Jessica Boynton

Introduction

The first settlers of Australia made there way to the island continent an estimated 30,000 years ago. It is still a mystery how they did it – seafaring ways sophisticated enough to grant safe passage from neighboring continents were long from being developed in other parts of the world. And it seems that, once the Australians got there, they stayed there. (Bryson, 2000)

This mystery, however, is earmarked for historians and archeologists more than for linguists. It is of interest to the examiners of language for one reason – it means that, whenever language did develop for these original inhabitants of the sun burnt country, it developed in isolation. Whether the very beginnings of the languages predated the excursion to the island continent, or the language didn't even begin development until settlement there, the languages evolved without outside influence. For millennia. Even when the postulated proto-Australian split off into dialects, then languages, then dialects of languages, certain peculiarities held fast and are still common in the Australian languages. (Dixon, 1980)

There is a remarkable degree of lexical similarity between Australia’s languages. This is due to the obvious geographical factors; the languages developed together for thousands of years and did not have much opportunity to borrow from non-Australian languages. (Dixon, 1980)

Lexical similarities are also due to the tabooing of words that sound like the name of one recently deceased. These temporarily forbidden words must be replaced somehow, so either a related word in the same language extends its meaning, or the language borrows a term from a neighboring tongue. This situation also tends to result in a multitude of synonymous terms in Australian languages. (Dixon, 1980)

While the body of this paper will center on these common peculiarities, in order to simplify presentation, examples will be drawn specifically from the languages of Queensland. Located at the North Eastern region of the Island, Queensland is the home of at least two dozen languages including Dyirbal, Guugu Yimidhirr, Wargamay and Yidiny. (Dixon, 1981)

Each of these languages is critically endangered. Many have only a handful of speakers remaining, only a handful have many (over, say, one thousand) speakers. (ethnologue) Recent revitalization efforts notwithstanding, the native tongues are doubtful to recover the effects of the oppression their speakers have endured.

Phonologically speaking, most Australian languages follow a CV(C)CV(C) pattern. Any languages that diverge from this norm do so only in minor ways. Rarely a monosyllabic word will be permitted, but it must more than one mora. Also, due to historic vowel-deletion, some onset clusters appear in some languages. (Dixon, 1980)

The phoneme inventory of Australian languages is diverse, but most languages have only three vowels - /i/ /u/ and /a/ - which are rarely reduced to schwa, and generally have no fricatives or silibants. They have an extraordinary number of stops, with a homorganic nasal complementing each plosive. (Dixon, 1980) The phonolgy and phonotactics of Australian languages will be more closely studied in section two of the paper.

Studying the nature of phonotactic and phonological differences between the languages of Australia helps to reconstruct proto-Australian; although, if scholars are right in supposing that the language predates proto-Indo-European (Dixon, 1980), any sketch of the protolanguage would be met with skepticism at best. Generally, any grammars imputed for a language more than a thousand years gone are too theory-dependent to be anywhere near accurate. (Grondona, personal communication)

Some of the notable aspects of Australian languages surface in their morpho-syntax. Australian languages have absolutive/ergative case markings – that is, the subject of an intransitive verb conditions like the object of a transitive verb and the subject of a transitive verb conditions differently – as opposed to the more common nominative/accusative markings where the subjects of transitive and intransitive verbs condition the same and the object of transitive verbs conditions differently. (Dixon, 1980)

There are actually twelve distinct cases used throughout Australian languages, although no single language uses all of them (most use five or six). Morphological markings indicate possession, roles in interaction, physical position, causation and aversion. (Dixon, 1980) This phenomenon will be more closely examined in section three of the paper.

Australian languages also often use avoidance dialects – a version of a dialect that shares syntax, morphology and function words with the mainstream dialect but has a totally different set of lexical words. (Dixon, 1989) Not only is this interesting on a pragmatic level – the dialects are often called ‘mother-in-law tongues’ because they are specified for use with one’s mother-in-law and in unfamiliar social settings – but provides interesting data from a Whorfian perspective.

Typically, one word in the avoidance dialect correlates to five or six in the everyday speech style. (Dixon, 1989) This illustrates a perceived similarity between the concepts embodied in these words. Put simply, somehow those five words are deemed similar enough by the culture to be lumped together into one word in the alternative dialect. This will be examined in greater detail in section four of the paper.

The rest of this paper is divided into three additional sections. Section two gives a more detailed account of Australian phonology and phonotactics. Section three surveys the case systems in Australian languages and will both explain their function and outline the various ways these markings are combined. Section four briefly studies speech styles that illustrate some social aspects of the languages as well as hint at semantic categories in the languages.

Aussie Phonology and Phonotactics

Every language incorporates a set of rules about the distribution of sounds within an utterance. General rules regulating this distribution are called phonological rules; rules specifically addressing word and syllable structure fall under the domain of phonotactics. These rules are useful in reconstructing protolanguages because sound correspondences, paired with a reasonable theory of a history of phonological changes, provide reliable evidence for formations in the protolanguage. These rules also account for accents; speakers of a language who apply different phonological rules to the same underlying form sound different.

This paper will begin by outlining general phonological and phonotactic traits of the Australian languages, then illustrate these phenomena with an analysis of a single language (Warrgamay).

2.1 Aussie Phonology

The phonemic sound inventory of Australian languages can range from a system of 13 phonemes to a system of 25. However, there are some radical exceptions to this generalization due to recent phonological change.

The vowel system in Australian languages is typically very simple. Most have only three vowels /i/, /u/ and /a/, although some have lax /e/ and lax /o/ as well. Also, some languages have contrastive vowel length in the first syllable of the word.

Consonants in Australian languages are conventionally categorized as either apicals (alveolar or postalveolar), laminals (interdental or palatal) or peripherals (dorsal or labial). The languages have no fricatives, but they more than make up for that lack with the remainder of the consonant system.

Australian languages lack voice distinction in plosives, and always each plosive is accompanied by its homorganic nasal in the inventory. There are four different combinations of place of articulation (represented as plosive/nasal pairs):

As shown in Table 1, some languages, such as Pitta-Pitta, have two apicals, two laminals and two peripherals, totaling in six places of articulation (the most possible for Australian languages) while some, such as Yidiny, allow only one apical, one laminal and two peripherals. All languages have at least one plosive/nasal pair (since, as stated before, the languages lack voice distinction in plosives, this is a pairing of a plosive with undefined voicing and a voiced nasal) for each category of place of articulation, and all have two peripherals. Most languages of Queensland lack an apical contrast.

Australian languages can have as many as four laterals – one for each non peripheral stop. Most languages of Queensland, however, only allow /l/, regardless of the stop inventory.

Australian languages have two phonemic rhotics - a voiced trill and a voiced continuant - and two semi-vowels – peripheral /w/ and laminal /y/.

Australian languages have a two-syllable word minimum, with the root structure C1VC2C3V(C4) or C1VC5V(C4). The restrictions placed on each consonant position are outlined in Table 2.

Notice that the restrictions guard against geminates and place the most restrictions on root boundaries and consonant clusters. Also note that, due to recent phonological change, C1 can sometimes be left blank, creating an onset-less word-initial syllable.

Stress is predictable in the Australian languages, with primary stress on the first syllable, secondary on alternating syllables thereafter (except on the last syllable, where stress is prohibited). Stress assignment can, however, be affected by morphological boundaries.

2.2 Case Study: Warrgamay

Warrgamay, spoken on the coast of Queensland, exemplifies many traits typical to Australian languages, especially to the languages of Queensland.

Warrgamay has three phonemic vowels /u/, /i/, /a/, like most Australian languages, with a phonemic length distinction in the first syllable, as shown in examples 1 - 6.

Warrgamay follows the simplest stop distribution (in regard to place of articulation) available to Australian languages – it has no apical or laminal distinction, and has two peripherals (labial and dorsal).

Unlike most Australian languages, but like many of those particular to Queensland, Warrgamay has only one lateral /l/. It has the same semi-vowels /w/ and /y/ as most Australian languages, and retains both distinctive rhotics.

Warrgamay allows word roots with the structure CV:CV(C) or CV(:)(C).

The structure guards against consonant clusters, including geminates, by essentially outlawing codas except in word-final position. Also, while word-final codas are possible, they may be replaced by a long vowel in single syllable words. It seems that the first syllable of each word must have at least two moras, probably because Warrgamay follows first-syllable stress assignment like the rest of Australian languages and it is common for stressed syllables to require more weight than non-stressed syllables.

2.3 Conclusion

Australian languages generally follow the same principles in sound distribution. Any major deviation from the set of norms (three vowels, lack of fricatives, many places of articulation for plosive/nasal groupings, CV(C)CVC structure) results from recent phonological change due to foreign influence.

The striking similarity between the sound systems of the languages, combined with the sound correspondences revealed by historical sound changes, provides evidence for the genetic relatedness of the languages.

Another interesting feature of the Australian languages lies beyond the scope of phonology – Australian languages provide prime illustrations for the morpho-syntactic phenomenon of case. These constructions will be evaluated in detail in the next section.

Case Morphology in Aussie Languages

Case systems provide languages with a means of expressing the theta role assignments in a sentence. Put simply, they express through morphological markings on the noun and/or verb which argument of a verb is doing what. Many languages employ a case system on some level – even Old English – but few have systems as rich as those in found in Australian languages.

This section begins with a summary of case systems in Australian languages, then outlines the case system of Yidiny for illustration.

3.1 Aussie Case

Dixon (1980) claims there are twelve cases used in the Australian languages; his book, however, lists only eleven. Each Australian language only uses five or six cases on average. Case systems are not only interesting because of their morpho-syntactic functions (the richness of the system allows word order to be pretty much free) but because of their semantic compositions as well. Since the possible cases can be generalized into twelve different sorts, but each language only actually marks case in a few different ways, different languages combine these sorts of cases in different ways. This, on some level, may reflect semantic similarities among the cases.

The cases can be split up into two major classes: core, which are those that mark phrases that must be included in a sentence for it to be complete, and peripheral, which mark noun phrases that merely add extra information to a sentence, but are not required for it to be complete. For example, cases that mark syntactic subjects and objects are considered core cases, because without a subject and, when a sentence’s verb is transitive, object, the sentence would be incomplete. Other cases, such as locatives, merely provide extra information (physical place, in the case of locatives) and are not necessary for a complete sentence. (Dixon, 1980)

The two core cases in Australian languages are absolutive and ergative. Absolutive case marks the syntactic subject of an intransitive verb and the syntactic object of a transitive verb. Ergative case marks the syntactic subject of a transitive verb. (Dixon, 1980)

The peripheral cases can be further subdivided into local peripheral cases and syntactic peripheral cases.

Local peripheral cases include locative, allative and ablative. Locative can be glossed as ‘in, at, on’ and is the least morphologically and typologically marked of the three. Allative can be glossed as ‘to, towards’ and ablative, the most marked of the three, can be glossed as ‘from.’

There are some semantic correspondents to these forms among the syntactic peripherals. Dative, which marks the indirect object, purposive, which marks the goal of an activity, and causal, which marks the cause of a state, semantically correspond to locative, allative and ablative respectively. (Dixon, 1980)

Locative and dative are linked only through their lack of markedness. Allative and purposive both mark a noun phrase that connotes the end point of something – the end of physical movement in allative, which marks the physical location being traveled to, and the end of an effort in purposive, which marks result being tried for. Ablative and causal both mark a noun phrase that connotes the start point of something – the start point of physical movement in ablative, which marks the physical location being traveled from, and the beginning of an effort (whether voliltional or not) in causal, which marks the cause of a state. (Dixon, 1980)

Other syntactic peripherals include instrumental, which marks the thing used by the subject of a verb, the aversive, which marks an NP which referent is feared and therefore should be avoided, and the genitive. Genitive case marks alienable possessions, with inalienable possessions being marked by syntax more than morphology. (Dixon, 1980)

3.2 Case Study: Yidiny

Of the cases available to Australian languages, Yidiny has: absolutive, locative/allative/instrumental, ablative/causal, ergative, dative, purposive, aversive and genitive alienable. (Dixon, 1980)

Absolutive, the object of a transitive verb (1) or the subject of an intransitive verb (2), takes zero realization (ø):

(1) yiŋu+ø waguuja+ø gali+ŋ
this+ABS man+ABS go+PRES
‘This man is going’

(2) mujaam+bu waguuja+ø wawa+l
mother+ERG man+ABS look.at+PRES
‘Mother is looking at the man’

Locative/allative/instrumental, the place, destination or instrument, is marked by –da ~ -la ~ -ŋga, as shown in examples (3), (4) and (5):

(3) mujam+ø wuna+ŋ digarra+la
mother+ABS lie+PRES beach+LOC
‘Mother lives at.the.beach’

(4) mujam+ø galiŋ digarra+la
mother+ABS go+PRES beach+ALL
‘Mother is.going to.the.beach’

(5) waguja+ŋgu wagal+ø bunja+ŋ baŋgaal+da
man+ERG wife+ABS hit+PRES axe+INST
‘The man hit [his] wife with [the back of] an axe’

Ablative/causal, the starting point of movement or the cause of a verb, is marked by –mu ~ -m, as in examples (6) and (7):

(6) mujam+ø gali+ŋ digarra+mu
mother+ABS go+PRES beach+ABL
‘Mother is.going from.the.beach’

(7) mujam+ø dubuurrj+øi wuna+ŋ minya+m
mother+ABS full.up+ABS lie+PRES meat+CAUS
‘Mother is.lying.down satiated from.meat’

Ergative, the subject of a transitive verb, takes –du ~ -ŋgu ~ -lu, as in example (8), where ‘wife’ is the subject of the transitive verb ‘look.at’:

(8) wagaal+du mujam+ø wawa+l
wife+ERG mother+ABS look.at+PRES
‘[My] wife is.looking.at mother’

Dative, the indirect object, takes –nda, as in example (9), where ‘wife’ is the indirect object of the verb ‘tell’:

(9) waguja+ŋgu minya+ø buji+ŋ wagaal+nda
man+ERG meat+ABS tell+PRES wife+DAT
‘The man is.telling.about meat to.his.wife’

Purposive, the goal of an activity, takes –gu, as in example (10), where ‘meat’ is the purpose for the verb ‘go.out’:

(10) yiŋu waguuja+ø gali+ŋ minyaa+gu
This+ABS man+ABS go+PRES meat+PURP
‘This man is.going.out for.meat’

Aversive, a thing to be feared and avoided, takes –jida ~ -yida, as in example (11), where ‘people’ are a thing to be feared:

(11) yiŋu+ø waguuja+ø garba+ŋ bama+yida
This+ABS man+ABS hide+PRES people+AVERS
‘This man is.hiding for.fear.of.the.people’

Genitive alienable, a possession that is not intrinsic, takes –ni, as in example (12), where ‘dog’ in the non-intrinsic possession of ‘wife’:

(12) mujam+bu wagaal+ni gudaaga+ø wawa+l
mother+ERG wife+GEN+ABS dog+ABS look.at+PRES
‘Mother is.looking.at [my].wife’s dog’

Conclusion

These are the eight case distinctions found in Yidiny. They serve not only to exemplify the semantics of their functions through the glosses, but to illustrate the amalgamation of various case markings that is found in Australian languages. Because of the richness of the case system in Australian languages, word order is pretty much free, so speakers have more control over what get emphasis (sentence initial position) in a sentence.

Having studied the phonology and the morpho-syntax of Australian languages, a more sociolinguistic aspect of the languages may be of interest. The final section of the paper examines two speech styles of use in Australian languages: avoidance dialects and secret languages.

Aussie Speech Styles

In any language, the particular words and structures used are influenced by the relationship of interlocutors. In English, young speakers might use more slang with their peers, more neutral language with their parents, and more jargon with their colleagues. In some languages, romance languages for instance, the second person singular pronoun used reflects the relative status of interlocutors. The familiar ‘tu’ form is used to address speakers of lesser or equal status while the formal ‘vu’ form is used to address speakers of higher status.

In the Australian languages, these phenomena take on a very structured form. In varying degrees, the language is drastically altered in certain situations. Avoidance dialects are used in specified social situations to reinforce social structures, and secret languages are an extreme form of “inspeak' that engender solidarity within a specific subgroup of a society. This section discusses both of these speech styles (avoidance dialects and secret languages) in turn, with exemplary case studies for each.

4.1.1 Avoidance Dialects

Avoidance dialects are speech styles used between members of a society who are socially required to be distant. The level of distance required varies from society to society; some must use carefully chosen words, others must talk to each other indirectly, using another speaker or sometimes even an inanimate object as a “translator,” and some are prohibited from speaking to each other at all. In many societies, the interlocutors must use the avoidance dialect - commonly termed Mother-in-Law Tongues because the mother-in-law / son-in-law pair is always one that requires an avoidance style. (Dixon, 1980)

Avoidance dialects differ from the common tongue of the community in varying degrees – some simply use obscure synonyms for intended terms, or pronounce common tongue words in certain ways, for instance in Warlpiri, only a few score words belong the the avoidance dialect alone – most are obscure synonyms from the common tongue. In some systems, all content words have different lexemes in the avoidance dialect, usually with a many-to-one correspondence between the common tongue and the avoidance dialect, for instance Dyirbal, which will be detailed later. Sometimes taboo subjects totally lack terms in the avoidance dialect, for instance in Guugu Yimidhirr, the avoidance dialect lacks terms for ‘clitoris’, ‘erection’, or ‘pubic hair’. (Dixon, 1980)

In all avoidance dialects, however, phonology and syntax are identical to the common tongue, as are function words. (There is a possible exception in Wik-Menh, which had /r/ historically, then lost it in the common tongue but kept it in the avoidance dialect). (Dixon, 1980)

Avoidance dialects serve to reinforce social boundaries within a society. According to Dixon (1980) “A primary social role of avoidance behaviour appears to have been to prevent sexual relations between certain kin.” Thus, the use of an avoidance dialect reflects the kinship system of the tribe that uses it.

4.1.2 Case Study: Dyirbal

Dyirbal, spoken in Queensland, provides an interesting example of a drastic difference between common speech and avoidance speech.

Dyirbal is divided into two styles – Jalnuy, the avoidance dialect, and Guwal, the common style. Janlnuy shares with Guwal its phonology, syntax and function words but has different words for adjectives, verbs and nouns – these are in no way derived from their Guwal counterparts. Jalnuy operates fully with generic terms, utilizing relative clauses to communicate further detail. (Dixon, 1980)

The groupings of Guwal terms under the generic Jalnuy counterparts reveal the underlying semantic system of the language. Many groupings satisfy the logic of the Western world, such as groupings of flora and fauna which, for instance, group the Guwal terms miyabur ‘black oak’, jungan ‘bull oak’, gurray ‘red oak’ and junyjurrjunyjurr ‘pink oak’ into the same Jalnuy term gurrunun because “they have the same sort of grain, and differ only in color.” (Dixon, 1980, pg 114)

Other groupings, however, are less transparent. For instance, Jalnuy gunduma-l translates to Guwal bugama-l ‘to chase’ and julma-l ‘to squeeze’ to reveal an underlying connection ‘to bring together.’ Just like the distinctive Guwal terms for oak fall into one Jalnuy term because of the semantic similarity of the terms, the distinctive terms for bringing things together fall into one Jalnuy term. (Dixon, 1980)

4.2.1 Secret Languages

Secret languages are less common than avoidance dialects and serve a related but different purpose. Where avoidance dialects are learned implicitly by all members of society and used in certain situations, secret languages are taught explicitly to newly initiated boys and used only by initiated men on ceremonial occasions.

The structure of secret languages varies – some mimic avoidance dialects in inheriting their phonology, syntax and function words from the common tongue. Others simply add secret morphology to existing words, or replace each word with its antonym. (Dixon, 1980) Unfortunately, due to the nature of the languages, little documentation is available to the general public, especially for the living secret languages.

Secret languages serve to distinguish between initiated and uninitiated members of a society, and to engender solidarity between the initiated members. Most secret languages have been lost, the most common type still in existence being the antonym languages.

4.2.2 Case Study: Damin

An interesting example of a secret language is Damin, historically spoken by men of the Lardil tribe. Spoken on Mornington Island, the language uses the same morpho-syntax as Lardil and has a strikingly peculiar phonology.

The lexicon, boasting approximately 250 words, includes general terms which meanings are extended to encompass multiple terms in the common tongue. In Damin adjectives, the root form is negative, with a prefix reversing the polarity – tijitjuu ‘small’ kuritijitjuu ‘large.’ (Dixon, 1980)

The phonology of Damin inherits the phonemes of Lardil, which has a phonological system typical of Australian languages. Added to that, however, are four nasalized clicks, an ingressive lateral fricative and an ejective bilabial stop. Phonetician J. C. Catford identifies five methods of initiating sound production (initiation types e.g. pushing air out of the lungs, pulling air into the lungs, ‘clicking’ by creating a slight vacuum in the mouth…) in Damin (in Dixon, 1980) and hypothesizes “that [Damin’s] sound system is a deliberately invented one.” Khoisan languages and some Bantu languages have clicks, but no other known language in the world has an ingressive lateral fricative or an ejective bilabial stop. Furthermore, no known natural language has this degree of variation in initiation types. (Dixon, 1980)

4.3 Conclusion

The alternative speech styles utilized by many Australian languages reflect the social structure of the groups that speak them, as well as semantic ties. Avoidance dialects reveal a society’s kinship structure because the dialects are used with community members that are, because of the kinship structure, sexually unavailable. Secret languages parse out the society into the initiated and the uninitiated by their very nature. Futhermore, through their generic lexicons and the mapping of multiple common terms onto a single term in the alternative speech style, they illustrate semantic links between more precise terms.

Because of their direct importance to the culture that speaks them, these styles must be maintained. Most have already vanished, the common speech styles following closely at their heels. Without the joint effort of politicians, linguists and community-members, these cultural voices will fade into oblivion, taking an entire worldview with them.

Conclusion

Australian languages illustrate a number of fascinating phenomena. Their categorizations of phonemes and phonological rules provide a valuable linguistic corpus of data that helps reveal typological norms and universal tendencies. Their complex case systems provide an excellent illustration of this morpho-syntactic phenomenon and an opportunity to study and compare this system with those used in other languages. The speech styles used encode a perception of social relationships and a worldview.

The vitality of these phenomena are imperative not only to the maintenance of the culture that speaks the languages, but to the preservation of a diverse linguistic corpus, which enables linguists to create better grounded theories about language. Since these languages are rapidly dying out, this area of the world needs more attention from field linguists and the irreplaceable data collected must be stored in such a way that it will endure. These voices are vanishing, and it is up to linguists and community members to strengthen them.

Bibliography