The research potential of Romney Marsh

Romney Marsh affords a unique opportunity for the study of rapid environmental changes throughout the past 10,000 years. In no other marshland of comparable size can the combined effects of changing sea level and the growth and decay of shingle barrier beaches be so effectively related to successive patterns of human settlement, transportation and land-use. The potential for studying a highly dynamic environment and man's response to it in historic and prehistoric times is outstanding. From both palaeogeographical and historical points of view, Romney Marsh needs to be studied in its regional as well as in its local context.

When the Romney Marsh Research Group was founded in 1984 there was no consolidated body of information on the evolution and history of the area, apart from that contained in the Soil Survey Memoir. Popular opinion is still often mis-informed on a number of major questions of palaeogeography and local history, and many general accounts are still based on assumptions of Camden (1586) and the findings and suggestions of Dugdale (1662), built upon by Victorian historians and engineers. The research published in this volume begins to rectify this situation and illustrates the potential for further research.

The shingle barriers of Dungeness provide an unequalled opportunity to determine the incidence and pattern of storms in this part of the English Channel during the past 3,500 years at least. Both in the Dungeness Foreland and in the area of Rye Harbour, investigation is needed of apparent abrupt changes in the levels of the shingle ridges, the changes in the orientation of the ridges and the origin of the alluvial areas separating the main groups of ridges. Radio-carbon dates are needed for organic sediments in the intervening alluvial tracts, in order to establish dates by which the respective groups of ridges had been emplaced. There is as yet no information by which to date the presumed breach in the shingle at New Romney, nor analysis of deposits which may be related to the thirteenth-century disruption of the Rye Bay barrier and the resulting loss of Old Winchelsea. Past erosion of the barrier beaches led to the construction of sea-walls to defend land already occupied: historical records of these remain as yet untapped. Present movement of the shingle is not only threatening to undermine and isolate the Dungeness nuclear power stations, but regularly exposes sea-walls to marine attack, and is leading to unacceptable flooding on MOD property at Lydd and Hythe.

The area has a significant contribution to make in resolving the problems of the pattern of land uplift and subsidence in the UK over the past few millenia. New data from the Romney Marsh area will lead to new insights which will have important implications for the present and the future management of the coastline. A scale of positive and negative tendencies of sea-level movement is therefore much needed, and it will then be important to distinguish regional from purely local factors.

Romney Marsh also has a contribution to make in understanding the evolution of the English Channel. Comparisons can be made between the sedimentary history and landforms of this area and those of the other marshland areas on the south and east coast of England and on the Continent. The value of studies of the sediments, the Foraminifera and the history of the vegetation and landforms has been demonstrated in this volume. In the valleys which drain towards Romney Marsh an early sequence of sand, clay and peat has been established, similar to that recognised by the Soil Survey on the marsh itself. This sequence indicates a shallowing of water and a decrease in environmental energy, in the period following the rapid Post-Glacial rise in sea level. In addition, the study of Foraminifera has proved, most importantly, that marine sedimentation extended as far upstream as Brede Bridge, confirming the pattern already established elsewhere in Britain of a contri- bution of marine sediment to tidal flats and estuaries. Preliminary work appears also to confirm the Soil Survey finding of a single major peat bed, with minor localised peats lower in the valley successions. A reliable chronology needs to be established by dating the organic horizons - while lithostratigraphic and biostratigraphic studies can be extended, to clarify palaeoenvironmental changes. Evidence of prehistoric human activity, including settlement and deforestation of the surround- ing upland, and its impact on the landscape, can be expected in these marshland peats and inorganic sediments.

All these studies should now be extended as widely as possible within the marshland area. The important distinction between the basal sands, the elevated sands at Midley and other sand deposits at Broomhill and Dymchurch (all termed Midley Sands by the Soil Survey) needs to be investigated.

The Marsh faces the Continent near the shortest Channel crossing. It has therefore always been in a position of national importance for communication, 176 Jill Eddison, Christopher Green and Andrew Woodcock trade and defence, as well as astride important arteries leading to the hinterland of the Weald. The marsh inlets, now silted, gave shelter to cross-Channel and up- Channel traffic. One of the Roman forts of the Saxon Shore was Stutfall, below Lympne; two of the all- important medieval Cinque Ports, Romney and Hythe, and the two 'ancient towns', Winchelsea and Rye, which were added to the Cinque Ports (significantly at the end of the twelfth century), were in the marsh area. Archaeologically and historically, therefore, there is a close relationship between the Marsh, the Channel and the hinterland. There is also considerable potential for study of maritime history and nautical archaeology.

Evidence has been presented of mesolithic occup- ation, at least in the river valleys. It has been suggested that the evidence of pollen analysis points to occupation and disturbance in the Neolithic. The recent discovery of Bronze Age axes near Lydd proves occupation on the outer fringe of the marsh some 3,700 years ago. The Roman presence on the marsh was almost undoubtedly greater than that indicated by present evidence, which is concentrated on the barrier beaches and at Stutfall, but otherwise limited to chance finds by observant farmers. A programme of intensive field-walking is needed to add to this evidence.

Undoubtedly an exceptional opportunity for interdisciplinary studies, incorporating geology, archaeology of settlements and sea defence works, documentary records and topography, lies in the medieval period. The extraordinary storminess of the middle to late 13th century, and the major disruption of'the shingle barrier across Rye Bay enabled marine conditions to encroach on reclaimed land as far north as Appledore, Fairfield and Midley, and allowed the tides to flow up the Rother valley beyond Newenden. Protection of occupied lands and subsequent land reclamation over the next four centuries gave rise not only to a complex of defensive walls and then to the largest sequence of medieval innings in the country, but also to an immense body of legislation, including the Laws and Customs of Romney Marsh which became the "pattern and exemplar" for drainage laws throughout the country, particularly in the Fenland and the Somerset Levels.

A vast, and almost untapped, reservoir of document- ation concerned with land-use, administration, drainage and sea-defence is to be found in the archives of the ecclesiastical land-owners, principally the Archbishop of Canterbury, the Dean and Chapter of Christ Church, Robertsbridge Abbey, Battle Abbey, and Bilsington Priory; of the lay land-owners, including All Souls and Magdalen Colleges, Oxford; and in national records.

The features of the landscape include the sites of numerous churches, some now in ruins (and ripe for archaeological investigation), and others deteriorating fast and in urgent need of recording; early sea-walls, fleets and creek beds, and particular patterns of drainage ditches. Most unfortunately a large amount of this valuable topographical evidence has been removed in recent years as a result of the change from sheep to large- scale arable farming. There is therefore an urgent need to record and, where possible, preserve what remains.

The excavation and documentary evaluation of Broomhill described here is an admirable illustration of the potential. In 1987 the continuation of that work was supported by a detailed sedimentological study.

The papers presented in this volume make a start, but only scratch at the surface of the potential for research. Major questions remain to be answered. The early history, between 9000 and 3500 BP should be sought in the record of the buried shingle and the other sediments. The two hypothetical models presented here, of the Roman period and the l l th-13th centuries, need to be tested. The Rhee Wall holds the key to many problems: who built it, when, and for what purpose? The courses of the Rother before 1250 remain unknown and all the various suggested courses need careful examination: that is, the Sedbrook Sewer across the north of the Marsh; the 'natural' watercourse running north of the Rhee Wall; and the Wainway channel and its postulated extension north-eastwards to the Romney estuary.

The Romney Marsh Research Trust has been set up to support the work of the Research Group, whose aim is to encourage research into the evolution, occupation and reclamation of the area. Priority will be given where possible to areas which seem likely to resolve the more important problems, and to sites in danger ofdestruction by, for example, change of agricultural use, quarrying and development. The Trust intends to ensure that each site is fully investigated in every relevant discipline and that the results are co-ordinated. Results will then be published, and made available to the general public.