The site of Cill Donnain lies in the coastal machair (windblown shell sands) of South Uist in the Outer Hebrides, Scotland. It was excavated from 1989–1991 as part of Sheffield University's long-running SEARCH Project (Sheffield Environmental and Research Campaign in the Hebrides), which focused on the multi-period landscape archaeology of South Uist and Barra. This volume forms the latest in a series of monographs that together provide a richly detailed picture of human settlement in the region from the Neolithic to the post-medieval period (see Branigan & Foster Reference Branigan and Foster1995; Parker Pearson Reference Parker Pearson2012).
This particular excavation, however, had an unhappy history. The original excavator, Marek Zvelebil, was a renowned specialist on the Mesolithic--Neolithic transition, who initially undertook the work thinking that he was dealing with a shell midden, akin to the Late Mesolithic examples known from Oronsay and elsewhere in the Inner Hebrides. Unfortunately (from Zvelebil's point of view at least), there was nothing remotely of that date at Cill Donnain, and he was left to dig a predominantly Iron Age site in which he had very limited interest. Unsurprisingly, he never wrote it up. That task instead fell to his colleague Mike Parker Pearson, following Zvelebil's untimely death in 2011.
Writing up someone else's excavation is always difficult. Missing sections, duplicated context numbers and misinterpreted stratigraphic relationships are all par for the course. Nonetheless, the problems here were (to put it mildly) unusually challenging. Recollections from some of the site supervisors, and from Parker Pearson himself, who had worked with Zvelebil during the SEARCH Project and had visited the excavations at Cill Donnain, paint a candid picture of an excavation in near total chaos. Despite this, Parker Pearson has managed to assemble a coherent report that contributes usefully to the wider settlement picture emerging from the SEARCH Project.
The main focus of the excavation was a small, stone-built Iron Age wheelhouse (so-called because the radial stone partitions resemble the spokes of a wheel in plan), but human occupation at Cill Donnain extended from the Early/Middle Bronze Age to the Late Norse period. This extraordinary longevity is by no means unusual in a Hebridean context and highlights the international importance of these Hebridean machair landscapes. Although the excavations hardly touched the earliest layers, which were associated with Cordoned Urn pottery, it was nonetheless possible to plot their approximate extent through coring, and Parker Pearson also argues for the presence of at least two associated houses. This tantalising evidence is enough to suggest that an exceptionally rare and potentially well-preserved Bronze Age settlement remains buried under the Cill Donnain machair.
The wheelhouse itself, at only 6.5m in diameter, was remarkably small and was probably, as Parker Pearson suggests, a peripheral element of a larger Iron Age settlement extending under the adjacent sand-hill to the east. Its rather slight outer wall seems unlikely to have supported the type of monumental roof construction seen at other machair wheelhouses, such as those at Sollas in North Uist (Campbell Reference Campbell1991) and Cnip in Lewis (Armit Reference Armit2006). The building's status as an annex or out-building might also explain some of the apparent peculiarities of the finds assemblage. The complete lack of querns, for example, is very unusual and quite distinct from other wheelhouses, such as Cnip, which produced numerous examples built into the walls and floors of the building (Armit Reference Armit2006). Unlike previously excavated examples in the Hebrides, which seem to have their floruit in the first centuries BC and AD, the Cill Donnain wheelhouse does not seem to have been built until at least the second century AD, although there may, of course, have been earlier wheelhouses on the site. Bayesian analysis of the numerous AMS dates suggests that it was probably not abandoned until AD 495–540. This chronological evidence is particularly interesting as it extends the period of the Hebridean wheelhouses closer towards that of the analogous structures built in Shetland during the Late Iron Age (e.g. Dockrill et al. Reference Dockrill, Bond, Turner, Brown, Bashford, Cussans and Nicholson2010).
Other elements of the assemblage do have close parallels. A fragment of an adult female skull, showing healed blunt force trauma, for example, was found in layers under the wheelhouse and may represent a deliberate foundation deposit. Given its date, as Parker Pearson suggests, this fragment may have been curated for several centuries before deposition, reflecting the complex treatment of human bodies and body parts across this region during the Iron Age (Shapland Reference Shapland2010). Echoing other machair sites, Cill Donnain also produced a wealth of faunal remains and artefacts made of bone and antler, producing useful insights into the economic basis of the site.
Although the rather glaring deficiencies of the original excavation leave some nagging doubts about the reliability of some aspects of the report, there is no doubt that Parker Pearson has done a remarkable job in recovering what was very close to being a lost cause. The volume makes a useful and distinctive contribution to the accumulating evidence for later prehistoric settlement in the Hebrides and is a worthy addition to the SEARCH monograph series.
