ARTICLES

Bruce Biggs, 1921–2000: A Tribute, 1-19
Andrew Pawley

A Reevaluation of Proto-Polynesian *h, 20-32
Lawrence Kenji Rutter

Proto-Polynesian (PPn) is divided into two primary branches: Tongic and Nuclear Polynesian. Between the two subgroups, only the Tongic languages regularly retain what is traditionally reconstructed as PPn *h, which derives from Proto-Oceanic *s. However, residues of PPn *h are found in certain words in Nuclear Polynesian languages. Furthermore, in some of these languages, it shows up as s, which is a violation of theories of sound change for which h > s is considered implausible. This paper shows that the traditional reconstruction of PPn *h actually conflates PPn *h and a lenis reflex of PPn *s, and that the sporadic retentions are not borrowings, but rather reflect a direct inheritance from Proto-Polynesian. Sporadic and irregular patterns of retention are attributed to lexical diffusion.

Some Shared Developments in Pronouns in Languages of Southern Oceania, 33-66
John Lynch and Françoise Ozanne-Rivierre

The languages of Vanuatu and New Caledonia manifest a number of innovations in the Proto-Oceanic pronominal system, the most interesting of these—and the most useful for subgrouping purposes—being in the nonsingular focal pronouns. We show that (1) they continue two changes begun in Proto–Eastern Oceanic—*t > *d in the first inclusive pronoun *kita, and the suffixing of the numerals *rua ‘two’ and *tolu ‘three’ to form dual and trial pronouns, and (2) they also continue the change *k > *g in the first exclusive and second person pronouns *kamami and *kam(i)u, which has been reconstructed for Proto–North–Central Vanuatu. However,the languages of Southern Vanuatu and New Caledonia apparently uniquely share two other innovations: (3) *kida > *gada (alternating with *gadi?) as the first person inclusive pronoun; and (4) restructuring of the system of number-marking, such that dual and trial are marked by greatly modified forms of the numerals *rua and *tolu, and a plural suffix deriving from an equally modified form of *pat(i) ‘four’ is added to the system. This paper documents these changes, suggests that they provide strong evidence for grouping Southern Vanuatu and New Caledonian languages into a single subgroup of Oceanic, and also suggests that changes (2) and (3) may have been part of a process—begun in Proto–Eastern Oceanic—by which deictics became fused with pronominal roots.

A Quantitative Study of Voice in Malagasy, 67-84
Edward L. Keenan and Cécile Manorohanta

This paper is a quantitative study of the voice system in Malagasy (a Western Austronesian language spoken on Madagascar). We show that nonactive verbs in Malagasy have a very different distribution in texts than nonactive verbs in English, German, and Dutch: they occur far more frequently and they typically present Agent phrases. This we claim reflects the very different role of the voicing system in the grammars of Western Austronesian and Western European languages. Part 1 reviews the voice system of Malagasy, classifying the various voice forms into active vs. nonactive, with the latter divided into passive and circumstantial; part 2 presents the results of our text study; and part 3 draws some conclusions regarding the nature of the voice system in Malagasy.

Rotuman and Fijian Casemarking Strategies and Their Historical Development, 85-111
Ritsuko Kikusawa

Although the Fijian languages and Rotuman are thought to be closely related genetically, and are all accusative languages, considerable differences are observed in their case-marking strategies. This paper describes the various strategies to be found in these languages, and discusses how they could have developed from a single protosystem. It is argued that in Rotuman, where it is word order alone that marks nouns as Nominative or Accusative, the preverbal position of clitic pronouns was generalized to become also the position of full noun phrases. On the other hand, in Fijian languages, different strategies resulted in the original clitic pronouns either remaining as clitics, or becoming grammaticalized as agreement features on the verb.

Mangarevan Doublets: Preliminary Evidence for Proto–Southeastern Polynesian, 112-124
Steven Roger Fischer

The Mangarevan language of the Gambier Islands, situated between Tahiti and Easter Island,displays one of the largest collections of doublets among the forty-odd Polynesian languages. These doublets indicate a pre-Proto–Central Eastern settlement of the island from the Marquesas via the Eastern Tuamotus. They also witness a later Proto–Central Eastern intrusion on the island of similar provenience. The cumulative weight of evidence suggests that Proto–Southeastern Polynesian, a hitherto unrecognized language that seemingly comprised Proto–Eastern Polynesian’s first differentiation, was spoken by the first settlers of the Eastern Tuamotus, Mangareva, Pitcairn, Henderson, and Rapanui.

SQUIBS

Negation in Saisiyat: Another Perspective, 125-134
Elizabeth Zeitoun

In a recent paper, Mei-li Yeh compares the distribution and function of eight Saisiyat negators and attempts to resolve the following questions: (1) What is the morphosyntactic relationship that ’oka’, ’okay, ’okik, ’amkay and ’amkik bear to one another? (2) Why are certain negators followed by a “ligature” (either ’i or ’ik) and others not? (3) Is it the negator or the ligature that determines the marking of the negated verb as dependent or independent? While the first of these questions is well handled, Yeh is unable to answer the last two questions. Based on my own fieldnotes, I suggest answers, showing that together with ’oka’ ’izi’ and ’i ’ini’, both dynamic and stative verbs occur in their [+dependent] form, and demonstrating that while ’i is a ligature, ’ik is not. I also account for the distributional differences between ’oka’, ’ʔizi’, and ’i ’ini’, on the one hand, and kayni’ on the other.

Ve as a Second-Position Clitic, 135-142
Ileana Paul

Contrary to previous claims in the literature, it is argued that ve the yes-no particle in Malagasy, is a second-position clitic. Although ve typically appears at the right edge of the predicate phrase, in sentences with nonstandard word order, it surfaces in second position. Ve is contrasted with other particles that are not second-position clitics but pattern with adverbs. The implications of the placement of ve for analyses of Malagasy clause structure are discussed.

Some Remarks on Stress, Syncope, and Gemination in Mussau, 143-150
Robert Blust

Contrary to theoretical expectation, Mussau, a language spoken in the St. Matthias archipelago in northwest Melanesia, has developed some geminate consonants through the loss of stressed vowels. It is argued that other considerations have overridden considerations of stress in geminate formation, not only in Mussau, but in many other Oceanic languages. Specifically, vowels tend to delete between identical consonants in partial reduplications, a tendency that accords closely with Odden’s (1988) arguments against the Obligatory Contour Principle, particularly as it applies to antigemination.

Malayo-Polynesian: New Stones in the Wall, 151-155
Robert Blust

Many of the shared phonological innovations that define a Malayo-Polynesian subgroup of the Austronesian languages are either sporadic or apply in very restricted phonological environments, and so are easily overlooked. This squib adds two new pieces of evidence for a Malayo-Polynesian subgroup that are of this type. The first is an apparent conditioned change by which PAn *RVj became PMP *lVj, leaving no instances of the former sequence in MP languages. The second is a complex of changes that altered PAn *SayaN to PMP *anay ‘termite’. Although some features of each innovation leave the direction of change indeterminate, others point clearly to the MP languages as the innovative set.

The Marking of Sex Distinctions in Polynesian Kinship Terminologies, 156-166
Per Hage and Jeff Marck

In Greenberg’s cognitive-linguistic theory of kinship universals, it appears that sex, unlike generational distance and genealogical closeness, may not be marked on a consistent basis. This paper presents some systematic evidence on the subject. In Polynesian kinship terminologies, the male point of view predominates. The widespread unmarked status of terms for male child in particular may reflect an emphasis on patrilineal succession in Polynesian societies.

The Numeral Confix *i- -(e)n, 167-176
David Mead

In his 1981 paper on Austronesian numerals, Otto Dahl adduced evidence from Paiwan, Ma’anyan, and Malagasy for positing a numeral confix meaning ‘so and so many days’, which he reconstructed as *(ma)ka- -(a)N. In this squib, I discuss the evidence for reconstructing another numeral confix of the same meaning but having the form *i- -(e)n. As far as I know, reflexes of this other numeral confix are limited to the Bali-Sasak languages and five language groups of Sulawesi. This distribution suggests that *i- -(e)n was a local innovation within Western Malayo-Polynesian.

BOOK REVIEWS

Jelle Miedema, Cecilia Odé, and Rien A. C. Dam, eds. 1998. Perspectives on the Bird’s Head of Irian Jaya, Indonesia: Proceedings of the Conference, Leiden, October 13–17, 1997, 177-162
Reviewed by Sander Adelaar

Laine Berman. 1998. Speaking through the silence: Narratives, social conventions, and power in Java, 182-186
Reviewed by Benjamin G. Zimmer

Terry Crowley. 1999. Ura: A disappearing language of southern Vanuatu, 186-189
Reviewed by Ross Clark