My interest in fluency originates from some different experiences. When I lived in Alice Springs among several of this country’s strongest languages I had the privilege of working with some extraordinarily patient and persistent teachers. Any positive outcomes I had in learning their languages were, I am quite sure, far more due to their ability than mine. Nevertheless our mutual success was such that I was eventually able to function across at least two languages in some very limited and highly predictable social settings. Included in these, at one stage, was a role coordinating the delivery of beginner classes in those languages for the Institute for Aboriginal Development (IAD).

268Although my primary responsibilities in this operation were logistic we always worked as a team delivering lessons in the classroom, probably because my teachers also saw this as an economical way of continuing my apprenticeship. And, on those occasions when unforeseen circumstances caused the real teachers to be absent, I knew I had their confidence, if not always my own, to keep the customers satisfied and pursue the scheduled activities until their return. One of the things I learned from this experience was, like the teacher who stays only one lesson ahead of their students, as long as your fluency is greater than someone else’s they really have very limited capacity to accurately assess yours and will often significantly overestimate it, especially when you’re the one standing in front of the class!

Continuing to work and socialise with my teachers and their friends and families over several years greatly improved my speaking and listening abilities and exposed me to a range of everyday expressions and interactions that were rarely touched upon in classes. Their close attention to my pronunciation also helped me minimise my English-speaker accent as much as I was able. A consequence of their persistence and my still quite limited capacity to hold a conversation was that native speakers, or local multilinguals, would sometimes mistakenly assume that I could speak a language right through. My teachers’ very flattering tendency to also assert this on my behalf, while no doubt largely intended to offer me encouragement, contributed significantly to creating that illusion. I knew I still had the language skills of a learner. But moderately better pronunciation than the average whitefella and the capacity to understand and make simple jokes, for example, meant that speakers could be misled about my fluency for a short time at least. This taught me not only the importance of a good accent and authentic expression, but that non-expert speakers can easily make inflated assessments of someone’s fluency in a language based on only a slight amount of evidence.

Subsequent travel in non-English-speaking countries has reinforced this awareness. I now understand only too well how a reasonable accent and a few memorised phrases can quickly get me into or out of some very difficult situations. And the effect operates in both directions; when local people are taught stock English dialogues for use with foreign visitors it can take a while for the traveller to realise that any unexpected answers or deviation from the script are largely incomprehensible to their new friend. While many bilinguals and linguists might consider these ideas self-evident they are, however, not at all obvious to those who dwell in a monolingual environment.

Now operating in Indigenous languages education in south-eastern Australia2 my contact is mostly with people who are engaged in a quest to develop fluency in their ancestral languages and supporting others to achieve similar goals. I am also directly involved in training and assessing those people who wish to be professionally recognised as teachers of those languages. In such contexts fluency is a central concern.

269While the majority of the owners of most languages in southern Australia are currently also non-speakers it can be especially difficult for them to establish who is fluent and to what extent. Most people, apart from a few elder speakers, are at an early stage of their journey towards fluency and therefore have limited capacity to accurately assess the fluency of others. There are also people who are taking matters into their own hands and endeavouring to teach themselves their language from learners’ guides, dictionaries and wordlists.3 While this is an admirable ambition, unaware of the sound and grammatical systems of their language they can end up making simple statements, but without the necessary detail to make clear who did what to whom and whether it happened yesterday, today, or is yet to occur. Coupled with a convincing accent and rapid delivery this can be very impressive to non-speakers, but any fluent speakers probably wouldn’t regard it as real fluency, or even real language, assuming they could recognise it at all. And this is clearly a concern for those who can tell and have an interest in revitalising their languages as faithfully as possible.4

Of course in languages education, teachers are required to constantly assess their students’ language abilities and the concept of fluency is directly relevant here. But even more importantly, in languages teacher training there is a justifiable assumption that accreditation has some connection to language ability. And learners of languages usually have an expectation that their teachers have an appropriate level of ability to perform the task, as do those who employ them.

So my interest in fluency stems from several positions – outsider and insider, language learner, speaker and teacher, linguist and trainer of languages teachers. It is definitely not as a gatekeeper with a desire to apply set standards, although I am required to deal with authorities that would very much like me to. Mostly it is as someone who is interested in seeing Australian languages survive and flourish and supporting individuals and communities to attain that goal.

What is fluency?

Fluency is an unfortunately vague term. It can be used to refer to both the ability to speak a language smoothly and a person’s overall capacity to communicate in a language as indicated by speaking it; two measures that have an obvious connection. Thus we can identify someone as a fluent speaker based on the lack of hesitation or interruption in their speech and the absence of particular disturbances such as stuttering. We can similarly identify someone as a fluent reader or even writer. To avoid confusion and focus particularly on the ability of a person to communicate meaningfully in a language through speech, linguists and language educators usually prefer to use the term oral proficiency.

270Of course a person’s total language abilities consist of more than just oral proficiency, and languages teachers are accustomed to dealing in terms of the macro-skills of speaking, listening (understanding), writing and reading. It is also widely recognised that learners’ abilities in listening typically precede those in speaking. However in everyday contexts the primary indicator of overall language ability is normally taken to be speaking, which is commonly discussed in terms of fluency, and I have chosen to follow that use.

Is fluency relevant to revitalising languages?

Clearly many people believe otherwise. Simply raising fluency as a topic of discussion with those involved in language revitalisation in Australia can bring a rapid halt to conversation or suspicion of gatekeeper motivation. In the context of a recent conference presentation that was, rather tellingly, quite poorly attended my audience reached their own consensus that it was simply too far removed from their circumstances to warrant consideration. Such reactions continue to provide me with concern and motivate my persistence.

Surely if the ability to speak a language is irrelevant, we are not discussing revitalisation so much as awareness. If a language is to be re-awoken to live again then a principal goal must be to have people speak it (Fishman 1991; Hinton 2002). And, if people begin to speak a language, then they must be expected to improve that ability to some extent, or we are only talking about language maintenance. Of course, for some languages that have little recorded information and no surviving speakers, the ultimate goal of revitalisation may be simply speechmaking or the mastery of a few fixed phrases. Even so the change from non-speaker to speaker in such contexts represents a positive change in fluency that we can at least observe and discuss, and assist people to achieve.

Where language revitalisation efforts are in their early stages and not many people have significant fluency, to focus on it might seem disheartening, even embarrassing, for some. Especially where claims for recognition and possibly even funding are involved, there might also be fears of negative outcomes if the truth about current levels of fluency in the community were known. Those concerns are understandable and not without some justification. However, in the long term, I believe they are also likely to be counter-productive. The assumption that progress is being made as long as some teaching-like activity is taking place and people are engaged and feeling good, may be quite reassuring. But unless people are actually developing greater fluency, it seems to me that revitalisation is not really happening.

To make a language vital again requires its speakers to progress from less to more fluent, both individually and as speech communities, even if the ultimate goal is not as lofty as restoring a first language speaker population (see Meakins, this volume). While such outcomes might conceivably occur naturally they are far more likely to be successful if they involve some language planning, and to plan for an increase in fluency requires some measure of both starting and end points as well as strategies 271to effect change. This is not to suggest that at either the individual or community level the measurement of fluency should be arbitrarily imposed. However, for those individuals and speech communities that can see benefit in knowing where their current skill level lies, it would certainly be useful to have the option and an appropriate mechanism available.

How is fluency measured?

Linguists and language educators have been measuring speakers’ fluency in many languages for many years and there is a wide range of highly developed testing methods available.5 Essentially all of them require the performance of some speaking task, the result of which is measured against some scale based on observation by someone with training and experience in the area. Tests of oral proficiency are also often married with tests of listening and, for written languages, with tests of literacy skills. Because test output at higher levels of fluency is more likely to be unique, its measurement is less likely to require specific words or strings to be uttered so much a judgement made regarding its overall communicative adequacy – is it only sufficient to perform basic fixed tasks like introduce oneself, enough to perform in a workplace, or sufficient to freely converse with native speakers on any topic?6

While the exact nature of the tasks may vary, the scales of measurement tend to be fairly consistent, although variously proposing finer or coarser grades of measurement. Usually each point on a particular scale is given a descriptive title, possibly a number, and an extended description of the functional indicators for assessment at that level. Some internationally popular and electronically accessible scales include the Canadian New Brunswick second-language oral proficiency scale (Government of New Brunswick, n.d.), the American Council on the Teaching of Foreign Languages (ACTFL) Proficiency guidelines: speaking (1999), and the Stanford Foreign Language Oral Skills Evaluation Matrix (FLOSEM) (Padilla & Sung 1999) that measures comprehension, fluency, vocabulary, pronunciation and grammar. The dominant scale in Australian settings, however, remains the Australian second language proficiency ratings (ASLPR) (Wylie & Ingram 1995a; 1995b)7 that provides the following developmental series: 272

But these are not the only means of assessing fluency. For example in languages education in schools, teachers should be familiar users of a range of assessment activities and measurable speaking objectives that derive from syllabus documents such as the New South Wales [NSW] K–10 Aboriginal Languages Syllabus (Board of Studies NSW 2003) and its associated support materials, although these may themselves have originally had some basis in scales like the ASLPR. Colleges, universities and community agencies are similarly providing courses that are generally recognised as indicating, at least, implied levels of fluency among other language skills.8 So the measurement and certification of fluency in revitalising Australian languages is already actively being undertaken both by government and community agencies. And that, rightly or wrongly, intentionally or not, affords them a considerable level of control.

273Of course scales such as the ASPLR have been principally designed for vital languages from around the world with the assumption that near-native fluency is achievable for second language learners. They are also clearly based in a modern Western worldview, a fact that raises questions of cross-cultural appropriatness. Some may also assert that the notion of formal testing itself is inherently non-traditional for Indigenous Australians. This may be true, but no more so than the idea of formal second language classes to learn one’s ancestral language, language centres, dictionaries or literacy.

McConvell (1994) addresses some of these concerns and provides a sample alternative testing instrument for one Australian language, Kija. Although the material discussed is specific to that language it provides an excellent model from which other language-specific tests could readily be developed. In the North American context the Ganöhsesge:kha:’: Hë:nödeyë:stha (Faithkeeper’s School) that teaches in the Seneca language has undertaken a comprehensive adaptation of the standard FLOSEM instrument to produce a culturally sensitive and appropriate scale for their own use (Borgia 2009).

How fluent do teachers need to be?

While the measurement of fluency can be construed as at least useful and relevant for individuals and communities engaged in revitalising their languages, it becomes critical for those who are required to use a language professionally. This is nowhere more so than for languages teachers. Both students and providers would normally have a justifiable expectation that someone working as a teacher of any language would have a reasonable level of fluency as well as being competent to foster its development in learners. Although in the early stages of revitalisation it is conceivable that the teacher might be literally only one lesson ahead of the class, or even on the same page, after languages education has been in effect for some time those who have a history of participation in the process would hopefully have significantly higher levels of fluency than beginners, and be able to feed their skills growth back into the community revitalisation cycle.

Australian primary (elementary) teaching qualifications do not normally require a languages component. However for secondary teaching the NSW Institute of Teachers (NSWIT), for example, currently specifies a minimum standard of a language major to qualify as a designated languages teacher; a major being, ‘a defined program of study in a designated area, generally comprising 3 years of degree level study of 6 semester long specified units of study or equivalent, including 4 units from later stages of the program (level 2 or above)’ (2008, p. 3), in addition to languages pedagogy requirements.

During initial discussions with the NSW Department of Education and Training’s Teaching Qualifications Advisory Panel (TQAP), a precursor to NSWIT, regarding recognition of the University of Sydney’s Master of Indigenous Languages Education (MILE),9 a figure of 200 hours post-secondary study in an Aboriginal language or

274languages was suggested as a minimum standard of fluency.10 Equally telling, but more functionally defined, the ASLPR scale for second language teachers does not commence until Level 2, Basic Social Proficiency, for regular modes of teaching with a minimum standard of Level 4, ‘Vocational’ Proficiency, for immersion or bilingual programs (Wylie & Ingram 1995b, p. iii).

In many Australian languages currently undergoing revitalisation such standards are unlikely to be achievable for even the most fluent teachers. If, for instance, a language is not yet offered to the final year of high school or in any university it is simply not possible for any teacher to have achieved to such levels themselves. And if there has been a breakdown in transmission over several generations and only a few elder speakers exist, or none at all, similarly skilled candidates are unlikely to exist in the community. Sadly this situation describes most, if not all, the languages of southern Australia and there would be few, if any, teachers who could realistically satisfy requirements for the lowest ASPLR teaching standard of Basic Social Proficiency. Given that each revitalising language is probably at a different point to every other, a standard to be applied across all, even within a single state, would be impossible to determine. And, if the process of revitalisation produces improved fluency across whole speech communities, both minimum and maximum standards must necessarily be expected to change over time.

Fortunately the various education authorities that permit Australian languages to be taught in schools have largely responded pragmatically to date and allowed languages to be taught by those who simply have some knowledge of the language and a preparedness to engage in school classrooms. These may be qualified Indigenous teachers but not normally with languages teaching accreditation (or training) or any certification of fluency. They may also be Indigenous community members without teaching qualifications but ‘some’ knowledge of the language working alongside a qualified teacher, who may or may not have languages education training themselves. In some cases they may even be non-Indigenous. The dedication and commitment of these people is not in question here, but their potential to continue without further fluency development as the languages are revitalised warrants consideration.

This situation is not likely to persist indefinitely and, as revitalisation and particularly school-based languages programs develop, it is increasingly likely that educational authorities will move to pursue a goal of parity for Indigenous Australian languages taught in schools with those originating from outside Australia. The limited accreditation of the MILE to 2010 is telling in this regard:

Such measures suggest that the imperative for government to apply ‘standards’ to Indigenous languages educators is looming large on the horizon, and the push for professionalisation should be anticipated, especially in states like NSW where a standardised state syllabus and expanding implementation is rapidly normalising them in the languages key learning area.11

What’s happening overseas?

Questions of fluency are not restricted to the Australian languages revitalisation process and it is of value to consider some of the responses from commonly compared situations overseas.12

The example of Aotearoa (New Zealand) is typically sophisticated and inspiring, but equally removed from the realm of possibility in Australia today. It nonetheless is worth considering as a possibly ideal goal. A single language and single state government together with legislative recognition of te reo Māori permits a formal testing regime applied by the Māori Language Commission (MLC):

Candidates initially sit a one-hour Level Finder Examination to assess general ability across all language macro-skills and may then undertake either of the two-hour

276Public Sector Māori or Teaching Sector Māori examinations. The standard for teachers is the highest. Accreditation of fluency for teachers and others is also possible through the university and college systems, much as for (non-Indigenous) languages education qualifications in Australia. The MLC exams provide an alternative means of certification for those who have not completed coursework, or who have increased their fluency by other means and wish to obtain a revised measure, or would just like to know how their current abilities rate.

The situation in the United States of America entails substantially greater diversity than here based on the sheer number of state jurisdictions and the considerable autonomy of local school boards, but has some similarities to both Australia and New Zealand at its extremes. Strong revitalising languages, especially those that have access to a substantial resource base, may implement their own fluency certification regime or have access to accredited university or college coursework options, as well as school-based programs. Thus for the Navajo (Diné) language, ‘Individuals seeking the Navajo Language Endorsement in New Mexico or Arizona are required to take the Navajo Language Proficiency Test. Diné College is authorised by the Navajo Nation to administer this test.’ (Diné College 2008, p. 32). Although, as Dean of Humanities and Social/Behavioral Sciences, Wesley Thomas pointed out, other community agencies operating in the Diné language offer fluency certification for teachers acceptable in some schools based on only a brief interview (pers. comm., 10 August 2007). Anecdotally, for smaller languages in the USA the situation is mostly similar to that in Australia; those who say they can, and are prepared to, can participate in teaching revitalising languages as long as the school community permits.

In Canada, self- or community-selection of languages teachers is also possible as is course-based certification through universities, colleges and seasonal institutes. However of greatest interest is an initiative from British Columbia (BC), a province that entails some linguistic situations directly comparable to many in Australia. Here the BC College of Teachers (BCCT) has developed in collaboration with Aboriginal community interests a system for the accreditation of First Nations language authorities recommended by a tribal council or other body acceptable to the College. These authorities may issue Interim First Nations Language Teacher Certificates to ‘… proficient First Nations language speakers … [whose] proficiency is determined by the recognized Language Authority, and the Language Authority recommends … for certification.’ (BCCT n.d., p. 1). Remarkably, as Beverley Maxwell, the BCCT director of certification advised, how the authorities determine proficiency is entirely their concern as it is their language (pers. comm., 16 July 2007). There is also a clear assumption that standards and certification methods will vary over time according to the current health of each language.

These certificates only permit the holder to teach classes in a specific language, and have potential to be made permanent. But the preferred outcome is for students to undertake formal teacher education through a program such as the laddered model auspiced by the University of Victoria. Through this program, certificate holders may take 277further training to obtain a Certificate in Aboriginal Language Revitalisation offered in partnership with the University’s Division of Continuing Studies, Department of Linguistics and the En’owkin Centre, an accredited First Nations language authority. Further study in languages and education leads to the award of the Developmental Standard Term Certificate issued by BCCT that allows the holder to teach in BC elementary (primary) schools, but has a standard term of four years within which a full teaching degree must be obtained. An additional two years full-time education coursework at the University of Victoria leads to the award of a Bachelor of Education (University of Victoria, Faculty of Education, n.d.)

Could community certification of fluency work here?

Devolving the certification of fluency to autonomous indigenous agencies might seem extraordinary viewed from within the current Australian environment. Yet it does not appear to have caused the downfall of Canadian Aboriginal education, or Navajo or Māori, and has much to recommend it. In fact, to a limited extent, community-controlled certification of fluency does already exist here. Indigenous language centres such as Muurrbay and the IAD have, through the provision of their accredited language courses, been acting as de facto certifying agencies for many years without apparent harm.

Acknowledging the right of Indigenous Australian communities to decide the standards for their languages and those who teach them would afford the potential for self-determination, in language revitalisation at least. It would put government authorities at arm’s length and give communities the status of ultimate judges of a cultural expression that should be undeniably theirs. It would relieve government, linguists and the rest of non-Indigenous Australia of any illusion that they need to be controlling the future of Australian languages and allow the transfer of responsibility back to community hands.

To broadly implement such a strategy would require a number of major steps, each requiring much consultation and negotiation. Existing community language agencies would need to develop language-specific materials and procedures for the local administration of testing. Where no such agency existed one would need to be established, possibly auspiced by other Indigenous bodies with a resource base and cultural role such as land councils, and with assistance from government or other interested institutions like universities. The potential to act as certifying authorities would, of itself, lend weight to the need for such agencies to be established and provide them with an immediate role in addition to the great deal of other valuable language work they could potentially undertake.

A system of accreditation for certifying agencies would need to be implemented together with a mechanism for meaningful and practical recognition of their authority. The BC example suggests that state-based professional teacher registration bodies would be suitable candidates, but school education boards of studies, vocational education and 278training authorities, and similar agencies could also be involved. Indigenous language authorities could then be given a place at the table in the accreditation of courses offered by those providers, further consolidating their role as well as determining how fluent a teacher of their language currently needs to be.

Mechanisms for articulation with training opportunities in languages education and other forms of language work could be developed, as in BC. Fishman (2001) has identified the critical role of sociocultural reward in motivating individuals to learn their language. If employment opportunities were aligned with the certification process the benefit of achieving fluency would be very clearly defined. But, for many, achieving a level of fluency in their language that was certified by their community would be reward enough, as it should be.

Of course such a system would need to allow for considerable variability by language and over time to take account of the dynamic nature of revitalisation. The optimum level for fluency in a specific language at a particular point in time would necessarily be different to another language and as community levels of fluency rose. For that reason the application of limited term certification might also be considered.

Conclusion

The measurement of fluency in Indigenous Australian languages is possible and is already being undertaken by schools, colleges, universities and community agencies. Culturally appropriate materials, methods and scales have been developed. As languages revitalise, assessing the fluency of individuals and communities has potential to assist in planning the future of that process.

Indigenous communities can pursue their current activity in revitalisation without regard to fluency or seek to exercise control. If they don’t it is probable that governments, particularly through education and teacher training, will increasingly do so.

Other indigenous populations have developed their own systems to deal with questions of fluency in their languages. A model that seems particularly appropriate is that applied in British Columbia where community-based language bodies have been established and exercise authority recognised by government for the measurement and certification of fluency in a framework of articulated qualifications for teachers and other language workers. Consideration of a similar model for the Australian context by language owners and other interested parties is suggested.

Postscript

As this volume was going to press, the NSW DET advised that continued recognition of the University of Sydney’s MILE as an acceptable qualification for Aboriginal languages teachers beyond 2012 would require the inclusion of ‘at least two units of study (or equivalent) in the Aboriginal language the applicant intends to teach’ (pers. comm., 21 September 2009). Fortunately, after several meetings where the Koori 279Centre asserted the current impossibility of compliance for any Aboriginal language in NSW, a practical way forward was found and the combination of linguistics and research units within the degree were deemed to satisfy this requirement (pers. comm., 28 January 2010).
 

References