Skip to main content

Changing patterns of routeways in the landscape of the eastern High Weald from the end of the Roman period to the building of the Turnpikes

Page number
Brendan Chester-Kadwell

Brendan Chester-Kadwell.2017.Changing patterns of routeways in the landscape of the eastern High Weald from the end of the Roman period to the building of the Turnpikes.Archaeologia Cantiana.138:227.



brendan chester-kadwell

Before the Roman came to Rye or out to Severn strode,
The rolling English drunkard made the rolling English road.
A reeling road, a rolling road, that rambles round the shire,
And after him the parson ran, the sexton and the squire;
A merry road, a mazy road, and such as we did tread
The night we went to Birmingham by way of Beachy Head.

G.K. Chesterton 19131

In his poem The Rolling English Road, G.K. Chesterton (1874-1936) played upon a cultural perception of the country’s roads being original, eccentric, and standing for ancient liberties. This easily engages with the English People’s innate romanticism about what they perceive as their ancient, idiosyncratic and timeless rural roads. Whilst it is interesting to observe the many fascinating cultural associations and poetical accumulations that landscape (and roads as a crucial aspect of landscape) has acquired over time, what is the true picture that emerges when examining the actual origins, development, and morphology of roads? This paper aims to explore the reality through a study of routeways in an area of the eastern High Weald that straddles the county boundary between Kent and East Sussex, centred on the upper levels of the River Rother.

The term routeway is used generically in this paper to include a wide range of different types of specific landscape features over which there is public right of passage by foot, horse or vehicle. Different kinds of routeway (such as common ways, the king’s highway) are explored more fully in the section on landscape context below, while terms such as lane, track, or road are used within the body of the paper to convey an idea of scale, relative importance and physical form.

The research has involved field investigation aided by historical maps and other documentary sources. The majority of the written sources, together with the parish and estate maps that record roads, mostly date to the seventeenth, eighteenth and nineteenth centuries, with little material relating to earlier periods. The earliest published county and national maps originating in the sixteenth and early seventeenth centuries either fail to record the roads altogether, or give only rudimentary notions of where they were located. This is a potential problem when trying to establish the actual age and precise geographical location of routeways. Consequently, landscape features, topography, and archaeological evidence are of especial importance in establishing earlier patterns of routeways. Logical deduction from all the available evidence has been an essential part of the methodology for this study.

In dealing with the complexity of change within the network of local lanes, it has proved impossible to detail every example over such a large area. Therefore, this issue has been illustrated through two short case studies located across the parishes of Rolvenden (Kent), Salehurst and Ewhurst (both in Sussex). Similarly, whilst the results of detailed field research have been deployed within the analysis, the author is aware that there has not always been the space within a short article to articulate this evidence in detail.

A number of plans have been produced to illustrate the points made in the text. As the study area is quite large, the routeways marked on the plans cannot include all the subtlety of detail that ideally might be desired. Furthermore, the historical time that many of the plans cover is quite broad and within that time frame there were variations that cannot all be illustrated. However, it is hoped that any shortcomings will not detract from the main arguments presented in the paper.

The focus of this study is a selection of parish landscapes adjacent to the River Rother and its tributaries as they flow through the eastern part of the High Weald, exiting at Rye. The location of the study area is shown on Figs 1A and 1B straddling the counties of Kent and East Sussex. On the Kent side it includes the parishes of Sandhurst, Newenden, Rolvenden, Wittersham, Benenden and Tenterden. On the Sussex side the study area includes the parishes of Bodiam, Salehurst, Ewhurst, Northiam, Beckley and Peasmarsh.

Figs 1 A and 1B Location and Extent of the Routeway Study Area.

© B. Chester-Kadwell (2016).


The study area covers approximately 222km2, although the focus is on those areas close to the River Rother and its local tributaries – the Kent Ditch, the Hexden and Newmill Channels. The county boundary follows the Kent Ditch and the River Rother. Fig. 1B delineates the parish boundaries and illustrates the routes and destinations of the modern A and B road system, set within the context of three major towns, Cranbrook to the north, Tenterden to the north-east and Rye to the south-east.

Fig. 2 illustrates the current topography of the study area using selected contours for clarity (0-10, 10-30, 30-60 and over 60m) showing the location of the main parish settlements. The study area contains the upper levels of the River Rother and its tributaries (its lower levels approach Rye and the sea). The upper levels comprise a tract of country shared by the counties of Kent and East Sussex, much of which lies below the 5m contour along the valleys of the Rother river system, before it adjoins the wetlands of Romney Marsh to the east. The landform of these parishes consists of a gently undulating countryside dissected by a series of streams, many within steep-sided gills in their upper reaches. These gills are wooded and this, together with the shaws and the small blocks of woodland around old field pits, give the landscape a moderately bosky appearance. By contrast, the reclaimed flood plain of the Rother and the lower slopes of the land that skirts them are open and relatively treeless. Fig. 3 shows the area of the upper levels in detail, denoting what are known as the West and East Maytham Levels, the Newenden Levels and the Wittersham Levels. The ‘Upper Levels’, as they are collectively called, currently constitute a low lying, reclaimed marshland, with channels for the Rother and its tributaries, and criss-crossed by drainage ditches. This land, the flood plains of the Rother and its tributaries, are still regularly inundated in winter, despite modern flood defence work. Fig. 3 also shows the Upper Levels in relation to the modern road system, the current configuration of local lanes, the main river crossings and settlements.

Fig. 2 Topography of the Study Area. © B. Chester-Kadwell (2016).

(Fig.1B Legend applies.)


Fig. 3 Topography of the Upper Levels. © B. Chester-Kadwell (2016).


Of the parishes containing territory on the Upper Levels, Rolvenden and Newenden are amongst the ones most affected by the history of this landscape. Rolvenden occupies a ridge of higher land orientated north-west to south-east that forms the watershed between the Hexden and Newmill channels. It is of particular interest because situated between the East and West Maytham Levels it was, in much earlier times, effectively a peninsular only accessible by road from either Benenden or Tenterden, and without any land routes over the extensive waterways. The contemporary parish routeway configuration is of a different order: the two nucleated settlements, Rolvenden Street and Rolvenden Layne, stand at significant points on a road system that enables passage north and south, as well as east and west, across waterways and higher terrain, bridging the Upper Levels at key points. For example, the A28 traverses south across the parish to Rye and Hastings, and north to Tenterden and Ashford. While many local journeys continue to make use of lanes of great antiquity, traffic through the territory mostly makes use of an extensive network of A and B roads, the fruit of the turnpike revolution made possible by the reclamation of land prone to inundation.

Newenden, however, has substantially declined with the diminution of the navigable waterways. At the time of Domesday (1086), Newenden was the most significant place on the Upper Levels with a market and one of the few with its own entry in the Domesday survey (Morris 1983, 2.27). With the decline in the waterways Newenden lost the strategic and commercial importance that it enjoyed in the eleventh century. However, it continues to be an important bridging point on the River Rother, situated south of Rolvenden at the termination of an east-west orientated ridge at the confluence of the River Rother and the Hexden and Newmill Channels. Currently the bridge at Newenden carries the A268, which commenced at its junction with the A21 at Flimwell, joining the A28 on its journey south for a small stretch, and thence onto Rye.

evolution of routeways in the environs of the upper levels

The High Weald is quintessentially ‘ancient countryside’ or, perhaps more accurately, one particular version of ancient countryside.2 Most farmland in the Weald has always been owned and exploited ‘in severalty’ – as property managed by individual families rather than managed in common by whole communities. This is the countryside of isolated farms or small hamlets, where villages are for the most part a late imposition on the landscape. The settlement pattern reflects its origins in the piecemeal enclosure of extensive commons into areas of wood pasture called dens originating during the Saxon period. By the Conquest the commons had dwindled considerably (Witney 1976, 31-55), and the broad settlement pattern of the eastern High Weald seems to have been established by the time of Domesday (Sawyer 1976, 1-2; Gardiner 1995, 94; Chester-Kadwell 2014).

However, it was not desirable to enclose all land because, in a pastoral economy, space had to be found within which driven herds and flocks could be managed along the routes of the droves – spaces within which common rights could continue to be exercised and which survive into modern times as greens. Greens are a frequent occurrence in the High Weald and in Kent especially, and there were a surprisingly large number of them – mostly small and unevenly distributed.3 Residual greens resulting from the enclosure of commons were often positioned centrally to the geography of the parish. Their location relates to their original function, the management of flocks of animals from the secondary settlement areas (the High Weald dens) to their primary settlement (the coastal manors) along the drove roads.4 Indeed, it is apparent from even a cursory perusal of early maps that roads and common land were different varieties of the same basic thing – unenclosed land, freely accessed by the communities. Roads connected greens and commons, flowing into them seamlessly, and on occasions it is difficult to distinguish between narrow linear commons and widened routeways.

The evolution of the High Weald landscape, such as the creation of individual farmsteads and the management of animal husbandry through greens and droves, have helped to determine and shape and structure of the routeway system. This particular socio-economic environment ensured, of necessity, a particularly dense mesh of local routeways connecting the greens with long distance drove roads, farmsteads and other locally recurring resources.5 Routeways emerged largely because farms and privately owned fields needed to be reached without taking stock across the land of other proprietors. These ancient lanes survive where patterns of land management and the organisation of field systems remain unaltered. Routeways in the High Weald should be understood, therefore, as part of the wider history of the landscape through which they run. Local networks of public rights of way form an integral part of the local pattern of settlement, the character of field systems, and the history of very early piecemeal enclosure. However, they subsist within a framework of long distance routeways that are essentially medieval in origin.6 They often also incorporate stretches of Roman roads and prehistoric trackways and have been modified, in much later times, by the turnpike trusts.

There are a huge variety of different types of routeway in the contemporary High Weald landscape, and this applies equally to their usage and physical form. Some are major roads or country lanes, metalled and open to all traffic, yet others are green lanes, trackways, footpaths or bridleways where public usage is restricted to non-motorised traffic. Some of these distinctions are of considerable antiquity. The pattern has evolved gradually, with some routes falling out of use in whole or part as local needs changed, but with many examples, although largely redundant, surviving as rights of way.

Because of the difficulty of dating roads they are commonly classified in the literature either according to road-building epochs or usage, and this is no different for the Weald (Wooldridge and Golding 1966, 221-224). In terms of formal road building in the High Weald there were two main historical phases prior to the establishment of the modern system of road construction and maintenance that followed the formation of the modern county councils in the late nineteenth century. The first and the earliest was the network of Roman roads with their associated trackways; the second, the roads built under the Turnpike Acts of the eighteenth and early nineteenth century.7 Roughly 1300 years separate these interpolations and between these phases of formal road building, the network of Wealden routeways has tended to be classified by function: for example, as drove roads, or long distance medieval routes (Witney 1976; Everitt 1986; Brandon 2003).

During the Middle Ages there was, in effect, a rather vague and shifting classif-ication of routeways generally. Firstly, there were major roads, ‘king’s highways’, which included both the nationally important roads connecting major centres of power, and also those that simply linked the main villages. The latter shaded off without clear definition into the more numerous group of lesser local highways, generally referred to in contemporary documents as ‘common ways’. It is also important to understand that medieval roads differed conceptually from later roads. This was because a route between two places or points made use of land that was known to be common land or waste, avoiding land under agricultural use. Common land or waste was normally manorial land, held and administered on behalf of the local community by the manorial court. The highway was less a physical object in the landscape than a legally enforceable right of passage. A traveller might wander off the beaten track onto private land in pursuit of this right if the way was obstructed or too poorly maintained to use (Webb 1913, 4-9). This common law concept of the right of passage is still seen in how footpaths are used,8 but does not sit easily with the way that contemporary highways are experienced. Modern roads, essentially land managed and maintained by the local state, are bounded public spaces beyond which the right of passage does not normally extend.

Historically, within the High Weald as a whole, through-routes tended to follow the principal ridges and provide multiple alternatives for routes across the area. No doubt, this was welcome when roads were poor and often impassable in bad weather (Witney 1976, 131). Although the through-routes favoured the ridgelines, at some point each had to descend to the valleys to cross one or other of the principal watercourses. These cross-routes were complemented by a multiplicity of local lanes whose purpose was to connect centres of habitation and allow access to associated farmland.

With the development of the High Wealden dens over time into ‘private’ space the remaining commons, especially the roadside greens, became particularly important for as long as there were large movements of animals in the area. Almost without exception greens are situated either along the main drove routes on the ridge tops, or at the principal cross-routes connecting the ridge-top drove ways. They occur on relatively level areas of ground and at the junctions of through routes with local lanes. Greens would have had the double utility of providing over-night stopping off points for livestock passing through, as well as collection and dispersal points for the flocks and herds belonging to local farmsteads. Green locations often also had the poorer soils that did not favour arable production.

In lowland enclosed countryside medieval roads were almost invariably bounded by hedges or some other form of stock-proof barrier, except where they widened to form greens and commons. These roads were used by wheeled traffic, travellers on horseback or on foot, and livestock. The boundary treatments indicated the limits of the common way and prevented the stock being moved along them from straying into adjacent fields. Conversely they prevented stock in fields from wandering off along the road. It was rather unusual for a public road to run unhedged through a field, because such an arrangement would rely on the good sense of users to keep the gates leading into and out of the field firmly closed. The public highways in anciently enclosed countryside can thus be likened to a complex network of sealed strips of common land, widening out in places to form greens and other more extensive areas of common grazing.

From the sixteenth century, individual parishes were responsible for maintaining the public highways within their boundaries and allocated more or less effort to surfacing and repairs, depending on the extent of use and perceived importance of the road in question.9 It was very largely the inefficiency of this system (especially along the major ‘national’ routes) that led to the formation of turnpike trusts that in the High Weald are rarely found to date before c.1760. For the most part the present divisions between lanes used by all traffic and ‘green lanes’ only crystallised out as tar-macadamised road surfacing began to be adopted in rural areas in the course of the twentieth century. This divided what had been a single spectrum of roads into two distinct groups – those suitable for general motor traffic and those not so.

The wide valleys of the Rother, the Hexden Channel, and Newmill Channel were lands that until major reclamation measures were undertaken during the seventeenth century were historically prone to inundation and tidal streams below the level of (approximately) the 5m contour.10 The Rother was navigable someway up-stream of Newenden and its tributaries were also navigable to a greater or lesser extent, as a consequence of which water transport was of considerable importance. Up until the early modern period settlements around the Upper Levels were often more accessible by water than by land with many places interconnected by ferry.

Prior to the reclamation in the seventeenth century the topography of the area made it a difficult terrain for road building and this was a significant influence on historical road networks. As has been said, most long distance routeways were of necessity confined at that time to the ridge-tops, which are orientated broadly east/west. Travel by road north/south, against the grain of the land, was difficult where wide valleys prone to flooding hindered the way. The focus was on finding reliable crossing points, often necessitating longer journey times. It is not surprising, therefore, that major routes largely avoided this area completely, whilst the local network of lanes provided access to the water for local resources and water transport, or connected local settlement to easier crossing points upstream.

The drainage of the inundated land and the canalising of the Rother, Hexden and Newmill channels brought about a very different potential to how the landscape of the Upper Levels worked generally, and to the landscape of Rolvenden especially. This was not achieved without a long period of contention and dispute. The land around the Isle of Oxney was a contested area between two sets of interests over a 300-year period, from the thirteenth to the early seventeenth century. On the one hand, those wishing to maintain the navigability of the waterways and, on the other, the land owners who wished to reclaim the inundated land for agriculture and protect their landholdings from flooding: each wanted to maintain, enhance and protect their interests. For example, Knelle Dam, built during the fourteenth century as part of a scheme to improve navigation between the ports of Rye and Winchelsea and their hinterlands, diverted the River Rother from one side of the Isle of Oxney to the other. This was a large and significant engineering intervention that solved the problems of water management for a while but eventually led to the permanent inundation of much of the Upper Levels. The situation became unsustainable after 1600 and attempts to drain the Upper Levels whilst also maintaining the course of the Rother north of the Isle of Oxney failed. In the early 1630s the decision was made – once and for all – to breach the Knelle Dam to send the waters of the Rother back south of Oxney through the Wittersham Levels and reclaim as much of the inundated land in the Upper Levels as possible (Eddison 1985, 95-106; Eddison 2000, 105-110). This effectively restricted the waterways to small craft (lightermen) and limited the movement of larger vessels between the coastal ports of Rye and Winchelsea and the hythes and wharfs further inland, such as Appledore, Small Hythe, Maytham, and Newenden.

The reclamation of the Upper Levels of the Rother was a defining process in the history of routeways in the area and a pivotal point in the present analysis. The network of routeways prior to this event was determined by the topography of inundation, marine activity and the human processes of landscape management as it was practiced at that time. The post-reclamation routeway network changed radically from what had gone before, spurred on by the opportunities presented by the better management of the waterways and the freedom to construct and successfully maintain roads over the previously inundated land. As will be seen, this affected both the way that long distance routeways were adapted to the terrain and how the network of local lanes developed. How major or long distance routes coped with the terrain before and after the reclamation of the Upper Levels, and how reclamation and changes to the major routes affected the local network of lanes are the two themes that define the analysis presented in this paper.

Early Long Distance Routeways in the Vicinity of the Upper Levels

Very little is known about pre-medieval routeways in this part of the High Weald, although it is reasonable to expect that they would have existed in some form or another. Palaeo-environmental evidence for early agriculture and deforestation also supports the view that the area was subject to human activity in the pre-Roman era but unequivocal evidence for the identification and dating of routeways to this time is not available. Speculation concerning the purpose of possible early routeways, often assumes that their use in some way anticipated later uses as drove roads. However, there is no way of being able to identify with certainty which (if any) of the lanes and tracks in the study area are definitely pre-Roman, but common sense suggests that some might be (Harris 2003, 22-23).

Although there is archaeological evidence for Roman activity in many of the parishes within the area of the Upper Levels – including industrial sites relating to the iron industry, interments, and a number of roads – the intelligence from this evidence remain patchy. In this part of the High Weald Roman roads were mainly secondary routes of relatively poor construction that tended to avoid the difficult topography of the Upper Levels. Additionally, they frequently adopted a sinuous course in order to deal with the challenges of the local terrain. Fig. 4 illustrates the two principal Roman roads; one took a broadly north/south alignment that passed to the west of the Levels, with a second going east/west that ran north of them (Margary 1973, 44-49). In terms of the modern topography, the road running north/south connected Rochester (Durobrivae) to the iron producing area north of Hastings, entering the parish of Benenden near Hemsted and leaving it south of Iden Green, crossing the Hexden brook at Wandle Mill. It then passed through Sandhurst parish and crossed the Kent Ditch at Bodiam into Sussex. From there it proceeded to Sedlescombe and on south towards Hastings. The east/west route branches from the Rochester to Hastings road at Hemsted, near where evidence for Roman settlement has been found (Pollard and Aldridge 2008, 301-307). This road proceeded east from here, eventually connecting to Canterbury (Durovernum Cantiacorum). Although the Rochester to Hastings route continued to be used in parts by later tracks and lanes, there were also many deviations from it. However, the Benenden to Canterbury route was largely ignored by later routeways, and other more local Roman roads that Margary speculates about have gone completely (Margary 1973, 47).

Fig. 4 Approximate Line of the Roman Roads Upper Levels.

© B. Chester-Kadwell (2016).


The Principal Medieval Routeways and some Post-Medieval Adaptations

Since medieval roads also avoided the inundated areas it is difficult to account for why more use wasn’t made of the Roman roads prior to the reclamation of the Upper Levels in the seventeenth century. It may be partly to do with the nature of the Roman roads themselves, which in the High Weald were usually constructed of material that would have needed continuous maintenance to ensure their use over time. Additionally, designed as through routes for an earlier period, the alignment of the Roman roads often put them in the wrong place to act as routeways connecting later local settlement. The siting of early Saxon settlement, for example, tended to avoid Roman roads and later settlement seems to have followed this pattern.11 As a consequence later routeways related to settlement developed differently. This does not mean that routes used by Roman roads were not used at all in the post-Roman period, but neither does the later use of Roman roads necessarily imply continuity of use. For example, a routeway may go out of use for a while before (at a later date) being re-used, especially where the nature of the local topography is such as to make this the logical choice.12

Whatever the reason for their abandonment, Roman roads do not appear to have shaped the framework of routeways over the long term and medieval routeways in this part of the High Weald seem often to have ignored them. Throughout the Middle Ages travel between London and the principal continental ports of Rye and Winchelsea was of considerable importance.13 One of the principal routes from London was through Tonbridge and across the Weald to Rye and Winchelsea via a number of alternative ways,14 but on these routes the all-weather roads still needed to circumvent the Rother Levels as they had done in Roman times.

Although in general terms medieval roads in the High Weald tended to keep to the higher land of the ridge tops, their actual course can be difficult to determine and where the evidence is forthcoming they often take rather surprisingly circuitous routes. In practice, the course of a long distance route was effectively the joining up of many local lanes rather than a planned long distance road as happened later under the turnpike trusts. In many cases there is simply not one major routeway but a number of parallel ways comprising a series of local lanes that are aligned in a similar direction through which traffic filtered: this was especially so where roads had to traverse against the grain of the landscape. Negotiating early road systems, therefore, would have required a degree of specialist knowledge, particularly as there was unlikely to have been proper signposting. The medieval traveller would need to know the route through personal experience, or have employed a guide who did (Furley 1874, vol. II pt 2, p. 256). In places like the High Weald where there was a complicated network of possible alternatives, the preferred option may have been to employ a series of local guides as the traveller progressed through the countryside. Local guides would have known the best routes to take according to local conditions at the time of travel – as indeed was still the case as late as the nineteenth century in some parts of south-east England (Cobbett 1983 [first published 1830], pp. 88-89).

The exact route taken by medieval roads across this area is frequently difficult to identify, although it is sometimes possible to speculate with some degree of certainty. Symonson’s map of 1596 gives a generalised view of how the major routeways crossed the area, but is of too small a scale to show the detail.15 Its significance rests on the fact that it shows the area prior to the improvements to the Levels of the early seventeenth century and may, therefore, reflect the late medieval system fairly well. Symonson shows the Rother flowing north of the Isle of Oxney a few years before the Knelle Dam was breached in a storm. Using Symonson’s map as a rough guide, with some detail added from fieldwork and the judicious use of later maps, Fig. 5A illustrates the most likely course of the major medieval routes through the study area, which may be summarised as follows:

Fig. 5A Major Medieval Routeways Upper Levels.


© B. Chester-Kadwell (2016).

a) The main route to the north of the Levels [A on the plan] ran from Cranbrook to Tenterden passing north of Benenden and Halden Park thus avoiding the Newmill Channel. The road then continued from Tenterden to Appledore and New Romney over Romney Marsh. Today the Tenterden road from Cranbrook is unclassified, but continues from Tenterden to New Romney as the B2080.

b) The route from London to Rye via Tonbridge and Flimwell (the modern A21) passes along the Sandhurst Ridge (where it becomes the A268) to the crossing point of the Rother at Newenden and then onwards to Northiam and Rye [B]. This route avoided the most difficult parts of the Levels but engaged with the valley of the Rother at Newenden. This reduced the journey time, bypassing the Rother crossing at Robertsbridge (the Hastings Road) and avoiding the longer southern route [C].

c) The route from north Kent that skirted the Levels to the west was more problematic. It seems that in the sixteenth century the Rochester road (effectively the route of the Roman road) divided at Camden Hill (just to the north of Cranbrook): one branch [D1] passed through Cranbrook to meet the Newenden road at Hawkhurst; the other [D2] proceeded to Benenden (continuing along the route of the Roman road, although along a different alignment) and eventually joined up with the Newenden road somewhere near Sandhurst.

How this latter routeway [D2] covered the area between Benenden and Sandhurst is not clear as there are a number of alternative routes that it may have followed, but it illustrates many of the issues concerning the workings of these early through routes over difficult terrain (see Fig. 5B).

Fig. 5B Conjectured Medieval Through-routes in the Benenden Area.

© B. Chester-Kadwell (2016). (Contours and Legend as per Figure 5A.)


Fig. 5B shows how the whole route illustrated by Symonson commences at Camden Hill, crosses the Cranbrook to Tenterden Road (almost certainly at Golford Green) and having passed through Benenden joins the Flimwell to Newenden Road to the east of Sandhurst. One possible route through Benenden would have been through Hemsted along the line of the Roman road to Iden Green (crossing by a paved ford at Stream farm) and from there via Standen Street and Hopemill Bridge to Ringlecrouch Green (the modern Crouch Lane) on the Hawkhurst to Newenden road [D2a, marked in red]. Alternatively, the road may have passed through Benenden Green and turned off the ridge near Pullington Farm along a now abandoned lane towards Iden Green and then on as before [D2b, marked in yellow]. It is also possible that the road may have turned south-east off the lane to Iden Green taking a now lost lane to Dingleden and onto Kensham Green in Rolvenden [D2c, marked in orange].16 From Kensham Green it would have crossed the Hexden stream either at Hopemill or possibly through a series of abandoned ways past Frenchhurst Farm to Lamberden Farm on the Newenden Road (not shown on Fig. 5B).17 It is as likely that these were alternative ways offering choice depending on the actual state of any single route, with some perhaps being favoured at different times of the year.

Finding convenient crossing points for the main rivers was another consideration in planning a route. Bridges were impractical across the broadest valleys in areas prone to inundation or tidal streams and even ferries could be problematic where there were seasonal variations of water level. Bridges were also expensive to build and maintain, which meant that few were constructed for pragmatic considerations.18 There were three crossing places on the Rother within the study area that resulted in bridges being constructed either during the course of the Middle Ages or in the Early Modern period; namely, Newenden, Bodiam (where there was also an earlier Roman crossing) and Robertsbridge; but the most significant for the purposes of this paper is that at Newenden. This crossing point was the lowest one on the Rother along the shortest route to Rye from Tonbridge (itself a key crossing on the Medway). It is not surprising, therefore, that as time went by local routes in the vicinity of the Upper Levels tended to gravitate towards the Rother crossing at Newenden.

Medieval crossing points, where attempts were made to carry the road across the river rather than rely on a ferry or simply fording the river, were often a combination of causeways and bridges. This was an effective technological approach as the valley of the Rother is flat and wide and the river would have had in many places a number of channels. The crossing point in Salehurst parish at Robertsbridge was probably typical of many medieval engineering projects of this type, which can still be seen here in the modern landscape. It is said that at one time there were as many as seven channels to be bridged at Robertsbridge and the process was to build a causeway between the channels, which were then individually bridged. This method made the process easier than attempting to bridge one large channel, but causeways also made the area more prone to flooding as they acted as a dam when water flows were exceptionally high. Where the river was forced into one larger channel bridges were more efficient, but they still had to be reached by a causeway used to carry the road above flood levels. In the lower reaches of the Rother such as at Newenden and Bodiam, where the demands of navigation required that the bridges could also allow the passage of larger craft, it was more advantageous to encourage the water into a single channel. Both Newenden and Bodiam are examples of how the technique of combining bridges with causeways was perfected during the eighteenth century.

Navigation in the East and West Maytham Levels on the Hexden and Newmill Channels (the two largest tributaries feeding the Rother in the Upper Levels) was more restricted than on the Rother itself and the craft able to use them must always have been lighter. However, although the demands of navigation were less than on the main stream, no attempt to bridge these valleys was made prior to the seventeenth century. This was largely because of the difficulty of the terrain and the width of the valleys that constituted the Maytham Levels. It was not until serious efforts were being made to cure the problems caused by the inundation of the Upper Levels that the establishment of new crossing points could be considered.

There were some modifications to the major routes as a result of the reclamation of the Upper Levels during the course of the seventeenth century and the early years of the eighteenth century. An attempt was made to solve the problem of inundation in the Upper Levels between 1600 and 1633 without sending the waters of the Rother south of the Isle of Oxney through the Wittersham Levels. Although unsuccessful in its principal objective, it did produce some results. One of the initiatives designed to manage the water in the Hexden Channel was to have a lasting benefit for the improvement to road transport in the long term. A raised causeway that was erected from below Frogs Hill in Newenden across the valley of the Hexden Channel enabled the stream to be bridged, with a sluice set above the causeway to assist in controlling the flow and enabling the stream above to be canalised. The causeway is shown in a plan of 1633, produced to show the disposition of lands before the Knelle Dam was breached the following year [ESRO: ACC 2806-1-9-02a]. This new bridge allowed access from the local network of lanes around Forsham and Kensham to the Sandhurst to Newenden road towards Rye and eventually presented a new opportunity for long distance travel as well.

Once the Rother had been allowed into the Wittersham Levels the Newmill Channel was diverted to the south-west to a new confluence with the Rother above Blackwall. To achieve this, a cut was made at Potman’s Hoath where the Knelle Dam was breached to allow the waters from the Newmill Channel to flow south. However, the reclamation of the Upper Levels and the establishment of the Rother to the south of the Isle of Oxney effectively took the best part of a hundred years to fully achieve. At some point during this time frame another causeway was built connecting a stretch of the Knelle Dam at Potman’s Hoath to a point between Maytham Wharf and Maytham Farm in Rolvenden parish. For the first time a land route was opened up between Rolvenden and the Isle of Oxney by utilising the aforementioned causeways and bridging over the breach in the Knelle Dam at Potman’s Hoath. This new road connected the existing Maytham Road from Rolvenden Layne that hitherto had stopped at the edge of inundated land from whence a ferry operated across the Maytham Levels, which it now replaced.

The Effect of the Turnpike Trusts on the Major Road System

A significant issue for the efficiency of the public highways was their maintenance and improvement, thus facilitating the safe passage of the sovereign’s subjects along the common highway.19 In 1555 a statute was passed by Parliament designed to remedy what was, by then, perceived as a comprehensive breakdown of the post-Conquest medieval system of road management (Webb 1913, pp. 9-26).20 From that date it became the responsibility of the local parish community and the county justices to undertake the maintenance and necessary improvements of the road system in their area. Each parish appointed a surveyor tasked with organising annually the labour service due by all parishioners to work on road repair. Not surprisingly this system proved inefficient and caused considerable resentment amongst the poorer members of parish society in particular. Despite the efforts of local parish surveyors, the road system in the High Weald remained largely unaltered from the medieval network until the introduction of local turnpike trusts later in the eighteenth century.21 Whilst the implementation of the turnpike system did not replace the responsibilities of the parish for general road maintenance, it did provide a more efficient process for upgrading and maintaining major routes.

Parliamentary Acts were used to create local road building trusts to improve selected key routes. These trusts enabled a programme of road building to be embarked upon that better reflected local needs and aspirations and, at the same time, improved the long distant routes required for trade and other essential functions both at the national and regional level. The tolls charged for their use were intended to pay for them and the trustees were empowered to build gates (known as turnpikes) across their roads for this purpose. Within the eastern High Weald it was usual for turnpike trusts to improve existing routes rather than constructing completely new roads.22 This meant making general improvements to the roadbed (such as widening, levelling and resurfacing), whilst building short stretches to straighten existing routes or cut off inconvenient diversions. Notwithstanding these limitations, a network of turnpike roads was generated within the study area during the latter half of the eighteenth century and the early years of the nineteenth. Fig. 6 illustrates how extensive this network of roads became over a period of about one hundred years (1753-1841). The earliest turnpikes in the area connected places like Rye [1A] and Hastings [1B] to the developing national system of roads centred on London. Similarly, the later turnpikes were intended to further improve local connectivity with other urban centres in the south-east region [2 & 3]. Once the economic benefits that the turnpikes generated was recognised, many cross-country routes were turnpiked as well. Some of these roads remained little more than country roads and today are nearly indistinguishable from other lanes. An example of a minor turnpike of this kind is the road from Staplecross through Bodiam to Silver Hill in Salehurst (although the deep cuttings constructed though the ridge in Bodiam are impressive) [4].

Fig. 6 Extent of the Turnpike Trust Roads c.1753-1841 in relation to


the Medieval Road Network. © B. Chester-Kadwell (2016).

After the reclamation of the Upper Levels had been accomplished there were more opportunities for the road system to be extended into this area. However, in practice only three turnpikes were constructed over land previously inundated. The first was the road from Tenterden to Appledore via Reading Street (originating at Cranbrook), authorised in 1761 and now known as the B2080 [5A]. This road was mainly an upgrading of an older medieval road that appears on Symonson’s map. The second was the extension from Tenterden to Newenden (1767), which was an improved and partially reconstructed route that seems to have been a rationalisation of a series of earlier lanes (now the A28) [5B]. The third was a completely new road constructed following an Act of 1841 between Cripp’s Corner in Ewhurst to Gill’s Green in Hawkhurst, which also provided a new bridge over the Rother between Salehurst and Bodiam (now the B2244) [5C]. Although few in number these three routes greatly improved passage through an area that had been, in previous times, more accessible by water. The work of the turnpike trusts created the system of roads that survive almost unchanged to the present day within the study area.

Impact of Changes in the Major Road System on Local Routeways

As the major road system evolved, especially in the eighteenth and nineteenth centuries, older patterns of routeways were modified in response to these changes. A particularly significant example can be seen within the parish of Rolvenden, which demonstrates very clearly not only how a system of routeways may change over time, but in this case also how the reclamation of inundated land and the formation of new major roads impacted on older patterns of communication. A later example of modification to a system of lanes on the Sussex side of the Rother in the parish of Ewhurst also demonstrates how a new road may effect change, evidencing the process that for earlier examples is often missing.

The Rolvenden Case Study

The functionality of the earliest routeways within the parish of Rolvenden was largely determined by the topography of inundation and the pattern of settlement. In the south of the parish (where the case study is located) it was the extent of inundation along the Hexden Channel and the distribution of agricultural settlement within very specific early land divisions that influenced how routeways were formed and maintained. Fig. 7A illustrates the extent of the framework of lanes that served the socio-economic needs of the medieval settlement pattern. The key factors were: the location of Kensham Green to the west and Layne Green to the east; the existence of a set of land divisions that stretched back northwards from the Hexden Channel and appear to have used as their boundaries a series of southward flowing local streams off the main ridge upon which the Den of Rolvenden was situated;23 and finally, the location of the lowest crossing point of the Hexden Channel south west of Kensham Green. The Hexden Channel impeded the establishment of land routes between Rolvenden, Sandhurst and Newenden below the established crossing point near to Kensham. This created a strong east/west bias for one set of local lanes connecting a range of settlements to the crossing point and further inland towards Benenden. A number of shorter north/south elements maintained the need for access to local resources along the Hexden or Newmill Channels for local communities (at a on Fig. 7A).

Fig. 7 Rolvenden Case Study Area: (A) Routeways Prior to the Reclamation of the Upper Levels; (B) Surviving Local Lanes and Turnpikes. © B. Chester-Kadwell (2016).


It is possible to largely reconstruct this older and more coherent pattern of routeways. Many of the earlier lanes are still extant, whilst others survived long enough to appear on earlier parish and estate maps. Some existing lanes may be quite late and it is uncertain whether they originated in the earlier landscape [marked b on the plan]. Other elements survive as abandoned features in the landscape or are conjectured from an interpretation of the form of surviving lanes and landscape features, many of which remain as footpaths or rights of way over private land.

Whilst Fig. 7A illustrates how the current network of lanes north of the Hexden Channel formed part of an earlier, more rational system of routeways, Fig. 7B shows the lanes that survive in the modern landscape in relation to the turnpike roads of the late eighteenth and early nineteenth centuries. However, the loss of earlier lanes happened over a long time frame and whilst many adaptations were made along the routes of the turnpikes at their time of construction, other of the now defunct lanes were still in existence until comparatively recent times. A small number of lanes were created in the post-medieval period [marked c on Fig. 7B] and other (once important ones) have been reduced to byways that survive as green lanes.

A series of lanes running east/west connected Layne Green to Kensham Green and this connection is retained in the present day landscape, but in a simplified (and rather circuitous) fashion. The reason for this simplification seems mainly to have occurred following the establishment of a new crossing point lower down the Hexden Channel at sometime between 1600 and 1630. This opened up the potential for a direct road link between the parish’s main settlement at Rolvenden Street and the bridge at Newenden. At first this may have been achieved through the existing network of lanes, but following the establishment in 1767 of the turnpike between Tenterden and Newenden (via Rolvenden) the route was rationalised.24 It is likely that the establishment of this important extension of the Tenterden turnpike involved some new stretches of road as well as improvements to existing stretches. It seems probable that the number of east/west-orientated lanes crossing the new road (of which there were many) was reduced and some may well have been diverted. This process effectively disrupted the earlier pattern of lanes and that this was a potent factor in the simplification of the previous system of routeways.

The earlier north/south routeways that connected local settlements to the shores of the inundated lands, giving access to ferries and other water-orientated resources became of less importance or relevance once the land had been reclaimed. The decline in the number of such lanes may, therefore, have commenced earlier than the disruption to many of the east/west orientated lanes caused by the interjection of the turnpike. However, the end result was the same and the system became simplified in what was, to some extent, a rather random fashion.

The Ewhurst Case Study

It seems that a similar loss of local lanes could occur wherever there was a radical intervention in how lines of communication were configured. A later example of a similar interruption of an established ancient system of routeways can be found south of the Rother on the border between Ewhurst and Salehurst. The insertion of a very late turnpike road following an Act of 1841 between Cripp’s Corner in Ewhurst parish and Gill’s Green in Hawkhurst led to a comprehensive re-organisation of local lanes, almost at a stroke.

Prior to the construction of what became known as the Junction Road turnpike a series of farmsteads located along the southern banks of the Rother Valley were served by a number of local lanes (see Fig. 8A). These lanes connected with two major routeways running north/south that served the crossing points at Robertsbridge (in Salehurst) and that on the borders of Ewhurst and Bodiam (both medieval routes, later turnpiked). Following the construction of the Junction Road, a newly laid out road that also established an additional bridging point over the Rother upstream of Bodiam Bridge, access to these farms was radically altered (Fig. 8B). The lanes that did not cross the new road survived (for example, Poppinghall Lane and part of Abbey Lane), but those that did were effectively reduced to access tracks serving local land resources, whilst remaining rights of way for pedestrians or as bridleways. New access ways were created for most of the farmsteads that either led onto the Junction Road, or towards Robertsbridge effectively removing the need to maintain the old lanes as public highways. Additionally, new settlement tended to be built directly off the Junction Road, reinforcing the new pattern. The greatest extent of the road closures occurred in Ewhurst parish itself, and the rationalisation happened there more thoroughly than elsewhere. This suggests that (not unexpectedly) there was a more holistic approach to transport planning in the mid to late nineteenth century than was generally found in earlier times.

Fig. 8 Configuration of Local Lanes between Robertsbridge and Staplecross:

(A) Routeways in 1805 and (B) Routeways Post 1841. © B. Chester-Kadwell (2016).


Because of the late date, this area had already been extensively mapped prior to the establishment of Junction Road Turnpike (now the B2244). Subsequent mapping has shown how the lanes decayed very swiftly following the construction of the new turnpike, whereas elsewhere in the earlier period secondary routeways within the High Weald (mainly represented by local lanes) often passed out of use almost without notice or public record. By the time that parish maps became more common, typically in the late eighteenth and nineteenth centuries, many of these lanes (or sections of lanes) were still extant but have subsequently disappeared. Evidence for abandoned routeways (frequently occurring within woodland) can be found in the general landscape that do not appear on any of the existing maps, suggesting that this simplification of the system of ancient lanes in the High Weald has been happening over a long period of time.

overview and conclusions

This study began with a challenge to the poetic and cultural perception of our country roads and routeways as unchanging, especially those in areas like the High Weald. After careful research, the reality seems to be much less romantic, but a great deal more complex and interesting. Located in ancient countryside (its formal appellation by landscape historians) with a maze of trackways and lanes, the routeways of the High Weald are regarded as part of its ‘historic’ character. The terms ‘ancient’, and ‘historic’ have encouraged the perception that routeways here have hardly changed over time – that what is experienced now has been bequeathed permanently by long-ago ancestors. However, the landscape of the Eastern High Weald, in particular, has altered in many ways over a long period of historical time. This is especially true in areas such as the Upper Levels of the River Rother where significant topographical developments have occurred as the result of previously inundated land being reclaimed for agriculture. As the lands of the Hexden and Newmill Channels were reclaimed this led to the re-configuration of through routes and the opening up of previously isolated areas. The narrative of this study has looked at the effect of these processes and the impact brought about upon the network of local lanes after the building of the turnpikes, using two detailed examples in Rolvenden and Ewhurst.

It is clear that routeways (roads, tracks, paths, highways) in the High Weald have evolved over time driven by the socio-economic needs of the communities they serve. Routeways are dynamic systems that need to change as economic and social needs change and, whether these changes are slow and piecemeal or rapid and sudden, the effect on the overall pattern may be considerable. Understanding the process of routeway creation and deposition in the High Weald is contingent on an understanding of settlement and the socio-economic realities of settlement. In previous times roads were not simply ‘objects’ in the landscape; they were passageways over land owned by the ‘community’ and they reflected the lives of those who travelled them as well as those who lived by their margins. Today, roads have become bounded public spaces within a predominantly privately owned landscape and because today’s road-makers have the technology to recast large swathes of road-scape within the blink of an eye, an older landscape such as the High Weald is often treasured for its apparently ancient unchanging ways. It may surprise us, therefore, to find this notion overturned: the High Weald routeways of today (together with their redundant features still discernable in the landscape) tells a complex and dynamic story of change in a changing landscape.

Routeways in the High Weald are an under-researched phenomenon. More use of recently developed techniques, such as LIDAR, can prove useful in establishing better landscape evidence, but more traditional methodologies could also be better deployed; for example, a systematic study of estate plans and early maps to learn how routes were regarded in the past. Further research needs to be done to establish the original nature, status and significance of relict routeways, where there remains an element of speculation. Although absolute dating may never be achievable, further research may establish a better understanding of earlier patterns of routeways, later stages of enclosure around greens and (following that) a sequencing of routeway generation and redundancy may be possible.


The writer would like to thank Dr Ernest Pollard and Professor Tom Williamson (University of East Anglia) for agreeing to read the draft of this paper. Their com-ments, suggestions, and encouragement have been invaluable and very much appreciated – the result would have been poorer without them. Any errors or shortcomings, however, remain with the author. The narrative would have been impossible without the accompanying figures. Therefore, thanks go to Pat Chester-Kadwell who not only composed the plans, but steadfastly coped with the many revisions that were required from time to time.

It should be noted that all of the plans were produced based on OS OpenData Contour data and OS VectorMap District (Vector) TQ, subject to the terms at


Albert, W., 1972, The Turnpike Road System in England 1663-1840, Cambridge.

Brandon, P., 2003, The Kent and Sussex Weald, Chichester.

Carley, J., 1970, The Turnpike Roads of Kent, Kent County Council.

Chester-Kadwell, B., 2010, ‘A Sense of Place in Rural Settlement: a locally oriented study of the Huntingdonshire Ouse Valley and the Eastern High Weald’, University of East Anglia, unpubl. doctoral thesis.

Chester-Kadwell, B., 2014, ‘A Reappraisal of Eleventh-Century Settlement in the Eastern High Weald’, Archaeologia Cantiana, Vol. 135, 105-130.

Chester-Kadwell, M., 2009, Early Anglo-Saxon Communities in the Landscape of Norfolk, BAR British Series 481.

Chesterton, G.K., 1914, The Flying Inn, London.

Cobbett, W., 1983 (original publ.: 1830), Rural Rides, Harmondsworth.

Eddison, J., 1985, ‘Developments in the Lower Rother Valleys up to 1600’, Archaeologia Cantiana, Vol. 102, 95-107.

Eddison, J., 1995, ‘Attempts to Clear the Rother Channel, 1613-1624’, in Eddison, J. (ed.), Romney Marsh: the Debatable Ground, OUCA Monograph 41.

Eddison, J., 2000, Romney Marsh: Survival on a Frontier, Stroud.

Everitt, A.M., 1986, Continuity and Colonization: the Evolution of Kentish Settlement, Leicester.

Furley, R., 1874, A History of the Weald of Kent: with an Outline of the Early History of the County, 3 vols, H. Igglesden: Ashford.

Gardiner, M., 1995, ‘Medieval Settlement and Society in the Eastern Sussex Weald’, unpubl. ph.d. thesis, University of London.

Gardiner, M., 2007, ‘Hythes, Small Ports, and Other Landing Places in Later Medieval England’, in Blair, J. (ed.) Waterways and Canal-Building in Medieval England, Oxford.

Harris, R., 2003, The Making of the High Weald, High Weald Area of Outstanding Natural Beauty Unit: Flimwell.

Harrison, S.A., 2005, ‘A History of Evolution and Interaction: Man, Roads and the Landscape to 1850, unpubl. ph.d. thesis, University of East Anglia.

Hasted, E., 1797-1801 (2nd edn), The History and Topographical Survey of the County of Kent, Canterbury.

Holloway, W., 1849, The History of Romney Marsh, London.

James, R., 2007, An Archaeological Desk-Based Assessment of the Strawberry Wood Culvert, Benenden, Kent, Archaeology South-East.

Keith-Lucas, B., 1984, ‘Kentish Turnpikes’, Archaeologia Cantiana, 100, 345-369.

Margary, I.D., 1946, ‘Roman Roads in West Kent’, Archaeologia Cantiana, Vol. 59, 28-63.

Margary, I.D., 1948, Roman Ways in the Weald: illustrated with maps, diagrams and photographs, Phoenix House: London.

Margary, I.D., 1973 (3rd edn), Roman Roads in Britain, London.

Morris, J. (ed.), 1976, Domesday Book 2: Sussex, Chichester.

Panton, F. and Lawson, T., 2004, ‘Turnpikes, Roads and Waterways 1700-1850’, in Lawson, T. and Killingray, D. (eds), An Historical Atlas of Kent, Chichester.

Pawson, E., 1977, Transport and the Economy: the Turnpike Roads of Eighteenth Century Britain, London.

Peak, S., 2009, East Sussex Turnpike Roads, Hastings Chronicle.

Pollard, E. and Aldridge, N., 2008, ‘An Early Boundary, probably Anglo-Saxon, associated with Roman sites in Benenden’, Archaeologia Cantiana, Vol. 128, 301-307.

Rackham, O., 2000 (first publ. 1986), The History of the Countryside, London.

Rippon, S., Smart, C. and Pears, B., 2015, The Fields of Britannia, Oxford.

Sanders, A., 2004, Cranbrook’s Turnpike Roads, Cranbrook.

Sawyer, P.H., 1976, ‘Medieval Settlement: Continuity and Change’, in Sawyer, P.H. (ed.), English Medieval Settlements, London, 1-7.

Scott-Giles, C.W., 1946, The Road Goes On, London.

Taylor, T., 1979, Roads and Tracks of Britain, London.

Ward, G., 1937, ‘Saxon Records of Tenterden’, Archaeologia Cantiana, Vol. 49, 229-246.

Warner, P., 1987, Greens, Commons and Clayland: the Origins and Development of Greenside Settlement in East Suffolk, Leicester.

Webb, S. and B., 1913, English Local Government: the Story of the King’s Highway, London.

Williamson, T., 2016, ‘The Ancient origins of Medieval Fields: a Reassessment’, Archaeological Journal, vol. 173, No. 2, pp. 264-287.

Witney, K.P., 1976, The Jutish Forest: a study of the Weald of Kent from 450 to 1380 AD, London.

Wooldridge S.W. and Golding, F., 1966, The Weald, London.

Other Reference material:

East Sussex Record Office: (a) Plan of the Upper levels c.1633: ACC 2806-1-9-02a; (b) Richard Budgen’s Map of Sussex 1724: ACC 3762/1.

Kent History and Library Centre: (a) Newenden Bridge, covering the period 1637 to 1749: U386/02/3; Q/AB/42; (b) Plan of the Upper Levels 1633: S/Ro P1.

Turnpike Trusts: County reports of the Secretary of State, under the Acts of Wm. IV. Cap. 80, No. 1 Kent; No. 2 Sussex, HMSO, 1852.

Maps and Plans:

1. Symonson’s map of Kent 1596.

2. A large number of parish maps and estate plans were consulted, including the tithe survey maps for the relevant parishes. There are too many to list separately here, but further information may be sought from the author.

3. OS maps, including the County Series 1:10560 1846-1969 and the County Series 1:2500 1854-1949, as well as the first edition 1in. (1813) and 2in. (1805) maps where appropriate.


  • 1. Although first published in 1912, this poem was reproduced in Chesterton’s novel The Flying Inn the following year, a futuristic allegory on modernistic erosion of traditional English values and culture.
  • 2. a. b. The High Weald landscape differs strikingly from those areas of England that had a ‘champion’ landscape in the Middle Ages. The density and character of the local route network was one of the key features used by Oliver Rackham to distinguish the ‘ancient’ countryside from the ‘planned countryside’ of eighteenth- and nineteenth-century enclosure (Rackham 2000, 4-5).
  • 3. The map evidence for the greens is more plentiful from the eighteenth century. Hasted records more than appear on the modern map. Most were either fully enclosed or mainly so by the time of the tithe surveys and continued thereafter to be growth points for habitation.
  • 4. Everitt claims that greens on the North Downs of Kent are often associated with ‘common pastoral woodlands’ (Everitt 1986, 147). Similarly, Warner proposes that relic greens on the claylands of East Suffolk are the result of the enclosure and cultivation of extensive commons in the late Saxon period (Warner 1987, 13-15). This area of East Suffolk is of particular interest because, like the Weald, it is one of apparent discontinuity between earlier Roman settlement and later Saxon colonisation (Warner, 1987, pp. 9-12).
  • 5. Routeways that have essentially local utility of this kind have been referred to as ‘resource linkage’ routes (Harrison 2005, pp. 147-149).
  • 6. ‘Medieval’ in this context is used to denote the period between the end of the Roman era and about 1535 (commonly considered the start of the Early Modern period). Medieval is used in a generic sense when the exact point of origin or usage of a routeway is not known. At other times it may be possible to attribute a timeframe within the medieval period, such as Anglo-Saxon (pre-Conquest) or post-Conquest, if not an actual date.
  • 7. The literature on these two periods of road building is not extensive for the High Weald. Margary is still the standard work for the Roman period (Margary 1946, 29-63; 1948; 1973, fig. 1, 54), although he has been added to and corrected on individual routes. The most informative overview of Wealden turnpike roads is found in the reports of the Parliamentary Commissions and select committees between 1830 and 1852. Besides a number of recent publications dealing with the history of turnpikes in England and Wales, some routes in the High Weald have been documented in detail; for example, the local turnpikes in the Cranbrook area (Sanders 2004).
  • 8. Footpaths often have no identifiable landscape features associated with them, but rely on the ancient common right of pedestrians to cross a particular stretch of land, even across enclosed, private fields. In contrast to highways, footpaths appear to have received no maintenance in the past, and still rely on the common law obligation of landowners to keep them clear.
  • 9. This obligation was formalised in an Act of 1555: 2&3 Philip and Mary c.8.
  • 10. The 5m contour has been taken as the limit of inundation. This fits well with the evidence, such as that for the tidal and high water data, as well as the historical map evidence.
  • 11. In a recent statistical analysis of Early Saxon settlement in Norfolk this tendency was clearly established (Chester-Kadwell 2009, pp. 135-142). Settlement usually related to medieval settlement of Saxon origins in the High Weald appears to reflect this also.
  • 12. The issue of continuity between the Roman and medieval periods is very much a live debate (Rippon et al. 2015 contra Williamson 2016; Chester-Kadwell 2010).
  • 13. Being essential for the Royal administration, the owners of the great estates, the Church, City merchants and others.
  • 14. Other principal routes from London for the Continental crossing were along the North Downs to Canterbury and then on to Sandwich or Dover and another going via Maidstone and Ashford to Hythe (Lawson 2004, 51).
  • 15. Symonson’s map of Kent (1596) was printed by Sturt in the seventeenth century. [See article on Symonson’s cartographic career, pp. 149-163.]
  • 16. This route survives as far as Dingleden as a footpath, although there is evidence for an abandoned lane through Strawberry Wood where there is a culverted stone bridge over a gill (only recently discovered and restored). This structure is of a substantial construction, but makes sense as a survivor from an abandoned road. Latterly, a similar stone culvert bridging another stream has been discovered in woodland to the north of Benenden. This example seems to relate to another lost routeway that is associated with a green lane at Mount-Le-Hoe Farm.
  • 17. There is evidence for a lost way from Kensham Green across the Hexden Channel downstream from Hopemill Lane out to the Sandhurst to Newenden road, but more research is needed.
  • 18. For a flavour of the difficulties see the documentation referring to the rebuilding and maintenance of Newenden Bridge covering the period 1637 to 1749 (KHLC U386/02/3; Q/AB/42).
  • 19. During the Middle Ages the processes for keeping the roads useable seem to have worked reasonably well, grounded in the common law practices based on manorial obligations; the benefaction of the religious orders; and the setting of specific obligations on individuals or corporations. By the middle of the sixteenth century the state of the roads had become so bad that a new approach to the problem was needed. An Act of 1555 (2 & 3 Philip and Mary c.8) placed the responsibility for maintaining the roads upon the parish and the labour for the work upon the parishioners. This system formed the basis for road maintenance until the passing of the turnpike Acts, and for roads other than the turnpikes until the General Highways Act of 1835 (5 & 6 William IV c.50).
  • 20. The 1555 Act was a general Act covering the whole country. However, there had been earlier Acts specifically concerning roads in the Weald. Famously the Act of 1523 (14 & 15 Henry VIII, c.6) for altering the highways in the Weald of Kent was followed by a similar Act for Sussex in 26 Henry VIII. A further Act of 27 Elizabeth c.19 1584/5 levied a charge on the owners of ironworks in Kent, Sussex and Surrey for the moving of iron along the roads in the Weald.
  • 21. Although the first Act that impinged on the High Weald was for the road between Sevenoaks and Tunbridge Wells in 1709, the Ticehurst to Hastings road (1753) was the earliest turnpike in the eastern High Weald.
  • 22. The major exceptions to the rule were the upgraded Hastings to John’s Cross section of the battle route to London and the St Leonards to John’s Cross turnpike, both largely new roads (approved 1836). The last turnpike to be built in the study area was the Cripp’s Corner to Hawkhurst ‘junction road’, which was also a mainly new road (approved 1841).
  • 23. These land divisions were (from east to west) Maytham, Milcham (now only preserved in a field name), Forsham, and Cassingham (an early name for Kensham). Their names were derived from the Saxon form –hamme, meaning open (meadow) land near to water/marshland.
  • 24. 1767 was the date of the Act that allowed the Tenterden Trust to extend the road from Tenterden to Newenden using Hexden Bridge. However, it does not follow that the new works required were actually carried out in that year and it is likely that some were carried out at a much later date.
This paper explores the English People’s innate romanticism about what they perceive as their ancient, idiosyncratic and timeless rural roads. It examines the actual origins, development, and morphology of roads exploring the reality through a study of routeways in an area of the eastern High Weald that straddles the county boundary between Kent and East Sussex, centred on the upper levels of the River Rother.

It concludes that routeways (roads, tracks, paths, highways) in the High Weald have evolved over time driven by the socio-economic needs of the communities they serve. They are dynamic systems that need to change as economic and social needs change and, whether these changes are slow and piecemeal or rapid and sudden, the effect on the overall pattern may be considerable.
Subject area