ae : : ; oENe mae sateen ie =
acre
pee
ane
Eee =
ae tes
= isos wera , peered ‘ is a - bi : : pie Series : He 3 eH 3 =o ~ 25 Se oye : J ae a : : : : eects : Sas > a5 : :
Perec
Pan Neen er
Pitenonacsn Ses
| PURCHASED GENERAL L BVON RW,
inate Se |
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
VOLUME 86 1993
Published by Wiltshire Archaeological and Natural History Society 41 Long Street Devizes SN10 1NS Telephone (0380) 727369
Registered Charity Commission No. 309534 V.A.T. No. 140 2791 91
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE VOLUME 86 (1993)
ISSN 0262 6608
© Wiltshire Archaeological and Natural History Society and authors 1993
Editor: Kate Fielden Hon. Natural History Editor: Patrick Dillon Hon. Local History Editor: James Thomas Hon. Reviews Editor: Michael Marshman Editorial Assistant: Lorna Haycock
Change of Title The journals issued to volume 69 as parts of The Wiltshire Archaeological and Natural History Magazine (Part A Natural History; Part B Archaeology and Local History) were from volumes 70 to 75 published under separate titles as The Wiltshire Natural History Magazine and The Wiltshire Archaeological Magazine. With volume 76 the
magazine reverted to its combined form and title.
We acknowledge with thanks publication grants for this volume from the following bodies: Wiltshire County Council, for C.K. Currie’s paper ‘Excavations at the Postern Mill Site, Malmesbury, 1986-87’; The Royal Commission on the Historical Monuments of England, for Grahame Soffe’s note ‘A Barrow Cemetery and Other Features Recorded by Air Photography at Beckhampton, Avebury’; Wessex Water plc, for the paper ‘Archaeo- logical Investigations at the Shrunken Village of Knook’ by Duncan Coe and Richard Newman; and the Ministry of Defence Land Agency for the paper ‘Recent Excavations of Iron Age and Romano-British Enclosures in the
Avon Valley, Wiltshire’ by Alan Graham and Caron Newman.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior
permission of the Society and authors.
Produced for the Society by Alan Sutton Publishing Limited, Stroud, Glos. Set in Plantin by Alan Sutton Publishing Limited Printed in Great Britain
WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY SOCIETY
COUNCIL AND OFFICERS (31 July 1992)
PRESIDENT
Professor S. Piggott, CBE,
VICE-PRESIDENTS Professor S. Piggott, CBE, FBA FSA Mrs C.M. Guido, FSA Professor W.H. Dowdeswell, MA, FIBio. R.G. Hurn
Chairman J.A. de Normann, OBE,
FBA, FSA
FOUNDATION TRUSTEES B.H.C. Sykes, MA R.G. Hurn H.F. Seymour, BA Dr T.K. Maurice, OBE J.F.Phillips, B.Sc.
MA (1990)
Deputy Chairman Professor W.H. Dowdeswell, MA, FIBio. (1989)
Elected Members
H.F.W. Cory, JP (1987) Dr C. Shell, MA, MMet. (1990) A.J. Lawson, MSc., FSA, MIFA (1987) M. Smith, BA, FICE (1990) Mrs G. Swanton, BA, Dip. Ad. Ed. (1987) D. Williams, MA (1990) Dr F.W. Hanford, MRCS, LRCP (1988) F.K. Annable, BA, FSA, FMA (1991) J.F. Phillips, B.Sc. (1988) Mrs S.J. Buxton (1991) P.R. Saunders, BA, FSA, FMA (1988) E. Elliott, FBPS, B.Sc. (1991) C.R. Elrington, MA, FSA (1989) J.M.G. Kirkaldy, Ph.D, M.Sc. (Econ.)(1991)
Ex-officio Members
*Mrs G. Swanton, BA, Dip. Ad. Ed. H.F. Seymour, BA
D. Shelton, BA, B.Sc., Dip. Ed., FRGS Mrs G. Learner, BA
M. Fuller, B.Sc. (Eng.)
*D. Williams, MA
Miss G. Fairhurst, ACIS
Mrs J. Friend
R.C. Hatchwell, FSA
Chairman, Archaeology Committee
Chairman, Buildings & Monuments Committee Chairman, Industrial Archaeology Committee
Chairman, Library Committee Chairman, Natural History Committee Chairman, Programme Committee Hon. Treasurer
Hon. Publicity Officer
Hon. Keeper of Prints and Drawings
*also Elected Member
Nominated Members
A.E. Edwards, Mrs V.C. Landell-Mills and Mrs P. Rugg T.J. Gaylard
D.J. Willmott, CBE, QBFM
Mrs L. Bennett
A.D. Sawyer
Mrs P. Slocombe, BA
R.L. Pybus, MA, FLA, FBIM, MILAM, FRSA
P. R. Saunders, BA, FSA, FMA
OFFICERS Secretary Curator Assistant Curator Assistant Curator (Natura! Sciences) Assistant Curator (Biological Recorder) Sandell Librarian Assistant Librarian Editor
Members, Wiltshire County Council Member, Devizes Town Council
Member, Kennet District Council Member, North Wiltshire District Council Member, West Wiltshire District Council Wiltshire Buildings Record
Director, WCC Library & Museum Service
Curator, Salisbury & South Wiltshire Museum
G.G. Brown, DMA
P.H. Robinson, Ph.D, FSA, AMA
Mrs H. Thorn, BA, AMA
A.S. Tucker, B.Sc.
Miss S. Scott-White, MA
Mrs P. Colman, Dip. ELH, FRSA
Mrs L. Haycock, BA, Dip. ELH, Cert. Ed. Miss K.J. Fielden, D.Phil.
Contents
John Thurnam (1810-1873) and British Prehistory by STUART PIGGOTT
Recent Excavations of Iron Age and Romano-British Enclosures in the Avon Valley, Wiltshire by ALAN GRAHAM and CARON NEWMAN with contributions by M.J. ALLEN, C. GAMBLE, J. EDE, J. EGERTON, P. HARDING, C. FITZGERALD, A.V.C. JENKINS, L.N. MEPHAM and S. WYLES
Excavations at the Postern Mill Site, Malmesbury, 1986—87 by CHRISTOPHER K. CURRIE with contributions by TRUDIE C. CURRIE, MARK A. LOFT and PAUL ROBINSON
Archaeological Investigations at the Shrunken Village of Knook by DUNCAN COE and RICHARD NEWMAN with a contribution by LORRAINE MEPHAM
Excavations at New Park Street, Devizes, 1990 by MILES RUSSELL with contributions by MARK GARDINER, DIANA JONES, BRONWEN LAUGHLIN, MICHAEL LAUGHLIN and JON WALLIS
A Survey of the Parish Church of St Swithun at Compton Bassett, Wiltshire by ANDREW REYNOLDS
A 12th-century Effigy in Salisbury Cathedral by D.I. STROUD
Bradford-on-Avon in the 14th Century by R.B. and B.K. HARVEY
The Poet as Historian: Fresh Light on Urban Fire Damage in Elizabethan Hungerford by NORMAN HIDDEN
The Revd William Lisle Bowles (1762-1850): The Need for a Re-appraisal by DOREEN SLATTER
Notes
A Barrow Cemetery and Other Features Recorded by Air Photography at Beckhampton, Avebury by GRAHAME SOFFE
58
73
102.
13(
Iron Age Coins From Cunetio and Mildenhall by PAUL ROBINSON
A 7th-century Gold Coin from Shalbourne by LORD STEWARTBY
A Note on the Identification of ‘Alvestone’ in Wiltshire Domesday by JASON ST JOHN NICOLLE Studies of Small Invertebrate Populations by BARBARA LAST and WILFRED DOWDESWELL
xcavation and Fieldwork in Wiltshire 1991
eviews Joan Blanchard, Malmesbury Lace (ROBERTA PRINCE) John Chandler, West Country Landscapes: The Vale of Pewsey (KEN ANNABLE) Albert W. Cooper, Benjamin Banks 1727-1795. The Salisbury Violin Maker (ALISON THIES) D.A. Crowley (editor), A History of Wiltshire, VCH Vol. 14: Malmesbury Hundred (K.H. ROGERS) Christopher Gingell, The Marlborough Downs: A Later Bronze Age Landscape and Its Origins (1.F. SMITH) Desmond Hawkins (editor), Wessex: A Literary Celebration (FELICITY GILMOUR)
‘Lorna Haycock, John Anstie of Devizes 1743-1830. An Eighteenth-Century Wiltshire Clothier and Pamela Colman, The Baker’s Diary: Life in Georgian England from the Book of George Sloper, A Wiltshire Baker, 1753-1810 (NORMAN ROGERS)
Phyllis Hembry (editor), Calendar of Bradford-On-Avon Settlement Examinations and Removal Orders
__ 1725-98 (STEVEN HOBBS)
‘Pamela M. Slocombe, Wiltshire Farm Buildings, 1500-1900 (J.E.C. PETERS)
‘Christopher Stell, An Inventory of Nonconformist Chapels and Meeting-houses in South-West England (JOHN CHANDLER)
Julian Thomas, Rethinking the Neolithic, New Studies in Archaeology (FRANCES HEALY)
\David Tylden-Wright, fohn Aubrey: A Life (MICHAEL HUNTER)
{Wroughton History Group, Wroughton History, Part 5: Between the Wars (MICHAEL MARSHMAN)
Idex
|] | |
i
147 150 151 152
158
165 165 167 168
168
170
171
173 173
174 15 176 MIME
179
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY SOCIETY
The Society was founded in 1853. Its activities include the promotion of the study of archaeology (including industrial archaeology), history, natural history and architecture within the county; the issue of a Magazine, and other publications, and the maintenance of a Museum, Library, and Art Gallery. There is a programme of lectures and excursions to places of archaeological, historical and scientific interest. The Society also maintains the Wiltshire Biological Records Centre at the Museum.
The Society’s Museum contains important collections relating to the history of man in Wiltshire from earliest times to the present day as well as the geology and natural history of the county. It is particularly well known for its prehistoric collections. The Library houses a comprehensive collection of books, articles, pictures, prints, drawings and photographs relating to Wiltshire. The Society welcomes the gift of local objects, printed material, paintings and photographs to add to the collections.
The Wiltshire Archaeological and Natural History Magazine is the annual journal of the Society and is issued free to its members. For information about the availabiliy of back numbers and other publications of the Society, enquiry should be made to the Secretary.
Contributions for the Magazine should be on subjects related to the archaeology, history or natural history of Wiltshire. There is no fixed length. Papers, notes and reviews should be typed on one side of a page only, with good margins and double spacing. The style for footnotes, references and so on should be that found in this issue. The author-date system is preferred for references and footnotes should be avoided unless essential. Contributions of article length should be accompanied by a summary of about 100 words. Two copies, one of which is a top copy, should be sent to the editor at the Museum, 41 Long Street, Devizes, Wiltshire, SN10 1NS. A further copy should also be retained by the author. The editor and subject editors will be pleased to advise and discuss with intending contributors at any stage during the preparation of their work. They will also supply notes, if requested, which may be helpful in explaining house style and in giving advice on the compilation of references and bibliographies, and the preparation of illustrations.
Proofs: authors will receive galley proofs only.
Offprints: ten offprints of each article will be given free or shared between joint authors of articles (not notes or reviews). Further offprints may be ordered from the printer at galley proof stage, when prices will be indicated.
Publication by The Wiltshire Archaeological and Natural History Society does not imply that the Society endorses the views expressed; the factual content and the opinions presented herein remain the responsibility of the authors.
Wiltshire Archaeological and Natural History Magazine, vol. 86 (1993), pp. 1-7
John Thurnam (1810-1873) and British Prehistory
by STUART PIGGOTT
John Thurnam, Medical Superintendent of the Wiltshire County Asylum (1851-1873), made an outstanding contribution to the study of prehistoric man. An assessment of his life and work is made in the light of the halting progress of 19th-century archaeological studies and the rise of professionalism in the field.
In 1989 I gave to this Society a lecture reviewing the circumstances of its foundation in 1853, and among the Founding Fathers I particularly singled out Dr ‘John Thurnam, at that time the Medical Superin- ‘tendent of the Wiltshire County Asylum, today -Roundway Hospital, for his outstanding contribution to the study of British prehistoric antiquities in the ‘mid 19th century (Piggott 1991). In the same year I ‘published a general study of the development of the ‘concept of prehistoric man — the ‘Ancient Briton’ — \from the Renaissance to the beginning of the 19th ‘century (Piggott 1989).
In this book I made the point that, in common with other disciplines, the story of prehistoric archaeology \(and indeed of archaeology in general, covering also |periods of time with historical documentation) did not ‘follow a neat progressive pattern of improvement, each generation building on the achievements of its \predecessors, but was a much more hit-and-miss affair, with troughs of stagnation followed often by an ‘exceptional individual contribution which brought ‘about a re-thinking of the subject. I quoted an historian of palaeontology, Martin Rudwick, on thinking in terms of ‘men of their own time, grappling with problems which they rarely had enough evidence to solve, and solving them, if at all, in terms of their own views of the world’. Such a man was John ‘Thurnam, whose work illuminated prehistoric jarchaeology, but shows all too well the limitations of ‘his age. A point I made (Piggott 1991, 109) was one |made by John Kenyon as an historian (1983, 96), and that was that no support for historical studies came from the universities until the end of the 19th icentury, with the result that ‘all the English historians of the High Victorian period were amateurs’ and not professionals, for no profession existed, and their livelihood had to be provided by other means.
| |
1. Biographical details where not otherwise documented are from | Dictionary of National Biography, 1975.
Archaeology lagged even further behind and, apart from the study of Greek and Roman sculpture and painting that passed for classical archaeology, prehis- tory was an amateur pursuit until well into the 1930s, with notable contributions made in the British field by country GPs. For instance, the first Chair of Archae- ology destined for prehistory was that founded in the University of Edinburgh by Lord Abercromby in 1927, to which Gordon Childe was elected. Before archaeologists came antiquaries. The Society of Anti- quaries of London, founded in 1710, never became a professional body until very recent times, and did nothing except provide a vehicle of publication for talented amateurs like John Thurnam.
Thurnam was born in 1810 of a Quaker family at Lincroft near York. He was privately educated and, debarred by his religion from entry to Oxford or Cambridge, took himself for his medical training to London, where he obtained Membership of the Royal College of Surgeons in 1834. In the same year he became resident medical officer at the Westminster Hospital, a post he held for four years, when he returned to York to become medical superintendent of the Friends’ Retreat. The Quakers had been pioneers in the treatment of mental illness, and the York Retreat had been founded by The Society of Friends and William Tuke in 1792 and opened in 1796. Thurnam was to publish papers on the treatment and statistics of insanity, and in 1849 Statistics of the York Retreat 1796-1840. Under the new Lunacy Acts County Asylums were set up, and Thurnam was appointed Medical Superintendent to that for Wiltshire at Devizes in 1851, a post he was to hold until his sudden death in 1873.!
But already he was embarking on what was to lead him into his entry into prehistoric archaeology — the study of ancient human skulls. As early as 1849 he
2 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
had published Anglo-Saxon skulls from Lamel Hill in Yorkshire (Thurnam 1849; Briden 1984), and more significantly, in the next year a paper on ‘the import- ance of collecting crania in tumuli’ (Thurnam 1850). In 1851, the year of his appointment at Devizes, he had applied for, and had been awarded, a research grant of £50 from the Royal Society ‘for the purpose of procuring exact drawings of well-authenticated skulls of early British races; — the drawings to be accompanied by systematic measurements and by a statement of the evidence on which the determination of the age of the skulls is founded, and to be placed in the Collection of the Royal Society’. Behind this grant lies an interesting episode in the history of science. In 1849-50 a Treasury grant (under Lord John Russell) had been made of no less than £1,000, a huge sum at the time, to the Royal Society for scientific research. The Society had the additional advantage of having as its President Lord Rosse, the third Earl, himself a distinguished scientist as an observational astron- omer. He had built his own great reflecting telescope on his Irish property at Parsonstown, and with it had discovered the first spiral nebulae; he was also an astute administrator who handled the government grant with great skill. Recipients were to be assessed by a panel of referees appointed by Council which in Thurnam’s case were three (pers. comm. from Alan Clark, Deputy Librarian R.S.; printed Council Minutes June 19, 1851; Hall 1984). The first of Thurnam’s three referees was John Phillips (1800— 1874), a geologist, Keeper of York Museum 1825-40, and later Deputy Reader in Geology at Oxford and Keeper of the Ashmolean Museum 1854-70 (Ovenell 1986, 211-229). Thurnam knew and quoted his Rivers, Mountains and Sea-Coast of Yorkshire (1853). His other two referees were William Bowman, an ophthalmic surgeon and anatomist at the Moorfields Hospital 1851-76, and William Sharpey, Professor of Anatomy and Physiology at University College London 1836-74.
The study which Thurnam was pursuing would at the ume be known as ‘craniometry’, a word used since the beginning of the 19th century to avoid any association with the contemporary fashionable pseudo-science of phrenology. This, described by an anonymous writer in the Edinburgh Review of 1815 as ‘a piece of thorough quackery from beginning to end’, was in the first place propagated by F.J. Gall (1758- 1828) and J.G. Spurzheim (1776-1832), and particu- larly in Edinburgh by George Combe (1788-1858)
2. R.G. Latham, a medical doctor turned philologist, had published his book, The English Language, in 1843.
(pers. comm. from Professor M.H. Kaufman; Kauf- man 1987). In brief the belief propagated was that the exterior conformation of the cranium accurately reflected the mental and psychological characteristics of the individual, and ‘feeling your bumps’ became a popular craze, appealing to the same section of the public which today turns to the irrational comforts of ley lines, UFOs, Ancient Wisdom and corn circles. In his professional reading Thurnam would certainly have come across Gall’s Anatomy and Physiology of the Nervous System in General and the Skull in particular, either in the French of 1810 or the subsequent English translation. But a more important influence was to modify his whole archaeological approach (and that of his eventual collaborator, Joseph Barnard Davis) which goes back to the great German anatom- ist J.J. Blumenbach (1752-1840), who has been described as ‘the true founder of craniology’ (Penni- man 1952, 55—56). If Thurnam or Davis encountered him directly, it must have been in the original German, as his works were not published in an English translation until 1865. But Thurnam could have come across his classification of mankind into five varieties of one species — Caucasian, Mongolian, Ethiopian, American and Malay — largely on skull form, in another source, that of J.C. Prichard. His professional reading would have led him to Prichard’s work on insanity, notably A Treatise on Insanity (1833 and subsequent edd.), and thence to his Researches into the Physical History of Man (1813 with later edd. to 1847; repr. 1973; Stocking 1973). Here Blumen- bach’s racial thesis is taken to an extreme form, and by the last chapter all the peoples of antiquity have been classified into ‘races’, including Scythians, Ger- mans, Pelasgians and Celts.
This bogus ethnicity was not only to have its deleterious effect upon students of prehistoric Britain, but became known to the general public. When in 1862 Anthony Trollope published his Orley Farm, he makes the priggish young Lucius Mason (educated in Germany) opt as a vocation for ‘Philology . . . or as a profession, perhaps literature. I shall devote myself to
philology and the races of man’. Later he is found | making drawings of cranial profiles with ‘both Pri- | chard and Latham’ by his side.” Trollope as a skilled |
novelist would not have introduced a topic not of contemporary interest to his readers.
But our main concern is its influence on Thurnam |
in his collaboration with Dr Joseph Barnard Davis in
the monumental Crania Britannica of 1865 (hereafter |
[
| |
JOHN THURNAM AND BRITISH PREHISTORY
CB). The genesis of the book is interesting, and largely set out in the eventual Preface to the 1865 CB. A Derbyshire barrow-digging squire, Thomas Bate-
man (1821-1861) of Middleton Hall, amassed not
only a large collection of artifacts from his exca-
‘vations, but also of human crania from inhumation
burials. These he encouraged a Staffordshire doctor of
his acquaintance, Joseph Barnard Davis, to publish in | August 1849. Craniometry was, as we have seen, very | much in the air at the time. Davis accordingly, in
1852, issued a prospectus for a Crania Celtica, model-
ling the title on the American physician S.G. Mor-
ton’s Crania Americana of 1839, and in the ‘Celtica’ indicating a racial or ethnic approach. This was spotted by Daniel Wilson, the author of the outstand- ing Archaeology and Prehistoric Annals of Scotland,*
‘who on seeing the prospectus wrote to Davis urging
an alternative and non-contentious Crania Britannica
which was in fact adopted. Thurnam, who, as we saw, had just received a Royal Society grant for a similar project, contacted Davis, and a partnership ensued.
_ Davis visited Devizes, and the book was published by
‘subscription in fascicles or ‘decades’, beginning in
1856, and forming the eventual two volumes of CB,
| with its fine lithographed plates of skulls, in 1865.
With individual entries, Thurnam’s main contri-
| bution was Chapter V, ‘Historical Ethnology of
| Britain’, a book-length prehistory of Britain of 111
| folio pages, something like 8,500 to 10,000 words.
| The book is a remarkable achievement, but unfor-
| tunately its basic approach is vitiated by two fallacies.
|The first is that the authors found themselves con- strained to assign each skull, from the Neolithic
if
_ onwards, to an ethnic group or ‘tribe’, in terms of the tribal names recorded in the Classical sources, notably Caesar and Tacitus. This was made possible by what | we would find an absurdly short chronology, in which _ prehistory was contained within a century or so before | the Roman Conquest. Thus Thurnam in his chapter i}
|
| | |
i
|
| dated Stonehenge to ‘the epoch of Divitiacus’ about
I
| Stonehenge monograph of 1876). |
In the collection of the crania for description and comment in CB, Davis of course had the Bateman | collection to draw on. Thurnam had received from | Daniel Wilson casts and originals from Edinburgh, | including the Phrenological Museum, but both he and | Davis regretted that Colt Hoare and William Cun- | nington in their excavations had left the skulls and
bones of inhumation burials in the ground, removing
, 100 BC (a date endorsed by William Long in his
| | 3. Published in 1851. Wilson incidentally invented the term ‘pre- | historic’.
5)
the grave-goods and then re-filling. Thurnam had access to some skulls from early excavation in the chambered long barrows of the Cotswolds such as Rodmarton and Uley, but the Colt Hoare material remained inaccessible. Thurnam, however, found a means of re-excavation through his professional ap- plication of the enlightened practice of his treatment of mental illness brought from the York Retreat. Believing devoutly in the beneficial effects of the contact of patients with the works of God in nature, he landscaped and planted the grounds of the Asylum (now Roundway Hospital) where all patients spent five hours a day out of doors, and many were encour- aged to take part, with the staff, in rural activities such as farming and gardening. With such an insist- ence on occupational therapy, Thurnam realised that he had a potential unpaid labour force for the re- excavation of Hoare’s and other sites which might yield prehistoric crania. Digging was just digging; there was no excavation technique then or for many decades to come. (Pers. comm. from Mr R.K. Newman of Devizes, formerly at Roundway Hospital.) In 1859 he and his team dug into what we now know to be the western chamber out of five in the West Kennet long barrow near Avebury, excavated in 1955-56 (Piggott 1962). The land owner gave permis- sion with the proviso that none of the visible capstones should be moved, so the 1859 diggers burrowed under that at the west, literally and metaphorically in the dark (a three-ton fragment fell in on them and had to be extricated). They found in fact what we now know to be a virtually unique feature shared by the un- touched chambers, a filling of chalk rubble with quantities of pottery and artifacts, with the coveted skeletons and their crania on the floor. Thurnam was not unnaturally puzzled by the Late Neolithic sherds in the Peterborough and Fengate styles which he recovered: they were ‘of the British or Celtic type. . . assimilated more to the style of the “drinking cups” . .and to that of the vases found in the Celtic barrows of North Britain and Ireland’, a comparison with Early Bronze Age ‘Food vessels’ by no means wide of the mark. For long his sherds remained the only Late Neolithic pottery from southern Britain from an archaeological context. When, in 1863, he turned his therapeutic team to unchambered earthen long barrows at Norton Bavant he was rewarded not only by crania, but among the skeletons left by Hoare the greater part of an Early Neolithic lugged bowl, the first of its type from such a context (see Kinnes 1992).
We now come to Thurnam’s main contribution to the Crania Britannica, Chapter V of the 1865 volumes. Though grandly dedicated to Queen Victoria, there is a subsidiary dedication to Blumenbach and Morton (of the Crania Americana). We must remember the ethnic and shortened chronology fallacies built into the work. Thurnam begins his chapter with Geolo- gical and Palaeontological Preliminaries, sensibly con- cluding that human occupation of Britain was post-glacial, with a modern fauna and subsequent to its insulation from the Continent, quoting Agassiz, Lyell, Wilson and inevitably, Prichard. His second section is on the Earliest Inhabitants, whom he brings from the north-east coast of Gaul and who are there- fore Celtic. He dismisses a non-Celtic substrate population despite the views of Wilson and Prichard, but the Silures on the authority of Tacitus could be Iberian, though he finds the cranial evidence uncon- vincing. In the third section, on Early Historical Notices, he is really at home, as in the subsequent section on Britain as known to the Romans. Of course, he has to start with the Phoenicians and the Cassiterides — ‘there can be no doubt Gaditanian vessels, from the emporium of Tarshish, traded to the south-west of Britain’ but there is no proof that they founded any colony on the shores of Britain. But we then have a long account of the Greek sources, and even a quotation from ‘the learned Bochart’ and his Geographia Sacra of 1674. In fact the attitude is still very near to the 17th century, and the same goes for the long section on the Roman sources, where Thur- nam shows a formidable knowledge of the original Latin texts, and fits the archaeological and cranial evidence into a chronology contained within a century or so before Caesar’s invasion, and embracing every- thing from the Neolithic to the Iron Age. In my general study of the development of archaeological thinking on prehistoric Britain (Piggott 1989) I pointed out how Caesar’s own statement on peoples of the interior as contrasted to those of the south coast, including the Belgae, had a disastrous effect on the understanding of British prehistory up to about 1800, and here we have it still accepted and enlarged on over six decades later. Despite his contacts with Daniel Wilson, Thurnam in CB ignores the Scandinavian ‘Three Ages’ system: Wilson had in 1846 arranged a visit to Edinburgh of the Danish archaeologist J.J. Worsaae, a leading proponent of this stone, bronze and iron taxonomy originally set out by C.J. Thomsen in 1836. Thomsen’s own book was translated into
4. Lubbock learnt Danish to read the relevant literature in the original.
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
English by Lord Ellesmere as A Guide to Northern Antiquities in 1848; Worsaae’s book developing the scheme by W.J. Thoms as The Primeval Antiquities of Denmark in 1849 (Daniel 1975, 38-56; Ash 1981). Wilson in his Archaeology and Prehistoric Annals of Scotland (1851, 18) wrote that ‘the system of classifica- tion now universally adopted in the nomenclature of archaeological science — the Stone, Bronze and Iron periods ... is justly esteemed the foundation of Archaeology as a science’. But not by the authors of CB in 1865: Thurnam (CB I, 55) quotes Wilson (1851, 177) ‘of the existence of a primeval race prior to the Celtic no doubt can now be entertained’ but only to dissent. But at least a valuable corpus of crania had been assembled, and Thurnam had already in 1856 (CB, Pl. 54) commented on the distinctive characteris- ucs of the craniometry of skulls from chambered long barrows. By 1863 (CB Pl. 59) he had got the skulls from the unchambered long barrows at Winterbourne Stoke and Tilshead and was able to write
For this part of Britain (and this is probably applicable to the rest of the island) the writer believes that, in regard to British sepulchral remains, a sort of axiom, though with certain limitations, has now been established, to this effect .. . Long barrows — Long skulls; Round barrows — Round or short skulls [his italics]. (See Kinnes 1992.)
Davis had to disagree — ‘our colleague dissents from this part of the writer’s views’ that the long barrow skulls were distinctively dolichocephalic; they were just examples of synotosis (Davis had disposed of the Neanderthal skull in similar terms in 1864 (CB 230-31)).
There could be no prehistory for Thurnam or Davis, working within the model of the Celtic tribes, but John Lubbock, as a young man still in his 20s was publishing five essays in the Natural History Review of 1861-64 which in the very year of CB, 1865, were published as one of the most influential archaeological books of the century, Pre-historic Times, as illustrated by Ancient Remains, and the Manners and Customs of modern Savages, which endorsed and gave a chrono- logical significance to the Three Ages system* and divided the Stone Age into two, calling that associated with an extinct fauna Palaeolithic, that with a modern fauna Neolithic — indispensable terms still with us today. Lubbock was a remarkable man: he inherited a baronetcy in 1865 and was elected a Fellow of the
JOHN THURNAM AND BRITISH PREHISTORY
Royal Society in the same year; he was an active and ‘progressive reformer in Parliament (he introduced Bank Holidays) and was created Lord Avebury in 1900. Crania Britannica represented an archaeology of the past in 1865; Prehistoric Times that of the future. ‘And it must be remembered that already in 1859 John Evans had established the association of humanly made flint tools with an extinct glacial fauna and Darwin had published The Ongin of Species. Thurnam’s main contribution to British prehistoric archaeology was embodied in two substantial papers published by the Society of Antiquaries of London in their Archaeologia (Thurnam 1869; 1871). Having ‘materially assisted in compiling a corpus of ancient crania, he saw that a complementary survey should be made of the grave-goods associated with the prehis- ‘toric skulls, and to this end made four visits for notes and drawings of the Colt Hoare collection, then stored in the cellars at Stourhead, in 1865-71, with the encouragement successively of Sir Hugh and Sir ‘Henry Hoare (Thurnam 1871, 120 n.). The collec- ‘tions were to come on loan to our Museum in 1878 and were acquired by purchase in 1883 (Piggott 1991, #112). The two papers were ‘On Ancient British Barrows’, ithe first on Long Barrows, the second on Round ‘Barrows. The material dealt with was primarily from Wessex and the Cotswolds, with such more northerly ‘material as he could glean from Bateman and a few other sources in print. His classification (1869, 168) was into a Stone period comprising Long Uncham- ‘bered and Long Chambered barrows: he had got as far as thinking in ‘periods’ but had not adopted Lubbock’s ‘Neolithic’ of 1865. This was followed in ‘Part II (1871) by barrows of a ‘Bronze period’, recognised as subsequent. He gives a table of 21 ‘unchambered long barrows and inevitably gives refer- ences to Caesar and Pomponius Mela to illustrate the likely rituals involved; for grave-goods he gives leaf- arrows and the Norton Bavant bowl. He notes that secondary burials include not only extended Anglo- ‘Saxon inhumations but ‘Drinking Cups’ ‘of the latest in decorated type, such as are usually met with in
ithe most modern circular British tumuli’. All the primary skulls are dolichocephalic. For the cham- bered long barrows he lists 21 in the Severn-Cotswold varea, essays a typology of chambers, and thinks that ‘infant burials are ‘the children of an aggressive enemy of the encroaching Belgic tribe’ (1869, 222). Grave- igoods again include leaf-arrows, and as for the pottery ‘at West Kennet, this was not of the dolichocephalic long barrow builders but ‘their successors the brachycephalic Belgae’; the long barrow people were
|
5
Dobuni ‘doubtless an aboriginal tribe . . . prior to the advent of the Belgae . . . this tribe of the “Britons of the interior”, may have lived on, perhaps down to Roman times’ (1869, 235-36).
For his Round Barrows he lists almost 400, all from Wessex, with a few from the north (as with the long barrows). He made a classification which was to endure — bowl, bell and disc — noting that the latter two were confined to Wiltshire and Wessex, though J.Y. Akerman had told him that there were a few in Sussex. Those of Salisbury Plain and South Wiltshire were of the Belgae, of North Wiltshire the Dobuni. The makers of the bell and disc barrows were the builders of Stonehenge, dating before but up to the Roman conquest. All inhumations in round barrows were crouched, cremations unaccompanied or in urns.
Thurnam then turns to the grave-goods, and for the first time makes a formal acceptance of the Three Ages system, writing
That division of primeval antiquities into those of stone, bronze and iron, which has of late obtained so much note; and which, since the researches of Thomsen of Copenhagen and Nilsson of Lund, became known to English antiquaries, has been more or less generally adopted (1871, 403).
But he still finds chronological difficulties:
If the barrows of Wiltshire and the adjoining counties be not the burial-places of the Britons
. Immediately preceding the Roman con- quest, but belong to one earlier by some cen- turies, it would result that the tombs of those who withstood Julius and were subjugated by the generals of Claudius are still to seek. [Except for the Yorkshire Iron Age barrows as at Arras, the rest must be] Ancient British, and as certainly, we think, of the bronze age of that people (1871, 486).
In dealing with the grave-goods in detail, Thurnam starts with pottery and then turns to implements and weapons but, for reasons which will become apparent, it is more suitable here to reverse this sequence. In dealing with stone objects, he distinguishes perforated battle-axes, unknown earlier, and similarly barbed- and-tanged flint arrows and daggers. He draws atten- tion to ‘bracers’ and their first identification as archers’ wrist-guards by A.H.W. Ingram in WAM 10 (1867), page 109, and lumps all bronzes together from flat to socketed axes, knives and spear-heads, and considers them indigenous rather than imported on
6 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
the ground of the occurrence of moulds. The technological change to iron working came ‘at the time of the invasion of Julius’ but bronze continued to be used, for shields for instance. He had already noted in his Chapter V of CB the Witham shield (found 1826) but ‘this example is however much surpassed in beauty by that found as we write, in 1857, in the Thames to the west of London’ — the famous ‘Batter- sea’ shield.
Thurnam’s pottery classification in the main has endured to recent times: Cinerary Urns, Food Vessels, and Incense Cups in various subdivisions of type, and above all, what he called Drinking Cups and which were re-named Beakers by Abercromby (1912) in order to stress the Continental affinities of at least some of them. Thurnam proposed a _ tripartite classification into Types a, (i and y, which piece of Greek affectation Abercromby sensibly reduced to A, B and C. Owing to the rarity of metal (copper and gold) associated with the Drinking Cups, Bateman thought they should come early in any sequence, but Thurnam thought the reverse. ‘It may rather be supposed that they belong to a late period when, from different causes, as the influence, perhaps indirect, of Roman culture, the burial of weapons had become rare, whilst that of vessels for food and drink had received no discouragement’ (1871, 389).
But the ABC division of Beaker types, essentially Thurnam’s, remains with us in modern archaeology. It was early seen that the B type was the earliest and virtually a component of the varied Late Neolithic pottery styles: it could be regarded as a ‘Pan European’ type. David Clarke (1970) produced a Beaker corpus with a complex and unworkable classification which was never found acceptable, and two Dutch archaeologists in a subsequent review- article (Lanting and van der Waals 1972) reverted to the tripartite division, with Bell Beakers (the old Type B) closely related to counterparts on the Continent, with the old Types A and C (Long and Short Necked) as insular developments in respectively the south and north of Britain.
There still remained (and remains) the problem of Thurnam’s original separation of ‘long’ and ‘short’ skulls: the dolichocephalic gracile population of the Neolithic as contrasted with the brachycephalic robust individuals of the subsequent Beaker and allied Early Bronze Age contexts, associated with the earli- est copper and gold metallurgy. The skeletal facts remain, but are they to be interpreted as the incoming of a new population group into a hitherto Neolithic Britain? The recent rapid advances in genetic science seem to confuse rather than clarify the problem,
especially in the current climate of archaeological thought that favours indigenous development within prehistoric societies as against population movements or ‘invasions’. Thurnam’s craniological observations first defined a situation which is with us as an unresolved problem today (see Kinnes 1992).
Thurnam’s work for prehistoric archaeology covered two decades, from his Royal Society grant of 1851 to his unexpected death from a stroke in 1873. With Davis, he was profoundly influenced by Pri- chard (and behind him, Blumenbach) into adopting a classificatory model of British prehistory in terms of the tribal areas recorded in the Classical sources, and was slow to admit the basic validity of the Danish Three Ages classification of the prehistoric past, and its potentialities. From the first he appreciated the value to archaeology of the corpus of classified information, from the skulls of Crania Britannica to the Bronze Age grave-goods of his two classic papers of 1869 and 1871. His absolute chronology, compress- ing prehistory into a few centuries before the Roman Conquest, may seem ludicrous today, but we our- selves continued to put up what we can now see to be absurdly short chronologies until the recognition and refinement of scientific techniques, beginning with radio-carbon in the 1950s. For the period 1850-70 Thurnam’s work for prehistoric archaeology in Britain remains unique, and as a corpus still not without its value today. The subject had to wait for several decades after the 1860s until work of a comparable calibre was undertaken.
BIBLIOGRAPHY AND REFERENCES
ABERCROMBY, HON. J., 1912 A Study of ihe Bronze Age Pottery of Great Britain and Ireland, Oxford
ASH, M., 1981 ““A fine genial hearty band”, David Laing, Daniel Wilson and Scottish archaeology’ in A.S. Bell (ed.), The Scottish Antiquarian Tradition, Edinburgh, 86-113
BRIDEN, C., 1984 ‘John Thurnam, 1810-1873’ in P.V. Addy- man and V.E. Black (eds.), Archaeological Papers from York, York Archaeological Trust, 163-65
CLARKE, D.L., 1970 Beaker Pottery in Great Britain and Treland, Cambridge
DANIEL, G., 1975 150 Years of Archaeology, London
DAVIS, J.B., and THURNAM, J., 1865 Crania Britannica, London
DICTIONARY OF NATIONAL BIOGRAPHY, 1975 (Micro-reduced edn.), Oxford
HALL, M.B., 1984 All Scientists Now, London
KAUFMAN, M.H., 1987 Death Masks and Life Masks (Cata- logue of an Exhibition, Edinburgh University 1987)
KENYON, J., 1983 The History Men, London
KINNES, I., 1992 Non-megalithic Long Barrows and Allied
Structures in the British Neolithic, B.M. Occas. Paper 52, |
London
_ JOHN THURNAM AND BRITISH PREHISTORY
LANTING, J.N., and
VAN DER WAALS, J.D., 1972 ‘British Beakers as seen from the Continent’, Helinium 12, 20-46
OVENELL, R.F., 1986 The Ashmolean Museum 1683-1894, Oxford
PENNIMAN, T.K., 1952 A Hundred Years of Anthropology,
London PIGGOTT, S., 1962 The West Kennet Long Barrow: Excavations 1955-56, London
- —— 1989 Ancient Britons and the Antiquarian Imagination,
London
—— 199] ‘The Background and Beginnings of the Wiltshire Archaeological and Natural History Society’, WAM 84, 108-15
PRICHARD, J.C., 1813 Researches into the Physical History of
Man, (facsimile repr. Univ. Chicago Press), London (1973)
7
STOCKING, G.W., 1973 (ed.), ‘From Chronology to Ethnology: James Cowles Prichard and British Anthropology 1800— 1850 in J.C. Prichard, 1813 (see above), ix—clxiv
THURNAM, J., 1849 ‘Description of an Ancient Tumular Cemetery . . . at Lamel Hill, Near York’, Arch. Fourn. 6, 27-39; 123-36
— 1850 ‘Observations on Danish Tumuli and on the Importance of Collecting Crania Found in Tumuli’, Arch. Fourn. 7, 34-35
— 1869 ‘On Ancient British Barrows . . Barrows)’, Arch. 42, 161-244
1871 ‘On Ancient British Barrows . . Barrows)’, Arch. 43, 285-544
WILSON, D., 1851 The Archaeology and Prehistoric Annals of Scotland, Edinburgh
. (Part I. Long
. (Part I. Round
Wiltshire Archaeological and Natural History Magazine, vol. 86 (1993), pp. 8-57
Recent Excavations of Iron Age and Romano-British Enclosures in the Avon Valley, Wiltshire
by ALAN GRAHAM and CARON NEWMAN
with contributions by M.J. ALLEN, C. GAMBLE, J. EDE, J. EGERTON, P. HARDING, C. FITZGERALD, A.V.C. JENKINS, L.N. MEPHAM and 8S. WYLES
Wessex Archaeology was invited by the Ministry of Defence Land Agency to excavate and observe four sites, A to D, along the line of a proposed gas main between Durrington and Netheravon, in the Avon valley, north of Salisbury. At Figheldean (Site A) the excavations revealed some evidence of Bronze Age activity, in addition to substantial evidence for a Ist to 4th century rural Roman settlement contained within an Iron Age enclosure. Near Durrington, the pipeline cut across another Iron Age enclosure, known as the Packway Enclosure (Site B), previously excavated in 1968. The remaining two sites, C and D, revealed very little of archaeological interest.
GENERAL INTRODUCTION
Background
During late January and February 1991, excavations took place to the south of Netheravon, Wiltshire, ahead of the laying of a Southern Gas main for the Ministry of Defence’s Property Services Agency. The proposed method of pipe-laying involved placing the pipe within the trench immediately after trench cut- ting, followed directly by backfilling. This did not allow for any observation of the trench during con- struction work. Following negotiations between Wilt- shire County Council, Wessex Archaeology, the Property Services Agency and Southern Gas, it was agreed that a period of four weeks should be allowed for excavation of selected areas in advance of the pipeline construction. Despite severe weather condi- tions, the excavation was completed on schedule, according to a brief set by the County Archaeological Officer for Wiltshire.
Location and archaeological background
The pipeline route (Figure 1) extends from the exist- ing gas main close to the Durrington roundabout (SU 152441), northwards along the west side of the Avon valley, parallel to the A345, to a point just south of the army camp at Netheravon (SU 148477), a distance of about 4km. At Netheravon the pipeline divides. One branch runs eastwards across the valley floor, before turning north to Netheravon, a distance of about 1.25km; the other branch goes north-west to the west
side of the camp, a distance of 700m. A separate stretch of pipeline, joining the existing gas main, runs for a distance of about 1.25km from Bulford (SU 175432), eastwards, following the line of the A3028.
Near Figheldean (SU 15054690 to SU 46974760), the pipeline crosses an area of dense archaeological activity (Figure 1), defined as Site A. There are a number of round barrows extant on the ridges, and across the valley to the west of Figheldean the line of the trench crosses the site of a tumulus visible on aerial photographs as a ploughmark (SMR No. SU14NW616). Another barrow (SMR No. SU14 NW6155), visible as a soilmark on aerial photographs 25m to the south-west, also survives as a low mound. To the west is a field system (SMR No. SU14NW655), recorded on early aerial photographs. A scatter of finds along the ridge to the north of the group of tumuli suggested the presence of a probable Romano-British settlement (SMR No. SU14NW302). The finds included imported pottery and metal objects, with over 150 coins, mainly dating to the 3rd or 4th century AD. A Romano-British building (SMR No. SU14NW301) was partially excavated in 1907 beneath the army camp at Netheravon, near to the pipeline.
Site B (Figures 1 and 18) was located to the north of Durrington Walls (SU 15154410 to SU 15154227), and was chosen to examine the Iron Age site known as the Packway Enclosure and its environs, parts of which were excavated in 1968 (Wainwright and Long- worth 1971, 307). The kite-shaped enclosure, consist-
\
iN WILJTSHIRE (¢
\
~~
NETHERAVON
“
47 —
Built-up area
Land over 91m OD
FIGHELDEAN
Route of pipeline
Field system
Romano-British finds scatter
Barrows
Probable villa
Packway enclosure
Y
YZ Cla DURRINGTON —”
AVE} \-
<
Durrington Walls
Figure 1. The Durrington—-Netheravon Pipeline: location in relation to local sites
10 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
ing of four sides each approximately 65m long, is defined by a deep, steep-sided ditch, and has an entranceway in its southern side (Figure 18). The excavations of 1968 took place during construction of the roundabout at the junction of the A3028 and A345, and work did not begin until a considerable amount of chalk had already been removed. It was thought that this was possibly the reason for the paucity of features found within the enclosure (limited to two pits in an area of c.1,800m’). The enclosure was dated to the later, pre-Roman Iron Age on the basis of pottery from the upper fill of the ditch sections and the contents of the two excavated pits which lay within it. A smaller ditch to the east of the enclosure contained earlier material, dating to the Neolithic and Bronze Age.
Site C (SU 17654320 to SU 18054185) lay to the north-east of an extensive spread of tumuli, on a section of pipeline that runs from a point just south- east of Bulford village, on the eastern side of the Avon valley, along the northern side of the A3028, for a distance of 350m (Figure 1).
Site D was located on the eastern side of the Avon valley (SU 15404825 to SU 15404855), along a north- eastern extension of the Durrington to Netheravon pipeline, leading up to Netheravon airfield (Figure 1). Confined to the line of the slope on the eastern side of the valley, along the scarp at the end of one of the ridges of Figheldean Field, it was placed along the edge of a field on the eastern side of the minor road which runs along the east side of the Avon valley, in order to cross the south-west end of a visible lynchet on the northern side of a dry valley running down to the River Avon.
Methodology
The chosen areas for excavation comprised 4m wide machine trenches of varying length, along the line of the easement, the width being determined by the size of the pipelaying trench. All features were recorded within the 4m width, but only those occurring within a 2m wide section, precisely on the line of the proposed pipe, were excavated, so as to avoid unnecessary disturbance of the archaeological depos- its. Three sites were excavated (Figure 1): Site A, 800m long; Site B, 150m long; and Site D, 200m long. Archaeological work along the Bulford pipeline was confined to the observation of a single site, C, 2m wide and 350m long.
The major effort was concentrated in Site A where a high density of features was encountered in the southern half of the trench and excavation continued throughout the four weeks. Resources were concen-
trated here for the last three weeks of the project but, because of the number of features and prolonged sub-zero temperatures, one area had to be left unexca- vated, although the features were recorded in plan. Within the central 2m of the trench, sections were excavated across at least 75 per cent of the features visible.
Site B was completed within the first week, as the density of archaeological features was low. Sites C and D proved unproductive, apart from a section through the end of a lynchet in Site D, and their excavation and backfilling were completed within a few days. The detailed description which follows relates only to Sites A and B, which were the major areas of archaeo- logical interest. Details of Sites C and D can be found in the site archive, which is deposited at the Royal Arullery Museum at Larkhiil.
Site A: Figheldean (SU 15054690 to SU 46974760; Figures 1 and 2)
Introduction
The site lay parallel to the A345, on the western side of the Avon valley, opposite the village of Figheldean. Its southerly part lay across the end of a ridge, above the 91m contour, forming a broad eastern extension of Netheravon Down and sloping gently down to the line of the A345, before dropping steeply to the river. The bedrock was Upper Chalk, overlain by about 0.3m of ploughsoil. The only exception to this was in the base of the dry valley, where colluvial deposits were encountered above the chalk, to a maximum depth of 1.6m. Once topsoil had been stripped to reveal the surface of the chalk, archaeological features were immediately apparent, as were furrows in the chalk caused by recent ploughing. In the dry valley at the northern end of the trench, parts of two negative lynchets were located on the valley sides. In the bottom of the valley, a machine-dug section was made through the layers of colluvium, in order to acquire molluscan evidence for the interpretation of the palaeoenvironment. The main concentration of features, however, lay on the fairly level ridge top in the southern part of the trench. They extended over a distance of some 360m, and almost without exception were contained within an area defined by two substan- tial enclosure ditches.
Most of the features belonged to a Roman settlement situated within the enclosure, and there appeared to be two phases of occupation: early Roman, with a possibility of continuity from an
_ IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY ll
4700 VA L/. rae Bethe Barrow Mound ef <a Ko Cie KEY W/,. | Se O-S Site grid points yy, \ | | 0 50 m 1500 Ditch 64
| [wa
97.08m OD
KEY TO SECTIONS
LT] Grey brown clay loam
Red brown clay loam
fl | | Pale brown silty loam
20-9] Flint
° SJ Burnt flint
Chalk rubble
ED . Limestone
Charcoal
Figure 2. The Bronze Age ring ditch at Figheldean (Site A): plans and sections
12 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
enclosed Iron Age settlement, and late Roman. The phasing has therefore been divided into the prehis- toric features which pre-date the enclosure (pre- enclosure features), and the evidence for the early and late phases of enclosure occupation (early enclosure features, and late enclosure features, respectively). Within the dry chalk valley, environmental samples were taken, providing information on the prehistoric landscape.
THE PRE-ENCLOSURE FEATURES (Figure 15)
Several parts of the trench provided evidence for features and developments within the prehistoric landscape.
The lynchets
A total of five lynchets was recorded along this section of the pipeline route. They were all negative lynchets, surviving as level shelves cut into the hillside.
Lynchet 317 lay along the south slope of the dry valley, at c.93m OD. It was c.0.7m deep and c.10m wide, with two apparent phases. Bronze Age, Iron Age and Romano-British pottery were found in its infill.
Lynchet 494 lay along the north slope of the dry valley. It was c.7m wide and 0.4m deep, and the fill contained no artefacts.
Three negative lynchets were recorded on the side of the ridge at the south end of the trench, spaced c.9m apart. They contained no dating evidence, but were comparable to the post-medieval lynchets in Site B.
Pit 319
At the top of the slope on the southern side of the dry valley, and to the south of lynchet 317, was a pit (319), rounded in shape and c.lm deep, with an uneven, slightly stepped profile (Figure 15). The fills contained weathered chalk rubble. The upper fill was loose flint and chalk (318), contained within a possible recut. The pit contained only undiagnostic flint flakes and fragments of bone but the environmental evi- dence provided an indication of date. The pit is probably early in the prehistoric sequence, perhaps Neolithic, on the basis of the land snail species within the fills indicative of a relatively wooded environment (see section on land-use history on page 45).
The ring ditch (Figure 2)
In the southern part of the trench (SU 14954707), the ring ditch of a ploughed-out Bronze Age round barrow was located and sectioned. The ring ditch
(SMR No. SUI4NW614) lies close to the south- eastern end of the ridge, c.40m from the top of the steep scarp down to the River Avon. The three other ring ditches (SMR Nos. SU1I4NW613, SUI4NW615 and SUI4NW616), and the surviving round barrow (SMR No. SUI4NW612) are known to lie adjacent, but none came within the line of the trench.
Sections 64 and 77 were on the north and south-east sides of the ring ditch, respectively. With an external diameter of c.27m, the ditch was 3.2m wide, 0.9m deep and steep-sided with a flat base. Large quantities of late Bronze Age flint-knapping debris were recov- ered from layers immediately above chalk rubble weathering. There were also a residual barbed and tanged arrowhead, a sherd of middle Bronze Age Deverel Rimbury type pottery, and a number of Romano-British finds from secondary fills.
There was no evidence of internal features, except for a few stake holes, which may or may not have been contemporary with the barrow.
The burials (Figure 3) About 300m to the north-west of the barrows were the remains of two inhumation burials, skeletons 402 and 405. Though no direct dating evidence was found with the burials, they pre-dated the later prehistoric enclosure, for the enclosure ditch had disturbed them. A Bronze Age date could be assigned to them from the nature of their burial, although they were not associated with any recognised form of Bronze Age funerary monument.
Skeletons 402 and 405 were within the remains of a shallow cut, 0.15m deep, immediately below the
4 Edge of Ditch 328 Skeleton 402
Skeleton 405
ey
4
C=
WA LIC Figure 3. Prehistoric burials from Figheldean (Site A)
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 13
ploughsoil. It was not possible to determine whether they were buried separately or in the same grave cut. Both were tightly crouched, with the knees drawn up. The legs, pelvis and parts of the spine and arms survived for skeleton 402; 405 had only the legs in situ. The remaining bones were cut away by enclosure ditch, 328, and human bone fragments were found in its lower fills.
THE EARLY ENCLOSURE FEATURES (Figures 1, 4-8 and Plate 1)
The enclosure was first identified from aerial photo- graphs taken by O.G.S. Crawford in the 1920s (Crawford Aerial Photographs 3544,4439, 4452 and 4482), which show the north-west corner of a sub- stantial ditched enclosure to the west of the pipe ‘trench, upon the same ridge (Figure 4). Untl recently, this parchmark had been included as part of an extensive field system to the west of the trench (SMR No. SU14NW 1230 and Figure 1), and its true nature had not been recognised. A more recent air photograph taken in 1976 (pers. comm. M. Corney, |RCHME), shows the southern end of the enclosure |(Figure 4), and cropmarks of internal features. This photograph was only brought to the attention of -RCHME, and then of Wessex Archaeology, follow- ing the conclusion of the excavation; it shows the enclosure to be located along the top of the relatively |steep slope on the south side of the ridge, with its northern side overlooking gently sloping ground. It appears to have been positioned to make use of the “steep scarp along the west side of the river valley at \ this point.
The internal features in the photograph are high in density and appear to be of similar form to those ‘excavated. It is possible to make out internal ‘divisions, pits and building terraces, examples of | which were excavated along the pipeline route. In | addition, there are a number of circular marks which appear to be roundhouses, some clearly showing /entrances with post pits on either side of the ‘doorways. Cropmark features which lay across the | line of the pipe trench match up exactly with features | within it (Figure 4).
_ The excavated features could be dated to two |periods: late Iron Age/early Romano-British (lst |century BC to 2nd century AD) (Figure 6), and later |Romano-British (3rd or 4th century AD) (Figure 8). 'Excavated features from the earlier of these periods were concentrated in the northern part of the enclos- ure, though some of the unexcavated features in the jgeuthern part could be of this date. They included
pits, small ditches and traces of structures in the form of slots and, probably, post holes.
The enclosure (Plate 1 and Figures 5 and 6)
Within the line of the pipeline, sections of two substantial ditches, 95 and 328, were encountered, marking the line of the north and south sides of the five-sided enclosure respectively, and matching the enclosure cropmark whose east side lay against the steep scarp above the river to the east, partly beneath the present road. Overall dimensions for the enclosure can be suggested as 360m north-south and c.300m east-west, an area of approximately 9 ha.
Ditch 95 was sectioned by machine, so there was no detailed examination of the infill sequence. It was c.7m wide and 3m deep, with steep sides and a rounded base; the upper profile was shallower, prob- ably through weathering and erosion. The lower fills (92 and 94) were chalk rubble, separated by loamy band 93. A small amount of Ist-2nd century AD pottery was in the upper fill, 81.
Ditch 328 was 5.4m wide, with steeply sloping sides. It was excavated to a depth of c.1.9m, with a projected depth of 3m, as in ditch 95. The lowest excavated fills were steeply dished coarse chalk rubble (388 and 386), containing human bone fragments (see above, pre-enclosure features), and separated by a band of soil, 387. The overlying layer, 378, lay across the whole width of the ditch, sloping steeply down from the south, and was thickest in the base of the ditch. There was one sherd each of late Iron Age and lst-2nd century AD pottery, and a single intrusive medieval sherd. The upper fill was a thick chalk rubble deposit, 326, with coarse and fine laminations tipping down from south to north, containing one Iron Age sherd and several Ist-2nd century AD sherds. The fill of the remaining slight hollow con- tained a small group of lst-2nd century AD sherds.
The enclosure was obviously once a major feature of the landscape, but the precise date of its construc- tion is unclear. The sequence of fills within the ditches demonstrates that after an initial period of fairly extensive weathering the profiles became stable. This is shown in ditch 95 by a band of loamy soil (93), and in ditch 328 by layer 378. No evidence of a true soil profile was observed within the section. The late Iron Age and Romano-British pottery recovered from below the chalk rubble ditch fills would suggest a possible construction date in the later Iron Age, and this would be supported by the evidence of internal features shown on the air photograph. The chalk rubble fills indicate deliberate infilling rather than weathering, and the tiplines show that it is likely, at
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
14
. Z a TT syseurdor9 0] uoNe[a1 UT YOUN UOMeARSXO :(Y 911) URappEYysiy “p o1n3I.7
OPT/VM orst 00ST 06bT SEES O8bl OLbT
puejsuq WHO uondisosues yder6o0j0yd jerzae yo ()
Say UL RRR pe RET a EEE] sg aa =| 002 Oor 0 069 Ajtsuap ul BurArea sy1ewyioS Lp a ee one LL/b9 WOUP Bury Z S6 491g TSS : s = ie MOLILG }ULLXY T ~ ste _~ _ squrod pub aig) L-d ee ey xaM ~s OOLD Vay \ i} ry Wey to fey Vovak {Ca ul | OLLD SZ61'2 yreuryoied OcLt> SOp pue
6OP sTeling ee
—————
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 15
‘ re
y pie ee PEA rik | DE a 4 pa 3) 2 Plate 1b. Figheldean (Site A): section through enclosure ditch 328
16 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Ditch 328
Projection based on BENS observed profile of Ditch 95 m
96.91m OD
Figure 5. Figheldean (Site A): section through enclosure ditch 328
least within ditch 328, that the material was pushed in from an internal bank. The presence of early Romano-British pottery, recovered from fills over- lying the chalk rubble, suggests that the ditches were filled in no earlier than the early Roman period. Within the trench there was no clear evidence for an entranceway, though observed anomalies in the infill of the excavated section of ditch 95, suggested at the time that the section may have been close to a butt end. The upper fill included bands of chalk rubble, but the darker and loamier upper fills of the ditch, visible in the east side of the section, were not present in the west side, which was consistently chalk rubble. This suggests that the layers sloped steeply down to the east, perhaps because of a ditch terminal a short distance west of the section. The aerial photograph, however, does indicate an entrance 42m to the west of the excavated section. It also shows differing densities of fill within the ditch in the area of the excavation, perhaps reflecting areas of deliber- ate chalk infill. It is likely that the layers noted in the east side of the section represent this sort of infill, rather than a ditch terminal. The photograph also shows an inturned entrance on the enclosure’s north- west side (Figure 4).
The pits (Figures 6 and 7)
Between the enclosure ditches, a number of pits were recorded. Of these, only two (346 and 353), on the basis of pottery, can be assigned to the Ist century AD or earlier, although there were also four pits which could be dated to the Ist or 2nd century AD. There was no demonstrable grouping of pits, though some of the six unexcavated pits close to pit 367 may have been contemporary with it.
Pit 47 was shallow, sub-circular, 1.4m in diameter with steep, even, concave sides, and 0.4m deep. It had two fills, 42 and 46, containing Ist-2nd century AD pottery. It cut an unexcavated feature, 54.
Pit 346 was shallow, sub-rectangular, and 0.6m deep beneath the chalk bedrock. It was partially truncated by a Romano-British platform 342. The fill (334) contained some large flint nodules and two sherds of middle to late Iron Age pottery.
Pit 353 was rounded, 1.7m in diameter, and c.0.9m deep with vertical or slightly undercut sides and a flat base. The primary fill, 364, contained chalk rubble and flint, including burnt flint and charcoal. The upper fill, 352, also contained a band of burnt flint and Ist century AD pottery.
Pit 367 was circular, 1.15m in diameter at the top,
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 17
1328
353
1485 *# 9715
WA/LJC
continued from below left
SITE A: NORTH SITE A: SOUTH
143. 4¥7e5 / Position of Ring Ditch — =e ee / 4 if ie if | | Wa \ \ Extant Barrow Mound ww S » / Ne ee J . acs x, ~ 429 oo i pee 471/470 : > a I '\ = 499 . NS + 4776 | \ s | 374 » \ KEY 448 : 46951500 459 os he Late Iron Age/Early Roman O 480°\ © we) * No evidence of date Wr a
E-T Site grid points
0 50 a | |
continued above right
Figure 6. Figheldean (Site A): late Iron Age and early Romano-British features
18 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Pit 346
97.07m OD TAN
Pit 353
n §S
Pit 367
98.15m OD TAS
Worked stone
Figure 7. Figheldean (Site A): sections through late Iron Age and early Romano-British pits
with curving, undercut sides and a rounded base, c.l.5m deep. The fill, 370, was flecked with chalk rubble, containing lst century AD pottery, bone and burnt flint. A fine chalk rubble layer, 369, c.0.1Sm thick, was probably used to seal off the pit after backfilling; the hollow above it was filled with layer 325 containing 3rd—4th century AD pottery.
Pit 374 was sub-rectangular, 2.3m by 1.7m, with steep sides, and a shallow rounded base 0.35m deep. The lower fills, 451 and 375, contained a high propor- tion of chalk rubble, and pottery of the Ist or 2nd century AD. The upper fill, 371, contained 3rd or 4th century AD pottery.
Pit 470/471 was probably a single, elongated oval, with sloping, often uneven sides; it was 2.6m long, c.lm wide, and a maximum 0.6m deep. The fills, 443 and 444, lay above an uneven layer of chalk rubble, and contained Iron Age and Romano-British pottery of the Ist century AD. It cut slot 428, described below.
The ditch (Figure 6) Only one excavated ditch, 459, could be assigned to this phase. It was aligned north-east to south-west, c.1.2m wide, with a sharp V-shaped profile, 0.65m deep. The lower fills contained late Iron Age pottery, and the upper fill 1st-2nd century AD pottery.
The ditch had no relationship to any other features
of this phase; it was, however, parallel to ditch 472 (see Figure 8 and following section on late enclosure features), 22m to the south. It is possible that the finds recovered from the fills of 459 were residual, and that it was contemporary with 472. The ditches may thus represent part of a small enclosure of the later Roman period, examples of which can be seen on the aerial photograph (Figure 4).
The structures (Figure 6)
Evidence for timber structures was found in most parts of the trench, in the form of post holes and slots. Many of the features could not be dated and the extent of the trench was insufficient for the recog- nition of structure plans, but the limited evidence suggested rectilinear rather than circular or sub- circular forms.
The nature of the excavated slots suggested two types of structure, one apparently using a horizontal sillbeam, the other more substantial, vertical timbers. Without more extensive excavation, no conclusions can be offered as to the nature or function of the structures. Most of the features appear to have dated
to the early post-Conquest period from the pottery |
found within them.
Slot 336 was straight-sided, 4m long, 0.55m wide, and 0.6m deep with rounded butt-ends, which could have held vertical timbers. Aligned approximately
north-south, with no evidence for individual post
positions, it contained lst-2nd century AD pottery.
The slot was isolated within the trench, although
aligned with post hole 348, 1.5m to the north, which
was 0.55m in diameter and 0.2m deep, with a possible flint packing stone.
Linear feature 398 was unexcavated but its section, noted in the sides of grave 397, was shallow, and
surface finds indicate a 1st-2nd century AD date.
Slot 429 was narrow, square-cut, aligned approxi- mately north-south and cut by pit 471/470. It was 0.4m wide, 0.2m deep, and extended for 7m from the southern butt-end. The north end was ploughed out. Similar, undated, shallow slots or gullies were recorded in other parts of the trench.
Slot 448 was U-shaped, 2m long within the trench,
0.4m wide and 0.1m deep. It contained one small
sherd of lst-2nd century AD pottery.
| THE LATE ENCLOSURE FEATURES (Figure 8)
Features dated specifically to the 3rd or 4th centuries
AD lay within the area encompassed by the two
| ditches, 328 and 95, spread over a distance of about | 270m along the line of the trench. In plan some of | these features are comparable with those excavated in | other parts of the trench, and it is probable that a
number of the unexcavated features are also of this
| date. The evidence for settlement of this period
included structures represented by terraces cut into
| the hillside, pits, kilns and a number of graves. The
limited nature of the excavations and the plan of features recovered again prevents any understanding of the wider layout and nature of the settlement.
The enclosure
There was no evidence for any later Roman material within the fills of the enclosure ditch. It may therefore be assumed that the ditch had been filled in by this phase.
| The pits
Sections of four rectangular, or sub-rectangular, pits were excavated, in addition to one group of intercut, rounded, fairly shallow hollows, which could be dated
| to the later Romano-British phase of settlement. A | specific function could not be ascribed to any of these | features.
Pit Group 351/357/359/361 comprised at least four
| intercutting hollows in the north part of the enclos-
ure. These were rounded, c.0.45 m deep, and the fills, spreading across the pits, possibly as one fill, con-
| tained scattered flint and limestone rubble, probable
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 19
ule fragments, animal bone and 3rd—4th century AD pottery.
Pit 374 was a sub-rectangular, 2.3m by 1.7m, with steep sides and a shallow rounded base, 0.35m deep. There was a high proportion of chalk rubble in the lower part of the pit, and it contained a small number of 3rd—4th century AD pottery sherds.
Pit 473 was square, 1.5m across, with vertical sides and a flat base, 0.8m deep. The fills, 469, 468, 465 and 440, contained varying amounts of flint and chalk rubble, with areas of whitish-grey, chalky clay, poss- ibly the remnants of a lining. The fills also contained relatively large quantities of 3rd—4th century AD pottery and animal bone.
Pit 480 was a shallow oval or sub-rectangular hollow with sloping sides 0.4m deep. The fill, 479, contained Ist-2nd century and 3rd—4th century AD pottery, and a coin of Antoninus Pius (AD 138-161).
Pit 490 was rectangular, and lay adjacent to the western side of the trench. It was only partly exca- vated, to 0.4m, and was cut by pit 473 on its west side. The fill, 461, contained 3rd—4th century AD and earlier pottery.
The ditches Two ditches were assigned to this phase.
Ditch 355 was 1m wide and 0.25m deep, with a rounded, U-shaped profile, aligned approximately east-west. It cut through the southern edge of the pit group, 351/357/359/361 (see above, and Figure 8). Its fill, 354, contained 3rd—4th century AD pottery.
Ditch 472 was 1.2m wide, with a rounded profile 0.5m deep, aligned south-west to north-east. In the top of the fill, 475, was a line of flint nodules, 474, along the southern lip. The ditch also contained 3rd—4th century AD pottery.
The structures (Figures 8-9)
Parts of five structures were excavated, each of which consisted of a shallow, flat-bottomed cut forming a levelled area, terraced into the hillside. Their fills contained pottery of a late Roman date.
Structure 72 lay at the southern limit of settlement spread. It was an extensive terrace cut into the Bronze Age ring ditch. Two parts were excavated: an area 8.5m wide across the full width of the trench, and two narrow sections across the east side of the trench. Its overall form is unknown. The larger area was a 0.3m deep flat-bottomed shelf with shallow sloping sides on the north and south sides. There was a possible post hole base in the southern edge. The southern part of the terrace truncated the ring ditch fills (Figure 2, layer 37); the northern part would have encroached
20 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
continued from below left a SITE A: NORTH \) SITE A: SOUTH 342 ‘ } 1485 <0 4 381 Ion Vo q Q ks N @-Grave 397 Me a 6, Grave 400 7 d Ze 47951495 50 q 0 ue oN 15 : q7 90 Grave 61 its \ AE #\ Grave 75 1/ 359/ a Grave 477-2 357 Pr-355 oA H Position of Ring Ditch iss ice we Es / /. \ \ Hh \ \ | | Vy el | | Extant Barrow Mound i“ x 72 yi \ / << S \ / Pt O aS ig -_ One 373 1199 / a 457 4716 / \ Grave 456 / 445 KEY 46951500 o # 490% 473 ge 3rd-4th century Romano-British Q one ee A a No evidence of date a) re E-T Site grid points L 0 50 472 m ee WA/LJC continued above right mm
Figure 8. Figheldean (Site A): late Roman features
|
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY ;
: upon the barrow mound, and was cut 0.25m below the chalk. The two sections were across the probable _ western edge of a more extensive platform, at least 0.4m deep, with part of a coarse flint rubble wall, 71, 0.4m wide. The fills, 71 and 36, contained 3rd—4th century AD pottery. There was no physical rela- tionship across the sections, but the fills, 36, 37 and -71, were similar, and thus probably the same layer.
Structure 342 lay at the apparent northern limit of settlement. It was cut into the chalk, 0.2m deep and _ 8m wide, with shallow, sloping edges. It continued _ west of the trench, but the edges turned in at the east end, probably forming the east side of the feature, although this was irregular and shallow, suggesting _ considerable erosion through ploughing. The north | and south sides were fairly straight, however, and the structure was probably rectilinear. The floor was uneven chalk bedrock, with no trace of wear or | compaction, and covered by a layer of chalk pea-grit.
Terrace 342
Stake Hole
a
21
It was overlaid by 341 and 329, containing concentra- tons of flint rubble, particularly along the northern edge, though these were probably not im situ walling. The fill sealed several features: a shallow gully on the north side, on the edge of which was a circular post hole, 377, 0.15m deep below the base; on the east side, two shallow post holes, 349 and 345, containing burnt chalk fragments; and two possible stake holes. The fill contained 2nd—4th century AD pottery, clay roof tile fragments, box-flue tile and animal bone. Structure 373 was comparable to 342. It was 7m by 3m, and 0.15m deep, with a sloping, slightly curving western edge within the trench; the eastern edge was probably eroded by ploughing, and its original extent is unknown. The fill, 372, contained scattered rubble of flint nodules, particularly concentrated along the western edge; 3rd or 4th century AD pottery; animal bone and oyster shell; burnt flint and fragments of burnt clay. There was no evidence of in situ walling.
Terrace 445
Fine chalk rubble
Limestone
Flint
Figure 9. Figheldean (Site A): late Roman building terraces
22 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Two shallow post holes were cut into the platform base.
Structure 38] lay against the western edge of the trench, and cut slot 336. It was 5.35m wide and 0.2m deep. The fill contained 1st-2nd century AD pottery, but the structure may be a late Roman feature as the finds were possibly redeposited from 336.
Structure 445 was rectangular, measuring 4m north-south by 3m east-west, with steeply sloping sides and a flat chalk bedrock base, 0.25m deep. The south-west corner cut an earlier feature, and against this were remnants of a wall or revetment, made of flint and sarsen rubble (446) in rammed chalk (447), surviving as a single course. Its purpose was to strengthen the sides, where the terrace cut the earlier feature. The fill, 433, contained burnt flint and clay
Kiln 50
Kiln 50
Section reversed SE 95.95m OD nN
fragments, animal bone and 3rd—4th century AD pottery.
The kilns (Figures 8 and 10)
Two kilns were found and excavated: one was a probably sub-rectangular feature, 457 (possibly waisted), and the other T-shaped, 50. The infilling of both can be dated to the later Romano-British period, and they may both have served as corn driers.
Kiln 457 was in the northern part of the settlement area. Grave 456 cut through the centre of the kiln, destroying the relationship between the two halves. The kiln was 2m long, 0.8m wide and 0.65m deep with vertical sides. The northern half was lined with flint and chalk rubble, and a single course of fired brick fragments was laid on the base which was
Kiln 457
Grave 456
Brick
Sarsen
Chalk lining
Figure 10. Figheldean (Site A): the kilns: plans and section
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 23
subsequently overlain by a thick black layer (483) containing burnt twigs and weed seeds. The southern half was lined with a band of flint and chalk rubble and chalk cob, 481, probably the remains of a collap- sed lining and superstructure, containing 3rd—4th century AD pottery. The base was also covered by a thick black layer, 482, containing numerous burnt grains (see section on carbonised seed remains on page 42, below).
Kiln 50 was a small, T-shaped structure with axial
arms, 3.2m long and 1m wide. The arms of the “T’ were relatively shallow, 0.2m deep, with sloping sides, and bore the remnants of a lining of flint nodules, 48. The body, or flue, was 1.6m long, c.0.8m wide, and 0.55m deep. Its southern part was lined ‘with brick and tile fragments, 43. Only the eastern half was excavated, so comparable linings in the 'western half were not exposed. There was no burnt deposit, and little charcoal in any of the fills. The chalk bedrock base, however, showed signs of burn- ing. The primary fill, 53, in the flue, was sealed with large limestone slabs, 56, at the south end, perhaps re-used rooftiles from the kiln superstructure, which | had subsequently collapsed. The upper layers (41 and | 49) contained 3rd—4th century AD pottery, an iron knife blade, and-a large quernstone fragment.
Dog skeleton
0 100 200 300
WA LJC
Figure 11. Figheldean (Site A): detail of burial 61 with dog’s tail
The graves (Figures 8 and 11; Plate 2)
A total of seven graves was excavated, of which four were spaced over an area of about 105m, within the northern part of the enclosure, and three within a limited area of the southern part. The graves shared common traits, and were probably elements of the late Romano-British phase of settlement, though there was little direct evidence to date them. All the graves contained hobnails, and all but two of the skeletons (400 and 456) appeared to have been buried wearing boots.
Grave 61 (Plate 2a) was 2.2m by 0.7m, with rounded ends, and 0.4m deep. The skeleton, 63, was aligned north-south, with the head to the north. The upper torso was twisted to the left, and the arms were raised. There was no evidence of a coffin. Laid within the curve of the upper arms and hands were four vertebrae from a dog’s tail (Figure 11). The upper grave fill contained the remains of a dog skeleton, 60, which comprised parts of the pelvis, legs and feet, and appeared to have been placed in a natural dog crouch. The backbone, ribs and skull were ploughed away; the tail appeared to be from the same animal, but was clearly detached at the time of burial.
Grave 75 was 1.8m by 0.7m and 0.4m deep, aligned east-west. There was no evidence of a coffin, and the skeleton, 74, lay on its right side, the legs flexed and drawn up.
Grave 363 was 2m long and 1.2m wide with vertical sides and a flat base, 0.7m deep, and aligned east— west. There was evidence for a wooden coffin c.0.8m wide, laid in the northern part of the grave, marked by concentrations of nails in the north-east and north-west corners (8 and 4, respectively) and along the south side. Skeleton 368 lay partially on its right side, with the head at the west end. There was evidence for some displacement of bones.
Grave 397 (Plate 2b) was 2.1m by 0.9m, and 1m deep, with vertical sides and square-cut ends, aligned east-west. The skeleton, 404, lay on its right side, with the skull at the west end. The right arm was beneath the body and the legs were slightly bent. There was no evidence for a wooden coffin.
Grave 400 (Plate 2c) was 2.2m by 1m and 0.85m deep, cut squarely into chalk bedrock and aligned north-south. The skeleton, 399, lay partially on its right side, with arms extended in front and the skull to the south. There was no evidence for a wooden coffin. There was a small group of hobnails against the edge of grave, close to the back of skull, but not enough for a pair of boots: perhaps a token in the form of a fragment of footwear.
Grave 456 was 2.2m by 1.2m and 1.1m deep with
24 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
ag BES
ae *
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 25
square-cut sides, aligned east-west. The skeleton, 466, lay partly on its left side, with the head to the east. The grave cut through kiln 457 and the fill of terrace 373. Iron nails were concentrated at the grave corners and along the sides, indicating a wooden coffin. A few hobnails were scattered across the grave.
Grave 477 was 2m long by 0.8m wide and 0.6m deep, with sloping sides, and aligned east—west. There was no evidence for a coffin. The skeleton, 478, lay with its skull in the south-east corner, on its left
side, and was supported by large flint nodules against its back and around the head and chest.
The four northern graves (363, 397, 400 and 456) were deeper and larger than the group to the south. Although their alignment varied, one north-south and three east-west, they were all deep, square-cut features. The location of iron nails in graves 456 and 363 indicated the presence of wooden coffins. The \three burials (61, 75 and 477) in the southern part of the settlement area were smaller and shallower than those to the north, and there was no evidence that wooden coffins had been used. In grave 477, however, large flint nodules had been placed around the head ‘and torso of the corpse. | Hobnails were found in all the burials, and are a fairly common find with inhumations in the Roman | period (Crummy 1983, 51; Manning 1985, 136). The type of footwear represented by these hobnails is ;unknown, although on the basis of finds from Colchester, a distinction has been suggested between sandals, with a small number of hobnails in a simple pattern, and more heavily studded shoes or boots (Crummy 1983, 53). Both types may be represented here, since the number of hobnails per foot ranges | from 11 (grave 75) to 113 (grave 61).
Dating evidence is slight; the presence of hobnails ‘in all the graves, however, suggests the later 2nd to 4th century AD (cf. Clarke 1979; McWhirr er al. | 1982). This is supported by stratigraphic evidence, as | grave 456 had been cut through the fills of kiln 457,
which contained late Roman pottery.
|
| UNDATED FEATURES
Most of the unexcavated 25 per cent of features are likely to have been similar in nature to those investi- | gated. The surface plan of the fills indicates the | presence of further linear features, such as ditches and | slots, pits and, almost certainly, building terraces. | The aerial photograph of the south end of the enclos- ure supports this: showing, for instance, a probable terrace 30m to the north of kiln 50 (Figures 4 and 8). _ Oneither side of the ring ditch, sections of two linear
| | } }
| ) | | | | | | |
features were located. The northern feature, 10m north of the ring ditch section, 64 (Figures 4 and 6), was unexcavated. The southern feature, a gully or ditch terminal, 488, produced no dating evidence. The aerial photograph reveals them to be part of asmall enclosure, and feature 488 appears to be the terminal end of one half of an entrance. The ring ditch is contained within the enclosure, as is one of the late Roman building terraces, 72. This small enclosure is probably an internal property division within the settlement, although there is a suggestion that feature 488 may continue across the enclosure ditch (Figure 4).
THE DRY VALLEY (Figure 15)
The valley is approximately 180m wide, with a smooth, concave profile. At its base, the chalky periglacial deposits are buried beneath a sequence of colluvial deposits, up to 1.5m in depth.
Layer 311, a buried B horizon (subsoil) which underlay layer 312, was a 0.3m deep band of weakly calcareous soil extending over about 25m in the base of the valley, overlying chalk.
Layer 312 was a band of fine, stone-free, worm- sorted soil, up to 0.1m deep, overlying layer 311. It comprised the remnants of a former land surface (buried soil) below layers 110 and 109.
Layers 109 and 110 were up to 1.1m deep, above layer 312. Both were colluvium, which became pro- gressively paler and chalkier towards the top.
A mollusc column was taken through these layers, and although there was no direct evidence of the date of deposition in the form of artefacts, the environ- mental evidence provided an interpretation of the landscape and possible dates of deposition (see section on land-use history on page 45).
THE POTTERY by L.N. MEPHAM
The pottery assemblage comprises 905 sherds (16923g). The majority of the assemblage is of late Iron Age/Romano-British date, but pottery of prehis- toric, medieval and post-medieval date is also present.
The pottery assemblage has been analysed in accordance with the standard Wessex Archaeology recording system for pottery (Morris 1991a). It was divided into five broad fabric groups on the basis of the dominant inclusion type: limestone-tempered fab- rics (Group C), flint-gritted fabrics (Group F), grog- tempered fabrics (Group G), sandy fabrics (Group Q), and fabrics of known type or source (Group E). These groups were then subdivided into 61 separate fabric
26 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
types, according to the range and coarseness of inclu- sions present, and each fabric type was allocated a unique, chronologically-significant fabric code.
The pottery has been quantified by fabric type within each context, and details of vessel form, surface treatment, decoration and manufacture have also been recorded. This basic information can be found in the archive. Prehistoric and Romano-British pottery fabric totals are given in Tables 1 and 2.
The prehistoric pottery
The prehistoric pottery has a potential date range of
middle Bronze Age to middle/late Iron Age although
positive dating within that period is by no means straightforward. Pottery attributed to the late pre-
Conquest period, i.e. early Ist century AD, is
discussed together with the Romano-British mater-
ial, although there may be some overlap in the dating. Eleven prehistoric fabric types were identified:
Cl Soft fabric; moderate (10-20%), poorly-sorted crushed oolitic limestone fragments <lmm; sparse (3-10%) quartz grains <0.25mm. Unoxi- dised.
Fl Hard fabric; moderate, poorly-sorted crushed calcined flint <lmm. Unoxidised; burnished surfaces.
F2 Soft fabric; sparse, poorly-sorted crushed calcined flint <2mm; sparse quartz grains <0.25mm; rare iron oxides. Unoxidised with partially oxidised exterior surface.
F3 Soft fabric; moderate, poorly-sorted crushed flint <2mm; rare quartz grains <0.5mm; rare burnt-out organic material, probably grass or straw; sparse iron oxides. Unoxidised with par- tial surface oxidisation.
F4 Soft fabric; sparse, poorly-sorted crushed flint <1.5mm; sparse quartz grains <0.5mm; rare calcareous fragments, probably chalk or lime- stone <1.5mm; rare burnt-out organic material.
FS Soft fabric; common (20-30%), poorly sorted crushed flint <3mm; sparse quartz grains <0.5mm. Unoxidised with oxidised surfaces.
G1 Soft, soapy fabric; moderate, fairly well-sorted grog <0.5mm; rare crushed flint <0.5mm; sparse iron oxides. Unoxidised with partially oxidised exterior.
G2 Soft, slightly soapy fabric; moderate, poorly- sorted grog <2mm; sparse crushed flint <2mm; rare quartz grains <lmm. Unoxidised with oxi- dised surfaces.
Ql Soft fabric; moderate, fairly well-sorted quartz grains <0.25mm; sparse mica and possible glau-
conite. Unoxidised with partially oxidised sur- faces.
Q2 Soft fabric; sparse, poorly-sorted quartz grains <0.5mm; rare crushed flint <0.5mm. Oxidised exterior only, or firing as for Ql. Vegetable wiping on surfaces.
Q3 Soft fabric; moderate, poorly-sorted quartz grains <0.5mm; spare iron oxides. Oxidised exterior only, or firing as for Ql].
Most of these fabric types are represented by very small numbers of sherds (see Table 1), and in some cases the lack of diagnostic material has meant that close dating is impossible. The association of various fabrics on the site is not particularly useful either, since the majority of the prehistoric material was redeposited.
One sherd of fabric F5 derives from a thick-walled vessel with a finger-impressed cordon (Figure 12: 1). Such fabrics and decoration are characteristic of Deverel-Rimbury vessels of middle/late Bronze Age date (Cleal with Raymond 1990, 240). Two sherds of the grog-/flint-tempered fabric G2 may also be of this date; both sherds are from similarly thick-walled vessels. The dating of the other prehistoric materials is more problematic.
While some of the flint-gritted fabrics, F2, F3 and F4, could also belong to the Deverel-Rimbury tradi- tion, it is just as likely that, together with fabrics Cl, C2, Gl, G2 and Q3, they could be ascribed to the post-Deverel-Rimbury traditions of the late Bronze Age/early Iron Age, as defined by Barrett (1980). Fabrics comparable to those from Figheldean have been found in late Bronze Age/early Iron Age vessel forms at other sites in Wiltshire, for example at Potterne, where flint-gritted fabrics are considered to be early in the sequence, superseded later in the early Iron Age by oolitic fabrics comparable to C1 and C2 (Morris 1991b).
The dating of the remaining fabrics (F1, Q1, Q2) is slightly less ambiguous. The flint-gritted fabric F1 can be distinguished from the other prehistoric flint-gritted fabrics on the basis of surface treatment; sherds in this fabric display a better finish, with burnished surfaces which are typical of the middle to late Iron Age. Comparable fabrics have been found at Boscombe Down West, near Amesbury, and Old Down Farm, near Andover (Richardson 1951; Davies 1981).
The two sandy fabrics, Q1 and Q2, are very similar in appearance, and may be subdivisions of one fabric type. Sandy fabrics such as these are common from the early Iron Age onwards in Wiltshire, for example at Potterne and Groundwell Farm (Morris 1991b; Gingell 1982).
M4 { | | } |
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 27 Table 1: prehistoric pottery by feature, Site A Middle Bronze Age/ Late Bronze Age/Early Iron Age Middle/Late Iron Age Late Bronze Age G2 FS Cl F2 F3 F4 Gl Q3 Fl Ql Q2 Feat Cont No Wt No Wt No Wt No Wt No Wt No Wt No Wt No Wt No Wt No Wt No Wt 65 39 = eS EK Zee ee ee hh KH hh 1 8D peated S08 = = SK CUCL RK OSE CUS 77 37.- 2): 35g 1 7gs= = e=0 > 4- = = = = =) 1 ge - - - - 121132 SMe Se Ba eae IS eS) bg ae 317 BOS Ga eet 2 de = = Og = Se as 328 326 - - - = = = = = = = = = = = = = = = = = lee Sg 386 - - - =~ =~ = = = = = = = = = - - 14 Fg 333 331 - - - -—- 3 23g - - - - = = = = = = = = =~ =~ 2. & 336 B88 = = = = 2 Fg = =—- - |= = = = = = = Ss FS = HK 346 334 Shoes Se eS eS RH SH he eH lh 1 CB 1 445 433 .- = - = = = = = = = = = = = = = = = 1 15g 1 18g 456 454 - - - = = = = = = = = = = = = = = = =~ = 1 4g 459 460 - - - = = = = = = =~ = = = = = = 2 26g = = 442 - - - = = = = = = = = = = = = -~ 4 6lg - - = = 473 468 - - - - = = = = = = = = = = = = = = = = 1 9g 807 = = Ss Hie =.= = L2lg - - = = ws = = = = SS = MOAI 2. 35¢ 2:°37¢ 5 30¢'2 14e I 2g 1 lle 2° 2ig-1 dilg 687g 3 105g 20 188¢
Sources for the prehistoric fabrics are uncertain. Most are likely to be of fairly local origin, although the presence of oolitic limestone in fabrics Cl and C2 would suggest a more distant source, the nearest outcrop of oolitic limestone being approximately 35km to the west.
The distribution of prehistoric pottery by feature is given in Table 1. In all but one case the sherds were associated with Romano-British material and must be considered as residual. Pit 346, in the northern half of Site A, was the only feature to contain non-residual pottery, which was of the late prehistoric Q2 fabric (Figure 12: 4).
The late Iron Age/Romano-British pottery Thirty-nine fabric types were identified, and these have been divided into fine wares and coarse wares.
Fine wares
Nine fine wares of known type or source were identi- fied. Non-British fine wares are represented by one sherd of terra rubra and a small quantity of samian. The terra rubra sherd is in a pale orange fabric with
self-coloured, polished surfaces (Rigby 1973, fabric TR 1(C) or TR 2), and derives from the base of a platter of uncertain form, with an unidentifiable fragment of a radial stamp surviving. Terra rubra is generally considered to have a distribution concen- trated in south-east England (Rigby 1973), but two stamped sherds are known from Oare in north Wilt- shire, in a group now considered to be military in origin and of immediate post-Conquest date (Swan 1975, 40).
No attempt has been made to attribute the samian to the various production centres, although one sherd with a distinctive orange fabric and comparatively poor quality slip may be an East Gaulish product. The only vessel forms recognised amongst the samian are two examples of plain bowls (Dragendorff 31), a common form of the latter half of the 2nd century AD.
Other fine wares of known source comprise various products of the Oxfordshire and New Forest produc- tion centres, in approximately equal quantities. From the New Forest, parchment wares and colour-coated wares of both red-slipped and stoneware type have
Table 2: late Iron Age/Romano-British pottery by feature, Site A
F1100-101 G100-105 SAV BB1 QCoarse QFine TRubra Samian NForest Oxford Feat Cont No Wt No Wt No Wt No Wt No Wt No Wt Wt No Wt No Wt No Wt 47 42 - - 1 7g - - - - 1 4g - - = = = 46 = _ 4 68g - ~ - - - - - - - - = - _ = 50 41 = - 10 279g —- - 3 112g 4 99g = - - - 1 13g - = 43 - - 1 80g - - - - - - - - ~ ~ = - = 49 - - 1 l4g - - 1 llg 13 197g - - - - - 1 2g = - 53 - - - - - - - - 5 57g = - - - a = = = 61 62 - - = - - - - - 1 2g = = = = = = = _ 63 = = = = - - - - 1 lg - - - - = = = - 64 38 - - 4 46g - - - - - - - - - - - - - - 45 2 - ~ - - - - - 2 7g - - ~ - - = = = i2 36 - - 5 54g 1 2g - 4 43g - - 1 2g 2 36g O- - 75 73 - - 3 S6g - - - 2 Slg - - = = - = = = 77 37 - - 8 185g 3 44g 2 log 28 301g - - - - 3 67g ll 114g 51 = = 1 l4g - - - - 5 57g - - - - = = = & 95 81 1 8g 2 26g - - 3 4lg - - - - - = = = - = 317 305 1 6g 3 32g - - - - - - - - - - - - - = 328 326 ~ - ll 49lg - - - - 8 70g - 10g - = = - = = 327 = = 6 97g - - 1 7g 7 = 183g — - = = = = = = 333 331 ~ - - - - - - - a 5g = = = = = = = = 336 335 - - 12 495g 9 175g 1 6g 22 270g = - - - 1 l4g - - 338 - - 3 56g 17g - - 2 135¢ - - - - - = = = 342 329 = - 6 80g - - 2 14g 26 562g = = = - 4 40g 1 17g 341 = - = = - - - - 4 109g 27g - - _ - - 1 4lg 345 344 - ~ - - - - l 3g - = ~ = = = = = = 346 330 - = 2 29g - - 1 log — - - - Lf i5e0 = = = = 351 350 - - 1 38g - - - 4 70g a - 1 2g 1 24g 2 og 353 352 - — 1l 405g - - 1 14g 1 17g - - - - = = - = 364 ~ - 1 28g 1 72g - ~ 2 982 - ~ - - - - - = 355 354 - - 1 7g - - - - 2 45g - - - - 1 3g 1 4g 357 356 - - - - 1 7g - - 7} 5g = = soe = = = = = 359 358 - - 1 384g 1 6g 1 5g Z 50g - - - ~ = = = = 363 362 = - = - - - - - 1 10g - - ~ ~ - - 1 2g 368 23 = = ~ - - 4 218g 1 2g - - - - - - = = 367 325 - - ~ - 1 8g - 11 = 123g - - - - ~ - = = 370 = - 45 1167g 16 160g 1 6g 91 1130g 43g ~ - - 1 2g = - 373 372 1 25g 22 45lg — - - - 38 = 454g - ~ 5 3lg - ~ 5 25g 374 371 2 29g 2 14g — - 11 9g 11 199g = — — = = = = = 375 - - = - 1 l7g - - 1 Sg - - 451 = - - ~ - - - - 1 2g - - - - - - - - 381 379 - - - - 1 l6g - - 2 36g - - - - - - - ~ 397 396 - - 2 20g 1 tIlg 1 482 1 4g - - - - - - - — 400 394 - - - - 1 6g - - - 445 433 - - 15S 458 7 134g 1 lg 80 12llg 8g - 1 lg 1 12g 1 6g 448 435 - - 1 9g - - - - - - - - - - - - - - 456 453 - - - - - - 4 47g 26 323g - - 1 2lg 1 64g | - 454 = - 13 312g 1 59g 7 119g 54 703g 3g - 1 3g 1 7g - - 458 - - 5 286g —- - 13. 133g = 33 489g - - - - 1 lg. - - 457 455 - _ 1 llg -— - 5 50g 40 642g - - - - - ~ - - 481 = - _ - - - ~ - 6 119g - - - - - - _ - 459 437 = _ 1 34g - - - 3 6g - - ~ ~ - - - - 460 - - 1 sg - - - - 2 14g - - - - - - = - 462 - - - 1 7g - 470 444 — - 2 20g - = = = = = = = = = = = = = 471 443 1 9g + - - - 1 25g - - - - - - - 1 2g 472 474 - - 3 29g 1 7g - - 17 281g - - - - - - - - 473 440 = _ 16 66lg — - 5 6lg 32 868g - - - - - - = - 465 - - 2 S7g - - 1 3lg 11 180g - - - - 1 9g + - 468 - ~ 13 334g) - - 2 21g 23 532g - - 1 22g - - - - 469 os = 5 ll4g - - - - ll = 167g - - - - - - - - 477 478 - - + - 1 8g 1 Sg 1 10g - - 480 479 = = 3 78g - 1 4g 34 483g - - - ~ 2 30g - - 490 461 = = 2 39g 1 37g 2 28g 2. 45g - _ - - 2 lig - - 306 as 1 Agree ed ee 1 6g = a oS ie 398 5 169g - - - - - - - - 427 - - ~ - - — = - 2 12g - - ~ - - - - - 452 _ - 2 165g = 9 79g 57 498g - - 1 6g - - - TOTAL 6 = 67g 255 7249g 50 793g 97 1227g 377 6212g 78g 10g 13 103g 24 335g 27 239g Key: Q Coarse = fabrics Q100-111, Q113, QI15, QI17 Q Fine = fabrics Q112, Q114, Q116, Q118, Q119 SAV = Savernake ware
Ql119
—Q112
-Qii4
|
—Q118
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY
been recognised; identifiable vessel forms include indented colour-coated beakers and one bow! with an internal flange and painted decoration (Fulford 1975, type 89; figure 13: 24). Oxfordshire products comprise both plain and colour-coated oxidised wares, including one mortarium (Young 1977, type C100) and several dish/bowl types (Young 1977, types C51, C55 and C94). The percentage of New Forest and Oxfordshire wares together is slightly less than that noted at the nearby site at Durrington (Swan 1971), but the same emphasis can be seen on the New Forest stoneware-type colour-coats in closed vessel forms, and the Oxfordshire red-slipped fabrics in open forms, although the dominance of the New Forest wares observed at the latter site is not apparent.
In addition, five fine wares of unknown source have been identified: Hard, pale-firing fabric; sparse quartz grains <0.25mm; sparse cream-coloured grog/clay pellets <lmm; sparse iron oxides. Oxidised pink/buff with creamy external ‘skin’; wheelthrown. Hard fabric; moderate, fairly well-sorted quartz grains <0.25mm; sparse iron oxides; rare possible glauconite and mica. Oxidised brick red; wheelthrown, with red colour-coat. Hard, pale-firing fabric; sparse, fairly well- sorted quartz grains <0.25mm; rare grog/clay pellets <2mm; sparse red iron oxides. Oxi- dised buff/pink; wheelthrown. Hard fabric; rare quartz grains <0.25mm; rare iron oxides. Unoxidised; wheelthrown, with possible quartz roughcasting on exterior surface. Soft, pale-firing fabric; moderate, well-sorted quartz grains <0.25mm; sparse red iron oxides. Oxidised pale salmon pink; wheelth- rown; painted decoration. None of these fabrics is represented by more than three sherds, and only one vessel form was recog- nised: the rim of a flagon of Hofheim type in fabric Q116, which would suggest a pre-Flavian date for this fabric type (Figure 12: 10). While sources for all five
Q116
fabrics are uncertain, a possible origin for fabrics
Q114 and Qi18 might be sought amongst the colour-
- coated ware industry of North Wiltshire, where a - postulated production centre in the vicinity of Wan-
borough was producing colour-coated beakers in
fairly coarse fabrics, some with roughcast decoration,
in the second quarter of the 2nd century AD (Ander- son 1978). However, one base sherd from Figheldean does not seem to belong to a beaker, which is as yet
29
the only vessel form known to have been produced in North Wiltshire colour-coated ware.
Coarse wares
Thirty separate coarse ware fabric types were identi- fied, including two of known source. Four of the fabrics are ‘catch-all’ types, and may include products of more than one source. The correlation of fabric types and vessel forms is given in Table 3.
E101 Black Burnished Ware (BB1); for fabric description see Williams (1977).
Savernake ware; for fabric description see Swan (1975, fabric 1).
Hard fabric; moderate, poorly-sorted crushed flint <Ilmm; moderate, fairly well-sorted quartz grains <0.5mm. Unoxidised with oxi- dised exterior; handmade with well-smoothed surfaces.
Hard fabric, moderate, poorly-sorted crushed flint <3mm;; rare grog <2mm; rare iron oxides and mica. Oxidised; handmade.
Hard, soapy fabric; moderate, poorly-sorted grog <Smm; sparse iron oxides; rare mica. Generally oxidised but some _ unoxidised examples; handmade. Surfaces smoothed, sometimes burnished, but with grog frequen- tly protruding through on interior.
Hard, soapy fabric; common, poorly-sorted grog <2mm; rare quartz grains <0.5mm; rare mica. Generally unoxidised; handmade with smoothed surfaces.
Soft fabric; sparse, fairly well-sorted grog <1mm; sparse quartz grains <0.5mm; sparse iron oxides. Unoxidised, but some patchy oxidisation; handmade, with well-smoothed surfaces.
Soft, slightly soapy fabric; moderate, poorly- sorted grog <lmm; moderate quartz grains <lmm; sparse iron oxides. Generally unoxi- dised; handmade.
Hard, slightly soapy fabric; sparse, poorly- sorted grog <2mm; sparse crushed flint <2mm; sparse quartz grains <0.5mm. Unox- idised; handmade with smoothed surfaces. Soft, soapy fabric; sparse, poorly-sorted grog <3mm; rare quartz grains <0.25mm; rare mica. Unoxidised; handmade, with poorly- finished surfaces.
Hard fabric; common, fairly well-sorted quartz grains <0.5mm; sparse probable glau- conite. Generally unoxidised but some oxi- dised examples; wheelthrown. A ‘catch-all’ fabric type for coarse sandy wares with clearly
E155
F100
F101
G100
G101
G102
G103
G104
G105
Q100
30
Key. 1: bead rim jar (Fig. 12:5 and 6); 2: high-shouldered bowl (Fig. 12:8); 3: butt beaker (Fig.
Q101
Q103
Q104
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Table 3: vessel forms by fabric (Romano-British coarse wares), Site A
Form 1 2 3 4
BBE 2. 1° = = Savernake I -—- —- — G100) -=72 = = = G101 1 Gl0Q2.. 4 = — = G103 1 GI04 “= = 2= = Q100 1 Q1l0l 1 - - - Ql102. 1 Q103; - = - = Q104 1 Q105 1 Lt Q106 l Ql07, - ‘= = = ©1099" = = = = Qll0 - - - -=- OMe a" => == Qll3. - - - =
TOTAL, 6 22. 1
5
6 7 98) 9 10 Ais 12182 4 ee ee WA een eee J 9) ee; ee Sr io ee ee ee eh ae eee ee PO. A ates oe eee ee = Se [ee ee eee eee ees = eg a ee ae. Sa ee ee Po” 2, Peas a ity eS, ate pace Fe ae ae a ae ee ey bees. $75 a eters. ee ae ae Pt eee a ss A eee oe ee ae ee ee le ee REN RS ee = ZV AD 21d 3. 13 ele o34e,
12:9); 4: carinated
bowl (Fig. 12:12); 5: lid-seated bowl (Fig. 12:11); 6: lid/shallow dish (Fig. 12:16); 7: reeded rim bowl (Fig. 12:15); 8: everted rim jar (1st-/2nd-C) (Fig. 12:7, 13, 14, 23); 9: dog dish; 10: flanged bowl (Fig. 12:21 and 22); 11: dropped flange bowl; 12: straight-sided bow] (Fig. 12:20); 13: flagon (Fig. 12:19); 14: everted rim jar
(3rd-/4th-C) (Fig. 12:17 and 18)
visible probable glauconite, giving a ‘speck- led’ appearance.
Hard fabric; moderate, fairly well-sorted quartz grains <0.5mm; sparse probable glau- conite; rare iron oxides. Firing as Q100; wheelthrown. A ‘catch-all’ fabric type; a finer version of Q100.
Hard fabric; moderate, fairly well-sorted quartz grains <lmm; rare probable glauco- nite; rare iron oxides. Unoxidised; generally handmade; frequent finger-smearing on sur- faces.
Hard, micaceous fabric; sparse, fairly well- sorted quartz grains <0.25mm; sparse mica. Oxidised; distinctive orange/brick-red colour; wheelthrown, frequently burnished.
Hard fabric; moderate, fairly well-sorted quartz grains <0.5mm; sparse grog <S5mm;
Q105
Q106
Q107
Q108
sparse iron oxides. Generally oxidised but
some unoxidised examples; probably wheelthrown. Hard fabric; common, fairly well-sorted
quartz grains <0.5mm. Generally unoxidised, but a few oxidised examples; wheelthrown. A ‘catch-all’ group for ccarse sandy wares; as Q100 without glauconite.
Hard fabric; sparse quartz grains <0.25mm; rare iron oxides and mica. Firing as Q105; wheelthrown. A ‘catch-all’ fabric type; a finer version of Q105.
Soft, slightly soapy fabric; sparse, poorly- sorted quartz grains <0.5mm; sparse iron oxides; rare mica. Okxidised; distinctive orange/brick red colour; wheelthrown.
Hard fabric; common, fairly well-sorted quartz grains <0.5mm; rare iron oxides.
| | |
|
‘RON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 3]
Unoxidised with partially oxidised core; wheelthrown.
2109 Hard fabric; moderate, poorly-sorted quartz <1.Smm; rare grog/clay pellets <3mm. Unoxidised; probably wheelthrown.
2110 Hard fabric; moderate, poorly-sorted quartz <1lmm; sparse mica. Unoxidised; handmade, with poorly-finished surfaces.
Yl11 Hard fabric; common, fairly well-sorted quartz grains <0.5mm. Unoxidised; disctinc- tive black colour; uncertain manufacture, with surfaces frequently burnished.
J113 Hard fabric; moderate, fairly well-sorted quartz grains <0.25mm; sparse iron oxides; rare mica. Unoxidised with oxidised core; wheelthrown. As Q106 but with obtrusive iron oxides.
Q115 Hard fabric; sparse, poorly-sorted quartz grains <0.25mm; rare probable glauconite and mica. Oxidised; probably wheelthrown.
Ql117_ Hard fabric; moderate, fairly well-sorted
quartz grains <0.5mm. Oxidised; wheelthrown.
Vessel forms recognised cover the whole Romano- British period. The continuity of native Iron Age forms can be seen not only in the use of flint-gritted fabrics, but also in the presence of bead rim jars and early Roman everted rim jar types (Figure 12: 13-14) in the grog-tempered fabrics and sandy fabrics (Figure 12: 5-7), a high-shouldered bowl in Black Burnished Ware (Figure 12: 8), and cordoned jars in ‘both sandy and grog-tempered fabrics; while Gallo- Belgic influence is reflected in the presence of butt beakers, lids and carinated bowls in finer sandy fabrics (Figure 12: 9, 12 and 16). Some of this material could indeed be pre-Conquest, particularly the flint-/grog-tempered fabric G104, and the poorly- finished fabric G105. Other sandy fabrics can also be assigned to the early Roman period: fabric Q111 (Figure 12: 6), fabric Q107 and fabric Q113 (Figure 12: 11 and 23).
The majority of the sandy fabrics appear to contain both early and late Roman material. This is particu- larly true of the ‘catch-all’ fabric types Q100, Q101, Q105 and Q106. However, each of these fabric types probably includes the products of more than one source, so this does not necessarily indicate continuity of production at any one kiln.
One fabric which does attest to the continuity of one source is Black Burnished Ware. The jars are dominated by the 3rd/4th century AD type flaring rims, but bead rims are also present. Bowls are almost equally divided between ‘dog dishes’, which have a
currency throughout the Romano-British period, and the 3rd/4th century AD dropped-flange bowls, with a single example of a bead-rimmed, high-shouldered bowl (Figure 12: 8), a form with a clear Durotrigian prototype (Cunliffe 1978, figure A:30, 13).
Only two fabric types could be positively attributed to a known source: Black Burnished Ware (BB1), from the Poole Harbour area; and Savernake ware from north Wiltshire. Sources for the other fabrics remain uncertain. Grey wares in particular are always difficult to characterise, and it is likely that the products of several different kiln centres are repre- sented within the Figheldean assemblage. One centre of grey ware manufacture was located in north Wilt- shire, where three kilns to the west of Swindon were producing such wares from the beginning of the 2nd century AD until the early 4th century AD (Anderson 1979). Another possibility is suggested by the pres- ence of probable glauconite in fabrics Q100, Q101 and Q115, which might indicate a source on or close to outcrops of Upper Greensand. These occur in west and north Wiltshire and contain deposits of glauco- nitic sand. The manufacture of grey wares is sug- gested by wasters and kiln furniture found at Westbury, located on Upper Greensand in the west of the county (Rogers and Roddham 1991, 51). Products of the New Forest and Alice Holt kilns may also be represented, but are not sufficiently well characterised to make identification possible.
Distribution
The distribution of late Iron Age/Romano-British pottery by feature is given in Table 2. Features dated by pottery to the Ist/2nd century AD are mainly pits and ditches (Figure 6). Both sections across the enclosure ditch (ditches 95 and 328) contained early Roman material, though not in primary fills. The early Roman sherds found in four of the seven Romano-British graves are almost certainly residual. Further pits and ditches can be dated by pottery to the 3rd/4th century AD, and other features which have been assigned to this period include the building terraces 72, 342, 373 and 445, and kiln 50 (Figure 10).
Discussion
The range of both fabrics and forms at Figheldean is typical of rural sites of this date in southern England, and there is nothing to suggest that the site represents anything other than a farming community of rela- tively low status. Fine wares constitute only a small percentage (less than 5%) of the total assemblage, which is dominated by utilitarian wares, with storage vessels well represented. A very similar assemblage
32 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
covering the early Roman period was recovered from Bower Chalke in the south of the county (Davies 1990), and the assemblage from Durrington contained comparable late Roman material, although the pro- portion of fine wares at the latter site was higher (Swan 1971).
Various sources for the Romano-British pottery have been discussed, and some contrast can be detected between the early and late Roman periods. In the early Roman period, there are known and postulated sources in north Wiltshire, with only a small amount of Black Burnished Ware from the south, and grey wares from the Greensand to the north and/or west. In the late Roman period, the connection with the south is strengthened, with greater quantities of Black Burnished Ware appearing, together with the New Forest fine wares. At the same time, fine wares are arriving from the north from the Oxfordshire kilns, and there is a greater quantity of Greensand grey wares. If the relative quantities of diagnostic rim sherds can be taken as a rough guide, the intensity of activity on the site would appear to have remained relatively constant throughout the Roman period. Such continuity is not evident at other nearby Romano-British sites in the Figheldean area. At Boscombe Down West, very little pottery later than 2nd century AD was recovered (Richardson 1951), while at Durrington, the emphasis is on the later Roman period (Swan 1971), and a similar pattern is emerging at Butterfield Down, Amesbury (Lehmann in prep.).
Illustrated sherds (Figure 12)
1. Body sherd, fabric F5; raised horizontal cordon with fingertip impressions; handmade. Context 45, fill of ring ditch.
2. Rim of jar/bowl, fabric Cl; handmade; unoxi- dised. Context 331, fill of ditch 333.
3. Rim of slack-shouldered jar, fabric Fl; hand- made; burnished ext. and int; unoxidised. Con- text 460, fill of slot/gully 459.
4. Rim of slack-shouldered jar, fabric Q2; hand- made; oxidised with unoxidised core. Context 334, fill of pit 346.
5. Bead rim jar, fabric G102; handmade; burnished over rim and ext. in narrow band below; unoxidised. Context 352, fill of pit 353.
6. Bead rim jar, fabric QI111; handmade; burnished ext. and over rim; unoxidised. Con- text 370, fill of pit 367.
7. Everted rim jar, fabric G102; handmade; unoxi- dised with patchy surface oxidisation. Context 440, fill of pit 473.
8.
10.
12.
13°
14.
15:
16.
17.
18.
19.
20.
21.
22.
23.
High-shouldered bowl, fabric BB1; handmade; burnished int. and ext. down to shoulder; unox- idised. Context 370, fill of pit 367.
Rim of butt beaker, fabric Q106; wheelthrown; burnished ext; unoxidised. Context 440, fill of pit 473.
Flagon of Hofheim type, fabric Q116; wheel- thrown; oxidised. Context 370, fill of pit 367. Lid-seated jar/bowl, fabric Q113; handmade; partially burnished ext. and over rim; unoxi- dised with patchy surface oxidisation. Context 370, fill of pit 367.
Carinated bowl, fabric Q105; wheelthrown; burnished int.; unoxidised with oxidised margin. Context 440, fill of pit 473.
Rim of large, thick-walled storage jar, fabric G100; handmade; burnished ext. and inside rim; unoxidised. Context 358, fill of pit 359. Jar, fabric G100; handmade; burnished over rim; burnished lattice decoration below; oxi- dised. Context 468, fill of pit 473.
Shallow, reeded rim bowl, fabric (Q105); wheelthrown; burnished int. and at least par- tially ext.; unoxidised with oxidised margins. Context 327, fill of enclosure ditch 328.
Lid, fabric Q106; wheelthrown; burnished ext.; unoxidised. Context 370, fill of pit 367. Upright-rimmed jar, fabric Q101; wheelthrown; burnished ext.; unoxidised with oxidised mar- gins. Context 455, fill of kiln 457. Narrow-necked jar, fabric Q106; wheelthrown; burnished ext.; unoxidised. Context 49, fill of kiln 50.
Flagon neck, fabric Q105; wheelthrown; burnished over rim; unoxidised. Context 458, fill of grave 456.
Top half of large, straight-sided bowl, fabric Q100. Wheelthrown; lightly burnished over and below rim, above matt zone with burnished decoration; unoxidised. Context 440, fill of pit 473.
Bowl of truncated conical profile, fabric Q103; wheelthrown; burnished ext. and int.; oxidised. Context 468, fill of pit 473.
Dropped-flange bowl, fabric Q103; wheel- thrown; burnished ext. and int. Context 468, pit 473.
Rim of large, thick-walled storage jar, fabric Q113; possibly handmade; rim decorated with oblique finger impressions; single perforation from inside of rim which does not completely pierce exterior surface; oxidised with unoxidised core. Context 479, fill of pit 480.
[RON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY
KLIS.
Op
DR AAAA => . ! “ 24
LJC
WA
Figure 12. Figheldean (Site A): prehistoric and Roman pottery
34 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
24. New Forest parchment ware bowl with internal flange. Wheelthrown; painted decoration inside rim. Context 453, fill of grave 456.
The medieval pottery
Only three sherds of medieval pottery were recovered from this site, each in a different fabric. The first is a coarse sandy fabric (E422), common in south Wilt- shire, and very similar to the wares produced at the Laverstock kilns near Salisbury, from the late 12th century at least until the late 13th century. The second and third sherds were fine wares: Coarse Border Ware, a white ware from the Surrey/ Hampshire border (E451); and coarse ‘gritty’ ware, containing poorly-sorted quartz grains (Q401), found widely in south Wiltshire; both are late medieval.
THE CERAMIC BUILDING MATERIAL by L.N. MEPHAM
A total of 61 pieces (3928g) of ceramic building material was recovered, twenty-nine of which can be identified as Romano-British. This includes ten frag- ments of flue tile, with scoring for plaster, and two imbrex fragments. A further 17 fragments have been tentatively identified as Romano-British on the basis of association with the diagnostic fragments and by similarity of fabric. This total does not include bricks from the structure of kiln 457, which were not collected, but does include a sample retained from the tile wall within the structure of kiln 50.
All the Romano-British ceramic building material was recovered from within the area of the enclosure. A concentration of 20 fragments came from contexts associated with the kiln (50) in the southern half of the enclosure; the remaining fragments were distributed throughout the northern half, with no noticeable concentration. In all cases the ceramic building material was associated with pottery of late Roman (3rd/4th century AD) date, although it is likely, especially in the case of the kiln, that building material had been reused from another nearby source, and may be of earlier date.
The remaining 32 fragments can be identified as medieval or post-medieval brick and roof tile frag- ments.
THE FIRED CLAY by L.N. MEPHAM
A total of 54 pieces of fired clay (1451g) was recovered from 10 contexts. Nine fragments had no recognisable features. All the remaining fragments derived from flat slab-like objects, with a thickness range of
14-22mm. Surviving curved edges on thirteen frag- ments indicate that these slabs were originally circu- lar, with a diameter of approximately 200mm. Most of the edge fragments are simply rounded (Figure 13: 2), although a few show a slight ‘lip’ (Figure 13: 3). The slab fragments generally have one surface, probably the upper, carefully smoothed and the other roughly finished, occasionally with superficial grass impres- sions.
The fabric of all the slab fragments is fairly similar, comprising a sandy matrix with rare calcined flint fragments and rare linear voids. All the fragments are oxidised, and several are of a very friable nature, perhaps indicating that these fragments have been subjected to greater degree or duration of heat.
The slabs are of uncertain function. Circular clay slabs have been found associated with late Iron Age/Ist century AD pottery kilns, and are usually interpreted as components of kiln floors, or spacers (Swan 1984, 64-6). Some have been found, for example, in late Iron Age features at Boscombe Down, where they are tentatively identified as ‘oven covers’ (Richardson 1951, pl. vi). These examples, however, are generally perforated with single or multiple holes, which are not evident in the Fighel- dean examples. Moreover, there is no supporting evidence for the presence of pottery kilns in the area.
The dating of the Figheldean slabs is ambiguous. All the slabs derive from features associated with pottery dated throughout the Romano-British period, although it should be noted that the majority of these features also contained residual sherds of late Iron Age pottery.
All the slab fragments were recovered from within the area of the enclosure, and show a marked concen- tration within a 50m stretch of the trench, associated with structures 373 and 445, and surrounding pits and ditches. All these features are dated on pottery evi- dence to the 3rd/4th century.
Illustrated objects (Figure 13)
2. Part of a circular clay slab; upper surface smoothed. Context 372, fill of platform 373.
3. Part of a circular clay slab. Slight ‘lip’ on under- side; upper surface carefully smoothed. Context 468, fill of pit 473.
THE METAL OBJECTS by L.N. MEPHAM
Copper alloy coin A copper alloy coin recovered from the fill of pit 480
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY
WA/LJC
Figure 13. Figheldean (Site A): fired clay, iron and worked bone objects
has been identified as a sestertius of Antoninus Pius, dated AD 161-80 (M. Rawlings, pers. comm.).
Copper alloy object One copper alloy object was recovered: a broken ring of circular section, from the fill of structure 342.
Tron objects
A total of 760 iron objects was recovered, all of which have been X-radiographed. Apart from six miscel- laneous items, including a knife blade and several sheet or plate fragments, the collection consists entirely of nails, both hobnails and coffin nails, deriving largely from the seven burials which are
assumed to be of later Romano-British date. Table 4 summarises the quantities of iron objects from grave contexts.
Five objects other than nails were recovered. One of these, consisting of the shank fragments from a large pin, possibly a shroud pin, was found in grave 397. A knife blade of short, wide, symmetrical type (Figure 13: 1) derives from the fill of kiln 50; it is of late Roman form (Manning 1985, figure 29, type 21). The other iron fragments are from objects of un- known form: two plate fragments, one with a nail or rivet hole, the other with a nail im situ; and one unidentifiable object of tapering form, possibly a ferrule.
36 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Table 4: ironwork from graves
Grave No. No. Hobnails No. Cleats No. Coffin Nails Other Objects
61 80(R); 113(L) 5(R); 6(L) 75 11(R);_ 18(L)
363 30(L) + 97 21
397 89(R); 57(L) 4(R); 4(L) 1 (pin)
400 30
456 6 34
477 74(R); 70(L)
TOTAL 675 55 1
Illustrated object (Figure 13)
14. Knife blade; short, wide and symmetrical, with tang on midline of blade. Dimensions: 95 x 28mm; obj. No. 1024, context 41, fill of kiln 50.
THE STONE by L.N. MEPHAM
A total of 92 pieces of stone was collected. These can be divided into portable objects such as querns; building material, including tiles; and unworked pieces. Stone from the structures of kiln 50 and kiln 457 was not collected on site, although fragments from the upper fills of these features may have derived from the original structures.
Six quern fragments were recovered, two of them conjoining. The querns fall into two categories: three small hand-querns, with a diameter of less than 0.5m, in both fine and coarse sandstone; and two large Greensand mill-stones, with a diameter of c.0.7m, of the type generally associated with mechanically- worked mills (see Cunliffe 1971, 153 and figure 71). The hand-querns derive from pit 367, and the fill of grave 456, which cut structure 373; the mill-stones both came from the upper fills of kiln 50.
Twenty-nine flat slabs, probably roof tiles, were recovered from a number of contexts. These occurred mostly in a ferruginous sandstone, but there are also three examples of limestone roofing slabs, one with an iron nail still i sitw. Other building material consisted of three blocks, two limestone and one greensand, probably architectural fragments, from the upper fill of corndrier 50, and from clearance layer 452.
The remainder of the stone consisted of fragments of greensand, limestone and sarsen, apparently unworked. The greensand and limestone fragments may also have derived from querns or building mater- ial; neither stone is local to the site.
THE FLINT by PHIL HARDING
A total of 1923 pieces of flint (30515g) was recovered (see Table 5). Residual material was found in 25 contexts of Romano-British date, and has not been examined in detail. The flint from pit 319 was undiag- nostic and of insufficient quantity to be informative. The largest quantity of flint from the pipeline was associated with the ring ditch (Figure 2; sections 64 and 77). Most of the stratified assemblage from contexts 45 and 51 has been analysed in detail and the results are described below.
Distribution
The flint assemblage was found in the base of the ditch hollow resting on the primary chalk silts. There was nothing to indicate that individual knapping clusters were present; in fact, it is more likely that flakes had been dumped and had moved downslope to their present position. The upper boundary merged with the overlying silts which contained Romano- British pottery. The flint from the two excavated sections has been amalgamated for the metrical analysis.
Raw material
Large nodules of fresh flint, which were particularly plentiful to the south of the barrow, outcrop at the surface of the Chalk. The flint is black to dark grey in colour with coarse grey cherty inclusions. Nodules of weathered flint, which may have been collected from the surface or from a barrow cairn, were also exploited for knapping. Raw material from the site has been tested to assess its workability, and is of sufficient quality for tool production. Thermal fractures are, however, plentiful and most nodules shattered in the early stages of flaking, isolating small pieces of solid
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 37 Table 5: flint by feature, Site A
Cores Broken
Unbroken — Broken Burnt Retouched
Scrapers Other
Cores Flakes Flakes Flakes Flakes Tools - Romano-British 1 2 78 64 1 5 5 = features ? Prehistoric pit - - 5 6 = = = Barrow unstratified 4 3 118 132 5 2 1 3 Barrow stratified 43 43 531 785 2 10 5 1 TOTAL 48 48 732 987 8 17 ll 4
flint which could be used as cores. Many of the excavated cores are characterised by thermal striking platforms and flaking surfaces with similar levels of patination. This suggests that the practice of using shattered nodules, which has also been observed at a _late Neolithic industrial site near Stonehenge (Hard- ing 1990) was used by the knappers at Figheldean. Eighty-nine broken unworked flint fragments, weigh- ing 3313g, were also recovered from the analysed contexts but were not retained.
Cores Of the 38 stratified cores, most are miscellaneous (39%) or single platform (37%) types. They are generally small and unproductive, and most are prob- ably failed pieces. Very little attention was paid to preparing the striking platform (only 25% of the 44 striking platforms) or the front of the core. Examin- ation of the causes for rejection indicates that at least 50% of cores show flaking angles above the level at which blanks could be removed successfully. This is often associated with the formation of hinge fractures, edge crushing and incipient cones of percussion, which result from mishits. In addition 50% of all cores are considered to show no signs of potential for further flaking. The quality of flaking and the limited number of removals from each core compared with the ratio of cores to flakes (1:32) suggest that success- ful cores may have been removed from the area.
At least 4 cores have small areas of battering which result from their use as hammers.
Flakes
The most complete flakes are squat or broad in plan, with a thick cross section and a broad plain butt (Figure 14). Flaking angles are generally low (av. 69°), a feature substantiated by the absence of faceting
to control the flaking angle. Platform abrasion, to strengthen the edge of the striking platform, is also virtually absent, although crude edge trimming was employed infrequently. Scar patterns on the dorsal surfaces of the flakes have confirmed that most were removed from unprepared cores, many of which included areas of shattered thermal surface. Flakes with these characteristics have consistently been shown to post date the early Neolithic period (Hard- ing 1990 and 1991).
A bulk sample was sieved through 2mm mesh, producing 24g of undiagnostic knapping chips, which is insufficient for an in situ knapping deposit even where platform abrasion is absent. This tends to confirm that the deposit represents dumped material. There are at least four pieces similar to those pro- duced by retouching a tool blank (Newcomer and Karlin 1987), suggesting that all phases of tool pro- duction are represented in the waste.
Retouched pieces
The ten retouched pieces from the stratified contexts are all undiagnostic. The retouch is mostly confined to individual notch removals or short lengths of denti- culate or irregular flaking. Retouch of this type would produce flakes similar to those noted above as by- products of tool manufacture. The five scrapers are all end scrapers made on the distal part of a thick blank with direct, continuous, irregular, semi-abrupt flak- ing. There is nothing to indicate that deliberate blanks were produced.
A barbed and tanged arrowhead was found in the south section of the ditch above, but not in immediate association with the knapping waste. It is of Green’s (1980) Sutton b type, defined as small miscellaneous arrowheads with unshaped barbs. There is nothing to indicate the form of the blank, both sides being flaked
38 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
by convergent covering retouch. The tip is unbroken although the tang is snapped.
Two flake knives were found in the overlying mixed horizons, 37 and 39, of the ditch. One has an abraded butt and was probably removed from a prepared core. Both implements have continuous lengths of low angle direct retouch along part of a convex flake edge.
Discussion
The composition and location of the flint assemblage from the barrow ditch suggests that it represents dumped industrial waste from a late Bronze Age industry which produced flake tools from blanks removed from unprepared cores. The barbed and tanged arrowhead, probably of Beaker date, and one of the flint knives from the upper fills of the barrow ditch, are residual.
The use of barrow ditches as dumps for knapping waste is a common phenomenon on later Bronze Age chalkland sites (Saville 1978). The construction of the monument provided fresh flint as well as weathered nodules from the surface.
Illustrated flint objects (Figure 14)
1. End scraper on flake. Context 45; obj. no. 1086. preservation. An exceptionally large but unmeasura- 2. End scraper on flake. Context 45; obj. no. 1087. ble red deer femur was recovered from context 462, 3. End scraper on flake. Context 51; obj. no. 1089. the fill of ditch 459, which is dated to the Ist or 2nd 4. End scraper on flake. Context 51; obj. no. 1090. century AD. 5. End scraper on flake. Context 51; obj. no. 1095. The assemblage of bones from the late Roman 6. Flake with inverse retouch. Context 51. contexts is influenced by the skeleton of a dog, 7. Flake with denticulate retouch. Context 51. recovered from grave 61, which forms 15.6% of the 8. Notched flake. Context 51. total. The skeleton is well preserved, and selected 9. Retouched flake. Context 51. parts of the hindquarters and front paws are present. Table 6: animal species by phase, Site A Phase Cow Sheep Goat Pig Horse Dog Red Deer Mallard — Indet. Totals Neo/BA 14 3 = 5 3 1 1 1 94 122 IA 1 2 e a w = = = 3 4 LIA/RB 1S 11 = ] 4 3 - = 56 90 ERB 100 149 1 19 14 6 2 = 421 712 LRB 93 20 - 20 15 115 8 = 467 738 TOTALS 223 183 1 45 36 125 11 1 1041 1666 % % % % % % % % % % ERB 14.1 21.0 0.1 237. 2.0 0.8 0.1 - 59.2 100 LRB 12.6 Ps 0.0 2.7 2.0 15.8 1.1 - 63.3 100.2
Key. Neo/BA: Late Neolithic or Bronze Age; IA: Iron Age; LIA/RB: Late Iron Age or Romano-British; ERB: Early
THE ANIMAL BONE by J. EGERTON, C. FITZGERALD and C. GAMBLE
The assemblage contains a total of 1743 bone pieces (13354g), of which 644 (37%) are identifiable. The description of the phases and the distribution of the material are presented in Table 6. Further details on the distribution of material by context in each phase, and a detailed account of the anatomical represen- tation can be found in the Archive.
The prehistoric and late Iron Age/early Romano- British phases produced very small quantities of bone and a high proportion of unidentifiable material (Table 6). Half of the unidentifiable fragments came from one context, 370, the fill of a 1st or 2nd century AD pit, 367. The condition of these fragments is variable, from well preserved surfaces to highly eroded.
A small collection of frog bones (Rana sp.) came from an early Romano-British pit (367), and from an Iron Age pit (353), but were considered to be intru- sive. The proportions of cow and sheep/goat are 14.1% and 21% respectively. The high number of sheep teeth may indicate a degree of differential
Romano-British; LRB: Late Romano-British
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY
Figure 14. Figheldean (Site A): worked flint
40 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
The lack of anatomical parts makes estimate of stature difficult; this would, however, be a large animal in comparison with other Roman dogs in the Wessex area (Maltby nd). The anatomical parts are recorded in the Archive. Notable in this phase is a considerable change in the proportion of cow to sheep bones compared with the early Romano-British phase. Cow comprises 12.6% of the assemblage and sheep/goat 2.7%. Despite a small sample size and problems with preservation, it seems reasonable to suggest that ovicaprids became less important in the late Roman period at this site.
Condition of the assemblage
The assemblage is highly fragmented and only one long bone, a sheep metatarsal from the fill of the drying kiln, 50, is measurable. Several bones are gnawed, the majority by dogs with a small amount by rodents. Weathering varies in degree, but the un- identifiable material shows very heavy amounts, which makes the identification of gnawing and butchery difficult. There are few burnt bones, and butchered bones are not common. Standard chop and cut marks are entirely consistent with the removal of meat from dismembered carcasses.
Ageing
The pattern of fusion is consistent with differential preservation and loss of unfused elements. The sample of bones and teeth which can be assigned to a wear stage is too small for economic reconstruction. The rarity of young animals in this collection is most probably due to preservation factors.
Discussion
The collection of faunal remains is too small to allow detailed economic statements to be made. The nature of the excavation was such that it is difficult to assess the representative value of the collection. Rural Roman sites, however, are still comparatively scarce, and the low density scatter across the features may well represent activity associated with the animal economy. The changing importance of sheep noted here during the Roman period has been noted else- where (Maltby 1981).
Pipeline excavation of this nature continues to provide an opportunity for sampling landscapes rather than sites. It may well prove, in the long run, to be an important means of assessing wider economic patterns through the nature of the samples it produces.
THE WORKED BONE by L.N. MEPHAM
Five pieces of worked bone were recovered, compris- ing two fragments of antler waste (Figure 13: 4), one needle (Figure 13: 6), one spindlewhorl (Figure 13: 5) and one animal tooth which appears to be cut horizon- tally.
None of these objects is closely datable on typo- logical grounds, although the needle is likely to be of Romano-British date, the type being most common during this period (MacGregor 1985, 193). Similar needles with spatulate heads from Colchester have been assigned to the later Roman period, although copper alloy examples are known earlier (Crummy 1983, 65, type 2). The spindlewhorl, made from the fused head of a cattle femur, represents one of the most common types of whorl. Femur head whorls are found sporadically from the Iron Age to the Norman period (Crummy 1983, 187).
Both the needle and the spindlewhorl were recov- ered from an upper fill of the barrow ditch 77 (context 51); the cut tooth came from the barrow ditch 64, again from an upper fill (context 45). All three objects were associated with Romano-British pottery. The two pieces of antler waste derived from pits 351 and 367; these features are dated by associated pottery evidence to the Ist/2nd century AD, and the 3rd/4th century AD, respectively.
Illustrated bone objects (Figure 13)
4. Length of antler, cut on two sides. Length: 144mm. Object no. 1124, context 350, fill of pit Eble
5. Spindle whorl, made from the fused proximal end of a cattle femur. This is one of the most common types of spindle whorl, appearing first in the Iron Age and continuing in use sporadically until the Norman period (MacGregor 1985, 187 and fig. 101, no. 6). Diameter: 44mm. Obj. no. 1092, context 51, fill of barrow ditch 77.
6. Head of needle; shaft circular-sectioned, with spatulate head; circular eye countersunk. Bone needles such as this example were particularly common in the Roman period (MacGregor 1985, 193 and fig. 101, no. 24); examples from Colches- ter are dated to the late Roman period (Crummy 1983, 65, type 2). Length: 35mm. Obj. no. 1025, context 51, fill of barrow ditch 77.
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 41
THE HUMAN BONE by A.V.C. JENKINS
Introduction and method
A total of nine persons is represented. The skeletons are reasonably complete except for 402 and 405 (the prehistoric burials), of which only parts of the lower bodies were recovered, and 478(b) which is repre- sented by no more than the distal portion of a humerus. Their condition of preservation varies from fair to very poor, which restricts the amount of information which can be obtained from them. Only one skull is sufficiently intact for a cranial index to be calculated.
Sex was evaluated by examination of the pelvic and cranial traits recommended by the Workshop of European Anthropologists. These were scored in the manner indicated by the same body and by Acsadi and Nemeskeri (1970). From these figures a Sexual Index was obtained in which a figure between 0.00 and 2.00 represents a masculine skeleton, while a figure between 0.00 and —2.00 represents a feminine one.
Age was assessed from the state of epiphyseal fusion, dental attrition and, in males only, the con- formation of the pubic symphysis. The material was compared with published stages of development (Brothwell 1981) and scored accordingly.
Stature was estimated by means of the regression equations of Trotter and Gleser (1958). Although the calculated heights are given to the nearest centimetre, they are accurate only within a range of three or four centimetres either side of the figure given.
Only those non-metric traits described by Finnegan (1978) were recorded when present. Pathological and abnormal conditions were considered under the head- ings of degenerative disease, infection, trauma and developmental or other conditions as advocated by Dr J Rogers (et al. 1987 and pers. comm.).
Results (Table 7)
Eight of the nine were adults, and one, 478(b) was an infant. This child is represented by a fragment of a single bone which was found in association with an adult, 478(a), and for this reason it must be con- sidered to be an inadvertent redeposition rather than a burial. Three were men and five were women. So far as can be determined, since established ageing tech- niques are of disputed accuracy, two men and two women lived out a normal lifespan, and three died comparatively young. The remaining adult is too poorly preserved for an age assessment, though the smooth articulating surfaces of the joints imply youth.
Table 7: pathological data for human remains
Number Sex Age Stature 63 M Elderly 1.76 74 F Elderly 1.53 368 F Elderly 1.63 402 F 20-25yr 1.56 404 F 17—25yr 1.56 405 F Adult Short 466 M Elderly 1.74
478(A) M 25—35yr 1.65
478(B) ? Perinatal? =
Only the prehistoric skeleton 405 among the adults, is too fragmentary for stature to be estimated, though her fine bones indicate that she was a short person. The range of the Romano-British skeletons is from 1.53m (5ft) for the shortest woman to 1.76m (5ft 9in) for the tallest man. This is comparable with the figures quoted for other, larger, Roman cemetery populations e.g. 1.48—1.82m at Cirencester (McWhirr et al. 1982).
As is usually the case in ancient populations, dental health was poor by modern standards. Every den- tition bears a deposit of calculus and has suffered a greater or lesser degree of alveolar recession. Dental caries and/or ante mortem tooth loss were also univer- sal. This means that dental hygiene was rudimentary or non-existent and that, as a result of the fermentable carbohydrates in the diet, dental problems were wide- spread, as was inflammation of the gums (gingivitis), the cause of much of the alveolar recession and ante mortem tooth loss.
Osteoarthritis was noted in three individuals (63, 368, and 466) with differing degrees of severity. The neck only was affected in 63; the neck and shoulders in 368. Skeleton 466 was afflicted throughout the body, including all of the vertebrae, the shoulders, hips, knees and ankles. On the femoral heads about 5mm of new bone had been laid down, with conco- mitant enlargement, eburnation and pitting of the acetabulum. No infective conditions were observed.
Skeleton 466 also fractured the first phalange of the little finger of his right hand. This had not been splinted and it had healed at a crooked angle.
Spondylolysis of the fifth lumbar vertebra was noted in skeleton 63. This is a condition which may result from a congenital weakness in which the neural arch becomes separated from the vertebral body as a result of continual stresses in the lower back. Spondy- lolysis is also a developmental condition, but a second
42 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
notable feature observed in the spine of skeleton 63 was the fusion of the second (axis) vertebra with the third. In addition, a gallstone was found with this skeleton. Skeleton 368 had a metopic suture and an osteoma on the left parietal bone.
CARBONISED SEED REMAINS by JOY EDE
Five samples were processed using standard flotation and sieving procedures. Residues greater than 5.6mm were sorted by eye for environmental material and
artefacts, and then discarded. All other fractions of residue were sorted and carbonised remains extracted. All flots were sorted by the author and all the carbonised remains identified using an incident light microscope up to X50 magnification. Nomenclature used follows Clapham, Tutin and Warburg (1952).
A summary of the carbonised seeds by feature is contained within Table 8. Two samples came from a late Iron Age or early Romano-British pit 353 and three from the late Roman drying kilns. The quantity of soil sampled varied between one and ten litres, although one litre is generally considered too small for carbonised remains.
Table 8: carbonised seed remains, Site A
Feature Type Pit Pit Corndrier Feature No. 353 353 50 457 457 Context 364 352 53 483 482 Quantity (1) 1 1 10 1 5 Botanical Name Common Name Habitat 2008 2009 2010 2020 2021 Triticum sp. wheat CF - ~ 1 29 3 cf. Triticum 5 1 - - - Hordeum sp. barley CF - ~ 1 20 91 Hordeum sp. twisted grain - - - - 4 Hordeum sp. straight grain - - - - 12 Hordeum sp. indet. grain - - - - 75 Hordeum sp. hulled CF - - - - 28 cf. Hordeum sp. CF >) - - 1 13 cf. Avena sp. CE - - - - 2 TOTAL ID. CEREAL GRAIN 5 1 7) 50 109 cereal indet. grain 28 14 3 341 219 frags. + + ar + + TOTAL CEREAL GRAIN 35 15 5 39] 328 CHAFF Triticum sp. glume base 3 6 - - - Triticum sp. spikelet fork 6 - 2 - Triticum sp. glume base 81 25 7 12 5 Hordeum sp. spikelet fork 20 2 - 8 - cf. Hordeum sp. spikelet fork - - - 2 - Avena sp. misc chaff frags. 2 - ~ - - Avena sp. awn frag. stick: ate - - - Indet. rachis frag. 33 6 - 7 ~ Indet. culm node - - - 1 - Indet. chaff frags. 4 - 7 Z - TOTAL CHAFF 114 39 9 25 5
(not awn/rachis/chaff frags/embryo)
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY
Botanical Name Common Name
WEEDS
Habitat 2008
2009
43
2010
2020
2021
BORAGINACEAE Lithospermum arvense CARYOPHYLLACEAE Silene sp. campions
indet.
CHENOPODIACEAE
Atnplex sp. orache Chenopodium sp. goosefoot
indet.
COMPOSITAE
Centurea cf. nigra lesser knapweed Tripleurospermum maritimum scentless mayweed indet.
CYPERACEAE
indet.
FUMARIACEAE
Fumana sp.
GRAMINEAE
indet. small grasses
indet. medium
LEGUMINOSAE
Medicago sp. medicks
cf. Medicago sp.
indet. legumes PAPVERACEAE
Papaver sp. poppies PLANTAGINACEAE
Plantago lanceolata/media _ plantain POLYGONACEAE
Polygonum aviculare agg. knotgrass Polygonum sp.
Rumex acetosella agg. sheep’s sorrel Rumex sp. dock/sorrel RANUNCULACEAE
Ranunculus sp. buttercups RUBIACEAE
Galium sp. cleavers SCROPHULARIACEAE Euphrasia/Odontites sp. eyebright bartsia UMBELLIFERAE indet. VALERIANACEAE Valerianella dentata Papaver sp. seedhead capsule frag. weed seed unident.
TOTAL WEED SEEDS
corn gromwell (silicified)
narrow-fruited corn salad
GDa 15
DaGMSW 9
DaG 26
Da 6
159
53
Wr NRK Ww
6
50
65
Key to habitat preferences: C = cultivated plant; D = disturbed ground; G = grassland; M = marsh; P = ponds, ditches, banks; S = scrub; W = woodland; F = food plant; a = disturbed land including arable; b = base rich; h = heavy soils;
n = nitrogen/phosphate rich soils; w = wet/damp soils
Notes on calculations: Fragments of grain have not been made up to equivalent number of grains represented as this would not
alter results significantly. Chaff total does not include rachis fragments, embryos, bits or awn fragments.
44
Pit 353
Two one litre samples were taken from this pit. They were rich in carbonised remains, and may represent two distinct events.
Layer 364 (sample 2008)
From the lowest fill, this sample contained a total of 35 cereal grains, only seven of which were identified: five probably wheat and two probably barley; this level of identification indicating that the grain preser- vation was poor. All the carbonised remains from this sample were fragile.
There were 114 items of chaff. Of the identifiable chaff, Triticum spelta (spelt wheat) was present, and Hordeum sp. (barley) spikelet forks tended to have quite long and slender rachi attached indicating a less than dense-headed form; twisted awn fragments, and two other fragments of chaff, were the only indi- cations of Avena sp. (oats, either wild or cultivated); the several rachis fragments came from either wheat or barley heads.
At least 18 species of weed seeds were present, all commonly found in arable fields, including seeds from plants whose seed head could remain intact in processing, e.g. Papaver sp. (poppies); also, small seeds of unheaded plants, present as free seeds in the unprocessed grain, e.g. Euphrasia/Odontites verna, and large seeds such as Lithospermum arvense (corn gromwell).
Layer 352 (sample 2009) The sample contained 14 unidentified cereal grains, one possible wheat grain, 39 items of chaff and 53 weed seeds; the density of carbonised remains was much less than in 364. The proportions of different elements of these assemblages (cereal: chaff: weed seeds) were similar (see Table 9).
Chaff indicated the presence of spelt wheat, barley
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
and oats (twisted awn fragments only), though there was proportionately less barley than in sample 2008.
The range of weed species was more restricted, with a minimum of 12 different species present. The assemblage was consistent with weeds growing in arable fields.
It is possible that for each sample the entire assem- blage of cereal remains and weed seeds comes from one episode of burning, though there may have been more than one source of carbonised material, the burnt rubbish from the various sources being thrown into the pit at much the same time.
The presence of carbonised material from this pit suggests that cereal processing occurred within the site. The majority, and possibly all, of the carbonised remains represent the waste material resulting from winnowing and/or fine sieving of crops. There may also have been present a small element of semi-clean grain which became burnt accidentally.
Drying kiln 50
The sample (2010) from layer 53 came from the bottom layer of feature 50, interpreted as a drying kiln by its T-shaped plan and structure. Although ten litres of soil were sampled, few carbonised remains were found. One grain of wheat and one of barley were present. There were seven wheat glume bases and six weed seeds from waste or cultivated land.
Drying kiln 457
Two samples came from this feature, also interpreted on site as a drying kiln. Both samples came from the base of the feature. Sample 2020 from layer 483 was very rich in carbonised remains (466 items per litre), and sample 2021 from layer 482 was quite rich (80 items per litre).
Table 9: seed ratios, Site A
Sample
2008 1 2009 1 2020 1 (2020) 15 2021 ]
(2021) 65.6
Chaff Weeds Density items per litre 3.2 4.5 308 2.6 3D 107 0.06 0.13 466 1 2 0.02 0.2 80 1 13
Figures given to the nearest decimal point, except the very small numbers where numbers are to the nearest two decimal points.
Density of carbonised items for sample 2003 was 0.4 per litre and for sample 2010, 1.8 items per litre.
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 45
Context 483 (sample 2020) The sample was not well preserved; half of the identified grains were wheat and half barley, chaff confirming the presence of both, possibly grown as a mixed crop (maslin), used either for human or animal consumption (Jones 1984).
Fewer chaff and weed seeds were present than cereal grain, implying that the majority of carbonised remains represented clean grain which became acci- dentally burnt, probably im situ; the lack of any evidence of germination suggests that barley was not used for malting. The grain may have been stored as free grain in a winnowed state before fine sieving, or enclosed by chaff (possibly still in spikelets in the case of wheat), winnowed but not fine sieved; the relative lack of chaff indicates that most of the chaff was removed previously.
Context 482 (sample 2021)
The sample contained better preserved grain, mostly unidentified, but including hulled six-row barley. The weed seed assemblage was small, and there was an under-representation of chaff.
Summary
The samples from this site indicate that the cereals utilised included spelt wheat and six-row hulled barley.
On the whole the grain appears to have been stored as free grain which had been threshed and winnowed at least once and relatively thoroughly, and possibly dried to prevent spoilage in store. The majority of the remains from pit 353 (dated to the Ist century AD) represent cleanings from grain processing whereas the majority of remains from kiln 457 (late Roman) represent clean grain. Whether the grain was cleaned in bulk prior to storage, or whether it was stored prior to fine sieving is unclear. It is possible that the site was a producer site, the crops present in these samples being produced by the inhabitants of the settlement, though it is not possible to be definite about this since grain may have been stored and therefore traded in a semi-clean state, that is before fine sieving had taken place.
THE LAND-USE HISTORY: THE MOLLUSCAN EVIDENCE by MICHAEL J. ALLEN and SARAH F. WYLES
A series of 20 samples from three locations was processed for Mollusca using the standard procedures outlined by Evans (1972) and detailed elsewhere (Allen 1989a and b; Allen ez al. 1990). The results are represented in standard histogram form (Figure 16)
and in Table 10. Mollusc nomenclature follows Waldén (1976). The samples analysed came from a single undated prehistoric, but possibly Neolithic, pit (see section on pre-enclosure features on page 12), a colluvial sequence (Figure 15) and the ring-ditch.
Prehistoric pit 319
Pits fills are particularly problematic for the interpre- tation of molluscan evidence because they may con- tain species which have eroded from old soil through which the pit was cut, fallen from the contemporary land surface, or associated with the material thrown into the pit or even species which lived in the pit (cf. Thomas 1977; Shackley 1976). Nevertheless, success- ful analyses of pit fills have been conducted (Allen 1989a; Evans 1984; Thomas 1977), and the poten- tially early date of pit 319 would provide a clue to earlier land-use history not available from other con- texts.
The coarse fills of the cone-shaped pit were sampled. Although the basal sample (321) produced few shells, the spot sample from the upper (possibly re-cut) fill (318) was relatively rich in Mollusca. Both assemblages almost exclusively comprised shade- loving species (as defined by Evans 1972) and no open country species were present (Figure 16). The preda- tory Zonitids, together with Carychium tridentatum, Discus rotundatus, the rupestral species, and the lack of open country species, are strongly suggestive of shady, probably woodland, habitats. It is, however, likely that some of the species, Oxychilus cellarius in particular, were exploiting the pit micro-habitats. The upper sample is dominated by C. tridentatum which, although a species that enjoys leaf-litter habitats, here probably represents tall mesic grassland as it is accompanied by catholic species.
These assemblages are markedly different from all others analysed from this area. The shady component, probably initially of a woodland environment, is the only such assemblage recorded and it is likely to be significantly earlier than that represented by all other samples; thus a Neolithic date is not improbable.
Colluvium
Immediately down-slope from pit 319 was an exten- sive colluvial deposit (Figure 15). Basic pedological descriptions taken on site indicate that the deposits probably represent a long chronological sequence. This sequence is unusual within Wessex and southern England as a whole in that an, albeit truncated, buried soil was preserved (Allen 1988; Bell 1983).
(Y aS) UBappeysty ie Aay[ea AIp ay] Jo uONdas pue ue[g “C{ en3ty
w Z N < Ort WM < w ODT O€T OcI OT oor 06 08 OL 09 Ov 0€ 02 Or 0 = == 1 it 1 —_1— 4 4 SS ——— | 1 kh = —— 0 S squy yuUM Ard aK v MoO]JOY UOTIN[OS © = +1 i 4 r& x a +€ a D6p yoYyoUAT Ee LIg Joyous tb <x c Z " 6T€E Wd 5 a " va Zz ) | dO Eeee6 L9 . N [los paring T1¢ — ; S [los paiing Z1g — > g wnianlod O1g — Se 5 wunLAnTIOD 60€ —~ ha a 5 — S pre AATIWA AUG AHL JO AMIOUd S) z : a 09Lb OSL j as a B SD) ig a a | 4 os 0 S en ial a SvEV Aqg-Aey a % ‘ “GS aS sjuiod pii6 a1ig¢ d-Vv \ \ Aa \ I ie 4, ajdwies uwnyjoo yo uole907 OTE Hd OTe Hd e \ & =i +41 \ \tY aoe ie ‘ ESE F10UIsOd | x 6D yoyouAT LI¢ JaYyIUAT
46
GNd HLYON ‘Vv ALIS
COLLUVIUM
PIT 319
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 47 & Pt $ ¢ ~ 2 Q 2 o o @& i! ® ® oe oS aN ® ¥ § 2 Ss = > 7 0 COS . ye OO 3 \ os 7 ows 2 S ° oo Go ® R € we V.o o SN Pd o sy o rf c) Py S Cae Q ° + 2 Ow 6 ha 2 : < © & ae Ae a © @ é& aes ¥ ° S < <) ee RY 2 oO & Re 2 ® o SS oO © = RY og és g é s < e s ee § 2 ° oS ow (ee) AS) NS \ ty = AS 2 Saas) Q vs ee x Ri & Le GENERAL 2 fe L L L c eee L Lt ci & ats (c ENVIRONMENT 412 + ogee | os a i i in Arable and pasture 438 ar | + 571 ar cu ar 516 + a. + To pata ih | é Arable 277, —C+ a af a ats ar Tillage and 154 7 ++ ie truncation 44 | Grassland e e e 207 fe | ed E | | | a | | a Long grass maim Woodland ile teal [ail pes Gaal foal = bala: ye ier hectn « Seateat SAL Talent fircteot {pet nme %® 0 300 40 0100 30 0 000100 20 0 40 0 20 0 20 00100
e — Species present
Figure 16. Figheldean (Site A): histogram showing relative abundance of Mollusca in samples from Pit 319 and colluvium
Sediment descriptions
0-30cm:
30-70cm:
70-110cm:
110-120cm:
120-150cm:
150 + cm:
topsoil. Humic moist loam, almost stone-free. Typical brown Rendzina. light buff, silty clay loam with common very small chalk fragments and common medium flints. Weak crumb structure. Highly calcareous.
dark silty clay loam, rare flint, common very small chalk pieces, gets darker and more humic with depth.
bA/B. Buried A/B horizon of the old land surface, humic stone-free hori- zon, medium weak blocky peds, weakly calcareous. Possibly truncated. bBt. Buried b horizon of old land surface. Clay loam with flints, non- or weakly calcareous. Some minor intra pedal coatings noted with hand lens; possibly lessivage/clay __ illuvation. Slightly rubified as a result of the iron rich tertiary clay.
periglacial solifluction deposits.
The basal bBt horizon was only weakly calcareous and almost devoid of Mollusca. Apart from this horizon, three main molluscan zones can be detected (Figure 16) which broadly follow the basic stratigraphy. The truncated A/B horizon of the buried soil, although only weakly calcareous, produced both charcoal and relatively high numbers of shells, especially from its upper surface. The mollusc assemblage is dominated by open country species, in particular Vallonia costata and, in the lower portion, Pupilla muscorum. The presence of Pomatias elegans, although probably over represented (Carter 1990), indicates disturbance which might be interpreted as the clearance of shrubs and a scrubby environment, especially as a number of charcoal fragments and flecks were recorded in the buried soil. The scrubby environment seems to have been replaced by a short-tufted, grazed, grassland. The lower, more humic, colluvium was rich in shells and represents erosion from the adjacent slope of the thicker, weakly calcareous soils of the type buried in the centre of the valley. P. elegans and P. muscorum almost disappear, while V. costata retains its
48 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
prominence, augmented by V. excentrica, Trichia hispida and Helicella itala. This is typical of many ploughwash deposits indicating arable conditions locally. The truncation of the buried soil profile can probably be attributed to this activity. In contrast, the lighter, highly calcareous upper colluvium indicates erosion of thin calcareous light brown-grey rendzinas, and the mollusc assemblages are characterised by a continued dominance of V. costata and a reduction in its congener V. excentrica. The increase in the number of Zonitids and the occurrence of a number of shade-loving species, including Carychium triden- tatum, Punctum pygmaeum and Nesovitrea hammonis, suggests a more diverse landscape ecology and may be taken to indicate a slightly more mesic environment concomitant with stable grasslands. This may repre- sent low intensity grazed downland and thus a signi- ficant reduction in the intensity of land use. The assemblages, therefore, probably represent both arable components and tracts of longer grassland which the more catholic species can inhabit, the latter possibly being in the centre of the valley.
Ring ditch (feature 64)
A series of nine samples through the ring ditch was analysed (Table 10 and Archive Figure 2). The basal deposits typically produced few shells. A limited number of shade-loving species in the lower fills indicate some localised shady environs, probably representing a combination of the shade afforded by the ditch micro-environs and chalk rubble, and a surrounding long, only lightly grazed grassland. The main ditch fills produced open country assemblages which are highly xerophilous in character. P. mus- corum is predominant, accompanied by V. costata and H. itala. Interestingly, two xerophilous species were confined to the secondary fill. The only indigenous obligatory xerophile Truncatellina cylindrica (Evans 1972) and the thermophile Abida secale occurred especially in the flint tumble in the centre of the ditch. T. cylindrica is a rare and local xerophile species (Ellis 1951) but it has been recorded in a number of Bronze Age sites in the immediate locality: Durrington Walls (Evans 1971), Woodhenge (Evans and Jones 1979), and also within the Stonehenge area (Evans 1972) although curiously it was never recorded in any of the analyses for the Stonehenge Environs Project (Allen et al. 1990). T. Cylindrica is found in very open dry habitats and on rocks (Ellis 1969) and A. secale also inhabits dry exposed places and avoids low-lying grassland. Both species may therefore to a limited extent reflect the ditch tumble niche as well as the very open dry surrounding arable landscape.
The tertiary fills contained similar open country assemblages but are more akin to the typical plough- wash assemblages recorded in the valley colluvium. The assemblages therefore indicate continuous open, predominantly arable, land.
Discussion
The almost total absence of open country species in the shade-loving assemblages from pit 319 may indi- cate evidence of a woodland and it is possible that no major clearance had occurred to allow colonisation by these species. Even if the assemblage is predomin- antly pit-dwelling, one would expect to find some open country species if they were living on the surface. This, therefore, represents an early (probably Neolithic) woodland environment which was cleared, probably in association with Neolithic activities including the digging of pit 319. This clearance would have led subsequently to an open, lightly grazed, long grassland environment.
The acid brown earth buried in the valley bottom produced evidence for disturbance (probably shrub clearance) followed by grassland. From Evans’ work at both Durrington (1971) and Woodhenge (Evans and Jones 1979) we know that a short-turfed grassland environment existed by at least 3965+90 BP (BM-399) at Durrington and 3755+54 BP (BM-678) at Woodhenge. It seems likely that the buried soil at Site A is of a similar date. Tillage and cultivation of the slopes ensued, resulting in erosion and the collu- vial build-up in the valley, an event probably broadly contemporary with the ring-ditch on the top of the adjacent slope. The ring-ditch provided evidence for a period of long grassland which may have been very localised and confined to the barrow. This was, however, followed by an intensive arable phase to which most of the ditch fill can be attributed. Open, largely arable or short-turfed downland has been recorded in association with many barrows (cf. Dur- rington G3: Allen et al. forthcoming; Butterfield Down: Wyles and Allen in prep; Amesbury G72: Kerney 1967).
We can therefore suggest that initial large-scale colluviation in the Avon valley occurred in the Bronze Age period and it is likely that this was also the case at Durrington Walls. Unfortunately, the relationship between the hillwash and Iron Age ditch at the southern circle at Durrington was not recorded and thus we cannot be sure of the date of the earlier colluviation there.
A reduction in arable activity is recorded at Fighel- dean in the upper hillwash with the introduction of a mixed arable and grass land use (Allen forthcoming).
49
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY
LOI 7EZ SOF 9Sh £62 807 £02 Ch I cv 6c£ 8tr ILS OTS LLZ +S +1 v L40z_ «(tl TWLOL Ok “Ol cl 20 81 9 AI Sh 6 6 6 ie 8 ra at or ss € Si ee exeL + + + + = + FT ~ + + + I I € + = 5 I Th ‘dds viupuy/vavdn) - - - - Zz - - - - - - - - - - - - - - - ‘dds vanday 1S 6G) IG ch = 9 b VAG ake I 16 IZ cll col £8 0s 17 = = € = (snaeuurT) opidsry D1yIu4 i SEE SS G8e 62) ZT OT Z LI oI IZ el Or ol 17 I ie im = (snoeuUry) 2jD11 01129119 - @ = ~ - - - - - - - - - - - - - - - - (B1SO7) BC) DIDSuIe DyjanuiaD = <I g = = = = = - - = - = ~ - - - = = = (JOIIOg) DIJassaqUuL DjNpIpuDD - - - - - - - - - - - - - - - - + aepnyisney) = =a ae eT =F if ae wall a = 7 = = = I € 5 (WIQNS) DIDIUapIg DYw:ND]D (n3eUOp) = = = = = = - - - - - - - - - I = - I = DIDUIMD] ~DULPO]YIO0D 18 ELE -49¢ Ol 2% L a = eS val c ie v c I I I ag ey (19]TNW) DynI190 saproyj19a_) = = = = = = = - = = = = = 3 = I = = € = (J2TTNW) snajnf snmnuomg_ 6 9 Ant 4 Zl 6 Mie v tp (4S 8e €€ 0€ ol G a Tc € = sepreun’y I = = pl! I = Sit. val = OF te Or a a = = = a + I (aT) snuvjjar snpiy96xO (pneusedeiq) =e te tefias | = al = wei! = = ce = = I = 1 PI vinpuiu vjjaurdosay zs i he aC I I Z = Se = = = > = = = - I O€ - (Japy) vand vjjaurdosay = 7 = 1 c = =i oC = a Z 0 = aa I = = = = =; (WIQNg) stuomuDy Dasm12089.N - - - - 7 - | - | - = = = = = - - - tl € (PuN]1eIsa A) 279D41U0I Dad A ma? go oe. PY Bee Fee | aoe ie aie = - 5 = z = Sas (atm) opioid oust A i, = = = il € LI om eure a = 7 Fs za a = = = 6r 72 (A937) SMppunios snasiq (pneuredeiq) v ( I v = S a = ra I L I c a a a a 6 G unavushd wniqun gy as = = = = = = = = - = = - - I = (19][NP) Vénosqo Dug - = - - - - - - - - - - - - - - - - + € (J9T[NW) VIvaynov vjnuryluvdy Mc Claapee 5ce OIG SCliae: S I Lz (a6 Sa 8el Lvl eZ 91 € > = = (THIAIg) DIMIUaIXA DIUOTID A Ser LS COM CT OSE ES: 8S) ols eit L481 8el 961 617 961 18 9 S c = = (19M) 7101809 vIU0]]D A 96 $8 «LIC 81 $9 -tL AI OT + = Z = b 9 Te 6l v a = = (snaeuury]) wnsossnu oypidngy =e ae 4 Il 9G 2a 52 = = = = = = = = = = = - - (pneusedeiq) 3020s ppiqy = 2 = = = = as = = = = # = = I - ‘dds of140/ = it ol 6t 7 2 + I i a i = = 8 € Z = a oe — (pneusedeiq) vavwmstd oss, (oess.I3,4) = = = —- | = = = = = = = = - = = = - - - pouupuyar DuYyjaiIvIUNs | So it GIR ee SG AG = = = = = I = eS = ‘dds ndooyy20 Ae ek yee alt eee = Sl = = = = = = = = = (JTW) PIugN] Ddor7Y90 aA = mae nC S I E € I S = ¢ = I = = — I I Z (Ossty) wniviuapun wniystn) 14 = +o aT I L 0) ee I I = z = + z 8 I = 4 ate (J2TNW) sundaja souvwog VOSNTIOW 00OT OOOT ODOT OOOT ODOT OOOT OOOT OOFT OO9T OOOT O00 0001 0001 000! 000T 00SZ 0002 000Z 0001 O00T (3) IA 7 a FS = = ez = = — [09-0S 04-09 || 08-0 06-08 001-06 OTT-OOTJ OZI-OIT 01-021 OsTOel = (wid) indeq WINTANTJOO 8E 6t «66h UH CS OS8b OS8h LBP (O8P qaddn WMIANTTOS JOMO] V/a9 v/d4 149 Ble 7e 1xa1U0) LOIE SOLE SOTE POTE OOTE EOTE ZOTE IOTE OTE} 610Z 8102 ZI0c 9102 STOZ _P102 E107 7107 110z__|[S00¢_ 4002] ajdureg $9 Youp Bury wWIMITANT]Oo 61¢ 1d oimeaf
VY aus wor eosnyow :O1 F921
50 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
It is likely that grassland was confined to the valley bottom. This land use history compares well with that detailed from the Stonehenge area (Allen et al. 1990) and previous analyses at Durrington Walls and Woodhenge (ibid.) The lack of well-dated contexts, particularly for the colluvial sequence, precludes any more detailed chronological statements. It is particu- larly unfortunate that no dating evidence was gained from the colluvial sequence.
SITE A: DISCUSSION
The prehistoric landscape
With the exception of a pit (319) tentatively assigned to the Neolithic period on environmental evidence, the evidence for prehistoric activity, prior to the Iron Age, is limited. It includes the Bronze Age ring ditch, the fills of which contained pottery of middle and late Bronze Age date, and two crouched burials of uncertain date. The filling of the ring ditch was associated with the introduction of intensive arable farming in the barrow’s vicinity. The evidence from the lower, undated layers of colluvium, which overlay the buried soil in the dry valley at the northern end of Site A, suggests that the landscape had become more open with the onset of an arable regime. Intensi- fication of arable farming in the Bronze Age may also be suggested by the lynchets on either side of the dry valley, which may be Bronze Age in origin. The evidence of farming, the albeit small quantities of Bronze Age pottery from the lynchets and residually in later features, and the flint knapping debris within the ring ditch, when taken together indicate the nearby presence of Bronze Age settlement.
The enclosure
The enclosure ditches at Figheldean almost certainly originated in the late Iron Age though the two small sections excavated provided little in the way of dating evidence. No occupation material of definite Iron Age date was recovered, though the evidence did suggest that these ditches had been infilled in the early Roman period. The air photographic record, however, has provided details of the enclosure’s shape and size, and information on its internal characteristics (Figure 4). The photograph shows numerous cropmarks which correlate with the excavated features and indicate many more buildings, pits and internal divisions. In addition to rectilinear buildings, there are clear circu- lar marks which belong to roundhouses, in some cases with well-defined doorways, none of which was evi- dent within the trench.
It is possible that the settlement continued throughout the Roman period. The excavated mater- ial can be attributed to two phases: Ist to 2nd centuries AD and 3rd to 4th centuries AD, the earlier occupation probably having continued from the Iron Age with no evidence for a break in occupation between the two phases. It is not possible to shed much light on the earlier phase, due to the limited nature of the excavation. Detailed evidence for struc- ture types in the earlier phase was not recovered, but gullies to possible rectilinear buildings were identi- fied. The roundhouses marked in the aerial photo- graph may be Iron Age, or early Roman, but they do not appear to occur in that part of the enclosure crossed by the trench.
The use of terraces for timber structures, in some cases probably on top of dwarf stone walls, seems to have appeared in the later Roman period; there is, however, no indication of the types of structures which may have been built on them. The base of the excavated section of terrace 342 was cut by three post holes: one on the north edge (377) adjacent to a small gully or slot, and two on the east side (345 and 349), which probably formed the uprights to a timber structure. Terrace 445 was partially lined on its west side by limestone, flint and chalk rubble, probably to stabilise the side of the terrace where it had been cut into an earlier feature. It contained no evidence for post holes, and the building may have rested on sill beams. Similar building terraces were excavated at Poundbury (Sparey Green 1987), where some were dated to the post-Roman period. These also revealed little information on the superstructures, although post holes were found which may have been for roof posts, doorways and internal features. One also had an infill of limestone and flint rubble, possibly deriving from a collapsed drystone superstructure. At Fordington Bottom, near Dorchester, and at Maiden Castle Road, in Dorchester, similar terraces were noted. The bases of these features were cut by a number of post holes which clearly outlined buildings within the levelled terrace (R.J.C. Smith, pers. comm.). At Skeleton Green, Hertfordshire (Partridge 1981), evidence for timber, non-post built structures was also found, although not within level terraced areas. They were in the form of level platforms of flint and gravel, which showed faint traces of longitudinal timbers. It is quite likely that the terraces excavated at Figheldean also contained similar sill beams, which would have left no trace on the bare chalk.
The late Roman burials The burials appear to belong to the last phase of
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 51
activity, as they cut some of the later Roman features such as a building terrace (structure 373), and a drying kiln (kiln 457; Figure 10). They do not appear to have been part of an organised cemetery, as they are spread sparsely over 270m of trench. It is likely that occupation on this part of the hilltop had ceased, the settlement having shrunk or shifted, and the vacant areas had then been used for burial. This occurred at other Roman sites, for example IIchester, where the suburbs went out of use in the 4th century AD and burials were then placed amongst the ruins (Leach 1982).
Summary
Too little is known from the aerial photograph and the limited excavation at Figheldean to be precise about the enclosure’s internal organisation and morphology, but certain points do emerge. The aerial photograph clearly shows a number of cropmarks which appear to be rectangular building terraces, similar to those excavated, as well as two possible streets, one of which may have been metalled. Small internal enclos- ures are also marked, and all these elements have been noted at several earthwork sites in Wiltshire.
The population of such a site may have been large. The settlement at Figheldean, considering the number of possible terraces shown on the aerial photograph, would probably have been a sizeable nucleated settlement, though it is not possible to say how many of the terraces would have held buildings for human habitation, or whether they were all con- temporary.
The excavation has revealed a high density of features across the entire eastern side of the enclosure. With this evidence, and that of the aerial photograph, it is reasonable to assume that the remainder of the enclosure may also contain a similar proportion of archaeological remains. Comparison with other sites on Salisbury Plain indicates that it is quite possible that the Iron Age or Roman settlement may extend outside the confines of the enclosure, and indeed the aerial photograph does show parts of another possible enclosure to the south.
The Romano-British settlement at Figheldean appears to have been an ordinary, rural working settlement. The pottery is typical of such sites, and is comparable to many other non-villa rural settlements in southern England. The environmental evidence has shown that its economy was in part based on arable farming. It does not appear to have been a particularly wealthy settlement on the basis of the artefactual evidence recovered from the excavation; the pottery is mainly local, and there were few metal artefacts.
Nevertheless, surface finds have been made by the farmer which include imported pottery, an enamelled brooch, tweezers and an ear scoop (SMR record, SU14NW302). The occurrence of stone and tile roof- ing materials in the excavation also shows that the buildings were of some substance.
The Figheldean site is one of several Roman sites in the Avon valley area. To the north is the possible villa site beneath the Netheravon army camp (Figure 1 and SMR No SU14NW301). To the south are two sites, similar to that at Figheldean but unenclosed. The first is at Durrington Walls (Wainwright and Longworth 1971), on the south-west side of the henge monument. Only a small part of this site was excavated, revealing a number of pits, post holes, a drying kiln, and hollows, although no definite structures were identi- fied. To the east of Amesbury is Butterfield Down (Figure 17), which appears to have been an extensive settlement. Revealed during an evaluation in 1989 (Wessex Archaeology 1991), prior to a housing devel- opment, only a small part of the site has been excavated to date. Cropmarks and geophysical survey, however, have confirmed that it extends over an area of at least 6 ha.
Elsewhere on Salisbury Plain, there is a high incidence of nucleated Romano-British — rural settlements (Figure 17) although there have been relatively few excavations, and those have tended to be on a small scale, such as at Durrington Walls, Boscombe Down West (Bowen and Fowler 1966) and Winterbourne Down (ibid., 48). Bowen and Fowler, in reviewing Romano-British rural settlements in Wiltshire and Dorset, listed seventeen sites, many of which survive as earthworks on or near Salisbury Plain and are associated with enclosures and Iron Age settlement. Many of these sites, such as Winterbourne Down and Chisenbury Warren had_ rectangular building platforms, of a likely Romano-British date, similar in plan to the terraces found at Figheldean, and overlying Iron Age features.
An exceptional group of earthwork sites is to be found on Great Ridge, between the Nadder and the Wylye valleys. These are Ebsbury, Hamshill Ditches, Hanging Langford Camp and Stockton earthworks (Figure 17). They are characterised by multiple and single ditch systems and enclosures (Corney 1989), and surface finds from these sites include late Iron Age material. They all, however, contain earthworks of Roman date and include building platforms and terraces organised around streets, and would appear to have been substantial rural settlements. Although similar in many ways, the Great Ridge sites are much larger than the site at Figheldean (about 9 ha), the
52 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
R. 4 “on 10 & 3 3 0 : oe
4 Figheldean @
6). Je 13 7 SALISBURY
R. Avon
8 Ebsbury
1 Berwick Down 2 Rotherley 9 Durrington Walls 10 Chisenbury Warren
11 Butterfield Down
3 Mancombe Down
4 Knook Down
5 Stockton Earthworks 12 Boscombe Down West 13 Winterbourne Down
14 Little Woodbury
6 Hanging Langford Camp
7 Hamshill Ditches
0 50 _— i S11)
WA LJC Figure 17. Romano-British chalkland sites in Wiltshire
largest encompassing 79 ha, and the smallest being 16 ha.
In general, Romano-British rural settlement on Salisbury Plain at present appears to be characterised by nucleated sites, of varying sizes, both enclosed and unenclosed. In many cases, these settlements ori- ginated in the Iron Age, for example Figheldean and Durrington Walls. Some, such as Figheldean and Butterfield Down, appear to have _ continued throughout the Roman period. Many of the sites, including Figheldean and the Great Ridge sites, also show elements of planning, with buildings arranged in streets.
The settlement pattern on Salisbury Plain appears to contradict the view of a dispersed pattern of villas and farmsteads which has been postulated as the
normal form of settlement pattern (Esmonde Cleary 1989, 106). It is acknowledged by others that throughout the Roman period ‘England’ was densely populated ‘with settlements of many sizes, forms and functions situated everywhere’ (Taylor 1983, 106). However, there is still very little evidence for disper- sal on Salisbury Plain, although this may be due to the difficulty of recognising that type of settlement. In other chalkland areas, such as Hampshire, a more dispersed settlement pattern has been identified. A number of such rural Romano-British sites have been excavated: at Viables Farm (Millett and Russell 1984), Cowdery’s Down (Millet and James 1983) and Ructstalls Hill (Oliver and Applin 1979), all near Basingstoke, and Chalton, near Petersfield (Cunliffe 1977). As all of these sites have evidence for Iron Age occupation, as do the sites on Salisbury Plain, it seems likely that the differences in settlement pattern were established in the Iron Age.
Salisbury Plain appears remarkable for its apparent lack of villas. Reasons put forward for this have included the light soils, which were probably ‘near- exhausted’ in prehistory (Esmonde Cleary 1989, 106), and the siting of villas being seen as favouring the best agricultural areas (Taylor 1983, 100). In fact, villas are not totally unknown on Salisbury Plain, but the recorded sites are in river valleys rather than on the chalk as, for example, Pit Mead villa at Sutton Veny, near Warminster (Cunliffe 1973, 447), and the prob- able villa at Netheravon (Cunliffe 1973, 447 and SMR SU14NW301). This paucity of villas may be due to other, perhaps social, reasons. Soil impoverishment is an unlikely reason, as shown by the large population that appears to have lived in the numerous settlements, and by the continued evidence for arable farming.
The excavations at Site A, Figheldean, revealed a late Iron Age and Romano-British nucleated rural settlement of a type which can be seen as common to the Salisbury Plain area. This nucleated settlement pattern found on the Plain appears to contrast with that of other, neighbouring chalklands, such as in Hampshire, where there appear to be more dispersed farmsteads and hamlets.
Site B: Durrington (SU 15154410 to SU 15154227; Figures 18-19)
Introduction Site B was situated at the southern end of the Durrington to Netheravon pipeline, just inside the
| | |
| I
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 53
SITE B
Lynchet 123.
Lynchet 120m
Lynchet 12 1g Pit 125-9
\ | Posthole;~ 34
Features excavated in 1968
Figure 18. Durrington (Site B): the Packway Enclosure, and trench location
field boundary on the north-west side of the Durring- ton roundabout, and ran downslope to the north (Figure 18). The trench was located on the north side of a ridge of Upper Chalk, running approximately east-west, falling to the north from approximately 110m to approximately 95m above sea level.
The pipeline presented the opportunity of examin- ing an area of undisturbed ground within the Packway Enclosure. The trench was opened up by machine on 28 January 1991, beginning with the southern half of the enclosure. It crossed the line of the enclosure’s defining ditch c.45m to the north, allowing the exam- ination of an area of c.140 square metres within the
enclosure but producing evidence for very few features. To the north of the enclosure, two small pits of uncertain date were excavated, as well as sections across the fill of three negative lynchets. In the lower part of the slope, the machine was used to dig a section through colluvium in the base of the valley.
THE PACKWAY ENCLOSURE
Excavations along the pipeline confirmed the line of the northern side of the enclosure, and the excavated section of the ditch (35) was identical to that from the earlier excavations (Wainwright and Longworth 1971,
54 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
307). There were, however, few features within the enclosure apart from a scatter of possible stake holes with a dark grey infill. Only a single, flat-bottomed pit c.0.8m deep, and a small post hole were found. The surface of the chalk bedrock on the upper part of the slope, within the line of the enclosure, was char- acterised by areas of weathering and disturbance which appeared to have been caused by tree-roots.
The enclosure ditch (35)
The line of the enclosure ditch crossed the trench at a slight angle, and a 2m wide section was excavated across it. The ditch was found to be c.3.5m wide, with a shallower slope on the inner, southern side (Figure 19). The sides were considerably steeper lower down, plung- ing down to a flat base, c.0.2m wide, at a depth of almost 2m. The lowest fill, of coarse and fine chalk rubble (33) represented initial, and probably rapid, weathering of the upper parts, and contained fragments of struck flint. It was overlain on the south side by a band of soil (32), above which was chalk rubble and loam (30), perhaps from the erosion of an internal bank and containing animal bone and pottery dating from the middle to late Iron Age. The top of the ditch contained a linear hollow, c.0.6m deep, the fill (29) of which contained flint nodules. The top fill was a finer soil (3).
The steepness of the ditch prompted the sug- gestion, in 1968, that it had held a palisade of vertical umbers. No evidence of individual post positions could be found, however, even when sought in the earlier, larger-scale excavations (Wainwright and Longworth 1971, 308 and fig. 100, I and II). Ditch 35
Ditch 35
ao
was similar in profile to the ditch excavated by Wainwright, with its narrow, flat base and steep walls; it too contained no evidence for individual post posiuons, while the silting within it also conformed with Wainwright’s findings.
Features within the enclosure Very few features were found within the line of the enclosure ditch.
Pit 26 (Figure 18) was 20m south of the enclosure ditch. It was circular, c.l.lm in diameter, with slightly undercut sides and a flat base c.0.8m deep. The fills, 4, 24, 25, 27 and 31, contained chalk rubble. The finds were sparse, and consisted of a few frag- ments of flint and animal bone.
Fill 25, which lay between the upper and lower fills, probably derived from a period of weathering of the sides of the pit, suggesting a period when the pit remained open after its initial infilling. In form the pit was broadly comparable to the two pits excavated within the enclosure in 1968 (Wainwright and Long- worth 1971, fig. 100, VI and VII), which each survived to depths of 0.68m and 0.91m. Pit 25 survived to a depth of 0.8m and had been partly damaged by ploughing.
Post hole 34 was 10m to the south of pit 26. It was c.0.5m in diameter and 0.08m deep. There was no dating evidence.
Stake holes 16, 17, 18, 19, 20, 21, 22, 23 had diameters which varied from 30mm to 40mm and they contained a grey-brown loam. Their distribution was apparently random.
S 106.52m OD ras
Figure 19. Durrington (Site B): section through the Packway Enclosure ditch
C2.
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY
Features to the north of the enclosure Beyond the line of the enclosure, to the north, were a single small pit and three negative lynchets (Figure 18).
Pit 125, c.13m north of the enclosure, was an
-elongated oval with sloping sides c.0.25m deep, with
patches of pea grit (126) in the base. This was overlain at one end by vacuous chalk rubble (124), probably the result of weathering. The upper fills (103 and 112) contained several sherds of Romano-British coarse pottery, and a few medieval sherds from the top of the fills.
Lynchets 125, 120 and 123 were shallow ledges cut into the north side of the hill, north of the enclosure. Their surviving width was c.4-Sm, and depth c.0.4m maximum. The fills contained post-medieval pottery.
THE POTTERY by L.N. MEPHAM
A total of 41 sherds weighing 364g was recovered from Site B. The majority of the pottery dated to the post-medieval period, and the Packway features pro- duced very few contemporary artefacts. The methods
_of analysis and the fabric and vessel form descriptions | used are described under the finds section for Site A
(ps 25).
_ Prehistoric pottery | Only three sherds, all of the probable late Bronze
Age/early Iron Age fabric C2, were recovered from Site B. They came from an upper fill of the Packway
| ditch (context 30). An oolitic fabric comparable to
fabric C2 was found in middle and upper fills of the Packway ditch during the earlier excavation and dated to the ‘Early pre-Roman Iron Age’ (Wainwright and Longworth 1971, ware A).
Fabric description
Soft fabric; moderate, fairly well-sorted crushed oolitic limestone fragments <0.5mm; rare (1-3%) quartz grains <0.25mm; rare mica. Unoxidised with partially oxidised exterior surface.
_ See Site A for general discussion of dating and source
for this fabric and other prehistoric fabrics.
Romano-British pottery
Romano-British pottery was recovered from the upper fill of the Packway ditch, from pit 112 and from three lynchets (120, 121, 123). In three of the four latter features, the Romano-British sherds were mixed with medieval or post-medieval sherds.
55
Medieval pottery
Seven sherds of medieval pottery were recovered. Six of these are in a coarse sandy fabric (E422), sometimes scratchmarked, which is common in south Wiltshire. Very similar wares are known to have been produced at the Laverstock kilns outside Salisbury, from the late 12th century at least until the late 13th century. A single sherd in a finer sandy, glazed fabric (Q400) was also found, but is of unknown source.
Post-medieval pottery
Twenty-one post-medieval sherds were recovered, comprising earthenwares, mostly of Verwood type, stoneware, white salt-glaze, and fine white wares of modern (19th/20th century) type.
Other Finds
Other finds included ceramic building material, fired clay, iron objects, flint and animal bone. Details of all the finds categories are contained within the site archive.
Discussion
The information recovered from the excavated evi- dence was limited, as so few features were found. The finds were few in number, and do not add any further information to our knowledge of the site. The profile of the ditch matches that excavated in the 1960s, as do the fills of the ditch. The lynchets to the north of the enclosure all appear to be medieval or post-medieval, and therefore unrelated to the rest of the site. In general, the excavation of Site B has added little to the evidence from Wainwright’s earlier excavation (Wain- wright and Longworth 1971). It does, however, sug- gest that the paucity of features encountered in Wainwright’s excavation may not have been entirely attributable to clearance for the road, but partly a result of features having been too shallow or delicate to survive ploughing, or to a genuine absence of archaeological remains.
Acknowledgements. The excavations were funded by the Ministry of Defence Land Agency, through the Property Services Agency, and thanks are due to Land Agent John Wooden for his cooperation and interest in the project. Wessex Archaeology would also like to thank the farmer, Mr Burgess, for his cooperation; and Nigel Allmark of Kings, for the provision of plant hire and site accommodation. The authors are grateful to Chris Place of Wiltshire’s County Archaeological Office, for the provision of background information, and especially to Mark Corney of the RCHME for his invaluable help and advice throughout the project. The plot of the cropmarks, shown in Figure 4, is reproduced by kind permission of the RCHME. Thanks are also due to Linda Coleman and Liz James who drew the illustrations, and to Susan Davies, Julie Gardiner and Richard Newman for their advice and comments on the text. Finally, we would like to thank the members of the excavation team and volunteers who coped admirably with the appalling weather conditions.
56 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
BIBLIOGRAPHY
ACSADI, G., and NEMESKERI, J., 1970 History of Human Lifespan and Mortality, Budapest
ALLEN, M.J., 1988 ‘Archaeological and Environmental Aspects of Colluviation in South-East England’ in W. Groenman-van Waateringe and M. Robinson (eds.), Man- made Soils, Brit. Archaeol. Rep. Supp. Series 410, Oxford, 67-92
—— 1989a ‘Land Snails’ in P.J. Fasham, D.E. Farwell, and R.J.B. Whinney, The Archaeological Site at Easton Lane, Winchester, Hampshire Fld. Club Archaeol. Soc. Mono. 6, Gloucester, 135-40
1989b “The Molluscan Evidence’ in S$. Howard, ‘Exca-
vation of a Round Barrow at Barford Farm’, Proc. Dorset
Natur. Hist. Archaeol. Soc. 111, 49-52
forthcoming ‘Prehistoric Land-use History and Collu- vial Products on the Wessex Chalklands’ in M.G. Bell and J. Boardman (eds.), Past and Present Soil Erosion, Oxford
——, ENTWISTLE R., and RICHARDS, J.Cc., 1990 ‘Molluscan Studies’ in J.C. Richards, The Stonehenge Environs Project, English Heritage Archaeol. Mono. 16, 253-8
, HEATON, M.J., and RICHARDS, J.C., forthcoming ‘Sal- vage Excavations at Bronze Age Barrow Durrington G3’
ANDERSON, A.S., 1978 ‘Wiltshire Fine Wares’ in P. Arthur, and G. Marsh (eds.), Early Fine Wares in Roman Britain, Brit. Archaeol. Rep. 57, Oxford, 357-71
— 1979 The Roman Pottery Industry in North Wiltshire, Swindon Archaeol. Soc. Rep. 2
BARRETT, J.C., 1980 ‘The Pottery of the Later Bronze Age in Lowland England’, Proc. Prehist. Soc. 46, 297-319
BASS, W.M.,1987 Human Osteology, Columbia, Missouri
BELL, M.G., 1983 ‘Valley Sediments as Evidence of Prehis- toric Land-use on the South Downs’, Proc. Prehist. Soc. 49, 119-50
BOWEN, H.C. and FOWLER, P.J., 1966 ‘Romano-British Settlements in Dorset and Wiltshire’ in C.Thomas, Romano-British Settlement in Roman Britain, Counc. Brit. Archaeol. Res. Rep. 7, 43-67
BROTHWELL, D.R., 1981 Digging up Bones, 3rd ed., London
CARTER, S.P., 1990 ‘The Stratification and Taphonomy of Shells in Calcareous Soils: Implications for Land Snail Analysis in Archaeology’, ¥%. Archaeol. Sct. 17, 497-507
CLAPHAM, A.R., TUTIN, T.G., and WARBURG, E.F., 1952 Flora of the British Isles, Cambridge
CLARKE, G., 1979 ‘The Roman Cemetery at Lankhills’, Winchester Studies 3: Pre-Roman and Roman Winchester, Part II, Oxford
CLEAL, R., with RAYMOND, F., 1990 ‘The Prehistoric Pottery’ in J. Richards, The Stonehenge Environs Project, English Heritage Archaeol. Mono. Ser. 16, 233-46
CORNEY, M., 1989 ‘Multiple Ditch Systems and Late Iron Age Settlement in Central Wessex’ in M. Bowden, D. Mackay, and P. Topping, (eds.) From Cornwall to Caithness: Some Aspects of British Field Archaeology, Brit. Archaeol. Rep. 209, 111-28
CRUMMY, N., 1983 Colchester Archaeological Report 2: The Roman Small Finds from Excavations in Colchester 1971-9, Colchester
CUNLIFFE, B., 1971 Excavations at Fishbourne, Vol. II: The Finds, Society of Antiquaries Res. Rep. 27
— 1973 ‘The Period of Romanization’ in E. Crittall (ed.), The Victoria County History of Wiltshire, Vol. 1, part 2, Oxford, 439-53
1977 ‘A Romano-British Village at Chalton, Hants’,
Proc. Hampshire Fld. Club Archaeol. Soc. 33 (1976), 45-68
1978 Iron Age Communities in Britain (2nd ed.), London
DAVIES, S.M., 1981 ‘Excavations at Old Down Farm, Andover. Part II: Prehistoric and Roman’, Proc. Hamp- shire Fld. Club Archaeol. Soc. 37, 81-163
1990 ‘Pottery’ in P. Rahtz, ‘Bower Chalke 1959: Excavations at Great Ditch Banks and Middle Chase Ditch’ WAM 83, 30-37
ELLIS, A.E., 1951 ‘Census of the Distribution of British Non-marine Mollusca’, 7. Conchology 23, 171-244
— 1969 British Snails, Oxford
ESMONDE CLEARY, A.S., 1989 The Ending of Roman Britain, London
EVANS, J.G., 1971 ‘Durrington Walls: the Pre-henge Environ- ment’ in G.J. Wainwright and I.H. Longworth, 1971 (see below)
— 1972 Land Snails in Archaeology, London
1984 ‘Land Snail Analysis’ in B. Cunliffe, Danebury, an
Tron Age Hillfort in Hampshire, Vol. 2. The Excavations
1969-1978: The Finds, Counc. Brit. Archaeol. Res. Rep.
52, 476-81
and JONES, H., 1979 ‘Mount Pleasant and Woodhenge: the Land Mollusca’ in G.J. Wainwright, Mount Pleasant, Dorset: Excavations 1970-1971, Society of Antiquaries Res. Rep. 37, 190-213
FINNEGAN, M., 1978 ‘Non-metric Variation of the Infracranial Skeleton’, 7. Anatomy, 152, 1, 23-37
FULFORD, M.G., 1975 New Forest Roman Pottery, Brit. Archaeol. Rep. 17, Oxford
GINGELL, C., 1982 ‘Excavation of an Iron Age Enclosure at Groundwell Farm, Blunsdon St Andrew 1976-7’, WAM 76 (1981), 33-76
GREEN, H.S., 1980 The Flint Arrowheads of the British Isles, Brit. Archaeol. Rep. 75 (i) and (11), Oxford
HARDING, P., 1990 ‘The Comparative Analysis of Four Stratified Flint Assemblages and a Knapping Cluster’, in J.C. Richards, The Stonehenge Environs Project, English Heritage Archaeol. Mono. 16, 213-24
1991 ‘Stratified Groups From Rowden and Cowleaze’ in P.J. Woodward, The South Dorset Ridgeway. Survey and Excavations 1977-84. Dorset Natur. Hist. Archaeol. Soc. Mono. Ser. 8, Dorchester, 73-86
HINGLEY, R., 1989 Rural Settlement in Roman Britain, London
JONES, M., 1984 ‘Interpretation of Archaeological Plant Remains: Ethnographic Models from Greece’ in W. van Zeist and W.A. Casparie, (eds.), Plants and Ancient Man, Rotterdam, 43-62
KERNEY, M.P., 1967 ‘Appendix III. Mollusca’ in P.M. Christie, ‘A Barrow-cemetery of the Second Millennium BC in Wiltshire, England’, Proc. Prehist. Soc. 33, 365-6
LEACH, P., 1982 Ilchester Vol. I: Excavations 1974-5, Western Archaeological Trust Excavation Mono. 3, Bristol
MACGREGOR, A., 1985 Bone, Antler, Ivory and Horn, Beckenham
MecWHIRR, A., VINER, L., and WELLS, C., 1982 Romano-Bnitish Cemeteries at Cirencester, Cirencester
MALTBY, J.M., 1981 ‘Iron Age, Romano-British and Anglo- Saxon Animal Husbandry: a Review of the Faunal Evi- dence’ in M. Jones and G. Dimbleby (eds.), The Environ- ment of Man, Brit. Archaeol. Rep. 87, 155-203
—— forthcoming ‘The Animal Bones from the Romano- British Deposits at the Greyhound and Methodist Chapel
IRON AGE AND ROMANO-BRITISH ENCLOSURES IN THE AVON VALLEY 57
Site in Dorchester’, in P.J. Woodward, S.M. Davies and A.H. Graham, Excavations at The Old Methodist Chapel and Greyhound Yard, Dorchester, 1981-84, Dorset Natur. Hist. Archaeol. Soc. Mono. Ser.
MANNING, W.H.,1985 Catalogue of the Romano-Bnitish Iron Tools, Fittings and Weapons in the British Museum, London
MILLETT, M., and JAMES, S., 1983 ‘Excavations at Cowdery’s Down Basingstoke, Hampshire, 1978-81’, Archaeol. 7. 140, 151-279
and Russell, D., 1984 ‘An Iron Age and Romano- British Site at Viables Farm, Basingstoke’, Proc. Hamp- shire Fld. Club Archaeol. Soc. 40, 49-60
MORRIS, E.L., 1991a Guidelines for the Analysis of Pottery, Wessex Archaeology Guideline 4, Salisbury
— 199lb ‘Ceramic Analysis and the Pottery from Pot- terne: a Summary’ in A. Middleton and I. Freestone (eds.), Recent Developments in Ceramic Petrology, British Museum Occasional Paper 81, 277-87
NEWCOMER, M.H., and KARLIN, C., 1987 ‘Flint Chips from Pincevent’ in G. de G. Sieveking and M.H. Newcomer (eds.), The Human Uses of Flint and Chert, Papers from the Fourth International Flint Symposium, Cambridge, 33-36
PARTRIDGE, C., 1981 Skeleton Green: A late Iron Age and Romano-British Site, Britannia Monographs No. 2
OLIVER, M., and APPLIN, B., 1979 ‘Excavation of an Iron Age and Romano-British Settlement at Rucstalls Hill, Basingstoke’, Proc. Hampshire Fld. Club Archaeol. Soc. 35, 49-92
RICHARDSON, K.M., 1951 ‘The Excavation of Iron Age Vil- lages on Boscombe Down West’, WAM 54, 123-68
RIGBY, V., 1973 ‘Potters’ Stamps on Terra Nigra and Terra Rubra Found in Britain’ in A. Detsicas (ed.), Current Research in Romano-British Coarse Pottery, Counc. Brit. Archaeol. Res. Rep. 10, 7-24
ROGERS, B., and RODDHAM, D., 1991 “The Excavations at Wellhead, Westbury 1959-1966’, WAM 84, 51-60
ROGERS, J., WALDRON, T., DIEPPE, P., and WATT, I., 1987 ‘Arthropathies in Palaeopathology: the Basis of Classifi- cation According to Most Probable Cause’, 7. Archaeol. Sc. 14, 179-93
SAVILLE, A., 1977-78 ‘Five Flint Assemblages from Exca- vated Sites in Wiltshire’ WAM 73, 1-27
SHACKLEY, M.L., 1976 ‘The Danebury Project: an Experi- ment in Site Sediment Recording’ in D.A. Davidson and M.L. Shackley (eds.), Geoarchaeology, London, 9-21
SPAREY GREEN, C., 1987 Excavations at Poundbury. Vol. I: The Settlements, Dorset Natur. Hist. Archaeol. Soc. Mono. Ser. 7, Dorchester
SWAN, V., 1971 ‘The Coarse Pottery’ in G.J. Wainwright, 1971 (see below)
—— 1975 ‘Oare Reconsidered and the Origins of Savernake Ware in Wiltshire’, Britannia 6, 37-61
— 1984 The Roman Pottery Kilns of Britain, RCHM Supp. Series 5, London
TAYLOR, C.C., 1983 Village and Farmstead: A History of Rural Settlement in England, Londen
THOMAS, K.D., 1977 ‘The Mollusca from an Iron Age Pit at Winklebury’ in K. Smith, ‘The Excavations of Winkle- bury Camp, Basingstoke, Hampshire’, Proc. Prehist. Soc. 43, 70-74
TROTTER, M., and GLESER, G.C., 1958 ‘A Re-evaluation of Estimation of Stature Based on Measurements of Stature Taken During Life and Long-bones after Death’, Am. 7. Phys. Anthrop. 16, 79-123
WAINWRIGHT, G.J., 1971 ‘The Excavation of Prehistoric and Romano-British Settlements near Durrington Walls, Wiltshire, 1970’, WAM 66, 76-128
—, and LONGWORTH, I.H., 1971 Durrington Walls: Exca- vations 1966-1968, Society of Antiquaries Res. Rep. 29
WALDEN, H.U., 1976, A Nomenclatural List of the Land Mollusca of the British Isles’, 7. Conchology, 29, 21-25
WESSEX ARCHAEOLOGY, 1991 Butterfield Down, Amesbury, Wiltshire. Archaeological Evaluations Stage 4A, W49
WILLIAMS, D.F., 1977 ‘The Romano-British Black-Burnished Industry: an Essay on Characterisation by Heavy Mineral Analysis’ in D.P.S. Peacock (ed.), Pottery and Early Commerce: Characterisation and Early Trade in Roman and Later Ceramics, London, 163—220
YOUNG, C.J., 1977 The Roman Pottery Industry of the Oxford Region, Brit. Archaeol. Rep. 43, Oxford
Wiltshire Archaeological and Natural History Magazine, vol. 86 (1993), pp. 58-74
Excavations at the Postern Mill Site, Malmesbury, 1986-87
by CHRISTOPHER K. CURRIE
with contributions by TRUDIE C. CURRIE, MARK A. LOFT and PAUL ROBINSON
Rescue excavation near the site of the former Postern Mill, Malmesbury revealed evidence for continued use of the area for industrial purposes since at least Late Saxon tmes. A wide range of activity was identified ranging from stone- and iron-working in the Late Saxon/Norman period to tanning or slaughtering of animals, milling, and brewing in the post-medieval period. These discoveries enhance our knowledge of the everyday life of a small market
town in Wiltshire over a thousand-year period.
INTRODUCTION (Figures 1 and 2)
Wiltshire Rescue Archaeology Project (WRAP) undertook rescue excavation ahead of development in the winter of 1986-87 on the site of the former Linolite factory at the end of Burnivale, Malmesbury (NGR ST 932872).
Incomplete clearance by the developers had left substantial concrete rubble lying across much of the site, limiting choice of areas to excavate. Small test trenches were dug in unpromising positions but these revealed no features and the only archaeological materials obtained consisted mainly of redeposited medieval and post-medieval ceramic.
Only one trench was feasible in a potentially promising position close to the site of the former Postern Mill, a structure incorporated into the Lino- lite factory and demolished with it early in 1986. This
Figure 1. Postern Mill Site, Malmesbury: location. (Scale approximate)
building is known to have been in existence in 1831 (Rees 1951, 191), but may be much earlier. The trench opened up was flanked to north and south by massive piles of concrete abandoned by the developers. To the west a number of modern drains were apparent, while to the east was a tarmac road. Because of these obstructions, the trench was restric- ted to an area of approximately 10m by 3.5m.
Cal \ \
TRENCH 1
t---7
Millpool Factory ‘site’ | =a aay 4 -
0 50m
Figure 2. Postern Mill Site, Malmesbury: location of trenches
EXCAVATIONS AT THE POSTERN MILL SITE, MALMESBURY
The site archive has been deposited with the County Field Archaeological Officer at County Hall, Trowbridge.
Topography (Figures 1 and 2)
The town of Malmesbury stands on a conspicuous imestone promontory surrounded on three sides by the River Avon. The Sherston and Tetbury tribu- ‘aries of the river meet on the south-east side of the ‘own. The only approach to this naturally fortified dosition is along a narrow neck of land on the aorth-west side of the present town, where a castle was sited in the 12th century.
_ The Postern Mill site is situated on the west side of the town between the bottom of a steep slope and the Sherston Avon. The river here forms a pool that must aave been associated with the mill. The distance detween the present course of the river and the dottom of the slope is about 75 metres. At the lower end of the site the remains of a former mill leet are visible halfway between the incline and the river.
| There seems to be some evidence that the slope of the hill on which the town stands has been made steeper by the quarrying of stone for building. The dood plain below this slope comprises river-borne silts, some of considerable depth, overlying clay. The ‘ow lying level of the Mill site indicates that flooding would have occurred here in the past. The normal ‘evel of the river is about 68m AOD by the mill pool while the top of the slope above it is 87.01m AOD; the incline was calculated at 1 in 2.5m.
Alstory
According to local tradition, Malmesbury took its qame from a hermit of Irish descent known as Mailduib or Madulph. He is believed to have settled on the hilltop because of its natural seclusion. Well- don Finn (1930, 44) gives the date as around AD 596 although the Victoria County History (Watkin 1956, 210) dates Mailduib’s arrival as ‘toward the middle of the 7th century’. From this humble origin grew a ‘monastery that came to be one of the richest in southern England, through royal patronage and rants of land from the late 7th century.
| Malmesbury’s urban status was ensured by the founding of a burgh there by Alfred the Great in the ate 9th century (Hunter Blair 1977, 293-4). In terms bf hide assessment, Malmesbury ranked tenth out of the 30 named places in southern England in the ‘Burghal Hidage. At the 1200 hide assessment, it
59
appears to have been a lesser settlement than other Wiltshire burghs at Cricklade (1500 hides) and Wilton (1400 hides), although it exceeded Chisbury at 700 hides (Biddle 1976, 126-7).
King Athelstan (AD 924-939) is reputed to have rebuilt the abbey (Welldon Finn 1930, 49), and during the monastic revival of Dunstan’s episcopate (AD 963-988), it received further attention. The pre-Conquest building, presumably of stone, was still standing c.1143 when William of Malmesbury described it as fairer than any Saxon church still standing in England (James 1925, 27). Fire subse- quently destroyed this building, and the present church’s Norman work appears to date from after 1150.
The 12th century saw much activity at Malmes- bury. Roger, Bishop of Sarum (died 1139) is supposed to have built a castle next to the abbey (Rees 1951, 184). The cause of frequent complaints, it was even- tually demolished in the reign of John (1199-1216). Local tradition attributes the beginning of work on the town walls to Roger’s episcopate (John Bowen, pers. comm.), although Rees considered that only on the east side of the town do remnants of this work remain (1951, 189).
It is thought that the Postern Mill may have taken its name from a small postern gate in the defences, allowing access to the narrow flood plain below the hill. The original wall is thought to have followed a line slightly higher than the present pathway known as King’s Wall (Rees 1951, 190-1).
The documentary evidence (see archive) suggests that Malmesbury was a town of local importance throughout the medieval period. Although its national status must have declined after the dissolution of the abbey, the town continued to thrive with the cloth- making industry of Wiltshire. As this declined from the 18th century onward, so the manufacturing side of the town’s economy gradually weakened. This was, however, a slow process and a number of industries survived in the town into the first half of the 20th century. The last two hundred years of Malmesbury’s economic history have paralleled that of many similar local towns.
It is only in the 18th century that any detail of the local economy can be determined. This seems to have been based largely on the products of the agricultural hinterland. It was hoped that archaeological exca- vation would show whether the economy had changed significantly since the foundation of the burgh in the late 9th century.
60 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
The Excavations
ROMAN
Although no features or layers were found attribut- able to this period, a small quantity of Roman mater- ial was re-deposited in early medieval levels (see specialist reports).
MEDIEVAL
Phase 1
Phase 1 comprised orange-brown clays overlying undisturbed clay. Although no features were found belonging to this phase, the clay contained a moderate amount of charcoal flecks, and fragments of burnt clay. A small number of animal bones were also recovered.
Phases 2 and 3 (Figures 34 and 7)
The earliest feature was a linear gully (F66), up to 0.6m wide, which cut into phase | levels. This was aligned east-west, and filled with a brown clay inter- spersed with moderate-sized chunks of limestone, many of which seemed to be aligned east-west on their long axes. The upper levels of the gully con- tained concentrated lenses of charcoal and burnt material, with some iron slag. Overlying this feature was a large limestone dump (F36) contained in a matrix of brown silty clays (context 37) which merged with context 56 to the south. It was considered that these levels were essentially the same, although con- text 56 contained less stone.
The stone dumping levels may have been truncated in antiquity. To the west of this truncation silty layers, contexts 39, 63 and 38, had built up.
Context 39 was a blue-grey silty clay containing much burnt material, especially waterlogged charcoal, and large quantities of small limestone chippings, as well as some stone slabs at its northern end. A much-used sharpening stone was recovered from within it. Context 39, like most of the levels exca- vated, slumped westwards towards the River Avon. This level rapidly became free of stone chippings, and the blue-grey staining less pronounced, towards the north and west. Where stone chippings were absent, the level was referred to as context 69. Following the line of feature 66 and the general disposition of the stone dump (F36), two large slabs, weighing about 30kg, overlay context 69.
Overlying context 39 were contexts 63 and 38. The latter was grey silty clay loam containing evidence for burning. Context 63 was a similar blue-grey silty clay
——
Figure 3. Postern Mill: trench plan, phase 2
Figure 4. Postern Mill: trench plan, phase 3
layer to context 39, the main differences being in the absence of stone chippings in any quantity, and the presence of large quantities of iron slag. Context 63 was overlain by context 38. The latter slumped sharply eastwards beneath the layers of the following phase.
EXCAVATIONS AT THE POSTERN MILL SITE, MALMESBURY
Discussion of Phases 1-3
Dating evidence for these phases was found in the form of crude limestone tempered ceramic fabrics. The typology of these wares indicated a 10th—11th century date. There was very little contamination from successive phases, and the fabrics present in phases 2 and 3 were not found in upper levels in any quantity. A lack of intermixing between these wares and the main fabric found for the early/mid 12th century (Great Somerford ware) seems to indicate that they had largely gone out of use before c.1100.
Although the evidence of stone working has been assigned to phase 2, owing to the deposition of stone chippings below the layers containing the most slag, it is possible that iron smelting or some other iron working activity was going on at the same time. Furthermore, moderate quantities of slag were found in the earliest feature (F66), which pre-dates the stone dump and context 39.
It is possible that the stone dump was once on the banks of the river, or an artificial leet. The stone chippings in context 39 may have washed into this river, or been deliberately dumped there but, by the end of phase 3, any watercourse that may have existed here had probably silted up. The presence of the phase 4 hearth (see below) over the silty deposits suggests that this area was dry land by the 12th century. From this it must be concluded that phase 3 had ended by the early years of the 12th century at the latest. It seems that the bulk of the iron working took place before this but after the stone chippings were deposited. The stone working layers are fairly safely assigned to a period before 1100. This eliminates the deposition of stone chippings from association with the recorded building programme of Bishop Roger of Sarum, who died in 1139, and suggests that it repre- sents earlier building activity, perhaps that docu- mented in the 10th century, by Athelstan and Dunstan.
Phase 4 (Figures 5 and 7)
This phase is focused on a limestone built hearth ‘feature 45) which stood on top of phase 3 levels. The aearth comprised small flat limestone blocks laid over one another. The blocks were set in a heavily burnt clay matrix, the bottom courses being built on top of context 67, a layer that appears to have been a Continuation of context 38. The remaining structure was approximately 1.30m by 0.8m, and about ).25-0.3m high.
| There were signs that the structure had slumped cowards the west. It is possible that it had once >xtended westwards, but may have been disturbed by
61
8+ +
——
Figure 5. Postern Mill Site, Malmesbury: trench plans. Above: hearth F45 (phase 4) cut through by F65 (phase 5); below: linear feature F30 (phase 5)
a later cut (Feature 65). The earliest fill of this later feature seemed to date, on evidence of the pottery, from the late 12th or early 13th century, thus determining the latest date at which the hearth was used. All around the hearth was a thick build-up of burnt clay and ash.
Discussion of Phase 4
Phase 4 is dominated by Great Somerford fabrics and cooking pots dating to the early/middle years of the 12th century (Vince 1981, 316). The build-up of ash and burnt clay around the hearth was quite consider- able and denotes intense activity over the years it was in use. The western part of the hearth appeared to have been removed by the cutting of Feature 65: as this seems datable to the late 12th or the 13th century, the hearth had probably ceased to be used by 1300.
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
62
suoNdas JO SUOTIRDO] puke saimieay / aseyd Surmoys ued youan :Ainqsoupey “US [TW UrIsog “9 ainsty
Ww?
+ + + + + + = + +
meer’
HIN3UL | qdId
/_ Ne
EXCAVATIONS AT THE POSTERN MILL SITE, MALMESBURY
Phase 5 (Figures 5-7)
This phase is associated with the digging of a large cut (F65) approximately along the line of the later wall F12/F13. There was no trace of the far side of this feature within the excavated area, and all levels to the east slumped towards it.
The fills within F65 were dark in colour and many contained a high percentage of silt. The lower layers were considerably water-logged. Continual flooding, combined with the danger of nearby rubble falling into the trench, prevented the feature from being fully sxcavated. The latest ceramics recovered from the lowest fills excavated seemed to indicate a late 12th- century or early 13th-century date.
The upper fills were slightly water-logged, and had relatively high silt contents. Ceramics from these evels contained a number of glazed jug sherds, and were dated to the 13th century. For the first time the Minety type fabrics were recognised in some quantity, iuthough residual 12th-century Great Somerford wares were still present. The bulk of the material, by ypology, appeared to be 13th-century, but some of he jug sherds could be attributable to the 14th.
. A linear feature (F30) cut through phase 4 levels on it north-south alignment. This feature lay about 1.5m o the east of F65, and was up to 1.5m wide. Its west side was sharply cut, and there was evidence of -onsiderable recutting throughout the later medieval, ind post-medieval periods. A well worn coin of idward I (1272-1307) was found in the lower fill. This was considered to have been lost long after its issue, ind was associated with late medieval ceramics, dating rom the later 14th century to the early 16th century. .
Discussion of Phase §
[t would appear that the cut, F65, may have been a arge and substantial channel. As it is known that here was a mill on the site in the post-medieval deriod, it is very likely that one existed on the site rom an earlier date. Certainly the stream feeding the nill was only a few metres to the west of the excavated irea until it was abandoned, and backfilled in the sresent century. It is possible that F65 represents the cutting of a mill stream or artificial river course after the middle years of the 12th century. This channel \ppears to have become backfilled by the 14th century. | F30 also appears to be a cutting associated with water-control though it seems to have been later than F65. The worn coin of Edward I found in it is inlikely to have been deposited before the mid-14th century. The ceramics from the lower fills of F30 >xtend from the later 14th to the 16th centuries. This dating indicates that F30 may have been dug after F65
63
had been filled in. There was some evidence that F30 had been recut on a number of occasions well into the post-medieval period, suggesting that the channel had had a long life, being dug at sometime after c.1350; it was still in use, in a heavily silted state, in the 18th century.
POST-MEDIEVAL
Phase 6 (Figure 7)
The lowest levels of this phase contained ceramics of the late 1S5th—-early 16th centuries. By the time most of the levels were deposited, however, post-medieval dates are confirmed. Although Cistercian and late Minety wares are present, the bulk of the material dates from the 18th century. Dark silty soils overlying F65 are dated mainly to the middle years of this century with some re-deposition of earlier materials. Possibly flooding contributed to this contamination, but these soils would also have been prone to earth- worm activity.
Building materials, in the form of ceramic tiles, were found in phase 6 levels, indicating that in the middle years of the 18th century, a nearby building dating from the medieval period was either demol- ished, rebuilt or re-roofed. The obvious candidate would appear to be the Postern Mill as it was less than 30 metres from the excavated trench. This dumping of materials represents the final filling of F30. Both F30 and F65 contained large quantities of animal bones, with a predominance of head and feet bones (see bone report, below).
Phase 7 (Figures 6—7) This phase is represented by the construction of the limestone wall, shown from 19th- and 20th-century map evidence to have been part of a building known as the Mill Brewery. No evidence was found for a foundation trench, although the layers were of such similar composition that it is possible one was missed. The initial foundation was made of large tabular blocks three courses deep and about 0.7m wide, the whole being laid on a bed of sandy mortar 20-30mm thick (context 50). This contained a piece of 18th century bottle glass and a small sherd of white salt glazed earthenware. Although it is possible to suggest a date of c.1730-80 for the construction of this building, thus relating it to the rebuilding/demolition evidence in phase 6, documentary evidence records no building on the site before 1831 except for the Mill. The wall foundation was capped by a wall (F13) approximately 9.22m narrower (that is 0.4648m
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
64
suondes :AInqsouep ‘9S [TW Usalsog */ amm3tJ
uoljgas Bulzey YJJON
UU ee 7°
voljas Bulses YjNos
IXCAVATIONS AT THE POSTERN MILL SITE, MALMESBURY
wide) than itself; made of smaller limestone blocks, it was set in an identical mortar to that of underlying F12. To the west of this wall was a stone floor made of massive slabs or tabular limestone, only roughly worked to shape. The floor surface was therefore uneven. Dividing the floor were two aisles approxi- mately 0.4-0.5m wide. There was no floor present here, only a dark silty loam surface, contexts 20/21. These continued under the stone floor, which was set on a thin layer of sandy mortar 2—3cm thick (context 26). This mortar and the interface between contexts 26 and 20 contained late post-medieval ceramics, dating to the late 18th and early 19th centuries.
In the far north-east corner of the excavated part of this building, the stone floor showed signs of wear in a circular form. It was thought this was caused by the continuous movement of a horizontal wheel against the floor and may be evidence for some form of machinery within the building. Immediately beneath the floor against the north baulk, a linear feature (F34) was observed, filled with loose-fitting lime- stone. It was up to 0.18m wide and 0.1m deep, with the level dropping towards the west. The end of the feature was situated beneath the area where the ‘conjectured wheel turned on the stone floor and it ‘might be suggested that water was used to cool this machinery, F34 being a drain to take this water away.
Above the stone floor was a dark silty layer, only a few centimetres thick. Immediately over it was a bed ‘of gravel, about 0.05m thick, covered by the thick ‘concrete floor of the Linolite factory that stood on the ‘site up until 1986. | CONCLUSIONS | Despite the small area excavated, the site displayed | evidence of almost continuous use for a range of ‘industrial purposes from the Late Saxon period through to the mid 1980s. | There was some evidence to suggest a Romano- British presence in Malmesbury. The small amount of /Roman objects recovered was not sufficient to make _any clear statements about the nature of this presence. | The earliest clear evidence for occupation on the ‘site dates from the Late Saxon period. The type of pottery recovered in the earliest phases can be assigned to the timespan AD 850-1100. The first “major activity on the site seems to have been associ- ated with stone-working although some slag and charcoal below a major stone dump (F36) may indi- | cate that iron working was being carried on alongside, _or prior to, the stone working. The largest quantity of waste slag, however, was discovered to be overlying
65
the earliest evidence for stone working. Documentary evidence for considerable building activity under- taken at the nearby abbey in the 10th century suggests that the phase 2 stone working was associated with these works.
It seems possible to date the hearth (F45) to the early/mid-12th century because of the prevalence of Great Somerford type wares associated with it. The exact function of the hearth is not known, and attempts to link it with smithing or iron working were not entirely successful: slag was recovered from phase 4 levels and so it is not possible to discount the idea, although quantities of this material were recovered from all levels from phase 2 onwards. It is possible, therefore, that iron working continued in the vicinity for many centuries.
In phase 5, two channels were dug running north— south through the area. The large channel (F65) cut through the hearth area, post-dating the mid-12th century. This probably represents either an artificial cutting for the main river, or a stream supplying a mill. Silting or backfilling seems to have been well underway by the later 13th century. The second, smaller channel (F30) also seems to have post-dated phase 4, and could also represent a further attempt at water control.
The predominance of animal head and feet bones in phases 5 and 6 suggests that this material was either slaughterhouse waste or associated with tanning. These activities may have taken place close to the site, although no features were found associated with either activity within the excavated area.
The dumping of materials, together with the rela- tively rapid backfilling of feature F65, suggests that this channel had fallen out of use by the end of the 14th century. A mill leet is known to have served the Postern Mill, on a line slightly to the west of F65, in the 19th century. It would appear that this mill stream and the mill had been in place since at least late medieval times, suggesting that F65 may have been abandoned before the mill of this period had been constructed. F30 seems to have silted up more gradu- ally, and its final in-filling appears to have been at the end of phase 6 (mid-18th century).
Finally, a late 18th or early 19th century building was constructed on the site. Documents suggest brewing started after 1831, although there is archaeo- logical evidence, in the form of malting tiles, to suggest that it may have taken place here from the late 18th century.
The conclusions drawn therefore suggest the foll- owing activities had taken place in the area occupied by the Postern Mill.
66
1. Corn milling: a mill probably existed from Saxon times. The first tentative evidence suggesting a mill dates from the formation of F65 in the late 12th or early 13th century.
2. Stone working: evidence for stone working pre- dates 1100. Documentary evidence could put this as early as the second quarter of the 10th century.
3. Some iron working, possibly on a small scale, seems to pre-date the major stone working phase. This activity seems to have continued until c.1100, and may have carried on into the 12th century and later. The hearth that existed on the site between c.1100—1250 may have been associ- ated with iron working or smithing, although the evidence is not positive.
4. The disposal of slaughter or tanning waste took place on the site from the 13th century, and may have continued into the 18th century. The site was well suited for a tanning industry with plenti- ful supplies of water, and clay to allow tanning pits to be dug within it.
5. Brewing and malting took place here in the 19th century, and on into the 20th century. It is possible these activities may have started in the later 18th century.
6. A factory complex known as Linolite took over from the brewery after a short abandonment of the site in the 20th century. The factory buildings were demolished in 1986.
There is a good case for suggesting that the area served as an industrial suburb of the town since Saxon umes. The location of the site, close to a water supply yet separated from the town proper by a tall escarpment, meant that it remained suitable for this function over a long period. Although site usage has changed little in general terms, the evidence for different industries from such a small area seems remarkable.
The Finds
BUILDING MATERIALS
Stone
Limestone was worked on the site during phase 2, as evidenced by large numbers of stone chippings present, and a dump containing large blocks of unworked limestone. It would appear that the worked stone had been taken from the site to help with extensive building programmes known from docu- mentary evidence to have occurred from the second
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
quarter of the 10th century. Locally quarried lime- stone was also employed for structures found in phases 4 and 6~7.
Ceramic Roof Tiles
A total of 40 fragments of ceramic roofing tile was recovered together with 8 fragments of ceramic floor ule. On the roofing tle, 80% by number were believed to be medieval, 20% post-medieval.
Three main fabrics were recognised; only one — the largest group — was provenanced, being a product of the medieval Minety industries. Further details of the fabrics can be found in the archive.
Ceramic Floor Tiles
Eight fragments of floor tile were recovered, mainly from contexts preceding the stone building in phase 7. Two fragments were possibly medieval. In appear- ance, none of the tiles appeared to be decorative, and it would seem they may have been part of the flooring of secular, industrial buildings.
Two perforated tile fragments are of particular note. The larger piece, which was glazed on both upper and lower surfaces, measured 100mm_ by 75mm. Six perforations survived whole, although the remains of eight more were visible along three sides. The holes on the upper surface were between 21 and 22mm across and tapered down to between 4 and Smm on the under-side. The tile was exactly 30mm thick, and made of an earthenware fabric similar to Ashton Keynes ware (see archive).
These tiles were malting tiles used in the brewing and malting industries (Dr R. Wilcox, pers. comm. ). It is known that a malt house existed on the site during the 19th century. The most complete tile was found in an 18th-century context.
THE POTTERY (Figures 8—9)
Roman Pottery by MARK A. LOFT
A few heavily abraded sherds of Roman date were | recovered from the site, re-deposited in medieval | contexts. Only one sherd was recognised as a known |.
fabric: an abraded piece of Samian recovered from the area of the stone dump (F36).
A few other sherds were identified as Roman, but | these were so small and abraded that little else canbe |
said about them.
|
_ EXCAVATIONS AT THE POSTERN MILL SITE, MALMESBURY 67
_ Medieval Pottery _ by CHRISTOPHER K. CURRIE
where it was found in 12th-century contexts. The fabric is unglazed and this particular form is considered to be earlier than the glazed rim Catalogue of Illustrated Pottery cooking pots known as Selsley Common ware Figure 8 which dominated 13th-century West Country 1. Cooking pot with flaring rim in reduced lime- coarseware assemblages (after Dunning 1949).
stone tempered fabric. Hand made with rim possibly finished on slow wheel. The limestone tempering of this pot is generally angular and apparently ill sorted. Inclusions are abundant, vary in size from less than 1mm to 7mm, and are usually grey in colour, although some exhibit brown staining. There is rare quartz present. Rim present 100%, context T1—37, phase 2. The type is similar to pots found at Gloucester in 12th-century contexts (Vince 1979, 177, fig. 9 no. 122; 1983, 126, fig. 76 no. 8) although the form was common throughout England in the 10th—12th centuries.
Cooking pot with flaring rim in limestone tem- pered fabric. Light grey core and light brown exterior with thin dark grey covering. Lime- stone inclusions are less frequent than in no. 1, and vary in size from less than 1mm to 3mm. Both rounded and angular particles are present. Also present, but rare, is quartz up to 1mm, shell up to 2mm and some reddish ore up to 0.5mm. Rim present 7%, context 37, phase 2. As with no. | the form was common throughout the Saxo-Norman period and could range from the late 9th century to the 12th century. Cooking pot with plain rim in limestone tempered fabric (similar to no. 2); grey core with pinkish buff exterior. Rim present 6%, context 56, phase 2. The form is paralleled at Gloucester (Vince 1979, 177, fig. 9no. 123), and at Huish, Wiltshire (Musty 1972, 127, fig. 7 nos. 20 and 21). The date range from these sites indicates late 11th—12th century, although an earlier date is possible. Cooking pot with straight rim in limestone tempered fabric. Fabric as no. 1; hand made pot. Rim present 6%, context 63, phase 3. Cooking pot with everted rim in limestone tempered fabric. Grey core with pinky brown exterior. Greyish limestone inclusions poorly sorted, ranging from 1mm to 5mm. Other inclu- sions include shell, rare quartz particles, occa- sional ore and frequent black or dark red rock fragments. Rim present 4%, context 33, phase 4. The form is common in the early medieval period and can be paralleled at a number of local sites, in particular Gloucester (Vince 1983, fig. 76 no. 6, fig. 77 no. 33; 1979, 177, fig. 9 no. 115)
Jug or pitcher with horizontal rim in limestone tempered fabric. Light grey core with buff exterior; traces of green glaze on outer surface. Rim present 15%, context 43, phase 4. Fabric provenanced as being produced by Minety industries. It is not considered possible to dis- unguish between coarseware jugs and pitchers from the rim alone amongst limestone tempered industries in North Wiltshire.
Cooking pot with slightly flaring rim and frilled lip in limestone tempered ware. Hard fired with light grey core and pinky brown exterior. Rim present 5%, context 55, early phase 5. The fabric is similar to that described by Vince (1982, 202) as belonging to the Minety indus- tries. There is, however, an absence of brown stained oolites in this vessel and it resembles the materials excavated by Thompson (1970, 167-9) at Great Somerford. The author has discussed the similarities between Great Somerford and Minety fabrics elsewhere (Currie, forthcoming). The Somerford assemblage was dated to the 12th century (Thompson 1970, 169). The form of this vessel is paralleled at Huish by Musty (1972, 127, fig. 7 no. 11) where it is also dated to the 12th century.
Vessel with slightly hooked rim in limestone tempered fabric. Hard fired with pinky brown exterior surfaces and light grey core. Rim present 7%, context 55, early phase 5. Similar fabric to no. 7. The form is paralleled at Winch- combe (Vince 1985a, 117, fig. 8 no. 3) where it is dated to the late Saxon/early Medieval period. Parallels with the Somerford fabrics, however, give it a 12th-century date (Thompson 1970). Cooking pot with flat rim in limestone tempered fabric. Hard fired with pinky orange exterior surfaces and light grey core. Similar fabric to nos. 7 and 8. Rim present 5%, context 49, phase 5. Cooking pot with slightly overhanging flat rim and well defined shoulder. Hard fired with pinky orange exterior and light grey core. Lime- stone tempered fabric similar to wares found at Great Somerford. Rim present 8%, context 49, phase 5. The form has parallels at Whittington Court (Jope 1952, 62, fig. 6 no. 11) where it is dated to the later 11th and 12th centuries.
68
BE
12:
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
J \
?
oe
Figure 8. Postern Mill Site, Malmesbury: medieval pottery, phases 2-5
Cooking pot with straight rim in limestone tempered fabric. Buff orange oxidised exterior and grey core. Limestone inclusions common and ill-sorted; sizes ranging from 1mm to 9mm, mostly angular types. Rim present 7%, context 49, phase 5. Fabric similar to nos. 1 and 4. Dated later 11th—12th centuries as nos. 3, 4 and 9:
Jug with slightly flaring rim in limestone tem- pered fabric, reduced grey throughout with thick green glaze on outer surface. Minety fabric. Rim present 11%, context 49, phase 5. Minety jugs are difficult to differentiate from pitchers although Musty (1973, 84) claims pitchers are generally unglazed. This contradicts the author’s own experience and is confirmed by
13:
14.
Vince (1981, 312) who has identified pitchers with glaze in the 12th century in other indus- tries. The dating of this vessel could be late 12th century.
Skillet handle with diagonally slashed decoration. Hard fired buff outer surfaces and light grey core with splashes of green glaze on upper surface. Limestone tempered ware of the Minety industries. Context 8a, phase 5. The type of decoration on the handle (diagonal slashing) replaced square stabbing at some time in the later 12th or early 13th century (Vince 1983, 126). Skillets and small handled cauldrons are known in London from the early 13th century (Vince 1985b, 46, fig. 14).
Bead rimmed vessel in limestone tempered
EXCAVATIONS AT THE POSTERN MILL SITE, MALMESBURY
15:
16.
fabric. Buff oxidised ware throughout. Similar to Great Somerford fabric. Rim present 6%, context 8a, phase 5. Nearest parallel is at Huish (Musty 1972, 127, fig. 7 no. 18) where it is dated to the 12th century.
Cooking pot with flaring rim in limestone temp- ered fabric. Reduced grey colour throughout. Similar fabrics to nos. 2 and 3. Rim present 10%, context lla, phase 5. This vessel is of similar type to nos. 1 and 2, and was made throughout the Saxo-Norman period. Date range most likely to be 10th—mid-12th centuries. In this instance, the piece is probably re- deposited in a later context.
Jug in limestone tempered fabric. Buff exterior with light grey core and splashes of glaze on both surfaces. Minety fabric. Rim present 14%, context lla, phase 5. Dated 13th/14th century.
Figure 9
Wie
18.
19:
20.
| 21.
Bowl with flat, slightly overhanging rim in limestone tempered fabric. Buff exterior with grey core and patchy green glaze on interior surface. Minety fabric. Rim present 5%, context 7, phase 5. The form is paralleled at Whit- tington Court (Jope 1952, 64, fig. 7) where it is dated to the later 11th or 12th century, and at Gloucester (Vince 1979, 177, fig. 9 no. 111, of 12th-century date). The Minety fabric makes a date before the early 12th century unlikely. Pitcher handle in limestone tempered fabric. Pinky brown exterior surfaces with grey core. Splashes of green glaze on inner edge of handle. Minety fabric. Context 7, phase 5. The handle is decorated with diagonal slashes, a style which replaced earlier decorative forms from the late 12th/early 13th century onwards (Vince 1983, 126).
Rod handle with buff exterior and grey core in hard fired fabric with occasional limestone inclusions less than Imm. Splashes of yellow- green glaze present. Context 7, phase 5. Cooking pot with everted rim in limestone tempered ware. Thin wash of yellow-green glaze on inner edge of rim. Limestone inclusions common, well sorted and generally less than 2mm. Rim present 7%, context 6, phase 5. This form was common in the West Country in the 13th century (Dunning 1949, 30), although it has been found in late 12th-century contexts in Gloucester (Vince 1983, 126).
Pitcher or jug with flat, overhanging rim in limestone tempered fabric. Pinky orange
DD
23:
24.
D5:
26.
69
exterior with dark grey core and occasional splashes of glaze on outer edge. The fabric is similar to Minety fabric except that the core is darker fired than is normal on Minety vessels. Rim present 16%, context 42a, phase 5. Pitchers of this type were common in the West Country throughout the late 12th and 13th centuries (Bruce-Mitford 1940, 104).
Vessel with slight flanged rim and thumb impressed strip on outer edge below lip of the rim. Limestone tempered fabric with buff exterior and light grey core. Rim present 12%, context 42a, phase 5.
Large pan? with overhanging flanged rim in limestone tempered ware. Another possibility is that this vessel is a lid. Buff exterior with light grey core and internal green glaze. Minety fabric. Rim present 6%, context 3, phase 6. Pans were known to have been manufactured at Minety but the rims recorded near the sup- posed kiln site excavated by Musty (1973, 82-84, 81, fig. 1 no. 1) are simpler forms than those found at Malmesbury. If this vessel is a lid, it has a more elaborate rim than those published from the Minety kiln site (Musty 1973, 84).
Vessels with wide flanges are thought to represent the later phase of the Minety indus- tries. They are provisionally dated to the late 14th—15th centuries by Musty (1973). At Trow- bridge they were found associated with demo- lition levels of the late 15th-early 16th centuries (Dr R. Wilcox, pers. comm.).
Large pan? with flanged rim in limestone tem- pered fabric. As for no. 23, there is a possibility that this vessel is a lid. Hard buff ware with grey core and internal green glaze. Minety fabric. Rim present 7%, context 8, phase 6.
Pitcher with slightly flanged rim in limestone tempered fabric. Diagonally slashed handle with splashes of glaze and vertical combing on exterior surface. Pinky brown outer surface, buff interior and light grey core. Rim present 35%, context unstratified. Similar vessels have been found throughout the West Country in late 12th and 13th century contexts (Bruce-Mitford 1940).
Everted rim cooking pot in limestone tempered fabric with traces of yellow green glaze on inner edge of rim. Buff exterior with light grey core. Minety fabric. Rim present 18%, context unstratified. Form known as Selsley Common ware after Dunning (1949, 30-44).
70
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
25 a 26 es
i oe
0
Figure 9. Postern Mill Site, Malmesbury: medieval pottery, phases S—6 (nos. 17-26) and post-medieval pottery, phases 6—7 (nos. 27-30)
Post Medieval Pottery
Lies
28.
29.
Jug with plain rim in dark red sandy fabric. Internal and external lead glaze. Cistercian type fabric. Rim present 9%, context 20, phase 7. These wares were made in Bristol during the 16th and early 17th centuries (Cirencester Museum type fabric file CM 208).
Base of wheel thrown jug in ‘Cistercian’ ware. Dark red sandy fabric with internal and external glaze. Base present 40%, context 40, phase 6. Jar in white salt glaze stoneware. Rim present 14%, context 20, phase 7. The base of a similar vessel was identified at Winchcombe, Glou- cestershire by Vince (1985a, 122, 123 fig. 11 no. 47) and dated to the mid-18th century. These
wares were made largely between 1730 and 1780 when they were superseded by the cheaper creamwares (Dr R. Wilcox, pers. comm.).
30. Dish or plate in white salt glaze stoneware. Rim present 3%, context 20, phase 7. DISCUSSION
The medieval assemblage was dominated by lime- stone tempered coarsewares (80.8% by weight, 88.5% by sherd count; for further details of fabrics, see archive). This was also the case in the assemblage recovered from Wootton Bassett (Currie forth- coming); there were, however, significant differences between the coarsewares recovered from the two sites.
EXCAVATIONS AT THE POSTERN MILL SITE, MALMESBURY
It would appear that the Malmesbury assemblage started at an earlier date than that of Wootton Bassett, with the earliest limestone tempered fabrics dating from the Saxo-Norman period. It is difficult, in the absence of other dating evidence, to put an exact date on these ceramics, but their typology ranges from the late 9th century to c.1100.
Phase 4 is dominated by coarsewares bearing simi- larities to Great Somerford wares (Thompson 1970). These are dated to the early/mid-12th century by Vince (1981, 316). This phase, like the earlier phases, is dominated by cooking wares, although the odd pitcher sherd suggests a continuation into the later 12th century.
Minety fabrics start to appear in phase 4 (mid/late 12th century), but only occur in numbers in phase 5. There are also unprovenanced limestone tempered wares in phases 4 and 5 that are not recorded else- where, although types found at Brinkworth (Currie 1992) were present. The Minety fabrics never completely dominated the assemblage although they are found in stratified contexts of the latest medieval periods early in phase 6 (late 15th—-early 16th century).
The conclusions drawn from an examination of the limestone tempered wares at Malmesbury indicate that there were a number of small industries serving very localised areas in the 12th and 13th centuries in North Wiltshire. It is possible that pottery was being manufactured very near Malmesbury, if not in the town itself. This conjectured industry could account for the unprovenanced fabrics. Certainly fabrics thought to be manufactured near Wootton Bassett or Cirencester were not positively identified at Malmes- bury. The nearest provenanced wares appear to be from Great Somerford. The presence of a kiln here is, however, not certain (Thompson 1970), and it may be that Somerford was supplied from somewhere nearer to Malmesbury itself. The author has seen at least two different fabrics in the Somerford assemblage. One, containing flint, was not recovered at Malmesbury. This suggests that the Somerford assemblage came from two different sources, as limestone temper with- out flint dominated the Malmesbury assemblage in the mid 12th century. Fabrics containing flint were almost totally absent at Malmesbury, although such fabrics are common at both Wootton Bassett and Cirencester.
From the 13th century onwards, Minety fabrics become increasingly common at Malmesbury. It is possible that this industry developed a near monopoly in the area as the medieval period progressed. There is evidence that it gradually supplanted the local indus-
71
tries at Cirencester and Wootton Bassett, and late Minety wares seem to continue right to the end of the medieval period, possibly into the early 16th century.
Of the medieval jug fabrics, there were 4.3% by weight and 7.8% by sherd numbers. No attempt was made to differentiate coarseware jug fragments from the bulk of the coarsewares. Provenanced jug wares recognised by odd sherds were from Ham Green and Naish Hill, although the bulk of the material was not identified with a particular site.
The jugs seemed to have been made largely with clays used as dug, and are typical of the types in use throughout the Oxford Clay Vale. Iron rich bands are common. Most of the sherds would appear to date from the 13th—14th centuries. One sherd of Surrey white ware was found, indicating an occasional wider trading area.
Earthenwares dominate the post medieval assem- blage as they do on most local sites. The commonest fabric, by far, was from Ashton Keynes. Other fabrics found are typical for the area, being similar to those from the Cirencester (Cirencester Museum Type fabric file) and Wootton Bassett assemblages.
The upper date ranges of the Malmesbury assem- blage would seem to be early in the 19th century from the inclusion of small quantities of Mocha and Willow Pattern wares, though these are possibly residual. Creamwares were relatively rare, considering the ubiquitous nature of this fabric, although white salt glazed wares were common. This would seem to indicate that the main bulk of the assemblage dates before 1780. Other late 17th and 18th century fabrics were common (Delft, iron glaze buff earthenwares etc.).
CLAY PIPE
A total of 46 fragments was recovered, most of these being pieces of stem. Only two diagnostic bowls were found, which were assigned to the early 18th century on bowl type; these probably date to the period c.1710-50 (Wood 1963, 272). Only one stamped fragment was recovered. This was badly damaged and it was not possible to distinguish the marking apart from the last two letters of one line: RY, and a letter in the bottom line: H.
The contexts containing clay pipe fragments, with one exception, were within phases 6 and 7, the greater majority being recovered in the latter phase in the level immediately below the stone flooring (F14) of the post-medieval building. This high concentration of clay pipe stem within the area defined by the wall compared with concentrations outside the wall (a ratio
ai. THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
of 41:5) may indicate that the line of the wall was a boundary before the stone floor was laid. This rein- forces other indications that the building was in existence in the 18th century.
SMALL FINDS (not illustrated) by PAUL ROBINSON
1. Reduced size imitation of centenionalis of Mag- nentius and Decentius with reverse type Victoriae DD NN Aug et Caes of the Lyons mint. Date AD 351-55. 2. Edward I (?); silver penny of the London mint. Type uncertain in North (1960) although struck between 1279 and 1307; much worn.
The rest of the small finds were largely unstratified, and of little interest. They can be referred to in the archive.
THE ANIMAL BONE by TRUDIE C. CURRIE
Introduction
A total of 1967 bones were recovered, 1938 from trench 1. Only the bones from this trench are con- sidered in the following discussion. 889 bones were identified to species; the rest were too badly frag- mented to be identified in any meaningful way. Sheep (61.1% by number), cattle (25.5%) and pig (9.3%) formed the bulk of the assemblage, with other identi- fied species being domestic fowl (1.7%), horse (1.3%), cat (0.4%), dog (0.3%) and fish (0.2%). Table 1 shows the number of bones found by phase.
A large proportion of the bones came from the skulls and upper/lower extremities of the animals concerned. Of the 227 cattle bones, 30% were from the feet alone. Of the 543 identified sheep bones, 69.4% came from the feet and 21% from the skull. The majority of the metapodial (feet) bones for both sheep and cattle came from phases 5 and 6. The high percentage of these particular bones indicates that the assemblage represents either waste from slaughtering or association with an industry such as tanning.
The presence of a tanning or slaughter industry is supported by the evidence from other documented sites producing a similar type of bone material (Rob- inson and Wilson 1987, 71). Metapodial refuse was used to make an oil for softening animal hide and it was common for tanners to ask for these bones to be left on the hide for this purpose. Calves’ foot oil was sull used for softening saddles until recently although it is now unobtainable in the Wiltshire area (Manager, Downton Tannery, pers. comm.).
Butchery
The cattle mandibles showed cut marks that may have been made by cutting out the tongue (masseter muscle). The mandibles were also frequently butchered on either side of the tooth row. Many of the ribs were cut away from the spinal column, and all of the vertebrae had been butchered through the centre.
All cattle scapulae found at Malmesbury in all phases were butchered either across the neck or laterally through the glenoid process giving the bones a distinctive appearance.
Upper and lower limbs were generally cut mid- shaft, leaving either the proximal or distal bone present in phases 3, 5, 6 and 7. Cattle extremities were cut on all surfaces of the metapodia. Ankle bones had been cut through the centre with cut marks visible on both lateral and anterior surfaces.
The sheep mandibles examined were usually left complete with only occasional cut marks noted on the bone surface. Ribs and vertebrae were butchered in the same way as in cattle. Scapulae were butchered across the neck leaving the glenoid cavity as the only form of identification. Upper and lower limbs were butchered mid-shaft, although during phase 5 humerii were cut vertically down the middle. No evidence was observed for butchery or cut marks on any sheep metapodials.
The pig remains showed evidence of butchery on all identified bones although there was little sign of any method used, with random cuts apparent on all recovered bones. Pig limbs were butchered mid-shaft on the long bones as with cattle and sheep.
Age of Death
Assessment of age was determined by examining the long bone and metapodial epiphyses of the cattle, sheep and pig together with tooth eruption from the recovered maxilla and mandibles. This information was obtained following the methods described by Silver (1969) and Cornwall (1956).
The cattle bones recovered from phases 1 and 2 were of mature adults. In contrast to this, cattle bones found in phases 5 and 6 showed that the animals were killed before they were two years old as many of the metapodial bones had not fused at the distal end.
Throughout all phases, sheep seem to have been killed at between one and a half years old and full maturity. Sheep bones from phases 5 and 6 were from mature animals at the time of slaughter as most of the metapodial bones (line 1) were fused at the distal end. The recovery of complete humerii and radii (fusion occurring between two and half and three years) supports this.
EXCAVATIONS AT THE POSTERN MILL SITE, MALMESBURY
73
Table 1. Malmesbury: total number of bones recovered from each phase
Phase
10th century?
10th—11th century 10th—11th century Early/mid 12th century Later 12th—15th century Mid 15th—18th century Late 18th—mid 19th century Modern
COND NB W LKB
Total
Total no. of bones Number identified
26 8 170 63 4] 14 80 28 483 176 922 508 213 90 3 2 1938 889
Table 2. bone percentages of each species calculated by phase
Note: details of percentages of type of bone in each phase can be found in the archive.
Identified species
1 7) 3
Sheep 25% 19% 35.7% Cattle 50% 49.2% 42.9% Pig 25% 25.4% 7.1% Domestic fowl = 6.4% 7.1% Horse = = 7.1% Cat - = = Dog = — = Fish - = = Total identified
bones per phase 8 63 14
The pig remains were mainly from young adults in all phases with tooth eruption indicating an age range of one to three years. One patella in phase 5/6 proved to come from a young suckling, but this was the only evidence suggesting early slaughter.
Conclusions
Until the 12th century the number of cattle seems to have declined compared with sheep but pig remained constant (see Table 2). After 1200 cattle stayed pro- portionally consistent, but sheep continued to increase, while pig declined.
It is assumed that cattle were kept to a mature age throughout most of the medieval period because of their use as draught animals, and for their dairy products, rather than as meat producers (Bell er al.
42.9% 17.9%
Phases
4 5 6 7 8
54.5% 69.3% 70% 50%
25.6% 22.6% 22.2% 50% 25% 14.8% 5.5% 3.3% = 7.1% 2.3% 0.6% 1.1% = = 1.1% 1.4% 2.2% = = 1.1% 0.2% 1.1% _ 3.6% = 0.4% = = 3.6% 0.6% = = = 176 508 90 2
1984, 94). This coincides with the mature age of cattle from bone evidence in phases | to 4. By the 16th century increasing proportions of cattle were reared for beef and veal as is evidenced from other sites in the West Country (ibid., 95). Again, this pattern is reflected at Malmesbury, with the majority of phase 5 and 6 cattle bones being from animals slaughtered before reaching one and a half years.
The increasing proportion of sheep throughout the phases probably reflects the growth of the cloth industry in the West Country, particularly notable after the mid-14th century when shortage of labour made sheep farming a more viable proposition than arable cultivation (Platt 1978, 130).
The increase in the number of sheep bones at the expense of pig and cattle matches the historical
74 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
evidence for changing land use in the Malmesbury area in the medieval and early post-medieval periods (see archive). Up until the 13th century, there were still large areas of the forest of Bradon to the east of Malmesbury where pigs are likely to have been grazed in some numbers. These areas were gradually assarted during the medieval period, reducing suitable pas- turage for pig keeping.
The bone evidence tentatively indicates that the pig numbers declined from the 12th century, whilst sheep increased. There are even suggestions that the domi- nance of sheep in north-west Wiltshire was already affecting land usage before the disruptions of the 14th century began to affect the local economy.
This evidence should, however, be viewed with some caution as there is no way of telling how typical the Malmesbury assemblage is. Although the figures look reasonably consistent in Table 2, the phase 4 cattle figure and the phase 3 pig figure seem to indicate anomalies that could derive from a number of outside factors. It is also accepted that the size of the assemblage may not be sufficient to give more than a general impression.
It would seem, none the less, that the proportional growth of the sheep percentage is evidence of the growing importance and, eventually, dominance of the wool and cloth industries in north-west Wiltshire in the later medieval and post-medieval periods at the expense of other land uses. Only comparison with other local assemblages yet to be excavated will reveal the reliability of these changes in relation to arable and woodland husbandry.
Acknowledgements. Acknowledgement of assistance is given to: mem- bers of WRAP, for carrying out the excavation, often in cold and wet conditions; Roy Canham, County Field Archaeological Officer, and Ron Wilcox, Department of Archaeology, Chippenham Technical College, for their patience and advice; the management committee of WRAP at Chippenham Technical College; and to the developers, Bowprine Ltd of Somerford, Wiltshire.
Special attention must be given to Eunice Wood of 53 Burnivale, who allowed us to use her outhouse in wet weather for shelter. John Bowen is acknowledged for imparting his knowledge of the town. Final thanks are offered to Haden Sheppard who gave up his time to search the contractor’s spoil heaps with his metal detector, permitting the recovery of items that would not otherwise have been found.
REFERENCES AND BIBLIOGRAPHY
BELL, M., 1984 ‘Environmental Archaeology in South West England’ in H.C.M. Keeley (ed.), Environmental Archaeo- logy: A Regional Review, Vol. I, London, 43-133
BIDDLE, M., 1976 ‘Anglo-Saxon Towns’ in D. Wilson (ed.), The Archaeology of Anglo-Saxon England, London, 99-150
BRUCE-MITFORD, R., 1940 ‘Medieval Tripod Pitchers’ in ‘Notes’, Antiquaries Journal 20, 103-12
CORNWALL, I., 1956 Bones for the Archaeologist, London
CURRIE, C.K., 1992 ‘Excavations at the Roman Kiln Site at Brinkworth, Wiltshire, 1986’, WAM 85, 27-50
— forthcoming ‘Excavations at Wootton Bassett 1986’, Wiltshire County Council typescript
DUNNING, G.C., 1949 ‘Report on the Medieval Pottery of Selsley Common, near Stroud’, Trans. Bristol and Glou- cestershire Archaeol. Soc. 68, 30-44
HUNTER BLAIR, P., 1977 An Introduction to Anglo-Saxon England, 1977 (2nd ed.), London
JAMES, M.R., 1925 Abbeys, London
JOPE, E.M., 1952 ‘The Medieval Pottery’ in H.E. O’Neil, ‘Whittington Court Roman Villa, Gloucestershire’, Trans. Bnistol and Gloucestershire Archaeol. Soc. 71, 61-76
MUSTY, J., 1972 ‘The Pottery’ in N.P. Thompson, ‘Exca- vations at Huish 1967-68’, WAM 67, 112-31
— 1973 ‘A Preliminary Account of the Pottery Industry at Minety, Wiltshire’, WAM 68, 79-88
NORTH, J.J., 1960 English Hammered Coinage, vol. 2, London
PLATT, C., 1978 Medieval England, London
REES, H., 1951 ‘Malmesbury, its Castle and Walls’, WAM 51, 184-92
ROBINSON, M., and WILSON, R., 1987 ‘A Survey of Environ- mental Archaeology in the South Midlands’ in H.C.M. Keeley (ed.), Environmental Archaeology: A Regional Review, Vol. IT, London, 16-100
SILVER, I.A., 1969 ‘The Ageing of Domestic Animals’ in D. Brothwell and E. Higgs (eds.), Science in Archaeology (2nd ed.), London, 283-302
THOMPSON, N.P., 1970 ‘A Medieval Pit at Great Somerford’, WAM 65, 165-71
VINCE, A., 1979 ‘The Pottery’ in C.M. Heighway, ‘Exca- vations at 11 Westgate Street, Gloucester’, Medieval Archaeol. 23, 170-81
— 1981 ‘The Medieval Pottery Industry in Southern England: 10th to 13th Centuries’ in H. Howard and E. Morris (eds.), Production and Distribution: a Ceramic Viewpoint, Brit. Archaeol. Rep. Int. Series 120, Oxford, 309-22
— 1982 ‘The Medieval Pottery’ in R.H. Leech and A.D. McWhirr, ‘Excavations at St. John’s Hospital, Ciren- cester’, Trans. Bristol and Gloucestershire Archaeol. Soc. 100, 191-209
— 1983 ‘Medieval and Post-medieval Pottery’ in C.M. Heighway, The East and North Gates of Gloucester, Bristol, 125-61
— 1985a ‘The Late Saxon, Medieval and Post-medieval Pottery’ in A. Saville, ‘Salvage Recording of Romano- British, Saxon, Medieval and Post-medieval Remains at North Street, Winchcombe, Gloucester’, Trans. Bristol and Gloucester Archaeol. Soc. 103, 113-23
— 1985b ‘The Saxon and Medieval Pottery of London, a Review’, Medieval Archaeol. 29, 25—93
WATKIN, A., 1956 ‘The Abbey of Malmesbury’ in R.B. Pugh and E. Crittall (eds.), Victoria County History of Wiltshire 3, London, 210-31
WELLDON FINN, R., 1930 Wiltshire, London
WOOD, E.S., 1963 Archaeology in Britain, London
Wiltshire Archaeological and Natural History Magazine, vol. 86 (1993), pp. 75-87
Archaeological Investigations at the Shrunken Village of Knook
by DUNCAN COE and RICHARD NEWMAN
with a contribution by LORRAINE MEPHAM
This report details and combines the results of two separate projects: excavations in advance of the construction of a pipeline and an assessment of the route of a proposed road scheme. Both these developments threatened to impinge upon an area of earthworks at the village of Knook in the Wylye valley. The investigations gave some indication of the extent of the 12th- to 13th-century occupation of the settlement. Both the archaeological and historical evidence suggest that the medieval settlement was bi-focal as a result of its lordship being split between two manors. Although occupation appears to have peaked in the 13th century, periods of contraction and expansion may have alternated until the settlement’s most recent shrinkage in the early years of this century.
INTRODUCTION
The village of Knook (ST 940419) lies in the Wylye valley, 7km south-east of Warminster (Figure 1).The present day settlement lies on the north-east bank of the river Wylye, with its nucleus around the small medieval church and the manor house (Figure 2). The geology of the area is dominated by chalk with substantial colluvial deposits in the valley bottoms. To the north-east of the current settlement lies a field of earthworks, field 6, (SMR number ST94SW455), | adjacent to a sunken lane leading from the _ Warminster-Salisbury road to the church. Recently surveyed by the RCHM(E) (Figure 3), these earth- works indicate that the settlement once extended up | to the Warminster—Salisbury road, the present A36, | possibly on both sides of the sunken lane. Although | any surface trace on the northern side of the sunken ‘lane has been obscured by modern local authority | housing (Figure 5), medieval pottery can be found in the embanked gardens of these properties. Until at | least the 1930s, housing also extended along the west | side of the Warminster—Salisbury road (see Figure 4). |The buildings on the east side are all part of West | Farm (Figures 2 and 7), where a few sherds of 11th- |to 13th-century pottery have recently been found (material shown to Wessex Archaeology by the owner, Mr Ian Duff). | In October and November 1990 two separate devel- ‘opments were proposed which threatened to impinge | upon the earthworks at Knook. The first of these was a pipeline to be constructed by Wessex Water plc
| | i
|
from Codford to Heytesbury, and the second was Wiltshire County Council’s proposed upgrading of the A36 between Codford and Heytesbury (Figure 3). Both of these proposals presented a threat to the presumed settlement earthworks, the pipeline at the south-western end of field 6 and the road at its north-eastern end. Wessex Archaeology contracted to carry out the archaeological work ahead of these developments. The detailed results of these projects are contained in two unpublished reports submitted to the separate clients, copies of which are lodged with Wiltshire County Council’s Sites and Mon- uments Record (Barnes and Allen 1990 and Coe 1990).
This report is an amalgamation of the results of the two separate projects, which were a response to two separate and specific briefs. The most detailed data were obtained from the salvage excavation and watching brief carried out for Wessex Water plc. The work for Wiltshire County Council was an evaluation intended to provide proposals for further work and as an aid to preservation. The level of investigation was largely concerned with locating the presence of archaeological remains and defining their nature; detailed analysis of archaeological deposits lay outside the scope of the works.
The most significant results obtained during the two projects related to the area of the shrunken settlement of Knook. This report is a synthesis of the data obtained and concentrates on defining the area of former settlement. In particular the extent and devel- opment of medieval settlement is assessed.
76 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE |
Warminster
We - — \ 2 z
ee
(
INS
= eo)
Built-up area
Land over 130 metres
Figure 1. Knook and the Wylye Valley
Historical Background
It is likely that a settlement existed at Knook near the present church by at least the 10th century. The fabric of the church, although largely restored in 1876, contains decorative elements of late 10th- to early 11th-century date (Taylor 1968, 54-7), possibly derived from an earlier building. By 1086 the settlement of Cunuche appears to have been divided into two manors, one held by Alfward Colling and the other by Leofgeat (Thorn, C., and F. 1979, 73c, d and 74b). By the 14th century Knook still appears to have been divided into two manors; although the honour of Knook was held by the Earl of Gloucester, it was divided into two lordships and held by the Earl of Gloucester’s vassals, the lords of Wardour and Heytesbury (the Lovel family). By the reign of Rich- ard II (1377-1399) only one manor is recorded at Knook, in the possession of Sir John Lovel, although the Earls of Gloucester continued as overlords (Colt- Hoare 1812). Knook appears, thenceforth, to have
remained a single manor. Knook’s early history as two manors may explain the bi-focal nature of the settlement as it appears on its earliest maps, dated to the 18th century. On an estate map of 1774 (Figure 4) it seems to have two foci, one around the church and manor, and the other near West Farm and the Warminster—Salisbury road. This latter focus may, from the field names surrounding it, have once been known as Garston.
In 1774 some settlement survived in the area of field 6. The population increased during the early years of the 19th century (Colt-Hoare 1812), and contraction of the settlement to its present size occurred mainly in the late 19th and early 20th centuries (I. Duff pers. comm.). To date, insufficient research has been undertaken to establish a firm link between the settlement shrinkage evident from the archaeological remains and any circumstances observed in the histor- ical record.
THE SHRUNKEN VILLAGE OF KNOOK
|
| /
| Bee | ark
\ Methodology
The pipeline route was archaeologically excavated Jacross the south-western end of field 6 (Figure 5). In Trench A, 82m in length and 1.7m wide, the topsoil |was removed by mechanical excavator and _ all archaeological layers and features were excavated “manually (Figure 6). In addition, finds were collected |from the plot to the north of the sunken lane (field 5) vafter topsoil stripping.
The projected line of the road improvement crossed the north-east end of the earthworks and also crossed
77
Knook Horse Hill
(
| Well Bottom °
Upton Lovell
Figure 2. Knook: present day local geography and settlement pattern
a number of fields to the north of the village. Throughout this area the route was examined in order to assess its archaeological potential. This evaluation involved fieldwalking, test pit digging, trial trenching and geophysical survey (Figures 3 and 5). Fieldwalk- ing was only possible in the fields to the north of the village (fields 2 and 4). All the test pits were excavated by hand with the intention of recording soil profiles/ features and retrieving finds. In field 6 two test pits were excavated, and in the fields to the north (fields 1, 3 and 4) a further ten test pits were dug. A trial trench
78 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE |
High density magnetic readings
---- Low density reading indicating a trackway
Figure 4. Knook and its common fields as shown in 1774. (Redrawn from original map, Wiltshire Record Office 101/174/1)
THE SHRUNKEN VILLAGE OF KNOOK
OA ; KEY O 1m Test pit without medieval pottery 0 100 @ Im Test pit with medieval pottery —aa pecracrl Im
@ 2m Test pit with medieval pottery 1-7 Field numbers
WA
Figure 5. Excavations around Knook in 1990: site locations
TRENCH B
Pe
79
KMN
80 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
[TRENCH A |
JNV
Figure 6. Knook: plan of Trench A
(B) was excavated in the field to the south-east of the village (field 7) in order to retrieve environmental data from the colluvial sequence. A geophysical survey using a magnetometer was undertaken in fields 1, 3, 4, 6 and 7. In addition to this work it was noted during the evaluation that a small group of earthworks existed in fields 3 and 4, to the north of the village, adjacent to the main road: these were surveyed (Figure 5).
Results
THE PIPE TRENCH (Figures 5 and 6)
The trench along the route of the pipeline across the earthworks (Trench A, Figure 5), produced little stratigraphic information and only five features.
The topsoil was generally 0.20m deep, overlying a clay loam deposit up to 0.40m in depth. Below this lay mixed chalk and loam and degraded natural chalk. At the north-western end of the field and running for approximately 25m into it, a different sequence was observed. Here a slight hollow appears to have been
landscaped to form a flat field; topsoil overlay a series of chalk dumps and loamy lenses, lying above a buried clay loam soil above natural chalk. This dumping would appear to be the result of modern activity.
The main earthwork feature, plotted by the RCHM(E) and crossed by the trench, was a low bank running north-east-south-west along the south- eastern edge of the field (Figure 3). This bank had been built over the top of a ditch, 32, which was 3.2m wide and 0.65m deep. The main point of interest about the ditch was its base: this was flat but three narrow gullies had been cut into it leaving low ridges of solid chalk. Small amounts of 11th- to 13th-century pottery were retrieved from all layers and fills of the bank and ditch. This feature appeared to mark the south-eastern edge of the earthworks, a hypothesis which may be confirmed by the absence of finds and features observed during pipe-laying in this area.
The other features found during the excavation were two gullies/ditches, a pit and a post hole/pit. The gullies, 22 and 38, were both aligned approximately north-south. Feature 22 was the larger of the two; it was 1.45m wide, 0.55m deep, and produced a large assemblage of llth- to 13th-century pottery, five sherds of which are similar to pre-Conquest material found elsewhere in Wiltshire (fabric C400, see below). The smaller feature, 38, was 0.45m wide and 0.40m deep and produced only a single sherd of pottery, probably from the Laverstock kilns. The pit, 41, was found on the edge of the trench and its full extent was not observed. It was also not possible to excavate this feature below 0.70m depth. Another large pottery assemblage, with a possible 11th- to 15th-century date range, was retrieved from this feature. The post hole/pit, 20, was oval in plan, 0.90m long, 0.56m wide, 0.30m deep and contained pottery of a medieval date.
THE EVALUATION (Figure 5)
Fieldwalking
Artefacts were collected from the surface of the ploughsoil in fields 2 and 4. In both of these fields a crop had recently been harvested and this resulted in poor surface visibility. Neither field produced any evidence of medieval activity, although modern ceramics, brick and tile affirmed the presence of the Great War army camp in field 2, and a row of houses in field 4, which were present on aerial photographs taken in 1930.
THE SHRUNKEN VILLAGE OF KNOOK
Test Pits
A total of twelve test pits were excavated in the immediate vicinity of the village of Knook: three in field 1, four in field 3, four in field 4 and two in field 6.
Field 1 No features were noted and only 3 sherds of medieval pottery were recovered, all from test pit 11.
Field 3. One feature was noted, a shallow slot, 0.1lm wide and 0.08m deep, cut into natural chalk in test pit 50. Sixty-five sherds of medieval pottery were found in test pit 13, 33 in test pit 50 but only 3 in test pit 14 and none in test pit 15.
Field 4 No features were noted. Six sherds of medi- eval pottery were recovered from test pit 51 only.
Field 6 Despite the presence of earthworks no features were noted. Sixteen sherds of medieval pottery were recovered from test pit 18, and in test pit 52 which was 2m’, 136 sherds were found. In comparison only 4 sherds of post-medieval ceramics were recovered, clearly indicating that the activity in the area was principally medieval.
Trial Trench (Figure 5)
A single trial trench, B, was excavated in field 7 to the south-east of the earthworks. This was one of three excavated along the route of the road but the only one | near the village of Knook. It was dug by mechanical excavator to recover details of the development of the colluvial sequence and to provide environmental evid- ence in the form of mollusc shells from the colluvium. _ The colluvial sequence suggested the presence of a relict older land surface of probable earlier prehistoric ‘date, whilst the main colluvial deposits developed in the Iron Age or later. The lack of colluvial material dating to the medieval period indicates that the valley ‘slopes were stable at this time or that hillwash was later eroded (information supplied by M.J. Allen).
Geophysical Survey (Figure 3)
|The area around the shrunken settlement was subjec- ‘ted to a geophysical survey following the excavation of \the test pits. Fields 1, 3, 4 and 7 were all partially surveyed, in an attempt to define the area of archaeological activity to the north and east of the |main area of earthworks. A second survey was later carried out in field 6 to gain more information about ithe main area of earthworks. The survey was con- ‘ducted by Geophysical Surveys of Bradford (Gater
| i
i
|
81
and Gaffney 1990a and b), using a fluxgate magne- tometer over a 20 metre grid. Detailed plots of their results are held in the site archive.
The survey isolated the south-east corner of field 1 as being badly disturbed, probably by the Great War camp or farming activity. The rest of the survey, in fields 3, 4 and 7, produced no significant anomalies. In fields 3 and 4 this may be attributed to the substantial depth of deposits, over 0.80m in test pits 13, 50 and 51, which may have masked features.
In field 6 the survey confirmed the findings of the earlier earthwork survey (Figure 3). In addition, a low resistance linear feature was noted. This was about 5m wide and ran approximately north-south, termin- ating at a large, low resistance rectangular feature. The linear feature may have been a sunken way; when compared with the surveyed earthworks it would appear that it ran towards a gap in the north-east corner of enclosure B. The three enclosures identified by the earthwork survey have been shown by the resistance survey to be different in character from each other. The resistance values in the interior of enclosure B were very low, suggesting a lack of structural evidence. The interior values for enclosures C and D were much higher. Enclosure C is particu- larly interesting, as the high resistance levels may indicate structural or possibly quarrying activity. It is likely that the high resistance readings associated with enclosure D may be associated with relatively modern activity, as it is near to the present track into the field.
The Finds (Table 1)
Tabulated lists of all artefacts recovered during the pipeline salvage excavation, and the road scheme assessment, are contained in the two site archives, both of which are to be deposited in the Salisbury and South Wiltshire Museum. For the purposes of this report only the most numerous and informative cat- egory, pottery, has been analysed in detail.
THE POTTERY by LORRAINE MEPHAM
A total of 563 sherds (3770g), from the area of the shrunken settlement at Knook, has been examined. This comprises 295 sherds (2028g) of pottery recov- ered from excavation in advance of the construction of the Codford-Heytesbury feeder main (field 6). It includes a small group of sherds recovered from surface collection in an adjacent field (field 5), and a further 268 sherds (1742g), which were recovered from test pits subsequently excavated within fields 1,
82
Evaluation Test pit
10 1] 12 13 14 15 18 50 51 52
Fabric type Prehistoric Romano-British
Medieval C400 C401 C402 C403 Q400 Q401 Q402 Q403 Q404 Q405 Q406 Q407 Q408
Total medieval
Post-medieval
TOTAL
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Table 1: The finds from Knook
Burnt flint Worked flint Ceramic Building Animal Bone Materials No. Wt. No. W. No. Wt. No. Wt. 3 24g - - - - - = 1 9g - - ] 58g 3 12g = _ - - 1 584g - = 1 Sg 1 2g = — 9 38g 7 66g 2 8g 10 130g 1 lg 12 215g 6 54g 3 25g 2 2g - - 3 7g l 252g _ - - - ~ - 3 Sg 2 18g i 22g 1 6g 1 9g 21 4lg 4 78g 13 6lg _ = = = Table 2: Pottery totals by fabric No. sherds Weight Mean sherd weight 3 9g 3.0g 3 29g 9.72 20 144g 7.2g 176 964g 5.82 7 56g 8.0g 13 124g 9.5g 107 725g 6.8g 11 94g 8.5g 4] 28lg 6.9g 84 649g 7.7g 4 12g 3.0g 2 17g 8.5g 1 4g 4.0g 57 297g 5.2g 1 2g 2.0g 524 3369g 6.4g 33 363g 11.0g 563 3770g
|
}
THE SHRUNKEN VILLAGE OF KNOOK
3, 4, 5 and 6, along the route of the A36 improvement (Barnes and Allen 1990). These two groups of pottery are closely comparable, and have been considered here as one assemblage. The majority of the pottery is of early medieval date, although some post-medieval material is also present, together with a few isolated sherds of prehistoric and Romano-British material. The whole assemblage is very abraded, and this is particularly noticeable for the medieval material, which has an overall mean sherd weight of 6.4g (see Table 2). No pottery has been illustrated because of the poor quality of the assemblage and the frequent occurrence elsewhere of the vessel forms present. The assemblage has been analysed in accordance with the standard Wessex Archaeology recording system (Morris 1991). The pottery was divided into 25 separate fabric types, which fall into four broad fabric groups: fabrics with calcareous inclusions (Group C), flint-gritted fabrics (Group F), sandy fabrics (Group Q), and fabrics of known type or source (Group E). The pottery is discussed by period below; total numbers and weights of sherds by period are given in Table 3. Fabric types are discussed in detail here only for the medieval period (13 fabrics); details of prehistoric, Romano-British and _post- medieval fabric types can be found in the archive.
Prehistoric
Three sherds have been identified as prehistoric: two with coarse (<1.5mm) calcareous inclusions, prob- ably chalk, and one with coarse (<1.0mm) flint inclusions. All three are small, featureless body sherds. On the basis of comparisons with other sites in the area, e.g. Potterne (Morris forthcoming), all are probably of late Bronze Age date.
Romano-Bnitish Three Romano-British sherds were identified: two
_ coarseware sherds of unknown source, one from a | flanged bowl of 3rd-/4th-century date, and the base of
a New Forest colour-coated beaker, also of 3rd-/4th- century date.
Medieval
Thirteen fabric types were identified as medieval. Fabric totals are given in Table 2. The vessel forms present were largely limited to cooking pots, similar to those found at other Wiltshire medieval settlements
_ such as Gomeldon (Musty and Algar 1986, 151).
Q400. Hard, coarse fabric; moderate to common (10-30%), poorly-sorted quartz grains <2mm; sparse (3—10%), poorly-sorted, sub-
Q401.
Q402.
Q403.
Q404.
Q405.
Q406.
Q407.
Q408.
C400.
C401.
C402.
C403.
83
angular flint <2mm; sparse mica and black iron oxides; rare (1—3%) fragments of carbo- naceous material, generally burnt out. Hand- made; irregularly fired orange-red/dark grey. Hard, very coarse fabric; common (20-30%), poorly-sorted quartz grains <1.0mm. Hand- made; firing as Q400.
Hard, coarse fabric; common, fairly well- sorted quartz grains <0.5mm. Handmade; firing as Q400.
Hard, coarse fabric; moderate (10-20%), poorly-sorted quartz grains <0.5mm; rare subangular flint <1.0mm; sparse black iron oxides and mica. Handmade; firing as Q400. Hard, moderately coarse fabric; moderate, fairly well-sorted quartz grains <0.5mm; rare red iron oxides and mica. Possibly wheel- thrown; oxidised orange surfaces with unox- idised core.
Hard, moderately fine fabric; common, fairly well-sorted quartz grains <0.5mm; rare frag- ments of carbonaceous material; rare black iron oxides. Possibly wheelthrown; unoxid- ised with oxidised white external surface. Hard, moderately fine fabric; common, fairly well-sorted quartz grains <0.5mm; sparse calcareous (probably chalk) fragments <0.5mm; rare black iron oxides. Possibly wheelthrown; unoxidised with oxidised buff internal surface.
Soft, fine fabric; rare quartz grains <0.5mm; sparse black iron oxides and mica. Wheel- thrown; oxidised brick red, occasionally with unoxidised core.
Hard, fine fabric; moderate, fairly well- sorted quartz grains <0.5mm; rare iron oxides. Wheelthrown; oxidised pale orange with unoxidised core.
Hard fabric; moderate crushed calcareous (probably limestone) fragments <1.5mm; sparse quartz grains <1.0mm. Handmade; firing as Q400. Generally soapy feel.
Hard, fine fabric; sparse calcareous fragments <0.5mm; sparse quartz grains <0.5mm; sparse iron oxides and mica. Handmade; firing as Q400.
Hard, moderately fine fabric; sparse calcareous (probably limestone) fragments <0.5mm; sparse quartz grains <0.5mm; rare fragments of carbonaceous material; rare black iron oxides; sparse mica. Handmade; firing as Q400. Possibly part of C401.
Hard, coarse fabric; sparse, poorly sorted
84 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
calcareous (probably chalk) fragments <1.0mm; moderate, poorly-sorted quartz grains <1.0mm; rare’ subangular flint <2.0mm; sparse mica. Handmade; firing as Q400.
None of these fragments is of known source, although it is likely that all are of at least fairly local origin. The nearest possible source is Crockerton, to the south of Warminster. At present only one kiln of 16th-century date has been discovered (Algar 1968), but documentary evidence suggests that potters were working there from at least the 13th century (McCarthy and Brooks 1988, 334). If it can be assumed, although this is not necessarily the case, that the most local source or sources will be represented by the most numerous fabric types within the assem- blage, then any, or all, of the fabrics Q403, Q400 and C401 (19.3%, 21.5% and 28.6%, respectively, of the total medieval assemblage by weight) could derive from this source.
Calcareous-tempered fabrics are widespread in the West Country, and are likely to derive from a number of different sources. One possible source has been suggested in the valley of the Bristol Avon near Bath (Vince 1979, 28). The sandy fabrics Q401 and Q402, frequently scratchmarked, are similar to material found in Salisbury, which is presumed to derive from the kilns at Laverstock, or nearby (Mepham in prep. a). These kilns were in operation from the 12th century (Musty et al. 1969).
A number of the fabrics are paralleled within the medieval assemblage from Trowbridge Castle (Mepham in prep. b). Fabrics Q400, Q403, C401 and C402 correspond to fabric types occurring in 11th- to
13th-century contexts at the Trowbridge site, while fabric C400 is very similar to a fabric which occurs in pre-Conquest (l0th-/1]lth-century) con- texts, although dating evidence from rim forms in fabric C400 is ambiguous. Parallels for fabrics Q404, Q406 and Q407 are also found at Trow- bridge, in later medieval contexts (13th- to 15th- century).
Vessel forms present at Knook confirm this date range, although the range of forms represented is very limited. The majority comprise cooking pots (33 examples), both short-necked, with simple everted rims (see Musty and Algar 1986, fig. 18); and long- necked, with thickened or clubbed rims, often with ‘dished’ neck profiles (see Musty and Algar 1986, fig. 15). Rim types present would suggest a broad 11th- to 13th-century date range. It is difficult to discern any chronological development within this range, although it should be noted that the simple everted rims tend to occur in the coarser, and potentially earlier, fabric types (Q400, Q401, C400). No bowl rims were recovered, but dishes of ‘West Country’ or incurved type have been recognised from base sherds (seven examples). This type of vessel is common in 12th-/13th-century contexts in the West Country (McCarthy and Brooks 1988, 125; Jope 1952) and also in 13th-/14th-century contexts in south Wales (Sell 1984).
Jug or pitcher forms are very scarce, and are limited to one possible rim sherd in fabric Q400, two strap handles, in fabrics Q403 and Q407, and one base sherd in fabric Q407; all these are unglazed. Glazed body sherds in various fabrics, two comb-decorated, probably also derive from jug/pitcher forms. None of these is closely datable.
Table 3: The medieval and post-medieval pottery from test pits
Test pit Medieval Post-medieval No. Wt. No. Wt.
13 66 421g
14 3 5g 3 38g
15 1 5g
18 16 109g
50 31 ISlg 2 6g
51 6 35g
52 133 908g 5 37g
TOTAL 256 1634g 10 8lg
THE SHRUNKEN VILLAGE OF KNOOK
Post-medieval
The post-medieval material consists of earthenwares, both pale-firing wares of Verwood type, and red wares of unknown source; single sherds of creamware and Jackfield ware, of late 18th-century date; and fine white earthenwares of 19th-century or later date.
Discussion
The fabrics and forms present within the medieval assemblage from Knook would indicate a broad 11th- to 13th-century date range, and fit well within the range of early medieval pottery known from the region, combining elements common in West Country assemblages with others more characteristic of material to the east. A small number of sherds, principally in finer sandy fabrics, frequently glazed, may extend the date range into the later medieval period, and there is a possibility of some earlier, pre-Conquest material. The assemblage, with its emphasis on coarse domestic wares, and almost complete absence of finer glazed wares, can be seen as typical of a small rural settlement.
It has not proved possible to use the stratigraphic data to refine this date range. The absence of stra- tigraphic relationships on the site has prohibited the construction of any sort of ceramic sequence, and the very abraded nature of a large proportion of the assemblage suggests that little, if any, can be regarded as having been found in situ. The limited areas excavated and the variety of investigative techniques used prevented any analysis of spatial variation, other than a plotting of presence and absence by test-pit
_ (Figure 5).
| DISCUSSION
| The results from the excavation on the line of the | water pipeline at Knook must be regarded as disap-
pointing. Although good dating evidence for the
_ earthwork complex was retrieved, the lack of struc-
tural features indicates that the south-western area of field 6 was not intensively occupied (Figure 6). The RCHM(E) earthwork survey, the geophysical survey and the large number of artefacts from test pits suggest that an area of more intense activity lay at the
| north-eastern end of the field (Figure 5), the area
| likely to be used for the re-routing of the A36.
At the north-eastern end of the field the earthwork survey had suggested the existence of a number of small rectangular enclosures. The results of the geo- | physical survey tentatively suggest that at least one of | these enclosures contains structural remains. This survey also indicated the presence of a large linear
85
feature beneath the earthworks, which ran diagonally across the area and may have been a sunken way. Clearly, field 6 was used for settlement over a con- siderable period of time, with the present visible earthworks reflecting the later stages of this activity and masking earlier developments. Some of the earth- works represent boundaries still in use in the later 18th century. The far greater proportion of medieval to post-medieval pottery suggests that occupation in this area peaked in the 12th to 13th centuries. The abraded nature of the medieval material indicates that it did not remain undisturbed and is further evidence of the continued activity in field 6 into the post- medieval period.
To the north of the main field of earthworks, another smaller group of earthworks was identified adjacent to the main road (Figure 5). Geophysical survey revealed the existence of both negative and positive features in this area, and test pits exposed one possible structural feature. The distribution of pot- tery suggests that medieval occupation, contemporary with that in field 6, occurred in field 3 concentrated around test pits 13 and 15.
The environmental trial trench, B, to the west of the village provided information pertinent to the medieval landscape. The lack of colluvial material in the medieval period suggests that the area on the slopes of the valley was essentially being used for pastoral activity at this time. Eighteenth-century estate maps show the dry valley to the north-east of the village under open field arable, with no cultivation on the slopes or hill tops; the palaeoenvironmental evidence indicates that the same situation existed in the medieval period.
The medieval settlement of Knook in the 13th century would appear to have been larger than the present day settlement (Figure 7). Two settlement foci can be postulated, one concentrated around the present manor house and church (ST 9380 4180) and the other along the southern flank of the Warminster to Salisbury road. By at least the 13th century these two foci were joined by settlement alongside the sunken lane (Figure 5), probably on both sides (Figure 7). Expansion in the Middle Ages may have led to a westerly extension of the settlement along the present Salisbury-Warminster road. A_ recently excavated pond at West Farm produced only minimal quantities of medieval pottery, consistent with a manuring scatter, and this, when taken with the evidence of the test pits on the north side of the A36, indicates that medieval settlement was probably con- fined to the south side of the road.
Even though a settlement peak may have been
86 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Area of probable
Medieval settlement
OT 4
West Farm
Knook Horse Hill
Figure 7. Probable extent of medieval settlement at Knook
reached in the 13th century, subsequent decline was probably not continuous from that time, and despite the present shrunken state of the settlement there are likely to have been periods of expansion alternating with decline since the 13th century. Abandonment of the area, represented by earthworks in field 6, was not completed until this century. The recent shrinkage of Knook can probably be related to the pattern of settlement desertion brought about by the agricultural depression of the late 19th century, and continued through changes in rural society until the present day. Such circumstances have led to desertions elsewhere in Wiltshire as at Snap (Watts 1989, 27-30) and in the Wylye valley at Chilhampton.
Acknowledgements. The production of this report was financed by Wessex Water plc and by Wiltshire County Council. Wessex Archaeology would like to thank all the landowners for allowing access to their land, in particular, Mr S.R. Pottow of East Farm who had to endure the greatest disturbance, and Mr I. Duff of West Farm for his interest and consideration. Chris Place of Wiltshire County Council was most helpful throughout the duration of the fieldwork. The authors wish to thank Julian Richards who coordinated the
pipeline project for Wessex Archaeology; Ian Barnes, who directed the fieldwork during the road scheme assessment and undertook some of the historical research; and Mick Rawlings who directed the excavations on the pipeline at Knook. The authors also wish to thank Mike Allen who provided information on the colluvial sequence, Sue Davies, Elaine Morris and Caron Newman for their helpful comments on the text; and the report’s illustrators, Linda Coleman, Karen Nichols and John Vallender.
REFERENCES
ALGAR, D.J., 1968 ‘Wiltshire: Crockerton’ in D.G. Hurst, ‘Post-medieval Britain in 1967’, Post-Medieval Archaeol. 2, 187-9
BARNES, I., and ALLEN, M.J., 1990 Codford—Heytesbury A36 Road Improvement: archaeological evaluation (Wessex Archaeology unpub. report)
COE, D., 1990 Codford—Heytesbury Pipeline: pre- construction excavation and watching brief (Wessex Archaeology unpub. report)
COLT-HOARE, SIR R., 1812 History of Ancient Wiltshire, Vol. 1, London
GATER, J., and GAFFNEY, C., 1990a Report on Geophysical Survey: Knook, Wiltshire (Geophysical Surveys unpub. report: 90/69)
THE SHRUNKEN VILLAGE OF KNOOK
1990b Report on Geophysical Survey: Knook, Wilt- shire (Geophysical Surveys unpub. report: 90/73)
JOPE, E.M., 1952 ‘Regional Character in West Country Medi- eval Pottery’, Trans. Bristol and Gloucs. Archaeol. Soc. 71, 88-97
McCARTHY, M.R., and BROOKS, C.M., 1988 Medieval Pottery in Britain AD900-1600, Leicester
MEPHAM, L.N., in prep. a ‘The Pottery’ in J.W. Hawkes, Excavations in Salisbury 1984-89
in prep. b ‘The Pottery’ in A. Graham, Excavations at Trowbridge
MORRIS, E.L., 1991 Guidelines for the Analysis of Pottery, Wessex Archaeology Manual No. 4, Salisbury
— 1991 ‘Ceramic Analysis and the Pottery from Potterne; A Summary’ in A. Middleton and I. Freestone (eds.), Recent Developments in Ceramic Petrology, British Museum Occasional Paper 81
— forthcoming ‘The Pottery’ in A.J. Lawson and C. Gingell, Excavations at Potterne
87
MUSTY, J.W.G., ALGAR, D.J., and EWENCE, P.F., 1969 ‘The Medieval Pottery Kilns at Laverstock, near Salisbury, Wiltshire’, Archaeologia 102, 83-150
MUSTY, J., and ALGAR, D., 1986 ‘Excavations at the Deserted Medieval Village of Gomeldon, near Salisbury’, WAM 80, 127-69
SELL, S.H., 1984 “The Medieval Incurved Dish — A Topic for Discussion’, Medieval and Later Pottery in Wales 7, 63-4
TAYLOR, H.M., 1968 ‘Anglo-Saxon Sculpture at Knook’ WAM 63, 54-57
THORN, C., and F., (eds.), 1979 Domesday Book, 6, Wiltshire, Chichester
VINCE, A., 1979 “The Medieval Pottery: Fabric Types’ in J.P. Greene, ‘Citizen House 1970’ in B. Cunliffe (ed.), Excava- tions in Bath 1950-1975, C.R.A.A.G.S. Excav. Rep. 1, Bath, 27-31
WATTS, K., 1989 Snap: The History, Depopulation and Destruc- tion of a Wiltshire Village, Trowbridge
Wiltshire Archaeological and Natural History Magazine, vol. 86 (1993), pp. 88-101
Excavations at New Park Street, Devizes, 1990
by MILES RUSSELL
with contributions by MARK GARDINER, DIANA JONES, BRONWEN LAUGHLIN, MICHAEL LAUGHLIN and JON WALLIS
An excavation at New Park Street, Devizes, revealed a sequence of three buildings. No firm dating evidence was found for the earliest structure which may have been built in the 15th century. The later buildings are attributed to the early 17th and mid-18th centuries. A number of post-medieval features, including a well and an icehouse were
also examined.
INTRODUCTION
In 1990 the Field Archaeology Unit of the Institute of Archaeology (University College London), commis- sioned by Wiltshire County Council, undertook an excavation at New Park Street, Devizes, in advance of construction of a link road. The excavation was intended to assess the level of settlement within the medieval town (Figure 1). The site, which measured approximately 40m by 40m by 16m, fronted New Park Street and lay in an area of the medieval town known at least from the 14th century as the ‘Old Port’ (Figures 1-3). The area available for investigation had, until recently, formed part of a garage forecourt.
Excavations had previously been undertaken on the site in 1986 by the Wiltshire Rescue Archaeology Project, whose trial trench, numbered here 66, appar- ently located part of a pit (context 31 on Figure 4) containing 12th- or 13th-century pottery. The full results remain unpublished (Currie 1986), though the trench plan is reproduced here as Figure 4.
The Excavations
The overburden from the area of excavation was stripped by machine and the features revealed were then excavated by hand. The major features or groups of features are described first in numerical order and then the development of the site is discussed.
Pit 51 (fills 52 and 251; Figure 2): a subrectangular pit cutting post hole 274 and partially overlain by a brick pillar base (82). Fills of both features contained pig bones and small quantities of 19th-century pottery.
Well 105 (fills 104 and 298; Figures 3 and 5): a subcircular chalk- and sandstone-lined well with an
average internal diameter of 1.3m. The fills of the well, containing late 18th-century rubbish deposits, were traced to a depth of 1.7m. Two sherds of l6th-century pottery were recovered from the con- struction trench (284).
Pit 141 (fills 90 and 142) and Pipe 146 (Figure 2): a square, brick-lined feature, possibly a cess pit. This was cut by pits 225 and 322. No datable finds were recovered from the remaining inner fill (142) or from the fill of the feeder pipe (146), though both are stratigraphically earlier than feature 264. The nature of the brickwork would suggest a 17th-century date. Pit 164 (fills 163, 226 and 265; Figure 2): the internal fill of this subsquare feature had been divided by a rough, unmortared brick and sandstone wall (226) aligned north-north-east to south-south-west. The northern half of pit 164 had been infilled at the time of the wall construction with a compact mass of green- sand (265). The southern half may have been left open for some time. The exposed face of the rough unmor- tared wall was supported apparently by a series of wood stakes. The purpose of this feature is unclear. It went out of use in the 17th century when the southern half was infilled with rubbish.
Pit 168 (fill 169; Figure 3): a large oval feature, partially overlain by walls 103 and 109 and brick drainage gulley 110. The foundations of 109 had destroyed all but the final 50mm of stratigraphy within the pit. Thirteenth-/fourteenth-century pot- tery was recovered from wall 169.
Pit 225 (fill 224; Figure 2): a straight-sided, rectangu- lar feature cutting the brick lining of pit 141. This pit was traced to a depth of 1.5m, but not to its bottom. The fills contained 18th-century rubbish. It may have served as a cess pit.
EXCAVATIONS AT NEW PARK STREET, DEVIZES
89
1990 —" Excavations
SHEEP ST
[| Ditch, probable line
e00m [| Bank, probable line
J
| Figure 1. Plan of Devizes, showing the location of the excavation and the probable lines of the banks and ditches of the castle
‘Pit 234 (fill 233; Figure 3): a shallow feature contain- ‘ing 13th-/14th-century pottery.
Pit 246 (fill 245; Figures 3 and 5): a subsquare feature {cutting pit 250 and partially removed by the later post hole 260. The fill contained 16th-/17th-century pot- ‘tery and four pieces of heavily corroded iron.
Pit 250 (fill 249; Figure 3): a rectangular-shaped feature disturbed by pit 246 and wall 120. Fill 249 produced small quantities of 13th-/14th-century pot- tery and 10g of possible copper alloy slag.
\Post hole 260 (fill 258; Figures 3 and 5): a subsquare
feature cutting the northern half of pit 246. The nature of the fill suggested that it surrounded a central post about 0.4m in diameter, the post evidently having rotted in situ. Seventeenth-century pottery and fragments of unworked sandstone were recovered from the post hole.
Icehouse 264 (fill 263; Figure 2): a brick-built icehouse or icewell set within a foundation trench (144). This was exposed in 1964 during the construction of the Wadham-Stringer garage. At that time the icehouse was infilled with concrete removed from the adjacent
youan Jo Jjey usaisea jo ueyd :joang yIeg MON °7 eNsty
9JAAIUO-)
CJ auolspuvg FF y ig] L am
a) youg
=~ = Ae 2 Ce
7 has
iO S Os y
S = d
tes roses) yet, ae
alt ‘= Na a % FAY.
3. Gnwit (Y) avaty ve vy wae we
a ei EY Ve vwwwewrvyeywwewyvyvyvy ¥ ~~
bi
vvvvy Repeat
094
A ae eee ee A
yess = — Sey al ta? ee See - == ich sSz4
fo) © ivi “anal
91
# LASYLS WYVd MAN 0} WO?E
EXCAVATIONS AT NEW PARK STREET, DEVIZES
Youan Jo Jyey usolsam Jo ued sloamg yeg MON ‘¢ an3ty
. Loe =
> od
> ae oS We L ie} iq Y sles Prats os
pur Y) BATTS BD eens
erect. ALLE TT Boo SF
TTT ry
bbs.
Lend |:
YY RyevVVy Ve vVvwvewvvyyy
eee Nyisssweees 3 a CT rc
: BI aoe
ores fore A 824 | ~ 4 924 =
Gel
92 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Bate aN Zr A
Pre-excavation
N KE ISCTr ae = {3 O s 30 35 0) Y/N Ea AN
Post-excavation
Figure 4. New Park Street: excavation plan of trench 66 by
Wiltshire Rescue Archaeology Project, 1986 (reproduced
with permission of Wiltshire County Council, Library and Museum Service)
1930s cinema, which lay on the site now covered by Chantry Court. It was noted then that the icehouse was bottle-shaped with an aperture 1.5m wide and an average internal diameter of about 3.5m. It was traced to a depth of about 6m, but not to its floor. Pottery recovered from the fill of the foundation trench dates to the mid-17th century.
A second brick-built icehouse was located in 1986 during the construction of Chantry Court 15m to the north-west (Newton pers. comm.). This was of simi- lar construction to the icehouse here, but was traced to a depth of 12m.
Pit 276 (fill 275; Figure 2): a small subsquare feature apparently cutting pit 362. The fill produced 16th- century pottery.
Pit/Post hole 296 (fill 295; not shown on Figure 3): a small, irregular-sided feature, overlain by chalk foot- ings (102). The fill contained two sherds of 13th-/ 14th-century pottery.
Pit 319 (fill 318; Figure 3): a shallow feature contain- ing small quantities of 13th- or 14th-century pottery.
The pit was badly disturbed by a 19th-century drainage channel (204).
Pit 322 (fill 321; Figure 2): a cylindrical feature traced to a depth of 2m. For reasons of safety the pit was not fully excavated. The fill contained some late 19th- century pottery. The feature was possibly a well, although it contained no evidence of a lining.
Pit 325 (fill 324; Figure 3): a shallow feature, badly disturbed by a 19th-century brick-lined drainage gulley (118). The fill 324 contained 13th-/14th- century pottery.
Pi 339 (fills 92, 338; Figures 2 and 5): a large subsquare pit badly disturbed on the eastern side by a 20th-century pit (58). The fill contained chalk, brick and sandstone rubble and included pottery of 19th- century date. The density of rubble within 339 and the capping of chalk may suggest that it served as a foundation base.
Pits 20, 119, 156, 160, 169, 177, 179, 186, 188, 190, 192, 196, 200, 207, 217, 219, 221, 223, 234, 240, 242, 246, 250, 258, 260, 274, 276, 280, 286, 289, 293, 305, 309, 315, 317, 319, 322 and 325, and the pipelines 181, 244 and 255 produced material of the 18th or 19th century.
Pits 57, 58, 66 and 121, and pipelines 60, 62, 73, 81, 84, 87, 111, 113, 132, 137, 149, 153, 162, 167 and 236 contained 20th-century material. Features 58, 62, 87, 111 and 137 formed part of a garage for the 1930s cinema complex. Pit 121 and pipes 81 and 236 belonged to the garage forecourt laid out in 1964. Pipeline 60 formed part of a temporary sewer that ran the length of the site.
Two buildings were recorded in the area excavated. Two phases of work could be distinguished in the first structure.
Structure 1: phase 1 (walls 101, 102, 103, 124, 129, 300 and 432; Figures 3 and 5)
The coarse chalk walls of Structure 1 survived to a maximum of two courses high and were apparently unmortared. They were presumably the footings of a timber building. No datable finds were recovered from the trench fills, though 102 and 103 disturbed two features (169 and 296) containing medieval pot- tery. Wall 129 may have been the footings for an internal partition. Feature 115 was formed of chalk rubble from wall 103.
Structure 1: phases 2 and 3 (walls 100, 108, 139 and layer 229; Figures 3 and 5)
The building was slightly altered in the second phase with the construction of a new wall (108) of mortared,
7XCAVATIONS AT NEW PARK STREET, DEVIZES 93 N Drain 204 SS) Structure 2
WwW Posthole 250 Posthole 260 E 56 C=)
W Pit 240 Pit 238 E
237
N Well 105 S
Brick
Y Sandstone
Mortar
Kk} Cotswold Stone Tile =
i er Mla tea
Structure 1 Structure 1
third phase 100
E Pit 339 Ww
Structure 1
second phase
E Ww 108
GBT,
Figure 5. New Park Street: section drawings
eee
94 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
roughly faced sandstone. In the third phase, the plan was changed by the construction of wall 100 against wall 101 which evidently closed a gap or entrance- way. No datable finds were recovered from the foundation trench fills. Layer 229, an area of compac- ted mortar, slate, tile and chalk, may have belonged to this phase. Pottery evidence suggested an early 17th- century date.
Structure 2 (walls 90, 91, 109, 121, 134 and 136; Figures 2, 3 and 5)
The building subsequently seems to have been completely reconstructed. New walls were built and the sandstone walls 100 and 139 from the second phase of Structure 1 were partially removed to install a brick drainage gully (56). Sandstone wall 108 may also have been superseded by a new brick wall (91; Figures 2 and 5). Pit 339 and the drainage channels 110 and 118 appear to belong to this phase. Pottery from the construction trenches suggests a mid-18th- century date.
DISCUSSION
Later building activity had removed most of the medieval features within the excavated area and only six features of that period remained. It is possible that one feature (Pit 31 in Currie 1986, reproduced here as Figure 4) contained 12th- or 13th-century material, though the present writer has not examined the finds. Despite the fact that the outer semi-circle of streets, New Park Street, Sheep Street and Bridewell Street was apparently set out early in the 12th century (Haslam 1980, 64), no material earlier than the late 13th century was recovered from the 1990 excava- tions. It is unfortunate that the one area that may have contained evidence of 12th-century settlement, the area fronting New Park Street and Monday Market Street, was completely destroyed in 1964 by the construction of a petrol storage tank, pit 135 (Figure 3).
The chalk walls of Structure 1 are the earliest building recorded. No dating evidence was obtained either from the walls or their foundation trenches, although one wall clearly overlies two 13th-/ 14th-century pits. The house which today lies immediately south of the excavated area, Great Porch House, has been dated to the early 15th century (Slocombe 1990, 31) and it is possible that the excavated footings of the first phase of Structure 1 might be contemporary with it.
By the mid-18th century the original ground plan of the building had been extensively modified. A brick-
lined icehouse and a series of cess pits had also been cut into an area to the north. Before its demolition in the early 1930s, Structure 3 had operated as a bakery (Slocombe pers. comm.). A ground plan in the pos- session of the Wyvern Partnership, Devizes archi- tects, shows the bakery shop as a single room fronting directly on to New Park Street. This room was largely destroyed in 1964 by cutting pit 135. Directly behind the bakery shop was the living area (walls 109, 134 and 136) and the bake house. The bake house ovens were situated over the brick rubble foundation base 339. The garden area beyond contained an ornamen- tal brick arch represented in the 1990 excavations by the pillar bases 82 and 83 (Figure 2).
The Finds
THE POTTERY by MARK GARDINER (Figures 6 and 7)
The pottery was divided into fabric types by naked eye and with the use of a hand lens when necessary, and was quantified by sherd number, weight and esumated vessel equivalents (EVEs) (Orton 1975). Fabrics were initially recorded by code letter and were subsequently compared with material in Devizes Museum and, where possible, then given common names. Some of the common names are those used by Alan Vince. They have been ascribed to the fabrics here by comparison with sherds labelled by him in the Museum. Full descriptions are given in the site archive in Devizes Museum.
The assemblage as a whole comes from very few sealed contexts and comprises a large proportion of residual material. The fabric types are listed and, where necessary, are described below. The fabric names are followed by three figures which indicate the total number of sherds, their total weight in kilograms and estimated vessel equivalent.
The dating of the pottery fabrics described here is not discussed in detail. For many of the medieval and some of the post-medieval fabrics the dating must await the publication of better stratified sites.
Medieval Fabrics
Fabric L (4, 0.7, 0.1): black faces and core with a slightly laminar fracture with 2 to 5 per cent unsorted angular, often rather platy, light-coloured flint up to 2mm across.
Fabric K (3, 0.2, 0.1): black or dark grey faces and mid grey core and margins. Distinguished by a
EXCAVATIONS AT NEW PARK STREET, DEVIZES
temper of 0.5 to 2 per cent unsorted sub-rounded quartz up to 0.5mm across and 0.5 per cent com- minuted shell.
North Wiltshire Oolitic Limestone-Tempered Ware (76, 1.86, 0.3). This fabric has also been called Selsley Common Ware after a site where it was first recog- nised (Dunning 1949) and Minety-Type Ware from the kiln which was one of the sources of production (Musty 1973). It seems to have been produced from the 12th to the early 16th centuries (Vince 1983a, 126, 130; Vince 1985, 56). Two variants were distin- guished, Fabric A, which had larger inclusions of limestone up to 1.0mm across, and Fabric J which had no limestone greater than 0.5mm.
South-East Wiltshire Ware (71, 0.70, 0). One sherd has applied bands and sgraffito decoration. Two variants of this were distinguished, Fabric B with about two per cent sub-angular quartz sand grains and Fabric Q, which was more sandy with 5 to 10 per cent quartz grains.
Bath A (Vince 1979, 27-8) (47, 0.82, 0.7). This fabric has also been called Cheddar J (Rahtz 1979, 317). Vessels in this fabric are often covered with a dark olive-green exterior glaze. Sherds are sometimes decorated with sgraffito wavy lines and thumbed bands around the neck. Common forms are jugs and pitchers.
Fabric P (27, 0.30, 0.1): buff faces and mid grey margins and core. Slightly sandy fabric with 5 to 10 per cent angular or sub-angular grains of quartz about 0.1mm across. This fabric compares quite closely with Bath A and may be a variant of it.
Newbury Fabric 3 (Vince) (14, 0.08, 0.1). Vessels in this fabric are rarely glazed.
Nash Hill B (McCarthy 1974, 107) (32, 0.19, 0.08). This was the most common of the Nash Hill fabrics. Vessels are usually glazed on the exterior with a light sreen glaze. Common forms are jugs.
tam Green Ware (Barton 1963; 1967) (11, 0.06, 0). 3arton has also reported finds of Ham Green Ware iearby at Bratton.
vabric I (42, 0.64, 0.6): mid to dark grey faces and jore. A hard fabric which has laminated in some herds revealing a temper of 5 per cent or more transparent quartz grains up to 0.25mm and up to 0.5 yer cent fragments of organic material.
ost-Medieval Fabrics
urrey-Hampshire Border Wares (24, 0.10, 0.3). ‘ustercian Ware (Brears 1971, 19) (13, 0.04, 0). For ating and finds in Gloucester, see Vince 1983b, 132. tonewares: Raeren (24, 0.40, 0.8); Frechen (4, 0.10, » Westerwald (7, 0.05, 0.2); Nottingham (1, 0.01,
a5
0); London source (3, 0.03, 0); Staffordshire white- glazed stoneware (58, 0.26, 0.5); and later stonewares (3, 0.02, 0).
Tin-glazed Earthenware (34, 0.11, 0.4).
Local slipwares (64, 0.46, 0.2). All the slipwares were in a similar fabric, and some had been decorated with trailed slip and sgraffito. The slip was covered with a glaze which appears yellow over the slip or a light green glaze. A low bowl in Devizes Museum (acc. no. 43.1972), found at Chippenham Library, in the same fabric has a sgraffito decoration and the date ‘1703’ inscribed on it.
Staffordshire Combed Slipware (21, 0.20, 0.4). Staffordshire Earthenware (6, 0.01, 0).
Fabric H (445, 7.10, 5.2): either oxidised mid orange or reduced mid grey with faces; margins and core a combination of these. Fine, very hard fabric temp- ered with quartz grains up to 0.lmm across. Two variants may be suggested by the glaze which is either dark olive green with large orange patches and slight brown flecking, or dull dark green. A broad range of forms is represented including bowls, plates and jugs.
Fabric R (21, 0.30, 1.2): a medium hard-fired earth- enware distinguished by its bright orange-red colour and slightly coarse fabric. The pottery is often glazed dark green or brown.
Fabric S (4, 0.06, 0.1) has black surfaces and core, and light grey-brown margins. It is tempered with very fine white sand quartz with platy dark orange- brown inclusions possibly of ironstone.
Fabric D (96, 3.46, 2.0): light buff faces and margins, and light pink core with occasional fragments of bright pink grog. White sub-rounded unsorted quartz grains up to 0.25mm. On internal surfaces this fabric often has a light yellow-green or green-yellow glaze with occasional brown flecks and sometimes, on the exterior, a thin colourless glaze. This is similar to products from the Verwood kilns in Dorset (Young 1979),
A slightly coarser version of this fabric was distin- guished by the inclusion of translucent sub-rounded quartz up to 0.25mm. This was termed Fabric M (153, 3.70, 1.8).
Pancheons and bowls are common products in Fabrics D and M.
Creamware (91, 1.08, 2.4).
Porcelain (3, 0.01, 0).
Basaltes (3, 0.02, 0.1).
Local Red Earthenware (84, 0.14, 0.4): hard, well- fired earthenwares, usually glazed internally with a mid brown glaze with occasional black flecks. Transfer-Printed China (55, 0.59, 1.4).
96 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE |
Illustrated pottery (Figure 6)
1. Wheel-thrown cooking pot. Fabric L. Context 168.
2. Foot from large, coil-built vessel in North Wilt- shire Oolitic Limestone-Tempered Ware. The interior is glazed and exterior decorated with two applied bands, one of which is thumbed and the other rouletted. 12th/early 13th century. Context 54.
3. Wheel-thrown cooking pot. South-East Wiltshire Ware. Context 168.
4. Tin-glazed earthenware bowl, with dark blue decoration, possibly made at Brislington (cf. Price and Ponsford 1979, 46-7, fig. 21, nos. 4, 5). Foot-ring is pierced for suspension. 17th/early 18th century. Context 201.
5. Tin-glazed earthenware bowl with dark blue chi- noiserie foliage. 18th century. Contexts 63, 74 and 301.
6. Rim of bowl decorated with an applied band and dark green glaze inside and out. Context 52.
7. Pancheon with orange-yellow glaze internally. 18th/19th century. Context 52.
CERAMIC BUILDING MATERIAL (Figure 7) by MARK GARDINER
Almost all of the ceramic building material was
medieval in date and was discovered as residual finds
in later contexts. The greater part of the finds was glazed tile. Four fabrics were distinguished:
Fabric 1. North Wiltshire Oolitic Limestone Temp- ered Ware.
Fabric 2. Fabric very similar to Fabric 1, but without the limestone temper, and with occasional inclusions of angular chert and flecks of iron ore.
Fabric 3. South-East Wiltshire Ware.
Fabric 4. Similar to Fabric 1, but without either the limestone, or the angular chert temper of Fabric 2.
The amount of building material recovered was limited and no quantification was attempted. Some clear trends were apparent from the examination of the material. Almost all the roof tile fragments were glazed and a number of the pieces were of uneven thickness. No peg holes or nibs were recorded and ridge tile fragments decorated with cocks-comb were very common.
The proportion of glazed fragments among the roof ules is unusual: roof tiles were rarely glazed and when they were, usually only the lower part of the tile, which showed beneath the overlap, was treated.
Ridge tiles on the other hand were commonly glazed. These observations may suggest that all the fragments of tile were from ridge pieces, a view supported by the absence of any peg holes or nibs, which would have been necessary to secure roof tiles. The roofs were presumably covered with the stone tiles described below.
Excavations at the Minety kiln recovered knife-cut cocks-comb ridge tiles similar to those at Devizes (Musty 1973). A feature of the Minety tiles was the slashes applied at the base of the cocks-comb and the same feature is found on many of the Devizes tiles, both those in Fabric 1 (e.g., Figure 7: 8 from context 154), which might have been made at Minety, and in other fabrics.
Other products represented in the building material were glazed floor tile (two undecorated pieces), oven tile (one piece) and a post-medieval pantile (one piece). Four fragments from two tin-glazed tiles were also recovered (Figure 7: 9).
METALWORK (Figure 8) by JON WALLIS
All the metalwork examined was of post-medieval
date. The more significant finds are discussed below.
10. Copper alloy globular-headed pins from the fill of foundation trench 68, possibly residual. 48 similar pins were recovered from this context.
11. Copper alloy pin head formed by wrapping a short wire strip around one end of the shaft. Sixteenth-century. Fill of pit 276.
12. Large globular-headed copper alloy pin. Sixteenth-century. From construction trench 284 for well 105.
13. Post-medieval copper alloy spectacle buckle. The tongue is missing. Traces of gilding evident on buckle face. Unstratified.
14. (Not illustrated) large, heavily corroded post- medieval copper alloy spectacle buckle with tongue. Pit 141. Residual.
15. Copper alloy thimble found within foundation trench 175 for brick-lined drainage gully (110). Possibly residual.
16. (Not illustrated) copper alloy lace end, length 26mm, width 2mm. Found within foundation trench 175 for brick-lined drainage gully (110). Possibly residual.
17. (Not illustrated) folded lead window came. Possibly medieval. Found within makeup of wall 108.
18. (Not illustrated) lead musket ball with casting flash. Diameter 18mm. From fill of 18th-
EXCAVATIONS AT NEW PARK STREET, DEVIZES 97
O 5 10cm
7
Figure 6. New Park Street: pottery: 1-3 medieval and 4-7 post-medieval. Scale 1:4
Figure 7. New Park Street: medieval and post-medieval ceramic building material. Scale 1:4
98
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Figure 8. New Park Street: post-medieval metalwork and clay pipes. Actual size
EXCAVATIONS AT NEW PARK STREET, DEVIZES
32 34
fe) 1
— Wy yy
| Figure 9. New Park Street: clay pipes. Actual size
century brick-lined drainage gully (56). Residual.
loin (not illustrated) George I. Irish copper halfpenny, 1723. Context 14.
|
CLAY PIPES (Figures 8 and 9) by BRONWEN LAUGHLIN
A considerable number of clay pipes were recovered during the excavation. The pipes were catalogued and examples are illustrated and discussed below. 19. Part of a bowl and stem; the heel is stamped with the initials ‘C B’ and is highly decorated. Its date and maker are unknown. Context 201. 20. Part of a stem with the heel stamped with a hand or glove. It was possibly made by Hugh Gaunt- let who was working in the 17th century, although it may be an imitation. Context 201. 21, 22, 23. Parts of three bowls with the heels stamped with a hand or glove, more elongated than the other examples. One example seems to show joints in fingers and design on the palm. Possibly made by Hugh Gauntlet, or perhaps an imitation. Contexts 171 and 201. 24,25. Part of a bowl and part of stem stamped ‘R G’ with a fleur de lys above the letters. The
26.
die
28.
29.
30.
315
99
an pyr
35 36
maker was Richard Greenland who was work- ing c.1660—80. Context 201, 204, 220. A bowl and part of stem, stamped with the initials ‘“G H’, with a fleur de lys above and below the letters. The stamp is recognised, but its origin is unknown. Context 201. (Not illustrated) part of a bowl; the heel is stamped with the mark of Nathaniel Howell who was working c.1640—50. Context 201. Part of a stem and bowl; the heel is stamped ‘JEFFRY HUNT”. The ‘N’ is reversed and is identified as being one of the earliest Jeffrey Hunt stamps. The earliest known stamp is dated c.1640. Context 201. A bowl with the heel stamped ‘JEFFRY. HUNT’. This is the most common Jeffery Hunt stamp and was widely distributed. This, with 23 other examples was found in context 201; one came from context 86; one from context 140; and one from context 108. A bowl with the heel stamped ‘JEFFRY.H’ for Jeffery Hunt and made c.1660—70. Context 173. 32. A bowl stamped on the heel ‘JOHN HUNT” with dot decoration above and below the wording. Three pipe makers all named John Hunt were working in Bristol in the second part of the 17th century. Context 201.
100 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE .
33. A bowl with heel stamped, possibly ‘K H’. Its maker and date are unknown. Context 86.
34. Part of a bowl with the heel stamped ‘T H’, a fleur de lys appears over the top of the letters. It is known that Thomas Hunt was working in Marlborough from at least 1667-96. Context 201.
35. Part of a stem. The heel is stamped with the mark of Thomas Hunt c.1640—96. Context 201. Another example was recovered from context 171.
36. Part of a stem; the heel is stamped with what could be a thistle, but the maker and date are unknown. Context 201.
THE ANIMAL BONE by DIANA JONES
All the bone from the New Park Street excavations was highly fragmented, reflecting the high level of modern disturbance of the medieval and_ post- medieval features. Pit 51 (19th-century) produced the only sizeable quantity of identifiable skeletal material.
ie) 5 10cm
38
41
A high proportion of pig metacarpals (n=96) and metatarsals (n=42) were recovered from fill 52. Such figures may reflect on-site butchery practices or the hoarding of bone material for later working.
WORKED STONE (Figure 10) by MICHAEL T.L. LAUGHLIN
Most of the rocks seen from Devizes were from
Jurassic strata which would have been quarried
locally. Two pieces were somewhat different. A
worked block of grey limestone may have been
imported from the Portland area. The whetstone is a
piece of siliceous grit, which may have been imported
from the Pennine area or from South Wales.
37. (Not illustrated) whetstone. A worked sample of a massive, fairly fine-grained grit with flecks of mica (muscovite). The slight reddening where weathering has occurred suggests that some iron-bearing minerals may be present. The sili- ceous nature and hardness would suggest that this may be Millstone Grit, possibly imported from the Pennine region or South Wales. This
39
40
Figure 10. New Park Street: worked stone. Scale 1:4
EXCAVATIONS AT NEW PARK STREET, DEVIZES
sample is of Carboniferous sediment. Context 96.
38. Oolitic limestone mortar. This dense grey rock contains fossil fragments and is Jurassic in age. Context 115.
39. Stone roof tile with bored hole. This consists of a shelly, sandy limestone with traces of oolitic structure. The fossils are very badly frag- mented, but are mainly lammelibranchs and brachiopods. These are consistent with those found in rocks of the Inferior Oolite series. Context 201.
40. Stone roof tile of a shelly, sandy limestone, probably from the Inferior Oolite series of Jurassic beds found locally. Context 263.
41. A worked piece of pale grey, fine-grained lime- stone which may have been imported from the Portland region of South Dorset. Context 201.
Acknowledgements. The assistance of Wiltshire County Council and particularly of Chris Place in organising the excavation is gratefully acknowledged. I am also appreciative of help from Mrs Pam Slo- combe, Dr Paul Robinson, Mr P.Newton, Mr H. Spercely and Sean Tye. Permission to reproduce Figure 4 was given by Wiltshire County Council Library and Museum Service. The publication drawings were prepared by Jane Russell. Mark Gardiner wishes to thank Clive Orton for his help in identifying the stonewares. The finds and site records have been placed in Devizes Museum, acc. no. 1990.24.
REFERENCES
| BARTON, K.J., 1963 “The Medieval Pottery Kiln at Ham | Green, Bristol’, Trans. Bristol Gloucestershire Archaeol. | Soc. 82, 95-126
| —— 1967 ‘A Note on the Distribution of Ham Green Pottery’, Trans. Bristol Gloucestershire Archaeol. Soc. 86, 201-2
101
BREARS, P.C.D., 1971 The English Country Pottery: its History and Techniques, Newton Abbot
CURRIE, C.K., 1986 Devizes: Wadham-Stringer 1986. An Interim Report (Wiltshire County Council)
DUNNING, G.C., 1949 ‘Report on the Medieval Pottery’ in H.S. Gracie, ‘Hut Sites on Selsley Common, Near Stroud’, Trans. Bristol Gloucestershire Archaeol. Soc. 68, 30-44
HASLAM, J., 1980 ‘The Excavation of the Defences of Devizes’, WAM 72/3, 59-66
McCARTHY, M.R., 1974 “The Medieval Kilns at Nash Hill, Lacock, Wiltshire’, WAM 69, 97-145
MUSTY, J.W.G., 1973 ‘A Preliminary Account of a Medieval Pottery Industry at Minety, North Wiltshire’, WAM 68, 79-88
ORTON, C.R., 1975 ‘Quantative Pottery Studies: Some Pro- gress, Problems and Prospects’, Science and Archaeol. 16, 30-5
PRICE, R., and PONSFORD, M., 1979 ‘Survey and Excavation Near St Peter’s Churchyard, Bristol, 1972’ in N. Thomas (ed.), Rescue Archaeology in the Bristol Area 1, City of Bristol Museum and Art Gallery Monograph 2, Bristol, 35-48
RAHTZ, P.A., 1979 The Saxon and Medieval Palaces at Cheddar B.A.R. British Ser. 65, Oxford
SLOCOMBE, P., 1990 Vernacular Architecture Group: Wiltshire Spring Conference Booklet 1990, 31-7
VINCE, A.G., 1979 ‘The Medieval Pottery’ in B.W. Cunliffe (ed.), Excavations in Bath, 1950-75, Committee for Rescue Archaeology in Avon, Gloucestershire and Somer- set, Excavation Report 1, Bristol, 27—31
1983a ‘The Medieval Pottery’ in C.M. Heighway, The
East and North Gates of Gloucester: Excavations 1974-81,
Western Archaeological Trust Monograph 4, Bristol,
125-31
1983b ‘Post-Medieval Pottery’ in C.M. Heighway The
East and North Gates of Gloucester: Excavations 1974-81,
Western Archaeological Trust Monograph 4, Bristol,
131-40
1985 ‘The Ceramic Finds’ in R. Shoesmith, Hereford City Excavations 3: The Finds, C.B.A. Research Report 56, 34-82
YOUNG, D., 1979 ‘The Verwood Potteries’, Proc. Dorset Natur. Hist. Archaeol. Soc. 101, 103-20
Wiltshire Archaeological and Natural History Magazine, vol. 86 (1993), pp. 102-112
A Survey of the Parish Church of St Swithun at Compton Bassett, Wiltshire
by ANDREW REYNOLDS
The building and churchyard were surveyed in order to produce scale plans. As a result, a previously unrecognised wall of possible Anglo-Saxon date was noted and seven major phases of structural alteration were identified dating
from the 11th to the 19th centuries.
INTRODUCTION
This paper forms part of the study of medieval settlement and land use currently being undertaken for the Compton Bassett Research Project, which is based at University College London, Institute of Archaeology. The present report will concentrate on the archaeological evidence for the phasing of the
church and yard, while detailed discussion of the documentary evidence and the function of the church in the interpretation of later Anglo-Saxon and medie- val settlement will be the subject of future work. The parish church of St Swithun lies at the extreme south-westerly end of the village of Compton Bassett (SU 031716), which is situated at the bottom of the scarp slope of the Marlborough Downs (Figure 1).
l
, pe fy NO
“ g~ VF) St.Swithuns ir <—=Church i to it
1
Figure 1. Compton Bassett: village map
THE PARISH CHURCH OF ST SWITHUN AT COMPTON BASSETT
Compton Bassett may be termed a settlement of the ‘interrupted row’ type as defined by Dyer (1990). The existence of a settlement by the Late Anglo-Saxon period is revealed by three entries in the Domesday Survey for Wiltshire (Thorn and Thorn 1979; 27-2, 32-3 and 67-63) which probably correspond to three separate manors. Earlier origins are suggested by casual finds dated to the Middle Anglo-Saxon period, including a quern stone from Freeth Farm just under lkm to the north-east of the church.
John Aubrey, writing in the later 17th century, was the first to mention the church, about which he wrote ‘In the church I find nothing remarkable . . .’ (Jack- son 1862, 43). Jackson (1bid., 43-44) elaborated to the extent of distinguishing between the Norman and Perpendicular work and went on to describe the fine stone rood screen that has since been the major focus of attention. He also made a note of the alterations to the church just before the large scale renovation and rebuilding of 1866 (ibid., 44).
A.E.W. Marsh, in A Aistory of the Borough and Town of Calne (1903), wrote in similar terms although, interestingly, he states that a rood gallery was still in use ‘less than thirty years ago. . .” (ibid., 305). Marsh was also the first to note that the stone screen originated from Winchester, although he gave no reference for this information.
The first attempt to interpret the development of the church was made by Ponting (1912). His observa- tions, as far as they went, were accurate, the screen receiving the most detailed description. It is interest- ing to read the Revd E.H. Goddard’s comments on | the screen’s origins, which appeared as a footnote in Ponting’s paper. He noted that ‘A curious tradition |seems to have arisen that this screen came from Winchester Cathedral’ (ibid., 431). As it happened, Goddard wrote the introduction to Marsh’s book which contained the root of the ‘curious tradition’. ‘The church was given the customarily short, but accurate, description in the Wiltshire volume of The | Buildings of England series, again with most attention ‘directed to a discussion of the screen (Pevsner and \Cherry 1975).
|
|THE CHURCHYARD
The church is approached via a sunken path which ‘runs from the north-east corner of the present ‘churchyard and is entered through a doorway in the north aisle. The path is up to 1m below the level of the ground surface in the churchyard and it seems prob- able that public access to the building has long
followed this course.
103
The present boundaries of the churchyard are mostly mid 19th-century. It is, however, possible to reconstruct the layout of earlier boundaries from the study of maps. The following sequence was estab- lished (Figure 2). The medieval boundaries may be extrapolated from Andrews and Dury’s map of 1773 and the 1901 1:10,560 Ordnance Survey map. The former shows the church standing in what appears to be a much larger yard than at present. The surviving boundaries on the north-east, south-east and south- west are, within 3m or so, probably those laid out at an early period as there is no evidence that any form of land modification has occurred in these immediate areas to facilitate significant boundary change. Mature yews are a feature of these boundaries.
The continuation of the north-eastern boundary, shown on the earthwork survey (Figure 3), appears on the ground as a sharp drop in slope of about Im. This lynchet-like accumulation of soil lends weight to the idea that it is early. The 1901 map shows two superimposed boundaries. The earlier one, formed of irregular rubble, is shown on the earthwork survey and can be seen to run around the outside of the present bounds of the church. If the line of the early boundary on the south-western side is. projected from where it disappears on the ground it can be seen to run in alignment with the north wall of the outbuild- ings associated with the now demolished Compton House. If the north-eastern boundary is also projec- ted, it follows a line which forms the rear boundary of properties fronting on to the road, including the possible priest’s house directly to the north of the church which is certainly late medieval and it may be reasonably presumed that its boundaries are of the same date. It may also be argued that the presence of the churchyard restricted these boundaries.
Andrews and Dury’s map was drawn up about a century after the construction of the stables for Compton House, dated to 1665-70 by Pevsner (1975, 189). The map shows the stables standing in open ground within the possible early bounds of the church. A division between two properties fronting the road, when projected along its south-westerly course, aligns with the western end wall of the stables, indicating the possible limit of the church’s enclosure on the north-western side.
The first edition of the one inch to one mile Ordnance Survey map, published in 1828, shows that access to the church was possible from Compton House, presumably by way of steps down the sharp slope of the south-western boundary. The course of this former entrance to the church shows clearly on the earthwork survey. The 1901 map indicates that by
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE |
104
Nasseg uojduro’) 1B preAyoINyo ayi jo satiepunog ay} jo juauIdojaaap oyy “7 2aN3Iy
sajqeys
JeAaipa
asnoH Yesseg uojdwod
THE PARISH CHURCH OF ST SWITHUN AT COMPTON BASSETT
| |
| } |
i]
e yew trees
25 50m
105
TENS d
ty,
ee
iI Nl
73 EAI ee
yb SS ys Ze we = =S oS i = eS = ‘ ° s Bae
Figure 3. The churchyard earthworks at Compton Bassett
the late 19th century the area including the church and the stables had been further subdivided, with a brick-built outbuilding forming the churchyard’s
{north-western limit. The access way from Compton | House was no longer in use and the new existing
boundaries had been laid out, without doubt reusing
‘stone from the early boundary walls.
A similar subdivision of the churchyard appears to
have occurred at Yatesbury, 3km to the east. Morris
(1983, 5) briefly discussed the secular occupation of ‘areas formerly used as churchyards and points out
i 0,16 5 |that it is most frequently a post-medieval phenom-
‘enon. The evidence here seems to conform to this general pattern. The churchyard was extended to the ‘north-east c.1930 and presently encloses some 2,300 square metres.
THE CHURCH
‘The church is situated on a level platform cut into the natural slope and is orientated some 17 degrees north of east from a precise east—west alignment.
Structural Materials
The fabric of the older structures — the nave, aisles and tower — is of limestone, mainly in ashlar form. The walls of the Victorian work are of a roughly- tooled limestone, although the quoins are well- finished. Generally, a higher quality, more durable stone has been used in the earlier work.
Sarsen stone was used in the possible Anglo-Saxon wall and at the foot of the tower on the northern side. Although very little sarsen is now evident in the immediate area, it was available in the medieval period. The churches at Yatesbury and Cherhill both have towers supported, in part, by large sarsen blocks.
William Stukeley, writing in the earlier part of the 18th century, noted that there were, ‘At Compton Bassett, not far from Marlborough, westward, houses made of stones as big as those at Stonehenge, standing endwise’ (Anon. 1867, 342). (In a coombe just over lkm to the north-east of the church is a large elliptical mound which is probably a great sarsen block covered with soil.) The stone screen, in contrast, is formed of Caen stone from Normandy.
106 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
The interior of the church is plastered, apart from the tower which is whitewashed internally. The nave has a cradle roof with an external pitch of about 40 degrees. Internally the roof structure is divided into twenty-four ceiled panels with carved floral bosses at the interstices of the panels. At wall-plate level are six carved wooden heads on each side of the nave. There are four wall posts on either side of the nave, running from the wall-plates and sitting on carved stone corbels. The nave and chancel aisles have lean-to roofs; that of the north aisle is enclosed, whereas the rest have exposed beams. The chancel and porch have simple pitched roofs, about 10 degrees steeper in pitch than the roof of the nave. The roof timbers are mainly oak.
The roof coverings are of three types. The nave and chancel are covered with limestone tiles; the north and south aisles of the nave and the tower are covered with lead cut in wide strips; while the north and south aisles of the chancel are slate-tiled. The flashings are of lead, as are some of the guttering components. All other drainage systems are cast iron.
Historical Development
A composite plan (Figure 4) shows the seven phases of structural alteration visible in the present fabric of the church at ground level. The historical development of the church is discussed with reference to the Buckler painting (Figure 5) which shows a phase not evident in the surviving structure. Certain events are not shown on the phase plans (Figure 6) as they do not affect the ground plan and, as a result, they are discussed with the phase to which they are most closely related in chronological terms.
Pre-Phase I (? Anglo-Saxon)
A substantial stone wall, formed with irregular sarsen stone rubble set within a soft silty mortar, was discovered in the 19th-century coal cellar to the north-east of the chancel. Although the wall is in alignment with the wall of the north aisle, it is unlikely to have formed part of any structure repre- sented by the surviving building. It would seem highly probable that the wall was encountered by Victorian labourers while excavating soil for the con- struction of the cellar, then left, as its removal would have been both difficult and time-consuming. It is highly unlikely that any other stone structure would have been erected on the site so close to an existing church. There is no evidence to suggest that the Phase 1 chancel was ever extended and, in any case, the wall is poorly situated to support such a claim. An earlier Norman church on a slightly different alignment is an
outside possibility but most unlikely. Where churches have been excavated, it is commonly the case that Anglo-Saxon structures have been completely demol- ished to make way for a new building. It is not unusual for the new building to be built without respect for the situation of the old as at St Mary’s, West Bergholt, Essex and St Paul-in-the-Bail, Lin- coln (Turner 1984; Gilmour 1979). On these grounds, a possible Anglo-Saxon date is proposed.
Phase | (late 11th or early 12th century)
This phase consisted of a two-celled masonry church. Parts of the nave survive at either end and possibly above the arcades, although there is no evidence for the position of windows and doors. The chancel is shown on the 1806 Buckler watercolour which, in terms of its portrayal of surviving features, can be presumed to be reliable. Although no Romanesque windows are evident in the chancel, its date is con- firmed by the round-headed doorway in the north wall and the clasping buttresses at the eastern end. Surviv- ing stonework of this phase suggests that well-coursed rubble formed the walls, with ashlar blocks used as quoin stones. The painting gives some idea of the appearance of the church as a whole at this time.
Phase 2 (late 12th century)
The north aisle, divided into three bays, was added, with half piers at either end of the arcade set against the stubs of the north wall of the nave. The piers are circular with square capitals. The base mouldings are closely matched by those in the nave at Framlingham, Suffolk (Rigold 1977, 125, No. 216), which are classified as mature forms of the Attic revival and are dated to the last quarter of the 12th century (ibid., 129). The original arches of the arcade have been replaced, while the two westernmost capitals are unfinished. The trumpet scallops, although formed, are not decorated. Examples of unfinished work are not common and their interest lies in that they show in situ working of architectural detail (zbid., 100).
A transitional date may be assigned to the arcade, on the grounds that the trumpet scallop decoration of the two easternmost capitals can be exactly paralleled by that on circular capitals of early Gothic style at St Lawrence’s church, Hilmarton, 4km to the north. At both churches the stone carving has been poorly executed with similar characteristics, very probably by the same mason. There is no other surviving masonry of this phase.
Phase 3 (mid 13th century) The south aisle was constructed in the same manner
107
THE PARISH CHURCH OF ST SWITHUN AT COMPTON BASSETT
Nasseg uoiduroy ‘ysinyD s,unyIMS 1§ Jo ued sytsoduiog “p a1nsty
ape jjem yesnjoeluoo - --- UJCLO PIN W USO PIN ~ yztoor YY wor E 4 Sto BES | yzroAuea /ylro ey ee
uyblO aa uoxeg-ojbuy ¢f
N
TL LASSVd NOLdWOD HOYNHO ae ne
108 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
ee
Figure 5. View from the north-east showing Compton Bassett Church before restoration in 1866. (Reproduction of a watercolour
by John Buckler, 1806, WANHS Library)
as its northern counterpart. The topography of the churchyard to the south of the nave necessitated the excavation of a quantity of subsoils so that level foundations could be dug. The capitals of the arcade are circular and are dated to the middle of the 13th century rather than earlier, by comparison with the Hilmarton capitals and the relative dating that these allow. The capitals at St Swithun’s are similar in dimension and form but do not have trumpet scallops. The aisle itself has been rebuilt completely. Above ground, no masonry of this phase has survived 27 situ. The Buckler painting shows a single lancet in the north wall of the Phase 1 chancel which was probably installed during this period.
Phase 4 (14th century) A new chancel arch was installed with mouldings in Perpendicular style. There are two orders of mould-
ings, the innermost of which rests on corbels con- sidered by Ponting (1912, 429) to be 14th-century.
Shortly after this, a newel stair was constructed. Its absolute dating is difficult on purely architectural grounds, but it clearly post-dates the chancel arch while ante-dating the Phase 5 screen which butts onto it. Presumably a rood loft existed, supported by a beam. There is, however, no surviving evidence for either feature.
Phase 5 (15th century)
The alterations in this period were substantial. The tower was the first addition and again the removal of subsoil was necessary. Unusually, the west end wall of the nave forms much of the lower part of the tower, as opposed to the more usual rebuilding of the west end of a church that occurs when a tower is added (Rodwell 1981, 63). The double-chamfered arches of both arcades are of this period. A clerestory was added with three windows on each side. The butt joint on to the tower and the presence of the ‘weathering’ of the earlier roof-line on the west end
St. SWITHUNS CHURCH COMPTON BASSETT
VII T Lise
----- conjectural walls
Figure 6. The historical development of St Swithun’s Church, Compton Bassett
110 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
wall of the nave clearly show that the tower is the earlier feature. The barrel vault roof belongs to this phase, as do the carved corbels on which the wall- posts sit. Both aisles were rebuilt in Perpendicular style, the doorway into the church was replaced and a porch added, the roof line being visible above the roof of the present porch. Subsequent rebuilding of the aisles has limited im situ masonry to that surrounding the doorway in the north aisle and the eastern part of the south aisle. The octagonal font, decorated with quatrefoils in panels, is of this period.
The screen deserves further consideration as its origin is of primary significance regarding the division of Phases 4 and 5 of the church. As the screen has been illustrated and discussed elsewhere (e.g. Val- lance 1936, pl. 72; Jones 1965, pl. 35, 4; Ponting 1912, 430-1) its architecture need not be explored here. It has been dated to the first half of the 15th century (Ponting 1912, 431) and, if made for the church, belongs to this phase. The perforation of the eastern stub of the north wall of the nave probably occurred at this point so as to allow more light to reach the screen. There are, however, a number of strong arguments for an alternative place of origin. The screen is constructed from Caen stone. The Box quarries, 20km to the west, provided first-rate building stone throughout the Middle Ages. When the difficulties and problems of long distance overland transport are considered, it seems perplexing that the oolitic freestones from Box were not used. The use of Caen stone is a characteristic of high status medieval buildings, mainly churches, along the south coast, where stone was brought in by ship in large quantit- ies. Significantly, the screen does not fit perfectly in place at the east end of the nave. If it had been built for this church, a perfect fit should have been achieved. This situation supports the notion that the newel stair was constructed for an earlier loft. Work of this standard seems somewhat out of place in the church; it would be more comfortable in a cathedral side chapel or some other high status religious estab- lishment. The suggestion of a Winchester origin was probably a local tradition recorded for the first time by Marsh (1903). There are parallels in style in Hampshire at, for example, Christchurch Priory. It is highly probable that the screen came from elsewhere, either as a gift or by purchase, possibly from a dissolved monastic house in which case it represents a l6th-century addition. It may be pure coincidence that both Winchester Cathedral and our church share the same dedication but a Winchester origin should not be ruled out. The necessary further research on this matter should start with a survey of Winchester Cathedral.
A Salisbury provenance for the screen has been proposed (Howkins 1969, 58) although no reference was given to support this claim. In the late 18th century, Salisbury Cathedral was restored by architect James Wyatt whose work has been much criticised for its severity (Vallance 1947, 80; Tatton-Brown 1989, 199). As part of this restoration, two 15th-century screened side chapels at the eastern end of the cathed- ral were demolished. Further screens were removed from the north-east and south-east transepts and the north and south arms of the great transepts (Vallance 1947, 80). It is therefore possible that the screen at Compton Bassett was installed in the late 18th century.
Limitations in time and resources have unfor- tunately meant that more comprehensive research into the origin of the screen has not been possible. It is intended that this will be rectified in a future paper.
Phase 6 (18th century)
The north aisle was largely rebuilt, as was the western part of the south aisle, where a break in the wall is visible. In both cases masonry and architectural detail from the preceding phase were reused. A date of 1767, inscribed on the exterior of the western end of the south aisle wall, may be related to its recon- struction. Round-headed windows were installed at the east end of the north aisle and at the west end of the clerestory. A new porch was built in classical style and is shown only on the Buckler painting. The painting also shows a rectangular two-light window in the east end wall of the chancel. This feature can only be dated as post 16th-century on stylistic grounds. It could have been put in before or after, but not during, the major work of this period as the style clashes with the round-headed windows.
Phase 7 (19th century)
In 1866 the church was restored by the architect Woodyer (Pevsner and Cherry 1975, 188). The Norman chancel was demolished and the present larger, aisled structure was built with the coal cellar below. The east end walls of the aisles of the nave were perforated to form continuous aisles. The tower arch was probably blocked at this point and the wooden screen taken from the rood loft and placed across the western end of the nave in front of a newly installed pew. The Buckler painting shows a diagonal buttress at the north-east corner of the north aisle. With the addition of the new chancel some rebuilding occurred at the western end of the north aisle, as the present buttress projects from the wall at an angle of ninety degrees. The study of relevant maps suggests
|
THE PARISH CHURCH OF ST SWITHUN AT COMPTON BASSETT 111
that the pathway to Compton House ceased to be used between 1828 and 1901. The newly installed pew was still accessible through the tower so it is probable that the pathway was blocked sometime after the major restoration. New windows, in Perpendicular style, were placed at the western ends of the aisles on the north and south side.
DISCUSSION
St Swithun’s probably originated as a proprietary church, although, as Morris (1983, 64) reminds us, this term ‘embraces a wide variety of types and circumstances’. The probability that three manors existed at Compton Bassett in the Late Anglo-Saxon period and the earlier Middle Ages suggests a complex system of patronage although the principal landowner was most likely to have had greatest interest in the church.
The dedication to St Swithun (d. A.D. 862) is supportive of an Anglo-Saxon foundation at Compton Bassett. Dedications to Anglo-Saxon saints have been considered by Butler (1986, 44) who places St Swithun in his first group, defined by the comparat- ively high frequency of their occurrence. A total of fifty-eight churches nationwide are dedicated to St ‘Swithun, whose feast day, 15 July, was celebrated widely in the medieval period (Smith 1987, 58). Only \three of these occur in Wiltshire, at Compton Bassett, Little Hinton and Patney. | In general terms it has been suggested that a church should reflect the economic status of its proprietors at a given time. Addyman, when describing the work of Hurst and Beresford at Wharram, noted their recog- nition that the church constituted a microcosm of the history of its associated settlement (Addyman 1976, 1). There are some flaws in these views, namely that the economic status of a church’s patrons may not be directly related to the settlement itself and that the reasons for the expansion of a church may be attribu- ‘able to the expression of individual status. The use of ustorical development as an indicator of population srowth or decline should be approached with caution. structural failure is a frequently overlooked factor vhich may explain replacement of structural elements yr contraction in plan. | The imprecise orientation of the church can be ‘xplained in a number of ways. To some extent, opographical features may account for this. As the ‘ite is on a slope, a suitable location for the first lhurch may have been selected for convenience. Jertainly in later periods the slope was modified to (ccommodate the south aisle and the tower, but these
structures were additions to an existing building and the creation of a level platform would have been unavoidable.
The time of year at which the first church was set out would have affected the orientation if the solar arc was used as a measure. Features in the landscape which were traditionally held to mark a specific orientation may have been used, although simple errors in setting out may be responsible. For all the importance attached to orientation, it can be seen that deviant churches are common and it is probably the case that alignment on a general east/west line was acceptable. At St Helen-on-the-Walls, Aldwark, York the church was orientated south-west/north-east, forty-five degrees out of precise alignment (Magilton 1980). In fact, very few of York’s churches are precisely aligned (Cave 1950, 50). Precise orientation does not seem to have been a stringent liturgical requirement.
The church is entered by a doorway on the north side, an uncommon arrangement. The reasons for this were almost certainly dictated by the layout of the early settlement. As discussed above, the area to the south of the church does not seem to have been occupied. If the site of Compton House represents the position of an early manor, then settlement to the north, east and west of the church can be envisaged.
Although seven phases of alteration have been identified at St Swithun’s, the limitations of purely above ground study of churches should be con- sidered. Excavations at St Martin’s, Wharram Percy, North Yorkshire and at St Lawrence’s, Asheldham, Essex have clearly shown that the structural history of a church may be considerably more complex than that suggested by analysis of the standing building only (Bell and Beresford 1987; Drury and Rodwell 1978).
Most frequently, phases recognised by excavation belong to the Anglo-Saxon or Early Norman periods. In this light, the sequence at St Swithun’s is fortunate in that the structural history can be traced from at least the Early Norman period and possibly the Late Anglo-Saxon period. There are a number of churches in Wiltshire that have surviving Anglo-Saxon fabric and architectural features. High status buildings, such as St Lawrence’s at Bradford-on-Avon, repres- ent the upper end of the scale for small churches. Settlements of lesser importance, however, could have had stone churches at an early date. Close to Compton Bassett such a church can be found at Bremhill.
From the construction of the Phase 1 building onward, the church follows a line of development common to many other churches. The addition of
12 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
aisles in the late 12th and 13th centuries is widely paralleled, as is the installation of a rood loft and newel stair in the 14th century. The large scale rebuilding and additions of the 15th century are probably best attributed to the wool boom which was responsible for funding much of the fine church architecture of Perpendicular style in south-western England. The cruck-framed building to the north of the church may have served as a priest’s house in the later medieval period as it has two diagonal stone buttresses, one at each corner on the northern side. It, too, may be attributed to the economic growth that many rural parishes experienced in the 15th century. The development of the majority of parish churches ceased at this point and only really began again in the 19th century, and often then only because of their poor condition.
Acknowledgements. I should like to thank Dr Ron Wilcox for teaching me much about medieval buildings and Christopher K. Currie for introducing me to their analysis. For their assistance in surveying the church and its boundaries I am grateful to Nigel Day, Alex Deacon, Anthony Hines, Robert Laver, Mark Loft and Matthew Reynolds, all of whom stood around in the cold at various points in February 1989. The illustrations were very capably produced by Matthew Reynolds. I should like to thank Judy Medrington and Gustav Milne for reading and commenting on a draft of this report at short notice and the Revd R. Butler for allowing me to survey the church. Sara Godward prepared the manuscript. The Buckler watercolour has been reproduced with the kind permission of the Wiltshire Archaeo- logical and Natural History Society.
REFERENCES
ADDYMAN, P.V., 1976 ‘Foreword’ in P.V. Addyman and R.K. Morris (eds.), The Archaeological Study of Churches, Council for British Archaeology Research Report 13, 1-2
ANON., 1867 ‘Extracts from a Common-place Book of Dr Stukeley’, WAM 11, XXXII, 341-44
BELL, R.D., and BERESFORD, M.W., 1987 Wharram: A Study of Settlement on the Yorkshire Wolds. Volume IIIT Wharram Percy: The Church of St Martin, Society for Medieval Archaeology Monograph Series No. 10
BUTLER, L.A.S., 1986 ‘Church Dedications and the Cult of Saints in Anglo-Saxon England’ in L.A.S. Butler and
R.K. Morris (eds.), The Anglo-Saxon Church, Council for British Archaeology Research Report 60, 44-50
CAVE, C.J.P., 1950 ‘The Orientation of Churches’, Antiquaries Fournal 30, 47-51
DRURY, P.J., and RODWELL, W.J., 1978 ‘Investigations at Asheldham, Essex. An Interim Report on the Church and the Historic Landscape’, Antiquaries Journal 58, 133-51
DYER, C., 1990 ‘Dispersed Settlements in Medieval England. A Case Study of Pendock, Worcestershire’, Med. Archaeol. 34, 97-121
GILMOUR, B.J.J., 1979 ‘The Anglo-Saxon Church at St Paul- in-the-Bail, Lincoln’, Med. Archaeol. 23, 214-17
HOWKINS, C., 1969 Discovering Church Furniture, Shire Pub- lications, Princes Risborough
JACKSON, J.E., 1862 Wiltshire: The Topographical Collections of Fohn Aubrey, F.R.S., A.D. 1659-70, Devizes
JONES, L.E., 1965 The Observers Book of Old English Churches, London
MAGILTON, J.R., 1980 The Church of St Helen-on-the-Walls, Aldwark, The Archaeology of York, Vol. 10/1, York
MARSH, A.E.W., 1903 A History of the Borough and Town of Calne, Calne
MORRIS, R.K., 1983 The Church in British Archaeology, Council for Briush Archaeology Research Report 47
PEVSNER, N., and CHERRY, B., 1975 The Buildings of England: Wiltshire, London
PONTING, C.E., 1912 ‘Notes on the Churches of Ashley, Berwick Bassett, Clyffe Pypard, Compton Bassett, Hil- marton, Lydiard Tregoze, Winterbourne Bassett and Winterbourne Monkton’, WAM 37, 428-31
RIGOLD, S.E., 1977 ‘Romanesque Bases in and South-east of the Limestone Belt’, in M.R. Apted, R. Gilyard-Beer and A.D. Saunders (eds.), Ancient Monuments and Their Inter- pretation: Essays Presented to A.J. Taylor, Chichester, 99-137
RODWELL, W., 1981 The Archaeology of the English Parish Church, Batsford, London
SMITH, W., 1987 ‘The Winchester Saints in the Calendar of the Wilton Psalter’, WAM 81, 57-62
TATTON-BROWN, T., 1989 Great Cathedrals of Britain, London
THORN, C., and THORN, F., 1979 Domesday Book: Wiltshire, Philimore, Chichester
TURNER, R., 1984 ‘Excavations at St Mary’s Church, West Bergholt, Essex, 1978’, in Four Church Excavations in Essex, Essex County Council, Occasional Paper No. 4, 43-68
VALLANCE, A., 1936 English Church Screens, London
— 1947 Greater English Church Screens, London
Wiltshire Archaeological and Natural History Magazine, vol. 86 (1993), pp. 113-117
A 12th-century Effigy in Salisbury Cathedral
by D.I. STROUD
The correct attribution of the 12th-century Purbeck marble effigy of a bishop in Salisbury Cathedral has long been in dispute. F.F.E. Raby, RCHM(E) and others attributed it to St Osmund (1078-99); H. de S. Shortt and the plaque in the Cathedral to Bishop Focelin (1142-84). The article examines the cases made by Raby and Shortt
respectively and concludes that the effigy 1s Osmund’s.
The 13th-century document commonly known as ‘the Register of St. Osmund’! states that on 14 June 1226 the remains of three former bishops of Salisbury — ~Osmund (1078-99), Roger le Poer (1107-39) and _Jocelin de Bohun (1142-84) — were brought down from the hill of Old Sarum and re-buried in the new | cathedral then being built in the Avon valley. It is | generally accepted that the two 12th-century effigies in bishops’ vestments, one of Purbeck and one of .Tournai marble, now in the south nave arcade of | Salisbury cathedral belong to two of these three bishops. But which two, and which is which? _ The Purbeck marble effigy has a verse epitaph cut | round the vertical edge of the slab which must clearly | be the principal evidence in determining the identity (of its subject. The last two lines state that the | individual in question had ‘dukes and nobles’ for his ancestors. This rules out Roger le Poer who was of | humble birth, leaving a choice between Osmund and _ Jocelin. F.J.E. Raby (1947)? made a strong case for | Osmund, Hugh de S. Shortt (1971)? a persuasive one i for Jocelin. Roy Spring (1979),* the Royal Commission | on Historical Monuments (England) (1980),° and D.I. Stroud (1984)° follow Raby in attributing the Purbeck tomb to Osmund. However, the attributive plaque in ‘the cathedral still (1992) labels this effigy ‘Jocelin’. No detailed examination of the conflicting argu- ‘ments put forward by Raby and Shortt has been ‘published to date. This paper aims to rectify this ‘omission and put the case for Osmund as the subject of the Purbeck effigy on a firmer footing.
| 1. W.H. Rich Jones (ed.), Register of St Osmund (Rolls Series 1883-4), Vol. 2, p. 55.
| 2. F.J.E. Raby, ‘The Tomb of St Osmund at Salisbury’, Archae- i ological Fournal 104 (1947), pp. 146-7.
| 3. H. de S. Shortt, The Three Bishops’ Tombs moved to Salisbury | Cathedral from Old Sarum (Friends of Salisbury Cathedral, | 1971).
i) 4. R. Spring, ‘Recent Discoveries in some of the Tombs in the
| | | |
The verse epitaph reads:
FLENT HODIE SALESBERIE QUIA DECI- DIT ENSIS
JUSTITIE PATER ECCLESIE SALESBER- IENSIS
DUM VIGUIT MISEROS ALUIT FASTUS- QUE POTENTUM
NON TIMUIT SED CLAVA FUIT TER- RORQUE NOCENTUM
DE DUCIBUS DE NOBILIBUS PRIMOR- DIA DUXIT
PRINCIPIBUS PROPE Q:TB’ QUASI GEMA RELUXIT
Some words are abbreviated and letters have been obliterated in a few places but the verse can be readily restored except for the last line where the correct expansion of the letters Q:TB’ has given rise to some controversy.
Literally translated, the inscription reads:
They weep today in Salisbury because the sword of justice, the father of Salisbury’s church is dead. While he was strong he fed the wretched and did not fear the arrogance of the powerful, but was a mace, the terror of wrongdoers. He took his descent from dukes and nobles; like a jewel he reflected lustre (Q:TB’) on princes nearby (in either time or place).
In his 1947 article, Raby wrote:
There can, however, be no doubt that the tomb belonged to St. Osmund (1078-1099) . . .. The
Nave of Salisbury Cathedral’, Hatcher Review 8 (1979), pp 3-14.
5. Royal Commission on Historical Monuments (England), Ancient and Historical Monuments in the City of Salisbury, Vol. 1 (London, H.M.S.O., 1980), pp. 19-20.
6. D.I. Stroud, ‘The Cult and Tombs of St Osmund at Salisbury’, WAM 78 (1984), pp. 50-54.
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Above: tomb effigy of a bishop, Salisbury Cathedral Below: detail of verse inscription, line 6
- SAN
(Photographs by J. Proctor, Salisbury Cathedral Works Department)
A 12TH-CENTURY EFFIGY IN SALISBURY CATHEDRAL
opening words of the inscription, ‘flent hodie’, show that the inscription was composed just after the prelate’s death. . . . The poet goes on to say that he derived his ancestry from ‘dukes and nobles’. . . . according to a late fifteenth- century document, the truth of which there is no reason to doubt, Osmund was son of Henry, Count of Seez and Isabella, daughter of Robert, Duke of Normandy, and sister of William the Conqueror. The document goes on to say that Osmund was lineally descended from seven Dukes of Normandy. The epitaph, therefore, correctly describes him as descended from ‘dukes and nobles’, for these were his maternal and paternal ancestors respectively. This is evidence which is really decisive in favour of the assignment of the tomb to Osmund.
Raby thus based his case for Osmund on the parallel
between the claim to descent from ‘dukes and nobles’ ‘in the verse and the statement in a late fifteenth- | century document ‘the truth of which there is no | reason to doubt’, that Osmund was the son of Henry | Count of Seez and Isabella, sister of William the Conqueror. The document in question is the Memo- | randa de Osmundo which appears in a miscellaneous | collection of documents in the records of the Bishop | of Salisbury.’ Unfortunately, it is of doubtful validity | and Dr Kathleen Edwards went so far as to say that it | is almost certainly a fairy story’, and that the persons | named probably never existed.* The Memoranda gives
ja largely fictitious genealogy of the ducal house of | Normandy and then continues:
| Et sic iste Osmundus descendebat linealiter ex nobili sanguine septem ducum Christianorum Nor- _ manniae et nepos fuit ut praescriptum est, Willelmi _ Bastardi, Angliae Conquestoris. (And thus | Osmund was lineally descended from the noble blood of seven Christian dukes of Normandy and, as stated above, was the nephew of Wil- liam the Bastard Conqueror of England.)
\The Memoranda appears to be an aide memoire \prepared after Osmund was canonized in 1457, for ‘people promoting the newly-created saint’s fame. It
i
jcan readily be accepted that the exact names and H
jrelationships quoted in such a document, drawn up for propaganda purposes more than 350 years after
7. W.H. Rich Jones and W. Dunn Macray (eds.), Charters and | ~ Documents Illustrating the History of the Cathedral, City and
Diocese of Salisbury in the 12th and 13th Centuries (Rolls Series 1891), p. 373.
| 8. Kathleen Edwards, The English Secular Cathedrals in the Middle
115
Osmund’s death, are unreliable. But it is only these suspect embroideries that need to be rejected out of hand. The Memoranda appears to contain a kernel of truth in its reference to the ‘noble blood of ... dukes’, where it echoes the wording not only of the tomb inscription but also of a far more reliable document, the Papal Bull of Osmund’s canonization.? The Bull did not mention Isabella or Henry of Seez, who are indeed unknown outside the Memoranda, but the relevant passage read:
Osmundus generositate praeclarus, stirpe regia ac etiam ducali_nobilissima ortus prosapia ...’ (Osmund, distinguished by high birth, of royal lineage and furthermore descended from a most noble ducal family . . .)
It will be seen that in describing Osmund’s lineage, the Bull used terms derived from the same words — dux, nobilis — as both the tomb inscription and the Memoranda. The tomb has ‘ducibus . . . nobilibus’; the Bull ‘ducali nobilissima’; the Memoranda ‘nobili . . . ducum’. It is difficult to believe that this correspond- ence of terms can be purely coincidental. The Bull must have relied on information supplied by the agents of the Dean and Chapter who handled the last stages of the canonization negotiations.!° The choice of wording suggests that their account was derived directly from the tomb inscription. And it must be remembered that in the 15th century no one can have been in any doubt as to which of the tombs in the cathedral was Osmund’s. Given the additional evid- ence of the Bull of Canonization, Raby’s claim, that the ‘dukes and nobles’ of the tomb inscription const- tute decisive evidence in favour of the assignment of the tomb to Osmund, should be accepted.
Raby offered two suggestions on the question of the correct expansion of the doubtful letters Q:TB’ in the last line of the verse: divitibus (he reflected lustre on rich princes); and quis vicibus (on whom in turn he reflected lustre . . .). Neither is wholly satisfactory (as Raby himself freely admitted) as both require an alteration to the lettering actually on the tomb. Diviti- bus fails to accommodate the ‘Q’ and quis vicibus the a Ue
A close scrutiny of the effigy suggests, however, that Raby was right in his suggestion that the mason rather than the poet may have been at fault. There
Ages (Manchester University Press, 1949), p. 13, n. 5.
9. A.R. Malden (ed.), The Canonization of St Osmund (Wilts Record Society, Salisbury, 1901), p. 227.
10. There is no reference to Osmund’s parentage in the formal canonization petition.
116 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
appears to have been a change of mason at the word PROPE in the last line. Up to that point the mason makes use of normal abbreviations and the device of placing letters inside other rounded ones such as O or P in order to fit the verse into the space available. At PROPE the mason seems to have gone to pieces. PROPE is given in full although the standard abbre- viation would greatly reduce it. Q:TB’ is then over- abbreviated to compensate, producing the conun- drum we now have to deal with. QI=quasi is given without the S inside the Q which would make its meaning definite (cf. QA = quia in the first line). Gema is spelt with only one M. The chance of saving space by putting the E of RELUXIT inside the R is missed. We seem to have here a mason who did not understand what he was doing. The Q:TB’ which causes us so much trouble may simply be a botched effort to cover up something he did not understand.
For these reasons quis vicibus (on whom in turn he reflected lustre) can be accepted as being most in accord with the general tenor of the verse (see below), in spite of its obvious difficulties.
H. de S. Shortt pointed out correctly, in 1971, that the reference to a noble ancestry fits Jocelin as well as Osmund, and then based his case for the former chiefly on his solution to the Q:TB’ problem. He suggested that these letters stand for QUISQUE TRIBUS (to each of three), a solution which has the great merit of incorporating all the letters actually to be found on the effigy. It should be noted, however, that this interpretation, which is crucial to Shortt’s case for Jocelin, is perfectly compatible with Raby’s case for Osmund; the ‘three princes’ would then be three contemporary (prope) members of the Norman ducal house — William I, William II and Duke Robert. Shortt then translated the last two lines of the verse:
Princes in hordes, dukes, nobles and lords as his sires he could muster.
Bishops were three who had sat in this see and to them he gave lustre.
Having translated principibus as ‘bishops’ Shortt claimed that, as the fourth incumbent of the see of Salisbury after the move from Sherborne, Jocelin must be the subject of the epitaph and the ‘three princes’ must be his three predecessors. However, he offered no reasons for the translation of principes by ‘bishops’. While bishops are sometimes referred to as
11. Diana E. Greenway (compiler), John le Neve, Fast: Ecclesiae Anglicanae 1066-1300, Vol. IV, Salisbury (University of
principes at this period, the word must be read in context and on this occasion a rendering of ‘bishops’ runs contrary to the whole gist of the verse. The last two lines are clearly meant to be read together and account has to be taken of the deliberate alliterative contrast between primordia — ancestry far off in me — of line 5 and the principibus prope— princes nearby (also in time) of line 6. It then becomes clear that a mutual relationship is being described. The ancestors confer honour by the blood relationship; the deceased confers back honour by his virtues. The introduction of bishops as recipients of the reflected glory destroys the mutuality and, in conse- quence, the coherence of the couplet.
Moreover it is difficult to believe that the composer of an epitaph in honour of Jocelin, writing presum- ably c.1184, would have classed together, as bene- ficiaries of his reflected lustre, the probably long forgotten Hereman, over 100 years dead, the worldly Roger whose final disgrace and death had occurred within living memory, and the revered Osmund, the subject of a well-established cult, with miracles reported from his tomb. In the case of Osmund, such a claim would have been presumptuous.
It shouid be noted not only that Shortt’s translation of the crucial last two lines of the verse is extremely free, but also that the exact Latin equivalents of the three words on which he rested his case — bishop (episcopus), three (tres) and see (sedes) — do not appear in the inscription. They are all a matter of interpreta- tion: principibus = bishops; TB’ = three; prope = in this see. In contrast, the words on which Raby relied — dukes, nobles — are exact translations from the Latin of the text.
Shortt concluded his case against Raby by suggest- ing that he ‘bases his argument to some extent upon the particular style of rhyming hexameters used in the epitaph, and cites Gualo Brito as writing similar verse in 1090’. Shortt claimed that an ‘exact parallel’ occurs on the tomb at Old Sarum of a certain Alward who ‘was buried beside Godwin the Precentor whose death appears to fall between 1160 and 1170’. Shortt had already given the date of Jocelin’s death as 1174 and thus contrived to imply that the Alward verse was written much closer to Jocelin’s death than to Osmund’s.
Unfortunately, the details Shortt gave are wrong in almost every particular. Jocelin died in 1184 not 1174.!! The two verses are not exact parallels — only two of the four lines on Alward’s tomb are end- rhymed and only one has an internal rhyme. The
London Institute of Historical Research, 1991), p. 3.
2
ee
Se SNA ee
hoe Coreen
Fini eens Rat
ars
A 12TH-CENTURY EFFIGY IN SALISBURY CATHEDRAL
tomb is fairly near but not beside the tomb of Godwin the Precentor who is thought to have died c.1130.'? The Godwin who died 1160-70 was a different man.!? In any event, Raby did not base his case on the style of
_ the verse but merely referred to Brito to prove that a _ date c.1100 was not too early for verse of this degree of
sophistication.
_ As Raby pointed out, the opening words of the epitaph, “They weep today in Salisbury’ imply that it was composed soon after the death of its subject. | Osmund died on 3 December 1099, but on stylistic
grounds the effigy must be ascribed to the second half
of the 12th century. It cannot therefore be part of
Osmund’s first tomb and Raby was mistaken in
_ believing that it was.
By the early years of the 13th century the Dean and Chapter were taking steps to apply for Osmund’s
| canonization. The petition was delayed by the up-
heavals of King John’s reign, but evidence eventually submitted to a papal commission in 1228 included a
_ report of a miracle said to have occurred at his tomb between 1155 and 1165.'* Accounts of later miracles
make it clear that the tomb was an established focus of
veneration from then onwards. It is certainly under-
standable that the cathedral authorities should have sought to encourage the cult by replacing the original coffin with a more imposing monument in the latest style. The epitaph, which had probably been cut so as to cover the whole surface of the old coffin lid _(Godwin’s and Alward’s were cut in this way), would then have been re-cut round the edges of the new effigy. The masons may have had some difficulty in fitting the inscription into the space available there. The 12th-century re-cutting would explain the use of Salesberie with an ‘I’. ‘Salisbury’ was gradually substituted for the Anglo-Saxon Saresberie in common speech as a result of linguistic changes following the Norman Conquest, although Saresberie, later corrup-
= 2
‘Report of the Excavation of the Cathedral Church of Old _ Sarum in 1913’, Proc. Soc. Antiquaries 26 (1914), pp. 112-13. (3. Diana E. Greenway, op. cit., p. 121.
'14. A.R. Malden, op. cit., pp. 35, 36.
(15. J.J. North, English Hammered Coinage, Vol. 1 (London, 1980), pp. 161, 184.
117
ted to Sarum, remained normal in written documents. Coins minted at Salisbury, however, have ‘Serbe’ in the reign of Henry I (1100-1135) but ‘Saleb’ under Henry II (1154-1189).'° It is possible that the tomb inscription cut c.1100 already used the ‘I’ form, but it is probably safer to assume that the spelling was altered at the time of the recutting to reflect the by then well-established change in common speech. The effigy has a second inscription running down the orphrey of the chasuble: Quisquis es affer opem devenies in idem (Whoever you are give alms you will become what I am).!° Preparations for submission of a canon- ization petition for Osmund and for building a new cathedral which would contain his shrine went hand in hand over a long period. The effigy was a cult figure, the first step in both processes, and it is suggested that this inscription is a request for contri- butions to the building fund.
Three conclusions may be drawn from this re- examination. The subject of the epitaph on the effigy is Osmund. Phrases in his Bull of Canonization and in the 15th-century Memoranda de Osmundo directly reflect the ‘dukes and nobles’ in line 5 of the verse. Secondly, Shortt’s case for Jocelin cannot be sus- tained. His expansion of Q:TB’ to QUISQUE TRIBUS is not the only possible one. His translation of the verse obscures the meaning of the last two lines and his rendering of PRINCIPIBUS by ‘bishops’ is not justified. It would have been presumptuous for Jocelin’s epitaph to have claimed that his virtues ‘gave lustre’ to Osmund, a candidate for canonization. Thirdly, the verse inscription was composed soon after Osmund’s death in 1099 for his first tomb and was later transferred to the effigy which dates from the second half of the 12th century. The effigy was a cult figure designed to promote plans for Osmund’s canonization. The inscription on the vestments appealing for funds supports this view.
16. The first two words are now illegible but the full inscription is recorded in Leland: L. Toulmin Smith (ed.), The Itinerary of John Leland (Oxford 1906-10), p. 167. In ‘The Cult and Tombs of St Osmund at Salisbury’, p. 50, I suggested that the inscription was a plea for the help (opem) of the reader’s prayers. I now think it is a request for material donations.
Wiltshire Archaeological and Natural History Magazine, vol. 86 (1993), pp. 118-129
Bradford-on-Avon in the 14th Century
by R.B. HARVEY and B.K. HARVEY
The paper gives a view of the town of Bradford and its manorial buildings, agriculture, and links with other manors in the Shafiesbury Abbey estate, as revealed in some late 14th-century accounts.
Introduction
This paper is concerned with the light shed on 14th-century Bradford by a roll in the Methuen archives now at the Wiltshire Record Office,! contain- ing portions of the accounts for the Abbey of Shaftes- bury’s manor of Bradford-on-Avon when Joan Formage was Abbess (1362-1393).
Bradford had been granted to the Abbey by Ethelred IT in 1001, the charter of the grant containing bounds? defining the extent of the land granted. It remained in the Abbey’s hands until the Dissolution. Apart from periods when this matter was in dispute with the King, the Abbey also held the Hundred of Bradford,? whose courts are referred to in the accounts. It should be noted here that Monkton Farleigh, Little or Cottles’ Atworth, Chalfield and some other vills were not in the Shaftesbury estate, though they were in the Bradford Hundred. Also mentioned is the prebendal or parso- nage manor, presumably set up before 1349 when arrangements for the Abbey’s impropriation of the church of Bradford were completed.*
There are not many other documents referring to the Abbey’s activities in the Bradford estate during the Middle Ages. The accounts can most usefully be compared with three 12th-century custumals in the Shaftesbury Cartulary,° dated c.1130, 1170 and 1190. The 13th-century Book of Fees® confirms the list of the Abbey’s vills — (Great) Atworth, (South) Wraxall, Holt, Stoke, Trowle, Winsley, Woolley and the town itself. Of later documents, there is a book of court rolls and accounts dated 1518 for nearly all the Abbey’s manors,’ and Bristol diocesan archives have prebendal court records from c.1563, especially a
—_
W[iltshire] R[ecord] O[ffice] 1742/6786.
G.B. Grundy, The Saxon Land Charters of Wiltshire, second series (1920), p. 101; R.B. Harvey, WAM 79 (1984), p. 235. V.C.H. Wilts., Vol. 7, p. 2.
WRO D1/2/3, Reg. Wyvill.
B[ritish] L[ibrary], Harleian MS. 61, ff.37-89.
Testa de Nevill (or Book of Fees) Record Commission (1807), p. 153b.
NR
Syne
Curia Recognitionis of that year.* A manorial survey of c.1630° also has links with the present accounts.
At various times several large portions of the estate were evidently farmed out and managed separately. Even in the 12th-century custumals, Stoke is said to have been leased at farm for a rent of 100s. per annum, and was still so in the Book of Fees. Two hides at Atworth had been claimed as a personal estate by Thomas, kinsman of the Abbess Cecilia (1107— 1120). Her successor, Abbess Emma, was able to reassert her right to this estate in 1122 under a writ of Henry I!° but Thomas and his descendants continued as tenants of one hide there. These two tthings or vills seem to have had some independence from then onwards. In the present accounts the ‘mansion of the court of Stoke’ was leased out, as were the pasture and the customary works. However, though Atworth had its own barn, granary, dovecote and a chamber, and its boon works, grain and livestock are always shown separately from Bradford’s, they were managed as part of the main estate. It is also clear from the accounts that Wraxall was farmed out to Richard Poyntz, probably a predecessor of the Longs of Wraxall.
There are important references in the accounts to the Hall family who had a town mill from before 1170, and to another large estate held by William Bezyle, ancestor of the Rogers family. These two freehold estates were later to become the (sub-)manors of Hall and Rogers, the latter being largely in Holt and Leigh. There remains Winsley, whose sub-manor does not seem to have been set up until the 17th century.!! The accounts therefore only record an early stage of the growth of sub-manors.
WRO 1728/deed no. 67.
Bristol Record Office DC/E/27/2.
WRO 959/1, undated.
B.L. Harl. MS. 61, ff.23, 30; Regesta Regum Anglo- Normannorum, 1066-1154 (HMSO, 1920), Vol. i, p. 131; Vol. il, p. 346.
11. WRO 947/1315.
SRO Co
: | | BRADFORD-ON-AVON IN THE 14TH CENTURY 119
| ; :
a ees ee ele ee F LSS =. 2 : c ; | ! . ATWORTH = / : ! 5 | JF ; \ 7 WRAXALL c UY; : t ; | ! one | NI i / ( all Nee i ! Vn y, ~ Ny echt Pama Ne J \ fe ay \ . = a ? EX a / en / x / a ; / @ Sots GON aN 7-7 » ye Ford Farm = \ Ue \ : : a : E ~ = LF SS ee q << ’ ‘ HOLT Bee ape WINSLEY f im V4 ‘ y ] Lai \ WOOLLEY Wp \ eee STOKE ! Ye [ YY BRADFORD : aN : oa Mo Bf ~ =e Avoncliff = | ce “\ Widbrook @ i MS \ ZS ? ( | Q J | z Trowle | vi. Farm__ . Aa
‘Figure 1. The manor of Bradford in the 14th century (not to scale). The 13th- and 15th-century tithing of Woolley was part of | Holt in the 14th century. (Drawing by Nick Griffiths)
The Document roll to protect it. Again, at some unknown date, the The roll contains accounts for five years — the 6th, ends of five of the six accounts were trimmed or worn 1th, 15th, 22nd and 30th of Abbess Joan Formage — off and so are missing. The only complete account is that is between 1367 and 1392. Each year’s account is one of the two parts for 1367. Of the others, that for 1383 ‘a fair copy, made for the annual Michaelmas audit, is shortest, with only 4 inches surviving. This is written on a number of parchment membranes which particularly unfortunate, as it is the only one for the were sewn together head to tail and presumably prebendal manor. Clearly the whole document is far too formed a roll. There were two such rolls for the 1367 long to reproduce here, and thus this paper can only accounts. introduce a number of items that illustrate manorial life. | At some later date these rolls were opened out, and The accounts keep to a set form, described as Phase their heads sewn together on to a thick cardboard 2 by Professor P.D.A. Harvey,'* and normal for the cylinder of c.3 inches diameter. The longest of them, time. Each year’s account was compiled in two parts, for 1372, is some 8 feet long; it was provided with a the one a cash account and the other for agricultural cloth tail which now wraps round the whole combined produce — grain and livestock on the demesne farm.
(2. P.D.A. Harvey, Manorial Records (British Records Assoc. no. 5, 1984).
120 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
The clerk occasionally refers to a Journal and the Bailiff's particulars, to support his statements. The accounts and the careful audit were made not so much to reveal the prosperity or otherwise of the estate as to prevent officers and tenants from misappropriating the Abbey’s property, and to show at the end of the year whether the reeve was in debt to the Abbess or conversely.
Each part of the account is divided into a number of sections. These are, on the whole, very similar to those listed by Harvey,!* though in the present accounts the distinction between Charge and Discharge is less clear. Each sectional account begins with the arrears at the end of the previous year, for which that year’s reeve was accountable. There follow the rents and other dues received or owed for the current year. It is clear that many of the entries in each paragraph did not change from year to year. These are required figures which should have been received during the year. If the actual receipt was different, the reason is usually stated and the reeve would have had to make good the deficit.
The farm stock account was similar. Each kind of grain and each kind of animal is dealt with in a two-part section. The first part shows how much of that grain or how many of those beasts the farm had, starting from what was left over from the previous year. The reeve was responsible for these. The second part shows how the grain was allocated, or what happened to the animals.
Administrative Staff
There is no evidence in the years for which accounts survive that the Abbess visited Bradford, although in 1368 she had received permission to leave the Abbey for a year and reside on her manors, for air and recreation.!* Her chief administrative officers were her stewards of the manor and the hundred. Only once is a name mentioned — that of Nicholas Thur- mond, manorial steward in 1392. His duty was to hold the manor courts twice a year, and to view the manor to see that all was well. His instructions are occa- sionally quoted.
Next in importance was the bailiff, who managed four manors, not identified, and Calveston, i.e. Kel- ston near Bath in Somerset. For all except the last account, the bailiff was Thomas atte fforde, whose family home was Ford Farm, Bradford Leigh. This he had held freely since 1352 on a chief rent of 1d., by agreement with the Abbess. He also shared a two-
13. Ibid., p. 32. 14. V.C.H. Dorset, Vol. 2, p. 78.
virgate farm, and had a house in Mill Street in the town. During his term of office, he had the custody of
Thomas Hall’s lands for the minority of the heir, ©
Thomas (junior). fforde married his daughter to young Thomas Hall, and so Ford Farm passed to the Hall family, who still held it at 1d. rent in 1630.
The reeve was normally a bondman of the manor, chosen annually by his peers, and holding the immediate responsibility for running the manor. For the first account the reeve was John le Eyr of Trowle Farm, a half-virgate farm in Trowle, and he was also an undertenant of fforde in Bradford Leigh. Such a man would have the necessary experience. The 1372 reeve was Walter Coleman whose namesake, and possible kinsman, built the Great House in Kington Langley in the late 17th century. Limpley Stoke had a part-time reeve, employed as a rent-collector for less pay.
The hayward supervised sowing the demesne fields, and was also responsible for summoning the workers at haytime and harvest. Four haywards were used in 1372, as appears by the wheat allowances they received at the time of sowing.
The Demesne Farm
Barton Farm (‘le Berton’) was the demesne farm of Bradford, and in the survey of c.1630 it is said to be the site of the manor. The courts, however, were held in a courthouse in the town, whose site has been identified by the pigeonhouse beside it. This gave its name to the field in which it stood, near the Saxon church. The court house must also have been near the site of the original monastery said to have been founded in Bradford by St Aldhelm, who was its Abbot in 705.!°
Barton farm lies beside the Avon, a little down river from the parish and Saxon churches, and is reached over the 14th-century Barton Bridge, mentioned in the accounts. The watermeadows here on the church side were called Innox Mead, a name usually indic- ating the earliest cultivation. The 12th-century custu- mals do not name the farm, but under the tithing of Trowle is a list of farm servants. Barton Farm was later in the adjoining Borough tithing. During the period of the custumals, however, the uthing bounda- ries were still changing, so it is reasonable to assume that the farm is on its original site.
Experience elsewhere suggests that the first buildings at Barton were timber-framed, later replaced in stone. Certainly stone was available from quarries developed in the hillside above the farm, and
15. V.C.H. Wilts, Vol. 7, p- 12.
RADFORD-ON-AVON IN THE 4TH CENTURY 121
)
ice
with tithe barn, in foreground. Lithograph by A.E. Tackle. (Photograph by Derek Parker, ibrary and Museum Serv
courtesy WCC L
Barton Farm,
3
_
Figure 2. Bradford-on-Avon in the mid 19th century
122 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE -
there was a ready supply of stone tiles in the 14th century, which were bought-in. By then, many of the buildings at Barton, including lesser ones such as the steward’s stable, were stone and stone tiled. All the repairs to the walls were made by masons, and to the roofs by tilers; thatch and timber framing are not mentioned. The sole surviving exception is the upper storey of the granary, which is timber framed, and is thought to date from the early 15th century. Although a granary is mentioned in the accounts, this did not figure in the repairs.
The present farmhouse has been altered several times, but appears to have its origin in the early 14th century, with a north wing added later in the same century.!© The earliest account includes some retil- ing, setting the tiles in fresh moss, so that the building had been tiled for some time. Two staircases are mentioned, which would be typical of a central open hall with a two-storey section at either end.
In 1367 the reeve gave an inventory of the Abbey’s possessions in the farmhouse, and listed 21 locks and keys, with their whereabouts (see Appendix). From these sources and the repair details, a list of rooms in the house and another of the farm buildings, can be compiled. The Abbess’ chief officers, steward, bailiff, reeve, hayward, and the shepherd who cared for her large flock, each had a locked room, emphasising that the function of the house was a head office, rather than the home of the lady of the manor. It was necessary, of course, to provide security for the Abbess’ possessions at Barton, and the senior officers might need privacy.
The contents of some of the rooms are revealing. In the bailiffs chamber were the rental and customs of the manor which he needed for his work. When the annual audit was made, the auditors judged estate performance against what they considered to be an acceptable level; any unexplained differences were charged against the reeve and some occur in the accounts. This system implies standardisation in measuring the produce. So the bailiff had a rod, a bushel and a gallon measure, each sealed with the royal seal as standard. He also had an iron rod for sealing standard measures of the hundred on behalf of the Abbess. The granary had a bushel measure agreeing with the Shaftesbury standard, and a non- standard bushel and peck.
The hall had the usual three tables and forms, and one chair. The top table guests were favoured with drinking cups and a table cloth, whilst the side tables
16. N. Moore and J. Reeves, R.C.H.M. unpublished report (1978). 17. Only one well is now known at Barton. A 19th-century
had a hemp cloth; both had towels, and there was a basin and ewer for washing. Neither here, nor in the kitchen, is there any mention of fire-irons, but the kitchen had a grid-iron, and a small furnace (‘fornax’) for cooking meat. Fish on Friday can have been no problem with the river so close, and the reeve issued oats and salt for the household pottage. There was a dairy where cheese was made, with two cheese racks, one a long hanging board, and the other of four round, stepped boards. The dairy was let out to successive tenants, called in one case ‘wickemannus’, and in the other ‘Daye’, both meaning dairyman. When, in 1376, some cheese and milk were sold, the cheese was jd. per lb and milk 3d. per gallon.
The steward’s chamber contained none of the Abbey’s possessions; possibly it was empty, as 1t was being re-tiled. Another omission from the inventory is any provision for those who had to sleep at Barton, and it may be concluded that they had to provide their own bedding.
In 1367 there was a brewhouse and a malthouse, but by 1392 two brewhouses existed. A separate malting account was kept. The malt kiln was men- tioned when it was repaired. The kitchen and malt- house each had a well,!’ the latter under a pentice which was repaired in 1392. It had a stone wall, perhaps low, a stone tiled roof, and woodwork which the carpenter spent fourteen days renewing. In the same year a new brewing trough was made. Two long elm boards were provided by Bradford, and sent to Shaftesbury, where plumbers lined the trough with lead. Possibly it was then loaded on to Bradford’s cart, which had a ladder body, and trundled home; this cart had harness for five horses.
Bradford is fortunate in the farm buildings which remain from the 14th century. Foremost is the Tithe Barn, a splendid fourteen-bay cruck barn built in the middle of the 14th century, probably as a result of the Abbey’s acquisition of the tithes. Its two keys belonged to the small personal doors in the sides of the two north porches. In the accounts it is called the great barn, to distinguish it from the small barn which had a single key and so, possibly, one porch. A cruck building which fell in the gales of 1990 may have been the altered remnant of this barn, or it could have been the byre, also mentioned. In the account of the prebendal manor, its reeve reports that he has threshed all his grain and passed it to the reeve of the lay manor, the amount being less than the demesne corn. Perhaps, then, the small barn should more
engraving shows what appears to be a brewhouse beside it, but these domestic buildings have not survived.
BRADFORD-ON-AVON IN THE 14TH CENTURY
properly have been called the Tithe Barn? In 1383, the parsonage barn was being leased out, with the curtilage next to the granger’s room. His duty was to receive and guard the corn, and he was probably expected to sleep with it for security.
Other farm buildings mentioned include the carthouse and the Barton dovehouse, this last presumably near the present converted cowsheds, since the field beyond is called Culver Close. These two were locked, but the byre and the cart horses’ stable were apparently unlocked; the stable had a rack 28 feet long. Excavation, and some comments in the repair accounts, suggest that the compound was walled.'® There was an outer gate and a garden, and the pound was on site.
The buildings at Atworth have already been men- tioned. The dovehouse was partly collapsed in 1383, and men carried stone from Bradford for repairs. The barn at Atworth still survives. It is the same width as the Tithe Barn, but about half the length.
In the repair accounts, the usual daily rates of pay were for the carpenter, 4d., mason 35d. and tiller 3d. Wheat was sometimes given as well.
Agriculture
Two aspects of the agriculture appear in the accounts: the produce and working of the demesne farm and the rents, services and other payments from the tenants. Seen from the reeve’s point of view, what happened _ on the tenants’ own land and plots was immaterial.
| Rents occupy three major sections of the accounts. | In the 14th century there were fixed rents called | ‘Rents of Assize’ with an annual total of £68 13s.4d., greatly exceeding the other kinds of rent. Unfor- tunately, only the quarterly totals of these payments _Wwere shown in the accounts, so defeating any
population estimation.
The second section — Farm of Lands — included leases of many kinds such as fisheries on the Avon, the farming out of the manorial bakery and dovecotes, as well as new rents and all changes in the Assize Rents. | Entries here give the name of the tenant of each holding and his predecessors, often with a brief description of the land, its whereabouts and of course the rent due. Thus ‘2d. from Agnes Corp for a place in Winsley, in which there is a pond for watering the horses’, or ‘4d. per annum from John Russel for 1 jacre of land in Wyndeslawe ... late of William Shernesham’. Wyndeslawe, now Winterleys, is a
group of fields above Elmscross, towards Westwood.
18. J. Haslam, ‘Excavations at Bradford-on-Avon, 1983’ WAM 78 (1983), p. 120.
123
Such rents added up to nearly £10 in 1367 and greatly increased in 1392.
A third type of rent occurs in ‘Exitus Operum’, proceeds of works, consisting of mainly cash sums received in lieu of former labour services. Again tenants are named, with some indication of the land. One such was:
2s.105d. from John le Eyr, reeve, for works arising from half a virgate of land, late Thomas Harold and earlier of John Spronks, in Troll, demised to him (le Eyr) yearly (10th year): besides 3s. of fixed rent, 94d. of larding money; so the said tenement pays 6s. 8d. a year in toto; also he will reap and mow 6 acres of the demesne corn at harvest every year at his own cost.
This entry is an interesting one, for there is still a field called Spronks at Trowle Farm near the Wingfield boundary, confirming that Trowle Farm was even then the principal farm of that tithing of the manor.
Comparatively few of the holdings in this section have labour services mentioned, and those which do are the larger ones. Another such was Thomas atte Welle, who paid 14s. 1d. for converted works and the usual taxes, and had also to plough 12 acres of the demesne, carry 12 quarters of grain to Shaftesbury and do reaping work in the demesne corn, providing 12 men for | day, with their food. However, it seems that most tenants who had exchanged their labour for money rents had managed to buy out all their duties. About three quarters of the tenants in this section paid cash rents only.
There are few references to the manor paying for labour to replace the exchanged customary works. There were manorial servants, the ‘famuli’, but there do not seem to have been many of them. Further- more, those named as receiving cash and grain allow- ances were not general labourers but had particular tasks. There was an annual grain allocation to 2 carters, 3 drovers, 3 ‘akermen’, | oxman, 2 shep- herds, 1 swineherd, 1 maltman and 1 ‘inhyne’. An akerman held a small acreage by service of ‘following the demesne ploughs’, and an inhyne was responsible for maintaining the enclosing of the fallow land for catch crops. Grain was allotted, too, to the bailiff for his team at harvest — the reeve’s boy, | hayward, 1 granger, 1 cook, 3 stackers, and 5 ‘houkars’. For the six weeks from early August to mid-September they received 5 quarters of wheat ‘as appears from the Journal’.
124 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
On top of their rents and customary works, tenants were liable to be called on for various boon works. After cash rents had replaced so many customary works, those would no longer suffice for the plough- ing needs of the manor farm or for bringing in the harvest. As a result, these extra boon works were demanded and duly recorded in the accounts. In 1367:
In payment of | plough-boon of men going with 24 ploughs, with 12 harrows ploughing and harrowing 24 acres for sowing wheat, with the expenses of the bailiff, reeve, hayward and others, 5 bushels of wheat.
This amounts to less than a fifth of the 136 acres of wheat sown. Boon work provided about a quarter of the ploughing for the oats and none for the barley. There was another boon that year to prepare 20 acres for fallowing by light ploughing. The tendency in later years was to have fewer plough boons, but on larger acreages.
The rates of sowing were fairly constant — wheat at 15 bushels an acre with 7 bushels extra needed that year on the above 136 acres, 23 bushels an acre for barley and dredge, and 35 for oats. These are rather less than was recommended by Walter of Henley!’ or was sown earlier in the century at Cuxham, Oxfordshire.*? The return in 1367 was 4.3 bushels per bushel of wheat sown, but less than 2 for oats. At Atworth the barley crop failed, causing much search- ing to replace it by the inferior dredge in the quota to be sent to the Abbey.
Records of the sowing unfortunately only survive for all the grains for 1367 and 1372 and for wheat in 1383. In those years there are lists of furlongs with the acreage of the particular grain sown there. Oats were sown in 1372 in ‘la Fyleche’, ‘Huggolston’, ‘Gavel- furlong’? and ‘Wyggebroukesfurlong’. The last of these is Widbrook Furlong, and Gavel Furlong is perhaps traceable, but not the rest. Comparison of the lists of names does not reveal any groups of furlongs recurring with the same or a different crop. With no information for consecutive years, no conclusions about crop rotation are possible.
The accounts reveal what was done with all this grain. Taking wheat in 1367 again as typical, the granger at the barn received nearly 169 quarters. Of these, 96 were allocated for local purposes, mainly payments in kind, 53 were sent to Shaftesbury and 20
19. D. Oschinsky (ed.), Walter of Henly (Oxford 1971), p. 307. 20. P.D.A. Harvey (ed.), Manorial Records of Cuxham, Oxfordshire, c.1200-1359 (Historical Manuscripts Commission, 1976).
to Philip Strange the Bradford maltman. With wheat brought forward, and other grains, Strange made about 241 quarters of malt. Nine quarters he brewed into ale at harvest but the rest, 232 quarters, — how many pints of beer would that provide? — went to Shaftesbury Abbey.
Livestock
The stock account at Michaelmas 1367 gives an idea of the number of beasts on the demesne farm. There were 6 horses and a foal, 129 cattle, 910 sheep, 96 pigs, 11 swans, 4 geese, 32 capons, 6 hens, 30 pullets, 2 pheasants and 3 partridges. Among the cattle were 51 oxen. As against the six horses, this would suggest that most, if not all, the ploughing was done with oxen.
The sheep were by far the most numerous animals, and the accounts have more to say about them than about the others. They were, of course, valued above all for their fleeces. The shearing is shown as the climax of the flock’s year. The sheep were also vital providers of fertility for the arable fields. The account contains five sheep sections, for the wethers, rams, ewes, hogs (i.e. hoggets, two year olds) and lambs, and it is possible that, as recommended in texts like the ‘Seneschaucy”*! of a hundred years before, each | group may have had a separate flock. Only two — shepherds are listed; even in the less busy parts of the | year they would surely have needed help with moving | the flocks around for fresh pasture? The number of | sheep varied widely from year to year. Thus in 1367 | and 1372, the only years for which these accounts survive, there were 910 and 1,072 sheep respectively | at Michaelmas. At the ends of the years there were 1,037 and 596 sheep, revealing big losses during | 1372-3.
Following through the year 1367-68, the story is | more complicated. In winter most sheep would be | kept in folds or in the open. Some sort of shelter was available for the more vulnerable animals, in the same way as there was the large sheephouse at Tisbury. | Other monastic estates are also known to have had | them. At Bradford in 1367, too, 15 acres of wheat | were to be sown on a field called ‘Sheephouscroft’. As late as the Tithe Apportionment schedule”’ of 1841 there are several fields called Shepherd’s close or croft, and in some cases the ‘Shepherd’s’ had been ‘Sheephouse’ in the 16th century.** Such sheephouses | were usually open-fronted, and the Tisbury one was
21. Oschinsky, op. cit., p. 261. 22. WRO T.A. Bradford. } 23. WRO 959/2.
BRADFORD-ON-AVON IN THE 14TH CENTURY
very long. Two wethers were taken in the ploughing season for the meals provided at the plough boons. There were losses too from ‘the murrain’, a disease from which 9 wethers, 29 ewes, 2 rams and 43 hogs died before the winter was through.
At lambing, two ‘Ekhurdes’ joined the shepherds. One was there for thirteen weeks from the feast of the Purification (2 February) to the ‘Invention’ (i.e. discovery) of the True Cross (3 May) and the other for the busiest five weeks. For treating the sheep, 93 gallons of tar and 8 of unguents were provided. Extra hurdles were made for shelter and 13 lb of candles (at a cost of 25d.) to give light to the fold. The ‘Lambs’ section of the account begins here. A total of 341 were born, for 84 of the ewes were found to be sterile or
infirm. Seventy of these lambs died before weaning, with 35 ewes. The account goes on baldly: one lamb became wages for the shepherd, and nine were the _vicar’s share of the tithe on the lambs. That left 261 in the flock, but before shearing 35 more had been added from the parsonage flock, presumably the tithes of _tenants’ sheep. Another detail recorded is the milk — the lactage of the 240 ewes at 13d. a head, was worth | 30s.
The next big event of the ovine year would be the ‘shearing. However, at some time before that, 140 /wethers were sent to the Reeve at ‘Ludynton’ (Lid- ‘dington), a Shaftesbury manor near Swindon, linked with Bradford. No reason is given; possible explana- ‘tions could be an outbreak of murrain or a lack of Bradford pasture compared with that of the downs above Liddington. Still, before the shearing the Lid- idington reeve sent back 97 kebs (barren ewes); 81 of them died shortly after, so perhaps the trouble may have been at Liddington.
| Shearing had been a customary duty, but in this year 3s.4d. was spent on ‘assistance in washing and shearing the Lady’s sheep in place of the said tenants’ - that is for 934 sheep and lambs this year ‘at Bradeford and Atteworth’. At the rate of 10 sheep sheared for 1d. quoted by Walter of Henley a century defore, the customary shearing of some 400 sheep had peen turned into cash payments by the time of the iccounts. Compared with this figure of 934, the 623 sheep and lambs sheared in 1372-3 indicates again hat it was a disastrous year. Another entry in 1367 is 2d. for half a cheese bought for the customary sheep hearing’. A more mysterious one is ‘by gift of the Lady, to certain fishermen guarding the Lady’s lambs a the water, 1 bushel of wheat’. Perhaps this was a orerunner of dipping?
The produce of the year’s operation was 627 fleeces and 13 more ‘found’) and accounting for these takes
125
seven more sections in the roll. The vicar’s tithe was 26 fleeces, the shepherd was allowed one and the rest were sent to the Lardinarius at Shaftesbury, as were the 296 lambs’ fleeces. The hides of the dead were shared out in the same way.
Cartage
The accounts show how much traffic there was between the various Abbey estates particularly, of course, from Bradford to Shaftesbury itself. The manor courts required visits by officials — the Sheriff twice a year for Hundred courts, the steward and bailiff more often. Each had his retinue and horses to be provided for — wheat bread and oats. Nothing in the accounts, though, seems to resemble the 12th- century task of accompanying nuns to Kelston or even to Gloucester.
Some movements of animals have already been noted. As another example, in 1392 two men drove six oxen to Barton — the starting point is unnamed — and then went on with 140 lambs to Donhead. Another task, perhaps an annual one, was to fatten 31 pigs with barley and beans for ‘Larding’ and then send them to the Abbey. Building work needed cartage too. New barn doors were taken on two carts to Kilmington and 150 feet of stone crests for the ridge of the sheephouse to Tisbury, both in 1392. In 1383 men went to Kelston with stone for the oxhouse and later there was a payment to the reeve there to help the work on his new barn. In 1376, the Bradford carpenters made a timber-framed porch with chamber over and this was sent to Shaftesbury, together with 600ft of palings for a new palisade. There is a more medieval air about the annual journey to obtain two packhorse loads of palm branches and take them to the Abbey on the eve of Palm Sunday. In 1392 it cost 4d. to find a source of palm at Boxleigh. Later, this duty was commuted into an annual payment called ‘Palmson money’, still being paid in the 17th century.
Naturally, the main traffic was of grain and malt from the great barn at Barton to Shaftesbury. In contrast to the 12th-century mentions of wagons, of small and large carts and pack horses, the 14th century only records ‘Averagia’ which are carriage services — perhaps standard loads, to cover both ‘carecta’ (carts), or ‘cariat®” (meaning ‘was carried’). The latter could be the pack horse journeys.
The section of the accounts dealing with these journeys survives only for 1367 when, during the year, most of the malt and one third of the wheat grain at Barton Farm were sent to the Abbey. Except for 43 quarters kept for brewing at harvest, all the 51
126 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
quarters of malted barley went to Shaftesbury in five convoys. The account reads:
Delivered to the 13 qr. 5bz. granger at Shafton by by | tally local carts
The same by ‘carecta 28 qr. by 2 Baronum’ tall.
The same by ‘cariat®” 5 qr. &c. of Tissebury
The same by ‘cariat® 3 qr. &c. of Donhevede
Delivered to the lsqr.
granger by ‘carecta Baronum’
Apart from the question of what a baronial cart was, why was the Tisbury, Donhead and sometimes Sedg- hill transport being used?
A section of the account is devoted to counting up the carriage duties owed by certain tenants — perhaps traditionally by certain holdings. First the available number of ‘averagia’ was counted up — in 1367, there were 165 unused in the previous year, plus 208 due from four tenants of two virgates each, who took a load of 4 bushels of corn or malt to Shaftesbury every Friday, each providing 52 journeys at 4d. each, sixteen one-virgate holders who went every other Friday, and so on. Overall, a total of 844 ‘averagia’ were due that year. Of these, 622 were actually used.
Bradford Town
By the mid-14th century Bradford had a thriving woollen cloth industry. Many who paid the poll tax in 1377 had names related to the trade, and there were fulling mills.2* Hall’s mill, which was freehold, has already been mentioned; at the end of the 16th century the family held Lower Greenland mill.’” The whole homage paid 13s. 4d. for suit of the fulling mill, released to them annually, presumably a second mill. A house, half virgate and a fulling mill in Stoke, which at the end of the 13th century had been in the possession of Adam Tucker, were let for 33s. 4d. and acorn rent. There was also a mill at Avoncliffe, which was mentioned in 1130. At that time the miller was entitled to one trunk from the woods annually for repairs, and his rent included 15 baskets of eels. These were still payable in the 14th century, and at that time were sent to Shaftesbury. The 30 baskets of
24. Public Record Office E179/234/193. 25. K.H. Rogers, Wiltshire and Somerset Woollen Mills (Edington, 1976), p. 148.
eels which formed part of Hall’s rent were sold locally, however.
The prosperity which the cloth trade brought to the town was no doubt affected by the Black Death of 1349. In the section of the accounts where the clerk has recorded those customary works which had been converted to a cash rent (proceeds of works), he has in some cases recorded how long this arrangement had run. The seven longest all dated from 1349; in no other year were more than three issued, and five years later only one of the long term tenants had changed. This suggests that shortage of labour in 1349 had allowed the tenants to bargain for a cash payment in lieu of service. In 1367 there were a number of farms in hand for want of a tenant.
Places mentioned in the text give some idea of what a visitor to the town in the late 14th century might have seen. If he approached the town from the direction of Holt, he would first have passed the home of the Hall family. This would probably have been an open hall house of the type which gave the family its name. The visitor then passed along what is now Woolley Street, a corruption of St Olave’s Street, to the church of that name at the corner of modern Whitehill. It was used in 1367 to identify the derelict house beside it. By 1392 this house had not been rebuilt, but the curtilage had been let for a small rent. Below St Olave’s the road became Fox Street, which had a number of burgages, besides two empty cottages and five new shops. A little further down, Mull or Mill Street turns off, now Kingston Road. Here Thomas atte fforde had a house, and there were again some empty cottages. There were in all ten empty houses and four derelict in 1367. All were mentioned again in 1372, when one more was empty with the tenant’s goods in possession of the bailiff of the hundred, and another had fallen down. The houses were no doubt timber framed, hence the use of the word ‘pros- tratus’.
In Fox Street the new shops were adjacent to the preaching steps. A late 17th-century document men- tions the High Cross of Bradford opposite the George Inn (now the King’s Arms) at this point.’° Perhaps the steps formed the base of the cross? Below this again was the old market place, the centre of the town. Here, later, stood the Market House, possibly to be equated with ‘la Tolselde’, which was leased out.
The tolls of the market, with the profits of ‘Stakpeny’, |
were also leased out to a collector. The pillory stood in
26. D.M. Anderson, Heritage (Bradford-on-Avon Preservation Trust, 1975).
| |
BRADFORD-ON-AVON IN THE 14TH CENTURY 127
Saxon Church
St. Mary Tory e LA
Bradford Bridge
CULVER
4 Barton Farmhouse
Small Barn - Granary
Tithe Barn
500 feet
‘igure 3. Sketch plan of Bradford. Whitehill and Church Street are not named in the Accounts. Fox Street is now Silver Street. (Drawing by Nick Griffiths)
he centre of the market place, and in 1372 there were held at Bradford Leigh. The fair was on the Monday 4 market stalls, 12 of which were new-built.’” after Trinity Sunday, perhaps because the parish | As has been mentioned, the Abbess held the preb- church was dedicated to the Holy Trinity. The earli- ndal manor, and the prebendal court house was in est mention of the fair is in the 1190 custumal.
ais region, opposite the pillory. Separate courts were At the end of Bradford Bridge was a vacant plot, eld for the lay and prebendal manors, and both where, for 6d. annual rent, John Carpenter had the ourts were leased, but the accounts are not strictly right to dig for marl. Two original arches of the bridge *parate, some prebendal rents occurring in the lay survive; it was much narrower then than now, and --counts. Whether Bradford’s annual fair was held in probably without parapets, since the town later je market place at that time is not known; later it was appealed for these to be added.”8
|
In the 19th century, the market moved to the present Market 28. V.C.H. Wilts., Vol. 7, p. 9.
Street, previously Pepit Street.
128
When Leland passed this way, he said that ‘there is a little streate over Bradford bridge and at the ende of that is an hospital of the kinges of Englandes fun- dation’. Aubrey confirmed that ‘a little beyond the bridge is a chapell and almshouse of an old date’. Protection was in 1235 granted to the brothers and sisters of the leper’s hospital of St Margaret.*” The site was in or near the present St Margaret’s Street, and the accounts record that the brothers and sisters of St Margaret received a bushel of wheat each week. Wheat bread was a luxury item, so this was generous. Continuing on his way, the visitor would pass the entrance to Barton Farm, with quarries perhaps on either side of the road, though these are not specif- ically mentioned.
If, instead of crossing the bridge, the visitor had remained north of the river, going along what is now Church Street, he would have reached the lay court house. At the gate of the court were four shops and the manorial bakery, also the court pigeon house. The court house was furnished with a striped covering or housing and four cushions, suggesting a fixed bench for the steward. He was well fed on his visits with wheat bread and good beer, while the household staff had theirs made from ‘inferior mixtures’. Geese and capons were also provided for the steward’s dinners. The Saxon church in this area is not mentioned, but the vicarage attached to the parish church was excused an annual rent of 4s.
Only three streets are named in the accounts; Fox Street and Mill Street have already been mentioned. The third was Padyhonestret which, to date, has not
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
been identified. Its name survived as ‘Paddow’ or ‘Pando’ Street until the mid-17th century.
The house of St Mary was also excused its rent, of 12d. It is tempting to refer this to the chapel of St Mary, Tory, set on the western heights above the town, and called, in 16th and 17th century deeds, the ‘Hermitage’.*! By 1877 only the east wall remained, and the whole was then rebuilt as a chapel and adjoining cottage. The Perpendicular east window looks rather later than 1367, however, and a deed dated 1587 gives the dedication as St Leonard.*” Per- haps this name was lost at the period of restoration, and the chapel was then named because of the spring called Ladywell below it. If this is so, St Mary’s has not been identified. When Leland came this way, he thought that the steep hillside below the chapel would be a good site for a vineyard. Bradford had a vineyard in Domesday, at an unknown site, and the 12th- century custumals mention duties to be performed in it. By the time of the accounts it seems to have disappeared, and its memory lingered only in one or two men called Viner.
Although, as explained above, medieval accounts were designed to establish whether the lord was in debt to his servant or conversely, there obviously was interest in the manor’s profitability. This has been worked out in the account for 1376, where at the end,
the value (profit) of the manor was said to be £201 5s. _
2d., and of wool £8 7s. This point safely established, we can imagine the officers and auditors going off cheerfully to the audit dinner!
Appendix
INVENTORY OF THE ABBESS OF SHAFTESBURY’S POS- SESSIONS AT BARTON FARM, BRADFORD (translated from a reeve’s account dated 1367)
In the Reeve’s chamber 1 coffer with lock, 1 axe, 2 forms with curtains, 2 baskets for oat flour, 1 barrel for salt, 1 balance for weighing wool and other things, 1 table with | pair of trestles, 1 form with a curtain in the Hayward’s chamber with 2 little stools, 1 barrel for tar.
In the Dairy 1 dome (?follow) for making and pressing cheese, 1 tumbrell of 4 round stepped boards for holding cheese, 1 long hanging board for the same, 2 stools.
29. J. Aubrey and J.E. Jackson, Wiltshire: The Topographical Collections of Fohn Aubrey (Devizes, 1862), p. 21. 30. V.C.H. Wilts., Vol. 3, p. 334.
In the Brewhouse | furnace, 1 vat, 1 chest for bolting flour.
In the Malthouse 2 cribs for carrying malt, 1 wooden shovel for turning malt, 1 iron bound bucket with iron coterell for the well of the malthouse.
In the Kitchen 6 brass pots, 2 cruets, 1 old tripod, 1
gridiron, | small furnace for cooking meat, 1 mortar |
with pestle, 1 copper pan containing 5 gals, | ‘owol’, 1 iron bound bucket for the kitchen well, 20 dishes, plates and saucers of wood.
In the Bailiff's Little Cellar 1 great chest for storing |
bread, 2 flasks or flagons containing 9 gals, 4 barrels each containing 8 gals.
31. WRO 1742/3245 et al. 32. I[bid., 1742/3265.
BRADFORD-ON-AVON IN THE 14TH CENTURY
In the Granary | iron bound bushel for measuring the grain of the manor, agreeing with the Shaftesbury standard, and 1 bushel of wood without iron, 1 wooden peck.
In the Hall 3 tables with 3 pairs of trestles, 1 chair, 3 forms with curtains, 1 ewer and basin, 2 tablecloths with 1 towel, 4 cups for the high table with 1 hand-towel and 1 hemp cloth for the side tables.
In the Pantry | trough for keeping bread in, 2 iron candlesticks.
In the Bailiffs Chamber | coffer and 1 locked chest, 1 custumal, 1 rental, 1 wooden cupboard, 1 iron rod sealed with the Royal seal, 1 standard bushel, sealed with the Royal seal, 1 iron for sealing bushels and other measures of the hundred on behalf of the Lady, 1 standard gallon with the Royal seal, 1 winnowing
129
fan, 5 sacks which each contain 5 bushels, 2 three- pronged dung forks, 4 iron shovels.
Item 21 locks and keys viz to the Steward’s room 1, pantry 1, granary 1, little cellar 1, malthouse 2, Bailiff's room 1, kitchen 1, reeve’s room 1, hayward’s room 1, hall 1, shepherd’s room 1, chamber next to reeve’s stairs 1, Bailiff’s stable 1, carthouse 1, dairy 1, great barn 2, small barn 1, dovecote 1.
At Atworth 1 table with 1 pair of trestles, 1 vessel containing 3 gals, 1 lock for barn door, 1 for granary door, 1 for chamber door.
At Bradford in the hall 4 cushions, | striped housing or covering. Item 1] iron mallet, 1 crowbar, 1 iron wedge.
This last item may refer to a courthouse in Bradford, mentioned elsewhere.
Wiltshire Archaeological and Natural History Magazine, vol. 86 (1993), pp. 130-135
The Poet as Historian: Fresh Light on Urban Fire Damage in Elizabethan Hungerford
by NORMAN HIDDEN
Poetry may seem an unlikely source for historical fact. A Latin poem entitled ‘Hungerforda’ by the Elizabethan poet Daniel Rogers records the 16th-century devastation caused by fire to this town on the Wiltshire-Berkshire border. Yet no other account of a great fire there exists. The author’s research into a wide range of 16th-century local records has revealed a mosaic of individual properties mentioned casually as affected by fire at this period, all located within a particular area, which gives credence to the poet’s veracity. Rogers’ interest in the town may have been inspired by
his link with the Earl of Pembroke at Wilton.
Although ancient Greek mythology named Clio as the muse of history and of epic poetry, present day historians may only occasionally find in poetry histor- ical evidence of a factual kind. Plato’s view that poets are fictionalisers is more likely to appeal, since the magic of poetry depends so much on a subjective and often hyperbolic view of events which it purports to describe. Recently the writer discovered the existence of a 16th-century poem which told of a great fire in a town on the Wiltshire—Berkshire border, of which no other record seemed to exist. Although it seemed unlikely that the poet would invent a disaster which never occurred, the question arose of how much credence could be placed in the poem’s reference to it and where might some documentary evidence be found to confirm it?
The poem itself is listed in the Historical MSS Commission report on the Marquess of Hertford MSS, which refers to'a small manuscript volume in Latin of poems by Daniel Rogers entitled Urbes. The volume is now in the possession of the Henry H. Huntington Library, USA, whose curator has kindly supplied a photocopy of the poem. !
The author, Daniel Rogers, was born c.1538 at Wittenburg, the eldest son of John Rogers, who was one of the Protestant martyrs under Queen Mary. Through an introduction to the court of Queen Elizabeth and his knowledge of European languages, he became useful to her ministers and others engaged in European affairs. He was employed by Sir Henry Norris, the English Ambassador in Paris, between 1566 and 1570. In October 1574 he went with Sir
1. Afistorical] Mf[anuscripts] Cfommission], Fourth Report, p. 251; Hunungton Library, USA: HM 31188, f. 205.
William Winter to Antwerp and a year later accompanied an important embassy to the Nether- lands to treat with the Duke of Orange. Between that date and March 1578 he was engaged in diplomatic business in the Low Countries. In September 1580 he was sent to Germany, was arrested at the request of Philip II of Spain, and detained there in prison for four years. In 1587 he was appointed clerk to the Privy Council. He died 11 February 1591 and was buried in Sunbury, Middlesex.?
Although the Rogers MSS became part of the papers of the Marquess of Hertford, it is not clear that Rogers had any special relationship with the Seymour family. Indeed, the Earl of Hertford himself was in disgrace with the Queen and if Rogers had had any connection with the family, it could not have been advantageous to him at court or in his career. He was, however, on close terms with leaders of the dominant Protestant faction such as Cecil, Walsingham, Leicester and Pembroke. His poems are dedicated to a wide range of Elizabethan court notabilities, and the twenty-eight urbes that he celebrated in his poems of that name seem to represent towns in which many of them had a special interest.
These included the cities of London, Bristol, Bath, Salisbury, Warwick, Coventry, Oxford, Cambridge, Canterbury, Winchester and York, not to mention the towns of Northampton, Banbury, Bedford, Reading, Rochester and Colchester. Among these historic and illustrious towns the inclusion of the small market town of Hungerford is — startling. Entitled Hungerforda, the poem reads as follows:
2. D.N.B.
|
URBAN FIRE DAMAGE IN ELIZABETHAN HUNGERFORD
Bercia Vilugis disterminat arva colonis Hungerforda sui voce celebris her1,
Quam mediam fluidis Cunetio dissecat undis Navigero primis flumine nostris avis
Aede, schola, domibus, numeroso et cive decora, Igne sed immodico pene perusta fuit
Manorem et cladem domini de clade recepit Corruit a civis crimine facta nocens,
Urbibus exemplo reliquis, hos esse beatos Quarum cives fidos regibus usque colunt.
In the last line the word cives has been underlined and in the right hand margin the word dominos has been written in the same handwriting, suggesting that Rogers had toyed with the idea of substituting the one word for the other.
_ Translated, the poem reads as follows:
Hungerford
Hungerford, celebrated through the fame of its lord,
| Separates Berkshire fields from the settlers of
Wiltshire,
As the Kennet, whose waters bore the ships of
| our ancestors,
, Divides the town with its flowing streams.
Graced with church, school,
| numerous townsmen
__ It was almost utterly destroyed by a devastating
fire.
Yet it received a greater blow by the downfall of its lord
And it fell disgraced by the crime of its own townsman.
It is an example to all surviving towns
That those citizens are blessed who remain faithful to their rulers.
houses and
1 What may have brought Hungerford to Rogers’ notice was the fact that the Earl of Pembroke was High Steward of the manor there, a fact which leads to the ‘supposition that Rogers’ poem was designed for his eyes (or ears).
The date of the poem’s composition is not known. The volume in which it is contained is said to be of thick quarto paper, ‘last quarter of the 16th century’, and the contents to be in Rogers’ own handwriting.? Most poems are undated, but some in the same volume have dates ranging from 1570 to 1579. The ‘nternal allusion within the poem, if correctly inter- yreted, to the crime of one of its townsmen means that
3. H.M.C., op. cit. ;4. P.R.O., DL37/13 (Duchy of Lancaster Chancery Rolls).
131
it cannot have been written before the latter end of
1573, and the allusion is sufficiently topical to suggest
a date c.1574.
This and other allusions suggest that Rogers is likely to have known the town from_ personal acquaintance with it. As it lay halfway on the direct route from London to Bristol, and also halfway on the direct route from Oxford to Salisbury, travellers found it a convenient overnight stopping place. If Rogers visited the Earl of Pembroke’s seat at Wilton near Salisbury, an overnight stay in Hungerford would bring him within thirty miles, or a comfortable day’s journey. As Pembroke was High Steward of the town, Rogers would have a reason to make himself acquainted both with its topography and its history and even its internal politics.
The picture of the town’s topography given in the opening lines of the poem is undoubtedly true and exact. For Hungerford is partly in Berkshire and partly in Wiltshire, the two counties having met at this point for as many centuries as we have record. The river Kennet even now, after modern irrigations and water workings, has many branches, small tribu- taries and ‘flowing streams’. One such stream which flows through the town is a small tributary nowadays called the Dun, but in past times regarded as part of the Kennet. The confluence of these streams had provided the early Saxon settlers with a last landing place upstream before navigation became impassable. Hence Rogers’ reference to the river as navigero (line 4).
Lines 5 and 6 which relate that the town was almost utterly destroyed by fire are among the most fascinating in the poem. For there is no other known document which refers to such a catastrophe, yet the statement is too precise for it simply to have been invented. In a town of open hearths and many thatched roofs, occasional conflagrations were to be expected, yet the poem underlines a much greater disaster than the occasional loss of one or two isolated houses.
There are two other specific historical allusions, which are compressed into lines 7 and 8. The first, in line 7, to the fall or disgrace of the town’s lord, must refer either to one or other of the several attainders which befell the family of Hungerford or to that of Edward Seymour, the Duke of Somerset. The lord- ship of manor, park, and borough had been granted by Henry VI in 1446 to Sir Walter, Lord Hungerford.* His grandson, Robert, third Lord Hungerford, like his forebears was a firm Lancastrian supporter. Attainted under Edward IV in 1464, he
132 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE :
was beheaded and his lands became forfeit to the Crown.° The lands were restored to the Hungerfords by Henry VII but fell again into Crown hands by attainder under Henry VIII. The manor was then granted by the young Edward VI in 1548/9 to his uncle the Lord Protector Somerset. Somerset was overthrown in 1552 and he too was beheaded and all his lands and honours forfeited.®
To which of these downfalls does Rogers refer as maioram cladem, a disaster even greater than the great fire? The most likely is that which befell Somerset, an event which occurred in Rogers’ own lifetime and one which would still be live in the memory of the local townsmen. Somerset’s seat at Wulfhall in Savernake forest was only a few miles from Hungerford.
What of the event hinted at in line 8? Only one instance is known of harm done to the town by one of its townsmen. This occurred in 1573 when the town’s charter was allegedly stolen by one of the townsmen, against whom a suit was brought by the town’s officers in the Duchy of Lancaster court. The court, however, found (and this was very much to the Duchy’s interest) that there was no evidence that such a charter ever actually existed.’ The townsmen con- sequently faced loss of all their ancient borough rights. In desperation they appealed to the Earl of Pembroke as High Steward of the borough to plead on their behalf with the Queen. This Pembroke found an opportunity to do when the Queen visited him at Wilton in September 1574, and he was successful in eliciting from her a masterly Delphic and diplomatic reply to the effect that ‘the inhabitants should have, use and enjoy without interruption all such liberties and profits and benefits as heretofore time out of mind and remembrance of man they had used and en- joyed’.* But what those liberties, profits and benefits actually were the Queen left the inhabitants and the Duchy to work out for themselves.
Returning to the reference in lines 5 and 6 toa great fire, there is no documentary record of a great fire as such in Hungerford, but a survey of town properties taken in 1573 reveals a disproportionate number of entries “decayed by fire’, ‘one void plot of ground late burned’, ‘a decayed piece of ground late burned’.? These entries concern six of the town’s burgage tenements out of a total of about 95, and the nature of the entries suggests they refer to buildings completely gutted. Six properties out of some 95 may not seem a
VCH Berkshire, Vol. IV, p. 188.
Loc. cit..
P.R.O., DL6/22 (Duchy of Lancaster Draft Decrees). VCH Berkshire, Vol. IV, pp. 185, 186.
CON NW
major disaster. The figures become more impressive, however, when the concentrated location of the burnt down dwellings is taken into account. They corre- spond to present day property sites on the east side of the old High Street (two burgages) and on the west side (four burgages). None of these affected proper- ties had been rebuilt in 1573 and must have been regarded as uninhabitable at the time of the survey.
All of them were situated at the lower or northern end of the town. The four properties lying on the west side of the High Street as it then was called (nowadays this portion is known as Bridge Street) stretched from the southern bank of the stream known today as the river Dun and extended over what is now the canal. It is at this point, from the southern bank of the canal, that the modern High Street begins. The southernmost of the four properties lay approximately on the site of present day nos. 6 and 7 High Street. Almost opposite this point were the two devastated houses on the east side of the street, where the Post Office now stands.
The town survey in 1573 gives no indication other than the word ‘late’ of how long previously the fire had occurred, but it must have been later than 1552 when a previous survey was made which contains no mention of any burnt dwellings.!° Another section of the survey of 1573 is concerned with the Queen’s water mill which, says the survey significantly, ‘is now in good reparation’. It may be assumed from this that the fire had damaged the town mill and that full repairs had been made which enabled it to continue its function by 1573. This assumption is reinforced by the enquiry of a commission in 1576 into the lands of the former priory of St John in Hungerford. This revealed that of three tenements which belonged to the priory, two were standing but one was ‘decayed by fire’.'' These three tenements were immediate neigh- bours to the mill.
We have here two additional items not specifically referred to as burnt in the 1573 survey, and it becomes clear that the fire was greater than the six particular references in the survey might at first suggest. Indeed, there was no necessity for the survey to refer to burnt properties at all, since quit rent was due to the Crown whether a site was occupied or vacant, whether it was used for dwelling house, business, or agriculture, or simply left idle.
Because it was Crown property and because of its economic importance to the town, it is likely that the
9. Berks. R.O., HMS/1.
10. P.R.O., DL42/108 (Duchy of Lancaster Miscellaneous Books).
1l. P.R.O., E178/2848 (Exchequer K.R. Special Commissions of Enquiry).
URBAN FIRE DAMAGE IN ELIZABETHAN HUNGERFORD
\ Berkshire 2oe IN ~ S88 #8 6 ¢ 6006, \EDDINGTON SN \
To Oxford & London
Y,
Wiltshire
on
HARNHA
; To Marlborough i ]
REET
HIGH ST
eeeee County boundary
To Salisbury
| Street map of Hungerford showing houses (in black) | reported as affected by fire
12. P.R.O., REQ2/240/19 (Court of Requests Proceedings). 13. P.R.O., DL4/15/6 (Duchy of Lancaster Depositions Series II).
133
mill was repaired or rebuilt with particular urgency and speed. A law suit in which miller John Yowle was involved in 1570 throws light on both the date and the origin of the fire. Yowle had acquired lease of the mill by marrying the widow of the previous miller John Austen, who died in 1560.!* As Yowle was only twenty years of age at the time of Austen’s death, his marriage to widow Austen and consequent tenancy of the mill may not have been immediate. In or about the year 1565, however, we know that he went to South- ampton to purchase a mill stone.! If this date indic- ates an approximate commencement of his interest in the mill, we may be able to place the date of the mill fire as between this purchase in 1565 and the date of the law suit referred to above in 1570. For in this suit Yowle claims that ‘by misfortune and negligence of his neighbours, the said mills were burnt and utterly consumed with fire. After which time . . . [he] did re-edify the said mills, which cost him [one hundred pounds] or very near thereabouts’.
This fire which ‘utterly consumed’ the mill and which began ‘by misfortune and negligence of his neighbours’ may well have been the starting point of the great fire of which Rogers’ poem makes mention. Neighbouring the mill on its northern side were a brewhouse and a dye house, both using furnaces for their operations.
Confirmatory evidence of other destruction by fire at this date is provided by the draft of a Crown lease dated 8 July 1566.'* By this lease the Crown let to Henry Edes a number of properties in Hungerford which had formerly belonged to two recently dissol- ved local chantries. These properties, about twenty in number, were scattered on either side of the old High Street. In a different hand from that on the main body of the lease the following postscript had been written: ‘Memo: there is six of the tenements belonging to these chantries burnt’. And Henry Edes was made responsible for the rebuilding or repair, at his own cost, of the houses which were ‘of late burnt’. A court case in 1569 confirms that he had covenanted in his lease ‘to build, make up, and re-edify certain burnt and decayed houses and tenements’ among the former chantry properties and it seems that he had as yet not done so.!°
It is not known which of the chantry properties were the six that had been burnt, but it is possible to determine the position of chantry properties in gen- eral from the 1573 town survey. Since the survey does
14. P.R.O., DL14/55/2 (Duchy of Lancaster Drafts and Particulars for Leases). 15. P.R.O., DL1/79E2 (Duchy of Lancaster Pleadings).
134 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
not mention whether burnt or not in relation to any of the chantry properties, it is clear that these burnt tenements acquired by Henry Edes must have been additional to the six non-chantry properties which the survey does mention as having been burnt. Assuming that the chantry tenements affected by the fire were most likely to have been in the area where all other reported damage had taken place, this would be in the northern section of the old High Street, below Church Street on the west and Cow Lane (now Park Street) on the east. On the west side, below Church Street, there were only three chantry properties used as dwelling houses. Two of them were in the immediate vicinity of the mill, while the third was on the site of present day nos. 13/14 High Street. On the east side, below the old Cow Lane, there were three chantry properties on the site of present day nos. 115, 117, and 118.
The difference of time between the date of the postscript to Henry Edes’ lease of 1566, the report of houses ‘late’ burnt in the survey of 1573 and of the commission enquiry into priory lands in 1576 is not greatly significant for, as may be seen from Yowle’s evidence, the cost of rebuilding could be considerable and few townsmen could afford to go about it so rapidly as circumstances had obliged Yowle to do. Edes’ court case in 1569 shows how reluctant land- lords were to set about re-building.
Indeed, the next town survey, 1n 1591, shows that of the six properties reported in 1573 as ‘decayed by fire’, at least four were still in the same condition. !° In one case the 1591 survey makes explicit that the ‘void plot of ground late burned’ (1573) was the site of a building ‘whereon sometime a house stood’. Even as late as 1609 a further town survey describes one of these same houses as ‘lately burned’. Restoration and re-edification was a delayed and expensive process in days when no insurance companies were to hand to meet the costs of rebuilding.
With the six chantry properties the total of houses burned has now increased to fourteen. In addition, there exists a set of surveys and rent rolls for the estate of the Hungerford family (at that time Sir Walter Hungerford Kt.). This estate was known as the manor of Hungerford Engleford (or Englefield). There is extant a manorial survey of 1583 which records two more instances of houses ‘lately burned’ that were not mentioned in the town survey. Both of these were at the northern end of the town. In one of the two cases some rebuilding had taken place, for the entry reads:
16. P.R.O., DL42/117 (Duchy of Lancaster Miscellaneous Books). 17. Wilts. R.O., 442/1.
‘one cottage lately burned and a_ backside whereon is now a dwelling house builded’. !”
Additional, but rather more circumstantial, evid- ence of the extent of the fire damage may be evinced by the Hungerford Engleford court rolls, which reveal a sharp concern for fire regulations to be observed by the tenants. These court rolls do not exist for the period suggested above as the probable date of the fire, but when they recommence at a court held in April 1593 it was presented that ‘Thomas Curr lit fires in his house without a chimney or “‘flewe’””’. For this he was threatened with loss of his customary tenancy in the premises. In the following year it was ordered that William Parre should not make a fire in his tenement called ‘le bakehouse’ without a chimney or flue on pain of his loss of the tenancy. George Burrowe was likewise warned to make himself a chimney or flue, the lord granting him an elm tree for the purpose. In the next court roll it was presented that George Burrowe still had not made himself a chimney or flue in his house nor in the section of it which he had sub-let to his mother-in-law. Accord- ingly, he was adjudged to have forfeited his tenancy.'®
These presentments and orders suggest an alertness to the dangers of fire which had impressed itself on the minds of the jurors and of the lord’s steward alike. Indeed, since no similar presentments occur in other Hungerford Engleford court rolls, neither earlier in the century, nor later in the next century, it may be assumed that their incidence at this particular date was a result of men’s memories, even some twenty- five years or more later, being still vivid of the great fire which had caused so much damage and inci- dentally had impressed a visiting poet.
Another remarkable instance of the length of time with which the memory of natural disasters remained impressed upon men’s minds may be seen in a landlord and tenant court case in Hungerford in 1610. The case also illustrates how slowly rebuilding after such a disaster could proceed. In this case! the landlord complained that his lessee had agreed to repair the building if the landlord would provide the timber. This the landlord says he did ‘to the amount of 20 tons at least’. The tenant then proceeded to work on the timber on what we would nowadays call a do-it-yourself basis, making a frame or fabric of the timber. He pulled down a greater part of the roof which was at that time covered with tiles, together with one double chimney, ‘intending to re-edify the
18. Wilts. R.O., 490/1539. 19. P.R.O., C2/Jas I1/M20/9 (Chancery Proceedings Series I).
URBAN FIRE DAMAGE IN ELIZABETHAN HUNGERFORD
house’, but suddenly all work stopped. The landlord complained that the tenant
has kept the frame un-set up and has made havoc and spoil of the timber, allowing the frame to be withoutdoors and uncovered and so subject to rain and weather for these many years . . . the several rooms where the building did formerly stand do yet lie open and uncovered for eight or nine years and [the tenant] has prepared and made fit only a little room for his trade and kept it thatched, to the hazard and danger of all other the inhabitants and neighbours their habitations, being once heretofore as he [the landlord] hath credibly heard set on fire and in great danger of consum- ing by means therefore.
The landlord would not himself have experienced this great fire, since he was a newcomer to the town, but he had apparently heard of it and if the reference is indeed to the great fire then his was one more house that was so affected. That these references to fire damage all relate to a great conflagration in the late 1560s seems probable not only for the reasons adduced, but also because all the instances of damage may be pin-pointed to an area at the northern end of the town. No reference has so far been found to fire damage which might have occurred to buildings beyond (i.e. south of) the Cow Lane — Church Street intersection. The supposition occurs that the width of these intersecting streets may have acted as a natural fire break.
At the time there were probably no more than 95 dwelling houses in the old High Street. The exact figure is uncertain because a few quit rent roll entries may be for closes of land on which no building was ‘sited. Of these estimated 95 dwelling houses, about half lay in the northern section of the street, and in ‘this northern section we have documentary evidence
for fire damage in the case of the following buildings: 1573 town survey 6 St John’s priory commission l mill 1 former chantry houses 6 ‘Hungerford Engleford Manor 2
This gives a total of 16 instances which have been discovered, a substantial proportion of the total buildings in this area. Even so, there are likely to have been more.
| The location of these instances on a street map
suggests a pattern of fire commencing in the neigh-
135
bourhood of the mill and carried by a strong northerly breeze up the main street. It seems inconceivable that some additional houses were not affected where they were located between those which are known to have been destroyed or damaged. The odd house or two may have been leapfrogged in such a wind-swept inferno and perhaps escaped destruction due to luck, or the possibility of a house being built of stone or brick or having a tiled roof, but there would not be many of these. Destruction must have been nearly total. The whole northern end of the town may have resembled something like the modern waste of a wartime bombed site. This northern end was the main entrance to the town from both London and Oxford and if Daniel Rogers passed through the town he would have seen at first hand some of the scars the fire had left, enough to justify his poetic licence in describing how the town igne sed immodico pene perusta fut.
A local historian may feel indebted to Rogers, for without his reference to the great fire such scattered allusions to burnt dwellings might have been passed over as a series of unrelated individual events. His reference to the existence of a school is also helpful, for we have no documentary evidence of the existence of a schoolmaster before 1633 or of a school before 1636, an absence which was puzzling in a market town the size of Hungerford.
In rendering thanks to Rogers for the otherwise unknown information he has given concerning these details in the history of the town, we should not lose sight either of his wider perspective as a courtier or of his skill as a poet. His perspective as a courtier may be seen in the concluding lines of the poem, which contain its moral. The compact Latin construction does not lead to easy translation; but the meaning though primarily expressing a general sentiment likely to be acceptable to Pembroke (or, if dominos is substituted for cives, to Elizabeth), is also a specific reminder to the townsfolk of Hungerford of their responsibilities to their High Steward. This compactness of language and tightness of structure while presenting vivid pictures and clarity of thought, are a tribute to the poet’s skill. A vast amount of story telling and moralising is compressed into a mere ten lines, and the firmness of the moral conclusion is admirable. Perhaps topographical poetry may be wel- comed as an adjunct to history after all?
Acknowledgement. I am indebted to Mr John Brooks for drafting the map which accompanies this article.
Wiltshire Archaeological and Natural History Magazine, vol. 86 (1993), pp. 136-141
The Revd William Lisle Bowles (1762-1850): The Need for a Re-appraisal
by DOREEN SLATTER
W.L. Bowles was a man of varied talents whose career embraced several different fields of activity, not all properly appreciated. The interest of his achievements as a clergyman, historian and musician has been obscured by the attention given to his literary work. Further research 1s needed into the details of the first half of his life in order to place the information given in Thomas Moore’s Memoirs relating to the second half in a proper perspective. Bowles’ relations with the Lake Poets and with Lord Byron have not been dealt with. It is suggested that Bowles’ importance in the intellectual, social and ecclesiastical life of Wiltshire will not be fully realised until a wider and more balanced
study of his career has been made.
Sir Uvedale Price, the author of An Essay on the Picturesque (1794), was invited to Bowood in the autumn of 1824! to meet a ‘Grand Chorus of Bards’, namely Samuel Rogers of London, the Revd George Crabbe, vicar of Trowbridge, the Revd W.L. Bowles, vicar of Bremhill, and Thomas Moore of Sloperton Cottage near Bromham. On both the national and the local scene, the reputations of these poets stood high with their contemporaries. Bowles had sprung into public notice as a young curate on the publication in 1789 of his Fourteen Sonnets written at Picturesque Spots on a Journey. He had continued to write and had incurred some notoriety because of his quarrel with Lord Byron and others over views expressed in his edition of the works of Pope, published in 1806. Bowles, Crabbe and Moore were invited to local literary gatherings as the famous Wiltshire poets” and they seem to have cherished great regard for each other. They were mentioned in biographical diction- aries of living authors,’ while The Gentleman’s Maga- zine of 1814 included a description by his friend, Archdeacon Robert Nares, of the ‘picturesque’ garden Bowles created at Bremhill vicarage.* Nares, one of many well known visitors, declared that “To the traveller of cultivated mind. . . there are no spots
1. P.W. Clayden, Rogers and His Contemporanes (2 vols, London, 1889), Vol. 1, pp. 387-8.
2. For example, Bowles, Crabbe and Moore attended a dinner in honour of the opening of the Bath Institute in January, 1825. A bronze tablet commemorating this event was unveiled by Louis Huchon in 1905: WAM 34 (1905-6), p. 230.
3. Such as J. Watkins and F. Schoberl, A Biographical Dictionary of living Authors of Great Britain and Ireland (London, 1816). Twelve other works, published between 1799 and 1885, are listed by D. Bank and A. Esposito (eds.), British Biographical
more attractive than those which are distinguished by the abode of Genius’. At the present time the three Wiltshire poets are not as widely known as they were, Bowles probably being the least appreciated. Such notices of him as have appeared in this century usually refer to the influence of his Sonnets on the young Coleridge or to his differences with Byron, while his own talents and interests have been largely neglected, damned with faint praise or even rejected as worth- less. Cecil Woolf, who published what he described as a ‘check-list’ of Bowles’ works in 1958,° considered that the events of his life were of little importance and that his prose writing was ‘now unreadable’. It will be suggested that this severity is not justified and that Bowles’ career should be examined on a broader front to show that it is of greater interest, especially for local history, than has been recognised.
Woolf made it clear that bibliographical work remains to be done to establish a complete list of Bowles’ printed works. Bowles’ name appears inter- mittently in modern histories of literature and espe- cially in biographies of Coleridge.° However, there has been only one full length book about him, published over sixty years ago, which is now inevit- ably in need of revision. This volume, by Professor
Index, 4 vols. (K.G. Saur, London, Melbourne, Munich and New York, 1990), Vol. 1, p. 213.
4. The Gentleman’s Magazine 84 (1814), pt. 2, pp. 203-4. Bowles included a detailed description of his garden in his book The Parochial History of Bremhill (London, 1828), pp. 249-53.
5. C. Woolf, ‘Some Uncollected Authors xviii: W.L. Bowles’, The Book Collector 7 (1958), pp. 286-94, 407-16.
6. Recently in Richard Holmes, Coleridge: Early Visions (Penguin Books, 1989).
7. The book was reviewed in WAM 43 (1925-27), pp. 562-4.
THE REVD WILLIAM LISLE BOWLES
Garland Greever of Harvard University, entitled A Wiltshire Parson and His Friends (Constable, 1926),’ was based on a collection of Bowles’ correspondence then in the hands of the Jones-Bateman family of Pentre Mawr, Abergele, North Wales. It comprises an appreciation and short life of Bowles, followed by transcriptions of letters from his more well known literary friends. It does not include letters from members of his family, and local interest is repres- ented only by letters between Bowles and his wife and the third Lord Lansdowne and his wife, about an un-named village school. In general, Wiltshire gentry and clergy are not represented. The detailed bio- graphy of Bowles contemplated by Greever® seems not to have been completed, a fact which may explain the selective character of the 1926 book and its lack of detailed references. As the result of a sale by Sothe- by’s in 1936,’ the Pentre Mawr collection has now unfortunately been dispersed to different repositories in England and America: the full story remains to be discovered. A few small collections of Bowles’ papers exist in this country, such as the album of poems and miscellaneous material which was purchased by the National Library of Wales in 1934!° and a collection of drafts, proof corrections and items collected by Bowles, acquired by the Bodleian Library in 1973 from the papers of Professor W.L. Renwick (1889- 1970).!' Manuscript material relevant to Bowles’
| career as a clergyman and a magistrate must also exist
iI
among parish, diocesan and quarter sessions records, held in local record offices, not in existence when Greever did his research.
Bowles’ published works!” provide a little direct autobiographical information and a number of per- sonal allusions can be found in his poems. The Revd George Gilfillan brought out a two-volume edition of Bowles’ collected poems in 1855, in the introduction to which mention was made of a Life about to be
| written by his ‘kinsman’, Dr John Bowles, assisted by | Alaric Watts.!? This seems never to have been pro-
duced but information about his family background
| can be found in county histories, the Alumni Oxonien- | ses and in volumes published by the Harleian Society. | Hutchins’ History of Dorset shows that W.L. Bowles |came from a respected old landed and professional
8. G. Greever, A Wiltshire Parson and his Friends, Preface, p. ix.
9. Sotheby and Co., Catalogue of Printed Books, Illuminated and other Manuscripts, autograph letters and historical documents, 1936, pp. 53-5.
10. National Library of Wales MS. 3600B.
\11. Bodleian Library, Arch. Hc.2(i) and MS. Eng. poet. c. 64-5. }12. See the introduction to Scenes and Shadows of Days Departed
Accompanied with poems . . . (London, 1837).
137;
family centred on the Dorset—Wiltshire border at Shaftesbury and Donhead St Andrew. Genealogical tables of this family and another branch at Salisbury and Heale House, Woodford, can be found in the History of Chalk Hundred (1833),'* written for Sir Richard Colt Hoare’s History of Modern Wiltshire by Charles Bowles, youngest brother of William Lisle and Recorder of Shaftesbury. Their father, the Revd William Thomas Bowles, vicar of King’s Sutton, Northamptonshire, and later rector of Uphill and Brean, Somerset, lived mainly at Barton Hill House, Shaftesbury,!° which he is said to have ‘planted and improved’. Their mother, Bridget, was the daughter of Dr Richard Grey, archdeacon of Bedford, author of Memoria Technica (1730). Their grandmother, Eliza- beth, wife of the Revd Thomas Bowles, was the daughter of William Lisle of LEvenley, North- amptonshire, M.P. for Brackley.'® William and Charles had a brother, Henry, an army doctor, whose career was cut short by his death at sea returning from Gibraltar in 1804. (He was commemorated by an inscription, one of several ornaments, in his brother William’s garden at Bremhill.)!” Charles died in 1837 leaving a daughter, Margaret, but W.L. Bowles’ papers descended to his niece, Marianne Burlton, daughter of his sister, Sarah, who married William Burlton of Wykin Hall, Hinckley, Leicestershire. Bowles’ youngest sister, Amy, married the Revd Peregrine Bingham, the writer, rector of Berwick St John. In his Scenes and Shadows of Days Departed (1837), Bowles himself suggested that he had inher- ited his love of landscape scenery from his father and his musical ability from his mother. From her father, he also inherited the fiddle once played ‘out of tune’ by Nathaniel, Lord Crew, Bishop of Durham! Coming from such a family and with such talents, Bowles was sure of a place in county society.
He was educated at Winchester College, of which his great-uncle, William, had been a Fellow, proceed- ing to Trinity College, Oxford, where he obtained his B.A. degree in 1786. At Winchester, Bowles was a favourite pupil of Joseph Warton, poet and critic, and at Trinity he came under the influence of Joseph’s brother, Thomas, historian of English poetry. His education and advancement in the Church were inter-
13. G. Gilfillan (ed.), The Poetical Works of William Lisle Bowles (2 vols., Edinburgh, London, Dublin, 1855), Vol. 2, p. xiv.
14. History of Chelk Hundred (London 1833), pp. 35-6.
15. J. Hutchins, History and Antiquities of the County of Dorset, 3rd ed. (4 Vols., 1861-4, repr. 1873), Vol. 3, p. 41.
16. H.I. Longden (ed.), Visitation of the County of Northampton, 1681 (Harleian Society, Vol. 87, London, 1935), pp. 122-4.
17. Hutchins, op. cit., Vol. 3, p. 112.
138 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
rupted at this point by the sudden death of his father in 1786 and an unhappy love affair, which caused him to travel and inspired his Sonnets of 1789. Printed references to him during the next fifteen years give only a sketchy account of his life and further research in diocesan, parish and other records is necessary to clarify it. According to Greever, Bowles was ordained deacon on 18 May 1788 to the curacy of East Knoyle but he went to live at Burlton’s, a farm house not far away in the parish of Donhead St Mary.!® He is said to have acted as curate at Donhead St Andrew in 1792 and the entry in the Dictionary of National Biography states that this was where he ‘first officiated’ as a clergyman. The incumbent at East Knoyle was Dr Charles Wake, prebendary of Westminster, to whose daughter, Harriet, Bowles had become engaged by 1792. She died, however, early in the following year and after some opposition, Bowles married her sister, Magdalen, in 1797. He seems to have continued to live at Burlton’s,!? although appointed rector of Chicklade in 1795 and rector of Dumbleton, Glou- cestershire, in 1797. His mother, Bridget, died this same year and he then probably sold Barton Hill House, Shaftesbury. He became vicar of Bremhill in 1804, though he may not have gone to live there until 1805.
His edition of the works of Alexander Pope was published in 1806 in ten volumes. The extent of his literary contacts during this period can be deduced from references in the works and letters of contem- porary writers, more editions of which now exist than were available to Greever. Acknowledgements and dedications in Bowles’ own works also provide the names of friends and correspondents, one of whom was the Roman Catholic, Lord Arundell of Wardour. For help with his edition of Pope, he thanked Henry Penruddock Wyndham, M.P., the Revd William Coxe, archdeacon of Wiltshire, the Revd Henry John Todd,”° David Parry Okeden,”! Sir Uvedale Price, the Revd Richard Polwhele and Alexander Chalmers. The names of other scholars and historians are given in Bowles’ later works. His Annals of Lacock Abbey
18. Greever, op. cit., p. 6.
19. The article on Donhead St Mary in the Victoria County History of Wiltshire, Vol. 13 (1987), p. 141, states that Bowles lived at Burlton’s farm between 1788 and 1797. Lord Arundell, in his History of the Hundred of Dunworth (1829), p. 36 (for Sir Richard Colt Hoare), calls the house ‘a small mansion commanding some very beautiful views’.
20. Printed as ‘Rev. Mr. Tod’, presumably Henry J. Todd (1763— 1845), librarian at Lambeth Palace.
21. Printed as Parry Okeden, probably David Parry Okeden (1774— 1833) of Turnworth, Dorset. (Hutchins, Vol. 3, p. 469).
22. Baker acknowledged information from Bowles and included a short tribute to him in his account of his birthplace, King’s
was produced in 1835 with the co-operation of the young John Gough Nichols and with a contribution from Henry Hatcher. Bowles, in his turn, was some- times able to give information to other writers, such as George Baker, the historian of his native county of Northampton.”
A most important contact, made at an early period, was with the playwright, theatre manager and politi- cian, Richard Brinsley Sheridan. In 1797 Coleridge consulted Bowles about his play Osorio, a copy of which was despatched by Bowles to Sheridan.?* The origin of the acquaintance that led to this is obscure, but it may have arisen from contacts between Bowles and the Revd John Savile Ogle, rector of East Knoyle. He was a brother of Hester Jane Ogle, daughter of Dean Newton Ogle of Winchester, who married Sheridan as his second wife in 1795. Dean Ogle was also precentor or rector of St Mary’s church, South- ampton and had a house there. Bowles, who kept in touch with Winchester College, would have known the family at least by reputation. He dedicated his Sonnets, reissued with seven more poems in 1789, to Dean Ogle. According to Thomas Moore’s biography of Sheridan,*+ Bowles frequently joined him and the Ogles on sailing excursions in the vicinity of South- ampton. Another companion was the Revd William Howley, later Archbishop of Canterbury, who had been at Winchester with Bowles. An undated poem by Bowles about a water party on the Beaulieu river?> mentioned Mrs Sheridan and her brother, Nathaniel. Sheridan admired Bowles’ poetry and also appreciated his interest in and talent for music. The name of another musical correspondent, Sir Henry Bishop, occurs in the catalogue of the 1936 sale of Bowles’ papers. This acquaintance is likely to have been due to Sheridan, as Bishop was closely connected with the Drury Lane Theatre. Bowles had his own contacts with William Linley, the writer and composer, brother of Elizabeth, Sheridan’s first wife. Linley also enjoyed Bowles’ poetry and would have liked to set some of his verses to music.7° There may be a connection between Bowles, Sheridan and the second
Sutton. The History and Antiquities of the County of Northampton, 2 Vols. (London, 1822-1841), Vol. 1, pp. 702-3.
23. M. Lefebure, Samuel Taylor Coleridge (Gollancz, 1974), p. 249.
24. Thomas Moore, Memoirs of the Life of the Right Honourable Richard Brinsley Sheridan (5th edn., 2 Vols., London, 1829), Vol. 2, p. 254.
25. Gilfillan, op. cit., Vol. 1, pp. 134-5.
26. Greever, op. cit., pp. 154-5, printed a letter from Linley to Bowles, 23 July 1797, saying he would like to have set some of his poems to music. Bowles wrote a poem on the death of Linley included with his obituary tribute in-The Gentleman’s Magazine 3 (1835), n.s., pp. 574-6.
THE REVD WILLIAM LISLE BOWLES
Lord Lansdowne, who abandoned his estate in Wil- tshire in order to live in the Gothick style castle he built at Southampton.?”? Bowles wrote a poem on ‘Southampton Castle’,?* emphasising its picturesque character: ‘The moonlight is without; and I could lose / An hour to gaze, though Taste and Splendour here, / As in a lustrous fairy palace, reign’. Sheridan and the second Marquess both supported the Whigs and both enjoyed sailing. Lord Lansdowne attempted to build an improved version of his ancestor, Sir William Petty’s ‘double-bottomed’ boat at Southampton. According to Thomas Moore, Bowles later told the story of the failure of this enterprise and the account suggests that he may even have witnessed the capsize of the new boat.’?? Two letters from the second Marquess to Bowles,*” dated 1806 and 1808, survive in the collection of letters between Bowles and the Lansdownes now in the W.L. Clements Library, Ann Arbor, U.S.A.
There is no doubt that through the fame of his own work and the friends and acquaintances he made in Wiltshire, Hampshire and social centres such as Bath, Bowles became known to a widening circle in the west of England, London and the country at large. It was thus possible for him to be of assistance to others. He may have helped to introduce Samuel Rogers to George Crabbe soon after 18107! and later, the ded- ication to him of Louisa Stuart Costello’s poems, Songs of a Stranger (1825),** acknowledged her grati- tude and esteem. It is not clear how Bowles came to know Thomas Moore but they were both important literary figures before they moved to live near Bowood. Bowles took part in the search to find Moore a suitable house.*? Bowles’ name occurs in the published correspondence of Samuel Rogers and Maria Edgeworth, but Moore’s Memoirs and Journal provide the most important information about him in later life and the place he occupied in the circle of literary and other figures gathered round the third Lord Lansdowne and his family. Moore was some years younger than Bowles but shared his love of ‘music and interest in ancient buildings. The Memoirs sive details of the social intercourse between the two
|
7. According to the Victoria County History of Hampshire, Vol. III, (1908), pp. 499-500, Lord Wycombe bought the Castle Hill site ! in 1804 before the death of his father, the first Lord Lans- downe. 3. Gilfillan, op. cit., Vol. 1, pp. 164-6.
?. Lord John Russell (ed.), Memoirs, Fournal and Correspondence of | Thomas Moore, (8 Vols., 1853-6), Vol. vi, p. 341. _). Mentioned in the sale Catalogue of 1936. I am grateful to Mr | R.S. Cox of the W.L. Clements Library for information. |. P.W. Clayden, op. cit., Vol. 1, p. 242. The author speculates | that Rogers met Crabbe about 1810. Crabbe writing to him
139
The Revd William Lisle Bowles. Engraving by J. Thomson, after Mullar, 1820. (Photograph by Derek Parker)
men and their wives at Bremhill Vicarage and Sloper- ton Cottage as well as at Bowood, but they do not deal with Bowles’ more serious activities, such as his work as a clergyman and a magistrate. Some anecdotes also reveal that Bowles had become mildly eccentric and was given to extreme absent mindedness. These stories have probably led to too much emphasis being placed on Bowles’ oddities of behaviour, which increased as he grew older and became deaf.** How- ever, the Memoirs leave no doubt that Moore had a sincere regard for his friend. On 1 September 1818 he wrote that he had been interrupted by Bowles ‘who never comes amiss; the mixture of talent and simpli- city in him delightful’.*° The ‘simplicity’ of manner may have been coupled with something slightly unu- sual in his appearance, as according to Moore, he was reluctant to be measured by a tailor for a suit of
about this time anticipates the pleasure of ‘meeting a gentleman to whom Mr. Bowles has given him hope that he should be introduced’.
32. A poem in Songs of a Stranger, p. 132, reveals that Louisa Costello (1799-1870) visited Mr and Mrs Bowles at Bremhill.
33. Clayden, op. cit., Vol. 1, pp. 250-1. Bowles wrote to Rogers on 1 Oct. 1817 about a possible house for Moore at Heddington.
34. See, for example, June Wilson (now Badeni), ‘A Georgian Eccentric’, Country Life, Vol. 120 (11 Oct. 1956), p. 771.
35. Moore, Memoirs, Vol. ii (1853), p. 153.
140 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
clothes.*° A portrait of Bowles was published by Henry Colburn in the New Monthly Magazine, Nov. 1820 (see Figure, above).*” It does indeed suggest that the coat worn by the kindly looking gentleman did not fit him very well. The sketch of Bowles in travelling clothes by Daniel Maclise, published in Fraser’s Magazine in 1836,*® in the series of portraits of literary figures, shows him wearing old fashioned knee breeches and a cloak.
Magdalen Bowles gave her husband devoted sup- port, both in his parish work and in the entertainment of his friends who often remembered her in their letters. She seems to have had creative talents of her own as she published three little books about village life between 1831 and 1836. She predeceased her hus- band on 7 May 1844 and his health then completely gave way. Bowles was obliged to resign his living of Bremhill and to retire to his canonical house, 21 The Close, Salisbury,*? the use of which he had enjoyed since becoming a canon residentiary in 1828. He died on 7 April 1850 and was buried, like his wife, in the cathedral. A lengthy obituary was published in The Gentleman’s Magazine,*® most of it taken from an address given by Henry Hallam, the historian, to the Royal Society of Literature, of which they had both been members. After praising Bowles’ Sonnets as being ‘among the first fruits of a new era in poetry’, Hallam went on to consider his life as a clergyman. ‘Whilst resident at Bremhill, Mr. Bowles was unre- mitting in his professional duties, zealous in the education of the poor’ and taking ‘a warm interest in the welfare of the rural population . . ., on more than one occasion he exerted his influence as a county magistrate . . . to demand the exercise of the prerog- ative of mercy’. This aspect of Bowles’ life has not
36. Ibid., Vol. vu (1856), pp. 162-3.
37. A portrait, apparently re-engraved for a smaller volume by William Humfrys after the same original, forms the frontispiece to Bowles’ Scenes and Shadows of Days Departed (William Pickering, London, 1837).
38. A Gallery of Illustrious Literary Characters (1830-38) ... (Reprinted from Fraser's Magazine), ed. William Bates (London, 1871), Portrait no. 70.
39. One of the oldest houses in Salisbury Cathedral Close, known from the fifteenth century as Aula le Stage. See article by C.R. Everett, WAM 49 (1940-42), pp. 288-312.
40. The Gentleman’s Magazine 33 (1850), n.s., pp. 672-6.
41. For example, he contributed to The Chimney Sweeper’s Friend and Climbing Boy’s Album, 1824, at the request of the editor, James Montgomery: Letters of William and Dorothy Wordsworth, 1821-50 (Clar. Press, 1939), Vol. 1, pp. 136-7, f.n. 2. Several poems and sermons were printed for the benefit of good causes, e.g. Verses on the Benevolent Institution of the Philanthropic Society for Protecting and Educating Children of Vagranis and Criminals (1790).
42. Bowles, A Word on Cathedral-Oratorios and Clergy Magistrates (London, 1830), p. 33.
been studied and it would be necessary to consult surviving local records and newspapers to have cor- roboration of Hallam’s opinion. There is evidence, however, that Bowles was always ready to preach or write on behalf of good causes and that he had a general sympathy with the unfortunate.*! In 1830 he defended the appointment of clergymen as magis- trates, saying ‘many of my poor people and many in the parishes around me would have died of hunger if the clergyman had not been a magistrate’.*?
Bowles’ varied talents, personal energy and enthusiasm made an important contribution to the intellectual and social life of Wiltshire and the adja- cent counties. More could be said of his musical abilities and interests*? and the manner in which he exercised them. A recent article shows that the last word has not yet been written about his importance in literary history. His prose works should be seen as examples of the intense interest in local history and archaeology in county families in the early 19th century. He moved in the circle of Sir Richard Colt Hoare and his collaborators who were assisted by Sir Thomas Phillipps from his collections.* Bowles cher- ished great regard for Bishop Thomas Burgess of Salisbury, the founder of the Royal Society of Liter- ature, which published a number of archaeological essays, particularly work concerned with inscriptions. Though his extensive knowledge was ill-disciplined in his writing, by drawing attention to source material, both documentary and archaeological, Bowles helped to arouse a more widespread interest in local history, which made possible the foundation of a county archaeo- logical society for Wiltshire in 1853. In another field, his Little Villager’s Verse Book*® should have a place in the history of hymn writing in the early 19th century.
43. He himself played several musical instruments and was much affected from childhood by musical sound. He wrote poems about musical performances and described himself on the ttle page of the pamphlet above as formerly a Steward of the Gloucester Music Meeting (The Three Choirs Festival). At Oxford he may have known the musician and artist, J.B. Malchair, who had important friends and pupils.
44. Howard Cooper in his article on Bowles in The Hatcher Review, Spring 1989, Vol. 3, no. 27, calls him ‘a founding figure of English Romanticism’.
45. The correspondence of Sir Thomas Phillipps in the Bodleian Library shows how this was encouraged by the willingness of Phillipps to give assistance to local scholars from his collections. He visited Bowles in 1821 to make notes from an emblazoned pedigree of the Lisle family then held by Bowles: Bodleian MS. Phillipps-Robinson d. 68, f. 50.
46. Woolf wrote in 1958 that he had been unable to trace the first edition of this little book which may have appeared as Hymns for Cottage Children c.1815: Woolf, op. cit., p. 288. Lady Lans- downe in a letter to Mrs Bowles, in 1818 or 1819, referred to it as the ‘Littke Hymn Book’: Greever, op. cit., p. 112.
|
THE REVD WILLIAM LISLE BOWLES
As Bowles wrote in 1836,*” he had been received, for half his life, as a guest in the houses of lords and ladies, archbishops, bishops, and ministers of state at a time when important changes in political and ecclesiastical organisation were being worked out. Though in general a traditionalist, and his attitude of mind sometimes reminiscent of the 17th century, he should have been very well informed in current affairs and in church matters was not indifferent to them. He spent much time commemorating others recently or
\7. Fraser's Magazine, 13 Jan.—June, 1836, p. 498.
18. See article by the Revd J.G. Jackson, WAM 1 (1854), pp. 251-64, for the interest of the third Lord Lansdowne and Bowles in the charity of Maud Heath’s causeway and the
141
long since dead, from his own old villagers to Bishop Ken and Maud Heath,**® but the memorial stone in the south choir aisle of Salisbury cathedral describes him only as canon residentiary. This contrasts with the well known monument by Flaxman in Winchester cathedral to his master, Joseph Warton. It is hoped that someone will fill out the work done by Greever and show that Bowles was a much more considerable figure than has been acknowledged.
erection of a column and figure of Maud Heath. Bowles composed the epitaph to Archdeacon Nares for the monument in Lichfield cathdral.
Wiltshire Archaeological and Natural History Magazine, vol. 86 (1993), pp. 142-157
Notes
A Barrow Cemetery and Other Features Recorded by Air Photography at Beckhampton, Avebury
by GRAHAME SOFFE
INTRODUCTION
This note is derived from an air photographic survey initiated in response to the requirements of the Trust for Wessex Archaeology for a record of the plough- levelled archaeological features in the upper Kennet Valley, in connection with a proposed sewer improve- ment scheme by Thames Water. The project was carried out by the Air Photography Unit of the Royal Commission on the Historical Monuments of England in April 1992. All available air photographs for the affected areas were examined (RCHME 1992b). One area, at Beckhampton, lying within the Avebury World Heritage Site, proved particularly interesting in that the survey provided a considerable enhance- ment of existing knowledge of one of the elaborate barrow cemeteries surrounding the Avebury group of monuments.
The collections of the National Library of Air Photography (RCHME), now based at Swindon, proved the most useful source consulted, supplemen- ted by the Cambridge University Committee for Aerial Photography collection (CUCAP). These photographs range from obliques taken by Major G.W.G. Allen in the early 1930s, and RAF vertical cover of 1946, to RCHME and CUCAP specialist cover taken in the 1950s and 1980s. Surprisingly, it was found that, despite the fame of the prehistoric monuments of the Avebury complex, the landscape surrounding them has experienced a relatively spora- dic history of aerial reconnaissance. Furthermore, the responsiveness of soils in this area is relatively poorer than that of chalk soils producing archaeological soil- or crop-marks elsewhere. Extensive mottled patches of deeper soil prevent the formation of a clear response to buried features. A major reason for this must be that the region has experienced a long history of ground disturbance through agriculture. South Street, Beckhampton, was cultivated in the Neolithic period and it is here that the earliest evidence in the
British Isles for the use of the plough has been recorded (Fowler and Evans 1967). This early activity must have been related to contemporary domestic settlements, the foci of which are identifiable (Holgate 1987), but very little of this evidence is detectable from the air and for much of this research we must rely on the use of ground-based techniques now well established in the Kennet Valley.
Another important factor, long recognised in the air reconnaissance of this area, is that year by year there will be differences in the appearance of certain features. For example, round barrows might appear as ring-ditches one year, chalky spreads in others, or not show at all. The Ordnance Survey has depicted some ploughed or otherwise destroyed barrows with antiquity crosses. At Beckhampton these locations have been based upon the same air photographic evidence as is discussed below.
The plan (see Figure) uses standard conventions to show soil- or crop-marks: solid lines for ditches or negative features, irregular stipple for banks and positive features, and dense regular stipple for scoops and hollows of archaeological origin. The archaeolo- gical detail has been plotted using a computer-based photogrammetric system published by the University of Bradford (Haigh 1989).
THE AIR PHOTOGRAPHIC EVIDENCE
Lying within the Avebury World Heritage Site, the area contains a major group of Bronze Age round barrows and other monuments occupying elevated ground overlooking Avebury lkm to the north-east and Silbury Hill 1km to the south-east. Most of the cemetery occupies the top of a spur known as Folly Hill at just over 170m above OD, extending east from West Down and Knoll Down and separated from Waden Hill by the Winterbourne. Most of the features lie just north-east of Beckhampton, a shrun- ken village in Avebury parish, and are enclosed by
NOTES
Beckhampton road (A4361) and the A4 which con- verge at the Beckhampton roundabout to the west.
Although this was one of the major barrow cemeteries clustering around Avebury, all but one of the barrows have been ploughed flat and air photo- graphy therefore provides the principal source of information about them. As has already been men- tioned, the immediate environment has _ been ploughed intermittently since the Neolithic period. Although the area was under pasture when most of the barrows were constructed in the Early to Middle Bronze Age, ploughing again seems to have taken place in the area from late prehistoric umes (e.g. Fyfield Down in Bowen 1961) up to the present. The recent extensive resistivity survey on the north edge of the cemetery has shown evidence for plough-levelled ridge and furrow of presumed medieval date (Ucko et al. 1991, 195-9). The present survey has identified and planned a series of narrow plough-levelled strip lynchets running along the slope to the south of the barrows.
The air photographic evidence shows that the cemetery, numbering at least twelve certain barrows, contains examples of elaborate or special form, prob- ably of Early to Middle Bronze Age ‘Wessex’ type. These elaborate barrows have their greatest concen- trations near Stonehenge, Everleigh and Avebury (Grinsell 1953, Fleming 1971) and occur together with bowl-barrows in nucleated, dispersed, linear and area cemeteries. In Fleming’s classification the ‘Beckhampton group is given as an example of a small nucleated cemetery situated at the north-east corner of an extensive area cemetery extending south- westwards. Although Fleming does not appear to have closely examined individual barrows in the Beckhampton group, he has included it in his ‘special area’ Wiltshire 2a, a region of 170 sq km containing 100 barrows, of which a relatively high 14.5% are special types, invariably bell-barrows. Although the 3eckhampton group is one of the closest to Avebury tself, the barrows are poorly preserved when ‘ompared with the spectacular examples on North own and Morgan’s Hill to the west and Overton Hill o the east (RCHME 1992a).
The new information provided by this survey has ighlighted the need for a more detailed account than 5 found in existing records. The numbering of each larrow in the following description relates to the jumbers shown on the transcribed plan (see Figure). | Only one barrow, 1, survives as a mound, tucked ito the north-east corner of a pasture field over- »oking the hollowed out A4361 road and the footpath inning off the road towards the Waggon and Horses
143
inn. This field was described as the Paddock in the mid-19th century (Wiltshire Record Office, Avebury Tithe Award 1845) and has been used for grazing and training racehorses since at least the beginning of the 19th century. Stukeley’s panoramic view of the Kennet and Beckhampton Avenues (Piggott 1985, pl. 13; Ucko et al., 1991 panorama) shows that barrow 1 and another barrow immediately east of it on the other side of a field boundary (presumably barrow 2 or 6) must have been extant mounds in the 1730s. By 1819 only barrow 1 appears to have survived and was surveyed, appearing on Crocker’s plan of ‘Abury and Silbury’ (Hoare 1821, pl. x). Presumably barrow 2 or 6 survived into the 18th century because of the proximity of the field edge, the remaining mounds having been plough-levelled by that time. In the latter part of the 19th century Smith (1884) carried out his survey of barrows and other monuments in the Avebury area. He records none of the Beckhampton group, suggesting that barrow | was missed or con- sidered unnoteworthy and that the others were ploughed flat at that time. Folly Hill, the field containing barrows 2, 3, 4 and 6 was under arable in 1845 (Avebury Tithe Award) and remains so today. Barrows 7, 8 and 9 were in Southfield, and 12 and 13 in South Mead Furlong, both also ploughed; the latter name suggests an open field up against former meadow, presumably close to the Winterbourne.
Of the plough-levelled round barrows, the air photographs show that barrow 2 originally possessed a very large circular mound defined by a very thin ditch and wider outer ditch. A series of dark marks in the area of the mound probably results from multiple burials, perhaps several secondary graves. These may have been of Roman or Saxon date, comparable with the three Roman burials found among the barrows on Overton Hill nearby (Smith and Simpson 1964). Beckhampton barrows 3 and 6 appear more elaborate still, with concentric ditches and banks surrounding mounds, possibly indicating multi-period structures or at the very least forms of bell- or disc-barrows. The very large barrow 4 is bisected by the footpath dividing the pasture paddock from the arable and both sectors have been ploughed flat. It was sur- rounded by a wide ditch and the dark crop-marks of the interior suggest a slight hollow. This may there- fore be a disc- or saucer-barrow or possibly a rare form of pond-barrow. Most of barrow 5 lies north of the road which cuts through it. It is overgrown by scrub but two thirds of the outer ditch can be detected on a limited number of air photographs. Barrows 7, 8, 9, 12 and 13 survive as ring-ditches with some evidence of interior mounds, indicating probable
— —~ 15 1 SS ee SN Lae ee ae CKox JO} 1x0] 338) nOIdwretppoag ye sydessoloyd yetsise Aq Pa[BaAaI sainieaj [BIdO;OoeYyoIe JIOGIO pu’ SMOLIeY
puej6ug yo syuawnNuoW Je91J0}SIH BY} UO UOISS}WWOD |eAOY ZEEE YG1sAdOD umoid 0) ‘ r = oS
[Nwooyggr =| wooggoe |
22 === 002 OOL
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
V
<< Moue@ 6u07 4aas}S YINES 40 BUS
sauo0js 6u07
144
LNY°OZ60" PAS SS \ : By ole ple N“ooy6g! | |
NOTES
bowl-barrows. Barrows 10 and 11 may not be barrows at all. They are exceptional in that the ditches of 10, certainly, and 11, possibly, form sub-rectangular or ovoid enclosures. Both are cut by the A4361 road. These can be paralleled with other monuments associ- ated with elaborate barrow groups. Apart from a similar example on West Down to the west of the cemetery (at SU 07056830), the group associated with Woodhenge provides an interesting comparison (Cun- nington 1929, 41-51, pl. 39; RCHME 1979, 23-4). Here a double ring-ditch has been shown to surround a Beaker burial, and an egg-shaped enclosure to be part of a Middle Bronze Age settlement. These sub-rectangular crop-marks could be in danger of morphological labelling and, as Loveday and Petchey (1982) have shown, shape is a dangerous criterion for dating. The very small number of excavated sites of comparable shape and size would suggest a Neolithic date (Ashbee 1970, 48—9) but crop-marks of this type have a wide distribution and their origin and dating may be equally diverse.
Three other features detectable on the air photo- grahs, 14, 15 and 16, may be ring-ditches, but this interpretation must be treated with caution as they occur on few photographs and as extremely faint marks. Features 5 and 10 are bounded to the north by a crop-mark of a ditch running nearly parallel to the A4361 road. It has two associated ditches running at right angles to it. These may be interpreted as medieval or post-medieval field boundaries. A pair of similar ditches run south from barrow 4 crossing the strip lynchets to the south. In the Paddock, the ploughed out remains of barrow 4 are overlain by a well preserved bank c.2.5m wide and 0.5m in height, cutting off the north-east corner of this field. This feature is not shown on any early maps or plans but is presumably a post-medieval field boundary.
The strip lynchets lying to the south of the barrows are recorded now as crop-marks on a gentle south- facing slope. They are quite narrow and closely spaced, particularly when compared with the plough- levelled examples further east on the west slope of Waden Hill and the lower slope of White Hill, West Overton. They were laid out in adjacent rectangular plots, the eastern one containing at least eight indi- vidual strips. They exhibit remarkably sharp edges between positive and negative crop-marks, probably showing how ploughing has started to erode the chalk bedrock at the base of the risers above each ‘positive’ lynchet accumulation. There is no reason to doubt that these lynchets are of medieval origin (Wood and Whittington 1959 and 1960; Taylor 1966), but fieldwork and excavation on better preserved
145
examples just north-west of Beckhampton at Comp- ton Bassett, should add significantly to knowledge of strip lynchets in this area (Reynolds and Wilkinson in prep. See p. 160 of this volume).
Immediately to the north of the barrow cemetery and in Long Stones Field lie the sites of the South Street long barrow (Ashbee et al. 1979) and the Long Stones. An oval enclosure, comparable in size to individual barrows in the Beckhampton group, and abutting on the north side of the Long Stones, was planned in the resistivity survey of this field (Ucko et al. 1991, 196-9, pl. 63), but evidence for the Beckhampton Avenue in Long Stones Field was inconclusive and no evidence for it was found in the present survey.
CONCORDANCE OF BARROWS
An 8-figure NGR is given for each barrow; the National Archaeological Record (NAR) number per- taining to the appropriate OS quarter sheet, the Sites and Monuments Record (SMR) number and Grin- sell’s number in his ‘Archaeological Gazetteer’ (1957) are listed where appropriate. In one case, barrow 1, a Scheduled Ancient Monument (SAM) number is given.
Barrow 1 (NGR SU 09156907); NAR 76B; SMR 654; SAM 673; Grinsell, Avebury 19b.
An oval turf-covered mound, probably a bowl- barrow, c.20m in diameter and 0.9-2.0m high. The oval shape is due to out-throw from central mutilation to its west side. Samian sherds and a cremation in a cist were found by Dean Merewether in 1849 (Craw- ford and Keiller 1928, 219). The SMR records that the ‘site is suffering some minor erosion and from some animal damage’. When visited in April 1992, it was found to have been recently dug into, to a depth of about 1m, presumably by treasure hunters. A flat, rivetted copper-alloy dagger (now in Devizes Museum) was found in the outcast material by the Ancient Monuments Warden, indicating the disturb- ance of a primary Early Bronze Age burial.
Barrow 2 (NGR SU 09196907); NAR 76A; SMR 655; Grinsell, Avebury 19a.
A plough-levelled circular mound, 40m in diameter, surrounded by a narrow ditch and an outer wide ditch. Dark marks in the mound area indicate a probable central grave and several secondary burials (see discussion above). On the east side of the mound, 3 ditches run out from the inner narrow ditch, perhaps forming a small secondary enclosure.
146 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Barrow 3 (NGR SU 09206903); NAR 76D; SMR 684; Grinsell, Avebury 19d.
A plough-levelled oval mound, 30m in diameter. Within the mound area is a sub-circular narrow ditch with 3 gaps in its circuit, the gap on the south side having enlarged ditch terminals. The mound area is defined by a narrow ditch with traces of a narrow bank on its north-west side. This in turn is sur- rounded by a wider outer ditch. The barrow may therefore be a special type, possibly a bell-barrow, and may contain a pre-mound palisade or stake circle (cf. Cunnington 1929, circle 1; Ashbee 1960, 60-65). The enclosed segmented ditches may be closely paral- leled with excavated examples such as the Beaker barrow on Stockbridge Down, Hampshire (Stone and Gray Hill 1949, 39-51), and examples known from air photography over the Thames River Gravels in Oxfordshire such as those at North Stoke (RCHME 1960, pl. 10a; Case 1982, 61, fig. 33) which have recently been rephotographed by RCHME.
Barrow 4 (NGR SU 09136900); NAR 76E; SMR 683; Grinsell, Avebury 19e.
A large wide ring-ditch, 46m in diameter, bisected by the footpath and existing pipeline. Both sectors of the barrow, under arable and pasture, have been ploug- hed flat, but the crop-marks of the interior area suggest slight hollowing and this may indicate a special type of ‘Wessex’ barrow, possibly a disc- or saucer-barrow, or perhaps even a rare and very large form of pond-barrow. After levelling the barrow was overlain on its south-west side by the earthwork bank mentioned above.
Barrow 5 (NGR SU 09146912); NAR 76K.
A ring ditch, 34m in diameter, mutilated by the cutting of the A4361 road on its south side and concrete structures over the site of the mound. It is overgrown with scrub.
Barrow 6 (NGR SU 09236908); NAR 76C; SMR 656/682; Grinsell, Avebury 19c.
An elaborate barrow consisting of the ploughed out traces of a circular mound 27m in diameter. The mound area contains an ovoid ditch surrounding a central pit, probably the primary burial. Possibly a bell-barrow or a multi-phase structure.
Barrow 7 (NGR SU 09276913); NAR 76G.
A plough-levelled ring-ditch with clear traces of central mound, 22m in diameter. A bowl-barrow similar to barrow 8 and forming a pair with it.
Barrow 8 (NGR SU 09296909); NAR 76H.
A plough-levelled ring-ditch with clear traces of a central mound, 22m in diameter. A bowl-barrow similar to barrow 7 and forming a pair with it.
Barrow 9 (NGR SU 09356914); NAR 76J; SMR 687. A plough-levelled ring-ditch, 32m in diameter. Little trace of mound surviving. Probably a bowl-barrow.
Barrow 10 (NGR SU 09246917); SMR 688.
A ditch surrounding a sub-rectangular area containing traces of a mound. Possibly not a barrow. Comparable with other prehistoric enclosures of similar shape and size (see discussion above). A similar feature has been recorded at SU 07056830 just to the west (Wiltshire County Council SMR AP AER 228). The enclosure is 42m in length, 32m wide, and is cut by the A4361 road at its south corner.
Barrow 11 (NGR SU 09286917).
Possibly a similar feature to barrow 10; 20m wide. Only parts of the north, east and west sides show as ditches on the north edge of the field against the road. The crop-marks are confused by turning marks of agricultural machinery.
Barrow 12 (NGR SU 09546907); NAR 77A; SMR 685; Grinsell, Avebury 19f.
A large ring-ditch containing traces of a mound. Completely plough-levelled and eroded. Diameter 43m.
Barrow 13 (NGR SU 09556902); NAR 77B; SMR 686; Grinsell, Avebury 19g.
A ring-ditch, slightly smaller than 12, with traces of mound. Completely plough-levelled and eroded. Dia- meter 32m.
Acknowledgements. The original project was carried out under the management of Roger Featherstone. Carolyn Dyer completed the drawing of the figure. Further thanks are due to Rowan Whimster and Simon Crutchley (RCHME), Chris Place and Helena Cave- Penney (Wiltshire SMR), Gill Swanton (WANHS Archaeological Committee), Sian Williams and Andrew David (English Heritage), and Michael Pitts. This paper is Crown Copyright: RCHME.
REFERENCES
ASHBEE, P., 1960 The Bronze Age Round Barrow in Britain, London
1970 The Earthen Long Barrow in Britain, London
ASHBEE, P., SMITH, I.F., and EVANS, J.G., 1979 ‘Excavation of Three Long Barrows Near Avebury, Wiltshire’, Proc. Prehist. Soc. 45, 207-300
BOWEN, H.C., 1961 Ancient Fields, London
CASE, H.J., 1982 ‘The Linear Ditches and Southern Enclos- ure, North Stoke’ in H.J. Case and A.W.R. Whittle
NOTES
(eds.), Settlement Patterns in the Oxford Region; Exca- vations at the Abingdon Causewayed Enclosure and Other Sites, CBA Res. Rep. 44, 60-75
CRAWFORD, 0.G.S., and KEILLER, A., 1928 Wessex from the Ar, Oxford
CUNNINGTON, M.E., 1929 Woodhenge, Devizes
FLEMING, A., 1971 ‘Territorial Patterns in Bronze Age Wessex’, Proc. Prehist. Soc. 37, 138-66
FOWLER, P.J., and EVANS, J.G., 1967 ‘Plough-marks, Lynchets and Early Fields’, Antiquity 41, 289-301
GRINSELL, L.V., 1953 The Ancient Burial-Mounds of England (2nd edn.), London
1957 ‘Archaeological Gazetteer’ in R.B. Pugh and E. Crittall (eds.), VCH Wiltshire I, Part 1, London, 21-279
HAIGH, J.G.B., 1989 ‘Rectification of Aerial Photographs by Means of Desk-top Systems’ in S. Rahtz et al. (eds.), Computer Applications and Quantitative Methods in Archaeology, BAR Int. Ser. 548, Oxford, 111-9
HOARE, SIR, R.C., 1821 The Ancient History of North Wiltshire, London
HOLGATE, R., 1987 ‘Neolithic Settlkement Patterns at Avebury, Wiltshire’, Antiquity 61, 259-63
LOVEDAY, R., and PETCHEY, M., 1982 ‘Oblong Ditches: A Discussion and Some New Evidence’, Aerial Archaeol. 8, 17-24
PIGGOTT, S., 1985 William Stukeley, An Eighteenth-Century Antiquary, London
RCHME, 1960 A Matter of Time: An Archaeological Survey of the River Gravels of England, London
147
— 1979 Stonehenge and its Environs, Edinburgh
— 1992a ‘West Kennett-East Kennett, Archaeological Survey in the vicinity of the Avebury World Heritage Site; Air Photographic Transcription and Analysis’ (Archive report in National Monuments Record)
1992b ‘Archaeological Survey in and around the Avebury World Heritage Site, The Kennet Valley foul sewer improvement; Air Photographic Transcription and Analysis’ (Archive report in National Monuments Record)
SMITH, A.C., 1884 Guide to the British and Roman Antiquities of the North Wiltshire Downs in a Hundred Square Miles round Abury, Marlborough
SMITH, I.F., and SIMPSON, D.D.A., 1964 ‘Excavation of three Roman Tombs and a Prehistoric Pit on Overton Down, WAM 59, 68-85
STONE, J.F.S., and GRAY HILL, N., 1940 ‘A Round Barrow on Stockbridge Down, Hampshire’, Antiqg. 7. 20, 39-51
TAYLOR, C.C., 1966 ‘Strip Lynchets’, Antiquity 40, 277-84
UCKO, P., HUNTER, M., CLARK, A.J., and DAVID, A., 1991 Avebury Reconsidered, From the 1660s to the 1990s, London
WOOD, P., and WHITTINGTON, G., 1959 “The Investigation in 1957 of Strip Lynchets North of the Vale of Pewsey’, WAM 57, 163-72
— 1960 ‘Further Examination of Strip Lynchets North of the Vale of Pewsey in 1958’, WAM 57, 322-38
Iron Age Coins from Cunetio and Mildenhall
by PAUL ROBINSON
In the past there has been a certain amount of confusion concerning the Iron Age coins found in the parish of Mildenhall. The recent acquisition by Devizes Museum of a plated silver coin of the Iceni from the site of the Roman town of Cunetio in the parish has prompted this note, listing coins which have been found for certain either on or very close to the site of the Roman town, coins which come from the parish of Mildenhall but are not from the site of the Roman town, and those coins, possibly from Mildenhall, which may or may not have been found on the site of Cunetio. There is no attempt to include Iron Age coins from adjacent parishes such as Rams- bury and Savernake. Coins purported to have been found in Marlborough have, however, been included. Most derive from the private collection of the Revd Charles Soames, vicar in Mildenhall in the last century, and the provenance ‘near Marlborough’ as it is used on their tickets and in the sale catalogue of his collection can be regarded as a probable synonym for
‘Mildenhall’, then a small village which was not well known outside Wiltshire. (We have excluded here a coin of type Evans G 7 (Mack 274) which appears under ‘Marlborough’ in Grinsell (1957, 85) but which in fact comes from Aldbourne.)
A. Coins found on or very close to the site of Cunetio
1. ICENTI. Plated silver coin of the Face-Horse type. Type Allen (1970) Face-Horse silver series d. Van Arsdell (1989) has attributed the Face-Horse type to Boudicca, though Burnett (1989) has argued that it is as likely to date before the Roman conquest. Although there is no precise location for the findspot, the coin is said to have been found before 1978 on the site of the Roman town in association with a number of Roman brooches and other metalwork. Devizes Museum (1991.72). Previously unpub- lished.
148 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
2. IRREGULAR DOBUNNIC. Silver coin of the ‘Upavon’ type, a type not in Mack (1975) or Van Arsdell (1989).
Found in 1977 at SU 215695 within the area of the walled town.
Devizes Museum (104.1978). Published in Robin- son (1981, 86, no. 11) and Haselgrove (1989, 57).
3. Uncertain copper or copper alloy coin akin to the British I series. The type is not in Mack (1975) or Van. Arsdell (1989). It resembles the Coriel- tauvian type C series but has a left-facing bust on the obverse and is much cruder and simpler than the Corieltauvian staters.
Found in 1967 at SU 217695 within the area of the walled town.
Devizes Museum (26.1967). Published in Allen (1967), Burchard (1968), Gunstone (1977, 29) and Haselgrove (1978, 26).
4. Silver coin of uncertain type with the design on one side of a standing figure with a staff. The type is not in Mack (1975) or Van Arsdell (1989). Found in 1978 at SU 215695 within the area of the walled town.
The present location of the coin is unknown. Published in Haselgrove (1984, 154).
5. Uncertain: perhaps either the copper alloy or copper core of a once plated British Qc quarter stater of uncertain type or the core of a plated example of the Upavon type coins in the Irregular Dobunnic series (as No. 2, above). Visible on this corroded coin are traces of the design of a horse facing right with perhaps two tails and a small circle between the legs of the animal.
Found in 1978 at SU 216693 in the course of investigating the circumstances of the deposition of the ‘Cunetio hoard’, a very short distance to the south of the walled town.
Devizes Museum (1991.85.1). Published in Besly and Brand (1983, 175).
B. Coins found elsewhere in the parish of Mildenhall
1. BRITISH Mb base gold stater, type Mack 62. Found in 1977 at SU 205689. Devizes Museum (33.1978). Published in Robin- son (1981, 85 no. 4) and Haselgrove (1984, 120).
2. BRITISH Mb base gold stater, type Mack 62. Found in 1978 at SU 218692. Devizes Museum (117.1978). Published in Ro- binson (1981, 85 no. 5) and Haselgrove (1989, 23), in both of which the grid reference for the findspot is incorrect.
3. CATUVELLAUNI, Cunobelin. Copper alloy coin of type Mack 225 (=Evans pl. XII 8).
The ticket with the coin merely records that it was ‘found near Marlborough’. Evans, however, states that it was found at Mildenhall in 1874 ‘within a mile of the Epaticcus hoard’ (i.e. the Savernake Forest hoard which was discovered in 1857). It cannot then have been found on the site of Cunetio which lies two miles away from where the hoard was found.
British Museum, ex Revd Charles Soames’ collec- ton (Sotheby’s 17 July 1903, lot 339). Published in Evans (1881, 570) and Grinsell (1957, 87). Central Gaul, AULERCI EBUROVICES. Copper alloy coin, type de la Tour 7021/7070. Found with the previous coin.
British Museum, ex Revd Charles Soames’ collec- tion (Sotheby’s 17 July 1903, perhaps either lot 339 or 340). Published in Evans (1881, 479), Grinsell (1957, 87), Allen (1960, 279) and Robin- son (1975, 7, n. 2).
C. Coins possibly found in Mildenhall
1.
DUROTRIGES copper alloy stater, type Mack 317-8.
Said to have been found ‘near Mildenhall’ before 1980.
Devizes Museum (28.1980). Published in Robin- son (1981, 86, no. 12) and Haselgrove (1984, 138). DOBUNNI uninscribed silver coin of type B or (Ge
Said to have been found ‘near Marlborough’ in 1980 or 1981.
Devizes Museum (35.1981). Published in Robin- son (1981, 86, no. 9) and Haselgrove (1984, 140). CATUVELLAUNI, Cunobelin: silver coin, type Mack 241la (=type 257 in earlier editions of this catalogue).
The sale catalogue of the Soames collection and the ticket with the coin state that it was ‘Found near Marlborough’.
British Museum, ex Revd Charles Soames’ collec- tion (Sotheby’s 17 July 1903, lot 339). Published in Evans (1881, 565 and pl. XXII, 8), Grinsell (1957, 85 and 87 — duplicated entry), Allen (1960, 235) and Robinson (1975, 7).
British Rb quarter stater in copper alloy, type Mack 68.
The only evidence for the findspot is the ticket with the coin which records that it was ‘Found near Marlborough’.
British Museum, ex Revd Charles Soames’ collec- tion (not identifiable in the sale catalogue of his collection). Published in Allen (1960, 203; 1961, 141 and pl. 37, no. 2) and Robinson (1977, 9, n. 1).
NOTES
5-7. IRREGULAR DOBUNNIC L (one coin) and M (two coins). Three silver coins of types Mack 377 and 384a.
The coins were found in about 1874. The findspot appears on the coin tickets as ‘near Marlborough’ and is discussed in Robinson (1975, 7, n. 2), who suggests that it was possibly in Mildenhall but not necessarily on the site of Cunetio. The closeness in date of the finding of this group and B 3 and 4 above might lead to the suspicion that they were found together but there is nothing concrete to support the idea.
British Museum, ex Revd Charles Soames’ collec- tion (not identifiable in the sale catalogue of his collection). Published in Evans (1881, 467f), Grinsell (1957, 87), Allen (1960, 249) and Robin- son: (1975, 7,.n. 2 and\.1977, 13=15).
Cunetio is at present the only Roman town in Wilt- shire at which Iron Age coins have been found. The exact dates when the coins were struck and the length of time after the Roman conquest during which they remained in circulation are uncertain. Nevertheless the coin finds from the site of the town suggest that the settlement may have begun during the late Iron Age, as at Siichester. No other finds of late Iron Age pottery or metalwork are recorded from the site of the town, although an important fragment of decorated metalwork has been found at Poulton Hill in Mil- denhall, while the findspot of the Marlborough bucket lies only a short distance outside the parish boundary.
The coins discussed above are essentially of three types. Most common are those with a distribution centred on or restricted to central to eastern Wiltshire (A2, perhaps AS, B 1 and 2, C4 and 5-7). Two coins of the Catuvellauni from eastern England reflect the general distribution of their coins as far as north- eastern Wiltshire. Three coins show longer distance contacts with the Iceni of East Anglia, perhaps the Corieltauvi in the east Midlands and with Gaul. The provenances of the two coins of the Dobunni and the Durotriges immediately to the west and the south-
149
west are not convincing. Absent also from both Cunetio and the parish of Mildenhall are coins of the Atrebates proper and of the northern Atrebatic kingdom of Epaticcus, although they are found else- where in eastern Wiltshire. Coins of Epaticcus were, however, common in the Savernake forest hoard, so their present absence from Mildenhall must be merely fortuitous.
REFERENCES
ALLEN, D.F., 1960 ‘The Origins of Coinage in Britain: a Reappraisal’ in S.S. Frere (ed.), Problems of the Iron Age in Southern Britain, Inst. Archaeol. Lond. Occ. Paper II, London, 97-308
1961 ‘A Study of the Dobunnic Coinage’ in E.M.
Clifford, Bagendon: A Belgic Oppidum. A Record of the
Excavations 1954-56, Cambridge, 75-149
1967 ‘Three Ancient British Notes’, BNF 36, 8-10
— 1970 ‘The Coins of the Iceni’, Britannia 1, 1-33
BESLY, E., and BRAND, R., 1983 The Cunetio Treasure. Roman Coinage of the Third Century AD, London
BURCHARD, A., 1968 ‘A new Celtic Coin from Cunetio’, WAM 63, 101
BURNETT, A., 1989 ‘Review of Van Arsdell (1989) BNF 59, 235-37
EVANS, J., 1881 Supplement to The Coins of the Ancient Britons, London
GRINSELL, L.V., 1957 ‘Archaeological Gazetteer of Wiltshire’ in R.B. Pugh and E. Crittall (eds.), A History of Wiltshire, Vol. 1, part 1, London, 21-279
GUNSTONE, A.J.H., 1977 Sylloge of Coins of the British Isles 24. Ancient British, Anglo-Saxon and Norman Coins in West Country Museums, London
HASELGROVE, C.C., 1978 Supplementary Gazetteer of Find-spots of Celtic Coins in Britain, 1977, Inst. Archaeol. Lond. Occ. Paper lla, London
— 1984 ‘Celtic Coins Found in Britain 1977-82’, Bull. Inst. Archaeol. Lond. 20, 107-54
— 1989 ‘Celtic Coins Found in Britain 1982-1987’, Bull. Inst. Archaeol. Lond. 26, 1-75
MACK, R.P., 1975 The Coinage of Ancient Britain, London
ROBINSON, P.H., 1975 “The Savernake Forest Find of Ancient British and Roman Coins (1857)’, BNF 45, 1-11
— 1977 ‘A Local Iron Age Coinage in Silver and Perhaps Gold in Wiltshire’, BNF 47, 5-20
1981 ‘Recent Acquisitions by Devizes Museum of Ancient British, Saxon and Norman Coins’, WAM 76, 83-92
VAN ARSDELL, R.D., 1989 Celtic Coinage of Britain, London
A 7th-Century Gold Coin from Shalbourne
by LORD STEWARTBY
In October 1989 a gold coin of great interest was found by metal-detector on farmland near the village of Shalbourne in north-east Wiltshire, close to the border with Berkshire and four miles south of Hungerford. Through the generosity of the land- owner and the finder, Graham Palmer, the coin has been presented to Devizes Museum.
Small gold coins of this kind, descended from the late Roman one-third solidus known as 2 tremissis or triens, were produced extensively in Merovingian Francia during the late 6th and much of the 7th century, and comparable coinage spread northwards into Frisia and England. An indication of the relative date of individual types can be gained from their gold content, since a gradual debasement took place from the end of the 6th century until by c.680 the coinage was of silver alone.! Some of the Frankish tremisses carry the names of kings, which enable an approxim- ate chronology to be constructed. On this basis, the Shalbourne coin with a specific gravity of 13.937, indicating for a gold—silver alloy a fineness of c.55%, seems likely to have been struck within a few years of 640.
The coin is difficult to describe because the dies were very crude, and the coin is not evenly struck (see Figure, below). It has also become slightly buckled. Its weight is 1.1026g (17.0 grains), against a norm of c.1.3g (20gr), but wear does not seem likely to have accounted for much of the deficiency. On the obverse is a short, thick bar, with a blob above to the left, and an arc with ‘spokes’ behind: these seem to be the nose, eye and hair of a coarse, degraded head, facing right. The reverse type, a cross with the ends potent, is surrounded by a_ meaningless inscription, OVTITVIT[.]V. The ‘letters’ on the obverse, above and before the head, are even rougher, perhaps to be rendered IILIIN’IT.
Many Merovingian gold coins with a profile head have crosses potent on the reverse, but usually moun-
1. J.P.C. Kent, ‘The Date of the Sutton Hoo Hoard’, in R.L.S. Bruce-Mitford (ed.), The Sutton Hoo Ship Burial, 1, 1973, pp. 588-607. The dating of coins of this period from their gold content can, of course, be no more than approximate.
2. C.H.V. Sutherland, Anglo-Saxon Gold Coinage in the Light of the Crondall Hoard, 1948, pp. 35 and 77-8; for the sixth specimen see I. Stewart, ‘Anglo-Saxon Gold Coins’, in R.A.G. Carson and C.M. Kraay (eds.), Scripta Nummaria Romana, 1978, pp.
ted on steps, or a globe, as on the Byzantine originals from which they were copied. There is, however, in the English series a type which could well have served as a model for the Shalbourne coin, Sutherland type I vi, nos. 17-19. Four specimens are known, from two linked pairs of dies, with a cross pattee on steps on the reverse and a runic inscription beginning Benu-, and two others with a cross potent and Roman lettering.’ The obverse head, with the hair rendered in stiff strokes inside an outline, on all of them, and the cross potent on the reverse of the two with Roman letters suggest that this type may have provided the proto- type for a crude local copy. The connection is rendered more probable by the find-spot in Wessex. One of the runic specimens was found in Glou- cestershire (S.17) and another (S.18b) at Dorchester in Oxfordshire. There was no example in the 1828 hoard from Crondall, Hampshire (on the Surrey border), which is our main source for the early English gold coinage. This raises the question whether, unlike most of the English coins from Crondall, the type may have come from outside south-east England. An origin further west, perhaps in the Thames valley, has seemed possible to those who have written on this coinage in recent years,’ and a north Wiltshire find-spot for an imitation of the type would be consistent with this.
143-72, no. A.95 (SCBI Mack 307). P. Grierson and M.A.S. Blackburn, Medieval European Coinage, I, 1986, p. 161, have suggested the possibility that A.95, purchased in Amiens, may be of Continental origin.
3. S.E. Rigold in Bruce-Mitford, op. cit., pp. 663 and 676; Stewart, op. cit., p. 149; Grierson and Blackburn, Joc. cit. I am grateful to Mr Blackburn for indicating that he is in general agreement with the content of the present note.
NOTES
Absence at Crondall could, however, also be due to the relatively early date of the type. Of the runic examples, the Gloucestershire coin is of gold about 72% fine, while one of the Roman-lettered specimens (S.19) is about 63% fine. The former is of a higher gold content than any of the English coins from Crondall, and suggests that the original runic type could have been introduced by 630, with the Roman- lettered derivative a little later. The Shalbourne coin, if copied from the latter and so representing a third stage in the process, is again, appropriately, a little less fine in its turn. There are few cases in the early English gold coinage of groups of related coins that could be described as a series, but this might perhaps be one.
Although it is possible that the design of the
151
Shalbourne coin was copied from a different model, or indeed that it was produced on the Continent, the combined inference from type, provenance and fineness suggests that it should be associated with the undoubtedly English coins of Sutherland type I vi and that it may therefore be of local origin. In the subsequent silver coinage, imitation of regular issues by less skilled die-cutters became quite common in the first half of the 8th century, and it need occasion no surprise to find this happening at a relatively early stage in the English gold coinage.
Acknowledgement. I am grateful to Mr Geoffrey Thoday of the Fitzwilliam Museum for making the plaster cast, with which this note is illustrated, from the wax impression I had taken.
A Note on the Identification of ‘Alvestone’ in Wiltshire Domesday
by JASON ST JOHN NICOLLE
The identification of Alvestone, an appurtenance of the large manor of Bradford on Avon, has long posed a problem for students of Wiltshire Domesday.! The implication would seem to be that it is to be located within the county, presumably in the proximity of Bradford. Yet no mention of a place or settlement with this name has been discovered among the extens- ive records relating to medieval Wiltshire. So where was Alvestone?
The Wiltshire manor of Bradford was granted to the nuns of Shaftesbury Abbey by Ethelred II in AD 1001.7 At the time of Domesday, the nuns also held property in three other counties: Dorset, Sussex and Somerset. According to Domesday, the Somerset property was limited to two small manors in the south of the county: Abbas Combe and Kilmington.* But we know from the Somerset Geld Rolls of c.1084 that
1. Domesday Book I, f. 67 b, c. P.H. Sawyer, Anglo-Saxon Charters. An Annotated List and Bibliography, Royal Historical Society Guides and Handbooks 8 (1968), no. 899. The charter includes a set of bounds corres- ponding to the hundred of Bradford, which was more extensive than the nuns’ manor.
3. Domesday Book I, ff. 91b; 98a.
4. The Victoria History for the County of Somerset [V.C.H.], I (London, 1906), p. 528. For the date of the Geld Rolls, see V.H. Galbraith, The Making of Domesday Book (Oxford, 1961), pp.
the nuns also held three hides in demesne in the hundred of Bath.*
It would seem that these three hides were at Kelston, a small village about four miles west-north- west of Bath (ST 698670). The manor of Kelston was the only property which the nuns are recorded as holding within the hundred of Bath. It was in their possession by 1122 at the very latest and there is no evidence of acquisition subsequent to the compilation of the Geld Rolls.°
But there is no mention of Kelston at all in Somerset Domesday. This is surprising, as the name itself — ‘tun where the calves are raised’ — points to a pre-Conquest settlement with an important economic role.° It may be assumed that the settlement was included in the entry for some other manor. As Domesday manors were arranged under their
87-101; and R.R. Darlington in V.C.H. Wiltshire, 11 (London, 1955), pp. 174-6.
5. Regesta Regum Anglo-Normannorum II, eds. C. Johnson and H.A. Cronne (Oxford, 1956), p. 263 (no. 1763). The writ can be dated 1129-33, but it refers back to the time of Bishop John (de Villula), who was Bishop of Wells from 1088 to 1122. He transferred his see to Bath in 1090.
6. E. Ekwall, Dictionary of English Place-Names (4th edn., Oxford, 1960), s.v.
152 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
respective tenants-in-chief, it would be natural to begin by looking at the entries for other manors held by Shaftesbury in the same area. It is surely more than a coincidence that the entry for Shaftesbury’s only manor in the vicinity includes an appurtenance which was considered administratively distinct from the manor under which it was entered. But there is much stronger evidence for identifying Alvestone with Kelston.
The best evidence comes from comparing the hidage totals in the Geld Rolls with those in Domes- day. The three demesne hides recorded in the Somer- set Geld Rolls are close to the four demesne hides in Domesday Alvestone. Small discrepancies such as this are by no means uncommon, and are only to be expected if, as seems probable, the inquests were separated by a couple of years. There is a similar one
7. V.C.H. Wiltshire, I, pp. 198-9.
8. S.C. Morland, ‘The Somerset Hundreds in the Geld Inquest and their Domesday Manors’, Proceedings of the Somerset Archaeolo- gical and Natural History Society 134 (1990), p. 103 n.
hide discrepancy in the number of demesne hides on the manor of Bradford, between the Domesday figure and the figure in the Wiltshire Geld Rolls.’ Equally suggestive is the fact that the total hidage assessment for Domesday Alvestone was seven hides, as it has been pointed out that if this figure were also the total hidage for the manor of Kelston, it would reconcile the difference between the total number of hides in the Domesday hundred of Bath with the equivalent number given in the Somerset Geld Rolls.°
The only remaining problem would seem to be the dissimilarity between the two place-names. An early 12th-century form of Kelston, however, was Calves- tona.”? Given the weight of the evidence identifying Alvestone with the manor of Kelston, is it too much to suggest that Bradford’s mysterious appurtenance is simply missing its initial letter?
9. See note 5, supra. The full text is printed in Two Chartularies of the Priory of St. Peter at Bath, ed. W. Hunt (Somerset Record Society 7, 1893), p. 55.
Studies of Small Invertebrate Populations
by BARBARA LAST and WILFRID DOWDESWELL
INTRODUCTION
The traditional approach to the recording of plant and animal populations has been to assess their numbers and occurrence over relatively large areas. Such a procedure, although useful as a general record of distribution, is of limited value in determining short term variations. Today, we are faced with climatic changes whose magnitude and effect on living organ- isms could be considerable. It is therefore important to study small areas in some detail so that relevant changes can be monitored quickly and effectively. The following account illustrates the kinds of inver- tebrates that can be found in two small areas of grassland, Middleton Down, near Salisbury, and Hazelbury Common, near Box. As they stand, the data are largely subjective, but notes are included as a guide to ways in which the relevance of such observa- tions could be increased by introducing elements of greater objectivity.
It is common knowledge that our natural environ-
ment is in a constant state of change. Much of this change is brought about by ecological factors whose effects are more or less predictable. But the gradual warming and pollution of the atmosphere could present an entirely new situation affecting the dis- tribution of environmentally sensitive species of plants and animals.
If we are to assess future changes effectively, it is important that we should acquire as much detailed information as possible on small localised areas which can be monitored relatively easily from year to year.
INVERTEBRATES OF MIDDLETON DOWN
Middleton Down nature reserve lies to the south of the village of Broadchalke, 8 miles west of Salisbury. The entrance to the reserve is at grid reference SU 046238. It was purchased by the Wiltshire Trust for Nature Conservation (now Wiltshire Wildlife Trust) in 1988 and comprises two steep combes running
NOTES
east-west (Lime-Kiln Bottom and Rat’s Castle Bottom) connected by the slope of Gorse Down, which has an east-facing aspect. The total area is 25 hectares, and is largely of superb quality downland turf. In the early years, the flora was surveyed, and notes have also been made of the invertebrate population, some of which are submitted here.
If some of the large flints lying on the surface of the turf on Gorse Down are lifted carefully—and replaced in like manner, the channels made by ants can be readily observed. Sharing this microhabitat are frequently found the common Isopods Armadillidium vulgare and Philoscia muscorum, and also, sometimes, a small white Isopod Platyarthus hoffmannseggi that lives as a commensal among the ants and is apparently ignored by them. Although supposedly blind, it moves into the soil with remarkable speed when light is admitted.
In May 1989, a number of brilliant iridescent green beetles, 2mm long, could be seen in the flowers of Ranunculus bulbosus. These were Meligethes aeneus whose preferred habitat is Brassica flowers where they may do considerable damage to rape and other crops. The female lays a single egg in a flower bud from which hatches a larva that feasts upon the anthers and pollen, and may cause deformity to the pod. In fully open flowers, the adult beetle feeds on pollen.
Meloe proscarabaeus is an interesting beetle resident on Lime-Kiln Bottom. The female is conspicuous at 3cm, entirely black, with elytra covering only one third of the grossly swollen abdomen, and lacking the power of flight. A large number of eggs laid in a hole in the ground hatch into mobile larvae which climb flower stems and wait to hitch a ride on a suitable host. If the larva attaches itself to a solitary bee, it may be transported to the nursery burrow of the bee, where the larva releases its grip. The exacting condi- tions for continued development are the presence of a bee’s egg floating in nectar, provisioned by the bee. The Meloe larva then parasitises the egg and feeds on the supply of nectar, going through ecdyses into a non-mobile bloated larva. At the third instar it remains dormant in the burrow during winter, moults again into an active larva, pupates and then gives rise to an adult beetle.
The life of a parasite is fraught with hazards. The chances of a pick-up by the right host and the burrow being in an optimum state to nurture the larva are very slim; hence the necessity for the fecundity of the beetle. Meloe beetles have been observed in coitus at _ Lime-Kiln Bottom adjacent to a bank favoured by the solitary bees Andrena cineraria, distinguished by white moustaches, and Colletes succinta, a small ginger bee, which may serve as hosts.
153
The blood of Meloe contains cantharidin, a naus- eous substance, which may be exuded from pores in articular membranes when the beetle is stressed. This is a deterrent to some predators but is ineffective against hedgehogs, bats and some birds. Cantharidin is also found in ‘Spanish fly’ — a related beetle in Central Europe where it was formerly prized as an aphrodisiac, and was prescribed by physicians Hippo- crates and Galen. Now this practice is in decline, partly due to the scarcity of the beetle, and partly because of the toxicity of cantharidin which can be lethal in a dose of 0.03gms.
There is a considerable population of Timarcha tenebriosa, a domed beetle, flightless because of its fused elyta early in the year. These beetles are colloquially known as ‘bloody-nosed beetles’ because they exude a drop of red fluid when alarmed. They and their larvae feed on bedstraws, abundant on the down. Bedstraw is also the food-plant of the Lesser Elephant Hawk moth, large numbers of which came toa U.V. light in July 1991.
Soldier beetles, or Rhagonycha fulva (Cantharidae), misnamed ‘blood-suckers’, are plentiful in late summer, feeding on the open flowers of Umbellifers. They are notable for the length of time spent in copulation. The larvae are carnivorous and feed on Mollusca.
Ontholestes murinus, a Staphilinid, has gold and black markings produced by a sheen of hairs; it is a dung-feeder.
Sometimes the empty shells of the banded ‘humbug’ snail Cepaea nemoralis are the result of larval predations of Glow-worms, the beetle Lampyrus noctiluca. Several male beetles, more mobile than the worm-like females, found their way into pitfall traps.
Large numbers of black and yellow Sexton beetles Necrophorus vespillo and the large black N. humator (28mm) were seen attracted to the flesh of two dead shrews, as were some black Carrion beetles, Silpha atrata, their projecting snout-like heads adapted for reaching into snail-shells.
The Violet Ground beetle Carabus violaceus, with iridescent elytra, is a large carnivore that feeds on slugs. Another Ground beetle Loricera_pilicornis sports an elaborate cage of stiff setae on antennae and mouth-parts to trap spring-tails on which they feed.
Among the Diptera are Robber-flies Aszlus crabro- niformis which fly in copulatory tandem in July. In late September, large numbers of Empis tesselata, large black carnivorous flies, lurk on flowers, waiting to capture nectar-feeding insects. Their bristly facial hairs protect them from the struggling prey which is
154 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
grasped firmly while the proboscis penetrates and sucks the flesh dry.
Hover-flies of several species abound in thousands in late summer. The Drone-fly Eristasis tenax, which rather resembles honey-bee drones, was particularly numerous in 1991, and feeds on flowers, especially Scabious. It has aquatic ‘rat-tailed’ larvae which are able to tolerate low oxygen levels and can reach air by means of an extendable breathing siphon. These flies probably breed in cattle troughs.
Several tiny flies, Acrocera globulus have been recorded. They have minute heads that seem com- posed entirely of eyes, and reduced mouth-parts as the adults do not feed. Behind the wings are a pair of inflated squamae resembling water-wings, and the globular abdomen is black and yellow. These flies are said to be rare and the larvae parasitise spiders.
Increasing Objectivity The preceding records of invertebrates on Middleton Down are essentially subjective. But the number and
diversity of species show clearly that, viewed in a more objective manner, they could provide just the kinds of evidence needed for the regular monitoring of a small community of plants and animals. Some ways in which greater objectivity could be achieved are as follows:
1. The survey needs to be conducted over a clearly defined and limited area.
2. Evidence is needed not only of the kinds of species present but also of their density.
3. The distribution of the different species ought to be mapped and any changes noted.
4. Interactions between the animals and adjacent plant populations should be noted.
5. The survey should be repeated as regularly as possible.
Perhaps these guidelines may be of help to anyone wishing to carry out a similar survey in future.
BUTTERFLIES OF HAZELBURY COMMON
Hazelbury Common is a small triangular area of grassland (see map) situated south-east of Box and
Chapel
ACs
Drive to Hazelbury Manor
—— Metalled roads
= Grass track
Map of Hazelbury Common, showing survey areas
NOTES
south-west of Chapel Plaister (ST 840678). The soil is well drained and overlies a layer of Jurassic oolitic limestone. There is evidence of an ancient strip cultivation running from east to west. The longest axis of the common is a southerly bank of 353m bordering a narrow tarmac road which joins the A365 to the south. The northern boundary consists of a dry stone wall 289m long while to the west, a thick hedge stretches for some 161m.
For our survey, we divided the common for refer- ence into four areas, each of which is bordered by one or more grassy tracks. The northern area is bounded by the wall and a number of adjacent trees and shrubs; the south consists of the grass bank with intermittent shrubs of hawthorn and dog rose; the west is bounded by a thick hedge of trees and shrubs, while the central area is grassland kept fairly short by a colony of rabbits. At its east end the common tapers
155
almost to a point and opens onto the drive leading to Hazelbury Manor.
The following records of the 20 species of butter- flies colonising Hazelbury Common are the result of regular fortnightly visits from late March until the end of August, 1991. As with the invertebrates of Middleton Down, the value of the data could be increased by a more objective approach, for instance, the more precise estimation of the numbers of the different populations.
The butterflies of Hazelbury Common can be divided for convenience into two groups (see Table). Vagrant species are periodic visitors which can occur almost anywhere. The food plants of their larvae are usually absent and no doubt most of them breed in adjacent meadows where their foods are plentiful. Resident species are those that probably breed in the area where the ecological conditions are favourable.
Species of butterflies colonising Hazelbury Common
VAGRANT SPECIES
Family Pieridae
Orange Tip
Anthocharis cardamines
Brimstone Gonepteryx rhamni Family Nymphalidae Large White Pieris brassicae Small Tortoiseshell Aglais urticae Small White P. rapae Peacock Inachis 10 Green-veined White P. napi RESIDENT SPECIES
Species Distribution
Family Hesperiidae Small Skipper
Family Lycaenidae
Small Copper Lycaena phlaeas
Brown Argus
Aricia agestis
| Common Blue
Thymelicus sylvestris
Polyommatus icarus
Common in all areas of long grass. Attracted particularly to flowers of knapweed Centaurea nigra.
Uncommon but widely distributed. Particularly attracted to flowers of marjoram Origanum vulgare in the SW corner.
A small colony in the SW corner bordering the hedge. Like the blues, the butterflies are much attracted to flowers of marjoram.
A flourishing colony in the SW area, the butterflies being particularly attracted to flowers of marjoram. The colony on the N side in 1990 declined sharply in 1991 to only a few insects.
156
RESIDENT SPECIES
Chalkhill Blue
Holly Blue
Family Nyphalidae Dark Green Fritillary
Family Satyridae Speckled Wood
Gatekeeper
Marbled White
Meadow Brown
Ringlet
Small Heath
LIVING ORGANISMS IN A CHANGING ENVIRONMENT
It could be argued that environmental variations take place so slowly and in such an irregular manner that fluctuations in plant and animal distribution are
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Species
Lysandra coridon
Celastrina argiolus
Argynnis aglaja
Pararge aegeria
Pyronia tithonus
Melanargia galathea
Maniola jurtina
Aphantopus hyperanthus
Coenonympha pamphilus
unlikely to prove sufficiently sensitive indicators of | change. Our past climatic history suggests, however, | that this argument may be invalid. Thus in the 13th century, conditions were warm enough for vineyards | to flourish as far north as York, while 100 years later |
Distribution
A small colony of perhaps 20 insects or fewer occurred in the SW corner of the area near the W hedge. Its dis- tribution was therefore unrelated to that of its foodplant, horseshoe vetch Hippocrepis comosa which is wide- spread.
A butterfly of shrubs and particularly numerous along the western hedge where there are a few trees of holly. The second brood in August also fre- quented the ivy covering the N wall.
A single freshly emerged male was seen on 14 July settled on knapweed on the south bank. The rarity of this species cannot be attributed to its foodplant, common dog-violet Viola riviana, which is abundant.
A small colony of a few insects restric- ted to vegetation bordering the west- ern hedgerow.
Common on all grassland.
Common, particularly in the central area. Much attracted to flowers of knapweed.
Common throughout the area. The butterflies are particularly attracted to flowers of knapweed and field scabi- ous Knautia arvensis.
A small colony in the long grass bordering the western hedge. Its dis- tribution overlaps that of the meadow | brown in this area. | Widespread at a low density in all | grassy areas.
NOTES
we were faced with a mini-Ice Age of extreme cold and snow lasting more than 100 days each year. During the short intervening period, changes in invertebrate distribution could well have been dra- matic. More recently, variations in our climate have been considerably less severe. Yet, of the 40 species of butterflies recorded in England by Moses Harris in 1766, four are now extinct and a number more are either on the verge of extinction or have drastically altered their range.
Today, we are faced with a new and largely unpre- dictable situation caused by man’s pollution of the environment, giving rise to the greenhouse effect, great storms and a hole in the ozone layer. At present we have little idea of the direction in which these influences are leading. Perhaps, if we are able to look back in say 25 years’ time, the distribution of living organisms such as invertebrates in localities like
ey)
Middleton Down and Hazelbury Common may play a significant part in aiding our assessment.
Acknowledgements. Our thanks are due to Mr Edward Elliott and Mrs Phillida Sneyd for kindly reading parts of the draft of this paper and for their helpful comments.
BIBLIOGRAPHY
CHINERY, M., 1979 A Field Guide to the Insects of Britain and Northern Europe (2nd edn.), Collins
FORD, E.B., 1957 Butterflies (3rd edn.), Collins New Natur- alist Series
GOLDSMITH, F.B. (ed.) 1991 Monitoring for Conservation and Ecology, Chapman and Hall
HARRIS, MOSES, 1766 (reprinted 1986) The Aurelian, Country- side Books
MANLEY, G., 1952 Climate and the British Scene, Collins New Naturalist Series
THOMAS, J., and LEWINGTON, R., 1991 The Butterflies of Britain and Ireland, Dorling Kindersley
Wiltshire Archaeological and Natural History Magazine, vol. 86 (1993), pp. 158-164
Excavation and Fieldwork in Wiltshire 1991
Alderbury: Alderbury and West Grimstead School (SU 194272); Post-Medieval
An archaeological evaluation was carried out by the Trust for Wessex Archaeology (TWA), on the pro- posed site of a new primary school on the north-east edge of Alderbury. The work revealed a low density of predominantly modern artefacts in the topsoil and an absence of archaeological features. A crop-mark feature, identified on aerial photographs as a possible ploughed-out barrow, proved to have no sub-surface components within the evaluation trenches.
Amesbury: Butterfield Down (SU 167412); Romano- British
An archaeological evaluation (Phase 4a), was under- taken by the TWA at Butterfield Down, Amesbury in December 1991 following a geophysical survey of the site. Previous archaeological investigation has revealed an undefended Romano-British settlement extending over an area of at least 6 ha (Richards 1991). The settlement dates to the 3rd and 4th centuries AD but there is also evidence for both prehistoric and earlier Romano-British activity.
The geophysical survey identified the location of a number of enclosures and smaller discrete features. The evaluation was designed to investigate these features and to determine the possible presence of structural elements im situ above the chalk bedrock. Seven trenches were positioned in relation to suspec- ted features. Areas within which no features were located by the geophysical survey were also investi- gated in order to check the accuracy of the survey results.
The evaluation trenches located features in trenches 28, 29, 31 and 32, but encountered none in trenches 27 and 30. This corresponds well with the results of the geophysical survey. Two intersecting ditches recorded in trench 33 were not, however, located by the survey. No artefacts were recovered during the evaluation but several sherds of Romano-British pot- tery were observed, both from the surface of features and within the overlying soil layers, as was a quantity of animal bone. On the sloping ground in the west and north-west of the area the remains of earlier plough- soils and some colluvial build-up were recorded below the modern topsoil. The archaeological features were all sealed by these soils rather than cutting through them. On this evidence it is unlikely that any struc-
tural archaeological remains exist above the chalk bedrock.
The evaluation has shown that within the southern part of the area designated as Phase 4A is a continua- ton of the undefended Romano-British settlement already observed to the east. Geophysical survey suggests that this is the edge of the settlement. The enclosures indicated by the survey are similar in size and form to others previously located by aerial photo- graphy and geophysical work on the eastern side of Butterfield Down. These may also lie on the edge of the settlement. No such enclosures have been located in the areas already observed and this distinction between centre and periphery emphasises the import- ance of Phase 4A.
The evaluation was commissioned and financed by Gleeson Homes Plc and the assistance of their Tech- nical Manager, Mr Nigel Hogg is gratefully acknow- ledged.
REFERENCE
RICHARDS, J.C., 1991 Butterfield Down: from Prehistory to the Present Day. Wessex Archaeology, unpublished report, Salisbury
Avebury: Avebury Manor (SU 09857000); Mostly Medieval
In autumn 1991 the TWA was commissioned by the | National Trust to carry out an archaeological | watching brief at Avebury Manor. The work, super- vised by C.A. Butterworth, resulted from the demoli- | tion of buildings constructed by the previous owner | and involved observing the reinstatement of the park- lands and gardens of the Manor, as well as the | selective excavation and recording of exposed archaeological deposits.
A high density of archaeological features was | recorded in areas where reinstatement involved the | exposure of natural subsoil. No features were cer- | tainly prehistoric in date and most were shown to be 11th-14th century. These features consisted mostly of boundary ditches, and the complexity of the strati- | graphy suggested several phases of activity. One of) the existing red brick garden walls was shown to rest} on a wide and solid sarsen foundation, part of an | earlier, substantial structure.
The watching brief was undertaken in conjunction
EXCAVATION AND FIELDWORK IN WILTSHIRE 1991
with earthwork and geophysical surveys within the grounds of the Manor.
Avebury: Overton Hill (SU 116685); ?Prehistoric Fieldwalking was undertaken by the Chippenham College Practical Archaeology Group (CCPAG) in October 1991 in the area close to the old Ridgeway Cafe site, west of the Ridgeway and up to the A4. A few plough-damaged flints were found, none of which has so far been identified.
Bishops Cannings: Easton Down Long Barrow (SU 064661); Neolithic, Early Bronze Age
Research excavations were undertaken by Dr Alasdair Whittle (School of History and Archaeology, Univer- sity of Wales, Cardiff) on high chalk downland south of the Kennet at the Easton Down long barrow, Bishops Cannings, as part of a project to investigate the Neolithic sequence and context in the Avebury area. The barrow has been previously described by Barker (WAM 79, 1985) and is scheduled. Limited cuttings were made to recover dating material and samples for environmental reconstruction.
Two ditch cuttings showed the usual sequence of primary and secondary infilling. Bone and antler samples from primary contexts have been submitted for accelerator dating. Sherds of Early Bronze Age style came from a soil at the top of the secondary fill, and the tertiary fill includes crude flintwork, probably of later prehistoric date.
A partial mound cutting (towards the west of the barrow) revealed a buried soil. This had been dis- turbed, probably by cultivation as well as by clearance; before the mound was built a thin turfline had developed. Molluscan analysis has shown open conditions, probably localised, in the area where the barrow was built, which closed in during the period represented by the secondary filling of the ditches. A
limited lithic survey around the barrow produced very little struck flint. It is likely that the barrow was set in arelatively little used part of the Neolithic landscape, which was not substantially cleared till the Late Neolithic or Early Bronze Age.
| Survey and excavation showed that the built barrow | was comparatively small. The sparse record of Thur- \nam’s investigation in the 19th century suggests that there may have been few burials at its eastern end. A ‘date late in the barrow sequence is predicted. Britford: Odstock Golf Course (SU 14102760); Multi- period, mostly Bronze Age
|A fieldwalking survey of a c.34.5 ha site at Odstock, jnear Salisbury, prior to the submission of a planning
159
application to construct a golf course on the site, was undertaken by the TWA. The archaeological poten- tial of the area had already been recognised owing to its general proximity to the Iron Age hillfort of Great Woodbury and observations made from aerial photo- graphs.
The survey revealed the presence of high levels of worked flint, probably of later Bronze Age date, across the site. The assemblage was dominated by flakes with only a few retouched forms, mostly scrapers, and a few cores and rejuvenation flakes. Other finds, including pottery and glass, were predominantly post- medieval or modern.
The evaluation was commissioned and financed by the Longford Estate.
Castle Combe: Castle (around ST 845778); Medieval
Further survey work was undertaken by CCPAG prior to golf course construction on and adjacent to the castle site. The outer faces of the banks of both baileys were found to be stone-revetted, and the castle mound to be crowned by the foundations of a shell keep with at least one tower. Traces of structures marked on the O.S. 25in map can still be made out. Golf course construction works have necessitated clearing a length of the outer bailey bank and the formation of a green within it. The remainder of the castle is overgrown with vegetation which is damaging the stonework.
Chippenham: Milbourne Farm and Showell Nurs- eries (ST 913714); Prehistoric and Roman
An evaluation undertaken by Gill Hey for the Oxford Archaeological Unit revealed activity of Late Neolithic/Early Bronze Age, Iron Age and Roman date over 25 hectares on a terrace of fluvial sands and gravel overlooking the River Avon. The work was funded by the potential developer, Greenham’s Con- struction Materials Ltd.
Evidence of Late Neolithic/Early Bronze Age activity included pits (one containing Beaker pottery) in the north-east of the evaluation area, and gullies, postholes, and two ditches on its southern edge. Other features of this date were two linear ditches, one possibly a boundary running north-south. Undated features in these areas could be contem- porary. Flints were also retrieved from the ploughsoil in these areas.
Two Middle Iron Age features were examined: a pit in the north-east and a ditch in the centre of the evaluation area.
Roman activity was present on the site in the two
160 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
areas where cropmarks had been observed from the air. In the north, enclosure ditches and gullies con- tained debris suggesting nearby domestic occupation. This occupation appeared to date principally to the late 1st and 2nd centuries AD, though some earlier Ist century and 3rd century material indicated that this settlement could have been longer-lived. Possible trackway ditches observed running south-east to north-west (to the river) were examined in the east of this area.
Further south lay another group of cropmark features tentatively dated to the Roman period. Field boundary ditches, ditches which were probably part of a large enclosure, and trackway ditches heading from the south-west to the river were all examined. Finds were scarce in this area indicating that they were situated away from domestic activity.
A medieval gully was excavated and a post- medieval field boundary ditch was located running approximately north-south through the southern fields.
Collingbourne Ducis: St Andrew’s Church (SU 24225362); ?Post-Medieval
At the request of the Sarum Partnership, the TWA inspected the dismantling of the east boundary wall of the churchyard, sections of which were to be rebuilt. The wall was of a typical post-medieval, Wiltshire chalklands construction, consisting in the main of mortared flint strengthened with courses of brick, and capped by brick. The wall height was c.1.3m on the outer face and 0.8m on the inner, with wall footings comprising four courses of brick headers on the outer face and two courses of stretchers on the inner, resting on rammed flint foundations. A portion of an oolitic limestone quernstone was incorporated into the walling.
Compton Bassett: Roach Wood (SU 041782); Medieval
Trial excavations on a flight of scarp-face strip- lynchets were carried out by Andrew Reynolds on behalf of the Compton Bassett Research Project. Two trenches were located in order to demonstrate the relationship between the lynchets and the bank and ditch system that encloses them. It was shown that the enclosure was the later feature and that it had prob- ably been constructed some time after cultivation had ceased. Although no foundation date was established for the lynchets, a jetton, dated to 1550-85, was found within the material forming the enclosing bank. Further work is to be carried out to establish the
purpose, date range, sequence and nature of develop- ment of the lynchets.
Compton Bassett: (SU 043723); Multi-period Members of the Compton Bassett Research Project under K.N. Wilkinson drilled five bore-holes into sediments accumulating in the bottom of a narrow coombe in order to locate and map colluvial deposits. It was found that up to two metres of colluvium had accumulated, probably caused by both prehistoric and historic agriculture on the chalk to the north. Within the lower 20cm of colluvium loessic deposits were found, probably deriving from a former soil on the chalk, whilst beneath it a possible soil was located. Work in 1992 will focus on the dating and mapping of the deposits and paleoenvironmental reconstruction using sedimentological and molluscan data.
Downton: Tannery House (SU 18082142); Pre- historic—Post-Medieval
An archaeological evaluation by the TWA, at Tannery House, Downton, produced evidence of activity ran- ging from the prehistoric to post-medieval periods. Test-pits were dug within Tannery House gardens, and machine trenches in an adjacent field. Downton Moot lies immediately to the east and south of the site and consists of a complex of earthworks forming one of the largest ringwork and bailey castles in England. There is evidence of prehistoric, Roman and Saxon activity in the immediate vicinity.
Very few archaeological features were recorded and there was no sign of the outer bailey ditch of Downton Moot, either within the garden or in the field trenches.
A small quantity of undiagnostic struck flint was recovered as well as 182 pottery sherds. The bulk of. the assemblage is of post-medieval date, but there is also a small group of medieval sherds (11th—12th centuries). A handful of possible Saxon material and) two sherds of Romano-British Black Burnished Ware} were also found, along with clay pipe fragments, roof! tiles and a piece of worked bone; the last, a fragment) from a narrow strip decorated with incised ring-and-| dot motifs, is probably part of a mount, perhaps from a casket, and is unlikely to be later than early 13th century in date.
The situation of Downton Moot adjoining the site might suggest that components of the castle survive within the grounds of the current development site, especially if the course of the outer bailey bank i: projected. The results of the evaluation apparently show, however, that little or no trace of this monu!|
EXCAVATION AND FIELDWORK IN WILTSHIRE 1991
ment survives within this area. (The garden was substantially landscaped in the post-medieval period. )
This project was financed by Barlows plc and commissioned through their consultants, Colwyn Foulkes and Partners.
Durrington: pipeline observations (SU 44001550— 44001485); Prehistoric, Post-Medieval
In November 1991, the TWA monitored the con- struction by Wessex Water of the final 600m of a double water main from Bulford Road, Durrington, to the Packway reservoir, and the construction of a pesticide filtration bed at the reservoir. At several points the southern edge of the easement runs within 120m of the northern edge of Durrington Walls and within 50m of the later Bronze Age-Iron Age rect- angular ditch known as the Packway Enclosure.
A number of Neolithic, possibly Bronze Age and post-medieval features were recorded. Prehistoric activity was represented by small pits or shallow scoops generally containing burnt flint, animal bone fragments, worked flints and, in some cases, flecks of charcoal. These are interpreted as having been hastily dug to receive the remains of camp fires, incorporat- ing flint knapping waste in some cases. Two pits contained a few sherds of pottery. Pit 184 produced a large number of struck flints, some Neolithic or Bronze Age sherds, animal bone and an antler pick, whilst pit 165 included a few sherds of ?Grooved Ware. None of the pits could be directly associated with either of the two major monuments in the vicinity though some transitory activity connected with the henge monument seems likely.
Post-medieval features consisted of a number of field boundary ditches and a probable trackway.
The TWA would like to thank Wessex Water, who funded the project, AVRON Construction, Raymond Brown Ltd and AMEY Construction for their help and cooperation.
Enford: East Chisenbury (SU 146532); Late Bronze Age to Early Iron Age
David McOmish and David Field report that during _ the course of fieldwork undertaken by members of the | Salisbury Plain Training Area Conservation Group, a surface concentration of Iron Age pottery and other | artefacts was noticed in a small plantation near East | Chisenbury. A surface collection of all visible ceramic material yielded c.900 sherds from an area of c.20m/?. | The volume, variety, style and chronological classi- | fication of this material indicates that the site is one of |major significance in the late Bronze Age-early Iron Age transitional phase. The assemblage contains both
161
decorated and undecorated fine and coarsewares, and is especially notable for the large number of short- necked furrowed bowls present, but there are no scratched-cordoned vessels or other later Iron Age forms.
A small scale exploratory excavation in the area of a large low mound, immediately east of the plantation, confirmed the presence of very substantial and complex midden deposits, more than 2m deep in places. The midden material is dark and greasy in texture and contains large quantities of pottery, bone, stone, wood and charcoal. The blade end of a sock- eted axe, probably of Sompting Type (Burgess 1969; Coombs 1979, 253-69), was found at the base of the trench. Within the excavated layers of the midden, there was a series of building floors of rammed chalk and other prepared surfaces combined with occupa- tional debris. An auger survey across the hilltop suggests that the deposit, which in places has been truncated by modern ploughing, covers an area of at least S ha (12 acres).
The midden appears to be associated with an enclosure to the east, previously noted by Colt Hoare as being partially defined by a pronounced rampart (1812, 192), in a prominent location at the west end of a chalk spur at 145m OD. Overlooking the River Avon valley to the west, it is flanked to the north and south by dry valleys and has extensive views to the east over the Salisbury Plain chalk plateau. The enclosure is univallate, now defined mainly by a wide, low bank, c.20m wide, enclosing a roughly circular area c.3 ha in extent. The relationship between this feature and the midden has not yet been established, though a ground modelling survey undertaken by RCHM(E) seems to indicate that the latest phases of the midden deposit overlie the enclosure boundary.
The evidence for settkement combined with the midden remains would suggest that the site at East Chisenbury is similar to a Near Eastern tell site, where successive phases of occupation, rebuilding and rubbish disposal on the same spot lead to the build-up of a mound, often of large dimensions. There are no direct parallels for the site at East Chisenbury. All Cannings Cross, the type site for the period, provides the most appropriate comparison (Cunnington 1923). Here, chalk floors and burnt clay spreads overlie midden deposits, though no associated enclosed settlement component was noted.
The East Chisenbury complex lies in an area of intense late Bronze Age-early Iron Age activity. Chisenbury Trendle hillfort (Cunnington 1932, 1- 3), which may be contemporary with the latest phases of the midden, lies c.600m to the north-east. Lidbury
162 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Camp (Cunnington 1917, 12-36), which does have a comparable assemblage, is c.2km to the east. Grinsell notes at least one other settlement site of similar date in the vicinity, near Upavon Aerodrome, at SU 147540 (1957, 69).
Middens of this date, 800 BC—600 BC, which represents the transition between the late Bronze and early Iron Ages, are very rare. The exceptional preser- vation of the massive midden and its incorporated settlement at East Chisenbury make this site unique and of extreme importance. Further investigation of the midden and the contiguous enclosure is planned and a full publication report is in preparation.
REFERENCES AND BIBLIOGRAPHY
BURGESS, C., 1969 ‘Some Decorated Socketed Axes in Canon Greenwell’s Collection’, Yorks. Arch. 7. 42, pt. 3, 267-72
COLT-HOARE, SIR R.C., 1812 History of Ancient Wiltshire, Vol. 1, London, 192
COOMBS, D., 1979 ‘The Figheldean Hoard, Wiltshire’, in C. Burgess and D. Coombs (eds.), Bronze Age Hoards, Some Finds Old and New, BAR 67, Oxford, 253-69
CUNNINGTON, M.E., 1917 ‘Lidbury Camp’, WAM 40, 12-36
1923 The Early Iron Age Inhabited Site at All Cannings
Cross, Devizes
1932 ‘The Demolition of Chisenbury Trendle’, WAM 46, 1-3
GINGELL, C., and LAWSON, A.J., 1984 ‘The Potterne Project: Excavation and Research at a Major Settlement of the Late Bronze Age’, WAM 78, 31-4
GRINSELL, L.V., 1957 ‘Archaeological Gazetteer’ VCH Wilt- shire, Vol. I, Part I, London
Enford: West Chisenbury and Compton (SU 1451); Medieval
Two shrunken settlements in the parish of Enford were surveyed by G. Brown and D. McOmish during 1991 as part of a study of medieval settlement in the Avon valley.
At West Chisenbury (SU 137527) settlement earth- works lie on the west bank of the River Avon. They comprise three elements of hut platforms and _ hol- loways. In the north there is a paired moated site with a dove cote and fish pond. This is probably the location of the demesne messuage documented in 1361.
At Compton (SU 134521) settlement earthworks are situated on either side of a dry stream which flowed from Water Dean Bottom to the River Avon. To the south, a planned element is indicated by property boundaries 35-40m wide extending from the stream to a back lane. A c.15m diameter depression by the stream may be the site of a former pond. A bank, with underlying lynchets, marks the western extremity of the settlement. To the north of the
stream there are a number of irregular platforms, one of which is in a prominent position; it measures 30m x 15m, and may be the site of the chapel recorded in 13:
Hamptworth: Golf Course (SU 237200); ?Mesolithic An archaeological watching brief by the TWA was carried out on the site of a proposed eighteen-hole golf course. No archaeological features were revealed but a quantity of struck flint from the extreme west end of the site included a bladelet core and several blades of Mesolithic affinity.
The project was funded by Hamptworth Park Leisure Ltd.
Marston Meysey: Roundhouse Farm (SU 135964); Prehistoric (Bronze Age and Iron Age), Roman and Medieval
The site, which covers approximately 55 hectares, lies on first terrace gravel on the north bank of the River Thames. It is at present agricultural land, part ploughed, part pasture. An application for gravel extraction by Greenham Construction Materials Ltd led to a planning enquiry, at which it was suggested from cropmark evidence that there was a Neolithic causewayed enclosure and a bank barrow on the site. Cropmarks also indicated at least two ring ditches and an extensive Roman field system, within which evid- ence of occupation was anticipated.
A field evaluation was carried out by Tim Allen for the Oxford Archaeological Unit in September and October 1991. The evaluation, using machine- stripped trenches followed by hand-cleaning and selective excavation, was aimed primarily at estab- lishing the origin and date of the cropmarks. Tren- ches were also targeted upon large areas thought to be covered in alluvium, including a possible former stream course crossing the site from west to east. |
The cropmarks of the ‘Neolithic enclosure’ proved, | however, to be of natural origin, and the parallel | linear cropmarks of the presumed ‘bank barrow’ late | medieval or later. Two circular cropmarks were con- | firmed as barrow ditches. The Roman field system | was indeed extensive, but hardly any finds were | recovered and any settlement must have lain outside | the area of the evaluation.
In contrast to this, a series of Iron Age settlements was found, only one of which had been suspected from the cropmark evidence. These were strung out along the former stream course, which proved to have | been active in the Iron Age. Alluviation began in the | Late Iron Age, and had sealed some of the Iron Age) features. |
EXCAVATION AND FIELDWORK IN WILTSHIRE 1991
In the Early Iron Age the site was divided by linear boundaries, and at its east end a settlement repres- ented by postholes, shallow pits and occupation debris lay alongside the stream. By the Middle Iron Age there were four separate areas of settlement, the largest of which was very well preserved, and very rich in both finds and features. This site, approxim- ately 100m by 70m, contained a number of round- houses and was enclosed by a ditch. The others were all of different character: one was an open settlement containing roundhouses and linear ditches, another was a single rectangular enclosure linked to a series of fields or paddocks, and the third consisted of small penannular enclosures.
The presence of both Early and Middle Iron Age settlements on the first terrace here, with the added environmental potential of the low-lying situation, makes this an unusual and important site in the Upper Thames Valley in the Iron Age.
Minety: (SU 014912); Medieval
A survey of upstanding features was made by CCPAG in a field containing a medieval pottery waster deposit excavated by John Musty (see WAM 68, (1973), pp. 79-88). The earthworks surveyed represent the remains of a small hamlet, comprising house plat- forms and a holloway, presumably inhabited by pot- ters whose kilns were sited in their back yards.
Old Sarum: Airfield (SU 13151033); Modern
An archaeological assessment was carried out by the TWA to the south-east of Old Sarum Airfield. The area has been used for the storage of gliders and light aircraft. The chief interest here lies in the proximity of the site to the Roman Road known as the ‘Portway’ and the junction of several Roman roads at Old Sarum. In the event, six strip trenches revealed only
modern disturbance and there were no archaeological finds.
Salisbury: Cathedral Close approach road (SU - 142294); Medieval—Post-Medieval In October 1991 the TWA undertook limited test- _ pitting along a proposed route for an access road to _ the Cathedral Close in Salisbury to assess the likeli- hood of survival of in situ archaeological deposits. No _ previous archaeological work is recorded in this area. Test-pitting to the rear of De Vaux Place revealed deep soil layers of post-medieval origin sealing appar- ently undisturbed medieval deposits. The lower soil layers in test pits 1 and 2, surviving at depths of approximately 0.55m below the present ground sur- face, contained pottery of only medieval date. No
163
evidence of associated structures was observed but the presence of tile and, particularly in the lower medieval layers, dressed building stone, suggests that there may be remains of structures within the vicinity of the site, perhaps protected by the depths of soil which have accumulated within the orchard through which part of the proposed route passes. The work was funded, and access granted, by the Dean and Chapter of Salisbury Cathedral via the Deputy Chapter Clerk, Mr Andrew Dawson.
Salisbury to Coombe Bissett: water pipeline (SU 1026—-SU 1328); Period uncertain
A watching brief by the TWA was undertaken in February 1991, covering 325m of pipeline trenching between Salisbury and Coombe Bissett for Wessex Water. The proposed route ran north-east through farmland roughly parallel with the A354, turning south-east along Green Lane to Howard Hospital. A number of field systems occur close beside the route but only one archaeological feature, a U-shaped ditch, 1.1m deep and 2.2m wide, crossed the route itself. No artefacts were recovered and the line of the ditch did not correspond with any known cropmark features.
Wanborough: Foxbridge Farm (SU 20208425); Late Medieval
An 80 metre long bank at right angles to Ermin Street was sectioned by the Thamesdown Archaeological Society, under the direction of Chris Chandler, Thamesdown’s Community Archaeologist. The earthwork was confirmed as a late medieval field boundary, consisting of a bank and ditch.
Pottery from the ditch was of 14th century ‘Minety’ type. Although a metal-detected hoard of Romano- British coins had earlier been found some 30 metres to the south, no evidence of this period was found.
Warminster: Horningsham Congregational Chapel (ST 81224115); Post-Medieval
Observations by the TWA, during the excavation of drains adjacent to the 16th century Congregational Chapel at Horningsham, revealed hitherto unre- corded architectural features within the fabric of the chapel itself. The work was funded by English Heri- tage. The architectural features comprised a blocked opening in the south wall, an exterior membrane wall serving the west and south walls, and a possible extension of the west wall to the south. The ‘mem- brane’ wall was 0.33—0.36m thick and extended along the west and south sides of the chapel, terminating at the western buttress. At its north-west end it was
164 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
butted by the graveyard wall. It consisted of rough courses of stone and brick fragments with a pale buff mortar set into the outer face of a narrow trench and braced on its inner face by narrow stone spacers. The courses along the southern extent were not horizontal, but mirrored the slope of the ground. Three features were observed behind the membrane wall: a blocked opening in the south wall beneath, and partly occu- pied by, the west window; a stub end of the west wall extending 0.1m beyond the end of the south wall; and a horizontal void in the west wall beneath the William Barnes Memorial.
The construction of the membrane wall probably relates to rebuilding works undertaken in the middle of the 19th century. Although it is not referred to
specifically in the available history of the building, it seems likely that it was included in the repairs undertaken in 1863 as an attempt to prevent ground water seeping into the west wall of the chapel by channelling it to a soakaway. The blocking of the opening in the south wall is unlikely to have been later than the construction of the membrane wall and was probably done at the same time. There is no previous record of a side entrance in the south wall and the ground level at this point must have been at least 0.5m lower than it is now for it to have been usable. The higher ground to the west may have been revetted by a southward extension of the west wall, now surviving only as a stub protruding at the south-west corner of the chapel.
Wiltshire Archaeological and Natural History Magazine, vol. 86 (1993), pp. 165-177
Reviews
Joan Blanchard. Malmesbury Lace. B.T. Batsford, 1990; 126 pages; illustrations and patterns. Price £14.95, hardback. ISBN 0 7134 66154.
Malmesbury Lace traces the history of the lace indus- try in the town during the 17th and 18th centuries, when the trade was a thriving cottage industry, through to its revival at the beginning of this century and up to the present day.
Joan Blanchard, a lace teacher from Chippenham, was encouraged in her researches through contact with the Athelstan Museum in Malmesbury. Her book follows the history of lacemaking in the district from the early beginnings of English Bonelace and the period of French and Dutch settlement in Wilt- shire. References to lace schools in the area obtained from the Wiltshire Record Office and Malmesbury parish registers, show entries as early as the 17th century.
In 1793, Old Cannop’s Mill in Malmesbury was purchased by Mr Hill, who opened a cloth factory, but there is evidence to show that local inhabitants could earn more by lacemaking than by working at the factory. The records show an active cottage industry at this period, when lace workers were earning eight to ten shillings a week. Eventually the mill changed hands and was later reopened as a silk mill in 1852 by Bridget Thomas & Co. The new mill contributed to a decline in lace making since better pay attracted many lace workers to employment there instead. By 1862 the silk mill employed 280 workers, including children, and by 1900 only six lacemakers were left.
A revival began in 1907, at which time a lace school was set up in the Kings Arms Hotel, Malmesbury, organized by the Countess of Suffolk. Classes were held once a week, for an hour, and young women and children attended from the surrounding villages. One of the pupils, Annie Goodfield, became known for her skills and in later life demonstrated lacemaking in the town.
Malmesbury lace is a fine geometric lace made with the Bucks Point technique, using very fine thread with unspangled bobbins. A strong association with Downton lace meant that many patterns were exchanged and copied, but several patterns are unique to Malmesbury. It is suggested that similarity
between the laces may only have developed owing to better communications between Malmesbury and Downton with the introduction of the Penny Post in 1840. Certainly the bobbins differ, though Downton bobbins are also unspangled.
The book contains many examples of the lace patterns held in the Athelstan Museum Collection (with local names), and others from various local sources. The 30 or so patterns in the book are clearly defined, with pricking instructions, photographs and techniques for making the lace explained. Some adaptations of the original designs are made into simple motifs suitable for jewellery and other decora- tive uses for the present day.
Today the Athelstan Museum holds lace demon- strations and the children are taught to make lace at Malmesbury Primary School. It is also pleasing to note that Malmesbury lace is now studied for the first time by City and Guilds Certificate students. Joan Blanchard’s book will broaden awareness among lace- makers and others of this fine lace, unique to the history of the ancient borough of Malmesbury.
ROBERTA PRINCE
John Chandler. West Country Landscapes: The Vale of Pewsey. Ex Libris Press, 1991; 160 pages; illustrations. Price £6.95, paperback. ISBN 0 948578 30 0.
If we can point to one thing which the author of this book holds in common with the earliest occupants of the Pewsey Vale, it is surely an eye for location. Seeking a starting point from which to begin his journey through time and change, he has chosen unerringly Adam’s Grave, the Neolithic ?chambered long barrow crowning the summit of Walker’s Hill, and ideally sited to command the breathtaking view southwards across the Vale. Follow the minor road leading to Wilcot from Alton Priors for a short distance until hill summit and barrow come into full view, dark in outline against the sky. What a point of vantage! If it were possible to envy the dead therein interred, one might grudge them their high eminence — this resting place of kings and lords of a dozen counties.
166 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
We are presented here with an account of the gradual evolution of a particular landscape, progress- ing in Dr Chandler’s own phrase from ‘the inanimate to the very animate’. His inanimate phase, spread over vast time and through earth movements, climatic change and glacial erosion, was ultimately to bring about soil conditions providing for the dissemination of plant and animal life. Man, the very animate branch of his scheme, emerges in the Neolithic period to take advantage of a now favourable environment and begin the process of crop growing and animal husbandry.
Early chapters summarise the slow progress towards an organised farming system, first observed in the surviving traces of field systems and ranch boundaries of the later Bronze Age. A leap forward from the tentative agriculture of prehistory began during Saxon times, developing towards the medieval system of large open fields divided into strips — the furlongs, hides, tithings and eventually modern-style parishes providing for cattle breeding and sheep and corn husbandry. Carefully organised to allow commu- nities to pursue a full farming economy, it was a pattern of agriculture that survived into the 19th century though subject to continual change as populations increased and higher food production was demanded. Thus land development was extended, relationships between tenants, lessees and landowners altered, and new uses were applied to the land. Inevitably the old-style medieval farming was out- moded, and with the passing of the 18th- and 19th- century Enclosure Acts, a new era of farming came into being. With the arrival of farm machinery we move into modern day agriculture.
After agricultural history, the principal forms of settlement in the Vale — the village and hamlet, are discussed; their location, layout, and some of the factors, natural or man conceived, which created prosperity or led to their decline. Some buildings were of especial significance, the church an obvious example, not only as a place of worship, but in earlier times, a vital centre of secular activity. Wealthy landowners and the monasteries played a formative role in instigating architectural improvements within villages but in later days, with the rise of industri- alisation, the railway and the Kennet and Avon Canal, there was greater change and expansion.
How were movement and travel, both long dis- tance, and locally within the isolated confines of the Vale organised? Stull indispensable it seems, in medi- eval times, were the great ridgeways, droveways and herepaths of earlier antiquity for driving cattle to market and fair; but local lanes linking villages to farm and field were created and maintained by indi-
vidual parishes. However, as wheeled traffic began increasingly to be used for long distance travel, the creation and turnpiking of a main road system became a necessity. The building of the Kennet and Avon Canal and, not long afterwards across the South- West, the railway drew wider horizons into view for the Vale people; such amenities in turn made for further change and development within the villages. With the arrival of the combustion engine, mobility was made complete.
In a small chapter, the place and purpose of folklore within the daily lives of early communities are con- sidered, how it arose and whether based on reality, or mere fantasy. Today, it is almost with surprise that we recognise its virtual necessity in an age when so many of the events and phenomena of everyday living were without rational explanation. Perhaps man, like nature, abhors a vacuum; do we then explain the origin and continuation of folklore as the need to fill the void at all costs? Thus it could dispel terror, uncertainty, forecast the weather, palliate disease, repel evil, even through folk custom exact rough justice. How for certainty then, do we thus accept the continuation of folklore into the present day, as instanced by the recent phenomenon of ‘crop circles’, particularly well recorded in Wiltshire? Has the pleth- ora of ‘suggestions’ put forward to explain them created a modern piece of folklore? It is of some interest to note that during a recent Devizes Arts Festival, a lecture given on ‘The Phenomenon of Crop Circles’ was heavily overbooked when other events were poorly attended. So, the urge for folklore still lingers, willingly helped along, we suspect, by archaeology’s lunatic fringe.
From conversations with surviving octogenarians, the author in a final chapter recalls for us the unhur- ried existence of Vale villagers at the turn of the century. Nervous of change or innovation, hardly affected by the world outside, their lives revolved unchangingly around the farming seasons of the year. Do I catch here a sense of regret echoed at the passing of a part of rural England, never to be recalled save in record or memory?
I commend this book to all students of landscape history. Dr Chandler is a fine scholar, thoroughly researching his subject at every period for data in support of his narrative. Field investigation and excavations, many carried out within and around the Vale, have provided scientific and artefactual evidence for its evolution through geological time to the begin- nings and progress of agriculture from prehistory up to modern history. Documentary sources, charters, place names, taxation surveys, census records and
REVIEWS
early maps, have been searched to explain the growth of communities, their settlements, buildings, social habits and customs.
But he is no dry as dust historian. He has an enthusiasm for his subject which goes beyond mere fact finding, and no less a delight in the beauty of the surroundings he describes — open skies, wide land- scapes, distant horizons, and all around a compelling sense of the past. Who indeed can remain unmoved by this piece of country, where at almost every step one comes upon some relic of antiquity, abiding barrows, encampments, dykes? There is a hint of poetry too within his prose. See for instance (chapter 2, p. 36) his delightful description of the downs escarpment fromTan Hill to Knap Hill. But in par- tcular, he evokes for us in his account a sense of continuity, of innovation, modification, amplification, running without pause across succeeding commu- nities.
Two small sections complete the book and guide the reader towards some of the attractions within and around the Vale, such as representative villages, notable gardens, churches and areas of archaeological and natural history interest. Four selected walks are described in detail, offering the walker some of the finest scenery of the region along with internationally known archaeological sites and monuments.
I have one complaint. The illustrations are almost wholly lacking in sparkle, and have thus lost much of their detail. The fault must lie in their reproduction and not the photography. The Vale and its downland slopes, so marvellously photogenic, surely deserved better treatment. It is much to be hoped that the publishers will improve on their illustrations within the further studies now in preparation.
KEN ANNABLE
Albert W. Cooper. Benjamin Banks 1727-1795. The Salisbury Violin Maker. Ashford Publications, _ Haslemere, 1989, in conjunction with Paul Cave | Publications Ltd., Southampton; 168 pages; colour and black and white photographs and illustrations. Price £25, hardback. ISBN 0 861 460766.
_ ‘The most eminent English maker of stringed instru- _ments’,! Benjamin Banks (1727-1795), was born, _ lived and worked in the city of Salisbury, Wiltshire. In addition to violins, Banks made fine violas, cellos,
1. Recorded on the epitaph restored in 1863 situated in the graveyard for the Church of St Thomas of Canterbury, Salisbury.
167
bows and guitars which are now in great demand, fetching high prices.
Beginning with an account of Banks’ apprentice- ship, Cooper refutes the traditional claim that Banks trained with Joseph Hill (1715-1784) in London, instead providing evidence which suggests that he trained with William Huttoft, music instrument maker of New Sarum.
The earliest recorded Banks instrument is a violin of 1755 and in 1757 his first advertisement appeared. By 1770 the business was thriving, with evidence of an increase in the number of craftsmen employed by the firm, amongst whom were his sons. An indication of Banks’ importance at this time is evident in 1784 when details of a serious fire in his workshop in Catherine Street were reported as far afield as Bristol, London and Winchester. The number of instruments existing after this date suggests a full recovery from the effects of the fire.
Banks died in 1785 and the business was continued by his sons James and Henry who in 1811 moved the firm to Liverpool. The deaths of the brothers in 1820 and 1830 brought about the end of the Banks dynasty but left for posterity a rich legacy of fine instruments.
Presentation of information about Banks is not limited to his career. As an eminent citizen of Salis- bury, he played important social roles and was priv- ileged to meet a number of well-known musicians. Amusing anecdotes and accounts treat the reader to a glimpse of contemporary life and temporary acquaintance with these musicians. Cooper’s interest in placing his subject in the relevant social setting means that this book provides fascinating reading as much for historians as local people, as well as for those with a violinist interest.
The remaining two-thirds of the book is devoted to a more detailed study of Banks’ violins, violas and cellos. Numerous photographs, a few in colour, pro- vide ample illustrations to the text and are a delight to the eye of any violinist. It is unfortunate that the prohibitive cost of colour illustration means that the majority of photographs have been produced in black and white. Although very readable and clearly expressed, this section of the book is written for those with a keen interest in stringed instruments.
This is a fascinating book, beautifully presented and unique in its subject coverage. It is a book to be enjoyed by specialists and non specialists alike and is warmly recommended. If the price of £25, reasonable for a book of this quality, seems too high, readers may
168 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
be interested in Albert Cooper’s abridged version.? Amply illustrated and priced at £1.75 it provides amazing value.
ALISON THIES
D.A. Crowley (editor). A History of Wiltshire, Victoria History of the Counties of England Series, Volume 14: Malmesbury Hundred. Oxford Univer- sity Press for Institute of Historical Research, 1991; xx + 267 pages; illustrated. Price £60, hardback. ISBN 0 19 722779 1.
What more can a reviewer say about the stately progress of our Wiltshire V.C.H.2 And the present reviewer has already asked that same question, twelve years ago now, in a review of Volume 1] (WAM 74-5 (1979-80), p. 209)! Our fourteen volumes exceed in number those published by the V.C.H. for any other individual county, and I am confident that no county history ever produced exceeds ours in quality.
This said, and my position as a V.C.H. admirer thus firmly established, may I be excused for continu- ing this review with two quite personal criticisms? The first concerns the history of the borough status of Malmesbury. Having long been a witness to the nonsensical claims regularly put forward there to ‘borough status’ supposedly granted by King Athel- stan or (even more fantastically) by Alfred, I was hoping that this volume would contain, once for all, a refutation of them in a succinct and categorical sen- tence or paragraph. This would then have superseded R.B. Pugh’s rather involved criticism of ‘Malmesbury and 1980’ in WAM 74-5. The V.C.H. article on Malmesbury presents all the facts, but does not draw them together in the form I hoped for, demonstrating that no historian of Malmesbury mentioned any charter by Alfred until 1969, and none is known to exist now, or ever to have existed; no charter of Athelstan now exists or is ever known to have existed, and the one presented by the borough of Malmesbury for confirmation in 1381 is clearly a forgery made up for the purpose at that time; no Saxon king is known to have granted a charter to any borough. The answer to this is perhaps that it is not the V.C.H.’s business to explode historical myths and, sadly, and no doubt correctly, that no amount of exposure will ever lead to the abandonment of Malmesbury’s claim to be ‘the oldest borough in England’!
My second criticism concerns the settlement his-
2. A. Cooper, Benjamin Banks Violin Maker and Citizen of Salis- bury, Salisbury and South Wiltshire Museum, 1989.
tory of many of the parishes dealt with in this volume. A look at Andrews and Dury’s map suggests immedi- ately that a very important factor in determining the settlement pattern shown on it is the vast amount of cottage settlement encroaching on the waste, yet this is nowhere, that I can see, made clear. Surely Dauntsey and Brinkworth, to name but two, are classic examples of the way in which this process has led to today’s diffuse villages?
K.H. ROGERS
Christopher Gingell. The Marlborough Downs: A Later Bronze Age Landscape and Its Origins. Wilt- shire Archaeological and Natural History Society Monograph I, in co-operation with the Trust for Wessex Archaeology, 1992; xv + 166 pages; 96 figures; 10 plates; 13 tables in text, 40 in microfiche. Price £30, paperback. ISBN 0 947723 04 8.
Air photographs taken before 1950 record a panorama of prehistoric field systems, associated enclosures and linear earthworks spread along the sides of dry valleys in the centre of the Marlborough Downs, North Wiltshire. Field walking and small excavations by the late Owen Meyrick and by Mrs C.M. Guido, now a vice-president of this Society, had already demon- strated the Bronze Age date of this relict landscape, untouched by the plough for some 2,500 years. In the mid-1970s, when modern cultivation presaged the imminent loss of all further information, Christopher Gingell devised and directed a programme of research/salvage excavations, carried out initially under the aegis of the WANHS Archaeology Research Committee and later under that of the Wessex Archaeology Committee, now the Trust for Wessex Archaeology. Most of the funds for the work in the field and for the preparation for publication were provided by the Department of the Environment, now the Historic Buildings and Monuments Commis- sion (England). Four sites were excavated during the years 1976-80, three on the Marlborough Downs (at Dean Bottom, Rockley Down and Burderop Down) and one on Bishops Cannings Down to the south- west, where ploughing had not begun until 1971. This volume, edited by Rosamund Cleal, now brings to a splendid conclusion the contributions made over the years by the numerous specialists and other participants in the project, especially by WANHS members and other volunteers whose commitment made possible the painstaking manual
| re-worked fragment
REVIEWS
excavation methods upon which the success of the whole undertaking depended.
Three of the sites retained traces of pre-Bronze Age activity. A Neolithic ploughsoil sealed beneath the bank of an Early Bronze Age disc barrow (Chisledon G1) adjacent to the Burderop Down site contained sherds of Earlier Neolithic, Peterborough and Beaker pottery. Grooved Ware, worked flints and other debris came from a small pit elsewhere on the site. Residual Beaker sherds were recovered from Bishops Cannings Down as well as from Dean Bottom where a sealed pit contained parts of some 20 Beakers, a bone ‘comb’, other bone and antler artifacts, animal bone and flint waste. All the Beakers from these sites, and all but one of the Beaker sherds collected by Owen Meyrick when fieldwalking on the Marlborough Downs, are attributed to the Middle Style. Small- scale, short-term activity is indicated, antedating the Bronze Age settlements by perhaps 500 years.
Relative dating of the four Bronze Age sites relies on ceramic and metalwork evidence. There were no vertically stratified deposits, and charcoal specimens from Bishops Cannings Down were unfortunately processed during the period of error at the British Museum’s Research Laboratory.
The open settlement on Bishops Cannings Down was crossed by a strip of undisturbed land along a modern fence-line where large potsherds and other remains lay at the base of the topsoil as they had been left when the site was abandoned. Elsewhere artifacts had been lifted but not dispersed. Consequently, the distribution patterns of settlement debris, especially pottery, provided information about working areas, domestic arrangements, and even details of house construction. The intersecting post-circles of two houses show that the first had a deep central post-hole and a south-facing out-turned porch; the larger replacement had lacked a central post and its south- east facing porch was defined by transverse post-slots. (Gingell does not accept Guilbert’s alternative inter- pretations of the plans.) Pairs of post-holes elsewhere on the site may have supported drying racks. David Tomalin’s analysis of the pottery identifies an Early Bronze Age component, suggesting foundation close to the transition to Middle Bronze Age ceramic styles. His wide-ranging discussion of the Deverel-Rimbury assemblage of globular, barrel and bucket urns estab- lishes its place within local and regional contexts. His identification of the probable sources of the Kimmer- idge clay used for the production of barrel urns at this
| site and elsewhere in Wiltshire is a notable contri-
bution. The nine pieces of metalwork include a of a Lisburn-type rapier.
169
Resources did not permit the additional work that might have established a relationship between the settlement and nearby field systems and linear earth- works.
Air photographs have enabled Gingell to recon- struct on paper the lost Bronze Age landscape of the Marlborough Downs and to replace the excavated Dean Bottom and Rockley Down enclosures in their original settings. These and similar ‘Ogbourne’ enclosures, previously interpreted as ‘cattle kraals’, now emerge as farmsteads, each set in its own field system, and attesting an adaptation to local conditions in order to exploit the more easily worked soil on valley slopes in an area where heavy soils cover the level uplands and the valley bottoms. The internally terraced enclosures, some defined by ditch and bank, others partly by scraped-up banks (as at Dean Bottom), were sited near the upper margins of blocks of ‘Celtic’ fields some 50 acres in extent which backed on to open grazing. Some units had associated tracks and/or dew ponds; the Dean Bottom pond was shown to be almost certainly a Bronze Age construction, remodelled in the 19th century. The Dean Bottom enclosure had at least five terraces or house platforms. One platform retained part of a setting of sarsen blocks which is plausibly interpreted as a foundation course for the cob wall of a rectangular building — another adaptation, in this instance to scarcity of timber. Barrel urns constitute the dominant vessel type in the Deverel-Rimbury pottery from this site; bronze finds include fragments of a bracelet which would not antedate the Taunton phase or ‘ornament horizon’ of Middle Bronze Age metalworking.
At Rockley Down, where the enclosure had been surrounded by a deep V-section ditch and internal bank, just enough survived to demonstrate similarity to the internal arrangements at Dean Bottom. A proportion of ‘Post-Deverel-Rimbury’ forms in the pottery suggests a somewhat later foundation date for this settlement.
The open settlement on Burderop Down was situ- ated on the north escarpment of the Downs, beside the ancient course of the Ridgeway, and had been linked to the ‘Ogbourne’ enclosures by a linear earth- work and a trackway which passes through the nearest one. The excavation of 500 square metres revealed three small four-post structures belonging to the latest stage of activity at the site. The largest measured 3.85m by 3.10m; each had a central hearth packed with heat-fractured sarsens. The structures are inter- preted as roofed shelters for cooking hearths associ- ated with communal activities. A remarkable feature of the area away from the structures was a layer of
170 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
heat-fractured sarsen fragments up to 0.15m thick and shown by magnetometer survey to extend for some 70m north and west of the excavation trenches; in this area the ploughsoil contained abundant potsherds. The pieces of sarsen seem to have been used to provide a metalled surface for a working area. Gingell suggests that the settlement may have been engaged in crafts other than farming, or perhaps in trading; its situation on the Ridgeway would have been appropriate for a market place or travellers’ rest. The pottery, with late-looking barrel urns and a considerable proportion of ‘Post-Deverel-Rimbury’ forms, indicates abandonment late in the Bronze Age, as does the metalwork, which includes bronze vase- headed and swan’s-neck pins. Part of a stone mould for a Stogursey (South Welsh) socketed axe was an unstratified find. (A cutting across the ditch of the disc barrow on the west side of the settlement affords a glimpse of the Late Bronze Age attitude to ‘her- itage’: about half of the natural fill of the ditch had been dug out for a purpose not now apparent and, after a short interval, the hollow had been used as a dump for household refuse.)
Philip Harding’s substantial contribution on the Neolithic and Bronze Age flint industries is of par- ticular interest for his reconstruction of the tech- niques that produced the attributes characteristic of Beaker and Bronze Age assemblages. Bronze waste suggests the possibility of metalworking at Bishops Cannings Down and Burderop Down. Those sites also yielded many stone implements: whetstones, rubbers and fragmentary saddle querns. Despite the ubiquity of sarsen in the area some were made of greensand or of rocks of more distant origin (slate, limestone). At Dean Bottom sarsen saddle querns had been manufactured on the site of the abandoned house.
Michael Allen’s environmental report, based on molluscan samples from the disc barrow on Burderop Down and from Dean Bottom, suggests a pattern of small-scale, localized clearances beginning in the Neolithic, followed by fluctuations in the intensity of management and ending as areas of open dry grass- land by the later Bronze Age. Carbonized remains of wheat and barley from the above sites were too fragmentary to provide detailed information. The 40,000 animal bone fragments from the four sites were also poorly preserved. Mark Maltby has, how- ever, been able to demonstrate that cattle outnum- bered sheep/goat at Bishops Cannings Down whereas sheep/goat were dominant at the other three sites. Little pork was eaten and bones of horses and wild animals were few. High proportions of calf bones at
Bishops Cannings Down and Dean Bottom have led Maltby to postulate the culling of most male calves in order to produce surplus milk for human consump- tion and to reduce pressure on pasture.
In his stimulating review of the evidence bearing upon the pastural arrangements Gingell puts forward an alternative hypothesis. Briefly, he suggests that distance from permanent water (3km and more) rendered these and similar sites unsuitable for year- round grazing. Dairy herds would have been wintered at the farms, where the calves were culled after spring calving, and were then moved to summer pastures in river valleys. Such an arrangement would have required the transfer of some labour, while most remained on the upland farms to attend to the arable crops. Evidently the full complexity of the farming economy remains undisclosed.
The painstaking and scholarly treatment of the evidence and the generally high standard of produc- tion (with just the odd mishap here and there) render this volume a major contribution to the archaeology of Wiltshire and to Bronze Age studies in general. It is one for which we have particular reason to be grateful because, in the author’s understated remark, ‘it is likely that any future excavation would be conducted in less favourable conditions of preservation’.
I.F. SMITH
Wessex: A _ Literary Celebration, edited by Desmond Hawkins. Century, in association with the National Trust, 1991; 224 pages; line drawings. Price £15.99, hardback. ISBN 0 7126 4962 X.
This anthology is the second in a series of regional anthologies published in association with the National Trust, East Anglia already having been covered. Unlike East Anglia, Wessex is a much more ambigu- ous term, signifying different things to different people in different contexts. Desmond Hawkins defines his Wessex as comprising Dorset, Wiltshire, Somerset and Avon, an area closer to Hardy’s Wessex than to the Saxon kingdom.
This region is one of great physical diversity, from the chalk downlands of north Wiltshire to the lush river valleys of Somerset and Dorset, and including two very contrasting coastlines. The large conurbations of Bris- tol and Swindon are in sharp contrast to the many rural areas in the region which seem virtually unchanged over the centuries while our earlier ancestors provide constant reminders of their activities in the richest archaeological landscape in Britain.
REVIEWS
The anthology is divided into chapters covering distinct areas within the region such as the Marlbor- ough Downs and North-West Wiltshire, and Bath and Bristol, or subjects such as Rural Life and Smugglers and Poachers. Each chapter begins with an introduction by the editor which sets the scene and draws together the extracts which follow.
Desmond Hawkins is well-qualified to compile such an anthology since he has lived in Wessex for over 45 years and has written many works on the area, being particularly well-known as an expert on Thomas Hardy. It is not surprising, therefore, although entirely appropriate, that a number of extracts from Hardy’s novels, diaries and poems are included, as well as a touching description of the effect of visiting Hardy from a letter by T.E. Law- rence to Robert Graves.
Hardy is not allowed to dominate, but takes his place among the writers, diarists, poets and travellers who provide a varied and stimulating collection, including the well-known and the neglected or obscure, all drawing inspiration from the landscape, people or events in Wessex.
I was fascinated by the extract from The Aeropleustic Art, by George Pocock (1827) in which a carriage drawn by kites competes with the more conventional horse-drawn mode of transport, and more surprised to discover that this was no inspired flight of fancy but a serious, though doomed, proposition! It was pleas- ing to find an extract from E.H. Young’s Fenny Wren in which the evocative description of Upper Rad- stowe, based on Clifton, Bristol, is typical of her well-crafted novels, in recent years finding a new and enthusiastic audience through the imaginative pro- gramme of the publishers, Virago.
Amongst many other enjoyable extracts are the very contrasting impressions of two visitors to the gardens at Stourhead: John Wesley unsurprisingly finds the statues mean, devil-images and indecent, while John Claudius Loudon greatly admires the combination of formality
| with fine natural features. Celia Fiennes, writing in 1687, provides a very detailed account of the hot baths at Bath. while R.D. Blackmoor (1869) contributes a spine- chilling tale of Exmoor in a winter fog.
_ It was disappointing to find three extracts from Llewelyn Powys but none from his brother John Cowper Powys, who wrote some of the most sensual | and memorable descriptions of the Wessex landscape in his epic novels such as Wolf Solent, A Glastonbury | Romance and Maiden Castle. Like Hardy, he was a writer intensely influenced by landscape, so that in | their novels it seems to dominate the characters and their actions.
171
Desmond Hawkins, however, admits that his choice of extracts is a very personal one, and there is much to interest and inspire the reader here. I felt it was a pity that the publishers had decided to list all the sources together at the end of the volume, so that although the author and date appear after each extract, to discover whether you are reading a novel, diary or letter, for example, you need to keep turning to the back. The index too is poor, including only people and places, not subjects, while a fairly useless map shows National Trust properties rather than places mentioned in the text. The line drawings by Philip Hood are very attractive and appropriate and my quibbles are only minor ones. I would strongly recommend this highly readable and enjoyable anthology which provides within its covers an excellent picture of the landscape and people of Wessex.
FELICITY GILMOUR
Lorna Haycock. John Anstie of Devizes 1743-1830. An Eighteenth-Century Wiltshire Clothier. Wilt- shire Archaeological and Natural History Society and Alan Sutton, 1991; 116 pages; illustrations. Price £9.50, paperback. ISBN 0 7509 0045 8.
Pamela Colman. The Baker’s Diary. Life in Geor- gian England from the Book of George Sloper, a Wiltshire Baker, 1753-1810. Wiltshire Archaeolo- gical and Natural History Society and Wiltshire County Council Library and Museum Service, 1991; 96 pages, illustrations; Price £5.95, paperback. ISBN 0 86080 221 3.
The last twelve months have seen the publication of the biographies of two citizens of Devizes, John Anstie and George Sloper. They were largely contem- porary, Anstie living from 1743 until 1830 and Sloper from 1730 till 1821. In their heyday they were both affluent and prominent townspeople, involved in the affairs of the borough, although Anstie, being a non-conformist, was not eligible to become a member of the Corporation. They were no doubt acquainted, being members of the philanthropic Bear Club, which met at the Bear Inn in the Market Place. At this time, the people of Devizes had the ear of the highest in the land as one of their two members of Parliament was Henry Addington, first Viscount Sidmouth, who was Speaker of the House of Commons and, later, Prime Minister and Home Secretary. Anstie and Sloper are the subjects of biographies by the two librarians of the Wiltshire Archaeological and Natural History
Nye THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Society. The two authors must have had much valu- able material at their fingertips, although they by no means confined their researches to the Devizes library of the Society as the impressive list of references at the end of each chapter in both books shows.
Although the name of Anstie is usually associated in Devizes with snuff, John was a clothier, a man who organised the various processes that led from raw wool to the finished piece of cloth. In his time, machines were being introduced that would speed up some of these processes and it meant that much of what had been done at home was now better done in factories. So Anstie built a new factory that still stands in New Park Street. However, the introduction of these machines led to fears of large-scale redundancies and there were riots in neighbouring towns. Anstie took a leading part in calming these fears. The next dispute he was involved in was on a national level, over the export of wool. It came as a surprise to this reviewer to learn that the export of wool was at that time forbidden. As prices were then three or four umes higher in France, there was a large amount of wool smuggled abroad. The wool producers were pressing for a relaxation of this ban but the cloth manufacturers were bitterly opposed to any change. Anstie led the manufacturers and eventually put forward a bill that was enacted in Parliament, tighten- ing the regulations against smuggling. His active role in this controversy led to his being consulted when Pitt wished to lower trade barriers with Ireland and France and he seemed to have been much in favour of facilitating the export of finished cloth. Mrs Haycock gives a sympathetic picture of Anstie but she has to admit that she finds his economic ideas ‘ambivalent’. He seems to have lived in some state in his house in New Park Street and played a very active part in the life of the town. However, in 1793 disaster struck when, as a result of heavy borrowing and bad debts, he was declared bankrupt. The rest of his long life was spent in straitened circumstances and increasing obscurity, much detail of his later years having been lost.
The facts of the wool export affair have to be inferred from Anstie’s part in it and I felt that I should have liked to be told more succinctly exactly what happened. For instance, when did the embargo on wool exports end, or how exactly did Anstie, a private citizen manage to get his bill through Parliament? But generally much of what is said strikes echoes in a contemporary mind: the fear of redundancies with the advent of more automation, bad debts, questions of free trade or protectionism, heavy borrowing and the subsequent payment of interest — with a little up- dating of the vocabulary Anstie’s problems could be
those of many a businessman of today. Even the introduction of his bill to Parliament revives the question of exactly who can initiate legislation in the European Parliament today.
George Sloper lived on quite a different plane. As a master baker in Devizes he kept an account book showing his daily takings for bread for fifty years, inserting in the margin notes concerning interna- tional, municipal and family affairs. He notes the first sailing with convicts to ‘Bottney Bay’ and the death of Captain Cook in ‘Owhyhee’ (Hawaii) — Sloper’s spell- ing is erratic. He mentions the descent on Devizes of 40 or more smugglers who retook a large quantity of tea from the Supervisor, and the death of his sister, Ruth, with a list of funeral expenses. Naturally the main information is on the price of bread — when the price of wheat was high, occasioning distress in the populace, Sloper made his highest profit, although it must be said that the price of a loaf was regulated by the Corporation. Sloper served the town he lived in for the most part of his long life, being elected to the Common Council in 1777. He filled many offices, including ‘Chamberlain’ (Treasurer) for two periods of four years, and was three times Mayor. He never married and when he died he was affluent (his estate was £10,000); he was well known, respected and on terms with the major family in the district, the Suttons of New Park (which gives Mrs Colman a reason to include as an appendix the story of the trial of Willey Sutton, mentioned by Sloper in his book.)
While it would be wrong to say that the book on John Anstie gives no picture of Devizes (chapter three is devoted to his life as Philanthropist, Citizen and Gentleman) the most memorable part describes his involvement with national affairs. By contrast, while the book on George Sloper gives insights into national concerns, the most important part is the picture of borough affairs — the elections of Mayors or MPs, the ‘feasts’, the magistrates courts and the subsequent sentences, and the various improvements made in the town.
Both books are well presented with numerous illustrations. There are interesting appendices includ- ing biographical summaries of persons mentioned: in the case of Anstie figures of national importance, in that of Sloper solid citizens of Devizes. This shows how these two books complement each other to give a full picture of the lives and concerns of the bour- geoisie of the town in the latter part of the eighteenth century. Devizes is lucky to have had such a picture painted and I hope it may inspire similar books from other towns in Wiltshire.
NORMAN ROGERS
REVIEWS
Phyllis Hembry (editor). Calendar of Bradford-On- Avon Settlement Examinations and Removal Orders 1725-98. Wiltshire Record Society, Vol. 45, 1990; xxii + 164 pages. Price £15, hardback. ISBN 0 901333 23 9.
Poor Law settlement examinations have long been considered by archivists, historians and genealogists as a premier source for social and family history in the 18th and early 19th centuries. It is entirely appro- priate that a particularly fine series from Bradford-on- Avon, preserved in the Wiltshire Record Office, should appear in print and thus be more accessible.
The documents were created as part of the pro- cedure to determine the rightful place of settlement of paupers and their families and, consequently, the responsibility for their maintenance. It was a system of social welfare in which, there being no central provider, individual parishes sought, not unreason- ably, to look after only their own poor. The system’s harshness and insensitivity were to earn the word parochial its modern meaning of being of narrow outlook. It did however generate a wealth of material making the labouring poor of this period one of the best documented of all English social groups.
The volume comprises abstracts of 414 settlement examinations with accompanying removal orders, a separate series of 141 removal orders and 6 related documents covering the years 1725-95. Dr Hembry is correct in stating that the documents provide vign- ettes of the labouring poor rather than that they are true autobiographies as claimed by Tate in The Parish Chest, for since they were restricted to the facts necessary to establish settlement much is omitted and, notably the underlying harshness, is left to the imagination. Nevertheless they provide a rich source for the individual and collective experience of the
labouring poor which in fact extends far beyond the | date of the earliest document since several examinants | in the 1720s and 1730s had already reached old age and recalled events from the last quarter of the | previous century.
| The scope of the documents is further extended due | to the fact that few examinants spent all their lives in Bradford. Mobility, particularly within a forty mile radius of the town, was considerable and a wealth of Personal details from the villages, towns and cities of | that area and beyond are provided. Such details relate | Mainly to birth, marriage, employment and residence but these are enlivened by the unusual, such as references to illegal marriages in Bath, cases of small- | pox and outlines of military careers on the battlefields of Europe.
173
The introduction and indexes more than adequately support the text. The occupational groups of the examinants, their degree of mobility together with the business of establishing settlement carried out by local justices, several of whom were wealthy clothiers, are well analysed. It is interesting to note the amount of movement by workers among the local centres of the cloth industry, notably Bradford and Frome, and the fact that many interrupted their trade to serve in the army. The separate indexes to place, person and occupation are comprehensive. The editorial method is also sound. Sensibly, all verbiage and negative statements are omitted. Nevertheless a clear impres- sion of the original documents and the procedure they recorded is provided.
The euphemistic description of the condition of a vagabond as lodging in the open air in a removal order issued in Westminster is an evocative example (392). The flow of text is unnecessarily interrupted by the inclusion of the original form of place names, justi- fied, perhaps, where a variant form occurs but not a variant spelling which can have little toponymic signi- ficance.
A few inconsistencies and inaccuracies occur in the identification of places which detailed research might have avoided. It is not clear why Sutton identified as S. Benger in (159) appears also as S. Mandeville or S. Veny in (261-2). Church Eaton in (220) and (93), where it appears incorrectly as Church Easton, is Yatton Keynell as a later examination of the same person confirms (256). Christopher Ingram appears in the land tax assessments for Teffont Magna between 1780-84 indicating that this is the place referred to in (374) rather than T. Evias.
These are only minor quibbles which should not detract substantially from the merits of this volume which is a notable addition to the publications of the Wiltshire Record Society. Dr Hembry will earn the praise of many researchers for editing so competently these ‘short and simple annals of the poor’.
STEVEN HOBBS
Pamela M. Slocombe. Wiltshire Farm Buildings, 1500-1900. Devizes Books Press, 1989; 80 pages; 113 photographs; 10 figures. Price £5, paperback. ISBN 0 9509099 6 3.
Wiltshire Farm Buildings, 1500-1900 is the second monograph to be published by the Wiltshire Buildings Record and is based on recordings it has made since 1979, work which is still in progress. The
174 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
book is lavishly illustrated with clear, well-chosen photographs and plans; a good proportion of the photographs are, usefully, of interiors or details. The conclusions are based on documentary and written sources as well as on the buildings themselves. There is a short bibliography, dealing with both general works and those particular to the county.
The book begins with an introduction on farming history and methods in the county, and its farm buildings. Descriptions of farmsteads dating from 1548 to 1898 follow, listing types and often sizes of buildings, and occasionally the materials. A short section on the farmstead as a whole is illustrated by photographs and plans of actual examples: the latter helpfully show the boundaries of yards and adjoining fields, features which are necessary to help interpret the use of the buildings, but sometimes omitted from published plans.
Descriptions of different types of buildings follow, beginning with the barn. The various plan types are clearly illustrated by photographs. The use of porches may have encouraged the provision of lean-to struc- tures against the barn, so saving in building costs. ‘L’ plan barns seem peculiar to the county, but those built on staddle stones spread rather further. The effect on the appearance of the barn of different building materials is well illustrated. Bracing in roofs continued much later in barns than in houses, as the former lacked the additional rigidity provided by upper floors. There is a clear photograph of tally marks cut by farmworkers in the wall of the barn by the threshing floor, a difficult item to photograph.
The reviewer is doubtful if reference in documents to ‘barns and stables’ or ‘barns and stalls’ means that the livestock was actually housed within the barn; it is more likely that the cow house or stable was attached to the end of the barn, or partly under it, but fully partitioned off from the crops, a feature illustrated in a different context in this book. Barns with a thresh- ing floor at one end may well have been for un- threshed corn before the 19th century, not just threshed straw, but these are minor quibbles.
Because of the importance of the horse in the farm economy some stables were more carefully, archi- tecturally designed; this point applies even more so to those provided for riding or carriage horses, rather than just working horses for the farm. Three illustra- tions show stables dating from the 17th century, one with a cowhouse attached. To find so early a cow- house surviving is unusual. Some of the early, open- fronted buildings for cattle which are illustrated may have been for tying cattle rather than providing shelter for loose, yard cattle as is suggested. However,
without surviving fittings or documentary evidence of their use it is not always possible to tell for which purpose the building was intended.
Dovecotes tended to be concentrated in the north- west of the county; provision was not necessarily in separate buildings, but might be nests in the outside wall of a barn or other building, or a small box hung on the wall. Even some small cottages in the early 19th century had a few nests in the gable! Unlike the dovecote, many granaries were timber framed: quite a few were separate buildings, a point linking Wiltshire to Hampshire and other southern counties; a few were two-storey. Less frequently than further north the granary was set above other farm buildings. Cartsheds and pigsties are discussed, with interiors of the latter (the photographer must have had a very good wide- angle lens!).
Dairies, cheese rooms and brew houses are dealt with briefly, related to detached examples; this is reasonable as they generally formed part of the farmhouse and as such have been well illustrated in the earlier, companion volume on that building.
This is a careful study of Wiltshire farm buildings, thorough in its coverage, both of building types and their details. It is essential reading for anyone studying farm buildings in that county, and a valuable work for those interested in other parts of the country.
J.E.C. PETERS
Christopher Stell. An Inventory of Nonconformist Chapels and Meeting-houses in South-West England. Royal Commission on the Historical Monu- ments of England, HMSO, 1991; xviii + 263 pages; 488 plans and illustrations. Price £60, hardback. ISBN 0 11 300036 7.
More than anyone, Christopher Stell has been responsible for stimulating interest in nonconformist architecture, and for alerting us to the dangers which threaten this poor relation of our ecclesiastical her- itage. During a distinguished career with RCHME he completed a survey (which he had begun privately) of over 4,000 chapels and meeting-houses in the whole of England. The archive resulting from this survey is deposited in the National Monuments Record, and derived from it Mr Stell edited for publication in 1986 an inventory of thirteen pre-1974 counties in central England. This handsome book was highly praised by | reviewers, and their encouragement, we are told, persuaded the Commissioners to publish the present volume. One hopes that no minor criticism let slip
REVIEWS
here will stifle their enthusiasm for extending this admirable series.
The present inventory describes 1,094 extant or recently extant buildings in seven historic counties (with the Isles of Scilly and Wight), from Berkshire and Hampshire westwards. Cornwall has the most (232); 164 Wiltshire chapels and meeting-houses are included. The inventory aims to include all significant surviving buildings of Protestant nonconformity earl- ier than 1850, and selected buildings up to 1914. Chapels demolished or altered since Stell carried out his surveys (he seems to have been in Wiltshire mostly between about 1969 and 1974) are described in the present tense, with a bracketed note about their subsequent fate. The text is laid out in the usual RCHME inventory format, and entries include thumbnail histories, with architectural descriptions and notes on monuments and fittings. Many buildings are illustrated with photographs, measured plans and elevations, and/or sketches by the author.
The bulk of the volume is taken up with descrip- tions and illustrations of individual buildings. Apart from a page of preface, and another of editorial notes, there is no general introduction, although a brief summary and synthesis head the inventory for each county. The preface to the central England inventory has promised us future general volumes in the series devoted to the growth of nonconformist denomina- tions, their architecture, fittings and plate. A useful and detailed index completes the present book — but note the curious treatment of surnames, all grouped together as sub-headings and without initials, column after column, under the heading ‘Surnames’.
Revising work for publication some twenty years after it was first carried out must have been a difficult
task, and often depressing, since many buildings have
|
|
been ruthlessly altered and some demolished. The worst Wiltshire casualty, in Stell’s opinion, has been
_ Manvers Street Wesleyan in Trowbridge — he de-
scribes its demolition as ‘inexcusable’. A few altera- tions known to this reviewer have escaped Stell’s attention, for example the conversion of Cricklade Baptist chapel into a museum; the internal alteration
| of Fisherton Street URC, Salisbury; and the conver-
sion of chapels at Woodborough and Corsley (Whit- bourne) into private houses. These are mere oversights, and there are also a few
| slips of the pen in the bibliographies — E.J. Cruse, , A.R. Stedman and T.R. Thomson are all spelled
wrongly. But a slightly more serious criticism con-
| cerns Stell’s use of published sources, especially the
Victoria History. Since the 1960s five volumes have
| been published, and in several cases, such as Devizes
175
in Volume 10 and Marlborough in Volume 12, he appears not to have used them, although they would have helped to improve his historical accounts. (Else- where, for example at Downton, the relevant volume (11) has been cited.) Occasionally, too, the present reviewer’s edition of meeting-house certificates (which is cited once) might have cleared up small points, for example at Bishopstone (south Wiltshire), where a Bishopstone (north Wiltshire) certificate is wrongly cited. It is curious too that no use appears to have been made of a much older book, Victory in the Villages, by W.C. Tonks (1907). This gives much detail about Primitive Methodist chapels in north Wiltshire, and would surely have proved useful. As it is, Brinkworth chapel is described with no mention of its key role in the spread of this denomination.
Nevertheless this book is a wonderful achievement, and Stell’s scholarship, enthusiasm, and dogged per- sistence have given us in Wiltshire a valuable new source of information and illustration. Nine Wiltshire chapels fall into the category of buildings deemed ‘most worthy of preservation’, viz. Avebury (URC), Bratton (Baptist), East Tytherton (Moravian), Derry Hill (Little Zoar Baptist), Corsham (Monks URC), Grittleton (former Baptist), Horningsham (Congrega- tional), Bradenstoke (Providence Baptist) and Tis- bury (URC/Methodist). But it is to be hoped that all the others in the county so lovingly described by Stell will be cared for and cherished. Now that this book is available we really have no excuse for neglecting them.
JOHN CHANDLER
Julian Thomas. Rethinking the Neolithic, New Studies in Archaeology. Cambridge University Press, 1991; xv + 212 pages; 60 illustrations. Price £35, hardback. ISBN 0 521 40377 4.
This work synthesises recent thinking and data on the Neolithic period in Britain, with particular reference to a region which could be called ‘greater Wessex’: Avon, Dorset, Gloucestershire, Oxfordshire, Somer- set and Wiltshire.
The strengths of the book lie in its capacity for synthesis and integration. The subjective nature of the study of remote and alien societies and the level of assumption entailed in the most familiar ideas about the period are spelt out to salutary effect. Faunal assemblages are related firmly and healthily to their contexts. Monuments are seen as components and modifiers of landscapes and as expressions of social
176
power. The liminality of causewayed enclosures is persuasively argued from both factual and theoretical considerations.
The character of pit deposits, the role of pottery in earlier Neolithic life and the distinct, even comple- mentary, uses of Peterborough Ware and Grooved Ware are worked into a persuasive, cohesive whole. Often-aired oppositions between wild and tame, traditional and novel, inward- and outward-looking, here mesh with behaviour and artefacts.
Analysis of Beaker burials affirms the validity of the main stylistic divisions within the tradition in terms of funerary use, distribution and associations, despite the recently demonstrated lack of correlation between radiocarbon determinations and Beaker typology.
A compare-and-contrast exercise emphasises the distinctive, perhaps unstructured and decentralised character, of Neolithic society in the Upper Thames Valley, the unity and co-ordination of the Avebury complex, and the distinct, perhaps conflicting, func- tions of the Salisbury Plain monuments.
Interpretation of spatial and temporal variation in the treatment of the dead betrays a naively confident reliance on ethnographic analogy. Most of the book’s other weaknesses stem from the time lag between conception and publication, in the course of which fresh ideas and information have begun to go stale. Thus, ‘The present understanding of the Neolithic in Britain rests upon its identification as a primarily economic phenomenon. Contesting this central assumption is the starting point for this book’ gives the impression that the author is setting up an already defunct Aunt Sally in order to reflatten her. One wonders too whether the Middle Neolithic agricul- tural recession proposed in the late 1970s has ‘slipped into orthodoxy’ (page 26), rather than into the history of prehistory. There is a very familiar ring to the conclusion of chapter 2: ‘a seasonal cycle of movement lay at the heart of both social and agricultural arrange- ments in the earlier Neolithic, with cultivated plants contributing rather little to diet . . . a real change in the fabric of the farming lifestyle would not take place until the mid to late second millennium. Only at that point did field systems and permanent domestic struc- tures become the norm in the British Isles’. This is the picture already painted by the results of the Stonehenge Environs Project, the Marlborough Downs Project and by recent work in the Dorchester area, including the South Dorset Ridgeway Project. The publication of so much new material from the region ensures that synthesis will remain fluid.
Detail too is sometimes out of date. The Hamble- don Hill flint mines, cited on pages 34 and 163, are no
THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
longer thought to be such, while Care’s model of the processing of raw materials brought from elsewhere at the Maiden Castle causewayed enclosure (also page 34) has been modified by Edmonds’ and Bellamy’s analysis of the lithic assemblage from recent excava- uons. The distribution of Neolithic Bowl pottery decorated in the Mildenhall style (Fig. 5.5) could have been shown extending north of the Wash as far as the Humber, on the evidence of finds from Dragonby, Wigber Low and Tattershall Thorpe.
It is a pity that the illustrations do not share the quality and originality of the text. The eye is distrac- ted from plans by over-heavy scales. Increasingly versatile and powerful computer graphics can convey information more informatively, attractively and diversely than can bar diagram after bar diagram. If photographs are reproduced as badly as some are here there is no point in publishing them.
Despite these trivial weaknesses the work is well worth reading and re-reading. It consistently stimu- lates, not least by provoking the reader to re-examine long-held ideas. Julian Thomas has created a colourful, life-filled tapestry which makes conven- tional works of synthesis look like unoriginal if worthy patchwork.
FRANCES HEALY
David Tylden-Wright. John Aubrey: A Life. Harper Collins, London, 1991; 270 + xviii pages; illustrated. Price £20, hardback. ISBN 0 00 215097 2.
This book is a tribute to the affection that Aubrey continues to inspire. Relatively few 17th-century per- sonalities apart from monarchs form the subject of more than one biography within a generation, and in Aubrey’s case, it might have been felt that Anthony Powell had already provided a virtually definitive life in his John Aubrey and his Friends, first published in 1948 and relatively recently reissued. Those who know Powell’s book will feel a definite sense of familiarity as they read Mr Tylden-Wright’s, which gives a Kaleidoscopic view of the people and places
Aubrey encountered during his life, on the basis of |
extensive use of Aubrey’s manuscripts, as well as a selection of secondary sources. The book will no doubt find its audience among readers of general biographies, perhaps particularly in Wiltshire, since Mr Tylden-Wright writes with enthusiasm about Aubrey’s native county: but it really has little to add to
Powell’s book, and in some ways one might question why | it seemed necessary to go over the same ground again.
|
REVIEWS
In one crucial respect Mr Tylden-Wright’s book is inferior to Powell’s, and this is in failing to give any sense of Aubrey’s originality as an author and thinker. It was in fact for this that Aubrey wanted to be remembered, rather than as a personality and raconteur whose tumblings up and down in life seem worthy of record. Powell showed his awareness of this side of Aubrey’s career to the extent of devoting a lengthy appendix to an account of Aubrey’s writings. Mr Tylden-Wright, on the other hand, has hardly anything to say about them, apart from Aubrey’s biographical and topographical collections, which fit well enough into the miscellaneous theme of his book. Mr Tylden-Wright would no doubt say that such matters are for scholars (who can consult my own book, Fohn Aubrey and the Realm of Learning (1975), to which he respectfully refers, though virtually ignoring its findings). But I regard this truncation of a major part of Aubrey’s career as an injustice to him, having a rather unfortunate trivialising effect which can only harm his reputation. For instance, Mr Tylden-Wright’s evaluation of Aubrey’s highly innov- ative Monumenta Britannica is quite inadequate: all he can see in this work — which blazed a new trail in antiquarian studies through its comparative studies of field antiquities — is ‘a guidebook’, and Aubrey’s other books are similarly depicted as ragbags of informa- tion, his aim in deploying which is almost entirely overlooked. This is a pity: a book of 270 pages could easily have found room for Aubrey’s ideas as well as his life, and an opportunity has been missed to give a sense of the man as a whole.
MICHAEL HUNTER
The Wroughton History Group. Wroughton His- tory, Part 5: Between the Wars. Wroughton History Group, 1991; 112 pages; photographs. Price £7, paperback. ISBN 0 9508393 4 5.
It has been interesting to watch the development of _ the Wroughton History from 1982 when Part 1 was printed from unjustified typescript. Part 2 was profes- | sionally typeset, with well-screened photographs, and _ production standards have continued to improve with _ every volume. The contents have also evolved from a | variety of themes represented in Parts 1 and 2; a structural pictorial record of Victorian and Edwardian times in 3; a collection of studies, mainly short | articles, in 4; to a comprehensive picture of a growing
177
village in the 1920s and ’30s presented in the latest volume.
Five volumes published in 10 years is a consider- able achievement for a group numbering around 20 members. Apart from the local history research, this reviewer was impressed by the obvious success of the promotion and marketing of the title. Part 5 has sponsorship from 24 bodies and has nearly 500 sub- scribers: an excellent testimony to the popularity of Wroughton History in north-east Wiltshire. A solid financial base is far more difficult to obtain than camera ready copy, and local societies wishing to publish their material could do far worse than to follow the Wroughton example.
Recent Wroughton life has been much involved with the training of racehorses and the greatest of these was Brown Jack, who has a public house named in his honour. He won the Queen Alexandra Stakes a record six times, the last at the advanced, for a flat racehorse, age of 10. He was in training at the Barcelona stables between 1927 and 1935 and it is pleasant to think that many a Wroughtonian must have won a few shillings through his successes. His story is told in the latest volume while racing estab- lishments feature in earlier parts.
There is much useful information that has come from the memories of older Wroughtonians and the Group is performing a valuable task by not only recording but publishing these recollections that would otherwise be lost to future generations. The story of an early radio in Wroughton and Grampy Huff's total disbelief that a band playing in London could be heard 84 miles away must have been repeated, with variations, in most villages. Unfor- tunately, interesting snippets of social history such as these are so easily lost.
A substantial archive of photographs has been built up by the Group and used to good effect in all five volumes. A pleasing feature of the latest work is pictures of some of the men and women who joined up, or were called up, for World War II. All articles have good relevant illustrations. Some show scenes seldom recorded by the camera, such as unemployed men working on local schemes and schoolchildren involved in gardening classes.
Wroughton must now be the Wiltshire village best recorded by local effort and I imagine that there are still a few volumes to come that will ensure that this enviable situation is maintained.
MICHAEL MARSHMAN
a
} |
| |
Index
NOTE: Wiltshire places are indexed under civil parishes.
Abbas Combe, Somerset, 151
Abercromby, Lord, 1, 6
Addington, Henry (Viscount Sidmouth), 171
Addyman, Peter, 111
Agassiz, J.L.R., 4
agriculture, 166, 174; Neolithic, 176; Bronze Age, 170; medieval, on Bradford manors, 123-5
Akerman, John Yonge, 5
Aldbourne, 147; Snap, 86
Alderbury, 158
Aldhelm, St, 120
Alfred, the Great, 59, 168
Alfward Colling, 76
Alice Holt, Hants, 31
All Cannings: Cross, 161
Allen, G.W.G., 142; Michael, 170; report on Avon Valley molluscan evidence, 45—50
Alton: Adam’s Grave, 165
‘Alvestone’, note on, 151-2
Alward, 116-17
Amesbury, 48; Boscombe Down West, 26, 32, 34, 51; Butterfield Down, 32, 48, 51, 52, 158; see also Stonehenge
animal bone: see bone
Annable, Ken, review by, 165-7
Anstie, John, 171-2
ants, 153
Antwerp, 130
army camp, Great War, 80, 81
Arras, Humberside, 5
Arundell of Wardour, Lord, 138
Asheldham, Essex, 111
Ashton Keynes, 66, 71
astronomy, 2
Athelstan, King, 59, 61, 168
Atworth, 118, 123-5, 129; Chalfield, 118; Little Atworth (Cottles), 118
Aubrey, John, 103, 128, 176-7
Austen, John, 133
Avebury, 142-7, 158-9, 175, 176; Beckhampton, note on barrow cemetery, 142-7; North Down, 143; Silbury Hill, 142; West Kennet Long Barrow, 3, 5
Avebury, Lord (John Lubbock), 4-5
Avon: see Bath, Brislington, Bristol, Ham Green, Kelston
Avon, River (Bristol Avon), 59, 60, 120, 122-3, 159; (Salisbury Avon), 162
Avon Valley, excavation of enclosures in, 8—57
Baker, George, 138
Banks, Benjamin, 167-8; Henry, 167; James, 167
Barford St Martin: Hamshill Ditches, 51
barrows, 5, 158; long, 5, 159; round, 5, 8, 12, 38, 142-6, 169; see also ring ditches
Bateman, Thomas, 3, 5, 6
Bath, 84, 95, 130, 139, 171, 173; hundred, 151
Beaulieu River, Hants, 138
bees, 153, 154
beetles, 153
Berkshire, 130, 131, see also Hungerford, Newbury
Berwick St John, 137
Bezyle, William, 118
Bingham, Peregrine, 137
Bishop, Sir Henry, 138
Bishops Cannings, 159: Down, 168-70; Easton Down, 159; Morgans Hill, 143
Bishopstone (north Wiltshire), 175; Little Hinton, 111
Bishopstone (south Wiltshire), 175
Blackmoor, R.D., 171
Blanchard, Joan, work reviewed, 165
Blumenbach, J.J., 2, 4, 6
Blunsdon St Andrew: Groundwell Farm, 26
bone: animal, 19, 21-2, 38, 40, 54, 60, 63, 65, 72-4, 82, 100, 159, 161, 170; dog, 23, 38, 40; frog, 38; human, 41-2; worked, 35, 40, 160
Bower Chalke, 32
Bowles, Charles, 137; John, 137; Magdalen, 138, 140; William Lisle, paper on, 136-41; family, 137
Bowman, William, 2
Bowood, 136, 139
Box, 110; Hazelbury Common, 152-7
Bradford-on-Avon, 111, 151-2, 173; 14th-century, paper on, 118-29; Barton Farm, 120-3; Boxleigh, 125; hundred, 118; Leigh, 118, 120, 127; Trowle, 118, 120, 123; Woolley, 118-19
Bradon Forest, 74
Brassica, 153
Bratton, 95, 175
Bremhill, 111, 136, 138—40
Brinkworth, 71, 168, 175
Brislington, Avon, 96
Bristol, 70, 99, 130, 131, 167, 170-1
Britford: Great Woodbury, 159; Odstock Golf Course, 159
Brito, Gualo, 116-17
Broad Chalke: Middleton Down, 152-7
Bromham: Sloperton Cottage, 136, 139
bronze artefacts, prehistoric, 5-6
Buckler, J.C., 106, 108, 110
Bulford, 8-10
Burgess, Thomas (Bishop of Salisbury), 140
Burghal Hidage, 59
burials: see inhumations
Burlton, William, 137
Burrowe, George, 134
butterflies, 154-6
Butterworth, C.A., 158
Byron, Lord, 136
Caen, France, 105, 110
Calne Without: Bowood, 136, 139; Derry Hill, 175
Cannings: see All Cannings, Bishops Cannings
Carpenter, John, 127
cartage, from Bradford manors, 125-6
Castle Combe, 159
castles, medieval, 159, 160-1
Cecilia (Abbess of Shaftesbury), 118
ceramics: see pottery, tile
Chalmers, Alexander, 138
Chalton, Hants, 52
Chandler, Chris, 163; John, review by, 174-5; work re- viewed, 165-7
chapels, nonconformist, 163-4, 174-5
180 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Cheddar, Somerset, 95
Cherhill, 105; Yatesbury, 105
Chicklade, 138
Childe, Gordon, 1
Chippenham, 95; College, 159, 163; Milbourne Farm, 159-60
Chippenham Without: East Tytherton, 175
Chiseldon: Burderop Down, 168-70
Christchurch, Dorset, 110
church, survey, 105—12
churchyards, 103-5, 160
Cirencester, Glos, 41, 71
Clarke, David, 6
clay pipe, 71-2, 98-100, 160
clay slabs, Romano-British, 34-5
Cleal, Rosamund, 168
cloth trade, Bradford, 126, 173
Coe, Duncan, report on Knook, 75-87
coins, Iron Age, 147~9; Romano-British, 8, 19, 34-5, 72, 163; Saxon, 150-1; medieval, 63, 72, 117; post-medieval, 99
Colburn, Henry, 140
Colchester, Essex, 25, 40, 130
Coleman, Walter, 120
Coleridge, Samuel T., 136, 138
Colling, Alfward, 76
Collingbourne Ducis, 160
Colman, Pamela, work reviewed, 171-2
Combe, George, 2
Compton Bassett, 145, 160; Research Project, 102, 160; survey of church, 102-12
Coombe Bissett, 163
Cooper, Albert W., work reviewed, 167-8
corn-driers: see kilns
Corp, Agnes, 123
Corsham, 175; Milbourne Farm, 159-60
Corsley, 175
Costello, Louisa Stuart, 139
Cotswolds, 5
Cowdery’s Down, Hants, 52
Coxe, William, 138
Crabbe, George, 136, 139
crania, craniometry, 1-6
Crew, Nathaniel (Bishop of Durham), 137
Cricklade, 59, 175
Crocker, Philip, 143
Crondall, Hants, 150-1
Crowley, D.A., work reviewed, 168
Cunetio, 147-9
Cunnington, William, 3
Curr, Thomas, 134
Currie, Christopher K., report on Malmesbury excavations, 58-74; Trudie C., report on animal bone from Malmes- bury, 72-4
Cuxham, Oxon, 124
Darwin, Charles, 5
Dauntsey, 168
Davis, Joseph Barnard, 2-4, 6
Derbyshire: see Middleton Hall
Devizes, 2, 3, 171-2, 175; Museum, 5, 94, 95, 145, 147, 150; report on excavations at New Park Street, 88-101
Diptera, 153, 155
dog-violet, common, 154
Domesday Book, 151-2
Donhead St Andrew, 137, 138
Donhead St Mary, 125, 126, 138
Dorchester, Dorset, 50, 176
Dorchester, Oxon, 150
Dorset, 151, 170; see also Christchurch, Dorchester, For- dington Bottom, Hambledon Hill, Maiden Castle, Poundbury, Shaftesbury, Sherborne, Verwood
dovecots, 162, 174
Dowdeswell, Wilfrid, note on invertebrate populations, 152-7
Downton, 160-1, 165, 175
Dumbleton, Glos, 138
Dun, River, 131, 132
Dunstan (Archbishop of Canterbury), 59, 61
Durrington, 29, 32; excavation of enclosure near, 8-10, 52-5; Walls, 48, 50-2, 161; Woodhenge, 48, 50, 145
East Knoyle, 138
Ede, Joy, report on Avon Valley seed remains, 42—5
Edes, Henry, 133-4
Edgeworth, Maria, 139
Edinburgh, 1, 3, 4
Edwards, Kathleen, 115
Egerton, J., report on Avon Valley animal bone, 38, 40
Elizabeth I, 130, 132, 135
Ellesmere, Lord, 4
Emma (Abbess of Shaftesbury), 118
enclosures: Neolithic causewayed, 162, 176; Bronze Age, 168-9; Late Bronze Age-Iron Age, 161; Iron Age-— Romano-British, 8-57, 158-60, 163; see also hillforts
Enford: Chisenbury Trendle, 161; Chisenbury Warren, 51; Compton, 162; East Chisenbury, 161-2; Lidbury Camp, 161-2; West Chisenbury, 162
Essex: see Asheldham, Colchester, West Bergholt, Witham
Ethelred II, 118, 151
Evans, John, 5
Everleigh, 143
Eyr, John le, 120, 123
farm buildings, 173-4
farming: see agriculture
Field, David, 161-2
field systems, 8, 162—3, 166, 168-9
Fiennes, Celia, 171
Figheldean, excavation of enclosure, 8-52
fire damage, Hungerford, 130-5
fishpond, 162
Fitzgerald, C., report on Avon Valley animal bone, 38, 40 Flaxman, John, 141
flies, 153-4
flintwork, prehistoric, 5, 12, 36-9, 50, 54, 82, 159-62, 170 folklore, 166
Fordington Bottom, Dorset, 50
Formage, Joan, 118-19
Framlingham, Suffolk, 106
Frome, Somerset, 173
Gall, F.J., 2
Gamble, C., report on Avon Valley animal bone, 38, 40
Gardiner, Mark, reports on pottery and tile from Devizes, 94-6
Gauntlet, Hugh, 99
Gilfillan, George, 137
Gilmour, Felicity, review by, 170-1
Gingell, Christopher, work reviewed, 168-70
glass, 18th-century, 63
Gloucester, 67, 69, 95, 125
Gloucester, Earls of, 76
Gloucestershire, 150, 151; see also Cirencester, Cotswolds, Dumbleton, Gloucester, Rodmarton, Selsley Common, Uley, Whittington Court, Winchcombe
|
INDEX
Glow-worms, 153
Goddard, E.H., 103
Godwin (Precentor of Salisbury), 116-17 Goodfield, Annie, 165
Grafton: Wulfhall, 132
Graham, Alan, report on Avon Valley excavations, 8-57 Graves, Robert, 171
Great Bedwyn: Chisbury, 59
Great Ridge, 51-2
Great Somerford, 61, 63, 65, 67, 69, 71 Great Wishford: Ebsbury, 51
Greenland, Richard, 99
Greever, Garland, 137, 138, 141
Grey, Richard, 137
Grittleton, 175
Guido, C.M., 168
Hall, family, 118, 120, 126
Hallam, Henry, 140
Ham Green, Avon, 71, 95
Hambledon Hill, Dorset, 176
Hampshire, 34, 52, 95, 139, 174; see also Alice Holt, Beaulieu River, Chalton, Cowdery’s Down, Crondall, New Forest, Old Down Farm, Ructstalls Hill, Silchester, Southampton, Stockbridge Down, Viable’s Farm, Win- chester
Harding, Phil, 170; report on Avon Valley flintwork, 36-9
Hardy, Thomas, 171
Harold, Thomas, 123
Harris, Moses, 157
Harvey, P.D.A., 119, 120; R.B. and B.K., paper on Bradford, 118—29
Hatcher, Henry, 138
Hawkins, Desmond, work reviewed, 170-1
Haycock, Lorna, work reviewed, 171-2
Healy, Francis, review by, 175-6
hearth, medieval, 61, 65
Heath, Maud, 141
Heddington, 139
Hembry, Phyllis, work reviewed, 173
Henley, Walter of, 124, 125
Hereman (Bishop of Salisbury), 116
Hertford, Marquess of, 130
Hertfordshire: see Skeleton Green
Hey, Gill, 159-60
Hidden, Norman, paper on Hungerford fire, 130—5
Hill, Joseph, 167
hillforts, Iron Age, 159, 161-2; see also enclosures
Hilmarton, 106, 108
Hoare, Sir Henry, 5; Sir Hugh, 5; Sir Richard Colt, 3, 5, 137, 140, 161
Hobbs, Steven, review by, 173
hobnails, Romano-British, 23, 25, 35-6
Holt, 118, 119
Hood, Philip, 171
Horningsham, 163-4, 175
horseshoe vetch, 154
Hover-flies, 155
Howell, Nathaniel, 99
Howley, William (Archbishop of Canterbury), 138
Huish, 67, 69
Humberside: see Arras
Hungerford, paper on fire damage in, 130-5
Hungerford, family, 131-2, 134
Hunt, Jeffrey, 99; John, 99; Thomas, 100
Hunter, Michael, review by, 176-7
Huttoft, William, 167
181
icehouses, 89, 92, 94
Idmiston: Gomeldon, 83
Ilchester, Somerset, 51
industrial activities, 65-6, 72
Ingram, A.H.W., 5
inhumations: Beaker, 176; Bronze Age, 12-13, 41, 50, 145; Romano-British, 23—5, 41-2, 50-1, 143; Saxon, 143
insanity, 1-3
invertebrate populations, small, note on, 152—7
iron slag, 60, 61, 65
ironwork: prehistoric, 5-6; Romano-British, 23, 35-6
Isopods, 153
Jackson, J.E., 103
Jenkins, A.V.C., report on Avon Valley human bone, 41-2 jetton, 160
Jocelin (Bishop of Salisbury), 113, 116-17
Jones, Diana, report on animal bone from Devizes, 100
Kelston, Avon, 120, 125, 151-2
Ken, Thomas (Bishop of Bath and Wells), 141
Kennet, River, 131; Valley, 142
Kennet and Avon Canal, 166
Kenyon, John, 1
Kilmington, 125, 151
kilns: Romano-British (corn-drying), 22-3, 25, 34, 42, 445, 51; medieval, 163
Kington Langley, 120
Knapweed, 154-5
Knook, report on investigations at, 75—87
Knoyle, East, 138
lace, 165
Lacock, 138; Naish (Nash) Hill, 71, 95; Showell Nurseries, 159-60
Lamel Hill, Yorks, 2
Lancaster, Duchy of, 132
Lansdowne, Lord, 137, 139, 141
Last, Barbara, note on invertebrate populations, 152-7
Laughlin, Bronwen, ‘report on clay pipes from Devizes, 99-100; Michael, report on worked stone from Devizes, 100-1
Laverstock, 34, 55, 84
Lawrence, T.E., 171
Leland, John, 128
Leofgeat, 76
Liddington, 125
Limpley Stoke, 118, 120, 126
Lincoln, 106
Lincroft, Yorks, 1
Linley, William, 138
Lisburn, 169
Lisle, William, 137
Liverpool, 167
Loft, Mark A, report on Roman pottery from Malmesbury, 66
London, 68, 72, 130-1, 135-6, 139, 167; Battersea Shield, 6; Institute of Archaeology, 88, 102; Royal College of Sur- geons, 1; Society of Antiquaries, 1, 5; Westminster Hos- pital, 1
Long, William, 3; family, 118
Longbridge Deverill: Crockerton, 84
Loudon, J.C., 171
Lovel, family, 76
Lubbock, John (Lord Avebury), 4-5
Lyell, Sir Charles, 4
lynchets, 10, 12, 50, 53, 55, 145, 160, 162
Lyneham: Bradenstoke, 175
182 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Maclise, Daniel, 140
MacOmish, David, 161-2
Maiden Castle, Dorset, 176
Mailduib (Madulph), 59
Malmesbury, 168; hundred, 168; lace, 165; report on exca- vations at Postern Mill, 58-74
Maltby, Mark, 170
marjoram, 154
Marlborough, 100, 147, 148, 175; Downs, 168-70, 176
Marsh, A.E.W., 103, 110
Marshman, Michael, review by, 177
Marston Meysey, 162-3
Meligethes aeneus, 153
Meloe proscarabaeus, 153
Mepham, Lorraine N., reports on Avon Valley excavations, 25-36, 40, 55; report on pottery from Knook, 81-5
Merewether, John (Dean), 145
metalwork: Bronze Age, 161, 169-70; Romano-British, 8, 34-6; post-medieval, 96, 98-9
Meyrick, Owen, 168, 169
middens, 161-2
Middlesex: see Sunbury
Middleton Hall, Derbyshire, 3
Mildenhall, note on Iron Age coins from, 147-9; Cunetio, 147-9
Minety, 63, 66-9, 71, 95-6, 163
moated site, 162
molluscan evidence, 45-50, 81, 159, 160
Monkton Farleigh, 118
Moore, Thomas, 136, 138, 139
mortar, 101
Morton, S.G., 3, 4
moths, 153
musical instruments, 167-8
Musty, John, 163
nails (hobnails), Romano-British, 23, 25, 35-6
Nares, Robert, 136, 141
National Monuments Record, 174
Netheravon, 8-10, 51-2
New Forest, Hants, 27-9, 31-4, 83
Newbury, Berks, 95
Newman, Caron, report on Avon Valley excavations, 8-57; Richard, report on investigations at Knook, 75-87
Nichols, John Gough, 138
Nicolle, Jason St John, note on ‘Alvestone’, 151-2
Nilsson, — (of Lund), 5
Norris, Sir Henry, 130
North Stoke, Oxon, 146
Norton Bavant, 3, 5
Ogbourne St Andrew: Dean Bottom, 168-70; Rockley Down, 168-70
Ogle, John Savile, 138; Newton, 138; family, 138
Okeden, David Parry, 138
Old Down Farm, Hants, 26
Old Sarum, 113, 116, 163
Orange, Duke of, 130
Osmund, St. (Bishop of Salisbury), 113, 115-17
Overton, West, 145; Overton Hill, 143, 159
Oxford, 130, 131, 135; Trinity College, 137
Oxfordshire, 27—9, 32; see also Cuxham, Dorchester, North Stoke, Oxford
Paris, 130
Parre, William, 134 Parsonstown, Ireland, 2 Patney, 111
Pembroke, Earl of, 130-2, 135
Peters, J.E.C., review by, 173-4
Petty, Sir William, 139
Pewsey Vale, 165-7
Philipps, Sir Thomas, 140
Phillips, John, 2
Piggott, Stuart, paper on John Thurnam, 1-7
pits, 54-5; Neolithic, 12, 45, 48-9, 50, 176; Neolithic—Early Bronze Age, 159, 161; Iron Age, 159, 163; Late Iron Age-Romano-British, 16, 18, 19, 25, 42, 44-5, 51; medie- val, 80; medieval—post-medieval, 88-9, 92, 94
Pocock, George, 171
Polwhele, Richard, 138
Ponting, C.E., 103, 108
Poole Harbour, Dorset, 31
Pope, Alexander, 136, 138
Portland, Dorset, 100, 101
Potterne, 26, 83
pottery, 25—6; Neolithic, 3, 6, 161, 169, 176; Beaker, 6, 169; Early Bronze Age, 3, 6, 159, 161, 169; Middle Bronze Age, 12, 26-7, 32-3, 50, 169; Late Bronze Age, 26-7, 32-3, 50, 55, 82, 83, 161, 169-70; Iron Age, 10, 13, 16, 18, 26-7, 31-3, 54-5, 161; Romano-British, 8, 13, 16, 18-19, 21-3, 27-34, 55, 66, 82-3, 158, 160; Saxon, 61, 65, 67-71, 80, 160; medieval, 13, 34, 55, 58, 61, 63, 65, 67-71, 79-85, 88-9, 92, 94-7, 160, 163; post-medieval, 55, 58, 63, 65, 70-1, 81-2, 84-5, 88-9, 92, 94-7, 160; see also tile
Poundbury, Dorset, 50
Powell, Anthony, 176
Powys, John Cowper, 171; Llewelyn, 171
Poyntz, Richard, 118
Price, Sir Uvedale, 136, 138
Prichard., ].C:,.2; 45,6
priest’s house, 112
Prince, Roberta, review by, 165
Pugh, R.B., 168
querns, 160; Bronze Age, 170; Romano-British, 23, 36 Raby, F.J.E., 113, 115-17
Ranunculus bulbosus, 153
Redlynch: Hamptworth, 162
Renwick, W.L., 137
Reynolds, Andrew, survey of Compton Bassett church, 102-12, 160
ring ditches, 11-12, 19, 25, 45, 48-50, 142-6, 162; see also barrows
Robert (Duke of Normandy), 116
Robinson, Paul, note on Iron Age coins, 147-9
Rodmarton, Glos, 3
Roger (Bishop of Salisbury), 59, 61, 113, 116
Rogers, Daniel, 130-5; John, 130; K.H., review by, 168; Norman, review by, 171-2; Samuel, 136, 139; family, 118
Rosse, Lord, 2
Roundway, Hospital, 1, 3
Royal Commission on the Historical Monuments of England, 142, 161, 174-5
Royal Society, 2, 3, 5, 6; of Literature, 140
Ructstalls Hill, Hants, 52
Rudwick, Martin, 1
Russel, John, 123
Russell, Lord John, 2; Miles, report on Devizes excavations, 88-101
Salisbury, 110, 130-1, 137, 140-1, 163, 167, 175; paper on cathedral effigy, 113-17; spelling of name, 117; see also Old Sarum
INDEX
Salisbury Plain, 5, 176; Romano-British settlement on, 51-2; Training Area Conservation Group, 161
sarsen, 105, 169, 170
Savernake, 28-31, 148-9
Scabious, 154
Sedgehill, 126
seed, carbonised remains, 42-5
Selsley Common, Glos, 67, 69, 95
settlements: Middle Bronze Age, 145, 159-60, 169; Late Bronze Age—Early Iron Age, 161-2; Iron Age, 162-3; Romano-British, 8, 10-52, 158; medieval, 75-6, 85-6, 162, 163, 168
Seymour, family, 130-2
Shaftesbury, Dorset, 137-8; Abbey (Abbesses), 118, 120, 122-9, 151-2
Shalbourne, coin from, 150-1
Sharpey, William, 2
sheep, on Bradford manors, 124-5
Sherborne, Dorset, 116
Sheridan, Richard Brinsley, 138-9; family, 138
Shortt, Hugh de S., 113, 116-17
Sidmouth, Viscount (Henry Addington), 171
Silchester, Hants, 149
Skeleton Green, Herts, 50
skulls, study of, 1-6
slag, iron, 60, 61, 65
Slatter, Doreen, paper on W.L. Bowles, 136-41
Slocombe, Pamela M., work reviewed, 173-4
Sloper, George, 171-2
Smith, A.C., 143; I.F., review by, 168-70
snails, 153; see also molluscan evidence
Soames, Charles, 147-9
Society of Antiquaries of London, 1, 5
Soffe, Grahame, note on Beckhampton, 142-7
soils, 45-8, 80, 160, 166
Somerset, 151, 170; see also Abbas Combe, Cheddar, Frome, Ilchester, Stogursey
Somerset, Protector, 132
Sompting, Sussex, 161
South Newton: Chilhampton, 86
South Wraxall, 118
Southampton, 133, 138, 139
spindlewhorl, 35, 40
Spronks, John, 123
Spurzheim, J.G., 2
Staphilinids, 153
Steeple Langford: Hanging Langford Camp, 51
Stell, Christopher, work reviewed, 174-5
Stewartby, Lord, note on coin from Shalbourne, 150-1
Stockbridge Down, Hants, 146
Stockton, Earthworks, 51
Stogursey, Somerset, 170
stone artefacts, 5, 36, 66, 100-1
Stonehenge, 3, 5, 37, 48, 50, 105, 143
Stourton: Stourhead, 5, 171
Strange, Philip, 124
Stroud, D.I., paper (113-17
Stukeley, William, 105, 143
Suffolk: see Framlingham
Sunbury, Middlesex, 130
Surrey, 34, 71, 95
Sussex, 5, 151; see also Sompting
Sutton, Willey, 172; family, 172
Sutton Benger, 173
Sutton Mandeville, 173
Sutton Veny, 173; Pit Mead, 52
on Salisbury Cathedral effigy,
Swindon, 31, 170
183
Tackle, A.E., 121
tanning, 65, 66, 72
Teffont, 173
tell sites, Near Eastern, 161
Thamesdown Archaeological Society, 163
Thies, Alison, review by, 167-8
Thomas, Bridget, and Co., 165; Julian, work reviewed, 175-6
Thoms, W.J., 4
Thomsen, C.J., 4, 5
‘Three Age System’ of prehistory, 4-6
Thurmond, Nicholas, 120
Thurnam, John, 159; paper on, 1-7
ule, 160; Romano-British, 21, 346; medieval—post- medieval, 34, 63, 66, 82, 96-7, 101, 122, 163; stone, 36; see also pottery
Tilshead, 4
Timarcha tenebriosa, 153
Tisbury, 124-6, 175
tobacco pipes, clay, 71-2, 98-100, 160
Todd, Henry John, 138
Tomalin, David, 169
Tonks, W.C., 175
trackways, 160-1, 169-70
Trollope, Anthony, 2
Trowbridge, 69, 84, 136, 175
Trust for Wessex Archaeology, 8, 75, 142, 158-63, 168
Tucker, Adam, 126
Tuke, William, 1
Tylden-Wright, David, work reviewed, 176-7
Uley, Glos, 3 Upavon, 148, 162
Verwood, Dorset, 55, 85, 95
Viable’s Farm, Hants, 52
Victoria History of Wiltshire, 175; reviewed, 168 villas, Romano-British, 52
Vince, Alan, 94
Viner, —, 128
Wake, Charles, 138; Harriet, 138; Magdalen (Bowles), 138, 140
Wallis, Jon, report on Devizes metalwork, 96, 98-9
Wanborough, 29, 163
Warton, Joseph, 137, 141; Thomas, 137
Watts, Alaric, 137
well, medieval—post-medieval, 88, 92
Welle, Thomas atte, 123
Wesley, John, 171
Wessex, 5, 45, 170-1, 175-6; Archaeology (TWA), 8, 75, 142, 158-63, 168
West Bergholt, Essex, 106
West Overton, 145; Overton Hill, 143, 159
Westbury, 31
Westwood: Avoncliffe, 126; Winterleys, 123
Wharram Percy, North Yorks, 111
whetstone, 100-1
Whittington Court, Glos, 67, 69
Whittle, Alasdair, 159
Wilcot: Oare, 27
Wilkinson, K.N., 160
William I, 115, 116; II, 116; of Malmesbury, 59
Wilson, Daniel, 3, 4
Wilton, 59, 131, 132
Wiltshire Buildings Record, 173-4
Wiltshire County Asylum, 1
Wiltshire Record Office, 118, 173
184 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE
Wiltshire Wildlife Trust, 152 Winchcombe, Glos, 67, 70 Winchester, Hants, 103, 110, 130, 137-8, 141, 167 Winsley, 118, 123
Winter, Sir William, 130 Winterbourne, Down, 51 Winterbourne Stoke, 4 Witham, Essex, 6 Woodborough, 175 Woodford: Heale House, 137 Woodyer, Henry, 110 Woolf, Cecil, 136
Wootton Bassett, 70-1 Worsaae, J.J., 4
Wraxall, South, 118
Wroughton, 177; History Group, work reviewed, 177
Wyatt, James, 110
Wyles, Sarah F., report on Avon Valley molluscan evidence, 45-50
Wylye, River, 75
Wyndham, Henry P., 138
Yatton Keynell, 173
York, 1, 3, 111, 130
Yorkshire: see Lamel Hill, Lincroft, Wharram Percy, York Young, E.H., 171
Yowle, John, 133, 134
in?
ee
ee
Steet
Easy
Sitie Ws Ciel ere t gee