- A dictionary of Owa, a language of the Solomon Islands by Greg Mellow
The languages of the Southeast Solomon Islands are among the best described in any part of the Pacific. This is particularly true with regard to lexicography, as dictionaries (some of them quite substantial) exist for no fewer than seven of the languages of Malaita, Makira (San Cristobal), and their immediate satellites. In order of publication year these are Sa'a and Ulawa (Ivens 1929), Arosi (Fox 1970), 'Āre'āre (Geerts 1970), Lau (Fox 1974), Kwaio (Keesing 1975), and Toqabaqita (Lichtenberk 2008); the seventh and most recent of these contributions is the dictionary under review.
According to the introductory material, Owa is spoken in the Star Harbour area of Makira, and on the smaller islands of Santa Ana1 (Owa Rafa) and Santa Catalina (Aorigi). It, thus, occupies the southeasternmost extremity of the Solomons chain (although the political entity also includes the more southeasterly Reef and Santa Cruz islands). The language is said to be part of the Kahua dialect chain, which extends well into eastern Makira, and the particular variety described is that of Santa Ana Island.
The Owa dictionary contains the following sections: 1. Acknowledgments, 2. Purpose and compilation method, 3. Owa people and language, 4. Dictionary layout, 5. Abbreviations, 6. Word classes, 7. A 105-page grammar sketch (phonology, nouns, noun phrases, verbs, verb phrases, clause structure, sentences, discourse), 8. Bibliography, 9. Owa- English dictionary (473 pages), 10. English-Owa finderlist (127 pages), 11. Owa words sorted by semantic domain (115 pages). Because of the substantial nature of the grammatical prologue, I will comment both on it and on the main body of the dictionary.
The description of phonology is brief (4 pages) and somewhat problematic. It should be noted that q represents a voiced bilabial implosive. There is one glaring quirk in stating that the Owa phone [t] in, for example, tao 'lie' is equivalent to the final consonant in English waltz, when its phonetic representation clearly is not [ts] (2). Likewise, one wonders why the phonemic vowels are written ܼ, ݜ, ܭ, ܧ, щ, rather than i, u, e, o, a with a phonetic explanation of their values (3)? Finally, the phonetically complex labial nasal ƾm is described as "labialized" when surely what is meant is "velarized."
The account of basic grammatical features shows much greater control of the data. I can touch on only a few topics, so my coverage will be selective. A good deal of attention is devoted to pronouns, which seem to be unusually elaborate in Owa. First, the preverbal subject pronouns or proclitics ("subject markers") differ in form depending on whether they are sentence-initial or sentence-medial (for example,1/2/3 sg au/o/e vs. kau/ko/ ke~ki). Second, they also differ in subordinate clauses (nau/no/ne). Finally, the same functional elements differ in form depending on whether the clause is in realis or irrealis mood (qai/ui/ka sentence-initially, qai/kui/ki sentence-medially, nai/nu~nui/ni in subordinate clauses). Like many other Oceanic languages, Owa also distinguishes singular, dual, [End Page 508] and plural numbers, so the combination of features defining subject pronouns with regard to position in the sentence, position in main or subordinate clauses, marking for mood, and marking for person and number has produced a rather elaborate system.
The most attention-drawing feature of nouns in many Oceanic languages is their relationship to possession, and this is also true of Owa, where nouns are directly possessed if inalienable, and indirectly possessed if alienable. Many Oceanic languages distinguish edible and drinkable possession from general possession through the use of possessive classifiers (these go under various names in different grammars) that carry the possessive pronoun, as with Fijian no-na, ke-na, me-na 'his/her thing (general possession)', 'his/her thing (to eat)', 'his/her thing (to drink)'. By contrast, Owa reportedly distinguishes only general and edible possession, and it does so with a partially...