All Austronesian languages are currently considered to derive from a single parent language, probably spoken on Taiwan something over 5000 years ago. Many scholars consider that the Austronesian language family has four highest order subgroups. Three of these subgroups comprise languages confined to Taiwan. The fourth subgroup — Malayo-Polynesian — includes all of the Austronesian languages spoken outside Taiwan. In effect, therefore, it is this subgroup of Austronesian languages that constitutes the predominant focus of this volume.
The principal method used for subgrouping the Austronesian languages is the classical comparative-historical method, largely developed in the comparative study of the Indo-European languages. This method is based on the systematic comparison of regular sound correspondences between languages as a first step towards reconstructing a proto-language from which it is possible to trace the derivation of daughter languages. Once the reconstruction of a proto-language is achieved, individual languages and sets of languages can be examined to determine the innovations they reflect relative to the proto-language. It is essentially upon shared innovations (phonological, morphosyntactic and lexical) between languages and sets of languages that subgroupings are established. Although the existence of the related Austronesian languages was already recognized in the seventeenth century, the systematic comparative research of Otto Dempwolff (1934-38) laid the foundations for much present-day linguistic research.
Comparative approaches to the study of the Austronesians in anthropology have been far more varied. Different regionally focused efforts at comparison have contributed to Austronesian studies and gradually the various separate strands of this research have begun to coalesce in a common set of comparative interests and approaches. L.H. Morgan’s investigation of Hawaiian kinship and his construction of a “punaluan” family (1870) could be considered an early contributor to this research, as could W.H.R. Rivers’ history of Melanesian society (1914). F. Eggan’s research on the Philippines which, in part, led to his paper on the method of controlled comparison (1954); the work of W.H. Goodenough in Micronesia that provided a basis for his influential paper on Malayo-Polynesian social organization (1955); Sahlins’ investigation of social stratification in Polynesia (1958) and I. Goldman’s comparative study of status systems in ancient Polynesia (1970) all contributed to common comparative concerns.
Another important strand in this comparative mix was the work of Dutch anthropologists in Indonesia. In 1935, at a time when Dempwolff was in the midst of publishing his Austronesian research, the Leiden anthropologist, J.P.B. de Josselin de Jong, delivered a programmatic call for the comparative study of populations of Indonesia. Inspired not by linguistic investigations but by Radcliffe-Brown’s study of “The Social Organization of Australian Tribes” (1931), de Josselin de Jong’s “The Malay Archipelago as a Field of Ethnological Study” (1935, 1977) set in train a program of research that has, under various guises, continued to this day.
The single most influential comparative study to draw on J.P.B de Josselin de Jong’s inspiration was that of his student, F.A.E. van Wouden, whose investigation of the societies of eastern Indonesia (1935, 1968) attempted to identify certain structural features of these societies as developments from an earlier proto-form of social organization — an organization that resembled Radcliffe-Brown’s model for Australia. Other Dutch anthropologists, including van Wouden himself at a later stage in his career, allowed this comparative approach to inform their ethnographic researches without committing themselves too rigidly to a single prototypic model of society. Later reformulations of this “Ethnological Field of Study” approach continued to insist on investigation of a shared “structural core” (P.E. de Josselin de Jong 1980, 1984), but also called for a linguistic focus directed to the study of a common set of shared social categories — the continuing preservation of similar metaphors for living (Fox 1980). A similar emphasis on the study of “historical metaphor” and its comparative significance was articulated and developed by Sahlins in his study of Hawaii and of the other Pacific island societies (1981, 1985). One evident inspiration for these perspectives was the work of the Indo-European comparativist, George Dumézil.
More explicitly in relation to the study of Indonesian societies, however, both Fox (1980, 1988) and Blust (1984) argued that to preserve the notion of an Ethnological Field of Study required reinterpreting it in relation to, and as part of, the comparative study of Austronesian languages. This notion was particularly critical in comparisons between Austronesian and non-Austronesian societies in areas, such as Halmahera, where contact has been continuous for periods of several millennia (Platenkamp 1984; Bellwood 1994).
The Comparative Austronesian Project under whose auspices this volume took shape was intended to draw together anthropological, archaeological and linguistic approaches for the study of the Austronesian-speaking populations and to fashion a general framework for the mutual interpretation of the complexities of the Austronesian heritage. The disciplines drawn upon to illuminate this heritage include some which focus mainly on the comparative analysis of phenomena of the present or the recent ethnographic past; these disciplines include linguistics, social anthropology, genetics and zoogeography. Cross-cutting are other disciplines which draw their data directly out of traces of humanity and human activity which survive from the remoter past. These disciplines include archaeology, palaeoanthropology and literary history.
The chapters have been organized into two sections, the first focusing on questions of origins and dispersal, the second on questions of the interactions and transformations which Austronesian peoples and societies have undergone since dispersal occurred.