Excavations at Dun Vulan: A Reinterpretation of the Reappraised Iron Age

Article excerpt


In a recent review of brochs and wheelhouses, Brochs and Iron Age society: a reappraisal (Parker Pearson et al. 1996), a model was presented whereby broch inhabitants were at the peak of a hierarchical system and wheelhouse inhabitants were base clients. The model proposed was based on faunal material recovered from an external midden at Dun Vulan broch and compared to southern English Iron Age evidence and medieval data (Parker Pearson et al. 1996: 66): This demonstrated a clearly different economy in comparison with the Kildonan III wheelhouse. This paper seeks to explore the evidence presented and concludes that there is, as yet, no evidence to link this external midden with the broch occupation. It is also the authors' intention briefly to review the evidence for Iron Age society from the occupation of brochs to the later 1st millennium AD secondary settlement. We would conclude that southern English models have little relevance to Atlantic Scoff and. Rather than try to propose wide-ranging syntheses, the relatively scant Iron Age evidence from Atlantic Scotland lends itself more towards the definition of local patterns.

Secondary structures

Mike Parker Pearson and Niall Sharples' excavations at Dun Vulan have revealed cellular buildings revetted into the central area of the structure, possibly at first-floor level (Parker Pearson et al. 1995). This evidence, however, appears to have been ignored in the subsequent publication dealt with here (Parker Pearson et al. 1996). The authors' experience and subsequent research into such structures, initiated during excavations at Loch na Berie broch, West Lewis (Harding & Armit 1990; Harding 1993; Harding et al. 1994 & 1995), can now be brought to bear on the evidence presented in the 1996 paper. Loch na Berie broch is a semi-submerged site where 10 years of excavation have uncovered extraordinarily well-preserved and detailed information on the development of secondary cellular structures. A very complex sequence of deposits spanning over 10 phases of occupation from the 2nd century AD to the 7th or 8th centuries AD have been excavated of which none is conclusively original to the use of the broch itself. The experience of this site, and evidence from the Howe, is relevant to the analysis and dating of the Dun Vulan material.

The lack of any published plans of the cellular buildings and external structures at Dun Vulan makes an analysis of their layout and specific architectural details difficult. However, the latter are stone buildings located in a forecourt to the broch, possibly erected by the 1st to 2nd centuries AD based on the MacKie system of dating brochs (Parker Pearson et al. 1996: 63). The original report noted that the entrances to the external structures face the same easterly direction as the broch itself (Parker Pearson et al. 1996: 63), and at least two of these were rectangular stone platforms with covered drains running in an east-west direction. Analysis of the soils on these platforms indicates their use as outhouses rather than dwellings although they incorporate some unusual features. These buildings were compared with the chronologically wide-ranging rectilinear structures at Cnip, Lewis, Tungadale, Skye and the Wag of Forse, Caithness (Parker Pearson et al. 1996: 63). The external midden is discussed during the same argument encompassing these later external structures (Parker Pearson et al. 1996: 63) which would seem to imply that it was certainly not a primary broch feature. Many excavations of Iron Age sites uncover evidence for rectangular buildings such as at Chip, The Howe, Wag of Forse and even Guineas. These are all dated to the 1st millennium AD and often occur prior to the appearance of figure-of-eight cellular buildings dating to the late centuries of the 1st millennium AD.

The appreciation of secondary structures within the dilapidated remains of broch structures is often sidelined by focusing on brochs and wheelhouses. …