Oceanic Linguistics, vol. 32, no. 1 (Summer 1993)
Contents
ARTICLES
The Drift Towards Final Open Syllables in Sulawesi Languages, 1-44
J. N. Sneddon
While removal of consonants from word-final position is common in Oceania, it is relatively rare in Western Malayo-Polynesian languages. An exception is the island of Sulawesi where final consonants have either been eliminated or considerably reduced in number in most languages. Strategies for achieving this have included loss, retention by means of a paragogic syllable, reduction to glottal stop, and velarization of nasals. While there are significant differences among these processes, with most of them not actually resulting in open syllables, there are compelling reasons for regarding them all as having the same target, and it is argued that they are all part of a drift towards final open syllables in this linguistic area. In a number of language groups, several of the processes occurred and appear to have been in competition as methods for "dealing with" final consonants. Of particular interest is the fact that addition of a support or paragogic syllable became the most favored method in this respect, replacing other processes in a number of linguistic groups. Even in languages in which all final consonants were earlier lost, borrowings are now regularized the addition of a paragogic vowel rather than the loss of the consonant. Despite the strength of the tendency towards final open syllables, there are changes counter to the general drift in some Sulawesi languages, in which closed final syllables have been reintroduced.
The PAN Phonemes *n and *N, 45-64
John U. Wolff
As many as six apical or laminal liquids and nasals have been posited for Proto-Austronesian. A review of the evidence weakens the case for several of the more marginal candidates and leads to the conclusion that only three can be justified: *l, *n, and *n. Only these three are found to be clearly in contrastive distribution when patterns of stress placement are taken into account. The resultant phonological system is more plausible than some were proposed earlier. This also allows us to bring doublets under one reconstruction in several instances. Further, one of the arguments proposed to support the existence of Formosan subgroup and a Formosan homeland of Austronesian is refuted.
PAPERS ON VERB SERIALIZATION
Joel Bradshaw, Guest Editor
Nuclear Layer Serialization in Lewo, 65-94
Robert Early
Lewo is spoken on Epi Island in Vanuatu. To the north is Paama Island, and to the south is the Shepherds Group. Crowley (1987) has described verb serialization for the former, as has Sperlich (this volume) for Namakir in the latter. In this area, linguistic relatedness closely correlates to geographical proximity, so an understanding of serialization processes in a good sample of the more southern North and Central Vanuatu subgroup is developing. There are important similarities and differences between Paamese and Lewo, so the Lewo data helps round out the picture. For example, while Paamese serial verb constructions can be distinguished from compound verbs, such a distinction cannot readily be made for Lewo.
In Lewo, nuclear layer serialization is very widespread. However, the process of incorporation has progressed to the extent that most of the serialized verbs no longer occur as verbs in their own right, and have become morphologically indistinguishable from a class of verbal suffixes. This raises practical orthographic and lexicographic issues for Lewo itself, but also wider questions of interpretation and analysis.
In 1982, Crowley described for Paamese a class of postverbal adjunct modifier particles. His 1987 reanalysis presented the same forms as nuclear layer serialized verbs. Crowley suggests that similar phenomena in other languages in the area, described as serial verb constructions. However, the Lewo data shows the ambivalent status of these putative serial verb constructions, and suggests that the earlier descriptions of these other languages may not have been so benighted after all.
Serial Verb Constructions in Namakir of Central Vanuatu, 95-110
Wolfgang B. Sperlich
Based on a review of the Oceanic and other theoretical literature on Serial Verb Constructions (SVCs), a descriptive model is adopted to describe SVCs in the Namakir language. Two main types are recognized: core layer and nuclear layer serialization. Argument-sharing possibilities for both types include same-subject, switch-subject, multiple-object, and ambient serialization. Analysis of the available data suggests that all except switch-subject SVCs occur in Namakir. These various types and subtypes of SVCs are exemplified and discussed.
Like other SVO serializing languages, Namakir shows more productive core layer serialization than nuclear layer serialization. The main analytical difficulty for the Namakir data is distinguishing between serialized verbs and adverbial constructions. A number of diagnostic tests argue that possible cases of ambient serialization are actually adverbial constructions. The question of free and fixed verbal slots in SVCs is also discussed, as is the notion of a serialization hierarchy.
Serial Verbs in Loniu and an Evolving Preposition, 111-132
Patricia J. Hamel
Serial verb constructions are very common in Loniu. This paper describes their various functions, examines their form, and compares some aspects of their structure with information available on other Oceanic serializing languages. In addition, it presents evidence of an ongoing process by which the most frequent serialized verb in Loniu, la 'go', is being reanalyzed from verb to preposition.
Subject Relationships within Serial Verb Constructions in Numbami and Jabêm, 133-162
Joel Bradshaw
As more and more languages are examined for evidence of verb serialization, more and more varieties of the phenomenon have turned up. This paper first outlines the evidence for verb serialization in Numbami, a poorly documented Austronesian language of Papua New Guinea, noting the parallels with its better-documented relative, Jabêm. The paper then focuses on subject relationships within serial verb constructions in Numbami and Jabêm. These two languages allow not only same-subject and switch-subject serialization, but also verbal-subject serialization. In addition, Jabem allows a fourth possibility ambient-subject serialization, in which the subject of a time-adverbial verb in initial position appears to be the general environment or "weather" subject.
Verb Serialization, Word Order Typology, and Zamboangueño: A Comparative Approach, 163-182
Michael L. Forman
Zamboangueño only rarely receives any notice at all from creolist theoreticians. Muysken and Bickerton both touch on it in their contributions to the Cambridge Survey of Linguistics. Contra Bickerton, this paper argues for a different view of Zamboangueño objects. Contra Muysken, the paper argues that the language does have serial verbs. Data are brought to bear from both Bidau Creole Portuguese and from Kristang of Malacca to use for comparative purposes in advancing the description.
REVIEWS
Susanne Holzknecht, 1989, The Markham languages of Papua New Guinea, 183-186
Reviewed by Bernard Comrie
Bruce Biggs, 1990, English-Maori Maori-English dictionary, 186-188
Reviewed by Ray Harlow
Susanna Cumming, 1991, Functional change: The case of Malay constituent order, 188-194
Reviewed by Mark Honegger
Merlan, Francesca, and Alan Rumsey, 1991, Ku Waru: Language and segmentary politics in the western Nebilyer Valley, Papua New Guinea, 195-197
Reviewed by Ger P. Reesink
|