13.1 Wooden two-piece bucket from Tower’s Fen, Thorney.
Reproduced with permission of BAR Publishing, www.barpublishing.com
Very often when pottery or wooden vessels are found on waterlogged sites, they have scraps of food adhering to their surfaces. This doesn’t indicate prehistoric slovenliness so much as the events leading up to the vessel’s final moments – like the collapse of a house into a lake. To gain a better picture, we should consider what mealtimes would have been like in an Iron Age household. We know that buildings were carefully laid out and that a good deal of care was taken in the growing and preparation of food. Iron Age pottery is of high quality, with a fine, smooth finish. So far as we know, ceramic plates had yet to appear, but we know wooden bowls and platters existed, and these were also well made. We also know that feasts were major social events, accompanied by much ceremony and probably attended by musicians. Given what we know about Celtic society, with its emphasis on ancestry and local heroes, it seems very probable that family gatherings took place regularly. Mealtimes would have provided a daily opportunity for households to come together, to cement relationships and provide discipline for growing children. So it seems to me inconceivable that people would willingly have eaten their meals off dirty or fly-blown platters. I am sure our Iron Age forebears would have done the washing-up! But now I want to look at another find from a well on the edge of the Fens, this time from our excavations in the 1970s at Fengate. The person who excavated it was my wife, Maisie Taylor, and she was quick to recognize its importance.
Maisie is a leading specialist in prehistoric wood and woodwork. Her interest in the subject began when she was working in the Fengate team while still a student at the Institute of Archaeology in London (now part of UCL). Her imagination was inspired by a small oak stake that she discovered at the bottom of an Iron Age well. There is an unwritten law in archaeology that the most interesting and difficult finds are invariably made right at the end of a dig, when most of the students and staff have returned to university and when the weather is starting to look grim. And that’s what happened in this instance. We only recognized the well for what it was very late, and we knew the site was going to be developed – and therefore destroyed – shortly. By now, everyone had departed except for Maisie and myself. So we excavated it ourselves, and I’m so glad we did.
What made that particular stake so unusual was that one side had been shaped into in a very distinctive, under-cut socket. Maisie worked out that it had to have once formed part of the socket, or housing, of a dovetail joint.5 At the time, dovetail joints were not known on Iron Age timbers, so the discovery caused much interest. I can remember being very impressed by the standard and confidence of the woodwork when I first saw it. But I didn’t realize that the well-executed joint cut into the wet stake in my hands would be the first of a train of discoveries that would change my perception of prehistoric craftsmanship for ever.
*
Prehistoric Britain is sometimes portrayed as a thickly wooded and very empty place, where people would have journeyed for days through dark forest paths linking far-flung settlements. I hope this book may already have dispelled some of these myths. For a start, the woodlands that covered the British Isles in post-Ice Age times were far from blanketing: certain river floodplains would have been naturally quite open, and on thinner, often sandy or limestone soils, a variety of heath-like plants, including shrubs, heather and bracken, would have formed a more open environment. And individual woodlands themselves would have been very different in nature: dominated by beech in the chalk landscapes of the south, by oaks in the Midlands, and by wet-loving species – such as willow, poplar and alder – in lower-lying areas of the Thames valley, Lincolnshire and East Anglia. To modern, urban eyes, woods are woods, but to people who truly understood their surroundings, these different types of woodland would have seemed very different: chalk and cheese.
The ‘clearance’ of trees and woodland has formed a major topic in many books devoted to landscape history. Indeed, the choice of the word is interesting, because it implies that people wanted to ‘clear’, or get rid of, trees, as if they were somehow in the way. Often, landscape historians have assumed that the aim of such clearances was to achieve open arable landscapes, which have since developed into the prairie-style ‘grain plains’ that have become such a sad feature of modern Britain. It’s a view I have never shared.
13.2 Four views of an oak stake from an Iron Age well, at Fengate, Peterborough. Note the dovetail housing joint cut into one side.
© Maisie Taylor
Having tried using them many times, I honestly don’t believe that early flint, stone or bronze axes were ever intended to ‘clear’ large forest trees. They’re simply too light and small for such heavy work. If large, well-established trees needed to be removed, their bark would have been detached in a ring around the base of the trunk (something light axes are well shaped to do). The tree would then die and eventually blow over. Pigs and grazing animals such as sheep are also very good at bark-ringing, which they do by nibbling away at the bark to enjoy the sweet sap beneath. It’s also worth pointing out that felled or ring-barked tree stumps will re-sprout vigorously, as coppiced stools. So ‘clearance’ would never have been a simple matter, though again, pigs and other grazing animals would have helped the process by munching on the fresh shoots.
Archaeologists approach prehistory through studying pottery, metalwork, flints, bones and other things that survive in the drier conditions found on most archaeological sites. But if through great good fortune you are able to work with waterlogged organic material, such as wood, you soon find yourself taking a rather different position from your colleagues. I firmly believe, for example, that much of the forest clearance that we know took place in later prehistory was a by-product of other activities such as the keeping of livestock, the search for coppice products (for use in hurdle-making), for fuel and timber for construction. Once the trees had gone, then the land could be ploughed and used to grow cereals. This way of viewing woodland ‘clearance’ sees the process as a natural by-product of the gradual growth and expansion of the population, the rapid decline of wooded areas in the later Neolithic and Bronze Age Britain (between c. 3000 and 1000 BC) reflecting the rise of local human populations.c
The idea of woodland ‘clearance’ makes the assumption that prehistoric communities somehow turned their backs on the woods. But we know that in the Middle Ages, communities treasured ‘their’ woods as a valuable resource, not just of firewood, but of sturdy poles for building and of hazel rods for hurdles, willow for baskets – and many other things. Oak for use in buildings would have been felled in older woodlands, probably reserved for hunting, and often part of the estate of an abbey or local aristocrat. I am convinced that broadly similar attitudes to trees and shrubs would have existed in the Bronze and Iron Ages. We can see this, for example, in the straightness and high quality of the roof timbers used for houses at Must Farm. This timber must have been grown and set aside for building purposes – people didn’t just stumble upon such well-formed young trees by accident. There is now abundant evidence that prehistoric woods were being carefully managed as long ago as 4000 BC, with the arrival of farming in the Neolithic period. As time passed, it is noticeable that huge, slow-grown oak planks, doubtless split from the massive trees that were still growing in the primeval post-Ice Age forests, were being used in certain important structures.6 They do occur occasionally at Flag Fen (1300–900 BC), but by the Iron Age, such planks had become very rare indeed, and instead we find narrower boards split from trees that had been deliberately encouraged to grow faster. This would suggest that stands of primeval forest had been progressively reduced; maybe by the Iron Age some would have been protected, perhaps as massive trees of special importance to the spirits of the ancestors.
Careful woodland management leads to better timber and higher-quality carpentry. But trees and shrubs have to be managed for other things as well. Indeed, some would
see such management as a variant of farming; take the production of hazelnuts. We know from numerous excavations that hazelnuts were an important food to the pre-farming, Mesolithic hunter-gatherer communities, some ten thousand years ago. Small pits filled with hazelnuts occur quite frequently on Mesolithic settlements and charred nutshells are often found in and around hearths. Clearly, hazel bushes must have been important, because we also know that pliable rods of hazel were used in the construction of house walls.
Over the past twenty-five years I have planted and grown about 400 hazel bushes in both woodland and hedgerows. The bushes that are surrounded by tall trees deeper in the wood grow long, straight rods when they are coppiced. They produce very few nuts. But the hazels around the edges of the wood and in hedgerows are best managed in a different way. Their stems are pruned back, rather than being simply coppiced to the ground, and in the sunshine they tend to branch out. These bushes produce a wealth of nuts, which can be picked, either off the stem or off the ground when the plant is given a good shake. Over the past ten years, nearly all our hazelnuts have been taken by grey squirrels in July and August – long before they are ripe. But the point is that hazel must be grown in different places and managed in contrasting ways to produce nuts or rods. I can’t prove it, but I have a strong hunch that this was something hunter-gatherer communities understood fully, back in Mesolithic times – many millennia before the first ‘real’ farmers arrived in Britain.
The working of wood and the management of woodland are among the oldest and most important skills we humans possess and yet today they are largely ignored in favour of two much cruder, bipolar opposites: fell trees and farm the land, or plant trees everywhere to mitigate climate change. Both are overreactions. Clearly, the latter would be preferable, but it is also very destructive to traditional landscapes. I would suggest instead a more considered approach and with it a real process of reskilling: we must reacquire woodland management skills and return to traditional materials, such as wood, or indeed to its modern variants, such as MDF, if we are to cut down on our use of plastics. Above all else, we must have the humility to learn from the past, if we want to enjoy a civilized future.
*
The discovery of the dovetail housing joint late in the Fengate project slipped into the background quite rapidly, as the team turned its attention to other threatened sites in the Peterborough area. Indeed, I had almost completely forgotten about it when in 1994 our excavations at Flag Fen revealed one-third of a Bronze Age three-part wooden wheel, dating to about 1300 BC.7 Technically speaking, such wheels are known as slab wheels because they are made from three shaped wooden planks, or slabs, which are held together by braces slotted into dovetail housing joints and by pegs or dowels, fitted into sockets in neighbouring slabs. This arrangement allows wheels to flex slightly, yet remain robust and secure.
The wheel fragment had been beautifully made, and although some of the pieces had shrunk after some three thousand years in the ground, it was clear that all the joints fitted together snugly and that the wheel would have been very functional – but also good-looking. All the wood used in its construction had been carefully selected, without knots and with good, straight grain. The choice of the woods used for its various components was very well informed and was clearly based on knowledge passed down from previous generations. The timber selected for the main slab or plank was alder. Unlike oak, alder is reluctant to split and although it isn’t as strong as oak, it is resilient and resistant to wet rot. The two dovetail braces that joined the three slabs together were made from oak sapwood, with the bark still intact. They would have looked very striking. The oak tree selected was young and straight-grained. Oak sapwood (the wood below the bark that carries the vessels that feed the tree and its roots) is far more flexible than the much harder heartwood that forms at the centre of the tree (from old sapwood) and provides the trunk and larger branches with a near-rigid internal frame. Both oak sapwood and alder would have benefited from occasional wetting, which would have helped maintain flexibility. The two dowels were made from carefully shaped oval pegs of ash. Ash was an excellent choice because it is strong, easy to shape and – most importantly – it is resilient, flexible and wears well.
13.3 Three views of one slab from a Bronze Age three-part slab wheel dating to about 1300 BC, from Flag Fen, Peterborough.
© Maisie Taylor
In the months leading up to the publication of the big report on Flag Fen in 2001, I gave dozens of lectures to archaeological societies, schools and local organizations. Whenever the slide of the Flag Fen wheel appeared on the screen, I had trouble trying to convey to the audience how the wheel actually fitted together. I discussed this with Maisie one evening and she suggested that we asked our resident draughtsman, Colin Irons, who had just completed a course in technical drawing, whether he would like to draw up a 3D isometric reconstruction of the wheel and the various joints used. And this is what he produced (see fig. 13.4).
It was so vivid and easy to grasp that we had to include it in the big report, but the reason I found Colin’s reconstruction so absorbing was that it suddenly reconfigured the wheel from being an ancient archaeological artefact to something far more up to date. Not only did the structure of the wheel make immediate sense, but you could think about it as if it were something that needed fixing in the garage, or out in the garden shed. I remember consulting the handbook of my first chainsaw, when the chain needed tightening, and I was struck by the similarity of the diagrams in the manual to Colin’s clear, concise drawing of the Flag Fen wheel. The past and the present had merged into one.
Not far from the wheel, but at a slightly different level, we found part of what had once been an oak axle, for a wheel of about the same size. It looked as if the vehicle to which the axle had been attached had been overloaded, because it had broken at its weakest point, a few inches in from the wheel at the place where the body of the vehicle would have been joined to the axle. After it broke, the axle had been removed. Then the broken end was very roughly chopped, to form a rather crude point. Finally, it was driven into the ground as one of the many stakes along the Flag Fen causeway. A square-section pin was found close by. It fitted the hole in the axle and it could well have been used to secure the wheel, or more likely its separate hub, to the axle in the manner of a lynchpin.d
13.4 An idealized drawing of the Flag Fen three-part wheel with a central hub inserted. Note the two dovetailed braces and the two locating dowels.
© Francis Pryor
13.5 Part of an oak axle with two views of a possible lynchpin, which retained the wheel, or its hub, on the vehicle. The axle broke in antiquity and was then roughly sharpened and reused as a stake or a peg.
© Maisie Taylor
When we were doing the research into the wheel for the Flag Fen report, we were struck by the similarity of slab wheels from wetland sites across large areas of north-western Europe, from Britain and Ireland to Scandinavia, Germany and the Low Countries, in the centuries on either side of 1000 BC. All the known wheels exhibit the same careful manufacture and other details, such as dovetail joints, dowels and the deliberate selection of different timbers. This suggests a degree of communication. Maybe wheelwrights had become an identifiable, specialized craft at about this time.
The rise of a specialized class of highly skilled craftsmen suggests that there were people who were prepared to pay for their services. I had always supposed that the first wheeled vehicles would have been rather crude, probably used for carting hay or straw on farms, and with big wheels and heavy superstructure. But the trouble is, there’s no archaeological evidence for such vehicles. When hay or straw needed to be moved, it was more likely to be carried to stacks by people with forks, or by oxen pulling sledges or slung containers. The vehicle used by our friendse The Flintstones featured wheels made from slices sawn off from large, round tree trunks. It’s an appealing idea, but unfortunately wheels of that sort would split wide open after a few days of use. All the careful
carpentry found in Bronze Age wheels is about avoiding such splitting. In other words, it’s impossible to fashion easy-to-make, heavy-duty wooden wheels. So what sort of vehicles were the Bronze Age wheels made for?
A similar, but complete, slab wheel was found near Flag Fen at Must Farm, in 2016.8 It dated to around 1100 BC and was about a metre (3 ft) in diameter – slightly larger than the Flag Fen wheel, but certainly not cart-sized. Other European Bronze Age wheels are of comparable size. This suggests that they were intended for vehicles smaller than farm carts. I had been pondering these questions while writing the Flag Fen report in the later 1990s, when in 1997 I was hired (I was rather broke and very freelance at the time) to direct excavations at an expanding gravel pit a few miles north of Peterborough, just across the county boundary in Lincolnshire, at Welland Bank Quarry. The site was perched on the gravel soils along the edge of the Witham Valley fens and it had produced good evidence for a farming settlement of the later Bronze Age.
One Friday, as we were returning to the site in relaxed mood after a lunchtime visit to the local pub, it started to rain quite hard but then, after a few minutes, began to ease off. I could see the black rain-bearing cloud heading in the direction of the site, which we reached about ten minutes later. A couple of diggers had a problem and they wanted my advice. We spoke for about fifteen minutes and once we had finished, I decided I should start making plans to expand the dig into an area of featureless silt, which my experiences at Fengate had taught me not to ignore. By this time, it was almost half an hour since the rain shower had passed over, and it struck me that any deep pits hidden below the silt patch might start to show through as the silt dried out after the sharp shower.
Scenes from Prehistoric Life Page 26
