Nature of the curriculum

4.3.1.1. Distribution of languages in the curriculum

Of primary concern for reflection is the time devoted to the L1 and L2 in the curriculum.

If the desired outcome goal includes maintenance and enrichment, there should be as much curricular support for the L1 (through L1-subject and L1-content teaching) as is necessary to avoid L1 impairment and as much support for the L2 (and further languages) as is possible. Concretely this means that unless the L1 is dominant in the wider community, it is probably advisable to devote no more than 50% of the time-table to the L2 in the first five to eight years of schooling (Baker 1996, Skutnabb-Kangas 1995).

Latvian program models 1, 2, and 4¹ emerge as examples of what Ramirez & Merino (1990) have called Late Exit Transitional Bilingual Educational programs. Such programs typically allow between 25-40% of classroom teaching in the L1 until the end of primary school. Program model 3, and particularly the LASOR model provide 50% or more of L1 teaching. This is a design feature of language maintenance programs as illustrated by the high proportion of L1 instruction in the European School program. This

1 See for a brief description of the models the introduction by Pieter Batelaan (ed)

13

is felt to be necessary for the many European School pupils who are in a minority situation (e.g. the Swedish, Greek, Danish, Portuguese, Finnish, Dutch) and who lack the rich linguistic out-of-school environment considered necessary for their normal L1 development. Also note that the inevitably smaller amount of time devoted to the L2 in the European Schools curriculum does not negatively affect L2 development, which is generally high and often even native-like by the end of secondary education.

Children in Latvian belonging to the smaller ethnic language groups (e.g. Polish, Yiddish, Estonians) are probably in a situation not unlike that of Danish, Greek, Finnish, or Portuguese pupils in the European Schools. Their L1 development, too, could be argued to benefit from intensive L1 instruction in primary school, probably more intensive than that provided by program models 1, 2 and 4. This would be particularly beneficial if the L1 is typologically unrelated to Latvian (like Estonian) and/or uses another spelling system (like Yiddish). The situation for Russian-speaking children is probably different.

Many (though not necessarily all) of the Russian-speaking children live in a richer L1 environment, providing substantial extra-curricular exposure and attitudinal and motivational stimulus for L1 development. Their L1 development therefore, seems currently less at risk, even if the total amount of time in the curriculum devoted to Russian instruction were to fall below the 50% criterion mark of a maintenance program.

Here it is the L2, Latvian, rather than the L1, that might benefit from additional curricular support, particularly in the light of the high levels of both oral proficiency and literacy in Latvian that are demanded from these children for academic success in monolingual Latvian secondary and higher education.

4.3.1.2. Extent of bilingual provision

If the desired outcome goals of a bilingual education policy include full bilingualism and particularly full biliteracy, then bilingual education should not only be allowed to start early (to take advantage of young learners’ natural language learning abilities) but also to span the entire education cycle, that is, primary and secondary education. This recommendation follows from what is known about the development of full proficiency and literacy in L1 and L2, which are complex, gradual and lengthy processes. Allowing bilingual education to extend into secondary schooling has the additional advantage of providing the element of continuity in the curriculum felt necessary for harmonious academic and socio-cultural development (Baker 1996).

14 Bilingual provision in Latvia is currently restricted to primary education only. There is no long-term continuity in the education system for minority children but an in-built switch from bilingual instruction to monolingual instruction in the L2. It may be queried whether this procedure is sensible if full and stable forms of bilingualism and biliteracy are indeed aimed for. Closer examination reveals that in many of the different program models this switch already occurs before the end of primary schooling. In program models 1 and 4 L1-medium instruction has virtually disappeared from the curriculum by grade 4. After that, the L1 is only offered as a subject, thus assuming a status in the curriculum similar to that of the foreign L3.

Surprisingly, L1-medium content teaching is lacking altogether from program model 2 until the last three grades of primary school. The rationale for this is unclear. Only program model 3 and the LASOR model provide significant L1-medium instruction throughout the primary cycle and thus offer the best possible chances for L1 maintenance.

4.3.1.3. Progression of L1 and L2 teaching and choice of content-matter subjects

Bilingual education means that both languages have to be used as media of instruction at some stage of the child's education, but the progression of how much of each is used and for which type of content matter seems to vary for minority and majority children. The theoretical rationale for the relevant choices is provided by Cummins (1984) who has identified three principles that capture the relationships between progression of L1 and L2 teaching, choice of content-matter subjects and successful linguistic and scholastic development in bilingual education. According to the Threshold Hypothesis, normal intellectual and scholastic progress in school requires that pupils attain minimal threshold levels of proficiency in the language of classroom instruction in order to successfully process content matter. According to the Interdependence Hypothesis, language tasks differ in terms of the degree of cognitive processing required and the degree of contextualization involved. Processing academic content-matter in a classroom involves decontextualized and cognitively demanding language use, requiring what Cummins (1984) has coined CALP (cognitive-academic language proficiency). This contrasts with BICS (basic interpersonal communicative skills), the type of language proficiency necessary for performing context-embedded and cognitively-undemanding language activities, such as carrying on a general conversation. BICS, which reflects a minimal threshold, is developmentally the most basic and provides the basis for the later development of CALP. Finally, the Common Underlying Proficiency Hypothesis assumes a common, language-independent basis for the development of decontextualized aspects of language proficiency, including literacy skills. Once firmly established in one language, such decontextualised language skills can easily be transferred to any other language being learned (provided there is sufficient contact with the target language and adequate motivation to learn it).

On the basis of these theoretical principles, several requirements can be proposed as necessary for bilingual curricula that aim at full bilingualism and biliteracy (Tucker 1998; Skutnabb-Kangas 1995; Krashen &

Biber 1988). For minority children the most stringent requirement appears to be that "the mother tongue must function as the medium of education in all subjects initially. At least some subjects must be taught

15 through L1 all the way, up to grade 12, but the choice of subjects may vary" (Skutnabb-Kangas 1995:14;

original emphasis).

The European Schools are often singled out as a program that is "close to ideal for minority children" in the way it has incorporated these requirements (Skutnabb-Kangas 1995:14). A comparison between the choices made in the bilingual programs in Latvia and those in the European School program may therefore provide an indication of the likelihood of success of the Latvian programs.

Progression and content-matter choices in relation to the minority L1 (i.e. the language least likely to develop to a high native speaker-like level) is as follows:

• In the European Schools all subjects are taught exclusively through the medium of the L1 during at least the first two years of schooling. The curriculum of the LASOR also follows this practice but the program models developed by the MoES do not (though, admittedly, the number of content-matter subject not exclusively taught through the L1 in the initial grades is small).

• In the European Schools, literacy is first developed in the L1. The situation in the Latvian programs is unclear. One would need to know more about what goes on in the "Latvian Language and Literature"

courses in grades 1 and 2. It would seem, however, that development of literacy is started in the L1 and L2 simultaneously.

• In the European Schools all cognitively demanding decontextualised content courses (mathematics, social and natural sciences) are taught first through the L1, again for a period of at least 2 years. The LASOR program also offers nearly all academic content matter first through the L1. The program models of the MoES vary in the extent to which they approximate this practice, depending on the type of course content involved.

• In European Schools the L1 is taught as a subject throughout schooling, from grade 1-12. This principle is also observed in all the Latvian programs.

For the majority L2, the progression and subject choices are as follows:

• In European Schools the L2 is taught as a subject throughout schooling, from grade 1-12. The same practice is also found in all the Latvian programs.

• In the European Schools the L2 is first introduced as a subject and is taught as subject for at least three years before being used as a medium of instruction for non-language content. This principle is only strictly observed in MoES program model 4. In all other Latvian programs, the L2 is used as medium of instruction, whether solely or in combination with the L1, either from the start or within the first three years of primary school.

• In the European Schools the L2 is first used as a medium of education for cognitively less demanding, contextualized subjects (Physical Education, Plastic Arts, Projects and Activity classes). The Latvian programs also adhere to this principle, with the exception of program model 4 where the first use of the L2 as the exclusive medium of education involves cognitively demanding and decontextualized content

16 matter (viz. Mathematics). In the other programs, exclusive use of the L2 is first reserved for less academic subjects like Sports and Home Economics.

• In the European Schools use of the L2 as a medium for teaching cognitively demanding and decontextualized subjects is delayed until grade 8, when the pupils have been taught the L2 as a subject for seven years (grades 1-7) and when they have been taught through the medium of the L2 in less academic subjects for five years (grades 3-7). In the Latvian programs, exclusive use of the L2 as a medium of instruction (i.e. excluding its use in 'bilingual teaching') for academic subjects is usually also delayed but not always as long as in the European Schools. The earliest use of the L2 for academic content matter is attested in MoES program model 4, where the L2 is used as sole medium of instruction for Mathematics from grade 4 onwards.

• In the European Schools, development of literacy in the L2 and of other decontextualized aspects of L2 is delayed until grade 5. By that time pupils have not only acquired the necessary basic interpersonal communicative skills (BICS) in their L2 but also CALP-related skills have been established in the L1, allowing for transfer to the L2. As was said earlier, it is unclear whether literacy in the L2 is also delayed in the Latvian programs.

In sum, curriculum structure and content of the Latvian bilingual programs seem adequate for the outcome goals desired from the policy. MoES program model 3, from which the LASOR model and many of programs currently implemented in Latvia's bilingual schools are derived, provides the most adequate curricular conditions for minority L1 maintenance and therefore seems particularly appropriate for implementation in areas where sufficient extra-scholastic contact with the L1 and motivational conditions cannot be guaranteed. In situations where L1 development is less at risk but where contextual support for L2 development cannot be guaranteed, bilingual education might be more profitably modelled after program model 4.

Finally, it may be queried whether restricting bilingual provision to primary education only is judicious if the aim is indeed to push both oral and literacy skills in both languages to their maximum capacity so as to become complete vehicles for full usage in all aspects of professional, public and personal life.