5.3.2 Wood and other plant material

Artefacts (other than the burnt or decayed remains of structural wood, or wood used in fires) of these materials are much rarer than those of bone, antler and ivory, and in some cases they exist only in the form of impressions left in pots (and in daub), or as skeuomorphs (in pottery). However, the few wooden artefacts that do survive include two of major national (and indeed international) importance, namely: i) a flatbow of yew from Rotten Bottom, Dumfriesshire (Sheridan 1992; 2007) and ii) an axe haft, of rosaceous wood, from Shulishader, Lewis (Sheridan 1992). Both were found in peat and had probably originally been deposited in pools; their waterlogging has preserved the wood. Both are on display in the National Museum of Scotland.

The Shulishader axe. Drawing by Helen Jackson for NMS

The Rotten Bottom bow, radiocarbon dated to 5040±100 BP (4040–3640 cal BC at 2δ, OxA-3540), is the oldest bow in Britain and Ireland and belongs to the Early Neolithic. Skilfully made from a single stave of yew, it now measures 1.36 m but would originally have measured 1.74 m, having broken in use, at full draw, almost certainly during a deer hunt. Its estimated draw-weight of 35 lbs (15.86 kg) for a 28" (71 cm) arrow indicates that it had been designed as a hunting bow, for use at fairly close range (10-15 m). It was found near the end of an upland valley in the hills above Moffat, in a position which would have been ideal for hunting deer (since the animals, having been driven along the valley towards the precipice, would have wheeled round at this point). Having broken, this precious possession could not have been mended, although its string would have been removed for re-use. It had probably been imported to Scotland since yew did not grow in Scotland at this period (except perhaps for one or two tiny areas in the south); it is most likely to have come from Cumbria, across the Solway Firth, but an alternative is Ireland. The early farming communities in Scotland were in contact with both of these areas, importing axeheads of tuff and porcellanite, and Antrim flint. Even though the bow was found in a hunting context, it is most likely to have been the possession of a farmer, given these links and the early Neolithic farmers' lifestyle would have included exploiting wild resources.


The Rotten Bottom bow, shown with its broken limb at the bottom of the illustration. Drawing: Marion O'Neil for NMS.

Yew is the best wood for making bows, being strong yet supple. This object shows that the early Neolithic inhabitants of Britain had a sophisticated knowledge of the properties of different woods and used these to their best effect. Furthermore, the small number of knots in the Rotten Bottom bow indicates that the growth of its parent branch had probably been managed. Bows are extremely rare in Neolithic Britain and Ireland, with only two others known (namely Meare Heath and Ashcott Heath, Somerset, both radiocarbon dated to c 3600-3100 BC. The earliest bow in Ireland post-dates the Neolithic, being radiocarbon dated to 2399-2042 cal BC (Barrysbrook, Co. Offaly: Murray 2004).

Neolithic arrow shafts are also extremely rare in Britain; one example, from Fyvie, Aberdeenshire (fig. x; Anderson 1876), is of viburnum and has a leaf-shaped arrowhead still in situ, along with traces of the mastic used to fix it in position. It may be that, as with other Early Neolithic examples found in the Somerset Levels, birch bark tar had been used to fix the arrowhead in position (Aveling & Heron 1998); only analysis can demonstrate whether this is indeed the case.

Leaf-shaped arrowhead with traces of shaft of viburnum, from Blackhillock, Fyvie, Aberdeenshire. Photo: Marischal Museum © University of Aberdeen

The Shulishader axe haft, which was found with its imported Antrim porcellanite axehead in position, represents an equally rare artefact type in Britain and Ireland. It has been radiocarbon dated to 4470±95 BP (3490–2910 cal BC at 2δ, OxA-3537) and is thus of Middle Neolithic date. It had been deposited complete but peat-digging had removed its bottom part; its extant length is 480 mm and it will have been made from a piece of roundwood at least 160 mm in diameter. Of the rosaceous woods, it is most likely to have been made from hawthorn. Most of its surface is covered with adze facets, which may have been left deliberately as a decorative feature. The haft swells towards the top, but has an unexplained deep square cut where the back of the axehead would have sat, thereby weakening the haft. When this axe was published in 1992, only nine other Neolithic axe hafts were known to have survived in Britain and Ireland (with a few others known from documentary references); since then, there may have been one or two additional finds, but these objects remain very rare, despite the fact that they would have been very common during the Neolithic, as a vitally important tool (and also a display item, as has been suggested for this object).

The only other wooden implement known from Neolithic Scotland is a finely-worked fragment of a handle, of willow, found at Skara Brae (Clarke 1976). In an Orcadian context, where wood was scarce, this would probably have been a valued object; it has even been suggested that it could have been a mace haft. Other pieces of wood were found in the waterlogged levels at Skara Brae in David Clarke's 1972–3 excavations, including twigs, fragments with tool marks and pieces of spruce – reminding us that driftwood would have been a valuable resource in the Northern and Western Isles, being washed across the Atlantic from virgin forests in North America.

A fragment that may have come from a decorated early Neolithic wooden bowl was found at the 'hall' at Warren Field, Crathes (Crone 1999), but this was so small that its identification must remain tentative.

The artefacts described above suggest that tree and shrub growth had probably been managed - as is clearly the case elsewhere in Neolithic Britain where, for example, the Somerset Levels have revealed evidence for Neolithic trackways. Woodland management is certainly implied by fragments of willow withies, used for building (e.g. as wattle and daub - as attested, for example, at Girvan Distillery: Sheridan 2009) and basketry. Fragments of probable withies were found in the waterlogged layer at Eilean Domhnuill, Loch Olabhat, and it has been argued that the design of some Grooved Ware pots - especially where vertical applied cordons are concerned - may constitute a skeuomorph of wicker baskets. The illustrated example from Powmyre Quarry, Glamis, Angus, is a reasonably convincing skeuomorph (or at least evocation), complete with the cross-hatching between the vertical ribs.


Large part of Grooved Ware pot from Powmyre Quarry, Glamis, Angus, showing basketry skeuomorphism. Photo: Headland Archaeology.

Wood was, of course, used to a major extent for building domestic, funerary and ceremonial structures throughout the Neolithic in Scotland (c.f. Noble 2006), and people clearly possessed the skills to fell trees, split them and work and join timber. The construction of the large Middle to Late Neolithic enclosures at Meldon Bridge, Dunragit and Forteviot (Brophy & Noble 2011) would have used large numbers of oak trees, as would the construction of post-defined cursus monuments (Thomas 2007). The construction of rectangular houses of various sizes (including the so-called 'halls') would also have required significant amounts of wood, and several of these show evidence for the use of planks. At Warren Field, Crathes, two large trunks appear to have been erected prior to the construction of a 'hall', and it has been suggested that these may have had a special symbolic significance (Murray et al. 2009); at any rate, they are reminiscent of the split trunk terminal posts used in Early Neolithic timber mortuary structures (i.e. Kinnes' 'linear zone' structures: Kinnes 1984, 1992). While the evidence for these structures consists of postholes and/or charcoal, extant wooden structures are extremely rare (although the waterlogged layers at Eilean Domhnuill, N Uist constitute an unexcavated example). One such structure – a hastily-constructed platform, perhaps used as a fowling/hunting hide – was excavated in 1999 at Parks of Garden, overlooking the Carse of Stirling in the Forth Valley (Ellis et al. 2002). This was made from roundwood trunks and planks, mainly of oak and alder, with birch brushwood and produced radiocarbon dates between c 3350 and 2900 BC, placing this structure within the Middle Neolithic (ibid., 250). A trackway found nearby during the 19th century could have been contemporary, and there may well be other Neolithic structures among the wooden structures listed in the Scottish Wetland Archaeology Database (SWAD: http://xweb.geos.ed.ac.uk/~ajn/swad/searches.html).

A final example of the structural use of timber is provided by the posts that had been set into pits and almost certainly used as a pit-fall trap at Mye Plantation, near Glenluce, Dumfries & Galloway). First excavated in the 1920s, this site was further investigated in the 1950s by Richard Atkinson and Roy Ritchie, but never published; a waterlogged post from one of the pits was radiocarbon dated in the 1990s and produced a date of 3913±39 BP (2560–2240 cal BC at 2δ, UB-3882). Although this places the site within the Chalcolithic period, pottery found between two of the pits shows elements of both Impressed Ware and Grooved Ware affinities. Atkinson & Ritchie had argued that the posts had been sharpened and set into the pits with the sharpened ends upright, although the extant posts have only natural weathering cones.

As is clear from the above, a variety of wood species were used in a skilful and knowledgeable way. It is highly likely that all parts of a tree would have been exploited, with bark usable to make containers and bast usable to make cordage, etc. However, direct evidence for this in Scotland is still lacking, even though examples are known from later periods (e.g. a birch bark cover for an Early Bronze Age log coffin from Dalrigh, Oban: Sheridan 2002).

Plants other than wood were indeed used, with mosses and rushes providing the raw materials for cordage and matting. A fragment of heather rope, exactly like recent Orcadian 'simmons' rope, was found at Skara Brae (fig. x) while the cord impressions frequently found on Middle and Late Neolithic pottery (e.g. from Glenluce: fig. x) may attest to the use of some kind of plant material; analysis would be necessary to determine the full range of materials used, and to see whether wool and sinew had been used.


(left) Matting impression on the base of a Grooved Ware pot (V275) from Forest Road, Kintore, Aberdeenshire, © AOC Archaeology
(right) Matting impression on the base of a Grooved Ware pot from House 3, Barnhouse, Orkney. Photo by Lynda Aiano, Orkney Museum.

There are a very small number of examples of basketry or (perhaps more likely) matting impressions on Neolithic pottery, all of it of Grooved Ware type. The clearest example is from Forest Road, Kintore, Aberdeenshire (acSween 2008), where a large flat pot base shares the same centre as the mat on which it had been placed; the impression could have been formed while the pot was being constructed, with the mat permitting the pot to be rotated. The spiral structure of the matting's construction is clearly visible.

A smaller Grooved Ware pot from Barnhouse, Orkney, also appears to have an impression of matting, again showing spiral construction (Richards 2005, fig. 4.17, no. 1890); and Gordon Childe had also observed a matting impression on the base of a Grooved Ware pot from Rinyo (Childe & Grant 1948, 34, pl. X,9). Among the Balfarg Riding School, Fife, assemblage of Grooved Ware, five pots bore faint traces of 'row upon row of unevenly twisted twine [sic], generally 4-5 mm thick' (McLellan 1993), on the lower wall and base of large, relatively thin-walled vessels. It was suggested that the pots may have been supported in a basket while being built up; re-examination by a basketry specialist should be able to determine whether that had indeed been the case.

Finally, although not strictly artefactual as such, the discovery of fragments of ten puff-balls in the waterlogged levels at Skara Brae suggest the deliberate collection of this fungus and its possible use as a styptic, to staunch the flow of blood. This reminds us of the various medicinal uses of plants which are bound to have been exploited during the Neolithic in Scotland.

The research questions arising from this review of the available material are as follows:

  1. Are there other Neolithic structures or artefacts represented in the SWAD database? (This can only be addressed through fresh fieldwork and dating).
  2. What material was used for the cordage whose impressions are found on Middle and Late Neolithic pottery? Were a variety of materials used, and do these differ from those used in later periods?
  3. How faithfully do Grooved Ware designs echo the design and construction of basketry? (This is easily addressed, and would build on past research by Linda Hurcombe in the University of Exeter.)
  4. Were the Balfarg Riding School Grooved Ware vessels made inside baskets, as has been claimed?
  5. What range of organic material was exploited at Eilean Domhnuill Neolithic settlement? (This can only be addressed through excavation, appropriately resourced)
  6. Do any other examples of Neolithic artefacts or structures made of wood and other plant materials exist among unpublished (or indeed published) excavation assemblages, or in museum (or other) collections?
  7. (Not so much a question, more an observation) Much more direct evidence is needed regarding organic artefacts and structures of all kinds: while we can see that organic resources were skilfully used, we only have a tiny snapshot of the full range of materials and uses. For example, we suspect that curragh-like boats had been used to transport the first farmers and their resources to Scotland, and that logboats had been used for onshore and inland water transport, but so far the earliest dated example of a logboat is Early Bronze Age (from Catherinefield).