From Home to School in Multilingual Arnhem Land: The Development of Yirrkala School’s Bilingual Curriculum

Abstract

This chapter presents a case study of Yirrkala School, a remote Indigenous school in the Northern Territory of Australia that has succeeded in maintaining its long-standing bilingual programme in spite of great pressure to move towards a mainstream, English-only model. Bilingual policy and education in the region is discussed, and the history and current realities of Yirrkala School itself are described. The chapter draws on a set of interviews conducted in 2015–16 with community members intimately involved with designing and implementing the local school curriculum, as well as on various community-based publications spanning several decades. These are used to articulate local perspectives on the role of language and culture in the classroom and the journey of two-way education in Yirrkala.

Introduction

The Northern Territory of Australia comprises a vast network of complex intersecting language ecologies (e.g. Mufwene and Vigouroux 2012; Meakins 2014), encompassing over 100 Australian Indigenous languages (NT Gov. 2016), as well as a wide range of contact languages: creoles, mixed languages and varieties of English. The region’s 1.35 million square kilometres is also home to some 245,000 people, around 70,000 of whom are Indigenous (based on 2011 data (ABS 2016)). Wilson (2014) reports that 65% of Indigenous people in the Territory use an Indigenous language as a home language, with the proportion much higher for those in remote communities.

Indigenous children live, learn and communicate within ‘shifting langscapes’ (Angelo and Carter 2015): linguistic contexts that have been shaped by extensive language contact and language shift. Furthermore, many are born into families and communities where English has little or no presence. And yet for the vast majority of Indigenous children, entering the school system means contending with an English-only or English-dominant environment. This is a result of a nexus of factors, including the apparent ‘invisibility’ of Indigenous children’s linguistic repertoires (especially for speakers of contact languages (see McIntosh et al. 2012; Sellwood and Angelo 2013)) and a broader ‘monolingual mindset’ (Clyne 2008) in Australia, reflected in most top-down national and state language policies.

Such linguistic juxtapositions are by no means particular to the Australian context; similar situations and consequences for Indigenous education are observable in many parts the world (see, e.g. Romero-Little et al. 2007). This is in spite of the acknowledged right of each child to be educated in his/her own language, as recommended by the UN Committee on the Rights of the Child,¹ the positive outcomes for children’s sociocultural well-being (Biddle and Swee 2012; Marmion et al. 2014; People Culture Environment 2014) and the proven efficacy of many programmes incorporating home language(s) for eventual literacy and numeracy outcomes (see, e.g. Baker 2011; Siegel 2007).

School responses within the context of these discourses and policies have varied, and this is what we concern ourselves with in this chapter. We present an exceptional case in the Territory—a school that has succeeded in maintaining its long-standing bilingual programme in spite of great pressure from many factions to move towards a mainstream English-only model. We first discuss the history and current reality of bilingual policy and education in the region, before turning to Yirrkala School itself. We draw on a set of interviews conducted in 2015 and 2016 with a range of community members intimately involved with designing and implementing the local school curriculum, as well as on various community-based publications spanning several decades. We use these to articulate local perspectives on the role of language and culture in the classroom and the journey of two-way education in Yirrkala.

The interviews were conducted by the first author, a non-Indigenous researcher/linguist who spent a total of 12 months onsite conducting research at Yirrkala School. The second author is also a non-Indigenous linguist working on multilingualism and linguistic variation in Arnhem Land. First-hand information comes from the chapter’s third author, who is a Yolŋu community member as well as the Yirrkala School principal-in-training.

Bilingual Education in the Northern Territory

Policy and Practice

In the Northern Territory, government policy has both at times created space for, and at other times firmly excluded, Indigenous languages from the schooling system. In the 1970s and 1980s, such top-down approaches empowered the classroom as a site for fostering community language use. However, the tendency in recent years has been towards the explicit or indirect dismantling of existing bilingual programmes and the continued privileging of English-language instruction; as Nicholls (2005) describes it, ‘death by a thousand cuts’. We do not provide a detailed discussion here of the history of bilingual policy and education in the Territory; we refer the reader to the numerous excellent accounts that exist already (e.g. Devlin 1995, 2011; Disbray 2014, 2015a, b; Devlin et al. 2017; Gale 1994; Harris 1995; Harris and Devlin 1999; Simpson et al. 2009; inter alia). Instead we include a short description of the trajectory of bilingual policy in the region over the last four decades in order to situate the discussion of the local Yirrkala context that follows.

In 1972 a bilingual education initiative was announced in the Northern Territory by the federal Whitlam government. In the initial years, the programme was rolled out across 24 schools, with each developing language and literacy resources in community languages, and with staff receiving support from the Australian Department of Education (NT Division) and, from 1979, the NT Department of Education (Devlin 1995; Disbray 2015a). Many of these programmes continued into the following decades, but the ideological space within which top-down policies were shaped shifted significantly over time. In the 1970s, the genesis of bilingual programmes emerged naturally from a broader global discourse of Indigenous rights and empowerment, although by the following decade, community language use in the classroom was largely framed in policy as a means by which to achieve better English literacy (McKay 2007).² As the bilingual programmes developed, interest in Aboriginal-language literacy and teacher training grew. This need was met by teaching programmes offered by Batchelor College (now Batchelor Institute for Indigenous Tertiary Education (BIITE)) and literacy courses at the School of Australian Linguistics, through which training was provided in the creation of teaching materials and curricula in Indigenous languages (see Black and Breen 2001). The 1990s, however, saw a severe decline in the number of trained Indigenous teachers, including those proficient in teaching literacy in their own languages. This was in large part due to a reduction in Indigenous training from BIITE, who had provided such important in-community training and support.

The late 1990s also saw a backlash from certain camps within the government and the education system: for many, the connection between vernacular-medium instruction and English literacy was difficult to understand, and the bilingual programme was seen to be expensive, requiring extra materials and staffing (in fact, the programme had already been broadly neglected and under-resourced). In the face of strong community support for the programme, however, the government commissioned a review, whose results were presented in what is known as the Collins report (Collins 1999). The report supported the programme’s continuation and recommended increased training and support for teachers. It also recommended a ‘rebranding’ of the programme as ‘two-way learning’. The government acquiesced to this name change, but to nothing further. Indeed, a number of programmes were shut down during this period, typically at the behest of school principals rather than community (Simpson et al. 2009). Furthermore, the programmes that remained were now mostly ‘transitional’ programmes rather than genuinely ‘two way’. Transitional programmes do not aim to develop home language and culture; rather, they use the home language in the early years as a tool to facilitate content learning and second-language acquisition. The home language ceases to be used for schooling once students have gained control of the second language (Baker 2011).

A yet more severe turn occurred in the following decade, with the 2003 Ramsey report calling into question the value of Indigenous language literacy and voicing concerns about children’s English skills. The 2008 decree from the then Northern Territory Minister for Education that ‘the first four hours of education in all Northern Territory schools will be conducted in English’ (Scrymgour 2008), effectively undermining the possibility of any truly bilingual programme, represents the unfortunate consequence of this ‘deficit discourse’ (Vass 2012). In an attempt to justify the decision, Scrymgour presented data from the NAPLAN³ national benchmark testing, which had been first administered that year, but this has since been shown to be incomplete and flawed evidence (Devlin 2009; Dixon 2010). In spite of the national, government-commissioned Our Land, Our Languages report in 2012, which recommended that ‘state and territory governments […] provide adequately resourced bilingual school education programs for Indigenous communities from the earliest years of learning’ (HRSCATSIA 2012, p. 4), the subsequent Wilson (2014) review of Indigenous education made no provision for Indigenous languages or bilingual education as part of its long-term recommendations.

Although the First Four Hours policy was scrapped in 2012, the damage had been done. A year later only eight schools in the Territory reported running a bilingual programme for more than two hours a week, with a further 17 incorporating some kind of Indigenous language programme (Wilson 2014, p. 115). This number was down from 60 total programmes in 2011 (including bilingual programmes, but largely language and culture programmes)⁴ (HRSCATSIA 2012).⁵

Curriculum and Local ‘Langscapes’

Shifting bilingual education policy and the privileging of English-medium instruction mean that the complexities of community language ecologies are typically not reflected in the formal design of classroom practice in non-bilingual programmes.⁶ While children may bring diverse linguistic repertoires with them to the classroom, educators tend to understand the space as monolingual. Thus there may be a mismatch between children’s home and community language(s) on the one hand, and the medium classroom information is transmitted in, and through which learning is expected to be assessed on the other (Angelo and Carter 2015, pp. 6–10). This is not to say that teachers are to blame for these challenges. By and large teachers are not prepared adequately for the complexities of their students’ linguistic repertoires and needs, as very few are trained in teaching English as an additional language and many arrive at schools unaware of the local language ecology. Yet it is within this challenging pedagogical context that teachers are expected to prepare children for standardised tests which do not adequately acknowledge their diverse language backgrounds in measuring achievement (Angelo 2013). To further complicate matters, many teachers working in a remote community will stay only for a year or two before moving on (Collins 1999, p. 75).

The general picture, then, is that top-down policy and the realities of the classroom are very often at odds with local discourses and language ideologies, which may value the teaching of community languages in schools alongside English. Curricula have been subject to criticism for not reflecting and incorporating local epistemologies. Prominent Indigenous educator and advocate Dr Marika commented:

I question whether current trends in Australia regarding curriculum and assessment, particularly the national profiles and the benchmarking process, are inclusive of other knowledge systems and languages and find them lacking. (Marika 2000, p. 46)

As Disbray (2015a, p. 10) notes, the prospect of incorporating more ‘place-based perspectives’ is daunting to policymakers who desire one-size-fits-all approaches that can be rolled out state-wide. And yet in spite of top-down pressures to replicate a Western, English-only curriculum, a number of schools have found or created spaces for multilingual and multicultural recognition and learning, demonstrating that educators and local stakeholders can, and do, exert their agency to resist dominant paradigms and acknowledge community priorities. Indigenous Language and Culture programmes (developed as a ‘learning area’ in the Northern Territory Curriculum Framework (Disbray 2015b) but not enshrined in policy) are employed in a number of schools, although they vary significantly in their implementation. Programmes may involve planned teaching, albeit of not more than a few hours a week, bush trips foregrounding important knowledge and language, and other cultural activities and instruction. A small number of schools, however, have succeeded in maintaining fuller bilingual programmes. In the following sections, we introduce one such programme: Yirrkala School in northeastern Arnhem Land has succeeded, within the historical trends we have seen, in continuing with their bilingual programme, even without appropriate funding and support.

Yolŋu and Yirrkala

Arnhem Land is a region located in the northern tip of the Northern Territory of Australia. The Indigenous people inhabiting the northeast of this region are known as Yolŋu. Yolŋu continue to follow the belief systems and rom ⁷ that were passed down to them by their ancestors. They have a complex system of society-wide relationships, including a classificatory kinship system. Everyone and everything in the Yolŋu world is allocated to one of two patrimoieties: Dhuwa and Yirritja. Dhuwa people belong to Dhuwa land, speak Dhuwa languages and perform Dhuwa ceremonies; Yirritja people are responsible for Yirritja land; they speak Yirritja languages and perform Yirritja ceremonies. Dhuwa and Yirritja are two halves of one whole. For example, Yolŋu must marry a member of the opposite moiety. The moiety groups are further divided into smaller family groups called clans. Each clan has its own language, homeland, totem, traditions, ceremonies, songs, creation stories, and dances. Yolŋu inherit their clan and moiety from their father.

It is critical to know one’s place in this complex system, as membership in these groups governs how Yolŋu relate to one another and to the rest of the world. For example, when a Yolŋu passes away, clan of their mother’s mother is responsible for carrying out specific ceremonies. Each Yolŋu child must understand this rich and complex system of cultural knowledge, rights and responsibilities to be a fully functioning member of the community. An individual’s social position within this cultural matrix is not merely ideological; it is lived and actively negotiated every day. Language is fundamental to these processes, as both the means and an end goal of socialisation processes (Ochs and Schieffelin 2011). As Merrkiyawuy Ganambarr-Stubbs, Yirrkala community leader and co-author of this chapter, pointed out in June 2016, ‘without our language, everything is meaningless’.

Yirrkala’s Language Ecology

Yirrkala is a very remote⁸ Indigenous community in northeastern Arnhem Land, 25 km southeast of the mining town of Nhulunbuy and 700 km east of Darwin. The Methodist church established Yirrkala mission in 1935, bringing together different clans of Indigenous people from the Yolŋu bloc. A homelands movement, whereby clans moved back to their original lands to live in smaller, more traditional communities (Yirrkala Literature Production Centre 1991), gained traction in the 1970s, although the region’s first homeland community, Gapuwiyak, had been established in 1948. Currently, the population of Yirrkala ebbs and flows as people live partly in the community and partly on their original homelands. If all members were present at one time, the population would be around 800.

Yolŋu Matha

Yolŋu Matha (YM) is a pandialectal cover term encapsulating approximately 30 language varieties spoken by Indigenous people in northeastern Arnhem Land. Yolŋu refers to the people while matha means ‘tongue/language’. Each clan has its own variety of YM, with many named after the proximal demonstrative, that is, the clan-specific word for ‘this’ or ‘here’. Although the different varieties of YM might, in linguistic terms, be considered dialects of the same language, the social reality for speakers is that these are separate languages. The different clan languages are mutually intelligible, distinguished largely by grammatical morphology and some minor lexical differences (Amery 1993, p. 47). Although there are two different branches of YM languages, shared features, as well as consistent contact between speakers of different clan languages, ensure mutual understanding between their speaker communities.

The result of this bringing together of different clans is that many distinct varieties of YM are spoken in Yirrkala, with speakers of at least 18 different YM clan languages currently residing in the community⁹ and several others no longer spoken. In Yirrkala, Gumatj and Djapu have the largest speaker communities among these clan lects, and another four have more than 20 speakers. The rest have between one and five speakers. There is a further YM dialect, Dhuwaya, spoken by all community members regardless of clan affiliations (see discussion below). Although English is not a daily language, many Yirrkala community members have some proficiency due to the schooling system and the community’s close proximity to Nhulunbuy; competence levels vary from very limited to relatively high.

Tradition dictates that children should learn their mother’s and maternal grandmother’s languages at a young age. As they become young adults, they begin learning their father’s language—their own clan language. Learning one’s esoteric clan lect is a crucial step in becoming strong in Yolŋu identity; it is through the acquisition of this language that Yolŋu learn about their rom and culture. They learn about their songs, their land, their ceremonies and cultural responsibilities. Through their language they learn how they are connected to different clans and different lands: ‘It was not until I spoke in my own language, Rirratjiŋu, that my view of the Yolŋu world became more meaningful’ (Marika 2000). Yolŋu also learn to identify an interlocutor’s clan affiliation by their speech, which allows them to determine their relationship with one another (Yirrkala LPC 1991).

Currently, however, all children are acquiring Dhuwaya among their main languages, and many continue to speak it during adulthood. Dhuwaya is different from the rest of the YM varieties in that it has no clan affiliation: it is spoken by members of all clans. It is a koine language—a variety that has emerged from prolonged contact between speakers of different dialects¹⁰ of the same language, and in this case originating at least in part from community baby-talk registers (Amery 1993, p. 55). All of the varieties contributing to the emergence of Dhuwaya are YM languages, and hence Dhuwaya comprises features from the different contributing clan lects. The process of koineisation tends to involve various processes of simplification (Siegel 1985), and Dhuwaya is indeed somewhat simplified (or, arguably, regularised (see Amery 1993, pp. 53–55)) in comparison to the clan languages (e.g. compare Dhuwaya’s four verb conjugation classes to Gumatj’s eight and Dhaŋu’s nine).

Dhuwaya is widely referred to as a lingua franca, as it was originally used as a common language in the community. Amery (1985) objects to the use of this terminology as the term ‘lingua franca’ designates a variety needed for successful communication between speakers of different languages. He argues that Dhuwaya was never needed to facilitate communication amongst different clans since the YM languages are mutually intelligible. Instead, Dhuwaya was created out of a social need ‘to stress solidarity within the peer group’ (Amery 1985, p. 128). As such, he prefers the term ‘communilect’, as Dhuwaya is only spoken in Yirrkala and its surrounding homelands.

Community members are therefore typically multilingual. Language choice in any given situation is dependent on a complex of factors, including interlocutor, domain and activity. Speakers may choose to speak their own language with everybody, although clan lects are most often spoken at home and with other members of the clan. Yolŋu also acquire the clan languages of other family members (certainly of their mother and grandmothers, but frequently also others) and will use them accordingly. While most Yirrkala community members continue to practice multilingual traditions, many younger Yolŋu are shifting towards more frequent use of Dhuwaya. The age at which people acquire their own clan lect is rising, creating changes in the home environment as an important site of language maintenance. This has led to concern among the older generation:

You have to like, look at yourself you know, and say “I’m a Djapu woman, my language is Dhuwal, it’s about time I have to speak my own language cause my grandmothers keep on telling me that I’m 31 years old and I’m still not speaking my own language, I’m still speaking Dhuwaya”. (Lirrina Munuŋgurr ¹¹ interview, November 2015)

[Y]ou’ve got young people growing up as parents speaking Dhuwaya who should be speaking two other languages… in addition to Dhuwaya. (Leon White ¹² interview, November 2015)

She was the first Yolŋu linguist over at Galiwin’ku and she’s a Wangurri lady […], speakers talked about her lamenting the fact that kids were growing up not speaking their clan languages […] [T]raditionally children would be growing up speaking their mother’s language and their father’s language, not just their father’s language so it’s really an important issue. (Leon White interview, November 2015)

Furthermore, within this complex language ecology, English is also used whenever speaking to non-Indigenous people who don’t speak YM.

Language and Curriculum at Yirrkala Community School

History of Yirrkala’s Bilingual Program

Yirrkala Community School (YCS) was established as a mission school in 1939. At first, YCS was English-only¹³ with the use of children’s home languages banned at school (Marika 2000). For many in the community, this linguistic barrier undermined existing local epistemologies and prevented real community engagement in the school programme:

[T]he missionaries didn’t realise that when they stopped us speaking Yolŋu language in the school, they were stopping our way of thinking. (Marika-Munuŋgiritj et al. 1990, p. 37)

The use of English made it difficult for Yolŋu children to understand what was happening in school since prior to the invasion of the missionaries, Yolŋu had had minimal contact with Europeans, and hence little exposure to English (Amery 1985).

The missionaries did, however, produce written materials in YM.¹⁴ Joyce Ross, a missionary linguist, had been translating Bible literature and hymns into Gumatj in the early 1970s. Thus, when the bilingual education initiative was launched, Yirrkala School stood out as a feasible location to initiate a bilingual programme: an orthography existed, the language had already been written down in church literature, a local linguist was available and community leaders were supportive of the idea. As a result, a team of linguists, educators, and community members made a concerted effort to develop the bilingual programme.

A major early challenge involved choosing which of the YM varieties would be the language of instruction. Gumatj emerged as the best choice for the bilingual programme for both practical and political reasons: Gumatj had already been studied and used in writing, and the language was understood by most members of the community. Moreover, the Gumatj were the traditional landowners of much of the surrounding area and among the most populous and powerful clans in the community (Yirrkala LPC 1991). The community understanding was that literacy skills in Gumatj would be easily transferable to other clan lects (Amery 1985, p. 10).

Gumatj literacy and curriculum materials then needed to be created, but this task proved challenging due to a lack of resources. The Literature Production Centre did not yet exist, and there were perhaps only two Yolŋu literate in their own language at the time. Furthermore, there was the significant challenge of harmoniously integrating the two languages’ distinct ways of talking about and classifying the world into one curriculum and expressing orally transmitted stories into a written format. For the Yolŋu teachers, this was a brand-new experience:

When we started biling. Ed the teachers found it hard to understand enriching first language. They had all been taught in a second language and teachers taught about another world. It took a while before they realised what we were on about – a completely new concept but once they caught on they loved it. (Beth Graham personal communication (email), February 2016)

As a result, the process of formulating well-written resources reflecting natural language use, and reshaping the curriculum to incorporate Gumatj, took the better part of the following decade. It was made possible only through constant cooperation between Yolŋu and non-Indigenous teachers and support from Yolŋu elders.

The use of Gumatj at the school, however, had always been contentious for both political and linguistic reasons. As already discussed, children learn culture and law through their own language, their mother’s language and their grandmother’s language; language is a core aspect of Yolŋu identity. While parents for the most part consented to their children speaking Gumatj, many were concerned about it threatening the use of other clan lects:

[A] lot of the people who live here are not Gumatj, and so when you start to tease it out, they could accept that Gumatj was used but they didn’t want it sort of replacing their own language. (Leon White interview, November 2015)

However, since Yirrkala had always been a community in which multiple related languages co-existed in a complex language ecology, Gumatj’s new status as a school language did not override entirely the existing dynamics of multilingual repertoires in contact. As a result, issues also arose due to the differences between the language used in the Gumatj readers and the language that children, and sometimes even teachers, were using. Reports soon emerged that children were experiencing difficulties with the language differences, particularly with the suffixes, and so while the school continued printing materials in Gumatj, teachers informally used Dhuwaya in the classroom, especially in the early years (Amery 1985).

In the mid-1980s, linguist Rob Amery came to Yirrkala to study and document Dhuwaya as a particularly linguistically interesting result of contact phenomena. In his interview, Leon White recalls how Amery’s work encouraged the community to talk about what the children were actually doing with language (i.e. developing Dhuwaya as a primary/first language) and to consider this variety as legitimate and worthy of attention. Moreover, by this time, classroom teachers, who had been reporting that the Gumatj stories were not always succeeding in engaging the students’ attention, began to request more literacy resources in Dhuwaya, the variety they were all using every day.

The school council¹⁵ discussed the use of Dhuwaya as a good alternative for the bilingual programme. It was a ‘neutral’ language in that it did not belong to a specific clan, and its use corresponded with the principle that children should be initially educated in the language they know best (Amery 1985) (see §1). It was also believed that using Dhuwaya at school would prevent further language change, as the variety would be subject to further codification and standardisation processes. Since Dhuwaya was understood to be ‘the closest to […] the baby talk, the lingua franca that the kids had sort of developed’, many feared that it was a transitional stage towards ‘further creolisation of Yolŋu Matha’ (Leon White interview, November 2015), seen as an unfavourable outcome (Amery notes that this concern is unfounded (1993, pp. 52–55)).

In 1987, Dhuwaya was established as the language of instruction (Yirrkala LPC 1991). Contention remains, however, over the use of Dhuwaya at school. Many Yolŋu still worry that the younger generation will not acquire their own clan language, and for some, Dhuwaya feels inappropriate for an academic setting:

That was one of my nhawi ¹⁶, arguments, because they have changed it to Dhuwaya and when I went to school everything was full-on Gumatj and it was more…how can I explain it? It was more, you had more challenging. […] [Gumatj] is strong and it’s more sort of…I think the way it’s structured, in a way it’s more of an academic way of language speaking. […] [I]t was more powerful than Dhuwaya…[Gumatj] is more like Standard Academic English. (Banbapuy Whitehead ¹⁷ interview, November, 2015)

For some teachers and community members, the ideal outcome would be the introduction of multiple clan languages to the classroom (Banbapuy Whitehead interview, December 2015), but it would be difficult to incorporate these into the curriculum due to limited funding and a shortage of qualified Indigenous teachers.

‘Aboriginalisation’ of the School

While older community members were on the whole pleased with the incorporation of YM language in the curriculum (Beth Graham interview, December 2015), many were not satisfied with the ‘Balanda’-oriented¹⁸ nature of many aspects of the school. Firstly, the school was heavily under Balanda control, and the non-Indigenous principal held sway over every decision regarding how the school was run (Marika-Munuŋgirtj et al. 1990). Furthermore, despite local efforts, the curriculum still focused almost entirely on Balanda knowledge and worldviews, with Yolŋu values and ways of knowing often undermined by the Western-dominant curriculum (Marika et al. 1990). Community leaders expressed a desire to restructure the school to incorporate a Yolŋu-oriented curriculum, one that would focus on community needs by building on topics deemed important by elders for fostering a strong Yolŋu identity (Marika-Munuŋgiritj et al. 1990).

In 1984, two groups were created with the intention of finding ways to exert Yolŋu control of the school (Marika-Munuŋgiritj et al. 1990; Stockley et al. 2017). The first, the Yolŋu Action Group, consisted of all Yolŋu staff at YCS regardless of position: administrative, clerical, ancillary, linguistic and teaching staff (Marika et al. 1990). This working body met weekly and made decisions regarding the day-to-day issues of the school (indeed it was at the instigation of this group in 1987 that Gumatj was changed to Dhuwaya as the language of instruction at the school). The second, the Nambarra School Council (now called Yambirrpa School Council), was made up of Yolŋu school staff from YCS and all homeland schools, as well as community members from all clans. The Council met several times a year and oversaw all major decisions across the schools, with the intention of ensuring that schooling respected Yolŋu beliefs and was in line with Yolŋu aspirations (Yirkala LPC 1991).

Together, these two groups formed an ‘Aboriginalisation plan’, with the ultimate goal of gaining complete ownership of the school for Yolŋu community members. School staff worked with community elders, writing down their ideas of how the school should run and what should be included in the school curriculum:

I knew that there was layers and layers of deep intellectual knowledge that we already had in the world that is connected to the land and our ancestors. We developed a Both Ways approach to education at our school that still exists today and I believe in the importance and relevance of embedding this way of learning, this way for our children. (Yalmay Yunupiŋu ¹⁹ presentation, December 2015)

By 1988 the School Council’s new constitution was officially accepted, marking the formal introduction of ‘two-way’ education. This new paradigm was designed to ensure that community language and epistemologies would take equal centre stage, producing:

Yolŋu students who are balanced in both worlds: strong in their Western knowledge and English and strong in their own identity, cultural knowledge and language. (Yirrkala LPC & Yolngu Action Group 2011)

Yirrkala’s Response to the ‘First Four Hours’ Policy

The 2008 decree that ‘the first four hours of education in all Northern Territory schools will be conducted in English’ (Scrymgour 2008) led to the closure of most extant bilingual programmes across the NT (see §2.1). While government funding was not pulled entirely, the Department of Education actively pushed English-only literacy tests, strategies, and programmes in the wake of this policy. This was met with great disappointment and frustration in the Yirrkala community:

[W]e, the old people, would be saddened by such an approach because our language comes from within the very essence of our being. It makes us who we are. (Interview with D. Marika, ²⁰ ‘Going Back to Lajamanu’ Four Corners, September 2009).

Language is the key to our education. It’s us, it’s a mirror of our soul and when you look into a mirror it’s you, so that’s what language is. (Banbapuy Whitehead interview, November 2015)

Nevertheless, in what had rapidly become a hostile policy context, the bilingual programme at Yirrkala continued to run. The government revoked all personnel and resource support for the school’s bilingual programme, and yet the school remained steadfast in their goal of educating Yolŋu children to be strong in their language and culture, through YM language as well as English. The importance of English instruction was not undervalued, but Yolŋu did not want general Australian culture to replace Yolŋu culture in an assimilationist manner, and for some community members, the First Four Hours policy was understood to be merely the latest instantiation of a larger project to westernise Indigenous people:

They want to try to westernise Yolŋu people. They want to leave us in a mainstream culture like a white man. That is a difficult part for us. We don’t want to live in that. We want to live in two worlds that we are comfortable and that’s one things our government are trying to close the gap to bringing us into a mainstream culture, into a mainstream world. And that is important that we as Yolŋu people need to be very strong in our own right identities. (Interview with D. Marika, ‘Going Back to Lajamanu’ Four Corners, September 2009)

School and staff members argued that the policy would hinder children’s scholastic success because of their unfamiliarity with the language. ‘[I]t’s important for children to be able to understand and compare because children, if we teach them in one language all the time, English, the children will be bored and children will never get attention to that. The language is very strange to them’ (ibid). In a letter to Scrymgour, one Yolŋu teacher wrote:

We have been told we are not to use our students’ first language, only English. Well, I already know that the children won’t understand what I’m saying, they will laugh at me, and they may even misbehave because they’ll be bored and won’t know what the lesson is about. So perhaps I will cheat and use some Yolŋu Matha – what will happen then? Will I have my mouth washed out with soap like in the mission times? Or will I have to stand on one leg outside the classroom? Or perhaps I will lose my job? (Y. Yunupiŋu 2010)

The First Four Hours policy was finally dismantled in 2012, but this has not brought much relief for many who work tirelessly to defend bilingual education:

One of the obstacles I’ve experienced in the continual politics that stands between bilingual programs, it brings me so much trauma and stress, and adds more strains and more pains. So often, our energy goes into defending the programs rather than improving them. I am an advocate for bilingual programs and I believe they are a good method to teach. They encounter Both Ways learning and shift the power balance and can empower Yolŋu teachers to contribute their knowledge. (Y. Yunupiŋu presentation, December 2015)

Indeed, the policy threat to bilingual education remains. The 2014 government-commissioned Wilson report advocates English-only instruction (leading in part to a ministerial decision to roll out direct instruction²¹ in remote schools) and recommends that all secondary students be sent to boarding schools in regional centres, further evidence that in attempting to ‘close the gap’, the government risks further undermining local community priorities and language maintenance.

Yirrkala Community School Today

Yirrkala School continues to run a two-way programme, incorporating both Yolŋu and Western language and knowledge systems throughout the curriculum. Community elders use gaṉma as a metaphor for the programme: a place where a current of water from the sea (non-Indigenous knowledge) meets a current of water from the land (Yolŋu knowledge). At this place, the two ‘currents engulf each other, flowing into a common lagoon and becoming one’ (Marika 2000, p. 47).

The school currently encompasses a ‘Families as First Teachers’ (FaFT) programme,²² preschool, primary school and secondary programme. It also offers extensive support to the nine homeland schools in the surrounding area. Around 100 students are enrolled across three classes in the primary school (Transition/Year 1, Years 2–4 and Years 4–6) and a further 90 in three secondary classes (Years 7/8, Years 8/9, and Years 10–12). This structure changes depending on the school’s needs, students’ attendance, the progression of students and staff changes (Yirrkala LPC 1991).

YCS aims to follow a bilingual step model where there is a strong emphasis on Dhuwaya instruction in the early years that decreases incrementally over time. Conversely, English instruction increases as students’ progress through their educational journey (see Fig. 4.1). Thus, while literacy is initially introduced in Dhuwaya, beyond Year 9, instruction is given largely in English. Implementation of the step model may be adversely affected by class groupings (with different year levels in one class) and team teacher attendance (sometimes low due to cultural events and community obligations).

Fig. 4.1

The bilingual education model at Yirrkala School (School Day: 5 hours 20 minutes; School Week: 26 hours 40 minutes; Preschool: 4 hours/day, 20 hours/week)

At present, the preschool is led by a qualified Indigenous teacher who is able to deliver both the English and Yolŋu programme. Often, preschoolers’ parents come to school to support their children. All primary classrooms have both a Yolŋu team teacher and a non-Indigenous classroom teacher who provides support in delivering the Dhuwaya programme. Team teachers in turn support the classroom teacher by facilitating communication when needed (Yirrkala LPC 2014). A Yolŋu team teacher delivers the primary school art programme. Secondary classrooms do not have Yolŋu team teachers, partly due to limited funding but also because of the transitional step-model nature of the programme. However, a Yolŋu secondary tutor splits her time between the Year 7/8 and Year 8/9 classrooms, teaching YM literacy (at least two hours a week) and maths in each classroom. The Year 10–12 class does not have an allocated Yolŋu teacher, but YM activities are included in their curriculum as much as possible. Year 10–12 students also participate in a three-day clan language workshop every school term (Fig. 4.1).

The two-way curriculum incorporates a number of innovative programmes developed by YCS. Galtha ²³ Rom lessons focus on vital cultural and developmental knowledge, and are delivered by elders in language in a more traditional setting (e.g. hunting, collecting paper bark) (Gale 1994; Marika-Munuŋgiritj 1990). The Garma ²⁴ Maths curriculum has been developed to incorporate both Yolŋu knowledge systems and Western concepts (Ŋurruwutthun 1991; Watson-Verran 1992; Marika 2000). The Yolŋu section of the programme encompasses two aspects: gurruṯu, the complex systems of kinship that connect individuals and clans to each other; and djälkiri (‘foot/footprint’), an individual’s connections to the lands and waters their ancestors passed down to their clan. In the Garma curriculum, gurruṯu is connected to maths (namely expressions of recursion), while djälkiri is connected to space/location.

All classroom resources required to run the two-way curriculum are produced by Yirrkala School’s Literature Production Centre (LPC). Available literacy resources include ordered readers, storybooks, story sequencing cards and vocabulary cards. School staff are in the process of creating iPad literacy training apps and iBooks in Dhuwaya. Classrooms are colourfully decorated and equipped with a wide range of Dhuwaya resources including alphabet wall cards, informative posters, and gurruṯu (kinship system) charts. Literacy worker staff positions are crucially filled by native Yolŋu speakers.

Yolŋu continue to take control within the school. The 2016 staff list includes 19 Yolŋu staff members, the same number as non-Indigenous staff members. Yolŋu fill all kinds of staff positions: teaching, linguistic, administrative, clerical, ancillary and janitorial. A Yolŋu principal-in-training works alongside a non-Indigenous principal. The teacher-linguist is a senior Indigenous woman who works closely with Yolŋu teaching staff on the curriculum delivery. A Yolŋu senior cultural advisor ensures correct cultural protocols are followed for any events occurring at the school and acts as a family representative in the school. The Action Group continues to meet weekly to discuss day-to-day matters of the school, and the School Council meets each term.

Yirrkala community members have worked hard to keep their language at school, and yet government budget cuts have time and time again resulted in the loss of vital staff members. The school lacks resources critical to the successful full implementation of a bilingual programme; this is in large part attributable to the continuing debate about the effectiveness of bilingual schooling (see §2), which is persistently deaf to academic research findings that demonstrate the efficacy and necessity of such programmes.

Conclusion

Yirrkala School has long existed in the crossfire of conflicting local, national and policy discourses. The range of top-down strategies intended to control and measure language use in schools has been fundamentally shaped by a ‘monolingual mindset’, deficit discourses and a broader assimilationist project in turn. As a result, state and federal government policy has routinely, and increasingly, undermined local priorities in cultural education and the maintenance of Indigenous languages more generally.

Yet while bilingual education policy has vacillated according to the vagaries of public and political ideology, local community priorities in Yirrkala have consistently privileged the importance of multilingualism and first-language(s) literacy and their rightful place in the classroom. The community has been vocal in the face of threats to bilingual education and has gone to great effort to express their dissent in a positive light, for example, through the ‘Don’t cut off our tongues’ campaign (1998–99) and the community event in April 2014 celebrating two-way education, designed in part to attract media coverage. Yolŋu have become increasingly ‘media savvy’ and are using these tools to effectively advance their local language policies and ideologies on a wider stage (Waller and McCallum 2014).

The two-way journey has been a constantly negotiated process that has had to be responsive to the needs of and changes in local language ecology. While it has not always been possible to achieve community-wide consensus on all decisions, the collaboration has been remarkable in its success in engaging in galtha to construct gaṉma together. In recognition of these achievements, on International Mother Tongue Day 2016, the prestigious International Linguapax Award²⁵ was given jointly to the Yambirrpa School Council and Yolŋu Action Group for their work in bilingual education in Yirrkala. The committee summarised their decision with these words:

These institutions carry on the struggle initiated more than 40 years ago by the community elders to convey the cultural and linguistic heritage of their people through bilingual teaching programmes in Yolŋu, in steady decline since 1980 due to government action.²⁶

Acknowledgements

The authors wish to thank three anonymous reviewers for their insightful and helpful comments. This research was supported by an Australian Research Council Discovery grant (DP0877762), the ARC Centre of Excellence for the Dynamics of Language (CE140100041), and an International Postgraduate Research Award to the first author.