Anglo-Saxon Immigration and the Early Medieval Landscape 

Nick Higham

Opinions popular in the early twentieth century envisaged England being populated in the fifth and sixth centuries almost entirely via mass migration from Denmark and western Germany, the so-called ‘Anglo-Saxon Settlement’.  Whole communities comprising numerous families were envisaged moving by boat across the North Sea and establishing a new and thoroughly ‘Germanic’ colony in Britain, cutting down the native forests to clear the land, found villages and plant fields, and cutting down as well, of course, any impertinent Britons who got in the way.  Their cemeteries, their houses, their metalwork, their language and the world of ideas which they inhabited were all deemed to be quite separate from what had been Roman Britain, befitting a ‘new’ people.

This picture has been subject to considerable change over the last half century, and has even been challenged to its core.  Some of the processes of change visible in fifth-century Britain can already be identified in the late Roman period, as for example the shift away from masonry and tiles and towards timber-framed building, and the landscape had already been long cleared by the time that Anglo-Saxon immigrants arrived.  The population of Roman Britain is generally now estimated at some 3-4 million, so wholesale replacement seems less probable than it once did. And pollen analysis suggests that there was no large-scale reversion to woodland; although regional variation occurred, the general trend was towards a de-intensification of land-use, as opposed to abandonment, suggesting population decline but not collapse, and the survival of large-scale Roman-period field systems in much of Eastern England right through to the twentieth century is very difficult to reconcile with any real break in exploitation of the landscape. 

Yet there certainly was a massive collapse in archaeological terms in the production of pottery, metalwork, coinage, masonry, sculpture and epigraphy in the late fourth and early fifth centuries, towns entered a period of terminal decline and there is still very little evidence for continuity at numerous sites, be they towns, villas or farms.  And new architectural styles appear connected with Germanic types of material culture, from the second quarter of the fifth century onwards.

These new architectural types centre on Sunken-Featured Buildings (SFBs) and hall-type buildings, which are often found in parallel on the same sites, sometimes stretching over considerable areas.  SFBs were clearly quite variable in size and complexity, their function is not entirely clear and may well have been quite various, and the precise nature of the pit over which they were constructed remains something of an issue.  Whether or not they were all built for storage and/or work space (particularly for weaving) remains unclear – they were initially diagnosed as habitative when first discovered by Leeds at Sutton Courtenay in the 1920s - and reconstructions vary enormously.  But their near ubiquity on early Anglo-Saxon settlements does imply a certain connection with the cultural shift that was occurring over the fifth century.  Yet the earliest insular examples are closely associated with late-Roman material culture, as at Monkton in eastern Kent, and these seem to be a style of building which should be associated with the frontier zones of the late Roman world in the North West, rather than specifically the barbarian world outside.    Whether SFBs are necessarily indicative of immigrants is therefore unclear. 

Hall-type buildings are similarly difficult to link very clearly with any particular ethnic group, occurring as they do in a thoroughly non-Germanic context at South Cadbury, for example, or Tatton Park.  Yet these post-in-hole or -in-construction- trench buildings are easily paralleled by examples across the Germanic world in the fourth-sixth centuries, as Helena Hamerow has demonstrated.  In England, they generally display a degree of organisation, so for example they are often oriented on a common axis and spaced out in a comparatively similar way, with or without clusters of SFBs in attendance.  Some fences have been identified dividing one from another but nothing in the way of new field systems, for example, and they seem to have fitted in to pre-existing and continuing systems of landscape exploitation.  There is a dramatic contrast between Romano-British villas and the hall-style buildings which effectively replaced them, but whether this was due primarily to migration or to fundamental shifts in social structures is difficult to say, and the two are in any case not entirely exclusive. 

A strong feature of the debate about hall-type buildings has been the assumption that they were the homes of farmers and representative of a social system without much in the way of hierarchy.  Yet at Mucking, for example, although the halls are comparatively uniform in size, there is evidence of a considerable gradient in the nature of goods buried with individuals in the two cemeteries, which may imply rather more social distinction than has been allowed.  There again, at West Stow for example, the nature of the archaeological material discovered – particularly the metalwork – makes it unlikely that these were just ‘farmers’ as the excavator surmised.  Rather, the site represents a community which was embedded in a system of elite exchange, rather than subsistence. 

At Mucking, there is also the view that this farming settlement shifted gradually over a couple of centuries, slowly moving from one end of the excavated area to the other, with the cemeteries in the middle.  But much of this is dependant on the life expectancy of these buildings and the durability of earth-fast posts, views on which have changed considerably in recent years.  There seems no obvious reason why a sixth-century hall should not have survived for 100 years or more, when similar structures from the later Middle Ages certainly did. 

Another difficulty with the view that these buildings were occupied en masse by immigrants is the discontinuity between architectural styles in Germany and England.  Massive long houses were characteristic of such terpen settlements as Feddersen Wierde across the third to fifth centuries, built to provide accommodation not only for people but livestock as well.  Such longhouses are virtually non existent in England, despite the clear association of material culture with the same regions.

The earliest Middle Ages displays, therefore, a confusing pattern of continuity and discontinuity as regards its settlements.  As regards material culture, such items as styles of clothing and decoration, pottery and metalwork all bear the stamp primarily of Anglo-Saxon manufacture, although these were not just lifted from any one settlement or locality in Germany and re-located in England.  But as regards architecture, there is a real fluidity, with new and remarkably consistent styles of housing being adopted which do not precisely map onto either Roman Britain or Germany, yet within an environment with strong roots in the Roman and British past.  We need to think of early Anglo-Saxon England as neither entirely immigrant in origin, nor entirely a local and insular development, but as a society which evolved out of processes of contact and social transformations which resulted in very different types of settlements than those which had previously dominated in either of the major donor societies.      

Sam Turner

Awaiting Synopsis

Fortifications in 12^th-century and early 13^th-century Ireland

Kieran O’Conor

Our knowledge of the development of castles built by the Anglo-Normans in Ireland in the years after 1169 is now quite detailed in comparison to what we know about contemporary fortresses built by Gaelic Irish princes and lords. Two aspects of Anglo-Norman built castles, in my opinion, however, need to be studied in more detail. Firstly, the long-held view in Ireland that mottes were ideal in an invasion context because they could be quickly erected needs to be questioned in the light of research carried out in Britain since the 1960s. There is growing evidence to suggest that there is little concrete evidence for motte building across Leinster and Meath during the early to mid 1170s. The first-large scale building of mottes in Ireland seems to occur in the early 1180s. The actual places used during the invasion period by the Anglo-Normans consist of a variety of different forms of fortification, including large ringworks, walled Hiberno-Norse towns and embanked and palisaded Irish monastic towns. These fortifications have one thing in common, despite their differences. They are all sufficiently large enough in area to adequately accommodate a field force bent on conquest – something that even a large motte and bailey could not do

Secondly, historians often bemoan the fact that there is little in the way of surviving documents detailing the social and economic development of the Anglo-Norman colony in Ireland during its first hundred years or so. In my opinion, they should look to the discipline of archaeology to fill this gap in our knowledge. The excavation of an important motte castle, for example, built in the last years of the 12^th century, would provide much needed information about the initial colonisation period, the developing manorial economy and rural life during the important first hundred years of the colony.  

There is relatively little evidence for European-style masonry and timber castles being built by Irish dynasts during this whole period. Most castles built in Ireland between the 12^th century and c. 1380 were constructed by the Anglo-Normans or their descendants. Irish princes and lords demonstrated their high status in different ways other than building great castles. Again, linked to this, combined archaeological, historical and literary evidence suggest that early medieval settlement forms such as crannogs, cashels and less clearly earthen ringforts continued to be used by the Irish into the high medieval period and beyond. The difficulty here lies in being able to identify which crannogs and ringforts continued to be occupied into the 12^th century and beyond without major excavation, which is both costly and time consuming.

One question remains unanswered. There are a number of pre-Norman references in the annals to princely fortresses that are described as caisléin or caisteoil. Again, certain historians and archaeologists, noting the increase in warfare in late 11^th- and 12^th-century Ireland and the centralizing tendency of various provincial kings, have argued that these sites represent something new in terms of fortification and were possibly a pre-1169 Irish importation of Norman-type castles from England and Wales. Some attempt has been made to suggest that these caisléin or caisteoil sites were mottes despite the fact that no evidence for such castles exists in the vicinity of these places. Certainly one site mentioned as a caislén in the annals appears to have been a crannog. Work being carried out in NUI, Galway by Paul Naessens in particular is suggesting that at least some 12^th-century caisléin or caisdeoil sites in Connacht consist of what can be described as impressive cashels whose extra-high walls were built of mortared stone blocks rather than drystone ones. No battlements, arrow-loops, gatehouses or stone keeps exist at these sites – they are simply very impressive cashels. These sites do represent an increase in defensibility over earlier and indeed contemporary cashels and ringforts and presumably this is why new words are used to describe them. Yet these sites are still demonstrably within the native tradition of fortification.

 It is imperative that scholars interested in reconstructing the landscape of the parts of Ireland that remained in some way under the control of Irish dynasts during the 12^th, 13^th and 14^th centuries realize that there was much continuity in terms of settlement type from the preceding early medieval period. Incidentally, the lack of Irish-built masonry castles before c. 1380 makes the widespread adoption of the tower house form of castle by them during the 15^th and 16^th centuries an important phenomenon that has yet to be fully explained.

‘Mountainous and rocky and almost wholly inarable’: conflicting cultural perceptions of landscape in 12^th- and 13^th-century Scotland 

Richard Oram

Conflicting perceptions of wealth, value, and economic development, medieval and modern, have together served to produce a badly skewed view of the socio-economic systems and settlement patterns of medieval Scotland.  Uncritical acceptance of models developed for lowland England, further coloured by the post-Improvement view of the supremacy of arable regimes over pastoral, has given rise to an image of agriculture in medieval Scotland as underdeveloped or primitive.  This image has been reinforced by the traditional historiography of 12^th- and 13^th-century Scotland, which presents the period as one characterised by cultural changes – social, economic, technological and artistic - brought about by the settlement of colonists of diverse backgrounds but generically labelled as ‘Anglo-Norman’.  Settler society and the cultural baggage which they imported have assumed the status of the norm against which the indigenous cultures of medieval Scotland have been measured.

One modern perception of conflict between a dominant colonial culture supported by the Scottish crown and a sullenly conservative Gaelic nobility is grounded in the language of the medieval records.  Produced by and for that same alien cultural group, the chronicles and charters provide prima facie evidence for a collision between native and newcomer that is viewed commonly as a clash of cultures and ideas rather than politically motivated conflict.  Wrapped in the language of Frankish cultural superiority and projecting the propagandist claims of the kings of Scots to mastery of all of what is now recognised as Scotland, these documents present conflict in terms of civilisation against barbarity and rebellion against legitimate authority, of an innovating ‘feudalising’ monarchy against a retrograde, conservative and xenophobic Gaelic nobility, rather than as a civil war fought for possession of the crown.  It is a representation which has helped to entrench a view of Scotland as a land caught between economic development and social or cultural backwardness, but this perception owes more to 19th-century socio-economic agendas than to the realities of 12th- and 13th-century Scotland. 

A growing body of research is emphasising the continuation of quite different perceptions of relative value in types of land between native and newcomer through the 12^th and 13^th centuries and beyond.  While some colonists may have gravitated towards zones of arable potential and preferred settlement as lords over peasant communities practising a mode of agricultural exploitation with which they could most readily identify, Gaelic lords maintained their interest in the uplands and a style of lordship supported by a primarily meat- and dairy-producing pastoral regime.  Post-medieval perceptions of the backwardness of the Gaelic cultural zones of Scotland, particularly in the 17th- and 18th-centuries, where ‘improved’ agricultural capacity was used as the measure of economic potential, have served to entrench the medieval image of an economic dichotomy expressed purely in terms of arable capability.

Negotiating the settlement of ancient commons in medieval Cornwall

Peter Herring

Medieval lives were enjoyed and actions performed in complex inherited places, shaped to varying degrees by prehistoric and Roman-period people.

Cornish commons had a particularly deep history, probably having been used as such since the mid second millennium BC, extended and reorganised in the early first millennium BC, and then gradually divided and organised on more local levels in later prehistory, and the Romano-Cornish and post-Roman periods. Throughout it seems that people took flocks and herds to the commons in the summer months, practicing a form of transhumance, and developing relationships with worlds in which there were numerous remains from earlier prehistoric periods – standing stones, circles, rows, cairns, etc – as well as the natural monuments like tors, pools, springs, marshes around which culturally meaningful associations developed.

When lords encouraged settlement of these uplands in the early second Millennium AD, they were therefore not simply compromising the established rights of commoners, gradually reducing summer grazing and fuel grounds. They were also disturbing individual and communal relationships with a familiar inherited world.

Most later medieval settlements on Bodmin Moor were small hamlets (2-6 households) established within ringfenced townlands in which hamlet-level ‘commons’ were separated from the manorial common beyond. Abandoned hamlets survive well on Bodmin Moor. Farmsteads of longhouses, individually held cowhouses, outbuildings, yards and mowhays were carefully arranged to be away from the communally defined and used townplace, reflecting what was probably a universal tension between the basic economic and social unit, the household, and the unit of agricultural cooperation and action, the hamlet.

Prehistorians have for some time been studying how memories, meaning and communally understood symbols framed and influenced action in Neolithic, Bronze Age and Iron Age Britain. Garden historians and archaeologists do the same for post-medieval ornamental landscapes and a small industry considers how our own lives are guided by a plethora of often conflicting symbols and messages.

For the medieval period good recent archaeological work shows how churches, castles, deer parks and early gardens etc were designed to be approached and appreciated in subtle ways to reveal similar cultural information.

An experiential archaeology of medieval peasant life, however, has yet to be fully developed. As a result, our interpretive models, still document-led, continue to place medieval peasants within controlled and functionalist worlds, denying them freedoms and agency as severely as any feudal lord ever did. And yet we know these people moved around, and left undisturbed many prehistoric monuments, curated them even. This visible, experienced and challenging inheritance from a distant past must have been discussed, mythologised and so negotiated, just as modern archaeologists help our own society cope with the very same monuments. We also know, perhaps too well, that medieval people carried with them structured but nevertheless personal views of power relations, sensitive appreciations of property and tenure, problematic schemes of Christian ideologies, and as we have seen in the hamlets, tense awareness of distinctions between the individual and the communal.

Daily or occasional encounters with material symbols or reminders of this problematised world must have reinforced or revised relationships, for instance setting off oppositional thoughts: owned/held; individual/communal; mine/his/ours; inherited/altered; certain/uncertain; safe/risky, etc. All of these must surely have influenced ways of experiencing, being, pondering and doing. We know from court rolls that medieval peasants certainly acted, often against convention, custom and law. They were not so very constrained. On Bodmin Moor we may see the apparently bottom-up, locally-determined reorganisations of hamlets and strip field systems that left the individual rather than the communal dominant as being a direct result of tensions between the two, tensions that had been reinforced virtually every waking moment.

So few of the processes that prehistorians study are now considered to have been unmediated by personal agency that functionalist economic and social factors are rarely resorted to for sole explanations of change. This may prove to be less the case in medieval Britain where commercial and political forces were probably stronger. But an experiential medieval archaeology will surely not only enrich representations of peasant lives, but also open up access to a much wider range of plausible causes of change.

Colonising the uplands of Wales: changing perceptions

Bob Silvester

Wales and its rural medieval settlement doesn’t slot readily into the concept of POMLAS, the reason being that the study has not advanced as fast nor as far as in England, though we are now probably at the end of an era, and one would like to think at the beginning of a new one, with the publication of Cadw’s Lost Farmsteads, and as importantly the compilation of the data that informed the various papers within it.

Much of the information on medieval rural settlement available to us comes from the uplands, and in one respect this is, perhaps, not entirely surprising as in the medieval period Wales exhibited vast tracts of unenclosed upland. While no reliable figures exist, in 1873 5.2% of England as a whole was classed as common land, in Wales it was 13.8%.  And it was not just about the great wastes extending along the central mountainous spine of Wales, but numerous smaller commons in rural areas and close to towns, many of which disappeared as late as the 18^th century. Neither the large upland commons nor their lower land counterparts were marginal, except in terms of lowland agricultural exploitation. They were lands to be used, providing a range of resources which played their part in an integrated agrarian system that had developed over hundreds of years without any fundamental political interruptions: not only grazing grounds, but also fuel, fodder, bedding materials, and water supplies.  

Of the two elements that constitute the use of the medieval uplands – permanent settlement and seasonal activity – only the former can be classed as a colonisation movement, and generally it seems to have been a small scale. Settlement growth implies population growth, the scale of which has yet to be adequately revealed for Wales. The evidence for medieval farms and fields extending onto the waste is not as widespread as might be assumed, rather it is localised. The Berwyn, still the upland with the best evidence, does not suggest that this was organised or controlled settlement, but opportunistic land intake, captured for agriculture as and when it was required. The waste thus provided an exploitable resource within striking distance of the core settlement in the valley below, and if there was manorial control, it is not apparent. 

Seasonal use of the waste ought to be straightforward, but isn’t. The associated settlements are the most widely spread elements of the medieval uplands, the hafod a part of a well-organised agrarian system, but arguably they remain the least understood. Existing perceptions of the use of the waste for seasonal grazing range widely, and quite a number of issues remain to be clarified, but what we can say is that it is not a colonising movement.  The uplands were not ‘new’ landscapes in the sense that the wetlands were, but have a long history, which perhaps originated in the prehistoric period, and was certainly in place before the 8^th century AD. It has been assumed that the Welsh were completely given over to seasonal grazing, with those who tended the herds travelling long distances into the hills in what might be considered classic transhumant fashion, but instead we should probably see a more pragmatic method of ensuring that the home farm could still be observed while the stock were grazed for the summer on the hills. This would seem to be the explanation for the presence of hafod sites around the edges of the commons. There seems to be little documentary evidence for this seasonal activity, and where it does exist it may be that the circumstances were atypical, as with seasonal usage of the Forest of Snowdon.

Within the compass of POMLAS the detailed area study may appear rather old-fashioned, but in Wales it is argued that there is a still an important place for it, not least because so few areal studies have been conducted.  The upper Vyrnwy valley is cited because it is an area where much systematic fieldwork has been completed, and where the balance between permanent agricultural settlements and seasonal hill settlements can be readily observed and assessed.

Perceptions and usage of marshland after Guthlac and Postan

Mark Gardiner

Wetlands, perhaps more than any other type of landscape in Britain, have evoked a wide variety of perceptions. In literature, for example, the Thames marshes are used to dramatic effect in the opening chapters of Dickens’ Great Expectations as a liminal place haunted by the escaped convict, Magwitch. The Fens, in a much earlier evocation of marshland, Felix’s Life of St Guthlac, is similarly used as a forsaken place haunted by demons which needed to be tamed by the saintly figure. Archaeologists and historians too have not been dispassionate in their appraisal of marshland. Often they have viewed it as watery waste, sometimes as an environment with a rich wildlife of fish and fowl to be exploited by hunters, but almost always they have praised the extraordinary late medieval works which tamed the floods and transformed the marsh into fertile farmland.

The more recent views of the medieval marshlands provide a starting place for consideration of this issue. Much of our thinking about waste has been framed in terms which might be characterized as ‘Postanian’. This is not to suggest that they owe their origin to the work of Michael Postan alone, but rather to the generation of rural historians who were writing in the early 1950s, but whose ideas had begun to crystallize in the late 1940s. Many of that generation were present at a seminal meeting convened by Postan in Cambridge in June 1948 when Axel Steensberg spoke about his experience of excavating medieval sites in Denmark. One of those was Maurice Beresford, who later summarized Postanian thinking with the phrase, ‘the journey to the margin’, describing the outward expansion of population under the pressure of growing numbers from the lighter loams and on to the heavy clays and poor sands. These marginal soils included also the uplands, and the areas still occupied by woodland and the marshes. Beresford illustrated this perspective with a number of photographs in Medieval England: An Aerial Survey (1958), including a depiction of assarts made high on the moors at Cholwich in Devon. Hoskins, another Postanian writer, shows a similar approach, depicting the advance of settlement into the Lincolnshire marshes in the direction of the coast. However, wetlands were rather different from uplands, for the land was not intrinsically infertile. Wetlands, Postan argued, lay on the ‘technological margin’, requiring considerable knowledge and ingenuity to drain and protect from flooding, but, like the poor soils of the uplands, were not exploited until the demand for food forced people into those areas.

This Postanian or neo-Ricardian view of medieval agriculture has been said to ‘resonate through the regional chapters of the second volume of the Agricultural History of England and Wales, published in 1988’. This volume represented a high-point of enthusiasm for the model. As historians began to take a critical view of the assumptions which underlay the model, they became increasingly critical of its explanatory value for the period between the tenth and sixteenth centuries. Bailey, in particular, argued that the concept of marginal land was based upon a number of unrealistic assumptions. There were not inferior lands awaiting colonization as the pressure of population increased. Marginality was partially a function of economy and society: it was not intrinsic to the soil. His critique led to the development of a more complex view of the operation of medieval agricultural economics, which included the costs of moving goods to the market and the demand represented by those markets or level of commercialization.

The Postanian view was intrinsic to the development of thinking about late medieval landscape history. The idea that medieval populations pushed outwards in the years before 1300 and expanded into new lands runs through Chapter 3 of Hoskins’, The Making of the English Landscape. Indeed, even now it remains at least residually in the common perception of landscape history, and marshlands, more than any other environmental type, seem to provide substance to the Postanian view. There, the frontier apparently can be mapped as a series of marshland embankments which progressively took in larger areas of land. Lands which were waste were transformed into fertile arable or pasture, and refinements of the chronology of embankment, which places much of the work in the late twelfth and early thirteenth century when the population was growing strongly, have appeared to support this view.

Recent work has questioned this view of marshlands. One misapprehension has been to imagine that both salt marsh and fen have limited value and usage before they were enclosed and drained. It is wrong to envisage these environments as dangerous or impassable terrain. Unenclosed marshland may be firm ground and often will provide grazing land suitable for cattle and sheep. Indeed, the act of grazing may improve the land, promoting the growth of grass instead of sedge. Studies of coastal salt marsh in Norfolk and excavation in Romney Marsh have shown that land was used in the past in this way. Trackways were laid out across the marsh, bridges constructed over the larger tidal creeks and some of the land was even divided into fields. Equally, work on the Fens have shown how the unenclosed freshwater marsh might be divided into long strips of meadowland known as ‘doles’. The purpose of embankments was to improve marshland further by controlling the water-levels, so that it might be used more reliably for arable or pasture. The risk of flooding could be reduced and the land used more productively.

We may substitute for the Postanian view of landscape history, a rather different one which considers land in terms of investment and intensification of usage. As the demand for pasture and arable rose as a consequence of a growing population, it was worthwhile investing in measures to increase the productivity of land. Instead of viewing the increase in the productive capacity in the eleventh, twelfth and thirteenth centuries in terms of extensification (the expansion of area of land) and intensification (getting more from the same area), we should regard the changes in landscape quite simply as different forms of intensification. In marshlands, a low stocking level might be replaced by higher numbers of animals, and arable or meadow might replace pasture.

We can turn now to medieval perceptions of wetlands and consider how far they reflect this rather different view of the transformation of marshland. The Life of St Guthlac stands apart from later medieval literature for it adopts an unusual, negative appraisal of marshlands. Henry of Huntingdon, who was a native of the Fenland edge, wrote in the early twelfth century about the area around Ramsey, ‘the marshland of which I am speaking is very wide and beautiful to behold’.  The marshes around Thorney are described by William of Malmesbury as a smooth plain on which trees for timber grew and grass flourished. Likewise Matthew Paris, seeking to explain the gap between Guthlac’s Life comments,

‘A marvellous thing has happened on these marshes in our time, which was, that where in past years they had been pathless and inaccessible, and where there were no means of travelling for men or cattle, and no habitation, only sedge, deep mud, and marshy beds of rushes, inhabited only by birds, not to mention evil spirits… those places are now converted into charming meadows, and even into arable land. Those parts of the same which do not produce corn or hay, supply an abundance of sedge, peat, and other fuel, useful to the inhabitants.’

No doubt the marshes had been transformed within living memory, but it seems that Paris was in fact trying to bridge the gap in perception between Guthlac’s Life and the then prevalent appraisal of marshland. Marshes were generally seen as places of abundance and potential, a view reflected, for example, in the Liber Eliensis which describes the conditions in Hereward’s camp, even though he was besieged by the Normans.

Postan and Guthlac have been used to typify some of the attitudes to marshland. We need to think carefully about the idea of the colonization of marshlands, since, as I have argued, this was not the movement of peoples into a new and empty space, but a process by which it was used in a progressively more intensive manner.

‘Uncommonly rich and fertile’ but ‘not very salubrious’: the perception and value of coastal wetlands

Stephen Rippon

In the 20^th century, coastal wetlands were often perceived as being of relatively low or moderate agricultural value, although in the past they were seen very differently. In the 18^th and 19^th centuries agricultural writers were impressed with the fertility of the soils and the rich pasture and meadows that they supported. In the 16^th century, John Leland described how the position of Banwell, on the edge of the North Somerset Levels ‘with the fennes close by, is not very salubrious, and Wick is worse’. This was, however, very much an outsider’s view as a survey of the manor of Congresbury - that extended across both the coastal wetlands and the adjacent dryland areas - rated the marshland as the best agricultural land. Clearly, perception of the value of a landscape depends on how well informed the writer is (and any preconceptions they might have about a particular environment), and the changing socio-economic climate within which agriculture is practiced.

By the end of the medieval period most coastal wetlands around Britain were protected from tidal flooding through the construction of earthen embankments and subsequently drained, a process known as reclamation. It is a common assumption that this was achieved as a deliberate policy of agricultural improvement with major landowners playing an important role. In some cases this may well have been what happened, but on the North Somerset Levels at least the process appears to have occurred in a more piecemeal fashion: while the end result was a substantial sea wall built all along the coast this was not the original idea. Initially, the landscape appears to have been modified through the construction of low embankments that acted as seasonal flood protection for relatively small areas. In some cases the construction of dams and sluices across creeks and rivers could have achieved the same thing. Over time, more land was enclosed, and the embankments raised to provide year-round protection. Eventually, the extent of individual reclamations grew to the point at which they could easily be joined up, with the result that there was a continuous sea wall all along the coast: a feature that, while vital to the present character of the landscape, was not created by design.

This very piecemeal approach towards reclamation suggests that it was local communities who took the lead, rather than the major landowner - the bishops of Bath and Wells - and this is also seen in the character of the landscape that was created following embankment. Areas of both nucleated settlement and common field, and dispersed settlement and closes held in severalty, were created on land belonging to the bishops suggesting that they took little direct interest in shaping how their estates were managed. These reclaimed wetlands were very much the product of local farming communities and their perceptions of how best to manage their environment.