnic groups in terms of percentages, others express reserva-
tions about the statistical validity of these figures.
Lambert (1996) has described Eritrea as a mosaic consistingof nine nationali-
ties, but the cultural, linguistic, and religious diversity is so complex and over-
lapping that these groups do not fit a single classification. While Claphams
(1996) reference to Eritrea as African Belgium is used in the political sense to
stress the artificiality of the state, Woldemikael (1993) has observed that Eritrea
shows a great deal of fluidity due to factors such as widespread ethnic mixing,
religious conversion, migration, conquest and wars in the region. A dichoto-
mous categorisationof Tigrigna-speaking Christianhighlanders versus Moslem
lowlanders assumes connections between language, religion and geography.
Multilingualism and Language Education in Eritrea 4 83
Residents of the Central Highlands generally belong to the Tigrigna nationality
and are predominantly Christians. In the lowlands on both the eastern and
western side of the highlands, formerly called Gash, Barka, Semhar and Senhit,
live the other nationalities who identify predominantly with Islamic religious
practices. While the majority of the highlanders are settled agriculturists, the
lowlanders are nomads, semi-nomads or settledagriculturists. Although no reli-
able census has been published as yet for the population of these communities
nor for Eritreas total population, the working figures generally set the Christian
and Moslem populations at more or less 50% each. The following data, taken
from EPLF (1977; 1987) and Pool (1997), attempt to describe Eritrean
ethnolinguistic diversity. With the exception of the Bilen (nationality name)
whose language is called Bilin, in the following enumeration the name of the
nationality also refers to the language of this nationality.
The majority of the Tigrigna nationality who live in the Central Highlands
belong tothe Coptic OrthodoxChurch; a minorityareCatholics andProtes-
tants.
The Tigre live in a large area around the Central Highlands in the north-
eastern and western lowlands in the regions called the Senhit, Sahel and
Semhar. The majority of Tigre are pastoral and agro-pastoral. They are
organised in tribes with varying degrees of solidarity. They are over-
whelmingly Moslem.
The Afar inhabit the Dankalia, the most arid region located along the
southern coast. With the exception of small groups who live off agriculture
or fishing and trading, the Afar are nomadic pastoralists. They are very
tightly organised socially, and entirely Moslem.
The Kunama occupy the region around a small town called Barentu. They
are an agrarianpeople with some cultural elements of a communal society.
While many Kunama are Christians and Moslems, animist practices are
still prevalent in the society. The Kunama are the minority group on which
Italian colonialism and missionaries left its strongest influence in terms of
religion and literacy.
The Nara live to the north and east of Barentu. They are followers of Islam
and live off agriculture. They are surrounded by the Tigre nationality and
are among the nationalities least influenced by colonialism. National soli-
darity is strong among this group.
The Bilen is a nationality that resides in areas around the towns of Keren
and Hagaz. They are surrounded by the two majority nationalities, the
Tigrigna and Tigre. They are followers of either Islam or Christianity.
The Saho live in a large area spread over the eastern escarpments stretched
downthe coastal areas of the RedSea, adjacent tothe Tigrigna nationalityof
Akeleguzay and the Tigre of Semhar on the east and north. The Saho are
predominantly Moslems and, other than those living in towns, have
remained relatively insulated from the influences of colonialism.
4 84 Journal of Multilingual and Multicultural Development
The Hidareb, who are called Bejas in the Sudan, are also one of the nationali-
ties least influenced by colonialism. They live in an area spread over the
northwestern part of the country that stretches up to the Sudanese border.
The Nara and Hidareb are affiliated culturally to the Tigre.
The Rashida are Arabic speakers living along the coasts of the Red Sea from
around Emberemi and Sheeb to Karora, the northern tip of the country.
Traditional agriculture is practised in some of the localities in the commu-
nity, but the Rashida are fairly mobile, engaging in pastoral farming and
trade.
According to UNICEF (1994) data the Tigrigna nationalityis the largest of all,
making up 50% of the total population. Demographically, the second largest
group is the Tigre which makes up 31%of the total. In terms of distribution, the
Tigrigna nationality is the most widely spread over the major towns of the
country. But geographically, the area where the Tigre language is spoken as a
first language is the largest. In a recent survey in 12 major towns in the country
(National Statistics and Evaluation, 1997: 23), the Tigrigna-speaking nationality
was found to comprise over 77% of the total population, followed by the Tigre,
which made up 13% of the population in those 12 towns. Another feature of
Eritrean linguistic diversity is the highlander and lowlander residential distinc-
tion, whichalsodenotes a religious distinctionbetween Christians andMoslems.
Notwithstanding, these social distinctions, no natural barriers mark linguistic
territoriality. Even before the new internal regional divisions were created,
geographic and linguistic borders did not coincide. The Tigre speakers, for
instance, are spread over the areas formerly called Sahel, Semhar, and parts of
Senhit and Barka. And the Saho nationality lives in both highland and lowland
areas. The religious, regional, and nationality boundaries also cut across
geographic boundaries.
Linguistically the nine most widely-spoken languages in Eritrea fall intothree
major language families. Afar, Bilin, Hidareb and Saho are Kushitic languages,
Tigrigna, Tigre and Arabic belong to the Semitic group, and Nara and Kunama
are Nilo-Saharan. The languages within each family are characterisedby dialect
differences that lead to different degrees of mutual intelligibility. As far as the
nature of bilingualism is concerned, it is not easy to state the degree of bilin-
gualismfor eachgroupwith any degree of confidence. It canbe generally argued
that the majorityof the westernlowlanddwellers speak Tigre as either a first or a
second language. The question of whether Arabic is a real lingua franca at a
national or sub-national level is in need of investigation.
It would be naive to generalise that all Eritrean Moslems understand Arabic
without proper qualification in terms of the level of understanding, the domains
of use, and the nationalityof the speakers. (The special role of Arabic in Eritrea is
discussed later in this paper.)
Existence of a literary tradition is an important factor in the description of
Eritrean ethnolinguistic diversity. Although the literary heritage of the Tigrigna
language is not a long one, it has been a written language since the arrival of
missionaries and the onset of colonialism (from approximately the 1880s).
Tigrigna is written in Geez orthography, as attested by religious books and
written laws. Geez script is the indigenous script with the longest literary heri-
Multilingualism and Language Education in Eritrea 4 85
tage, and is considered the parent language of Tigre, Tigrigna, and Amharic.
Tigre dates the beginning of its literary development to 130 years ago.
In general it can be said that Tigrigna and Tigre together make up 80%of the
total population and are most widely distributed throughout the country. In
addition, by virtue of the urbanisation of Tigrigna and Tigre speakers, they can
be said to belong to a socioeconomic elite. Both languages serve as languages of
inter-ethnic communication, in that manymembers of other nationalities use one
of the languages as a second language. Thus, they can serve as lingua franca at
either national or sub-national levels.
Mazrui andMazruis (1992) distinctions are useful here as a clue tohownative
languages coexist in a multilingual setting. The Eritrean languages, apart from
Tigrigna and Tigre, could thus be categorised as intra-ethnic languages of
communication and solidarity to which members are sentimentally attached.
The interplay of their functional roles as vernacular languages, as media of
intra-ethnic, inter-ethnic and/or official languages, is claimed to be very impor-
tant (Mazrui & Mazrui, 1992).
Another dimension of Mazrui and Mazruis analysis of African
multilingualism which may be used to describe the Eritrean sociolinguistic
patterns is their dimension of relative competition and complementarity. They
delineate four categories: Afro-ethnic, Afro-Islamic, Afro-western, and
Western. According to this rubric, Tigrigna and Kunama, and to a certain
degree Bilin, could be counted as examples of Afro-ethnic languages. Despite
their family origins, the remaining Eritrean languages, including Arabic, are
Afro-Islamic. The influence of Arabic is due to its use as a common religious
language. Historically, the spread of Arabic southwardover the last few centu-
ries and its assimilating effects are also relevant here. Recent developments in
the Tigre language are illustrative. Conditions of war in Eritrea led to a massive
exodus of refugees to the Sudan and resulted in lexical influence of Arabic in
terms of borrowingandcode-switching, especially in the speech of the younger
generation.
In Eritrea, Arabic, Tigrigna andEnglishare special status languages, eachwith
different functions in different domains. Though none of the languages is
declared an official language, two of them, Tigrigna and Arabic, function as
working languages, or de facto official languages. The nature of this status needs
further elaboration for the relevant contexts. These languages may be described
in Coopers (1989) terms as working, symbolic or statutory official languages.
These statuses may not be mutually exclusive, however.
Tigrigna is a symbolic official language in Eritrea; it represents the state of
Eritrea in the sense that its revitalisation is associated with the success of the
national liberation struggle, which accommodated Tigrigna linguistic nation-
alism. The impositionof Amharic as the only national language andthe resulting
resentment helped Eritrean nationalism gain momentum. The effect of this
process was to engender among Tigrigna speakers commonromantic memories
of language and aspirations for its revitalisation. The fate of the language was
very much associatedwith the success of the political struggle, earning Tigrigna
a symbolic significance. This process can be compared to the symbolic function
Swahili played during the political movement in Tanzania (Cooper, 1989).
4 86 Journal of Multilingual and Multicultural Development
When the British assigned Arabic a co-official status along with Tigrigna, it
was given a statutoryofficial language function. This decision was based strictly
on demographic considerations. Although the use and spread of Arabic had
been suppressed, its survival had not been threatened by Ethiopian rule, since
the language had an alternative setting where it could develop and be used by
native speakers. It is a sacred and prestige language for Moslems in Eritrea.
According to Araya (1990), it is the preferred spoken language among the
Moslem elite. Despite cultural biases and stereotyping which promotes polari-
ties of Arabic-Moslem versus Tigrigna-Christian, future sociolinguistic
research may clarify the role of Arabic as a national lingua franca. This research
should indicate, among other things, what percentage of the Eritrean Moslem
populationuses spoken Arabic, at what level of proficiency, andin what kinds of
situations. Whatever the findings of this kind of research and whatever differ-
ences between reported and actual language use may be, Arabic remains a
language of official ceremonies, national gatherings and government declara-
tions. In this sense, therefore, both Tigrigna and Arabic enjoy both statutoryand
official functions.
English is the medium of instructionfrommiddle and secondary school up to
the university level. It also serves as the language of international communica-
tion and business. Unlike the situationin many other African nations, it does not
serve as a gatekeeper to filter upward movement in the political domain. But it is
the language of the educational elite. Unlike other developing nations of Africa,
no other private European community or English language schools for the chil-
dren of the upper class exist. Withthe growing need of international contacts, the
advantages of knowing Englishbecome apparent tolarger segments of the popu-
lation.
The influence Italianlanguage even 60 years after the end of colonial period is
still strong. A good percentage of the last generation living in towns had a func-
tional knowledge of Italian. The contribution of the Italianlanguage towards the
enrichment of the Tigrigna lexicon is significant. Presently there are a couple of
Italian kindergartens and one Italian community secondary school. But, the
present generation, whichwent through Amharic-mediumschools at some level
of its schooling, or worked in Ethiopian civil service can be said to speak reason-
ably good Amharic.
Finally, the role of the classical language, Geez, deserves mention here. It is a
language now generally limited to religious affairs, particularly in the Christian
Orthodox Church. Ethiopists considered it to be a carrier of the Abysinian
culture and literary traditions (Ghebremedhin, 1989). It has the same function in
Ethiopia. Its status andfunctionis somewhat comparabletothat of classical Latin
(Bloom & Tamrat, 1996).
Language Policy in Post-independent Eritrea
In the plenary sessionof the First National Conference on Eritrean Languages
(Asmara, 1618 August 1996), the Eritrean Peoples Front for Democracy and
Justice (PFDJ) acknowledged that the national language policy was not a
completely worked out perfect document and that it should be open to improve-
ments and modifications (PFDJ, 1996). That was the first and only forum where
Multilingualism and Language Education in Eritrea 4 87
the language issue was debated by participants from all walks of life and ethnic
groups. It was also one of the hotly discussed issues, mostly in non-Tigrigna
areas, in the debate leading up to the approval of the draft-constitution
(19951997). Apart from these public forums, the Eritreans in the Diaspora,
particularly those in North America and Europe, have debated the issue on the
Eritrean web-site Dehai. The web-site offers a means for criticism by those who
prefer using aninternational language at the expense of an indigenous language.
The language policy issue has so far not been dealt with in academic circles.
The essence of Eritreas language policy is its recognition of multilingualism
and its decision to deal with its complexities. The present government, aware of
the problems of other multilingual nations, seems committed to ensuring
equality of nationalities, and considers equality a right that is to be secured
through public policy. On this background, language policy may support,
tolerate or reject multilingualism, and in doing so, it may give special emphasis
to one or more languages. Adoption of a language policy, which is meant to
promote multilingualism, is based on motivations which are essentially social,
cultural, political and economic in nature (Clyne, 1997: 304).
Eritrean language policy has its ideological roots in the EPLFs National
Democratic Programme adopted at its first congress in 1977 (EPLF, 1977). This
program, which was also endorsed at the second congress in 1987, takes the
language issue as an important element of its commitment to ensuring equality
of all nationalities (EPLF, 1987). Accordingly, no official status is offered to any
one language in order not to marginalise the speakers of other languages. The
policy onlanguages andplanning which has followed in Eritrea conforms witha
decision-making process in which the Government of Eritrea, the successor of
the EPLF, sits at the top of the organisation chart and implements language
policy decisions through its agent, the Ministry of Education.
Decisions about language policies movetop-down, andthe public, whichmay
applaud or resent those decisions, expresses its reactions and exercises its
evaluative judgement. The forms of public evaluation vary according to the
nationality in question, the social and educational status of the members of that
group, and the religion or political ideology the group values. Whatever reasons
and intentions, no policy can be successfully implemented without being sanc-
tioned by the government. In order to provide a more practical discussion and a
better understanding of Eritrean language policy, in what follows we focus on
the implementation process in the field of education and the school systems,
consideredtobe institutions for the constructionof social andcultural identities.
Language-in-education Planning
The ways in which the global objectives of language policy and planning
processes are related to their implementation can be summarised in Ingrams
(1989) statement that education, and particularly the classroom, are sites where
all the entities of the whole policy system converge and generate tension:
Language-in-education planning is that field of planning which exists
between language policy-making and the classroom and its curriculum.
The boundary between language policy-making in general and
language-in-education planning is often unclear, but the latter seeks how
4 88 Journal of Multilingual and Multicultural Development
within the education planning system the ideas, the goals, and content of
the former to the extent they are relevant to education, can be realised.
(Ingram, 1989: 53)
The educational institution is recognised not only as an agent of implementa-
tion but also as an arena for the interface of macro-level ideologies and
micro-level linguistic behaviours in the everyday life of speakers. Particularly in
a multilingual setting like in Eritrea, it is a place where relationships of power
among ethnolinguistic groups maybe formed. Accordingto Pool (1997: 12), cul-
tural, educational, and language policies in the post-liberation period derive in
the large part fromthe EPLF practices during the liberation struggle. In keeping
with its egalitarian practices of the pre-independence period, the Ministry of
Education of the Government of Eritrea seeks to realise the more general
macro-level national objectives of education policy in the school systems in a
disciplined and structured manner. Generally stated, the aims of the language
policy are:
To foster Eritrean national consciousness and national unity, ensure social
justice, commitment to nation building and the cultivation of good citi-
zenry. (Garahtu, 1996: 1)
On the background of this philosophical basis for the policy, the government
works towards its realisation through school programs based on multilingual
and multicultural considerations. The language in education policy of the
Government of Eritrea provides the following:
(1) Each ethnic group has the right to use its language as the language of
instruction at elementary school level.
(2) The language of instructionat post elementary school level shall be English,
while other languages could be introduced as subjects.
(3) At the elementary school all those students whose medium of instruction is
Arabic will take up Tigrigna as a subject, and those whose medium of
instruction is Tigrigna will take up Arabic as a subject. Arabic is anyway
taught at both levels as a subject. (Garahtu, 1996: 2)
The underlying philosophy of this policy is the recognition and promotion of
multilingualism in the state of Eritrea. With mother tongue education at the
centre, the policy is based essentially on a pluralist approach to education. The
goal is preservationof minority languages and enhancement of multilingualism.
Generally speaking, this approach fits into what is known as enrichment bilin-
gual education, a variant of a Developmental Maintenance Bilingual
Programmewhichaims togo beyond the staticmaintenance andpromoteindi-
vidual andgroupuse of the minoritylanguages, withthe intent that it will leadto
cultural pluralism and to the social autonomy of the ethnic group (Baker, 1996:
173). In contrast, both the switch from mother tongue instruction in primary
school to English at the secondary level and the teaching of Arabic and Tigrigna
as subjects of instruction characterise bilingual approaches with a clear transi-
tional flavour. That transitional orientation may be somewhat mitigated by the
multilingual nature of Eritrean society and by the fact that English is a foreign,
rather thana secondlanguage for most of the Eritrean population, whereas most
Multilingualism and Language Education in Eritrea 4 89
cases of transitional bilingual education involve transition to a language spoken
natively by a large segment of the indigenous population (e.g. English in the
United States, French in France).
Above we claimedthat language policy constitutes a decision-making process
that is political in nature. Effects of that policy, by their very nature, take a long
time to be realised in observable and quantifiable terms. During the process,
reactions of language policy consumers may provide feedback which deci-
sion-makers then use tomake alterationswithinthe overall strategyof the policy.
Whether and how these reactions and debates affect the decision-making
process depends on how they are presented. There is an inherent tension
between the need for public evaluation and discussion and the orderly, efficient
implementationof the policy. Emotionallychargedarguments, sometimes based
on ungrounded assumptions, may stem from prejudices and mistrust. Such
emotion will affect implementation of even the most ideal policy. In a climate of
rhetoric, the symbolic value of a language may become inflated beyond its objec-
tive proportions, for example, in terms of number of speakers, frequency of use,
etc. Cooper (1989: 86) writes about the symbolic potential, as follows:
When a language serves or can be made to serve, as a symbol of a glorious
past, or of the unique genius of a people, the elites and counter-elites who
manipulate this symbol can use it to maintain or acquire legitimacy in the
name of authenticity and tradition.
Language policy and planning, a relatively young discipline, may not yet be
capable of prescribing magic formulas to offer to the planner as a panacea for all
language problems, especially in the case of a multilingual society on which little
research has been done. The discipline does, however, suggest frameworks
within which variables for the analysis of language planning can be studied. In
that light, variables relevant to the particularity of the situation in which the
policy is to be implemented deserve careful attention.
Discussion and Conclusions
Within the overall conception of language policy as a decision-making
process, anattempt has been made in this paper to identify andtodescribe which
actors attempt to influence what behaviours, of which people, for what ends
(Cooper, 1989). Two more questions, which need further investigation and anal-
ysis, remain: under what circumstances and with what effects are those policies
implemented. To these ends, we feel the brief account of the evolution of
language policy in each of Eritreas colonial periods and the relevant historical
data which led to these social contexts provides some background knowledge of
the circumstances under which current Eritrean policy functions. With regard to
the question pertaining to the effects of Eritrean language policy, not much can
be said at this point in time. Field-based studies, such as the Netherlands-Israel
Research Programme (NIRP) sponsored project Unity and Diversity via
Multilingualism in which the authors are engaged at present, are expected to
provide some indication. Reports coming out from the Ministry of Education
point to the problems of implementation of various forms of language policy.
Apart from the administrative challenges in terms of textbook preparation,
4 90 Journal of Multilingual and Multicultural Development
teacher training, supervision, lack of motivation of teachers due to low salaries
and difficult living conditions in the countryside, one important issue at present
revolves around the attitude of the parents of language minority children. From
informal reports available, there are indications that some parents fromthe Afar,
Tigre and Saho nationalities prefer Arabic to their own vernaculars.
Although nation-building is not a function of language policy decisions, the
connectionbetween language andnationalityanda sense of political community
offers insight into the political process. Language policy (official and/or educa-
tional) is a decision-making process with complex economic, social, cultural and
political ramifications. Careful articulationof the cultural, religious and political
dimensions of language choice may help clarifysome of the issues. Language is a
delicate subject in Eritrea; it faces all kinds of reactions, ranging from the senti-
mentally charged applause of government supporters to outright disapproval in
the polemics of counter-elites. The direction fromwhich these reactions emanate
may be less important than the question of whose interests these discourses
represent. A critical question for language planning in Eritrea should focus on
bringing together the macro and micro level perspectives on language issues.
Solutions will depend on appropriate sociolinguistic investigations which look
at language use at different levels of communication vis-a-vis the effects of the
policy on the various facets of the societys life.
Correspondence
Any correspondence should be directed to Dr Sjaak Kroon, Babylon, Center
for Studies of Multilingualism in the Multicultural Society, Faculty of Arts,
Tilburg University, PO Box 90153, 5000 LE Tilburg, The Netherlands
([email protected]).
References
Abbay, Alemseghed (1998) Identity Jilted or Re-imagining Identity: The Divergent Paths of the
Eritrean and Tigrian Nationalist Struggles. Trenton: Red Sea Press.
Abdulaziz, M.H. (1991) Language in education: Comparative study of the situation in
Tanzania, Ethiopia, and Somalia. In O. Garcia (ed.) Bilingual Education: Festschrift in
Honor of Joshua A. Fishman (pp. 7586). Amsterdam/Philadelphia: John Benjamins.
Araya, M. (1990) The Eritrean question: An alternative explanation. The Journal of Modern
African Studies 28 (1), 79100.
Baker, C. (1996) Foundations of Bilingual Educationand Bilingualism. Clevedon: Multilingual
Matters.
Bamgbose, A. (1989) Issues for a model of language planning. Language Problems and
Language Planning 13 (1), 2434.
Bloom, T. and Tamrat, W. (1996) Issues in Ethiopian language policy and education.
Journal of Multilingual and Multicultural Development 17 (5), 321337.
Bokru, H. (1988) A short history of Eritrean literary tradition (Tigrigna version). Journal of
Eritrean Studies 3 (1), 3039.
Clapham, C. (1988) Transformation and Continuity in Revolutionary Ethiopia. Cambridge:
Cambridge University Press.
Clapham, C. (1996) Eritrean independence and the collapse of Ethiopian centralism:
Causes, consequences, and implications. Geopolitics and International Boundaries 1 (2),
115129.
Clyne, M. (1997) Multilingualism. In F. Coulmas (ed.) Handbook of Sociolinguistics (pp.
300314). Oxford: Blackwell Publishers.
Cooper, R.L. (1989) Language Planning and Social Change. Cambridge: Cambridge
University Press.
Multilingualism and Language Education in Eritrea 4 91
Coulmas, F. (1994) Language policy andplanning: Political perspectives. Annual Reviewof
Applied Linguistics 14, 3452.
ELF Foreign Information Centre (1977) The EritreanRevolution: 16 Years of Armed Struggle.
Beirut.
EPLF (1977) National Democratic Programme, adopted at the First National Congress of
the EPLF.
EPLF (1987) Political Report andNational Democratic Programme, adoptedat the Second
and Unity Congress of the EPLF and the ELF-Central Leadership.
Erlilich, H. (1983) The Struggle over Eritrea, 19621978: War and Revolution in the Horn of
Africa. Stanford, CA: Institution Press.
Eyob, R. (1995) The Eritrean Struggle for Independence: Domination, Resistance, Nationalism,
19411993. Cambridge: Cambridge University Press.
Fasold, R. (1984) The Sociolinguistics of Society. Oxford: Blackwell Publishers.
Fellman, J. (1979) Language development in Tigrigna. Language Problems and Language
Planning 3, 2527.
Fishman, J.A. (1968) Nationality-nationalism and nation-nationism. In J.A. Fishman, C.
Ferguson and J. Das Gupta (eds) Language Problems of Developing Nations (pp. 3951).
New York: Wiley.
Fishman, J.A. (1974) Language modernisation and planning in comparison with other
types of planning. In C. Kennedy (ed.) Language Planning and Language Education (pp.
3754). London: George Allen and Unwin Ltd.
Garahtu, T. (1996) Language policy and mother tongue instruction (Tigrigna version).
Paper at the First National Conference on Eritrean Languages, Asmara, Eritrea, 1618
August.
Ghebre-Medhin, J. (1989) Peasant and Nationalismin Eritrea: A Critique of EthiopianStudies.
Trenton, NJ: Red Sea Press.
Grabe, W. (1994) Foreword. Annual Review of Applied Linguistics 14, xivxii.
Habte Selassie, B. (1980) Conflict and Interventionin the Horn of Africa. New York: Monthly
Press.
Hameso, S. (1997) The language of education in Africa: The key issues. Language, Culture
and Curriculum 10, 113.
Ingram D.E. (1989) Language-in-education planning. Annual Reviewof Applied Linguistics
10, 5378.
Lambert, R. (1996) Language policy: An overview. Paper presented at the AILA
Conference, Jyvskyl, Finland, 49 August.
Leonard, R. (1988) Popular participation in liberation and revolution. In L. Cliffe and
Davidson, B. (eds) The Long Struggle of Eritrea for Independence and Constructive Peace
(pp. 105135). Nottingham: Spokesman.
Mahmud, S. (1996) Advantages and disadvantages of using the Geez Script in writing
Tigre language (Tigrigna version). Paper at the First National Conference on Eritrean
Languages, Asmara, Eritrea, 1618 August.
Mansour, G. (1993)MultilingualismandNationBuilding. Clevedon: Multilingual Matters.
Markakis, J. (1994) Group conflict and human rights in the horn of Africa. Issue. A Journal
of Opinion 22 (2), 58.
Mazrui, A.M. and Mazrui, A.A. (1992) Language in multilingual context: The African
experience. Language and Education 6, 24.
National Statistics and Evaluation Office (1997) EritreanHousehold Income and Expenditure
Survey: Population and Housing Tabulation Report. Asmara.
Negash, T. (1987) Italian Colonialism in Eritrea 18821941: Policies, Praxis and Impact.
Stockholm: Almqvist and Wiklsell International.
Negash, T. (1997) Eritrea and Ethiopia: The Federal Experience. New Jersey: Transaction
Publishers.
Ottaway, M. (1998) Africans new leaders: Africas solution or African problem? Current
History: A Journal of Contemporary World Affairs 97, 209213.
Patman, R. (1990) Eritrea: Even the Stones are Burning. Trenton: Red Sea Press.
4 92 Journal of Multilingual and Multicultural Development
Peoples Front for Democracy and Justice (PFDJ) (1996) Compilation of papers presented
at the First National Conference on Eritrean Languages. Asmara, Eritrea, 1618
August.
Pool, D. (1997) Eritrea: Towards Unity in Diversity. A Minority Rights Group International
Report. London: Minority Rights Group.
Rubagumya, C.M. (ed.) (1990) Language in Education in Africa: A Tanzanian Perspective.
Clevedon: Multilingual Matters.
Stewart, W. (1968) A sociolinguistic typology for describing national multilingualism. In
J.A. Fishman (ed.) Readings in the Sociology of Language (pp. 531545). The Hague:
Mouton.
Teklehaimanot, B. (1996) Education in Eritrea during the colonial period. Eritrean Studies
Review 1 (1), 122.
Tesfay, A. (1997) Diversity, identity and unity in Eritrea. Paper presented at the
Conference onIdentity andConflict inAfrica. University of Leeds, 1517September.
Trevaskis, G.K.N. (1960) Eritrea, a Colony in Transition: 19421952. Oxford: Oxford
University Press.
Tseggai, A. (1981) The economic viability of independent Eritrea. PhD thesis, University
of Nebraska.
Tseggai, A. (1988) The history of the Eritrean struggle. In L. Cliffe and B. Davidson (eds)
The Long Struggle of Eritrea for Independence and Constructive Peace (pp. 6784).
Nottingham: Spokesman.
UNICEF (1994) Children and Women in Eritrea.
Woldemikael, T.M. (1993) The cultural construction of the Eritrean nationalist
movements. In C. Young (ed.) The Rising Tide of Cultural Pluralism (pp. 179199).
Madison: University of Wisconsin Press.
Multilingualism and Language Education in Eritrea 4 93