LEISTON SUBSTATION 132kv CABLE ROUTE, SIZEWELL LEISTON LCS 150

LEISTON SUBSTATION 132kv CABLE ROUTE, SIZEWELL LEISTON LCS 150 Post-Excavation Assessment Report SCCAS Report No. 2012/016 Client: South East Electricity Substation Alliance Authors: Anthony Breen, David Gill and Richenda Goffin December 2014

LEISTON SUBSTATION 132kv CABLE ROUTE, SIZEWELL LEISTON LCS 150 Post-Excavation Assessment Report SCCAS Report No. 2012/016 Authors: Anthony M. Breen, David Gill and Richenda Goffin Contributions By: Sue Anderson, Tom Cousins, Julie Curl, Richard Darrah, Val Fryer, Dr Ben Gearey, Dr Tom Hill, Ian Riddler, Dr David Smith and Ian Tyers Illustrator: Beata Wieczorek-Oleksy Editor: Richenda Goffin Report Date: December 2014

HER Information Report Number: 2012/016 Site Name: Planning Application No: Leiston Substation 132kv cable route, Sizewell C/06/2191/FUL Date of Fieldwork: 20th May 26th June 2008 Grid Reference: TM 4719 6316 TM 4693 6272 Client/Funding Body: South East Electricity Substation Alliance Client Reference: - Curatorial Officer: Project Officer: Oasis Reference: Jess Tipper Rob Atfield suffolkc1-119095 Site Code: LCS 150 Digital report submitted to Archaeological Data Service: http://ads.ahds.ac.uk/catalogue/library/greylit Disclaimer Any opinions expressed in this report about the need for further archaeological work are those of the Field Projects Team alone. Ultimately the need for further work will be determined by the Local Planning Authority and its Archaeological Advisors when a planning application is registered. Suffolk County Council s archaeological contracting services cannot accept responsibility for inconvenience caused to the clients should the Planning Authority take a different view to that expressed in the report.

Contents Summary Drawing Conventions 1. Introduction 1 1.1 Project background 1 1.2 Sequence of archaeological work 1 1.3 Original project aims 5 2 Geological, topographic and archaeological background 5 2.1 Geology, topography and recent land use 5 2.2 Archaeology and historical background 6 2.3 Scope of the project 9 3 Results of the fieldwork 10 3.1 Methodology 10 3.2 Excavation summary 11 3.2.1 Overview 11 3.2.2 Phasing 14 3.3 Site description by chronological phase 17 3.3.1 Phase 1 Prehistoric 17 3.3.2 Phase 2 Pre-Conquest-10-11th century 18 3.3.3 Phase 3 11th-12th century to Phase 4 18 3.3.4 Phase 4 12th-13th century 21 3.3.5 Phase 5 mid-13th century 23 3.3.6 Phase 6 late13th-14th century 34 3.3.7 Phase 7 post 14th century 44 4 Quantification and assessment 47

4.1 Post-excavation review 47 Factual data 47 4.2 Quantification and assessment of the bulk finds archive 48 4.2.1 Introduction 48 4.2.2 Pottery 49 4.2.3 Ceramic building material 52 4.2.4 Fired clay 53 4.2.5 Clay tobacco pipe 54 4.2.6 Worked flint 54 4.2.7 Burnt flint 55 4.2.8 Stone (unworked) 55 4.2.9 Quernstone 55 4.2.10 Iron nails 55 4.3 Quantification and assessment of the small finds archive 55 4.3.1 Introduction 55 4.3.2 Factual data 56 4.3.3 Finds associated with the boat 57 4.4 Quantification and assessment of the environmental evidence 59 4.4.1 Re-used boat timbers 59 4.4.2 Wood technology 59 4.4.3 Dendrochronology 61 4.4.4 Animal bone 61 4.4.5 Shell 67 4.4.6 Charcoal 67 4.4.7 Charred plant macrofossils and other remains 68 4.5 Quantification and assessment of the documentary records 71 4.5.1 Introduction 71

4.5.2 Research 72 4.5.3 The progress of the research 73 5 Significance of the site data and potential for analysis 76 5.1 The significance of the site record 76 5.2 The boat timbers 79 5.3 The potential and significance of the finds data 80 5.3.1 Pottery 80 5.3.2 CBM 81 5.3.3 Fired clay 82 5.3.4 Clay tobacco pipe 82 5.3.5 Worked flint 82 5.3.6 Burnt flint 82 5.3.7 Slag 82 5.3.8 Quernstone 82 5.3.9 Stone 83 5.3.10 Iron nails 83 5.3.11 Small finds 83 5.3.12 The organic finds from the boat timbers reused in the well 85 5.4 The potential and significance of the environmental evidence 86 5.4.1 Wood technology 86 5.4.2 Animal bone 87 5.4.3 Fishbone 88 5.4.4 Charred plant macrofossils and other remains 88 5.4.5 Soil micromorphology 89 6 Updated Project Design 89 6.1 Revised research aims 89 6.2 Updated project design 90

7 Preliminary publication synopsis 91 7.1 Suggested layout 91 9 Acknowledgements 92 10 Bibliography 93 List of Figures Figure 1. Site location 2 Figure 2. Excavation areas LCS 148 and LCS150 plotted over the LIDAR data 4 Figure 3. Historic maps 8 Figure 4. Site plan showing all cut features 15 Figure 5. Phase 3 and Phase 4 features 19 Figure 6. Phase 5 features 25 Figure 7. Detail plan of Phase 5 building 1092 28 Figure 8. Phase 6 features 37 Figure 9. Phase 6 Building, ovens and cisterns detailed plan 42 Figure 10. Phase 7 features 45 List of Tables Table 1. List of physical records 47 Table 2. List of digital records 48 Table 3. Finds quantities. 48 Table 4. Pottery quantification by fabric 49 Table 5. Pottery quantification by feature type 51 Table 6. Pottery in structural groups 52 Table 7. CBM by fabric and form 52 Table 8. Quantities of fired clay by fabric 54 Table 9. Quantification of the faunal assemblage by context type 62 Table 10. Quantification of species by context type 64 Table 11. Small finds by category and material 84 Table 12. Summary of tasks for analysis and publication 95 Table 12a. Summary of boat specific tasks for analysis and publication 96

List of Plates Plate 1. General view of the eastern end of the site 12 Plate 2. General view showing the line of the palaeo- channel 12 Plate 3. Phase 5 posthole building 1092 27 Plate 4. Timber wall plate from pit 1503 27 Plate 5. Fragment of oven floor 1311 27 Plate 6. Well 1365 containing a suite of pottery vessels 33 Plate 7. Pit 1025 contained a coarseware jug 33 Plate 8. Water tank 1219 40 Plate 9. Bottom of water tank 1730 40 Plate 10. Side of water tank 1219 made from a section of boat hull 40 List of Appendices Appendix 1. Appendix 2. Appendix 3. Appendix 4. Appendix 5. Appendix 6. Appendix 7. Appendix 8. Appendix 9. Appendix 10. Appendix 11. Appendix 12. Appendix 13. Appendix 14. Appendix 15. Brief and specification Context List Harris matrix Documentary evidence Bulk finds catalogue Pottery catalogue by context CBM catalogue by context Fired clay catalogue by context Small finds catalogue Leiston boat timbers report Dendrochronology report Animal bone by context Charred plant macrofossils and other remains Paleo-environmental assessment OASIS summary report

Summary A program of archaeological fieldwork, culminating in an open area excavation of 3,700sqm, was completed in advance of laying underground 132kv cables between the site of the substation for the Greater Gabbard Windfarm and the Sizewell Power Station. The cable route passed through part of medieval Sizewell, the once burgeoning market town that flourished under the benefaction of Leiston Abbey until the 14th century. Time and tide has since seen Sizewell reduced to a fishing hamlet; a decline that was the result ultimately of a coastal squeeze, caused by erosion and subsequent (and repeated) inundations by the sea. Resignation of the settlement s diminishing status was first signalled by the relocation of the abbey convent to its present inland spot at Leiston in 1361, which was followed, within thirty years, by the re-siting of the market. Despite this, fishing remained vibrant, due in part to the addition of Dutch expertise, and Sizewell s population was still comparable in size to Leiston in the early 16th century, but in the face of a relentless sea, numbers waned rapidly through the 16th and 17th centuries. The archaeological excavations offered an opportunity to chronicle a part of Sizewell s past and through this study inform our understanding of the changing shape of our towns. The spread of the archaeological features recorded during the excavation seems to represent the pinnacle of the settlement s westerly expansion, which peaked at or soon after the turn of the 14th century, and attested to a period of industrial activity centred on a freshwater marsh. The site produced evidence of workshop-type buildings, ovens and an assortment of timber-lined wells and sunken water-tanks together with a large assemblage of finds to help characterise the site as a place of work and a place where cereals crops were being processed. The cut-off in the archaeological record occurs during the first half of the 14th century; this is abrupt and coincides closely with what is historically the start of Sizewell s decline. The excavation data has the potential to shed light upon the type of occupation that occurred here and a site sequence can be reconstructed. The finds assemblages reflect the settlement s location on the beach hinterland and include fishing equipment (hooks, weights, boat nails and possible net fragments) along with the bones of both freshwater and marine species of fish. Of particular pertinence to this coastal milieu was the discovery of sections of planking from a small inshore boat of between 6-9m long. The planks were riveted together and caulked with sheep s wool and its construction

seemed to follow a British (rather than a Scandinavian) tradition. Some of the boat fragments are worthy of museum display and retain enough details to enable the boat to be reconstructed (virtually). The boat s timbers (if not the entire vessel) were sourced from Ireland and were from trees felled sometime between AD1241 and AD1266; this is compatible with few examples of early 13th century boat remains found along the east coast which suggest that Ireland was the source of oak boards before the development of the timber trade with Hanseatic League in Northern Europe. Beyond the beach, the presence of boar, deer, and rabbit suggested the population successfully exploited the natural hinterland or derived direct benefits from the infrastructure of the abbey of Leiston. The plant macrofossil record, although not plentiful, showed that heathers and bracken, probably gathered from the nearby commons, were being used as fuels within ovens and corn driers on the site and hemp, was being processed. The pottery assemblage is particularly important as it produced a number of complete or near-complete vessels dating to 13th/14th-century; most notably jugs but also some jars and bowls, which will form a basis of a type series for this part of Suffolk. In addition the preservation of a number of examples of medieval joinery, including barrel staves, a building wall plate and well-linings offer an opportunity for the study of wood-working technology. In addition to the archaeological evidence, there is an unusually complete set of medieval land records for the manor of Sizewell which indicate that the land to the east of the town was divided into a high proportion of small copyholds. The records include the area sampled by the excavations and date back to the period it was occupied. The properties and tenancies described within these records have the potential to be precisely located geographically and, together with close dating gained from timber and organic remains recovered from the site, show that there is a high potential to relate the written records to the archaeological one, greatly enhancing the academic value of both. In summary the site, together with the neighbouring excavations(lcs148), has provided a relatively large sample across several medieval plots which have a high potential to contribute to the knowledge of the town s early layout, character, development history and subsequent decline. The study of the development of medieval towns is a key research topic, contributing to a better understanding of the region s historic landscape; as such the excavation findings are of regional significance and merit academic

publication. It is therefore suggested that the results of this and the adjacent site, LCS 148 are published together in the East Anglian Archaeology (EAA) monograph series. A full preliminary publication synopsis is included with the report.

1. Introduction 1.1 Project background A program of archaeological work, consisting of paleo-environmental investigation, evaluation by trial trench and ultimately an open area excavation was undertaken in advance of laying the underground 132kv cables which were to connect the proposed substation for the Greater Gabbard Windfarm to the Sizewell Power Station and thence the national grid. The cable route lay within an area of high potential archaeological interest and the archaeological work was a condition of the planning consent (on application C/06/2191/FUL), in accordance with Planning Policy Guidance 16. This document contains an assessment of the potential for analysis and publication of the archaeological fieldwork results. It summarises the field and post-excavation work already carried out and details the additional work needed to publish the excavation results. This is required in order to bring the project to a conclusion in line with the discharge conditions on the planning consent. The assessment is consistent with the requirements of the English Heritage guidelines as set out in The Management of Research Projects in the Historic Environment (English Heritage2006). The field work was undertaken between May and June 2008 by SCCAS Field Team. The project sponsor for all of the work to date is South East Electricity Substation Alliance, a partnership between National Grid and construction companies AREVA, Skanska and Mott MacDonald, which has funded all stages of the fieldwork to date and this assessment. 1.2 Sequence of archaeological work The proposed cables were to be laid along the route using a combination of Horizontal Directional Drilling (HDD) and, wherever possible, conventional open trenches. Where trenched, the cables were to be laid in four parallel channels; each 0.6m wide, 1.4m deep and spaced at 3m intervals. The cable trenches were to be excavated within a working corridor 20m wide which was to be stripped of the topsoil exposing potential archaeological deposits to considerable damage. Thus the heritage appraisal of the site was concentrated on the areas of the proposed open excavation namely along the south edge of Pill-box field and the launch pits of the HDD near Suffolk Coastal District Council s car park alongside the beach. 1

A Norfolk B A SUFFOLK Essex 0 0 25 km 2 km 10m LCS 052 262900 LCS 049 LCS 112 LCS 051 LCS 050 LCS 053 262800 Site LCS 148 S ll we ize p Ga LCS 054 262700 2 LCS 057 L 10 m ET L ET 262400 647600 5m LCS 114 5m LCS 056 262500 N m 10 LCS 073 263000 647500 647400 647300 647200 647100 647000 646900 646800 646700 263100 B 1 0m LCS 117 5m 262300 262200 TM Crown Copyright. All rights reserved. Suffolk County Council Licence No. 100023395 2012 0 200m Figure 1. Location of site and Historic Environment Record entries as mentioned in the text (green)

An initial desk-based assessment of the area, commissioned as part of the Greater Gabbard wind farm project, (Maritime Archaeology Ltd 2006) identified it as one of high archaeological potential. Suffolk Coastal District Council (SCDC) demanded further field evaluation to test this potential, with the aim of assessing the extent and quality of the possible archaeological resource and its vulnerability to the impact of the development. To this end a staged program of archaeological work, consisting of the monitoring of engineering test pits along the full 1km cable route followed by a targeted palaeoenviromental assessment by auger survey and an archaeological evaluation by trial trenching was recommended. A series of Briefs and Specifications was drawn up by Dr. Jess Tipper, the Suffolk County Council Archaeological Service Conservation Team (SCCASCT) officer advising the planning authority, detailing how each aspect work was to be done. The field surveys and evaluation identified a sequence of ditches, pits and postholes across the south end of Pill Box field indicative of a settlement site. The features produced closely dateable medieval pottery from a variety of local manufacturing sites, including Hollesley-type wares, suggesting an occupation of the settlement starting in the 12th and ending at the start of the 14th centuries. Oyster shell, animal and fish bones were also found and the remnants of clay surface, damaged by modern ploughing, indicated that part of the earlier ground surface was likely to remain intact (Atfield and Gill 2008). The palaeo-environmental surveys found organic-rich deposits symptomatic of a backwater lagoonal floodplain. The deposits contained an abundance of well-humified peats, characteristic of a wet environment which indicated a high potential for well-preserved plant and animal macro-fossils (Hill, Gearey and Smith, Appendix 14). The evaluation trench within the area of the launch pit near the beachside car park and the monitoring of the engineering test-holes established that the archaeological potential on the cable route beyond Pill-Box field was low. No deposits were identified of sufficient importance to warrant preservation in-situ, but preservation by excavation and record was stipulated by SCDC to mitigate for the loss of the archaeological remains within the development. A further brief was prepared by Dr. Tipper which outlined an area of excavation totalling 3680sqm (Appendix 1). The open-area excavation confirmed that the features were evidence of settlement and part of the once burgeoning town of Sizewell that developed around the landing place 3

4 ro Gap well Size Flood Zone ad Plan Scale 1:10,000 0 LCS 148 and 150 excavation areas 500m Figure 2. Excavation areas plotted against the LIDAR and flood zone data. The Sizewell Gap road is a relatively recent addition, the medieval route between Leiston and Sizewell skirted the edge of the common and is shown in red; the orginal bridge across the brook that is now Sizewell Wents is marked with the red star just to the east of the site; the bridge was short-lived and relocated further to the north in the later medieval period. N

and market under the administration of Leiston Abbey. The settlement spread was further recorded in the adjacent field to the west as part of the same overall project (Greater Gabbard Wind Farm Onshore Works excavations (LCS148) funded by GGOWL (Gill 2013). 1.3 Original project aims The broad aims of the project to assessment stage were to: provide a record of all archaeological deposits which would otherwise be damaged or removed along the cable route. assess the significance of the archaeological data and its potential for reconstruction of the site s history and land use, with particular reference to the date of its establishment, subsequent development and end. assess the site s character and status and place it in the context of the local historical landscape. provide data for the study and characterisation of the local palaeo-enviroment. The academic objective centred upon understanding the nature of medieval settlements; their layouts and their relationship with the landscape and the sea, establishing the nature of the wet environment, dating when the lagoon was flooded and the settlement s relationship with this very particular setting. These themes were considered pertinent at the beginning of the project based on the limited understanding of the site at the evaluation stage. More specific and updated objectives, determined after the review of the excavations findings, are presented at the end of the report in Section 6. 2 Geological, topographic and archaeological background 2.1 Geology, topography and recent land use The site lies in the parish of Leiston, 2.8km west of the centre of Leiston town (Fig. 1) and in an area of open farmland. The site is situated within the Suffolk Coast and Heaths Area of Outstanding Natural Beauty on the North Sea coastline and the mean high tide mark is c.350m to the east. The parish incorporates the remnants of the medieval hamlet of Sizewell which has been largely lost to coastal erosion that occurred most dynamically during the 16th century. 5

The geology of the area is of deep, well drained, sandy soils overlying glaciofluvial drift. The site is one of level ground at a height of c.4m AOD. Ground levels rise to the south, west and north-west, placing the site in a shallow valley, currently an area of low-lying waterlogged farmland that leads to the coastline and the Sizewell Belts which is a designated S.S.S.I. The sites of the Sizewell A and B nuclear power stations lie 400m to the north on an area of relatively high ground overlooking the coastline. 2.2 Archaeology and historical background Based on documentary research by A.M. Breen The site lies within an extensive, multi-period archaeological landscape recorded in the Suffolk Historic Environment Record (HER). This has been previously highlighted in a desk top survey of the Sizewell Belts S.S.S.I. (Newman and Ridgard 1995), a desktop assessment and field survey of the proposed Sizewell C area (Richmond 1994) which includes the current site, and in a desk based assessment for the neighbouring site to the west, LCS 148 (Maritime Archaeology Ltd, 2006). In summary the HER records evidence of prehistoric, Roman and medieval activity in the nearby area (Fig. 1). Possible Bronze Age ring ditches and other earthworks (LCS 052, LCS 053, LCS 055, and LCS 057) have been identified within 350m to the northwest and west. A Roman finds scatter is recorded 200m to the north-west (LCS 051) whilst medieval finds have been collected from areas 200m to the north (LCS 0049), 300m to the west (LCS 054) and 300m to the north (LCS 073). In addition various undated crop marks have been recorded on the surrounding agricultural land (LCS 050 and LCS 056). A post-medieval boundary bank is recorded 400m to the north-east (LCS 114). In the wider vicinity there are numerous sites relating to the World War I and II defences of the Suffolk coastline (e.g. LCS 112, a WWII command post trench and pill box). The site also lies in close proximity to known areas of medieval settlement. The village of Leiston existed at least as far back as the Anglo-Saxon period and is recorded in the Domesday Book as Leistuna. The manor and Soke of Leiston was granted to Leiston Abbey at its foundation in 1182 by Raulf de Glanville; the Soke of Leiston was a territorial unit, which probably pre-dated the Norman Conquest and that included the whole of the parish of Theberton and parts of Aldringham cum Thorpe and Knodshill. The original Abbey convent was located near Minsmere, 3.4km to the north (LCS 002) 6

before moving to its present site 3km to the north-west in 1363 and part of the current development area (site LCS148) is within the Abbot s own demesne lands. Minsmere Haven, once the mouth of the Minsmere River was a shelter for boats and the chapel at the original abbey s Minsmere site was dedicated to St Nicolas, the patron saint of sailors, reflecting the local population s strong association with the sea. By the 13th century a medieval settlement also existed at Sizewell; this settlement rivalled Leiston in size and was granted a market in 1237. Following on from the relocation of the Abbey, the market was also removed from Sizewell to Leiston in 1391 because it had ceased to be of any value and the manor had become impoverished through misfortunes. These misfortunes are likely to have included loss of land through inundations from the sea and records attest to the disappearance of at least sixty acres and, in a separate occasion during the 1340 s, 300 houses were sunk. Fishing however remained vibrant, in part due to the addition of Dutch expertise, and in the early 16th century the settlement was still of a comparable size to Leiston as indicated by numbers of taxpayers. The decline of Sizewell through the 16th and 17th centuries appears to have been at times a rapid process largely caused by flooding and coastal erosion (Breen 2013). The focus of the medieval geography of the immediate area was different from that of today; the Sizewell Gap Road was an insignificant drift way, a short cul de sac which only became the road linking Sizewell and Leiston in 1806. Prior to this the route between the settlements ran north from Leiston and followed the south side of Leiston Common before dropping south (diagonally across Pill box field) to a bridge that crossed a watercourse (now part of Sizewell Belts) immediately north-west of the excavation (Fig. 2). Early on the bridge was replaced by one positioned slightly further to the north; this change is likely to have occurred in the medieval period as both bridges are named after tenants who are known to have lived in the 14th century (Breen 2013). The site west of Sandy Lane (LCS148) formed part of Leiston Warren and part of the Abbot s own demense lands which following the Dissolution became Crown Land, whereas the area to the east of the lane (LCS150) had become separated from the Abbot s demesne prior to the Dissolution and become manorial copyhold which was subsequently sold on to a number of copyholders. 7

a) Hodskinson Map of Suffolk 1783 Sizewell Gap appears as an annotation on the map just to the right of the site which is outlined in red and the junction between the Gap road and Sandy Lane is shown just as they appear today. Sandy Lane provided access to Leiston Wet and Dry Commons and as a route was clearly more important than it is today. Leiston Commons were enclosed in 1824-5 b) Tithe map for the parish of Leiston cum- Sizewell1841 The site lies in the corner of Field 254 which at the time was Crown Land and not listed in the tithe apportionment, but is named as Sizewell Vent on maps of 1777 and 1806. Field 271 was recorded as Pit Field and was part of an unnamed farm of 180 acres, one of several holdings in the ownership of Francis Hayle (IRO ref FDA164/A1/1b) c) First Edition Ordnance Survey 1882 This suggests that entire area of the excavations were open fields. Broom Covert, is shown as a smaller area and does not include that area of trees that covered the north end of field 202 that were sampled in the evaluation. Butchered rabbit bones (once a luxury foodstuff and indicator of high status) were found in some of the medieval contexts and it is interesting to note the proximity of the area designated The Warren to the site Figure. Historic Maps

Hodskinson s map of Suffolk published in 1783 (Fig. 3a), depicts Sizewell Gap and the Gap Road to the junction with Sandy Lane just as they are today, while to the north of the site the extent of the Wet and Dry Common are shown; Sandy Lane provided access to the commons and was clearly more important than it is today. When the tithe map was drawn in 1841, the site (LCS 150) was recorded as 271- Pit Field and was part of an unnamed 180 acre farm owned by Francis Hayle. (Fig. 3b). The adjacent site (LCS 148) was exempt from tithes as Crown Land and was not listed on the apportionment, but is named as Sizewell Vent on maps of 1777. During the modern period the site appears to have been open farmland and is shown as such by the First Edition Ordnance Survey of 1882 (Fig. 3c). 2.3 Scope of the project During the medieval period, Sizewell was a much larger and more influential market town and trading gateway than it is today. The site lies on what was, in the Middle Ages, the western edge of the town close to the earliest known bridging point of Chapel Brook, the access into the town. Together with the neighbouring fieldwork (LCS 148), the excavated areas exposed a relatively large sample (8200sqm) of the settlement area; this ranged across several medieval plots that exhibited evidence of a range of crop processing, semi-industrial and fishing-related activities on the margin of the town. The archaeological features were well dated by finds, demonstrating a relatively short occupation of the site that peaked at the advent of the 14th century. The period of demise, identified as a drop in archaeological evidence on the site, coincides with the date of the removal of Sizewell s market to Leiston, the relocation of the Abbey convent inland and a time of encroachment by the sea. Alongside the archaeological evidence there is an unusually complete set of medieval land records for the manor of Leistoncum-Sizewell which cover the area sampled by the excavations. The properties and tenancies described within these records have the potential to be precisely located geographically and, together with close dating gained from timber and organic remains recovered from the site, show that there is a high potential to relate the written records to the archaeological one, greatly enhancing the academic value of both. The research agenda for East Anglia (Medleycott 2011) recognises several areas of study which aim to improve understanding of the formation of our medieval urban 9

centres, including changes in their internal layouts, housing densities, and their role as centres of supply and demand. At Sizewell an almost complete cycle of change, accelerated by the effects of the sea, has occurred which has reduced a medieval market-town s physical and commercial prospects to small fishing hamlet that we see today. The aim of this report is to summarise the results of the archaeological fieldwork against this framework and assess the potential of the site s findings to address these study topics. 3 Results of the fieldwork 3.1 Methodology An area of 3680sqm was stripped by a mechanical excavator fitted with a toothless ditching bucket, under the supervision of an archaeologist, which removed the ploughsoil to the top of the archaeological levels. Unstratified finds were collected during the machining and recorded under context number 1000 during the excavation. Sites and spoil heaps were thoroughly surveyed by an experienced metal-detectorist both during the machining and subsequent hand-excavation of features. A temporary cessation in the stripping was caused by the discovery of live ordnance, an incendiary device dropped by the Luftwaffe during WWII. The ordnance was removed and the site assessed by BAC TEC International Ltd, a specialist bomb disposal company, who maintained a presence on site for the remainder of the soil strip. Archaeological features were normally clearly visible. Areas were cleaned and all features were then investigated by hand excavation; generally 50% of pits and postholes, 10% of ditches and 100% of features that could be interpreted as structural or of other specific interest. Additional sections were also placed where required to investigate stratigraphic relationships. Sixty-three samples, consisting of environmental bulk soil samples and preserved timbers, were collected from selected contexts for further analysis. Both evaluation and excavation were recorded using a single context continuous numbering system; numbers 5000-5036 related to the evaluation, 1000-1728 to the excavation and small finds were allocated a specific block of numbers, 2001-2037. The trenches, excavation areas and features were planned with an RTK GPS and Total Station Theodolite. Individual hand drawn feature plans were recorded at a scale of 1:20 10

or 1:50 onto A1 permatrace sheets. Feature sections and profiles were recorded at a scale of 1:20 onto A1 permatrace sheets. Digital colour and black and white print photographs were taken of all stages of the fieldwork, and are included in the digital and physical site archives. The majority of site data has been input onto separate MS Access databases. Bulk finds have been washed, marked and quantified, with the resultant data also being entered onto databases. The site archives are kept in the main store of Suffolk County Council Archaeological Service at Bury St Edmunds under HER No. LCS 150. 3.2 Excavation summary 3.2.1 Overview The site lay in an arable field and archaeological deposits or the natural geology were visible following removal of the ploughsoil. The excavation area was situated on the southern margins of a low-lying palaeo-channel that ran east west. It was sampled by trial trench in the adjacent site (LCS148) and palaeo-environmental survey as part of the evaluation of LCS150. Radiocarbon and environmental samples from the deeper parts of the channel indicate that this was once a water course which became silted facilitating the slow accumulation of peat from the mid-late Bronze Age (cal date1130-930 BC). The uppermost peat dates to the Middle Saxon period but there is a clear and sharp interface between the organic layers and the silt-sands that seal it suggesting that it may have been truncated and buried by ploughing in the medieval period. Whilst the palaeo-channel had become completely infilled by the time of the occupation of the site it still existed as a discernible hollow and prone to flooding, as occurred during the excavation (Pl.2). One hundred and nine cut features were recorded across the entire area of the excavation with particular concentrations over the eastern half and, less so, the west end (Fig. 4). Although during the medieval period the Gap Road did not exist as the link between Leiston and Sizewell it is interesting to note the features occur where the site is closest to the road edge; at the eastern end, the putative Phase 5 building lies within 17m of the road whereas at the west end the site is no closer than 20m from Sandy 11

Plate 1. General view of the eastern end of the site looking south showing a pattern of ditches and features Plate 2. General view of the east end of the site looking west which shows the line of the lowlying palaeo-channel following summer rain 12

Lane. Features were most sparse at the centre of the site where the natural ground level was at its highest and there is a strong likelihood that the features from this area have been a casualty of truncation through modern ploughing and it was noted that the remaining ditch runs were shallow and incomplete. Where best preserved (at the east end of the site) the archaeological deposits had been partly protected by soil build up, the result of plough movements, and discrete areas of patchy clay, indicating the former medieval floor/ground level and the remains of surface-built ovens, which were recorded 0.5m below the existing surface. A limited vertical stratigraphy existed in these areas which were left upstanding during the mechanical soil strip and hand-excavated within soil blocks; over the vast majority of the site however the archaeology could only be identified as cut features once the surface of the natural sand had been exposed. The low-lying situation, at the margin of a linear hollow at the tail of the Sizewell Belts wetlands, meant that the site was waterlogged within 300-400 mm of the excavation surface and flooded across the lowest-lying parts of the site. With the exception of those features excavated specifically to access water, all of the cut features were shallow stopping at or just within the existing groundwater level. This suggested that the height of the water table at the time of the excavation was similar or slightly above that of the medieval period. Where features breached the water table, preservation of organic material was good and this together with the concentrations of mineralised iron within the natural sand was indicative of the site being historically waterlogged. The archaeological features were the result of occupation between the 11th and 14th centuries. The early medieval phase consisted of a series of a sequence of narrow ditches running down to the edge of the palaeo-channel. The early ditches were welldated by pottery to the 11th-12th century although other feature-types relating to this initial occupation were sparse. None of the building evidence which was encountered on the adjacent site (LCS148) for this period was recorded here. This apparent low intensity of occupation was reflected in the pottery totals and more than twice as much early medieval pottery was collected from the LCS148 site than from here; the early medieval pottery accounted for only 2.6% of the total LCS150 assemblage a ratio that was lower than that of the adjacent site. From around the end of the 12th century, a more intense occupation phase occured; initially with pit-digging on the (marginally) higher ground on the south edge of the excavation, but during the 13th century the activity developed on the lower lying area 13

and included two posthole buildings, clay-built ovens, a timber-lined well and similarlylined large cisterns or steeping tanks; a second less concentrated and defined group of occupation features also occurred near to the corner of Sandy Lane. The main group of features suggested semi-industrial activity exploiting the freshwater resource and was confined to a narrow section of the site spanning c.30m. The extent of the intense feature spread was narrow and it is thought that these features lay entirely within a single property; the area covered by the spread of features was similar to the plot widths observed at LCS148 site (Gill 2013) although here not enclosed within a bounded area. For this high medieval period, the pottery quantities were greater than the adjacent site and the variety of vessel types broader. The pottery assemblage included caches of complete or near complete vessels, mainly jugs, which had been discarded into two of the wells when they were abandoned. Amongst the other finds the animal bone assemblage despite being small shows a rich diversity of species, including fish and sea birds reflecting the coastal position; there were good organic remains and most significantly jointed planking from the hull of a 13th century boat, which had been reused to create the well linings, and a wall plate from a modest building of similar date. There are no features and very few finds that postdate the mid-14th century indicating a probable total abandonment of the site by this time. 3.2.2 Phasing Approximately 56% of the cut features could be dated by pottery and or small finds, all of which were attributed to a relatively short-lived period of occupation of the site between the 12th to 14th centuries. The wet environment and unstable nature of waterlogged sand meant that the open features existed fleetingly and this transience was apparent in the multiple re-cutting of some of the ditches and short date range of some of the pottery assemblages. Short stratigraphic sequences within local feature groups allowed the evidence of this high medieval activity to be sub-divided further; particularly in the low-lying areas in the centre of the site where a very limited vertical stratigraphy existed. Using a combination of pottery dating and stratigraphic and spatial relationships it was possible to assign 82% of the features to one of the seven chronological phases listed below. 14

N palaeo-channel Crown Copyright. All rights reserved. Suffolk County Council Licence No. 100023395 2014 0 20m Plan Scale 1:120 Figure 4. Overall plan showing all features 15

The site appears to have been open farmland since the 14th century and its only known land use history is one of continuous cultivation. Modern ploughing has resulted in the damage and removal of the upper levels of archaeological deposits. Other factors however, such as animal disturbance, buried services or drainage trenches, were limited meaning that the majority of features were well preserved and provide secure contexts for the material dating evidence recovered. The occupation is clearly contemporary with the activity on the LCS148 site and the phase numbering is common to both sites. A site matrix has been prepared (Appendix 3) and the site is described by phase in the next section. Phase 1: Prehistoric Phase 2: Pre-Conquest 10-11th century Phase 3: Early medieval 11th 12th century Phase 4: Medieval 12th 13th century Phase 5: Medieval mid-13th century Phase 6: Medieval late 13th-14th century Phase7: Post 14th century 3.3 Site description by chronological phase 3.3.1 Phase 1 Prehistoric The environmental assessment (Hill, Geary and Smith, Appendix 14) suggested that the dynamics of water in the palaeo-channel slowed, probably due to shifts in local drainage patterns, to a lower energy depositional force sometime in the later Bronze-Age (1130-930 Cal yrs BC; SUERC- 19651) which allowed the development of peats along this former water course. Radiocarbon dating shows a continued accumulation of peat throughout the following 1500 years and the preliminary assessment of pollen grains, beetles and diatoms preserved within the peat profile indicates that the palaeo-channel continued to exist as a shallow body of fresh water and throughout this time was surrounded by grazing meadows. The earliest evidence of human activity on the site comprised fifteen struck flint flakes; the flint working was poor quality and therefore likely to be of an Iron Age date, and all were recovered as residual material in later contexts. 17

3.3.2 Phase 2 pre-conquest-10-11th century The organic accumulation within the palaeo-channel stopped just after AD440-630 (SUERC 19649) and the shift from organic to inorganic silts/clays at the termination of the pollen record suggests increased fluvial influence during the early medieval period which led to increased erosion of material onto the site and it is suggested that this could signal agricultural activity destabilising the soil or it may be linked to factors such as climate and sea-level change. The earliest pottery was 10th-11th century Thetford-type ware; the pottery was collected from two contexts which also produced later medieval material. Six sherds (five of which came from one context, 1265) were recovered from this site to consolidate the twelve recovered from LCS148 and attest to an Anglo-Saxon presence in the vicinity, but there were no features assigned to either of these phases. 3.3.3 Phase 3 11th-12th century to Phase 4 The earliest features on the site, as determined by either stratigraphic relationship or pottery spot date, are shown in pale blue in Figure 5 together with the slightly later Phase 4 features in green and undated features in grey. The plan shows a low density of small, shallow ditches running either at approximate right-angles towards the edge of the palaeo-channel or skirting the edge of the palaeo-channel just above the wet ground. None of the ditches were more than 250mm deep and all were filled with pale, leeched, grey silts mottled with iron panning indicative of past waterlogging. Some of the ditches appeared to be paired (1473 with 1372 and 1468 with 1669) suggesting track or drift ways; the pattern of ditches at the west end of the site bears some resemblance to the layout of the existing Sandy and opposing Home Farm lanes, both of which are historic routes whilst the trackway implied by ditches 1372 and 1399 is orientated towards the putative site of the earlier crossing of the Sizewell Belts. All of the features were either ditches or other boundary markers and there was no evidence of pits, buildings or other occupation-type features. The suggested trackways were 6m and 11m wide; the ditches of the narrower one funnelled out at the junction with the east-west track suggesting that the two were linked. The ditches which defined the tracks were shallow and the recorded lengths represent only the truncated bottom of surviving segments and there was no evidence of the track surface itself. Although not continuous the shallow cuts 1399 (identified as 18

N palaeo-channel 1255 1279 1018 1559 group 1446 1201 1016 1354 1338 1409 pit group 1387 1377 1150 1033 1399 1372 1473 1468 1669 Crown Copyright. All rights reserved. Suffolk County Council Licence No. 100023395 2014 0 20m Plan Scale 1:120 Figure 5. Overall site plan highlighting the Phase 3 (pale blue) and phase 4 (green) features. Undated features are shown in black and subsequent phases in pale grey. 1

5025 in the evaluation), 1454 and 1473 were all considered to be variations of the same ditch and paired with 1372. No individual ditch section produced more than three sherds of pottery, mostly early medieval and medieval coarsewares but the combined assemblage from all of the ditches assigned to this phase tended towards the early end of the date range for the pottery implying that the ditches were open during the 12th century. Ditch 1468 produced eight large pieces of fired clay that were similar in appearance to triangular loomweights of Iron Age or early Roman date, but may be pieces of some other rectilinear object, such as a kiln bar. The E-W track way ditches terminated where they fed into a contemporary north-south aligned ditch 1377. Ditch 1377 was paralleled by ditches 1016 and 1279, which were also the earliest features in their respective sequences, and together formed the boundaries of an enclosed area. Within this area and aligned with its boundaries was a fence, in the form of an alignment of postholes (Group 1446) which ran down to the edge of the palaeo-channel which continued and extended along a line of a narrow slot 1150. The postholes produced very little in the way of finds but none dated later than 11th-12th century. 3.3.4 Phase 4 12th-13th century Features that were assigned to Phase 4 (Fig. 5) all produced pottery that was spotdated to the 12th-14th century and was generally later than that recovered from Phase 3. The features were however the earliest within their respective stratigraphic sequences and therefore it is likely that the pottery dates from the beginning of this date range and Phase 4 denotes the start of the high medieval period of occupation. At the western end of the main site the occupation evidence comprised two north-south aligned ditches 1338 and 1354. The ditches were 15m apart and shared the same alignment as the Phase 4 ditches on the eastern side of the adjacent site. The south end of the ditch 1338 terminated within a baulk that crossed the site; this was also similar to the end-point of the LCS148 ditches suggesting that they were laid out in common. Each ditch produced only a single sherd of pottery, ditch 1354 was cut by crossing ditch 1352 whilst 1338 was cut by 1335 and 1344. 21

Pit Group 1387 At the eastern end of the site restricted to the slightly higher/drier ground close to the south edge of the site were a group of pits (Group 1387). None of the pits were greater than 0.5m deep and many were less suggesting that the high ground water, experienced during the excavation, was similar to conditions in the medieval period. The group consisted of pits 1007, 1009, 1011, 1013, 1033, 1036, 1058, 1101, 1232, 1269, 1289 and 1572. The fills were all sterile looking, pale leeched out sands, and did not contain the hearth and oven debris that was prevalent in the later phased features. The pits produced some pottery, mainly coarsewares, but in general the pottery quantities were low with no more than four sherds coming from any one pit. The exception to this was pit 1033 which had been used for the disposal of rubbish, deposited in a single fill, and over seventy sherds pottery along with food waste in the form of animal bone and shell were recovered from it. A minimum of thirty-six vessels were represented by the pottery fragments produced largely (86% of them) in the same micaceous fabric and spot-dated to the 12th-14th century. Pit 1033 was cut by Phase 5 pit 1062. North of the pits a small cluster of postholes (Group 1201), part of an unknown structure, were also assigned to this phase. The postholes were sealed by pale silts and sealed and/or cut by later features associated with Phase 5 and Phase 6 ovens but were otherwise undated. The group was associated with a sinuous linear hollow 1307, not a cut feature but possibly an eroded one from the posthole structures use, that was filled with a similar silt. The posthole group was located centrally between a pair of narrow, gently diverging ditches 1255 and 1559 and because of this they have been phased together; the ditches combined produced three sherds of pottery post-dating the start of the 12th century. Well 1018 To the east of the postholes was a possible well 1018. It was a broad, shallow circular pit 2.0m in diameter which had become silted up with pale washed sands; the upper fill, 1019, was stained heavily with mineralised iron. The pit was cut into the water table and the lower fill 1020 was running sand which contained several fragments of preserved timber. The wood pieces were the remains of a collapsed lining from the south side of the well and consisted of two plank fragments, each approximately 1m long, and three round wood stakes (Group 1223, pieces1451-1453 and 1456-1457). The timber remains suggested a simple structure; square in plan and built without apparent 22

carpentry jointing, whereby the planks, laid horizontally were retained behind upright stakes that had been were driven into the natural sand at the base of the pit. The well was no more than 1m deep and the top of the wood was 320mm below the soil strip level (top of the natural). The timber remains are the bottom planks of one side and the absence of further timbers suggests that the remainder of the lining structure was retrieved when the well silted up or collapsed. The plank fragments were reused, they had been salvaged and cut down from a previous structure but retained some woodworking technology from their past use. Plank 1451 had a peg-hole but top and bottom edges were missing and Plank 1452 was shaped, possibly a 'prow piece' (a brace/seat/holdfast from a small boat Robert Simper, local maritime historian pers comm) that was fastened with dome-headed pegs. On specialist advice piece 1451 was recorded and discarded, 1452 has been retained for analysis. The well pit was cut by the Phase 6 ditch 1183 and cut the Phase 3 ditch 1016; the find was a residual sherd of early medieval shelly ware pottery. 3.3.5 Phase 5 mid-13th century The most developed phase of occupation occurred during the second half of the 13th century (Phase 5) through to the first half of the 14th century (Phase 6). During this period there appears to have been a progression in the spread of feature-digging northwards towards the line of the palaeo-channel and the erection of a building close to the channel edge (Fig. 6). The spread of features towards the channel may indicate that the ground was drier, as a result of a drop in the ground water level, and the two ditches 1106 and 1374, which run away from the building, are likely to have been originally excavated to maintain it so. The most significant development in Phase 5 was the construction of a small timber building (1092) within the plot in the eastern half of the site and the increase in activity which was reflected in the increased number of finds and occupation debris, in the form of more organic fills as well as the occurrence of charcoal and burnt clay, which made up the pit fills close to the building. The presence of charcoal was ubiquitous across all features in Phase 5; to the extent that it appeared that the features (including the building postholes and the ditches) all shared a common fill, which prompted thoughts 23

that there had been a fire or major burning/clearance event. The charcoal-filled features pre-dated the evidence of oven-building that subsequently occurred in Phase 6. At the western end of the site there were fewer features and evidence of activity was less common; the features counted amongst their number an incoherent spread of small postholes and east-west ditch, 1352. Finds quantities were generally low, the exception to this was a wood-lined well 1365 which produced a cache of buried pottery vessels; the single largest assemblage of pottery from the whole site. Building 1092 Evidence for building 1092 survived as an almost complete footprint; it covered an area of 9.5m x 5.6m and was orientated SW-NE, parallel to the line of the Gap Road. The building s superstructure was composed of seventeen postholes which made up the perimeter walls and included a possible further seven extraneous postholes within its footprint (Fig. 7 and Pl. 3). The north and east walls were made up of four and three posts respectively; the post being widely-spaced at regular 2m intervals. In contrast the south wall is made up of an array of closely-spaced smaller posts; there were two distinct fill types and possibly two structures, or a repair phase, are represented in this particular wall. The earlier postholes included much burnt material (1076, 1099 and 1131) and the supposed later ones (1127, 1180, 1289, 1293, 1297, 1299, 1301, 1303, 1459, 1479 and 1538) were packed with yellow clay; this was similar to the vestiges of clay that made up the surface to the south of building and extended, in traces, over the building s floor area. The clay infilling of the postholes was in the form of a solid block, rather than packing around/securing a post, suggesting that these were foundation pads. The position of the twin postholes on the south wall are not paired with, or opposed to, postholes in the north wall. All of the postholes were c.300mm across; the floor height was indicated by the level of the subsequent Phase 6 hearths which suggested that the postholes were 300mm deep. Any evidence of the building s west wall had been lost, removed by a later ditch which also removed one of the north wall postholes. A 2m long fragment of a wall plate jointed to accept roof-rafters (Pl. 4), was recovered from a nearby pit and was conceivably part of the building (see pit 1503). Outside the line of the building s walls, but near the southeast corner was a waterlogged pit 1591 which contained a large piece of preserved timber, 1593. This lay flat across the base creating a floor to the pit and possibly acted as a post pad or a foot; a load- 24

N palaeo-channel 1248 1106 1352 1092 1365 1103 1503 1374 1712 1515 Crown Copyright. All rights reserved. Suffolk County Council Licence No. 100023395 2014 0 20m Plan Scale 1:120 Figure 6. Overall site plan highlighting Phase 5 features (in blue);subsequent phases are shown in pale grey and unphased features in black. 2

Plate 3. Phase 5 posthole building 1092 looking east with an earlier ditch in the foreground. The divisions on the scales are 0.5m. Plate 4. (left) Part of a wall plate from pit 1503, mortised for roof rafter along its top face; an extremely rare survivor of vernacular joinery from a humble building of the c.13th century, Plate 5. (right) Fragment of oven floor, 1311, pierced with closely-spaced stake holes; photographed looking west, the scale divisions are 0.5m. 27

1297 1099 1293 1299 1131 1301 1303 1171 1076 1127 1103 1180 1023 1289 1103 1106 0 2m 5m Plan Scale 1:75 N 1581 1092 2 1591 1538 1459 1479 1380 1503 1374 1220 Crown Copyright. All rights reserved. Suffolk County Council Licence No. 100023395 2014 Figure 7. Plan of Phase 5 Building 1092 and associated features.

spreader to prevent sinking into the soft wet sands. This was the only one of the pits/postholes which included or required this apparent reinforcement of its base. Pit group 1445 A succession of pits was located close to the building and was thought to be directly related to the building s occupancy or use; five pits in all have been assigned to this group of which four are closely spaced together alongside (c.2m from the south of) building 1092. The pits were phased together on basis of their stratigraphic relationships supported by their pottery spot-date. Unlike the sterile-looking, silt-filled pits which have been assigned to the earlier phases in this area, the Group 1445 pits (1023, 1103, 1171, 1220 and 1581) were all filled with occupation debris. Charcoal and burnt clay/daub was common in each and all produced a mixed finds assemblages (pottery, animal bone, shell and burnt clay) in which the pottery post-dated the late 12th century. The pits were all shallow sided and cut into the natural sand without the timber lining or structure seen in some of the more specialised features, and their function was thought to be simply for the disposal of domestic rubbish. In Phase 6 there were several ovens in this area with associated pits in which there was oven debris; the clay and charcoal fill of the Phase 5 pits looked similar to that of the later pits suggesting that they too were derived from ovens, however, no such hearths were identified within the area during Phase 5 (the pits within group 1445 were separated stratigraphically from the Phase 6 ovens beneath a patchy surface of clay (1105)). Two of the pits within group 1445, pits 1103 and 1220, cut Phase 4 features. The pits varied in size from c.0.9m across (pits 1023 and 1103) to a large elongated pit 1171 which was 4m x 2m across; all of the pits however were 0.5m deep or less and it seem likely that pit-digging beyond this depth was restricted by the level of the ground water. The feature depths were recorded from the top of the subsoil but originally the pits were cut from a ground surface higher up; this is dramatically illustrated by pit 1023 in which a coarseware jug(1025), laid on its side on the base of the pit, had been sliced off along its mid-line where the pit had been truncated by post-medieval ploughing. Pit 1503 and associated ditch 1374 A large waterlogged pit, 1503, was located 17m west of building 1092. The pit was very broad (4.0m wide x 9.0m long) and 0.7m deep so that it was cut below the current 29

(summer 2008) water table. The pit was connected to contemporary ditches 1374, 1380 and 1399 that fed into the pit and the ditches extended eastward to terminate within 7m of the building. The line of ditch 1374 was extended by a shallow linear depression 1145 within the building which may have been a continuation of the ditch. The primary fills of the ditches, the pit and depression were all similar and consisted of charcoal with burnt clay which appeared to be a common single deposit. Within pit 1503 this burning layer was numbered 1585 and was up to 0.45m deep and therefore represented a large amount of material. There was no indication of in-situ burning and the charcoal seems to have been a dumped deposit. Over the charcoal the top of the ditches and pit were infilled by a contrasting pale silt, an alluvial or waterlogged deposit impregnated with iron staining. Forty five sherds of medieval pottery were collected from the pit 1503 in total; these were mainly medieval coarsewares but the presence of Hollesley-type and Hollesley glazed wares indicate a late 13th century date. The most significant find was a fragment of a building wall-plate; a component of a building s timber-frame recognisable from angled mortises on the outer edge of its top face which would have housed the rafters of a pitched roof. The joints were 0.45m apart and the wall plate fragment was 2m long (the timber s end had rotted off so the true length is unknown); it was made from a wood susceptible to insect attack rather than oak; this use of an inferior timber is indicative of lower status vernacular from the early years of the medieval period and it is therefore a very rare survivor. There was no mortising on the underside of the wall plate for wall studs suggesting it was perhaps part of an open-sided shed. In the pit the timber lay on the bottom, angled across the entrance to the adjacent ditches and it was speculated that may have acted as a dam or a sluice, although there was no other structural evidence to support this. The ditches extending from pit 1503 produced similar pottery assemblages to that of the pit, but in much-reduced quantities. The ditches, 1374 and 1380, were sealed by the Phase 6 oven (1310) and cut by clay-filled postholes 1434 1472; pit 1503 was cut by pit 1430 which produced no finds. Ditch 1106 Ditch 1106 ran eastward from the edge of building 1092. It was aligned NE to SW, approximately parallel with the Gap Road, and set back 18m from the current road 30

edge. The west terminal of the ditch turned southward close to the building to leave a 2m interval between it and the building s NE corner post. The ditch was 2m wide with a square cut sides, a flat base and was no more than 0.6m deep. It was sectioned in two places close to the building and the soil profile was similar in both sections; the lower fills (1107 and 1167) were made up of dark grey/black silt and sand (waterlogged) and produced similar distinct bone assemblages (made up of only cat and fish bones (cod) a cameo of a domestic drama?), wood fragments and some heat-altered flint. The upper fill (1168 and 1108) was a mid-brown silt which included burnt and unburnt clay lumps, heat-altered flint, charcoal flecks, oyster and other shells. The pottery collected from the ditch was predominately from the upper fills (forty-six sherds as opposed eight from the lower fill); the upper fills producing Hollesley-type wares from late 13th century whereas the lower fills tended to be earlier greywares. Ditch 1106 cut the Phase 3 ditch (1016) and was in turn cut by ditches from Phase 6 (1183) and Phase 7 (1441); part of the later Phase 6 ditch network (ditch 1003) paralleled the line of ditch 1106. Well 1248 Ten metres to the north of ditch 1106 was an isolated pit 1248. The pit was broad and shallow (1.7m x 0.24m deep) and excavated into the pale leeched out sands within the margins of the palaeo-channel. The pit fill was brown sand with iron pan staining which was waterlogged and unstable at the time of the excavation. The pit was probably created to access ground water, but it was no more than a simple hole without a lining or structure and it produced only ten pottery sherds which were dated to the late 13th century; no animal bone was found. Western area Evidence of occupation in the western end of the site was sparser and made up of a low density of dispersed pits and postholes, a well and part of a ditch system numbered 1352. The south edge of the palaeo-channel encroached onto the northern end of the site here and was identified as an area of leeched white and heavily mineralised sand. Two soil changes were identified in the natural subsoil within the depression that was the palaeo-channel and these were in effect tidemarks attesting to fluctuations in the water level. The E-W ditch 1352, and later the Phase 6 ditch 1335, lay near and paralleled these expressions of the channel edge and apart from well (1365), the 31

channel was devoid of cut features from the later phases. Only a 0.35m depth of ditch 1352 survived; it was filled with pale grey sand and produced only two sherd of pottery from context 1357. The pottery date was early medieval (11th-13th century) but is likely to be residual as the ditch cut and post-dated Phase 4 ditches 1338 and 1354. Well 1365 Timber-lined well 1365 had been created by sinking the sides of a redundant wooden cask (1662) into the low-lying ground on the margins of the palaeo-channel. It was constructed in waterlogged ground so that the sides and the fill of the features became running sand when opened, which made it challenging to excavate and record. The feature was particularly significant in that it included a cache of whole and near whole pottery vessels. The well was 1.02m deep from the excavation (subsoil) surface at which point it was 1.35m in diameter; this width however included the well s construction pit and the lined well shaft itself was only 0.7m across. The lining was first encountered at 0.5m below the surface; at this depth the barrel was set tight against the natural sand and the funnel shape of the cut of the feature above this point suggested that the cask had been worked into the waterlogged ground, to half its depth, from the bottom of a more bowl shaped pit; the well builders presumably having to contend with the same liquid sand as the archaeologists below this level. A 300mm depth/height of the barrel/cask lining survived pressed into the bottom and consisted of thirteen vertical staves (all of the barrel s component pieces are individually numbered 1634-1673). The structure seems to have been placed into the well as a complete cask (minus the lid and bottom) as the fragmentary remains of the wooden hoop (1646 and 1647) was recorded, pegged around the outside of the staves. An inner hoop (1640 and 1649) was also recorded, located 180mm up from the bottom of the staves, which was presumably an adaptation for the well rather than an original feature of the barrel. The barrel was approximately circular, but flattened on the SW side where the well sides had begun to cave in; it was 0.68m across and the staves were 150mm thick. Six staves were selected by wood technologist Richard Darrah to be assessed for dendrochronological dating but were not suitable for reliable analysis (Appendix 11). 32

Plate 6. Well 1365 containing a demonstrably contemporary suite of pottery vessels; the forms represented are jars, jugs, bowls and a cistern. The fabrics are mainly 12th-14th century coarsewares but the deposition of the group can be more narrowly dated to the L13th-14th by the presence of a Hollesley-type jug. Plate 7. Pit 1025 contained a coarseware jug that had been sliced off along its mid-line by subsequent post-medieval ploughing; a dramatic illustration of the depth of truncation. 33

Although only the bottom half of the well lining survived, the pattern of infilling indicated that the timber had originally extended to the full height of the excavated hole. This was apparent in both the excavation cross-section which showed the fill which had been within the barrel (context nos 1364 and 1367) as distinctly column of dark soil; the vertical-sides extended up from the surviving lining and manifested as a dark circle in plan on the surface. Outside the line of the barrel sides the construction pit had been backfilled with pale sand (1366); presumably this was spoil from the original excavation of the hole, re-packed behind the barrel once it had been set in place. The fills inside the lining were waterlogged and rich in organic material; bulk samples (S136, 137 and 138) were taken and have been processed. The survival of the plant macrofossils were good but the findings were unremarkable and probably reflected the surrounding environment rather than a particular well use. The pottery was a primary discard deposit, which accounted for c.40% by weight of the total recovered from the entire site, including several distinct broken pottery vessels and five complete or near complete examples. The pottery was deposited in a layer midway down the well shaft, demonstrating that the well had already silted up by half its depth before the pottery was dumped and occurred above the surviving timber lining at a point where the well shaft was dry. The pots were planned and photographed in situ and each vessel given a context number (1609-1626) to locate them on the drawing; they occurred only in the north half of the well, packed together against the NE side and apart from pot 1626 which lay beneath, were arranged in a layer one pot deep. Deposited all together the pottery is a demonstrably contemporary suite of vessels and the forms represented are jars, jugs, bowls and a cistern. The fabrics are mainly coarsewares that normally can only be dated broadly to 12th-14th century but in this case the deposition of the group can be more narrowly dated to the L13th-14th by the presence of a Hollesley-type coarseware jug. The construction pit backfill (1366) and the basal layer (1651) which relate directly to the foundation of the well, produced only two sherds of medieval coarseware. 3.3.6 Phase 6 late13th-14th century The greatest level of activity, measured in the number of features, occurred around the turn of the 14th century (Fig. 8). The Phase 5 building 1092 was probably demolished and replaced by a group of at least three external ovens (Group 1035). These are 34

thought to have co-existed with a twinned pair of sunken, cistern-like structures (1219 and 1730), made from re-used boat timbers, and a small out-building constructed around earth-fast post (1117); features which together suggest an industrial working yard. The features within this group extended across an area of c.30m immediately west of ditch 1183, but it is not immediately clear if the buildings, ovens and wells were contained with a bounded plot or not, but the extent of the feature concentration largely mirrors that of the previous phase. The oven and cistern structures seemed to exist in conjunction with a spread or surface of unfired yellow clay and were located on the clay s north and western edge; the clay in effect formed a promontory extending out into the marginal wet environment and the ovens and cisterns accessed via trampled track-way 1041. The ditches which surround the features were shallow and showed signs of being waterlogged and are probably in effect dykes or drains managing the water around the working-area rather than demarcating a boundary. The clay did not exist as a coherent solid surface but rather survived as a slumped deposit, capping off the pits and features of the previous phase; as a coherent layer it was best seen in section on the south edge of the site. The presence of a large angled post (1044) on the eastern side of the clay and recorded in section on the edge of the excavation, suggests that the clay and the sand track-way 1041 were retained within a possible post revetment. At the western end of the site the features are less intelligible and consist of a low density of pits and postholes and part of a ditch system (ditch 1335) which paralleled the edge of the palaeo-channel. The postholes and pits from this area were assigned to this phase because they either contained at least one sherd of pottery that post-dated the first half of the 13th century, or were the latest features within their immediate stratigraphic sequence. Beyond this it is difficult to identify associated features but post holes and pits (1679, 1681, 1687, 1698, and 1707) have been grouped together under the number 1447 based on their shared alignment and even spacing. The arrangement of ditches 1704 and 1465 suggests part of an enclosure with an entrance at the corner; ditch 1465 is on the projected line of ditch 1106 of the east end of the site suggesting a possible association between the two. 35

Oven Group 1035 The remains of three large ovens, 1141, 1310, and 1442 were recorded within a 10m space, the ovens aligned with and were located on and followed approximately the edge of the wet palaeo-channel (Fig. 9). The ovens would have been constructed on top of the medieval ground surface, over the then topsoil, and the structure of their firing chambers would have been wholly above ground. As such the remains existed relatively high in the soil profile and normally would be vulnerable to subsequent plough damage. Their part survival here was due to the low-lying nature of this part of the site; had they been elsewhere they would no longer exist. The remains suggested two oven forms and these are described by type below. Ovens 1141 and 1310 Oven 1141 was built directly over a short ditch (1145) from Phase 3 and had slumped into the ditch; giving the appearance that it followed the ditch s contours. An oblong construction pit numbered 1221, 1m x 0.45m, was cut into the top of the ditch fill and filled with a thick deposit of clay, 1222. This was pierced by a series of small stakeholes, 1139 which presumably once secured the timber-formwork on which the oven s superstructure was built. Sixteen stake-holes were recorded in all, located on the edges of the clay. The holes were 45-50mm in diameter and all but two pierced the full depth of the clay 40-60mm deep. In plan the pattern of holes enclosed an incomplete rectangle c.0.80m x 1.0m. The clay at the centre of the oven, 1129, was vitrified and a dark maroon/black colour and overlying this was a variety of clay deposits deriving from the collapse/demolition of the oven s dome that were recorded in sections 1057 and 1140. Consisting of charcoal rich sands and scattered pieces of clay these deposits held a range of medieval material and lay in a broad spread either infilling the oven itself (1130, 1428, 1429) or lying above it (1081, 1083, 1105, 1580, 1587) and extending southwards. Lying c.8m to the west, 1310 was an overall number issued to the remains of a second clay-built oven. The oven structure, 1623, was very similar to 1035 although only the oven floor, which appeared as a surface of partially heat reddened clay, survived. The area of the clay measured c.2.2m by 1m but this was merely a fragment of what was likely to have been a much larger structure (Pl. 5). Oven 1623 was built over the infilled terminus of ditch 1374 and lay above a basal layer of mixed sands, 1625. A series of sixteen small stake-holes, 1311, were cut through the oven base. The holes were 36

N palaeo-channel 1141 1533 1531 1142 ovens 1035 1183 1335 1219 1730 1041 group 1447 building 1117 1465 1601 1461 1604 1704 1726 0 20m Crown Copyright. All rights reserved. Suffolk County Council Licence No. 100023395 2014 Plan Scale 1:120 Figure 8. Overall site plan highlighting the Phase 6 features (red) and clay surface in yellow. The Phase 7 features are shown in pale grey and unphased in black.

between 35-50mm across up to 100mm deep and filled with brown silty sand with charcoal; they were closely spaced and located within the burnt/reddened are of the clay and but did not fall into a discernible pattern. Oven 1442 Phase 7 ditch, 1441 sliced through the centre of oven 1442 destroying it almost completely leaving only vestigial remains to be recorded in the ditch s sides (section 1057). Oven 1441 post-dated oven 1141 which was constructed 1.5m to the east and cut the spread of clay layers associated with the earlier ovens use. Oven 1442 was a type thought to be for drying or the malting of grain and took the form of a pit, 1.3m and about 0.6m deep; this is a different kind of oven from those described above (ovens 1141 and 1310). The pit had tapering-sided which were lined all around with clay to create walls 100-150mm thick, the clay (layers 1078 and 1080) was burnt red to the full height and the depth of burning was in places through to the full thickness of the clay. The later ditch cut though the centre of the oven chamber removing the floor of the oven pit and any evidence of its firing. The oven was located at the west end of the Phase 5 building 1092 on its centre-line but this appears to be coincidental and the two are not thought to be related. Water-pits/cisterns: pits 1133 and 1172 To the south of the ovens and west of the clay spread lay two large irregular pits, 1172 and 1133, each about 3m across. These were construction pits or enabling works which allowed two, square, timber box-like structures to be set into the underlying waterlogged sands to create sunken water tanks or cisterns (Pls. 8 and 9). The two tanks (numbered 1219 and 1730 respectively) were c.1.5m square; the surviving timber remains were first encountered at 240mm below the excavation surface and the tank in total would have been about 1-1.2m deep. The tank s sides were assembled from recycled sections of a boat hull, cut to length and held in place against an internal frame. The frame was post and rail type; the post made from stout square-sectioned timber, the rails (in 1219) from bent-wood branches ( in 1730 the wood was more processed). The posts were mortised and the horizontal rails tennoned, with a simple single-faced shoulder, and pegged into place and the whole structure exhibited a measure of woodworking skill. The frame structure (at least) had been prefabricated and lowered into the construction pit complete; the squared-off post bases rest on the floor of the excavations. The planks were positioned behind the frame, shoring up the sides of the excavated hole, and were 39

Plate 8 Water tank 1219 (pit 1172), assembled from recycled section of boat, during excavation looking north. The horizontal scale is 2m long Plate 8. Water tank 1219 (pit 1172), which was created from the hull of a boat, during excavation. The photograph is taken looking north and the horizontal scale is 2m long Plate 9. Base of water tank 1730 viewed from the SW. The frame was prefabricated outside the excavation for the well before it was lowered in to simply rest on the floor of the construction pit. 1m scale Plate 9. Bottom of water tank 1730 showing the planks and internal frame, this is the bottom of the structure which simple rested on the floor of the well pit Plate 10. Side of water tank 1219 made from a section of boat hull. The shaped planks were fastened together along their long edge with iron roves, the joints caulked with twisted sheeps wools. Plate 10. One side of water tank 1219 made from a section of boat hull. The carefully-shaped planks were fastened together along their long edge with iron rivets and the joints caulked with twisted sheep s wool. 40