Attempts to clear the Rother Channel, 1613-1624

The beginning of the seventeenth century was a critical time for the drainage of the Rother Levels (Fig. 12.1). Three thousand acres (1200ha) in the valley were described as "deeply drowned lands" (perennially under water) and a further two thousand acres (800ha) were "summer lands7' (useful only in summer) and even these were under threat of further inundation. Between 1613 and 1624 major attempts were made to improve the situation, involving very high expenditure. This paper describes the work undertaken, the expenditure laid out, and the provision of the necessary finance. It aims to illustrate the problems of river management and of the attempted solutions. Mention is made of the sources and transportation of materials used in the works. Finally, in view of the failure of all the schemes the decision to undertake the work in the first place is questioned. The Sources The principal source is a book of accounts for the "Intended works upon the River of Rother or Channel of Appledore" recently deposited in the East Sussex Record Office'. This book consists mainly of long lists of items of expenditure kept by the expenditors and clerks to the Commission of Sewers, the drainage authority, each list being known as an account. There is no regularity in the period covered by each account, although the duration of exactly one year from 1 March 1623124 for the last account, combined with assessment by a committee which met on similar dates in successive years (30 May 1623 and 22 May 1624) suggests evolution towards an annual routine. The content of the individual entries is inconsistent, and the degree of completeness of the different accounts is irregular. One long account is missing altogether. Details are given in Table 12.1. The book also contains numerous useful short calculations, notes about income, details of payments of rent, and a summary entry recording payment of £270 for timber. Of particular interest are seven sides concerning an inquiry set up in September 1619. The questions and answers in this have been of great value not only in clarifying details of income and expenditure up to that date, but also in throwing light on some of the technical issues. Apart from occasional references to the outfall of the Five Waterings Sewer in the Decrees of Romney Marsh proper2, the only material found to amplify the information in the Account Book before March 1622123 comes from second-hand accounts dating from 1659-63 (Powell1659, 1662; Harlackenden 1663). Since these publications resulted from one of the many disputes in the Levels, there is a strong suspicion that the evidence on both sides may be biased. In March 1622123 the sequence of surviving Decree Books of the Rother Levels begins, and the combination of this with the Account Book provides much fuller information than is available before that date3.4. In addition, an untitled map of this period, although undated, has thrown more light than would otherwise have been possible on the location of various features in the channel and estuary5. A facsimile of part of this map is reproduced in Figure 12.5. The Problems The Rother reaches marsh level at Udiam, 1.5km west of Bodiam (Fig. 12.1). From that point, some 20km up the valley from Appledore, the river bed was (as it still is) below High Water Neap Tides (OD Newlyn), which meant that the outflow of the river was stopped for a significant length of time at every high tide. Every incoming tide tended to deposit sand and silt in the channel and on the adjacent marshlands, especially in the reaches downstream from Maytham. In this way the marshlands at the seaward end of the valley became * The dates for this period were Old-style, i.e. the year began on 25th March. In this paper all dates between January 1st and 24th March are given as, for example, 1st March 1623124. Explanation of terms not now in common use is given in the glossary on page 162. In quoting from original sources the spelling has been modernised throughout. Attempts to clear the Rother Channel, 1613-1624 Sea and rlver walls Fig. 12.1. The Rother Levels, with approximate extent of Appledore Water c. 1600. raised above the level of those further inland. The ebb tide would have helped to remove sediment deposited in the channel, but as the marshlands in the valley were progressively reclaimed and marine influence retreated, that vital tidal ebb diminished. Great seasonal variations in the freshwater flow of the river caused further problems. In winter, swollen by heavy rain or melted snow, it helped to scour silt from the channel and sluices, although at the same time flooding the marshes. But in summer the flow was often minimal and afforded no such help, and the balance was tilted in favour of further silting. There were thus three main causes of the flooding - the flatness of the valley, the deposition of sand and silt at the seaward end, and the seasonal variations in the flow of the river and therefore of its scouring effect. In these complex circumstances the Rother channel had two vital functions. It served as the principle watercourse conveying water from the uplands to the sea. It also functioned as a reservoir to contain that water at such times as it was held back by the tide. In 1613 neither of these functions was being adequately fulfilled. L Past History A summary of the medieval history of the drainage of this valley has been published previously (Eddison 1985). By 149 NEY MARSH the middle years of the sixteenth century the valley was suffering serious freshwater flooding, which may have resulted partly from reclamation and embanking of various parts of the valley (Hipkin this volume, 138). In 1562 it was reported that at least 6,000 acres (2,400ha) from Reading up to Robertsbridge were flooded every winter, and great floods were lying on both sides of the channel. Some of the marshes near Reading stood above the water in the summer, but were so "moist and watery" as to be of little benefit to their owners. It was said that in times past the channel between Reading and Rye had been 200 to 300 feet (60 to 90m) wide, but by 1562 it was reduced to 16, 20 or 24 feet (5, 6 or 7m). "The fresh (water) had neither time nor term to issue as in times past". The problems were attributed to silting of the channel between Reading and Rye, and to the straightness of that part of the channel. It was also noted that the channel was "swerved to a higher level at the seaward end of the valley than the fresh water standing further inland6. A petition by the mayor of Rye concerning the decayed condition of that harbour in c. 1573 referred to standing water at Newenden and to watercourses which were so overgrown with "slub, bull-rushes, reed and suchlike" that the water could not flow away. They needed to be scoured from Newenden to Appledore and on to the sea7. Another document refers to a plan to divert the river as early as c. 1577 through Wittersham Level, a shorter route Jill Eddison Table 12.1. Details of the accounts contained in the Account Book, ESRO A2806 1/4/1 Folio 4 l5 76V 24 25 26 ref. on 37 39 47 57 Accountant R. Sheppard Expenditor R. Sheppard Expenditor apparently R.Shepherd l entry T.Leech Clerk R. Sheppard Expenditor R.Sheppard Expenditor R. Sheppard Expenditor Alex.Fowle Expenditor Alex. Fowle Expenditor Type of Expenditure Admin. Work Work in Back Sewer Timber from Wigsell Work and rents General Missing General General General Years 1613-1617 1614-1617 1614-1617 pre Aug. 1617 1616 Some 1617, mostly 1619-20 1617-1620 1620 1620-1623 1623-1624 Amount £ 29 1 2074 22 1 270 200 1123 or 1048 1958 158 1513 3795 Complete Yes Yes Yes Yes Yes No Missing Yes Yes Yes Signed Yes Yes Yes No Yes No No No Yes No Date 22 Sept. 1617 22 Sept. 1617 22 Sept.1617 before Sept. 1617 27 Nov. 1620 26 May 1621 30 May 1623 22 May 1624 Attempts to clear the Rother Channel, 1613-1624 to the sea, but evidently this came to nothing8. The map of John Stoneham of 1599 shows graphically, if cartographically somewhat inaccurately, the valleys as "Drowned Lanes" (sic)Y. From all these primary sources it appears that freshwater drowning was prevalent in the valleys at least fifty years before 1613. A more detailed -and somewhat different - account is given at second-hand in a publication by Powell in 1662, though this is likely to be biased because of his interest in the Upper Levels. According to him, the Newenden Levels were in relatively good condition at the end of Henry VIII's reign, when neither East Maytham nor any of the Levels downstream from it were reclaimed. Subsequent progressive inning in a seaward direction of Dumborne, Farnhill and Kenchill Levels and of salt marshes in Appledore and Stone resulted in all the levels upsteam suffering drainage problems in turn, which he described 151 as a 'progressive gangrene" engulfing the valley. By 1602 a Back Sewer had been cut from Acre Brook (a place not identified) to Reading Ferry, to improve the drainage of East and West Maytham, Lossenham and Farnhill. In order to attempt to solve the problem and achieve the economic advantages which would follow from draining the Levels effectively, it was necessary that all the land in the valley should be administered by the same drainage commission. This was achieved by a decree of 1st April 1609, which united all the deeply drowned lands (previously known as the Newenden Levels) in one commission of sewers with the summer lands i.e. the Levels of Ebony which, according to Powell, included Stone, and Shirley Moor which included Redhill and Appledore (Fig. 12.2). In due course this united commission became known as the Commission for the - Embankments d~v~d~ng 0 1 2 3km - ---- Embankments dividing the Levels (doubtful) Fig. 12.2. The River of Rother or Channell of Appledore and major tributaries, showing the private levels amalgamated in 1609 to form the Upper Levels. Tributary channels: a. Kent Ditch b. Hexden Channel c. Newmill Channel. Note: The Hexden Channel follows the line shown on ESRO A2806/1/9/2, which is straighter than the present channel. Deeply Drowned Levels: l. Bodiam 2. Ewhurst 3. Dixter 13. Farnhill 14. Peening 4. Northiam 5. Knelle 6. Mensham 15. Dumborne 16. Kenchill 7. Ethnam 8. Hemden 9. Newenden Common Brook Summer Lands: 10. Lossenham 11. West Maytham 12. East Maytham 17. Ebony (including Stone) 18. Shirley Moor (including Redhill and Appledore) Based on O.S. 1:25000 First Series, with information from ESRO A2806/1/9/2 and ESRO A2806/1/9/6 152 Jill Eddison Upper Levels. This unification may have introduced further -political - problems, since the interests of the owners of the Newenden Levels were by no means the same as those of the owners of the summer lands (see Hipkin this volume, 139). It was thus this administrative device rather than an abrupt physical crisis which sparked off the massive programme of work in 1613. The Income The work was administered by the new commission of sewers, and the finance came entirely from local sources, i.e. from the owners and occupiers of the marshlands concerned. All the commissioners and the members of committees whom it has been possible to trace were either owners of the estates along the sides of the valley, owners of marshland in the valleys, or long-time residents of Tenterden or Hawkhurst, who probably anyway had interests in marshland in the valleys. In spite of the implications of the tidal flow in the Rother for the harbour of Rye, which was of national importance and was failing critically in the second half of the sixteenth century, neither the Commission for Rye Harbour nor the central government made any contribution to this work on the river channel (see Hipkin this volume, 143). Consultations on the subject of navigation did, however, take place with both the commissioners for the Harbour of Rye and with the Lord Warden of the Cinque Ports. Contributions towards the work were made in the long-established form of water-scots. These were raised by proclamation at thirteen churcheslO. Throughout this period each channel scot was paid at the rate of 12d. per acre for the drowned lands and 4d. for the summer lands, presumably on the understanding that the drowned lands, although earning less at the time, stood to benefit more if the works were successful. The number of channel scots decreed depended on requirements, and the fact that the commissioners who decreed them were themselves going to pay out these sums implies that a check was kept on expenditure. A second form of tax, the indraught scot, was instituted in August 1615 and was to be paid regularly twice a year at the rate of 6d. for the drowned lands and 2d. for the summer lands. Although this tax was ostensibly to pay for the costs of the indraughts (see below, this page and 157), the income seems to bear little relationship to the capital costs of setting them up or to the rents paid for the lands used for them. For example, six months' rent paid in March 1621122 amounted to £18 8s.4d. for Ebony and £21 8s.2d. for Peening, while the scot for the same period brought in £79 l s.8d. The indraught scot was nevertheless a valuable steady source of income from May 1616 until the end of the period under consideration. In the eleven years covered by the account book 61 channel scots were levied, raising a total of £9,781 6s.8d. In the same period 17 indraught scots were levied, producing a total of £1,362 3s.ld. A grand total of £1 1,143 12s.9d. was thus raised in scots, as shown in Table 12.2. It is interesting to note that farmers with leases for ten years or longer were expected to pay their own scots. Where the tenants had leases shorter than ten years, the owners were expected to contribute two-thirds of the scots (Powell, 1659, quoting a decree of 1st October1 l Jas. I, i.e. 1613). The Operations Until the beginning of 1623 all the operations took place upstream from Appledore, within the confines of the valley (Figs. 12.2 and 12.3). In 1623 work in that part of the channel was virtually abandoned, in favour of massive activity further downstream (Fig. 12.4). At a session of sewers held on 1st October 1613 it was decreed that 20 channel scots should be paid towards the work, indicating that several years' work was anticipated. These scots were collected at intervals between November 1613 and September 1617. In those four years the whole length of the channel was widened and deepened systematically, one reach after another (Figs. 12.2 and 12.3). Harlackenden (1663) stated that at this stage the channel from Bodiam to Reading was to be 40 feet (12m) wide; from Reading to Redhill, 50 feet (15m); and from Redhill to Oxney Ferry, 60 feet (18m), though it should be noted that he was writing fifty years after the event. Information about the depth of the channel is scarce, but in 1615 it was to be seven feet (2.lm) deep between Lossenham Gutt and Lords Gutt -though from what level within the channel this measurement was to be taken is uncertain. Work in the channel was made possible by the construction of pairs of temporary dams across the channel, usually made of earth, occasionally reinforced with boards. The river was then diverted either into adjacent low-lying marshes, or into the Back Sewer (see below, 153) or on one occasion through Wittersham Level. (The last course is somewhat surprising since Wittersham Level, which lay south of the Isle of Oxney, had wisely chosen to maintain its independence and remain under its own separate commission, see Rendel1962 and Eddison 1988). Supplementary work included maintaining the tow- paths and their bridges, strengthening the banks, weeding and removing slid-butts (slumped material detached from the banks along concave lines of weakness) from the channels. Removal of vegetation and slidbutts was of prime importance because both of these would have encouraged silting. Simultaneously two indraughts were set up, at Ebony and Peening (Fig. 12.3). These were areas of marshland adjacent to the river into which sea water was admitted through a sluice at high tide, and probably also fresh Attempts to clear the Rother Channel, 1613-1624 Fig. 12.3. River Rother, Lossenham Gutt to Appledore; Channels, Gutts, Indraughts and Ferries, 1613-1621. a. Hexden Channel, b. Newmill Channel, c. Tenterden Sewer, d. Cradlebridge Sewer. Based on O.S. 1:25000 First Series, with information from ESRO A2806/1/9/2. water on occasions. The water was then released when the tide was well down, when it was hoped that it would have sufficient force to scour away the silt deposited in the channel by recent tides. The scouring process was assisted by the use of iron rakes and harrows: the accounts record provision of such implements and replacement parts. Ebony Indraught was located between the conical Isle of Ebony and the upland to the west of it and probably filled that embayment".12. An earth wall 40 rods (200m) long and armed with faggots and needles was constructed to contain the water, the work involving the use of hired carts and oxen. Damages had to be paid for trampling of land by the oxen and by workmen carrying in the arming materials, and for admitting the channel water. This indraught was brought into operation in the winter of 16 15, and rent for the land occupied by it began to be paid on a regular basis at Michaelmas 16 16. At Peening the situation seems to have been rather different, because there is no record of any embanking. The whole of Peening Level was (and still is) particularly low-lying, and there already existed Peening Flote, well- known as an area apparently permanently nnder water. In the summer of 1614, without any record of preliminary work, salt and fresh water were admitted to this Level. The first recorded rent was paid for the land occupied by this indraught on 25th March 1616. Two very small adjacent fields, shown on Figure 12.4, are named the Inner and the Outer Indraughts on the Ebony Tithe Map. In the same period maintenance work and alterations 153 -Rother channel and major tributaries - Relief channels were carried out in the Back Sewer (also known as the New Sewer, or Cut), which ran along the narrowest part of the valley north of the main channel (Fig. 12.3)13. This sewer was used to accommodate excess water in time of floods, and was used as an alternative to the main channel when repairs were needed there, but it is not likely that it increased the amount of water which could be voided at low tide: that was limited by the dimensions of the channel further downstream and the length of time between high tides. In the tidal reaches precautions had to be taken, and dams were watched at times of spring tides. There was also always the danger of fresh-water flooding, even during the working months of the summer. A crisis was narrowly averted at Dixter in August 1616, when two loads of cleaved timber were brought in hastily to reinforce a dam. Eighteen men were employed at the unusually high rate of 2s. for a day. Then three men were employed on a Sunday - a practice unusual enough to receive special mention in the book of accounts. When the accounts were finished off and examined in August 1617 the work in the Channel had cost £2074 5s.ld., of which at least £96 was for erecting the wall to contain the Ebony Indraught. Separate accounts show that a further £220 17s.lOd. had been spent on the Back (Kenchill) Sewer and its gutts; £291 2s.2d. on administration; and £270 on timber bought from Mr Culpeper of Wigsell, an estate north-west of Bodiam. The last item is assumed to be the 1,000 tuns of timber ordered Jill Eddison Table 12.2~. Income from Channel Scots The scot rate of l2d. the acre for drowned lands and 4d. the acre for summer lands was established when one scot was raised in 1609, when Mr Thomas Culpeper was BailijJ But the decree granting this scot was already missing by the time ofthe inquiry in 1619 $36). Sum raised by each scot E S. 39 scots were granted by decrees as follows: 1 Oct 1613 22 Sept 1617 14 Oct 1617 18 April 1618 8 Jan 1618119 7 April 1619 24 May 1619 20 at Tenterden 1 Boughton Monchelsea 4 Boughton Monchelsea 4 Boughton Monchelsea 4 Ashford 3 Maidstone 3 Boughton Monchelsea The last of these was due to be paid on 30th Aug. 1619 (f.36). Single scots were payable as follows (f.56): 7 May 1621 2 July 1621 20 Aug 1621 12 Nov 1621 26 Aug 1622 Further single scots were payable as follows (f.67): 28 April 1623 19 May 1623 26 May 1623 2 June 1623 30 June 1623 14 July 1623 4 Aug 1623 18 Aug 1623 1 Sept 1623 15 Sept 1623 30 Sept 1623 21 Oct 1623 28 Oct 1623 4 Nov 1623 11 Nov 1623 18 Nov 1623 155 18 11 155 18 1 1 155 18 11 155 18 1 1 155 18 11 155 18 11 155 18 1 1 158 158 158 158 158 172 172 172 172 172 172 172 172 172 172 172 172 172 172 172 172 Totals d. & S. d. 3,118 18 4 2,962 19 5 3 3 3 3 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 4 4 4 4 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 790 16 8 2,720 13 4 Attempts to clear the Rother Channel, 1613-1624 Table 12.26. Income from Indraught Scots The indraught scots were instituted in August 1615, to be paid regularly twice a year at the rate of 6d. an acre for the drowned lands and 2d. an acre for the summerlands. Sum raised by each scot S. £ Single indraught scots were due as follows (f.36): 19 May 1615 1 Oct 1616 1 April 1617 13 Sept 1617 31 May 1618 5 Oct 1618 5 April 1619 * 77 77 77 77 77 d. 19 19 19 19 19 5% 77 19 5% 77 8 Oct 1619 27 May 1620 One indraught scot was received by the expenditor (f.38~): 4 Oct 1620 Single scots were payable as follows (f.56): 2 April 1621 8 Oct 1621 1 April 1622 14 Oct 1622 . . . . . . . .and as follows (f.67): 1 April 1623 14 Oct 1623 1 April 1624 Notes: 77 77 79 79 79 79 79 86 86 86 19 19 19 2 1 0 0 0 Totals £ 5% 5% 5% 5 % 5% 5 % 5 % S. d. 710 14 9 9 1 1 1 8 8 8 8 5 5 5 79 316 258 2 6 1 9 8 3 1. * This date seems to have been incorrectly recorded. It must have been March 1615/16 or May 1616, since the indraught scots were not instituted until August 1615. 2. Although the status of the information varies, i.e. some scots were "granted", and others "due" or ')ayable", and there are some minor discrepancies in the dates, there are no grounds for doubting that this list is a complete record of the income from scots assessed for this work. Jill Eddison in March 1613114 (Powell 1659). A total of £2,856 5s.ld. had been spent over a period of three years, and the income provided by the 20 scots allocated in 1613 had been used up. The work undertaken so far had clearly not achieved its objective, and further work was necessary. In the period from September 1617 to 1st March 1620121 a further nineteen channel scots were decreed. But this time they were decreed in groups of only three or four at a time, spread over six sessions of sewers between 22nd September 1617 and 24th May 1619, which suggests a somewhat less confident, more ad hoc, approach than was initially taken by the commissioners in 1613. Although unfortunately most of the details of expenditure in this period are missing, the total amount spent was £3,006 8s.Od. (Table 12.1). This sum was remarkably similar to that for the previous three-year period, indicating that the rate of work did not change significantly. The one account which has survived includes a small fraction of the total. It records that in 1619 the channel was widened upstream from the confluence with the Kent Ditch, at a cost of £130. Other major works in 161 9-20 concentrated on the reach between Peening and Reading, and consisted of clearing the bottom of the main channel for 700 yards below Smallhythe (£179); making yet another channel parallel to the main sewer, this one on the south side of the valley near Ebony (£177); and enlarging an old sewer in Peening (£103). By the summer of 1619 there were clearly serious doubts about the effectiveness of the latest programme of work, because at a session of sewers held in Maidstone on 29th September a committee was appointed to carry out a searching inquiry14. After several meetings the committee reported their findings to a session attended by nine commissioners at Tenterden on 13th September 1620. Details of the income and expenditure since 1613 of the expenditor, the deputy bailiff and the two collectors were presented and accepted. £390 7s.ld. was reported due in uncollected scots, which constituted 5% of the total due. No payments were outstanding to any of the workmen. Minor adjustments were made to the scottable acreage. In all, three sluices or gutts, known variously as the Navigable Sluice, the Great Sluice, and the Great Gutt at Kenchill are mentioned in this short account, but it is unfortunately by no means always possible to distinguish between the various references to them. In response, apparently, to a recommendation of the committee a 'navigable sluice' was laid at Peening Brack near Smallhythe in the autumn of 1619, the objective being to prevent the salt water flowing further upstream, and to "draw the fresh water through Peening in the winter". This appears to mean that Peening Level was being used as a reservoir, functioning in the same way as, but probably for longer periods than, the present-day Wet Level, which occupies 245ha of Wittersham Level. For this structure 26% tuns of timber were carried down the river from the west end of Hernden (now Heronden) to Newenden, where the sluice was framed and laid at a cost of £34. The structure, then weighing 30 tuns, was carried by water down to Peening Brack. In spite of the fact that a model of "the Sluice" was commissioned and paid for as early as 1613, it seems that this was the first sluice to be built. There is, however, some doubt as to whether it was laid across the river or simply at the entrance to the Brack, because although it was clearly referred to as a "navigable sluice", numerous later documents refer to the first sluice across the river being that of 1623 at Thorney Wall (see below, page 158). In 1619, too, a Great Sluice complete with a sea gate was laid at Reading, presumably at the mouth of the Ebony Indraught. Carpenters and labourers were paid £203 for laying it, and labourers received a further £30 for work on dams relating to it. Since this sluice was loaded on and off a lighter, it must have been constructed elsewhere. It may have been an old sluice which was taken out of Peening Level that year. Otherwise it was probably brought down the river from Newenden. The wall of Ebony Indraught received further arming, which required 2000 bush faggots and 6 bundles of edders. The committee advised widening the channel from Bodiam Bridge to Peening Flote, which was duly carried out in the summer of 1621. They also discussed widening the channel from Kenchill Gutt to Oxney Ferry but failed to reach a conclusion: some members favoured this course of action; others had a more fatalistic approach and argued that it would only silt up again. The decision was referred back to the commissioners who, with their minds on other schemes, apparently ignored the proposal. The only evidence of the width of the channel achieved as a result of all these works comes from Harlackenden, writing 45 years later. He claimed that during this period the channel was cleansed and widened "at the top" to 90 to 85 feet (27m to 26m) from Thorney Wall (see Fig. 12.4) up to Redhill; to 80 feet (24m) from Redhill up to Kenchill; narrowing from Kenchill "after like proportions" down to 40 feet (12m) at Bodiam Bridge. There must, however, be some doubt about some of these figures since it seems clear from the 1619 inquiry coupled with the accounts for 1619-20 that the work was not carried further downstream than Kenchill. Unfortunately, but perhaps significantly, there is no record of any answer to what was probably the most important question put to the committee. This concerned the amount of benefit which had been received by any of the Levels as a result of the work already carried out. It is possible that this omission conceals differences of opinion and personal interest among the commissioners. Alternatively it may have been considered better not to give an official answer at that time, since it is clear from the subsequent account that by 1621 the commissioners' intentions had turned towards schemes in another direction. Attempts to clear the Rother Channel, 1613-1624 The account for this period covers just over two years, from 1st March 1620121 to 25th March 1623 (the first day of the new year). The sum total of expenditure was £1,502 13s.2d., two-thirds of which were spent in payment of rents for the occupied lands and on administration in preparation for two new ventures. Only two operations were carried out in the valley. The largest was to widen the Channel from Bodiam Bridge down to Peening Flote, following the advice of the committee in the Inquiry of 1619, at a cost of £465 157 Woodruffe (also known significantIy as Low Woodruffe) out on Walland Marsh (Fig. 12.4). The account for the 12 months from 25th March 1623 records outstandingly high expenditure, £3794 17s.7d. Construction of the Woodruff indraught involved wholescale transformation of an area which measured nearly 1.5km from west to east. The walls round the perimeter of the indraught measured three and a half kilometres, and those along either side of the neck of the 1 ls.8d. The other significant development was the removal of the Great Sluice from Ebony Indraught. For this £30 was paid to a carpenter, £30 to two men for scavel work, and £27 for adjustments to the channel and for reinstating the tow-path. Otherwise, confidence that adequate improvements could be made to the channel in the valley had waned, and the commissioners' attention had turned to a completely new sphere of activity further down the estuary. Negotiations were carried out and plans were made for installing a freshwater sluice across the channel at Thorney Wall, and for establishing a vast new indraught at Old indraught totalled another two kilometres. New walls were constructed and other pre-existing walls were heightened. These walls made a major contribution to the local topography until mid-twentieth century, and some were remaining standing in 1992. The pre-existing surface of this reclaimed marshland was evidently deeply cut across by several old creeks, which had either been preserved when salt marsh was primarily inned, or had been incised during subsequent inundations, or which may have been a complex product of both processes. An essential preliminary to the construction work was to fill in several of these creeks up to the general Fig. 12.4. Woodruffe Indraught and the Sluices at Thorney Wall. Information from ESRO A2806/1/9/2, based on O.S. 6" First Edition. Jill Eddison level of the marsh. This proved a very expensive operation: more than £30 was spent making up one such creek; and £26 for making up two other creeks and setting up a wall across them. A 'carrying-way' was built up to provide access from Thorney Wall, and other more local tracks were established. Existing sewers were widened. The Great Sluice which had previously stood at the Ebony Indraught was stored over the winter and then 89 tuns of it was carried on lighters down to Thorney Wall, at a cost of £1 1 17s.8d. Thirty-three man days were spent loading and unloading it at a further cost of £3 6s.Od., and eight man days were needed to carry it "from the place where it was framed to the waterside". It was reconstructed by a carpenter for £30, and installed at the mouth of the neck of the new indraught. Associated scavel work cost a further £80. The freshwater sluice at Peening was taken up, transported and rebuilt across the river channel near Thorney Wall, and a thatched sluice house was built beside it. After some changes of mind as to whether or not this sluice should have a facility for navigation, it was decided that it should not, and provision was made to compensate the mariners of Rye for any delay caused by loading and unloading their goods there. A proviso was added that "if the stop be hurtful to the haven of Rye or to navigation on the river, other than increasing the price of carriage, the stop be removed" (see also Hipkin this volume, 145). There seems to have been, in addition, a third "New Sluice" which it has not been possible to locate with any certainty: it probably stood between the neck and the main part of the indraught. Outside Thorney Wall lay a wide band of salt-marsh, and beyond that the mud-flats of the river channel. Two cuts were made across the salt marshes, the second extending out across the mud-flats to the low tide mark. Digging of 90 rods (450m) of the Great Cut is recorded, but whether this was for both sides of a 450-metre trench, or whether it was for one side only of a 225-metre trench is impossible to determine. Thirty-seven and a half rods' (186m) work is recorded on the Shorter Cutt, but again it is impossible to be sure of the length of the cut. These must have been the two channels shown on the undated map reproduced as Figure 12.5, which shows no discernible difference in their length. The recorded work in the Long Cutt cost between 30 and 42 shillings per rod, and totalled £180. In the Short Cutt it cost between 26 and 29 shillings per rod and totalled £70. All the work on the sluices and the cutts was carried out under limitations imposed by the tides. The mud-flats would have been covered at every high tide and the salt marsh by spring tides. Some insight into the problems of working under such conditions is given by the payment of 18s, to a carpenter for nine days' pumping at the Great Sluice, and the same amount to a blacksmith for mending the pump on various occasions. On occasions dams were built to keep the tides out. The accounts also record that walling was also built on the salt marsh. In fact, it must have been necessary to build walls to link the new channel sluice to the estuary embankments on the east and to the Isle of Oxney to the west, but unfortunately they do not appear on the contemporary map (Fig. 12.5). Without such a cordon of sea defences, spring tides would have flowed round the channel sluice and rendered it useless. Construction work in these new situations, most of which were exposed to the erosive power of waves, required large quantities of, and heavy expenditure on, materials. A total expenditure of £1,684 5s.2d. is recorded on large quantities of faggots, needles, keys and edders, which were needed to protect both the walls and the cuts. One thousand piles were supplied, presumably for use at the sluices and along the cuts, at a cost of £60 16s.8d. One hundred and fifty-four tuns of new timber and 23 oak trees came from Gate Court, Northiam, and further timber came via the yard at Newenden. Planks and boards were provided, some from Newenden, some from Bodiam, and some from across the river at Stone in Oxney. As has already been shown, where existing structures had become obsolete they were dismantled and transported down the river for use second-hand. From a financial point of view, iron work used on sluices, pumps, and hoists was a minor consideration, but references are made to the smith at Sandhurst and to Brenchley furnace. Seven thousand bricks were brought down from Tenterden for use at the Thorney Wall sluice, the bricks costing £5 12s.Od. and "carriage" £3 10s.Od. Hay and reed were provided for ramming the sluices and thatching the sluice house. The available evidence, which is relatively full on this point, shows that all these materials came from local sources. Timber and coppice wood came from estates bordering the Rother valley or from Oxney, the ironwork from relatively close sources in the Weald, and the bricks from Tenterden. The Rother is shown to have provided an essential artery for transport, which is hardly surprising considering the notorious nature of Wealden roads. The undated map reproduced as Figure 12.5 indicates the relative location of the channel sluice and the indraught sluice. It shows Thorney Wall and the point where all the construction materials and also, presumably, the labour force were disembarked. Two cuts are shown, one being the salt-water intake for the indraught. The other leaves the river channel at its nearest point to Thorney Wall, which is upstream of the new sluice in the channel and was therefore presumably a fresh-water channel. It appears to terminate against the side of the indraught sluice, which raises the possibility that there was some mechanism by which fresh water was introduced this way into the indraught. A further stop lies in the centre of this cut, and on the manuscript map the section furthest from the river is coloured a deeper shade of blue, suggesting a locked- in area where boats could stay afloat whatever the level of river or tidal water. Lastly, there are two elliptical features on the wall near the river channel, which may either represent a pair of small boats drawn up on to the Attempts to clear the Rother Channel, 1613-1624 Appledore ! >B - Scale Perches L? ! I T~~ \c, \ \ +y 2 : G?' > / lJT:I:'J~T1b 7' ,l; v, Fig. 12.5. Facsimile of part of map ESRO A2806/1/9/2. wall, or alternatively some form of lifting device. The accounts mention expenditure on gins and gin ropes. One other topographical alteration was necessitated by the installation of the Woodruff indraught. This concerned the Five Waterings Sewer, which drained the westernmost triangle of Romney Marsh between Appledore, Hamstreet, and Snargate into the Rother channel. The Patent Roll of 1544 shows that since that date this sewer had crossed the Rhee Wall at Arrowhead re_---- ,---_ -_-- ___ -- , ' ,' Gutt, and then followed a course which lay down the middle of the proposed new indraught. It passed through a sluice at Chappelhook Bridge (which was used as a second line of defence in case the sea broke through that far), then down the corridor which was now to be used as the neck of the indraught, and entered the Appledore Water through a gutt which stood approximately where the Great Sluice was to be re-erected (Cal. Pat. Rolls, 28 March 32 Henry VIII, vol. 19 pt. 1, 172) (Fig. 12.6a). 160 Jill Eddison Fig. 12.6~. Early courses of the Five Waterings Sewer. Based on O.S. 6" First Edition with information from All Souls CTM 226~165~ As early as August 1615 the Lords of Romney Marsh, in whose responsibility lay the drainage and outfall of the Five Watering Sewer, had decided that it would be feasible to lay out 200 acres of the "drowned lands" of Woodruff as an indraughtI5, which suggests that the passage of the Five Waterings Sewer through Woodruffe was already not without its problems. With the installation of the new indraught, the Sewer had to be moved into a new course just east of the Indraught, which it still occupies today (Fig. 12.6b). The account book records a journey by the expenditor and one other to attend a lathe of Romney Marsh at Dymchurch in Whitsun week 1622, at which it was agreed that the sewer should be moved to a convenient new course. The cost of its removal, however, lay with Romney Marsh and not the Rother Levels. Romney Marsh was to pay up to £40, and £700 was loaned by Sir Alexander Culpeper. Epilogue The Account Book comes to an end in March 1624125, and it seems likely that the reason for discontinuing this series of accounts was the cessation of construction work. However, the Decree Book carries the story forward and throws light on the very unsatisfactory outcome of all this (Thomas Gull), late 16th century. work. The administrative infrastructure was soon wound up. The last record of rent paid for Peening Indraught is dated Michaelmas 1622, and that indraught apparently went out of use soon after that. The lands which had been used for Ebony Indraught were returned to the owners at Michaelmas 1624. At the time it was said that these two indraughts were too far from the sea to take in the salt waterI6, although with the wisdom of hindsight it seems very probable that their failure was due at least in part to the deposition of silt within them by the tides, which could have raised their ground level critically. The Woodruffe Indraught had a very short life. Storms and "an extraordinary tide" which occurred sometime between 26th May 1626 and 10th April 1627 endangered the Great Sluice at its mouth, and fretted and broke down the walls of the indraught itself. The walls were no doubt eroded by waves generated in the 200 acre (82ha) expanse of water which the commissioners themselves had created: a short length of the Cross Wall survives today with an extraordinarily irregular profile. Repairs costing £2,000 were essential, to prevent widespread flooding of Walland and Romney Marshes. In 1627 it was ordered that the indraught be walled off, and the neck alone (some 35 acres, 14ha) was designated a fresh-water indraught. In 1635 the Appledore Channel was eventually Attempts to clear the Rother Channel, 1613-1624 Fig. 12.6b. Course of the Five Waterings Sewer since 1624. Based on O.S. 6" First Edition with information from ESRO A2806/1/9/2. abandoned as the main river channel (but retained as the local Reading Sewer) and the Rother was re-routed south of the Isle of Oxney. This greatly improved the flooded condition of the Upper Levels, but simply transferred the drainage problems into Wittersham Level (see Eddison 1988). Even today, with the advantages of modern pumping stations, it is impossible at times of high run-off to get all the river water away to the sea immediately, and one third of Wittersham Level, 600 acres (245ha) of low-lying land adjacent to the river, has to be used as a temporary storage reservoir. It is only by the deliberate flooding of this large area that inundation of the rest of the Levels is avoided. Conclusions In the period 1613-1624 a total of £1 1,418 was spent trying to restore the outflow of the Rother and relieve the permanent flooding of the Upper Levels. From 1613 to 1619 work concentrated on the reaches of the river enclosed in the valley. After a pause for administrative preparation, the operations were resumed in 1623 with increased vigour - and increased expense - in the area south and east of Appledore. At the end of this period very little, if any, improvement had been achieved. 161 VlNAL BRIDGE All the three methods employed in these attempts to manage the Rother i.e. enlarging the river channel, the use of indraughts, and the use of sluices to keep out the silt-laden tides had been employed on these marshes for centuries. The work was put in hand by the commissioners of sewers, all local land-owners who were working in their own interests. At first sight, they can scarcely be expected to have committed expenditure of this magnitude without reasonable hope of success. Why did all their attempts fail so dismally? There is no doubt that this period marked a critical point in the swing of the geomorphological pendulum. In the thirteenth and fourteenth centuries Walland Marsh was probably wide open to the forces of the sea, and the tides flowed right up the Rother Levels. By the fifteenth century the shingle barrier had once more partially enclosed the southern shore of Walland Marsh and created behind it a sheltered environment particularly favourable to silting. Both the silting and the reclamation that it encouraged reduced the area open to the tides, and the inevitable cycle of silting, then reclamation, then further silting continued. As a result of this interaction between physical processes and human intervention, a critical point had been reached in the history of the valley. In these circumstances it is difficult to explain why the 162 Jill Eddison commissioners persevered for so long with their attempts, and especially why they embarked on the last, most expensive, phase. The account book sheds light on the work which was done, but not on the reasons for decisions taken by the commissioners, or on the interests of individual landowners. That is a question which must remain open, and is one which might be investigated more thoroughly in the future, using other sources and information. Whatever the politics behind their decisions, it is reasonable to conclude with the wisdom of hindsight that the physical cards were stacked against the commissioners. They seem to have been attempting the impossible. Acknowledgements This research was carried out with the support of a grant from the British Academy. The author is most grateful to Dr Joan Thirsk for her advice and encouragement, and to Mark Gardiner who commented on a draft of the paper. Christopher Whittick kindly drew attention to the existence of the account book. Dorothy Beck discussed the structure and materials used for arming the walls. Clifford Bloomfield kindly drew up a fair copy of Figure 12.5 and discussed the implications of it, Jane Russell made fair copies of the other maps, and James Eddison set the tables. Glossary To arm: to protect and strengthen a bank of soil with a lattice work of timber products, i.e. faggots, edders, needles and keys (see below). To blow up: to lift the whole structure of the frame and gates of a sluice or gutt clear of the ground, a phenomenon likely to be caused by an exceptionally high tide. Cut, cutt: a new channel. Edders: light flexible rods of wood such as osiers or hazel. Expenditor: a paid officer appointed by the commission of sewers to supervise expenditure of the money collected in scots for the repair and construction of drainage and/ or sea defence works. References (Superscript numbers in the text refer to unpublished sources, listed below.) Published sources Calendar of Patent Rolls Eddison, J. 1985: Developments in the lower Rother valleys up to 1600. Arch. Cant. 102, 95-1 10. Eddison, J. 1988: 'Drowned Lands': Changes in the Course of the Rother and its Estuary and Associated Drainage Problems, 1635-1737. In Eddison, J. and Green, C. (editors), Romney Marsh: Evolution, Occupation, Reclamation. OUCA Monograph 24, 142-162. Harlackenden, Thomas 1663: Animadversions on several passages Faggots: bundles of sticks or twigs bound together, laid along the sides of the walls to arm them. Fleet, float or jZote: 1) an area always flooded. These were often in the form of long narrow strips, being old creeks which remained below the water-table after the area had been reclaimed. 2) a dock, e.g. Newenden Flote. A Gin: A mechanical apparatus for hoisting heavy weights, a crane. Now usually in tripod form, one leg being movable for variations in height, and the other two set a certain distance apart with a winch beetween them, round which a rope is wound. Gin-ropes: see gin, above. Gutt: an outfall. An early name for a sluice. This name seems to have been applied originally to outfalls of fresh into salt water, though as the extent of the tides diminished, this meaning often became obsolete for specific structures and included outfalls into fresh- water channels. Indraught: a (generally) large area of marshland used as a reservoir into which fresh or salt water was admitted through a sluice, or gutt. The water was then released in a rush with the object of scouring the sluice and the channel downstream. Keys: pieces of wood which were inserted through the holes in the needles to hold them in place. Lathe: a session of the Lords (the sewer commission) of Romney Marsh proper. Needles: stakes, driven down into the soil of the walls, with holes punched through them to accommodate the keys. A scavel: a spade. Slidbutts: sections of the bank which became detached along concave lines of weakness and slumped into the channel. Slub: thick unconsolidated mud. To be swerved: choked with sediment. in a book written by Sir Nathaniel Powell, Bart, together with a more exact narration of the state of those levels. London. Hipkin, Stephen 1995: The impact of marshland drainage on Rye Harbour, 1550-1650. In Eddison, J. (editor), Romney Marsh: the Debatable Ground. OUCA Monograph 41, 148-163. Powell, Nathaniel, 1659: A remonstrance of some decrees, and other proceedings of the Commissioners of Sewers for the Upper Levels in the counties of Kent and Sussex, touching the proportioning of water-scots upon the said levels. London. Powell, Nathaniel 1662: A summary relation of the past and present conditions of the Upper Levels, lying in the counties of Kent and Sussex. London. Rendel, W.V. 1962: Changes in the course of the Rother. Arch. Cant. 77, 63-76. Attempts to clear the Rother Channel, 1613-1624 Unpublished sources Abbreviations used: BL British Library CCA Canterbury Cathedral Archives CKS Centre for Kentish Studies ESRO East Sussex Record Office PRO Public Record Office 1. ESRO A2806 11411 2. CKS S/Rm/SOi 3. ESRO A2806/1/2/1 Decree Book, Kent Sewers, River of Rother. The excellent organisation of this series seems to reflect the appointment on 11th March 1622123 of Ambrose Cogger, a man of exceptional ability and organisation, as Clerk of the Levels. 4. ESRO A2806111312 An abstract of the Books of Orders and Decrees made by the Commissioners of Sewers of the Marshlands sewing into the River of Rother and Channel1 of Appledore, commonly called the Upper Levels in the Counties of Kent and Sussex, 11 March 1622 123 - 15 July 1731. 5. ESRO A2806111912. This map covers the Rother valley from Thorney Wall to Bodiam Level. 6. PRO S.P.Dom. Eliz. XXIII, no.41 7. PRO S.P.Dom. Eliz. XCIII, no.22 8. PRO S.P.Dom. Eliz. CXX, no.28 9. John Stoneham: Map of Romney and Walland Marshes and Rother Levels, 1599. Hastings Museum MA 189; ESRO Rye 13217,8. 10. The 13 churches where the scots were proclaimed were Beckley, Bodiam, Ebony, Ewhurst, Northiam, Newenden, Rolvenden, Sandhurst, Stone, Smallhythe, Tenterden, Wittersham, and Woodchurch. 11. CKS U41 P2, Ebony Court Lodge Lands, 1710, by Francis Hill. This map shows two fields, the Little Indraught of 10 acres, and the Great Indraught of 11 acres. The indraught was probably larger than just these. 12. CCA, Ebony Tithe Map. 13. BL Cotton Augustus 1 i 25, map temp. Eliz. ESRO Rye 1319 is a reduced copy of this map. 14. ESRO A2806111411, ff.34~-37~. 15. CKS S/Rm/S02. 16. ESRO A2806111411, f.7.