The Cristobal-Malaitan Subgroup of Southeast Solomonic

Abstract

1.1 THE AIM AND THE SCOPE OF THE STUDY. That the languages of the islands of Malaita, San Cristobal, and their small offshore neighbors, together with two languages spoken on Guadalcanal exhibit features indicative of a close genetic relationship is well known. This was pointed out, although without explicit evidence, already by Codrington (1885:498). Subsequently, virtually all linguists dealing with the Austronesian languages of the Southeast Solomons have assumed, implied, or argued that those languages form a genetic (See, e.g., Ray 1926, Fox 1947, Pawley 1972, Levy 1980, Tryon 1982, Tryon and Hackman 1983, Blust 1984. But see Dyen 1965 for a different view.) The group is generally known as the group. The Cristobal-Malaitan languages are among the best described members of the Oceanic family. With the publication of Tryon and Hackman's (1983) large-scale lexical survey of the languages of the Solomon Islands, some information has become available, for the first time, for all the languages of the Cristobal-Malaitan The purpose of this study is a detailed investigation of the CristobalMalaitan group, the phonological innovations that define it, and its internal subgrouping and external relationships. Only phonological evidence is considered here. No account is taken of grammatical, syntactic, or lexical evidence. Accordingly, the internal and external relationships of Cristobal-Malaitan posited here on phonological grounds constitute a set of hypotheses to be tested against other kinds of evidence. The groups and subgroups have been identified on the basis of uniquely shared innovations, the strongest kind of subgrouping evidence. The conclusions reached in this study conflict with the standard assumption of a binary split of Cristobal-Malaitan into a Cristobal and a Malaitan subgroup. Instead, a tripartite division is suggested: Central and North Malaitan, South Malaitan-Cristobal, and Longgu. On the

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 16 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.