Issue no 49, November 1999

Air photographs reveal that Hadrian's Wall was built in a settled, farmed landscape, writes Tim Gates

For years archaeologists have believed that the Romans built Hadrian's Wall in a landscape that was - and had always been - a thinly populated or desolate `waste'. This preconception, first propounded in the 17th century by the antiquary William Camden, is still commonly held today.

New work on part of the Hadrian's Wall corridor in Northumberland, however, indicates that in fact the Wall was built through a settled landscape that was being farmed long before the Romans arrived in the North in the closing years of the 1st century AD.

The new evidence comes from an air photographic survey which I undertook on behalf of the Northumberland National Park. Within a sample territory of just over 100 square kilometres, extending from the river North Tyne westwards over the open fells to the Cumbrian border, more than 200 previously unidentified earthwork sites have been mapped from air photographs, including ones taken especially for this project. Prominent amongst the newly discovered sites are six late Iron Age or Romano-British settlements of a type made familiar in the western dales of Northumberland by the work of the late Prof George Jobey. These take the form of a more or less rectangular enclosure, defined either by a stone wall or a bank and ditch, and containing the remains of a number of round stone houses.

Almost all of the new sites adjacent to the Wall are accompanied by organised field systems or by patches of `cord rigg', an early form of cultivation in which soil is thrown up into narrow ridges measuring only 1.0-1.5 metres across. Present indications are that these ridges were formed with spades or hoes, rather than by the use of a mould-board plough.

Four of these recently identified settlements are prominently situated within a few hundred metres north of the Wall or south of the Vallum (the ditch which runs parallel to the Wall to the south). Two will serve as examples. The first is at Green Brae which lies barely a kilometre to the south-west of Housesteads Roman fort. Here two enclosures, containing between them at least four round stone houses, are accompanied by a field system and several small patches of cord rigg. Although the crag top on which the site stands is clearly visible from the fort, the settlement was not detected until 1992 when it was spotted from the air - a salutory reminder of what could yet remain to be discovered in other, less well explored parts of the Wall hinterland.

The second example is at Fold Hill. This farmstead is situated on the northern side of the Wall, just two kilometres north-east of Sewingshields Crag, and consists of a well preserved enclosure with a clearly defined entrance in the east-facing side. While in this case there are no identifiable houses in the interior, their absence can be explained in terms of 19th century stone-robbing connected with the building of the circular sheepfold, or `stell', after which the site is named. On the other hand there is a pair of hut circles just outside the western perimeter of the enclosure which could indicate some expansion in the size of the resident population during the life of this farmstead.

As the appearance of stone-founded houses on these sites is a phenomenon that is not thought to have taken place before the mid-2nd century AD, which is to say about the time that the Hadrianic frontier came into being or even a little later, this would suggest a relatively long period of occupation, extending well into the Roman period.

Be that as it may, the existence of a well-defined field of cord rigg next to the site almost certainly indicates that cereals or other crops were being grown here by native farmers. If this was indeed the case, we may have to think in terms of a more hospitable landscape around Hadrian's Wall than the one we have been accustomed to imagine.

Patches of cord rigg have been identified at more than 70 different locations within the survey territory around Hadrian's Wall, and the size of separate plots can be anything from that of a small allotment to something larger than a football pitch.

Its date has been suggested by a number of instances where late Iron Age or Roman period contexts have been established by excavation or field survey. For example, narrow-ridged soil surfaces have been found beneath the Hadrianic levels of several forts along the Wall. At Denton, west of Newcastle, a field of cord rigg was found to have been under cultivation right up to the point where the land was appropriated by the Roman army in order to build the Vallum in about AD130.

Similar early Roman or pre-Roman contexts can be inferred at Greenlee Lough where a 1st or 2nd century marching camp overlies an extensive tract of cord rigg; and also at the Roman fort of Great Chesters where, as recent air photography has shown, the aqueduct carrying water to the fort cuts through plots of cord rigg at several points on its ten kilometre journey.

Since this form of cultivation seems to have required nothing very sophisticated in the way of tools, it would be no surprise if it proved to have a long prehistoric ancestry. Certainly, a strong case exists for its widespread use by native farmers in the later Iron Age and Roman periods not only in Northumberland but also in parts of southern Scotland. The evidence derives largely from air photographs which record many instances of cord rigg close to stone-built settlements similar to the ones described above.

Nor is the Hadrian's Wall corridor the only place in Northumberland where this phenomenon occurs in the immediate vicinity of Roman forts. In upper Redesdale, for example, there are no fewer than 15 native settlements within a five kilometre radius of High Rochester that are accompanied by fields of cord rigg, including those at Yatesfield and Barracker Rigg which are amongst the best preserved field systems in the county.

The dispersed pattern of native Romano-British settlement now emerging in the Wall corridor, with small farmsteads and fields spaced at irregular intervals over a landscape that seems mainly to have consisted of unenclosed and uncultivated land, is paralleled in other upland parts of Northumberland. Where the amount of enclosed or cultivated land belonging to particular settlements can be estimated, the acreages involved fall a long way short of what would be required to generate anything in the way of a marketable surplus of grain, even on the improbable assumption that all such land was given over to the production of crops.

On this basis we may speculate that arable farming (as opposed to stock rearing) did not rise above subsistence level and was not normally undertaken for purposes of trade or, for that matter, the payment of taxes. Interestingly, the same general conclusion has been reached independently by the archaeologist Marijke van der Veen of Leicester University, in her study of grain samples from excavations on native sites between the Tyne and the Forth. This has led her to conclude that, during the late Iron Age and Roman Periods, the local economy was characterised by `small-scale, intensive, subsistence cultivation'.

In the light of this increasing evidence for native settlement and agriculture in the Wall corridor, it is now plausible to suggest that native farmers may have played a part in the substantial reduction in tree cover detected in several pollen diagrams for the region. This is thought to have commenced around the time of the Roman occupation or, in the case of one site, Fellend Moss, perhaps a little earlier. The evidence has been the subject of much recent debate, all of which has centred on the destructive impact on woodlands brought about by the Roman army in search of timber for building purposes.

Given that there are now no fewer than ten certain or probable Romano-British settlements within a four kilometre radius of the site in Fozy Moss from which one recent peat core was obtained, it is now clear there is scope for wider debate on this issue.

Tim Gates is an aerial archaeologist based in York. Copies of his project report and the accompanying maps can be seen at Northumberland National Park. Contact Paul Frodsham, park archaeologist, on 01434 605555.

Experimental reconstructions can provide surprising insights into everyday life in the distant past. John Lord and Jacqui Wood describe recent work

Today if we lose or break a piece of equipment, we can simply ring for assistance or buy a new one. These options were not available to our early prehistoric ancestors, for whom personal ingenuity in craftsmanship was a way of life.

We can be sure that people in the Palaeolithic regularly had to face the unexpected. For example, implements and weapons will have been lost or damaged during a hunting expedition. The only materials available for repair or replacement were those lying close to hand. In such a situation, what would a Stone Age huntsman have done?

The properties and inherent problems possessed by the materials that our ancestors used have not changed significantly, neither have the skills that are required to manipulate them successfully. Through `experimental archaeology' we can therefore often shed light on possible ways in which emergency repair or replacement problems were solved in the past.

This summer I was given the chance to investigate how to make a Lower Palaeolithic wooden spear, from scratch, using only local materials at a site which I had never previously visited. A television company commissioned the work, and the materials were to be those found at the film location site. The spear had to be comparable in size to the roughly 400,000-year old `Clacton' spear - a large yew-wood weapon suitable for attacking fairly large mammals.

The prospect of cut hands, blisters and splinters were set aside, and after the removal of a few flakes from a flint nodule which was destined to become a chopper, I began the felling of a suitably straight, two-and-a-half-inch diameter sycamore sapling. The stave was down in approximately six minutes. The side shoots were removed with the chopping tool, and the bark was stripped off easily by using one of the flakes.

The next stage proved to be more problematic. I started holding the stave with one hand and chopping with the other, the aim being to produce a finely tapering point; but this proved to be too slow and laborious. More would be achieved if a large flake could be used as a draw knife - which is a two handed operation. I initially thought I could clamp the stave between my feet, and weave it through branches in order to gain stability; but these were all vain moves, as the spear became mobile every time that the flake started to bite.

A prehistoric vice was needed. After some thought, the answer became obvious - the stump of the sapling, which had been left standing about two feet high. After reducing the rest of the chopping nodule into a series of substantial flakes, a cleaver-like flake was hammered into the top of the stump. The stump split down with ease, and fortunately there were no knots to impede progress. The stave was then forced down into the newly made split, and sideways movement was now under control.

The same could not be said for forward movement. If only there were someone around to hold the parted apex of the stump together, or if there were some string to lash around the stump, then the stave would be trapped and stable.

Examples of cordage do not survive well in the prehistoric record but this does not mean it was non-existent. There would in fact have been no shortage of thread, cord and rope, as many plants have a structure which contains strong fibrous material, and certain animal products such as strips of hide, gut and sinew would also have been used. Using animal products was not an option for me - not least because there were no animals in sight - but it was summer and a nearby patch of tall stinging nettles saved the day.

The next stage proved to be the most painful. Six long nettles were uprooted and stripped of their leaves, after which the stems were split open and the woody material was peeled from the outer skins. It is the outer skin of the nettle which contains the strong flexible fibres. Two of the nettle skins were knotted at one end, then while holding the knot with one hand, I used my other hand to roll the two skins which were lying parallel on my thigh. When the knot was released, the two skins plied themselves into a strong cord.

The process was continued by regripping the nettle where the plying had ceased with the `knot hand', and parallel rolling with the other. As each nettle skin neared its end, a fresh skin was introduced until all six stems were used. This process produced a cord about three metres long, but in order to increase its strength, I doubled the cord, gripped it in the centre, then rolled and plied it again to form a shorter,stronger length. From start to finish the cord took about 15 minutes to produce.

I then used the cord to lash the top of the stump together. The prehistoric vice was completed and the spear stave was firmly held. I had both hands free to shape the stave to a fine tapering point. To achieve this, I selected a flake which had an obtuse-angled cutting edge, as flakes with acute-angled edges, although initially sharp, soon become chipped and ineffective. The carving was over in less than half an hour, and the vice then became useful for holding the spear firmly while any slight twists were pulled straight.

One can only wonder: did this process ever take place in the prehistoric past? (JL)

John Lord is a specialist in the reconstruction of prehistoric tools, based in Norfolk

The archaeology of 11th century Scotland is almost entirely undiscovered, writes Nick Aitchison

Macbeth, immortalised in Shakespeare's compelling tragedy of ambition, betrayal and the supernatural, may be one of Scotland's most famous kings. The myth of Macbeth has universal appeal. The historical Macbeth (1040-57), however, is less well known, largely because of the scarcity of reliable source material.

Historians have traditionally been dependent on a combination of contemporary Anglo-Saxon and Irish sources, Scottish king lists that survive only as much later copies and, in particular, the medieval Scottish chronicles which provided the inspiration for Shakespeare's Macbeth. All these sources present problems of interpretation and/or reliability.

Nevertheless, Macbeth emerges from the shadows as a remarkable character in several respects. The known facts about his life are, briefly, these: he belonged to the hereditary aristocracy of Moray, an extensive and turbulent northern province centred around Inverness, Elgin and Forres, and separated from the rest of Scotland by the Grampian Mountains. His father was mormaer (`great steward') of Moray but was murdered by Macbeth's cousins. Macbeth later seized the province, probably by burning his cousin and his cousin's warband to death, and took his cousin's wife as his own. Macbeth also belonged to the Scottish royal kin group through his mother and was probably a grandson of Malcolm II, with whom he was closely associated; Macbeth was present at the meeting between Malcolm and Cnut, King of England and Denmark, in 1031.

Macbeth was by no means exceptional in seizing the kingship by killing his predecessor, the hapless Duncan (1034-40). But Macbeth does stand out by then holding on to the Scottish kingship for 17 years. He defeated an uprising led by Duncan's father, the (lay) abbot of Dunkeld, in 1045 and his grip on power was sufficiently strong for him to make the pilgrimage to Rome - a demonstration of royal piety made popular by Cnut - involving an absence of up to a year. Macbeth made an opulent entrance to the Eternal City by scattering silver to the poor, according to the chronicler Marianus Scotus.

Macbeth was defeated by a combined Scottish and Northumbrian army led by Malcolm, Duncan's eldest son, and Siward, Earl of Northumbria, in 1054. Medieval chronicles place the encounter at Dunsinane but it is recorded in contemporary sources only as the `Battle of the Seven Sleepers', after the festival on which it occured (27 July). Macbeth survived, but his grip on the kingship was probably broken. Although the end of his reign is conventionally dated to 1057, Macbeth's final years probably saw his power limited to Moray. Macbeth was killed by Malcolm's forces at Lumphanan, near the southern limit of this territory, in 1057. His foster son, Lulach, held on to Moray for another four months before he, too, was killed by Malcolm.

Macbeth's reign straddles the midpoint of a turbulent but formative century in Scottish history. The Scots' victory at Carham in 1018 wrested Lothian and the Borders from Northumbrian control and established the Anglo-Scottish border at the River Tweed, where it remains to this day. Although the Northern and Western Isles and Caithness were settled by the Vikings, the extent of the kingdom of Scotia or Alba, as it was known in Gaelic, was similar to that of modern Scotland. This was the zenith of the Gaelic language in Scotland, its use extending into the Borders as place-names attest.

Scottish society during this period was emphatically hierarchical. Kings were selected alternately from different royal kin groups, which frequently resulted in dynastic instability and aspiring kings killing their predecessors. The accession of Duncan, a grandson of Malcolm II, marked a change in the pattern of royal succession and the adoption of primogeniture, descent through the direct line. This resulted in further instability and led to Macbeth killing Duncan for the kingship.

Below the king were two levels of lordship. Mormaers were often cadet members of the royal kin group and were powerful territorial magnates, ruling over extensive provinces, such as Moray. Thanes were royal officials, who administered a royal estate or thanage, collecting the dues and tribute that the king and his entourage then lived off as they travelled around the kingdom. The title of mormaer was replaced by that of earl during the 12th century while thanages gave way to sheriff-doms, although some survived into the 17th century. Below them were probably several levels of both free commoners and unfree or semi-free serfs and bonded peasants, although these social strata are scarcely represented in the historical record.

If the documentary sources for 11th century Scotland are so limited, can the archaeological record help to fill the gap and enhance our understanding of Macbeth's Scotland? The reality is disappointing. Little is known about the archaeology of 11th century Scotland. The absence of indigenous and scarcity of imported artefacts that are closely datable, such as coins and pottery, make it difficult to identify sites or phases of occupation belonging to this period. Exchange was non-monetary and - a few imported Anglo-Saxon examples aside - coins start to appear only in the following century.

This is not to suggest that Scotland was a cultural backwater, simply that its material culture differed from that of Anglo-Saxon England and was more similar to that of the Irish, with whom the Scots enjoyed close cultural and linguistic ties. The lack of evidence may partly reflect the preponderant use of organic materials in this period - such as wood for utensils - and the destruction caused by later wars. Scotland's known waterlogged sites, where organic materials survive, all seem to date to other periods.

The lack of evidence may also reflect the shortage of excavations on sites of the right period. Scotland's towns largely began in the 12th century and it is not always clear where the settlements of the previous century should be sought. Excavations at the known royal site of Forteviot in Perthshire during the early 1980s proved inconclusive.

The most impressive surviving monuments of this age, however, remain under-rated. The splendid ecclesiastical round towers of Irish type at Abernethy in Perthshire, and Brechin in Angus - possibly fortified bell towers - are poorly studied and imprecisely dated to anything from the 10th to the 12th century. This is part of a perceptual problem. Eleventh century Scotland has a low visibility in comparison to other periods, falling into a `no-man's-land' between the more fashionable early medieval period (Picts and Vikings) and the Scottish kingdom of the high Middle Ages (Braveheart and all the rest).

With few exceptions, notably Stephen Driscoll's research at Glasgow University on lordship and landholding in Strathearn in Perthshire, archaeologists have failed to address this period and the far-reaching changes it saw within Scottish society. Because so little material evidence can be firmly consigned to the period, it is hardly surprising that kings and battles should dominate our perception of it.

Macbeth presents an excellent opportunity for channelling widespread public awareness and interest into the archaeology of 11th century Scotland. Some sites are obvious candidates for investigation. Dunsinane, closely associated with Macbeth in myth and drama, is an impressive Perthshire hillfort about eight miles northeast of Perth, probably of the late Bronze Age or Iron Age. Nevertheless, the Scottish regnal lists place Kenneth II (971-95) at Dunsinane, indicating that it may have been refortified and/or reoccupied only 50 years or so before Macbeth's reign. Dr Driscoll's work concluded that hillforts were generally abandoned as high-status settlement sites by the end of the 1st millennium but there may have been exceptions.

Macbeth is associated with other sites. The Prophecy of Berchán, an 11th century verse history of Scottish and Irish kings, predicts a bloody death for Macbeth at Scone, the inauguration place of the Scottish kings. The mound on which the Scottish kings were inaugurated still survives. But 18th and 19th century folklore associations with prehistoric monuments around Dunsinane and nearby Meigle - probably the site of an early medieval monastery - are probably mythological in nature and reflect increased antiquarian interest in Macbeth at that time.

The period between about 1790-1860 saw numerous excavations on Dunsinane. Although poorly conducted and published, these excavations attest to an imaginative and proactive approach to the study of Macbeth which has yet to be equalled. It would be naive to assume that archaeology could provide physical evidence of Macbeth himself. But the potential of archaeology to enhance our understanding of places with which Macbeth was associated and illuminate the age in which he lived is considerable.

Dr Nick Aitchison is the author of Macbeth: Man and Myth, published by Sutton last month at £20.00 (ISBN 0-7509-1891-8)