ECR Case Study: Carving out identity in Late Neolithic Sardinia

Kirsty Lilley, University of Edinburgh

Introduction

Around the world, monuments are a means of making places and understanding landscapes, acting as reference points for communities to construct and reiterate identities across time and space. This is particularly true of funerary monuments, which are often more permanent – and more archaeologically visible – than the settlements of their associated living populations. In the Neolithic period, tombs and burial cairns are usually understood as the result of collective community efforts, representing the beliefs and values of the groups responsible for their creation (Fowler 2005, 120; Jones 2005, 215).

Somewhat in contrast to the other (Scotland-based) case studies presented here, my PhD research focuses on how funerary monuments contributed to the construction and expression of identity and community in prehistoric Sardinia. Using a combination of statistical and geospatial techniques to consider differences in tomb carving and decoration, I am attempting to bring a quantitative approach to research questions usually subjected to qualitative considerations. I was first introduced to Sardinia and its incredible archaeological record during my undergraduate studies in 2017 when I joined a small field team (led by Dr Guillaume Robin of the University of Edinburgh), to conduct fieldwalking surveys in and around several cemetery sites, and to record the tombs using photogrammetry. Six years later and I have remained a part of this team, although our research focus has shifted towards settlement patterns in the Late Neolithic and Chalcolithic periods (called the ‘pre-Nuragic’ period in Sardinian chronology). Currently, there is a noticeable lack of settlement archaeology for this period, contrasting with the numerous funerary monuments still surviving; however, studies of the tombs have largely been confined to typo-chronological and cultural-historical approaches, with little room for social questions (Webster and Webster 2017, 15–30). Therefore, for my PhD research, I chose to focus on these monuments, known as domus de janas, both to ensure that my dataset is comprehensive and to ask questions of the social topography of prehistoric Sardinia.

Sardinian rock-cut tombs: the domus de janas

As mentioned above, Sardinia’s rock-cut tombs have the local name of domus de janas, meaning ‘houses of the fairies’, on account of their diminutive size and their internal architectural decorations, giving them the appearance of houses in miniature (Webster and Webster 2017, 17). From data collection and a review of existing information, the number of tombs across the island currently stands at around 3600, although only a minority are embellished in some way (approximately 490). Below are distribution maps of all tombs on the island (left), and decorated examples (right).

Whilst these elaborations vary in complexity and style, there are some richly decorated tombs featuring motifs such as carved roofs, pillars, false doors, architraves and mouldings, geometric and abstract patterns, and carved bull’s horns or heads known as bucrania. In exceptional cases, these carvings appear to reproduce entire structural interiors – complete with ‘sagging’ ceilings evocative of timber and thatch architecture – and so providing better evidence for Late Neolithic and Chalcolithic/Eneolithic dwellings than the rarely-excavated remains of pre-Nuragic settlements have thus far offered (Hayden 1999, 115; Robin et al 2021).

Close up inside a small stone structures, showing a solid stone beam and sloping roof
Inside a stone structure, face a square shaped hole below the sloping roof
The interiors of two ‘house-like’ domus de janas, at Sant’Andrea Priu, Bonorva (top) and Noeddale, Ossi (bottom) © Kirsty Lilley
Four square images in a larger square collage, each of which showing a close-up of a detail carved into stone - such as spirals and straight lined patterns
A selection of the carved motifs or ‘cultural traits’ found in the tombs © Kirsty Lilley

Rediscovering past identities

There are many ways in which one could pose questions of identity and group affiliation for prehistoric societies, but for the domus de janas, their embellishment is a means by which we can look at subtler differences between communities undertaking similar practices. In this sense, we can also assess the flexibility of its execution, and identify what aspects were essential within the broader decorative tradition.

The motifs or ‘cultural traits’, despite varying in presentation style across different areas, generally comprise a limited set of depictions. Hence, their use in combination in funerary settings is likely due to different factors, including socially-driven practices and perhaps the length of time tombs were in use. The research methodology involves grouping the decorated examples statistically, using techniques derived from ecology and social sciences and based on the presence or absence of carved motifs for each tomb. This generates a numerical ‘similarity’ value between each tomb pair across the entire dataset, which then enables tombs to be placed into clusters based on their similarity to one another. The groups are then mapped and displayed in QGIS, so that any geographical clustering can be identified visually. Whilst several factors can influence similarity and the associated groupings – including topography, geology, or cultural factors – these clusters are interpreted in this research as culturally-similar communities or ‘identity’ groups. Tomb similarity (effectively a ‘cultural distance’) was also compared with geographical proximity to assess the level of correlation, and edge-bundling techniques were used to create spatial visualisations that, arguably, approximate routes of cultural information transfer.

Distribution map of Sardinia with clusters of data circles using yellow, blue and red oblons.
Hierarchically-derived clusters corresponding to similar decorations in tombs, and perhaps ‘identity’ groups © Kirsty Lilley
A vivid visual representation of the decorated dataset using edge-bundling, which looks like a purple network of connected dots and lines against a black background.
Edge-bundling techniques applied to the decorated dataset, which creates a network of connections across the sites and can be seen as an approximation of routes of information flow amongst prehistoric communities – the main roads of Sardinia today still follow similar paths © Kirsty Lilley

These methods are based in statistics, but are carried out with the understanding that identities cannot simply be equated to figures and artistic differences: this would be too reductionist, and would fail to recognise that differences also emerge between people and material culture in the same cultural setting. For this study, this effect is somewhat reduced by the large number of tombs included in the analysis, which permits the identification of common patterns across the dataset and not only between a few selected sites. Furthermore, by considering the results together with other influencing factors (exchange networks, geology, material and site distributions), and comparing them with other archaeological and ethnographic parallels, we built a more rounded picture of how tombs may have both consciously and unconsciously projected social identities. Additionally, the approach taken is multiscalar, considering smaller case study areas across Sardinia as well as the island as a whole. One of the key advantages of statistical methods is that it allows for comparisons to be made quickly and in a standardised way across a very large dataset, removing some of the bias involved in more traditional typological analyses.

Current thoughts and future directions

Overall, the application of statistical methods to this dataset allows me to explore identities qualitatively, alongside considerations of why the tombs were decorated in this way, and how art acted in these contexts. Moreover, I have been able to suggest interpretations of the different ways in which tombs were understood and used across the island, and it is clear that although the practice of tomb-carving was overall a binding factor in pre-Nuragic society, there was room for communities to rework this tradition within their own social setting. An area of particular interest is that around the municipality of Mamoiada, where tombs are numerous, but largely devoid of decoration; however, any embellishment that is present appears to be very localised, with the use of motifs only seen in these areas. This indicates a secondary level of engagement with this tradition, and perhaps more expedient use of funerary monuments in this region.

To conclude, and to bring the discussion back to Scottish archaeology, the domus de janas comprise an insular island tradition that developed in its own way during the prehistoric period, but the more general practice of rock-cut tomb creation is known around the Mediterranean, in Europe, and beyond. In Scotland, the only example of a Neolithic rock-cut chamber tomb is the Dwarfie Stane on Hoy (Orkney), carved into a prominent sandstone block located in one of the central glens of the island. Whilst this does not necessitate contact between these regions, it implies similar concerns with life, death, and monumentality in different Neolithic societies – and, especially given Scotland’s impressive chambered cairns from this period, we may ask similar questions despite geographical and cultural differences. As one of ScARF’s student bursary recipients for the SIRFA Orkney symposium in March 2023, I was thrilled to visit the Dwarfie Stane and discover Scotland’s own domus de janas for myself!

Two people wearing weterproof clothing and hats pose on opposite sides of a doorway in a low, solid stone structure. A large boulder sits in front of them, seemingly a closure for the doorway.
Exploring the Dwarfie Stane on Hoy during the SIRFA Symposium in Orkney, March 2023 © Kirsty Lilley

Bibliography

Fowler, C 2005 ‘Identity Politics: Personhood, Kinship, Gender and Power in Neolithic and Early Bronze Age Britain’, in Fowler, C and Casella, E C (ed), The Archaeology of Plural and Changing Identities: Beyond Identification. New York, NY: Kluwer Academic/Plenum Publishers, 129–134.

Hayden, C 2001 ‘Houses and monuments: two aspects of settlements in Neolithic and Copper Age Sardinia’, in Brück, J and Goodman, M (ed), Making places in the prehistoric world: themes in settlement archaeology. London: UCL Press, 112–128.

Jones, A 2005 ‘Lives in fragments? Personhood and the European Neolithic’, Journal of Social Archaeology 5(2), 193–224.

Robin, G, Soula, F, Tramoni, P, Manca, L and Lilley, K 2021 ‘The Dead are Watching Us’: A Landscape Study of Prehistoric Rock-cut Tomb Cemeteries in Ossi, Sardinia, Italy’, Proceedings of the Prehistoric Society 87, 1–30.

Webster, G and Webster, M 2017 Punctuated insularity: the archaeology of 4th and 3rd Millennium Sardinia. Oxford: BAR Publishing.