|
|
Paper presented at the British Educational Research Association
Annual Conference
(September 11-14 1997: University of York)
1. BACKGROUND
Establishing bilingual schools in Wales
The provision of an appropriate language education for children of different language backgrounds in Wales has been put in place over a number of years and in piecemeal fashion. The Education Act of 1870 introduced the principle of education for all, but Welsh was not admitted to the system until the Report of the Royal Commission of 1888 recommended that Welsh be taught as a subject. The 1928 government report The Welsh Language in Education and Life laid the basis for the principle of bilingualism in the education system in Wales, and this was confirmed by the report of the Ministry of Education in 1953, The Place of Welsh and English in the Schools of Wales. That report proposed the teaching of both languages to all pupils in Wales, and this policy was adopted by all the Welsh counties except Monmouthshire and Radnorshire. In practice, the policy usually meant teaching Welsh as a subject at primary and secondary level to first-language Welsh-speakers, and as a second language to other pupils, although several counties found difficulty in implementing the latter aim, particularly in primary schools.
The Welsh Office, however, urged 'Welsh education and education authorities and schools to visualise Wales as a bilingual country.' The Gittins Report (1967) advocated a bilingual education system for non-Welsh speakers to complement the Welsh medium schools established for Welsh-speaking pupils. In response to Gittins, a Schools Council Project was implemented in some seventy schools in anglicised areas. The project created a situation, therefore, whereby separate systems could facilitate pupils' acquisition of a second language, either English or Welsh, whilst maintaining competence in the first language.
However, the movement towards the establishment of bilingual schools on a national scale was halted by the demand in anglicised areas for Welsh medium education. Dodson (1995) offers the following explanation for this change:
The pressure for total Welsh-medium, rather than bilingual, schools was fuelled by the apparent success of French-medium 'immersion' schools in Canada and the United States, where all the intakes consisted of monoglot English-speaking children. It was considered that if the French immersion programmes worked satisfactorily, then Welsh medium schools with mixed intakes of both Welsh-speakers and originally monoglot English-speakers would be even better for Welsh-learners from a linguistic point of view. (p. 113).
Immersion bilingual education was first attempted in St. Lambert, Montreal, in 1965, with the founding of an experimental kindergarten which would teach French to English-speaking children. The aims of the project, according to Lambert and Tucker (1965) were to help children become bilingual and bicultural through:
Baker (1993) argued that it would be dangerous to generalise from the successful Canadian experiment to Wales. In Canada, the two languages considered are both "high-status, international languages". In Wales, one language (English) is the majority language within the country, while the native language (Welsh) is very much a minority language, and without the international status of English. Whilst the influence of the Canadian movement is important, attempts had been made to develop systems for Welsh medium education even before the St. Lambert experiment. Morgan (1966) notes the existence of 43 designated Welsh schools in 1963, a majority of whose pupils were from non-Welsh speaking homes.
Welsh medium schools have come to be defined as those where Welsh is the sole or main medium of instruction of first and second language pupils. Welsh-medium education is based on the Canadian immersion method, with immersion occurring early but the time spent in immersion declining between the ages of seven and eleven as the use of English increases. However, it never attains partial immersion levels of 50% as Welsh remains the main language used. The challenge for Welsh medium schools is to help pupils achieve equal attainment in each language, and to develop strategies which foster this balance.
Language as a medium of instruction
The one characteristic that is common to both the bilingual and immersion approaches is the emphasis placed on ensuring that language learning and acquisition occurs through both medium and message-orientated activities. According to Price (1993):
In lessons, the teacher should establish the principle of basing language on an activity. It is important to ensure that there is a purpose to the lesson apart from teaching language. (p. 83).
This reflects a view of learning in the Welsh context, elaborated for example by Professor Jac L. Williams and Carl Dodson, the latter attributing special importance to children's language in developing message-orientated activities.
Children's interactions with others are message-orientated communication, because in those situations the message takes precedence over how to use the language or the medium. First language acquisition can thus be seen as a fluctuating process between medium and message-orientated communication, with medium-orientated communication decreasing as an individual's (grasp of) language increases. (Dodson, 1995).
The message orientated approach lies at the heart of the philosophy of Welsh medium education, as Baker (1993) indicates:
Immersion education is based on the idea that a first language is acquired relatively unconsciously. Children are unaware that they are learning a language in the home. Immersion attempts to replicate this process in the early years of schooling. The focus is on the content not the form of the language. It is the task at hand that is central, not conscious language learning. (p. 230)
In the research project reported in this paper, it was possible, however, to discover a high incidence of direct instruction in schools. Medium-orientated rather than message-orientated approaches were very much in evidence. Another important question regarding Welsh medium education is whether approaches towards instruction are similar for Welsh and English. Can schools foster development in two languages merely by adopting the message-orientated approach, or, as children grow older and more proficient, is there a need for more formal teaching? Swain (1996a) notes, 'Our observations in immersion classrooms suggested that there is a lot of content teaching that occurs where little or no attention is paid to students' target language use; and there is a lot of language teaching that is done in the absence of context laden with meaning'.
Baker (1993) poses other operational questions relevant to the use of two languages. Should languages be mixed in the same lesson? In which language should particular subjects be taught? How much time should be devoted to the two languages within the curriculum? He further asserts from his research into the Canadian situation (Baker, 1988) that the immersion process greatly aids second language learning, and, although the child's first language skills may remain underdeveloped during early years, children tend to be fully bilingual by the end of elementary schooling:
" ...immersion children learn French at no cost to their English. Indeed, not only is there the gain of a second language, there is also good evidence to suggest that immersion results in possible extra benefits in terms of English proficiency"
While Baker's research suggests that this applies equally to Wales, it is not always clear how this is achieved. In particular, it is uncertain how teachers monitor the balance of language use and the individual pupil's developing competence. The subject-based nature of the National Curriculum has further complicated the situation regarding Welsh medium schools, in that schools are required to teach a prescribed programme of study in both Welsh and English at Key Stage 2, possibly leading to more medium-orientated instruction.
The NFER study as indicated in the main section of this paper suggested that the use of both languages has led to tensions in attempting to integrate language and content teaching. Indeed, the visits to Welsh-medium schools confirmed the observation made by Swain that there was a distinction to be drawn between language medium teaching (i.e. focused on content) and language teaching (focused on language form). In this respect, Welsh immersion classrooms displayed similar characteristics to those in Canada.
It is these issues, and in particular the questions of equal linguistic development and parity of provision, which have been identified as critical in the literature on bilingual education, in reports of Office of Her Majesty's Chief Inspectors (OHMCI) on individual schools and in a recent NFER study (Lewis and Powell, forthcoming). It also prompted the NFER to initiate this study into the ways in which Welsh medium schools dealt with the use of Welsh and English as the medium of instruction.
2. THE RESEARCH STUDY
The project schools
Researchers from the Welsh Unit of the NFER visited five Welsh medium primary schools providing education for pupils aged 4-11 years of age. The research was based on interviews with staff in schools and on a review of relevant documentation. The schools differed in size, nature of language background and location in Wales. The characteristics of each school are outlined in Table 1.
Table 1 Sample of Welsh medium primary schools
Number of pupils Percentage from Welsh Location
speaking homes
School 1 85 90 North West
School 2 100 50 South West
School 3 226 2 North East
School 4 294 3 South East
School 5 426 14 South East
Two schools were chosen in south east Wales to reflect the nature of Welsh medium education in an area which has seen rapid growth in demand for this type of school. Whilst one was a medium size school, the other was the largest Welsh medium school in Wales and both provided Welsh medium education in a context in which the majority of pupils had English as their first language. The school from north east Wales dealt with a similar intake, as far as language background was concerned, and was located in a similar area to School 4. The two smallest schools were traditional Welsh medium schools, reflecting the high proportion of Welsh speakers in the schools' catchment areas. The natural Welsh speaking intake of School 2 had been diluted by an influx of English speaking families, so that Welsh and English speakers were in approximately equal numbers. The local authority had made attempts to compensate for this by establishing a unit where non-Welsh speakers could undergo intensive Welsh teaching in order to be able to participate in Welsh medium education.
Research Methods
At each school, the NFER conducted a number of face-to-face interviews with key staff - the headteacher, language coordinators, and Year 3 and Year 6 teachers. In total, thirteen teachers were interviewed - five headteachers, each language coordinator (one of whom was also a headteacher and two, teachers of Year 3), two other teachers of Year 3 and one teacher of Year 5 and Year 6 respectively. Researchers also collected documents such as language policies, planning grids and local education authority guidelines and these were analysed in conjunction with the interview data.
The NFER saw this project as an exploratory study which could address the following aims:
The way in which schools approached the development of competence in both languages, both institutionally and at classroom level, and the means by which they ensured a balanced provision became a major focus of the study. The remainder of this paper provides a summary of the results of the investigation organised under headings which reflect the main issues of concern to both practitioners and policy-makers.
3. THE USE OF ENGLISH AND WELSH IN WELSH MEDIUM CLASSROOMS
Underlying principles
The rationale for the use of English in Welsh medium education, in particular the concern that total immersion in the Welsh language might hinder the development of the use of good English amongst pupils, was an initial focus of the study. However, the project schools readily emphasised the fact that being a Welsh medium school did not detract from a commitment to ensure that there was adequate provision for pupils to develop their English. All the schools had an explicit policy aiming for all pupils to become equally competent in both languages by the age of eleven. This was stated clearly in the language policy documents, for example:
'The main aim of the school is to ensure that each child is totally bilingual.'
The schools' staff regularly referred to Key Stage 2 test results as evidence that they had succeeded in these aims. English and Welsh test results of pupils were often similar, and this was seen as an indication that performance in both languages was at the expected level for this key stage. In addition to this, teachers noted that pupils' performances in end of key stage tests were on a par with those in English medium schools, justifying the value of Welsh medium education. End of key stage test results were also invoked by schools to disprove claims from some quarters that focusing on the Welsh language cluttered pupils' minds and detracted from their academic performance in other subjects in the curriculum.
A minority of schools expressed additional aims for the teaching and use of English. As well as the statutory obligation to foster parity of competence in two languages, one headteacher of a school in an anglicised area acknowledged that the majority of pupils would need eventually to find employment exclusively in an English language environment. The vocational argument was therefore one reason for ensuring that pupils possessed the required skills in English. Pupils at another school in a Welsh speaking area fed into two comprehensive schools, one of which was English medium, and this was another reason for developing adequate skills in both languages by the age of 11.
Welsh medium schools, are not statutorily required to introduce the teaching of English until Year 3, but, in recognition of the need to sustain levels of proficiency in English, they normally use Welsh as the main language of instruction in key stage 1 only, introducing English in the later stages of key stage 1 or the early part of key stage 2. Indeed, this was a pattern adopted by most of the project schools, as described below.
Rationale for the deployment of two languages
All staff interviewed were clear that the two languages included in the curriculum at key stage 2 in Welsh medium schools were 'teaching media and not separate subjects'. However, it became apparent that whilst most teachers would hold to this principle in theory, in practice Welsh and English were used both as media and also taught as subjects in their own right. One headteacher stated unequivocally that 'Welsh and English are used to teach other subjects and they are taught as languages'. This distinction is important and needs to be highlighted as the teachers often experienced difficulty in reconciling these two positions. One explanation may be that teachers were more at ease in planning the use of Welsh as a medium of instruction than the use of English, and were focusing exclusively on their role as teachers who needed to focus primarily on the use of the Welsh language in their classrooms. The problems encountered by teachers are described in the next sections.
The Introduction of English
The introduction of English varied from school to school - or more accurately from area to area according to the language situation in their locality. In the Welsh-speaking areas, all teachers introduced the use of English towards the end of the second term or during the third term of Year 2. Only one of the schools in the anglicised areas did so - although to a limited extent - towards the end of Year 2. Otherwise, the most usual procedure was for class teachers to begin to use and teach English in Year 3. This approach is unlike some other immersion language programmes in other countries where a different teacher assumes responsibility for teaching the second language. It also differs in the fact that individual class teachers teach both content and language, although as Swain (1996a) indicates, 'this gives immersion teachers an advantage in knowing both the language and content needs of their students; it also puts a heavy load on immersion teachers as they try to do the task of two teachers by teaching both content and language'. Key Stage 2 teachers in Welsh medium schools have a particularly heavy load in that they need to teach content through the medium of two languages as well as teaching those languages.
Teachers need to consider carefully which strategies they deploy from the outset, as they introduce and use an extra language which pupils would have only met incidentally during their school lives. One Year 3 teacher in a school with a high intake of pupils from an English speaking background described his method of introducing English into his classroom. 'English work is introduced gradually into the class during the first weeks of Year 3, mainly by the means of reading aloud to the pupils'. This approach was said to be welcomed by the children, as many of them would already have been used to reading English books at home with their parents. It also eased the transition process for pupils who had heard and used no English at all within the school walls prior to this point.
In the light of the overriding need to ensure that schools functioned as Welsh medium institutions, it became apparent that the use of English, throughout key stage 2, would not receive a more generous time allocation than Welsh. Teachers from all schools emphasised that the use of English gradually increased throughout the junior school in order to ensure that pupils can carry out academic work equally well in both languages. However, it was clear that there was always a far greater level of the use of Welsh in all of the schools. One school attempted to quantify the proportion of Welsh and English medium teaching: 70 per cent of the curriculum was delivered through the medium of Welsh in Years 3 and 4, falling to 60 per cent in Years 5 and 6. However, this pattern was not evident in the other schools, even if staff expressed the view that the use of English did increase considerably by the time pupils reached Year 6, specifically by the use of timetabled English lessons.
In addition to the use of English in the classroom, schools often invited non-Welsh speakers into schools to address assemblies or talk to groups in order for pupils to experience listening to different registers of language. These sessions were also viewed by schools as a means of reminding pupils that they lived in bilingual communities. In a minority of schools, attempts had been made to integrate the use of external speakers into the English teaching programme, with some follow-up being carried out by teachers. A systematic approach such as this was rare and schools often failed to maximise the opportunities for additional and informal English language use. Schools in general were also uncertain about the level of English to be used informally and incidentally around the school. One headteacher deliberately used English in school assemblies whilst other teachers felt this would undermine the position of the Welsh language as the 'official' language of the school. Schools also varied in their use of posters and advertisements in English displayed around the school, underlining in some cases an unease felt by some staff that too much exposure to English would affect adversely the Welsh ethos of the schools.
The Use of English as a medium of instruction
In the majority of schools there were few examples of systematic planning for the use of English as a medium of instruction. It was evident that determining the best means for deployment of another language in the classroom often demonstrated the confusion that existed in teachers' minds over the effectiveness of indirect acquisition of language as opposed to formal instruction. In practice, however, teachers found a way out of this maze by engaging in increasing levels of direct instruction throughout Key Stage 2, possibly because they had access to English language schemes and a wide variety of reading books. Nonetheless, teachers still claimed that English was used primarily as a medium rather than taught as a separate subject. In effect, teachers deployment of English was governed by two factors. Firstly, they had to deal with the challenge facing all primary classroom teachers of integrating a number of curriculum subjects into the teaching of a single topic, made all the more difficult by the need to decide which of the two languages to use and when. Secondly, teachers often held their own beliefs about the effectiveness of particular approaches; for example, many felt that the formalised instruction provided by the English language schemes would enable pupils to catch up on lost ground in English. However, the picture was far from uniform and systematic.
An example of a structured approach resulting from clear planning was seen in a school which had received much support from the local education authority. The advisory service input had helped the school to develop its own guidelines on using particular languages in topic work. This guidance was intended to help teachers decide what subject to teach in English or Welsh in the half termly plan. Within the yearly six topic cycle, two of these topics were based on language work, one for Welsh and one for English, drawn from the text book. Teachers at this school, as at other schools in the study, noted that they could only treat these guidelines as a help to planning and that considerable freedom had to exist at classroom level. The LEA advice had helped staff to consider how to apportion the use of English amongst all subjects taught and to ensure that pupils encountered a comprehensive range of language in their work across the curriculum. The language coordinator pointed out a need to create continuity through the whole-school thematic cycle (which details the topics that each class teacher covers) - a need which year plans had not met before implementing the LEA guidelines.
It was evident that even schools that were clear in their intention of structuring the use of English accepted a lack of systematicity in favour of the flexibility required and afforded at the classroom level. Whilst schools had often attained a greater level of planning for the use of English as a medium of instruction, it was evident that tactically teachers could often change from the use of English to Welsh, depending on the nature of the classroom interaction. In Welsh medium classrooms on the other hand, Welsh would always be the dominant medium with English intended to be introduced at predetermined intervals. It was apparent, however, that even planned lessons are subject to change, with different subject matter often being introduced and a different language of instruction used. This is common in practice and it raises problems for teachers whose instincts are often at odds with their aim to control exposure to English and to achieve some form of balance between the languages within their classrooms. To maintain such control teachers would need to record methodically the amount of English and Welsh actually used to continually redress the balance. This is an unrealistic expectation and, given the fact that teachers for the most part remain flexible about the choice of activities and the language to be used, it means that parity of language use in objective terms is probably unattainable.
Although all of the schools continued to adopt a thematic approach to teaching, an increased emphasis on individual subjects was evident and appeared to be a consequence of the introduction of the National Curriculum. Two schools had achieved some degree of structure in their approach by denoting which content, in terms of subjects, was taught through the medium of English. In one school it was Science and History, in the second Mathematics and Geography and in the third, Mathematics was taught solely through the medium of English - a decision dictated by the availability of course books. In this way, therefore, teachers could be clear about which part of the curriculum would be delivered in English.
In the other schools, the choice of medium was far less systematic; one teacher remarked that he used English as the medium of instruction according to 'convenience, as there is no real structure'. It was felt that it was easier to teach some subjects or topics in English because of the availability of resources appropriate to the linguistic ability of the pupils. Other staff in the same school indicated that English was more widely used as a medium for teaching across the curriculum since the introduction of the National Curriculum was first introduced, indicating that the subject based approach did offer some means of planning which language to use. Nonetheless, they admitted that they approached this in an unstructured manner and did not always use subjects as the basis for planning which medium to use. For instance, switching to English was common in classes which dealt with topics in local history where there were many English language resources which teachers thought it more natural to discuss in that language. Whilst teachers welcomed this flexibility of approach they were willing to acknowledge that a greater degree of planning was necessary in order to ensure an appropriate balance between the two languages.
The largest Welsh medium school visited, again in an anglicised area, had retrospectively mapped what had been taught to pupils over the past four years by analysing teachers' records. This audit had also proved to be a means of detailing what language objectives - both in Welsh and English - should be taught in order to foster language development within the teaching of all subjects. Teachers often felt isolated as they attempted to undertake such intricate planning and stated that they needed clearer guidelines, and examples of good practice, both locally and nationally. One headteacher acknowledged that schools had not allowed time for teachers to consult with each other on such important issues. The tension identified in this case arose from ensuring that pupils had an adequate and balanced language education without undermining the coverage of specific subject content.
This matter was graphically illustrated in another school located in an anglicised area. The school's language policy outlines the items to be covered in language work such as punctuation, vocabulary, spelling, types of writing and comprehension. This list was then used as a basis or framework for ensuring that there was adequate scope for language development in the subjects taught in topic work; this was referred to by the headteacher as a 'checklist of language aspects'. However, a recent OHMCI inspection of the school felt that this approach adversely affected the nature of Geography lessons viewed by the inspection team. This highlights the challenge of presenting content and language, and the inspection comments point to the potential conflict between the demand of each. The headteacher maintained that the detailed procedure was necessary to foster language development more systematically within content teaching of subjects, something that would not be possible at Key Stage 3, where the division between content and language teaching is more clearly delineated. In this particular case, teachers had a sense of having to perform a juggling act in deciding what language to use and how to structure its use.
It is evident that the level of planning needed to ensure compatibility of language use, and the means by which teachers can implement plans, are major issues for bilingual teaching. The evidence presented in this study indicates a lack of systematicity in English medium teaching, partly because teachers have a readily available resource in published textbooks and also because of a belief that direct instruction is more effective. Another factor which can undermine attempts of curriculum planners to achieve compatibility is the fact that teachers regularly switch languages in an ad hoc manner.
Swain (1996a) emphasises that immersion teaching needs to be highly structured and systematic in its planning if it is to succeed: 'if immersion pedagogy is as we claim it to be - both content teaching and language teaching...we need to be doing a lot more fundamental planning about how to integrate language and content teaching.' In the instances observed in Wales, such high level integrated planning was variable, with teachers often isolating teaching of language from content by introducing periods of formal instruction. This was true of both languages and the next section considers teachers' motives in including substantial separate periods of language teaching for English.
The teaching of English as a subject
Many teachers felt that the function of their school was primarily to facilitate the development of pupils' Welsh language skills by means which could then be adapted to the teaching and learning of English. However, it became apparent that this was rarely the case in practice and that approaches to the use and teaching of English differed from Welsh. Schools were more confident that pupils' development in English could be assured by separate periods of instruction rather than in teaching content. In fact, far more sequential planning of activities was evident in the implementation of a programme of English instruction than any other aspect of the use of language in these schools. All schools timetabled English lessons and deployed a commercial scheme, such as Nelson Primary, the Primary Language Programme, the Oxford Reading Tree or Ginn 360, which provided a structure for the organisation of the teaching.
The use of structured teaching material allowed staff to monitor closely the specific language work covered by classes. It was evident that the greater the involvement of staff in teaching the language as a subject, the more confident they were that children's language development needs were catered for. This situation has been referred to as 'the teacher's paradox' (Edmonson, 1985, p.162). Specifically, teachers would agree that 'we seek in the classroom to teach people how to talk when they are not being taught' yet often in practice seek to ensure that the learner behaves as a learner rather than in other roles. Ellis (1990) sees the difference as a question of 'whether the participants are engaged in the act of trying to learn or trying to communicate'. This distinction illustrates one of the principles underlying medium education - of promoting language learning through communication - whilst also acknowledging that practitioners often favour formal instruction.
The view of one headteacher, who felt that there had been too much emphasis on teaching through the medium of a language and too little on teaching as a subject was typical in this respect. He believed that advisers and inspectors were at fault for this, and that this had deterred teachers from increasing the number of periods of direct instruction. Thus, teachers would feel confident about their English language work when pupils were placed in the role of traditional learners, working their way through structured language courses. With such methods, teachers could feel secure about the language objectives being covered, but relatively insecure when they attempted to integrate content and language.
It is was also apparent that teachers deployed formal teaching strategies, effectively separating language from content, in order to attend to accuracy as well as fluency. This was true of both English and Welsh, as is shown later in this paper. Teachers in these schools were particularly concerned about pupils' incorrect use of language, both in oral and written forms. They also commented that pupils were often deficient in their knowledge of vocabulary. Research in Canada has noted these same weaknesses in pupils' language and has proposed similar strategies to those observed in the Welsh classrooms 'Many believed the obvious solution to the grammatical weaknesses observed in the spoken and written French of the immersion students was to provide the students with written and oral exercises focusing on the grammatical problems observed' (Swain, 1996b). Whilst the commercial courses available for English contained meaningful and sophisticated exercises, these were isolated from the English used in content teaching. Swain argues for more integration and the use of specific tasks during content teaching that enable pupils to reflect on language form - an approach that needs to be planned: 'situations must be contrived to ensure that students both hear and read the language we want them to learn, and to ensure that students are given opportunities to be pushed back beyond their current abilities in the target language through the provision of feedback on the accuracy, coherence and appropriateness of the immersion language they use.' (Swain, 1996a). This would be a different type of methodology from that described by teachers in the schools visited, as this paper demonstrates when describing teachers' methods in teaching English and Welsh in the following sections.
Reading in English received much attention in these schools as teachers were aware that there was a need to compensate for any time that may have been lost in concentrating on reading in Welsh. It was generally acknowledged that the development of English reading was aided considerably by the quality of books and reading schemes available. As one headteacher indicated, 'attempts to foster children's reading are assisted by the standard of reading materials which are more attractive to children than books in Welsh aimed at this age group'. Generally, teachers noted that children did not experience any difficulties when reading English for the first time in the classroom, especially those from English speaking backgrounds. Reading English was not even considered a challenge for those in Welsh speaking areas as the language they encountered was similar to that taught in schools. This was in contrast to the difficulties met by pupils from Welsh-speaking homes in making the transition from speaking a colloquial form to recognising and dealing with standard Welsh in written form.
Teachers thought that writing English posed more problems than reading; often children who had little contact with the language had little idea of correct orthography. It was often mentioned that the 'phonetic' nature of Welsh (that is, the close and fairly regular correspondence between sounds and symbols), caused children to write words as they pronounced them. When children encountered good models of English this tendency was said to decrease. Again, in order to compensate for such deficiencies teachers advocated the use of direct language instruction rather than the use of strategies embedded in content teaching.
The teaching of Welsh as a subject
As with English, teachers arranged separate periods of Welsh language instruction. The time allocation for these was usually comparable to that for English but the approach was often less well structured - mainly because of the lack of suitable teaching materials. During periods of Welsh instruction, teachers in most schools reported that they 'drilled pupils' in certain aspects of grammar - this was said to ensure that they met and used correct language patterns, and because there was a tendency for children from English speaking backgrounds to use Welsh inaccurately. The intensive methods deployed were deemed to have led to more correct usage of the language. The teaching of the Welsh language was seen to be complementing the content teaching in Welsh, rather than providing the main thrust, as was the case with English. Whilst teachers were more confident that they could develop pupils' language skills through the Welsh content teaching, they still needed to provide time for explicit language teaching, or for remedial work. Although these lessons had slightly different aims from the English ones, they again reflected the principle of separating the teaching of language from content, as this was seen as the only way of compensating for deficiencies in vocabulary and grammar.
Other factors affecting equal linguistic development
The schools felt that the effects of formal language instruction were evident in the improvement observed in pupils' performance. In one school, in an anglicised area, the headteacher noted that the English spoken by the children on entry to the school showed less concern for accuracy than their Welsh, which, as a second language, had been learnt in a structured fashion. In this situation, the school was actively considering ways of maintaining a progressive development in English. The language coordinator of a school with a similar intake felt that the childrens' grasp of Welsh (their second language) was stronger than their English, not only because of the nature of language instruction in Welsh, but also because pupils were more careful in their use of Welsh. Similarly, teachers in a school located in a natural Welsh speaking area were concerned about the standard of pupils' Welsh (their first language).
The ability of pupils to develop similar if not equal skills in both languages was also affected by the home background and social context of the school, factors which were referred to constantly by teachers. It is here that the ambiguity of the term first language arises. Mayor (1994) suggests that (The term) first language ... needs to be used with care when referring to children from multi-lingual families or societies ... What is more useful to know is who speaks what language to whom and under what circumstances ... The so-called 'first language' may not remain the primary or dominant language throughout life and may not be the preferred language in a wide range of contexts. The suggestion is that, soon after entering the education system, in a bilingual environment, the lines between first and second language become blurred. Certain facets of language skill may be stronger in one language than the other, and it is often true that no one language is truly 'stronger' than another in all respects.
In common with other studies of medium education it appears that children were in no way disadvantaged as far as their development in either language was concerned. Evidence such as external test results support what researchers have formerly noted about first and second language development. Studies of the French immersion programmes (for example, Lapkin et al., 1990) found that pupils attained near native-like performance in French as a second language in receptive language skills, although speaking and writing were not so well developed. Similar evidence of first language ability was also found; in fact Swain and Lapkin (1982) found that pupils educated in a bilingual situation displayed greater proficiency in their first language than those taught in a monolingual situation. However, the study highlights some important considerations for the development of an effective pedagogy relevant to Welsh medium education and other bilingual education systems. These are discussed in the next section.
DISCUSSIONS
All schools contacted for the purposes of this research were confident in and comfortable with the rationale governing the teaching of English in Welsh medium schools. It had always been accepted that while the Welsh medium sector had an important function in developing the use of the Welsh language, it was equally important that pupils in Welsh medium schools did not lag behind their counterparts in English medium schools in any area of the curriculum, including English. This rationale is based on the pragmatic realisation that all young people in Wales need fluency in English and all round educational achievement in order to be able to compete in the job market.
Teachers claimed that using both languages as a medium of instruction was consistent with these aims. However, it became apparent that, in practice, teachers taught subject content through the medium of Welsh and English, and deployed additional strategies to teach these languages. There were a number of factors which determined the balance for this, including lesson planning, teachers' perceptions of the best ways to ensure language development, the course of classroom interaction and the type of resources available.
Ensuring the integration of teaching content and language in both languages required a sophisticated and detailed level of planning, which schools attempted to achieve but were often unable to deliver in the classrooms. Successful integration would require clear principles for deciding on which language to use to teach particular aspects of the curriculum and an understanding of how language development is to be achieved. This would involve schools in developing an explicit structure for lessons that enabled teachers to programme carefully the language used. This would need a high level of awareness of language and content objectives on the part of teachers when managing classroom interaction.
Another factor which contributed to teachers' uncertainty about effective planning was confusion regarding the conditions for language acquisition as opposed to language learning: many felt that pupils needed to be placed in the role of conscious learner before being able to make progress in the language. Teachers were more confident that language learning was taking place when they were in control of the content of the language learning and when the sequence of that learning was made explicit.
The predisposition of teachers to favour formal instruction undermines one of the central tenets of medium education and its message orientated emphasis. This was more obvious in the teaching of English where Key Stage 2 teachers felt that they needed to use intensive published courses that could compensate for the late introduction of the subject. Teachers also allowed for periods of formal instruction in Welsh. This emphasised the fact that with both languages teachers deliberately separated language and content because they were convinced this was necessary to promote language development.
Formal language teaching was typified by the use of exercises requiring pupils to consider aspects of language outside the subject context. If Swain and others are right (in their analysis of the immersion situation), teachers need to reconsider these strategies and develop methods to enable pupils to examine their use of language as it occurs within the framework of the subject content. This approach would be more consistent with teachers' stated objectives and result in a return to the founding principles of medium teaching, based on the cohesive development of both message and language. However, more development may be needed to establish principles governing the integration of language and curriculum content.
The lack of commercial courses to teach Welsh contributed to the fact that teachers were more comfortable in their Welsh content teaching. They had nowhere else to turn for systematic language teaching provision, and as a result produced remedial exercises to deal with pupils' deficiencies. It would be interesting to find out whether the approaches to Welsh and English content and language teaching would be better aligned if there were similar courses available for Welsh. Despite differences in approach, a general conclusion is that teachers see the teaching of language and curriculum content mostly as objectives requiring different kinds of attention.
The evidence presented in this paper derives from an investigation of school practice and teachers' beliefs about language teaching, and it is apparent that current practice in Welsh medium schools differs considerably from stated policy. In order to obtain a better understanding of the relationship between policy and practice the investigation would need to extend to an assessment of their impact on pupils proficiency. This is work that remains to be done.
References
GITTINS REPORT. GREAT BRITAIN. DEPARTMENT OF EDUCATION AND SCIENCE. CENTRAL ADVISORY COUNCIL FOR EDUCATION (WALES) (1968). Primary Education in Wales. London: HMSO.
GREAT BRITAIN. MINISTRY OF EDUCATION. CENTRAL ADVISORY COUNCIL FOR EDUCATION (WALES) (1953). The Place of Welsh and English in the Schools of Wales. London: HMSO.
GREAT BRITAIN. BOARD OF EDUCATION. WELSH DEPARTMENT (1927). Welsh in Education and Life.
BAKER, C. (1993). Foundation of Bilingual Education and Bilingualism. Clevedon: Multilingual Matters.
DODSON, C. (1995). The Effects of Second-language Education of First/Second Language Development. In: MORRIS JONES, B. and SINGH GHUMAN, P.A. Bilingualism, Education and Identity. Cardiff: University of Wales.
PRICE, E. (1993). 'Y Gymraeg fel ail iaith yn yr ysgol gynradd.' In: WYN, G. Agweddau ar Ddysgu Iaith. Llangefni: Canolfan Astudiaethau Iaith.
MAYOR, M. (1994). 'What does it mean to be bilingual?' In: STIERER, B. and MAYBIN, J. (Eds) Language Literacy and Learning in Educational Practice. Clevedon: Multilingual Matters.
MORGAN, G. (1996). The Dragon's Tongue. Cardiff: Triskell.
SWAIN, M. and LAPKIN, S. (1982). Evaluating Bilingual Education. A Canadian Case Study. Clevedon: Multilingual Matters.
LAPKIN, S., SWAIN, M. and SHAPSON, S. (1990). 'French immersion research agenda for the 90s, Canadian Modern Language Review, 46, 4, 638-74.
SWAIN, M. (1996a). 'Integrating language and content in immersion classrooms: research perspective', The Canadian Modern Language Review, 52, 4, 529-47.
SWAIN, M. (1996b). 'Discovering successful second language teaching strategies and practices: from: programme evaluation to classroom experimentation,' Journal of Multilingual and Multicultural Development, 17, 2/4, 89-104.
BAKER, C. (1988). Key Issues in Bilingualism and Bilingual Education. Clevedon: Multilingual Matters.
LAMBERT, W.E. and TUCKER, S. (1972). Bilingual Education of Children. The St. Lambert Experiment. Rowley, MA: Newbury House.
ELLIS, R. (1990). Instructed Second Language Acquisition. Oxford: Basil Blackwell.
LEWIS, G. and POWELL, R. (forthcoming). Welsh medium Education and the Teaching of Welsh: the Introduction of the National Curriculum. Slough: NFER.
EDMONDSON, W. (1985). 'Discourse worlds in the classroom and in foreign language learning', Studies in Second Language Acquisition, 7, 159-68.
This document was added to the Education-line database 06 February 1998