A reappraisal of Eleventh-Century Settlement in the Eastern High Weald

A REAPPRAISAL OF ELEVENTH-CENTURY SETTLEMENT IN THE EASTERN HIGH WEALD BRENDAN CHESTER-KADWELL The High Weald is usually considered to be an area of poor soils and relatively late settlement, overshadowed by the more prosperously settled coastal areas to the north and south. Many believe that the High Weald was in times past a marginal land, difficult to subdue, which even though permanent settlement had already been established 'was in the late llth century grossly under-exploited' (Brandon 2003, 52). The evidence appears to support a marked distinction in wealth and density of population between the settlement of the coastal fringes of Kent and Sussex and the High Weald. However, does this mean that High Weald settlement was particularly sparse? A reappraisal of the evidence questions whether it was especially sparse, particularly when compared to other parts of the country: notably with places that were not exceptionally prosperous, but not marginal either. lbis article sets out to re-examine the evidence for the extent and density of eleventhcentury settlement in selected hundreds and parishes of the eastern High Weald as represented in Domesday and other contemporary sources. It explores how much of the eventual settlement pattern existed and how populous the area had become. The lack of archaeological evidence for early medieval settlement means that the documentary sources are of particular importance: 1 these sources are investigated so as to estimate the nature of settlement in the eleventh century, either directly, or by implication. The existence of a church, for example, can be an indicator of associated permanent settlement, and the evaluation ofthe different sources for Kent and Sussex help to understand how well this part of the High Weald was settled. Settlement is defined by its attributes; for example, settlement pattern describes how settlement is distributed within the landscape, whether it is dispersed or nucleated. Settlement form describes its structure, or morphology - especially how different elements relate to each other spatially. Rural settlement is taken to mean the combination of its essential parts, its built environment (habitation elements) as well as the spatial resources (fields and woods etc) that makes it economically and socially viable. How these elements have developed over time is of particular interest to settlement studies, especially in relation to those settlements that have persisted to this day. Eleventh-century settlement in lowland England was sparse in comparison to later times: even the population of its greatest cities were smaller than many country towns today.2 The issue about the state of development of settlement is, therefore, one of comparison and informed judgement between areas at any one time: would 105 BRENDAN CHESTER-KADWELL Eastern High Weald 40 -=-=--==--Kllomelers 0 5 10 20 Location of the Study Ar.eas N C> Crown copyright 2009 An Ordnance Surwy/EOINAl[lglmap oupplled service Fig. 1 Location of the Study areas. © Crown copyright 2009. An Ordnance Survey/ EDINA/Digimap supplied service. 106 REAPPRAISAL OF ELEVENIH-CENTIJRY SETTLEMENT IN THE EASTERN HIGH WEALD the eye of a travelling man, experiencing the agricultural realm of England in the eleventh century, have seen the Eastern High Weald as an area of settlement more sparsely developed than elsewhere? l11e article, therefore, attempts (maybe for the first time for the High Weald) to compare the level of resources recorded in Domesday with another comparable area outside the South-East, on this occasion the townships in the Huntingdonshire Ouse Valley. Thereby, it is hoped to throw light on how 'marginal' High Weald settlement actually was. However, the results of this comparison remain tentative, partly because the Wealden data is less abundant than should ideally be the case, but also because a comparison with only one other area (the Huntingdonshire Ouse Valley) is insufficient to be certain about the conclusions. However, the results are encouraging and give new insights to the state of contemporary settlement in this part of the High Weald. Overview of the Study Areas The study areas (see Fig. 1) featured in this article are ones that were used in an e>..1ensive longitudinal research project looking at origins and development of some settlements along the eastern Rother valley of the High Weald and the Huntingdonshire Ouse valley (Chester-Kadwell 2010). The areas were selected partly because of their topographical similarity, but also because of a shared original settlement pattern of wood pasture communities; although by Domesday in Huntingdonshire a system of common agricultural management was practiced, causing the decline of its dispersed settlement pattern. The framework for eleventhcentury settlement in each of the areas is discussed as follows. Eastern High Weald. This was an area of small hundreds and large parishes. The study area featuring in this article looks at selected settlements along the catchment area of the River Rother. It focuses on the Domesday hundreds of Rolvenden and Selbrittenden in Kent (looking at the parishes of Benenden, Rolvenden, and Newenden), and the hundreds of Shoyswell, Henhurst, Hawkesborough and part of Staple in Sussex (looking particularly at the parishes ofEtchingham, Salehurst and Bodiam).3 Although the connection between hundreds and Domesday parochia is not always clear, it is possible to relate eleventh-century churches to the later medieval parochial structure that has come down to the present day in the form of civil parishes. This establishes a basis for a comparison with later settlement structures. Unfortunately, knowledge of the High Weald hundreds in Kent is very limited, so that the data available for an area comparison of the eastern High Weald with the Ouse valley is effectively restricted to the Domesday record for the three adjacent Sussex hundreds of Shoyswell, Henhurst, and Hawkesborough in the Rape of Hastings. The entries for these hundreds contain a fuller record than is usual for the High Weald (although regrettably, an incomplete one) because of an administrative technicality involving the 'Pevensey outliers' (explained below). The Huntingdonshire Ouse Valley. Huntingdonshire was a county oflarge hundreds containing townships that at Domesday were served by well established parochia, many in the process of dividing into the smaller parishes generally established 107 BRENDAN CHESTER-KADWELL by the thirteenth century. The townships in the study area fall mainly within the hundreds of Toseland and Hurstingstone, but excepting one that falls within the hundred ofLeightonstone. l11e Ouse Valley lies on the eastern edge of the Midland open field system and by the eleventh century the settlement pattern was becoming more nucleated and the system of common field management was being extended and developed throughout the region. However, it had previously been an area of distinctive wood pasture (for which the place name and landscape evidence still remains) and this had not been completely extinguished by the eleventh century. At Domesday this part of the Ouse Valley, whilst not being as well populated as parts ofNorfolk (for example), had riparian settlements that were of moderate prosperity, but also regions above the river valleys that would not be fully exploited until the thirteenth century. These factors make it a good comparator for the eastern High Weald. The two areas chosen for the comparison study are both centred on a defining river system, which has created significant alluvial deposits amenable to early settlement (see Figs 2 and 3). The river valleys in both areas contrast with once heavily wooded hills, and both are associated with significant wetlands that give eventual access to the sea (Romney Marsh for the High Weald and the Fens for the Ouse valley). The High Weald is part of the Wealden region geologically, although topographically it forms a distinct sub-region that needs to be considered on its ovvn merits (Thomas 2013, 4). The eastern High Weald is in some respects different to its more western reaches. It was, historically, accessible from the sea with the Rother being navigable as far as Bodiam or Robertsbridge for much of the Middle Ages (Eddison 1985). It was also rather less wooded than many other parts of the region and there is a higher percentage of better soils, especially along the course of the Rother and its tributaries. These factors may also be exl)ected to have affected the manner and development of its settlement. The chosen section of the Ouse Valley is situated in the historic county of Huntingdonshire (now in Cambridgeshire). The River Great Ouse enters Huntingdonshire at St Neots, the site of a Saxon monastic foundation, and flows north and then eastwards for about twenty miles to Earith, a fen edge settlement, before leaving the district. It fonns a distinct sub-region that geographically has characteristics in common with the eastern High Weald. The Settlement Debate Bedevilling the whole issue of the establishment of permanent settlement in the eastern High Weald, and its extent at any particular time, is the lack of clear archaeological evidence for early medieval settlement generally, and including the immediate post-Conquest period. Earlier writers like Witney, Everitt and (to some extent) Brandon have chosen to read this lack of evidence as proof that no early permanent settlement existed and that later settlement was especially sparse. Later writers like Gardiner, Harris, and Thomas are more cautious: they maintain that the lack of evidence is not evidence of a lack of early settlement, as there are good reasons why the evidence may be hard to locate (Gardiner 1990, 33-35). Early Saxon habitation is notoriously difficult to find and easily missed, and there is the potential for earlier settlement evidence to be hidden under, or destroyed by, later 108 REAPPRAISAL OF ELEVENIH-CENTIJRY SETTLEMENT IN THE EASTERN HIGH WEALD 1 N Ticehurst 0 Benenden 0 B=Bodiam E = Etchingham N=Newenden S = Sandhurst Crown copyright 2009 An Ordn.illnc@' Surv@'y/EDINA.IDigm\ap sup pi􀀬 Sl!!NIC@ o -- - - - _____________ _ _ _ 20 kilometres = c. 12 miles LEGEND @) River Rother land liable to inundation ... Hexdon Channel D land above 30 m * Newmill Channel • Kent Ditch Fig. 2 T he Topography of the Eastern High Weald. © Crown copyright 2009. An Ordnance Survey/EDINA/Digimap supplied service. development. Ibis lack of evidence may partly arise from the comparatively little developer-funded archaeology in the High Weald. Field walking and metal detecting, which have been very productive in other parts of the country, are difficult in an area of mainly pastoral fanning (as the High Weald has reverted to in recent years), and the Portable Antiquities Scheme has not yet recorded much in terms of Saxon finds. For whatever reason, archaeological evidence for Saxon settlement in the Weald has been less forthcoming (Gardiner 1990, 47) and the search less systematic with fewer resources applied than elsewhere. Unless or until more archaeological evidence is available the question of settlement continuity and the chronology of settlement formation will always be an area of difficulty. The lack of archaeological evidence means that there has to be a greater reliance 109 l LEGEND BRENDAN CHESTER-KADWELL N 0 -----------------20 __________________ Kilometers = c 12 miles 􀀡 fenland below 7 m D land above 30 m « Fig. 3 The Topography of the Great Ouse Valley study area. Crown copyright 2009. An Ordnance Survey/EDINA/Digimap supplied service. on other kinds of evidence, such as place-name studies, and the landscape itself (including the results of environmental archaeological analysis). However, it is the documentary evidence that, as it becomes more abundant post-Conquest, is of potentially greater significance when making assessments of settlement in the eleventh century. For many historians the High Weald as a whole was an area of Saxon colonisation by piecemeal advance driven by a process of resource exploitation by the peripheral estates. In this view, estates in the coastal fringes of Kent and Sussex sought seasonal pannage for their swine within the forest that led them further and deeper into the uninhabited Weald. Within this process Wealden settlement was typified by impennanent, seasonal settlement in the Early Saxon period, followed by the gradual establishment of pennanent settlement into the Late Saxon period. 110 REAPPRAISAL OF ELEVENIH-CENTIJRY SETTLEMENT IN THE EASTERN HIGH WEALD However, the pattern of pennanent settlement was still incomplete in some areas of the High Weald at the time of Domesday. l11is emphasis on a lack of continuity with pre-Saxon land use in the High Weald, the impennanence of early settlement, together with the importance of transhumance in tenns of economic and social activity, became the dominant model for the origins and development of settlement morphology of the High Weald until very recent times. l11e origins of this traditional view of Wealden settlement can be found in the work of Edward Hasted (1798) and Robert Furley (1871), both historians of the Kent Weald. Hasted, writing in the latter years of the eighteenth century at a time when the role of the old Lathe and Hundred administrative units were still (just) more than a memory, recorded a wealth of contemporary information still relevant to modem scholarship. Furley's History of the Weald of Kent (1871) is very much in this tradition but with a later style of scholarship and also recorded an extensive list ofWealden dens (wood pastures) still considered valuable reference material today. Others followed, but amongst the most influential texts supporting the traditional view has been Kenneth Witney's The Jutish Forest: a study of the Weald of Kent from 450 to 1380 AD (Witney 1976), Alan Everitt's Continuity and Colonization: the Evolution of Ken fish Settlement (1986), and a number of contributions by Peter Brandon (1974; 1978; Brandon & Short 1990), including his recent The Kent and Sussex Weald (2003). Witney's research was, by his own admission, strongly influenced by Jolliffe's Pre-Feud al England: the Jutes ( 1933) and Du Boulay's The Lordship of Canterbury (1966). He relied upon a wide range of evidence, including place name evidence such as that supplied by Wallenberg in The Place Names of Kent ( 1934), the large number of early Saxon charters, Domesday Book, as well as later historical sources including manorial records. Witney's great contribution to our knowledge ofWealden settlement was his detailed analysis of the development of the Wealden dens and the drove ways linking them to the northern manors. Comparable research conducted in the Sussex High Weald, looking at the relationship between the manors on the Sussex coastal plain and their outliers in the High Weald also shares many ofWitney's conclusions. Everitt believed that in Kent there was a degree of continuity between Roman and Saxon settlement in the coastal fringes, but did not find evidence for continuity in the Weald - although some of Everitt's arguments have been called into question by more recent authorities (Thomas 2013, 2-3). During the last fifteen to twenty years the development of improved techniques and understanding of the strength and weaknesses of different evidential resources has led to a re-evaluation of the evidence itself. This, more than the turning up of completely new sources of evidence, has led to a shift in how early settlement patterns are understood. For example, Mark Gardiner has argued that in the eastern Sussex Weald permanent settlement was more widely established than had previously been considered the case and that most of the High Weald as a whole had permanent settlement by 1086, 'even in the most distant areas of the Weald' (Gardiner 1995, 68). Roland Harris, in his overview of the most up-to-date Kent and Sussex Weald research, generally supported the idea that pennanent settlement occurred earlier rather than later, and that there was most probably some level of continuity with the Romano-British practice of transhumance (Harris 2003, 25). Gabor Thomas in a paper given in 2007 to the South East Research Framework 111 BRENDAN CHESTER-KADWELL challenges earlier assumptions that the Weald was colonised piecemeal from the edges inwards, but instead suggests that colonisation was an expansion of settlement from within - that is, the Weald was fully explored and known early on (possibly through transhumance) and more permanent settlement was produced on the back of this (Thomas 2007). In conclusion, it is fair to say that the early history of High Weald settlement is still very much open to debate. This article is a modest contribution to this debate: it suggests a re-appraisal of some of the evidence can provide new insights into the level of settlement development in the immediate post-Conquest period. Churches as Evidence of Settlement Churches are an indication of permanent and developed settlement that is at least substantial enough to bear the cost of supporting and maintaining the ecclesiastical infrastructure required for the local church to function. The sources for churches in the eastern High Weald are well kn0\\'11 and have been commented upon by a number of scholars over the years, but often with the sources from one county being explored in isolation from the other. This is an opportunity to summarise the evidence for both Kent and Sussex, specifically in the context of wider settlement studies. Domesday is very uneven in its recording of churches nationally. In some counties (such as Huntingdonshire, for example) it gives an almost complete record of the number of churches existing at the time. For the eastern High Weald generally, it is not particularly good, but other contemporary evidence exists that helps to overcome this deficiency. They are the Domesday Monachorum, the Textus Rojfensis for Kent,4 as well as other sources relating to the existence oflocal churches for Sussex, including the Chichester Cartulary, the foundation charter for St Mary's at Hastings, and the later, but still relevant, Taxatio of 1291. The aim here is to examine this evidence in order to establish where churches existed on the assmnption that these places must also have had communities to support them. The collected list of churches in the Domesday Monachorum identifies where contemporary churches were situated, but it is what can be inferred from it about local Wealden settlement that is of especial interest (Fig. 4). Examination of the Domesday Monachorum helps to explain not only what places were extant in the eleventh century, but also something of their relative importance. The first list in the collection is of those churches owing dues to the archbishop for chrism at Easter and the amount is an indication of the status of the church concerned, minsters and parish churches paying more than chapelries, for example. Thus, Appledore - a late Saxon minster church on the eastern boundary of the area under discussion (Riddler 2004, 33) - had a customary due of 7s., which was twelve times the minimum payment listed in the Domesday Monachorum of 7d. Other churches in the eastern High Weald listed there were Sandhurst, Rolvenden, Woodchurch, Benenden, and Cranbrook, which were all assessed at 28d. paid directly to the archbishop and probably indicating their status as parish churches. St Peter's, Newenden, appears in the second list under the churches that paid their dues to the Abbey of St Augustine, Canterbury, and is listed as a church subordinate to Lyminge. Considering the importance ofNewenden in Domesday this subordinate status might seem surprising (see below). The third list (after mentioning the dues 112 REAPPRAISAL OF ELEVENIH-CENTIJRY SETTLEMENT IN THE EASTERN HIGH WEALD • Domesday Churches In bolh counthJs Oth9r KENT CHURCHES 9 Domesday Monachotum churches 11st 1 􀀷 DM Churches liSI 2 ., DI.I Chut<:hos list 314 EAST SUSSEX Other SUSSEX CHURCHES Q Churche• 1/$/ed ;,, the CMchester Ce'1111a,y rlJ Churches llsted Jn the Tsx-atlo Cranbrook • EAST KENT Tenrerden Woodchun;h • • Battle Abbey Boundary or lhe Rape of Hasting11 ::::c.12 miles Fig. 4 Eleventh-century churches in the Eastern High Weald. Map constructed using infonnation from Morris 1976; 1983. owed by the minster churches 'before the coming of Lord Lanfranc as Archbishop' - i.e. superseded by the dues in the first list) aJso includes the churches 'of the tenure of St Augustus and beyond' (considered as a fourth list in the collection by Neilson (1974, 257a)). Amongst them are Stone in Oxney and Tenterden, both owing 7d., which may indicate that they were not yet established parish churches. T11is evidence suggests that most churches recorded in later medieval records in this part of the Kentish High Weald were already in existence by the eleventh century. The inference must be that there was sufficient permanent settlement to support these churches, and this paints a more encouraging and complete picture of settlement distribution than is implied by Domesday alone. Effectively, the eastern High Weald in Kent can claim to have attained a nearly complete complement of its medieval churches by the eleventh century, a feat that was only achieved in the Huntingdonshire Ouse Valley by the early thirteenth century. In the Huntingdonshire study area fifteen 'mother' churches and two others later identified as chapelries were mentioned in Domesday. By the thirteenth century, there were thirty-four churches and chapels of which twenty-one were parish churches. The average (estimated) size of Domesday parochia was about 4,000 acres, whilst the average size of parishes in the thirteenth century was about 3,000 acres. 113 BRENDAN CHESTER-KADWELL Unfortunately, the evidence for eleventh-century Sussex churches is piecemeal. Other than churches recorded in Domesday, it relies either on information gathered from charters (such as that connected with the foundation of St Mary's free chapel at Hastings) or deductions from the list of churches contained in the Taxatio of Pope Nicholas IV of 1291 (Rushton 1999, fig_ 2, Appendix 2). Domesday mentions a church at Salehurst in Henhurst hundred and another in Shoyswell hundred at Hazelhurst in what is now the parish of Ticehurst (Morris (ed.) 1976, 9, 82; 9, 60), although it is unclear whether the churches at Hazelhurst and Ticehurst were different churches on separate sites, or an earlier and later name for the same site . Churches at Ewhurst and Bodiam were mentioned in the Chichester Cartulary in the eleventh century, and the former church is believed to have been a late Saxon minster (Rushton 1999, 141 fig_ 5). Ewhurst was also one of those granted as a prebendary to Hastings College at its foundation (or re-establishment) by the Count of Eu sometime before 1086 (Gardiner 1989, 44). It is possible that Ewhurst's original parochia could have ex1ended over the hundreds of Henhurst and Shoyswell prior to the establishment of Salehmst and Hazelhurst, which each seem to have originally been the churches for their individual hundred. However, if this was so their previous relationship left no sign in the evidence that has survived from the eleventh century. Etchingham does not appear to have had its own church until later and an earlier church mentioned in the Taxatio as being at Burgham, a short distance away from the present church, appears to have been a chapelry in Salehurst. Etchingham was not formally established as a separate ecclesiastical parish until it obtained burial rights in 1362, and the present church was built soon after (Saul 1986, 140). It is impossible to arrive at a verifiable chronology for the Sussex churches in the Rape of Hastings, but the evidence suggests that the process of church building and parish formation in the Kent High Weald was in advance of that in East Sussex. However, this apparent discrepancy between the two counties may be as much a matter of the quality of the records that survive or the manner of church foundation and need not imply that settlement in the Sussex eastern High Weald was substantially different to that in Kent. Evidence for parish formation was emerging in the eleventh and twelfth centuries in both Sussex and Kent. Parish boundaries became clearly demarcated soon after their foundation, driven by the need to know the land from which tithes could be derived (Morris 1989, 210) and parish boundaries once established tended to remain fixed for long periods. However, the boundaries of the emerging parishes rarely conformed to the territorial divisions of the hundred. This complexity makes the hundred difficult to deal with within eleventh-century settlement studies and, therefore, being able to relate the contemporary parochial strncture to the hundred system is useful: it helps in the establishment of continuity and geographical location of settlement over time. Domesday Settlement in the Eastern High Weald (Fig. 5) Settlement as recorded in Domesday is expressed as manors or townships within a particular hundred. The parish (ecclesiastical or civil) had not yet taken on the significance it later had (and in fact still has) for defining the boundaries of local 114 REAPPRAISAL OF ELEVENIH-CENTIJRY SETTLEMENT IN THE EASTERN HIGH WEALD * Ootnesda:y settlements mentioned In lhe study area * Domesday settlements outside the slu