Stephen May

Professor Stephen May- 
Ulimasao Bilingual Education Conference
Alexander Park, Auckland 3 October 2002
Bilingualism or language loss? 
Pasifika communities and bilingual education
Language shift and loss is increasing exponentially in the world today as speakers 'shift' from speaking their first language to speaking the dominant language of the wider society in which they live. This process occurs for many minority language groups, but particularly for those who have migrated to another country. 
This keynote address explores the implications of this wider process of language shift and loss for Pasifika communities in Aotearoa/New Zealand. It addresses some of the following key questions: What should Pasifika families and communities be doing in relation to first language maintenance? Can first languages be retained alongside English; indeed, should they be? What are the benefits of bilingualism and bilingual education? 

Talofa lava, Malo e lelei, Fakalofa lahi atu, Kia orana, Bula vinaka, Taloha ni, Halo aloketa, Greetings, Kia ora koutou katoaIt is a great pleasure and privilege to be able to speak to you all today. I also think that the theme of this conference, and the discussions that are going on within it, are crucially important for the future of Aotearoa/New Zealand generally, and Pasifika peoples in New Zealand in particular. 
As such, I hope that the wider issues and concerns discussed here are heard, listened to, and crucially, acted upon -not only by those of us here, but also our communities, schools, policy makers and politicians. 
I think this is particularly important because, as I will try to show today, there is so much at stake in these discussions for Pasifika peoples in Aotearoa/New Zealand - particularly with respect to issues of language maintenance, but also crucial issues like increased or improved educational success for Pasifika students and related social mobility for Pasifika adults. 
There is also the particular problem that we have in the fields of bilingualism and bilingual education of having to constantly explain to people the benefits of bilingualism and bilingual education, while at the same time, having also to correct them about views on these issues that might seem like common sense but are, more often than not, simply wrong. This level of misunderstanding on issues of bilingualism and bilingual education is especially problematic because it often leads to the promotion of educational and social programmes that are actively unhelpful or counter-productive to those they are intending to help - Pasifika students.

Given this, what I want to do today is to discuss and respond directly to two key 'myths' or 'misunderstandings' about bilingualism and bilingual education, as well as a whole range of related issues.
Myth 1: Our (first) language will always be spoken (if not by me or my immediate family, at least by my wider family, or those in the wider community)

In fact, what we are seeing today, at the beginning of the 21st century, is the exponential loss of many of the world's languages. Of the estimated 6000 languages spoken in the world today, it is predicted on present trends that between 20-50 percent will 'die' - or no longer be spoken - by the end of this century. 
Of course, such language loss and language shift have always occurred - languages have risen and fallen, become obsolete, died, or adapted to changing circumstances in order to survive, throughout the course of human history. But what is qualitatively (and quantitatively) different at the beginning of the twenty first century is the unprecedented scale of this process of decline and loss - some commentators, such as Tove Skutnabb-Kangas, have even described it as a form of 'linguistic genocide'. 
These claims, though they might at first appear overwrought and alarmist, at least to the sceptics out there, are supported by hard data. For example, a survey by the US Summer Institute of Linguistics, published in 1999, found that there were 51 languages with only one speaker left, 500 languages with fewer than 100 speakers, 1500 languages with fewer than a 1000 speakers, and more than 3000 languages with fewer than 10,000 speakers each. The survey went on to reveal that as many as 5000 of the world's 6000 languages had less than 100,000 speakers. It concluded, even more starkly, that 96 per cent of the world's languages were actually spoken by only four per cent of its people. 
These figures graphically reinforce a suggestion made by the socioloinguist, Michael Krauss that, in addition to the 50 per cent of languages that may die within the next century, a further 40 per cent of languages are 'threatened' or 'endangered'.The end result of this process of language shift or loss, if Krauss and other commentators are to be believed, is that perhaps only 600 languages out of 6000 (10 per cent) will survive in the longer term - perhaps even as few as 300.
Most, in fact nearly all, of the languages that are currently threatened are so-called 'minority' languages - that is, languages that have little social prestige and power in the contexts in which they are spoken. This includes all indigenous languages and many so-called 'community' languages, or the languages of migrants. Consequently, I think, all Pasifika languages in the New Zealand context, as well as Mäori, fit into this category of endangered minority languages. But even majority languages - such as national languages - are no longer immune to such processes, especially because of the rise of English as a global language. So we are seeing in some Pasifika nations - e.g, in the Cook Islands - an increasing use of English at the expense of Cook Islands Mäori. 
How does this process of language shift or loss occur? In three broad stages. 
The first stage sees increasing pressure on minority language speakers to speak the majority language, particularly in formal language domains; a state of affairs described in sociolinguistics as 'diglossia'. This stage is often associated with the introduction of education in the majority language (in the Aotearoa/New Zealand context this is, of course, English). It leads to the eventual decrease in the functions of the minority language, with the public or official functions of that language being the first to be replaced by the majority language. This is what happened historically, of course, to te reo Mäori in Aotearoa/New Zealand, as with indigenous languages the world over.
The second stage sees a period of bilingualism, in which both languages continue to be spoken at the same time. However, this stage is usually characterised by a decreasing number of minority language speakers, especially among the younger generation, along with a decrease in the fluency of speakers, as the minority language is spoken less, and employed in fewer and fewer language domains. This pattern often occurs in Pasifika families in Aotearoa/New Zealand, where the parents may continue to speak a Pasifika language to each other and older relatives, but not to their children. 
Likewise, children will increasingly speak English, both to their parents and to each other, unless a Pasifika language is actively maintained in the home. (The use of English is also often seen to increase from the oldest to the youngest children - i.e the younger children speak English more than the older ones).
The third and final stage - which may occur over the course of two or three generations, and sometimes less - sees the replacement of the minority language with the majority language. The minority language may be 'remembered' by a small (almost always, older) group of language speakers, but it is no longer spoken as a wider language of communication. The end result of this process is particularly evident in the 2nd and 3rd generations of migrant families the world over where there has been a transfer from the first (or minority) language to the second (or majority) language, rather than the ongoing maintenance of both.
So, in short, if these patterns of language shift continue - and they show no signs of abating - then the vast majority of the world's languages will no longer be spoken - perhaps within as little as 100 years from now. What this means for Pasifika communities specifically - especially in Aotearoa/New Zealand, but also perhaps in the Pacific Islands itself - is that English will eventually replace Pasifika languages.
And we are not just talking about language shift and loss here either but cultural and social change as well. Now, of course, cultures always change and adapt, including shifting from one language to another (cf. Irish), and not all adaptation is negative. But there are two key questions that are often overlooked here 1) is the change inevitable, or even necessary, and, more importantly perhaps, 2) who exactly controls this process, or calls the shots? 

The first questions the idea of language replacement that is behind most language shift and loss. No-one, least of all bilingual advocates, dispute the importance and significance of learning English, but what we do question is why it should be necessary to learn English at the expense of one's first language - why can't both be maintained? Or, to put it another way, why should cultural and linguistic adaptation and change always be from a minority language/culture to a majority one.
This leads us to the second question, which highlights the issue of power since, as the famous linguist, Noam Chomsky asserts: 'Questions of language are (always) basically questions of power'. And this helps to explain why language death seldom occurs in communities of wealth and privilege, but rather to the dispossessed and disempowered.
So, what (if anything) can be done about this? There are two key areas that can be addressed in order to ensure that this process of language shift does not occur.
The first (and, arguably, most important) is that the language must continue to be spoken regularly and meaningfully in the home - not only among adults but between adults and children. The prominent sociolinguist, Joshua Fishman, argues that only when this 'intergenerational transmission' of the language, as he calls it, continues to occur, will languages survive, or continue to be spoken, over time. This is pointedly, and poignantly, illustrated by the sociolinguistic definition of a moribund (or dying) language - that is, one that is no longer passed on to or spoken by children. 

What this also means is that - important though the church, or even the schools might be, in maintaining a first language (and they are important, as I will go on to discuss), they cannot act as a substitute for the family. That said, where languages continue to be spoken in other language contexts, or domains, like the church, they clearly can and do act as a support for the language of the home. We see this in the Pasifika context, for example (both in the Cook Islands, and particularly in New Zealand) by the more rapid language loss of Cook Islands Mäori, perhaps because the church is not such a prominent feature of social organization in this community.

The second key area where this process of language shift and eventual language loss can be addressed is through education, or the schools. And this brings me to the second key myth that I want to discuss today.

Myth 2: In order to learn English, the language of the school must be English
Wrong. This is perhaps the most widely held misperception, or misunderstanding of all. As I have already said, no-one is disputing the importance of acquiring a majority language - in our context, English. We clearly need English to function effectively in a society like Aotearoa/New Zealand where English is dominant - and increasingly in the international arena as well of course, given that English is the current world language. Where people go wrong though is in taking the next step - that is, to assume that the best way to learn English - if it is not our first language - is through English, particularly in schools.

The thinking behind this view is often based on the idea that our first language is a problem or obstacle to educational success. This amounts to what has been termed in the literature a 'subtractive' view of bilingualism. That is, a view that assumes that the first language of the students is an educational obstacle to be overcome - usually by excluding the use of the language within schools - rather than as an educational and social resource to be valued and used within the school.

This view was most closely associated historically in Aotearoa/New Zealand with the overtly assimilationist educational approaches adopted towards Mäori throughout the nineteenth and much of the twentieth century, but is by no means limited to them. Today, for example, we continue to define NESB (Non English Speaking Background) students (many of whom are Pasifika) as simply deficient in English - as the term itself implies - rather than as bilingual learners with control of more than one language.

So, what we have here is a subtractive view of bilingualism that is wrong, both cognitively and educationally. It is wrong on both counts because educational research over the last 40 years has consistently found the following:

1. Active bilingualism is a cognitive, academic and social advantage (rather than a deficit)

For example, there are now close to 150 major research studies, carried out since the early 1960s, that consistently report significant advantages for bilingual students on a range of metalinguistic and cognitive tasks. As a result, it is now widely recognised that bilinguals mature earlier than monolinguals in acquiring skills for linguistic abstraction, are superior to monolinguals on divergent thinking tasks and in their analytical orientation to language, and demonstrate greater social sensitivity than monolinguals in situations requiring verbal communication. 

In fact, this shouldn't come as any surprise really, since knowledge of two or more languages necessarily lends itself to a greater level of knowledge and understanding than simply knowing one. But it nonetheless still does surprise many. In response to this ongoing scepticism about the advantages of bilingualism (most usually, by monolingual English speakers), I usually make one comment and ask one question.

When anyone describes a bilingual learner as simply being deficient in English - or, as I still hear some teachers saying, as having 'no language' - I always comment that, however 'limited' the student's grasp of English is at this particular point, it is almost always likely to be far greater than the teacher's own knowledge of the student's first language! 

And when people argue that bilingual education - particularly for Mäori and Pasifika students - is problematic and/or unnecessary, I always ask a question that Jim Cummins first raised: Why is bilingualism good for the rich but not for the poor? 

In other words, why do many of these very same people argue for the benefits of what is termed 'elite bilingualism' - learning French or German (or even Latin, as I did) in school, and yet can dismiss or oppose bilingualism and bilingual education in - say - Mäori, Samoan or Tongan? The reason is that they are basing this argument not on the linguistic benefits of bilingual education, and certainly not on the educational benefits (as I will make clear in a moment) but rather on the perceived 'value' or 'prestige' of these particular languages - that is, a social and political judgement which appears to value French and German over Mäori or Samoan. When I point this out, the response I often then get is that French and German and other so-called international languages are more 'useful', to which I again reply by asking how often have they actually ever used such languages, and, anyway, what could be more useful in Aotearoa/New Zealand than knowing other languages that are regularly spoken here, particularly Mäori, but also clearly Pasifika languages? 

But I digress! So let me turn to the next key finding from the research literature which is that, contrary to popular belief:

2. The least effective way of teaching a majority language (English, in our context) is by problematising and/or excluding first languages

International research over the last 40 years clearly demonstrates that minority language students who receive most of their education in English rather than their first language are more likely to be unsuccessful in acquiring literacy in English, and consequently are therefore also more likely to fall behind and drop out of school. The consequences of this limited educational success for such students are also equally clear. Let me take just two examples here. 

1. In the recent MoE report on adult literacy, More than Words (2001), it was noted that current adult literacy levels in English are consistently lower overall for both Mäori and Pasifika adults when compared to the New Zealand population as a whole. I am sure that this is largely the result of the assimilationist educational approaches that have been adopted towards Mäori and Pasifika students over the years (and educational research from around the world consistently supports this conclusion). In other words, it is the result largely of the historical (and in many cases, still ongoing) devaluing and exclusion of the first languages and cultures of Mäori and Pasifika students in schools.

2. These patterns of relative disadvantage in English language literacy, are specifically confirmed when we examine school level analyses of English literacy acquisition. For example, while New Zealand students perform very well internationally in relation to the acquisition of school-based literacy in English at both ages 9 and 14, as measured by the regular international (IEA) literacy valuations of all OECD countries, they only do so ifthey are already first language speakers of English. 

In fact, these IEA evaluations have now highlighted for some time that the 'home language gap', as it is called, - that is, the gap between the literacy achievements of students whose home (or first) language corresponds with that of the school and those students for whom it does not - is the largest of any OECD country at both 9 and 14 years. In other words, New Zealand has the poorest performance across the OECD in the successful acquisition of English language literacy for those students who do not speak it as a first language. And this is why so many Pasifika students continue to feature prominently in the poorly performing end of these statistics.

What is the alternative to all this? Again, research points in only one direction:

3. Bilingual education programmes, particularly when they are based on an additive view of bilingualism, are recognised internationally as the most effective and successful educational programmes for language minority students

Why is this? Because the process of learning a second language, and particularly becoming literate in that second language, is best achieved by first becoming literate in one's first language and then transferring that knowledge and skill to one's second language. Or, put another way, high level second language proficiency depends fundamentally on well developed first language proficiency - what has been described in the literature as the 'interdependence principle'.

In contrast, research also tells us that ESL withdrawal programmes (involving individual, often remedial instruction in English) - the most common language programmes in Aotearoa/New Zealand schools, and often specifically targeted at Pasifika students - are far less effective in achieving proficiency in English literacy. 
So, before turning directly to bilingual education, let me briefly address the potential problems around ESL instruction here. There are two key reasons why ESL withdrawal programmes often don't work well, or at least as well as one might expect: 

  • 1) The usual exclusion of the first language as a means of bridging to the second language  

As I have already suggested, achieving literacy in one's first language, and then transferring these literacy skills and knowledge to a second, is a key factor in successfully achieving literacy in a second language, or becoming biliterate. This is because achieving literacy, or more broadly, 'academic proficiency' in a second language is quite different from simply being able to speak a second language conversationally. Most ESL/EAL can pick up communicative competence, or conversational control of a second language within 1-2 years of being exposed to it, particularly if it is the language of the wider community. 

But the mistake many people (including teachers and educational policy makers) then make is to assume that being able to speak conversational English is the same as having control or mastery of academic literacy in English. This is not so, since ESL/EAL students actually tend to take at least 5 years, often longer, to achieve academic language proficiency in English. This is because academic, or classroom-based language is often quite different to conversational language - it is usually highly decontextualised and abstract, involves complex vocabulary and grammatical constructions (e.g. the use of the passive constructions) and is also heavily print-oriented. None of these features is present - at least to the same degree - in conversational English. 

To take just one example here - vocabulary. Academic English uses much more low frequency vocabulary (often Graeco-Latin in origin rather than Anglo-Saxon) 

Frequent A-S nounsFrequent G-L nouns
Frequency (in relation to Collins 6 point scale)
A-S top (most frequent) category of usage
G-L mean average of 3.25]

What is the implication of this distinction between conversational and academic English? Simply this: the specificity of academic language registers provides a strong basis for the advocacy of bilingual education at least until middle childhood - in order for students to specifically master these academic language proficiencies in their first language (although these do not need to be 'fully developed' before the introduction of a second language).

The second key weakness of ESL withdrawal programmes is that
  • 2) ESL withdrawal programmes operate in isolated rather than communicatively rich language environments 
In other words, such programmes work in relative language isolation, with little emphasis on active and experiential language learning, and of language learning in authentic and meaningful contexts. Consequently, they avoid or overlook the greatest language resource they could draw upon - the language-rich classroom environment. 
Both these limitations of the ESL withdrawal method could be easily and effectively addressed, as they are beginning to be in Britain for example, by the integration (rather than segregation) of bilingual/EAL students, and their associated teaching and learning, in mainstream classrooms and, crucially, the use of bilingual teaching assistants in classrooms, involved in individual support of learners, translating, and whole class work within these classes. This allows for both the use of first languages and the language rich environment that ESL withdrawal methods lack.
So now that I've dealt with ESL, let's turn again to bilingual education. 
I'm aware that John McCaffery has already spoken to you directly about the various approaches to bilingual education that can be adopted, and of those, the ones which have been identified clearly in the research literature as being the most effective (although even the least effective programmes, I would suggest, are more useful than the ESL withdrawal method just discussed). Given this, I don't want to repeat that discussion here - except to point out, briefly, the following key points about what makes, or contributes to a successful bilingual programme.
Bilingual education programmes differ widely in a number of ways:
- Educational philosophy/approach (subtractive, transitional, maintenance or enrichment bilingual programmes)
- Level of immersion (full; partial)
- Student population (predominantly L1; L2, or a combination)
- School/programme relationship (whole-school or targeted). 

This has often led to confusion about what constitutes a 'proper' or effective bilingual education programme. The debates here are complex, but what the research literature does tell us is that while there is clearly more than just one way of doing bilingual education effectively, there are also clear research indicators about what constitutes a good or effective bilingual programme, as opposed to a less effective/ineffective one. What are these?

a) Effective leadership (at school level) and administrative support
- the principal should support/understand/be able to articulate / explain the aims/philosophy and methods of the programme
- a programme leader is also essential for advocating, overseeing, coordinating and providing training for the programme

b) School environment
- an additive approach to bilingualism
- an active commitment to equality (addressing wider power relations)
- positive teacher pupil and pupil-pupil (L1/L2) relationships
- cooperative learning and teaching approaches

c) Teachers Need:
- native or near-native fluency (or have this input in the classroom from teacher assistants)
- to understand the research and theory underpinning bilingual education generally, and their approach/model in particular
- to understand second language development (e.g. distinction between conversational competence and academic literacy)
- instructional strategies in second language development (the difference between being able to speak a language and teach it)
- training/awareness of/commitment to educational equity
- training in cooperative learning approaches
d) Instructional Design
- duration needs to be at least 4-6 years at primary level (in order to ensure acquisition of academic language proficiency/literacy in the first language, and thus facilitate 1st-2nd language transfer. (This length of time is not necessarily sufficient though for students to achieve native-like proficiency in the language)
- literacy should begin in the target/minority language - e.g. via reading - this is crucial for higher immersion programmes (cf. interdependence; Kohanga/kura)
- 50% minimum of target/minority language use, though the higher the target/minority language use, particularly in the early grades, the more fluent students will eventually be
- context embedded/scaffolded approaches to begin with; balanced against the need to introduce cognitively extending/demanding language input (in order to acquire academic language proficiency)
- some explicit language teaching instruction
- separation of languages of instruction (key)
e) Home/School relationships
- strong community support
- strong parental involvement

f) Educational Outcomes
Well implemented bilingual programmes achieve bilingualism and biliteracy at academic levels (in both languages) comparable to first language speakers in monolingual programmes 
L1 English speakers do equally well in their (good) English language development in partial and full immersion additive bilingual education programmes

L1 Minority language speakers achieve greater proficiency in that language at higher levels of immersion

The higher the immersion in the L2 minority/target language, the higher the proficiency in that language for L1 English speakers
Less instruction in English for minority language speakers did not negatively effect acquisition of English
Bilingual proficiency was greatest for those in higher minority/target immersion programmes
g) Additive vs. subtractive/transitional bilingual education
The overall implications of the research on bilingual education are equally clear - bilingual education programmes that are based on an additive view of bilingualism (that see the development of bilingualism as a cognitive, educational and social advantage) and that are long term (and at least 4-6 years in duration) are the best means of not only successfully maintaining and fostering the first languages of students, but also the best means by which students can acquire English. 

This is because, as Jim Cummins (1986) has observed:

Widespread school failure does not occur in minority groups that are positively oriented to their own and the dominant [language] and culture, that do not perceive themselves as inferior to the dominant group, and that are not alienated from their own cultural [and linguistic] values.
In contrast, transitional bilingual education programmes have been proven to be less successful educationally. This is because they: 
1) continue to hold to a 'subtractive' view of individual and societal bilingualism. 
In assuming that the first (minority) language will eventually be replaced by a second (majority) language, bilingualism is not in itself regarded as necessarily beneficial, either to the individual or to society as a whole, and this limits the educational effectiveness of the approach

2) are usually only 1-2 years in duration and thus do not for the development of literacy in the first (Pasifika) language (and thus the use of this as a bridge to the second language)
I just want to make one final clarification in relation to the findings in the research literature. While level of immersion must be at least 50% for bilingual education to be effective, and must involve the minority or target language being used meaningfully and regularly as an instructional language (i.e. - the language is used to teach rather than simply to organise), this clearly still does allow for models of bilingual education that are not necessarily full-immersion. Indeed, we have seen the development of highly successful 'dual-medium' bilingual education approaches in the US that are either 90/10, or 50/50 models. 

It's my view that while clearly higher levels of immersion result in more fluent speakers in the minority language, these partial immersion programmes could also be successfully pursued in Aotearoa/New Zealand as long as an additive and long-term approach is taken to them. In other words, we can develop a variety of bilingual education programmes for Pasifika students in terms of levels of immersion which would still be highly effective if these other research principles are also central to such programmes.
There is still a distinct possibility that we might be able to accomplish this in Aotearoa/New Zealand for both Mäori and Pasifika students, although the current MOE proposals for Pasifika bilingual education do not, it appears, allow for this as yet - since they remain wedded to a 'transitional' and 'subtractive' view of bilingualism and bilingual education.
Should these MOE proposals be implemented as they currently stand, they are not likely to address, in either the short or longer term, the ongoing language shift and loss currently facing Pasifika communities in Aotearoa/New Zealand, nor are they likely to significantly improve the educational experiences and outcomes of Pasifika communities.
Which brings me to my final point - if the process of language shift and loss, as well as the longstanding (and ongoing) patterns of Pasifika educational disadvantage, are ever to be successfully addressed, challenged and changed, it requires all of here, but particularly Pasifika educators and communities, to argue publicly and privately for the merits of bilingualism and (effective) bilingual education. This involves not only highlighting some of the key educational issues underpinning bilingual education, but also requires all of us to directly confront and correct the various myths and misperceptions about bilingualism and bilingual education that I've discussed with you today.
This is crucial, because these myths are clearly still regularly promoted - not only by teachers, educators and policy makers, but also by and within Pasifika families and communities themselves. 

Unless we change those attitudes within our families and wider communities, the battle for Pasifika languages and the better education of Pasifika children, may already be lost. 
What the various myths and misperceptions that I have discussed today clearly indicate is that the battle for bilingualism and bilingual education will be ongoing for some time to come and that victory, or even progress, is by no means assured. All the more reason then, why we need to continue the fight - not only in specific relation to education, but also simply in order to ensure that we can continue to speak a first language as a basic human right. Why is this so important? Because, as the political philosopher, Will Kymlicka has observed: 'Leaving one's [language] and culture, while possible, is best seen as renouncing something to which one is reasonably entitled'.
Contact details:
Professor Stephen May BA Hons, M.Ed, PhD, Dip T (Sec) MRSNZ
Foundation Professor and Chair of Language and Literacy Education

No comments:

Post a Comment