Pottery is one of the most commonly recovered artefacts from archaeological sites. Despite more than a century of relative dating based on typology and seriation, accurate dating of pottery using the radiocarbon dating method has proven extremely challenging owing to the limited survival of organic temper and unreliability of visible residues. We developed a method to directly date archaeological pottery based on accelerator mass spectrometry analysis of absorbed food residues using palmitic (C16:0) and stearic (C18:0) fatty acids purified by preparative capillary gas chromatography. Accurate radiocarbon dating of pottery vessels can reveal: (1) the period of use of pottery; (2) the antiquity of organic residues, including when specific foodstuffs were exploited; (3) the chronology of sites in the absence of traditionally datable materials; and (4) direct verification of pottery typochronologies.
Our ambition was to bring compound-specific radiocarbon dating of lipid residues preserved in archaeological pottery vessels into routine use in archaeology. The workflow required the following steps: (i) rigorous testing of the robustness of this new method with a range of well-dated archaeological pottery assemblages and material types, (ii) work closely with our commercial archaeology and cultural heritage sector partners on how best to operationalise this innovation for routine use using the Bristol Radiocarbon Accelerator Mass Spectrometer Mass Spectrometer facility, (iii) analyse in detail the global market for 14C dating to assess the potential for routine adoption, (iv) produce an IPR strategy that will maximise global adoption.
In this project we tested the robustness of the method by dating the exploitation of dairy and carcass products in Neolithic vessels from Britain, Anatolia, central and western Europe, and Saharan Africa. This includes references sites or cultural phases from different burial environments (e.g. soil, sand) well-dated by dendrochronology or by a suite of radiocarbon dates and their modelling using Bayesian statistics. The results demonstrated the dates on pottery vessels to be entirely compatible with the pre-existing chronologies and support the robustness of the method.
We also used this method for dating directly dairy residues from the first farmers in central Europe from the Western and Eastern end of the agricultural expansion. This allowed us to demonstrate the antiquity of lipid residues and showed these to be present with the very earliest LBK settlement in the respective regions. Critically, this study demonstrates the power of dating directly pottery vessels for refining chronologies involving the inception of new food procurement practices rather than assuming timing indirectly based on associated materials which could be subjected to taphonomic uncertainties.
We also applied the method at archaeological sites located on islands or near the coast, the potential exists for lipid extracts of potsherds to contain fatty acids from both aquatic and terrestrial organisms, meaning that consideration must be given to marine reservoir effects in radiocarbon analyses. We studied the site of Bornais (Outer Hebrides, UK) where lipid analysis orevealed ruminant and marine product mixing for 71% of the vessels. Compound-specific radiocarbon analysis of fatty acids yielded intermediate radiocarbon ages between those of terrestrial and marine bones from the same contexts, confirming a marine reservoir effect existed. We showed that lipid radiocarbon dates can be corrected using a determined percentage of marine products, such that pottery vessels from coastal locations can be radiocarbon dated.
The results were we have obtained have demonstrated to the global archaeological community the wide applicability of our new protocol. Our primary results have been published in leading peer reviewed international journals, notably Nature and Radiocarbon, thereby achieving our IP objectives.
Furthermore, we used this method as part of six collaborations and commercial analyses to identify the need of the community. The two main reasons for using the method by collaborators/commercial need was (1) dating sites lacking conventional materials, and (2) verifying the antiquity of the food residues within sherds.
As part of the commercialisation of the method, we worked on the development of new international radiocarbon standards. The use of standards of similar nature of unknown age material is of primary importance for quality assurance in radiocarbon dating. Based on our extensive research we identified archaeological ‘bog butters’ preserved in Irish peat bogs as ideal standard materials for testing the preparative-GC processing of commonly occurring fatty acids from archaeological materials for AMS analysis. Through negotiations we accessed c. 0.5 kg of two bog butter specimens from the National Museum of Ireland for their preparation as radiocarbon standards. We freed them from visible particulate contaminants and homogenised the bog butters. We then verified their homogeneity by characterization of their lipid composition, compound-specific stable carbon isotope values, bulk and compound-specific radiocarbon dates. We produced two standards which posses dates in archaeological and historical time periods. These standards will be deposited with the National Museum of Ireland and will be made available to international radiocarbon laboratories on request so that they can validate the use of our methods in their laboratories.
The method is now validated and proven to be a robust method for dating archaeological pottery vessels. Our new approach is considered a "game changer" by the archaeological community.
