By Stephen Rippon (Archaeology Department, University of Exeter)
and Benjamin Morton (School of History, Classics and Archaeology, Newcastle University)
The aim of this paper is to review progress in medieval settlement research published between 2007 and 2016, and as such it follows on from Gardiner’s (2006) previous review that covered 1996–2006. The geographical scope is Britain (England, Scotland, and Wales), and while the structure is broadly thematic a division is sometimes made between the early (c. 410–1066) and later medieval (c. 1066–1500) periods. Examples are provided of published research that illustrates a particular issue, but it is impossible to provide a complete bibliography. Due to the volume of data now being published, the focus has been on rural settlements and their associated landscapes (as opposed to towns, castles, and monasteries). Important collections of papers published within this review period that embrace the wealth of evidence we have for studying the wider medieval landscape include Gardiner and Rippon 2007; Gilchrist and Reynolds 2009; Higham and Ryan 2010; 2011; Hamerow et al. 2011; Turner and Silvester 2012; and Christie and Stamper 2012.
During the long history of research into medieval settlements and their associated landscapes there have always been regional disparities in the areas where fieldwork has occurred. In the 1960s and 1970s, the golden era of deserted settlement research, there were distinct concentrations of surveys and excavations where particular researchers were based, such as the pioneering fieldwalking surveys of East Anglia and excavations of deserted settlements on Dartmoor. There was, however, a general bias towards England’s Central Province where so many well-preserved DMVs were to be found and were threatened by ploughing and development.
In recent years the context in which archaeological fieldwork takes place, and where it occurs, has changed dramatically with the vast majority of projects now being development-led and in response to conditions imposed as part of the planning process. Mapping the distribution of fieldwork that has revealed medieval rural settlement in recent years is beyond the scope of this paper, but it will be broadly similar to that related to Romano-British rural settlement (Smith et al. 2016, fig. 1.1) with notable hotspots such as the environs of Bedford, Cambridge, Milton Keynes (in Buckinghamshire), and Stansted Airport (in Essex) where major urban, industrial, and infrastructure developments are concentrated. Regional Research Frameworks provide summaries for each region of England (https://historicengland.org.uk/research/support -and-collaboration/researchframe works- typologies/research-frameworks/).
An exciting outcome of the recent expansion in developer-funded archaeology is that surveys and excavations are also now happening in parts of the landscape that have previously seen very little archaeological investigation. Major linear developments – such as new roads, railways, and pipelines – in particular provide important transects through the countryside (e.g. HS1 through Kent: Andrews et al. 2011; Booth et al. 2011; A1 in North Yorkshire: Brown et al. 2007).
While development-led archaeology is leading to the investigation of some wholly deserted sites, there is also a growing body of data from currently occupied but shrunken medieval settlements that is starting to transform our understanding of village origins. One example is Wright’s (2015) demonstration that ‘Middle Saxon’ occupation beneath later medieval villages is far more common than previously thought. Another approach towards investigating currently occupied medieval settlements – in particular as part of community archaeology projects – is the digging of small test pits in back gardens (e.g. Shapwick in Somerset: Gerrard with Aston 2007; the English CORS Project in eastern England: Lewis 2016).
Although there are still some major university-led research- and training-based excavations of medieval settlements (e.g. later Saxon Bishopstone, in Sussex: Thomas 2010), these are now a very small proportion of the total fieldwork being undertaken. The focus within the higher education sector has instead shifted towards programmes of non-field-based research. The latter includes a series of recent ‘big data’ projects, synthesising the results of published reports and grey literature, although their focus has been on the Roman and early medieval period (the Roman Rural Settlement Project: Smith et al. 2016; English Landscape and Identities Project (EngLaId: https://englaid.wordpress. com/); The Fields of Britannia project: Rippon et al. 2015; and Blair’s Building Anglo-Saxon England published outside this review period in 2018).
In a period with so much data it is easy to focus on Britain, or more often just one region of Britain, but there is also a need to broaden our horizons. A number of important studies placing medieval Britain into its wider European context have been published (e.g. Renes 2010; Curtis 2013; Loveluck 2013), and more comparative studies of that type are needed.
Important progress has been made on publishing some key backlog sites (e.g. Caldecote in Hertfordshire: Beresford 2009; Flixborough in Lincolnshire: Dobney et al. 2007; Evans and Loveluck 2009; Loveluck 2007; Loveluck and Atkinson 2007; Glastonbury Abbey in Somerset: Gilchrist and Green 2015; Raunds in Northamptonshire: Audouy and Chapman 2009; Chapman 2010; St Peter’s church, Barton-upon-Humber in Lincolnshire: Waldron 2007) and long-term research projects (e.g. Brixworth in Northamptonshire: Parsons and Sutherland 2013; Shapwick in Somerset: Gerrard with Aston 2007; Wharram Percy in Yorkshire: Mays et al. 2007; Wrathmell 2012). There remains, however, a growing problem of developer-funded work not being published (although many reports are lodged with the ADS Grey Literature Library), and too many palaeoenvironmental reports on important assemblages of faunal remains and plant macrofossils being left at assessment stage (see Van der Veen et al. 2013 for a review).
While recent research has shown that the Romano- British landscape was not as spatially or temporally uniform as we might once have thought (Smith et al. 2016), the pace of change – and the development of regional variation in landscape character – appears to have accelerated during the early medieval period. There has been much debate about what happened to Britain’s countryside as it ceased to be part of the Roman Empire, with perspectives ranging from crisis through to continuity (e.g. Gerrard 2013; Haarer 2014). The following centuries were characterised by three major phases of invasion and migration – Anglo-Saxon, Scandinavian, and Norman – that led to changes in the physical structure of the countryside and its place- names in some parts of Britain, against a backdrop of the survival of a substantial native population. There is growing evidence that large parts of Britain shared in an intensification of agriculture during the ‘long eighth century’ that was associated with the emergence of the Christian church, the re-structuring of territorial arrangements, and the revival of trade (e.g. Hanson and Wickham 2000; Wickham 2005). This was followed by the period of profound social, political, and economic upheaval brought about by the Scandinavian invasions and subsequent settlement of parts of eastern and northern Britain, the establishment of the unified English State, and, finally, the Norman Conquest. The Oxford Handbook of Anglo-Saxon Archaeology (Hamerow et al. 2011) provides excellent introductions to the diverse evidence from the early medieval period as a whole and includes discussion on the themes of identity; rural settlement; mortuary ritual; food production; craft production and technology; trade, exchange, and urbanism; the body and life course; religion; signals of power; and the relationship between archaeology and other disciplines.
Debate over the extent to which the Romano-British landscape – in terms of its land-divisions and patterns of land-use – survived into the early medieval period continues, although discussion is hindered by the traditional periodization of archaeology that leads to separate discussions of the Roman and the medieval periods. The Fields of Britannia project, however, explicitly set out to bridge this divide by exploring the relationship between the Roman and medieval countrysides (Rippon et al. 2015). Using the large corpus of archaeological evidence which has been generated since the introduction of developer-funded archaeology in the early 1990s and a synthesis of pollen evidence, it argues for a strong degree of continuity albeit with some marked regional variations (although Williamson (2016) has suggested that at least some of this potential continuity could have been topographically determined).
The broad regional overviews of The Fields of Britannia are supported by detailed local studies which suggest that, in many areas, the fifth century saw little change to the general character of land-use (e.g. the Upper and Middle Thames Valley: Booth et al. 2007) although, as with all of the ‘big data’ projects that have sought to establish broad national and regional trends, there is now a need to establish patterns at the more local level of individual pays.
The nature of Anglo-Saxon settlements dating to the fifth and sixth centuries has been explored by Hamerow (2010; 2012) and Loveluck (2013). A distinctive feature of these sites is that they were predominantly isolated farmsteads and small hamlets that lacked a regular layout, boundary enclosures, or internal divisions (that are archaeologically visible). This changed from around the seventh century when an increasingly stratified settlement pattern started to emerge alongside the demarcation of space and regularity in layout. In both phases, however, security does not appear to have been a major concern in eastern areas – as settlement were undefended – in contrast to some western and northern areas.
The extent of the Anglo-Saxon migrations, and of the survival of a Romano-British population, has long been discussed. Until recently the identity of those buried in ‘Early Saxon’ cemeteries was the subject of speculation, such as several female graves in the Mucking cemeteries that would appear to have been native Britons based on the jewellery they were wearing (Hirst and Clark 2009). Advances in archaeological science, however, are starting to provide more definitive answers such as at Oakington, in Cambridgeshire, where the analysis of ancient DNA has established that both immigrants and natives were present (Pitts 2016).
The prevailing view remains that settlement patterns at the start of the early medieval period were dispersed, and that the Central Province of England then saw a process of nucleation. This model is most closely associated with Northamptonshire where a series of large-scale fieldwalking surveys and excavations have occurred (e.g. Raunds: Audouy and Chapman 2009). The publication of the Atlases for Rockingham Forest (Foard et al. 2009) and the rest of Northamptonshire (Partida et al. 2013) is a particularly remarkable achievement since, together, they reconstruct the medieval landscape across an entire county (and see Hall (2014) for a general overview of the evidence for open fields).
Interpreting the evidence for village origins in Northamptonshire and across the Central Province more widely has continued to generate much debate over both its chronology and causes. That there was a major discontinuity in settlement around the ‘long eighth century’ is still borne out by examples of ‘Early Saxon’ or ‘Early to Middle Saxon’ settlements that were abandoned around that time. There is also growing evidence for seventh- to eighth-century occupation beneath later villages (e.g. Cambridgeshire: Cuttler et al. 2011; Gloucestershire: Stratford 2012; Stansbie et al. 2013). While some archaeologists have suggested a relatively early date for protracted changes which eventually resulted in the nucleation of settlement (e.g. Wright 2015), other scholars have argued that the entire process started later (e.g. Williamson et al. 2013).
There also remains a debate between those who see social agency as driving landscape change and others who argue for a greater role having been played by the natural environment (e.g. Williamson 2013). Among those who favour social agency, the debate continues between those who argue for a ‘top-down’ process driven by the landed elite, notably the Church (e.g. Blair 2013) or ‘bottom-up’ with a greater input from wider communities (e.g. Loveluck 2013; Williamson 2013). There has also been a revival of discussion into whether villages were planned – for example the discussion of Raunds (Audouy and Chapman 2009) and West Cotton (Chapman 2010) – or developed organically (e.g. Williamson’s (2013) suggestion that the regularity evident in some village plans results from them having spread over the strips and furlongs of earlier open fields). Blair (2013) has suggested that from the seventh to the eleventh centuries a range of individual buildings and entire settlements across much of eastern England were laid out with the aid of carefully surveyed and geometrically precise grids, often employing a short perch (about 4.6 m) in four-perch square modules.
This was also the period when we start to see significant social stratification in the countryside, with estate centres for the collection and processing of agricultural surpluses starting to appear (e.g. Higham Ferrers in Northamptonshire: Hardy et al. 2007; Lyminge in Kent: Thomas et al. 2016). The development of estate centres is not restricted to England’s Central Province, with a potential example in the West Midlands being Rotherwas Industrial Estate near Hereford (Miller 2011). Other investment in infrastructure during this ‘long eighth century’ included causeways across wetlands (e.g. Street in Somerset: Brunning 2010; 2014), intertidal fish traps (e.g. the Thames in London: Cowie and Blackmore 2008, 115–124), and watermills (e.g. Ebbsfleet in Kent: Andrews et al. 2011).
While the debate about the origins of villages and open fields in England’s Central Province is important, large parts of south-eastern, western, and northern Britain did not undergo this transformation of the countryside. Instead, settlement patterns remained predominantly dispersed, and field systems mostly enclosed, throughout the early medieval period.
The reason why the landscape of medieval Britain witnessed the development of these marked regional variations in character has seen much discussion, within a broader debate over the drivers of change within society (summarised in Lambourne 2010). There has been a trend toward seeing peasants and lords as both having a role as the instigators of landscape and settlement change (e.g. Stocker and Everson 2012; Hamerow 2012; Wrathmell 2012; Lovelock 2013; Dyer 2015), while Williamson (2013, 243) also champions ‘the importance of the productive base rather than the extractive superstructure’, arguing that landscapes developed gradually through the interaction of local communities with their environments. The role of the natural environment – notably topography, soils, and climate – in shaping human behaviour and landscape character is a topic that most archaeologists have generally avoided for many years, in fear of being labelled an ‘environmental determinist’, but various studies remind us that geology, topography, soils, and climate were important factors in shaping the choices that human communities made (e.g. Rippon et al. 2014; Williamson 2013).
A major issue is that relatively few early medieval sites have been excavated in the largely aceramic western and northern regions, although radiocarbon dating is starting to make a difference (e.g. Dale View Quarry in Stanton- in-Peak, Derbyshire: Brightman and Waddington 2010; Ingleborough in North Yorkshire: Johnson 2012; 2015a; 2015b).
An important part of landscape character is the language used for its field- and place-names, and while the study of these has traditionally been a discrete discipline, a number of interdisciplinary studies have started to explore them in their wider landscape context which is to be welcomed (e.g. Cullen et al. 2011; Higham and Ryan 2011).
Major advances have been made in our understanding of agriculture through the increased amount of palaeoenvironmental data coming from large-scale developer-funded excavations. While the acidic soils in many western and northern areas will always mean that there is a dearth of faunal material, important charred cereal assemblages are being recovered in these areas (e.g. Parc Bryn Cegin, near Bangor in Gwynedd: Kenny 2008; South Hook in Herbranston, Pembrokeshire: Crane and Murphy 2010).
Several important overviews of early medieval farming have been published including Banham and Faith’s (2014) Anglo-Saxon Farms and Farming and Holmes’ (2014) Animals in Saxon and Scandinavian England. The picture that is emerging across all regions is one of non-specialised mixed farming strategies intended to supply mainly local needs in the fifth and sixth centuries, with cropping regimes based on a variety of cereals, cattle and sheep/goat dominating the animal bone assemblages, and relatively little evidence for hunting (Crabtree 2010).
From the Middle Saxon period there is emerging evidence for some specialisation within animal husbandry strategies – such as wool or meat production – alongside increasing specialisation in particular crops that in part may result from the increasing cultivation of a wider range of soils (e.g. Banham 2010; Crabtree 2010; 2012; McKerracher 2016). This shift away from economic self-sufficiency to the production of surpluses is also reflected in the emergence of elite sites with evidence for large-scale crop and meat processing. McKerracher (2014, 82) has shown how the Middle Saxon period was marked by the re-emergence of large- scale grain ovens, the ‘conspicuous monumentality’ of which ‘suggests large-scale and/or centralized crop production and processing at these sites’ (e.g. at Higham Ferrers, Feltham, and Gillingham). There is a high degree of conformity in their construction and their form mirrors Late Roman examples, perhaps reflecting the ideology underpinning the emergence of new forms of lordship. The re-introduction of the heavy plough – at least on some high-status sites with contacts to mainland Europe – can also now be dated to this period based on the discovery of an early seventh-century coulter at Lyminge in Kent (Thomas et al. 2016, 752). The wider context of this intensification appears to have been the progressive fragmentation of large folk-based territories.
The shift from field cemeteries to graveyards associated with churches – an association that would form the central focus of later medieval communities in landscapes within both nucleated and dispersed settlement patterns – remains a profoundly important phase in the development of our modern countryside. Several studies have examined the relationship between Early Anglo-Saxon cemeteries and settlements (e.g. Hamerow 2012, 121–136; Sofield 2015) and Middle to Late Saxon burials (e.g. Buckberry and Cherryson 2010). The opportunities to carry out large-scale excavations of medieval parish churches and cemeteries in order to explore their origins are few (although we await the results of HS2 that will cut a transect across central England), but there has been important work on sites where early medieval churches and cemeteries fell out of use (e.g. Sedgeford in Norfolk: SHARP 2014). A particularly significant Middle Saxon cemetery has been excavated at Great Ryburgh, also in Norfolk, where the site’s waterlogged conditions preserved the remains of 81 log coffins and 6 plank-lined graves as well as a small timber structure, measuring about 4m wide and 6m long, that was aligned east-west and had a southern entrance: although presently undated, it seems likely that it is a church or chapel contemporary with the cemetery (Current Archaeology 2016).
The evidence for the early medieval Church in western and northern Britain was reviewed by various papers in Edward’s (2009) The Archaeology of the Early Medieval Celtic Churches, and excavations are also shedding important light on the development of individual sites (e.g. Brownslade Barrow and West Angle Bay in Pembrokeshire: Groom et al. 2011).
The later medieval period in England traditionally begins with the Norman Conquest whose immediate consequences were felt most acutely in specific places such as major towns and some rural areas in northern England (e.g. Robert 2008). The Norman Conquest also had a longer-term impact through the acceleration of social and economic trends that had started in the tenth century such as the intensification of agriculture, the expansion of settlement into physically more marginal environments, the growth of towns, foundation of new monasteries, and the increasing wealth of the Church as a whole. The conquest of Wales was a particularly long drawn-out process.
Although Creighton and Wright (2016) have recently reminded us of the potentially significant impact of short- term events – such as the ‘Anarchy’ under King Stephen – the eleventh to thirteenth centuries were largely ones of stability and economic expansion against a backdrop of rapidly rising population (Dyer and Schofield 2007; Hinton 2013). This trend was reversed in the first half of the fourteenth century by the dramatic and devastating Black Death and other episodes of human and animal disease that preceded it (Dyer 2010). Nevertheless, the late medieval period as a whole was one that made a significant contribution to many character-defining features within our present countryside including the initial enclosure of open fields, the construction of a significant number of domestic houses, and the rebuilding of many parish churches.
While settlement nucleation appears to have already begun in some regions during the ‘long eighth century’, the process continued into the post-Conquest period when it becomes better documented. In areas such as the East Midlands and East Anglia where nucleation appears to have started relatively early, these villages may not have assumed their present form until the post-Conquest period, while in other regions the process of nucleation itself did not start until the later eleventh to thirteenth centuries (e.g. Williamson 2013).
These centuries were ones of growth in population and farming, which is reflected in various documentary sources (e.g. Broadberry et al. 2015). Fieldwork and documentary research have continued to provide examples of how settlement expanded into environmentally marginal landscapes both in uplands (e.g. Alnham Moor in Northumberland: Dixon 2014; Westmondhalgh in the Pennine foothills of south-west Yorkshire: Hay 2014; Camp Shiel Burn in Peeblesshire, Liddesdale, and Samsonshelis in Channelkirk, all in Scotland: Oram 2011, 258; Dixon 2009; Durham 2009), and in lowland areas (e.g. Poolgate Green, in Sussex: Stevens 2007). There was also extensive wetland reclamation in coastal and estuarine areas (e.g. Rippon 2008; Gardiner 2009; Oram 2011).
The expansion of settlement into new areas was not, however, the only reaction to rising population and food prices, as there is also growing evidence for how farmers tried to improve productivity of their existing land, including the greater use of manure (Jones 2012). The eleventh- to twelfth-century lime kiln at Crummack Dale in North Yorkshire probably reflects attempts to improve the quality of the acidic upland soils and make them suitable for arable cultivation (Johnson 2015b).
In Scotland it has proved difficult to identify medieval rural settlements, although large turf-walled long houses have now been recognised in highland Perthshire (Carver et al. 2013; and see Dodgshon (2015) for a recent general overview of research into the Scottish landscape). In lowland areas there were significant variations in the degree of nucleation, with ‘champion’ countryside of villages and open fields developing in some areas (Dixon and Fraser 2007; Dalglish 2012; Hindmarsh and Oram 2013). The extent of regional variation in medieval settlement in Wales is yet to be explored, although some areas did see the development of villages and open fields following the Anglo-Norman Conquest, including in Pembrokeshire where such developments were associated with Flemish colonists (Rippon 2008). Scotland also experienced Flemish settlement during a period of wider immigration in the twelfth and thirteenth centuries in areas such as south Lanarkshire and Aberdeenshire (Oram 2011).
Archaeology is increasingly contributing to our understanding of how farming communities adapted to the changing economic circumstances of the late medieval period. The dominant trend was the increasing diversity in local agrarian strategies. At Dudley Castle, in Worcestershire, for example, the management of pigs changed from free-range grazing of large herds in areas of woodland and wood-pasture in the eleventh to thirteenth centuries (pannage) to the more intensive management of fewer pigs in enclosed land (which is reflected in the increased size of the animals and isotopic evidence for their having had a more varied diet including legumes: Hamilton and Thomas 2012).
There has been a long-term trend away from fieldwork and excavation on deserted medieval villages towards a far wider range of settlement types (reflected in the papers in Dyer and Jones’ (2010) Deserted Villages Revisited). Development-led archaeology in particular has led to relatively little new work being carried out on wholly deserted medieval villages (although see Goldicote in Warwickshire: Thomson and Palmer 2012), but it is shedding important light on two broad areas: first, areas with more dispersed settlement patterns, and second, villages that shrank during the late medieval period but which are now seeing renewed expansion.
Of the former group some of the most revealing work has been carried out in advance of major linear developments providing transects through areas of landscape that have seen little previous work such as the A120 across the claylands of central Essex (Timby et al. 2007a), and the M1 across the clay-with-flints that cover the Chiltern dipslope in Hertfordshire (Stansbie et al. 2012). In both cases excavation revealed isolated farmsteads which were characteristic of the region as a whole. In contrast, major road schemes through landscapes that had been characterised by villages – and that are themselves avoided by the new highways – found very little evidence for later medieval settlement (e.g. along the A1 in North Yorkshire; Brown et al. 2007), although an isolated smithy was found during the construction of the Great Barford Bypass on the Bedfordshire claylands (Timby et al. 2007b).
Where settlements were abandoned, we continue to see that this was a long drawn-out process with a wide variety of causes. Some farmsteads appear to have been abandoned in the late thirteenth century, perhaps due to tenurial changes such as the amalgamation of tenements (e.g. Allcourt Farm in Lechlade, Gloucestershire: Stansbie et al. 2013; M1 Junction 8 in Hertfordshire: Stansbie et al. 2012). In some cases a simple retreat from environmentally challenging environments – that had only recently been occupied due to high population pressure during the twelfth and thirteenth centuries – occurred in the fourteenth century (e.g. Stebbingford Farm in Essex: Timby et al. 2007a), while other settlements survived only to succumb to social and economic changes in the fifteenth century or later (e.g. Alnhamsheles in Northumberland: Dixon 2014).
Understanding the relationships between different groups within society – most notably lords and peasants – is just as important in later medieval landscapes as it is in the early medieval period. Smith’s (2009b) study of Wharram Percy, for example, reminds us that lords were often not resident within their manors and this increased the agency of the peasantry, although they could still stamp their authority upon a landscape (e.g. the construction of the North Manor). This is also a period for which there is rich documentary sources for some places, and an example of their value in understanding the dynamics of the medieval countryside is Kilby’s (2015) exploration of peasant attitudes through an analysis of incidences of trespass on demesne and peasant land in the Suffolk vill of Walsham-le-Willows.
A theme within the study of domestic buildings has been continuity, reflected in Blair’s (2015) review of high-status houses in England from the tenth to mid- eleventh centuries. The continued significance of dendrochronology is reflected in work in Wales (Silvester and Kissock 2012), while Alcock and Miles’s (2013) The Medieval Peasant House in Midland England uses it as the starting-point for a far wider analysis of these structures including construction techniques and social contexts.
There have also been important advances in our understanding of lower-status domestic houses. Gardiner (2014) has reviewed the archaeological evidence, showing that most buildings constructed between 1000 and 1200 were of earth-fast construction with timbers placed into the ground, whereas from around the thirteenth century there was a shift to ground-set timber construction.
Standing buildings start to survive in significant numbers from the late medieval period. Matthew Johnson’s (2010) English Houses 1300–1800: Vernacular Architecture, Social Life has shown how buildings were not only the setting for daily life but also played an active part in creating social structures. Dyer (2013) has integrated archaeological and documentary evidence to explore how buildings were used, and what this tells us about the way of life and outlook of the peasantry across the Midlands and northern England, while Richard Suggett (2013) has reviewed how domestic architecture was used to express peasant identities in Wales.
For too long the study of vernacular architecture was separate from research into the wider medieval landscape, and Dyer’s (2006) call for greater integration has been addressed in a series of studies at different scales. Rippon (2012), for example, used data from standing buildings on and around the Blackdown Hills in eastern Devon and southern Somerset to confirm that regional variations in settlement patterns depicted on nineteenth-century maps were broadly the same in the late medieval period, and that the ways in which new architectural styles were adopted reflects other long standing differences in landscape character. At a more local scale, Mileson (2015) explored the positioning of houses within village plots in Ewelme hundred, south Oxfordshire, which was used to signal important differences in social standing and relationships between households.
In Scotland, it has proved difficult to distinguish medieval buildings from post-medieval ones due to a lack of distinctive plan-forms and the use of perishable building materials such as turf and timber. This gap in the evidence has been alleviated most successfully in the Western and Northern Isles (e.g. Turner et al. 2013) and in lowland areas of eastern Scotland (e.g. Hindmarsh and Oram 2013). Despite the concerted efforts of a large-scale multi-disciplinary project at Ben Lawers in the central Highlands, our knowledge of medieval settlement in the Highlands remains patchy, and largely dependent on extrapolating from medieval documents and post-medieval remains.
Important work has also been carried out on farm buildings. Dendrochronology at Charlton Court Barn, at Steyning, in Sussex, for example, shows that it was erected from trees felled a few years after the manor passed from the crown into private hands in 1403 (Aldsworth 2007).
The ways in which the aristocracy used the landscape to make and sustain notions of lordship continue to attract attention (e.g. Hansson 2009), although whether the term ‘designed landscapes’ is appropriate for the medieval period has also been questioned (Creighton 2009; cf. Liddiard and Williamson 2008). The debate over the construction of moats around some high-status farmsteads has also continued with a combination of functional and social factors being posited (Creighton 2009; Platt 2010; Creighton and Barry 2012; Johnson 2015). Elite landscapes, such as parks and forests (e.g. Liddiard 2007; Mileson 2009), were also associated with specialised settlements related to particular aspects of countryside management such as lodges (e.g. Simonsbath on Exmoor: Hegarty and Wilson-North 2014; Colwyn Castle, Powys: Silvester 2010; Buzzart Dykes Park in Perthshire: Hall 2013).
The majority of the population in medieval Britain lived in farming communities where there was a mixture of arable and pastoral land-uses, although some landscapes were characterised by more specialist forms of subsistence. Pastoralism is mostly often associated with uplands, and the MSRG volume on Life in Medieval Landscapes (Turner and Silvester 2012) – a tribute to the greatly missed Harold Fox – contains a series of papers on wood-pasture, upland farms on Dartmoor, transhumance in Cornwall, and seasonal settlements in northern England (and see Fox 2012). The seventh Ruralia conference – based in Cardiff in 2007 – explored Medieval Rural Settlement in Marginal Landscapes (Klápště and Sommer 2009) and the resulting volume contains papers on Scotland, the Yorkshire Dales, and Cornwall that also reflect the traditional focus on upland areas. Recently published surveys and excavations include Upper Pasture in Horton-in-Ribblesdale, North Yorkshire (Johnson 2012), highland Perthshire (Carver et al. 2013), and Hen Caerwys in Flintshire (Davies and Silvester 2015). Crouch and McDonagh (2016) remind us of how highly valued areas of lowland pasture were, using Wallingfen in Yorkshire as a case study.
Wetlands were a particularly distinctive type of landscape, but rather than being perceived as ‘marginal’ – because they required considerable investment in drainage and flood defence before they could be farmed – medieval communities clearly saw them as important landscapes (Rippon 2009). At the start of the early medieval period most wetlands were simply exploited for their rich natural resources such as fish, salt, and seasonal grazing (e.g. Fishtoft in Lincolnshire: Cope-Faulkner 2012), although from around the tenth century settlements started to become permanent as embankments were constructed to keep the sea at bay (e.g. eastern England: Gardiner 2007). Wetlands remained, however, prone to flooding, and phases of expansion could be interspersed with periods of inundation and retreat (e.g. Titchwell and Thornham, in Norfolk, and Broomhill, in East Sussex: Gardiner and Hartwell 2006).
In their natural state rivers, estuaries, and their associated wetlands might be a barrier to communication, although this could change (Rippon 2007). Watercourses and river valleys more broadly became important communication arteries, and John Blair’s (2007) Waterways and Canal Building in Medieval England brings together a series of papers that reveal the extent to which rivers were canalised, and wholly artificial canals were constructed, in the medieval period. Settlements on reclaimed wetlands relied on agriculture as their primary means of subsistence, although in coastal areas specialised fishing villages started to develop during the late medieval period (Fox 2007).
Agriculture was, of course, the primary means of subsistence for most rural communities, although evidence continues to accumulate for the significance of extractive and manufacturing industry in some areas. These industries are relatively well known in the later medieval period, but their extent during the pre-Conquest period is starting to become apparent. Evidence for iron smelting during the early medieval period in Northamptonshire’s Rockingham Forest and
Gloucestershire’s Forest of Dean (Pine et al. 2009) should come as no surprise as these were the centres of important iron industries in both the Roman and later medieval periods (e.g. Brady et al. 2015), but less well known are the more scattered, small-scale, local industries (e.g. Erglodd, Llangynfelyn, Ceredigion: Page et al. 2012; Mytton Oak Road near Shewsbury, Shropshire: Bradley 2016). Other industries have seen important work such as the royal silver mines in the Tamar Valley on the Cornwall-Devon border, where Bere Ferrers may have been a particularly early specialist mining town (Rippon et al. 2009).
Radiocarbon dating is of particular importance on iron production sites since other artefactual material can be scarce – particularly in the early medieval period – but even pottery production sites can benefit from scientific dates: a Stamford Ware kiln in Pontefract, West Yorkshire, for example, was producing wares dated typologically to the late eleventh or twelfth centuries, but archaeomagnetic and AMS radiocarbon dating suggest an early eleventh-century date for the last firing of the kiln (Cumberpatch et al. 2013).
Pottery production occurred in many areas, and archaeological investigations have been carried out in various locations from Sussex on the south coast (e.g. Chichester and Ringmer: Sulikowska 2014; Wallis 2013) up to Cumwhinton in Cumbria (Railton et al. 2014). Textile production was another important rural industry, and a particularly early example of mechanisation has been recorded at Barrowburn in Northumberland where a fulling mill dating to the later twelfth and thirteenth centuries appears to have been built by the monks of Newminster Abbey: the mill had a low breast-shot wheel
– a more efficient configuration than an undershot wheel
– which only became common in England during the eighteenth century (Carlton and Jones 2014; 2015).
Interpreting material culture is central to the dating and understanding of medieval settlements both in individual site reports and general syntheses (e.g. Smith 2009a; Hinton 2010; Hyer and Owen-Crocker 2011; 2015; Gilchrist 2012; Stanley 2013; Jervis et al. 2016). There is still much to be gained through the writing up of assemblages collected in the past (e.g. Glastonbury Abbey, in Somerset: Gilchrist and Green 2015; Meols, in Cheshire: Griffiths et al. 2007), in particular where this is combined with applying modern scientific analysis (e.g. the use of Inductively Couple Plasma Spectrometry to establish the production centres for imported tin- glazed pottery at Glastonbury Abbey: Allan et al. 2015).
Pottery is the most common artefact type, and important studies have been published of individual industries (e.g. Ipswich Ware: Blinkhorn 2012) and particular regions (e.g. Cambridgeshire: Spoerry 2016). Key advances in theoretical approaches towards the analysis and understanding of ceramics, such as Ben Jervis’ (2014) Pottery and Social Life in Medieval England that uses relational archaeology to focus on ‘the processes through which humans and non-human form relationships and the effect of these relationships on the emergence of social contexts, identities or objects’ (Jervis 2014, 17).
Particularly significant are moves towards a well- integrated approach towards studying material culture alongside documentary sources such as Jervis et al.’s (2015) analysis of the material culture of rural households in the south and east Midlands, and Dyer’s (2012) study of Wharram Percy that also embraces palaeoeconomic data. Dyer’s (2014) wide-ranging paper on the ‘the material world of English peasants’ shows them to have behaved rationally and been resourceful and enterprising, and while some peasants were impoverished others were improving their living conditions.
This review period began with the MSRG’s Perceptions of Medieval Landscapes and Settlements (POMLAS) seminars, whose focus was on the people who organized and lived in the medieval landscape, with the aim of understanding their experiences, their attachments and loyalties to places and regions, and the context of their activities and decision making within the landscape (Dyer 2007, 6). These themes have since been explored in several broad interdisciplinary studies, for example Jones and Semple’s (2012) Sense of Place in Anglo-Saxon England and Whyte’s (2009) Inhabiting the Landscape: Place, Custom, and Memory, as well as through case- studies of specific types of site (e.g. Johnson (2015) on moated sites in Kent). There have also been advances in the study of material culture and how it was used to communicate information, such as identities (e.g. Smith 2009a), and a growing awareness of the significance of ritual activity within medieval settlements (e.g. Morris and Jervis 2011; Knox 2016).
With field-based investigation of medieval settlements now dominated by developer-funded programmes of work, the approaches being adopted are increasingly formulaic: a desk-based assessment of existing databases, aerial photographs, and easily accessible documentary/cartographic sources, a limited range of surveys (notable geophysical), and trial trenching, sometimes followed up by open-area excavation or ‘strip, map and record’. Test-pitting within currently occupied settlements also now has a proven track record. There is, however, a danger within British archaeology that comprehensive programmes of survey and evaluation will only be carried out where there is a perception that it can be justified, running the risk that fieldwork will increasingly become concentrated in areas where archaeology is already known and/or easily visible. One type of landscape in which archaeology appears to be readily visible are the types of light soil that produce good cropmarks, while the dominant form of survey that is now carried out in advance of development – geophysics (usually magnetometry) – will also pick up large buried features. There is, however, a danger in the over reliance on aerial photography and geophysics as some sites are not associated with major below-ground features that will show up using those techniques, which is particularly a problem for the early medieval period when most settlements were not associated with ditched property boundaries or the use of stone for buildings; the problem is even greater from the eighth century when Grubenhäuser go out of use (e.g. Hamerow 2012). The decline in fieldwalking as a survey technique is very marked, and we must wonder how many early medieval sites are being missed because of it.
The past decade has seen the increased use of LiDAR data as this has become more freely available. Such imagery has huge potential for mapping earthworks where air photographic coverage is poor (e.g. the Forest of Dean: Gloucestershire County Council 2012), although it must be remembered that large areas of medieval earthworks have been destroyed through ploughing (notably of ridge and furrow) and urban development (affecting shrunken settlement earthworks but also ridge and furrow) since the 1950s and as such early air photographs – especially those taken by the RAF in the late 1940s – still provide important evidence. In addition to high-tech methods of recording there is still an important place for traditional earthwork surveys, shown, for example, in Creighton and Wright’s (2016) study of Anarchy-period castles and related settlements.
The development of the Portable Antiquities Scheme database (PAS) and Corpus of Early Medieval Coin Finds continues to go from strength to strength. By the end of 2015 the PAS held 27,731 records (containing 39,637 finds) of early medieval date (c. 410–1066) and 161,405 records (192,643 finds) of later medieval date (c. 1066– c. 1500: see Naylor 2016). This provides researchers with a huge dataset, for example in improving our understanding of ‘productive sites’ (e.g. Richards et al. 2009). The later medieval period has the second largest dataset after the Roman period but has been somewhat neglected (but see Oksanen and Lewis 2015 for a distribution of these finds, and an interim report on their ‘Placing Medieval Markets in their Landscape Context through the Portable Antiquities Scheme Data’ project). PAS data does, however, need very careful interpretation both spatially (e.g. Bevan’s (2012) study of Anglo- Norman coin finds) and in terms of the types of artefacts being reported (e.g. certain types of Early Anglo-Saxon brooches are greatly under-represented: McLean and Richardson 2010).
A major development in recent years has been the growing sophistication of scientific analyses that have the potential to transform our understanding of the past. The highest profile work is being done within the field of migration studies, and early results from the application of new techniques suggest that we may be on the brink of being able to answer long-standing questions over the extent of Anglo-Saxon, Viking, and Norman immigration into Britain. One approach is the study of DNA (e.g. Pitts 2016; Schiffels et al. 2016), while another is the analysis of stable isotopes that are absorbed by humans and animals from the food that they eat, with the chemistry of the plants reflecting the geology on which they grew. Hughes et al. (2014), for example, studied strontium and oxygen isotopes from nineteen human skeletons in what has traditionally been regarded as the ‘Early Saxon’ cemetery at Berinsfield, in Oxfordshire. This showed that the majority of the deceased had been born and brought up in that locality, with only a small number of immigrants from elsewhere in Britain and just one individual who may have come from mainland Europe. In contrast, at Bronslade and West Angle Bay in Pembrokshire, isotopic analysis of early medieval burials indicates a highly mobile population with the possibility that some individuals originated either from the extreme west coast of Ireland or north-west Scotland but others possibly from the Mediterranean region (Groom et al. 2011).
Isotopic analysis can also be used to study other aspects of the life cycle such as human diet. Mays and Beavan’s (2012) study of human remains from twelve fifth- to seventh-century sites showed that these communities made relatively little use of wild resources within what were predominantly agrarian economies. Lamb et al. (2012) found a very similar picture at the remote later medieval Scottish coastal village of Auldhame where the largely static population had surprisingly little marine protein in their diet and relied instead on land- based agriculture for their subsistence. Studies of human remains have also provided a better understanding of when weaning occurred (Mays et al. 2002; Privat et al. 2002; Haydock et al. 2013).
Isotopic analysis is starting to shed new light on animal husbandry practices, such as Evans et al.’s (2007) study of strontium isotopes within cattle teeth at sixth- to seventh-century Empingham and tenth- to twelfth-century Ketton, in Rutland, which suggested that the areas across which animals consumed food had expanded over time (see also Hamilton and Thomas’ 2012 study of pig management noted above).
The archaeological study of medieval settlements and their associated landscapes has become increasingly dominated by development-led fieldwork which brings both advantages, such as the increasing volume of data and new work being carried out in areas that have previously been neglected, but also disadvantages, such as a bias in where fieldwork is now occurring towards a relatively small number of development ‘hotspots’. With so much development-led archaeology now being carried out in a restricted number of lowland areas, there is a need for the very limited research funding available to be targeted in areas that will otherwise continue to be neglected. While all regions have important research questions to answer, of the three nations in Britain it is Scotland and Wales that are most in need of more research, and within England it is regions outside the Central Province. The decline in fieldwalking within development-led archaeology is very noticeable, and this is one area in which research projects – particularly those involving the community – could make an invaluable contribution.
Fieldwork only improves our understanding of the past if post-excavation analysis is carried out, and full reports made easily accessible. While some backlog sites have now been published, and large amounts of grey literature is archived with the Archaeology Data Service, there is still a problem of some important sites remaining unpublished and some grey literature reports not being made publicly accessible (including all the specialist reports and appendices). While Blair’s (2018) Building Anglo-Saxon England has just been published – albeit outside this review period – we need a major synthesis for the later medieval period. The timeframe of this does, however, need careful consideration. The medieval period was one that saw major episodes of change – sometimes sudden – but also centuries of relative stability and we must avoid creating false divisions within our datasets. From a landscape perspective, across most regions, 1066 is not a good point at which to divide the study of the medieval landscape in England, let alone Scotland and Wales.
We are increasingly aware of regional variation within the medieval countryside which is a major theme in all of the ‘big data’ projects that explored the landscapes of Roman and early medieval England and Wales. There is, however, a need to go further and try to establish trends at a more local – pays – level. This needs to include more reconstructions of the physical fabric of the medieval countryside (following on from the Atlas of Northamptonshire), and a greater understanding of how regional variations in landscape character came about. It is also important to see trends in Britain within their wider European context, something that British medieval archaeology has not been very good at.
A long-established aspect of medieval settlement research that makes it so exciting is the great diversity of source material, and accordingly stronger interdisciplinary communication with historians, historical geographers, vernacular building specialists, and place-name experts must be an aspiration for the future. There is also a growing body of scientific research being carried out now, and this is to be welcomed. We are starting to see the potential of studies of ancient DNA to transform our understanding of migrations, and the movement of people on a smaller scale through trade and other social interactions. We also need more high-resolution pollen analysis and even greater use of radiocarbon dating as it is starting to show its value in identifying aceramic early medieval sites in particular.
Overall, this review period 2007–16 has been dominated by a dramatic increase in the amount of data, and the challenges this presents. Some progress has been made towards synthesising this material – particularly for the early medieval period – but the key challenges for the future must be to ensure that fieldwork continues to be published, and for a major overview for the later medieval period. Such is the increase in data and new techniques that we now have at our disposal, however, the future is bright.
For their help in writing this paper we thank Neil Christie (University of Leicester), Oliver Creighton (University of Exeter), Chris Dyer (MSRG Vice President), Ben Jervis (University of Cardiff), Michael Lewis (PAS), Gundula Muldner (University of Reading), Andy Seaman (Canterbury Christ Church University), Naomi Sykes (University of Exeter), Stuart Wrathmell (MSRG President), and Duncan Wright (Bishop Grosseteste University).
Part of the Research Frameworks Network