Communities and linguists working in language revival face a common and constant challenge: how to balance the possibilities of linguistic analysis and the knowledge latent in archive sources with contemporary knowledge, usage and priorities for the language. The working solutions range (at various stages, with different groups of people, and from different starting knowledge bases) from ratifying the remaining orally transmitted knowledge exclusively, through to referring to linguistic advice on authentic grammar as an ongoing primary strategy.

In Victoria, while different language programs have taken divergent paths, the tendency is for communities to prioritise community knowledge and to reclaim their languages step-by-step as understanding develops. Community knowledge may start from the memories of Elders, words and meanings embedded in local varieties of Aboriginal English, and particular records of the language valued by individuals. As language awareness increases the knowledge broadly held in the community may expand to greetings and other set phrases, relatively fixed speeches, and sets of words such as the names of animals or elements of a traditional practice such as eel trapping. The emphasis is on rapid release of what is available into community use, community control of language products and processes, accessibility of the language to community members, maintaining cultural appropriacy of teaching content and approaches, and authoritative lines of transmission (see Eira & Stebbins 2008 for a detailed exploration of this last element). Linguists work closely with very few programs, but are more generally available as a support resource for training and consultancy on specific projects. Grammatical and phonological sketches are available for a number of languages, and wordlists or dictionaries produced by communities, linguists or both.

The emphases indicated above give rise to languages which have great value for Aboriginal people and high importance for identity and community strengthening. They are also languages-in-process, being expanded and revised at each new stage of development and each time information from an archival source or a linguist gains acceptance by language workers, Elders, and others. For example initial research by 374Taungurung community language workers resulted in a wordlist widely distributed among the community, generating some language use at community events and in public arenas. While some understanding of the phonology of the language was gained through community workshops, for the most part people simply pronounced the words as the written form suggested to them. The importance of this stage was access to the words as such, as a means to reconnect with language, identity and culture. It was a few years later that a new language worker, Lee Healy, began a painstaking reconstruction of the pronunciation of each word from a comparison of all available sources, proposing an orthographic system as part of the process. The greatest challenge here will be to bring the results to the community in a way that promotes their acceptance, without undermining the achievements of the first stages of language reclamation (L. Healy, pers. comm., 14 August 2009). Here is an example of a house already lived in. Any new stage of development must be carefully grafted in without loss to the pride and confidence in their reclaimed knowledge and connection to their heritage that people have already gained. Because the house is lived in it is used – the fundamental groundwork, the framework, the roof, the rooms are there, but it needs renovations and new rooms added as it is expanding and growing.

How do we do this and stay true to the look of the house?

First and foremost is to work with that fundamental groundwork which is the Elders past and present, and the community still living who know these words and speak these words, sometimes without even noticing that they are speaking the language.

Gunnai/Kŭrnai language reclamation2

The Gunnai language program is a case in point. It is a strongly community-based program often held up as a model for other Victorian programs due to its continuation over 20 years and its establishment throughout the education system in the region. The initial materials for teaching arose from an Elders’ workshop in 1991 focused on plants and their uses. From here a community wordlist (Dent 1997) was eventually developed which remains the basis of teaching in community and formal education to the present. Formal language teaching began at Gippsland Institute of Advanced Education (later Monash University, Gippsland campus) before the community language program was established in 1996 through the Gippsland and East Gippsland Aboriginal Cooperative. Teaching gradually expanded to schools including Woolum Bellum Koorie Open Door Education, and then preschools. In 2004 Gunnai was introduced into the Victorian Certificate of Education (Indigenous languages of Victoria, revival and reclamation: Victorian Certificate of Education study design), and

375into certificate programs at the Central Gippsland Institute of Technical and Further Education (TAFE) Koorie Unit in 2006. Informal language teaching is also developing, for instance in a Sunday school class. Gunnai/Kŭrnai speaker/learners range in age from three to 77 years old and live across the full extent of south-east Victoria.

The presence of Gunnai language is now evident in forms including public signboards (some quite extensive), speeches given by various Gunnai/Kŭrnai people, casual use by people who have been through the teaching programs, and an expanding range of language resources including illustrated books, a CD and accompanying learning guides. As teachers of the language are largely graduates of one or more of the above programs their teaching is quite homogenous and uses the same basic set of materials as a reference and resource kit.

The central principle underlying teaching and development of Gunnai is the value of oral transmission. This is the way Lynnette was taught. We need to use that oral information that we have; then we look at the oral documentation of our Elders, such as tapes. Next we look at written documentation approved by the Elders. Lastly, and only when we need to get further help or support, we use additional material documented by non-Aboriginal early recorders and linguists, but only when it has been talked about with Elders and community learners to see if it fits with our way of using language. The principle is that the written supports the oral language knowledge of plants, medicines and so forth, the stories passed down, and the speaking knowledge of sounds and the way those words were said before written documentation. This often causes problems and divisions in the community because the sound appears changed in the writing. Relying on the archive and academic sources can mean that the written takes away the oral.

This system of priorities establishes the lines of authority in language as firmly within the community, and maintains a traditional practice of learning from your own Elders according to their decisions about what is available to be learned, who by and how. There are obvious benefits here for identity and community strength and cohesion, as well as the maintenance of values such as respect, patience, deep rather than fast learning, and the role of Elders in directing and mentoring the community. While Elders freely acknowledge that their language has changed from various influences including English, many in this community have a view of living languages that can accommodate change. What is most crucial here is that the knowledge and views of Elders remain central to decisions and practice, and the community is in control of their language. The downplaying of archival records does mean however that, aside from those sources which some Elders appear to draw on as part of their own knowledge base, additional storehouses of Gunnai/Kŭrnai language records have remained largely unutilised to this point.

Gunnai/Kŭrnai pronouns: A case study

In this chapter we describe the process by which Lynnette (language worker and Gunnai teacher) and Christina (Victorian Aboriginal Corporation for Languages 376[VACL] community linguist) recently reviewed the current pronoun system by incorporating an analysis of archival sources into current knowledge. This differs radically from an approach whereby analysis of archival sources together with cross-linguistic comparison and reconstruction are considered primary. In this system a much higher value is placed on current community knowledge and practice, and oral transmission principles as described above, with archival sources brought in to support and expand the language to its next stage of development.

Pronouns in current use

Tables 1 and 2 set out the pronouns as listed in the current community wordlist (Dent 1997). The core set of words most commonly taught and used at present is highlighted in bold.

377All the core teaching set are glossed with meanings parallel to those of English. However the wider set of pronouns here reveals some indications of the possibilities latent in the list. A distinction among one, two and more is evident. There is an apparent possessive suffix {-lung}, sometimes {-l }. This is confirmed by the names of subgroups and/or dialects within Gunnai/Kŭrnai such as Braiakaulung or Tatungalung (approximately, of the men of the west and of the sea, respectively). There are apparent choices of word for a number of English meanings such as you, suggesting that more detailed meanings might be buried in this listing. In this regard the implied distinction between ngaju as I and ngio as me is significant.

These indications within what is already familiar open a door to expanding the system for greater reclamation of the breadth of meaning and complexity latent in the language.

Analysing the historical sources

Historical collections and recent analyses

Nineteenth century Gunnai speakers were recorded anonymously in over 30 documents, some of which are revisions and publications of earlier notes. The most important of these for our present purposes are those with texts, sentences and paradigms: the sections in Smyth (1878) contributed by Bulmer (pp. 24–39, 96–97), Hagenauer (pp. 97–98) and Howitt (pp. 48–49), an additional manuscript by Hagenauer (n.d.), and the extensive work of R.H. Mathews (1902, n.d.a & n.d.b). Some wordlists also include individual pronouns, such as Crouch (1863) and the survey response by Miss Henry, collated by J. Mathew (n.d.).

Little analysis of this material has been carried out, and still less published. A masters thesis by Fesl (1985) collates and discusses some of the grammatical information evident in historical sources. The chapter on pronouns draws almost exclusively on various work by R.H. Mathews (including 1902, n.d.a & n.d.b) and, while this certainly achieves some inroad into the complexities represented by the full range of documentation, it necessarily leaves a considerable amount of data, and hence questions and possibilities, unconsidered. Information in the cross-linguistic tables in Blake & Reid (1998) follows Fesl, though more cautiously. In an unpublished analysis of the textual sources for Gunnai/Kŭrnai, focused primarily on case and verbal morphology, Morey contributes some more complex consideration of both free and bound pronouns, but states wisely that ‘… a comprehensive discussion of Gippsland pronouns is beyond the scope of this paper’ (n.d., p. 55).6 378

Analysing the pronouns: the next stage

Due to the status of research on pronouns in Gunnai we decided to start from scratch. Christina compiled a list of all tokens glossed with pronominal meanings in the historical sources, parsing all sentences, phrases and texts in Toolbox. This resulted in a ridiculous 120 putative pronouns, counting possible bound, inflected or derived forms separately, but not counting obvious spelling variants. A few possible additional tokens may also be embedded in currently analysable phrases. Clearly it is beyond the scope of this chapter (to echo Morey) to discuss the analytical process involved in considering all of these candidates. Instead we restrict ourselves to exploring the issues posed by first person singular (1sg) forms to give a general picture of what was involved, then summarise the least problematic choices for the whole paradigm as one of the bases for our proposal for a new contemporary paradigm. Forms recorded only as clitics or bound forms are not included in this paper.

The following forms are recorded for 1sg:

R.H. Mathews also glosses ngal as 1sg, but this is surely a misunderstanding of first person dual (1du) ngalo or nalloo. Both Mathews and Bulmer also record ngal as 1du, Mathews listing it as inclusive.7 It is easy to understand a 19th-century speaker of English struggling with a lexicalised concept of I and you.

379We have assumed ngaiu, ngio and ngaju to be alternative spellings for the same form. Ngi can also be added to this set, interpreting i as /aI/ or /ajI/ and assuming the last sound was quiet or dropped in the speech context. Ngioma appears to display the Gunnai clitic {-ma}. The functional range of {-ma} includes, but is not limited to, possessive marking. It would not be expected to cliticise to a pronoun. Without a context it is hard to understand Bulmer’s analysis of ngioma as causative. He also includes nindoma as second person singular (2sg) causative (by thee). Causative is a common 19th century description of the ergative function, but ngioma and nindoma do not match easily with other tokens for the ergative singular forms.8 Regardless of the final analysis of {-ma} it seems clear that it can be treated as a suffix or clitic, rather than an integral part of the pronouns themselves. This brings ngio(ma) also into line with the ngaiu set. Ngan and watha, listed by Bulmer as having object function, also remain a little mysterious at this point suggesting that further cross-linguistic comparison may be needed. Watha may related to wert, a form collected as first person plural (1pl). Ngat, nat and ngaty (Fesl /ŋad̪/) match well with the apparent clitic ngadha (Bulmer; Hagenauer and Mathews), recorded in both subject and agent contexts. (A possible reduced form {‑ndha} also appears in Mathews.)

Historical sources suggest an ergative/absolutive or possibly nominative/accusative distinction in at least 1sg and 2sg with (inconsistently) different forms listed for S(ubject), A(gent) and O(bject) by R.H. Mathews and especially Bulmer. (From this point we will use terms employed in teaching; active for ergative, and non-active for absolutive). Cross-linguistically, forms cognate to the Gunnai candidates ngaiu and ngaty suggest the former as non-active and the latter as active, which is at least compatible with the Gunnai evidence.

Following the kind of investigation indicated above for all tokens in the sources, we made a heavily reduced summary of the most useful and likely pronoun forms recorded, to discuss in relation to the contemporary list (Tables 4 and 5): 380

While even this reduced paradigm is clearly not without discrepancies, it highlights for present-day speakers and learners some of the extended meanings possible in Gunnai. The distinction among singular, dual and plural is partly clarified. An inclusive/exclusive distinction is clearly evident and there are indications of a partial active/non-active distinction. Importantly one or more candidates are now available for every expected slot in the paradigm.

For accessibility to the contemporary community it is at least as important that there is a significant degree of overlap between this list and Dent (1997). Some words are identical in probable pronunciation, if not also spelling; others are similar such as 1du nangoo (current list nanangoo). The suffix {-lung} apparent in Dent (1997) is attested here for more of the pronouns, offering regular alternatives for all members of the paradigm.

The next step forward

With a relatively clear picture both of contemporary usage and the contribution of historical sources, it was now possible to develop an expanded set of pronouns 381to propose to the Elders who formed the reference group. We designed a set of working principles to support the knowledge and confidence already built up in the community while also providing people with a new level of access to the richness of their language:

1. Keep what’s familiar
2. Reclaim the full range of Gunnai meanings for pronouns
3. When there’s more than one word in the current wordlist, choose the one that matches the historical sources
4. Select just one spelling for each word (or morpheme) each time it appears
5. Fill in ‘missing’ pronouns by using the patterns we can see, then from the historical sources.

Tables 6 and 7 list the newly expanded pronoun paradigm developed on this basis.

First it is important to note the degree to which we have been able to affirm pronouns already in use (Principle 1). Ngaju, ngaiu (respelt from ngio), ngallu (subsuming nalloo and nalu), werna, njinde, noonga (subsuming noong), thana, ngetal, ngingal, noongal, limbauloong, waruloong and wernaloong (spelling adjusted to match werna), are all 382present and accounted for. Slight adjustments adapt nanangoo to nangoo as the simplest option in a set of apparent variants, and limbaook to limbau as a solution to inconsistencies in the historical evidence. Ngurtana is added from archival sources but was found in Dent (1997) as every (every one of you). We have also made a couple of adjustments for consistency: we can safely assume that booloonga (their plural) should be specifically dual as it relates directly to boolaman (two),9 a common strategy in other languages, while thanal seems misplaced as theirs dual as this morpheme is associated everywhere else in the wordlist with plural.

We have consistently applied the possessive {–loong}, evident in Dent (1997) and further attested in the archival sources, across all dual and plural pronouns (Principle 5 extended). It is also offered as an optional alternative for the singular paradigm, retaining the more familiar ngetal, ngingal, noongal as primary. ‘Missing’ pronouns can be backformed by removing this suffix, yielding boola and waru (supported also by the archival sources), as well as supporting our choice of limbau (Principle 5).

Spellings have been regularised such as {-loong} and noongal (Principle 4). Note that this is done at lexical or morphological level, not at phonological level as in a standard orthography.

We have been able to account for the apparent choices in the current wordlist for both I/me and you. Ngaju and ngio match well with both historical sources for Gunnai and active/non-active pronouns in languages across the continent, as discussed.10^. Looking at the options for you, we have a slightly more complex problem. None of the sources shed much light on the multiplicity of words given or their possible shades of meaning. Njinde is already very established in community use, being part of the standard greeting Wunman njinde?, and so has to be retained. Comparison with other Aboriginal languages indicates that a nginda-like word is more likely as the active pronoun, while a nginna-like word is more likely as the non-active. All things considered, we have opted for njinde as the active pronoun and ngingu as the non-active. In its contemporary pronunciation /nInʤI/, the former approximates the expected form of an active 2sg – although in the 19th-century spelling nj is probably intended to represent, not an n followed by English j, but a palatal nasal (as in Spanish señor). Ngingu is a compromise between Dent (1997) and a historical/comparative representation of the word. In practice, since njinde is so thoroughly established as the general word for you (extending also to dual and plural by analogy with English), this will probably be the slowest pronoun to shift to the proposed meaning. Current usage may, in the end, override the revision in this case.

383Similar procedures were applied to identify the inclusive and exclusive pronouns. All four words for we (and their possessive counterparts) are apparent in both community wordlist and historical sources with minor adaptations as above. Historical sources are fairly clear about which is which for the plural pronouns. While they are less clear about the duals, discussions by R. H. Mathews result in slightly better evidence in favour of ngallu as inclusive and nango as exclusive. Data analysis for language revival has a particular purpose: it has to result in a workable decision that people can use now. Where the available evidence leaves issues in doubt, in many cases a best guess or even simply a choice has to be made. In this sense, as well as in the sense of community processes, language revival is necessarily an ongoing phenomenon.

Where additional pronouns in Dent (1997) are unexplained by this process of expanding the range of meanings, we have not included them in the basic pronoun paradigm (Principle 3). This does not entail their removal from the wordlist, as they are easily explained as alternative means of referring to someone. That one, or similar, is commonly used even in Aboriginal English to refer to a third person, which makes sense of the many words given for he and they. Some of these demonstratives, such as gindi, also match well with forms found in other languages. Lynnette suggests that other possible explanations may account for other forms, such as nindethana, misunderstood as a core pronoun, which may be simply you and they.

This completes the basic singular, dual and plural paradigms for all three persons. As some readers may have noticed our proposal is not completely finished. For example it is not clear how to spell thana when it appears as part of another pronoun, {-tana}. Noonga may be better represented as ngoonga, as this form does appear in the historical sources, and it is well known that ng at the start of a word was commonly overlooked by collectors. We have not even touched on pronouns as suffixes on the verb though these abound in the historical sources. These and other questions will serve to raise discussion topics for the next generation of language students and a way for them to participate actively in the development of their own language.

Aboriginal people have been told over nearly two centuries that they and their languages are ‘primitive’. While people today may know at some level that their language is as rich and complex as any other, this fact is usually talked about with considerable emotion indicating that the wound is far from healed, and the indictment of both language and people is still in need of strong resistance. The tangible evidence of this richness in the form of complex distinctions between exclusive and inclusive, singular/dual/plural and so on, is an important contribution to finally overturning the power of this label.

The Elluminate session

As a linguist and language worker in partnership we have had many discussions about the words of the language – what’s available, what’s missing, what’s clear from the historical sources and what’s tentative – but these discussions and the language development process which results is generally unavailable to others. For this reason 384we decided to record an interactive session in Elluminate Live!11 targeted to advanced students and language workers. Elluminate provides a virtual classroom environment with interactive whiteboards, breakout rooms and so on. Lynnette has already been using Elluminate for distance education through the Central Gippsland Institute of TAFE, so it seemed a logical next step to record a staged conversation between ourselves complete with PowerPoint slides and tables of the proposed revisions at various stages. The one-and-a-half-hour session documents our discussions about historical sources, how to match them with the knowledge already in circulation, and the issues which are raised by the process. Viewers can observe the process and apply what they understand to their own language, in terms of possible gaps and how to fill them, in ways which are readily traceable to the language as recorded by 19th century speakers, while at the same time supporting contemporary knowledge and practices. The session can be pulled apart to form digestible pieces for students and intertwined with additional training material as relevant. For example, at a VACL language workers’ workshop in 2009, we used a framework of about half the session to raise issues of ongoing language development, interpretation of historical sources, and identification of morphological patterns.

Conclusion

Our case study illustrates how it might be possible to continue developing the house already lived in with minimal cost to the ‘residents’. The fact that the language is known and in use in the community means that current community knowledge and usage has to be privileged if any further development is to be successful. In the Gunnai context an important principle for this is to value the oral above the written. Written or archival sources are viewed as supporting knowledge that has been transmitted orally, and for seeking words and meanings missing from current knowledge. All language decisions are referred to Elders. The principles we followed in developing the pronouns aim to ensure that: (a) the confidence of learner-speakers in their current knowledge can be maintained, (b) community authority in their language is maintained, and (c) the contemporary language is validated as a 21st-century living language, regardless of the completeness of that language and the theoretical challenges this presents for notions such as authenticity and language change.

For the linguist in this partnership the point of the collaborative process is that it allows me to gain a better understanding of what the community knows they need to do. It helps me to work with what’s happening, thereby smoothing out potential blocks to collaborative productivity as we go. When this is working well the collaboration also gives the community good access to the kinds of interpretations that linguistics can bring to historical sources, returning more of the ancientness of the language’s structures and meanings to the language of the present.

In addition the way of working trialled here has potential to take the principle of collaborative research to a deeper level. We are not simply proposing a partnership

385model, which has been suggested and implemented many times before, but a merging of the principles by which we determine what is correct. For Christina as linguist, correctness can be determined by careful analysis of data. For Lynnette as language worker, correctness can be determined by listening to those with the authority to know. We see no benefit in pitting these principles against each other. Instead, in the interests of accessibility and acceptability of the research to its end users, we simply prioritise as data what is already validated and in use in the contemporary community. In all probability we will need to reconcile ourselves to a separation between the methods, goals, and validation systems for reconstructing a historical language, and those targeted to a functional analysis for a contemporary emerging language.

To include contemporary usage and knowledge in assessing what is correct challenges the assumption often held by both linguists and communities, that the only correct or authentic form of the language is what was spoken at the time of colonisation. For linguists this represents a theoretical shift in our understandings of language loss and change. For communities it represents a process of recognising and then coming to terms with that loss and change – what Jeanie Bell (2009) has called ‘the grieving phase of research’. This research also underlines the need to accept the staged nature of language revival – again, an issue faced continually by both communities and linguists. It is clearly neither feasible nor desirable to wait until language analysis and language decisions are final before using what is accessible. Thus both analysis and language planning decisions are necessarily a work in progress. The solution we propose is to embrace what is known and accepted now, and use it loudly and proudly, while also understanding that if change is an intrinsic part of living languages, it is even more a part of living reclaimed languages.
 

References