British Archaeology, no 24, May 1997: Features


Mapping the forgotten remains in towns

The archaeology of English towns is becoming better understood, writes Simon Denison

How much is known about the archaeology of Kidderminster? Or Ludlow? Or Evesham? Surprisingly, perhaps, the answer is not very much. That, however, is about to change.

A basic outline of the archaeology of 64 small towns in the Welsh Marches has now been drawn by an innovative five-year survey. Several industrial small towns, such as Kidderminster and Redditch, have been shown to have unexpected medieval origins, and the archaeology of some historic market towns and other sites has been mapped for the first time.

Meanwhile, an intensive survey of the archaeology of Newcastle upon Tyne has shed new light on the city's Roman origins and produced detailed maps of its surviving medieval and post- medieval deposits.

Each of these two projects is the furthest advanced of two series of urban surveys - one of small towns, the other of major cities - funded since 1992 by English Heritage. In addition to the Welsh Marches, small-town surveys are also underway in Somerset, Avon, Gloucestershire and Hampshire, while 30 cities are taking part in the intensive survey project.

The Central Marches Towns Survey, now drawing to a close, covered small towns in Herefordshire, Worcestershire and Shropshire, where the absence of much previous archaeological work had left the archaeology either little-understood, or assumed to be non- existent, despite the historical importance of some of the towns. The survey covered historic market towns, industrial towns, and shrunken or deserted towns such as the former Marcher boroughs of Clun and Wigmore on the Welsh border.

The project was intended primarily to produce a `historic landscape map' of each town, to help guide planners' decisions on the siting of future building developments. However, the survey has also produced new information about the history of some of the towns, by gathering together and mapping the results of historical research, records of standing buildings, known sites and monuments, and information taken from maps.

Redditch, for example, has long been considered a 19th century product of the Industrial Revolution, developed as a centre of the needle-manufacturing industry, and then largely redeveloped in the 1960s. It was assumed the town had little surviving pre- 19th century archaeology. The project, however, has mapped the results of recent research by Prof Chris Dyer of Birmingham University, which suggested Redditch was a semi-urban trading centre in the medieval period. Subsequent excavations over the past couple of years have proved Prof Dyer's predictions correct.

`For the first time ever, we are now finding medieval deposits in the town centre,' said the survey director, Hal Dalwood of Hereford and Worcester County Council.

Leominster, on the other hand, was well-known to have medieval origins, but the extent of its archaeology was little understood, according to Mr Dalwood. The project has now shown that the town's suburbs expanded and then retracted in the 12th and 13th centuries, not to spread out again until the 17th century. The later expansion, however, was not as intense as in the medieval period, leaving dense medieval deposits potentially undisturbed.

In Newcastle, the intensive urban survey has emphasised the city's Roman military origins, suggesting that civilian settlement may have been less widespread than was previously thought, according to Dave Heslop, County Archaeologist for Tyne and Wear. The survey's main contribution, however, has been to produce the first detailed, computerised map of the town's archaeology. `Not much synthesis of Newcastles archaeology had previously been done,' Mr Heslop said.

The survey has established, for example, that although Newcastle was the fourth largest town in England in the 14th century, no houses of the period have yet been identified. Remains of the town's medieval nunnery, however, were found to still exist. Once its former location was mapped, the nunnerys boundary was shown to run through the site of a proposed development in the city centre. As a result, in 1995 an excavation was recommended, and it produced evidence of the boundary wall, as well as rubbish pits dating from the nunnery's demolition, with large amounts of pottery and window glass.

`This was in a part of town where the archaeology would not have been expected to survive, but we located it precisely with a very small trench,' Mr Heslop said.

According to Roger Thomas, who supervises both survey projects at English Heritage, the intention is not to grade towns or parts of towns into areas that can or cannot be developed, but simply to give planners advance warning of the likely archaeological response to any development proposal. The result may be that, as each survey approaches completion, the prospect of accidental destruction of archaeological remains in England's towns will recede.


Return to Table of Contents Return to CBA Homepage


New evidence suggests how the Dover Boat was built and used. Peter Clark reports

Lessons from Bronze Age boat-building

When a large Bronze Age boat was discovered in Dover in 1992, about one-and-a-half times as long as a double-decker bus and substantially complete, it was immediately regarded as one of the most spectacular prehistoric finds of recent times. About two thirds of the boat was recovered, and since then it has been studied in detail by the Canterbury Archaeological Trust, and a part of it reconstructed, in order to answer questions about how the boat was built and where it sailed.

The boat (tentatively dated to around 1300BC in the Middle Bronze Age) consists, in essence, of four oak planks: two flat bottom planks and two curved side planks - although additional side planks had been removed in antiquity. The bottom planks were fixed together without nails or carpentry joints, but by ramming wedges and cross-timbers through a pair of upstanding ridges on either side of the main joint and through a series of cleats (or semi-circular wooden hoops). This technique seems strange, as the main joint appears to be a line of weakness where we would expect a strong keel. We know from other Bronze Age boats that carpentry jointing was known at the time, but for some reason it was not used here.

The side planks were stitched to the base by yew withies, and the seams made watertight by compressed moss. It seems, also, that the boat, some 15m long, was made out of single lengths of oak timber, requiring the use of massive trees with long, straight sections of trunk uninterrupted by branches.

This much was clear from studying the boat itself. However, only through reconstruction of a 3m section, using replicas of Bronze Age tools, did we begin to understand how the boat was built. First, two logs about 1m in diameter were split into four half- logs, and these were sculpted into the four base and side planks by hammering in wooden wedges to split off sections of timber. The roughly shaped planks achieved through this process were then finished off with metal tools - palstaves (bronze heads that can be socketed to form either axes or adzes), a socketed axe, a chisel and gouges.

Although the process was largely trial-and-error, the tool-marks on the reconstructed boat matched those on the original precisely, including parallel grooves running along the base of the boat, which had originally been thought to be possibly deco rative. This suggested that we had correctly identified the stages of construction and the toolkit used by the original boat- builders. Furthermore, we actually produced a better finish than on the original boat, suggesting the original builders were more interested in getting the boat completed quickly than in producing a fine finish.

The sheer scale of the boat suggests that it was a sea-going vessel, though this question is still hotly disputed. We know that boats did cross the channel in this period, because part of the cargo of a Bronze Age boat - mainly broken metalwork - was found on the seabed off Dover in 1974. Some, however, believe the Dover Boat was used only on rivers, as the contemporary boats from North Ferriby on the Humber are thought to have been. Yet the river Dour today is a small stream, flanked by steep hills, and it is difficult to imagine it being a broad tidal estuary like the Humber, where a sizeable boat was needed simply to cross from side to side.

The grooves on the bottom of the boat had been worn away in places, suggesting either that the boat had been beached - evidence for use at sea - or that it had rubbed against the bottom in shallow water. Preliminary study of environmental evidence surrounding the boat indicates it was abandoned in fresh water, but if the boat had been beached, its great weight (particularly when loaded with cargo) would have required the assistance of the tide to refloat it. A scrap of unworked shale found in the boat has been proved to come from Kimmeridge Bay in Dorset, which may suggest the boat was plying along the south coast 3,000 years ago. The waters below the high cliffs at Dover are little different from those out to sea, so even if the boat had hugged the shoreline, it would have needed to have had sea- going capabilities on all but the calmest days.

The sight of the finished reconstruction itself, however, was perhaps more evocative than all the lessons the boat has taught us about the past. The sturdy timbers, golden brown in colour and with the distinct odour of fresh green oak, could not fail to fire the imagination of even the most cynical onlooker or archaeologist. Eventually, we hope the finished piece will be displayed alongside the original boat in its new gallery in Dover Museum.

Peter Clark is Deputy Director of the Canterbury Archaeological Trust. The reconstruction project was funded by English Heritage.


Return to Table of Contents Return to CBA Homepage


England's towns were first planned at the same time as its villages, explains David Palliser

On the earlier origins of English towns

The origin of the towns and villages of England is one of the most disputed subjects of medieval history and archaeology. The traditional view is that most villages were created in the early Anglo-Saxon period and most towns - at least, most towns `planted' at one time - after the Norman Conquest. Increasingly, however, research is showing that this needs to be revised. It now seems that many towns and villages were created broadly at the same time, with the earliest towns actually preceding the earliest villages.

Earlier this century, the pioneer landscape historian WG Hoskins suggested that most villages were created from the 5th century onwards, but recent work in the East Midlands by the historian Christopher Dyer (see BA, June 1995) is finding the 10th and 11th centuries the most likely period for village formation. Similar conclusions are being provided by the archaeologist Mick Aston's ongoing project at Shapwick in Somerset.

No one believes any longer that villages were all created by the Anglo-Saxon invaders, but older views about the origin of towns are proving more tenacious. That is a pity, because towns had a key role to play in the commercialisation that made England so wealthy and such a tempting prize for the Norman invaders of 1066. Furthermore, since England was the first country in the world to become fully urbanised - with more than half its population living in towns - in the 19th century, it is important to understand the origins of English towns aright.

Medieval towns and villages were often similar in their form and fabric, but the similarities have often escaped comment, because they are usually considered by separate specialists on urban and rural settlement. The similarities did not, however, escape Christopher Taylor, who wrote in Village and Farmstead that the period from the 9th to the 13th century was one `when both towns and villages were being planned, perhaps for the same reasons' (my italics).

The creation of many towns between the Conquest and the Black Death is well enough known, with hundreds of places acquiring trading privileges and borough status, and in some cases with the laying out of new towns on virgin sites. The process, however, was well under way when the Normans arrived: `The decisive phase in early urban history came in the time of the Vikings and the West Saxon reconquest of England', as Edward Miller and John Hatcher put it in their recent book on towns and commerce, Medieval England. The period included not only the fortified towns (burhs) of King Alfred and his successors, and the trading towns of the Danelaw like York and Lincoln, but also the more numerous towns which were established next to major churches, the so-called minster towns.

Furthermore, as the historical geographer Brian Roberts has observed, many villages `adopt the same planning principles' as towns, and `belong to the same planning phase'. This is often overlooked, because villages seem often to have started with a simple plan (often a single or double row, a green, or a cluster), whereas towns tend to be divided between a majority with no very obvious planning, and a minority with regular layouts, usually discussed in terms of examples with a gridded layout. The two groups are often distinguished as `organic' and `planned', which is misleading, since towns do not `grow', they are planned - whether piecemeal or as a whole.

Many villages are now categorised as either simple in plan, or polyfocal (with several originally separate nuclei balling together), or composite (with later units added to an original core). The same is true of towns. Some, chiefly the smaller market towns, are of simple plan, but many seem superficially irregular because they are either polyfocal or composite.

Large cities like London or Coventry did not `grow' (the false biological analogy again) but were composite in plan. Careful analysis shows that they were planned in phases, sometimes of one or two streets, sometimes with a whole new quarter or suburb. Other towns were polyfocal in origin, including Cambridge, Shrewsbury and York (though Norwich, which has been claimed as polyfocal, is now reinterpreted as composite). The most striking example is Durham, where five or six separate communities (three of them called `boroughs') grew up at the same time as, or even before, the central `borough' on the Wear peninsula.

Other towns were planned more simply. At the humblest end were small market towns which were physically almost indistinguishable from agricultural villages. Towns like Battle or Chipping Campden were urban equivalents of the two-row or street village. The true gridiron-plan towns laid out in a single phase, like Salisbury or New Winchelsea, were always exceptional. Many so-called smaller grid towns were really nothing but cross-road settlements, with back lanes parallel to both main roads. And some larger grid-plan towns were less simple in layout than they seem. Ludlow and Hedon, both described as single-phase grids in Professor Beresfords New Towns of the Middle Ages, have been reinterpreted as complex plans of several successive phases.

In short, many towns and villages were apparently planned on similar lines and over a similar time-span. It is interesting that although most towns and most nucleated villages - where dating evidence of any sort is available - were created between the 9th and 13th centuries, at least some towns go back further. By contrast, no planned and nucleated villages (with the possible exception of West Heslerton in North Yorkshire) do so.

Seventh and 8th century England, therefore, sees already some cathedral towns like Canterbury, some minster towns or proto- towns such as Oxford and Reading, possibly some Mercian fortified towns like Cambridge and Hereford, and earliest of all the trading emporia of London (the Aldwych/ Strand site), Hamwic (Southampton), Ipswich and York. Some of these early `towns' may have been very straggling and scattered, but regular planning can be attested at Hamwic by about 700, at Ipswich about 800, and at Winchester and York about 900. The walled City of London, reoccupied in 886, also seems to have had a planned street grid originating in King Alfred's time, although the system as a whole may be 10th century in date. In view of all these dates of early urban planning, Brian Roberts was surely right to ask, in The Making of the English Village: `Did regular town plans precede and form models for the rural village?'

That question may seem absurd to those who still assume that small, rural settlements must emerge before large, complex towns. The American writer Jane Jacobs, however, has suggested that in the ancient world the emergence of cities may have preceded and helped to produce settled agriculture, rather than the other way round. Could the same have been true in the economy of the post-Roman West? The mechanisms might have included royal initiatives, from the interest of early kings in ports-of-trade, such as King Ine of Wessex at Hamwic, through the military and economic interests of the Mercian and West Saxon kings in creating networks of burhs in the 8th, 9th and 10th centuries.

Christopher Dyer's article in this magazine postulates that `with the growth of the state from the late 9th century, and the multiplication of the number of lesser aristocrats . . . there came an increased demand for taxes, rents and labour', leading to pressure for more efficient patterns of agriculture and rural settlement - but before that, we might suggest, leading to fortified centres for defence, markets and mints. Such a scenario would almost take us back to the view of TF Tout, the pioneer interpreter of medieval town planning in England, who wrote in 1917 that `the political necessity for town making arose earlier than the economic need'.

Admittedly, such a tentative model would fit some regions better than others. Royal sponsorship of towns is better attested for Wessex and Mercia than for East Anglia and much of the Danelaw. Some regions acquired planned towns and planned villages; others - notably East Anglia again - had few obviously planned towns and very few nucleated villages at all. James Campbell, in The Anglo-Saxons, has some fascinating suggestions about the differences between Winchester's `neat grid' and Norwich's `sprawling tangle', as part of `a pattern of distinction between East and West, which owes much to the policies and attitudes of the kings of the house of Wessex'.

Nevertheless, what has been suggested here would fit very well with Mick Aston's interpretation of Wessex at the very least, where the 10th century was a `boom time' for new towns and growing commercialisation, and where one consequence was that `go-ahead' lords like Glastonbury Abbey were concentrating peasants into new nucleated villages like Shapwick, and replacing their former scattered holdings by `cereal farms' (Current Archaeology, February 1997). In some regions, then, at least, new planned villages were preceded and inspired by the regular plans of newlyplanted towns.

David Palliser is Professor of Medieval History at the University of Leeds. He is currently writing a book on English medieval towns, and is editing the forthcoming medieval volume of the Cambridge Urban History of Britain.


Return to the British Archaeology homepage

Return to the CBA homepage


© Council for British Archaeology, 1997