A REVIEW OF ARCHAEOLOGICAL DATING EFFORTS AT CAVE AND ROCKSHELTER SITES IN THE INDONESIAN ARCHIPELAGO Hendri A. F. Kaharudin1,2, Alifah3, Lazuardi Ramadhan4 and Shimona Kealy2,5 1School of Archaeology and Anthropology, College of Arts and Social Sciences, The Australian National University, Canberra, ACT, 2600, Australia 2Archaeology and Natural History Department, College of Asia and the Pacific, Australian National University, Canberra, ACT, 2600 Australia 3Balai Arkeologi Yogyakarta, JL. Gedongkuning No. 174, Yogyakarta, Indonesia 4Departemen Arkeologi, Fakultas Ilmu Budaya, Universitas Gadjah Mada, Yogyakarta, 55281 Indonesia 5ARC Centre of Excellence for Australian Biodiversity and Heritage, College of Asia and the Pacific, Australian National University, Canberra, ACT, 2600 Australia Corresponding author: Hendri A. F. Kaharudin, [email protected] Keywords: initial occupation, Homo sapiens, Island Southeast Asia, Wallacea, absolute dating ABSTRACT ABSTRAK In the last 35 years Indonesia has seen a sub- Sejak 35 tahun terakhir, Indonesia mengalami stantial increase in the number of dated, cave peningkatan dalam usaha pertanggalan situs and rockshelter sites, from 10 to 99. Here we gua dan ceruk dari 10 ke 99. Di sini, kami review the published records of cave and meninjau ulang data gua dan ceruk yang telah rockshelter sites across the country to compile a dipublikasikan untuk menghimpun daftar yang complete list of dates for initial occupation at lengkap terkait jejak hunian tertua di setiap si- each site. All radiocarbon dates are calibrated tus. Kami melakukan kalibrasi terhadap setiap here for standardization, many of them for the pertanggalan radiocarbon sebagai bentuk first time in publication. Our results indicate a standardisasi. Beberapa di antaranya belum clear disparity in the distribution of dated ar- pernah dilakukan kalibrasi sebelumnya. chaeological sites across Indonesia, which seem Temuan kami mengindikasikan disparitas yang to be mostly influenced by ease of access, inter- jelas pada distribusi situs yang telah national collaboration focus, and the history of dipertanggali di seluruh Indonesia, yang prior research success in a region. In addition, sebagian besar kemungkinan dipengaruhi oleh our review of the literature revealed a clear kemudahan akses, fokus kolaborasi lack of standardization in the presentation of internasional, dan kesuksesan penelitian radiocarbon dates and their usage in publica- sebelumnya di area yang bersangkutan. Sebagai tions. Despite the impressive increase in dating tambahan, tinjauan pustaka kami menemukan across Indonesia, our review of the literature kurangnya standardisasi dalam suggests numerous excavated prehistoric sites mempresentasikan dan cara menggunakan data in Indonesia remain undated at this time. Stud- pertanggalan dalam publikasi. Meski terdapat ies such as this, and possible others focused on peningkatan yang sangat mengesankan dalam Indonesia’s other archaeological sites, are use- jumlah pertanggalan di Indonesia, masih ful for providing researchers with a dataset for banyak situs yang sudah diekskavasi namun investigations of some of the bigger questions in belum dipertanggali hingga saat ini. Studi archaeology in the region. seperti ini, dan beberapa yang lebih berfokus pada situs-situs arkeologi di Indonesia lainnya, sangat bermanfaat untuk menyediakan data yang lengkap demi menjawab pertanyaan yang lebih besar. JOURNAL OF INDO-PACIFIC ARCHAEOLOGY 44(2020):80–112 INTRODUCTION each site can potentially be accumulated to help Archaeology has benefited enormously from the answer larger, more generalized archaeological development of scientific dating methods. Most questions such as migration, trade, cultural dis- modern societies see time as a scale unit to tribution, and regional occupation trends. For measure events and determine their order, dura- instance, dating records in relation with pa- tion, and/or interval. In archaeology however, leoenvironmental or paleogeographical data al- time is most often used as a tool to conceptual- low us to estimate the type of landscapes which ize the development of human communities, Homo sapiens encountered as they migrated cultures, and technologies, based on their arti- from mainland East Asia into Sunda (continen- facts, unearthed remains, ruins, and other ar- tal Southeast Asia; Bird et al. 2005), or from chaeological data (Simonetti 2013). Archaeo- Sunda to Sahul (Australia-New Guinea; Kealy logical interpretation is very dependent on the et al. 2016). In terms of artifactual analysis, dat- context of the material data. Having the answer ing records allow us to see cultural and techno- for ‘what’ and ‘where’ is not enough to continue logical distribution among sites during the same to ‘how’ and ‘why’ questions without having period, and the possibility of trading activities the answer of ‘when’ already understood. With (Reepmeyer et al. 2019; Shipton et al. 2020a). knowledge of the temporal context related to An effort to make a list of absolute dating re- select archaeological data, we are able to meas- cords (i.e., radiocarbon) in Indonesia was made ure their lifespan and form inferences on the by Bronson and Glover in 1984. Thirty-five story behind it. For instance, an artifact has its years after the method was developed, they suc- own lifespan beginning from when it is being cessfully gathered 65 radiocarbon records from made, then transported, marketed, used, and fi- sites all over Indonesia. Even then this number nally discarded (Adams 2003). of dated sites was considered very few com- Two different dating methods are used by ar- pared to other countries. As they stated: “… chaeologists, both of which have been devel- [the] single site of Ban Chiang in north-east oped to serve the same purpose; providing the Thailand has more radiocarbon dates than the best temporal explanation about the archaeo- whole of Indonesia.” (Bronson and Glover logical materials in question. The earliest 1984:37). More than three decades after that method available to archaeologists was ‘rela- paper was published, the numbers of archaeo- tive’ dating, whereby the age of an object is in- logical sites, and sites for which there are corre- ferred based on its association with other mate- sponding dates, have increased substantially rials in a sequence from oldest to youngest; across Indonesia. However, large regions of the hence it is dated as older or younger relative to country remain largely unexplored by archae- the other material. Absolute dating on the other ologists and many sites have very limited or no hand refers to the direct, actual age of the object absolute dates (Mansyur 2007; Prasetyo 2014; in question in a quantitative rather than qualita- Kealy et al. 2018a). More recent attempts in tive manner (Michels 1972; Walker 2005). cataloguing archaeological dating records in In- Modern relative dating techniques rely heavily donesia have tended to focus on just a single on data obtained with absolute dating methods. island and/or temporal period, such as the list Absolute dating techniques have been ap- made by Bulbeck (2018) on Holocene sites in plied in archaeological research all over the Sulawesi. He compiled 73 known sites in Su- world in various degrees. Dating records in pre- lawesi with compatible radiometric determina- historic sites have a practical application to es- tions. By using the dating records, he was able tablish the period of occupation, changes in oc- to analyze the level of site use at 500 year inter- cupation intensity or site abandonment, changes vals (Bulbeck 2018). to subsistence strategies (technological and die- Here, we review published archaeological tary), and to mark the deposition of important dates from Indonesia, focusing on the dating discoveries such as a burial layer or unique arti- records of initial occupation by Homo sapiens fact. Furthermore, partial dating records from of cave and rockshelter sites. We focus on cave- 81 sand rockshelters not only because they repre- zontal deposition of sedimentary layers through sent a reliable source of shelter for prehistoric gravity; the Principle of Lateral Continuity, communities, but they also have a good chance which describes original sedimentary beds as for preserving archaeological materials. While continuous layers that extend laterally in all di- cave stratigraphies are often complex (see Suti- rections; and lastly, the Principle of Cross- kna et al. 2016; O’Connor et al. 2017), over- cutting Relationships, which states that a sedi- hangs and chambers offer greater protection to mentary feature which cuts across another must archaeological deposits, and enable a greater be younger than the sediment it cuts across degree of sediment build-up and maintenance of (Kravitz 2014). All four of these laws of strati- stratigraphic layers, than do open sites. Simi- graphy have been adapted to interpret strati- larly, dating of burial and shell midden sites of- graphic sequences in archaeology. ten only captures a brief moment in time and is In the early 1950s, the radiocarbon revolution at even greater risk of stratigraphic disturbances brought a new wave of dating methods into ar- or redeposited materials than most cave sites chaeology. Absolute (also known as chrono- (Attenbrow 1992; Bedford et al. 2011). This metric) quantitative measurement is now the ‘time capsule’ capacity makes cave and rock- favored method to estimate the age of archaeo- shelter sites ideal for prehistoric archaeological logical data. The moment Arnold