Journal of Language Contact 4 (2011) 106–140 brill.nl/jlc

Reappraising the Eff ects of Language Contact in the Torres Strait

Jessica Hunter , Claire Bowern , and Erich Round * Yale University

Abstract Th e contact history of the languages of the Eastern and Western Torres Strait has been claimed (e.g. by Dixon 2002, Wurm 1972, and others) to have been suffi ciently intense as to obscure the genetic relationship of the Western Torres Strait language. Some have argued that it is an Australian (Pama-Nyungan) language, though with considerable infl uence from the Papuan language Meryam Mir (the Eastern Torres Strait language). Others have claimed that the Western Torres language is, in fact, a genetically Papuan language, though with substantial Australian substrate or adstrate infl uence. Much has been made of phonological structures which have been viewed as unusual for Australian languages. In this paper we examine the evidence for contact claims in the region. We review aspects of the phonology, morphology, syntax and lexicon of the Eastern and Western Torres Strait languages with an eye to identifying areal infl uence. Th is larger data pool shows that the case for intense contact has been vastly overstated. Beyond some phonological features and some loan words, there is no linguistic evidence for intense contact; moreover, the phonological features adduced to be evidence of contact are also found to be not specifi cally Papuan, but part of a wider set of features in Australian languages.

Keywords Australian languages ; areal diff usion ; Papua New Guinea ; metatypy ; convergence ; grammatical contact

1 Introduction and Background Information

Th e Western Torres Strait language (WTS) and Meryam Mir are spoken on western and eastern islands respectively in the Torres Strait between Cape York, Australia and Papua New Guinea. Th e genetic affi liation of WTS and its degree of linguistic similarity to Meryam Mir, a Papuan language of the Eastern Trans-Fly family, have been contentious. Some, including Capell (1956 ) and Dixon (2002 ), have argued that WTS is genetically Papuan, but

* We thank Rod Mitchell and Barry Alpher for insightful comments; however, all errors of interpretation are our own. Th is work was funded in part by NSF grant BCS-844550.

© Koninklijke Brill NV, Leiden, 2011 DOI 10.1163/187740911X558798

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 107 strongly infl uenced by Australian languages, while others, including Wurm (1972a , 1972b ) and Dixon (1980 ), have argued that WTS is Australian and a member of the Pama-Nyungan family. Alpher et al (2007) show that WTS is Pama-Nyungan, but do not discuss the extent of convergence between WTS and Meryam Mir. Here we examine structural similarities between WTS and Meryam Mir in the areas of phonology, nouns and pronouns, modifi ers, verbs, deixis, and syntax. We argue that the degree of structural similarity between the two languages has been overstated and is not borne out by a systematic comparison of features. Previous investigations of Torres Strait language con- tact have focused heavily on certain areas of Western Torres phonology and extrapolated from it without, it seems, looking into other areas of grammar in detail. A thorough comparison reveals that the case for structural convergence is weak. Section 2 outlines details of the features listed above and includes a brief description of WTS, a description of Meryam Mir, and a comparative evalua- tion of the similarity between the two languages based on the preceding descriptions. Following the investigation of structural similarity, Section 3 is a brief investigation of lexical similarity between the two languages. Section 4 summarizes the fi ndings.

1.1 Th eoretical Context

Contact linguistics combines elements of both historical linguistics and sociolinguistics in order to study the “linguistic consequences” of interaction between two or more languages (Winford 2003 : 9-10). Perhaps the most obvious result of contact between languages is the borrowing of lexical items, but borrowing of non-lexical, structural features also occurs. According to Th omason’s (2001: 70) borrowing scale, the borrowing of content words occurs in situations of casual contact, but as the intensity of contact increases, more and more categories of lexical items can be borrowed and structural borrowing can also occur (see also Th omason and Kaufmann 1988 for detailed discussion). Examples of borrowed structure include the repurposing of native syntactic structures for new functions at the fi rst level of intensity above casual contact, the loss or addition of phonemes based on the inventory of the source language, and a shift in stress placement at the second level above casual contact, and the “loss or addition of agreement patterns” and wide- spread changes in the language’s typology at the most intense level of con- tact, a type of language change termed “metatypy” by Ross (e.g. 1996, 2007). Preceded by both lexical borrowing and grammatical calquing, metatypy arises in a situation of bilingualism when morphosyntactic constructions

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

108 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 are “restructured” such that they imitate constructions modeled in the source language, resulting in a matching of constructions in “both meaning and mor- phosyntax” (Ross 2007 : 124).1 Language contact has long been a focus of Australian historical work (see Bowern 2011 and the references therein). For example, Dixon ( 2002 : 22-27) attributes many of the characteristics typical of Australian languages to a process of structural diff usion for languages “where a signifi cant proportion of the speakers of one also have some competence in the other.” “Phonetic and pho- nological” features, “grammatical categories, construction types and techniques,” “grammatical forms” (such as pronouns), and lexical items can all diff use.

1.2 Methodology of the current study

In this article, the phonology, nouns and pronouns, modifi ers, verbal systems, and deixis of WTS and Meryam Mir are investigated and compared in an attempt to survey the complete language systems, rather than pointing out a few similar or dissimilar features arbitrarily. Th at is, we examine grammatical features of the languages as a whole for an overall picture of degree of similar- ity. Shared features must be found in both languages in order to demonstrate contact-induced change. Th at is, we maintain that it is not suffi cient to claim that WTS is “un-Australian” or “Papuan” without describing the areas of the language which are claimed to be the result of contact. Discussion here focuses in particular on the claim that various structural and phonological features in WTS are ‘un-Australian’ or ‘Papuan-infl uenced’, and specifi cally whether the infl uence comes from Meryam Mir. We recognize that this approach has some drawbacks. For example, in con- fi ning the discussion to whether the features in question are similar or not, we rule out identifi cation of features which have obliquely diff used, or which have subsequently undergone change such that source and result are now dis- tinct. Th is is of necessity; if two features are not similar, there are many reasons why they might diff er, and in most cases identifi cation of the cause of diff er- ence would be impossible (at least at this stage), or would draw us into specu- lative histories well beyond the immediate scope of this paper. We also limit discussion here to WTS and Meryam Mir, not because we wish to imply that these are the only two languages involved in potential regional contact, but

1 We recognize that there are cases of language contact that, beyond metatypy, result in the borrowing of structural features in a process which may come to render the source and result dissimilar. However, identifi cation of such features requires a solid foundation of knowledge of the languages under consideration.

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 109 because previous debate in the literature has focused specifi cally on these lan- guages. We are concerned here primarily with the reevaluation of the claims of Wurm, Dixon, Capell, and others, that WTS is unclassifi able because of con- tact-induced change with Meryam Mir .

1.3 Why WTS and Meryam Mir?

WTS a member of the Pama-Nyungan language family, while Meryam Mir is uncontroversially classifi ed as an Eastern Trans-Fly Papuan language. However, the classifi cation of WTS has been the subject of debate. Capell (1956 : 89) considers Meryam Mir to be a Papuan language, but classifi es WTS as “Australian-infl uenced Papuan” rather than an Australian language, on the basis of physical resemblances between Islanders and “Papuan easterners” and on facts about WTS’s phonemic inventory, namely its voicing contrast in its stops and its possession of the fricatives /s / and /z/, which he considers proto- typically non-Australian. (Th at is, two of the features are phonological and one is not linguistic.) Dixon ( 2002 : 130) agrees with Capell’s assessment and also considers WTS to be “Australian-infl uenced Papuan” with an “Australian sub- stratum.” Th is is a reversal of Dixon’s (1980: 234) standpoint, in which he clas- sifi ed WTS as Australian, because “although it has some unusual characteristics, there are enough correspondences with mainland [Australian] languages to leave no doubt concerning the genetic connection.” Here, Dixon notes that WTS and Meryam Mir share some similarity in terms of phonological inven- tory and lexical items, but have “few similarities at the grammatical level.” Wurm ( 1972a : 151) classifi es WTS as Australian, but notes that it has “a largely un-Australian phonological system” and “has been lexically infl u- enced by the neighboring Miriam language [Meryam Mir],” which he consid- ers Papuan. Wurm ( 1972b : 349) concludes that there is strong evidence of language contact but confi nes discussion of evidence to phonology and a brief review of lexical material. Wurm further states that “Australian infl uence on Miriam is negligible in contrast to the quite strong Papuan infl uence in Mabuiag [WTS]”. Alpher et al. (2008 ) also provide evidence that WTS is Pama-Nyungan; their argument is based on Pama-Nyungan retentions of lexicon and mor- phology. Alpher et al. (2008 : 15) note that WTS has often been thought to have “heavy linguistic infl uence from Meryam,” but they call for a more serious investigation of the facts before such a claim is perpetuated: Over the last hundred years, the literature has suggested several classifi cations of the WTS language: ‘Papuan’, ‘Australian’, ‘Australo-Papuan’ and ‘Papuo-Australian’. Each of these classifi cations implies a diff erent history for the language speakers,

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

110 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140

and each is in principle demonstrable or falsifi able through detailed linguistic comparison. To make claims of mixed language descent and substrate infl uence without the detailed work required, however, trivializes the notion of comparative linguistics. It is the purpose of this paper to provide such a “detailed linguistic compari- son” of WTS and Meryam Mir. In order to limit the scope of the paper, we focus specifi cally on claims about the contact status of these two languages (see also below for evidence of population contact in the region). Other contact scenarios are of course possible (such as the participation of both languages in a wider Torres Sprachbund), but since previous literature has focused on this one, and because the previous claims have been so categorical, we focus here on these languages.

1.4 Torres Strait Sociolinguistic Situation

WTS is spoken on the islands in the western and central portion of the Torres Strait, including Saibai Island, Mabuiag Island, and Th ursday Island, as well as on mainland Australia in Bamaga on Cape York. WTS is a name that encompasses four dialects, including Kala Lagaw Ya (KLY) and Kalaw Kawaw Ya (KKY), meaning ‘Western Island Language’ or ‘Back Island Language.’ Ray and Haddon ( 1893 ) also list subdialects. Th ere are approximately 3500 – 4000 speakers of the language as cited by Ford and Ober ( 1991 : 118). Th ere are also speakers in major north Queensland towns (such as Townsville and Cairns) and in Brisbane. Meryam Mir was traditionally spoken on the islands of Erub, Ugar, and Mer (Murray Island), in the eastern Torres Strait, although at the time that Piper ( 1989 : 1) did her fi eldwork on Meryam Mir in the 1980s, it “had been seriously impinged upon” by Torres Strait Creole. In 1977, there were only 700 speakers of Meryam Mir, including speakers living on Cape York. Figure 1 shows the islands mentioned above in relation to Cape York, Australia, and the island of New Guinea. Th e Strait spans a north-south distance of about 150 kilometers. In addition to WTS and Meryam Mir, two other languages are spoken in the Torres Strait. Th ey are English, the language of education and administra- tion, and, as mentioned above, Torres Strait Creole (TSC), an English-based creole which serves as a regional lingua franca (Piper 1989 : 1, 5). From the 1770s onward, there is British documentation of intermittent shipwrecks and interaction with Islanders, but permanent contact between Islanders and Europeans was not established until 1871 with the arrival of the London Missionary Society in the Strait (Ray 1907 : 1-3). Writing in 1907 of his

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 111

Papua New Guinea 50 km Daudai

Eastern Trans-Fly

Saibai Ugar Western Erub Torres Strait Meryam Mabuiag Mir Murray Island (Mer) Badu Island Moa Island

Horn Island Papua Thursday New Island Yatay Guinea Torres Strait Pantyinamu Yantaykenu Bamaga Cairns Townsville Utudhanamu Northern Yaraytyana Territory Atampaya Cape York Queensland Yadhaykenu Peninsula Brisbane Angkamuthi Wudhadhi Northern Paman Uradhi Languages Victoria

Figure 1: Map of the Torres Strait and neighboring Paman languages (after Crowley 1983)

experiences with the Cambridge Anthropological Expedition of 1898, Sidney H. Ray, tasked with documenting the structures of the WTS and Meryam Mir languages, noted that “nearly all of [the Islanders had] more or less acquaintance with English” (5). Historically, before contact with English-speaking Europeans and before the development of TSC, bilingualism between WTS and Meryam Mir was not widespread. Shnukal ( 1995 : 123) paints a sociolinguistic picture in which Western Islanders and Eastern Islanders conceived of themselves as diff erent groups and where contact “appears to have been quite circumscribed.” Bilingualism was limited to a few select males from each group who were

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

112 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 responsible for conducting “reciprocal exchange between the east and west.” Th ere is some direct evidence for culture contact, however. Ray (1907 :50-51) gives the illustration of the spread of ceremonial kamut and kolap songs which came from the Western Islands into Meryam Mir. MacGillivray (1847) provides a description of the contact situation in the 19th Century between mainland Cape York Peninsula and the Torres Strait Islands. Ray and Haddon ( 1893 :506-508) give trade routes and some diff used trade vocabulary. Shnukal’s and Ray’s views are diffi cult to reconcile, though Shnukal refers specifi cally to widespread multilingualism, while the earlier sources concern culture contact more generally.

1.5 Sources and Source Limitations

Th e main sources used here are sketch grammars of WTS and Meryam Mir. For WTS, Ford and Ober (1991 ) and Bani and Klokeid (1978) were the main sources, with supplementary material provided by Kennedy (1981 , 1985a , 1985b , nd) and Alpher et al. ( 2008 ). Ford and Ober and Kennedy’s descrip- tions are based on the Kalaw Kawaw Ya dialect of WTS, while Bani and Klokeid is a description of Kala Lagaw Ya. We use the general term WTS here but diff erentiate as to dialect where relevant. Kennedy (nd), a dictionary of Kalaw Kawaw Ya, was also useful. For Meryam Mir, Piper (1989 ) was used almost exclusively, though with reference to Ray (1907 :51ff ). Ray (1907 ) treats his Meryam Mir materials as highly tentative, and has overlooked a number of complexities in morphology which makes his analysis fl awed in places. However, on the substantive points of constituency, phonology, and case alignment, the sources are in agreement.2 Th ere is a considerable body of nineteenth century work on both Torres Strait languages, including Ray ( 1907 ) and Ray and Haddon ( 1893 , 1894); lexical and dialectal material can be found in King (1837), von der Schulenburg (1891), MacGillivray (1852 ), and Murray (1876). Th ese represent, to our knowledge, all the main published sources on the languages. Brief comparison is also made to the northernmost Paman languages spoken on the tip of Cape York Peninsula (e.g. Crowley 1981 , Sutton 1978), but since the research question we ask concerns ‘Papuan’ vs ‘Australian’

2 A further potential source is Mitchell (1995), which is based on extensive recent fi eldwork. However, we received a copy of this work too late to be able to refer to his insights systemati- cally. His description diff ers in numerous small respects from the other sources. None of these diff erences would appear to materially aff ect the analysis presented here.

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 113 characterization and the shared features between ETS and WTS, we do not systematically compare Uradhi or other northern Paman languages here.

2 Descriptions and Comparisons of Languages

2.1 Phonology

2.1.1 WTS Phonology Th e phonology of KKY is discussed in Kennedy ( 1981 ) and Ford and Ober ( 1991 ), and is mentioned in Kennedy ( 1985a , b ) and Alpher et al. ( 2008 ). Bani and Klokeid (1971) do not discuss phonology, but a good deal can be inferred from their orthography and primary data. Th e consonant inventory of WTS is shown in Table 2. Stops are distin- guished at four places of articulation, fricatives at one, and nasals at three; there are two liquids (one rhotic and one lateral) and two semi-vowels. Th e stop and fricative phonemes contrast voiced and voiceless. Although / s, z / are classed as ‘fricative’ phonemes, / s / is reported by Kennedy ( 1981 ) as having an aff ricate allophone [ts] word initially, in free variation with [s]. Both Kennedy and Ford and Ober (1991 ) report an aff ricate variant of /z/. Ford and Ober (1991 :120) transcribe the latter as [ʥ], raising the question of whether /s, z/ might be better classed as alveo-palatal rather than alveolar. 3 , 4 Th e rhotic / r / is reported by Kennedy ( 1981 ) as having a trilled and non-trilled allophone, although whether the latter is a tap, or a rhotic approximant is not stated; Alpher et al. ( 2008 ) show / r / refl ecting both */r / and */ɻ / in Pama-Nyungan cognates.

Table 2 WTS Consonant Inventory, after Kennedy ( 1981 ) Bilabial Laminodental Alveolar Alveo-palatal Velar Stops Voiceless p th [t]t̪ k Voiced b dh [d]d̪ g Fricative Voiceless s Voiced z Nasal m n ŋ Liquid l, r Glide w j

3 Th is analysis is implicit in Alpher et al’s (2008:23) arrangement of WTS consonants. 4 Kennedy (1981 :§3.2.2) describes /t, d, n/ as having a ‘pronounced apico articulation’, which seems to us to be unlikely to be compatible with the production of a sibilant fricative.

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

114 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140

Table 3 WTS Vowel Inventory i [i] œ [ɨ, ə] u [u] e [ɛ] a [a, ɐ] 6 o [ɔ]

WTS distinguishes six vowel qualities, shown orthographically and pho- netically in Table 3 . KKY appears not to distinguish length.5 KLY does dis- tinguish length, with examples in Bani and Klokeid (1971) containing instances of long / iː, eː, aː, uː/ . Th e phoneme written ‘œ’ (or ‘oe’) is characterised as a high central vowel [ɨ] by Ford and Ober ( 1991 ), and is classed as a phoneme / ə / by Kennedy ( 1981 ), though with two allophones [ɨ,ə]. Morphophonemically, ‘œ’ often alternates with /a/, with Kennedy (1985:99) reporting ‘a great deal of idiolectal varia- tion’ between the two vowels in verb paradigms. Alpher et al. (2008 :26–7) identify several instances of pPN *a, *aː > WTS ‘œ’. Th e mid, non-central vowels /e / and /o / are less common than other vowels, occurring in just fi ve and two words respectively in a 100 word, KKY vocabulary list (Ford and Ober 1991 :139–140), whereas the other four vowels occur in at least seven- teen words each within the same list. Mitchell (1995) provides more informa- tion about these alternations. WTS syllables may contain rhymes with two (semi-)vocalic qualities, the second of which is [u/w] or [i/j]. Sources diff er in their analysis of these vocoids as phonological vowels or as glide consonants, and in their ortho- graphic rendering of them as ‘u, i’ or ‘w, y’. In the following, we refer to them for convenience as glides. Th e facts pertaining to syllable contact clusters in WTS are not entirely clear. Intra-morphemically there are very few cluster types that occur frequently, but whether this extends to outright bans on other types is unclear. Ford and Ober (1991 :120) state that KKY only permits clusters of {liquid or glide } + { obstruent or nasal } or liquid+glide , though presumably this statement refers only to intra-morphemic clusters given that clusters such as /bg/, / dg/,̪ / nk / and / mp / occur across morpheme boundaries in their own data, as in (1).7

5 Th ough see Alpher et al. 2008 :23 for some doubts. Kennedy (nd) states that length is not phonemic, but nonetheless provides a number of minimal pairs which diff er in vowel length and/or quality. 6 Kennedy ( 1981 ) uses [ʌ] for a mid-low central vowel, corresponding to the modern IPA symbol [ɐ] used here. 7 Kennedy reports intra-morphemic clusters / pk, pm, dm, dp, mp, dz / but does not give examples.

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 115

(1) Evidently polymorphemic words from Ford and Ober ( 1991 :140) a. maludh-goemul b. bamidh-goemul c. goerab-goemul ‘green’ ‘yellow’ ‘white’ It is clear that the overwhelming majority of intra-morphemic clusters in both KKY and KLY are as Ford and Ober ( 1991 ) stated for KKY. Interest- ingly, clusters of nasal+obstruent, which are very common in Australian languages, do not rate among these. Alpher et al. ( 2008 :26) adduce several instances in which a historical nasal+stop cluster has become a voiced obstruent in WTS. Also worth noting is the apparent absence of any clus- ters, even inter-morphemically, which end in a glide. Clusters ending in a liquid are also either rare or absent in KKY, though a / kr / cluster is attested in the frequentative aspect suffi x 8 -aikruig , -ekruig in KLY (Bani and Klokeid 1971:21). It is not uncommon for stems in WTS to exhibit morphophonological complexities, such as a lexically determined ‘epenthetic’ or ‘augment’ strings which appear between the fi nal segment of a stem and a following suffi x, as in (2, 3).

(2) Vocalic epenthetic augments (KKY, Ford and Ober 1991 :121)9 a. awn ‘stingray’awn-a-n ‘stingray- aug-erg ’ b. yoepkaz ‘woman’yoepkoez-i-n ‘woman- aug-erg ’ c. thurik ‘axe’ thurik-u-n ‘axe-aug-erg ’ (3) Other epenthetic augments (KKY, Ford and Ober 1991 :121, 129) a. a ‘great-great-grandparent’a-l-pu ‘g.g.g’parent- aug-dat ’ b. nga ‘who’nga-be-pa ‘who-aug-dat ’ (KLY, Bani and Klokeid 1971:2) c. zaa ‘thing’zapun ‘thing.erg’ d. ai ‘food’aidun ‘food.erg ’ Native words in WTS may begin with any consonant, and with any of /i, e, a, u / (though not / o, ɨ/). Th ey may end in any consonant or in any vowel other than /ɨ /; only one word-fi nal cluster is attested, /rs/. Primary stress in WTS falls on the fi rst syllable, with secondary stress on alternate syllables thereafter, although secondary stress tends to skip over the ‘epenthetic’ vowels mentioned above (Kennedy 1981 :§8.1).

8 KKY lacks a cognate suffi x. 9 Epenthetic vowels of kind are not found in KKY. Th eir absence results in there being signifi cantly more inter-morphemic clusters in KLY than in KKY: compare ‘caught (distant past)’ in KLY / gasam-din̪ / with KKY / gasam-a-din/.̪

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

116 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140

2.1.2 Meryam Mir Phonology Th e consonant inventory of Meryam Mir is shown in Table 4 , following Piper (1989 ). Phonemes which are absent in Meryam Mir but found in WTS are marked with a dash. Stops are distinguished at just three places of articulation, fricatives at one, and nasals at just two; there are two liquids — one rhotic and one lateral — and two semi-vowels. Th e stop and fricative phonemes contrast voiced and voiceless. Von der Schulenburg (1891), working from a translation of Mark’s and John’s gospels held at the time by the British Foreign Bible Society, adds that certain foreign words contain / h / and / f/; he also lists a phoneme / θ / which, while not singled out as foreign, appears in his data only in the foreign word /θri / ‘three’. Haddon (1907:2,52) also notes that some words were recorded by King (1837) with /ŋ / and / f/, but with / g / and / p / by him. Unlike in WTS, there is no indication that /s, z/ in Meryam Mir possess aff ricate allophones. Th e rhotic / r / is reported to be a tap in most positions, but either a tap or approximant when followed immediately by a consonant.10 According to Piper (1989 ) Meryam Mir distinguishes fi ve vowel qualities, shown orthographically and phonetically in Table 5 . Vowel length is noncon- trastive and conditioned by stress.

Table 4 Meryam Mir Consonant Inventory, after Piper ( 1989 ) Bilabial Interdental Alveolar Alveo-palatal Velar Stops Voiceless p – t k Voiced b – d g Fricative Voiceless s voiced z Nasal m n – Liquid l, r Glide w j

Table 5 Meryam Mir Vowel Inventory, after Piper ( 1989 ) i [ɪ] u [ʊ] e [e]̞ a [a] o [ɔ]

1 0 Piper ( 1989 ) lists a trill allophone in her allophonic rules (p.8), but later states (p.9) that this ‘may be erroneous because the rhotic is never really trilled.’

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 117

In a minority of words there is free variation between /e / and /i / (Piper 1989 :15–16). Von der Schulenburg’s (1891) sources distinguish six ortho- graphic vowels: ‘i, e, ë, a, o, u’. Th e vowels ‘e’ and ‘ë’ are distinguished consis- tently throughout his corpus; both of them correspond to /e / in Piper (1989 ), as in (4). Ray ( 1907 :52) lists seven monophthongs and fi ve diphthongs.

(4) v.d.Schulenburg (1891:8,11,22,34) Piper ( 1989 :16,38,80,163) a. kekëm ‘zuerst (initially)’kikém ‘initially, fi rst’ b. nerkëp ‘innere (interior)’nerkep ‘heart’ c. ikëdi ‘legen (put down)’ikedi ‘put down’ d. ikëli ‘thun (do)’ekeli ‘do’ Th e syllable structure of Meryam Mir is (C)(G)V(G)(C), where G is a glide and C a non-glide consonant. Th ere appear to be few absolute restrictions on clusters formed across syllable boundaries, though certain types do fail to appear intra-morphemically in Piper’s data. Cluster types which are freely permitted and amply attested intra-morphemically are {obstruent or nasal }+ liquid (almost entirely absent in WTS), obstruent+{stop or nasal} (absent intra-morphemically in WTS), liquid +{ obstruent or nasal } (com- mon in WTS, although unlike in WTS lateral-initial clusters in Meryam Mir appear to be particularly rare (Piper 1989 :24,26)). Most alveolar+glide clusters are prohibited in Meryam Mir, specifi cally {/n, d, l/}+/w/ and {/t, d, l/}+/j / (whereas WTS freely admits / lw / and / lj / clusters). Stress, or ‘accent’ (Piper 1989 :16) is a lexically contrastive property of mor- phological roots. In a word, accent falls on either the fi rst or second syllable, and thus it may shift as prefi xes are added. Th e accent need not fall on the root itself: even an unprefi xed, monosyllabic root is able to select for accent that falls on its suffi x. Words in Meryam Mir may begin with any vowel, and with any consonant besides / r/; they may end with any consonant or vowel. Initial and fi nal con- sonant clusters are restricted only to the extent provided by the general pho- notactic considerations outlined above.

2.1.3 Evaluation of Similarity – Phonology Alpher et al. (2008 : 23) note that WTS is “somewhat unusual for a Pama- Nyungan language” because it lacks retrofl ection and has a voicing contrast in its stops and fricatives. WTS also lacks the two rhotics found in most Australian languages, and unlike most Australian languages does not have a nasal phoneme at the same place of articulation as all of its stops. When these observations are juxtaposed with our knowledge of the Meryam Mir consonant inventory, there is an impetus to attribute the typologically unusual

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

118 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 features of WTS (as seen from an Australian standpoint), to infl uence from Meryam Mir. We consider these issues now in turn. Most Pama-Nyungan languages contrast two apical places of articulation: apical alveolar and apical post-avleolar (retrofl ex). However, within a geo- graphical band around 500km wide running up most of the east coast of Australia, languages fail to exhibit that contrast. Th erefore lack of retrofl ection is a feature of the Australian languages of that area, and need not be ascribed to contact with Meryam Mir per se . Approximately 30% of Australian languages exhibit a phonological contrast between two stop series (whether in voicing, length or fortition). Austin ( 1988 : 4) characterises WTS as having “the best established voicing contrast and the widest range of contrasting environments of any Australian language,” yet goes on to write that it is “unclear” if these facts are due to infl uence from Papuan languages. In fact Cape York, immediately to the south of the Torres Strait and WTS, is a hot spot in Australia for phonological voicing contrasts. Austin (1988 : 4–6, 21) cites Mbiywom, Wik Muminh, and Umbuygamu, all Paman languages on Cape York, as languages with voicing contrasts. Much the same argument applies to WTS’s fricatives. Sibilant fricatives are exceedingly rare in Australian languages, being restricted to a handful languages of the Daly family in the Northern Territory and to languages in Cape York. In Cape York, Morrobolama/Umbuygamu (Ogilvie 1994 ), Mpakwithi Anguthimri (Crowley 1981 ) and Linngithigh (Hale 1997 ) are all documented as possessing phonemes with sibilant allophones. Among the six fricative phonemes of Morrobolama is a voiceless, apico-postalveolar sibi- lant /ʃ/. Mpakwithi contrasts a voiced palatal sibilant /ʒ /, while its voiceless stop /ƫ / is optionally realised by a voiceless sibilant allophone [ʃ] preceding continuant consonants. In Linngithigh the obstruent phoneme labeled /c / takes as its primary allophone a lamino-alveopalatal sibilant ‘[sy ]’, “though the stop variant (somewhat aspirated) is also heard” (Hale 1997 :209–10). Linngithigh /c / thus possesses an allomorphy remarkably close to that of WTS /s / (which is [s ~ ts]), while Mpakwithi contrasts two phonemes with sibilant allophones, one voiced and one voiceless. Th us, while voicing contrasts and sibilant fricatives may be considered rare in the overall context of Australian languages, neither is quite so rare in the context of the languages of Cape York.11 Moreover, the sibilants of WTS have

1 1 Note that since we do not have a detailed history of Pama-Nyungan internal subgroup- ing, voicing could be a shared innovation between Paman and Western Torres languages. It could also be a Paman-West Torres contact feature. It could be an independent (but parallel) innovation in both groups. Th ere are irregularities in descent of the voiced and voiceless series

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 119 aff ricate allophones. Allophony between stops and sibilants is attested in other Cape York languages, and it is not found in Meryam Mir, which if anything paints the WTS sibilant phonemes as closer to those in certain nearby Australian languages than to those in Meryam Mir. WTS has only one rhotic phoneme /r/. Dixon (2001 : 67) cites approxi- mately 85% of Australian languages as having two rhotics. Yet in Hale’s (1976) survey of ten northern Cape York languages, Luthigh and Mpalityanh have only one rhotic, and while the other eight languages do contrast two rhotics, one of these is often highly restricted in its distribution. In Ngkoth for example the contrast is neutralised word-medially and in Aritinngithigh, only one of the two rhotics is attested in refl exes of Hale’s proto Paman lexemes. According to Dixon (2001 : 67), 98% of Australian languages possess a nasal phoneme at each place of articulation (other than glottal) occupied by a stop. Just as in Meryam Mir, this parallelism fails to obtain in WTS, although the missing nasal phoneme is diff erent in each case — */n/ ̪ in WTS (which like all interdentals is absent from Meryam Mir) and */ŋ / in Meryam Mir (which does occur in WTS).12 WTS has a six-vowel system, though morphophonemic alternations suggest that / ɨ/ (orthographic ‘œ’) may have arisen primarily via a split of an earlier phoneme */a/. Th is would seem to bring the WTS vowel inventory closer to that of Meryam Mir, especially in light of the fact that Australian languages typically have a three-vowel system, consisting of / i/, / a/, and / u/. Dixon ( 2001 : 67) gives the rate of three-vowel systems in Australia at about 67%. Moreover, the WTS mid vowels / e, o / (and / o / in particular) are relatively uncommon. Th is situation is certainly consistent with a hypothesis of infl u- ence in WTS from Meryam Mir. However in the Cape York context the vowel inventory of modern WTS is not unusual, but rather highly typical. In addi- tion to /i, a, u/ all ten northern Cape York languages surveyed in Hale (1976 ) possess a non-high front vowel (compare WTS / e/), and four of the ten con- trast a non-high back vowel (compare WTS / o/); three contrast other qualities in addition. WTS and Meryam Mir diff er rather starkly in the intra-morphemic consonant clusters they allow. Meryam Mir freely permits { obstruent or

(that is, some apparent refl exes of Pama-Nyungan cognates appear with voiced stops, while oth- ers appear with voiceless ones – see Alpher et al (2008 ) for details) which imply that language contact of some type is likely to be involved. 12 In Cape York, Hale ( 1976 :16) hints that Awngthim may lack a / ny / phoneme correspond- ing to the stop /ƫ /. Conversely though, Aritinngithigh (Hale 1976 :20) distinguishes / ny / despite lacking a parallel stop */ƫ/.

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

120 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140

nasal }+ liquid and obstruent+{stop or nasal} clusters, all of which are absent or nearly so in WTS. Meryam Mir prohibits many alveolar+glide clusters which are per mitted in WTS (though subject to other, cross-cutting constraints). On the other hand, both languages permit clusters of {liquid or glide } + {obstruent or nasal } and liquid+glide. To a large extent, the fact that both languages share these clusters is unremarkable, since almost all Australian languages possess liquid +{ obstruent or nasal or glide} clusters ( Hamilton 1996 ). However, this still leaves the question of WTS clusters of glide +{ obstruent or nasal }. Glide +{ obstruent or nasal } clusters are rare in Australian languages, though once again Cape York is something of a special case. Clusters of /j/+{ obstruent or nasal } are relatively common among Paman languages, including those of Cape York (e.g. Hamilton 1996 :Appendix B contains 10 such languages 13 ). Th is leaves clusters of /w/+{ obstruent or nasal}, which both WTS and Meryam Mir permit. Th ese clusters are exceedingly rare in Australian languages and even though they do appear in the Cape York languages Alngith and Awngthim (Hale 1976 ), the latter languages also allow {obstruent or nasal }+/w / clusters which WTS does not. It is there- fore entirely plausible that WTS has developed / w/+{ obstruent or nasal } clusters — perhaps elaborating upon pre-existing / j/+{ obstruent or nasal } clusters — following contact with Meryam Mir. Th e stress/accent systems of the two languages are unalike, with regular word-initial stress in WTS but lexically determined stress in Meryam Mir. In sum we fi nd very little in the WTS phonological system which from an Australian typological perspective is both ‘unusual’ and which calls for an interpretation in terms of language contact with Meryam Mir. With the one possible exception of / w/+{ obstruent or nasal } clusters, all of what is ‘unusual’ about WTS phonology when taken within a wider Australian per- spective is in fact unremarkable within the context of Cape York and Paman. We do not rule out the possibility that WTS and northern Paman languages exhibit these phonological characteristics ultimately because of contact with languages from a Papuan family. Demonstrating that is beyond the scope of this paper. Our point is simply that the phonological system of WTS does not provide incontrovertible evidence for contact specifi cally with Meryam Mir, and that contact between these two languages is neither the only, nor the best, explanation for WTS phonology.

1 3 Th e languages are Djabugay, Dyirbal, Gaalpu, Guugu-Yimidhirr, Kuku-Yalanyji, Mbabaram, Nyawaygi, Olkol, Uradhi and Yidiny.

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 121

2.2 Nouns and Pronouns

2.2.1 WTS Nouns and Pronouns WTS nouns exhibit a split case coding system, where a distinction is made between the core cases applied to common nouns and to proper nouns (Comrie, 1982 ). Proper nouns take nominative and accusative (-na ) case, while common nouns take ergative (-an, -in ) and absolutive case. Oblique cases are possessive (-w ), instrumental (-an, -in), dative (-ka (KKY), -pa (KLY) ), ablative (- ŋu), locative (-nu ), and comitative (-ya ) cases; it is, however, diffi cult to get a true sense of each case’s full semantic range. Th e suffi x which marks the ergative core case and that which marks the instrumental oblique case are the same in WTS, a phenomenon that Dixon ( 2001 : 68) cites in about 85% of Australian languages. Nouns distinguish between singular and plural num- ber through the use of a plural suffi x affi xed to the singular form. Nouns are marked for either number or case, but not both.14 WTS pronouns take ergative, nominative, and accusative as their core cases and distinguish between singular, dual, and plural number. Th ere is a gender distinction in third person pronouns with Ford and Ober ( 1991 : 121-122) and Bani ( 1987 ) marking the distinction as masculine versus feminine (Alpher et al. ’s [2008: 17] feminine versus non-feminine). Gender marking is partly determined by sex, but also by factors such as specifi city and number (Bani 1987 ). Th ere are also morphological declension classes (Bani and Klokeid 1971). WTS has a mixed case system, in which proper nouns follow a nominative- accusative pattern, common nouns follow an ergative-absolutive pattern, and pronouns mark three diff erent cases for three diff erent syntactic functions: the subject of a transitive verb, the object of a transitive verb, and the subject of an intransitive verb (as illustrated in Table 4 ).

Table 6 Summary of Case Coding in WTS Pronouns and Nouns. Pronouns Common Nouns Proper Nouns A X X X S Y Y X O Z Y Y

1 4 An alternative way to state this is that a morphological distinction between singular and plural is neutralized in non-core cases. Alternatively, only arguments are diff erentiated by number.

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

122 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140

2.2.2 Meryam Mir Nouns and Pronouns Meryam Mir also has a split case coding system between its common and proper nouns. Common nouns take ergative (-(i)de, -gize ) and absolutive case, while proper nouns take ergative (-(i)de ), nominative, and accusative (-i ) case. Piper (1989 : 34-42) lists the oblique cases for Meryam Mir as locative (-ge) , instrumental (-u ), allative (-em, im ), ablative (-lam ), genitive (-(i)ra, -i ), and associative (-kem ). She describes the case functions as follows: the locative case marks the place of an event, the end point of a motion verb, specifi c points in time, or the physical or emotional state of an experiencer, and also has a sec- ondary comitative meaning when used with animate nouns. Th e instrumental marks the nouns which are the means with which an action is performed. Th e allative typically marks the goal or place of a motion, but also can have a pur- posive meaning, and mark an unrealized end point of a state such as dying, or an unreached point in time. Th e ablative marks motion away from a place or time, and also has a causal meaning. Th e genitive marks benefi ciaries of an action as well as the possessor or owner of something. Th e associative marks a person or an object associated with an event, but not a participant or instru- ment in the event. Meryam Mir personal pronouns take nominative and accusative core case. Th ey feature a singular versus nonsingular number distinction, with the fi rst person nonsingular further breaking down into a inclusive versus exclusive distinction. Each combination of number and person is marked by an inde- pendent pronominal form, while oblique cases are formed from a pronoun’s accusative form through the use of suffi xes. Table 7 below shows whether the three syntactic functions, subject of a transitive verb (A), object of a transitive verb (O), and subject of an intransi- tive verb (S), are marked by diff erent morphological case forms or show some syncretic combination within the categories of pronouns, common nouns, and proper nouns in Meryam Mir. Number marking in Meryam is confi ned to a free word giz , which can be used to collectivize a concept. Number is not regularly marked in the Noun Phrase (Piper 1989 : 44).

Table 7 Summary of Case Coding in Meryam Mir Pronouns and Nouns Pronouns Common Nouns Proper Nouns A X X X S X Y Y O Y Y Z

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 123

2.2.3 Evaluation of Similarity – Nouns and Pronouns Both WTS and Meryam Mir have a split case coding system between proper and common nouns, but this is not necessarily a sign of relatedness or contact between the two languages. Split case coding systems are not uncommon in Australian languages (Dixon 1980 : 286-291) and the split between the cases assigned to pronouns, common nouns, and proper nouns is diff erent in the two languages, as seen below in Table 8. WTS and Meryam Mir common nouns both take ergative and absolutive case, but WTS proper nouns distin- guish between nominative and accusative case, while WTS proper nouns make a three-way core case distinction. Conversely, WTS pronouns show tri- partite marking, while Meryam Mir pronouns follow a nominative-accusative pattern. Similarly, both languages have a split number coding system, where the number categories available to pronouns versus those available to nouns diff er. But the way that the languages make the split is diff erent, since WTS nouns have singular and plural number, while Meryam Mir nouns do not distinguish number, and WTS pronouns have singular, dual, and plural number, while Meryam Mir pronouns distinguish between singular and nonsingular, as well as nonsingular inclusive and exclusive in the fi rst person. Th e terms used for the oblique cases of WTS and Meryam Mir are not equivalent. Since the information available on the functions of WTS cases is sparse, an attempt has been made to match them based on function with their equivalent cases in Meryam Mir in Table 9 . Finally, it is important to note that no evidence was found during the course of this comparison to suggest that the morphological forms of one case system or pronoun system have been borrowed into the other language. Th ere is not a single comparable form between the two languages. Th e same applies to pronominal forms. Table 10 shows the diff erence in singular pronoun forms, taken from Ford and Ober ( 1991 : 138) and Piper (1989 : 68). For WTS, the forms are given as they would appear as the subject of an intransitive (intr.) verb (S) and as the subject of a transitive (tr.) verb (A).

Table 8 Summary of Case Coding in Pronouns, Common Nouns, and Proper Nouns Pronouns Common Nouns Proper Nouns WTS Tripartite Erg/Abs Nom/Acc Meryam Mir Nom/Acc Erg/Abs Tripartite

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

124 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140

Table 9 Functions Fulfi lled by Oblique Cases in WTS and Meryam Mir Function WTS Meryam Mir Possession Possessive -au, -iu, Genitive (and Locative -(i)ra, -i (=accusative in -u for a person’s place some classes) where event occurs) Comitative -ya NP at which Dative KLY: Allative -em, -im action is directed -ka KKY: -pa Object used to Ergative/ -an, -in Instrumental (distinct -u complete an Instrumental from ergative) action Motion towards Dative -ka Allative -em, -im Motion from Ablative -ŋ u Ablative -(i)lam Comitative Comitative -ya Associate -(ip)kem nonparticipatory Comitative [not marked by Locative -ge participatory case?] Place or endpoint Locative -nu Locative -ge of action

Table 10 Diff erence in Singular Pronominal Forms in WTS and Meryam Mir WTS Meryam Mir 1sg.ŋ ay (intr.)ka ŋath (tr.) 2sg. ŋi (intr.) KKY ma ni (intr) KLY ŋidh (tr.) KKY nidh (tr.) KLY 3sg. ‘she’ na (intr.) e nadh (tr.) ‘he’ nuy (intr.) nuydh (tr.)

As Meryam Mir pronouns follow a nominative-accusative pattern and do not have grammatical gender, only one form is given. As Evans ( 2005 :255) notes, the WTS pronouns are refl exes of Proto-Pama-Nyungan forms. Th e Meryam forms are clearly cognate with forms in other Eastern Trans-Fly languages

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 125

(cf. Wipi 1sg kon and 2sg man (Dondorp and Shim 1997 ) and Gizrra 1sg ka and 2sg ma (given without source in Evans 2005 )). Number marking is also diff erent between the two languages, with number integrated morphologically with the case system in WTS, but marked with a separate word in Meryam Mir. Meryam has a separate defi niteness marker –et , which WTS lacks as a category (though defi niteness can be signaled through gender marking). Th ere are no similarities indicative of language contact in the nominal and pronominal systems.

2.3 Modifi ers

2.3.1 WTS Modifi ers In WTS, Ford and Ober fi nd “a large class” (p123) of adjectives. Adjectives can be used both attributively and predicatively, though there are some that can be used only as attributes. When used attributively, most WTS adjectives are invariable. When predicative, adjectives “may take one of two series of suffi xes,” each with a singular and plural form. Th e series are -ngal/-mayl and - g /-gal . Ford and Ober (1991 : 124) state that there is not a “systemic [semantic] diff erence” between the series except in nonfi nite relative clauses. In addition to the class of elements that are exclusively adjectives, adjectives can be derived by applying the participial suffi x - zi to verbs. Bani and Klokeid (1971) describe predicate markers of the form –ma and -ŋ a . Ray ( 1907 :13) gives some other “terminatives”, including a privative marker. WTS has a class of indeclinable adverbs that can modify any constituent, including the sentence itself. Adverbs can be derived by applying a vocalic suf- fi x to nouns. Nouns in the locative case function adverbially, but Ford and Ober ( 1991 : 124-125) consider them still to be nominals. Th ey give examples of both temporal and locational nouns being used to derive adverbs or used adverbially, including the adverb dapara ‘in the sky’ derived from dapar ‘sky’ using the vocalic suffi x and the locative form goeyg-i-nu ‘during daylight’ of the noun goeyga ‘day.’

2.3.2 Meryam Mir Modifi ers Th ere is no clear class of adjectives in Meryam Mir. At best, there is a general class of modifi ers which includes number words, quantifi ers, determiners, and four specifi c words, meaning ‘big/old,’ ‘little/young,’ ‘good,’ and ‘bad,’ which Piper tentatively labels adjectives (but which are not syntactically distinct from the other items listed). Piper ( 1989 : 75, 77) further notes that these words can also serve as nouns and that “several items functioning as adjectives are

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

126 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 homophonous in form with nouns.”15 Th ey can be used attributively, in a position preceding the noun which they modify, and predicatively when they “precede the existential/locational verb.” Meryam Mir ‘adjectives’ can also be used to intensify other adjectives; for example, the adjective aw ‘big’ coupled with the intensifi er kaka modifying the adjective kebi ‘little’ means in totality ‘very small’. Nominals can function as modifi ers through means of reduplica- tion. Piper also claims that modifi ers can be derived from nouns through the use of -kem , a suffi x homophonous with the associative case. Piper identifi es two words which are solely adverbs: nab ‘(attempt) unsuc- cessfully’ and mirem ‘try.’ She suggests that there maybe other elements that fi t into this category of adverbs, since they serve only an adverbial function. Meryam Mir ‘adverbs’ are invariable and have no fi xed position in the clause. As with adjectives, elements functioning adverbially can also be derived from nouns through reduplication. Reduplication may also serve a separate intensi- fying function.

2.3.3 Evaluation of Similarity – Modifi ers In WTS there is evidence for a class of adjectives and a class of adverbs, while in Meryam Mir, there are only a few words that can be said to be exclusively adjectives or exclusively adverbs. Th e process by which adjectives and adverbs are derived in these languages also diff ers. While WTS and Meryam Mir both derive adverbs from nouns, WTS does so through use of a suffi x and Meryam Mir through reduplication. In the case of adjectives, WTS derives them from verbs by affi xing a participial suffi x and Meryam Mir uses reduplication or a suffi x homophonous with the associative case to derive them from nouns. WTS also has singular and plural suffi xes for its adjectives, while number is not marked on Meryam Mir adjectives, whose nouns also do not distinguish number. WTS and Meryam Mir modifi ers diff er in the degree to which they constitute distinct classes and in how they are derived. Th ese languages have no particular structural similarities in nominal modifi cation.

2.4 Verbs

2.4.1 WTS Verbs WTS verbs are divided into two conjugation classes based on transitivity. Following an ergative-absolutive pattern, transitive verbs take suffi xes whose

1 5 It is unclear whether the latter statement is a reformulation of the former or is meant to signal something slightly diff erent.

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 127

Table 11 Singular allomorphs of tense markers in KKY and KLY Transitive Intransitive KLY KKY KLY KKY Remote Past - dhin - oedhin - aidhin - aydhin Yesterday past - ŋul -oe ŋu -ai ŋul -ai ŋu Last night - nbuŋel - buŋel Near Past - nu - nu -ma -ma Present -n -an -ø -iz Future -aik -oene -ka -idhe Near Future -aikai -oene + kay -ekai -oene + kay Remote future -aikakai -oene + bal -ekakai -oene + bal Imperfect -adh -oepa -adh -ipa Past Imperfect -oedha -idha Usitative -arŋul -ar ŋu -aiŋul -ay ŋu Frequentative -aikruig -ekruig Potential -e -re Potential-future -dhekai -dhekai Negative -aigiŋa -oeygi ŋa -aigiŋa -oeygi ŋa Imperative -r -ar -ø Negative -aig -ayg -aig -ayg Imperative form is conditioned by the verb’s object, while intransitive verbs take number suffi xes which agree with the subject. WTS verbs do not mark person, though there are some suppletive stems which are conditioned by number, such as ‘go’, which has singular uzaray but plural loedhuy (Kennedy nd). Forms of the suffi xes are given in Table 11 (from Ford and Ober, Kennedy and Bani and Klokeid, using the terminology of Kennedy (nd)).16 While Table 11 gives many forms, not all are used equally frequently. Marking in the WTS tense and aspect system seems to be largely determined by discourse. Th e main distinction made by verbs is between completive and incompletive aspect, which make basic references to the past and to the present and future, respectively, “whenever the time of the action does not

1 6 In the interests of clarity only singular number forms are given. Note also that complex vowels of vowel reduction and epenthesis (which are imperfectly described in the sources) make morphemic segmentation doubtful. For example, it is plausible that the tense suffi xes listed here are not, in fact, vowel-initial; the vowel could be analyzed as either part of the verb root, or as a linker which is added to the stem. Additional complexities in the description are introduced by the terminology used by the various authors in describing WTS verb morphology.

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

128 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 need to be indicated with great precision” (Kennedy 1985b : 84). Th ese mark- ers are labeled here as ‘present (perfect)’ and ‘imperfect’. Th ese markers are extensively used; other markers are rarer. In addition to the completive and incompletive aspect, WTS also has habitual, uninterrupted continuous, ter- minated continuous, and immediate striving aspects and today past, yesterday past, historical past, and future tenses. Any use of these additional aspects or tenses (instead of the completive or incompletive aspects) puts special focus on the manner or time of an action. All three of the past tenses are marked on the verbs through the use of suffi xes, while the future tense is marked with a verbal suffi x and clitics which indicate whether the action will take place later that day or on a future day. Ford and Ober (1991 : 123) call the future con- struction periphrastic because of the appearance of clitics in addition to the infl ected verb, but Kennedy (1985b : 87) lists the words Ford and Ober term clitics as optional in the formation of the future tense. Th e additional aspects are also marked by verbal suffi xes. WTS has indicative and imperative moods. Th e imperative suffi xes mark the number of the subject in the case of intransitive verbs and of both the subject and the object in the case of transitives. Haddon (1907:28ff ) describes a series of compounded verbal items; these appear to use the items which are normally used in Complex Predicate constructions (Bowern 2008, Butt and Geuder 2003), such as body parts. Kennedy (nd) also has extensive numbers of phrasal verbs; an example is sibpoegay “get cramps from overeating”, comprising sib ‘liver’ and poegay ‘pierce’.

2.4.2 Meryam Mir Verbs Th e Meryam Mir verbal system organizes verbs into two types, which Piper terms “Atelic Stative” and “Telic Active”, based on telicity, though the stative verbs are also all intransitive.17 Th e stative verbs take suffi xes that distinguish between present and nonpresent tense, while the telic verbs distinguish between the present imperfective, nonpresent imperfective, and perfective. Future tense is marked by prefi xes and can appear on either type of verb. Th ere is also a remote past in Meryam Mir. Th e irrealis is also marked on both classes of Meryam Mir verbs and can be used to refer to both past events that might

1 7 Th e description of the Meryam verb is here based entirely on Piper’s (1989) analysis. Haddon (1907:65) writes that his analysis of verbal morphology is tentative and incomplete. Von Schulenburg (1891) lists some affi xes but contains numerous fl aws (for example, he con- fl ates the future and 1/2Sg object prefi xes (both of the form na- , but they occur in diff erent positions).

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 129 have, but did not, occur and events whose continuation into the future is uncertain (Piper 1989 : 115, 119). Number is marked in two ways on Meryam Mir verbs. While some verbal roots do not change based on number, there are also verbal roots that have suppletive or morphologically related forms for diff erent numbers (many more than for WTS). According to Foley (1986 : 128), marking number through changes in the form of the verbal root is fairly widespread in Papuan languages. Additionally, Meryam Mir verbs also cross-reference number through the use of affi xes. According to Piper (1989 : 139), “it is the combina- tion of markers and verb roots themselves which achieve specifi c number dis- tinctions.” Meryam Mir cross-references the number of subject and object arguments on verbal stems through the use of both prefi xes and suffi xes according to both telicity and transitivity. Stative verbs cross-reference num- ber and person on the verbal root by means of prefi xes, while telic verbs may use both prefi xes and suffi xes. Th e ordering of morphemes in a fully infl ected verb in Meryam Mir is com- plex. Tense, aspect, and mood morphemes are affi xed in multiple places, including initially and fi nally. Th e tense marking is both bound with and dependent on the subject person. Th e order of the future tense morpheme is dependent on the marker of intransitivity and plural subject of an intransitive. Deletion of the initial CV sequence in a Meryam Mir verbal root is possible; the initial syllable can be replaced by certain tense or agreement markers. Th ere is also an infi x na- which marks certain 3rd person nonsingular intransi- tive subjects or transitive objects. A summary template of Meryam verbal marking is given in Table 12 . Prefi x positions are numbered, following Piper ( 1989 :89). Intransitive verbs can be derived from transitives by prefi xing the intran- sitizer ba - to verbal roots. Th e intransitizer is homophonous with a morpheme that cross references plural subjects of intransitive verbs on the verbal root; it also marks refl exive/reciprocal verbs. Piper ( 1989 : 140) suggests that there is also a “transitiviser morpheme” i- , which is found word-initially in some tran- sitive verb roots.

Table 12 Meryam Mir Verbal Template (adapted from Piper 1989 : 89) 4. 3. 2. 1. TAM- Deixis- Future- [(C)V- 3rd person- rest of root] -TAM Tense- -Agreement Agreement-

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

130 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140

2.4.3 Evaluation of Similarity – Verbs Th e verbal systems of WTS and Meryam Mir are radically diff erent. WTS has an intransitive and a transitive conjugation class, while Meryam Mir’s conju- gation classes are based on telicity. In WTS, intransitivity or transitivity appears to be an inherent feature of verbs, while in Meryam Mir, Piper claims that intransitive and transitive verbs can be derived from each other using specifi c markers on the verbal root.18 WTS verbal roots are infl ected only through the use of suffi xes, while Meryam Mir has prefi xes, suffi xes, and infi xes to its verbal root. Th is is mark- edly diff erent from both WTS specifi cally and Australian languages broadly, since no Pama-Nyungan language is known to have infi xes. WTS has a straightforward verbal template comprising a root, number suffi x and TAM suffi x (with some fusion between number and TAM). In Meryam Mir, there are three prefi xal positions (two marking tense and one deixis); agreement and further tense marking is either infi xed to the verb root or fused with it (e.g. replacing the initial syllable of the verb root). Th ere is an additional tense, person and number portmanteau suffi x, with forms varying according to con- jugation. Examples (5) - (7) are illustrative of the diff erence in verb marking; the fi rst two are from WTS, while the last is from Meryam Mir.

(5) Th ana pathe-mi-nu. Th ey-PL embark-PL-TP ‘Th ey left by vehicle (earlier today)’ (Kennedy1985b:83-84)

(6) Nuy-dh wapi lumay-pa. He-ERG fi sh search-INC ‘He is searching/will search/searched for the fi sh.’ (depending on the context) (Kennedy1985b:83-84)

(7) wi- wa- ta- bakyamu- lam 3n SgS Fut3 Deix Sg/DualS go- Fut2/3Dual ‘Th ey (2) will be coming.’ (Piper 1989 :89)

One similarity between the two languages is that both follow an ergative- absolutive pattern when marking number on their verbs. Th e WTS forms cannot be attributed to language contact, however, since the morphology

1 8 Th ere is some doubt about the basis of the WTS conjugation classes. While they are described as transitive and intransitive in the sources, Kennedy (nd) gives a substantial number of verbs which are morphologically but with transitive uses, and vice versa; Mitchell (1995) also doubts that transitivity determines the conjugation class. However, whatever the basis of the WTS conjugations, it is not telicity.

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 131 appears to be inherited. Alpher et al. (2008 :23) give discussion of cognates for the WTS dual suffi x –ma ; such a category is archaic within WTS and recon- structible to Proto-Pama-Nyungan. See further Keith (1998). Alpher et al ( 2008 ) also provide evidence for cognates in a number of other WTS verbal morphemes. A comparable analysis of Meryam Mir is beyond the scope of this paper, but from available materials for related languages (e.g. van Bodegraven and van Bodegraven 2004 and Dondorp and Shim 1997 ), Meryam verb mor- phology is typologically similar to other Eastern Trans-Fly languages. For example, Gizrra, Bine, and Wipi all have both prefi xal and suffi xal agreement, with the subject suffi xes compounded with tense markers. Th ere is no indication that either structural elements or specifi c forms in the verbal systems of WTS and Meryam Mir have been borrowed from one lan- guage into the other.

2.5 Deixis

2.5.1 WTS Deixis WTS has spatial deixis marking and distinguishes the following spatial loca- tions in relation to the speaker: proximate, remote, ‘over there,’ ‘up there,’ ‘down there,’ ‘up at the front,’ and ‘down at the back.’ Th e proximate and remote locations break down further into ‘in view,’ ‘not in view,’ and ‘loca- tional’ categories. ‘Over there,’ ‘up there,’ ‘down there,’ ‘up at the front,’ and ‘down at the back’ have both nominal and locational forms. All categories distinguish between singular, dual, and plural numbers, with the singular hav- ing both masculine and feminine forms. WTS deictic markers occur within the noun phrase. In example (6), the remote, locational, masculine deictic marker senawki occurs within the noun phrase of the direct object, agreeing in gender with the masculine noun ‘pig.’

(6) Ngath burum senawki waaydhin I-ERG pig-ABS deictic chase-remotepast ‘I chased a pig there.’ (Ford and Ober 1991 :127)

Ray (1907 :11) also gives deictic forms for WTS involving windward/lee- ward directions, implying that the language has an absolute system for direc- tions in addition to the forms mentioned above.

2.5.2 Meryam Mir Deixis Th e Meryam Mir deictic system refers to both time and space. Piper’s analysis is heavily qualifi ed as to its accuracy and completeness, but she suggests that

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

132 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 deictics are not independent words since they “nearly always” appear before a verb (Piper 1989 : 146). Th e deictic markers, while certainly expressing spatial reference, seem to be primarily organized according to temporality since the temporal location of an event relative to time of utterance determines the choice of deictic marker. As given by Piper ( 1989 : 148) deictics in Meryam Mir refer to the categories of location, shared argument/consecutive/contigu- ous, goal, and source, relative to the following time points: remote past, past, present, future, punctual, and non-punctual. Piper (1989 : 96) also cites a verbal deictic marker ta- and its allomorphs which is only spatial – it marks either an event “moving towards the speaker” or an event “happening away from the speaker.” Example (7) illustrates the use of pe , a deictic marker which refers to location and is used for an event happening at the present relative to the utterance.

(7) ná- gerger pe irdi? Q.clitic dayS Deix n PlS be (time, space) ‘What day is it today?’ (ie. What day is it now?) (Piper 1989 :153)

Th ere is no information available regarding the use of absolute direction terms in Meryam Mir. Ray ( 1907 ) does not provide an Eastern language equivalent to the WTS ‘windward’ direction term.

2.5.3 Evaluation of Similarity – Deixis Th e WTS and Meryam Mir deictic systems are quite diff erent in that the WTS system expresses a spatial relation to the speaker’s utterance while Meryam Mir can relate both a temporal and spatial relation. Piper ( 1989 : 146) is equivocal on the status of deictic markers and about where they occur in a sentence, but she does note that they are “always in conjunction with a verb.” Meryam Mir also has a verbal prefi x ta- that fulfi ls a deictic function, while in WTS all deictic markers occur solely within the noun phrase.

2.6 Syntax: constituent and morpheme order

2.6.1 WTS Basic information about constituent order and major phrasal order is given in Haddon (1907:48) and Ford and Ober ( 1991 :132). Th e word order is SV in intransitive clauses and SOV in transitive ones, with adjectives preceding the noun and adverbs preceding the verb. Possessors precede their possessed nouns. Adjuncts and adverbs appear, according to Ford and Ober (1991 :125) in directly preverbal position. An example is given in (8).

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 133

(8) Lawnga Ezoera woerab-al poer-a-y-gi-nga=a nu-ngu ira-pa gumi ma-y-gi-nga. No, E coconut-pl pluck- inf-neg-pred= and he- poss in.law- dat secretly give- inf-neg- pred. “No, Ezra didn’t pick the coconuts and give them secretly to his in-laws.” (Ford and Ober ( 1991 :132)) Negation is marked in the verb, and negative interjections appear fi rst in the clause. Th is is also illustrated in (8). Indirect objects may appear either side of the verb (Haddon 1907:48). Ford and Ober describe a ‘penultimate’ focus posi- tion in the clause; normally the verb is last and the item preceding it receives sentence focus, but if the verb is in focus another element is fi nal in the clause.

2.6.2 Meryam Mir Meryam Mir is verb-fi nal, with SV, SOV order and modifi ers preceding their noun (Piper 1989 , Ray 1907 :87). Th ere is a strong preference for the verb to be fi nal in the clause, but other items (such as adverbs) may follow it. Th ere is evidence for a verb phrase with adverbs preceding the OV complex. In clausal negation, the negator element appears in second position in the clause.

(9) Kára nole bakir-kak na-gri. 1 sg.gen neg stone-priv SgS.PlS- be.thing “I do not have money.” (Piper 1989 :185) 2.6.3 Comparison Here again there is little that is particularly indicative of close contact. Both languages are fairly consistently head-fi nal, but so are many other languages in the greater region ( http://wals.info/feature/81 shows the distribution of SOV order). Other details of syntax are also diff erent. Adverb placement in Meryam Mir provides some evidence for a verb phrase; the ungrammaticality of adverbs intervening between the verb an its object implies that they form a constitu- ent. Th ere is no such evidence in WTS, where adverbs appear between the object and the verb. Meryam Mir allows a single case infl ection per noun phrase, whereas WTS has case concord. Negation is distinct in the two lan- guages, with WTS having an optional initial negator (and the main marking of negation on the verb), and Meryam Mir using a negator which appears in second position in the clause.

3 Lexical Comparison

While our focus in this paper so far has been evidence (or lack of it) for con- tact induced grammatical change in the Torres Strait, we also briefl y survey

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

134 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 the evidence for language contact in the lexicon. First we survey previous work on this topic. Wurm (1972b :349) claims that WTS basic vocabulary includes a Papuan component, though he does provide any examples:

Th e Mabuiag diff erences [in phonology-JH/CB/ER] are fundamental and deep- going, and this and the Papuan nature of some of its basic vocabulary makes it less Australian, and more Papuan, than the Cape York Peninsula languages which look quite un-Australian at fi rst glance … Further work is needed to establish the exact nature and origin of the Papuan vocabulary in Mabuiag. Much of it is clearly Miriam… Apart from some Mabuiag loan-words of Australian origin, Australian infl uence on Miriam is negligible in contrast to the quite strong Papuan infl uence in Mabuiag. Haddon and Ray (1892:505ff ) provide discussion of the comparative vocabulary of the Eastern and Western Torres languages (and the language they call Daudai, which is probably Bine, the closest relative of Meryam Mir on the Papua New Guinea mainland). Th ey speak in terms of “agreement” in vocabulary between the languages, rather than “loans” or “shared inheritance”, though they draw conclusions about relationships between the languages which imply that they do understand the distinction. Th ey point to a number of words for trade items, including mai ‘pearlshell’, and kadik ‘armguard’. Ray (1907 :167-69) gives introduced (i.e., loan) words categorized by language of origin. Th e words given as introduced from Greek, Hebrew and Latin are mostly loans from English, or from biblical languages through bible transla- tion (e.g. lipano ‘frankincense; cf. Gk λίβανος and fi lakteri ‘phylactery’); oth- ers are said to be from Samoan. Kennedy (nd: 2) off ers the following breakdown of WTS vocabulary, based on his comparison of “100 or so basic vocabulary items” (we have no informa- tion about what items were compared):

Australian word stock 30% Austronesian word stock 50% Australian or Austronesian 6% Papuan word stock 5% Loans (from Indonesian/Meryam Mir/Papuan etc.) 1%

Kennedy gives no source for his ‘Austronesian’ comparisons, though the small number of examples he quotes (e.g. susu ‘milk’) are from Malay or Tok Pisin. In this section we review the evidence for lexical contact between Meryam Mir and WTS, and other potential sources of vocabulary (such as loans into one or both languages from Austronesian).

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 135

3.1 Comparison of Ray’s (1907) ‘Mabuiag’ and ‘Miriam’ materials

Ray ( 1907 :170ff ) provides a comparative index to his wordslists for WTS (which he calls Mabuiag, after the island name) and the Eastern Language. He lists WTS and Meryam Mir words “correspond[ing] in a general sense to the English” words. Th ese lists were compared and likely loans extracted. (Th e 19 th Century materials are not phonemically notated; therefore we looked for close resemblances in form and meaning, or items which were morphologically parseable in a donor language but not the recipient language. We are not yet at the stage of being able to do exhaustive etymologization of lexicon in either language.) Items include ‘bat’ (WTS sapur, Meryam Mir saper ), ‘bamboo’ (morap : marep), ‘cassowary (samu : sam), ‘cough’ ( kobaki : kobek), ‘cone shell (kaimi: kaimeg ) ‘fi shing line’ ( ariga : ariag), ‘inside’ ( mui : mui) and ‘mat’ (waku : wakoi ). Flora and fauna and trade items predominate; there are also a few kinship terms, including ‘sibling-in-law’ (ngaubat : neubet ~ naiwet), ‘aunt’ (apu : apu), ‘uncle’ ( awade : aua) and perhaps ‘brother’ ( babat : berbet ). Some items are shared beyond Meryam Mir and WTS; these include items possibly introduced by Kiwai traders (Lawrence, 1998 ). Such shared vocabulary com- prised less than 10% of the wordlist. A few authors have noted words which are similar to Austronesian lexical items in the Torres Strait. Ray and Haddon (1892:512-16) give a list of words which are formally similar to words in Austronesian languages such as Motu. Th ese include ‘banana’ (WTS daui, Motu dui ), ‘forehead’ (Miriam mat , Motu mata ‘eye’), ‘lungs’ (Miriam pereg , Motu baraki ), and ‘milk’ (WTA susu, Miriam sus ‘nipple’; susu is widespread in the Austronesian languages of the area).

3.2 Basic Vocabulary Comparison

WTS and Meryam Mir basic vocabulary were compared and coded for loan status. Th e comparison was based on a 204 item wordlist used by Bowern et al. (2010), which was in turn based on that used by Greenhill et al. (2008) for comparison and reconstruction within Austronesian. Th e WTS list was missing 5 items, for a total of 199, and the Meryam Mir list was missing 2 items, for a total of 202. Surrounding languages were also compared (that is, the Eastern Trans-Fly languages Gizrra, Bine and Wipi (Fleishmann and Fleishmann 1977), and the Pama-Nyungan languages in Bowern (2010 )). Th is allows us to draw some conclusions about the directions of loans, where the two languages share an identical (or near-identical) form. Th e list was also compared to various Austronesian and Papuan languages in the vicinity, and

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

136 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 to the Proto-Oceanic reconstructions of Pawley and Blust given in Greenhill et al (2008). Note that for the wider comparison, data were substantially incomplete and so conclusions should be regarded as tentative. WTS was found to have 13 loans from Meryam Mir and one from a Paman language (either Atampaya or Uradhi). Th e Meryam Mir loans included umai ‘dog’ and mapu-asi ‘heavy’. Th e Paman loan is moi ‘fi re’ (cf. Mpakwithi mwa , Alngith mae , from Proto-Northern Paman *thuma). Meryam Mir was found to have two loans from WTS; lid ‘bone’ and tot ‘roof’. Th ere were also three additional items which occurred in both languages (usar ‘kangaroo’, baz ‘cloud’ and sip(i) ‘root’) where the directionality of the loan cannot be determined.19 A subsample of 100 items from the 204-item list were compared to neigh- boring Austronesian languages. Th is was singularly unfruitful. Beyond susu ‘breast, milk’, not a single similar form was found. A typical example of the degree of similarity in items is in the word for ‘head’, and we are therefore unable to explain Kennedy’s claim that 50% of WTS basic vocabulary is Austronesian in origin. While there are no doubt loans from Austronesian in other parts of the vocabulary, we found absolutely no evidence for Austronesian infl uence on basic vocabulary. Th us while we do fi nd evidence for loans between the languages in fl ora and fauna, trade items, and religious terms, evidence from basic vocabulary does not support a claim of pervasive contact-induced language shift (from any language). Th e loans confi rm contact through trade (which is also described in

Table 13 Words for ‘head’ in selected languages of the region Family Language Form Eastern Trans-Fly Meryam Mir kerem Bine mopo Gizrra siŋɨl Wipi mop WTS Mabuiag kwi:ku, kuwikul Austronesian Proto-Oceanic (Blust) *qulu, *bwatu(k) Motu kwara Makassar ulu

1 9 It should be noted that WTS appears to have a high level of ‘unique’ vocabulary; that is, vocabulary for which no etymology can be found at present. Some of those ‘unique’ items will no doubt subsequently be found to have etyma in other parts of Australia.

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 137 the ethnographic literature; see Haddon (1891) and Lawrence ( 1998 ) for example). Th e loans in basic vocabulary do show some infl uence from Meryam Mir into WTS, but a fi gure of under 10% places the language in Haspelmath and Tadmor’s (2009) ‘low borrowers’ category.

4 Summary and Conclusion

We have investigated the similarity between structural features of WTS and Meryam Mir in an eff ort to contribute to the goal of substantiating or disprov- ing the claims made in the literature as to the nature of shared genetic history or contact between the two languages. Following Th omason’s (2001: 93-94) guidelines for proving the existence of structural change in a language due to contact, a broad survey of structural features was conducted. Th e fi rst explored were those relating to the phonolo- gies of WTS and Meryam Mir. Phonology is the most often cited evidence for contact in the Torres Strait, since WTS seems to be closer to Meryam Mir than it is to prototypical Australian languages in regards to voicing contrast, frica- tives, and rhotics, nasals, and vowel system. In his argument for classifying WTS as Papuan, Capell ( 1956 ) points specifi cally to the phonemic inventory and the voicing contrasts present in both languages as evidence for his view. Wurm (1972a , b ) also points out the “un-Australian” nature of the WTS pho- nological system. While the facts of the WTS system are undisputed, it cannot be assumed that these similarities are in fact signifi cant, since the neighboring Paman languages on Cape York paint an areal picture in which the features of WTS phonology are not particularly rare or unusual. While both languages have a split case-coding system between common nouns and proper nouns, and also code their pronouns in a third way, the split does not match up in the two languages in terms of how A, O, and S are marked. Number and case coding on pronouns and nouns also diff er between the two languages, and the case markers are formally distinct and show diff er- ences in function. Th ere is no evidence for convergence here. Adjectives and adverbs are very diff erent in these languages, particularly in terms of whether or not they can be said to form discrete classes and in how they are derived. Th e verbal systems diff er both in the divisions on which their conjugation classes are based and in their verbal templates. Tense and aspect also diff er in the languages, with WTS being predominantly organized around completive and incompletive aspect, with other aspects and tenses implying special focus, while Meryam Mir applies tense and aspect diff erently to its two conjugation classes. A review of the deictic systems of the languages was

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

138 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 undertaken. Th is comparison yielded no similarity, as Meryam Mir organizes its system along both space and time, while WTS makes references only to a spatial relation between the event and utterance. Th e languages also diff er in the location of deixis within the sentence. Th ough some basic syntax is similar, with both languages being verb-fi nal and having many of the correlates of head-fi nal languages, such as pronominal (rather than postnominal) modifi ca- tion, this is not unusual in the region, and is not typologically unusual. A comparison of the WTS and Meryam Mir lexicons was undertaken. Using Ray (1908), the number of shared items between WTS and Meryam Mir was calculated at less than 10%, while a comparison with a basic vocabu- lary list yielded a fi gure of 6.5% loans from Meryam Mir into WTS, and negligible loans from other sources. In conclusion, the similarity between WTS and Meryam Mir has, as Alpher et al. (2008 : 28) previously suggested, “been overstated.” Th ese results show the benefi ts of questioning and investigating the veracity of axiomatic claims made and accepted in the literature. Individual attempts in the litera- ture to classify WTS have seemed quite confi dent in their assertions; reading just one of them might lead the reader to believe that WTS is Australian- infl uenced Papuan, for example, a claim proved false by Alpher et al. ’s (2008) demonstration of WTS’s ties to Pama-Nyungan. Surveying a wider number of claims made about the history of WTS leaves one with the impression, no matter where the sources might place WTS’s genetic affi liation, that there is a large degree of signifi cant similarity between WTS and Meryam Mir. By systematically comparing features of WTS and Meryam Mir, we have shown that this impression is a false one; there is little similarity between the two languages, and what appears to be surface similarity, such as that of the phonemic inventories, is not necessarily signifi cant given the features of other Australian languages in the area. At best it is indicative of a more general areal pattern. Claims made by linguists in the past as to the nature of WTS and Meryam Mir’s shared history relied on what turns out to be superfi cial or selective evi- dence for support. A consideration of a wider variety of evidence reveals that whatever the history of culture contact in the Torres Strait, it has not resulted in language convergence.

References

Alpher , Barry , Geoff rey O’Grady and Claire Bowern . 2008 . Western Torres Strait language classifi cation and development . In Claire Bowern , Bethwyn Evans , and Luisa Miceli (eds.), Morphology and Language History: in Honour of Harold Koch , 15 - 30 . Amsterdam : John Benjamins .

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140 139

Austin , Peter . 1988 . “Phonological Voicing Contrasts in Australian Aboriginal Languages.” La Trobe University Working Papers in Linguistics . 1 , 17 - 42 . Bani , Ephraim , 1987 . Garka ga ipika . Australian Journal of Linguistics 7 ( 2 ): 189 - 201 Bani , Ephraim and T. Klokeid . 1971 . An outline of Mabuiag morphology . Papers on the Western Island language of Torres Strait. (Final Report to the Australian Institute of Aboriginal Studies) ed. by Ephraim Bani and Terry J. Klokeid . Canberra : AIATSIS, MA , pp 1 - 34 . van Bodegraven , Nico and Elly , 2005 . Phonology essentials Gizrra language. Data Papers on Papua New Guinea Languages 47. oai:sil.org:48164 Bowern , Claire . 2010 . Pama-Nyungan etymological database . Version 6, August 2010 . Capell , A . 1956 . A New Approach to Australian Linguistics . Sydney : University of Sydney . Comrie , Bernard . 1982 . Ergativity and grammatical relations in Kalaw Lagaw Ya (Saibai dialect) . Australian Journal of Linguistics . 1 : 1 - 42 . Comrie , Bernard . 2005 . Numeral bases . In Martin Haspelmath , Matthew S. Dryer , David Gil , & Bernard Comrie (eds.), Th e world atlas of language structures , chapter 131 . Oxford & New York : Oxford University Press . Crowley , Terry . 1981 . Th e Mpakwithi dialect of Anguthimri . In R.M.W. Dixon and Barry J. Blake (eds), Handbook of Australian languages , volume 2 . Canberra : Australian National University Press , 146 - 94 . Dixon , R.M.W . 1980 . Th e Languages of Australia . Cambridge : Cambridge University Press . Dixon , R.M.W . 2001 . “Th e Australian Linguistic Area . In Alexandra Y. Aikhenvald and R.M.W. Dixon (eds.), Areal Diff usion and Genetic Inheritance , 64 - 104 . New York : Oxford University Press . Dixon , R.M.W . 2002 . Australian Languages . Cambridge : Cambridge University Press . Dondorp , Anne , and Shim, JW . 1997 . Wipi grammar essentials. ms, SIL Ukarumpa. Evans , Nicholas . 2005 . Australian languages reconsidered: A review of Dixon (2002) . Oceanic Linguistics 44 (1) . 242 - 286 . Fleischmann , Lillian and Sinikka Turpeinen . 1976 . A dialect survey of Eastern Trans-Fly languages . In Richard Loving (ed.), Surveys in fi ve Papua New Guinea languages, 5-50. Workpapers in Papua New Guinea Languages , 16 . Ukarumpa : Summer Institute of Linguistics. Foley , William A . 1986 . Th e Papuan Languages of New Guinea . Cambridge : Cambridge University Press . Ford , Kevin and Dana Ober . 1991 . A sketch of Kalaw Kawaw Ya . In Suzanne Romaine (ed.), Language in Australia , 118 - 142 . Cambridge : Cambridge University Press . Greenhill , S. , R. D Gray & R. Blust . 2007 . Th e Austronesian Basic Vocabulary Database . http:// language.psy.auckland.ac.nz . Haddon , A.C . 1890 . Th e Ethnography of the Western Tribe of the Torres Straits . Journal of the Anthropological Institute , 19 , 297 - 440 Hale , Ken . 1976 . “Phonological Developments in Particular Northern Paman Languages.” In Peter Sutton (ed.), Languages of Cape York , 7 - 40 . Canberra : Australian Institute of Aboriginal Studies . Hale , Ken . 1997 . A Linngithigh Vocabulary . In Darrell Tyron and Michael Walsh (eds.), Boundary rider: essays in honour of Geoff rey O’Grady, 209-246. Pacifi c Linguistics , C-136 . Canberra : Australian National University . Hamilton , Philip . 1996 . Phonetic constraints and markedness in the phonotactics of Australian Aboriginal languages . Toronto : University of Toronto Ph.D. thesis . Hammarström , Harald . 2010 . Rarities in numeral systems . In Jan Wohlgemuth & Cysouw Michael (eds.), Rethinking universals: How rarities aff ect linguistic theory , 11 - 61 . (Empirical Approaches to Language Typology 45) . Berlin & New York : Mouton de Gruyter . Howitt , A. W . 1904 . Th e Native Tribes of South-East Australia . London : Macmillan and Co (facsimile edition, Aboriginal Studies press, 1996) .

Downloaded from Brill.com09/27/2021 08:04:08AM via free access

140 J. Hunter et al. / Journal of Language Contact 4 (2011) 106–140

Kennedy , Rod . 1981 . Phonology of Kala Lagaw Ya in Saibai dialect . In Bruce Waters (ed.) Australian Phonologies: Collected Papers , 103 - 137 . Work Papers of the Summer Institute of Linguistics, Australian Aborigines Branch A, 5 . Darwin : Summer Institute of Linguistics . Kennedy , Rod . 1985a . Clauses in Kala Lagaw Ya . In Sandra Ray (ed.), Aboriginal and Islander Grammars: Collected Papers , 59 - 77 . Darwin : Summer Institute of Linguistics . Kennedy , Rod . 1985b . Kalaw Kawaw Verbs . In Sandra Ray (ed.), Aboriginal and Islander Grammars: Collected Papers , 81 - 103 . Darwin : Summer Institute of Linguistics . Kennedy, Rod. nd. Kalaw Kawa Ya/English Dictionary. MS. King , 1937 . A voyage to Torres straits in search of the survivors of the ship Charles Eaton by CM Lewis. d. Grantham , North Shore , Sydney. Lawrence , D. 1998 . Customary Exchange in the Torres Strait . Australian Aboriginal Studies. MacGillivray , J . 1852 . Narrative of the Voyage of HMS ‘Rattlesnake’, commanded by the late Captain Owen Stanley, RN, FRS . London : T & W Boone . Murray , A W 1976 . Forty Years’ Mission Work in Polynesia and New Guinea . London : Nisbet & Co . Ogilvie , Sarah . 1994 . Th e Morrobolama (Umbuygamu) language of Cape York peninsula, Australia . Canberra : Australian National University Masters thesis . Piper , Nick . 1989 . A sketch grammar of Meryam Mir . Canberra : Australian National University MA thesis . Ray , Sidney H . 1907 . “Linguistics.” In Alfred C. Haddon (ed.), Reports of the Cambridge Anthropological Expedition to Torres Strait. Reports . Vol. 3 . Cambridge : Cambridge University Press . Ray and Haddon , 1893 . A Study of the Languages of the Torres Straits (I) . Proceedings of the Royal Irish Academy . 463 - 616 . Ray and Haddon , 1897 . A Study of the Languages of the Torres Straits (II) . Proceedings of the Royal Irish Academy . 119 - 373 . Ross , Malcolm , 1996 . Social networks and kinds of speech-community event . In Roger Blench & Matthew Spriggs (eds.), Archaeology and Language I , 209 - 261 . London : Routledge . Ross , Malcolm . 2007 . “Calquing and Metatypy.” Journal of Language Contact . THEMA 1 , 116 - 143 . von der Schulenburg , A. C. Graf . 1891 . Grammatik der Sprache von Murray Island . Berlin : Friedrich Wilhelm Universität zu Berlin Doctoral dissertation . Shnukal , Anna . 1988 . Broken: An Introduction to the Creole Language of Torres Strait . Canberra : Australian National University . Shnukal , Anna . 1995 . “From monolingualism to multilingualism in Australia’s Torres Strait island communities.” International Journal of the Sociology of Language . 113 , 121 - 136 . Sutton , P . 1976 . Languages of Cape York . Canberra : Australian Institute of Aboriginal Studies . Th omason , Sarah G . 2001 . Language Contact: An Introduction . Washington, D.C. : Georgetown University Press . Th omason , Sarah Grey and Terrence Kaufman . 1988 . Language contact, creolization, and genetic linguistics . Berkeley : University of California Press . Winford , Donald . 2003 . An Introduction to Contact Linguistics . Malden, MA : Blackwell Publishing . Wurm , S. A . 1972a . Languages of Australia and Tasmania . Th e Hague : Mouton . Wurm , S. A . 1972b . Torres Strait: a linguistic barrier? In D. Walker (ed), Bridge and Barrier: Th e Natural and Cultural History of Torres Strait . Canberra : Australian National University . Zentz , Jason , and Claire Bowern . 2010 . Diversity in the numeral systems of Australian hunter- gatherers . MS, Yale University .

Downloaded from Brill.com09/27/2021 08:04:08AM via free access