BOOK REVIEWS

Prehistoric Settlement of the Pacific. Ward Goodenough, ed. Transactions of the American Philosophical Society Held at Philadelphia for Promoting Useful Knowledge Vol. 86, Pt. 5. Philadelphia: American Philosophical Society. 169 pp.; figures, references, index.

Reviewed by ATHoLL ANDERSON, Australian National University

The papers in this volume are from two tions. On the matter of why these popula­ small symposia on the settlement of the tions should have eventually expanded Pacific, evidently built around a distin­ southward, Chong and Goodenough reject guished lecture by Ben Finney, from which Bellwood's agricultural-demographic hy­ the main article stems. Delivered in 1993, pothesis in favor of an argument about but updated for publication, they are state­ luxury . Interesting as this is, however, of-the-art surveys of several of the central and complemented as it is by Pamela themes of Pacific colonization, although in Swadling's recent study, Plumes from Para­ a slightly unexpected combination. There dise (1996, Papua National are three papers that deal primarily with Museum), it is difficult to see why Aus­ colonization in the Pacific: tronesians would expand so far and so fast, Finney's review of island discovery in the with so little sign of the return of pro­ light of experimental voyaging, Jim Allen ducts. Other archaeological examples of on Quaternary settlement of the western expansion in this manner, for example the Pacific and its significance for understand­ eastern Vikings (Rus) , are replete with ing the Lapita phenomenon, and Pat Kirch material evidence in both directions, de­ on the nature of Lapita culture and its spite much of the trade being in perishable distribution. The other articles focus on goods (in the case of the Rus, slaves, silk, the origins and relationships of the Aus­ honey, wax, and so forth). An expanding tronesian language phyla in various ways: Austronesian trading system is a fair propo­ Robert Blust on the cultural inventory of sition, but it remains far from demonstrated early Austronesians, Kwang-Chih Chang by the archaeological evidence. and Ward Goodenough on the archaeology Blust's first paper is a brief and some­ of southeast China and the origins of Aus­ what casual argument for cultural devolu­ tronesian cultures, and Blust again on the tion as Austronesians advanced south and more distant relationships of Austronesian east. It asserts the advantages of compara­ languages in continental Asia. tive linguistics over archaeology for recon­ Taking the latter group first, the Chang structing the cultural inventory of early and Goodenough paper is a data-rich and Austronesian speakers, but it needs to be lucid review of the likely archaeological balanced by the counterargument from correlates in south China and of archaeological evidence of innovation in the proto-Austronesian speaking popula- material culture and settlement patterns. Blust's second paper, which looks again at Schimdt's "Austric" and Benedict's Asiall Perspectives, Vol. 37, No.2, © 1998 by University of "Austro-Tai" hypotheses, is situated more Hawai'j Press. 3°2 ASIAN PERSPECTIVES remotely from Pacific prehistory, but it In his paper, Finney reviews the accom­ is a thoughtful analysis of where Proto­ plishments of his project in experimental evolved-the mid­ archaeology, particularly in the construc­ dle Yangtse valley-as well as their rela­ tion and voyaging of the Hokule 'a. These tionships with other languages of southeast are so extraordinary and have provided Asia that spread down the major river val­ such an impetus to the renewed analysis of leys from a broadly common source. Oceanic voyaging, including by Geoff Turning to the Pacific papers in this Irwin and others, that they need no enco­ volume, Allen reviews his argument for a mium here. In writing about the project, largely indigenous origin of Lapita culture however, Finney does rather gild the lily in western . The tide is moving here and there. I doubt very much that against an emphatic statement of this opin­ there is an authentic Maori tradition (that ion, as Allen recognizes, but it does not hurt is, of probable pre-European origin) that to be reminded that much of the content states that the passage from the Cooks or of Lapita culture could have multiple ori­ Tahiti to should be started in gins or could have been significantly rear­ early spring. The argument that Nainoa ranged during an Asian sojourn in Island Thompson navigates in a traditional way Melanesia. Clearly many incomers also but for his inability to rid himself of the remained behind. It is worth noting that concepts of miles and degrees is highly du­ the rates of natural increase of island colo­ bious too. From modern maps, Thompson nists can be extraordinarily high, up to 4 knows all about Pacific geography, the dis­ percent per annum, as on Pitcairn Island. position of all the islands, the directions Even at half that figure, a colonizing popu­ and strengths of the winds and currents, lation of 1000 would have exceeded and the sequences of weather systems and 100,000 in little more than two centuries, their likely effect on planned passages. The leaving ample scope both for continuing Hokule 'a team uses satellite images and rapid expansion eastward within the Lapita other meteorological data to plan its voy­ timescale as well as for extensive coloniza­ ages. It is doubtful indeed whether any tion of western Melanesia-the two pro­ prehistoric navigator ever had even a frac­ positions are not contradictory. tion of that background knowledge to go Kirch's paper is a fairly brief but masterly on. Experimental voyaging in the Pacific summary of the prevailing hypothesis of has contributed immensely to current ideas Lapita origins. In this, as in Pacific pre­ and needs no exaggerated estimation of the history generally, descriptive models, in verisimilitude of its procedures. this case Green's Triple I formulation, are All in all, this a worthwhile and stimu­ more prominent than testable explanations lating volume. It is up-to-date, written by about why things happened. I do not find scholars of undoubted standing, and neatly the trade argument entirely convincing, but unobtrusively edited by Goodenough. partly for reasons already alluded to, nor It points to both the achievements of Pa­ propositions about developing maritime cific prehistory in the modern era and the technology or preadaptation of agriculture. continuing areas of difficulty. It will be Why Lapita colonization breached the valued by students of Pacific prehistory as western boundary of remote , after well as their teachers and it is written suffi­ people had lived so long in such close ciently plainly to be accessible to the inter­ proximity to it, is a matter that eludes us yet. ested general reader. BOOK REVIEWS 3°3

Prehistory oj the Indo-Malaysian Archipelago. Peter Bellwood. Revised ed. Hono­ lulu: University of Hawai'i Press, 1997. 384 pp.; illustrations, maps, plates, bibli­ ography. Softcover.

Reviewed by RICHARD SHUTLER JR., Simon Fraser University

Since the first edition of Prehistory of the for mud volcanoes that create lahars (mud Indo-Malaysian Archipelago was published in flows), which mix sediments from different 1985, a significant amount of new research time periods, confusing the stratigraphy. has been reported in many of the areas Third, none of the fossil hominids found considered in this book. This information have a definitely known provenance (per­ provides new insights into the prehistory sonal communication, Geoffrey G. Pope, of the region over the past two million Dec. 4, 1997). I can personally vouch for years. At the same time, some of the the latter, as I have been to Peming twice recently reported data make long-standing and was shown a different location each disputes more controversial. Bellwood time where the Mojokerto skull is sup­ clearly states that the objective of this book posed to have been found. The dating of is to present "a multidisciplinary recon­ Homo erectus in Java is far from resolved. struction of the prehistory of the modern On page 45, Bellwood cites a claim for a nations of and as hominid mandible and possible stone tools viewed from the perspective of the whole dating to almost two million years at Southeast Asian and Australasian region." Longgupo Cave in Sichuan, China. Again, Chapter 1, "The Environmental Back­ there is a need for caution in accepting this ground: Present and Past," covers the in­ early claim for a hominid in east Asia. Pope fluence of climate, landforms, soils, (personal communication) expressed the levels, volcanoes, fauna, and flora over the opinion that the mandible is not of a past two million years. Since 1891, when hominid. It is clear that Bellwood fully Dubois discovered the first bones of Pithe­ understands the implications of the dating canthropus erectus on the Solo river at TriniI, of Homo erectus wherever it is found, when Java, the time of the first arrival of the first he sums up the dating problems saying, hominids in Java has never been clearly "There are big problems of chronology defined, and has remained extremely con­ concerning the radiation of Homo out of troversial. On page 30, Bellwood mentions Africa, and I see no easy resolution at the the recently reported 1.6 million year date moment" (p. 45). for the Pithecanthropus mandible C from In pages 47-49 Bellwood presents a Pucangan at Sangiran, and the argon date well-balanced discussion of the Pucangan of 1.8 million years for the Mojokerto and Kabul hominids of Java. While the skull from the Pucangan deposit at Peming, jury is still out on the exact age of the east Java. If these dates are correct, the Ngandong skeletal material because of entire story of Homo erectus in east Asia as stratigraphic problems, Bellwood sees it as we understand it today will have to be re­ dating somewhere between 40,000 and written. A number of experts in the field 100,000 years ago. He sees the DNA evi­ challenge these dates for a variety of rea­ dence favoring an "out of Africa" model sons. First, the sediments from which the for all anatomically modern humans. How­ dating material was obtained contain forms ever, on page 53, he clearly states that the that are time-transgressive, some dating issue between continuity and replacement back to the Eocene. Second, Java is known is far from settled. In this same chapter, on the issue of whether or not Homo erectus made tools, Bellwood accepts the two small Asian Pcrspcctil/t'S, Vol. 37. No.2, © 1998 by University of cores, five flakes, and two blades found at Hawai'j Press. Sangiran as having been made by Homo 3°4 ASIAN PERSPECTIVES . 37(2) . FALL 1998 erectus, thereby accepting as fact that Homo The section is very useful for nonlinguists. erectus did make tools in Java. These lithic I highly recommend that anyone who is specimens apparently came from valid not fully versed in linguistic terminology Kabul deposits at Sangiran. Here Bellwood and who has reason to discuss Oceanic/ has made a definitive statement about a Southeast Asian prehistory read pages 100­ controversial subject. Claims for stone tools 105. Following detailed discussions related associated with a Pleistocene skull at Sam­ to Austronesians, on page 110, Bellwood bungmacan depend on the correct dating clearly states his position that Taiwan was of that skull, and a clearly demonstrated the "location of the proto-Austronesians," association of the tools and skull. I have that is, "the place where the first split within been to the site, have seen the lithic speci­ reconstructed proto-Austronesians occur­ mens, and am convinced that the lithics are red." At the same time, he makes it clear tools, but I am not convinced that the tools that the "true homeland" of the proto­ and skulls came from the same level, if in­ Austronesians was on the mainland of deed the spot I was shown is the true loca­ southern China. On page 123, Bellwood tion of the finds. acknowledges the Bismarck Archipelago as On page 67, Bellwood describes two the immediate homeland of the Oceanic finds of stone tools in apparent association languages. In the rest of this chapter and in with stegodon in Flores. The first one was in Chapter 5, Bellwood discusses language, 1970, and the most recent in 1991-1992. culture, and migrations. If these claimed associations of stone tools Chapter 6, "The Hoabinhians, and Their and Pleistocene fossil fauna are verified, Island Contemporaries," contains useful Bellwood points out two important possi­ summaries of the prehistoric cultures of bilities: first, "that contemporaries of the the preceramic and early Holocene period Ngandong hominids may have been able to of the Indo-Malaysian Archipelago, a large venture along the Lesser Sunda chain," and area to cover. One new site reported is the second, that"... it would put a possible pres­ Balobok rockshelter, located in the Sulu ence of middle Pleistocene Homo erectus in Archipelago. Bellwood places this site in strikingdistance of theAustralian continent." the Indo-Malaysian pebble and flake reper­ In Chapter 3, "Indo-Malaysians of the toire. In my view this site has other poten­ Last 40,000 Years," after discussing the tial importance. Balobok has red-slipped modern populations of the Indo-Malaysian and lime-impressed decorated . region, Bellwood refers to the work of one There are radiocarbon dates of 7000-8000 of his students, D. Bulbeck, who has car­ B.P. from this site; however, which artifacts ried out extensive studies of these modern the dates are associated with is unclear. If populations. Bellwood states that Bulbeck further excavations at the site resolve this "has ·considered the whole question of local problem, and the dates turn out to be asso­ evolution within Southeast Asia in great ciated with the pottery, then there is a pos­ detail in order to provide stronger support sibility of Balobok figuring in the ancestry for a continuity hypothesis." Bulbeck sees of the proto-Austronesians. the main problem as how to explain the In Chapter 7, "The Archaeological obvious modernization that has taken place Record of Early Austronesian Commu­ within Southeast Asian populations. After nities," Bellwood begins by presenting a examining a large amount of Southeast brief version of his overall model for the Asian cranial material, Bulbeck stresses that later stages of Indo-Malaysian prehistory, there is "nothing in the evolutionary rec­ which serves as an introduction for his dis­ ord of recent Southeast Asia that demands cussion of the later prehistory of China, a migration of Southern Mongoloids from Taiwan, the , and Indonesia. the north" (page 91). On pages 235-236, Bellwood presents a In pages 100-101 (Chapter 4), Bell­ summary of the Lapita pottery culture. wood defines linguistic terms used by lin­ This chapter is very important reading for guists, anthropologists, and archaeologists. Pacific archaeologists. BOOK REVIEWS 3°5

Chapter 8, "The Archaeological Record first edition. To me this indicates an up­ of Early Agricultural Communities in Pen­ to-date effort by the author. For anyone insular Malaysia," is a new title, and is con­ interested in the prehistory of the Indo­ cerned mostly with mainland Southeast Malaysian Archipelago and the Pacific, Asia. Chapter 9, "The Early Metal Phase: A whether academic or layman, this is a must­ Protohistoric Transition toward Supra­ read book. It will surely be the standard Tribal Societies," begins with a discussion reference on the subject for years to come. of the Dong Son culture of , fol­ The book contains very few errors. lowed by a brief outline of the metal cul­ In fact, it is the only book I have read ture of mainland and island Southeast Asia, that contains no typographical mistakes. At with reference to India. least I did not find any, and I read every One could write a book reviewing this word. One thing puzzles me, though. On one. In my view, Bellwood has accom­ page 169, "(Zuraina 1994)" and "(Zuraina plished the objective he stated in his pref­ 1991)" should be "(Majid 1994)" and ace in a grand manner. This book repre­ "(Majid 1991)." Neither Zuraina nor Majid sents a monumental effort on the part of appears in the index. In the bibliography the author and his helpers. The subject she is listed as "Zuraina, Majid" instead of matter of this book is current up to the "Majid, Z." Finally, some of the maps, time it went to press. I took the trouble to charts, and tables, when reduced for publi­ count the number of references in the bib­ cation, have lettering too small to be read liography: there are 1231 references, of easily. These are minor problems, however, which 417 date after 1985; thus, roughly and they in no way detract from the great one-quarter of the references postdate the value of this book.