Transcription of Australian Aboriginal languages
Prior to the arrival of Europeans, Australian Aboriginal languages had been purely spoken languages, and had no writing system. The Latin script of the colonizers was inevitably used for the transcription of Australian Aboriginal languages, but the details of how the sounds were represented has varied over time and from writer to writer, sometimes resulting in a great many variant spellings of the same word or name.
At first, Australian languages were written following English spelling conventions, as it sounded to the writer. This meant that sounds which were distinguished in Australian languages but not in English were written identically, while at the same time sounds which were allophones in Australian languages but distinct in English were written differently.
Most Aboriginal words used in English follow these early conventions, and therefore do not usually give a good idea of how the word was pronounced in the original language.
|Guugu Yimidhirr||“tongue”||unjar (1770)
|nganhdhaar (1979)||[ŋan̪d̪aːɻ]||Early spellings may miss the word-initial [ŋ], and fail to properly distinguish dental consonants.|
|Gamilaraay||“honey”||wuddul (1903)||warrul (1993)||[waɾul]||Early spellings may fail to distinguish between [ɾ] and [d], which are allophones in English but distinct in Australian languages.|
Writers with more linguistic knowledge sometimes employed symbols such as ŋ or ġ for /ŋ/, ñ for /ɲ/, macrons (¯) or circumflexes (^) for long vowels, breves (˘) for short vowels, but these were often applied inconsistently.
Modern practical orthography
Linguists working with Australian languages today purposely use unambiguous phonemic orthographies based on detailed phonological analysis of the language in question. In orthographies of this kind each spoken word can only be written one way, and each written word can only be read one way.
Usually, but not always, practical orthographies use just the letters of the basic Roman alphabet. This necessitates the use of digraphs for sounds that do not have a standard character. In some cases this can lead to ambiguities, for example where the single sound /ŋ/ and the consonant cluster /nɡ/ could both be written as ng. These are commonly distinguished by writing the cluster n.g (inserting a full stop), n’g (inserting an apostrophe), or nk.
Vowels and semivowels
Most Australian languages distinguish just three vowels, which are written i, a and u. Even though they may sound like e or o at times, they are not written e or o, e.g. the Martuthunira word wirrirri "flame" is roughly pronounced werrerree [weɾeɾɪ]. Long vowels are represented by double letters, i.e. ii /iː/, aa /aː/, uu /uː/.
The semivowels w and y are usually pronounced as in English. In some languages, w may not be pronounced next to u, and y next to i, but for various reasons a linguist may still choose to write them, so that e.g. Gamilaraay yinarr "woman" is actually pronounced inarr [inar].
A handful of languages have a dental semivowel, which is written yh (see Place of articulation below).
In languages that have only one of the two r's, it's simply written r.
Place of articulation
The bilabial, velar and alveolar consonants are usually written the same as in English, i.e. p /p/, b /b/, m /m/, k /k/, g /ɡ/, ng /ŋ/, t /t/, d /d/, n /n/, l /l/. Ng may also be written using the non-English letter ŋ, called eng. Note that ng sounds like the ng in singer, not as in finger; the latter would be written ngg.
Palatal consonants are often represented by a digraph made of an alveolar consonant + j or y, i.e. tj /c ɟ ɲ ʎ/ can be written tj/ty, dj/dy, nj/ny, and lj/ly. C and j are other possible ways of writing the palatal stops.
Dental consonants are represented by a digraph made of an alveolar consonant + h, i.e. th /t̪/, dh /d̪/, nh /n̪/, lh /l̪/. Note that th is not a fricative as in Australian English, but a stop as in Irish English.
Retroflex consonants are usually represented by a digraph made of r + an alveolar consonant, i.e. rt /ʈ/, rd /ɖ/, rn /ɳ/, rl /ɭ/, as in Swedish. Sometimes, as in Pitjantjatjara, the alveolar is underlined instead.
Voicing of stops
Most Australian languages do not distinguished between voiced and voiceless stops, so that e.g. t and d both occur as variants of the same sound. Both the voiced and voiceless allophone will usually be written the same way, but whether to use the voiceless symbol or the voiced symbol varies depending on which occurs more frequently in the language. Some languages have been written using the voiced symbols by one linguist and the voiceless symbols by another. Moreover, some linguists choose to use voiceless symbols for some consonants in a language and voiced symbols for others.
Some languages do distinguish between voiced and voiceless stops, however.
Some languages have prenasalised consonants, a stop preceded by a nasal sound which is considered one consonant. In Yanyuwa these are written mb /ᵐb/, ngk /ᵑɡ/, nj /ᶮɟ/, nth /ⁿd̪/, nd /ⁿd/, rnd /ᶯɖ/.
- Cook, James (1955). The Journals of Captain James Cook. Cambridge: Cambridge University Press.
- Roth, Walter E. (1901). The structure of the Koko-Yimidir language. Brisbane: Government Printer.
- Haviland, John (1979). "Guugu Yimidhirr Sketch Grammar". In R. M. W. Dixon and B. Blake. Handbook of Australian Languages Vol I. pp. 26–180.
- Mathews, R. H. (Jul–Dec 1903). "Languages of the Kamilaroi and Other Aboriginal Tribes of New South Wales". The Journal of the Anthropological Institute of Great Britain and Ireland (The Journal of the Anthropological Institute of Great Britain and Ireland, Vol. 33) 33: 259–283. doi:10.2307/2842812. JSTOR 2842812.
- Austin, Peter (1993). A Reference Dictionary of Gamilaraay, northern New South Wales. La Trobe University.
- Bradley, John; Kirton, Jean (1992). Yanyuwa Wuka: language from Yanyuwa country. unpublished.
- Dixon, R. M. W. (2002). Australian Languages: Their Nature and Development. Cambridge: Cambridge University Press.
- McGregor, William (2004). The Languages of the Kimberley, Western Australia. London, New York: Taylor & Francis. pp. 21–26.