English Historic Carpentry E-Book

Cecil A. Hewett

• Herausgeber: Phillimore & Co Ltd
• Kategorie: Lebensstil
• Sprache: Englisch

Veröffentlichungsjahr: 2022

Beschreibung

A true magnum opus, Cecil A. Hewett's English Historic Carpentry appeals to every type of architectural historian, from the connoisseur of cathedrals to the vernacular 'barn and cottage' enthusiast. It also offers practical insight into the structure and age of the old homes and outbuildings of Britain, and will reveal that much timber-work is appreciably older than their owners might think. This book will be a source of pride and inspiration to all who work in timber today, showing that the master carpenters of the past achieved their highest levels of craftsmanship well before the master masons, and that they were their equals as medieval architects. Beginning as a young upstart with wild theories, Hewett grew to become the accepted authority on all matters of historic carpentry and a pioneer of a whole new technology in the dating of ancient buildings. In English Historic Carpentry, he provides the definite statement of his work, superbly illustrated by his own drawings. As Hewett maintained, 'the building is the document' and this book will teach you how to read the evidence that is written in wood.

Leseprobe

First published 1980

Second edition first published 2009

This edition first published 2022

The History Press

97 St George’s Place, Cheltenham,

Gloucestershire, GL50 3QB

www.thehistorypress.co.uk

© Cecil A. Hewett, 1980, 2009, 2022

The right of Cecil A. Hewett to be identified as the Author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988.

All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without the permission in writing from the Publishers.

British Library Cataloguing in Publication Data.

A catalogue record for this book is available from the British Library.

ISBN 978 1 80399 208 2

Typesetting and origination by The History Press

Printed and bound in Great Britain by TJ Books Limited, Padstow, Cornwall.

eBook converted by Geethik Technologies

CONTENTS

CHAPTER

List of Plates

Glossary

Acknowledgements

Bibliography

Introduction

1 The Anglo-Saxon Period (A.D. 449 to 1066)

2 The Norman Period (c.1050 to c.1150)

3 The Great Transition (c.1150 to c.1200)

4 The Early English Period (c.1150 to c.1250)

5 The Decorated Period (c.1250 to c.1350)

6 The Perpendicular Period (c.1350 to c.1450)

7 The Tudor Period (c.1450 to c.1550)

8 The Renaissance and After (1581 to 1890)

APPENDICES

1 Scarf Joints in Chronological Succession

2 Forms of Post-Head and Tying Joints for Tie Beams

3 Joints used for the Framing of Floors

4 Lap Joints for Secondary Purposes (such as Bracing)

5 Posts having Carved Capital and Base Treatments

6 String- or Band-Type Mouldings

_____________________

The Development of Storey Posts

The Development of Mortises

Some Technological Changes Illustrated by the Examples

Carpenters’ Mouldings

The Succession of Roof Designs

The Succession of House Types

_____________________

Conclusion

LISTOF PLATES

(between pages 182 and 183)

I Matrix of a notched lap joint on a re-used timber in the belfry of the church at West Bergholt, Essex. (Author’s photograph)

II Joint from the church of St Thomas the Apostle, Navestock, Essex.

III The roof of Little Hall, Merton College, Oxford after restoration. (Photo: J.W. Thomas, A.R.P.S., Oxford)

IV A queen post in the roof of Little Hall, Merton College, Oxford. (Photo: J.W. Thomas, A.R.P.S., Oxford)

V Vertical section of Salisbury Cathedral Spire. (Photo: John McCann)

VI The roof of Westminster Hall. (National Monuments Record)

VII The roof of Westminster Hall, showing the longitudinal arcades. (National Monuments Record)

VIII The chancel roof of the parish church at Saffron Walden, Essex. (Photo: Frank Joel)

IX Blackmore Belfry. (Photograph of author’s watercolour)

X The loading arrangements for joist-shear tests. (Photo. supplied by Dr D. Brohn, Bristol Polytechnic)

XI The Barn, Prior’s Hall, Widdington, Essex.

XII Plan and vertical section, scale-drawn, of the spire of the parish church at Upminster, Essex. (Author’s watercolour)

THE ILLUSTRATIONS

All the line drawings reproduced are by the author; their quality varies considerably because the dates of execution cover three decades. Those illustrating the D’Arcy Chantry, at Maldon, are by my colleague, Mr M.C. Wadhams, to whom I am indebted for numerous discussions of the subject. Figs. 225 to 233, and 236 to 249 are reproduced with the kind consent of the Greater London Council, and that authority’s Historic Buildings Section. The photographers whose work constitutes the half-tone plates are individually named in the List of Plates, and to them I am most grateful.

C.A.H

GLOSSARY

Addorsed: From heraldry, meaning placed back to back – the opposite of ‘addressed’, or face to face.

Angle tie: Tying timbers placed across angles, normally the returns of wall plates. These were widely used during the 18th century as a means to step hip rafters, which were seated in a third timber, the dragon piece.

Abutment:Abut, O.Fr. abuter, to touch at the end (à, to, bout, end). Any point in timber jointing where one timber’s end touches another constitutes an abutment. A ‘butt-joint’ is, therefore, one where ends meet; no integration is implied.

Arcature: The curvature of an arch, as segmental, ogee or lancet.

Arcade: A range of arches. Term applied also to the series of posts standing inside an aisled timber building, because they are sometimes arch-braced in their longitudinal direction.

Arch braces: Term generally applied to braces beneath tie beams, which were frequently curved, or arched.

Arris: The edge at which two surfaces meet.

Arris-trenched: Trenched (q.v.), so that the trench is cut obliquely through an arris and affects both adjacent surfaces.

Ashlar pieces: Short, vertical timbers at the feet of rafters, generally standing upon sole pieces. These continue the internal wall surface until it meets the underside of the rafters, avoiding a visual discontinuity and greatly strengthening the rafters’ base.

Barefaced: With the face uncovered, without a mask; avowed, open. Term used to denote a timber joint possessed of only one shoulder, but which normally possesses two.

Base crucks: Timbers placed as wall posts and containing the naturally grown angle of the eaves, above which they may rise to collar height.

Bays: The divisions, normally postulated by the material used for construction of the lengths of buildings. In the case of arcades, each arch is taken as one bay.

Bird’s-mouth: Term used to describe joints bearing a visual resemblance to an open bird’s beak.

Blade, -ing, -ed: Term used to specify scarfs that are face-halved and terminated in inset, barefaced tongues. See Fig. 270.

Bole: The butt of a tree trunk, normally of concave conoid form, used to provide jowls by inverting the timber.

Bowtell: Small roll moulding, or bead.

Brace: Any timber reinforcing an angle, usually subjected to compression.

Bridle: Term applied to timber joints having open-ended mortises, and tenons resembling a horse’s mouth with the bit of the bridle in place.

Bressummer: Breast-summer, a timber extending for the length of a timber building, normally forming the sill of a jettied storey.

Bridging-joist: Floor timber that supports the ends of common joists, and normally bridges the bays from one binding-joist to the next.

Broach: A spit or point.

Butment cheeks: The timber left on either side of mortises, against which the shoulders of tenons abut.

Cant post: Posts that converge upwards; see Navestock belfry.

Camber beams: Beams sufficiently cambered to form the basis of the simplest type of roof, their curvature serving to drain their surfaces when clad.

Chamfer: The slope or bevel created by removing a timber’s corner or arris. These are termed ‘through’ if run off the end of the workpiece, ‘stopped’ if terminating in a decorative form before the end of the piece or its conjunction with another.

Cant: The oblique line or surface which cuts off the corner of a square or cube. The term is applied to soulaces in roofs, because they produce a canted plane; roofs possessing soulaces, collars and ashlar pieces are thus described as ‘of seven cants’.

Chase: From chasse, a shrine for relics. In carpentry a score cut length-wise, a lengthened hollow, groove, or furrow.

Chase, or chased mortise: A long mortise into which a tenon may be inserted sidewise.

Cladding: The external covering applied to a wall or a roof.

Clamp: A term applied variously to timbers depending upon the type of building. In houses the term denotes horizontal timbers attached to the wall studs in order to support floors; these clamps normally indicate the later intrusion of such floors.

Clench, clinch: Either to turn the point of a nail or spike, and re-drive it back into the timber through which it has passed; or to form its end into a rivet, or clench, by beating it out upon a washer or rove.

Coak: A peg or dowel of a diameter almost equal to its length, used in 19th-century shipwrighting to join futtocks and timbers because it was cheaper than scarfing them.

Cogging: A method of housing an entire timber’s end, sometimes used to prevent its rotation – as in door cases.

Collar beam: A roof timber, placed horizontally and uniting a rafter couple at a point between the bases and the apex. Collar beams can act either as ties or strainers.

Collar purlin: A longitudinal timber uniting and supporting the collars; it is normally carried by crown posts.

Compass timber: A term denoting timber of natural and grown curvature, as distinct from relatively straight-grown timber from which curves are cut.

Common joist, -rafter: The majority of either kind, and normally those of the least cross section in any floor or roof.

Corner post: The post standing at the return of two walls, as at the end and adjacent side of a building.

Counter rebate: See Fig. 57.

Crenellate: To furnish with battlements, a decorative device used much in timber buildings.

Crown post: A vertical timber standing at the centre of a tie beam and supporting a collar purlin.

Cruck blade: The elbowed timber forming one half of a pair of crucks.

Cusp: In Gothic tracery the pointed shape or form created by the intersection of two concave arcs.

Ctyma: Ogee. Formed by a concave and a convex arc in a single linear association.

Dormer: An upright window protruding from the pitch of a roof.

Double tenons: Two tenons cut from the same timber’s end and placed in line; if side by side they constitute a pair of single tenons.

Dragon beam, -piece: A timber bisecting the angle formed by two wall plates. If a beam it supports a jetty continued around the angle; if a piece it normally serves to step a hip rafter.

Draw knife, or -shave: A hand tool possessing handles at either end of its blade, used to produce chamfers.

Durns: Timbers with grown bends suitable for the manufacture of door-ways of Gothic arcatures; two were frequently sawn from one piece of the requisite form.

Eaves: The underside of a roof’s pitch that projects outside a wall.

End girt, -girth: Horizontal timber in an end wall placed halfway betwixt top plate and groundsill, thereby shortening the studs and stiffening the wall.

Fascia: A board forming a front, frequently used to cover a number of timbers’ ends, as joists at a jetty, or rafters at an eaves.

Fillet: In moulding profiles a small raised band, normally of square section. Also a small squared timber.

Fish: A length of timber with tapering ends which can be used to cover and strengthen a break in another timber.

Fished scarf: A scarf that relies upon the introduction of a third timber.

Footing: Foundation.

Free tenon: A tenon used as a separate item, both ends being fitted into mortises cut into two timbers to be joined; often used to effect edge-to-edge joints.

Gambrel: Perhaps Old French (Norman). ‘Also gambrel roof … so called from its resemblance to the shape of a horse’s hind leg’ (O.E.D., 1933).

Girth, girt: Horizontal timbers in wall frames, placed at half height, which shorten and thereby stiffen the studs.

Groundsill: The first horizontal timber laid for a timber building. As the name implies these were in ancient times laid directly upon the ground, as was the case at Greensted church, Essex.

Halving: In jointing the removal of half the thicknesses of two timbers, as in cross halving.

Harr, arr: The edge timber of a door leaf or gate nearest the hinges, the opposite edge timber of which is the head.

Hanging knee: Term denoting a knee placed beneath a beam. Knees placed above beams are ‘standing knees’, or ‘standards’, and those in the horizontal plane are ‘lodging knees’. All three derive from shipwrighting and were used in ships as early as the Viking period.

Haunch: Adjuncts of tenons, designed to resist winding; they may be square or diminished.

Header: Short timber to carry rafters’ tops at the exiture of chimney-stacks.

Hewn knee: A knee, or angle, cut from a timber’s end, as distinct from a separate and applied piece.

Hip rafter: A rafter pitched on the line of intersection of two inclined planes of roof, forming the arris of a pyramidal form.

Hogging: Stress caused by supporting the centre of a beam and leaving the ends unsupported, as when a wave rises amidships of a vessel, and beneath her.

Housing: In jointing a cavity large enough to hold an entire timber’s end.

Jamb: The side of a doorway, archway or window.

Jetty: The projection of a floor outside its substructure, upon which the next storey was built. This resulted in floor areas that increased as the storeys ascended.

Joggle: Said of two timbers, both of which enter a third but which have their joints out of line in order to avoid excessive weakening.

Joist: The horizontal timbers supporting floors; these are binding, bridging, common and trimming. Binders unite storey posts; bridgers span or bridge each bay from binder to binder; commons are the most numerous and actually carry the floor boards. Trimmers are used to frame the edges of voids, such as stair wells.

Jowl: (Jole) ‘The external throat or neck when fat or prominent … the dewlap of cattle’. (O.E.D.) Term applied to the thickened ends of such timbers as storey posts which facilitate the jointing of several other timbers.

Kerf: The cut produced by a saw.

Key: Tapered piece of dense hardwood transfixing a scarf, used to close its abutments.

King stud: A stud placed centrally in a gable, normally supporting the collar purlin.

Lap dovetail: That form of dovetail that overlaps, and is not finished flush. The alternative form is the ‘through dovetail’ used by cabinet makers; they may also be ‘secret’, ‘secret-mitred’ or double.

Lap joints: Any jointed timbers which overlap each other.

Lodged: ‘To put and cause to remain in a specified place.’ (O.E.D.) A term applied to floors retained in place by their weight alone.

Main span: In aisled buildings this is the central and greatest distance spanned.

Midstrey: The porch-like structure at the front of a barn, derived from middle-strey. Each bay of a barn was a strey, and ancient barns normally had one such porch at their centre.

Mitre: Abutments at 45 degrees, producing square returns.

Muntins: Vertical members of panelled areas; the term may derive from mountants.

Mullions: Vertical components of windows, placed in the void.

Nogging: The material used to infill a framed wall betwixt sill and top.

Notched laps: A category of lap joints having V-shaped indentations on plan-view to prevent their lengthwise withdrawal.

Outshot, outshut: An area of space added to a building’s bays, normally at the sides: when at the end of a building they are called by the ancient term ‘culatia’.

Passing brace: A brace uniting several successive members of a frame and passing them by means of halved jointing; of mainly Early English and Decorated usage. (Author’s coinage, 1962, without historical validity.)

Plate: A horizontal timber laid at the base of a timber frame; the term implies a footing, as distinct from a groundsill.

Prebend: The share of the revenues of a cathedral or collegiate church allowed to a clergyman who officiates in it at stated times.

Prick post: Any vertical timber placed in compression, but not a storey post.

Principal-rafter: A heavy rafter placed at bay intervals, normally associated with side purlins.

Purlin: A longitudinal timber in a roof.

Queen posts: Posts set in pairs between tie beams and collars and acting in compression.

Raking struts: Inclined struts used in pairs between tie beams and principal-rafters.

Reversed-assembly: Indicates a system of rearing transverse framing units, the lengthwise timbers of which (top plates) are laid last. In these cases the tie beams are under the top plates. (Author’s coinage, 1962, without historical validity.)

Rive: To split timber lengthwise, i.e. along its grain.

Rove: The circular plate, or washer, upon which the clench, or rivet, in boat- or ship-building is formed.

Sagging: Stress caused by supporting the ends of a timber and applying weight to its centre.

Sally: An obtusely angular and pointed projection, normally on a timber’s end. Alternatively a ‘tace’.

Samson post: ‘Pillar erected in a ship’s hold, between the lower deck and the Kelson’. (O.E.D.) The term alludes to the strength of Samson (Judges XVI, 29), and is applied to similar posts used to support early floors.

Scarfing: The jointing of relatively short timbers into continuous lengths, by means of various expedients; the four faces are smooth and continuous.

Scarfed cruck: A cruck blade having a scarf-jointed angle, as distinct from a grown angle.

Set: The divergence of the sides of a dovetail.

Shore: An inclined timber supporting a vertical one, acting in compression.

Shuts: The edges of a door leaf collectively form the ‘shuts’ of that door.

Side girt, girth: See end girt.

Soffit: Underside or archivolt.

Sole pieces: Short horizontal timbers forming the base of any raftering system that has a base triangulation.

Soulace: A definitive term (Salzman, 1952) for secondary timbers connecting rafters with collars, and placed under the latter.

Spur tie: A short tie such as connects a cruck blade and a wall plate, or a collar arch and a wall plate.

Spire mast: Central vertical timber of a framed spire.

Squint: Angle other than 90 degrees.

Storey post: A wall post of a multi-storeyed timber building that continues through the floor levels.

Straining beam: A horizontal beam between two posts, acting in compression to keep them apart.

Strut: A timber in a roof system that acts in compression, in a secondary capacity.

Stub tenon: A short tenon that does not entirely penetrate the mortised concomitant timber.

Studs: From O.E. studu, a post; the vertical common timbers of framed timber walls.

Table: A raised rectangular portion on a worked timber, normally a scarf adjunct.

Tace: See sally.

Tail: The male part of a dovetail joint.

Tie beams: Beams laid across buildings to tie both walls together; they must have unwithdrawable end joints for this purpose.

Tongue: A fillet worked along the edge of a plank to enter a groove in another.

Top plate: A horizontal timber along the top of a framed wall.

Transom: A cross beam acting as a support for the superstructure.

Trench: A square sectioned groove cut across the grain.

Tusk: The wooden key driven through the protruding end of a tusked tenon, an unwithdrawable form of that joint.

Waney: Used to describe timber, the squared section of which is the greatest that can be cut from the rounded trunk, when any missing sharp arrises are said to be ‘waney’ edges.

Winding: In carpentry the result of torque or twisting, or the result of drying a spirally-grained tree.

Wind braces: Braces fitted into the angles of either roofs or walls to resist wind pressures.

Only terms that are likely to be strange to the general reader are here given, and the correct reference is, of course, the Oxford English Dictionary.

ACKNOWLEDGEMENTS

I wish to record my gratitude and indebtedness to all who have facilitated my studies during the past 30 years: the bishops, deans and chapter clerks, master masons and masons, carpenters and librarians of our great cathedrals – and the numerous architects responsible for their fabrics. Also to the many vicars and rectors of parish churches, and their secretaries to Parochial Church Councils. Among these I particularly thank Mr L.S. Colchester of Wells Cathedral, for minutely detailed information concerning the documentation of Wells; Dr J.H. Harvey for his unfailing readiness to assist with all matters that affect the dating of the cathedrals or monastic great churches of England; and Mr R.O.C. Spring of Salisbury Cathedral, for his sustained kindness and assistance with studies and ascents of that cathedral. Also Dr H.M. Taylor, for much advice and assistance with the assessment of pre-Conquest buildings. I am also grateful to the late Mr S.E. Rigold for much assistance with documentary researches concerning many manor-houses, and parts of the historical background information for these I quote from his writings. I am particularly indebted to Miss Linden Lawson of Phillimore [an imprint of The History Press] for her immense care and attention to detail throughout the period of this book’s production.

I thank the innumerable owners and occupiers of countless historic buildings, both domestic and agricultural, who have so kindly allowed their roofs to be searched, or their floors excavated in search of early bases for posts. Also my wife, for continual assistance and tolerance during the long and irksome compilation of this and previous books upon the subject.

BIBLIOGRAPHY

Arnold, T., Memorials of St Edmund’s Abbey, Rolls series, 96 (London, 1890)

Barker, F., ed. and trans., Vita Aedwardi Regis (London, 1890)

Beaumont, G.F., A History of Coggeshall in Essex (London, 1890)

Berger, R., ed., Scientific Contributions to Medieval Archaeology (1970)

Bilson, J., ‘Notes on the earlier Architectural History of Wells Cathedral’, Archaeological Journal (1928)

Brown, G. Baldwin, The Arts in Early England II, Anglo-Saxon Architecture (London, 1903. 2nd ed. 1925)

Bruce-Mitford, R.L.S., The Sutton Hoo Ship Burial (B.M. Trustees, London, 1954)

The Builder, Vol. LXXXVII (1904)

Calendar of Patent Rolls, Vol. 11, Edward VI (H.M.S.O.)

Cave, P., Hospital of St Cross (1970)

Chapman, F.R., ed., The Sacrist Rolls of Ely, Vol. I (Cambridge, 1907)

Chisenhall-Marsh, T.C., Domesday Book, relating to Essex (Chelmsford, 1864)

Christensen, A.E., jnr., Vikingskipene (Oslo, 1970)

Chubb, Rev. N., All Saints, Brixworth (1977)

Clapham, A.W., English Romanesque Architecture, After the Conquest (1964)

Clowes, G.S. Laird, Sailing Ships, their History and Development (H.M.S.O., 1932)

Colchester, L.S. and Harvey, Dr J.H., ‘Wells Cathedral’, Arch. Jnl., Vol. 131 (1974)

Colchester, L.S., Wells Cathedral Library (1978) 2nd ed.

Colvin, H.M., ed., History of the King’s Works (1963)

Cristie, H., Olsen, O. and Taylor H.M., ‘The Wooden Church of St Andrew at Greensted, Essex’, Antiquaries Journal (1979), Vol. LIX, part I

Davey, Dr N., A History of Building Materials (London, 1961)

Davis, R.H.C., ‘The Norman Conquest’, Hist. Ass. Jnl., Vol. LI (1966)

Deneux, H., ‘L’Evolution des Charpentes du xie au XVe siècle’ (C.R.M.H.)

Downes, K., Hawksmoor (1969)

Drinkwater, N., ‘Old Deanery at Salisbury’, Antiquaries Journal, 44 (1964)

Dugdale, W., Monasticon Anglicanum (London, 1718)

Everett, C.R., ‘Notes on the Decanal and other houses in the Close of Sarum’, Wiltshire Archaeological & Natural History Magazine, 50 (1944)

Finn, R. Welldon, ‘Changes in the Population of Essex’, Essex Archaeol. /Historical Society, Vol. 4 (1972)

Fletcher, Sir Banister, History of Architecture (1956)

Forrester, H., Medieval Gothic Mouldings (Chichester, 1972)

Gethyn-Jones, Canon D., Kempley (1957). Available at the church

Gibb, J.H., Sherborne Abbey (1972)

Hale, Archdeacon W.H., The Domesday of St Paul’s (Camden Society, 1858)

Harvey, Dr J.H., ‘The King’s Chief Carpenters’, Arch. Jnl., 3rd series, Vol. XI (1948)

Harvey, Dr J.H., The Master Builders (London, 1971)

Harvey, Dr J.H., The Perpendicular Style (1978)

Harvey, Dr J.H., English Cathedrals (1961)

Harvey, Dr J.H., The Medieval Architect (1972)

Harwell, Berkshire Carbon 14/Tritium Dating Laboratory. Certificate dated 2 March 1976. HAR – 1258. Age bp (yrs) 1020±90, bp – 1950, 930

Hewett, C.A., ‘The Barns at Cressing Temple, Essex’, Journal of Society of Architectural Historians, March 1967, Vol. XXVI, No. 1. (U.S.A.)

Hewett, C.A., ‘Aisled Timber Halls and Related Buildings, Chiefly in Essex’, Transactions, Ancient Monuments Society, Vol. 21 (1976)

Hewett, C.A., and Smith, J.R., ‘Faked Masonry of the Mid-13th Century at Navestock Church’, Essex Journal (1972), pp.82-5

Hewett, C.A., ‘The Carpentry’, Archaeology in the City of London, No. 3 (1975)

Hewett, C.A., ‘The Dating of French Timber Roofs’, Transactions, Ancient Monuments Society, new series, Vol. 16 (1969)

Hewett, C.A. and Tatton-Brown, T., Archaeologia Cantiana, Vol. XCII

Highfield, Dr J.R.L., ‘The Aula Custodis’, Postmaster, Vol. IV, No. 4 (1970)

Hope-Taylor, B. (ed. by C.A.R. Rawley-Radford), ‘The Saxon House, A Review and Some Parallels’, Med. Arch., No. 1. (1975)

Hunt, E.M., The History of Ware (Hertford, 1949)

Jervis, S., The Woodwork of Winchester Cathedral (Friends of Win. Cath., 1976)

Kent, Rev. J.A.P., History of Selby Abbey (London, 1968)

Kent, W., The George Inn, Southwark (1970)

Kidson, Dr P., A History of English Architecture (1965)

Larking, L., The Knights Hospitallers in England (Camden Soc., 1857)

Leeds, E.T., Early Anglo-Saxon Art and Archaeology (Oxford, 1936)

Lingpen, A.R., ed., Master Worsley’s Book (London, 1910)

McHardy, A.K., The Church in London, 13 75-1392 (London Record Society, 1977)

Mills, D., Lambeth Palace (Church Information Board, 1956)

Morant, P., History and Antiquities of Essex (London, 1768, Chelmsford, 1816)

Morgan, F.C., A Short History of Abbey Dore (1949-51)

Ogborn, M.E., Staple Inn (Institute of Actuaries, 1964)

Parsons, D., ed., Tenth-Century Studies (Chichester, 1975)

Pevsner, Sir N., An Outline of European Architecture (1943, 1963, 1973)

Pevsner, Sir N., Buildings of England – various dates of separate County publications

Postan, M.M., The Mediaeval Economy and Society (1972)

Rackham, Cl., Blair, W.J. and Muriby, J.T., Med. Arch., Vol. XXII (1978).

Ray, Rev. P.W., History of Greensted Church (Ongar, 1869)

Repton, J.A., Norwich Cathedral (Farnborough, 1965)

Rickman, T.H., ‘The Farmhouse, Thorpeacre, Loughborough’, Med. Arch., Vol. 12 (1968)

Rigold, S.E., ‘Romanesque Bases, in and South-East of the Limestone Belt’, Antiquaries Journal, Occasional Papers

Rodwell, Dr W., ‘The Archaeological Investigation of Hadstock Church, Essex’, Antiquaries Journal, Vol. LVI, part I (1976)

Royal Commission on Historic Monuments, Inventory, Essex, 1926 (H.M.S.O.)

Rudder, S., A New History of Gloucestershire (1779)

Salzman, L., Building in England (1952)

Sherley-Price, L., trans., A History of the English Church and People (1968)

Simpson, Sir J.W., Some Account of the Old Hall of Lincoln’s Inn (Brighton, 1928)

Stenton, Sir F., English Families and the Conquest (1968)

Stranks, Ven. C.J., Durham Cathedral (London, 1971)

Taylor, Dr H.M. and J., Anglo-Saxon Architecture (Cambridge, 1965)

Teledyne Isotopes, 50 Van Buren Avenue, Westwood, New Jersey, 07675. Dated

Thorpe, B., The Anglo-Saxon Chronicle trans. Longman, Green (London, 1861)

Victoria County History of Essex (London, 1956)

Whitham, J.A., The Church of St Mary of Ottery (1968)

Wilson, C.G., Winchester Cathedral Record (Friends of Win. Cath., 1976)

Wilson, D.M., ed., The Architecture of Anglo-Saxon England (1960)

Wood, R.G.E., County Advisory Officer, History; Essex County Council

INTRODUCTION

A decade has elapsed since an introduction was written for my first book on structural carpentry, and since this is so short a time it is still appropriate to begin with the same remarks. There is a view, widely held, even among those having special knowledge of and interest in historic timber buildings, that carpenters had, at any time in the past, the whole range of timber joints known at the present time, from which they were free to select the one they preferred for their purpose. That is to say that the carpenter’s craft, unlike any other, had had no history of development towards perfection; this is but one of a number of irrational opinions that gained a general acceptance in this country for reasons that are obscure. A similar view was that medieval buildings were not attributable to individual architects of genius, because ‘Mediaeval men had an innate instinct for co-operative design’ (Dr J.H. Harvey, 1972, 9), or that man must have first built in stone. Such opinions probably originated in the higher centres of learning, since they obtained during a century of compulsory general education, but they are immediately ridiculous upon serious consideration.

The first to realise the differences of styles of carpentry, or styles of jointing and assembling frames, seems to have been Henri Deneux, of the French Historical Monuments Service. He published in July 1927 a volume entitled L’Évolution des Charpentes du XIè au XVIIè Siècle This work should have revolutionised studies of historic timber buildings from that time onwards, since after reviewing over five hundred important structures he was able to show that: ‘by examining all these examples of framework we have been able to prove, despite their great variety, that each period is characterised by definite assembly-methods (dispositions)’. It is now over half a century since this publication appeared, and its message is gradually coming to be recognised.

Deneux’s methods cannot be improved upon, and the subject is approached by way of a selection of timber buildings that are representative of the technique of their times, reviewed at the end of each recognised period, and more synoptically reviewed at the end of the book. The periods of English architecture ‘are based partly on historical periods and partly on architectural character’, and ‘as they have held the field for so long in all descriptions of English architecture, they have become, as it were, an integral part of architectural phraseology’ (Sir Banister Fletcher, 1956, 347). In view of the inescapable truth of this statement these divisions have been used, with this overriding qualification – ‘What is important about the Styles is not their dates, which fluctuate (in different parts of the country) but their very recognisable general character’ (J.H. Harvey, 1979, letter to author).

It becomes clear as the examples are considered that the effects of these changes, which were essentially stylistic, upon the works of structural carpenters are not always obvious; and it also seems that certain styles directly affected structural thought, and not always to its mechanical advantage. The examples are described and illustrated, their structural joints and decorative features are listed in chronological sequence at the end of the book, and from these sequences the logical conclusion becomes clear – that carpenters’ jointing techniques had of necessity to attain mechanical efficiency if their structures were to endure for any length of time. This is entirely obvious, and it was equally true of the masons’ techniques, yet it has only been recognised in the second instance. It follows from this assessment of the historical succession of developments that timber buildings are datable by the techniques employed for their construction, a fact that has gained least acceptance among students of the subject, but which is gaining some recognition.

It is possible to establish when certain improvements were effected by studying numerous high quality buildings without regard to regional characteristics, thereby discovering what constituted the history of this particular technology at the national level – but it is difficult to show when out-of-date techniques lapsed into obsolescence. The use of jointing for the dating of buildings is therefore limited, but its value can hardly be over-estimated, since typological dating (which is our principal method) cannot be applied in part – that is, without regard to the most important type-series relevant to the study. Dating is fundamental to architectural history, and it is often problematical, since even incised dates such as occur on date stones or the side girt of Rooks Hall may record either the date of commencement or of completion. The same uncertainty attaches to cathedrals (which are generally the most fully documented buildings we have), because the records are primarily accounts, and as such concerned with the costs of the operations. The inception, possibly at Wells, of the single tenon mounted on spurred shoulders can only be broadly dated by Bishop Bubwith’s will, but such records do assist with tolerably close approximations. It has in the past been considered that a building was dated if ascribed to a particular century, but the inadequacy of this is apparent when the present selection is considered – and the majority of works described cannot be dated with respect to each other. Two alternative and scientific methods of date determination are known, carbon14 dating, and tree ring analysis – normally given the name of dendrochronology – and for a concise exposition of the principles of both the book by Professors Horn and Berger is recommended (R. Berger [ed.], 1970, 17-21 and 183-212). The first is a valuable but very expensive method, and the second is not yet available to the extent that is desirable. How useful these methods will become has yet to be established, and typological assessments that have due regard to the technological typology herein proposed are the best method available at the present time.

As in previous books it has been found most convenient first to describe and illustrate either the whole building in question, or such of its features as furnish evidence supporting the general thesis. This has been done with regard to the accepted periods of the successive styles of English architecture insofar as it has seemed possible, for not all timber buildings conform to those styles. At the end of each period the evidence provided by the selected structures is generally discussed, and after the selected buildings have all been introduced the evidence they provide is reviewed under the heading of a single category of joints or other building components, such as storey posts.

The important matter of cruck building has not been dealt with as those familiar with that building type would require; this is because only a few examples exist in the south-eastern counties. Some such buildings have, however, been mentioned as an aspect of the overall pattern. There has been only one historical course of development in English structural carpentry in the writer’s opinion, and the preference for either straight or crooked timber was, irrespective of the regional pattern of distribution, one of the basic options that were open to carpenters. The use of base crucks cited in this text, in the capital and at relatively recent dates, endorses this view. It has been found in such cruck buildings as the writer has examined with the necessary thoroughness – such as the barn at Bishop’s Cleeve in Gloucestershire – that the technological aspects of the carpentry, such as the scarfing, relate precisely to the general thesis expounded, and such buildings should be considered as an integral part of timber-building history rather than as something separate and distinct.

It has always been accepted that fieldwork can never be final, and while this text has been in production important new discoveries have been made which elaborate and confirm the thesis expounded; I have chosen to omit these, however, rather than alter the layouts and delay the production. In the event of a second edition many amendments will be appropriate, since this is a field of study that is in its infancy.

CHAPTER ONE

EXAMPLES FROM THE ANGLO-SAXON PERIOD (A.D. 449 TO 1066)

THE CHURCH OF ST ANDREW, GREENSTED-JUXTA-ONGAR, ESSEX

This building is, in the present state of knowledge, unique in England. Its walls are made of oak trunks, split into halves and reared with their flat surfaces inward, a method of timber building for which the Venerable Bede seems to provide some evidence, referring to the year A.D. 664 (L. Sherley-Price, 1968, 185): ‘He (Bishop Finan) built a church in the Isle of Lindisfarne suitable for an episcopal see, constructed, however, not of stone, but of hewn oak thatched with reeds after the Scots manner’. Furthermore: ‘Eadbert, a later Bishop of Lindisfarne, removed the thatch and covered both roof and walls with sheets of lead’, implying two important factors: a substantial building to carry such weight, and a metal technology able to produce sheet lead. Nothing else can be inferred from Bede, unfortunately, than that some churches were of timber during his time, i.e., the 7th century. Concerning the date of Greensted church there has been much speculation, and more doubt; but that the most acceptable of the half-logs in its walls are of a date between the 7th and the llth centuries seems apparent, the terminal limit being provided by the translation of St Edmund’s remains in 1013. This event was recorded in a document of disputed date which is no later than c.1300 (BM. Add. MS. 14847, f. 20. and V.C.H., 1856, 60), and which has been much quoted: Idem apud Aungre hospitabatur vero ejus nomine lignea capella constructa permanet usque hodie – and seems to indicate Greensted church.

The association with Edmund, king and martyr, has clouded the matter of dating the structure, since it has been thought that it was built for the purpose of housing the remains, rather than that it was already in existence and at a convenient point along the route from London to Bedriceworth, afterwards Bury St Edmunds. That it could not have been built hurriedly for the occasion is evident from its complex assembly method, and long-standing use of its site by the church has been established by a limited excavation, conducted in 1960 (H.R. Christie, O. Olsen and H.M. Taylor, 1979, 92-112). This was reported as follows: ‘Recent excavations in the chancel have established the former existence of a small wooden chancel, of upright logs set in the ground, without any sill, and of a larger wooden chancel which replaced it; the larger chancel had a wooden sill and may therefore be assumed to be of the same general type as the existing nave’. This was approved, at the time, by all the excavators (Dr H.M. and J. Taylor, 1965, 263). The last ‘restoration’ occurred in 1848-9, when it was thought necessary to rebuild the nave because the ground sills had rotted and the half-logs were said to be partially suspended from the top plates; a total rebuild was rapidly carried out, the Builder magazine of the day observing that this was of ‘undue severity’. It certainly was, and the greater part of the western gable, which had survived until that date intact, was removed to provide an unnecessarily wide vestry door; whether the timbers were destroyed or went into some private collection has eluded the records of this disastrous event. A recent re-examination of what remains confirmed that all evidence for the original roofing method was lost at that time.

Fig. 1 The top joints of the Greensted half-logs with the top plate; perspective from Ray’s drawing.

Study of the accounts that were published concerning this restoration serves only to prove disagreement between the various observers who wrote them, but these and the remains of the fabric are the only basis for a reappraisal of the building. The incumbent of the parish during the rebuilding was the Reverend P.W. Ray, who produced a small book about his church that ran to five editions (Rev. P.W. Ray, 1869, 19-20), in which he wrote: ‘There are 24 timbers on the South side, and 25 on the North. The nave is 29ft 9ins long, 14ft wide, and 5ft 6ins high to the top of the plate. The west end was carried up in the middle as high as the ridge of the roof, and consisted of two layers of planks fastened together with treenails. The planks are not long enough to reach the whole height, they are therefore so arranged as to break both the perpendicular and horizontal joints’. This information can be expanded by a more detailed account (Essex Record Office, Mint Portfolio, Greensted), which was also contemporary, and where accounts agree with each other and the remains of the fabric, grounds for deduction as to its original form and detail exist.

Another observer said of the walls: ‘They are about 6 feet high, including the sill and the plate, and are formed of rough half-trees, averaging about 12 inches by 6 inches (the greatest length on the base line being 18 inches by 9 inches, and the least 8 inches by 6 inches). Mr Suckling does not believe them to have been “half trees”, but that “they had a portion of the centre, or heart, cut out, probably to furnish beams for the construction of the roof and cills; the outsides or slabs thus left being placed on the cills”. We see no evidence of this, for the timbers were evidently left rough, and the dimensions prove them to have been, as nearly as may be, “half trees”. These uprights were laid on an oak cill, 8 inches by 8 inches, and tenoned into a groove 1½ inches deep, and secured with oak pins. The cill on the south side was laid on the actual earth; that on the north side had, in two places, some rough flints, without any mortar driven under. The roof plates averaged 7 inches by 7 inches, and had a groove corresponding with the cill, into which the uprights were tenoned and pinned. The plates were also of oak, but they and the cills were very roughly hewn, in some parts being 10 inches by 10 inches, and in others 6 inches by 6 inches or 7 inches.

Fig. 2 Greensted, scale-drawn elevation of the north-west part of the gable, with horizontal section at base beneath the scale.

‘There were twenty-five planks or uprights on the north side, and twenty-one on the south side. The uprights in the north side were the least decayed. Those on the south side required an average of 5 inches of rotten wood to be removed, those on the north about 1 inch only, and the heights of the uprights, as now refixed, measuring between plate and cill, are, on the north side, 4 feet 8 inches, on the south side 4 feet 4 inches, the cills being bedded on a few courses of brickwork in cement to keep them clear of damp. The uprights were tongued together at the junction with oak strips, and a most effectual means it proved of keeping out the wet, for although the interior was plastered, there was no evidence, in any part, of wet having driven in at the feather edge junction of the uprights – a strange contrast to many of our modern churches, where, with all the adjuncts of stone and mortar, it is found no easy matter to keep out the driving weather from the south-west. The roof was heavy, and without any particular character; it consisted of a tie-beam, at less than 6 feet from the floor, with struts. The covering was tile’.

An alternative and conflicting account from the same portfolio (E.R.O., Mint Portfolio, Greensted) says: ‘Each was cut with a tenon going into the cill; and the top of each was cut to a thin edge and pinned into the roof-plate’. This last detail was illustrated in several of the editions of the book by the Rev. Ray, and is redrawn as a perspective in Fig. 1. This is an important difference between the records quoted and the surviving fabric, which now has tenons along the tops; it is felt that these were probably cut from the thin edges by the carpenters affixing the new top plate. It is difficult to believe that the incumbent would have taken the trouble to have a block engraved and published in order to propagate a falsehood. There was, at the restoration, a south doorway, 4ft 7ins wide, with door-posts, and a northern doorway, 2ft 5ins wide – this last was found blocked and was reblocked with original timbers from some part of the structure.

The northern part of the west gable is drawn to scale in Fig. 2, in which a horizontal section showing the assembly testified by Ray is drawn at the base. No part of this wall that can be examined today confirms Ray’s statement, but outside measurements set against inside measurements taken from the actual corner suggest the construction drawn – which, however, cannot be proved. In fact, at least two half-logs with grooved sides can be seen from within the west tower, one of which is scarfed simply and was fixed with two nails. This area, when checked and compared with the complementary south-western area, shows enough original peg holes to form a horizontal line across the wall at a height that was probably between the two top plates at each side; this indicates the original existence of a tie beam at this end, without which the whole would have been too flimsy to survive.

Fig. 3 The south-west angle at Greensted, raised above the grooved sills.

The south-western corner is drawn as Fig. 3, in which the ‘thin edges’ shown by Ray and other contemporaries are drawn. Both western corner posts have a quarter removed to form their internal angles, and this is a feature which no contemporary plans show, but on-site examination has failed to prove that the northern post is not an original, due allowances having been made for the partial replacement of its timber at the top. The most curious detail of these corners is the elaborate arrangement for mounting two top plates, each rebated to carry a plank-on-edge, between them; it is difficult to envisage a Victorian carpenter inventing this elaborate system, and one suspects that it is a reflection of the original. The original uprights of this end are taller than the corner posts by the corresponding amount, a fact supporting the authenticity of the method.

Fig. 4 Greensted, method of assembly against the corner logs.

Fig. 5 Greensted, suggested method of top plate insertion.

Fig. 6 Greensted, hypothetical reconstruction of the missing central tie beam.

The timbers shown in Fig. 2 with their tops cut obliquely flat have survived in that form, and suggest that the verge rafters were set on to the same form of tenons as were used for the side walls. As shown in Fig. 4 the studs (half-logs) probably had to be reared against the corner posts, into which the fillets would have been previously inserted with respect to the side walls, the same corner logs being rebated to accommodate the end wall. The top plates could then have been advanced, endwise, into the rebates cut in the corner posts (Fig. 5). Alternatively, all the side walls complete with sills and top plates, pegged together and each with terminal corner posts, could have been reared by sheer manpower, and the whole stabilised by the tie beams at that point. The method of fitting the tie beam shown in Fig. 6 was, in fact, hypothetical – but the most recent on-site examination proves that this was the method used by the 1849 restorers. It is probable that a tie beam was fitted at the eastern end, but nothing survives there today. The former existence of the western tie beam is not only substantiated by the peg holes now approximating to its line; it was also obviously necessary because if, as Ray described it, it was built of two layers all vertically placed and slightly overlapping, it could have possessed absolutely no strength with which to survive until 1849. Fig. 7 shows such a tie beam. Against this beam the western wall could have been reared, either of the two alleged layers or of single half-logs – a structural point upon which the visible evidence is equivocal. Figs. 8, 9 and 10 illustrate possible processes for this operation. Fig. 10 shows the west as though built of grooved half-logs, together with the scarf visible on one half-log that survives in the tower at south-west enlarged on the left. Fig. 11 is a purely hypothetical impression of the whole, with assumed ridge piece and purlin slots.

Fig. 7 Greensted, hypothetical reconstruction of assembly method for the western end, against the tie beam.

Fig. 8 Greensted, the western plank wall, viewed from inside.

Fig. 9 Greensted, the western wall viewed from outside, with internal layer complete and external half-logs being applied to it.

Fig. 10 Greensted, reconstruction of the complete western gable, with exploded view of joint in a halflog.

Fig. 11 Greensted, hypothetical reconstruction of whole carcase.

The plank-on-edge system of wall plating perpetuated by the 19th-century restorers is accounted for in Fig. 6, wherein it may be seen that this plank fills the gap betwixt the top plates which was caused by fitting tie beams between them. These suggestions go a little way towards understanding what little remains at Greensted, and indicate. a remarkably close relationship with the grave chamber of the Gokstad ship.

Fig. 12 The framing of the Rhenish helm at Sompting.

THE CHURCH OF ST MARY, SOMPTING, SUSSEX

The western tower of this church retains a rare example of the spire form known as the ‘Rhenish helm’, a square-planned and pyramidal spire that rises from the apexes of four gables. This is a survival that has generated many ‘scholarly’ misinterpretations, including a tradition that its height was reduced in 1762, and another, more recent and even less probable, that it was rebuilt after the Conquest. The present state of knowledge does not admit of a date ascription closer than somewhere between c.A.D. 950 and c.1050. It is unlikely to be later than this.

No other Anglo-Saxon tower in this country has a contemporary spire of this type, but there are structural indications that St Benet’s, Cambridge, was originally of the same form; and it is suspected that others were replaced by more conventionally English types right up to the 19th century. The architectural and structural concept of the Rhenish helm is extraordinary, but its execution in carpentry at Sompting is a work of such assurance and competence, achieved with such economy of means, that it both indicates the work of a master and suggests the previous existence of a tradition of framing such works.

The entire framing of the helm is shown in Fig. 12, in which enough components have been omitted to clarify the matter. Until such time as the cladding is removed for repairs there are points of construction that cannot be verified, and these are omitted. They include the framing of the purlins to the eaves rafters, and the ‘valleys’. It is apparent from both structure and situation that the helm had been previously worked and fitted together at ground level, when it was numbered for reassembly with chisel-cut Roman numerals that are still clearly visible. Late Saxon method, therefore, anticipated ensuing carpenters’ methods. The verges of the gables were not finished in the level plane, but were inclined at the same pitch as the surfaces of the spire, for which reason the gables’ truncated tops produced triangular flats.

The whole structure was soundly designed, as its survival proves; and the freestone columns formed by the quoins and the mid-wall pilasters were placed under the heaviest loads, the helm being designed to stand on eight bearings, four at the gable apexes and four at the valleys, and to have a central steady in the crossing of the transoms, in which the spiremast was seated. This system produced a spire that wind pressures have failed to dislodge from its tower. The first unit of framing assembled in the tower top is shown as Fig. 13. The diagonal braces fitted in the horizontal plane would have been fitted next and jointed (as shown in Fig. 14) with chase tenons and, more important, modified lap dovetails in pairs and addressed. This assembly greatly stabilised the four gables and provided locations for the feet of the four principal-rafters and the central spire mast: the latter was lowered through the two square voids formed by the four rising braces and the crossings of the paired straining timbers, when its foot tenon could enter the mortise on the upper transom (Fig. 13).

Structural details of the four gable posts are given as Fig. 15, in which the direct and simple techniques are of interest; for example the arris trenches cut into the posts’ tops to receive the principal-rafters are unique to date in the writer’s experience. When assembly had reached this point the spiremast would have been fitted (the two unrelated kinds of scarfing it shows will be discussed elsewhere), and when positioned it was face-pegged (Fig. 16). This last is a recognisably Saxon technique, and one which can be traced through the ‘Saxo-Norman Overlap’ and into the 13th century. Note the hewn outsets, on the faces of the mast, designed to rest on the tops of the braces. Hereafter the four rafters forming the arrises of the helm were fitted, and then the common rafters, five to each facet, each being made of two lengths. The central rafters which run diagonally down to the lowest points of the helm were wall-anchored (Fig. 17). These are unusual examples, in timber, of functional members that were later made in wrought iron. They presage the base triangulation of common rafters which was to become invariable during the medieval period; and the lap joints used to make them, whilst adequate for their purpose, were unable to resist withdrawal and may be the origin of the later notched lap joints.

Fig. 13 Sompting, the first unit of the timber framing, with ‘free’ tenon inset.

Fig. 14 The lap dovetails used at Sompting.

Fig. 15 Sompting, exploded view of one gable post.

Fig. 16 Sompting, central spiremast section, showing both scarf joints and outsets with face-pegging.

Fig. 17 Sompting, a wall anchor, showing lap joints.

Fig. 18 Sompting, perspective of tower floor.

The framing of the floor beneath the bell chamber is also of interest (Fig. 18), and appears to be original to the tower. Unlike other floors this was laid upon four bridging-joists aligned north to south, and spaced in such manner as to leave a central interstice equal to the diameter of the largest intended bell. Across these, and apparently trenched into their upper faces, were laid two more joists of a cross section intermediate between the bridging-joists and the common joists. The latter, insofar as has yet been determined, were located in cross coggings.

THE CHURCH OF THE HOLY TRINITY, COLCHESTER, ESSEX

The western tower of this church has many features indicating a date shortly before the Conquest (G. Baldwin Brown, 1925, 447; H.M. and J. Taylor, 1965, 143-5). The tower and its internal timber work make a confusing spectacle from within, giving the impression that the 18th-century timbers were inserted to support the floors; but closer examination proves that they were added to provide vertical stability to the rubble walls, which have dressings of brick, some of which must be Roman. The points at which the joists enter the walls seem mainly undisturbed, and if this is accepted then the floors must be considered original and late Saxon. Their framing does not resemble the Sompting example and has one heavy bridging-joist from east to west, across which are laid eight common joists. All timbers are squared to Roman imperial measurements, the biggest being 12ins square, with a 2in stop chamfer along its soffit arrises, whilst the common joists measure 7½ins deep by 8ins wide – a section presaging the later medieval use of timber laid ‘flat’ instead of on-edge. This floor merits no illustration, being simple, as described.

Fig. 19 Hadstock, internal view of north door.

THE CHURCH OF ST BOTOLPH, HADSTOCK, ESSEX

Recent excavations carried out by Dr W. Rodwell have established that this church has passed through at least three phases of building, the last of which was during the 11th century. The fabric retains at least five pieces of Saxon carpentry – four mid-wall window frames and its famous north door leaf. The latter has long been accepted as Anglo-Saxon because it is Romanesque in style, but has no structural affinities with Norman door-leaves, a larger number of which have been identified. This is shown as Fig. 19. It was constructed from four wide planks which must have been seasoned at the time of use, because the long edge joints spaced by ironwork have never opened due to shrinkage; the joints are splayed rebates. The ledges forming the rear frame are made of D-sectioned oak, the top one being bent to the arcature of the door’s head. The whole is fastened together by means of iron roves and clenches, the roves being so elongated as to encircle the wood and prevent it splitting when the clenches were formed (Fig. 20). These techniques all derive from shipwrighting, and have survived in use until today, when they may be seen in the building of clincher-built small craft.

Fig. 20 Exploded view of the Hadstock door assembly.

Fig. 21 Hadstock, construction of mid-wall window frames.

The mid-wall window frames, four of which survive to be seen, are high up in the double splayed openings of the nave walls; they are the only recorded examples of the kind that were constructed from four carpentered timbers, jointed together and pegged (H.M. and J. Taylor, 1965, 676-7). One of these is drawn as Fig. 21. They exhibit carpentry that was skilled, and each have two chamfered stiles, a sill and an arched head timber, assembled by stub tenons transfixed with two pegs each.

The head-and-sill timbers penetrate the rubble walls for a considerable distance, as indicated by dotted lines in Fig. 21. I am indebted to Dr W. Rodwell for this information. The pegs are of the curious section illustrated at left and right of the drawing. The cutting of such holes suggests the use of three tools: an auger, a scribing gouge of the sides’ curvature, and a second scribing gouge to cut the rounded ‘corners’. A close study of this work proves that these tools were of good steel, very sharp, and handled by a craftsman proficient in their use.

THE BARN AT PAUL’S HALL, BELCHAMP ST PAUL, ESSEX

This is one of the ancient manors of the Chapter of St Paul’s in London, regarding which remarkably detailed farm leases have been preserved. Some of the leases – no more than a sample – were published by Archdeacon Hale, in the case of this holding a lease from the time of Dean Hugh de Marney, 1160-81 (W.H. Hale, 1858, 138-9). The surviving barn is the last of three that existed within living memory; it is aligned from north to south and sited west of the house, whereas the lease describes only two barns, both of which were aligned east to west. The given measurements in the lease, however, are not irreconcilable with this barn, which may therefore have been moved when rebuilt in c.1200, as there is structural evidence to show. It is this survivor that incorporates the features most important for the historian of timber building.

The whole has been rebuilt more than once and greatly extended in length, some of the later works including re-used timbers, as a result of which it contains a diverse series of ‘developmental’ features. At its northern end are what appear to be the remains of a three-bay building that used structural principles more archaic than any hitherto recorded. The only undamaged post at this end stands closest to the southwest angle of the barn, and is earth-based upon a lime/cement pad placed 6ins beneath the floor. Its foot is completely independent of the side wall behind it. This post is scale-drawn as Fig. 22, wherein absent components are drawn in chained line. It has been established by a small excavation that the post was moved at an early date and re-erected in its present position. It seems from a packed layer of pebbles that at that time the remnant of the earthfast shore, its foot rotted off, had been left in situ. A carbon14 date was obtained for this post (for which I am indebted to B.B.C. television’s Chronicle programme) and the determined age was 924 years, plus or minus 95 years, before A.D. 1950. The central date is therefore A.D. 1026, and the dating range is 931 to 1121. It can be proved that the existing tie beam is later than the post because the empty mortises in the post have no equivalents in the beam, which together with the existing top plate point to major rebuilding, c.1200.

This example provides a pair of late Anglo-Saxon posts belonging to an aisled timber building (which was probably a barn since it was never sooted), one of which retains a basal standing of a probably pre-Conquest type. It suggests the existence of elaborate aisled buildings which possessed many of the structural features, such as passing braces, that were to predominate for two centuries after the Conquest. Furthermore, such structures if razed from their sites would have left evidence neither of their elaborate designs nor of the skills of their carpenters.

Fig. 22 The barn at Paul’s Hall, scale-drawn rear and side elevations of a post.

Fig. 23 The Westminster Abbey Saxon door leaf.

DOOR LEAF IN WESTMINSTER ABBEY, LONDON

This is illustrated in Fig. 23