30 Yan-nhanu language documentation and revitalisation

The purpose of this paper is to discuss revitalisation prospects for the Yan-nhaŋu language of Eastern Arnhem Land, northern Australia. We review previous work on the language and outline some issues to consider for language revitalisation. We tackle the difficult question of evaluating ‘success’ in revitalisation. We argue that language revitalisation projects should not be judged successful or otherwise purely on the basis of linguistic outcomes; as such programs may produce valuable outcomes in the socio-cultural context of language use even if they do not increase the number of speakers of the language. Linguistic, social and geographical background Yan-nhaŋu is a Yolŋu (Pama-Nyungan) language of the Crocodile Islands of NorthEastern Arnhem Land. It is a member of the Nhaŋu dialect cluster spoken from the Crocodile Islands in the west to the Wessel Islands in the east.3 The language name literally means this language; yän (tongue or language), nhaŋu (this). This naming convention is common to most Yolŋu language varieties.4 1 Linguistics, Yale University. 2 Anthropology, Australian National University. 3 Information on the classification of Yolŋu (Yolngu, Yuulngu) languages can be found in Bowern (2005), Schebeck (2001) and the references therein. 4 This paper contains the names of people who have passed away. These names should not be spoken aloud in the presence of family members. In this paper we quote all Yan-nhaŋu and Yolŋu words in the widely used Yolŋu Matha orthography (used, for example, in Zorc, 1986). Underlining indicates retroflection, ŋ has the same value as its IPA value, ä is IPA /a:/. Nh and 362 Re-awakening languages Many Yan-nhaŋu people now live at the ex-mission settlements on their homelands at Milingimbi and Galiwin’ku, although some also live at Maningrida (the next community to the west) and surrounding outstations. The founding of the Milingimbi Mission in 1922 brought extensive changes to the Yan-nhaŋu traditional lifestyle, not least because it involved the permanent settlement of a large number of people on Yan-nhaŋu clan lands from other Yolŋu groups. Much of the day-to-day business of Milingimbi community is run by groups other than the Yan-nhaŋu. Nonetheless Yannhaŋu proper names are still used by the Yolŋu (Aboriginal people) of Milingimbi to refer to sites on the Islands and in the sea. Yolŋu living at Milingimbi acknowledge the sacred links among Yan-nhaŋu and the seas of the Crocodile Islands, although the Yan-nhaŋu are one of the least politically powerful groups in the area. Migrations of larger clans from the east and a legacy of marginalisation from the day to day running of the missions provide a background to the diminution of Yan-nhaŋu language use. There is also a Yan-nhaŋu outstation settlement on the largest of the outer Crocodile Islands of Murruŋga, some 50 kilometres from the northern Australian coast. During the period of intense inter-clan fighting immediately following the Mission settlement many Yan-nhaŋu withdrew to this island. More recently the North-Eastern Arnhem Land homelands movement of the 1970s made it possible for Yan-nhaŋu people to return more permanently to their customary outer island home, as well as to travel more easily among Murruŋga outstations and the larger settlements on their other island homelands. Murruŋga Island is these days a focal point of Yan-nhaŋu identity and a large part of language work has involved recording subject matter related to this place (Yan-nhaŋu Language Team, forthcoming; James, forthcoming; Bagshaw 1998). Historically Yan-nhaŋu speakers have had extensive ceremonial, cultural and economic links with other Yolŋu groups as well as with speakers of genetically unrelated languages further west. They are active participants in the extensive social networks that crisscross the whole of the Arnhem Land region. For example, Yannhaŋu women marry into other language groups including Dhuwal and Dhuwala speaking groups in the east, Djinaŋ and Djinba language groups to the south, and Burarra to the west (Keen 1978, pp. 130, 138; Bagshaw 1998 pp. 156–77). The linguistic situation at Milingimbi is complex and many people are bior multilingual. Yan-nhaŋu people now generally speak Dhuwal (also known as Djambarrpuyŋu) in day-to-day interaction. Prestige languages in the area include local variants of Dhuwala (Gupapuyŋu) and Dhuwal at Milingimbi, Ganalbiŋu (Djinba) at the nearby community of Ramingining, and those residing at Maningrida regularly speak Burarra (Gun-nartpa) and English; all of these languages are exerting pressure on Yan-nhaŋu. Many Yan-nhaŋu people speak some English and most also know something of other more distant languages in the region, including Rembarrŋa and Gunwinygu. dh are lamino-dental consonants; ny, dj and tj are palatal consonants. This paper is based on Bowern and James (2006) but revised, expanded and updated. Language documentation 363 Yan-nhaŋu itself is not a homogeneous language (Bowern 2008). There are six patrilectal or clan varieties; three are Dhuwa, three Yirritja.5 Not all the varieties are still spoken and most of the speakers involved in language work come from the Mälarra and Gamalaŋga clans. In addition to the small number of fluent speakers between the ages of 40 and 80 there are approximately 150 heritage owners with patrilineal ancestral connections to Yan-nhaŋu language, land, sea and madayin (sacred paraphernalia), and a further 120 Yirritja Burrara/Yan-nhaŋu (Gamal, Gidjingali, and Anbarra) people with language ownership rights. Table 1 provides information on the Yan-nhaŋu groups, their moiety, and the number of people belonging to each (see also Bagshaw 1998, p. 157). Name Patri-moiety Linguistic affiliation(s) Full speakers Partial speakers Total Walamaŋu Gamal Yirrchinga Burarra/Yan-nhaŋu 27 89 116 Ŋurruwulu Yirritja Yan-nhaŋu 2 4 6 Bindararr Yirrchinga Burarra/Yan-nhaŋu 5 10 15 Gorryindi Dhuwa Yan-nhaŋu 8 30 38 Mälarra Dhuwa Yan-nhaŋu 10 36 46 Gamalaŋga Dhuwa Yan-nhaŋu 9 35 44 Table 1: Yan-nhaŋu language groups. The complex relationships among groups are mapped through the idiom of kinship. Marriage in this area is exogamous so husband and wife will always be from different moieties and different clans. The Yan-nhaŋu groups signify their identities as separate from more distant groups primarily through reference to language rather than any distinct cultural practices. This linguistic identification includes groups speaking languages other than Yan-nhaŋu, so that purely linguistic classifications are not without ambiguities. The Gamal and Bindararr are referred to with the Ŋurruwulu as the Walamaŋu bäpurru (patrigroups) consistent with the logic of their ritual linkages (James, forthcoming, p. 92). Gamal people identify as Yan-nhaŋu but speak Burarra as ‘their’ language. This is relevant in a revitalisation program when part of the target group for language revitalisation expresses intellectual property of the language and 5 The Dhuwa or Yirritja moiety categories fundamentally divide and classify every aspect of the Yolŋu universe. Everything is either one or the other, so that every person or animal is Dhuwa or Yirritja and belongs to a Dhuwa or a Yirritja clan. 364 Re-awakening languages wish to have a say in the revitalisation and description process, but have no intention of shifting towards speaking the language themselves.6 Dhuwa Yan-nhaŋu patri-groups may also call themselves Märinga based on ritual associations. The three clans refer to each other as yapa-manydji (sister-dyad).7 That is, kinship terms are used to denote the relationship among the clan groups. In certain contexts they may also refer to each other as märi-manydji (grandchild/grandparentdyad). We include this information about the way that the patri-groups talk about their relationships to one another because it shows the cohesiveness of the Yan-nhaŋu speech community, despite evident patri-linguistic differences. The same type of cohesion exists among the Yirritja Yan-nhaŋu groups, which are known collectively as Malkurra. Myths and stories, shared country and secular ceremonial and marital links further strengthen alliances among these groups who also refer to each other as sister or company in Aboriginal English. Despite the small number of Yan-nhaŋu speakers in each patri-group there is a great degree of cooperation among the Yan-nhaŋu-speaking patri-groups and the different varieties can be treated as a single language for the purposes of linguistic description. We leave aside for the moment the problems involved in deciding how much of the variation among speakers should be attributed to idiolects and how much to differences in clan language, although we note the considerable technical problems in providing a coherent description of a language where each variety is spoken by perhaps only a few family members. Previous research on Yan-nhaŋu Almost all of what has been recorded for Yan-nhaŋu before the last few years comes from incidental notes in ethnographic descriptions. Between 1926 and 1929 Lloyd Warner carried out fieldwork at Milingimbi Mission. In 1937 he published his ethnography, A Black Civilization. His account of Yolŋu life is primarily concerned with Yolŋu groups that in-migrated to Milingimbi Mission from the east. He produced extensive discussion of local social organisation, material culture, technology and warfare. Despite living on the Yan-nhaŋu island of Milingimbi, his focus on the whole Murngin (Yolŋu) culture bloc largely obscures the differences between Yan-nhaŋu and the more numerous speakers of Central Yolŋu varieties such as Dhuwal and Dhuwala. Later ethnographers – among them Thomson (1939, 1949), Berndt (1951), and Keen (1978, 1994) – also describe the characteristics of the larger terrestrial group which they call the Yolŋu, touching only briefly on the Yan-nhaŋu and again glossing over the linguistic peculiarities of the most western of the North East Arnhem La