Some Problems of the North Downs Trackway in Kent

SOME PROBLEMS OF THE NORTH DOWNS TRACKWAY IN KENT By REV. H. W. R. LILLIE, S.J., M.A.(Oxon.) THE importance of this pre-historic route from the Continent to the ancient habitat of man in Wiltshire has long been recognized. In the Surrey Archceological Collections of 1964 will be found an attempted re-appraisal of its route through the county of Surrey. Although the problems connected with its passage through Kent are fewer owing to its being better preserved, there are some points which I think still deserve attention—the three river crossings of the Darenth, the Medway and the Stour; the crossing of the Elham valley; and the passage to Canterbury of the branch route from Eastwell Park, laiown as the Pilgrims' Way. It may be worth while, before dealing with the actual crossings, to note a few general characteristics. Mr. I. D. Margary—our most eminent authority on ancient roads in Britain—has pointed out the dual nature of this trackway. It comprises a Ridgeway and a Terraceway. The first runs along the crest of the escarpment. The second runs parallel to it, usually at the point below the escarpment where the slope flattens out into cultivation. In Kent for the most part the Terraceway has survived more effectually than the Ridgeway. I t is for much of its length used as a modern road, marked by the familiar sign 'Pilgrims' Way'. Except at its eastern terminus the Ridgeway has not been so lucky, although it can be traced fairly accurately by those who take the trouble to do so. When these two paraUel tracks come to the crossing of a river, we find in Surrey that for a steep descent to the river-bed the two tracks tend to unite—as they do on both banks for the crossing of the Wey at Guildford and for the east-west crossing of the Mole down the steep edge of Boxhill. However, the approach to the Mole from the west being more gradual, both the Ridgeway and the Terraceway tend to keep their own individuality, (cf. SyA.C, lxi (1964), 26.) I. THE CROSSING OE THE RIVER DARENTH With the above in mind we can more easily understand the dual crossing of the River Dareth between Knockholt and Otford. The gradients on both sides of the river being gradual, the Ridgeway and the Terraceway wUl not unite but each will have its own crossing. 203 SOME PROBLEMS OE THE NORTH DOWNS TRACKWAY IN KENT The Ordance Survey maps all mark what is the accepted route of the Terraceway across the valley from the London-Tonbridge road (A21) to Otford. And the route of the Ridgeway crossing could run from Polhill down to the neighbourhood of Filston Farm and thence across the valley to climb the opposite ridge at Greenhill. However, it should be noted that the accepted route for the Terraceway leads it across a wet and sticky patch of gault. Perhaps both these routings for Terraceway and Ridgeway are later variations of earlier routes which seem to cross the valley at Shoreham. Here the width of the valley is less than half its width at Otford. There is no gault. Firm ground runs down to the river on both banks, and there are easy gradients both for the Terraceway and Ridgeway up from the river on either side. Moreover, though there is undoubtedly a track down the slope from the Ridgeway at Polhill, the natural trend of the Ridgeway seems to suggest a further run of two miles northwards along the spur, to descend by an old well-marked track to an ancient ford just north of Shoreham whence there is an easy ascent to Greenhill on the eastern side of the valley. If this more natural route for the Ridgeway crossing be accepted, it will be seen that the Terraceway's route will run a mile or so to the north of the present accepted crossing and use one of the fairly numerous fords a little to the south of Shoreham, or even at Shoreham itself. 2. OTFORD TO THE MEDWAY Between Otford and the crossing of the Medway a third route appears to the south of and running parallel to the Terraceway. Its line can be traced through Otford, Kemsing, Wrotham, Trottiscliffe, Coldrum, Birling Place, Paddlesworth and Snodland. I t is probably a much later development of the Terraceway, running, due to practical needs, closer to the line of human habitation. I t touches three ecclesiastical palaces of Otford, Trottiscliffe and Wrotham, so it could well have been something in the nature of a service road originating from Otford —the southern end of the Darenth valley—and leading eastwards to other ecclesiastical establishments, including Rochester and Canterbury. However, its marked ecclesiastical character should not exclude more secular uses which such a valuable road could serve. East of the Medway it seems to connect with an old road linking up with the modern A20 near Harrietsham and leading to the channel ports. I t should be noted here that in early times Otford, owing to its geographical position at the head of the Darenth vaUey, must have been a place of some importance. I t hes not only on the main trackways leading east and west, but on branch trackways leading to the north along both sides of the vaUey to the crossing of the Thames at Dartford. 204 SOME PROBLEMS OF THE NORTH DOWNS TRACKWAY IN KENT 3. THE MEDWAY CROSSING This part of the route has been the object of much discussion. Hilaire BeUoc, in his book The Old Road, believes Snodland to be the old crossing point. Mr. I. D. Margary, our modern authority, accepts Snodland as one of the possible crossing points, but himself prefers a more northerly point at Holborough where the ground is firmer and there are indications on both sides of the river of well-made tracks leading down from the hiUs. I t would seem that Snodland hes really to the south of the earher tracks of both Ridgeway and Terraceway. In fact, as we have observed in the preceding section, it seems to be sited on the more modern route directly connected with Otford, and parallel to the accepted trackways. As in the Darenth valley, it appears more probable that the earlier crossings of the Medway lay to the north of Snodland where the valley is narrower. Undoubtedly Holborough is a suitable passage point, but an even more likely spot seems to be Halling-Wouldham, where the crossing is narrower than that at Holborough and the river banks on either side equaUy firm. Both on the eastern and western sides the chalk stratum approaches right to embarkation point. Further, we must take into consideration the trend of the tracks on both sides of the river. On either side the Terraceway lies at the foot of the Downs which here sweep to the north along the Medway estuary. Reckoning from south to north, junction tracks from it to the river certainly run to Snodland, Holborough, Wouldham-HaUing, and further north to Cuxton, even to the crossing now bridged at Rochester. Possibly, on the western side, we can trace the Terraceway even further to the north to the Isle of Grain, suggesting a lead to Cliffe-at-Hoo and thence across the Thames into Essex. Above, along the crest of both escarpments, run the Ridgeways, paraUel to the line of the Terraceways below. But it wiU be noted that they do not appear to run so far north as the Terraceways. The Ridgeway on the scarp east of the river carries its northwards passage only as far as the point where the scarp peters out opposite Cuxton. Similarly, on the western side of the river, the Ridgeway does not seem to continue its northerly direction beyond Cuxton where the scarp also terminates in a vaUey. This crossing at Cuxton may appear unhkely because it hes near the point where the width of the river and of the vaUey begins to expand. But it was doubtless a crossing that could be made good use of at low tide. Perhaps traveUers on the Ridgeway would prefer the Halling-Wouldham crossing to which there is a strong track leading from the Ridgeways on both banks. The Terraceway crossing seems happily placed by Mr. Margary at Holborough, though I fancy he would like to see the Ridgeway cross here too. He speaks of a track leading down from the ridge on the west 205 SOME PROBLEMS OE THE NORTH DOWNS TRACKWAY IN KENT bank and of one on the east bank leading up by Scarborough Farm to the ridge. I can find no evidence for this latter track on the eastern side. On the whole I think the topographical evidence leads us to conclude that the Halling-Wouldham crossing suits the Ridgeway best, and that at Holborough the Terraceway. The crossing at Snodland, like the crossing at Otford, seems a much later variant and one too gaulty and too far south to be considered one of the earlier trackway crossings. 4. THE STOUR VALLEY CROSSING This is the widest valley crossing along the whole of the trackway in Kent. Its general features are similar to those of the Medway and the Darenth. From the Channel ports a Ridgeway and Terraceway come westwards along the Downs and follow the curve of the hills to a point east of Wye where the hills veer northwards. From the west come a similar Ridgeway and Terraceway which (according to Mr. Margary) unite at Dunn Street, near Westwell, throw off a branch towards Canterbury in Eastwell Park, and then proceed together by Boughton Lees to cross the Stour at Wye. Thence they proceed up the eastern slope to divide again a mile or so east of Wye at the foot of the Downs and at the point where the Ridgeway and Terraceway come to meet them from the east. Let us examine this. Undeniably there is an ancient crossing at Wye (a significant name) where the Stour is a sizeable river, though easily forded, and not tidal. But, as at Snodland and at Otford, there is a considerable amount of gault on this passage to the crossing, also the vaUey here is much wider than it is a mile or two further north near Godmersham and Chilham where the opposite sides of the valley come closer together. This suggests— as in the Medway and Darenth crossings—that there was (or were) more ancient crossings to the north of Wye. If we look at the course of the Ridgeway and Terraceway on the eastern bank of the river, though there is undoubtedly a passage (or passages) leading down from them to the river at Wye, yet these tracks seem of themselves to be pushing on quite purposely to the north; rather as though the passages down to Wye were an afterthought and not the original track. The Ridgeway, owing to considerable indentations in the escarpment, proceeds a little away from the crest to a point beyond Godmersham, just north of Pope's Farm, where there is a weU-marked descent to an old ford adjacent to a prehistoric earthwork. The crossing leads immediately to a spur carrying a road leading up to the ridge above the west bank of the river. At a point just outside the fence of Chilham Park the traveller finds himself on the Ridgeway-track usuaUy known as 'the Pigrims' Way' which wih take him along the ridge, either to the east or to the west. 206 SOME PROBLEMS OF THE NORTH DOWNS TRACKWAY IN KENT The Terraceway appears to cross the Stour at Godmersham where it joins the Terraceway on the opposite bank. Mr. I. D. Margary, in his book Roman Ways in the Weald (1948), 253-256, neatly describes this passage, though perhaps in the context he is more expressly concerned with its Roman rather than with its prehistoric aspect. The suitability of these crossings will be apparent if we note how closely at these points the opposite sides of the valley approach together, less than a mile apart and with firm ground on either side of the river, and I submit that they are clearly more ancient crossings than that at Wye. These conclusions will, I think, be confirmed if we examine the course of the Ridgeway and Terraceway on the western bank of the river. Take the Ridgeway first. Mr. Margary thinks that from Charing HiU eastwards it proceeds through Longbeech Wood to join the Terraceway at Dunn Street whence it continues with it through EastweU Park to the crossing at Wye. Certainly there is a downhill track to Dunn Street, but is it more than a link-track from Ridgeway to Terraceway for traveUers along the ridge who want to transfer to the Terraceway? If the Ridgeway comes as far as Charing Hill, there is no reason why it should not continue further. For the ridge of hiUs itself certainly continues further. And so it would seem that in fact it not only ought to do so, but actually does so. It continues along the line of the modern crest-road from Charing HiU to Challock Lees, though a little way off the crest to circumvent the considerable indentations of the escarpment in EastweU Park. At Challock it turns slightly south-east to join the modern road from Faversham to Boughton Lees and Ashford. When it reaches the descent to Boughton Lees, it veers to the left so that it can remain on the crest. This brings it to Soakham Down whence it continues along the ridge in Godmersham Park till it reaches the spur on which stands the Castle and village of Chilham. At the fence of Chilham Park it turns downhill to the crossing of the Stour by the ford we have mentioned above. From Soakham Down to Chilham the track is on the hne marked in the Ordnance Survey as 'The Pilgrims' Way'. Also it is worth noting that at Soakham Down there is another branch track on the hne of a modern road, down White Hill to Boughton Corner and thence on to the river-crossing at Wye. This is a much firmer approach to the river than that from Boughton Lees. There is no gault, and the chalk spur of White Hill gives a firmer approach on chaUc to the river. It seems that this is perhaps a more ancient route to the crossing than that from Boughton Lees. It now remains for us to diagnose the passage of the Terraceway from the point where it enters EastweU Park at Dunn Street to the crossing at Godmersham. 207 SOME PROBLEMS OF THE NORTH DOWNS TRACKWAY IN KENT From Dunn Street the track runs through the Park to the ridge on which the Home Farm stands, thus avoiding the marshy ground near the lake to the south of the Park. I t proceeds along the ridge for a quarter of a mile, then veers north-east keeping to the north side of the mansion till it coincides with the modern drive leading from the mansion to the north-eastern exit from the Park. Here it crosses the Ashford-Faversham road (A251) and proceeds through a patch of scrub to cross the smaU vaUey below Warren Farm. At White Hill it crosses the road from ChaUock to Boughton Corner and continues by a weU-marked passage to Soakham Farm. Thence, keeping its northeasterly direction, first by a short road, then by a line visible on the turf, it reaches a point below Godmersham Spur where it merges into the old main road to Godmersham, passing the gates of Godmersham Park. I t then crosses the Stour by the old ford to meet the Terraceway coming from the south-east. It wih be noticed that we have ignored the route marked in the Ordnance maps, leading from Boughton Lees by Boughton Aluph church across to Soakham Down. This could be a later variation of the old trackways. It may be a church-goer's short cut, or a medieval pilgrim's short cut, but it is quite out of character with the undulating habits of a prehistoric terraceway. Between Soakham Down and Soakham Farm lies a fairly sharp escarpment bearing the remains of an ancient track. This is, I think, a link-way between the Ridgeway on Soakham Down and the Terraceway running below. 5. THE ELHAM VALLEY CROSSING Both Ridgeway and Terraceway foUow their usual course along Wye Downs south-eastwards till they reach Stowting Combe. This is a large indent, a mile or so wide. The Terraceway can be easily foUowed across the mouth of the combe, round the spur and across the Roman road, Stone Street, and on to Posthng. The course of the Ridgeway proceeds along the crest of the escarpment, which is here so irregular that we must expect it to keep weU away from the actual scarp. Mr. Margary traces its course round the combe and brings it down to Etchinghill there to cross the Elham vaUey, then up on to the ridge overlooking Folkestone and eventuaUy reaching the heights above the Folkestone-Dover main road, passing through Paddlesworth and along a modernized road by the name of Creteway. For various reasons it seems to me more probable that the older Ridgeway route runs through Lyminge rather than through EtohinghiU, whioh—as I hope shortly to show—lies on the Terraceway, not on the Ridgeway. The more characteristic route for the Ridgeway—and 208 SOME PROBLEMS OF THE NORTH DOWNS TRACKWAY IN KENT traceable on the Ordnance maps—lies round the top of the combe through Stowting Common where it turns east across the 600-ft. contour, crosses Stone Street then on by a well-marked track to Rhodes Minnis. Thence it proceeds to cross the Elham valley at Ottinge and by Great Shuttlesfield Farm makes its way to Paddlesworth. There is I think an alternative route from Stowting Common, running a little nearer the escarpment to cross Stone Street near Hemsted; thence on to Woodland, Lyminge, up on to the ridge again by Eastbrook Lodge, and thence along to Paddlesworth. This is about half a mile shorter than the route by Ottinge, but steeper. I must confess to being a little uncertain about Paddlesworth as a point on the route. Paddlesworth is not on chalk, but on a patch of gravel. It would appear more probable that the earher Ridgeway passage kept along the crest, near the escarpment, when it reached the ridge east of Ottinge or Lyminge. Indeed this actual track is wellmarked on the map. Thus it would pass by Cheriton Down, the Norman Castle, behind Sugar-Loaf, along Creteway to the spot where now stands 'The Vahant Sailor'. It now remains to chart the passage of the Terraceway. East of Postling I think it would be an error to bring it round the south side of Tolford Hill and the adjacent Summerhouse HiU. From there the line by Frogholt and Pean lies across a fairly extensive patch of gault. Clearly the route lies to the north of this group of hills, by-passing, as it does at Reigate in Surrey and elsewhere, the projecting spurs. From Posthng the correct route lies eastwards, not southwards, through Staple Farm, then round the spur to EtchinghiU. Here it crosses the valley, which at this point is at its narrowest, and turns south-east along the foot of the escarpment more or less on the old track of the Elham VaUey railway. At Pean (where there is a link road to the Ridgeway) it turns east to Danton Pinch, then wends its way, nestling close to the escarpment for there is much gault here, to its crossing over the shoulder of the Folkestone-Canterbury road. Then it proceeds along the course of a modern road, drops into East Wear Bay and continues near the Folkestone-Dover main line to the foot of the cliffs at Dover, where it unites with the terminus of the Ridgeway. The passage from Folkestone to Dover has probably been obscured through the oenturies by sea-terosion. Perhaps the original route terminated at East Wear Bay. Some doubt has been expressed about the existence of the Terraceway eastwards from the Elham valley. But it is plain for aU to see, and can be easily descried from the railway carriage window between West Folkestone and FoUcestone Central stations! 209 SOME PROBLEMS OF THE NORTH DOWNS TRACKWAY IN KENT 6. 'THE PILGRIMS' WAY' EROM CHILHAM TO CANTERBURY This branch of the main North Downs Trackway has been known by this familiar title for a good many years. Whether pilgrims ever traversed it or not to Canterbury does not concern us here. I t is much older than even the beginning of the Christian era. Tradition has always represented it as a single track running eastwards from Chilham to Canterbury, entering the city by Watling Street (London Road) just to the east of the village of Harbledown. But, from what we have seen in section 4 above deahng with the crossing of the Stour, it was apparent that though the main east-west trackways crossed the river there, the Ridgeway and Terraceway on both sides of the valley actually continued north-eastwards along the river valley. So if any prehistoric traveller wished to continue his path in that direction, there were four paths for him to choose from—a Ridgeway and Terraceway on either bank—not just the one path known as the Pilgrims' Way. The two routes on the south bank of the river move along in their customary positions, parallel to one another, to debouch into the city at the western end of Wincheap. The Ridgeway runs by Chartham Down, the Terraceway through Chartham itself. There are also two tracks along the northern bank of the river. The Terraceway proceeds eastwards from the point near Godmersham where the main trackway crosses the Stour (cf. above, section 4) reaching Chilham by a route shown plainly in the older Ordnance maps through Godmersham Park to Mountain Street. Thence, passing round the north side of the lake (now enclosed in the grounds of Chilham Castle) it emerges at the Woolpack Inn. Thence it proceeds along the line of the Ashford-Canterbury road, diverging from it where the modern road turns south to cross Milton Bridge. I t holds its course along the north bank of the river through Tonford, reaching the city at Whitehall. I t passes onwards by North Lane, Broadoak Road and Sturry to the coast, or perhaps to terminate at Fordwich, the ancient port of Canterbury. The Ridgeway presents some difficulties, famihar enough to readers of Belloc and those who have written about this part of the track. We are now on the Pilgrims' Way. We have aheady traced the route through Chilham Park to the viUage of Chilham. How does it proceed from this point? The Ordnance maps trace it as descending Chilham Spur to the north of the church, crossing the Canterbury-Maidstone road and proceeding up the opposite slope to the village of Old Wives Lees. Then, after turning east for about half a mile, it returns southwards to the Ashford-Canterbury road. This apparently fatuous route has attracted a good deal of ridicule. Among the more constructive criticisms is the suggestion that instead 210 SOME PROBLEMS OF THE NORTH DOWNS TRACKWAY IN KENT of turning south after passing through Old Wives Lees the road continues eastwards down into a vaUey, up the other side, then down again to use a crossing stile over the railway, then past Nickle Farm to join what is thought to be a part of the old track leading up to Chartham Hatch. Let us examine aU this. First of all, I do not think the Ordnance maps can be entirely wrong, once it is grasped that this track is a Ridgeway. Chilham stands on a spur which commands the entry to two valleys, one leading north-west, the other south-west along the Stour vaUey. From early times it has been regarded strategically as a strong military point. If the Ridgeway passes through this strong point, to regain its ridgeway position it must first descend into the valley and then ascend the massif on the other side. It would in fact be expected to do this because it must maintain its character as a Ridgeway. So I think the Ordance maps are right in as far as they trace its route up to Old Wives Lees. Where does it go from this point? The suggestion that it turns back to the road it has just crossed does indeed appear ludicrous. Nor does the route eastwards across to Nickle Farm really commend itself. It is too undulating and out of character with the usual habits of the Ridgeway. I think the solution lies in our remembering that at Old Wives Lees the Ridgeway has not yet attained its usual position on the summit of the escarpment. The village stands on a spur projecting southwards from a higher ridge running east to west, about a mile to the north. Again, this spur has quite -a substantial valley on its eastern side, running up from Shalmesford as far as Upper Ensing. In order to avoid this valley (and to keep its character as a ridgeway) its route would have to go northwards from Old Wives Lees to Upper Ensing, crossing to the higher ground near Selling railway tunnel, then turning eastwards through Joan Beech Wood and Bower Wood north of Nickle Farm till it reaches Hatch House, thence to proceed to Bigbury by the familiar route through Howfield Wood. That this suggested passage is correct can be realized perhaps more convincingly if one travels westwards from Bigbury. Except as a hnk to the Terraceway down below, there seems no reason to take the route down the hiU from Hatch House to Nickle Farm. It is obvious that traveUers wanting to keep to the Ridgeway would not turn south at Hatch House, but keep straight on in the same westerly direction as the track has already been leading them. It is true that this route hes over gravel, but this is the quickest way to the chalk spur at Old Wives Lees, and in fact the track has been travelling on gravel since it left Canterbury. Bigbury Camp is itself on gravel, not on chalk. To reach most advantageously a solid 211 19 SOME PROBLEMS OF THE NORTH DOWNS TRACKWAY IN KENT chalk underfoot this route I think must be accepted as correct. Thus I venture to suggest that we have solved the age-long problem of the missing track between Old Wives Lees and Nickle Farm which puzzled Hilaire BeUoc and his successors. I do not think that the passage of these trackways through the Stour valley should be treated too lightly, as if they were unimportant off-shoots from the main route to Dover. Apart from the Thames estuary, the Stour estuary is the most comfortable approach for shipping from the Continent wishing to land in south-east England. The distance from Dover, Folkestone and Hythe to their opposite ports on the French coast is shorter than the crossing to the Stour estuary. But Dover, Folkestone—and to a certain extent Hythe—were artificial harbours. The Stour estuary was the only real natural harbour in this area, and I feel sure was much used by shipping in ancient times, even before the time of Julius Caesar and St. Augustine. Consequently the tracks along the Stour vaUey must have been of great importance, even perhaps of greater importance to early man than that part of the main trackway from Dover to Wye along the southern face of the North Downs. If we were to chart the whole network of prehistoric tracks from ports in south-east Kent, connecting up with the main trackway to the west of England, I think it would be clear that the routes along the Stour valley were of primary importance. 212