Niall Sharples (ed.). A Late Iron Age farmstead in the Outer Hebrides: excavations at Mound 1, Bornais, South Uist.

Henderson, Jon

Niall Sharples (ed.). A Late Iron Age farmstead in the Outer Hebrides: excavations at Mound 1, Bornais, South Uist. xviii+419 pages, 213 colour & b&w illustrations, 112 tables. 2012. Oxford & Oakville (CT): Oxbow; 978-1-84217-469-2 hardback 37[pounds sterling].

[ILLUSTRATION OMITTED]

This extensive monograph describes excavations at Mound 1, one of a complex of artificial settlement mounds which dominate the coastal machair plain at Bornais on the island of South Uist. Between 1994 and 2004 excavation was undertaken on the principal mounds of the complex as part of a wider project examining the archaeology of the southern Hebrides, by the University of Sheffield and others (see also review by Jane Downes, this volume). This is the second volume in a planned trilogy, each one dealing with a particular mound. The first volume described the extensive Viking/Norse deposits of Mound 3 which, taken together with the evidence from Mound 2, forms part of one of the largest and most important Norse settlements in Scotland (Sharples 2005). Although there is also Norse occupation on Mound 1, the stated aim of excavation here was to examine the evidence for earlier settlement activity. Between 1996 and 1999 the fragmentary remains of a Late Iron Age house dating from the fifth-sixth centuries AD were discovered underlying the Norse deposits.

The volume is divided into 8 chapters. After setting Bornais within its southern Hebridean Iron Age context in Chapter 1, discussion quickly moves in Chapter 2 to describing the structural evidence for the Late Iron Age settlement. This evidence is slight: an arc of edge-set slabs cut into wind-blown sand, two projecting piers, two hearths, two possible entrance thresholds and a range of stones, holes and pits outlining the original existence of an oval structure which would have had a maximum diameter of c. 6.5m. In addition to the two projecting piers Sharples postulates the existence of four more and interprets the remains as a wheelhouse (Fig. 33) from which the stone had been symbolically removed after abandonment in the first half of the sixth century AD. Chapter 3 deals with the Norse reoccupation of the mound which begins in the ninth century AD culminating in the construction of a rectangular house in the eleventh century which was subdivided and subsequently filled with midden in the twelfth and thirteenth centuries. Discussion of the Norse deposits is not the main focus here, though it does contribute to our understanding of the extent of the Norse occupation at Bornais.

The significance of Mound 1 lies not in the recovery of imposing structural material but in the survival of well-preserved deposits from which detailed environmental and artefactual evidence has been extracted. For example, two Late Iron Age periods of occupation were identified, separated by a charcoal rich conflagration deposit containing a large number of carbonised timbers, most probably the remains of a roof. The identification of the timber as spruce driftwood not only proves that such wood was readily available but also underlines the lack of locally grown timber. A large assemblage of animal bone and carbonised grains was recovered, providing detailed insights into the economy of a mid-first-millennium AD settlement. As is usual the crops were dominated by barley but there is also evidence for diversification through the introduction of oats and flax for the first time in the area. Cattle and sheep make up the majority of animal bones but includes some pigs and red deer (with very young animals particularly targeted), and there is clear evidence for fishing of saithe and salmon. The environmental evidence is augmented by a very large assemblage of stone tools, worked bone objects and diagnostic ceramics, including Dun Cuier ware, which are all firmly dated by a suite of radiocarbon dates. All this is analysed in detail in the following chapters: assemblage (including artefacts, pottery, plants, charcoals and bones) in Chapter 4, the exploitation of local resources in Chapter 6 and an extended interpretation of the activities at the site in Chapter 7. Most significantly the evidence can be dated to a precise period of activity from the fifth to the sixth centuries AD as outlined in Chapter 5 which deals with the site chronology.

Chapter 8 is a general discussion and overview of the site in context and here the main problem with the volume becomes evident. Rather than the wheelhouse envisaged by Sharples, the structural evidence at Bornais is more in keeping with that of a mid-first-millennium AD cellular settlement. These are defined by vertical slabbing revetted into other material as seen at sites such as Beirgh on the Isle of Lewis and Buckquoy on Orkney. Crucially the projecting piers at Bornais do not appear to be associated with substantial internal walling, as is the case at wheelhouse sites, but are instead connected only to vertical slabs more characteristic of cellular forms. Roughly built piers like those at Bornais are known from other cellular sites where they are used to further subdivide spaces.

Sharples' interpretation depends on a scenario where almost all traces of the stone wheelhouse were surgically removed leaving no stones behind and, more crucially, no trace of a cut or damage to the floor deposits. Ali this was supposed to have been done while symbolically leaving behind two piers and the discontinuous eastern arc of slabs while protecting the floor deposits to ensure the survival of Hearth 2 and its associated cattle bone arrangement. A cellular settlement interpretation is less reliant on such special pleading and better fits the surviving evidence. Interestingly, Sharples also seems less sure of his interpretation in the final phase of occupation, pointing out that the precise size and shape of the house in its second phase was difficult to distinguish stating that "it was not a standard wheelhouse" (p. 54) but was similar in shape to the associated trapezoidal hearth.

To suggest that Bornais was a cellular settlement is important not on grounds of typological pedantry but so as not to confuse it with the monumental wheelhouse forms which have been radio-carbondated to the Middle Iron Age (200 BC to AD 400). Throughout his discussion Sharples tends to compare the evidence from Bornais to these earlier sites rather than consider it in its contemporary context. For example, be comments that the trapezoidal hearths at Bornais are very unusual in a wheelhouse context but does not consider that they are well attested in cellular settlements (see Harding & Gilmour 2000, figs. 29a & b). While it is true that there are Late Iron Age aisled roundhouses at Scamess on Shetland, these differ in construction from Middle Iron Age wheelhouse forms of the Western Ides and in any case occur within a context of cellular settlement.

This problem of interpretation aside--to which one could add frustration that excavation did not extend to identifying the primary phases of occupation on the mound--Sharples' engaging interpretation of the function, meaning and life-cycle of the site is important in terms of our understanding of mid-first-millennium AD settlement in the Outer Hebrides. The thorough investigation and dating of the well-preserved deposits and artefacts presented in this volume ensure that it will make a substantial contribution to a very much understudied phase in the development of Atlantic Scottish settlement.

References

Harding, D.W. & S.M.D. Gilmour. 2000. The Iron Age settlement of Beirgh, Riof, Isle of Lewis. Volume 1: the structures and stratigraphy (Calanais Research Series 1). Edinburgh: Department of Archaeology, University of Edinburgh.

Sharples, N. 2005. Bornais: a Norse farmstead in the Outer Hebrides: excavations at Mound 3, Bornais, South Uist. Oxford: Oxbow.

JON HENDERSON

Department of Archaeology, University of Nottingham, UK

(Email: jon.henderson@nottingham.ac.uk)