Oral tradition and the creation of Late Prehistory in Roviana Lagoon, Solomon Islands. In A Pacific Odyssey: Archaeology and Anthropology in the Western Pacific. Papers in Honour of Jim Specht

The use of oral tradition or oral history in archaeology is often a contentious issue. In this paper we briefly review methodological issues surrounding the use of such data and follow this with a case study using our research into the last 1,000 years of prehistory in Roviana Lagoon (New Georgia Group, Solomon Islands). We argue that it is not possible to generalize cross-culturally about the historicity of oral tradition/history. However, in the Roviana case, careful use of ethnohistory and archaeology together indicates that: (a) Roviana oral history is linear; (b) there is a close relationship between genealogical age and radiocarbon age; and (c) the modern uses of the oral tradition by Roviana provide a theory of their use in the past. We conclude that the model for the formation of the Roviana Chiefdom which emerges from the working back and forth between archaeology and ethnohistory has much more explanatory power than one based on either source of data by itself. SHEPPARD, PETER, RICHARD WALTER & SHANKAR ASWANI, 2004. Oral tradition and the creation of Late Prehistory in Roviana Lagoon, Solomon Islands. In A Pacific Odyssey: Archaeology and Anthropology in the Western Pacific. Papers in Honour of Jim Specht, ed. Val Attenbrow and Richard Fullagar, pp. 123–132. Records of the Australian Museum, Supplement 29. Sydney: Australian Museum. Records of the Australian Museum, Supplement 29 (2004): 123–132 ISBN 0-9750476-3-9 www.amonline.net.au/pdf/publications/1408_complete.pdf Archaeologists generally acknowledge the importance of incorporating into our explanations or interpretations data that move beyond the economic and material to the ideological and symbolic, and which encompass notions of agency and structure. Even noted processual archaeologists (e.g., Flannery & Marcus, 1993; Renfrew & Zubrow, 1994) have turned to cognitive archaeology, cosmology and ideology. At the same time, post-processualists have pulled back from the relativist abyss and acknowledged that the material world studied by archaeologists is not totally malleable or arbitrary in interpretation (Hodder, 1994: 73). Today we see the potential in bringing together the large scale, long-term materialist arguments of the evolutionary models with the short-term variety generating processes of daily cultural behaviour that are foremost in idealist approaches (Preucel & Hodder, 1996: 311). However, as archaeology comes to adopt a realist philosophical position, it is left requiring standards of proof which, although they may not be as methodologically rigid as the positivism of the 1970s, nonetheless require explanation to be based on 124 Records of the Australian Museum, Supplement 29 (2004) arguments whose strength can be evaluated by some nonarbitrary means. What this means in practice for archaeologists interested in ideology and symbolism is the existence of a body of reliable historical or ethnohistorical data (Flannery & Marcus, 1993; Trigger, 1995). But how can these data be evaluated? Archaeologists have long been wary of the uncritical use of oral tradition. Many argue that there is no scientific way to test the “truth” of such data and often suggest that oral tradition or history is subject to political manipulation and is accordingly more about the present than the past. This, of course, is the fundamental philosophical position of the post-processualists, although they extended it to refute the processualists claims of doing objective science. In an attempt to move beyond the relativist impasse for archaeology in general, Wylie (1993) has proposed a realist philosophy where strength of argument is improved, in part, by the convergence of multiple lines of independent evidence. Upon reflection, this appears to be the way in which most effective archaeological explanation is done or attempted. We suggest that oral tradition/history, ethnology and linguistics can all be used as independent lines of argument in the critical “cables and tacking” methodology suggested by Wylie (1993, 2000). Of course, uncritical use of any lines of evidence by themselves in a simple direct reading of the past is unsound, but denying roles to large bodies of relevant data is, at the very least, unwise and unproductive. In the following we discuss our experiences with the use of oral tradition/history and ethnology while investigating the prehistory of the Roviana people as part of our larger project on the prehistory of the Western Solomon Islands (New Georgia Archaeological Survey). Oral tradition/history and archaeology Ethnohistory has often been a minor aspect of archaeological research, but has either existed as an add-on to the main archaeological database or as a parallel study with little archaeological cross-over, with notable exceptions (e.g., in the Pacific, Green & Davidson, 1969; Kirch, 1996; papers in Torrence & Clarke, 2000). In practice, however, much archaeological interpretation has made use of analogical arguments from ethnography, which in turn are often heavily reliant on oral tradition. In the Pacific region most ethnography attempting to describe “traditional” snapshots of cultural systems are describing entities, which changed dramatically in the last 100 to 150 years (Carrier, 1992). While processualists have been reluctant to incorporate direct oral history in their narratives, they have been much less reluctant to use the summary results of ethnographic research, if only in model formulation, although often it appears to be used in a simplified analogical fashion which masks both history and recorded variety (Feinman, 1997). Unfortunately, much of the debate over the use of oral history has become confused with the political debate over the “ownership of the past” and negotiations between indigenous peoples, archaeologists and historians. A recent example is the series of papers in American Antiquity that are clearly issues related to NAGPRA (the North American Graves Protection and Repatriation Act passed by the American Government in 1990) and other social currents loosely described as the “Science Wars” (Wylie, 2000). These papers reflect polarized positions around the use of oral history, with the positivist archaeologist Mason (2000: 264) recommending after reviewing the issue that it not be used. This is followed in the same issue with the Native American archaeologist Echo-Hawk (2000) arguing that some Native American oral history provides literal history back to the colonization of North America 13,000 years ago. He concludes his paper, by stating that oral history must be subject to critical scientific evaluation. In his evaluation of the use of oral tradition/history, Mason (2000: 242) has presented the following as major problems: 1 Oral tradition is not trustworthy as it depends on memory and verbal transmission; 2 The genre by its nature is more an artifact of contemporary culture than a record of the past; 3 Oral traditions are closed belief systems, beholden to authority and impervious to external challenge; 4 Access to oral tradition may be limited by the keepers. To these we would more specifically add: 5 Much oral tradition should not necessarily be conceived of as literal or lineal history; 6 Formulaic ways of relating to time or space may be characteristic of large culture areas and therefore not be reliable accounts of specific past events. In response, we would argue that the data provided by oral tradition needs to be analysed and interrogated in much the same critical fashion (Vansina, 1985: 186) as any archaeological data, if the goal is the creation of a richer understanding of the past. When such data are available it is counter-productive to ignore it. Working back and forth between archaeology, ethnography and oral tradition/history provides a rich field of data and a product of greater use to an anthropological archaeology (c.f. Green, 2000 on holistic archaeology in the Pacific) and an indigenous community. The problems noted by Mason (2000) are often present with the use of oral tradition, although it is equally not possible to generalize about the historicity of oral accounts. Societies vary greatly in the extent to which they consider the past important and attempt to remember or manipulate it. Similarly, the notion of history and its use in the present can vary widely. Evaluation of collected information is required to ascertain what kind of data can be created from it. Vansina (1985) has reviewed the methodology by which such evaluation should be carried out. He suggests the utility of the information is dependent on a variety of factors. These include the familiarity of the collector with the culture, his or her competence in the native language, and understanding of the context under which the information was collected. He also points out the importance of using multiple lines of evidence to cross-check the stories, both to determine how variable they are within the society and to assess, if possible from independent evidence, the historicity of the claims. Vansina also defines different classes of data, which may have different kinds of use in the construction of the past. These include specific descriptions of historical events or processes, myths or charters which can inform on cultural structure and/or power relationships and testimony to the function, use or name of things or places in the past. All of these have been used in our study of the last 1,000 years of Roviana development. Sheppard et al.: Oral Tradition and Late Prehistory 125 Roviana: oral tradition and archaeology Speakers of the Austronesian Roviana language are today found living beside the Roviana and Vonavona lagoons that stretch 70 km along the southwest coast of New Georgia Island in the Western Solomon Islands (Fig. 1). Since 1850, European traders have been living in Roviana, but it wasn’t until 1902 that Methodist missionaries established a mission at Munda in central Roviana. Early missionaries (Goldie, 1909) an