|Kalau Lagau Ya, alt. Kalaw Lagaw Ya|
|Western Torres Strait|
|Region||Western and Central Torres Strait Islands, Queensland|
|Ethnicity||Badu, Mabuiag, Kaurareg, Mualgal|
(Torres Strait Islanders)
|957 (2016 census)|
|Western Torres Strait Islander Sign Language|
Range of Kalau Lagau Ya (orange) in the Torres Strait
Kalau Lagau Ya, Kalaw Lagaw Ya, Kala Lagaw Ya([kala(u) laɡau ja]), or the Western Torres Strait language (also several other names, see below), is the language indigenous to the central and western Torres Strait Islands, Queensland, Australia. On some islands, it has now largely been replaced by Torres Strait Creole.
Before colonisation in the 1870s–1880s, the language was the major lingua franca of the area in both Australia and Papua New Guinea, and is fairly widely spoken by neighbouring Papuans and by some Aboriginal people. How many non-first language speakers it has is unknown. It also has a 'light' (simplified/foreigner) form, as well as a pidginised form. The simplified form is fairly prevalent on Badu and neighbouring Moa.
The language is known by several names besides Kalaw Lagaw Ya, most of which (including Kalaw Lagaw Ya) are names of dialects, spelling variants, dialect variants and the like — and include translations of the English terms, Western Island Language and Central Island Language:
One term used by Eastern Islanders and neighbouring Papuans for Kala Lagaw Ya is Yagar Yagar, from the word yagar (yá 'speech, etc.' + gár 'sympathy clitic' ('dear', 'please', etc.), often used by Western and Central Islanders in speech to show a sympathetic or nostalgic frame of mind.
In literature on the language the abbreviations KLY (Kalaw Lagaw Ya), KKY (Kalau Kawau Ya), KulY (Kulkalgau Ya), MY (Muwalgau Ya) and KY (Kaiwaligau Ya) are often used as abbreviations. The name Mabuiag /mabujaɡ/, in English pronounced //, is fairly widespread as a name for the language, this having been established by the Cambridge Expedition to Torres Strait, whose main research on the language was with Mabuiag material. Though the preferred term in English in Academia for some time was Kala Lagaw Ya, according to Ober (2007), the form was always regarded as "colloquial" by native speakers. In a High Court decision on 7 August 2013, the decision was taken to officially term the language Kalau Lagau Ya, using the formal form.
When speaking to each other, speakers generally refer to the language as Langgus 'language' or use phrases such as KLY/KulY ngalpudh muli, MY-KY ngalpudh/ngalpadh muli, KKY ngalpadh muliz, e.g. KLY/KulY ngalpudh muuli, thanamunungu tidailai!, MY-KY ngalpudh/ngalpadh muuli, thanamuningu tidailai!, KKY ngalpadh muli, thanamulngu tidaile! 'Speak in our language so they don't understand!'. Ngalpudh/ngalpadh literally means 'like us'. The construction X-dh mula+i- 'speak X-like' is used to refer to speaking in a language, e.g. KKY markaidh muliz 'speak [in] English', zapanisadh muliz 'speak [in] Japanese', dhaudhalgadh muliz 'speak [in] Papuan', mœyamadh muliz 'speak [in] Meriam Mìr', thanamudh muliz 'speak like them, speak [in] their language'. It is otherwise common for speakers to use nominal phrases like KLY/KulY ngalpun ya, MY-KY ngalpun/ngalpan ya, KKY ngalpan ya 'our language' to refer to the language when speaking to each other.
Kalau Lagau Ya is spoken on the western and central islands of Torres Strait, between Papua New Guinea (Naigay Dœgam Dhaudhai "North-side Mainland/Continent", also called Mœgi Dhaudhai "Small Mainland/Continent", KKY Mœgina Dhaudhai/Dhawdhay) and the Australian mainland (Zey Dœgam Dhaudhai "South-side Mainland/Continent", also known as Kœi Dhaudhai "Big Mainland/Continent"), though on some islands it has now been largely replaced by Brokan (Torres Strait Creole).
Before Colonisation in the 1870s–1880s, the language was the major lingua franca of the area in both Australia and Papua, and there is some folk history evidence that the language was spoken as a first language in a few villages neighbouring Torres Strait in Papua. It was also formerly spoken by the Hiámo (Hiámu, Hiáma) of Daru (Dhaaru) to the north-east of Torres Strait, who were originally settlers from Yama [Yam Island] in Torres Strait, Hiámu/Hiámo/Hiáma being a Kiwai pronunciation of Yama. The main body of the Hiámo moved to the Thursday Island group to escape the Kiwai colonisation of Daru some centuries ago.
The language is classified as being part of the Pama–Nyungan languages. Mitchell (1995) and Mitchell (2015) regards it as a mixed language with an Australian core (Pama-Nyungan) and Papuan and Austronesian overlays, while Capell (1956) and Dixon (2002) classify it among the Papuan languages. The personal pronouns are typically Australian, most kin terms are Papuan, and significant sea/canoe and agricultural vocabulary is Austronesian.
Kalaw Lagaw Ya has only 6% cognation with its closest Australian neighbour, Urradhi, with a further 5% 'common' vocabulary (loans of various origins) — and about 40% common vocabulary with its Papuan neighbour, Meriam Mìr. Of the 279 Proto-Paman forms given in Sommer (1969, pp. 62–66), only 18.9% have definite realisations in Kalaw Lagaw Ya, with a further 2.5% which may be present. One word that illustrates the problems of 'may-be' relationship is kùlbai (KKY kùlba) 'old', which may be a metathetic realisation of CA *bulgan 'big; old'. Potentially 80% of the vocabulary of the language is non-Australian, and includes Papuan and Austronesian items. Bouckaert, Bowern and Atkinson (2018) found that Kalaw Lagaw Ya had the highest number of 'unique' (that is, otherwise unshared) forms of any Australian language in their sample.
|*gamo 'belly' |
|*p[ae]- 'that, there'
pi-/pe- 'specifically yonder'
|*waura 'south-east' |
kœnara 'k.o. tree snake'
|*boro-ma 'pig' |
bero 'rib; side of boat, hillside, river bank, etc.'
|*pu[lr]i 'magic' |
puy(i) (older puuři) 'magic, plant'
Oral tradition and cultural evidence recorded by Haddon (1935) and Laade (1968), backed by archaeological evidence and linguistic evidence, shows that Austronesian trade and settlement in South-West Papua, Torres Strait and Cape York occurred; the languages have significant Austronesian vocabulary content, including items such as the following:
|Kalaw Lagaw Ya||meaning||Meriam Mìr||meaning||Bine
(Kalaw Lagaw Ya loan)
(Kalaw Lagaw Ya loan)
Some of the Austronesian content is clearly South-East Papuan Austronesian:
|word||Kalaw Lagaw Ya||Gudang
(Central District, Papua)
SE Kiwai harima
|upirri 'magic'||—||hui (< *fui) 'magic'||*pu[rl]i||*(m)puluŋ 'magic'|
The linguistic history of the Torres Strait area is complex, and interaction of well over 2500 years has led to many layers of relationship between the local languages, including many words that are obviously common, such as the following 'trade' words in Torres Strait area languages.
|Kalaw Lagaw Ya||Meriam Mìr||Kiwai
tusk, knife, tusk/knife-life formation
tusk, knife, tusk/
long house, hall; church
|ìut (alt- eut)
any strange four-legged animal
However, the question of external relationships of Kalaw Lagaw Ya is also complicated by resemblances between both the Paman (Pama-Nyungan, Australian languages) and the Trans-Fly (Papuan) languages. Though few, these may be significant, and include forms such as those noted below, not all of which appear in Kalaw Lagaw Ya. Such resemblances could point to a deep-level relationship dating back to before the flooding of Torres Strait at the end of the last age, as claimed by Mitchell , or they could point to genetic inheritance and subsequent language contact, as discussed by Alpher, Bowern, and O'Grady 2009.
(or a specific North Cape York language)
|meaning||Proto-Trans-Fly||meaning||Kalaw Lagaw Ya||meaning|
kùrusai- (compounds only)
|*mini||good||*mi:nji||id.||miina||real, true, very|
(wara 'one of a group')
|*pama||man, person||*pyama||id.||(mabaig lit. 'walker')||id.|
A comparison of the Kalaw Lagaw Ya, Meriam Mìr, Kiwai and Uradhi personal pronouns show similarities and differences in typology. In comparison to Uradhi, Kalaw Lagaw Ya has an archaic typology — or, rather, Uradhi has innovated, having lost the Common Australian 1, 2 and 3 plurals. Kiwai does not have 1–2 pronouns, while Meriam Mìr does not have a dual and trial/paucal set of pronouns which correspond to its verb system. The Kalaw Lagaw Ya system, like that of Uradhi, is Australian:
Note that except for Meriam Mìr, the Trans Fly languages also have two-gender masculine-feminine systems, though not marked on the pronouns themselves.
|plural||(same as dual)||ana(va)||(same as dual)||(same as dual)|
However, even though the system has no real surprises for Australian linguistics, it is clear that Kalaw Lagaw Ya has innovated in the 1st and 2nd pronouns, which have the following CA origins:
The 2nd person dual and plural pronouns are based on forms that literally mean 'you dual' (ngipel) and 'you-they' (ngitha[na]), in much the same way as the demonstratives mark the dual and plural (see further below in Nominal Morphology).
|you and I||ngœba||ngœba||ngœba||ngœba||ngœba||*ngana+pulV|
'we dual, exclusive'
|we dual (exclusive)[a]||ngalbay||ngalbai||ngalbai/ngalbe||ngalbe||ngalbai/ngalbe||*ngali+[?]|
'you and me, you and us'
|*NHiin+pulV 'you dual'|
|*NHiin + *DHana 'they plural'|
|*pula 'they dual, two'|
|*DHana 'they plural'|
|mi-||mi-||mi-||mi-||*miNHa 'food; what'|
An examination of the various sub-systems (vocabulary, syntax, morphology) suggests the following:
Some basic and abstract vocabulary, all personal pronouns (inc. who and what/which), some verbs. Some grammar, such as nominal and verb morphology (subject, agent, object, genitive, -l locative, -ka dative, perfective attainative, imperfective, -i/-iz(i) perfective active. These typological categories also exist in the Trans-Fly languages; the forms in Kalaw Lagaw Ya are clearly Australian.
Some basic and abstract vocabulary, some verbs. Some grammar, such as verb number and different stems for different number forms of some verbs. Use of state/movement verbs as existential and stative 'be' verbs. Two non-personal pronominals: naag/naga 'how', namuith 'when' (both in KKY, the dialect of the islands off the Papuan coast).
Some basic vocabulary, terminology dealing with agriculture, canoes, the weather, the sky and the sea, some abstract nouns, some verbs. Possibly some grammar in the form of function words, such as waadh (KKY waaza) 'existential emphasis' (i.e. 'it is true that ... '), Proto Oceanic Austronesian *waDa 'existential'.
The Australian word forms and structure found in Kalaw Lagaw Ya appear to be retentions, i.e. inherited; the original Australian forms appear to be unchanged at the core level. This suggests that the language is not a pidgin/creole in origin, but an Australian language which has undergone strong external lexical and grammatical influence. The language appears to be a classic case of shift, whereby speakers of one language retained multilingualism over a long period of time, absorbing aspects of another language. The Austronesian and Papuan overlays modified the Australian phonology and syntax profoundly. The contrast of Australian laminal nh/ny and lh/ly and apical n and l has been lost, voicing has become phonemic and s, z, t, d, o and òò have developed. This also affected the phonology of Australian vocabulary, where these 'foreign' sounds also occur.
The non-Australian content appears to be mainly lexicon (including verbs), particularly dealing with the sea, farming, canoe and sky/weather/astrology, with possible some syntactic words. This presents a picture of a typically extensive borrowing situation with much lexical borrowing and some structural borrowing with a large amount of passive bilingualism and little active bilingualism.
Laade's picture (1968) of Australian and Papuan settlement in Torres Strait supports the above scenario of Papuan and Austronesian speakers who shifted to an Australian language over a long period of time, the Austronesians being culturally a superstratum, however not in a position to impose their language. He presented folk history evidence that a few Austronesian traders (men) settled at Parema (north-east of Daru) and married local [Proto–Trans Fly speaking] women. To avoid further miscegenation, they soon moved and settled in Torres Strait, first to the Eastern Islands, then to the Central Islands, then to Moa, Badu and Mabuiag. At Mabuiag, Badu and Moa they found Aboriginal people, killed the men and kept the women (and presumably the children). Some moved on up to Saibai, Dœwan and Bœigu to avoid this new miscegenation, hence the lighter colour of many Saibai, Dœwan and Bœigu people. Bœigu folk history collected by Laade also shows direct East Austronesian genetic influence on Bœigu.
The social context was that of a few Austronesian men who settled on the outskirts of an East Trans-Fly group, intermarried, and whose children were either bilingual, or speakers of their mothers' language, with some knowledge of their fathers' language. The local people did not need to speak the traders' language, who in turn had to speak the local language. The children in turn would then speak the local language, with varying ability in the fathers' language, particularly in areas that were culturally important for the fathers.
These people then shifted to Torres Strait — maintaining established ties with Papua as well as with Austronesian speakers further east (this latter being suggested by various characteristics of the Austronesian content in Kalau Lagaw Ya) — and overlaid an Australian population in such a way that the majority of women spoke an Australian language, with a significant number, mainly men, who spoke a South-East Papuan Austronesian language, accompanied by their Papuan wives and their perhaps bilingual children. Over time, the core structure of the local mothers' language dominated, with retention of the newcomers' Papuo-Austronesian content in the appropriate cultural subsystems. In essence this would have been a 'replay' of the original settlement by Austronesian traders at Parema, with the women understanding the language of the men, but not really needing to speak it while retaining parts of their language for significant areas. The children then created a new language shift to an Australian language with a Papuan-Austronesian admixture.
Kalaw Lagaw Ya is thus a mixed language in that a significant part of its lexicon, phonology and grammar is not Australian in origin. The core nominal, pronominal and verb morphology is Australian in both form and grammar — though a certain amount of the grammar is common to Trans-Fly and Paman languages in the first place. Some semantic categories, verb number morphology, and some other morphology are non-Australian in origin. Potentially 80% of its vocabulary is non-Australian. The interplay of the above within the subsystems of Kalaw Lagaw Ya lexicon, phonology and grammar points more to mixing through shift and borrowing rather than pidginisation and creolisation.
The language also has some vocabulary from languages outside the Torres Strait area, from the Indonesian, Malay, Philippine, English and other 'outsiders'. Where loan words from the Western Austronesian (Indonesian, etc.) loans are concerned, it is possible that some such came into the language in pre-European contact days, with the Makassans and similar fishermen/traders who visited northern Australia and Torres Strait.
Examples of post-European contact Western Austronesian loan words:
|word||Kalaw Lagaw Ya||origin|
|coconut toddy||thúba||tuba (Eastern Indonesian or Philippine language)|
|trumps (in cards)||záru||zaru/jaru (Eastern Indonesian or Philippine language)|
|mate, friend, brother||bala
Boigu variants: bœra, baya
|bela/bala (Eastern Indonesian or Philippine language)|
Some words in the language, assuming that they are Western Austronesian loans, appear to be pre-contact words. This is suggested by form and use in the language and in neighbouring languages (some of these words may ultimately be from Arabic and Sanskrit).
|Kalaw Lagaw Ya||meaning||possible source||meaning|
|come! (singular)||Malay: ayo||come!|
(Sanskrit: वायु, romanized: vāyú)
(Sanskrit: अधि, romanized: adhi)
|huge, great |
(also as an honorific)
|pawa||deed, action, custom||Malay: paal [paʔal]
(Arabic: فَعَلَ, romanized: faʿala)
In the KKY dialect of Kalaw Lagaw Ya, 'waterspout' is markai gùb(a) 'spirit wind'; waterspouts were one of the weapons of the markai who mainly came from the west/north-west (i.e. from what is now Indonesia) in the NW monsoon season (when waterspouts are common), and went back to the west/north-west with the SE trades.
Loans from modern Eastern Austronesian (Polynesian and Melanesian) into the language are mainly of religious or 'academic' use. In general, such words are terms for objects that are strictly speaking European goods. One exception is the last in the following table, which is commonly used instead of the traditional words imi 'spouse's opposite-sex sibling', 'opposite sex sibling's spouse' and ngaubath 'spouse's same-sex sibling', 'same-sex sibling's spouse'. These have also similarly been replaced in common usage by the English loan woman (pronounced [woman]) in the meaning of 'sister/daughter-in-law'.
|Kalaw Lagaw Ya||meaning||source||meaning in originating language|
|thúsi||book, document, letter, etc.||Samoan: tusi||(same meaning)|
|laulau||table||Samoan: laulau||plaited coconut leaf used as a tray|
|wakasu||anointment oil||Drehu: wakacu||coconut oil|
(emotive form thawi)
|Kalaw Lagaw Ya||meaning||source||meaning in originating language|
|basalaya||kingdom||Ancient Greek: βασιλείᾱ||id.|
|aretho||holy communion||Ancient Greek: ἄρτος||wheaten bread|
|Sathana||Satan||Biblical Hebrew: שטן||Satan, opponent, adversary|
|Sabadh(a), Sabadhi||Sunday||Biblical Hebrew: שבת||Saturday (Sabbath)|
Two early English loans of interest show back formation from what in the language appeared to be a plural. Most nouns (a) form the plural with an -l suffix, and (b) in the nominative-accusative singular elide the stem final vowel, thus tukuyapa- 'same-sex sibling', plural tukuyapal, nominative-accusative tukuyap. Under this model 'custard-apple' became katitap, plural katitapal, and 'mammy-apple' (pawpaw/papaya) became mamiyap, plural mamiyapal.
There are four main dialects, two of which are on probably the verge of extinction, one (Kaiwaligau Ya) through convergence to the neighbouring Kalaw Lagaw Ya. Within the dialects there are two or more subdialects. The average mutual intelligibility rate, based on a Swadesh count, is around 97%.
The Southern dialect has certain characteristics that link it closely to the northern dialect, and folk history dealing with the Muralag group and Mua reflects this, in that the ancestors of the Kowrareg (the Hiámo) originally came from Dharu (Daru, to the north east of Torres Strait) — and who had previously settled on Dharu from Yama in Central Torres Strait.
They cut down a big tree earlier today to make a canoe.
|They pl+nom||today||big||tree+acc||chop+att+sing+today pst[+loc]||canoe+dat||make+vn+dat|
Some isolect markers of the four dialects of Kalaw Lagaw Ya:
|Kalau Kawau Ya||Kaiwaligau Ya||Kalaw Lagaw Ya||Kulkalgau Ya||Kauraraigau Ya |
|-pa, -pari |
|-i (Badhu -in),
-izi (Badhu -izin)
Phonological differences between the dialects are rare, and in general sporadic. The only regular differences are the following:
Found in Kulkalgau Ya and Kaiwalgau Ya:
Such elision is rare or sporadic in Kalau Kawau Ya.
In Kalaw Lagaw Ya, such final vowels in correct language are devoiced, and deleted in colloquial language, except in a small class of words which include bera 'rib', where there is a short vowel in the stem and in which the final vowel is permanently deleted, with compensatory lengthening of the final consonant (thus berr).
Strictly speaking, the process is not final vowel devoicing, but rather stressed vowel lengthening accompanied by final vowel devoicing — except in the case of words such as bera 'rib' > berr, where the process is final consonant lengthening by the final vowel being 'incorporated' into the consonant. Note that in the following the word-final capital letter represents a devoiced vowel:
In declined forms of such words, the long vowel is shortened, and the final vowel voiced, and in words like ber 'rib' the final vowel often reappears:
This vowel shortening in affixed/modified forms exists in all dialects, however the other dialects have retained contrastive length to some extent, whereas Kalaw Lagaw Ya has largely lost it for 'morphophonological' length, where the stressed vowel in non-emotive words (see below) of one or two syllables is automatically lengthened in the nominative-accusative; this also applies to words of three syllables with second syllable stress (as in mœrààpI 'bamboo').
One of the very few length contrasts in the Kalaw Lagaw Ya dialect is kaaba 'dance performance, knot in bamboo etc.' vs kaba, kab 'paddle, oar' (Old Kaiwaligaw Ya (Kaurakaigau Ya) kœRaba. œRa has regularly given short a in Kalaw Lagaw Ya in kaba, kab). Such length contrasts are more widespread in the other dialects.
The exceptions are (1) the small class or words that include ber 'rib' and kab 'oar, paddle', and (2) emotive words. Emotive words are those that equate to a certain extent to diminutives in languages such as Irish, Dutch and German, where specific suffixes are added to show 'diminutive' status (-ín, -je and -chen respectively). Emotive words in the Kalaw Lagaw Ya dialect include familiar kinship terms [the equivalent of English Mum, Dad and the like] and words used in emotive contexts such as singing/poetry.
|Mum||(apuuwa, apùù, àpu — mother)||Ama|
|Dad||(thaathi, thaath — father)||Baba|
|home (island)||laaga, laag||laga|
|dust, spray||pœœya, pœœy||pœya, paya|
|bamboo||mœrààpi, mœrààp||mœràpi, marapi|
|head||kuwììku, kuwììk||kuwìku, kuiku|
A small class of words in Kalau Kawau Ya do not have the final i-glide found in the other dialects, including the following:
Word forms in neighbouring languages as well in the Kauraraigau Ya (Kowrareg) of the mid-to-late 19th century, such as the Meriam Mìr kopor and Kauraraigau Ya kupar/kopar 'birth cord' show that in such words the final -i/Ø are the modern forms of older *ɾ.
The main syntactic differences are:
In all dialects except Kalau Kawau Ya, the verb negative is the nominalised privative form of the verbal noun. As this form in itself a noun, its subject and direct object are cast in the genitive:
The Kalau Kawau Ya dialect uses the verbal noun privative form as an invariable verb negative:
The Kalau Kawau Ya dialect has the tenses and aspects listed in the section on verb morphology. The other dialects have largely lost the remote future tense, using the habitual instead; the remote future in the other dialects is retained most commonly as a 'future imperative', where the imperative refers to a vague period in the future. The Kalaw Lagaw Ya dialect also has a 'last night' tense, where the adverb bungil/bungel (reduced form bel) 'last night' has become a verb postclitic, following the model of the adverb ngùl 'yesterday', which had previously become grammaticalised as a 'recent past' tense marker in all dialects, with reduction to -ngu in Kalau Kawau Ya. In the other dialects bongel 'last night' is a fully functioning temporal adverb used in conjunction with either the today past or the recent past.
The dialects differ in the forms of the following affixes:
The main nominal affix difference is the dative ending, which has the following forms in the various dialects:
The plural/HAVE suffix -LAI (underlying form) also shows a small amount of dialect variation with stems of two syllables, where Kulkalgau Ya differs from the other dialects in retaining the full form of the suffix -lai reduced to -l in the other dialects. In stems of three or more syllables, the suffix is reduced to -l in all dialects, while retained as -lai (variants according to noun sub-class -thai, -ai, -dai) with stems of one syllable.
burum 'pig', stem: buruma-, plural burumal
lag, KLY laaga 'place, home, home island', stem: laga-, plural lagal, KulY lagalai
The main differences between the dialects are to do with vocabulary, as can be seen in the following examples:
Kala Lagaw Ya is the only Australian language to have the alveolar fricatives /s/ and /z/. However, these have allophonic variants /tʃ/ and /dʒ/, which are the norm in Australia languages (usually /c/ and /ɟ/ but non-contrasting). These latter two are allophones in that in all environments /s/ and /z/ can appear, while /tʃ/ and /dʒ/ can not appear at the end of a word; note that this allophony is very similar to that of the neighbouring Papuan language Bine. All the stops, except for the alveolars ⟨t⟩ and ⟨d⟩, have fricative allophones, thus ⟨p⟩ can be [p] or [ɸ], ⟨k⟩ can be [k] or [x], ⟨b⟩ [b] or [β], and so on. Furthermore, it is one of the few Australian languages with fully functioning voiced-voiceless distinctions (⟨p/b⟩, ⟨t/d⟩, ⟨s/z⟩, ⟨k/g⟩, ⟨th/dh⟩) — and one of the few without retroflex stops.
The language is also one of the few Australian languages with only one rhotic, one ⟨l⟩ and one ⟨n⟩. The earliest recorded dialect, Kaiwalgau Ya (Kauraraigau Ya [Kowrareg]), however, did have two rhotics, the tap and the glide; the rhotic glide has in general become /j/, /w/ or zero in the other dialects (and Modern Kaiwaligau Ya), rarely /r/. Neighbouring languages retain an /r/ in related words, such as:
However, in singing, /s/, /z/ and /r/ are pronounced [s], [z], and [ɹ], and virtually never as [tʃ], [dʒ] and [r].
|Labial||p ⟨p⟩||b ⟨b⟩||m ⟨m⟩||w ⟨w⟩|
|Dental||t̪ ⟨th⟩||d̪ ⟨dh⟩||n̪ ⟨n⟩||l̪ ⟨l⟩|
|Alveolar||t ⟨t⟩||d ⟨d⟩||—||r ⟨r⟩|
|Alveo-palatal||s/tʃ ⟨s⟩||z/dʒ ⟨z⟩||—||j ⟨y⟩|
|Velar||k ⟨k⟩||ɡ ⟨g⟩||ŋ ⟨ng⟩||—|
|Close||i ⟨i⟩||iː ⟨ii⟩||u ⟨u⟩||uː ⟨uu⟩|
|Close-mid||e ⟨e⟩||eː ⟨ee⟩||ʊ ⟨ù⟩||ʊː ⟨ùù⟩|
|Open-mid||ə ⟨œ⟩||əː ⟨œœ⟩||o ⟨o⟩||oː ⟨oo⟩|
|Open||a ⟨a⟩||aː ⟨aa⟩||ɔ ⟨ò⟩||ɔː ⟨òò⟩|
Internal reconstruction and comparison with neighbouring languages suggests an underlying four vowel structure with contrasting vowel length, where underlying *i typically gives surface ⟨i⟩ and ⟨e⟩, underlying *a typically gives surface a and œ, underlying *ò typically gives surface ⟨o⟩ and ⟨ù⟩, and underlying *u typically gives surface ⟨ù⟩ and ⟨u⟩ (there are other realisations as well, depending on rules of assimilation etc.):
The language undergoes low-level vowel shifts, caused by stress domination within words and phrases. Long vowels are shortened, and short vowels raise when the word is preceded by morphemes such as adjectives, demonstrative articles, prefixes and the like; the changes also occur within words when these are suffixed:
The processes are low-level in that they are not 'automatic' — the changes do not have to occur and can be consciously 'blocked'. In normal speech, vowel shortening and the change of a to œ are the norm, which the changes of e to i and o to u are sporadic, and most common in unstressed syllables.
Assimilation of vowels to other vowels in the vicinity and consonants is also widespread, particularly of the vowel œ:
The following summary of the phonology of Old Kauraregau Ya is compiled from MacGillivray (1852), Brierly (in Moore 1978), Ray and Haddon (1897) and Ray (1907). In general, there does not to appear to have been any great phonological difference between OKY and the modern dialects of Kalau Lagau Ya (apart from the retention of ř).
Stress appears to have been similar to that of the modern dialects, with stress patterns being most similar to that of modern Bœigu and Ngœrupai speech, the most conservative dialects in this respect.
Stress is initial:
A few forms (such as gru: gœrú 'sugar cane') show that contrastive stress existed in bisyllabic words.
Stress is either on the initial or second syllable:
Shifted stress also appears to have occurred as in the modern dialects:
These appear to have been the same as in the modern language. Vowel length in general appeared in the same environments as in KKY, though some amount of vowel lengthening under the KLY model is evident, as in kawp: kaapu 'seed', Kalau Kawau Ya / Kulkalgau Ya kapu, Kalau Lagau Ya kaapu.
The exact extent of retention of underlying vowel length and the development of variant forms is difficult to measure, as the spelling systems used by Brierly and MacGillivray did not always mark vowel length. Further, as they obtained words through elicitation (which has a common 'lengthening effect' on vowels when words are 'slowed down'), there are a few cases where they marked vowel length wrongly. Ray marked vowel shortness in stressed syllables.
The various sound changes that the vowels and diphthongs undergo in the modern language also occurred in OKY. One change that occurred much more than in the modern dialects was that of ai monophthongisation to e. The resulting e then often raised to i in open unstressed syllables.
In the modern dialects, these forms are:
The change of ai to ei appears to have been very common elsewhere in the dialect:
One form shows optional i insertion:
gassumu-, gassima-: gasama- ~ gasœma- ~ gasima- 'catch, get', modern dialects gasama- ~ gasœma-
OKY had one more consonant than modern WCL, transcribed ř. Though the actual pronunciation of this sound and its difference from r was not given by any early writer, it most likely was a rhotic glide [ɹ], perhaps with a palatal 'hue'. The loss of this sound in the other dialects (and in modern KY) occurred in the following rules; the changes were beginning to be evident already in OKY:
Ř between like vowels or in [ə]__V deletes.
Ř sporadically becomes [+hi] when in ə__a and the following syllable is stressed.
Ř becomes a [+V] glide when between [-hi] and [+hi] vowels, and between [+bak] and [-bak] vowels.
Vuř becomes /w/ when intervocalic.
Ř optionally becomes /i/ when syllable final and following [-hi] vowels; in at least words metathesis first occurred.
Ř deletes when syllable final following high vowels and non-final.
Ř disappears when followed by unstressed i and more than one syllable.
Early spellings (e.g. möaga [məaga] 'sweat' and neet/naat/nöat/niet [nejat], [nat], [nəat], [nijet] 'platform' show that ř disappeared first, leaving a hiatus (except in those cases where ř > y~i), with reduction of [V1-V1] and [ə-V1] to [V1], and [ə-VV] to [VV].
OKY underwent the same allophony and sound changes as the modern dialects, though z ~ dh and s ~ th variation appears to have been more general in OKY, as in the following (perhaps evidence of older allophony in the language which is now levelling out):
An instance of optional r deletion before s is also attested in the following example, unless the first i in myaichipp is a misprint or misreading of *myarchipp:
maayi-arsipa 'wail, keen, weep': Brierly myaichipp, MacGillivray maierchipa, OKY mayarsipa, mayasipa
Various forms in OKY showed metathesis of ř and r in the environment of u, i and au:
Syllabification occurred as in the modern dialects, with the addition of ř also attested as a syllable final consonant. One word was recorded by Brierly and MacGillivray with a [+nas][-son] cluster, namely enti 'spider', however this appears to be a confusion; enti is probably Gudang (Australia) ant[h]i 'sore'.
Syllables were vowel final or end in r, ř, l, glide i or glide u. Otherwise surface syllable final consonants have an underlying following vowel, in which case all consonants could be syllable initial.
There is no strict standard spelling, and three slightly different orthographies (and often mixes of them) are in use.
The Mission Spelling (established at first by Loyalty Islands missionaries in the 1870s, then modified by Polynesian missionaries in the 1880s): a, b, d, e, g, i, j, k, l, m, n, ng, o, ö, p, r, s, t, u, z, sometimes also th, dh, dth, tr, dr, oe, ë, w, y, j, and sometimes double vowels to show length. This spelling system was based on that used for the Drehu (Lifu) language, though later with the change to Polynesian mission staff, as well as the growing number of indigenous Torres Strait missionaries, the overtly Drehu forms tr, dr and ë were lost; these had no phonological basis in Kalaw Lagaw Ya. The mission system is used in the Reports of the Cambridge Expedition to the Torres Strait (Haddon et al., 1898 and on, University of Cambridge) and in Myths and Legends of Torres Strait (Lawrie, University of Queensland, 1971). Ray, the linguist of the Cambridge Expedition, also used various diacritics to represent short vowels and vowel quality.
Established in the 1970s: a, aa, b, d (alveolar), dh (dental), e, ee, g, i, ii, k, l, m, n, ng, o, oo,oe (/ə/), ooe (/əː/), p, r, s, t (alveolar), th (dental), u, uu, w, y, z
Established in the late 1970s: a, b, d (alveolar), dh (dental), e, g, i, k, l, m, n, ng, o, oe (/ə/), p, r, s, t (alveolar), th (dental), u, w, y, z (vowel length, though it exists, is rarely represented).
People not only use these three slightly differing spelling systems, but also write words more or less as they pronounce them. Words are therefore often spelt in various ways, for example sena/sina 'that, there', kothai/kothay/kothei/kothey/kothe 'back of head, occiput'. Such variation depends on age, family, island, village and other factors such as poetic speech. It can be difficult at times to decide which is most correct — different people have different opinions (and sometimes have very strong opinions).
In general the pronunciation of older people has priority; however, some people can actually get quite offended if they think the language is written the 'wrong' way. Some insist that the mission spelling should be used, others the Bani spelling, and still others the KKY (Saibai etc.) spelling, and still again others use mixes of two or three, or adaptations thereof. Some writers of the Mabuiag-Badhu dialect (Kalaw Lagaw Ya), for example, write mainly in the Mission system, sometimes use the diagraphs oe, th, dh (variant dth) and sometimes use capital letters at the ends of words to show devoiced vowels, such as ngukI 'fresh water/drinking water, fruit juice' /ŋʊːki̥/. In the Bani/Klokheid orthograophy nguki is written nguuki, and in the other dialects the final vowel is either fully voiced, nguki /ŋʊki/), or elided, nguk /ŋʊk/).
The biggest bone of contention between the advocates of the 'modern' orthographies and the 'traditionalist' orthographies is the use of w and y to show the semi-vowels. In general native speakers in literacy classes seem to find y and w very difficult to learn, and that u and i are the 'logical' letters to use. Syllabification of words by untrained speakers suggests that u and i are really the underlying sounds. Thus, a word like dhaudhai/dhawdhay 'mainland, continent' syllabifies as dha-u-dha-i, not dhau-dhai. In songs, the glide-u/i can also be given full syllable status. Historical considerations also point to the semi-vowels often being vocalic rather than consonantal. Thus, lagau, the genitive of laag[a] 'place' is in underlying form <laaga+ngu>; the full form of the genitive ending -ngu is only retained where the nominal has a monosyllabic stem (see the section on Nominal Morphology). Similarly, verbal nouns end in -i, e.g. lumai, stem luuma- 'search, look for, seek, hunt'. The mid-19th century to early 20th century records of Kauaraigau Ya show that the verbal noun ending was previously -ri (thus lumari), where the -r- was presumably the rhotic glide rather than the rhotic tap/trill.
A dictionary now in preparation (Mitchell/Ober) uses an orthography based on detailed study of the surface and underlying phonology of the language, as well as on observation of how people write in real life situations. It is a mix of the Mission and Kalau Kawau Ya orthographies with the addition of diacritics (the letters in brackets) to aid correct pronunciation, since many of the people who will use this dictionary will not be speakers of the language:
a (á), b, d, dh, e (é), g, i (í), k, l, m, n, ng, o (ó, ò, òò), œ (œ'), r, s, t, th, u (ú, ù), w, y, z
Within this orthography, w and y are treated as consonants — this is their phonological status in the language — while u and i are used as the glides where phonological considerations show that the 'diphthong' combination has vocalic status.
The typewritten forms of œ and œœ are oe and ooe.
The English pronunciations given in the list below are those of Australian English, and are only meant as a guide. The letters in square brackets () are the IPA.
Combinations of vowels ('diphthongs', such as ai, au, œi, eu etc.) are pronounced as written. Thus, for example, ai is a-i (basically very similar to 'i' in 'mine' with a posh accent). In singing and sometimes in slow speech, such vowel combinations can be said separately. In the Bani and Saibai (etc.) orthographies, the last elements can be written as y and w instead of i and u. The diphthongs are:
Where the morphology is concerned, the language is somewhere along the continuum between agglutinative and fusional. Nominals have the following cases: nominative, accusative, instrumental (subsumes ergative), dative (subsumes allative, purposive), ablative (subsumes elative, avoidative), specific locative, nonspecific locative (subsumes perlative and comitative) and global locative. Nominals also have the following derived forms: privative, similative, resultative and proprietive, which also forms the noun nominative-accusative plural. All stems end in a vowel or a semi-vowel, except for a few monosyllables ending in -r and -l (which includes a very few reduplicated words, like tharthar 'boiling, seething', as well as ngipel 'you dual' [a compound of ngi 'you singular' and -pal 'two']). For many nouns the surface nominative(-accusative) undergoes a final stem-vowel deletion rule; in the Kalaw Lagaw Ya dialect the rule results in final devoiced vowels accompanied by main vowel lengthening. There are three numbers, singular, dual and plural. Singular and dual are the same form in all nominals except the personal pronouns. Furthermore, the plural is only distinguished in the nominative-accusative — except for the personal pronouns, where the difference in number is shown by the stem.
There are two nominal classes, Common Nominals (common nouns, demonstratives, locative/temporal/etc. adverbs) and Proper Nomals (Proper names [personal names, boat names, emotive kinship terms], pronouns). The major difference between the two classes are 1) semantic — Proper nominals have pronominal characteristics, and, 2) declensional, for example Proper Nominals have one locative case rather than the three of Common Nominals.
Note that the following are in the Kalau Kawau Ya dialect.
|Case/Suffix||Hoe/Adze||Place/Home||Knife||Water||Mud||Middle||looking||giving, getting, being, moving, doing, etc.|
-i glide final
|sp-loc||pabunu, pabu'||laganu, laga'||gilai, ginu||wœrai, wœrnu||saithai, saithe||dhadhal, dhadha'||nœgainu||mainu|
|gl-loc||pabuyab||lagayab||gipu||wœrab, wœrpu||saiyab, saipu||dhadhayab||nœgaiya||maiya|
There are few irregular nouns, the most common being:
The language has a closed class of demonstrative morphemes with special morphological characteristics:
The Kauřařaigau Ya forms recorded are the same as in the modern dialects, with the exception of ka-/kařu- 'non-specific here, this', se-/si-/seřu- 'there, that', kařa- 'non-specific yonder', modern dialects kai-, %ka- and -puwai 'ahead there', modern dialects -pai/-pa.
These demonstratives can take masculine, feminine and non-singular morphology (as such are pronominal) as well as case forms. Í- 'here, this' and se/si- 'there, that (not too far away)' take the gender/number morphemes as suffixes, and the other demonstratives take them as prefixes. Note that ka- 'non-specifically here' and kai- 'there in the distance in a non-specific position' cannot appear with the gender/number morphemes, these latter being specific. Í- and se/si- also take an article forming affix -bi to become demonstrative articles (e.g. KLY senuubi kaazi, KKY senaubi kaz 'that boy', KLY senaabi kaazi, KKY senabi kaz 'that girl', KLY sepalab kaazi, KKY sepalbi kaz 'those two children', sethabi kœzil 'those children').
|senau masc, |
|gen||kœu, kœwau||—||seu, sewau||—|
|dat||kœpa, kœwupa||—||sepa/sipa, sewupa||—|
|abl||kœzi, kœwuzi||—||seizi/sizi, sewuzi||—|
|sp-loc||kai, kœwa||in masc,
|senau masc, |
|n-sp-loc||kaiki, kawuki/kœwuki||inuki masc,
|seiki/siki, sewuki||senauki masc, |
|kedhabi||senaubi masc, |
|nom-acc-inst-sp-loc specific[a]||(pi)nugui masc,
|(pi)nupun masc, |
|nom-acc-inst-loc non-specific||kaigui||kaika||kaingap||kaipai/kaipaipa||kaipun, kaipawapa|
|dat specific[a]||(pi)numulupa masc,
|(pi)nupawapa masc, |
|n-sp-loc/gl-loc neutral[a]||(pi)nuguiki masc,
|(pi)nupuniki/(pi)nupawapa masc, |
The personal pronouns three nominative-ergative-accusative in declension. Note that the third person pronouns are also used as definite articles, e.g. Nuidh garkœzin nan yipkaz imadhin 'The man saw the woman'.
|nom||ngai||ngi||nui||na||nga||mi- (miai, miza)|
|acc||ngœna||ngin||nuin||nan||ngan||mi- (miai, miza); |
|inst||ngath||ngidh||nuidh||nadh||ngadh||midh (miaidu/miden/midu/midun, mizœpun)|
|gen||ngau masc, ngœzu fem||nginu||nungu||nanu||ngœnu||mingu (miaingu, mizœngu)|
|dat||ngayapa||ngibepa||nubepa||nabepa||ngabepa||mipa (miaipa, mizœpa)|
|abl||ngaungu(z) masc, ngœzungu(z) fem||nginungu(z)||nungungu(z)||nanungu(z)||ngœnungu(z)||mingu(zi) (miaingu, mizœngu)|
|priv||ngaugi masc, ngœzugi fem||nginugi||nungugi||nanugi||ngœnugi||miaigi, |
|sim||ngaudh masc, ngœzudh fem||nginudh||nungudh||nanudh||ngœnudh||midh (miaidh, mizœpudh)|
The dual and plural pronouns are nominative-accusative, the accusative being the same in form as the genitive, except in KKY, where they are do not mark the nominative or accusative.
|Case/Suffix||we dual||you and I||you dual||them dual
|(as for singular)|
|(as for singular)|
|(as for singular)|
|(as for singular)|
|(as for singular)|
Ngawal 'who dual' is constructed from nga 'who' plus the clitic -wal 'both (dual conjunction)'.
|Case/Suffix||we (exclusive)||we (inclusive)||you||they
|gen||ngœimun||ngalpan||ngithamun||thanamun||(as for singular)|
|dat||ngœimulpa||ngalpalpa||ngithamulpa||thanamulpa||(as for singular)|
|abl||ngœimulngu||ngalpalngu||ngithamulngu||thanamulngu||(as for singular)|
|loc||ngœimuniya||ngalpaniya||ngithamuniya||thanamuniya||(as for singular)|
|sim||ngœimudh||ngalpadh||ngithamudh||thanamudh||(as for singular)|
Ngaya 'who many' is constructed from nga 'who' plus the clitic -ya 'and others (plural conjunction)'.
Familiar kinship terms are the equivalent of English kin terms such as Dad and Mum, while non-familiar terms are the equivalent of Father and Mother; these latter are treated as common nouns in the language.
|Case/Suffix||Tom (mas.)||Anai (fem.)||Dad/Uncle
The earliest grammatical records of the language are those of the mid-1800s Kauřařaigau Ya dialect. This dialect is identical to the modern dialects, apart from retaining more archaic forms of some endings and suffixes.
No early writer recorded declined feminine forms, apart from the genitive. Ray (1907:20-21) implies (by default) that the OKY paradigm is basically the same as that of OKLY.
Brierly (B), MacGillivray (M) and Ray (R) recorded the following forms of the singular pronouns of OKY:
Based on the above forms and the modern dialects, the OKY pronouns are reconstructed as follows:
The accusatives, the ablatives and imitatives underwent optional final vowel deletion, while the ergatives optionally transformed the final u to a or œ, or deleted it, thus ngathu > ngatha > ngathœ > ngath.
The recorded dual-plural forms are:
These can be reconstructed as:
Verbs can have over 100 different aspect, tense, voice, mood and number forms. Verb agreement is with the object (i.e. 'ergative') in transitive clauses, and with the subject in intransitive clauses. Imperatives, on the other hand, agree with both subject and object in transitive clauses.
There are three aspects ('perfective', 'imperfective', 'habitual'), two telicity forms ('active', which focuses on the verb activity and subsumes many intransitives, many antipassives and some transitives, and 'attainative', which subsumes many transitives, some antipassives and some intransitives), two moods ('non-imperative' and 'imperative' [which resembles a subjunctive in some uses]), 6 tenses ('remote future', 'today future', 'present', 'today past', 'recent past', 'remote past' — KLY has developed a 7th tense, a 'last night' tense) and four numbers ('singular', 'dual', 'specific plural', 'animate active plural' — in form the animate active plural is the same as the singular, and is only found on certain verbs).
In most descriptions of the language the active and attainative forms have been mistermed transitive and intransitive respectively. Transitive, intransitive, passive, antipassive and 'antipassive passive' in the language are syntactic categories, and are formed by the interplay of nominal and verbal morphology, clause/sentence-level characteristics such as word-order, and semantic considerations.
Verb morphology consists of prefixes (aspect, positioning, etc.), suffixes (telicity, number, and two fossilised multiplicative/causative suffixes) and endings (tense, aspect and mood, and a very limited extent number and telicity). The structural matrix of theverb is as follows. Note that the two fossilised suffixes are mutually exclusive; if a suffix is in the A slot, a suffix cannot appear in the B slot, and vice versa:
(prefix) + (prefix) + stem (+FOSSILISED SUFFIX A) + (TELICITY) (+FOSSILISED SUFFIX B) + (number) + ending (+ending)
prefix: pa- 'telic prefix'
prefix: bal- 'positional — across'
stem: kabutha- 'place, lay'
telicity suffix: -Ø 'attainative', -i 'active'
number suffix: -ma 'dual' (absolutive agreement)
tense-aspect-mood ending: -dhin 'remote past perfective'
prefix: gar- 'collective'
stem: wœidha- 'place, put'
Fosslised suffix: ma 'intensive
telicity suffix: i 'active'
number suffix: ma 'dual'
tense-aspect-mood ending: dhin 'remote past perfective'
The verb here is íma- 'see, observe, supervise, examine, try, test'
|Case/Suffix||Perfective Attainative||Imperfective Attainative||Perfective Active||Imperfective Active|
|remote future singular||imane||imaipu (imaiparui)||imedhe||imepu (imeparui)|
|dual||imamane||imampu (imamparui)||imemadhe||imempu (imemparui)|
|plural||imamœine||imamœipu (imamœiparui)||imemœidhe||imemœipu (imemœiparui)|
|near future singular||imaipa||imaipu (imaiparui)||imepa||imepu (imeparui)|
|dual||imampa||imampu (imamparui)||imempa||imempu (imemparui)|
|plural||imamœipa||imamœipu (imamœiparui)||imemœipa||imemœipu (imemœiparui)|
|today past singular||imanu||imadha||imema||imedha|
|recent past singular||imangu||imarngu||imaingu||imairngu|
|remote past singular||imadhin||imar||imaidhin||imai|
|plural||imamœidhin||imamœi (imamir)||imemœidhin||imemœi (imemir)|
|Attainative Habitual||imaipu (imaiparui)||imampu (imamparui)||imamœipu (imamœiparui)|
|Active Habitual||imepu (imeparui)||imempu (imemparui)||imemœipu (imemœiparui)|
|Perfective Attainative Imperative (Singular Subject)||imar||imamar||imamœi (imamir)|
|(Non-Plural Subject)||imau (imaziu)||imamariu||imamœi (imamœiziu, imamiu)|
|Active Imperative||imi||imemariu||imemœi (imemœiziu, imemiu)|
|Imperfective Attainative Imperative||imadha||imamadha||imamœidha|
|Imperfective Active Imperative||imedha||imemadha||imemœidha|
|independent impersonal form (nom-acc)||imai||imailnga||imaiginga||imaizinga|
|independent personal form (nom-acc)||—||imailaig||imaigig||imaizig|
These were the same as in the modern dialects.
The only suffix differences with the modern dialects were in the form of the plural and verbal noun suffixes. In OKY these were maři and ři respectively. The dual was ngauma on ma- 'take, give, move etc.' and otherwise uma.
Class 1: wœidha- 'put, place, cook'
Class 2: ni-, niya- 'sit, stay'
|ATTAINATIVE INDICATIVE||perfective||singular perfective active
monosyllabic stem: -iziři
|ATTAINATIVE IMPERATIVE||-r(a) SgS, -u PlS, -riu Dual||-i||-adha|
On the whole, the OKY verb seems to have been declined like the Kalau Lagau Ya verb. This includes the loss of the suffix ma in the intransitive imperfective present/perfective today future singular. This loss, however, appears to have been optional in the today past equivalent:
Vowel/diphthong deletion and reduction in class 1b verbs was optional in OKY where it is now optional or obligatory:
The irregular verb yœwi- / iya- / yœuna- 'lie/slant/lean over/down' was recorded in the form iipa (eepah), indicating the stem ii- (the remote past form iir is found in modern KY, though not recorded in OKY). Otherwise, only yœuna- was recorded for OKY.
Three paradigms that have irregular morphology are:
The Torres Strait Islanders, neighbouring Papuans and neighbouring Australians have a common sign language,, though early records did not make a detailed study of this (e.g. Australian Aboriginal sign languages). Simple conversations and stories can be carried out in the sign language; however, it does not attain the sophistication of a fully developed sign language. It's had some influence on Far North Queensland Indigenous Sign Language.