Language, Education, and Public Policy in Eritrea
African Studies Review, Apr 2003 by Woldemikael, Tekle M

Given such uneven acceptance of mother-language instruction among different language groups, the question naturally arises: How or why did this differential implementation of the policy came about?

Tigrinya and Tigre

Tigre- and Tigrinya-speakers together constitute the majority (more than 80 percent) of the Eritrean population. It is estimated that both languages originated about one thousand years ago. Tigrinya is a written language and is spoken by over 50 percent of the population. Tigre is spoken in the Red Sea coast and eastern Sudan and Metahit, the name given to the northwestern and western arid lowlands of Eritrea, and is spoken by an estimated one million people. One-third of the Eritrean population speaks Tigre as a primary language. It is the second language of some ethnic groups in Eritrea such as the Bilen, Tigrinya, Nara, and some Saho. Tigre is also a written language. Some Catholic and Swedish evangelical missionaries first wrote Tigre in Ge'ez script around the 188Os, producing texts mostly for religious purposes. They wrote books that describe the Tigre culture, folklore, and grammar. Tigrinya, however, is the more widely published of the two languages, It has a wide range of written texts including novels, poetry, instructional books, and newspapers, and has a longer history of being used for mass communication and instructional purposes. One Eritrean writer contended that if Tigre, rather than Arabic, had been made one of the two official languages (with Tigrinya) in 1952, then Tigre would have been as developed as Tigrinya is now (Ahmed Dahli 1996:
Since at least 50 percent of the Eritrean population speaks Tigrinya as their mother tongue, it is to be expected that the Tigrinya-speakers would have the lion's share of students attending primary schools. But with 68 percent (370 out of 549) of the total number of schools teaching in Tigrinya in 1996-97, the absolute dominance of Tigrinya as a medium of instruction in Eritrean schools exceeded our expectations (table 3). This also meant that the Tigrinya speakers have made a remarkably successful switch from using Amharic as the language of instruction in primary education to using Tigrinya in their communities.

Because of the size of the Tigre-speaking community (31 percent of the population) and the long-term involvement of Tigre-speakers in the Eritrean nationalist struggle, one would have expected strong support for teaching in their mother tongue in predominantly Tigre-speaking areas. Tigre also was one of the first two languages to be introduced and used as a medium of instruction after 1976 in primary schools in the EPLF-controlled base area. The EPLF and the government of Eritrea have shown consistent commitment to developing Tigre as a language of instruction. Both have conducted seminars and public discussions to persuade Tigrespeakers to accept Tigre as a medium of instruction in primary schools in their communities. From 1977 to 1996, the EPLF and the government produced over forty teaching materials, including textbooks, for primary and adult schools, teachers' guides, and some general literature (Negusse Woldu 1996:1-12). The significance of their commitment becomes clear when we compare their efforts with that of the rival liberation front, the ELF. Eyewitnesses have informed me that the ELF banned teaching and writing in Tigre, and that it burned teaching texts that were produced by its cadres who were unaware of the organization's ban on the development of the language.3 It was very surprising, therefore, that only 5 percent of the schools (29 out of 549 schools) were using Tigre as a medium of instruction in 1995-96 and that in 1996-97 only a small number of children (3 percent, or 7139 students of the total number of 240,737 students) were learning in Tigre, as opposed to those taught in Tigrinya (82 percent, or 196,350 students) and in Arabic (12 percent, or 28,452 students in 20 percent of the schools) (tables 3 and 4).
Clearly many Tigre-speaking parents have chosen Arabic instead of Tigre for the primary education of their children. From 1993-94 to 1996-97, there has been an 8 percent growth of schools using Arabic, while those using Tigre have shown a growth of under 1 percent for the same period. As one of the Arabic panel members who himself was a Tigre-speaker stated, "We found people reject Tigre; they said no to Tigre or no to Nara.... They say we want schools in Arabic. The government policy is based on identity...." One reason for the rejection of Tigre may be its close affiliation to Amharic, the national language of Ethiopia. Although Tigre is arguably the closest existing language to the Ge'ez language from which both are derived, I have found most Tigre speakers unaware of this close relationship. A number of Tigre-speaking informants have told me that the Tigre-speakers have been suspicious of the attempts to write their language in Ge'ez script, for they associate the Ge'ez script with Amharic. They view Amharic negatively, as a language of domination that they were forced to learn and speak until Ethiopia was defeated in 1991.
On the other hand, Tigre-speakers often overemphasize the relationship between Arabic and Tigre. Islam in Eritrea is connected with Arabic and evokes strong feelings among Eritreans. The Arabic language came with Islam over one thousand years ago and has influenced the religious and communal life of the Muslim communities in Eritrea. The moral principles of all Muslim Eritreans are derived from the Qur'an. They pray in Arabic. Their religious leaders use Arabic in mosques, teachers use it in Qur'anic schools, and judges use it in administering the Shari'a law in Islamic courts. Thus the demand for Arabic comes from inside of Eritrean communities and not as an outside influence. As Jabir Sa'ad, one of the panel members in the Ministry of Education, described the situation: "We conducted seminars among the elders and religious leaders, and local leaders to overcome resistance on the use of mother tongue. We were told that we were pushing them out of their religion. They say Our children should be able to pray in Arabic'" (interview, Asmara, Dec. 17, 1997).

Thus the use of Arabic in public schools has been growing fast because of the overwhelming public demand for it. From 1993-94 to 1996-97, there was a net increase of thirty-two schools using Arabic, surpassing the rate of growth of all other languages, including Tigrinya (twenty-two net increase), Kunama (zero increase), Saho (seven net increase), and Tigre (eight net increase) (table 3). The lack of public enthusiasm for using Tigre in schools can be seen if we look at the language of instruction in schools that are supported by private initiative. From 1993-94 to 1996-97 only one private school was using Tigre, whereas eleven private schools were using Arabic and seventy-one private schools were using Tigrinya (table 3) (Government of Eritrea 1997:49).
Ahmed Dahli (1996:13) claims that the Tigre language "deserved to be a partner of Tigrinya in the constitution of Eritrea,... for the Tigre language is not less than Tigrinya as a language that can be used for development and growth." The Language Panel members in the Ministry of Education have insisted that for pedagogical and social reasons there will be greater acceptance of Tigre as a medium of instruction in the long run. They cite studies that show that although parents want Arabic, their children like studying in their mother languages because they can better understand the educational materials. The Ministry of Education also has claimed that their own studies show that the sixth graders who study in Tigre perform better academically, know more about their culture, and have a stronger sense of pride in their identity than those who study in Arabic. In addition, they believe that studying in their mother tongue increases the speed with which the students acquire a second language such as English. Nevertheless, the acceptance of Tigre as a language of instruction faces deep resistance. As Dahli states, "The Tigre speakers zealously love their language but are blocked by their blind embrace of the Arabic language and thus, they are depriving Tigre speakers all elements to move freely to capture diverse fields of knowledge and technology."
The Case of Minority Languages

The various speakers of minority languages-including Kunama, Bilen, Saho, Afar, Hadareb, and Nara-have chosen different paths in implementing the language policy directives of 1991. In the process, the preexisting uneven development among the Eritrean languages has grown. The Kunama people have been more receptive than all the other minorities to the new policy of instruction in their mother tongue. In 1995-96, there were twelve schools conducting classes in Kunama. After grade five, the teaching continues in English as a medium of instruction, and Tigrinya and Arabic are taught as subjects. Many Kunama speakers have become bilingual, with a predominance of Tigrinya as a second language. They believe Tigrinya is more valuable for upward social mobility than Kunama or Arabic because it is more widely spoken and has attained a national stature. In the border areas with Sudan, however, Kunama speakers know three or four languages, including Arabic, Kunama, and Tigre and Tigrinya. Even Amharic is spoken in some communities as many children learned it in their primary schools during the Ethiopian occupation.

Kunama is perhaps the most developed of all minority languages. Its development is linked to the coming of missionaries. In order to conduct their missionary work, the Swedes started writing the language around 1867 using the Latin script. Italian Roman Catholic missionaries established residence around Barentu and translated the Gospel into Kunama. In the 1940s they opened an elementary school with boarding facilities in which they used Italian as a medium of instruction. Until the Ethiopian Revolution of 1974, the missionaries offered teaching from first to eighth grade for future priests and teachers, many of whom were sent to Asmara for further training. With the growing political unrest in Eritrea after 1974, the Swedes left the country and did not return until 1978, when they reopened some of their elementary schools, giving instruction in Kunama. In 1988, the schools closed again and did not reopen until the end of the nationalist war in 1991.

In 1997, the Curriculum Department of the Ministry of Education set up a Panel on Kunama language to study the variations in dialects, working on standardization and orthography of the language. The missionaries had developed two dialects in a written form, including the Barka dialect used by the Catholic missionaries and the Marda dialect used by the Swedish evangelical missionaries. The Panel on Kunama followed the pioneering work of these missionaries. It reduced the dialectical differences from seven or eight to two: the Barka and the Marda dialects. "The Voice of the Masses," which started broadcasting in Kunama in the mid-1980s, has influenced the language shift toward the Marda dialect.4 Many government and mission schools in the Kunama region have accepted the government's language policy without resistance.

Like Kunama, Bilen developed as a written language in connection with missionary activity. Missionaries used the language for religious purposes, and some Bilen intellectuals have written essays, language training manuals, dictionaries, novels, grammars, and religious and educational texts in the language. The first text in Bilen language, written in 1882, was a description of the grammar of the language and the culture of the people and was written using the Ge'ez script (Kiflemariam Hamde 1996:8-12).
The Bilen speakers live in close proximity with Tigrinya-speaking Christians and Tigre-speaking Muslims, and they tend to know both Tigre and Tigrinya as their second and third languages. Bilen-speakers are divided almost equally between Muslims and (mostly Catholic) Christians. Rather than demanding the use of the Bilen language as the medium of instruction in their communities, those Bilen who are Muslim have opted for Arabic primary schools and those who are Christian have opted for Tigrinya. The present underdevelopment of Bilen as a language of instruction is partly a result of the manner in which education was introduced to the region. The beginning of formal education in the area came with the setdement of the missionaries around Keren in the late nineteenth century (Olsen 1997:11). Although missionaries were the first to write Bilen using Ge'ez script, they never used the language for teaching purposes. Instead, as in most schools in Eritrea, they taught in Italian during the Italian period, in Tigrinya and Arabic during the British administration, and in Amharic during the Ethiopian rule (Government of Eritrea 1996:49
There has been a controversy over which script to use for writing Bilen (Kiflemariam Hamde 1996:11-12). There is a compelling argument that Bilen can be written and read easily with only a slight modification of the Ge'ez script. In 1992, some interested Bilen-speaking individuals produced two grammar books and a Bilen-English dictionary using Ge'ez script. The EPLF and the government of Eritrea have used the Latin script, however. In a Conference on Eritrean Languages sponsored by the EPLF in 1985, the Front made a decision to use Latin script in writing Bilen and all other minority languages, putting the policy into practice in 1987. The explanation for the choice of Latin over Ge'ez script is mostly political, having to do with the decision of the Eritrean government to accommodate public sentiment rather than to build on the existing experience of writing the Bilen language in Ge'ez script. Like other Muslims, the Muslim Bilen see Ge'ez as part of the Christian religion and the Ethiopian government. The government hoped that both Christian and Muslim groups would use the Latin script without feeling that one group was favored over the other.

In 1993 the government set up a committee to deal with technical and scientific terms and to standardize the Bilen language. By 1996 texts for math and science were published in Bilen, but there still was no single school that used Bilen as a language of instruction. The government schools gave instruction in Tigrinya and the teachers were Tigrinya-speakers. Since 1997 there has been one first-grade class in a government school offering instruction in Bilen language following the government-recommended curriculum, script, and orthography.

Many of the other speakers of minority languages in Eritrea, including Afar, Saho, Nara, Hadareb, and some Bilen, prefer instruction in Arabic to their mother tongues (tables 2 and 3). As with the Tigre-speakers, the resistance of the communities to giving instruction in their mother tongue is related to the popularity of Arabic as a language of choice among Muslims On the other hand, the Saho, although mostly Muslim, have partially accepted the Eritrean government's policy on language use in their schools. Like most Muslim Eritreans, the Saho have ambiguous feelings about the use of their mother tongue in schools, and many have opted for schools in which Arabic is the medium of instruction (Government of Eritrea 1997). There have been, however, a large enough number of schools teaching in Saho to warrant further analysis (see tables 2 and 3). In 1996-97, there were eighteen government-run primary schools, constituting 3 percent of all the schools, offering instruction in Saho. This represented an increase of seven schools compared to the period from 1993-94 to 1996-1997 (table 3). These schools were mostly new, for there were hardly any schools in Saho districts before independence.
There are three possible explanations for this partial success. First, the EPLF and the government have followed a flexible policy toward the Sahospeakers. For example, although the EPLF started using Ge'ez script when it began using Saho as a language of instruction in 1983, it stopped its policy right away when it faced public resistance. In 1985, the EPLF introduced Latin script, claiming that the problems of pronunciation were minimized when Saho was written in Latin. A more likely reason for the switch to Latin script was political, however, because the Saho speakers, just like the Tigre- and Bilen-speaking Muslims, associate Ge'ez script with Ethiopian Christian faith and practices. By 1990 the EPLF had prepared Saho textbooks in all subjects using Latin script. It was able to offer instruction in Saho in grades one to five in predominantly Saho regions.

Second, the Saho are a small nationality living in close proximity to Tigrinya-speaking Christians. They interact with Tigrinya speakers in their everyday life, and most Saho speakers are bilingual and trilingual in Saho and Tigrinya and/or Tigre. Third, the Ministry of Education presented many workshops, seminars, and lectures to communities, religious leaders, and village elders to explain its policies of using one's mother tongue.
In many cases, resistance to the policy of use of the mother tongue in Eritrean schools has to do more with seeking direct economic and political rewards and benefits from learning regional and international languages such as Arabic and English In fact, one can make a compelling argument that other than Tigrinya and Arabic, which have been designated as working languages, there is hardly any material benefit attached to being literate in the other Eritrean languages. Since English, Tigrinya, and Arabic are widely used in official communication, a person knowing all or a combination of these languages has an advantage in gaining economic and political opportunities and rewards in the modern sector of Eritrean society. In fact, one can function effectively in the lower official circles of the Eritrean government knowing only Tigrinya but not knowing only Arabic or English. Tigrinya is the language of military training and communication. In the higher circles of the governmental hierarchy, however, English is an essential medium of internal and international communication. Thus, learning Tigrinya, English, and Arabic are advantageous in the job market as well as in governmental and international transactions.

Moreover, learning Arabic and English opens up opportunities for employment outside of Eritrea. Since the citizens whose mother tongue is Arabic are few, the existing system works to the advantage of the Tigrinya speakers at the expense of other language speakers. In addition to learning their mother tongue, the Tigrinya-speakers have to learn English and Arabic while the non-Tigrinya-speakers have to learn English, Arabic, and Tigrinya. Since Tigrinya is a written language with a comparatively more developed body of literature, and is widely used in official and unofficial transactions, the incentives for wanting to learn and master Tigrinya are great. Non-Tigrinya-speakers, whose mother languages are still at an early stage of being written and developed, are expected to learn Tigrinya, Arabic, and English. Thus, for non-Tigrinya-speakers, learning in their mother language is an additional burden they have to bear. As an Eritrean expert remarked, instead of focusing on teaching major foundational and essential subjects, most Eritrean schools devote a large part of their time to teaching the various required languages; therefore, schools have practically become "institutes of languages" (Berharie Woldemichael 1995:38). That more than 50 percent of the material taught in the Eritrean schools in 1996-97 was language-related attests to this fact (Hoyle 1997:128).
Conclusion

Most writers are pessimistic about state-sponsored social engineering (Scott 1998; Fardon & Furniss 1994; Van Binsbergen 1994; Laitin 1992). They believe that centrally planned language policy will often have little effect. James Scott (1998) has presented an incisive analysis of why large-scale plans to redesign and improve social and economic arrangements in various states have failed with tragic consequences to millions. He argues that large-scale authoritarian plans have gone wrong when they tried to enforce schematic visions that violated the complex interdependencies that were not-and could not-be fully understood. He also argues that the success of designs for social change depend significantly upon the recognition that local, practical knowledge is as important as formal, epistemic knowledge (Scott 1998). As Fardon and Furniss state poignantly: "Experience suggests that the state can sponsor dramatic language reform only under exceptional circumstances (as in the revival of Hebrew in Israel) but that its capacity to engineer perverse effects is always remarkable" (1994:24). They recommend language sponsorship that is pragmatic, with "a modest and constantly revised understanding of the changing patterns of language use combined with a willingness to seize opportunities that arise at levels between the state and the immediate locality" (1994:24).
The Eritrean experience is still at an early stage. In order to succeed in its social engineering, the government has shown considerable flexibility in its application of the language policy. And the policy seems to follow some of the guidelines that experts such as Richard Fardon, Graham Furnish, Wim Van Binsbergen, and David Laitin have recommended. For example, Fardon and Furnish praise the Nigerian policy of assuring knowledge of at least one and often two or three major languages at the secondary-school level. This policy is aimed at encouraging language versatility as well as reducing the link between language, ethnicity, and political affiliation. Eritrea, however, has to find ways of making knowledge of Eritrean languages other than Tigrinya and Arabic emotionally and materially rewarding for its citizens. There is also a need to minimize the trend of turning Eritrean primary schools into institutes of languages.

Whether Eritrea will follow the lead of other countries and abandon its language-planning policy only time will tell. So far, its attempt has had a mixed outcome. It has been successful in gaining acceptance of its mother-tongue policy among the Tigrinya-, Kunama-, Arabic- (Rashaida), and to a limited extent, Saho- speakers. Others, mostly but not exclusively Muslims, including Tigre, Nara, Afar, Bilen, Hadareb, and some Saho, have resisted the government's initiative and have opted for Arabic, and in the case of Christian part of the Bilen-speakers, Tigrinya.

The outcome that Eritrea expects from its language policy is that every person will be well versed in her or his mother tongue and also will have the ability to communicate in one of the working languages of EritreaArabic and Tigrinya-and in the international language of English. This expectation is now close to realization and actually resembles the outcome that most African countries aim for, aptly characterized by David Laitin (1992, 1994) as the "3 ± 1" solution. According to Laitin, market forces in many African countries demand that upwardly mobile citizens have facility in their vernacular (primary language), an African lingua franca, and a colonial or international language. If the individual's vernacular is the same as the lingua franca, then the person will need to learn only two (3 1) languages. If the person's vernacular is different from the language taught in the region's primary schools, the citizen has to learn four (3 + 1) languages (Laitin 1992, 1994; Pardon & Furnish 1994). In Eritrea, a "3 ± 1" outcome would give all Eritreans the ability to communicate effectively with each other and with most of the rest of the world.

In the long run, Eritrea might abandon its policy of giving instruction in all Eritrean languages, for the simple reason that the increasing burdens of economic, social, and cultural development encroach on the limited resources available for language-planning policy and implementation. As Williams points out, in all cases of language planning, the delicate choice of whether to promote one; or many African languages in the educational domain and public agencies of new states is becoming less and less a matter of free choice. The dominant trend toward the globalization of economic, political, and cultural relationships is a major factor in language choice. But there is also a countertrend, which emphasizes the value of cultural diversity and the worth of each specific language and is especially concerned with endangered languages in Africa. Since the coming of independence to African countries, the truly endangered languages in Africa are threatened more by policies favoring the strongest African languages over the others than by strong international languages (Williams 1996:50).
In the case of Eritrea, the threat comes from Arabic and Tigrinya, the two most dominant languages. With an overwhelming consent of the Muslim and the Christian communities, Arabic and Tigrinya are capable of replacing most of the Eritrean languages. Given a chance, Arabic could easily become the dominant language of instruction and communication among Muslim Eritreans, in the same way that it has come to dominate the Sudan and North Africa; the same can be said for Tigrinya among the Christians. The government is fully aware that Arabic, as an international language and the language of countries surrounding Eritrea, may eclipse Tigrinya, which is only a regional language at best. This does not mean that the various mother tongues will disappear, but it does mean that their languages will not develop to their full potential within the Eritrean political spectrum. And while people will not lose their distinctive identities, this situation might create social pressures that will force Eritrea to revisit its recent troubled history of the Christian-Muslim split, which arguably has fractured the society for the last fifty years. Whether Eritreans' sense of unity is strong enough to withstand divisions based on religion, language, and other cultural sentiments is an open question. What we can say confidently is that the government of Eritrea has deliberately followed a policy that is variable enough to accommodate the conflicting demands of Eritrean communities without abandoning its policy of teaching children in their mother tongue at the primary school level.

Acknowledgments

I would like to thank members of the Ministry of Education who provided me with substantial information for this research, including Tesfamikael Gerahtu, head of the Curriculum Branch, the late Negusse W. Ghebru and all the coordinators of Panels of the Mother Language Instructions in the Curriculum Branch, arid Isayas Merhatzion of the Department of Research and Human Resources Development. I am also grateful to Ababa Habtom, head of the Preschool Panel, Ministry of Education, and Zemehret Yohannes, director, Ministry of Information and Culture, Government of Eritrea. I have benefited greatly from the comments I received from M. Crawford Young, Tom Killion, and Salaheddin Nur on an earlier version of this article.

Notes

1. I conducted an interview with Kidane Tseggai, a representative of the Ministry of Local Administration of the Government of Eritrea, on July 9, 2001. From the information I gathered, the approximate population of Eritrea in 2000 was 2,926,401. This figure excluded soldiers and those who lived abroad in refugee camps and as exiles and immigrants around the world. See table 1 for details.

2. While it is common to use the same name for a language and an ethnic group in Eritrea, the use of Tigrinya to refer to a language and to the people who speak it is a new development, which made its appearance since the establishment of the state of Eritrea. It is beyond the scope of this paper to speculate why the government started to refer to the Tigrinya-speakers as the Tigrinya people.

3. This information is based on a conversation I had on December 16, 1997, in Asmara with a former ELF official who is now an official with the government of Eritrea.

4. "Dimtsi Hafash" (Voice of the Masses) started broadcasting in Tigrinya, Tigre, and Arabic in 1979 and added programming in Afar, Kunama, and Amharic in mid-1980s. See Killion (1998:351-52).

References

Ahmed Hasan Dahli. "Philosophical and Political Dimension of Language." Paper presented at the Conference of the Eritrean Languages, Asmara, August 16-18.

Berhane Teklehaimanot. 1996. "Education in Eritrea during the European Colonial Period." Entrean Studies Review 1 (1): 1-22.
Berhane Woldemichael. 1995. "Education for the Pastoral Communities of Eritrea." Research Study of the Ministry of Education and Swedish Save the Children. Asmara: Ministry of Education.

Eritrean Peoples Liberation Front (EPLF). 1982. "Education uunder the EPLF." Eritrea: Branch of Education, EPLF.

_____. 1988. "Building a National System of Education: Basic Tasks and Main Problems." Eritrea: Branch of Education, EPLF.

Fardon, Richard, and Graham Furniss. 1994. "Introduction: Frontiers and Boundaries." In Fardon and Furniss, eds., African Languages, Development and the State, 1-29. London: Routledge.

Government of Eritrea. 1996. "Government Schools Using Arabic as a Medium of Instruction." Asmara: Ministry of Education.

_____. 1997. "Eritrea Basic Education Statistics and Essential Indicators, 1995-96." Asmara: Ministry of Education.

Ghebru, N. W. 1996. "Mother Tongue Education: MT Instruction, Problems and Means of Tackling." Asmara: Ministry of Education.

Habtu Ghebre-Ab. 1993. Ethiopia and Eritrea: A Documentary Study. Lawrenceville, N.J.: Red Sea Press.

Heine, B. 1992. "language Policies in Africa." In R. K. Herbert, ed., Language and Society in Africa. Johannesburg: Witwatersrand University Press.

Hoyle, Peggy Ann. 1997. Eritrean National Identity: The Role of Education and the Constitution. Ph. D. diss., University of North Carolina.

Idriss Abback. 1996. "Rough Draft on the Formation of Central Languages and Traditions in Eritrea." Paper presented at the Conference of the Eritrean Languages, Asmara, Aug. 6-18.

Kiflemariam Hamde. 1996. "Culture, Language and Society: The Role of Language in Social Development." Paper presented at the Conference of the Eritrean Languages, Asmara, Aug. 16-18.

Killion, Tom. 1998. Historical Dictionary of Eritrea. Lanham, Md.: Scarecrow Press.

Laitin, David D. 1992. Language Repertoires and Stale Construction in Africa. Cambridge: Cambridge University Press.

Negusse Woldu. 1996. "The Difficulties of Teaching in Tigre Language." Paper presented at the Conference of the Eritrean Languages, Asmara, Aug. 16-18.

Olsen, Karin. 1997. "Implementation of Mother Tongue Education in Eritrea." Unpublished.

Provisional Government of Eritrea. 1991. "Declaration of Policies on Education in Eritrea." Legal Notice, Number 2-1991 (Oct. 2). Asmara: Department of Education.

Scott, James. 1998. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven: Yale University Press.

Sherman, Richard. 1980. Eritrea: The Unfinished Revolution. New York: Praeger.

Trevaskis, G. K. N. 1960. Eritrea: A Colony in Transition. Oxford: Oxford University Press.

Van Binsbergen, Wim. 1994. "Minority Language, Ethnicity and the State in Two African Situations." In Fardon and Furniss, ed., African Languages, Development and the State. London: Routledge.

Wedekind, Klaus, and Nadine Dutcher. 1996. "Assessment of Mother Tongue Education in Eritrea." Published in Collaboration with the Language Panels of the Curriculum Research and Development Division of the Ministry of Education. Asmara: Ministry of Education.

Williams, Colin H. 1996. "Ethnic Identity and Language Issues in Development." In Denis Dwyer and David Drakakis-Smith, ed., Ethnicity and Development: Geographical Perspectives. West Sussex, U.K.: John Wiley and Sons.
Zemehret Yohannes. 1996. Eritrea Profile 3, no. 33 (November): 6.

Tekle M. Woldemikael is an associate professor and Chair of the Department of Sociology and Anthropology at the University of Redlands, Redlands, California. He is the author of Becoming Black American: Haitians and American Institutions (AMS, 1989). He has published a number of articles on ethnicity and nationalism in Eritrea and on Eritrean and Ethiopian refugees in the United States.

Copyright African Studies Association Apr 2003
Provided by ProQuest Information and Learning Co