Theme 1: Organic residues of milk and other products in LBK pottery

Researchers: Richard Evershed (PI), Mélanie Roffet-Salque (PDRA), David Altoft (PhD student), Emmanuelle Casanova (PhD student), Technician (to be appointed).

Background to our research

Figure 5. Examples of the principal LBK vessel forms.Over the past 20 years our research group have pioneered a range of new techniques for the analysis of organic residues preserved in archaeological pottery with the aim of addressing major research questions in archaeology.

The success of all the techniques rests on the remarkable capacity of the porous fabric of unglazed pottery to absorb and preserve lipid residues of the animal and plant products processed therein. Such residues have been shown to survive in the earliest Neolithic pottery dating to the 7th millennium BC from the Near East. Critical to the success of this project is the fact that animal fat residues are the dominant (>90%) category of organic residue detected in European Neolithic pottery. Molecular and stable isotopic techniques now exist for identifying the fats of the major classes of domesticated animals, i.e. ruminants versus non-ruminants, dairying fats and of some hunted terrestrial species (e.g. Copley et al. 2003; Evershed et al. 2002; Outram et al. 2009, 2011).

  • Marker compounds have been identified that allow the processing of fish and other aquatic resources to be detected in pottery at high sensitivity (Hansel et al. 2004).
  • Lipids in pots also preserve climatic information in the form of the hydrogen (δD values) they acquire through biosynthesis from the water the animals drank and the plants they consumed (Outram et al. 2009).
  • Pottery can be directly dated by dating of individual lipids by accelerator mass spectrometry (AMS) (Berstan et al. 2008).

Remarkably, all these new methods are based on the same class of compounds, namely fatty acids (the building blocks of all fats; see Fig. 5), that are ubiquitous to all animals and preserved widely in archaeological pottery. A further, equally important, development is our new approach to research design, in terms of scale and multi-disciplinarity. As shown by our work in the Near East and SE Europe by analysing organic residues in potsherds from several thousand vessels recovered from tens of sites covering a wide geographical range, patterns emerge that offer new insights into the different ways animals were managed in different regions by early Neolithic farmers (Evershed et al. 2008).

Integration of these findings with faunal evidence provides synergy, with new levels of specificity achievable, in terms of the transitions between hunted and herded species, i.e. which animals were providing meat, which were providing milk, and how this balance varied with culture, environment and through time.

Specific objectives

To investigate:

  • the origins and quantitative contributions of the major animal products acquired and processed in LBK cooking pottery, via fatty biomarkers and compound-specific stable isotopes of animal fats preserved in food residues;
  • the date, via compound-specific 14C analysis, and location of the origins of dairying based on the detection of milk fats in LBK cooking vessels, to test the hypothesis raised by the most recent genetic studies, which indicate the LBK as the core region for the emergence of the LP gene variant;
  • the application of compound-specific deuterium isotopes values of fatty acid to record high resolution regional climatic signals.

Three further hypotheses will be tested:

  • Milking was introduced concomitantly with the LBK material culture, as part of its Neolithic package, and practised from the very beginnings of LBK;
  • Milking developed in temperate Europe during the LBK, essentially emerging as a new technology in a specific region(s) and at certain times, which then spread more or less continuously and increased in intensity through time, reaching a maximum in the later LBK;
  • Milking was not an intensive practice of the management of animals in the LBK, and displayed a discontinuous pattern in time and space.

Materials and Methods

Pottery selection and sampling for organic residue analyses

The LBK pottery has the widest distribution of any (early) Neolithic culture, ranging from nearly the Atlantic in the west, via the Baltic in the north, to the Black Sea in the east. In the south, it reaches the rivers Sava and the lower Danube at the 44th degree of northern latitude. Sampling will also include sites from all ecological zones and environmental variants (e.g. lowland and upland sites in Poland), as well as sherds of La Hoguette and Limburg – vessels found within LBK sites of the upper Rhineland and Paris Basin regions – which potentially represent the existence of persistent hunter-gatherers living alongside the LBK population.

We will sample sites and pottery sherds from all chronological phases of the LBK in the respective regions. We will sample sherds of all available pottery categories (hemispherical cups, bottles and amphora; sieves/strainers where they exist); excavated from the lateral oblong pits accompanying the typical LBK longhouses. This will give us the opportunity to investigate variation on a house-by-house basis as well as comparing settlements and regions. A smaller proportion come from other types of LBK settlement pits.

The scale of this part of the project is unprecedented (> 5000 vessels will be investigated), which will provide wholly new insights into the way LBK pottery vessels were used and nature of the animal products acquired and processed therein.

Analyses will involve sub-sampling, cleaning and grinding small pottery fragment then extracting with organic solvent to yield total lipids extracts. Screening by gas chromatography (GC) identifies the animal fat containing sherds. Further analyses will focus on individual fatty acids as carriers of information regarding fat type, date and environment.

Identification and quantification of major fat types

Use of stable carbon isotope values of fatty acids from pottery vessels to identify milk use in SE Europe and the Near East (each datapoint on the plots represents a residue from an individual pottery vessel). The red shaded box shows the region of most intensive milk use at sites dating to 6.5 ky BC (Evershed et al. 2008, Nature).

Use of stable carbon isotope values of fatty acids from pottery vessels to identify milk use in SE Europe and the Near East (each datapoint on the plots represents a residue from an individual pottery vessel). The red shaded box shows the region of most intensive milk use at sites dating to 6.5 ky BC (Evershed et al. 2008, Nature).

We will use the compound-specific stable isotope approach developed in our laboratory, which links the isotopic compositions of the fatty acids to the metabolism of the animals from which they derived, based on comparisons with reference fats from modern animals. The δ13C values of individual fatty acids are determined using a GC linked to an isotope ratio mass spectrometer (IRMS). Consideration of the distributions of fats in the vessels from across the entire LBK range will reveal differences between region(s) and time periods in which different animal products were being processed in pottery vessels.

Different vessel types will be analysed to test hypotheses relating to the processing of different commodities.

The findings from these analyses will be correlated in Theme 3 with the results of the faunal analyses obtained under Theme 2 to connect production and processing. In regions where skeletal remains are not preserved the organic residues provide vital missing information on animal exploitation.

Hydrogen isotope analysis

These δD values of fatty acids carry isotopic information relating to the climate (temperature/precipitation) of the region in which the animals were raised. GC-thermal conversion-IRMS will be used to be used to determine compound-specific deuterium isotope values for the two major fatty acids (C16:0 and C18:0). These new δD values, recorded across a wide geographic range will provide a novel source of  spatiotemporal evaporation/precipitation information, which will be correlated in Theme 3 will other environmental proxies, in order to determine the major drivers of animal management. These new compound-specific δD values will be compared with Greenland ice core δ18O values, to test much debated theories concerning whether climatic fluctuations drove the waxing and waning of the LBK (Gronenborn, 2007). These data will feed directly into the modelling component of Theme 3 in order to test hypotheses concerning the impacts of environmental variables on patterns of animal exploitation in through the LBK.

Compound-specific 14C dating

The lipid analyses performed above of sherds from each assemblage will identify those vessels containing sufficient fatty acid for radiocarbon dating using accelerator mass spectrometry (AMS; Berstan et al. 2008). These fatty acids (C16:0 and C18:0) offer a unique source of robust radiocarbon dates, derived directly from the commodities processed in the vessels. Compound-specific pottery lipid 14C dates will be correlated with new faunal-based collagen 14C dates and literature values (obtained from charcoal, seeds or collagen from the same sites and/or regions) to provide precise chronologies for the animal-based hunting, herding and dairying activities deduced from the molecular and carbon isotope analyses of organic residues and faunal analyses. Again these 14C data will provide primary inputs to the statistical modelling to be performed in Theme 3.