Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain [1 ed.] 9781784912383

The appearance and revival of handmade grog-tempered ware producing pottery industries during the late 3rd and 4th centu

149 36 2MB

English Pages 188 Year 2016

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

Late Roman Dorset Black-Burnished Ware BB1: A Corpus of Forms and Their Distribution in Southern Britain, on the Continent and in the Channel Islands 178969955X, 9781789699555, 9781789699562

Much has been written about Roman Dorset Black-Burnished Ware (BB1) and its Late Iron Age Durotrigian origins since the

145 1 6MB Read more

Roman Britain

560 102 383KB Read more

East Asian Screen Industries 9781838711375

East Asian Screen Industries is a guide to the film industries of Japan, South Korea, Taiwan, Hong Kong and the PRC. The

177 113 9MB Read more

South-East Asia in Turmoil

642 56 27MB Read more

The Stakes of Democracy in South-East Asia (RLE Modern East and South East Asia) 9781138901599

226 72 549KB Read more

Embassies – Negotiations – Gifts: Systems of East Roman Diplomacy in Late Antiquity 3515106324, 9783515106320

This book offers an original approach to late Roman/early Byzantine diplomacy as a system. Assessing both official and c

101 45 2MB Read more

Wall-painting in Roman Britain 0907764150, 9780907764151

109 106 190MB Read more

Ear-rings in Roman Britain 9780860546054, 9781407318233

182 112 29MB Read more

Agricultural Buildings in Roman Britain 9780860540656, 9781407323428

170 31 123MB Read more

Defence and Decolonisation in South-East Asia: Britain, Malaya and Singapore 1941-1967 1136839011, 9781136839016

This book explains why British defence policy and practice emerged as it did in the period 1941-67, by looking at the ov

175 57 1MB Read more

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain [1 ed.]
9781784912383

Author / Uploaded
Malcolm Lyne

0 0 0
Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

File loading please wait...

Citation preview

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain Malcolm Lyne

Archaeopress Roman Archaeology 12

Archaeopress Publishing Ltd Gordon House 276 Banbury Road Oxford OX2 7ED

www.archaeopress.com

ISBN 978 1 78491 237 6 ISBN 978 1 78491 238 3 (e-Pdf)

© Archaeopress and M Lyne 2015

All rights reserved. No part of this book may be reproduced, stored in retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying or otherwise, without the prior written permission of the copyright owners.

This book is available direct from Archaeopress or from our website www.archaeopress.com

Contents List of Figures�� v Introduction and Acknowledgements�� iv 1: The Late Iron Age and Early Roman Background��1 1.1: East Sussex��1 1.2: Hampshire and the Isle of Wight��1 1.3: East Kent��1 1.4: West Kent��2 2: Industry Group 5. East Sussex Wares��3 2.1: Introduction��3 2.2: A history of previous research��3 2.3: East Sussex Ware fabrics��5 Introduction�� 5 The fabrics�� 5 2.4: Sources��6 2.5: Industry 5A (Bardown ware)��6 The evidence�� 6 The exploitation of raw materials��7 Technology�� 7 A Corpus of forms produced�� 7 Jars�� 7 Bowls�� 9 Dishes�� 9 Lids�� 9 Distribution�� 9 2.6: Industry 5B. (East Sussex Wealden Ware)��9 Sources and exploitation of raw materials��9 Technology�� 10 A corpus of forms produced��10 Jars�� 10 Bowls�� 10 Dishes�� 10 Lids.�� 10 Asham pots�� 12 Distribution (Figure 6)�� 12 The end of the industry�� 13 2.7: Industry 5C. East Sussex Downland ware��13 Centres of production�� 13 Sources of clay, filler and fuel��14 Fabrics�� 14 A corpus of forms produced��14 Jars�� 14 Bowls�� 17 Dishes�� 18 Dishes with handles�� 20 Cups�� 20 Lids �� 20 Large dry-storage jars�� 20 Distribution.�� 20 c. 200-270 AD.�� 20 c. 270-370 AD.�� 23 c. 370-400+ AD. .�� 25 2.8: Industry 5D. (Beddingham/Ranscombe ware)��27 Source and exploitation of raw materials��27 i

Technology�� 27 The date of the industry�� 27 The forms produced�� 28 Jars�� 28 Bowls�� 28 Dishes�� 30 Distribution�� 30 2.9: Industry 5E. Pevensey Grit-and-Grog-Tempered ware��30 Introduction�� 30 Fabrics�� 30 Source�� 30 A Corpus of forms�� 32 Jars �� 32 Bowls�� 32 Dish/Bowls�� 32 ?Jugs�� 32 Distribution�� 32 2.10: The relationship between Pevensey ware and East Sussex wares.��33 2.11: Industry 5F. Barcombe/Burgess Hill vitrified ware��34 Introduction�� 34 Dating�� 34 A Corpus of forms�� 34 Jars�� 34 Bowls�� 34 Dishes�� 36 Flasks�� 36 Mortaria�� 36 Distribution�� 36 3: Industrial Grouping 6. Hampshire Grog-Tempered Wares��37 3.1: A history of previous research��37 3.2: Fabrics��37 Group 6A. Hampshire Pipe-clay tempered wares��37 Group 6B. Hampshire Camouflaged-grog tempered wares��37 3.3: Firing Technology��37 3.4: Industry 6A. Hampshire Siltstone-Grog Tempered wares��38 Source and exploitation of raw materials.��38 Technology�� 38 A corpus of forms produced��39 Jars�� 39 Bowls�� 39 Dishes�� 41 Beakers�� 41 Everted-rim storage-jars�� 41 Lid-seated bead-rim storage-jars�� 43 Rope-rim storage-jars�� 43 Lids�� 43 Trading patterns�� 43 c. 250-270 AD�� 43 c.270-300 AD�� 43 c. 300-370 AD�� 47 c.370-430 AD.�� 47 Long distance coastal trade.��51 3.5: Industry 6B. Hampshire Camouflaged-Grog-Tempered Wares��51 Sources of and exploitation of raw materials.��51 Technology�� 51 A corpus of forms produced��53 Jars�� 53 Bowls �� 53 ii

Storage-jars with everted rims.�� 55 Distribution �� 55 c. 250-300 AD.�� 55 c. 300-370 AD �� 55 c. 370- 400+AD�� 55 4: Industrial Group 7. East Kent Grog-Tempered Wares��60 4.1: A history of previous research. ��60 4.2: Industry 7A. East Kent Siltstone-Grog-Tempered ware��60 The Fabrics�� 60 Sources�� 60 Technology�� 61 A Corpus of vessel forms�� 61 Jars �� 61 Bowls�� 61 Dish/Bowls�� 61 Lids�� 63 Distribution�� 63 c. 270- 370 AD�� 63 c. 370-400+ AD�� 67 4.3: Industry 7B. Richborough/Canterbury Grog-tempered ware��67 The Fabric�� 67 Sources�� 67 Technology and utilisation of natural resources�� 68 A Corpus of forms produced��68 Jars�� 68 Jugs�� 68 Bowls�� 68 Dishes�� 68 Distribution�� 70 5: Industrial Group 8. West Kent Grog and Grit Tempered Wares��73 5.1: Introduction ��73 5.2: Industry 8A. West Kent Grog and Grit-and-Grog-Tempered wares��73 Fabrics�� 73 Sources�� 73 A Corpus of forms produced��74 Jars�� 74 Bowls�� 74 Dish/Bowls�� 74 Beakers �� 74 5.3: Industry 8B. West Kent Sand, Shell and Calcite Tempered wares��74 The Fabrics�� 74 Source and dating�� 76 A Corpus of forms produced��76 Jars�� 76 Bowls�� 76 Dishes�� 76 Flasks�� 77 The Distribution of Industries 8A and 8B wares��77 c .270-370 AD�� 77 c. 370-400 AD�� 79 c. 400-430/450 AD�� 80 6: The Late Roman Grog-Tempered Ware Industries: A Discussion��81 6.1: The organisation of the industries��81 6.2: The reasons for the revival of handmade grog-tempered pottery during the Late Roman period��84 The period to c. 270 AD.�� 84 c. 270-370 AD �� 85 c. 370-400+ AD�� 87 iii

7: The End of Roman Pottery Production in Britain��89 7.1: Previous research.��89 7.2: A decline in the use of pottery during the fourth-century?��89 7.3: The end of Roman Pottery production in southern Britain��92 Introduction�� 92 Kent�� 92 London�� 95 Hertfordshire�� 96 South-Central England�� 97 Appendix 1. Schedule of Quantified Site Assemblages including Late Roman Handmade Grog-Tempered wares�� 100 Appendix 2. Schedule of Site Assemblages examined but not quantified and those quantified but lacking grog-tempered wares�� 109 Appendix 3. The grog-tempered ware industries: gazetteer of examples seen�� 123 Appendix 4. Fabric breakdowns of quantified East Sussex Ware assemblages�� 161 Appendix 5. Breakdown of the East Sussex Ware percentages of quantified assemblages as per vessel type�� 162 Appendix 6. Breakdown of the Hampshire Grog-Tempered Ware percentages as per vessel type�� 164 Appendix 7. Breakdown of Industries 7A, 7B, 8A and 8B percentages of quantified assemblages as per vessel type �� 170 Bibliography�� 174

iv

List of Figures Figure 1: Industry 5A forms.�� 8 Figure 2: Industry 5B forms and 5C.36 storage-jar�� 11 Figure 3: Industry 5C jar and bowl forms�� 16 Figure 4: Industry 5C dish and storage-jar forms�� 19 Figure 5: The distribution of early 3rd century girth-cordoned jar and bowl forms �� 21 Figure 6: Percentage distribution of Industries 5B and 5C wares c.200-370 AD�� 22 Figure 7: Distribution of Industry 5C developed beaded-and-flanged bowl forms�� 24 Figure 8: Percentage distribution of Industry 5C wares. c.370-400 AD.�� 26 Figure 9: Industry 5D forms�� 29 Figure 10: Percentage distribution of Industry 5D wares. c.300-400 AD. �� 31 Figure 11: Industry 5E forms�� 33 Figure 12: Industry 5F forms�� 35 Figure 13: Industry 6A jar and bowl forms�� 40 Figure 14: Industry 6A dish, beaker and storage-jar forms�� 42 Figure 15: Distribution of late 3rd century Industries 6A and 6B jar and lid forms�� 44 Figure 16: Percentage distribution of Industry 6A products c.270-300 AD..�� 45 Figure 17: Distribution of late 3rd century Industries 6A and 6B storage-jar and bowl forms.�� 46 Figure 18: Percentage distribution of Industry 6A products c.300-370 AD.�� 48 Figure 19: Distribution of 4th century Industry 6A bowl and convex-sided dish forms�� 49 Figure 20: Percentage distribution of Industry 6A products c.370-400+ AD.�� 50 Figure 21: The eastern distribution of vessels in fabric 6.1A.�� 52 Figure 22: Industry 6B forms�� 54 Figure 23: Percentage distribution of Industry 6B products c.250-300 AD..�� 56 Figure 24: Percentage distribution of Industry 6B products c.300-370 AD.�� 57 Figure 25: Distribution of vesicular Industry 6B products�� 58 Figure 26: Percentage distribution of Industry 6B products c.370-400+ AD�� 59 Figure 27: Industry 7A forms�� 62 Figure 28: Percentage distribution of Industry 7A products c.270-370 AD�� 64 Figure 29: Distribution of 4th century Industries 7A, 7B, 8A, 8B and 6A.18 bowl forms�� 65 Figure 30: Industry 7B forms.�� 69 Figure 31: Percentage distributions of Industries 7A and 7B products c.370-400+ AD�� 71 Figure 32: Industries 8A and 8B forms�� 75 Figure 33: Percentage distribution of Industries 8A and 8B products c.300-400+ AD�� 78 Figure 34A: North-south transect through Industries 6A and 6B marketing zones. B: East-west transect through Industry 5C marketing zone. C: North-south transect through Industries 7A and 7B marketing zone. D: Transects through Overwey/Portchester D ware marketing zone..�� 82 Figure 35A: Transects through the New Forest greyware marketing zone c.300-370 AD and c.370-400+ AD. B: Transects through the combined Industries 6A and 6B marketing zones c.300-370 AD and 370-400+ AD�� 83 Figure 36: Transects through the Alice Holt/Farnham industry marketing zone c.300-370, 370-400 and 400+ AD.�� 91 Figure 37: Dishes with solid hemispherical bosses.�� 98 Figure 38: The distribution of dishes with bosses, Industry 5E products and Roman towns with grubenhauser�� 99

v

Introduction and Acknowledgements This publication is taken from the author’s PhD thesis presented at Reading University (Lyne 1994) and revised to incorporate the results of further research during the 21 years since. It deals with the Late Roman handmade grog tempered ware industries of East Sussex, the Hampshire basin, East Kent and West Kent, presenting corpora for these various wares, discussing the reasons for their appearance during the late 3rd century AD, increasing popularity during the 4th and disappearance during the early 5th century AD. The original numbering system for the various industrial groupings is retained here, explaining why they they run from 5 to 8 in this publication. Industrial Groupings 1 to 4 are BB1, BB1 imitations, Vectis ware and Other Handmade Sand –and-Grit-Tempered wares respectively: Groupings 9 and 10 concern Rawreth ware and Handmade Shell-Tempered wares. The section on BB1 in the original thesis has been published in modified form within the Bestwall Quarry, Wareham monograph; taking into account more recent work on the subject (Lyne, in Ladle 2012). The sections on BB1 imitations, Vectis ware and the other industries are the subjects of ongoing research. There are frequent references in this publication to the fabric and form breakdowns of numerous site pottery assemblages without any reference to published reports. These pottery assemblages are either unpublished or published without detailed quantification but were examined by this author in museum and archaeological unit stores in a programme of research between 1989 and 1994. I am indebted to the staff of the numerous museums and archaeological units, as well as amateur groups and private individuals visited and worked for between 1989 and the present day: a full list of these can be found in Appendices 1 and 2 (p.---). I must also thank Professor Fulford at Reading University for supervising the original PhD work and making useful suggestions as to improving it.

vi

1: The Late Iron Age and Early Roman Background 1.1: East Sussex

1.2: Hampshire and the Isle of Wight

Unlike the ‘Belgic’grog-tempered wares of Kent, East Sussex Ware does not appear to have had any Continental inspiration. Cunliffe termed it Eastern Atrebatic but its connections with both the Southern Atrebatic industries of West Sussex and the Northern Atrebatic of Northern Hampshire are minimal and perhaps non-existent. East Sussex Ware production seems to have commenced at some time during the mid 1st century BC, with the earliest fabrics including those with additional calcined flint and soapy fine grog filler. Jars make up most of the output and are usually small, poorly made and plain. Some of the more elaborate examples are decorated with black paint and have incised eyebrow and other motifs on their shoulders and elsewhere. These more distinctive forms include the bulbous narrow necked Asham pot type and large storage vessels with raised finger-impressed girth cordons. The combing and furrowing of jars, characteristic of the ‘Belgic’ grog tempered wares of Kent, was not practised by East Sussex Ware potters.

Handmade grog tempered wares feature little in the Late Iron Age of Hampshire and are mainly confined to the north of the county in and around Silchester. The rest of the county was dominated by the Middle Iron Age saucepan pot Worthy Down ceramic tradition up until the beginning of our era. Sand and sand and flint tempered Northern and Southern Atrebatic wares began to replace the Worthy Down saucepan pots after c. 25 BC and appear to have resulted from a fusion of the ‘Belgic’ grog tempered ware and Middle Iron Age traditions. Grog tempered wares ceased being made even in the Silchester area soon after the Roman Conquest and do not feature again in the Roman ceramic tradition of the area until the mid 3rd century AD. These Late Roman grog tempered wares owe nothing to previous ceramic traditions in the area, leading us to ask as to their origins. The answer seems to lie on the Isle of Wight where there was a tradition of producing handmade pottery extending through the period from the Middle Iron Age to the end of the Roman occupation.

There are major problems in the dating of Late Iron Age East Sussex Ware in that very few imported amphorae and later Augustan Gallo Belgic and Central Gaulish finewares are found in the region before the Roman Conquest. The local East Sussex Ware forms are also very conservative and change little until the 2nd century AD apart from the limited copying of Gallo Belgic platters, girth beakers and other forms after AD 43.

Excavations at Havenstreet and Mersley Farm on the Isle of Wight indicate that the Middle Iron Age saucepan pot tradition continued on the Isle of Wight into the earliest years of our era and, unlike on the mainland, made use of pre fired clay grog as a filler. The pottery in this tradition was supplanted by handmade Late Iron Age sand tempered Vectis ware c. 20 AD, production of which continued into the early years of the 4th century AD. Vectis ware began, in turn, to be supplanted by handmade grog tempered wares during the late 3rd century AD. The high percentages of grog tempered of Industry 6A wares in 4th century AD pottery assemblages from the Island strongly suggest that at least some of them were made there.

Production of East Sussex Wares took place on both coastal sea salt production sites in the Newhaven area and at the various iron producing settlements in the eastern Weald: both of these industries were capable of yielding large quantities of underfired clay grog filler from the grinding up of old furnace material. The distribution of East Sussex Wares appears to have been more extensive during the Late Iron Age than later on, with significant quantities being traded over the coastal plain as far west as the River Arun. This western boundary of the trading area shrank back to the River Adur during the early Roman period with Stane Street marking this boundary north of the South Downs.

1.3: East Kent As in East Sussex, there was a long tradition of handmade grog tempered ware production in Kent, with its origins going back to the early years of the Late Iron Age. This so-called ‘Belgic’ pottery is characteristic of the Late Iron Age in Kent, Essex and Hertfordshire and makes its first tentative appearance c. 100/75 BC but does not become predominant until c.25 BC. The wares are characterised by the frequent use of diagonal, vertical and horizontal combing on cooking vessels and storage jars, which make up the overwhelming bulk of the forms produced. These were joined by polished fineware copies of imported Gallo Belgic platters, jugs and other forms after c. 25 BC, with the jugs being frequently and intentionally fired red: cordoned and plain fineware cups

There is no clear division between Early Roman East Sussex Wares and those produced during the Late Roman period. The mid 2nd century AD saw an increasing influence on East Sussex Ware forms by the products of neighbouring wheel using Romanised pottery industries. Jars began to have better formed rims and flanged bowls, lid seated examples and straight sided dishes made their appearance: at the same time, however, Asham pots and girth cordoned storage jars continued being made with little change until the mid 3rd century AD. Because of this, an arbitrary date of c. 250 AD has been selected as the boundary between early and late products. 1

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain and bowls also make their appearance at this time but tend to be fired black like most of the coarse cooking vessels produced. Many of the finewares were made on the potters’ wheel but some were handmade or produced on a turntable.

grey with glassy surfaces and red ‘scorched’ patches. It supplied significant quantities of such wares to much of East Kent during the early to mid 3rd century AD before production tailed off and ceased during the early years of the 4th century.

Several centres of production can be distinguished in Kent: there is one centred on the oppidum at Canterbury and others centred on another oppidum at Loose, Ashford, Isle of Thanet, Folkestone, West Kent and the hill-fort at Oldbury in the Otford area. Most of these production centres used ground up underfired clay or grog as filler but the Folkestone, Loose and Isle of Thanet centres of production had an altogether wider range of fabrics, which in the case of Folkestone included grog, chalk, quartz sand and glauconitic sand in various combinations (Thompson 1982, Lyne Forthcoming C). The wares made in and around the Loose oppidum in the upper Medway valley are overwhelmingly in fabrics with glauconitic sand filler (Kelly 1972) but also include some with such sand combined with sparse calcined flint. The production centre at Minster on the Isle of Thanet made both grog tempered and silty wares (Lyne 2011): those along the shores of the Thames estuary at Highham and elsewhere in West Kent made use of a wide variety of fillers, such as shell, grog and quartz sand both by themselves and in various combinations (Pollard 1988: 39-42). The use of calcined flint as a filler has its origin in earlier periods but lingered alongside the new Late Iron Age ones, employed on some of the new coarseware forms, until sometime between the beginning of our era and c. 25 AD.

1.4: West Kent The best known Early Roman source of grog tempered wares in West Kent was the so called Patchgrove pottery industry, which was probably produced at several centres in the Otford area. These production centres supplied Surrey and Kent west of the River Medway with distinctive bead rim and everted rim jars, as well as other forms, in soapy grey cored oxidised fabrics with or without crushed black grog filler. The jars include large storage vessels, frequently decorated with finger tip jabbed or incised horizontal cordons on their shoulders. Such storage vessels had a considerably wider distribution than the other forms, suggesting that they may gave been used as packaging for an as yet indeterminate local product. Patchgrove ware is regarded as a post Conquest development by Champion (1976: 71) but Pollard (1988: 39) argues that its presence in early pit groups with an absence of Romanised fabrics at the North Pole Lane, West Wickham site (Philp 1973) is indicative of such wares having their origins in the latest pre Roman Late Iron Age. The various production centres west of the Medway went into decline during the period after AD 70 and finally ceased most activity towards the end of the century. They were replaced by a variety of fine sanded, wheel turned greyware producers, including the early Thameside industry, Highgate Wood and other more local centres. As regards handmade storage jars, however, the small Romanised kilns of the late 1st and 2nd centuries AD could not accommodate such large pots: they retained their old coarse fillers and handmade natures so that they could still be fired in bonfires and clamps. The Thameside industry continued producing large shell tempered bead rim storage jars and the Patchgrove one oxidised grogtempered examples; the former until the 170s AD and the latter until c. 270 AD. The shell tempered storage jars frequently have resin on their rims, indicating that they, like the Patchgrove ones, were also used as packaging for some kind of traded commodity.

The distribution zone for these Late Iron Age wares in Kent and elsewhere in the south-east of Britain is an insular extension of a more substantial one in Gallia Belgica on the other side of the Channel, making similar forms in a wide variety of shelly and sand tempered wares from c.100 BC onwards. It may be indicative of the importation of ideas followed by movements of people into Britain during Caesar’s Gallic wars and later. In that part of Kent east of the River Medway, Late Iron Age pottery assemblages are dominated by wares in the Aylesford-Swarling ceramic tradition. Here the handmade grog tempered ware producers continued to flourish throughout the late 1st and 2nd centuries with little change other than the abandonment of body combing and furrowing during the mid 2nd century, the increasing inclusion of sand in the grog filler and firing to slightly higher temperatures than previously. Some of these handmade ware producers operating on the banks of the Wantsum channel adopted superior kiln firing technology during the the mid to late 2nd century AD, resulting in the production of high fired semi vitrified ‘Native Coarse Wares’ in a variety of grog, sand and flint tempered fabrics (Pollard 1988: 98). This industry produced little other than everted rim cooking-pots fired

Significant quantities of handmade East Sussex Ware had been supplied to and were produced in the Weald of Kent during the early Roman occupation and upped their share of markets in the upper Medway valley during the early 3rd century AD before being slowly supplanted by similar products in siltstone grog tempered ware from a new supplier, Industry 7A, after AD 250-270.

2

2: Industry Group 5. East Sussex Wares 2.1: Introduction

(Holleyman 1936) questioned how long the decorative tradition continued into the Roman period:

Roman coarseware assemblages in Sussex east of the River Adur are dominated from the beginning by relatively crude handmade wares of Late Iron Age character with crushed clay-grog filler. Whereas such native wares did not usually outlive the first 20 or 30 years of the Roman occupation in lowland Britain, here in East Sussex and Wealden East Kent they continued to account for a considerable element in pottery assemblages until the early 5th century.

‘The finding of native-made wares side by side with Roman pottery is interesting. How long native potters continued producing vessels embodying native tradition in shape and design in competition with Roman kilns is difficult to say; the association of finger tip impressed wares with a wide variety of purely Roman types principally belonging to the 2nd and 3rd centuries in Cutting III seems to imply that the practice continued (perhaps sporadically) well into the occupation’,

2.2: A history of previous research The crude nature of these handmade grog-tempered wares has led to their study being neglected until fairly recently. In the case of East Sussex Wares, there was an assumption current until the 1960s that they were unlikely to be Roman but were instead of entirely Late Iron Age date. Although the origins of the East Sussex Ware tradition do lie in the Late Iron Age of the region, we now know that the wares continued to be made until after the end of the Roman occupation. The report on the Caburn hill-fort excavations of 1926 (Curwen and Curwen 1927) is well in advance of the usual standards of archaeological reporting for that period. The appearance of this report marks the beginning of the study of what is now called East Sussex Ware and its Late Iron Age antecedents. Amongst the Iron Age wares illustrated in the report are a number of jar body-sherds with finger-impressed and slashed raised cordons. This decorative technique attracted the interest of the Curwens in that they thought that it might be a survival of Bronze Age urn decorative treatment, through the finger-impressing of ‘Halstatt’ type vessel rims, into the later Iron Age.

Excavations were also carried out on Thundersbarrow Hill behind Shoreham during the early 1930s (Curwen 1933) and a number of Roman pits and corndryers located. The pottery report (Oakley 1933) divided the handmade native wares into three groups; firstly, those associated with the Late Iron Age La Tene III tradition, and secondly, wares connected with a period of transition marked by the introduction of ‘Romanising techniques’. The third group was associated with 4th century coinage and was clearly late. Oakley regarded it as belonging to a period when Roman culture was on the wane and native ceramic traditions were reasserting themselves. He admitted to finding it difficult to distinguish much of this late pottery from the pre-Roman wares but drew attention to three other distinctive late fabrics within this third grouping. Two of these are of little significance but the third one, Thundersbarrow ware, was subjected to what is probably the first scientific analysis of a RomanoBritish coarseware on record. It was shown, amongst other things, that the clay used in Thundersbarrow storage-jars did not come from clay-pits beside a track leading to the site.

Four years later, the publication of an Iron Age and Roman site at Kingston Buci (Curwen and Hawkes 1931) included a further fragment with a raised slashed cordon. Although such pieces were still regarded as of La Tene III character, it was conceded that this particular piece was of such good fabric that the type may well have survived until a later date. The report concluded that full Romanisation of pot fabrics in the area may not have been achieved until after AD 70.

The 1930s also saw a radical shake-up in the study of Iron Age pottery in Britain. Hawkes reclassified such wares, hitherto fitted into a Hallstatt / La Tene framework, into categories A, B and C. The first was regarded as being introduced by Hallstatt, the second by Marnian culture invaders in the 3rd and 2nd centuries BC and the last by the Belgae at the time of Caesar’s Gallic wars: these basic subdivisions were further subdivided according to regional variations.

During the 1930s, there was a great deal of archaeological activity on Iron Age and Roman Downland sites in East Sussex, with particular attention being paid to the vessels decorated with fingered or slashed raised girth cordons. The pots bearing such decoration were still regarded as being almost entirely of Late Iron Age date, although the author’s conclusions in the Highdole Hill report

Ward-Perkins noted in his report on excavations at Crayford, that the Patchgrove grog-tempered wares present had cultural affinities with the grog-tempered wares from iron-working sites in the Weald of Kent and East Sussex (1938). Using Hawkes’s new system, he combined all of these wares under the heading of ‘SouthEastern B’. 3

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain Hawkes himself produced two articles on pottery assemblages from East Sussex on the eve of the Second World War. In his report on the pottery from Castle Hill Seaford (1939), he re-stated the idea that the SouthEastern B wares were entirely of Late Iron Age and earliest Roman date and that the pottery with raised finger-impressed cordons was of different cultural origin and descended from the Iron Age A tradition.

Roman period, although the poor stratification of most of the Bishopstone material meant that

Frère (1946) questioned Hawkes’s theory as to the origin of the raised-cordon decorated pottery and suggested instead that the decorative technique owed much to the cultural influence of Patchgrove wares from Surrey and West Kent. Patchgrove storage-jars are also decorated with finger-impressed shoulder cordons, although raised cordons are lacking. Frere also pointed out that Patchgrove and Asham type vessel production appeared to survive into the 2nd century AD, with an Asham pot from the Hassocks cemetery being associated with Central Gaulish Samian. He also attributed the longlived survival of the South-Eastern B traditions to the lack of an urban Romanising centre in the area.

The raised cordon decorative technique, employed on certain types of jar, was now seen to be common on East Sussex Ware jars until the mid 3rd century; thus confirming Holleyman’s suspicions over the Highdole pottery, registered so long before. Green also discussed Thundersbarrow ware, as well as the problems involved in transporting the large storage jars manufactured in this very friable fabric.

‘isolation of the fourth century types and investigation of a possible later decline in the industry must await the excavation of a solely fourth century site in East Sussex under modern conditions, since most sherds are undiagnostic as regards date’.

Green produced his seminal paper on East Sussex Ware in 1980, discussing its origins and attempting to produce a corpus of dated forms. He also expanded Frere’s hypothesis on the survival of the Late Iron Age ceramic tradition in East Sussex being due to the lack of an urban centre in the area to disseminate Romanising influences and produced a series of convincing maps to back up his argument.

‘The Weald, therefore, remained for one reason or another skirted rather than permeated by full Roman influence…… Thus the survival of our South Eastern B derived jars into the second century AD is seen to be no isolated phenomenon. It illustrates the characteristic conservative tendencies of a peasant culture. The Wealden folk had been anti Belgic: they seem to have remained unresponsive to the hand of Rome’.

In the last 20 years or so, the excavation of a stratified 2nd and 3rd centuries AD sequence of rubbish dumps in successive Beddingham villa bath-block rebuilds, a 2nd to 4th century AD sequence at the Barcombe villa and detached bathhouse and the author’s work on the associated pottery have resulted in much new information about East Sussex Ware (Lyne Forthcoming A and B). The writing up of the pottery from the 1936 excavations at Pevensey (Lyne 2009) has also thrown much illumination on changes in the ware and its marketing during the 4th and early 5th centuries.

The excavations at West Blatchington during the 1940s and early1950s revealed a number of 3rd century AD corn-dryers containing quantities of pottery from that period. This showed conclusively that the handmade grog-tempered native wares were still being made in quantity during the 3rd century AD.

A recent re-examination of the surviving pottery from the 1926 Caburn excavations has led to a re-interpretation of some of the excavator’s findings. On close scrutiny of the calender of Caburn pit contents at the end of the 1927 report, it was found that some of those containing the published examples of East Sussex Ware sherds with decorated raised cordons also contained North Kent Fineware (Pit 133) and vessel glass (Pit 77). The original finds still survive in Lewes museum and further investigation showed that the other pits containing raised cordon decorated fragmentrs also had 2nd to 4th century AD pottery associated. Such was the strength of the then current belief that such handmade ware could not be Roman, that four 3rd and 4th centuries AD coins in the fill of Pit 82 were concluded by the excavators to be intrusive: this despite the fact that one of Constantine I was on the very bottom of the feature (Curwen and Curwen 1927:11).

During the 1970s, Chris Green wrote a number of articles on the subject of native wares in Roman East Sussex. In the first of these (1976) he described such wares from the Meeching School, Newhaven excavation and coined the term ‘Cooking Jar Fabric’. This was the first detailed account of the very variable composition and surface treatment of the ware. Because of its imprecision, the term had a very short lease of life and was changed to a scarcely more meaningful, but geographically correct ‘East Sussex Ware’ by the time Green produced his Bishopstone pottery report the following year. This Bishopstone pottery report (Green 1977) improved on the fabric description and tentatively subdivided it: changes in fabric composition during the Roman period were also described. It was now recognised that the ware was produced throughout the

4

2: Industry Group 5. East Sussex Wares 5-5. Fabric with a predominance of coarse subangular off-white siltstone grog alongside smaller amounts of clay grog, ironstone and other filler. The size of the siltstone inclusions ranges between 0.30 and 3.00mm with thin-sectioning showing it to be kaolinite rich pipe-clay (Williams Forthcoming). The use of what is a relatively rare type of clay as a filler is difficult to explain but may, for this reason, help to narrow down the areas un which such wares are likely to have been made. White-firing clays are known at Fairlight near Hastings, occur as bands elsewhere in the lower Weald Clay and are also a feature of the Eocene Woolwich and Reading Beds, outcropping in patches on the coastfacing slopes of the South Downs between Shoreham and Newhaven. Fabric 5-5 is the most common East Sussex Ware variant in use during the 4th century AD. but unfortunately is nearly identical to the late East Kent grog-tempered ware fabric 7A-1 (p.60): some of the post c. 350-70 AD dated pots in siltstone grog tempered ware from Pevensey and elsewhere in East Sussex are of East Kent forms (Lyne 2009:111) and come from their source in the Lympne/Folkestone area (Lyne Forthcoming C).

2.3: East Sussex Ware fabrics Introduction Despite the shortcomings of Green’s East Sussex Ware terminology, it has been retained and elaborated on in this publication. The various pottery fabric variants recognised by him have been given new identifications. Several of these fabric variants have been recognised as peculiar to 3rd and 4th centuries AD East Sussex Ware products. With the exception of the grit-and-grog tempered 5-6, 5-7 and 5-8, however, individual fabric variants do not correlate with specific East Sussex Ware industries but are often associated with more than one such industry and were in use alongside other fabric variants. The earliest detailed account of East Sussex Ware fabric in general is that of Green in describing the Newhaven pottery (1976). This was soon followed by others on the Bishopstone and Ranscombe Hill pottery (Green 1977, 1978). Four different handmade fabrics were identified in the Ranscombe report: East Sussex Ware proper, chert, ironstone and grog tempered, crushed ironstone tempered and Thundersbarrow wares. The first of these fabrics corresponds with fabrics 5-1, 2 and 3 below, the second and third with fabric 5-6 and the fourth with fabric 5-4. Up to six fabrics were identified at Bishopstone, including a high-fired version of fabric 5-5, referred to by Green as G-T 2.

5-6. A variant of fabric 5-2 but with a predominance of crushed black ironstone, chert and quartz grit. 5-7. This is associated with the very late Pevensey grit-and-grog-tempered wares of Industry 5E and, because of its variability, will be described under that heading (p.30).

The fabrics 5-1. Soapy fine fabric with few inclusions visible to the naked eye but including ironstone, quartz and clay particles. This fabric is uncommon after the 2nd century AD but sporadically occurs during the early and mid 3rd century, particularly on Wealden iron-working sites.

5-8. Hard high-fired black-to-blue-grey fabric with coarse, crushed angular pre-fired siltstone grog. Associated with late 4th century Industry 5F (p.34). East Sussex Ware was fired to a variety of colours. The intention seems to have been to produce black carbonsoaked vessels but sherds can come in shades of dirty grey, brown and orange or patchy-fired with all or some of these colours. The patchy-fired pots are indicative of poor atmosphere control during the firings and of fuel coming in direct contact with the pots. Such conditions would be found in kilns without separate combustion chambers, bonfires and poorly-sealed clamps.

5-2. Fabric with multicoloured filler, including subangular pellets of crushed buff and grey fired clay grog, limonite, shale, chert and brown/black ironstone. The fabric occurs in a fine version (5-2A) with inclusions up to 1.00mm and a coarse version (5-2B) with up-to 3.00mm inclusions. 5-3. Black fabric filled with grog and other inclusions of similar colour to the matrix in which they are embedded and only revealing themselves in the texture of the surfaces of sherd breaks. This is not a true fabric in itself but is a result of low-temperature carbon-soaking in a pottery fabric which is usually of types 5-2 or 5-5.

Fabrics 5-6 and 5-8 are, however, often fired to a very high temperature, so as to be semi-vitrified. This pottery feels very rough and hackly to the touch and is frequently fired to a clean blue-grey colour with orange or whitish patches. The blacks, dirty browns and dirty greys are less common with this fabric, due to the burning out of carbon at the high temperatures reached during firing. The patchy-firing, however, indicates continued poor atmosphere control and random pot stacking but in these cases inside some kind of kiln capable of reaching high temperatures.

5-4. Thundersbarrow ware. This is a very coarse version of 5-2 and mainly used in the making of large storage jars during the late 4th century. The inclusions in this mainly oxidised fabric can be up-to 2.00 cm across and include small flint pebbles and lumps of chalk, as well as ironstone and fired clay grog.

5

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain was believed by the excavator to have been founded c.150 AD and abandoned during the second half of the 3rd century AD. Although iron-working appears to have been the main industry, there is evidence for both pottery and tile manufacture as well.

2.4: Sources The very nature of East Sussex Ware, handmade with poor fabric definition and displaying considerable variability in forms produced, makes it difficult to determine centres of production. The fabrics associated with the East Sussex Ware tradition were also, for the most part, fired to low temperatures and probably in bonfires and clamps. Such ephemeral structures leave little trace in the archaeological record and the variability of the pot fabric fillers make it very difficult to isolate sources of such material. Nevertheless, it has proved possible to identify six industrial groupings:

The evidence for pottery manufacture takes the form of pot wasters (Ibid.:20). Initially, at the time of the production of the PhD thesis on which this publication is based (Lyne 1994), evidence for local pottery production was based on examination of several surface collections of sherds from the site. The largest such collection was amassed by Ascherson in the 1940s and is now in Lewes Museum: Money’s and Haverfield’s surface assemblages (Cleere 1970, Haverfield 1917) are in Tunbridge Wells Museum and together with the pottery in Lewes leave no doubt that Cleere’s views on pottery production are essentially correct. This pottery incorporates a variety of wasters including bloated, discoloured and distorted fragments and tends to be in the fine fabric 5-2A .

5A. Bardown ware 5B. East Sussex Wealden ware 5C. East Sussex Downland ware 5D. Beddingham/Ranscombe ware

More recently, this author has written up the pottery from Cleere’s excavation (Lyne Forthcoming D) and obtained a good overall understanding of pottery supply to what appears to have been a small industrial town like that at Westhawk Farm, Ashford, 54 kilometres to the east and linked with it by road and trackway (Booth et al 2008).

5E. Pevensey Grit-and-Grog-tempered ware 5F. Barcombe/Burgess Hill ware with high-fired grog The production sites of only the first of these (5A) have been identified with any certainty, although it is probably that the bulk of East Sussex Wealden ware was made on iron-working sites and East Sussex Downland ware on sites around Newhaven at the mouth of the River Ouse, with production later transferred to around the Pevensey Saxon Shore fort.

Excavations at Bardown were carried out in three main areas; the slag bank, an industrial area and the ‘barrack block’. Production of Bardown ware appears to have started in the mid-2nd century AD and continued until shortly after AD270. The highest percentage of the coarse pottery (28%) comes from the slag bank, where not only iron slag was dumped but pottery kiln wasters as well: the fresh, partially complete vessels are similar to those in Tunbridge Wells museum and it is probable that Haverfield’s surface collection also came from the slag bank. Bardown ware accounts for an average of 18% of the coarse pottery from the ‘barrack block’ and 6% of the pottery from the industrial area Examination of the pottery also indicates that Bardown ware Industry 5A vessels were produced by potters who also made coarser inferior products in East Sussex Wealden ware fabric 5-2B.

Because of the similarities in fabrics and the presence of forms common to both the East Sussex Wealden and Downland ware groupings, it has proved virtually impossible to distinguish between the bulk of their jars, bowls and dishes during the period up to c..270 AD where their distribution zones overlap. As a result, the division between the marketing of these two wares is made on an arbitrary geographical basis indicated by their names. In the case of East Sussex Downland ware, its true distribution zone during the earlier 3rd century AD can be distinguished through the plot for its distinctive girthcordoned storage-jar form 5C.36 (Figure 5).

This demand for superior quality East Sussex Ware vessels at Bardown is not encountered on other ironworking sites in the Weald other than at Westhawk Farm, Ashford, where similar finewares were made during the 2nd century AD but had ceased being produced by AD.200. It may be no coincidence that these two sites were large industrial urban complexes, whereas as most of the Wealden iron-producing sites were probably not. A considerable amount of pottery came from Cleere’s excavations and was quantified by vessel type (Lyne Forthcoming D,Tables 1 to 5). This indicates that the majority of East Sussex Ware vessels produced at

Beddingham/Ranscombe, Pevensey Grit-and-Grog and Barcombe/Burgess Hill wares are very distinctive and present us with no problems of identification, although the actual production centres are unknown. 2.5: Industry 5A (Bardown ware) The evidence The Classis Britannica iron-working site at Bardown near Wadhurst (TQ 663293) was excavated between 1960 and 1968 (Cleere 1970). The associated settlement 6

2: Industry Group 5. East Sussex Wares Bardown were in East Sussex Wealden Ware, with just a small minority in Bardown Ware. The products of both industries display an overwhelming predominance of cooking pots, with just a few open and other forms.

if so, they have still to be discovered. Some of the pottery was carbon-soaked black in the manner of Dorset BB1, although many of the wasters are oxidised honey-brown in colour and others patchy grey or brown on black.

Using Willis’s work on Samian form assemblage breakdowns being indicative of types of occupation (2005), it was found that the Bardown Samian was unusual in combining aspects of both military and low status occupation. This would be appropriate for a site worked by slaves but with the additional presence of significant numbers of Classis Britannica personnel. That some of these personnel were of high rank is indicated by the discovery of a medallion of Antoninus Pius in the slag bank (Robin Hodgkinson pers. Comm.). Perhaps the superior Industry 5A vessels were for the use of Classis Britannica personnel and the East Sussex Wealden Ware ones for the use of the iron producers.

A Corpus of forms produced

The exploitation of raw materials

5A.1. Cordoned-and- necked jar form in grey-black fabric 5-2A. c. 150-270 AD.

As much as 16% of the 3rd century AD pottery from Cleere’s excavations originated in the Thameside potteries, 35 kilometres to the north of the site. The Bardown ceramic form range, and in particular jars and bowls, was heavily influenced by the forms associated with that industry. In this respect it can scarcely be regarded as typical East Sussex Ware. Jars Figure 1.

Cleere has drawn attention to the importance of the local Wadhurst clay for local pottery and tile manufacture during recent times (1970:2). There are large numbers of pits in the neighbourhood of Bardown and it is almost certain that this was the clay exploited by the Roman Bardown potters. The grog used by the potters was probably derived from the crushed fabric of old iron furnaces and the ironstone from the material used in iron manufacture.

5A.2. Necked jar form in dark grey fabric 5-5 fired patchy off-white/black. c. 150-270 AD. 5A.3. Cordoned-and-necked jar form waster in dirty buff-grey fabric 5-2A. c. 150-270 AD. 5A.4. Everted rim jar form in patchy buff/black fabric 5-2A. c. 150-270 AD. 5A.5. Everted rim necked-jar/bowl form in pinkish-grey fabric 5-2A fired black internally. c. 150-270 AD.

The fuel supplies would have been the same as those used for iron production, although in the latter case the wood was converted into charcoal before use. Cleere (Ibid.:16) identified a variety of tree species in the charcoal from the iron furnaces. The samples were dominated by oak, with lesser quantities of hornbeam, birch, hazel and hawthorn. Cleere thought that the mainly branch-derived charcoal suggested that a policy of tree lopping was operated: there does not appear to be any clear evidence for encoppicing.

5A.6. Similar vessel form but with cavetto rim in black fabric 5-2A fired brown/buff externally. c. 150-270 AD. The cordoned-and-necked jar forms are similar in general appearance to the Thameside necked-bowl form 4A-2, dated there to c. 110-200 AD (Monaghan 1987), and to jars of 2nd and early 3rd centuries AD date from a variety of West Kent, Surrey and Hampshire sources.

Technology

Cordoned-and-necked jars are far more common in the Bardown industry jar repertoire than in those of the other East Sussex Ware variants. They make up 41% of the jars in the examined Bardown assemblage in Tunbridge Wells museum, compared with the 13% of all East Sussex Wealden ware jars in the similarly dated Bodiam one (p.12).

The Bardown pottery industry employed hand-working technology on its bowls, dishes and lids. The jars vary somewhat in quality, with some being quite clearly coil built and others possibly drawn up on a turntable. These latter jars tend to be those with neck cordons and single horizontal grooves around their girths or shoulders: they are, however, inclined to be facet-burnished externally and what appears to be evidence for the use of a turntable may simply reflect superior finishing.

Jar type 5A.4, with its flat rim and stabbed shoulder cordon, is similar in appearance to late 2nd and early 3rd century AD. examples of Alice Holt Class 3A (Lyne and Jefferies 1979). This, along with some of the neck cordoned jars, may reflect influence from that quarter. Examination of an early 3rd century AD assemblage from Sanderstead on the Surrey/ Kent county boundary

The distorted nature of some of the wasters suggests that higher firing temperatures were reached by this industry than was normally the case with East Sussex Wares. It is probable that kilns were used by the Bardown potters but,

7

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain

Figure 1: Industry 5A forms.

8

2: Industry Group 5. East Sussex Wares shows that Alice Holt wares were being traded there; only 42 kilometres from Bardown. Some of the jars represented in the Bardown material (5A.5 and 5A.6) are of slack and ill defined profile; similar to vessels from and representing the stylistic influence of the main East Sussex Ware 5B industrial grouping.

5A.15. Shallow dish in honey-brown fabric 5-2A with patchy black exterior. c. 150-270 AD.

Bowls

5A.16. Perforated lid form in patchy black/chocolatebrown fabric 5-3. Perforations are the exception rather than the rule as regards Industry 5A lids. c.150-270 AD.

Lids Figure 1.

Figure 1. 5A.7. Flanged bowl form in lumpy black fabric 5-2A with external, linear-burnished acute latticing. c. 150270 AD.

Distribution A complete lack of significant 3rd century pottery assemblages from the area around Bardown means that it is as yet impossible to assess the distribution of the wares. It may be, however, that the potters supplied no more than the iron-working settlement there with their finer wares. In support of this notion, it is noted that no certain examples of Industry 5A products have been seen in pottery assemblages from other Wealden ironworking sites. The Bardown settlement, along with many other Wealden iron-working sites, was abandoned during the mid /late 3rd century AD and its pottery production terminated.

5A.8. Plain variant of 5A-7 in grey-black fabric 5-2A. c. 150-270 AD. 5A.9. Incipient beaded-and-flanged bowl form in buff fabric 5-2A with external linear-burnished acute latticing. c. 200-70 AD. 5A.10. Cavetto-rim bowl form in patchy black/buff fabric 5-2A with external linear-burnished acute latticing. c. 150-270 AD. Convex-sided flanged bowls of types 5A.7 and 5A.8 with lattice decoration would seem to be influenced by the c. 140-200 AD dated Gillam 39 BB1 bowl form (1976) and the incipient beaded-and-flanged bowl of type 5A.9 by the c. 220-280/90 AD dated bowl of Lyne type 6-2 in similar fabric (2012).

2.6: Industry 5B. (East Sussex Wealden Ware) Sources and exploitation of raw materials This is not so much a specific industry as a potting tradition; probably common to several production centres manufacturing similar forms in similar fabrics. The products in this tradition make up the bulk of East Sussex Ware on Wealden iron production sites: hence the name.

The cavetto-rim bowl type 5A.10 and its derivative dish type 5A.11 may owe something to the slightly earlier Thameside classes 5B-3 and 5B-4 dated c. 70-130 AD (Monaghan 1987) and in both cases be ultimately based on samian forms Dr,35 and Dr.36 (c. 70-230 AD). A similar bowl form was also made at an unknown centre near Barcombe during the late 2nd and early 3rd centuries AD in wheel-turned white ware fired blue-grey.

The distribution pattern, described below, indicates the presence of a major production centre in the Hastings area; probably on an iron-working site: Cleere and Crossley (1985:295) have drawn attention to the abnormally large quantities of 2nd and early 3rd century AD East Sussex Ware pottery from the iron ore mining and roasting site at Petley Wood, Battle. Another major pottery production centre was at Bardown, which also made the finer Industry 5A products. Other, more westerly, Wealden iron working sites probably had their own small pottery industries. Until the production centres are more precisely located, there is little that can be said about the exploitation of raw materials by the potters belonging to this tradition. It is, however, characterised by the use of fabrics 5-1, 5-2 and 5-3; both coarse and fine. These three fabrics probably indicate, from their inclusions, that most pots, like those of Industry 5A, were made on iron production sites by potters exploiting the same fuel resources and other raw materials as that industry.

Dishes Figure 1. 5A.11. Dish form with out-turned rim in patchy buff/ grey fabric 5-2A with external latticing. c. 150-270 AD. 5A.12. Plain variant of 5A-11 in grey fabric 5-2A fired brown and polished externally. c.150-270 AD. 5A.13. Flanged dish in dirty grey-black fabric 5-2A with burnished diagonal lines on its exterior. c. 150-270 AD. 5A.14. Deep dish in black fabric with crushed tile and coarse off-white grog filler. c. 150-270 AD.

9

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain 5B.10. Incipient beaded-and-flanged bowl form in grey fabric 5-3 fired black with black internal patches. c. 150270 AD. Beauport Park.

Technology East Sussex Wealden ware products differ from Bardown ware in being exclusively handmade and fired to a lower temperature. As with Bardown ware, the intention seems to have been to produce black pots, although a significant number ended up being patchy orange/brown/black. This black finish to most of the pots may indicate that they were smoked over a fire before firing (Hodges 1964:.35).

5B.11. Variant of the above in black fabric 5-3 fired patchy buff/black externally with overall polish. c. 150270 AD. Beauport Park. 5B.12. Everted rim bowl in black, soapy fabric 5-1. c. 150-270 AD. Bodiam.

A corpus of forms produced

The bowl forms are similar to, but more poorly finished than those associated with Bardown industry 5A. Forms 5B.8 and 5B.9 are probably influenced in style by Thameside BB2 products; some of which were traded down the Roman road from Rochester via Maidstone and Bodiam to the iron-working sites behind Hastings.

Jars Figure 2. 5B.1. Slack-profiled jar form in dirty-grey fabric 5.3 fired black with overall external burnish. c. 150-270 AD. Bodiam.

The incipient-beaded-and-flanged bowl forms 5B.10 and 5B.11 could not be more rudimentary in form. There is no true flange; the lid-seating being achieved by the creation of a finger-impressed hollow in the top of the rim of a simple wall-sided bowl. Identical bowls were made at the Ouse Valley production centre or centres of the East Sussex Downland ware industry: their combined distribution zone is shown in Figure 5 (p.---). The everted-rim bowl form 5B.12 is similar to the Bardown industry form 5A.10 and probably had a similar origin.

5B.2. Form 5B.1 variant in soapy black fabric 5.3. c. 150-270 AD. Bodiam. 5B.3. Form 5B.1 variant in blue-grey fabric 5-3 variant with a little fine sand added. c. 150-270 AD. Bodiam. 5B.4. Necked bowl/jar form in lumpy brown-black fabric 5-3 fired black. c. 150-270 AD. Bodiam. 5B.5. Jar form with everted rim, in fine grey-black fabric 5-3 fired buff-brown. c. 150-270 AD. Bodiam.

Dishes

5B.6. Similar form but lacking the weak off-set below the rim. In coarse fabric 5-3 fired blue-grey. c. 150-270 AD. Bodiam.

Figure 2 5B.13. Dish form in coarse black fabric 5-3 fired patchy grey/black internally. c. 150-270 AD. Beauport Park.

5B.7. Jar form with clubbed everted rim. In coarse black fabric 5-3. c. 150-270 AD. Bodiam

5B.14. Convex-sided dish form in fine black fabric 5-3. c. 150-270 AD. Bodiam.

The jar forms associated with this pottery grouping are characterised for the most part by slack profiles and weakly everted and poorly-finished rims. There is a total leck of decoration on the pots from Bodiam, although some of the similar forms from the Beauport Park bath-house have linear-burnished acute latticing. Neck-cordoning on jars is rare and on the whole poorly executed.

Form 5B.13 is similar in style to Bardown industry’s 5A.11 in also having a slightly-out-turned rim. Lids. Figure 2 5B.15. Simple form in dirty-grey fabric 5-1 fired black. c. 150-270 AD. Bodiam.

Bowls Figure 2.

5B.16. Variant of 5B.15 in fine grey fabric 5-1 fired black. c. 200-70 AD. Footlands, Sedlescombe.

5B.8. Bead-rim bowl form in black fabric 5-3 fired brown externally with overall polish. c. 150-270 AD. Bodiam.

Lids seem to have been manufactured from the inception of the East Sussex Wealden ware industry until its demise. There is little change in form throughout this period.

5B.9. Bowl with weak bead-rim in black fabric 5-3. c. 150-270 AD. Bodiam.

10

2: Industry Group 5. East Sussex Wares

Figure 2: Industry 5B forms and 5C.36 storage-jar

11

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain An assemblage of 2nd to mid 3rd century AD pottery from Ditch 10610 running behind the main villa building had grog tempered wares accounting for 17% of it (Ibid. Assemblage 8). It was, however, difficult to distinguish local fabrics from East Sussex Wealden ware ones but vessels from the latter source, including a dish of form 5B.14, were clearly present. It can be roughly estimated that 5% of the 3rd century AD coarse pottery from the site was from East Sussex Wealden ware sources.

Asham pots Figure 2. 5B.17. Neck of storage vessel with weak cordon and an Asham pot profile, in coarse, dirty grey fabric 5-2B with additional sparse 5 to 10mm ironstone fragments. c. 150270 AD. Bodiam. Specific storage-jar forms are not a normal feature of this industry during the 3rd century AD and it is probable that large cooking pots were sometimes used for this purpose. The Asham pot form (Curwen and Curwen 1930) was, however, a characteristic late 1st and 2nd century form of both the East Sussex Wealden and Downland ware industries and, in view of its constricted neck, may have been used for storage. Limited production of Asham pots by both industries seems to have lingered on well into the 3rd century through debased forms such as the example illustrated here.

The East Malling villa, seven kilometres to the west of Maidstone, was destroyed c. 270 AD. Its latest occupation had East Sussex Wealden ware making up 10% of all of the coarse pottery, of which jars accounted for 9.5% and dishes a nominal 0.5%. Further west still, a 3rd century AD East Sussex Ware girth cordoned storage jar of form 5C.36 was noted amongst the pottery from Lullingstone villa in the Dart valley (Pollard 1987: Figure 75,147). It was the only certain piece seen in the total pottery assemblage and indicates that the villa lay beyond the distribution area of the East Sussex Ware industries other than for these specialised vessels, possibly used as packaging for produce (p.23).

Distribution (Figure 6) There is evidence for the trading of East Sussex Wealden wares from the Hastings area, by road through Bodiam to the Medway valley. Their domination of the total pottery assemblage from the Footlands, Sedlescombe iron-working site (87%) must surely indicate a pottery manufacturing centre, such as Petley Wood, in the vicinity (p.22). This is further borne out by a similar percentage of the coarse pottery in an AD.200-270 dated pottery assemblage from the nearby iron exportation port at Kitchenham Farm, Ashburnham (Lyne Forthcoming H, Assemblage 5, 84%).

The large iron-producing settlement at Bardown lies 25 kilometres to the south of these sites and was a centre for the production of not only Bardown wares but East Sussex Wealden wares as well. The former accounted for 18% of the coarseware assemblage from the mid to late 3rd century AD ‘barrack block’ and the latter for a further 45%. This 45% can be further be subdivided into 37% jars, 3% bowls, 3% dishes and 2% lids. What may be an iron working settlement at Wye in the valley of the River Stour 16 kilometres south west of Canterbury has the unpublished 1st century AD coarse pottery assemblages dominated by handmade grog tempered wares, including an Asham pot decorated with black paint. Most of Kent during the years immediately after the Conquest was dominated by grog-tempered wares in the ‘Belgic’ Aylesford/Swarling tradition. This unpublished assemblage is, however, so strongly influenced by the Eastern Atrebatic tradition as to suggest that the early Roman ceramic tradition of the Wye area was connected with that of the eastern Weald rather than with the rest of Kent.

East Sussex Wealden ware jars accounted for more than half of all coarse pottery in the assemblage from Footlands (62%) and bowls and dishes for considerably less (13%). The presence of three beakers (a rare East Sussex Ware form) is further indication of the proximity of a production source and, with lids, account for a further 15% of the assemblage. Six kilometres to the north at Bodiam, grog-tempered vessels accounted for a slightly lower 77% of all coarse wares. Jars were again the predominant East Sussex Wealden ware form (61%), with much smaller percentages of open forms (4%) and lids (12%).

Such grog tempered wares ceased being supplied to the Wye area by the end of the 1st century AD, but the early 3rd century. saw the reappearance of such vessels in East Sussex Wealden ware forms and fabrics and indicating renewed contact from that direction. This pottery includes examples of form 5B.1 and may originate in the Hastings area: it could equally well come from a production area nearer at hand, such as the iron-working settlement at Westhawk Farm, Ashford four kilometres to the south-west.

Further up the road to Rochester, the mid to late 3rd century AD pottery from Phase III at the Maidstone Mount villa has East Sussex Wealden ware jars accounting for a much lower 6% and bowls and dishes for a nominal 3% of the total coarse ware assemblage. At Thurnham, seven kilometres east of Maidstone, the CTRL excavations by Oxfordshire Archaeology indicated that small amounts of East Sussex Wealden ware were getting to the site from the late 1st to the mid 3rd century AD (Lyne 2006A).

12

2: Industry Group 5. East Sussex Wares Industry 5B wares make-up 25% of all of the coarse pottery from the Harville,Wye villa rubbish midden, with a predominance of jars and much smaller percentages of other forms, such as bowls and lids. This is similar to the situation at Maidstone Mount and East Malling.

amounts of Arun Valley greywares at both this site and at Alfoldean (Winbolt 1927) , however, indicates that Stane Street was exploited as a significant trade route for that industry’s wares during the 1st to 3rd centuries AD. This trade effectively blocked off that in East Sussex Wealden wares to the west of Stane Street, other than a few stray pots found on West Surrey greensand sites such as Rapsley (Hanworth 1968: Figure 25-153).

Some of the grog-tempered wares from elsewhere on the site are later in date than AD 270 and include developed beaded-and-flanged bowls of types 7A.8 and 7A.9. These forms were manufactured by Industry 7A, which appears to have been based at Lympne and supplied the shore fort there with most of its coarse pottery after c.AD.280 and smaller amounts to other sites in East Kent coastal areas, during the late 3rd and 4th centuries AD (p.63-7). It appears therefore that some of the post AD.270 grog-tempered pottery at Harville was coming in from the Folkestone area and supplanting that coming in from the Sussex Weald, where most grog-tempered pottery manufacture ended during the mid to late 3rd century AD.

The end of the industry The Wealden iron works achieved their greatest output during the late 2nd and early 3rd centuries AD under the control of the Classis Britannica. Computations based on volumes of slag in waste dumps suggest a combined annual output of 750 tons of iron between c. 150 and 250 AD (Cleere and Crossley 1985: 81). The Classis Britannica ceased to exist during the late 3rd century and most of the iron-working centres were closed down, with the annual output of iron dropping sharply to an estimated 200 tons. With the departure of the iron workers, most of the production of East Sussex Wealden ware came to an end.

East Sussex Wealden wares made their appearance at Westhawk Farm, Ashford during the mid 2nd century AD and achieved 17% of the coarseware element in early 3rd century Assemblage 35 alongside a late ‘Belgic’ grogtempered ware variant with sparse fine siltstone grog filler (23%) and the first examples of Industry 7A jars in coarse siltstone-grog-tempered ware (3%). The late 3rd century AD Assemblage 40, dating to just before the near abandonment of the site, has 19% East Sussex Wealden ware and a greatly increased percentage of Industry 7A products (33%): this provides further evidence for Industry 7A supplanting East Sussex Wealden ware in south-east Kent during the late 3rd century . The East Sussex Wealden ware products can be subdivided into jars (14%), bowls (1%), dishes (3%) and lids (1%): they include examples of forms 7B.5, 7B.13, 7B.14 and 7B.15 (Lyne 2008: Figures 6-10,181, 6-11,213,215 and 216).

East Sussex Wares are rare at the shore fort founded at Pevensey between AD 293 and 296 before the mid 4th century. (Lyne 2009), reflecting both the collapse of the East Sussex Wealden ware producing industries during the late 3rd century AD and the distance of the fort from the surviving East Sussex Downland and Beddingham/ Ranscombe wares centres of production. The little grog tempered ware that is present at Pevensey before the mid 4th century probably comes from East Sussex Downland ware sources, although an incipient-beaded-and-flanged bowl of type 5B.10 from fort construction deposits appears to indicate that some East Sussex Wealden ware production continued into the 290s or later. This continued production of East Sussex Wealden wares may also have taken place at the Kitchenham Farm site at Ashburnham, recently excavated by the Hastings Area Archaeological Research Group, as a few early 4th century coins and ceramic imports are present there.

It is unlikely that much of the grog tempered ware from the more westerly Wealden iron working sites came from the Hastings area. The Bardown pottery has indicated some variability in the grog-tempered ware producing industries operating in the Weald and the material from Garden Hill includes a number of cordoned-and-necked jars in superior fabric but differing in detail from the equivalent Bardown 5A range (Eade 1976:12). The Broadfields site at Crawley is the most north-westerly of the Roman iron-working sites within the Weald and lies only ten kilometres to the east of Stane Street (Cartwright 1992). The surviving pottery is of 1st to 3rd century AD date and is characterised by high percentages of both East Sussex Wealden ware and Arun Valley wheelturned products. A great deal of the pottery from this site has been lost, however, since the excavator’s death, meaning that accurate quantification of assemblages cannot be carried out. The presence of significant

2.7: Industry 5C. East Sussex Downland ware Centres of production This is perhaps the best known and studied of all of the East Sussex Ware industrial groupings and probably supplied most, if not all, of the East Sussex Ware pottery found on Downland sites east of the River Adur and some of that from Wealden sites to the north and east. As with Industry 5B, there was more than one centre of production, although the distribution plot for the distinctive jar form 5C.36 (Figure 5) suggests that there was one main pottery source, somewhere near the mouth of the River Ouse during the period c. AD.200 - 270.

13

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain This is further supported by Green’s observations on the Bishopstone grog-tempered wares (1977:155). He noted that the pottery contained ironstone inclusions similar to those found naturally in the Eocene deposits of the Newhaven area; the most easterly outlier of that formation in Sussex. The possibility of this filler coming from the same formation in the Brighton area is largely negated by the fact that percentages of East Sussex Ware fall off in that direction. Small production centres there, at sites like West Blatchington cannot, however, be completely discounted.

Fabrics Fabrics 5-2 and 5-3 are the most common fabrics associated with 3rd century East Sussex Downland ware, although fabric 5-5 became increasingly significant towards the end of the century. This siltstone grog tempered fabric became much more significant during the 4th century, to the extent of becoming the predominant one. During the later 4th century AD, the very coarse fabric 5-4 became regularly used in the manufacture of large rope-rimmed, dry storage jars. This vessel form was also produced by a variety of pottery industries in southern Britain during the 3rd and 4th centuries, particularly at the Alice Holt/Farnham industry kilns (Lyne and Jefferies 1979,Class 10) and those of the New Forest and Rowlands Castle pottery industries.

The percentage distribution map (Figure 6) and the Beddingham villa sequence (Lyne Forthcoming A) indicate that the Ouse Valley potting centre went into decline after c. 270 AD. The focus of East Sussex Downland ware production seems to have moved west into the Brighton area and remained there until c. 370 AD. The evidence for this is backed up by the distribution of developed beaded-and-flanged bowl types (Figure 7).

Appendix 4 (p.161) illustrates the growth in the significance of fabrics 5-4 and 5-5 within the late East Sussex Ware tradition. The analyses are divided into three dated groupings: the left-hand column under each fabric heading shows the EVE percentage of all pottery and the right-hand one, in brackets, the fabric as a percentage of all East Sussex Ware in the assemblage.

The map of c. 370-400+ AD East Sussex Downland ware distribution (Figure 8) indicates that a new centre of production, possibly based at Pevensey, appeared east of the River Cuckmere after AD.370. Fabric 5-5 is the most common one associated with this production centre.

Fabric 5-5 was insignificant before c. 270 AD but achieved predominance between that date and c. 370 AD. It is noticeable that its share of the range of East Sussex Ware fabrics tends to increase in a westerly direction, whereas the opposite is the case after c. 370 AD. This suggests that the potters in the Brighton area were using white-firing siltstone grog filler from Eocene clays before c. 370 AD. The situation is complicated after that date in that some of the Pevensey pottery appears to emanate from the Industry 7A source in the Folkestone area; an industry which was making use of siltstone grog filler indistinguishable from that in East Sussex Ware fabric 5-5.

Sources of clay, filler and fuel Late Iron Age occupation sites around the estuary of the River Ouse have produced clear evidence for seasalt production: briquetage in the form of bars, pillars, wedges and containers has been found at both the Castle Hill Newhaven and Bishopstone sites (Bell 1977:122). The briquetage at the latter site was also shown to have been made from Eocene clays taken from Castle Hill, as was similar clay used in Iron Age grog-tempered ware manufacture at Bishopstone. There is no certain evidence for East Sussex Ware production on either of these sites during Roman times but the evidence certainly indicates that at least one major source was not very far away.

Some of the bowls from Pevensey and Frost Down are of Industry 7A types, as are deep, convex-sided dishes from Pevensey, Bishopstone and other sites east of the River Ouse. There is as yet no way of distinguishing between the bulk of the Industries 5C and 7A jars in fabric 5-5, so it is probable that the presence in East Sussex of the latter wares after AD.370 has been understated. It may be that there was enough of the Kentish Industry 7A pottery getting into the Pevensey area to create the predominance of siltstone grog-tempered at the shore-fort and the increase in their significance east of the Ouse.

The Ouse had an open estuary in Roman times, which extended as far upstream as Lewes. Just as Late Iron Age salt production seems to have been associated with that of pottery, the same arrangement may have continued in Roman times. The salt production industry would have required the same fuel resources as pottery production and it is probably no coincidence that coastal pottery production sites of Iron Age, Roman and later date were almost always in or near large areas of saltern. Such areas of saltern existed around the Ouse estuary and would have supplied peat for brine boiling and for pottery clamps as well. Just as the iron-working sites in the Weald were capable of producing a regular supply of fired clay grog from old furnaces for use as pot filler, so the briquetage from brine boiling could fulfill a similar need.

A corpus of forms produced Jars Figure 2. 5C.36. Globular jar form in grey-black fabric 5-3 with raised girth cordon decorated with finger-impressed decoration. c. 100-270 AD. Findon well. 14

2: Industry Group 5. East Sussex Wares decoration. c. 270/300-400+ AD. Slonk Hill, Shoreham. Postholes 212-214.

Figure 3. Types of girth-cordon decoration found on 5C.36 jars.

5C.10. Everted rim jar form in dirty patchy grey fabric 5-5. c. 370-400+ AD but occasionally found in assemblages as early as c. 270-300 AD. Pevensey. Salzman excavations. Unstratified.

5C.A. Cordon cut by stabbing and rough moulding into crude, pyramidal sections. c. 100-270 AD. Beddingham, Context 101. 5C.B. Finger-impressed decoration. c. 0-270 AD. Beddingham, Context 101.

5C.11. Slack-profiled everted-rim jar form in brown fabric 5-2 with external wiped self-slip fired black. c. 270-400+ AD. Pevensey 1936, Trench II, Context 2.

5C.C. Slashed herringbone decoration. c. 100-270 AD. Beddingham, Context 11. A variant of this has simple diagonal slashing.

During the 1st and 2nd centuries AD, this industrial grouping’s range of jar forms was dominated by a variety of handmade examples with weak everted rims; sometimes with neck cordons but usually not. Two more distinctive jar types were also present and were probably used for storage. The first was the sub-biconical Asham pot form (Green 1980: Figure 28:17) with a shoulder sloping up to a constricted short neck with weakly-everted rim. The second form was a large, globular jar with a raised and decorated girth cordon separating scribed or burnished arcading on the rounded shoulder and burnished acute latticing on the lower half (Ibid.: Figure 27:1).

5C.D. Flush girth cordon edged with horizontal grooves flanking band of vertical stabbing. c. 220-70 AD. Beddingham, Context 101. 5C.E. Decorative girth band consisting of a broad, shallow, horizontal groove with crude stab marks. c. 220-70 AD. Beddingham, Context 101. 5C.F. Girth band of irregularly placed small oval stab marks. c. 270-300 AD. Beddingham, Layer 14.

Both of these vessel forms are characteristic of the East Sussex Downland ware industrial grouping, although the Asham pot is also found in the repertoire of East Sussex Wealden wares. During the early 3rd century AD, the Asham pot form disappeared from the Downland ware form range: it was absent from the fills in the cold-plunge of the Beddingham bath-house. The same deposits do, however, contain decorated girth cordon fragments from globular jars, indicating that they continued being produced during the 3rd century (Lyne Forthcoming A). The deep Findon well (Lyne Forthcoming J.) contained very large quantities of mid to late 3rd century pottery, along with coins of the same period. This pottery includes a solitary example of a girth cordoned jar, providing further evidence for manufacture well into the late 3rd century. Such jars had been made from the pre-Roman Late Iron Age and saw changes in decorative technique during their long period of manufacture.

5C.1. Jar form in grey fabric 5-3 with crudely stabbed decoration of type 5C.E on matt band around its girth. c. 220-70 AD. Beddingham, Context 101 5C.2. Jar form in dirty brown fabric 5-5 fired black with srabbed decoration of type 5C.F around its girth. c. 270300 AD. Beddingham. Context 357 in Pit 358. 5C.3. Cordoned jar form in black fabric 5-2. c. 220-70 AD. Beddingham, Context 101. 5C.4. Cordoned jar form in dirty-grey-brown fabric fired black. c..220-70 AD. Beddingham, Context 101. 5C.5. Everted rim jar form in black fabric 5-5B. c..220400 AD. Bullock Down, Feature 8 (Drewett 1982: Fig.66:161). 5C.6. Everted-rim jar form in brown fabric 5-5A fired black. c. 270-400 AD. Beddingham, Layer 27.

The use of paint was thought by Green to have been at an end by c. 100 AD, along with eyebrow decoration (1980). The Beddingham sequence suggests that there was a decline in the frequency and quality of decoration on the girth cordons of such jars during the 3rd century AD. Raised girth-cordons with finger-impressed and slashed decoration of types 5C.A to 5C.C disappeared around the middle of the century, whereas flush girth cordons with stabbed decoration went in into the last quarter, albeit in the form of crude, degenerate types 5C.E and 5C.F. Jars decorated with girth cordons are virtually absent from the shore-fort at Pevensey, suggesting that the use of this decorative technique had all but ended by the time that the fort was constructed during the period AD 293 - 296.

5C.7. Jar/necked bowl form in black fabric 5-3 sandwich-fired patchy brown/black with orange rim edge. Decorated with burnished acute-latticing. c. 270400 AD. Beddingham, Layer 251. 5C.8. Jar form with everted rim, in soapy grey-black fabric 5-5B fired patchy brown/black externally, black over the rim and brown internally. c. 270-400+ AD. Slonk Hill, Shoreham. Postholes 212-214. 5C.9. Jar form with everted rim, in dirty grey-black fabric 5-5A fired patchy black/brown with 90 degree lattice 15

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain

Figure 3: Industry 5C jar and bowl forms

16

2: Industry Group 5. East Sussex Wares Some 3rd and 4th centuries East Sussex Downland ware everted rim jars were decorated with burnished acute or 90 degree latticing, the dating of which differs from that on BB1 and BB2 cooking pots. In the case of BB1, acute lattice decoration on cooking pots was joined by 90 degree lattice towards the end of the 2nd century AD before they were both replaced by obtuse latticing during the earliest years of the 3rd century AD. With East Sussex Ware, acute lattice decoration appears on cooking-pots during the 2nd century. and continues in use into the late 4th century. It was only employed on a minority of jars, the bulk of examples having either overall external facet burnishing or no surface treatment at all.

5C.15. Beaded-and-flanged bowl form in brown fabric 5-3. c. 270-350 AD. West Blatchington, Hearth over Corndryer 6. 5C.16. Beaded-and-flanged bowl form in grey-brown fabric 5-5. c. 270-350 AD. West Blatchington, Hearth over Corndryer 6. 5C.17. Beaded-and-flanged bowl form in brown fabric 5-3. c. 270-350 AD. West Blatchington, Hearth over Corndryer 6. 5C.18. Beaded-and-flanged bowl form in soapy grey fabric 5-5A fired brown. c. 270-350 AD. West Blatchington, Hut in Findon Close.

Surface decoration on jars became rarer during the 4th century AD: neck-cordons finally disappeared, so all we are left with are plain examples. At Pevensey, founded during the 290s, East Sussex Ware jars show little sign of decoration other than a fragment from a fingerjabbed jar neck and a lightly-combed girth sherd. The latter decorative technique may have been inspired by that on 4th century AD Alice Holt/Farnham industry jar form 3B-14 and similarly-dated New Forest greyware form 30.6 (Lyne and Jefferies 1979, Fulford 1975).

5C.19. Beaded-and-flanged bowl form in grey-brown fabric 5-5B fired black. c. 300-400+ AD. Slonk Hill, layer 2 within western barrow ditch. 5C.20. Beaded-and-flanged bowl form in coarse grey fabric 5-3 fired black externally. c. 370-400+ AD. Pevensey 1936. East Gate. Fill of Gully 1. 5C.21. Beaded-and-flanged bowl form in coarse orange fabric 5-3. c. 370-400+ AD. Truleigh Hill 1949-50.

Increased trading of Alice Holt/Farnham industry wares into East Sussex may also have influenced some of the 4th century jar forms of Industry 5C, many of which now had well-formed everted rims, reminiscent in cross-section of contemporary Alice Holt Class 3B ones.

The East Sussex Downland ware industries produced a variety of bowl forms during the 2nd and 3rd centuries AD. One of the more distinctive forms was the reeded-rim variety 5C.12: this form probably copied contemporary Rowlands Castle and Canterbury greyware ones and is sometimes decorated with stabbing along the upper rim surface. The presence of examples in the early3rd century cold-plunge bath fills at Beddingham and similarly dated deposits elsewhere suggests that the form continued to be made well into the 3rd century AD.

A distinctive feature of some East Sussex Downland ware jars is a simple scribed cross quartering the underside of their bases. This has no Christian significance but is merely a simplification of the more elaborate cruciform decoration found on the bases of Late Iron Age East Sussex Ware pots from the region (Curwen and Curwen 1927: Pl.XVI).

The deep bowl form 5C.13 has a slightly-everted lip and may be East Sussex Downland ware’s equivalent of Bardown Industry 5A type 5A.11: it is similarly dated to the late 2nd and early 3rd centuries AD. As with the other East Sussex Ware industries, the trading of Thameside industry BB2 bowls of Monaghan Class 5D (1987) across East Sussex led to the production of copies like 5C.14 during the same period.

Bowls Figure 3 5C.12. Reeded-rim bowl form in patchy buff/black fabric 5-1 with linear-burnished lattice decoration on the exterior. 2nd century to c. 270 AD. Beddingham Context 101.

The incipient beaded and flanged bowl form appeared at the Meeching School, Newhaven site as early as the late 2nd century AD. These bowls are, for the most part, identical in form and fabric to the East Sussex Wealden ware forms 5B.10 and 5B.11 and highlight the problem encountered in separating out some of the pre AD.270 products from the two industrial groupings where their distribution zones overlap. For this reason they are not given separate East Sussex Downland ware corpus numbers and are combined with East Sussex Wealden ware forms 5B.10 and 5B.11 in their distribution plot (Figure 5)

5C.13. Convex-sided bowl form with slightly-everted rim, in soft medium-grey fabric 5-2. c. 150-270 AD. Beddingham, Context 101. 5C.14. Bowl copying those in BB2 fabric of Monaghan Classes 5C and 5D, in grey-black very coarse fabric 5-2B fired black externally and brown internally. c. 130270 AD. Beddingham, unstratified.

17

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain Developed beaded and flanged bowls made their appearance at Beddingham after 270 AD. They are totally absent from the early 3rd century fills of the cold plunge bath but are present in occupation layer 14 above, which is associated with the rebuilt villa. At the base of this occupation layer was a coin of Victorinus (AD.268-270), suggesting that this occupation dates to c. 270-350+ AD.

5C.25. Dish form in grey-black fabric 5-3 fired patchy black/orange externally and brown internally with external burnished latticing. c. 150-350 AD. Beddingham, Context 11. 5C.26. Dish form with out-turned rim in black fabric 5-5B. c. 200-370 AD. Truleigh Hill 1949-51. 5C.27. Deep, convex-sided dish form in dirty-grey fabric 5-3B fired black. c. 350/70-400+ AD. Pevensey 1936, East Gate, Context 3.

The form found in layer 14 (5C.17) has its bead and flange formed by pressing a finger down into the top of the bowl rim, drawing it round the vessel and forming a concave-sectioned hollow. The clay is squeezed out to form a stubby flange over the body wall, thickened just below it by pressure from above. Use of this technique is characteristic of most East Sussex Downland ware beaded and flanged bowls but it was also employed on some similar Hampshire Grog tempered Ware forms of late 3rd century date (Figure 13).

5C.28. Deep, convex-sided dish form in dirty-grey fabric 5-5B fired patchy black/orange. c. 370-400+ AD. Truleigh Hill 1949-51, Unstratified. The simple, straight-sided dish forms associated with this industry appear very early on in the Roman period. The excavation at the Meeching County Primary School site in Newhaven (Green 1977) produced nine substantial pottery assemblages spanning the period c. 60-270 AD.

The type specimens for bowls of form 5C.17 and its variant 5C.15 come from the late 3rd to early 4th century hearth over the early 3rd century AD Corndryer 6 at West Blatchington (Norris and Burstow 1951: Plate IX, 89-90). The distribution plot for the types (Figure 7) suggests that these vessels were made by new centres of grog tempered ware production in the Shoreham and Brighton areas.

Group 1 is of c. 55-100 AD date and included a few such dishes, differing little from the 3rd century forms illustrated here. The 2nd century groups from Newhaven have much greater numbers of such dishes, all handmade and rarely decorated until the second half of the century, when acute latticed forms 5C.24 and 5C.25 made their appearance (Ibid: Figure 33225). Such latticed forms may owe their inspiration to contemporary BB1 forms and continued being produced until the early 4th century AD.

The pottery from the Findon Close hut site at West Blatchington appears to be of mainly early 4th century date and includes another example of 5C.17, as well as examples of form 5C.18. The latter has had the body thickening below the flange removed to improve its definition.

The transition from latticing to burnished arcading on BB1 dishes at the beginning of the 3rd century AD was not followed by the East Sussex Downland ware potters. They continued to use lattice decoration throughout the century before abandoning all forms of linear burnished decoration by c.350 AD.

Bowl form 5C.19 is a typical East Sussex Downland ware form of the later 4th century AD. As with 5C.18, the bead and flange were still formed by finger pressure, with the resultant thickening of the upper part of the bowl wall removed for better flange definition. In the case of this form, however, the thinning of the wall has been overdone, resulting in a constriction below the flange.

The flanged dish form 5C.22 may have been inspired by a late 2nd century AD BB1 form and appeared at the Newhaven Meeching School site towards the end of that century. The Beddingham pottery sequence has the form appearing at around the same time and indicates that the form continued to be made until c. 270 AD.

Dishes

Deep dishes of Bardown Ware type are rare in 3rd century East Sussex Downland ware assemblages but there are occasional examples like form 5C.23. Such dishes do, however, become far more common after AD 370, in the forms of convex-sided versions 5C.27 and 5C.28. These are the last dish forms to be massproduced in East Sussex Ware and, being concentrated mainly in the east of the region, are probably influenced by the contemporary East Kent industry 7A.16, now being traded into that area (p.67). Shallower versions of convex-sided forms 5C.27 and 5C.28 are also found

Figure 4. 5C.22. Flanged dish form in dirty-grey fabric 5-3. c. 180270 AD. West Blatchington, Corndryer 6. 5C.23. Deep dish form in patchy buff/grey fabric 5-2. c. 200-350 AD. Beddingham, Context 101 5C.24. Shallow dish form with external latticing in patchy buff/black fabric 5-3. c. 150-350 AD. Beddingham, Context 101

18

2: Industry Group 5. East Sussex Wares

Figure 4: Industry 5C dish and storage-jar forms

19

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain in 4th century assemblages and are distinguished in the gazetteer by the suffix (S). They appear earlier and may also have been produced by the East Sussex Downland ware potteries in the Shoreham/Brighton area.

5C.33. Rope-rim storage-jar form in patchy-fired fabric 5-4. c. 350/70-400+ AD. Bishopstone (After Green 1980: Figure 31-8). 5C.34. Rope-rim storage-jar form in patchy-fired fabric 5-4. c. 350/70-400+ AD. Bullock Down (After Green 1980: Figure 31-5).

Dishes with handles Figure 4.

5C.35. Hole-mouthed storage-jar form in patchy-fired fabric 5-4. c. 350/70-400+ AD. Thundersbarrow (After Green: 1980: Figure 31-10).

5C.30. Oval or circular dish form with at least one handle. c. 370-400+ AD. Thundersbarrow Hill 1933, Pit 2. Handled dishes are an extremely rare element in the East Sussex Wealden ware form range but there are examples from Thundersbarrow and Pevensey. The form copies BB1 ‘fish dishes’ (Lyne 2012: Class 9) known to have been traded in East Sussex during the late 3rd and 4th centuries AD.

The very large storage-jars associated with the 4th century AD East Sussex Downland ware industries were invariably made in the very coarse Thundersbarrow ware fabric 5-4 and are exclusively of late 3rd to 4th century date, with an emphasis on the late 4th century AD. An alleged early example in the c. 75-100 AD dated pottery Group III from the Meeching Primary School site in Newhaven can be discounted as the fabric description indicates that it is not in fabric 5-4 and probably intrusive in the pot group (Green 1977: Figure 27-89).

Cups A vessel from rubbish-dumping over the latest road surface inside the east gate at Pevensey appears to be copying a samian Dr.33 cup. It is particularly crude in having a thick, heavy wall with vertical draw-marks on its exterior surface, brought about by pulling up a clay coil to make the vessel. The layer in which this pot was found also contained a number of late 4th century coins, including one of Honorius. Much of the pottery contained in this rubbish-dump could well have been deposited after AD 400, including the vessel under discussion (Lyne 2009: Figure 29, 39).

Perhaps the earliest well-stratified ‘Thundersbarrow ware’ storage-jar is the rather aberrant 5C.31 from the c. 293-320 AD dated occupation in Trench XIII against the interior face of the north wall of the Pevensey shorefort. The more typical storage-jar forms (5C.32, 33 and 34) are a phenomenon of the late 4th century AD and characterised by heavily finger impressed rims. Distribution.

Figure 4.

c. 200-270 AD. (Figures 5 and 6, Appendix 4)

5C.29. ?Dr.33 cup copy in dirty medium-grey fabric 5-5 with external oxidised patches. c. 370-400+ AD. Pevensey 1936, East Gate, Context 3.

The percentage-based distribution plot for East Sussex Ware Industries 5B and 5C (Figure 6) shows that during this period as later the main distribution zone lay to the east of the River Adur. The highest percentages of such wares are found in pottery assemblages from sites in the vicinity of the Ouse valley and provide additional evidence for manufacture in that area, as does the distribution of the distinctive jar form 5C.36 (Figure 5).

Lids It is possible that lids continued being made during the 3rd century AD but the evidence is inconclusive. The early 3rd century bathhouse fills at Beddingham contained large quantities of pottery, including only four small lid fragments, which from their condition could easily have been residual.

At the Meeching School site in Newhaven (Bell 1976), an EVEs quantification of the early and mid 3rd century AD pottery assemblage from the upper fills of the Site 6 enclosure ditch had East Sussex Ware accounting for 60% of all of the coarse pottery. The similarly dated cold plunge bath fills at the Beddingham villa (Context 101) produced much pottery with East Sussex Ware making up a similar 59% of the coarseware assemblage.

Large dry-storage jars Figure 4. 5C.31. Bead-rim storage jar form with vertical linear burnish on its exterior. In grey-buff fabric 5-4 fired black internally. c. 293-320 AD. Pevensey 1936, Pit 20.

Twenty two kilometres west of Newhaven, the West Blatchington site near Hove had a number of late 2nd and 3rd century corndryers containing large assemblages of pottery (Norris and Burstow 1951). The excavators dated the abandonment of these corndryers to dates

5C.32. Bead-rim storage-jar form in patchy-fired fabric 5-4. c. 350/70-400+ AD. Thundersbarrow (After Green 1980: Figure 31-9). 20

Figure 5: The distribution of early 3rd century girth-cordoned jar and bowl forms (Scale 1:800000).

2: Industry Group 5. East Sussex Wares

21

Figure 6: Percentage distribution of Industries 5B and 5C wares c.200-370 AD (Scale 1:800000)

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain

22

2: Industry Group 5. East Sussex Wares ranging between the late 2nd and late 3rd centuries AD but the associated pottery assemblages suggest that all of the dryers ceased being used simultaneously during the 270s. East Sussex wares make up 43% of the coarsepottery from Corndryer 6.

Appendix 4, listing combined East Sussex Downland and Wealden ware form percentages of the various assemblages, may give us some clues as to trading mechanisms driving the distribution of these wares during the earlier 3rd century. The most obvious feature in this table is the fall of in the percentages of jars west of the Ouse estuary. The pattern of bowl and dish percentages is less coherent, but suggests that the longer distance trade was driven not only by the commodities carried in the girth-cordoned storage-jars but, in the manner of contemporary BB1 marketing, by a shortage of and resultant demand for open forms.

The westerly fall-off in percentages of East Sussex Downland wares in the Brighton area was largely due to competition from a wheel using pottery industry in the Barcombe area producing slate-slipped white and red wares and other pottery and trading them south along the Roman London to Brighton road. The presence of these wares in the assemblages from the two Ouse Valley sites is minimal but at West Blatchington they make up 22% of the Corndryer 6 assemblage.

c. 270-370 AD. (Figures 6 and 7, Appendix 4) The period between AD.250 and 270 saw not only the abandonment of most of the iron-working sites in the Weald and a sharp decline in East Sussex Wealden ware production but considerable disruption to occupation in Downland and coastal Sussex. The Meeching School, Newhaven occupation site was abandoned, as was most of the West Blatchington site; where use of the corndryers ceased and occupation became confined to the small aisled hall farmstead and a couple of ephemeral huts. The production of East Sussex Downland Ware at the Ouse Valley centre went into terminal decline and smaller East Sussex Ware potteries, situated further to the west in the Brighton area, increased in significance.

Any decrease in percentages of grog tempered wares from sites to the west of Shoreham can only be guessed at. There are no significant c..200-270 AD dated pottery assemblages until we reach the ritual site at Muntham Court,Findon on the South Downs behind Worthing. A late 2nd to early 3rd centuries midden deposit at this site (M5) has a mere 1.5% East Sussex Downland ware and indicates that very little such pottery was being marketed west of the River Adur. Most of the few East Sussex wares which did travel west beyond that river consist of 5C.36 girth-cordoned storage-jars: an example is known from the Findon ritual shaft and another from the Shipphams site in Chichester and raise the possibility that these large jars were used as packaging for some kind of produce from the Ouse valley such as salt.

The pottery from occupation layer 14 over the fills of the cold-plunge bath at Beddingham shows the decline of the Ouse Valley industry well: there was a fall of over 40% to 21% in the East Sussex Downland ware share of the coarseware assemblage; dated by coins to c. 270300+ AD.

Fragments from at least 15 girth-cordoned storage-jars are present in 3rd century assemblages at the Barcombe villa and a further three at the detached bathhouse in Church field next door. Assemblages of similar date at the Pond Field site further to the north have East Sussex Downland wares making up 39% of the pottery in a late 3rd to early 4th century ditch, including fragments from another girth cordoned jar. Further north still at the Bardown iron working site on the Sussex/Kent county boundary, another girth cordoned storage jar has traces of resin sealent on its neck and shoulder, providing further evidence that jars of this type were used as packaging for some kind of commodity.

The Findon Close hut site at West Blatchington was of early 4th century date and the large pottery assemblage from it also shows a fall by 21% to 22% from the percentage of grog tempered ware in the 3rd century Corndryer 6. At Slonk Hill behind Shoreham, a small c. 270-300 AD dated pottery assemblage from Pit 32 had grog tempered wares making up just 10% of its coarsewares. It is likely that small grog tempered ware production centres, falling into Peacock’s category of Household Industry (1982:17), were already established in the Brighton area before AD 270 but remained insignificant during the last 30 years of the 3rd century. The 4th century, largely pre AD 370 dated, pottery assemblage from the west barrow ditch at Slonk Hill does, however, indicate an increase in the significance of locallyproduced grog-tempered wares on this site to 26% of all of the coarse pottery

No assemblages large enough for quantification have been located from sites on the South Downs east of Newhaven but fragments from 5C.36 type storage-jars are known from Bullock Down and Jevington. Fragments from other jars of this form are known from sites much further to the east and north such as Tongs near Hastings and Westhawk Farm, Ashford (Lyne 2008, 239). Another such jar is known from Lullingstone villa in the Dart valley (Pollard 1987, Figure 75-147) and together with the other examples gives girth cordoned storage-jars a considerably greater distribution zone than other East Sussex Downland ware forms.

There is a progressive decrease in the amounts of grogtempered wares marketed west of the River Adur. At Chanctonbury on the northern escarpment of the South 23

Figure 7: Distribution of Industry 5C developed beaded-and-flanged bowl forms (Scale 1:800000)

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain

24

2: Industry Group 5. East Sussex Wares Downs, the excavations (Bedwin 1980) revealed that the stratification on the temple site was greatly disturbed but that Context 108 was of late 3rd century AD date, with a small coarseware assemblage including 8% East Sussex Ware. Ten kilometres to the north west at Wiggonholt, the late 3rd and early 4th centuries AD occupation contexts on Site B had a nominal 3% of the wares, comparable with the 1% present in the contemporary pottery assemblage from the Highdown bathhouse to the south.

The c. 300-370 AD figures suggest a fall off in the percentages of jars away from the now established Brighton /Shoreham area source, in a similar manner to the pattern in relation to the Ouse Valley in the earlier 3rd century. A similar fall off in the percentages of bowls and dishes is not evident although the presence of such vessels is constantly insignificant in assemblages during this period other than in that from the west barrow ditch at Slonk Hill

Although the amounts of East Sussex ware are small on sites between the Adur and the Arun, there does seem to be more than in the earlier 3rd century AD. This can be seen in the distribution map for developed beaded and flanged bowl forms 5C.15 and 5C.17 (Figure 7). This greater westward penetration by the wares at a time when their production was in decline is almost certainly due to both the main production centres being situated somewhat further to the west than previously and the collapse of the Arun Valley pottery industry.

c. 370-400+ AD. (Figure 8, Appendix 4). There was a revival in the fortunes of the East Sussex Ware industries during the mid 4th century AD. The late 4th century AD site assemblages with the highest percentages of such wares come from sites to the east of the River Ouse. The highest percentage of such wares is from Feature 8, Layers 1 and 2 at Frost Down, Beachy Head. East Sussex Wares account for 55% of the coarsewares in this assemblage, followed closely by the 50% of the Group IX assemblage at Bishopstone, 16 kilometres to the west.

The Pevensey shore fort was founded between AD 293 and 296: the earliest occupation deposit in Trench 1A and the early 4th century AD layers 4 and 3E in Trench XIII both have less than 10% East Sussex Downland ware. Here at Pevensey, coarseware assemblages up-to c. 350 AD are dominated by local wheelturned greywares, produced specifically to supply the fort and nowhere else.

Only 15 kilometres to the north-east of Frost Down is the Saxon Shore fort at Pevensey. The make up of the second road inside the East Gate contained pottery and coins of the period c. 340-370/80 AD. The ceramic assemblage is small but includes a considerably increased East Sussex Ware EVE percentage of 26%, compared with the 10% of the early 4th century assemblage from Trench XIII. The grog tempered wares consist almost entirely of jars but fragments from a Thundersbarrow storage-jar are also present. Comparison of these two assemblages suggests that the revival in the East Sussex Ware industries took place between 360 and 380 AD.

It would appear that production of fine white and red wares fired grey in the Barcombe area, at the junction of the London-to-Lewes Roman road and the Sussex Greensand one, ceased during the early 3rd century AD. They are present at Beddingham and in contexts with early 3rd century AD pottery at West Blatchington and Chanctonbury. There are none in the Pit 32 assemblage at Slonk Hill or at Pevensey, although they are present in late 2nd and 3rd century assemblages at Polhills Farm, Arlington only a short distance west of Pevensey. The four main competitors for East Sussex Downland wares were now handmade grit and grog tempered Industry 5D wares fron the Lewes/Glynde area, Dorset BB1 arriving by sea in significant quantities at the mouth of the River Adur and at Pevensey, coarse sanded greywares from the Wickham Barn, Chiltington kilns near Barcombe (Butler and Lyne 2001) and variable quantities of Alice Holt/ Farnham ware distributed throughout East Sussex.

The rubbish deposit over the second road inside the East Gate (EG.3) can be dated by coins and pottery to between AD 370/80 and the early 5th century AD. In this deposit, the grog tempered wares account for 30% of the coarse pottery; a percentage almost equal to the 32% Alice Holt/Farnham greywares in the same assemblage. Pottery from the Alice Holt kilns had become the biggest single competitor of East Sussex Downland wares east of the Adur, with the highest Alice Holt/Farnham ware percentage shares of pottery assemblages being reached on sites to the north of and on the northern edge of the South Downs at Pevensey and Wolstonbury (30%).

The pottery form percentages for c. 270-370 AD are tabulated in the same manner as previously and subdivided into c. 270-300 and c. 300-370 AD dated assemblages (Appendix 5). The Corndryer 5 assemblage from West Blatchington is strictly speaking a c. 200-270 AD dated one like that from Corndryer 6 but incorporates later 3rd and early 4th century pottery from occupation above it. The considerable early 3rd century element in the assemblage from Corndryer 5 may be responsible for the high 39% East Sussex Ware and should perhaps be treated with caution.

A small early 5th century pottery assemblage from the fort ditch outside the West Gate at Pevensey was associated with worn Theodosian coinage and was dominated by East Sussex Downland wares (45%). Much of the pottery in this assemblage was comminuted and probably residual but may indicate that these wares were the most significant at Pevensey in the years just after AD 400, when they were joined by Industry 5E products. 25

Figure 8: Percentage distribution of Industry 5C wares. c.370-400 AD. (Scale 1:800000)

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain

26

2: Industry Group 5. East Sussex Wares Further west, a farmstead site on Truleigh Hill, overlooking the Adur walley, was ploughed in 1949-50 for the first time in centuries. A very large pottery assemblage was salvaged, of which the bulk was of late 4th to early 5th century date and was accompanied by a buckle of Hawkes and Dunning Type 1A and a clipped siliqua of Gratian. The material also included some 3rd century and earlier pottery, which was disregarded in the quantification where detectable. The East Sussex Ware share of the pottery was 30% but it is possible that this figure may have been inflated slightly by the accidental inclusion of earlier material. Overwey/ Portchester D type coarse-sanded buff rilled jars and other types are very strongly represented (23%), as they are at Wolstonbury further east. Some of these Overwey/ Portchester D jars are atypically white cored or blackened through and some of the convex-sided dishes lack an external groove below the rim. This suggests that most of these wares did not in fact come from the Alice Holt/ Overwey area but from a production centre much closer at hand. Smaller percentages of this Overwey/Portchester D related fabric are found elsewhere on the East Sussex Downlands to the south and east of these two sites.

The distribution of this ware suggests that it was made on a Wealden iron-working site which had survived the mass-closures of the mid 3rd century AD; perhaps one situated somewhere to the east of Lewes. This notion is further supported by the nature of the fabric filler, with its high percentages of crushed black ironstone and fired clay. Ironstone is uncharacteristic of the chalk and alluvium in the vicinity of the Beddingham villa but is found in profusion in the Weald clays and sandstones.

There is only one late 4th century pottery assemblage of any note from between the Arun and the Adur. The Canada Bottom site at Findon was excavated in 1948. Surviving quantities of pottery are small, but were nevertheless quantified and shown to have East Sussex Wares accounting for 10% of the coarse pottery.

The production of iron requires a firing technology capable of reaching temperatures somewhat higher than those needed for pottery production and it is possible that aspects of the former technology were acquired by these potters. Kilns of some description must have been used but the atmosphere control inside the oven must have been poor and stacking done in a random manner to produce the patchy finish found on most of the pots. It is highly likely that any such kilns had combustion taking place in and around the pot load rather than in a separate combustion chamber below an oven.

Technology The products of this industry are characterised by the exclusive use of the crushed-ironstone rich fabric 5-6. The handmade vessel forms are, for the most part, somewhat similar to those made by the East Sussex Downland ware potters but differ in the nature of their firing. The fabric tends to be fired to a very high temperature, burning out the carbon which is present so that clean bluegrey colouration is achieved, with whitish and orange patches, of a vitrified fabric with rough, hackly texture not dissimilar to that of Rowlands Castle ware.

The percentages indicate that many of these post AD 370 grog-tempered wares were emanating from the extreme east of the region and probably from Pevensey. It is, however, probable that the Brighton-Shoreham area potteries remained in production until the end of the 4th century AD or even later and were a producer of Thundersbarrow storage-jars.: these may very well have been used as beehives. These vessels in oxidised fabric 5-4 are present on most of the sites belonging to this period. They display a considerable variety of rim forms and may not all originate from the same source. The considerable number of such jars from Thundersbarrow would, however, suggest that the Brighton-Shoreham production centres were the main suppliers of such wares.

The date of the industry The only 3rd century pottery sequences of any note from East Sussex north of the South Downs are those from the Beddingham villa, Barcombe villa, Pond Field Barcombe and Bardown. Assemblages of 2nd century date from Beddingham are totally lacking in Beddingham/ Ranscombe ware sherds. The c. 200-270 AD dated fills of the cold-plunge bath (Context 101) have such wares making up a mere 4% of the coarse pottery but the late 3rd century occupation deposit immediately above (Context 14) have Beddingham/ Ranscombe wares accounting for 45% of the coarsewares from it. All this suggests that the industry commenced production at about the time that the bulk of East Sussex Wealden ware pottery production centres closed down.

The breakdown of vessel types per site emphasises the improved fortunes of the East Sussex Ware potteries during the late 4th century AD. With one exception, there was still an overwhelming emphasis on jars, although bowls and dishes were slightly more significant than before. 2.8: Industry 5D. (Beddingham/Ranscombe ware)

The excavations at Ranscombe Hill (Bedwin 1978) produced a late 4th century AD pottery assemblage from a corndryer (Green 1978), which indicates that the industry was still in production towards the end of the 4th century. and may have continued until the very end of production of pottery in the Romano-British tradition in East Sussex.

Source and exploitation of raw materials The largest volumes of sherds in the distinctive fabric 5-6 associated with this industry are in assemblages from the Beddingham villa dated to the period c. 270-300+ AD. 27

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain versions of East Sussex Ware during that period. Some such jars of 5D.4 and similar form have rim edge blackening similar to that encountered on the large Hampshire Grog-tempered ware everted-rim storage-jar type 6A.27, manufactured during the late 3rd and early 4th centuries (p.43). Vessels with such rim edge blackening are given the suffix REB in the gazetteer (Appendix 3).

The forms produced Jars Figure 9. 5D.1. Jar form in hard grey fabric 5-6 fired patchy orange,with raised girth-cordon decorated with overlapping finger-impressed Type A motif and burnished latticing on the lower half. c. 220-70 AD. Beddingham Pit 77.

This industry produced globular jars of form 5D.1 with finger-jabbed girth-cordons during the 3rd century AD, similar to those produced by the East Sussex Downland ware industry. During the late 3rd century AD, this form developed into forms 5D.2 and 3, lacking raised cordons and having a simple finger impressed band around the girth. Unlike the East Sussex Downland ware industry, this one appears to have decorated its jars with stabbed or slashed girth decoration until the late 4th century; such decoration being present on a jar from Ranscombe in a deposit of that date (Green 1978).

5D.2. Jar form in hard blue-grey fabric 5-6 fired patchy beown/buff/orange/black, with finger-impressed motifs of Type B around the girth, burnished-line chevrons on the shoulder and crude latticing on the lower half. Later 3rd century. Bormer cemetery. 5D.3. Reeded-rim jar form in hard grey fabric 5-6 fired patchy orange/grey/off-white with finger-impressed motifs of Type B and wavy-line burnishing around the girth. c. 270-300+ AD. Beddingham Context 14.

A greater percentage of Beddingham/Ranscombe ware jars have neck cordoning or corrugation, compared with those produced by Industries 5B and 5C. In this respect they have something in common with Industry 5A products, although the neck-cordons are less well defined than those on the jars made by that industry. Body latticing is also present on many jars from the 3rd through to the late 4th century AD.

5D.4. Slack-profiled jar form in buff-grey fabric fired brown-black externally and black along the edge of the rim. c. 270-300+ AD. Beddingham, Context 27 5D.5. Everted-rim jar form in black fabric fired brown with dark brown exterior surface. c. 270-300+ AD. Beddingham, Context 1/27.

Bowls

5D.6. Jar form with corrugated neck in pale blue-grey fabric fired buff-grey with rim edge blackening. c. 270300+ AD. Beddingham, Context 14.

A shortage of material means that we know very little about this class of vessel. The bead rim bowl from Beddingham (5D.9) is of a type similar to those associated with the other East Sussex Ware industries during the 3rd century and is interesting in that it comes from a context post-dating the end of the Wealden 5B industry. We have already remarked on the conservatism of this industry, in its continued use of girth decoration on jars well into the 4th century AD.

5D.7. Jar/necked-bowl form in hard black fabric with facet-burnishing on the shoulder and burnished latticing over the lower half. c. 350-400+ AD. Ranscombe Hill (Green 1978:Figure 5-25). 5D.8. Jar/necked-bowl form in hard blue-grey fabric fired patchy brown/orange/black. c. 270-400+ AD. Bormer cemetery.

Bowl form 5D.10 from Ranscombe suggests that a modified version of the bead-rimmed bowl may have continued being made into the late 4th century. The author has not, however,seen this sherd and it may be that the rim is from a lid.

Jar girth decoration motifs. 5D.A. See form 1D.1 5D.B. See forms 1D.2 and 3.

No developed beaded and flanged bowls have been seen in the Beddingham pottery assemblages but form 5D.11 from Ranscombe shows that the industry was producing types similar to those in East Sussex Downland ware during the 4th century.

5D.C. Stabbed herringbone decoration flanking a burnished horizontal line. Ranscombe Hill (Green 1978: Figure 5-26). The jar forms associated with this industry have much in common with the East Sussex Downland ware ones but there are subtle differences. The bulk of the 3rd century jars are of the simple, slack profiled variety, characteristic of both the Wealden and Downland

Figure 9. 5D.9. Bead-rim bowl form. Beddingham, Context 14.

28

c.

270-300+ AD.

2: Industry Group 5. East Sussex Wares

Figure 9: Industry 5D forms

29

Late Roman Handmade Grog-Tempered Ware Producing Industries in South East Britain 5D.10. ?Bowl form with hammer-head rim. c. 350-400 AD. Ranscombe Hill (Green 1978: Figure 5-29).

The ware is absent from the 3rd century AD occupation at Polhills Farm, Arlington to the east of Beddingham but the fill of a late 4th century corndryer at Ranscombe Hill immediately to the north of that villa included a number of large fresh potsherds (Green 1978: p.251). Vessels in two fabric 1-6 variants (Ibid: p.252) make up 36% of the pottery from this corndryer and indicate continuing production by this industry until at least the end of the 4th century.

5D.11.Developed beaded-and-flanged bowl form. c. 350-400 AD. Ranscombe Hill (Ibid.:Figure 5-30). Dishes All of the few dishes that have been examined are seen to be of the deep convex sided ‘dog-dish’ variety. Figure 9.

2.9: Industry 5E. Pevensey Grit-and-Grog-Tempered ware

5D.12. Deep dish form. c. 270-300+ AD. Beddingham, Context 13.

Introduction Pottery assemblages from post AD 370 dated contexts excavated at Pevensey in 1936 include small amounts of handmade grit and grog tempered ware. Most of this material comes from post AD 400 dated contexts, suggesting that we are dealing with Sub Roman pottery production.

5D.13.Similar but more irregular form c. 350-400 AD. Ranscombe Hill (Green 1978, Figure 5-26). Distribution (Figure 10) The Beddingham villa has more pottery from this source than any other site excavated so far. The c. 270300+ AD dated occupation over the cold-plunge bath fills (Context 14) has such pottery making up 45% of the coarse pottery from it, whereas the c. 200-270 AD dated fills beneath had a mere 4.5%. This in itself suggests that the industry came into being shortly before AD 270. When we move south, north and west from Beddingham, we find such wares rapidly diminishing in importance, supporting the idea that the source was to the east of the site and not very far away.

Fabrics Three fabric variants can be distinguished in the limited amount of pottery available for study. The presence of black ironstone inclusions in all of the fabric variants makes them superficially similar to Beddingham/ Ranscombe ware, although Pevensey Grit and Grog Tempered ware has a rougher, more irregular finish. 5.7A. Off white to light grey fabric fired dirty grey to black with subangular 1.00