ISSN 1357-4442Editor: Simon Denison

Issue no 31, February 1998


Roads that ramble, and roads that run

In the first of a new series of features on the historic landscape, Paul Hindle explains some of the peculiarities of British roads

Roads are one of the commonest features of the landscape, but probably the most ignored; they are taken for granted as we rush from one place to another. But as we travel around it can be fascinating to see what each section of road can tell us about its history.

Perhaps the most obvious feature of a road is its direction, and in particular how straight or sinuous it is. Straight lines are rare in the landscape, and straight roads are equally rare, usually being either Roman or related to enclosure. As Sellar & Yeatman said in 1066 and All That, ‘the Roman roads ran absolutely straight in all directions and all led to Rome’! In fact Roman roads are better described as being ‘direct’, and usually proceed in a series of straight alignments from one Roman site to another. They were designed this way for simplicity of construction. Many are marked on OS maps; but otherwise, the existence of Roman sites en route is a fair clue to a road’s Roman origins. Some still retain traces of the aggers, or ditches, that originally ran along each side; good examples can be seen along Ermine Street as it crosses Lincolnshire.

Straight enclosure roads, by contrast, laid out between 1760 and 1840, run through the then newly created enclosed landscapes of straight walls or hedges, which are mainly found in midland England (and much less in Wales or Scotland). Whereas Roman roads were often later used as parish boundaries, enclosure roads rarely run along the boundaries but were designed to give access from each village to its new fields and to neighbouring villages. These roads often bend and change width at parish boundaries, reflecting the work of two different surveyors who had each built a straight road, often of different widths, from their village to the boundary. If the roads did not meet up exactly at the boundary, a double-90° bend might result.

Most British roads are sinuous, though GK Chesterton’s poem probably does not reflect the true cause:

The rolling English drunkard made the rolling English road,
A reeling road, a rolling road, that rambles round the shire . . .
A merry road, a mazy road, and such as we did tread
The night we went to Birmingham by way of Beachy Head.

More likely is that country roads bend in response to the landscape, for example to avoid sometimes long-gone obstacles such as woods, ponds, bogs, depressions, even buildings. Some bends may also have been reinforced by the traditional traveller’s right to divert from a difficult section of road, even to the extent of trampling crops. Even modern motorways rarely go in a straight line for more than a couple of kilometres.

Enclosure also produced a distinctive width of road; standard widths of 60ft (18m), 50ft (15m) or 40ft (12m) were common, to allow horses and carts to get around any obstructed sections. With modern road surfacing, enclosure roads often look rather odd with a narrow strip of tarmac running, with wide verges, between walls or hedges up to 60ft apart.

There is a vast range of road widths, from packhorse tracks less than a metre wide (often paved, as in the Calder Valley in Yorkshire, or cobbled in parts of Lancashire), through country lanes and main roads to motorways. As a general rule, the wider a road, the more recent it is, though the physical remains of older narrower roads are often destroyed in the widening process. An exception to the rule is Roman roads, which were built to standard widths of up to eight metres. The standard minimum width of a road nowadays is four metres, designed to allow two cars to pass, and OS maps mark roads that fall below this width with a narrower yellow line. Many non-enclosure roads in country areas were originally ‘single-track’, but most were widened on a piecemeal basis during the first half of this century when they were first surfaced with tar. The continuing existence of single-track roads in some rural areas today is simply a sign of little traffic.

A curious feature of some old roads is the creation of multiple tracks running in parallel, either where there was nothing to contain the road, as in a drove road crossing high ground, or where travellers had chosen the easiest route available, such as when climbing a hill. The Icknield Way is thought once to have been a mile wide, and on maps one can see a whole set of parallel tracks below the Chiltern escarpment near Lewknor in Oxfordshire. In some places a modern road runs right down the middle of a set of old drove tracks, as at Winsford in Somerset. Again, in aerial photos one can often see a set of roads climbing a hill, only one of which is tarmacked as the modern road – an example is at Twyford Down in Hampshire (it has not been destroyed by the new motorway). Modern descendents of this old multi-track habit are dual carriageways and motorways.

The surface of roads varies from modern concrete and tarmac, compacted stone (as prescribed by McAdam), to unsurfaced tracks, whether of stone or compacted earth. In the last category are the ‘green roads’ which traverse many parts of upland Britain. The reason why some roads were tarmacked, and others not, is related to need. The numerous roads that lead up into fields, for example, were rarely needed for motor vehicles; while the huge network of drove roads, which was in any case always more or less independent of the rest of the transport system and by-passed the majority of settlements, fell into disuse as modern transport haulage systems were developed. A few drove roads have continued to be used, however, such as the ‘Welsh Road’ and Banbury Road in the Midlands which are tarmacked in places.

Street furniture is a ubiquitous feature of urban roads. In the country, there is much less of it, but what there is can be very interesting. Stone or metal bollards indicating distances to nearby places may be turnpike milestones, showing that the road was once turnpiked, probably in the mid-18th century. A few Roman milestones survive in situ, while quite a number of parish boundary-markers survive at the roadside. These often go unnoticed, and can be cairns, crosses, stones, iron plates, even trees. One example can be seen at the boundary between Ingleton and Bentham in North Yorkshire.

Virtually all urban roads have names, which may give dues to the history of the road and the area it served. Simple words like ‘Old’ and ‘New’ in road names can indicate which road came first, while landowners often have roads named after them, and many long-gone features may also be recalled by road names. In the country, fewer roads are named, but any names can be instructive. Most simply say where the road is going. A Welsh Road in the Midlands, or a Galloway Road in northern England were probably once drove roads, whilst a Coal Road probably led to a coal mine, and a Turbary Road to peat cuttings.

What lies beyond the road boundaries may also give clues to the road’s history. The sheer number of roads in an area will be determined by economic features in the landscape, such as towns and villages, mines or industry; while in earlier times the presence of forts and castles was also a reason for a road to develop. The existence of cultivated fields also requires roads to allow agriculture to operate.

But it is the physical landscape which has the strongest control on the routes roads take. Slopes are the most obvious control. Some roads contour around slopes, whilst others have to go up them, perhaps using curves or zigzags, the steepness of the roads being determined by the traffic that used them, and by the technology available. Thus Roman roads are often quite steep, as they were designed for men on foot and horses. Until this century, wheeled vehicles could not tackle steep slopes, but cattle could; so drove roads sometimes had steep sections, but the main roads did not. Footpaths, horse-tracks and cart-tracks are original, not modern, distinctions between roads, and they reflect not only the different widths but also the different steepness of roads. Rivers also control the routes which roads take, either by forcing them to travel along their banks, or by the choice of fording and then bridging points.

It may seem odd that height is a feature of roads, but there is a height limit to most roads today, reflected in the height of bridges and tunnels, determined by the size of vehicles in use, such as lorries and double-decker buses. Overhanging trees are cut to this same ‘loading gauge’. The reverse, depth, is also a feature; where an old road runs down a hill, centuries of traffic plus the effects of rainfall often lead to an over-deepening, and the creation of a ‘holloway’.

So when you travel around Britain, whether for business or pleasure, keep an eye on the very roads beneath your wheels and the tracks beneath your feet. They have a fascinating history, and many clues can easily be seen.

Dr Paul Hindle is a Senior Lecturer in Geography at Salford University

Further reading:
BP Hindle, Roads, Tracks and their Interpretation (Batsford), 1993;
A Raistrick, Green Roads in the Mid-Pennines (Moorland), 1978;
GN Wright, Turnpike Roads (Shire), 1992.

Return to Table of Contents | Return to CBA Homepage

Executions and hard Anglo-Saxon justice

Hanged or beheaded Saxon skeletons belonged to excecuted criminals, who were condemned to a tormented afterlife, writes Andrew Reynolds

When groups of early medieval burials are found with mutilated or dismembered skeletons, they are usually regarded as the victims of battles or massacres. The heroic campaigns of conquest undertaken by the early Anglo-Saxon kings eager to extend their territories, and the ravaging of the Viking armies from the late 8th century are commonly thought to provide the historical background.

The dramatic written records of early England give an impression of a violent society accustomed to strife, banditry and lawlessness. Such accounts, however, are contained in sources such as the Anglo-Saxon Chronicle, which tend to neglect the apparently more mundane aspects of life. The truth is that many of these groups of burials, especially those with decapitations and tied hands, probably represent criminals executed within an organised system of later Anglo- Saxon civil justice.

Once the larger kingdoms became more securely established from the late 7th and 8th centuries, the maintenance of peace and social stability required an enforceable and practical system of justice. The poetic and prose sources tell us about prisons at royal manors, open-air judicial courts held every four weeks, the ‘judicial ordeal’ overseen by the Church, and, finally, about the range of punishments meted out to those found guilty.

The law codes issued by English kings from the 7th to the 11th century mention a range of capital offences and, from the 10th century, state that executed wrongdoers were not to be buried in consecrated ground. The administrative structure of later Anglo-Saxon England is well-documented and has been studied in depth by historians, geographers and archaeologists, with the general conclusion that the Anglo-Saxon landscape became increasingly rigidly organised. By the close of the 10th century at the latest, each administrative district or ‘hundred’ can be seen to have had its own prison, court and place of execution.

Each hundred comprised a number of estates, and in southern and western England at least, Old English descriptions of the boundaries of many estates have survived. Significantly, such descriptions frequently mention places of execution as boundary markers. It follows that execution cemeteries must have been widespread in later Anglo- Saxon England.

About 20 cemeteries are known which fit the criteria of judicial execution sites. These are found in southern and eastern England, from Malling Hill near Lewes, Sussex, to Walkington Wold in the East Riding of Yorkshire. They can be recognised by the characteristics of the burials, and by landscape associations such as elevation and proximity to boundaries.

In recent years, osteologists have begun to consider in detail the pathological traces that hand-to-hand combat might have left on the human skeleton. Research suggests various blows to the upper body, particularly the arms and the head (see BA, October). Decapitations from the Anglo-Saxon execution cemeteries, however, are characterised by untidy, and in many cases excessively violent beheading from behind, probably with a sword. Blows intended for the back of the neck were delivered with far from clinical precision, to points anywhere from the shoulder blades to the top of the head.

A fear of the corpse rising up from the dead has been offered to explain the practice of decapitation. Particular fear appears to have surrounded the man from the Roche Court Down cemetery in Wiltshire whose head had been smashed and buried separately from the body within a ring of flints. Evidence for the public display of heads comes in the form of weather-worn skulls buried without the lower jaw, which had presumably dropped off, and from boundary descriptions which refer to heafod stoccan (head stakes) as markers.

Execution burials are also found facedown, and include the adult male from Meon Hill, Hampshire, buried with a large boulder on his back, and an inhumation from Guildown, Surrey, whose legs had been cut off at the knees. A superstitious motive also seems likely with the prone burials, especially as they are often the deepest graves at execution sites. In a few instances, bodies are found bent over backwards, perhaps suggesting that such individuals were put to death whilst kneeling in their graves.

The most common mode of execution, however, was hanging. There is evidence for two-post gallows structures from execution sites at Sutton Hoo, Suffolk, and South Acre, Norfolk, among others, but the majority of evidence is provided by the large number of corpses found with the hands tied behind the back. In certain instances, flexion (ie, tight clasping) of the hands has been noted and interpreted as an indicator of a violent death.

Execution victims were commonly buried in twos and threes implying contemporary execution, but in general little care was taken with the burials of wrongdoers, with no apparent concern for grave orientation, and the minimum effort invested in the digging of the graves (with the exception of the ‘superstitious’ burials).

Two-thirds of the excavated execution sites are associated with barrows of prehistoric and Anglo-Saxon origin, whereas the remainder are located upon linear earthworks. In general the sites afford commanding views, frequently within sight of important routes of communication by water and road. All of the excavated sites lie on boundaries, about a third on county boundaries, and all but two on hundred boundaries. Such strong locational tendencies suggest that the choice of earthworks and principal boundaries was common practice in the later Anglo-Saxon period.

Descriptions of the limits of Anglo-Saxon estates prove a fruitful source for contemporary terminology, and give further confirmation that execution cemeteries were located on hundred boundaries. The bounds (or boundary descriptions) of Havant, Hampshire, of 980, refer to the wearrihtan stocc, or ‘stakes where felons are righted’, those of Stanton St Bernard, Wiltshire, of 957 and 960, mention the wearhroda, or ‘felon’s cross’, atop the great linear earthwork Wansdyke. Other terms include heašenan byrgels (‘heathen burials’) and cwealmstowa (‘killing place’). The great majority of the places of execution and burial recorded in boundary clauses lie upon hundred boundaries, and many exhibit the same characteristics as the excavated sites including elevated location and inter-visibility with routeways.

The choice of the hundred boundary as a fitting repository for executed offenders probably reflects the desire to banish social outcasts to the geographical limits of local territories. The choice of elevated locations suggests a further visual motivation, as those travelling by execution sites would have looked up to see corpses hanging, in effect, between heaven and earth, being deemed unworthy of both.

The re-use of prehistoric features is of interest and the topic has attracted much recent attention (see BA, November 1997). The choice of such sites as suitable burial places for outcasts, seemingly reflects a fusion of Germanic lore and Christian imagery. Associations between barrows and dragons are well known in Germanic epic poetry such as Beowulf, where barrows were widely perceived as the houses of dragons and demons. It appears that, in addition to exclusion from consecrated ground, offenders were interred in places where they would endure eternal torment from supernatural monsters.

As for Christian imagery, a Biblical motivation for a number of aspects of Anglo-Saxon kingship has long been accepted, and an analysis of the machinery of the judicial system suggests strong influence here also. The lengths of prison sentences prescribed in the laws of King Alfred and his successors are, in Biblical fashion, 40 and 120 days. The choice of barrow mounds for execution sites could be interpreted as a representation of the Hill of Golgotha where Christ was hanged with the two thieves. At first a link between the imagery of the death of Christ and the execution of common criminals seems hard to accept, but the following passage from the Old English poem Christ III gives a contemporary viewpoint:

Those seduced by sins, dark evil-doers, will fearfully stare in distress [at the crucifixion scene] . . . They will see as their ruin that which would have best befitted them.

The frequent use of the term rod (‘rood’, or the crucifix) in the terminology applied to execution sites in estate boundaries, and the finding of triple inhumations at execution cemeteries, further indicates the use of the powerful image of the crucifixion by the authorities responsible for the maintenance of order.

These cemeteries, in short, reflect not civil unrest and disorder, but the efficient administration of systems of social control by the later Anglo-Saxon kings.

Andrew Reynolds is a Lecturer in Medieval Archaeology at University College, London

Return to Table of Contents | Return to CBA Homepage

When there is no end to a good game

Some board games have been played for 1, 000 years. Ian Riddler reports

It has been said that good games never die, but they continue to evolve. It is certainly the case that some modern board games have had a remarkably distinguished history – in particular chess and backgammon, which have been played in Britain continuously for about 1,000 years, and elsewhere for much longer.

The historical study of board games, rather like the games themselves, was long considered a frivolous occupation, beneath the dignity of serious scholars. Yet in recent years it has acquired a new respectability, with the compilation and re-evaluation of historical sources, accompanied by reviews of the burgeoning archaeological evidence.

Medieval and earlier games can roughly be divided into ‘war games’ involving skill, and ‘race games’ relying more on luck. The favourite Roman war game was ludus latruncularum, in which two players battled with a number of pieces of equal value. A complete set was recently found in Colchester. How exactly it was played is unclear, though, and the game had died out by the medieval period.

The Anglo-Saxons had a different war game, which they referred to as taefl. It was, it seems, the only board game they knew. Here, the outnumbered associates of a king, initially grouped at the centre of the board, attempted to escort him to the edge of the board, and to defend themselves against being taken by assailants. The game has Germanic origins which go back to the 1st century AD at least, and its unusual feature is the use of two sides of unequal numbers. The game is associated, in the Viking Sagas, with nobility. A late medieval source has King Cnut playing chess, which is most unlikely; but he almost certainly played taefl. Everyone did. In the Orkneyinga Saga, skill in the game is accounted one of a noble’s nine accomplishments, along with pulling an oar well and walking on snow-shoes.

At the turn of the millennium the situation began to change radically, and two centuries later England and Europe had entirely succumbed to the appeal of an Indian war game known as chess and to the reintroduction of a Mediterranean race game called tabula – an early form of backgammon. This, in an earlier incarnation as ludus duodecim scriptorum, had helped to satiate the Roman love for gambling.

The origins of chess remain obscure but its development can be traced from India westwards to the Islamic world. It appeared in southern Europe shortly before the millennium and spread northwards, acquiring the familiar chequered board as it did so. Historical and archaeological evidence indicate that it reached England after the Norman Conquest.

At first, chess was not a popular game. It was played with abstract figures and some, including the rukh (or chariot) and the elephant, were unfamiliar to a northern audience. The ‘old chess’ was a slower game than the modern variant, and both the queen and the bishop were relatively powerless. Games could last for days and occasionally, as with two 13th century incidents in London involving men of Essex, they could lead to deaths. By the end of the 12th century, however, the conquest of taefl by chess was complete. Figurative pieces had been introduced, the rukh was gradually supplanted by the castle or its keeper (although the name ‘rook’ persists), the bishop replaced the elephant, and the queen evolved from the vizier.

The Church, however, didn’t approve. A letter written in c 1061 by the Italian bishop Petrus Damiani, for example, denouncing his fellow Bishop of Florence for chess playing, prefigured a series of official ecclesiastical prohibitions all over Europe from the 11th to the 13th century. The game was associated with frivolity and the sin of gambling. The Bishop of Florence’s defence, that chess involved skill and was therefore ‘unlike other games’, formed a precedent for the counter-arguments of the following decades. During the 13th century, however, the Church gave in. The game had become too popular amongst the educated classes, many of whom were clerics. Finds of 12th and 13th century pieces come mostly from castles, manor houses, and ecclesiastical contexts such as abbeys and priories.

For many people, the Isle of Lewis pieces personify medieval chess. The remains of at least four sets – together with pieces from other games, such as tabula – were found in a sand dune at Uig, sometime before April 1831. Recent work has confirmed that they are of Scandinavian origin, and they can be dated to somewhere between AD1150 and 1200. It is now thought they were intended for sale in Scotland, and were deposited in a container in the dune for safekeeping. In their figurative design, however, the Lewis chessmen were unusual. Abstract sets were the norm, and complete or near-complete sets of abstract pieces are known from Adelsdorf in Germany, Sandomierz in Poland, and Venafro in Italy; while the fragmentary remains of a late 11th century whalebone set from Witchampton in Dorset represent the best surviving early English pieces.

By the late 15th century, chess had more or less reached its modern form. Draughts and dominoes also first appeared in western Europe at this time. Draughts was a derivative of chess initially played only with pawns or tabula counters, and finds of early pieces are extremely rare. Dominoes are slightly less so. Recently Europe’s earliest known domino, from the late 15th century, was found at Einbeck in Germany, while 16th century pieces are known from Oxford, very similar in form to those used today.

The other great medieval game, tabula, was played with 30 discoidal counters and with several dice, although these are seldom found with discarded sets, possibly because they were also useful in other games. In the last 15 years complete sets have been discovered at Gloucester (c 1120), St Denis near Paris (c 1150–1200), and Freiburg in Germany (c 1300), and these have revolutionised the study of the game. The Gloucester tabula set is elaborate, with carved antler figure scenes and highly decorative board points. It was an expensive, noble piece of equipment. Why it was thrown away remains a mystery.

The Freiburg set was played on a hinged wooden board, similar to those seen today, and the assemblage includes both chess and tabula pieces. The association between these two games was controversial for hundreds of years. In the earliest discoveries of chess and tabula pieces, and in the earliest documentary references to them (such as in wills), the games remain separate. Indeed, this separation presumably supports the Bishop of Florence’s defence of his actions.

However, the Lewis hoard, and an early–mid 12th century assemblage from Loughor Castle in Wales, contain pieces of both games. They are also found together in the ecclesiastical residence at St Martinin-Palace Plain at Norwich of c 1180. A German nobleman, meanwhile, referred to both games in his will of c 1180. By the end of the 12th century, therefore, tabula seems to have become socially as acceptable as chess, even if it was not yet fully accepted by the Church. The English remains of both games were sometimes burnt, however, and their remains are fragmentary. In some cases the board games are burnt, but other objects from the same deposit are not, and it is possible that they were destroyed as ludi inhonesti, games which everyone knew it was rather naughty to play.

Tabula required dice and some variants of chess were also played with them, as a means of accelerating a sometimes tiresome game. Dice, of course, were well known throughout this period as a vehicle for gambling, and the 12th century cleric and scholar John of Salisbury refers to ten different dice games, including tessara, calculus, dardana, pugna, taliorchus and vulpes. We know nothing of these games. Medieval dice are almost invariably cubical, whilst earlier Germanic dice included rod dice with four long sides and numbers placed in irregular arrangements, as with a recent example recovered from the Roman fort at Birdoswald.

From the Roman period onwards, dice could be loaded for cheating. Several Roman examples from Trier have been tested and found to roll mainly to specific numbers. A hoard of late medieval false dice have been found in London, with lead weightings skilfully applied within them, assuring that they would roll to particular numbers.

In 1734 Karl Linnaeus, the great botanist, travelled to Lappland and witnessed the local game of tablut, a game derived directly from taefl and adapted to local circumstances, with the Swedes defending their king against the Muscovites. A modern version of the game, based on his account, evolved in Britain this century, and it remains a testament to the durability of a good game.

Ian Riddler works at the Canterbury Archaeological Trust and is a specialist in medieval games

Now try this one

The most difficult medieval board game, devised in the early 11th century and known from books of the 16th century, was rhythmomacia.

It was played with round, square and triangular counters on an 8 × 16 board. In one version of the game, 28 of these were white, 29 black. Based on the idea that all numbers are multiples of one, except one itself, the aim was to form a harmony (either arithmetic, geometric, or harmonic) of three pieces on the opponent’s half of the board. The pieces had different values and moves, and captures of pieces involved mathematical calculations of the relationships between the numbers on them.

The game never really caught on.

Return to the British Archaeology homepage

Return to the CBA homepage

© Council for British Archaeology, 1998