Geochemical analysis of an obsidian tool mistakenly attributed to Rapa Nui challenges current views about societies in the ancient Pacific.
Ethnographic artefacts housed in museums for over a century can sometimes get misclassified. The example of a large obsidian tool from the Bishop Museum, which was mistakenly attributed to Rapa Nui (Easter Island) and even published in academic papers about the island’s archaeology, illustrates the value of applying new methods to historic collections. Using portable x-ray spectroscopy (PXRF) to measure the chemical composition of the obsidian, we showed the tool actually comes from New Britain in Papua New Guinea, but from a place that was unexpected. The research therefore returned an artefact to its rightful place in the museum, but, more importantly, added complexity to current views about social interactions in Papua New Guinea, more than 3000 years ago.
Peering into a poorly lit museum case on Rapa Nui, I could hardly believe my eyes. Sitting right in front of me was a large obsidian artefact that looked identical to the ‘stemmed tools’ from Papua New Guinea that I have been studying for some years. Due to their large size and carefully flaked shape that resembles a leaf with a stem, archaeologists think these beautiful objects were ceremonial items exchanged between trading partners to create and solidify social links over a vast area within Papua New Guinea.
Except for the serrations along the handle, the tool in the museum case would have been totally at home among artefacts from Papua New Guinea, dated by my archaeological excavations to between about 3000-10,000 years ago, long before humans had first colonised Rapa Nui. This beautifully crafted tool looked completely out of place among the local implements known at mata’a, because they are typically much smaller and made with very little flaking. If genuine, the artefact raised important questions about the presence of craft specialists on Rapa Nui. There was no label to explain where the tool had been found or why it was in the museum display and I left the island very intrigued. I showed my photo of the tool to a number of archaeologists with expertise in the region, but they had no information.
I had to wait several years to find out more about the artefact in the museum case. Shortly after the archaeologist Dr. Mara Mulrooney was appointed at the Bishop Museum, I wrote to inquire about the unusual stemmed tool because I knew she had done her PhD on Rapa Nui. My timing was perfect. Mara was just embarking on a study of all the obsidian artefacts in the Bishop Museum that had been collected on Rapa Nui. It turns out the tool I had seen in the museum was a cast of the original artefact in the Bishop Museum. Her research found it had been collected by Rev. William H. Cox. He had lived in New Britain, Papua New Guinea for nearly 20 years and had donated a number of objects from that region. So the first question raised was whether some time after the acquisition, the stemmed tool had mistakenly been re-classified as an Easter Island mata’a. This is not a surprising error since stemmed tools from Papua New Guinea are very rare and unpublished until recently.
To find out where the stemmed tool was made, Mara conducted a geochemical study using a portable x-ray spectrometer (PXRF). This technique is revolutionising the study of museum artefacts because we can analyse artefacts with out damaging them by cutting off samples required by older methods. By comparing the results with those from her collections of Easter Island obsidian, it was clear the tool had been misclassified. To find its original home, she then needed to measure samples from my collection of obsidian from geological outcrops in Papua New Guinea. The results of the PXRF study were not surprising in one sense, but totally unexpected in another. Yes, the tool was made of obsidian from New Britain, but it was not from the Kutau-Bao obsidian source used for all other large ceremonial stemmed tools from New Britain. Instead, the geochemistry of the Bishop Museum tool closely matched another poorly known obsidian source from the Mopir region of the island.
Given the manufacture of the Mopir tool demanded great experience and skill, it seems likely there was a second production centre for stemmed tools on the island of New Britain. Why are Mopir tools so rare? Did a group from Mopir challenge the authority and control of the chiefs who owned the obsidian at the more commonly used source at Kutau-Bao? The production of ceremonial tools by highly skilled craft specialists in two different locations opens up the possibility that during this early period in Papua New Guinea, the organisation of society was much more complex than we had previously assumed.
Thanks to the capability of PXRF for analysing precious museum objects in a safe and relatively inexpensive manner, the complex past of social interaction in ancient Papua New Guinea is gradually coming to life. The production of specialised obsidian tools at several centres and their distribution over vast areas indicates the presence of an organised society with status differences and intricate social networks prior to the beginning of pottery in the region. The next step is more fieldwork in the Mopir region to find out more about the people who made this exquisite stemmed tool.
But wait, there is more ahead... There was another group of unusual obsidian tools in that same museum case on Rapa Nui. Mara has plans to bring a PXRF to the museum to analyse them soon. Watch this space!
Dr Robin Torrence
Senior Principal Research Scientist
- Mulrooney, M., Torrence, R. and McAllister, A. 2016 The demise of a monopoly: Implications of geochemical characterisation of a stemmed obsidian tool from the Bishop Museum collections. Archaeology in Oceania DOI: 10.1002/arco.5069.
- Mulrooney M., McAlister A., Stevenson C., Morrison A. Gendreau L. 2014 Sourcing Rapa Nui mata‘a from the collections of Bishop Museum using non-destructive pXRF. Journal of the Polynesian Society 123(3), 301–338.
- Torrence, R., Kelloway, S. and White, P. 2013 Stemmed tools, social interaction, and voyaging in early-mid Holocene Papua New Guinea. The Journal of Island and Coastal Archaeology 8: 278-310. doi.org/10.1080/15564894.2012.761300
For more on the activities of AMRI, subscribe to our newsletter: