Language Attitudes and Schooled Education
The Case of Karagwe
ÅsaWedin
Högskolan Dalarna and Örebro University
Abstract
In this paper I argue that language policies for education have effects on pupils’ educational possibilities. With the case of Karagwe district in Tanzania, the paper suggests that the policy of ‘Swahili only’ in primary school education in Tanzania favors the small minority of the school children that live in a context where Swahili is used. This appears to lead to inequality in pupils’ chances in education and to a low level of achievement of academic content in schools. This also promotes the development and use of what I call safety strategies among teachers and pupils that hide failure and prevent pupils’ learning.
1. Introduction
In a poor country and in a setting with poor educational opportunities, it is important that resources devoted to education, both economic and human, are effectively used and that education is relevant for pupils. This is particularly the case in Karagwe district in Kageria region in Tanzania. It is a well-known fact that educational outcome in primary school in Karagwe is poor. It is also well known that there are many factors behind this; such as lack of well-trained teachers, teaching materials and relevant school buildings (see for example Wedin [2004]). However, the role of language medium in primary schools in Tanzania is seldom discussed. The fact that the use of English as medium in secondary school, where neither teachers nor students master English at a sufficient level, causes great problems for the educational outcome in secondary school in Tanzania, is discussed elsewhere (see for example Lwaitama and Rugemalira [1990], Trappes-Lomax [1996]) whereas the question of language medium on primary school level, is seldom a topic for discussion. Among educational policy makers in Tanzania, two assumptions seem to be commonly held, namely that most children in Tanzania master Swahili when they enter primary school and that using a second or foreign language as medium in school is a good way of teaching students that language. Thus, there seems to be an assumption that in the case children do not know Swahili, the use of ‘Swahili only’ in primary school is an effective way to teach them the language and that an effective way of teaching children English is to make it the
medium also of primary school. In this article, I aim at showing that neither of these assumptions hold true, particularly not in Karagwe.
Karagwe district is situated in the north-west corner of Tanzania, on the borders to Uganda and Rwanda. The people in the area, the Banyambo, number about 500 000. Karagwe was traditionally a hierarchical society, with ruling pastoralist clans and a number of agriculturalist clans. The king of Karagwe was one of the strongest kings in the Western Lacustrine area from the 16th up to the middle of the 19th century. With the arrival of first
the Arab traders and later the European colonizers and missionaries the strength of the king of Karagwe had started to decline. Iron mining and cows were important assets for the king. (For further information on the history and traditions of the Banyambo, see Wedin [2004]). The Banyambo spoke Runyambo, which is closely related to languages such as Kinyarwanda, Kinyankole and Ruhaya. The Arab traders introduced Swahili and also Arabic for religious matters. As the area was first colonized by the Germans and then later by the British, both German and English influenced language patterns in the area.
Today, more than 30 years after independence and the creation of the state of Tanzania, most Banyambo over ten years of age are functionally bilingual in Swahili and Runyambo. This means that both languages are frequently used and that most adults can use both languages for their daily communicative needs. English is a high status language that is mastered by very few in the upper levels of society, such as those in higher administrative positions and others with higher education.
During the period of 1999-2003, I carried out a study of literacy practices in Karagwe. I studied people’s use of and attitudes to written language and tried to find ways to make literacy education in school more relevant by linking what happens in school with what happens in society. I spent a total of ten months in the area, and worked on five field studies, observing literacy practices in different settings; in schools, offices, homes, churches, market places and in other domains where people spent their days. One important obstacle for effective and relevant literacy education I identified was language patterns and language choice in primary school (see Wedin [2004, 2005]).
2. Education in bilingual settings
The situation in Karagwe with a multilingual setting, with one language being used mainly in private domains and one mainly being used in official domains is not unique in the world; this is the case in many settings all over the world. In Karagwe nearly all children master Runyambo when they begin school but not Swahili. I estimate that nearly all children in the countryside and perhaps more than 90 % of the children in towns do not know more than a few words in Swahili when they begin school. Neither is this very uncommon in the world. There are several settings in the world where children master one language when they enter school and are supposed to master another language when they finish primary school. This is thus a bilingual setting where children are supposed to learn a second language through schooling.
The research on bilingual education is quite impressive today. Studies in similar settings as Karagwe, a poor setting where children are supposed to learn a dominant language through primary schooling, have been carried out by among others Fafunwa (1990), Hornberger and Chick (2001) and Ndayipfukamiye (2001). A study by Tomas and Collier (1997, 2001) showed very impressively what earlier studies had indicated, that the most important factor for children’s success in bilingual education, is that their mother tongue is strongly supported throughout their schooling (see also Cummins [1996, 2000], Gibbons [2002]). This is particularly so in the case where children are supposed to learn a second dominant language while they simultaneously are expected to acquire basic academic education, such as the three R’s (reading, writing and arithmetic). It should not be very controversial to claim that children learn best when they are taught in a language that they understand. Another important factor for children’s success in second language acquisition in school is that the language teaching is based on their earlier knowledge and experiences. This means that, in the case of Karagwe, children should gain a lot if they were to be taught Swahili as a second language, which is actually what it is for these children, while simultaneously having support in their mother tongue, Runyambo. According to official school policies, Swahili is treated as the children’s first language, and therefore appropriate teaching methods are not discussed at all.
Before we look further into the question of language use in primary school in Karagwe, we will look a bit closer at language history and contemporary language attitudes in Karagwe.
3. Language attitudes and language use in Karagwe 1500-2000
As earlier mentioned, Runyambo was the language traditionally spoken by the Banyambo. This was a language that showed ethnicity, by differing those within the area, in this case Karagwe, from those outside. Both the ruling clans and the agriculturalists, the commoners, spoke the same language (Hydén [1968], Seitel [1972], Carlsson [1986]). As men traditionally traveled to earn money and as the ruling clans married with ruling clans from the neighboring areas, they probably also mastered some of the neighboring languages to some extent. During the 19th century, when the Arab traders arrived in the area, one may
assume that those who came into contact with them also learned some Swahili. With the coming of the westerners there was a turn in both language use and language ideologies. The first westerners are believed to have arrived in Karagwe in 1864 and by the turn of the century the influence of the Germans was seen mainly in the domains of education, administration and religion. The traditional Omuteko-schools, that were arranged by the Abakama, the kings, where young boys had learned war-fare, traditions, hut-construction and good manners among other things, where forbidden by the Germans and the role of the king was substantially weakened. The last traditional king, Ntare VII, was hung in front of the post-office in Bukoba (Hydén 1968). Instead of the Omuteko-schools, a Western type of schooling was introduced, with Western Christian worldviews. This was the period when the first grammars and dictionaries of Swahili were written and when Arab script was exchanged by Roman script for the official writing of Swahili. As Swahili was the language used in schools and as school results were used for selection for minor administrative posts,
Swahili came to be used as a gate-keeper for economic status. German was still reserved for only a very small minority, for example for some selected sons of the king, and was used for selection for higher posts. This is where we can see the turn where language was made a marker for status through education.
With the take-over by the British, who seized power in the Kagera region 1916, the area west from Lake Victoria, which is a few years earlier than in the whole of Tanganyika, there was another change in language policies. The British colonizers wanted to keep a distance to the powerful Muslim Swahilis on the coast. Therefore they introduced local languages in primary school with a shift to English after a few classes. In the case of Karagwe this meant that Ruhaya was used in the first years of schooling, Ruhaya is close to Runyambo and Banyambo may read Ruhaya without many problems. Thus, the role of Swahili was played down during the British period, 1916-1961, which resulted in an increase in the status of both Runyambo and Ruhaya. At the same time, English became a high status language as the gate-keeper for higher education and posts in the administration. With independence, and the strong promotion of Swahili in official domains as a uniting tool for the new state, Tanzania, there was a decrease in status for both Runyambo and English. Swahili was developed, new terms coined, standardization was carried out and Swahili was to be used throughout primary school. Officially, Swahili was also to exchange English as medium of instruction in secondary and tertiary education (Trappes-Lomax [1996], Rubagumya [1990, 1994]). This has, however, not been implemented during the more than 40 years of independence. From the 1980’s and following the Structural Adaptation Plans we have seen a strong increase in the status and importance of English all over Tanzania. This has not been less so in Karagwe with its close connections to Uganda and Rwanda. Throughout the latest centuries it has been common among ruling clans and rich families in Karagwe to send their children to Uganda for education, invariably English medium education. This education has always been highly valued in the area. Although quite a high portion among the élite in Karagwe has attended high posts in Dar es Salaam, both politically and privately, there has always been some sense of suspicion against the coastal Swahilis among the Banyambo. Thus it is not surprising that when English medium primary schools started to mushroom in the 1990’s, Karagwe was an area that saw many private English medium primary schools being constructed.
In Table 1 (from Wedin 2004) there is an overview of language policies in Karagwe during the years 1500 – 2000. The table is constructed with the theory of triglossia as a base, see for example Martin-Jones (1988) and Rubagumya (1991), which takes social inequality as the root difference in status for different languages. The theory distinguishes between high varieties (H) and low varieties (L). In the case of the history of Karagwe we see that the situation is complex.
Table 1. The language policy in Karagwe 1500 – 2000
Period Runyambo Swahili English German 1500 – 1900 1900 – 1916 1916 – 1961 1961 – 1980 H L L2 L Additional L1/H2 L1 H1 -H H2 -H1
-1980 - 2000 L2 L1/H2 H1 H = language with high status, a dominant language.
L = language with low status, a dominated language.
L1 /L2 = one of the low languages (L1) was dominating the other (L2). H1/H2 = one of the high languages (H1) was dominating the other (H2). = the language was gaining new domains.
= the language was losing domains.
Thus, in the beginning of the third millennium, the language situation in Karagwe is one where English is the high status language, although very few know more than a few words in English. Swahili is still very strong but those who can take every chance to put their children in private English medium schools.
If we look closer into how languages are used in Karagwe today we get a clearer picture on language patterns in the area. In Table 2 (from Wedin 2004) an overview is given on language use in different domains in contemporary Karagwe. We can see that Runyambo does not hold many domains but that it holds the domains where most inhabitants spend most of their time. For most people in rural areas the only place where they are likely to meet Swahili outside school, is occasionally in the market-place and if they happen to visit a hospital or an office of any kind. As most Banyambo go to school and as children actually acquire functional skills in Swahili before they leave standard seven, most people can be assumed to master Swahili. However, one has to remember that the increase in number of school drop-outs and the number of children who do not go to school, may result in an increase in the number of Banyambo that do not know Swahili, at least in rural areas.
Table 2. Language use in different domains in contemporary Karagwe1
Domain Level/Event Language(s) mainly used Commercial National/International Swa/Eng
Regional Swa/Ru
Small trade Ru/Swa Education Informal Ru (Swa)
Primary school Swa/Ru2
Vocational training Swa/Ru/Eng Secondary school Eng/Swa Tertiary level Eng/Swa Adult education Swa/Ru Official National/International Swa/Eng
District/Regional Swa/Ru/Eng
Informal Home Ru (Swa)
1 Key: Swa = Swahili, Eng = English, Ara = Arabic, Ru = Runyambo, including also Ruhaya in the
case of cultural domains. Languages are given in order of estimated frequency that is Ru/Swa means that Runyambo is used more often. Languages given in brackets are used in certain cases, such as in some families, in towns or in some congregations. In regional trade other languages such as Kinyarwanda and Luganda are also used but they are not mentioned in the table.
Neighbors Ru (Swa) Workplaces3 Ru/Swa
Cultural Worship Swa/Ara /Ru Literature Swa (Ru)
Music Swa/Ru /Eng
TV/Video Swa/Eng/Ru Social events Swa/Ru Newspapers Swa/Eng Village/Family related Swa/Ru
Swahili is the dominant language for official domains at the local and regional level, although most officials also master Runyambo and usually switch to that language with poor Swahili speakers. However, this is very often done in a humiliating manner. English is mastered by very few and its status is very high. We see in the table that English is used in most domains, mainly on regional and national level. The high status of English is seen in how people strive to “sprinkle” their talk with English words or expressions. Although someone may only know a few words, he or she is likely to use them to show their command of English. People of higher status commonly use English expressions in their talk without caring about whether the interactant understands them or not.
The different status may also be understood by observing language attitudes among teachers. In a questionnaire I asked teachers to express their language use and their language attitudes. 51 teachers from seven schools answered the questions. In table 3 the results from the questionnaire are given (from Wedin 2004).
In the table we can note some interesting information on teachers’ language use through which we can obtain some understanding of their language attitudes. All teachers claim to use Swahili in classrooms. Many also claim to use English. Only three claim to use Runyambo at all. We should remember that this includes teachers of pre-school and the first years, where most children do not understand Swahili. This is what we could expect teachers to answer as the setting where they are asked is the school and as this is according to regulations. If we look at teachers’ self-report on language use in the teachers’ office we can see that teachers claim to use ‘Swahili only’ or Swahili and English. However, as I spent much time in teachers’ offices, among other places, I know that the languages they actually use are Swahili and Runyambo, approximately equally. None of the teachers answering the questionnaire mastered English enough to use it for conversation of the type used in their offices. Also when we look at other domains and compare with the results from my observations in different areas, I state that teachers over-report on their use of Swahili and English while they under-report on their use of Runyambo. For example, Runyambo is the most frequently used language in market places with Swahili as the second, while half of the teachers claimed to only use Swahili there. In homes Runyambo is the language for daily communication and for interaction with children, also in teachers’
3 Workplaces here include both the most common workplace, the mashamba, market places and
homes, while twelve teachers reported to use only Swahili. The far majority of the teachers, 43 out of 53, answered that the language they preferred to use was Swahili.
Table 3: Teachers’ self-report on language use in different domains 4
Domain Swahili Swahili/
English
English Swahili/ Runyambo
Runyambo
In classrooms 32 18 0 3 0
In the school yard 36 1 0 7 2
In teachers’ office 43 9 0 0 0
At home with children 12 0 0 235 15
With neighbours 3 0 0 25 25
At home to elders 1 0 0 18 34
At the market and in shops 26 0 0 22 3
Preferred language to use 43 2 1 2 1
The answers of these teachers are in accordance with the official policy of Swahili-only in school and the results show that these teachers are officially loyal to this policy. The fact that the questionnaire was carried out by the researcher in their schools, probably had the effect that teachers rather answered even more in accordance with official policies than they would probably have done if asked elsewhere, for example in their homes. However their actual behavior, using Runyambo to a great extent also on the school compound, showed that there is a discrepancy between what teachers expressed in their answers and their actual linguistic behavior. The pattern of language use and the over-estimation of their use of high-status languages tell us something about their language attitudes. These attitudes are important to take into consideration when language use in the schools is discussed.
4. Children’s language skills in primary school in Karagwe
Through interviews with children, teachers and parents and through observation in different areas where children spend their time, I have found that children in Karagwe generally do not master Swahili until the last classes in primary school. I carried out formal interviews with children from different classes, apart from all informal contacts I had with numerous children. According to my results it is not until in standard four or five that most students master Swahili enough to be able to interact in ordinary conversation. This was verified by teachers. I asked teachers, for example, if they told children in lower classes stories but they answered that they could not do so as the children would not understand a story told in Swahili until earliest in standard five. This means that teachers’ assumptions in this case coincide with mine. However I have found that probably nearly all students that finish
4 N = 53 although not all teacher answered all questions. In the table Runyambo stands for both
Runyambo and Ruhaya. No teacher used the word “Runyambo” for the local language, instead they used Kilugha (a Swahili derogative for language), Kinyambo (Swahili for Runyambo), lugha ya
mama (Swahili for mother tongue) or lugha kieneji (Swahili for vernacular or local language). 5 One of the teachers also reported the use of Kinyarwanda with children and elders in the home.
standard seven, that is the whole primary school, do master Swahili afterwards on a functional level. I carried out interviews with some of the students in standard seven with the poorest results and they could all discuss with me in Swahili on a level that I estimated as proper for their age. Some of the most low-performing students I also met in follow-up interviews one year after they had finished school and they could still discuss with me in Swahili at an age-proper level.
The fact that children did not know Swahili in lower classes was made clear for me in different ways. I carried out formal interviews with pupils from grades three to seven, both high-performing and low-performing pupils. I also talked to children in different settings such as in homes, in the street and in the fields. When I tried to talk to younger children, older children or adults often told me that the child did not know Swahili, usually showing some embarrassment for this.
To acquire a picture of their level of understanding of what was taught at school, I observed classes, studied the pupils’ exercise-books and carried out follow-up interviews a few days after the lessons. Generally the understanding was very poor, which of course has many reasons. However, the fact that nearly no children, apart from some in the highest grades, showed any understanding of what was taught is to me a sign of linguistic problems. They did not seem to have understood what had been taught at all. What also became clear was that pupils and teachers had developed what I call safety strategies. These strategies were developed by teachers and pupils to create a picture of successful learning and to hide failure. Teachers tended, for example, mainly to address successful learners and not to follow up low results in exercise-books. They also used interactional patterns that guided students to give the expected answers without actually understanding. Less successful students tended to hide in the back of the classroom, not handing their exercise-books in and even failed to show up at school at all. For further discussion on safety strategies used by teachers and pupils in primary schools in Karagwe see Wedin (2004).
5. Conclusions
The two assumptions that I wanted to contest in this paper were: Children in Tanzania master Swahili when they start school
Using a second or foreign language as a medium in primary school, is an effective way to teach that language
In the case of Karagwe, my research shows clearly, not only that the far majority of children in Karagwe do not know more than perhaps a few single words in Swahili when they start school but also that it takes many years in primary school before they know enough to actually be able to understand the academic content of what they are taught. My research also shows clearly that pupils did not understand much of what they were taught and that the language used was one important factor contributing to this.
I claim that the outcome of the language policy of Swahili-only in primary school in Tanzania, is one important factor behind the low outcome in primary school in Karagwe. I also claim that the Swahili-only policy together with the policy of allowing private schools
only as English medium, has strong effects on the sorting of pupils. The result of this is that the conditions that were introduced by the colonizers making language an important factor for sorting students in school, by constituting language as a gate-keeper for higher education and job opportunities and a marker for status, has further become strengthened by language policies in Tanzania after independence. As I see it, it is not only unjust that language background is made to constitute a sorting factor for schooled education but it is a huge waste of resources that so much effort are ineffectively used in education. In Karagwe, education holds an extremely high importance. Vast resources are put into education, many teachers work extremely hard, many children spend lot of time in school with hard work and many parents spend more money than they can afford to educate their children. However, I claim that through the prevailing language policies much of this is wasted. If the question of language medium in primary school could be openly discussed so much better outcome could be the result.
If policy makers would admit that most pupils have another language than Swahili as their mother tongue, then the language question could be open to discussion. There are, for example, much relevant research that show how second language education for children in primary school could be made relevant and effective. There are examples of successful teaching both of a second language and academic content in situations similar to the one in Karagwe. There are even examples that children have been successfully taught in a second language, such as would be the case of Swahili in Karagwe, while they simultaneously have also learned the academic content. There is no reason why this could not be done also in Tanzania, if policy makers would start to discuss the situation with the actual situation as a base. However, I also want to stress that the new ways of using language in primary school in Tanzania should be developed in cooperation with Tanzanian teacher educators and linguists. To copy examples from other contexts uncritically should be avoided.
REFERENCES
Carlson, Robert G. 1989. Haya vorldview and ethos: an ethnography of alcohol production
and consumption in Bukoba. Doctoral thesis, University of Illinois, Urbana-Champaigne.
Cummins, Jim. 1996. Negotiating identities: education for empowerment in a diverse
society. Ontario: California Association for Bilingual Education.
_______2000. Language, power and pedagogy. Bilingual children in the crossfire. Clevedon: Multilingual Matters.
Fafunwa, Babs A. 1990. Using national languages in education: a challenge to African educators. In: Karsten Legère (ed.) The role of language in literacy programmes with
special reference to Kiswahili in Eastern Africa. Bonn: German Foundation for
Gibbons, Pauline. 2002. Scaffolding language scaffolding learning. Teaching second
language learners in a mainstream classroom. Portsmouth, NH: Heinemann.
Hornberger, Nancy. and Keith Chick. 2001. Co-constructing school safetime. Safetalk practices in Peruvian and South African classrooms. In: M. Heller, and M. Martin-Jones, (eds.) Voices of authority. Education and linguistic difference. Westport, Connecticut: Ablex publishing.
Hydén, Göran. 1968. TANU yajenga nchi. Politic development in rural Tanzania. Doctoral thesis. Uppsala: Nordiska Afrikainstitutet.
Lwaitama, Azaveli F and Josephat M Rugemalira. 1990. The English language support project in Tanzania. In: Casimir M Rubagumya (ed.) Language in education in Africa: A
Tanzanian perspective, pp 36-41. Clevedon Philadelphia: Multilingual Matters.
Ndayipfukamiye, Lin. 2001. The contradiction of teaching bilingually in a postcolonial Burundi: from Nyakatsi to Maisons en étages. In: M Heller and M Martin-Jones (eds.)
Voices of authority. Education and linguistic difference, pp 101-116. Westport,
Connecticut: Ablex publishing.
Rubagumya, Casimir M. (ed). 1990. Language in education in Africa: A Tanzanian
perspective. Clevedon Philadelphia: Multilingual Matters.
_______(ed). 1994. Teaching and researching language in African classrooms. Clevedon: Multilingual Matters Ltd.
Seitel, Peter I. 1972. Proverbs and the structure of the metaphor among the Haya of
Tanzania, Doctoral thesis. University of Pennsylvania, Pennsylvania.
Thomas, Wayne and Virginia Collier. 1997. School effectiveness for language minority
students. NCBE Resource Collection Series, No. 9. Washington: George Washington
University. Retrieved 2008-10-01 from
http://www.ncbe.gwu.edu/ncbepubs/resource/effectiveness.
Wedin, Åsa. 2004. Literacy practices in and out of school in Karagwe – The case of
primary school literacy education in rural Tanzania. Stockholm: Centre for research on
Bilingualism, Stockholm University, Stockholm
_______2005. Language Ideologies and Schooled Education in Rural Tanzania – The Case of Karagwe. International Journal of Bilingualism and Bilingual Education, 8:6, pp 568-588.
