Bronze Age Fuel: The Oldest Direct Evidence for Deep Peat Cutting and Stack Construction?
Branigan, Keith, Edwards, Kevin J., Merrony, Colin, Antiquity
Peat has been used as a fuel and as an additive to arable fields to aid fertility since prehistoric times in many parts of northern Europe (e.g. Fenton 1986; Whittle et al. 1986). The cutting of deep peat and the construction of peat stacks as part of the drying process has been documented from Medieval times, but the antiquity of such activities is unknown. Peat stacks are ephemeral structures whose purpose is to aid the drying of hard-won, wet peat in areas where other fuels such as wood and coal are expensive or unobtainable. They are typically cleared within a few months of construction and leave no traces of their former presence. Here we report the unprecedented discovery of a `fossil' pyramidal peat stack dating to the 2nd millennium BC, from the Isle of Barra in the Outer Hebrides of Scotland. Individual turves contained finger and thumb impressions and pollen analysis reveals environmental conditions at around the time of cutting. The method of extracting and stacking the peat used some 3500 years ago may be similar to that used today.
The site and analyses
Archaeological field survey (Branigan & Foster 2000) in the Isle of Barra, at the southern end of the Scottish Outer Hebrides archipelago (FIGURE 1), entailed work in the township of Balnabodach (NGR: NF 7114 0165) on the eastern side of the island. A visit in June 1999 coincided with peat cutting by a crofter (Mr Ken MacKinnon) who drew our attention to the fact that his treisgear (Gaelic: peat-cutting spade) had been meeting resistance around 0.8 m below the cut face of his peat bank, located about 200 m west-southwest from his croft house. The resistance had been caused by a hard slab of dried peat (a peat or turf) upon which the clear impressions of two fingers and a thumb were evident. Further investigation prior to our arrival revealed that the turf was one of a cache in quasi-pyramidal formation (FIGURE 2). At least 14 turves had survived in the stack and a second stack apparently stood alongside the first. The slabs we removed also preserved finger and thumb impressions (FIGURES 3 & 4). There were no stratigraphic signs of any pit or intrusion by which the dried peat blocks could have been inserted to such a depth from above and it was clear that the `fossil' stack was of considerable antiquity.
[FIGURES 1-4 OMITTED]
The peat was cut across the roots of the constituent decomposed peat-forming vegetation rather than parallel to them, indicating that the peat was cut into an open face, rather than from above on the bank surface. On one turf, the mark of an implement about 3.8 cm wide with a rounded end was visible--presumably deriving from a cutting tool which may have had a wooden, stone, bone or metal blade.
In an effort to determine the age of the buried stack, a 1-cm thick peat sample, with its lowest point 2mm above the base of the deepest turf (no. 8 in FIGURE 2), was taken for radiocarbon ([sup.14]C) dating, on the assumption that it would represent the first peat growth around the abandoned stack. Dating by accelerator mass spectrometry produced a radiocarbon age of 3310 [+ or -] 50 BP (Beta-137614) and a [sup.13]C/[sup.12]C ratio of -26.6 [per thousand], which is within the expected range for peat. This calibrates (Stuiver et al. 1998; CALIB 4.3) at the 2-sigma level to 3640-3440 cal BP (1690-1490 cal BC) (highest probability value (0.954) of Method B and rounded to the nearest 10 years).
In order to obtain evidence for environmental conditions during the time of peat winning, samples of peat from the basal turf and from a monolith of its enveloping material were analysed for their pollen content. Standard KOH and acetolysis pretreatments (Faegri & Iversen 1989) were used and selected pollen and spore percentage taxa are presented in FIGURE 5 (palynomorph type nomenclature follows Bennett et al. 1994). For the turf, the 3 samples were taken from the top, centre and base of turf no. …