ARTICLES

Some Comparative Notes on Proto-Oceanic *mana: Inside and Outside the Austronesian Family
Juliette Blevins, 253

In a recent article, Blust (2007) presents a comprehensive summary of the etymology of Proto-Oceanic *mana ‘potent, effectual; supernatural power’, highlighting an ancient association between this word and the powerful forces of nature. Here I present new lexical data from Oceanic, Central Malayo-Polynesian, and South Halmahera–West New Guinea languages that support the reconstruction of Proto–Central-Eastern Malayo-Polynesian *mana ‘supernatural power, associated with spirits of the ancestors and the forces of nature’. Lexical comparisons from non-Austronesian languages of New Guinea suggest significant prehistoric contact between Austronesian and non-Austronesian speakers, and may support a connection between the
meaning of Proto–Central-Eastern Malayo-Polynesian *mana, and the
semantics of Proto–Western Malayo-Polynesian *mana(q) ‘inherit(ance)’.

A Reanalysis of Wuvulu Phonology
Robert Blust, 275

Wuvulu, a member of the Admiralty branch of the Oceanic subgroup of
Austronesian languages, has been reported in earlier publications as having as many as four allophones of the velar stop /k/, all of which appeared to be
in free variation. It is now clear that the allophony of /k/ involves both complementation and free variation. Surprisingly, the complementation of
velar obstruent allophones is phonetically conditioned, but follows no obvious phonetic principle. Wuvulu thus presents on the level of the allophone a challenge to phonological theory similar to that presented by sound changes that do not appear to be linguistically motivated.

Liquid Vocalization and Loss in Central Vanuatu
John Lynch, 294

A number of languages in central Vanuatu show merger of Proto-Oceanic *l
and *r, but also show a split in the merged phoneme. Although reflected as a
liquid in certain environments, especially before or adjacent to a high vowel
and also word-initially, *l and *r are also reflected as i, zero, or zero accompanied by fronting and raising of one of the adjacent vowels in the neighborhood of nonhigh vowels. The languages that show this context-sensitive vocalization and loss are geographically fairly contiguous, being spoken in southeast Malakula, Paama, and Epi, but belong to different genetic subgroups of Central Vanuatu. This paper attempts to explain these facts in their historical context.

Whence the Austronesian Indirect Possession Construction?
Mark Donohue and Antoinette Schapper, 316

Possession in some Austronesian languages shows levels of elaboration far in excess of cross-linguistic norms, while in others it is strikingly unelaborated. The appearance of alienable/inalienable contrasts has been assumed to result from contact with Papuan languages, and the existence of a paradigm of indirect possessive classifiers is cited as one of the pieces of evidence for the Oceanic subgroup, while acknowledging that indirect possession constructions can be found in Malayo-Polynesian languages further west. We argue that the appearance of possessive classifiers in these languages is also the result of contact with Papuan languages west of New Guinea.

The Interaction of Syntactic Structure and Postlexical Prosody in Saisiyat of Taiwan
Wen-yu Chiang and Fang-mei Chiang, 328

Postlexical prosodic phenomena in Austronesian languages have received relatively little attention, and consequently their patterns remain unknown. This paper aims to bridge this gap by investigating how syntactic structure interacts with postlexical prosodic phenomena in Saisiyat, an endangered language spoken in Taiwan. Several significant findings are made. First, Saisiyat sentential fundamental frequency (F0) patterns are based largely on its original lexical-level word F0 contour. The accent of a content word in sentences usually falls on its ultimate syllable, while trisyllabic or quadrisyllabic content words may sometimes undergo postlexical accentual modifications such as accent spreading, accent fronting, and accent adding. Function words, in contrast, play a role of interpolation as an intermediate site in bridging the F0 of their preceding and following syllables. Second, a yes-no question exhibits substantial prosodic modification by influencing its word preceding sentence-final interrogative particle aj, as compared with its counterpart in a declarative sentence. Third, agents in agent-focus sentences and patients in patient-focus sentences demonstrate higher values with respect to F0 peak, mean F0, and mean intensity. We provide typological explanations for these findings and explore the theoretical implications of postlexical prosodic patterns of Formosan languages in Taiwan.

Daniel Macdonald and the “Compromise Literary Dialect” in Efate, Central Vanuatu
Nick Thieberger and Chris Ballard, 365

Daniel Macdonald, a Presbyterian Church of Victoria missionary to the New
Hebrides from 1872 to 1905, developed a particularly strong interest in language. A prodigious author, he published widely and at length on the languages of Efate, and especially those of the Havannah Harbour area where he was stationed. But if his work is recalled today, it is as something of a curio, both for his insistence—archaic even for the times—on a link between ancient Semitic and Efate, and for his vigorous promotion of the use by the mission and its converts of a single, hybrid Efate language. This paper addresses and seeks to analyze what Macdonald himself called this “compromise literary dialect.” By identifying distinctive features of the three main varieties of Efate languages known today (Nguna or Nakanamanga, South Efate, and Lelepa), we aim to move beyond the lexical comparisons that have been the sole means of gauging relationships among these languages thus far. This enables us to begin the process of investigating the claim of Captain Rason, British Deputy Commissioner for the New Hebrides during Macdonald’s last years on Efate, that the “compromise literary dialect” was in fact a spoken dialect particular to the area of Havannah Harbour. We hope to reconsider and perhaps recuperate some of Macdonald’s writing as a rare if often distorted window on indigenous life and language at a pivotal moment in the transformation of Efate communities.

Subject Agreement in Marovo: Diachronic Explanations of Synchronic Conditions
Bethwyn Evans, 383

In many Oceanic languages, subject arguments are indexed by preverbal markers within the verb complex. In Marovo, an Oceanic language of the Solomon Islands, such subject markers show unusual synchronic behavior, in that their presence is conditioned by both morphosyntactic and pragmatic characteristics of the clause. Thus, in Marovo preverbal subject markers occur obligatorily with certain clause-initial discourse connective particles and with the negative particle. Subject markers also occur outside of these morphosyntactic environments, where their use contrasts with that of other expressions of the subject argument, including lexical or pronominal noun phrases or the lack of overt expression. Within this context of subject expression more generally, the occurrence of Marovo subject markers can be seen to be determined by the discourse role of the subject argument. It is argued here, through comparison of Marovo subject marking with that in closely related languages, that these synchronic conditions have diachronic explanations.

A Conservative Vowel Phoneme Inventory of Sumatra: The Case of Besemah
Bradley McDonnell, 409

In general, the Malayic languages of Sumatra show a vowel phoneme inventory that is either equal to or greater than that of the five phonemes in Standard Malay/Indonesian (not including the so-called pepet vowel); dialects of Jambi Malay, Palembang Malay, and Minangkabau all show five phonemes. However, from recent fieldwork in Besemah, a Malayic language in the highlands of southwest Sumatra, I shall describe an inventory that evinces the smallest number of vowel phonemes in any Malayic language of Sumatra described to date. That is, Besemah has three vowel phonemes. Nevertheless, the analysis of Besemah vowel phonemes is not straightforward; vowel lowering in closed syllables, vowel harmony, and raising of the word-final low central vowel all cloud the analysis of a three-vowel system. Furthermore, Besemah currently is experiencing intense pressure from other standard and nonstandard varieties of Malay/Indonesian, which demonstrates how intense contact involving diglossia and increasing bilingualism in these other varieties of Malay/Indonesian appear to be leading to an emergence of mid vowels in Besemah.

SQUIBS

The Language of Lapita: Vanuatu and an Early Papuan Presence in the Pacific
Mark Donohue and Tim Denham, 433

The languages of Vanuatu are uniformly Austronesian, but have long been
described as “aberrant.” Blust (2005) points out a number of morphosyntactic features of the Vanuatu languages that might provide evidence for a Papuan element in their history. We add to that argument, presenting phonological evidence that links the languages of Vanuatu and New Caledonia with the non-Austronesian languages of New Guinea. Accepting that the earliest archaeological sites in Vanuatu are Lapita sites, we suggest that this implicates non-Austronesian speaking Melanesians in the earliest occupancy of the islands, calling into question assumptions that the Lapita expansion in the Pacific can be unproblematically associated with the expansion of Austronesian languages of the Oceanic subgroup.

Remote Melanesia: One History or Two? An Addendum to Donohue and Denham
Robert Blust, 445

Blust (2005) proposed that certain typological traits in the Austronesian languages of Vanuatu and New Caledonia—here called “Remote Melanesia”—
suggest Papuan contact influence in situ. Given the absence of any pre-Lapita archaeological tradition in this area, it now seems best to frame this hypothesis in terms of two closely spaced migrations that appear to be archaeologically indistinguishable. The first wave brought Austronesian speakers of southern Mongoloid physical type into Remote Oceania. The second wave brought Papuan speakers who had acquired the outrigger canoe complex, pottery, and some other elements of material culture from the incoming Austronesians in Near Oceania, but who remained biologically and culturally distinct from them in other ways. In time, the Papuan languages of Remote Melanesia were abandoned in favor of the more uniform and widely dispersed speech of late Proto-Oceanic speakers, but not before leaving traces of their former presence in the form of typological divergence toward a pattern that is more typical of Papuan languages than of Austronesian languages outside Melanesia.

BOOK REVIEWS

Ruben Stoel. 2005. Focus in Manado Malay: Grammar, particles, and intonation.
Reviewed by Anastasia K. Riehl, 460

Fritz Schulze and Holger Warnk, eds. 2006. Insular Southeast Asia: Linguistic and cultural studies in honour of Bernd Nothofer.
Reviewed by Franz Mueller, 468

Elizabeth Zeitoun, ed. 2004. Fait de langues: Revue de linguistique no 23–24: Les langues austronésiennes.
Reviewed by Paul Geraghty, 470

INDEX

Index of Languages in Volume 47, 474

Electronic facsimiles of all out-of-print volumes of Oceanic Linguistics Special Publications are now available online in the Arts & Sciences Complement of the JSTOR journal archive. The titles still in print and available from University of Hawai‘i Press are hyperlinked by number and by title to the Press’s website. The rest are linked to JSTOR.

35. Bill Palmer. 2008. Kokota Grammar. 448 pp.

34.Thomas John Hudak. 2008. William J. Gedney’s Comparative Tai Source Book. 232 pp.

33. Nicholas Thieberger. 2007. A Grammar of South Efate: An Oceanic Language of Vanuatu. 416 pp.

32. Otto Dempwolff; tr. & ed. by Joel Bradshaw and Francisc Czobor. 2005. Otto Dempwolff’s Grammar of the Jabêm Language in New Guinea. 132 pp.

31. Terry Crowley. 2004. Bislama Reference Grammar. 206 pp.

30. Juliette Blevins. 2001. Nhanda: An Aboriginal Language of Western Australia. 170 pp.

29. Videa P. De Guzman and Byron W. Bender, eds. 2000. Grammatical Analysis: Morphology, Syntax, and Semantics. 298 pp.

28. Graham Thurgood. 1999. From Cham to Modern Dialects: Two Thousand Years of Language Contact and Change. 498 pp.

27. Terry Crowley. 1998. An Erromangan (Sye) Grammar. 294 pp.

26. John W. M. Verhaar. 1995. Toward a Reference Grammar of Tok Pisin: An Experiment in Corpus Linguistics. 470 pp.

25. Midori Osumi. 1995. Tinrin Grammar. 304 pp.

24. Jerold A. Edmondson and Kenneth J. Gregerson, eds. 1993. Tonality in Austronesian Languages. 178 pp.

23. Robert Parkin. 1991. A Guide to Austroasiatic Speakers and Their Languages. 198 pp.

22. Samuel H. Elbert. 1988. Echo of a Culture: A Grammar of Rennell and Bellona. 306 pp.

21. Charles Randriamasimanana. 1986. The Causatives of Malagasy. 684 pp.

20. Veneeta Z. Acson and Richard L. Leed, eds. 1985. For Gordon H. Fairbanks.

19. Paul A. Geraghty. 1983. The History of the Fijian Languages. 484 pp.

18. Frantisek Lichtenberk. 1983. A Grammar of Manam. 648 pp.

17. James Patrie. 1982. The Genetic Relationship of the Ainu Language. 174 pp.

16. Videa P. De Guzman. 1978. Syntactic Derivation of Tagalog Verbs. 414 pp.

15. Fang Kuei Li. 1977. A Handbook of Comparative Tai. 390 pp.

14. John E. Reinecke, Stanley M. Tsuzaki, David DeCamp, Ian F. Hancock, and Richard E. Wood, comps. 1975. A Bibliography of Pidgin and Creole Languages. 804 pp.

13. Philip N. Jenner, Laurence C. Thompson, and Stanley Starosta, eds. 1976. Austroasiatic Studies, Part I (pp. 1–692) and Part II (pp. 693–1344).

12. Timothy M. Manley. 1972. Outline of Sre Structure. 240 pp.

11. Morice Vanoverbergh. 1972. Isneg-English Vocabulary. 640 pp.

10. Leatrice T. Mirikitani. 1972. Kapampangan Syntax. 264 pp.

09. Susumu Nagara. 1972. Japanese Pidgin English in Hawaii: A Bilingual Description. 322 pp.

08. Lawrence A. Reid, ed. 1971. Philippine Minor Languages: Word Lists and Phonologies. 242 pp.

07. David D. Thomas. 1971. Chrau Grammar. 258 pp.

06. David R. Counts. 1969. A Grammar of Kaliai-Kove. 170 pp.

05. Albert J. Schütz. 1969. Nguna Grammar. 188 pp.

04. Albert J. Schütz. 1969. Nguna Texts. 326 pp.

03. Richard E. Elkins. 1968. Manobo-English Dictionary. 356 pp.

02. Lawrence Andrew Reid. 1966. An Ivatan Syntax. 160 pp.

01. Stanley M. Tsuzaki and John E. Reinecke. 1966. English in Hawaii: An Annotated Bibliography. 62 pp.

ARTICLES

A Typology of First Person Dual Pronouns and Their Reconstructibility in Philippine Languages
Hsiu-Chuan Liao, 1

This paper provides an overview of the distribution of first person dual pronouns in Philippine languages and addresses the issue as to whether or not first person dual pronouns can be reconstructed for the ancestral language of all Philippine languages. Based on data from different microgroups of Philippine languages, the following conclusions are reached. First, no first person dual pronouns can be reconstructed for the parent of the Philippine languages. Second, the wide distribution of first person dual pronouns in different microgroups of Philippine languages is due to drift rather than direct inheritance from the parent of the Philippine languages.

Transitivity Discord in Some Oceanic Languages
Anna Margetts, 30

Some Oceanic languages have clause types that feature intransitive verbs cooccurring with what looks like an object argument. Such constructions are sometimes described as noun incorporation, but there is evidence for two distinct constructions: noun incorporation, and clauses with what I will call transitivity discord, featuring intransitive verbs and object nouns. Such discord constructions have transitive and intransitive features that are manifested on different structural levels and they can be described as showing a mismatch between verb-level and clause-level transitivity. They share features with noun incorporation, but they are structurally different in that the object noun has syntactic independence rather than being part of the verb.

Is there a Bima-Sumba Subgroup?
Robert Blust, 45

For some seven decades a number of Austronesian languages in the Lesser Sunda islands of eastern Indonesia have been assigned to a “Bima-Sumba” subgroup. No evidence has ever been presented for this group, yet through sheer repetition it has come to be accepted by many scholars. A comparative analysis of “Bima-Sumba” languages shows clear support for a Sumba-Hawu group, and limited evidence for a larger genetic unit that includes many or all of the languages of western and central Flores. However, there is no support for a more inclusive subgroup that incorporates Bimanese, unless it also includes languages that were not assigned to the original Bima-Sumba group.

Yet More on the Position of the Languages of Eastern Indonesia and East Timor
Mark Donohue and Charles E. Grimes, 114

The line dividing the Austronesian languages into Western Malayo-Polynesian (WMP) and Central-Eastern Malayo-Polynesian (CEMP) is drawn east of Sulawesi and through the middle of Sumbawa. A number of phonological or semantic changes are claimed as forming the basis of this distinction, as well as the typological profile of the languages to the east being different from those to the west, and a number of lexical items being attested only east of the line. We examine the phonological and semantic innovations, as well as the erratic morphological ones, showing that none of them define the CEMP line, but indicate that (a) the Central Malayo-Polynesian (CMP)–area languages do not convincingly meet the criteria commonly accepted for a subgroup or even a linkage, (b) some of the WMP-area languages exhibit more of the same features found in at least some of the CMP-area languages than do others, and (c) many of the traits ascribed to the CMP- or CEMP-area languages can be found in more conservative WMP-area or Formosan languages as well.

The Undergoer Focus ma- in Kavalan
Shuping Huang and Li-May Sung, 159

This study aims to explore the functions and grammatical status of the Kavalan preverbal affix ma-. In addition to agent focus, locative focus, and referential focus (mainly used to focus the instrumental or benefactive case), Kavalan ma- is found to mark specific sentential focus. This marker behaves like agent focus in terms of its grammatical behavior, and is mainly used in two scenarios: spontaneous events, when the event is conceived of as happening spontaneously without an extraneous causer; and anticausatives, when the patient is the focus and the agent is conceived of as insignificant. In some limited cases, ma- is also used for naturally collective/reciprocal events. In ma-marked clauses, the clausal subject is typically conceived of as a spontaneously affected role, and this marker is therefore termed “undergoer focus” in the present study. Similar grammatical devices are found in two other Formosan languages, Paiwan and Amis, both of which share semantic and syntactic similarities with Kavalan ma-. It has been suggested by Evans and Ross (2001) that the same form ma- is commonly found in Oceanic languages, with different functional manifestations cross-linguistically. Further investigation into the manifestations of similar devices in other genetically related languages may help us to gain a better understanding of internal relationships within the Austronesian language family.

Reefs–Santa Cruz as Oceanic: Evidence from the Verb Complex
Åshild Næss and Brenda H. Boerger, 185

Recent research has shown that the long-standing assumption that the Reefs–Santa Cruz languages have a non-Austronesian substrate is unlikely to be valid: the languages do show regular sound correspondences with Proto-Oceanic (Ross and Næss 2007), and the alleged noun classes cannot felicitously be analyzed as such (Næss 2006). This paper addresses the third argument given in previous work for a non-Austronesian substrate: the complex verb structures. Presenting data from both Natügu (Northern Santa Cruz) and Äiwoo (Reefs), we show that while some of the verb morphology has clear cognates in Proto-Oceanic, other parts can be understood as deriving from an earlier productive process of verb serialization followed by reduction of the forms found in such serialized constructions. Given that both verb serialization and grammaticalization of elements of serializing constructions are well known in Oceanic languages, this leaves no linguistic evidence for a non-Austronesian substrate in Reefs–Santa Cruz.

SQUIB

Possession in Irarutu
René van den Berg and Takashi Matsumura, 213

This short paper offers a description of possessive constructions in Irarutu, an Austronesian language spoken in Indonesian Papua, which belongs to the South Halmahera–West New Guinea subgroup. Possession in Irarutu follows a typical East Malayo-Polynesian pattern distinguishing alienable and inalienable possession, but some of the morphology used to code possession is unusual, including what appears to be possessive infixation.

Typology, Areality, and Diffusion
Mark Donohue, Søren Wichmann, and Mihai Albu, 223

Dunn et al. (2007) state that their typological comparisons do not demonstrate genealogical relatedness in the usual sense, but that the technique does accurately recapitulate trees established by the comparative method. We demonstrate that the signal picked up by their method is areal, rather than genealogical, and suggest that the method, when tested on known language families, will also show a high sensitivity to the effect of diffusion.

IN MEMORIAM

In Memoriam, Françoise Ozanne-Rivierre, 1941–2007
Darrell Tryon, 233

BOOK REVIEWS

John Bowden and Nikolaus Himmelmann, eds. 2004. Papers in Austronesian subgrouping and dialectology.
Reviewed by Alexander Adelaar, 240

Kunio Nishiyama and Herman Kelen. 2007. A grammar of Lamaholot, Eastern Indonesia: The morphology and syntax of the Lewoingu dialect.
Reviewed by John Bowden, 247

ARTICLES

325, Universal Uses of Demonstratives: Evidence from Four Malayo-Polynesian Languages
Jessica Cleary-Kemp

Across the languages of the world, elements recognized as “demonstratives” perform a wide range of functions, including reference tracking, discourse deixis, and recognitional functions, as well as simple pointing in the immediate physical context. It has long been assumed that this last “situational” use is the most basic, and that all other uses of demonstratives are derived from it. However, the primacy of situational use has recently been questioned, and evidence has been presented that suggests the four abovementioned uses of demonstratives are pervasive in all languages, and are therefore equally basic. This article examines further evidence from four non-Oceanic Malayo-Polynesian languages. The data presented here tentatively support the theory that the above four uses of demonstratives are universal, but they also confirm that situational use should indeed be considered the basic use of demonstratives.

348, Typology and the Linguistic Macrohistory of Island Melanesia
Mark Donohue and Simon Musgrave

Recent years have seen much discussion on the use and meaning of typological argumentation when reconstructing language history and language relations. We address the conclusions and methodology of a paper “Structural phylogenetics and the reconstruction of ancient language history” (Science, Sept. 23, 2005), which claims that, on the basis of a typological comparison, the non-Austronesian languages and (Austronesian) Oceanic spoken to the immediate east of new Guinea can be shown to belong to two unrelated genetic entities. We argue that the data and discussion in this paper do not allow us to conclude that the non-Austronesian languages in the study form a valid linguistic group in any historical sense, or that the methods they apply can be used to make claims about linguistic relatedness.

388, Statistical Reasoning in the Evaluation of Typological Diversity in Island Melanesia
Michael Dunn, Robert Foley, Stephen Levinson, Ger Reesink, and Angela Terrill

This paper builds on a previous work in which we attempted to retrieve a phylogenetic signal using abstract structural features alone, as opposed to cognate sets, drawn from a sample of Island Melanesian languages, both Oceanic (Austronesian) and (non-Austronesian) Papuan (Science 2005[309]: 207-2–75). Here we clarify a number of misunderstandings of this approach, referring particularly to the critique by Mark Donohue and Simon Musgrave (in this same issue of Oceanic Linguistics), in which they fail to appreciate the statistical principles underlying computational phylogenetic methods. We also present new analyses that provide stronger evidence supporting the hypotheses put forward in our original paper: a reanalysis using Bayesian phylogenetic inference demonstrates the robustness of the data and methods, and provides a substantial improvement over the parsimony method used in our earlier paper. We further demonstrate, using the technique of spatial autocorrelation, that neither proximity nor Oceanic contact can be a major determinant of the pattern of structural variation of the Papuan languages, and thus that the phylogenetic relatedness of the Papuan languages remains a serious hypothesis.

404, Proto-Oceanic *mana Revisited
Robert Blust

Few linguistic terms in the history of anthropology have had greater currency than mana. While anthropological debate about this term has tended to center on the correct interpretation of the native concept, little attention has been given to the etymology of the word. When this is pursued, a novel perspective on this pivotal concept emerges. Cognates meaning ‘thunder’ and ‘wind’ suggest that Proto-Oceanic *mana did not refer to a detachable spiritual or supernatural power that could be possessed by humans, but rather to powerful forces of nature such as thunder and storm winds that were conceived as the expression of an unseen supernatural agency. As Oceanic-speaking peoples spread eastward, the notion of an unseen supernatural agency became detached from the physical forces of nature that had inspired it and assumed a life of its own. It is argued that the process that gave rise to the canonical sense of mana, as this is commonly understood in the anthropology of religion, is part of a larger process in which widespread and apparently arbitrary features of human cultures were inspired by a prescientific attempt to understand the forces of nature.

424, Lexical Perspectives on Voice Constructions in Tsou
Huei-ju Huang and Shuanfan Huang

Very few verbs in Formosan languages are known to carry a full set of voice affixes, but it remains unclear how and to what extent the lexical meaning of a verb influences constructional variability. In this study we sort out where the gaps may lie, and probe the complex and interwoven nature of the relationships between the semantic roles of nominative arguments and various non-Actor Voice clauses in Tsou. We argue that voice forms are not automatic mechanical processes based on lexical selection alone, and we often we insights that show that both lexical and constructional meaning interact in licensing Benefactive Voice (BV) clauses. The argument is also made that two types of BV clauses must be distinguished, causativized and noncausativized. Verbs in noncausativized BV clauses “subcategorize” for nominative NPs that would normally function as “noncore” or peripheral arguments in a European language like English, which thus constitutes important evidence that Tsou is a language that does not grammatically make the core/oblique distinction. Causativized BV clauses, on the other hand, have the effect of bringing in an extra argument to the source clauses and marking it with oblique case, while retaining the case markings on the nominal arguments in the source clauses, violating the well-known causee accessibility hierarchy.

456, An Oceanic Origin for Äiwoo, the Language of the Reef Islands?
Malcolm Ross and Åshild Næss

Whether the languages of the Reefs–Santa Cruz (RSC) group have a Papuan or an Austronesian origin has long been in dispute. Various background issues are treated in the introductory section. In section 2 we examine the lexicon of the RSC and Utupua-Vanikoro languages and show that there are regular sound correspondences among these languages, and that RSC languages display regular reflexes of Proto-Oceanic etyma and are therefore Austronesian. We also show that together the RSC and Utupua-Vanikoro languages form an Oceanic subgroup, which we label “Temotu,” and that the Temotu group is probably a first-order subgroup within the Oceanic family. In section 3, we examine a variety of constructions and morphemes in Äiwoo, the language of the Reef Islands, to see whether they have plausible Oceanic sources. The answer in most cases is that they do. This is important, as several of these constructions have in the past been given as evidence that the RSC languages have a Papuan origin. We conclude that the RSC languages are Austronesian and that there is no need to posit a Papuan element to explain their origin.

499, Imperfective Aspect and the Interplay of Aspect, Tense, and Modality in Torau
Bill Palmer

Torau displays a highly complex system of aspect, tense, and modal marking. One of the most complex elements of this system is the marking of imperfective aspect. Imperfective in Torau is marked by a construction employing a choice of two overt imperfective markers and the possible presence of reduplication. The range of imperfective semantics encoded by this construction varies widely, encompassing progressive, habitual, persistive, and progressive inchoative or inceptive. Which reading is given depends not only on the choice of imperfective marker and the presence or absence of reduplication, but on a complex interplay of these factors with other aspectual, modal, or tense marking, and the aspectual semantics of the verb itself. This paper teases apart each of these highly interdependent factors to determine the independent functional characteristics of each imperfective marker and of reduplication.

520, The Papuan Language of Tambora
Mark Donohue

I present data from Tambora, a now extinct language of central Sumbawa, and argue from the lexical data and the inferred phonology, compared with areal norms, that it was a Papuan language spoken by a trading population of southern Indonesia. The existence into historical times of a large and nonreclusive Papuan political entity this far west forces a major revision of our ideas about the linguistic macrohistory of Eastern Indonesia.

538, Definiteness and Specificity in Mavea
Valérie Guérin

Specificity and definiteness are universal semantic categories, but not all languages express these categories morphologically. In this paper, I present data from Mavea, a language spoken in northern Vanuatu, which show morphological expressions of these two semantic categories. I argue that in Mavea, the article le denotes specificity, aite encodes indefiniteness, te … aite refers to indefinite nonspecific expressions, while the lack of an article expresses definiteness.

554, Cebuano Passives Revisited
Michael Tanangkingsing and Shuanfan Huang

The view that the gi-clauses and/or their equivalents in other Philippine-type languages, specifically in Cebuano and closely related Bisayan languages, are active constructions has been widely accepted by a number of Austronesian linguists. In a recent study on the gi-verb clauses in Cebuano, however, another linguist reinterprets those with Verb-Patient-Agent (VPA) word order as passive. In this paper, we argue against such an interpretation, based on analyses of the semantics and discourse pragmatics of the gi- and na- clauses in spoken data. A gi- attached to a verb implies a deliberate intention of an Agent; a na-clause directs attention to the often accidental effect of an action on a Patient without emphasizing any reference to an Agent. This renders a na-verb construction, especially one in which the Agent is missing, as a much more plausible candidate for passive in Cebuano.

585, Conjunctive Reduction and its Origin: A Comparative Study of Tsou, Amis, and Squliq Atayal
Wei-tien Dylan Tsai

This paper discusses the issue of how coordinate structures evolve into subordinate structures in both syntactic and semantic terms. I call this type of process “conjunctive reduction.” It is well established in the literature on Chinese historical syntax that some modifier-head and verb-complement compounds actually derive from coordinate structures in Ancient Chinese. Based on this finding, I suggest that a similar process is also at work in Formosan languages, but on a quite different scale. That is, while Chinese encodes conjunctive reduction in compounding morphology, the same process involves full-fledged syntactic operations in Formosan languages. I propose that there are two general directions of conjunctive reduction. On the one hand, the coordinator may become a modifier marker, where the first conjunct becomes a marked adverbial; then the modifier marker may disappear completely, making the first conjunct an unmarked adverbial. I call this “adverbialization.” On the other hand, the coordinator may become a complementizer, introducing either an infinitive complement or an adverbial adjunct such as a conditional or temporal clause. I take Squliq Atayal, Tsou, and Amis to represent the Northern, Tsouic, and Paiwanic groups respectively, which in turn points to the existence of a protolanguage with extensive coordinate construals along the line of Neo-Davidsonian semantics,
very much like Ancient Chinese.

SQUIB

603, A Typologically Unusual Interrogative Word in Toqabaqita and Other Oceanic Languages
Frantisek Lichtenberk

In Toqabaqita and some other Oceanic languages the reflexes of Proto-Oceanic *sapa ‘what?, which?’’ are used to inquire about parts of whole and/or about kinship relations; for example, Toqabaqita tafa ‘which part of person’’s or animal’’s body?’’. While these interrogative functions may be unusual cross-linguistically outside of Oceanic, in Oceanic they follow naturally from a grammatical property of Proto-Oceanic, specifically the use of one type of attributive possessive construction to encode inalienable-possession relations.

BOOK REVIEWS

613, Nicholas Thieberger. 2006. A grammar of South Efate: An Oceanic language of Vanuatu.
Reviewed by Catriona Hyslop Malau

618, Gabriele H. Cablitz. 2006. Marquesan: A grammar of space.
Reviewed by Giovanni Bennardo

623, Carl R. Galvez Rubino. 2006. Intensive Tausug: A pedagogical grammar of the language of Jolo, Philippines.
Reviewed by John U. Wolff

624, K. Alexander Adelaar. 2005. Salako or Badameà, sketch grammar, texts and lexicon of a Kanayatn dialect in West Borneo.
Reviewed by Daniel Kaufman

INDEX

634, Index of Languages in Volume 46

ARTICLES

1, Òma Lóngh Historical Phonology
Robert Blust

The phonology of Òma Lóngh Kenyah as described by Soriente (2006) shows striking typological differences from its nearest relatives. Contrary to a pattern of avoidance that is almost universal in Austronesian languages, it has developed final palatals, including a voiceless unreleased palatal stop (written -j), and a palatal nasal (written -ny). In violation of universal tendencies in phonological systems, it has also innovated a voiceless velar nasal (but no other voiceless nasals) in final position. Out of a Proto-Kenyah six-vowel system in which tense mid vowels occurred only word-finally, it has developed three new vowels and an unusual system of double vowel harmony that requires both High-Mid avoidance and Tense-Lax agreement. Even more surprisingly, a typologically bizarre connection between the tenseness/laxness of the penultimate vowel and the shape of the final syllable is present in one subclass of bases, but emerges clearly only through a historical analysis. Together, these innovations add to an already impressive picture of north-central Borneo as a “hot spot” for rapid phonological change, including changes that do not appear to be phonetically motivated.

54, An Unusual Passive in Western Oceanic: The Case of Vitu
René van den Berg

Passive constructions are rare in Melanesia and none is reconstructed for Proto-Oceanic. This paper reports on a passive in Vitu, a Meso-Melanesian language spoken in West New Britain, Papua New Guinea, describing its structural properties (some of which are unusual) and attempting to provide a diachronic scenario through which it may have arisen.

71, Three-participant Events in Oceanic Languages
Anna Margetts

In this study I investigate the linguistic strategies available in Oceanic languages for the encoding of events with three participants (such as expressions of ‘sending’, ‘giving’, ‘showing’, ‘telling’, or doing something for someone’’s benefit). The notion of three-participant events is traditionally associated with the concept of ditransitive clauses, but there are, in fact, a variety of other strategies found cross-linguistically, and only some of these involve ditransitive constructions. Languages may differ considerably in which of these methods of encoding they productively use. In the present study I explore which of the strategies are used in the Oceanic language group. This may be a first step toward establishing whether language families or groups differ in their preferences for certain strategies and whether such preferences correlate with other typological features. While in some strategies all three event participants are encoded by syntactic means, in other strategies the involvement of a third participant is essentially evoked by pragmatics. The Oceanic language group seems to show a greater preference for such pragmatic strategies than is familiar from the study of the better-known European languages.

128, On Ergativity and Accusativity in Proto-Polynesian and Proto––Central Pacific
Douglas Ball

This paper continues the debate begun over three decades ago about the nature of the case-marking system of Proto-Polynesian, whether it possessed an ergative or an accusative system. After consideration of extra-Polynesian comparative evidence, putative counterexamples in some Polynesian pronominal systems, and the history of the current accusative marking, it finds in favor of the accusative system view.

154, A Long Lost Sister of Proto-Austronesian? Proto-Ongan, Mother of Jarawa and Onge of the Andaman Islands
Juliette Blevins

This paper applies the comparative method to two related languages of the southern Andaman Islands, Jarawa and Onge, leading to the reconstruction of a protolanguage termed “Proto-Ongan” (PON). The same method is used to argue that Proto-Ongan may be related to Proto-Austronesian (PAN). Lexical and grammatical evidence suggests that Proto-Ongan and Proto-Austronesian are sisters, daughters of a Proto––Austronesian-Ongan (PAO). The implications of this discovery are wide-ranging, from potential solutions to problems in PAN grammar, to new hypotheses regarding ancient speaker migrations. While few of these implications are examined here, an extended Austronesian phylogeny is proposed in the hope that it will seed new avenues of research, and highlight the potential importance of Andamanese studies in understanding Austronesian prehistory.

199, Heads in Oceanic Indirect Possession
Bill Palmer and Dunstan Brown

In many Oceanic languages the “indirect” possessive construction, which is typically associated with alienable possession, uses special forms to host person and number agreement indexing the possessor. This can be contrasted with the “direct” possessive construction, typically associated with inalienable possession, where a lexical possessum noun itself carries possessor-indexing agreement. The host forms used in the indirect construction are often referred to as “classifiers.” We argue that this term should not be applied to indirect possession marking in many Oceanic languages, and present evidence to show that indirect possessor-indexing hosts in such languages do not have the properties typically associated with classifiers. In contrast with this, we further argue that these indirect possessor-indexing hosts should be treated as the syntactic head of the noun phrase in which they occur, thereby allowing treatment of the syntax of NPs with indirect possession that is consistent with those with direct marking. In both instances, the person and number indexing morphology simply attaches to the syntactic head.

210, Is Kazukuru Really Non-Austronesian?
Michael Dunn and Malcolm Ross

Kazukuru is an extinct language, originally spoken in the inland of the western part of the island of New Georgia, Solomon Islands, and attested by very limited historical sources. Kazukuru has generally been considered to be a Papuan, that is, non-Austronesian, language, mostly on the basis of its lexicon. Reevaluation of the available data suggests a high likelihood that Kazukuru was in fact an Oceanic Austronesian language. Pronominal paradigms are clearly of Austronesian origin, and many other aspects of language structure retrievable from the limited data are also congruent with regional Oceanic Austronesian typology. The extent and possible causes of Kazukuru lexical deviations from the Austronesian norm are evaluated and discussed.

232, Another Look at the Marking of Plural Personal Noun Constructions in Austronesian Languages
Lawrence A. Reid

In a recent note in this journal, Robert Blust, using data from Philippine and Formosan languages, proposes a functional difference for Proto-Austronesian between the forms of genitive common noun phrase markers, such that PAn *nu marked ‘genitive of common nouns’, while PAn *na marked ‘genitive of plural personal nouns’. This paper examines the Philippine and Formosan evidence for these reconstructions and concludes that the evidence provided is the result of convergent development in the languages cited, and cannot be considered evidence for the proposed reconstructions. Alternate reconstructions that better account for the Philippine evidence are proposed.

253, Papuan Malay Pronominals: Forms and Functions
Mark Donohue and Yusuf Sawaki

Papuan Malay, the easternmost variety of Malay/Indonesian, has received even less attention than other nonstandard varieties of Malay/Indonesian. Papuan Malay has innovative forms and functions for its pronominals that have not been described in detail for other varieties of Malay/Indonesian, though they are present over a wide area. We examine both the bound and the free pronominal forms, describing the status of the different members in each paradigm as they are used with different functions such as possessor, subject, and object. In addition to noting these different uses, we discuss a trivalent construction in the language with an exceptional use of pronominal forms, and propose an ongoing path of grammaticalization that can account for it.

277, Missing Complement Clause Subjects in Malagasy
Eric Potsdam and Maria Polinsky

This paper presents and analyzes a biclausal construction in the Austronesian language Malagasy in which the subject of a complement clause is not expressed but is interpreted as coreferential with a higher noun phrase: Manantena Rabe_i [mba/fa hividy fiara Ø_i] ‘Rabe_i hopes that Ø_i will buy a car.’ We show that there are two different structures here depending upon the choice of complementizer. With the complementizer mba, the construction instantiates finite control: an obligatory referential dependency between an unexpressed subject in a finite clause and a higher antecedent. With the complementizer fa, despite some apparent similarities, the construction is not control. We propose instead an NP Drop analysis in which the missing subject is a null pronominal licensed by an independent process of topic drop.

BOOK REVIEWS

304, Albert J. Schütz, Gary N. Kahāho‘‘omalu Kanada, and Kenneth William Cook. 2005. Pocket Hawaiian grammar: A reference grammar in dictionary form.
Reviewed by Ray Harlow

306, I Wayan Arka. 2003. Balinese morphosyntax: A lexical-functional approach.
Reviewed by Paul Kroeger

313, Andrew Pawley, Robert Attenborough, Jack Golson, and Robin Hide, eds. 2005. Papuan pasts: Cultural, linguistic and biological histories of Papuan-speaking peoples.
Reviewed by Angela Terrill

321, Alexandre François. 2003. La Sémantique du prédicat en Mwotlap (Vanuatu).
reviewed by Darrell Tryon

ARTICLES

269, Bound Nominal Elements in Äiwoo (Reefs): A Reappraisal of the “Multiple Noun Class Systems”
Åshild Næss

The little-described Reefs-Santa Cruz (RSC) languages are usually assumed to be of mixed Papuan-Austronesian origin, though attempts at linking them systematically either to known Papuan or Austronesian languages have yielded meager results. One of the main arguments in the literature for the presence of “Papuan structures” in the RSC languages has been the claim that the languages have complex systems of noun classes, described in any detail only for the largest RSC language, Reefs or Äiwoo. This paper examines the claim that Äiwoo has one or more noun class systems, based on fieldwork material. It draws two main conclusions: First, the phenomena in question cannot be felicitously analyzed as noun classes in the usual sense of the term, and bear no obvious resemblance to the Papuan-style gender systems to which they have been compared. Rather, they are bound nominal elements of which some have a mainly nominalizing function, whereas others show characteristic properties of classifiers or class terms. Second, there is little or no evidence that the presence of these elements in the language indicates a non-Austronesian origin or influence. On the contrary, the classifier or class-term system, in fact, has obvious parallels in a number of Oceanic languages of Vanuatu. While this does not entail any conclusions about the genetic status of Äiwoo, or of RSC in general, it is clear that the so-called noun classes do not constitute evidence of a Papuan link.

297, The Benefactive Construction in South Efate
Nick Thieberger

The benefactive construction in South Efate employs a prepositional phrase in the position immediately preceding the main verb. This position facilitates the expression of an additional participant in a sentence without competing for slots held by other participants (core arguments or adjuncts). Possessive morphology encoding the benefactive has been noted for other Oceanic languages, with distinct word-order marking a final stage of grammaticalization of the benefactive. While South Efate shares features with southern Vanuatu languages, it is shown that a preverbal benefactive is an areal feature of several languages to the north of South Efate, potentially supporting South Efate’s position in the Central Vanuatu subgroup.

311, The Origin of the Kelabit Voiced Aspirates: A Historical Hypothesis Revisited
Robert Blust

Kelabit, an Austronesian language spoken in northern Sarawak, has a typologically rare series of true voice aspirates. Contrary to claims in the general phonetics literature, these segments cannot be analyzed as consonant clusters. In earlier publications such as Blust (1969, 1974a,b), the origin of the Kelabit voiced aspirates was attributed to ancestral clusters of voiced obstruent + sibilant that arose from syncope in the reflexes of PAn *bVS, *dVS, *jVS, *zVS, and *gVS. This hypothesis required the expansion of a number of Proto-Austronesian reconstructions through the addition of a vowel after *S, or the addition of a syllable with *S, as in PAn *tebuSu > Proto–North Sarawak *təbSu > Kelabit təb^huh (for earlier PAn *tebuS) ‘sugarcane’, *qapejuSu > Proto–North Sarawak *pədSu > Kelabit pəd^huh (for earlier PAn *qapeju) ‘gall (bladder)’, or PMP *ijuSuŋ > Proto–North Sarawak *idSuŋ > Kelabit id^huŋ (for earlier *ijuŋ) ‘nose’. Problems with this analysis were pointed out by other scholars, but the alternatives that they proposed were not entirely satisfactory. Some aspects of these alternative proposals, together with additional observations, have now led to a new analysis, in which Kelabit b^h, d^h, g^h, and the historically related segments in other North Sarawak languages are derived from earlier voiced geminates.

339, The Semantics and Pragmatics of Irrealis Mood in Nyulnyulan Languages
William B. McGregor and Tamsin Wagner

The languages of the small Nyulnyulan family of the far northwest of Western Australia all exhibit a grammatical category traditionally dubbed irrealis. In this paper we describe the grammatical expression of this category, and its range of meanings and uses. It is argued that these can be accounted for as contextual senses or pragmatic inferences based on a single encoded core meaning, that the referent situation is construed by the speaker as unrealized. This semantic component remains invariant across all uses of the category, and is not defeasible. Contra claims by some investigators, the realis-irrealis mood contrast is fundamental, and encapsulates a viable conceptual contrast between real and unreal events; epistemic and deontic notions of probability, necessity, desirability, and the like are secondary pragmatic inferences. The irrealis is thus a modal category that can grammaticalize in human languages; indeed, it is a communicatively useful category. We explicate the nature of the conceptual contrast between the construed real and unreal. It is further argued that the notion of scope is essential to an understanding of the irrealis, and its interaction with other mode-like categories. Finally, we situate the Nyulnyulan irrealis in the wider cross-linguistic context of irrealis.

380, Numeral Classifiers and Counting Systems in Polynesian and Micronesian Languages: Common Roots and Cultural Adaptations
Andrea Bender and Sieghard Beller

Polynesian and Micronesian languages inherited a decimal number system from Proto-Oceanic, and individually extended it on one or more dimensions: in length by adding terms for larger numbers, in breadth by specifying numeral classifiers for certain objects (prevailing in Micronesia), and in factor by introducing a larger counting unit (prevailing in Polynesia). Specific counting systems are characterized by a combination of these features: They are based on larger counting units (multiplication function) and apply to certain objects only (object specificity). This paper surveys the distribution of each extension type in Polynesian and Micronesian number systems, characterizes the features that they share, and analyzes the constitutive role that numeral classifiers play for specific counting systems. It is concluded that in most of these languages, number systems are composed according to similar principles, while the divergence in classifiers, objects of reference, and factors chosen results from cultural adaptations, some of which might have been responses to socioeconomic requirements and served purposes of cognitive facilitation.

404, On the Multifunctionality of Compound Prepositions in Indonesian
Dwi Novirini Djenar

This paper examines the assumption that the function of compound prepositions in Indonesian (e.g., di atas ‘on top of’, ke dalam ‘into’) is to add specificity. It is argued that, although compounds indeed serve this semantic function, they are not limited to it. Based on a study of a corpus of spoken and written Indonesian, it is shown that in many instances compounds do not add any semantic information other than what is expressible by their simple form counterparts. Compounds may be used simply to render explicit what is inferable from general knowledge. It is further argued that compounds are a marked category. Their distribution tends to be correlated with medium and type of discourse. In addition, as the more explicit or marked member of the simple/compound opposition, compounds serve a similar anaphoric function to full NPs. They can be considered as nonnominal markers that track locations and serve a number of other discourse functions, which include confirming a previously mentioned location, adding an affective dimension to an utterance, and marking discourse transitions. Compounds are therefore claimed to be multifunctional; and while in some instances either the semantic or pragmatic function predominates, in other instances both functions interact to produce the desired interpretation.

429, Niuean and Eastern Polynesian: A View from Syntax
Yuko Otsuka

While the Tongic subgroup of Polynesian, consisting of Niuean and Tongan, is defined by a number of shared innovations, Niuean is known to exhibit phonological and lexical features that can be taken as evidence of Eastern Polynesian (EPn) influence. Niuean also shows syntactic features that are prominent among EPn languages, but absent in Tongan. This paper examines whether these syntactic features can be taken to be shared innovations and hence constitute a basis for a subgroup including EPn and Niuean. The current study suggests that convergent development is the probable explanation for the syntactic similarities between Niuean and EPn.

457, On the Origin of Philippine Vowel Grades
Lawrence A. Reid

The concept of vowel grade by which morphological features in some Indo-European languages are signaled by change in the quality of the vowel of a given form has long been recognized. More recently, the term has also been applied to the variation in vowels that occur in some case-marking prepositional forms in Austronesian languages. The purpose of this paper is to demonstrate some of the processes by which vowel grades developed in some Philippine languages. These processes include what has been referred to elsewhere as vowel-grade harmony, an assimilatory process by which the vowel of one case-marking preposition copies that of another. Noun phrases in many Philippine languages are commonly described as being introduced by “phrase markers” that specify certain syntactic and semantic features of the noun phrase they introduce. These are typically unstressed clitic forms having a CV or CVC shape. However, the quality of the vowel varies from language to language. Thus, in Ivatan, the forms that introduce common noun phrases all have an u vowel, while those that introduce personal noun phrases all have an i vowel; in Tagalog the forms that introduce common noun phrases all have an a vowel, while those that introduce personal noun phrases all have an i vowel, like Ivatan. Recognizing that the similarity in vowel quality of “phrase markers” in these languages is commonly the result of vowel-grade harmony and not necessarily the result of regular phonological change provides an explanation for the multiple irregularities that are found in attempting to reconstruct the protoforms of “phrase markers.”

474, The Pronoun System in Galeya: Arguments against a Clitic Analysis
Fumiko S. Yamada

There are two types of pronominal elements for marking subject and object in Galeya (an Oceanic language spoken on Fergusson Island, Papua New Guinea): those that occur independently (full pronouns), and those that must be attached to verbs. As the latter show characteristics shared by both pronouns and agreement affixes, they can be considered to be pronominal clitics. Although the distinction between pronominal clitics and agreement affixes has not always been made explicitly in previous studies of Oceanic languages, it is important for the analysis of the sentence structure of Galeya, in which both types of pronominal elements are present. If we assume that one type of pronominal element is an independent pronoun and the other is an agreement affix, the former should be an argument. If, on the other hand, we take the former to be an independent pronoun and the latter a pronominal clitic, the clitic is taken to be an argument and the independent pronoun an adjunct. By applying certain tests, it is determined that the subject and object markers attached to verbs are grammatical agreement affixes rather than pronominal clitics. This analysis leads us to the conclusion that the word order in Galeya is SOV, the same pattern as is found in neighboring languages.

491, In Memoriam, Per Hage, 1935–2004
Jeff Marck

NOTES

497, Some Observations on Proto-Austronesian *t to k
Anthony P. Grant

BOOK REVIEWS

501, I Wayan Arka and Malcolm Ross, eds. 2005. The many faces of Austronesian voice systems: Some new empirical studies.
Reviewed by Norvin W. Richards

505, John Lynch, ed. 2003. Issues in Austronesian Historical Phonology.
Reviewed by R. David Zorc

INDEX

517, Index of Languages in Volume 45

ARTICLES

1, Reference Point Constructions, the Underspecification of Meaning, and the Conceptual Structure of Palauan er
Michael B. Smith

The Palauan grammatical morpheme er is a preposition-like word whose wide variety of uses seem unrelated to each other and whose semantic function (if any) is obscure. Reminiscent of English of in signifying an intrinsic relation between two entities, the meaning of er appears to be even more schematic and context dependent. It is argued that er’s basic conceptual structure, and therefore its meaning, resides in its designation of an abstract reference point construction in which its object serves as a reference point with respect to which other entities (either things or relations) are construed to be located in some kind of physical or abstract domain. Consequently, the meaning of er is highly schematic and underspecified. Its apparently unrelated senses are related to each other in reflecting instantiations of this construction when construed against different backgrounds in particular contexts.

21, The Phonetics and Phonology of “Definitive Accent” in Tongan
Victoria Anderson and Yuko Otsuka

The so-called definitive accent (DA) in Tongan has been analyzed in various ways in the literature: as stress shift from penultimate to final vowel, as simultaneous stress reduction on a penult and stress addition on an ultima, and as addition of a syllable by repetition of the final vowel. This study investigates each of these analyses empirically in order to establish the phonology of DA in Tongan. Our findings support Melenaite Taumoefolau’s proposal that definite NPs are formed by repetition of the NP-final vowel, and thus a morphological analysis of DA as reduplicative suffixation. Moreover, our findings substantiate an account of Tongan in which stress is unexceptionally penultimate in a foot, and in which “long vowels” and “diphthongs” are to be considered sequences of two syllables, as suggested by Taumoefolau.

43, Retained and Introduced Final Consonants in Vitu
René van den Berg and Peter Bachet

In this article we present data on final consonants in Vitu, a Western Oceanic language spoken in West New Britain (PNG). Although Vitu has as a rule lost Proto-Oceanic final consonants, we show that such consonants still emerge in nouns that are inalienably possessed. Surprisingly, a number of nouns that ended in a vowel in Proto-Oceanic also show a final consonant in Vitu. The reason for the introduction of these consonants is unclear.

53, Grammatically Marked Ideophones in Numbami and Jabêm
Joel Bradshaw

Numbami and Jabêm, two Austronesian languages of Morobe Province, Papua New Guinea, share a trait rarely recorded for Oceanic languages: they each have a rich class of morphologically marked ideophones. In Numbami, the marker is a suffix -a(n)dala unique to ideophones, but clearly related to NUM andalowa ‘path, way, road’ and Proto-Oceanic *jalan. In Jabêm, the marker varies depending on the length of the ideophone itself, just as it does on other classes of adverbs. The shortest forms are usually followed by tageη ‘one, at once, only’, longer forms take the regular adverbial enclitic -geη, and fully reduplicated forms are usually left unmarked.

64, Is There Pasif Semu in Jakarta Indonesian?
Peter Cole, Gabriella Hermon, and Yassir Tjung

This paper examines whether pasif semu (P2), one of two passives in Standard Indonesian, exists in Jakarta Indonesian. In P2 in Standard Indonesian the verb appears in bare stem form, the theme has been promoted to surface subject, but—unlike European-style passives—the actor has not been demoted to adjunct. This is to be contrasted with the di-passive (P1), in which di- is prefixed to the verb, the theme is promoted to surface subject, and the agent is demoted to adjunct. Standard Indonesian is to a large extent an artificial language, the creation of language planners rather than of its speakers. In contrast, Jakarta Indonesian is the native language of the natives of Jakarta, and, through the influence of TV, movies, and radio, is heard with ever greater frequency throughout Indonesia. In our study (drawn from a variety of corpora), a clear contrast is seen between child and child directed speech, on the one hand, and adult to adult speech, on the other. P2 is essentially nonexistent in the former but exists robustly in the latter. We argue that child speech and child directed adult speech represent basilectal Jakarta Indonesian, in which P2 has been lost, and that the adult to adult corpora represent a mesolectal variety of Jakarta Indonesian that shows a number of influences from Standard Indonesian not found in the child and child directed corpora.

91, The Pragmatics of Case Marking in Saisiyat
Fuhui Hsieh and Shuanfan Huang

Saisiyat, an Austronesian language spoken in northwest Taiwan, has an elaborate case marking system for nominals and pronominals, but the nominative case is often zero marked. Based on natural spoken data, we demonstrate that this absence of nominative case markers is closely tied to the ongoing word order shift from a V-initial language to a strongly subject-initial language, especially in Agent-Focus sentences, rendering any marking for the nominative case pragmatically redundant. Nominative case marking remains a pragmatic option for the presentative construction to introduce a new referent into discourse. A second issue addressed concerns the coding of the Recipient in a ditransitive sentence. We present an unusual case of biaccusative constructions where the semantic role of the Recipient is marked by either the dative or the accusative case marker, the choice being pragmatically determined by the spatial or psychological distance between the Agent and the Recipient. If the Recipient is perceived as being within the spatial or psychological sphere of influence of the Agent and consequently likely to be affected by the action of the Agent, the accusative case is preferred; elsewhere the dative is used. The effect of this is to produce biaccusative constructions, because the Theme in ditransitive sentences is always coded by the accusative case. Case marking in Saisiyat therefore cannot be dissociated from the discourse-pragmatics of language use and an understanding of the nature of the word order change the language is currently undergoing.

110, The Prosodic Realization of Negation in Saisiyat and English
Wen-yu Chiang, I Chang-Liao, and Fang-mei Chiang

This study investigates the prosodic realization of negation in Saisiyat, an endangered aboriginal Austronesian language of Taiwan, and compares the prosodic properties of its affirmative and negative sentences with those of British English. In order to test Yaeger-Dror’s “Cognitive Prominence Principle,” according to which cognitively prominent items (such as negators) should be prosodically marked, we measure the F₀ peak, the intensity peak, and duration of lexical items appearing in affirmative and negative sentences. Our results indicate that sentential subjects are the most acoustically prominent items with respect to F₀ height and intensity in Saisiyat negative sentences, whereas the negator itself is the most acoustically prominent item with respect to F₀ in an English sentence. In addition, the presence of a negator does not significantly change the prosodic parameters of contiguous words in Saisiyat. English, in contrast, exhibits relatively large-scale prosodic differences in both F₀ and intensity between affirmative and negative sentences. This paper suggests that the following typological features can account for the differences observed between Saisiyat and English: (1) the relationship between prosodic prominence and syntactic subjects in Saisiyat, (2) transparency of the negation system in Saisiyat, and (3) the relationship between prosodic prominence and semantically defined focus in English.

133, Numerals in Formosan Languages
Paul Jen-kuei Li

This is a general survey of cardinal numerals in Formosan languages. Most languages distinguish between human and nonhuman numerals, not only in cardinal numerals, but also for terms that have to do with number, such as ‘how many/much’, ‘many/much’, and even ‘few/little’. Some languages have a third set of numerals as used in counting, different from both human and nonhuman numerals. Tables of the numerals 1–10 in the still extant Formosan languages are given in the appendixes. Most Formosan languages retain a decimal system, although a few numerals may have been modified in some of the languages. Pazih is the only language that has nearly a quinary system. Numerals may be derived from other numeral stems by addition, subtraction, or multiplication, and some are unique to Formosan languages. Numerals may function either as nouns or verbs, depending on their syntactic position, and they may appear in simple or derived form.

153, Argument-marking and the Distribution of wh-Phrases in Malagasy, Tagalog, and Tsou
Paul Law

This paper argues that postverbal wh-phrases in Malagasy, Tagalog, and Tsou are subject to the same general constraint on marking of trigger arguments. It shows that the trigger is not necessarily definite or specific; the nonoccurrence of wh-phrase trigger in postverbal position therefore cannot be reduced to the definite/specific constraint on the trigger argument. There exists evidence, little noticed in the literature, that wh-phrase trigger is sometimes possible in postverbal position. I claim that postverbal wh-phrase trigger is possible just in case it can be marked independently in the same fashion as postverbal non-wh-phrase triggers.

191, On the Edge of Grammar: Discourse Particles in Niueans
Diane Massam, Donna Starks, and Ofania Ikiua

This paper provides a first examination of discourse particles in Niuean, a Polynesian language of the Tongic subgroup, using data from interviews from the Pasifika Languages of Manukau project. Although there has been some research on grammatical particles in Niuean, there has been little or no reference to the structure and function of discourse particles. We provide an initial framework for categorizing the various types of discourse particles in the language, and give a preliminary account of the function of two affirmative discourse particles, haia and mitaki.

SQUIBS

206, The Phoneme Inventory of the Aita Dialect of Rotokas
Stuart Robinson

Rotokas is famous for possessing one of the world’s smallest phoneme inventories. According to one source, the Central dialect of Rotokas possesses only 11 segmental phonemes (five vowels and six consonants) and lacks nasals while the Aita dialect possesses a similar-sized inventory in which nasals replace voiced stops. However, recent fieldwork reveals that the Aita dialect has, in fact, both voiced and nasal stops, making for an inventory of 14 segmental phonemes (five vowels and nine consonants). The correspondences between Central and Aita Rotokas suggest that the former is innovative with respect to its consonant inventory and the latter conservative, and that the small inventory of Central Rotokas arose by collapsing the distinction between voiced and nasal stops.

210, Anomalous Liquid : Sibilant Correspondences in Western Austronesian
Robert Blust

The present contribution is meant to draw attention to two or more apparent sound correspondences among the languages of the Philippines and western Indonesia that involve liquid phonemes in one language or set of languages and sibilants in another language or set of languages. The data are problematic, and in my view do not justify the reconstruction of new protophonemes. At the same time, the recurrent regularities appear to be greater than chance, and so deserve a public airing. In either case, there would seem to be a lesson to be learned from the set of comparisons considered here: either we must reckon with at least two protophonemes that were overlooked by Dempwolff, or we must recognize that factors other than divergent development from a common ancestor may sometimes produce recurrent phonological similarities that bear a striking mock resemblance to true sound correspondences.

217, Tutuba Apicolabials: Factors Influencing the Phonetic Transition from Apicolabials to Labials
Maho Naito

The Tutuba language, spoken in the Republic of Vanuatu, has speech sounds known as apicolabials, a sound-type found in very few languages worldwide. It is thought that the apicolabials found in various languages of Vanuatu shifted as follows: 1. from labials in the protolanguage (*labials) > apicolabials, and 2. *labials > apicolabials > dentals/alveolars. However, the shift 3. *labials > apicolabials > labials has also been hypothesized. A phonetic change from apicolabials to labials, equivalent to 3., is currently taking place in the Tutuba language. It is thought that the main factor behind this change is the influence of other languages, including Bislama, the lingua franca of the area. A geographical analysis of available information shows that while the languages in which phonetic change 1. occurred are spoken on isolated islands and on the coasts of islands, the majority of the languages that have undergone phonetic changes 2. and 3. are spoken inland. This suggests the possibility that hypothesized phonetic change 3., from apicolabials to labials, occurs spontaneously as the result of external factors—the exposure to other languages.

229, Coronals and Velars: Support for Blust
Mark Donohue

Robert Blust raises the issue of the *t > k change that is widely attested in Austronesian languages, but infrequently in other language families. He offers both structural and perceptual explanations for the “naturalness” of this change, but admits that the data raise more questions than can be answered. I offer support for the view that this change is not unnatural, based on the distribution of stop types cross-linguistically, and the patterns that are found. I introduce another kind of argumentation, that of typologically determined systemic naturalness, in the spirit of Evolutionary Phonology.

NOTES

242, Some Notes on Nhanda, as Spoken by Mrs. Lucy Ryder† (1919–2003)
Juliette Blevins

BOOK REVIEWS

247, Isabelle Bril and Françoise Ozanne-Rivierre, eds. 2004. Complex predicates in Oceanic languages: Studies in the dynamics of binding and boundedness.
Reviewed by Joel Bradshaw

250, Donald C. Laycock. 2003. A dictionary of Buin, a language of Bougainville, ed. by Masayuki Onishi.
Reviewed by Peter C. Lincoln

257, Senft, Gunter, ed. 2004. Deixis and demonstratives in Oceanic languages.
Reviewed by Paul Geraghty

262, Robert Blust. 2003. A short morphology, phonology and vocabulary of Kiput, Sarawak.
Reviewed by Adrian Clynes

ARTICLES

307, Reference to Motion Events in Six Western Austronesian Languages: Toward a Semantic Typology
Shuanfan Huang and Michael Tanangkingsing

The language of motion events is a system used to specify the motion of objects through space with respect to other objects. Recent research has shown that languages differ in the relative saliency of manner or path they focus on in motion event descriptions. These can be thought of as different strategies dedicated to specifying the spatial relationship between objects in motion and the landmark object. We propose a four-way typology based on the narrative data from six western Austronesian languages. Evidence is presented that each of the languages examined typically has a preferred strategy for describing motion events and that each has a distinct narrative style. These six languages are shown to share the commonality of giving greater attention to path information in motion events. Path salience in the encoding of motion clauses appears to exhibit a strong diachronic stability, suggesting that Proto-Austronesian was probably also a path-salient language.

341, Iconicity as Evidenced in Saisiyat Linguistic Coding of Causative Events
Shuping Huang and Lily I-wen Su

Causatives have been subjected to intensive scrutiny by linguists in recent years. Cross-linguistic studies suggest that the formation of causatives reflects the real world perception of cause-result relations, though some studies contradict this finding. The aim of this study is to explore the underlying principles that determine the way an event is encoded linguistically in Saisiyat, a Formosan language. Following Croft’s model of idealized single events, we promote the study of causatives to the discourse level. The results show that the iconicity of language is reflected in the ordering of linguistic elements as well as their grammatical integrity.

357, Malayo-Sumbawan
Alexander Adelaar

Using phonological and lexical evidence, this paper seeks to demonstrate that Balinese, Sasak, and Sumbawa (which form an exclusive subgroup) are more closely related to Malay than they are to Javanese. It concludes—especially on the basis of phonological evidence—that the previously posited Malayo-Javanic subgroup should be replaced by a “Malayo-Sumbawan” subgroup that includes Malayic, Chamic, and the Balinese-Sasak-Sumbawa group in one branch, and Sundanese and Madurese in two other branches. Javanese is excluded from this subgroup.

389, The Apicolabial Shift in Nese
John Lynch

Nese is one of a dozen or so languages/dialects spoken in the south Santo–north Malakula area of Vanuatu that reflect original simple bilabials before nonround vowels as apicolabials. In some of these languages, the apicolabials subsequently became dentals/alveolars. Nese is unusual, however, in the inconsistency of its reflexes: the most frequent reflex of Proto-Oceanic *b in this environment is indeed the apicolabial stop b, but the most common reflex of *m is alveolar n, while with *p both apicolabial v and labiodental v occur with roughly equal frequency. This paper attempts to explain this variation, and also attempts to explain why *p behaved far less consistently across a range of languages in this area than did *b and *m.

404, Saisiyat as a Pitch Accent Language: Evidence from Acoustic Study of Words
Wen-yu Chiang and Fang-mei Chiang

This paper investigates the acoustic realization of lexical-level accent in Saisiyat, an endangered aboriginal language of Taiwan. Accent in Saisiyat usually falls on the ultimate syllable of content words. This phenomenon has been described in previous studies as either “stress” or “accent.” Our measurements and analysis of various prosodic parameters of syllable rhyme (Fo height at onset, offset, peak, and valley, as well as pitch range, duration, slope, peak alignment, and intensity peak) suggest that accent in Saisiyat should be classified as pitch accent, because lexical accent is realized by means of specific Fo patterns, rather than duration and intensity. Thus, among three typological categories that have been proposed for languages (lexical tone, lexical stress, and lexical pitch accent), we propose that Saisiyat belongs to the category that has lexical pitch accent.

427, Syntactic and Lexical Factors Conditioning the Diffusion of Sound Change
Mark Donohue

A sound change may propagate through a language in different ways. Different studies attest sound changes spreading at different rates through different phonological and/or phonotactic environments, diffusing through the speaker population (or through different dialects) in different ways, or simply spreading differentially through the lexicon. In Palu’e there is evidence for a sound change applying at different rates for different grammatical categories, with the sound change advancing in the small set of bound grammatical morphemes perhaps more completely than in free lexemes. This is evidence that syntactic information on parts of speech can affect the diffusion of a sound change through a language, and that bound forms are not necessarily more conservative than free lexemes when it comes to phonological change.

443, Unraveling the History of the Vowels of Seventeen Northern Vanuatu Languages
Alexandre François

Data collected on the 17 languages spoken in the Banks and Torres Islands (northern Vanuatu) reveal strikingly diverse vowel systems, differing both in the quality and the quantity of their phonemes. Except for Mota, which still perpetuates the five vowels of Proto-Oceanic, the languages of this area have historically increased their inventories to as many as 13 and even 16 vowels. The aim of this paper is to track the systematic correspondences between modern languages and their common ancestor, and to reconstruct the processes that led to the present-day phonemic diversity. The phonemicization of new vowels, including diphthongs and long vowels, is shown to result from stress-induced vowel reduction and metaphony. This general process of “vowel hybridization” yielded results that differed from one language to another, and sometimes within the same language. After describing and classifying the various patterns of sound changes attested, this paper shows how a proper understanding of vowel hybridization proves indispensable for the reconstruction of both the lexicon and the historical morphology of these northern Vanuatu languages.

SQUIBS

505, Syntax of Verbal Nouns in Marquesan
Viktor Krupa

Verbal nominalization is a productive process in Marquesan and in other Polynesian languages. The markers are predominantly suffixal in nature. The transient character of the verbal nominalizations is underscored by their partial compatibility with both nominal and verbal particles. Their frequency of occurrence varies from text to text, as can be seen when the two main types of data used in this study are compared: folkloristic material, and a translation of St. John’s gospel. Attention is given to the syntactic functions of verbal nominalizations and to the ways in which their logical subjects are marked.

517, The Role of Phonological Predictability in Sound Change: Privileged Reduction in Oceanic Reduplicated Substrings
Juliette Blevins

Phonetic reduction is more likely when a word is predictable or recoverable independent of acoustic information. Attributing higher rates of phonetic reduction to lexical predictability has implications for subword domains. A range of historical developments in Oceanic support this position. In reduplication, where the content of the reduplicated substring is wholly predictable and recoverable from the base, the reduplicant undergoes leniting sound changes more readily than the base, and more readily than other prosodically comparable domains. Synchronic consequences of these developments challenge models associating reduplication with the emergence of unmarked structures.

527, Word Order in New Guinea: Dispelling a Myth
Mark Donohue

It has been claimed that the appearance of SVO order in a non-Austronesian language of New Guinea and its environs is evidence of contact and influence from Austronesian languages. I suggest that, because SVO is an innovative order in Austronesian languages as well, the influence might well be in the other direction, from a period before the popularization of SOV in New Guinea.

537, Liver and Lungs: A Semantic Dyad in Austronesian Languages
Robert Blust

A number of Austronesian languages reflect PAn *qaCay ‘liver’ either singly or in combination with a modifying word in the meaning ‘lungs’. A smaller set of languages reflects PAn *baRaq ‘lungs’ in the meaning ‘liver’. These terms are paired in the formal dyadic ritual language of Roti, and so raise the question whether the observed semantic crossover might be a historical residue of culturally determined semantic relationships in a similar ritual language that existed by at least Proto–Malayo-Polynesian times. However, because the interchange of meanings in reflexes of *qaCay and *baRaq is unparalleled by other examples of a similar type, this interpretation is rejected, and it is concluded that the crossover linking liver and lungs is a practical consequence of similarities based on shape as seen in the internal organs of butchered animals, particularly pigs.

BOOK REVIEWS

544, John Lynch, Malcolm Ross, and Terry Crowley. 2002. The Oceanic languages
Reviewed by Robert Blust

559, Bethwyn Evans. 2003. A study of valency-changing devices in Proto Oceanic
Reviewed by Frantisek Lichtenberk

INDEX

564, Index of Languages in Volume 44

ARTICLES

1, Some Notes on the Proto-Austronesian Words for ‘Water’
Isidore Dyen

Dempwolff’s reconstruction of *wayeR ‘water’ is reviewed. The meaning ‘fresh water’ is suggested. The initial *w may have been dialectally distributed with a probably older *kw. Chamorro’s treatment of *w is formulated and its relevance to the evidence for *kw is challenged. The articulation of *R may have not been a velar spirant in all PAn dialects in view of the agreement of Bonfia, Palauan, and Yapese on sibilance. In some dialects there may have been a paragogic vowel after the *R and that vowel itself may have been followed by a glottal stop. Finally, evidence for a medial *S has been found.

12, What the Rawas Dialect Reveals about the Linguistic History of Rejang
Richard McGinn

Core vocabularies of five major dialects of Rejang are derived from Proto–Malayo-Polynesian etyma mediated by a reconstructed protolanguage, Proto-Rejang, crucial evidence for which is provided by Rawas, a previously neglected dialect. Every etymon turns out to be reconstructable on the basis of just two dialects—either Rawas and Pesisir or (more often) Rawas and Kebanagung. One clear conclusion is that the Rawas area represents the oldest Rejang settlement in Sumatra. While a few minor claims made in previous work on Rejang historical phonology are corrected or refuted by the Rawas evidence, the most important findings are confirmed. Thus, (a) individual Rejang dialects have undergone more changes (splits and mergers) of Proto–Malayo-Polynesian vowels than any other known Austronesian group; (b) Proto-Rejang underwent two accent shifts, first to a Malay-type pattern (a modified form of penultimate word-stress) and then to the contemporary word-final stress pattern; and (c) complex conditions, including vowel harmony conditions, are needed to preserve regularity in the majority of sound changes affecting Rejang dialects.

“The dialect [that] is probably most important from a historical point of view . . . is the Jang Abeus dialect, spoken in the upper reaches of the river Rawas. . . . In 1941 it still had the final -l in such words as biyol, Lebong biyoa water.”—P. Voorhoeve, pers. comm. reported in Blust 1984:448, n.2.

65, Positional Slots in Saliba Complex Verbs
Anna Margetts

Saliba, an Oceanic language of Papua New Guinea, has complex predicates in which two or more stems combine to form a single word. The stems in these complex verbs can express a number of functions including cause and result, manner, directionality, and other adverbial-like functions. It is possible to identify a number of positional slots in these constructions, based on the sequential ordering and cooccurrence restrictions of stems. The slots in complex verbs host different classes of stems, and different types of relations between the slots can be distinguished.

90, Final Consonants in Remote Oceanic
John Lynch

This paper examines retention and loss of Proto-Oceanic final consonants in three different groups of Remote Oceanic languages in the light of Blevins’s recent discussion in this journal of the unnaturalness of final C-loss within the wider Austronesian family. Languages of (i) northwest Malakula, (ii) southern Vanuatu, and (iii) the Loyalty Islands and northern New Caledonia do not experience the total or near-total loss of final consonants that is commonplace elsewhere within Remote Oceanic. Each group shows partially different patterns of retention and loss from the others, and there are also some differences between members of the same group; but in no case was there a rule deleting all final consonants. There were, however, rules deleting all final vowels, and this V-deletion process may have created a situation in which final consonants were more resistant to loss than in languages with predominantly open final syllables. I will suggest that in each of these three areas consonants were lost by means of natural rules, and that a series of natural rules, rather than a single unnatural rule, may be the explanation for cases in other subgroups where all final consonants have been lost.

113, On the Notion “Adjective” in Toqabaqita
Frantisek Lichtenberk

Toqabaqita has three classes of lexemes that function as modifiers of nouns. One class, the largest one, contains stative intransitive verbs. These lexemes (and similarly functioning lexemes in other Oceanic languages) are a counterexample to the typology of parts-of-speech systems of Hengeveld, Rijkhoff, and Siewierska (Journal of Linguistics 40:527–70, 2004), according to which verbs can only function as the heads of predicate phrases. The second class contains only two members, which are nouns. The third class contains only one member. This lexeme, which developed historically from a noun, can function only as a noun modifier and is the sole adjective in the language. In the development of the noun-modifying function of the lexemes in the second and the third classes, dependency reversal took place, whereby the original head position in a noun phrase became the modifier position, and vice versa.

145, Sulka of East New Britain: A Mixture of Oceanic and Papuan Traits
Ger Reesink

Sulka, spoken along Wide Bay in the East New Britain province of Papua New Guinea, is known as an isolate Papuan language. In the area where Austronesian and Papuan languages have been in contact over the last three-and-a-half millennia, this is one of a number of languages that are difficult to classify. The question is often raised whether a language is basically Papuan or Austronesian, with some kind of borrowing from the other linguistic stock. In this paper it is argued that clusters of features can shed light on the genealogical or contact history of a language. On this basis, Sulka can be typified as having ancient Papuan (non-Austronesian) roots, but with a number of morphosyntactic constructions and some vocabulary that are associated with the Oceanic branch of Austronesian, in particular the languages of the St. George linkage. It is hypothesized that some Western Oceanic innovations may actually have originated in Papuan languages such as Sulka.

194, Kena as a Third Type of Malay Passive
Siaw-Fong Chung

Two types of passive in Malay are compared using corpora examples, the di- passive and the kena adversative passive. It is proposed that the latter is constrained by pragmatic specifications related to the contexts in which it appears. In one analysis, 50 instances of each of these two passives collected from two newspapers were compared as to their frequency, degree of transitivity, and two pragmatic functions: their connotations and their register. The results show that the kena adversative passive is lower in frequency, higher in transitivity, less formal in register, and that it has a negative connotation for recipients of the actions when compared with the di- passive. In a second analysis, an additional 100 examples of the kena adversative passive were collected (50 from classical Malay manuscripts and 50 from Internet sources) and compared with the instances from newspaper articles. The results show that the kena passives from all three sources reflect similar features of transitivity and pragmatic use in discourse. Some of the kena instances from the Internet display more colloquial usage than those from the other sources, manifest, for example, by simplified spellings in kena phrases and by the frequent occurrence of code-switching whereby an English past participle form is used following kena (as in kena caught).

SQUIB

215, A Note on the History of Genitive Marking in Austronesian Languages
Robert Blust

Three forms of the genitive phrase marker have been proposed for Proto-Austronesian: *ni, *na, and *nu. While there is universal agreement on the form of this reconstructed system, reconstruction of the meanings/functions of these forms has been far more problematic. It is argued that *nu marked the genitive of common nouns, while *ni and *na marked the genitive of singular and plural personal nouns respectively. The evidence supporting this reconstruction forces a reconsideration of the *ni-phrases posited by Robert Blust in this journal in 1974: “Proto-Austronesian syntax: The first step” (13:1–15).

IN MEMORIAM

223, In Memoriam, Terry Crowley 1953–2005
John Lynch

Early in the third week of January this year, Oceanic linguists and other colleagues and friends received with utter disbelief the devastating news of Terry Crowley’s sudden death at his home in Hamilton, New Zealand, on the weekend of January 15–16. There had been no warning signs, no immediately preceding period of hospitalization, no known illness to give us warning. A fit fifty-one-year-old nonsmoker (some might say virulent anti-smoker), jogger, moderate drinker, careful of what he ate, he died suddenly from a severe heart attack.

Terry made major contributions in a number of areas: to descriptive studies of Vanuatu languages; to the study of Bislama and, more generally, to creole studies; to the study of the history of the Oceanic languages; to literacy and other sociolinguistic and applied studies in the Pacific; to theoretical and historical linguistics; and, in his younger days, to Australian Aboriginal linguistics. He was, perhaps, the most prolific publisher among all Oceanic linguists. In a reference written in support of his application for a full professorship, Andrew Pawley said of his productivity: “I don’t know how he does it. The quantity is staggering. . . . And the quality is uniformly high. His books are meticulously researched and well written, several are based on his own extended fieldwork, and all will stand for a long time as important reference works.”

At the time of his death, he was directing a rather large research project on the languages of Malakula, and was himself actively involved in working with about half a dozen languages on that island, most of them moribund. His sudden death has cut short what should have been another fifteen or twenty years of that same productivity, as well as taking away a friend and admired colleague. David Walsh summarized the situation in an email message to me: “He was a good linguist and a good bloke—there’s no bloody justice in the world!”

REVIEW ARTICLE

242, Australian Languages Reconsidered: A Review of Dixon (2002)
Nicholas Evans

Is the Australian linguistic area, because of its unique history, one in which the established methods of historical and comparative linguistics have limited appropriateness? Do neighboring languages in this situation come to share an “equilibrium level” of 50 percent basic vocabulary regardless of their degree of genetic relatedness? Is the Pama-Nyungan grouping totally without foundation and something that must be discarded if any progress is to be made in studying the nature of the linguistic situation in Australia? Are Australian scholars more hesitant than scholars elsewhere to criticize the work of colleagues? These and other “deliberately unorthodox” views of R. M. W. Dixon set forth in Australian Languages (Cambridge Univ. Press, 2002) are countered, while conceding that the book brings together an enormous amount of historically and typologically relevant material in one place.

BOOK REVIEWS

287, Ger P. Reesink, ed. 2002. Languages of the Eastern Bird’s Head
Reviewed by Mark Donohue

301, Terry Crowley. 2004. Bislama reference grammar
Reviewed by Miriam Meyerhoff

ARTICLES

277, Phonation Types in Javanese
Ela Thurgood

The somewhat misnamed breathy voiced vowels of Javanese have been retermed “slack voiced” by Ladefoged and Maddieson (1996). This paper extends the analysis of Javanese vowels to a wider range of vowels, describes in precise detail what the acoustic characteristics of Javanese slack voice are, and examines the acoustic characteristics of the “emphatic” voice quality often used in Javanese–the acoustic qualities of which are usually associated with what is more typically called breathy voice. I find that it is necessary to extend the range of voice qualities found in Javanese from stiff and slack to include the distinct breathy voice used for emphasis.

296, The Phonetics of Paicî Vowels
Matthew Gordon and Ian Maddieson

This paper presents results of a phonetic study of the vowel system of Paicî, an Austronesian language of central New Caledonia. The Paicî vowel system is of phonetic interest for both its three-way lexical tone contrast, rare among Austronesian languages, and its relatively large inventory of both oral and nasalized vowels. The large number of nasalized vowels is rare not only from an Austronesian perspective, but also is typologically atypical throughout the world. This paper focuses on the analysis of qualitative aspects of both the oral and nasalized vowels of Paicî. It is shown that vowel qualities posited in previous research on Paicî are phonetically differentiated, with the contrast between certain nasalized vowels being more subtle than the contrast involving their oral counterparts. In addition, the phonetic realization of the three tones of Paicî is discussed.

311, The Efate-Erromango Problem in Vanuatu Subgrouping
John Lynch

While the languages of Erromango clearly belong to the Southern Vanuatu family, and the languages of Efate also clearly belong to the Central Vanuatu subgroup, there are quite a few nonlexical innovations shared by just these languages and none of their close relatives (and, remarkably, apparently very little evidence of lexical sharing). This paper investigates these innovations, and proposes a settlement pattern for central and southern Vanuatu that explains these changes while still recognizing the distinct identity of the two subgroups.

339, The Argument Structure of Verbs with the Suffix –kan in Indonesian
Peter Cole and Min-Jeong Son

The verbal suffix -kan in acrolectal Indonesian gives the appearance of being a homonymous form with multiple functions. In many sentences the suffix seems to be a causative morpheme; in others it appears to be an applicative affix, while in yet others it seems to be an object marker. We show that these functions are in fact predictable if -kan is a derivational morpheme affecting the argument structure of the verb to which it is affixed. We argue that the role of -kan is to indicate the syntactic licensing of an argument in the argument structure that is not licensed syntactically by the base verb. Thus, the distribution of -kan provides evidence that there exist linguistic generalizations that need to be stated with respect to a distinct level of argument structure rather than with respect to such syntactic levels as S-Structure and Logical Form.

365, *t to k: An Austronesian Sound Change Revisited
Robert Blust

Although the change of *t to k in Hawaiian has been known and commented on for over 150 years, the widespread driftlike character of this development within Austronesian as a whole has generally gone unappreciated. This paper examines 20 historically independent instances of a *t > k change in at least 43 languages. Twelve of these changes are confined to Oceanic languages, seven to languages of eastern Indonesia, and one to western Indonesia. Almost without exception, the change *t > k has followed the loss of *k. In four languages *t > k took place only word-finally, and in two others it appears to be dissimilatory. Both structural and perceptual motivations for the change are considered, and it is concluded that *t > k usually begins as free variation within an enlarged phonological space created by the loss of *k. A few instances are difficult to reconcile with this explanation, and continue to present a challenge to linguistic theory.

411, The Higher Phylogeny of Austronesian and the Position of Tai-Kadai
Laurent Sagart

This paper presents a new higher phylogeny for the Austronesian family, based on three independent lines of evidence: the observation of a hierarchy of implications among the numerals from 5 to 10 in the languages of Formosa and in PMP; the finding that the numerals *pitu ‘7′, *walu ‘8′, and *Siwa ‘9′ can be derived from longer additive expressions meaning 5+2, 5+3, and 5+4, preserved in Pazeh, using only six sound changes; and the observation that the phylogeny that can be extracted from these and other innovations–mostly changes in the basic vocabulary–evinces a coherent spatial pattern, whereby an initial Austronesian settlement in NW Taiwan expanded unidirectionally counterclockwise along the coastal plain, circling the island in a millennium or so. In the proposed phylogeny, Malayo-Polynesian is a branch of Muic, a taxon that also includes NE Formosan (Kavalan plus Ketagalan). The ancestor language, Muish, is deemed to have been spoken in or near NE Formosa. Further evidence that the Tai-Kadai languages, contrary to common sense, are a subgroup of Austronesian (specifically: a branch of Muic, coordinate with PMP and NE Formosan) is presented.

445, From Implicature to Construction: Emergence of a Benefactive Construction in Oceania
Anna Margetts

It has long been observed that in many languages around the world possessive relations and benefactive relations are expressed by the same morphemes. There are examples of functional extension in both directions, with possessive constructions developing from benefactive ones, and vice versa. But there are also claims in the literature about the unidirectionality of this process, predicting that the development from possessive to benefactive constructions should not occur. This paper presents a detailed case study of the development of specialized benefactive expressions in Oceanic languages, where they commonly derive from expressions of attributive possession. The development starts with a possessive construction carrying a pragmatically implicated benefactive reading that gradually becomes grammaticalized and manifested in the morphosyntax of the language. This process may finally result in a benefactive construction that is syntactically and/or morphologically distinct from the expression of possession from which it originates. Syntactically, the process sets off from an object NP consisting of a noun and its modifier. In the process of grammatical change, this modifier is reanalyzed as a separate constituent, syntactically and semantically independent of the object noun. Based on data from Oceanic languages, three stages in the extension from possession to benefaction are identified. Also discussed are the contextual prerequisites for the benefactive implicature to arise in the first place.

469, Old Bikol -um- vs. mag- and the Loss of a Morphological Paradigm
Jason William Lobel

A semantic contrast between verbs taking the -um- paradigm and those taking the mag- paradigm is known to exist in Tagalog and Waray-Waray but is virtually absent in all varieties of Bikol and most Bisayan languages. Evidence is presented from Fr. Marcos de Lisboaís seventeenth-century Vocabulario de la Lengua Bicol that many Old Bikol verbs had contrasting -um- and mag- conjugations. The -um- and mag- conjugations of Old Bikol are compared with those of Waray-Waray, Southern Tagalog, and Rinconada Bikol. After a discussion of the categories of semantic contrast between -um- and mag- verbs in Old Bikol, and how these contrasts were restructured in Modern Bikol, a progression of stages is proposed to explain how the -um- vs. mag- contrast has been lost in Bikol and other Central Philippine languages.

498, Linguists, Literacy, and the Law of Unintended Consequences
Kenneth L. Rehg

In 1970, the Pacific and Asian Linguistics Institute of the University of Hawai’i launched a fourteen-year effort designed to document and support the languages of Micronesia. The first goal of this undertaking was to prepare grammars and dictionaries of these languages, the second was to train Micronesian educators in the principles and practices of bilingual education, and the third was to develop vernacular materials for use in Micronesian schools. This paper assesses the consequences of those endeavors, both intended and unintended. In particular, it focuses upon the concept of “standard orthography” and how that notion, in Micronesia and elsewhere, has sometimes impeded the development of vernacular language literacy. More contentiously, it considers the possibility that the conventional goals of vernacular literacy programs might, in some circumstances, be counter-productive; that is, rather than enhancing linguistic vitality, they might, in fact, diminish it.

BOOK REVIEWS

519, Catharina Williams-van Klinken, John Hajek, and Rachel Nordlinger. 2002. Tetun Dili: A Grammar of an East Timorese Language
Reviewed by Aone Van Engelenhoven

522, Bert Remijsen. 2001. Word-Prosodic Systems of Raja Ampat Languages
Reviewed by Graham Thurgood

525, Nikolaus P. Himmelmann. 2001. Sourcebook on Tomini-Tolitoli Languages: General Information and Word Lists
Reviewed by Robert Blust

531, Karen Davis. 2003. A Grammar of the Hoava Language, Western Solomons
Reviewed by Ulrike Mosel

INDEX

535, Index of Languages in Volume 43